O:

li

ru
m

-D

.-q

o
o

-=1
a
m
a

/

SECOND SEEIES: PULMONATA.

MANUAL

OF

CONCHOLOGY

STRUCTURAL AND SYSTEMATIC

WITH ILLUSTRATIONS OF THE SPECIES

VOL. XXII
ACHATINELLID^E

BY

HENRY A. PILSBRY

ASSISTED BY

C. MONTAGUE COOKE j

GENEALOGY AND MIGRATIONS OF THE ACHATINELLID^J

BY

ALPHEUS HYATT

PHILADELPHIA :
Published by the Conchological Department

ACADEMY OF NATURAL SCIENCES OF PHILADELPHIA

1912-1914

PUBLICATION COMMITTEE

SAMUEL GIBSON DIXON, M. D. , LL.D. , President of the Academy.
HENRY SKINNER, M. D., Sc. D., WITMER STONE, A. M., Sc.D.,.

HENRY A. PILSBRY, Sc. D., WILLIAM J. Fox,

EDWARD J. NOLAN, M.D.
Treasurer : S. RAYMOND ROBERTS.

A

CONTENTS.

PAGH

PREFACE v

INTRODUCTION

Achatinellidos defined and compared with other families. ix

Structure and interpretation of the teeth x

Classification xiii

Color-patterns xix

Shape and structure of the shell xxiv

Conceptions of species and minor races xxv

Habits, stations, conditions of collecting, etcetera .... xxxiv

Supplementary observations on Hawaiian zoogeography. . xli

Remarks on the synonymy of Achatiriettidce 1

Historical notes on the literature of Achatinellidos .... lii

SYSTEMATIC PART

Genus NEWCOMBIA Pfeiffer 1

Genus PARTULINA Pfeiffer 14

Section Perdicella Pease 15

Section Partulina s. str 23

Section Eburnella Pease 67

Section Baldwinia Ancey 90

Observations on Hawaiian Achatinellidse, by H. W. Henshaw. 91

Genus ACHATINELLA Swainson 117

Section Bulimella Pfeiffer 118

Section Achatinellastrum Pfeiffer 180

Section Achatinella s. str 274

APPENDIX I. Additional notes, undefined species etcetera . . 355

APPENDIX II. Genealogy and migrations of the Achatinellidse

in the Hawaiian Islands, by Alpheus Hyatt 370

EXPLANATION OF PLATES . . , 400

INDEX 423

DATES OF ISSUE OF THE PARTS OF VOLUME XXII 428

(iii)

PREFACE.

This volume treats of the tree snails of the Hawaiian Islands
belonging to the family Achatindlidae ; those of the family Amas-
tridas (the genus Laminello) having been dealt with in Volume
XXI, and arboreal forms of Tornatettinida (Auriculella) remain-
ing to be considered in Volume XXIII of this series.

The Achatinellidse hold an important place in general biological
literature for the reason that a contribution of the first importance
to the doctrine of evolution arose from a study of these snails.
Gulick's theory of "divergent evolution through cumulative seg-
regation ' is now generally recognized as setting forth one of
the conditions invariable for the "origin of species,' 1 having no
necessary connection with natural selection, Lamarckian factors
or mutation theory, but everywhere essential to speciation. For
this reason, among others, the account of the group has been
made somewhat more elaborate than usual in purely systematic
treatises, in order that zoologists may be in a position to form
intelligent ideas of the facts of distribution and variation of the
group. It will readily be understood that having to cover the
whole field, it was impossible to enter upon a detailed study of
any single group of forms in a restricted area. The necessary
limits of this work permit only brief discussion under each
species. For an exposition of the facts bearing on evolution in
general, a far more limited field should be chosen, two or three
adjacent valleys, where the data could be made nearly complete.
Even so, a certain amount of breeding under control would be
essential to solid conclusions. As matters now stand, we infer
where we should have exact knowledge.

The name of Dr. C. Montague Cooke is placed upon the title-
page of this volume in recognition of his collaboration in eluci-
dating the involved synonymy of the genus Achatinella. To the
systematist this is perhaps the more important work of the vol-

v

VI PREFACE.

ume. Indeed, the main features of this synonymy (excepting so
much as had been recognized by earlier naturalists) were original
with my colleague, having been worked out before we went over
the ground together in Honolulu. While for the accuracy of the
synonymy the senior author assumes equal responsibility with
Dr. Cooke, it would be unfair to hold the latter responsible for
the manner of presentation, or for the treatment of questions left
in abeyance for want of time or type material in Honolulu. It
was not practicable for my collaborator to go over the work in its
final form; and for this reason many statements of fact or opinion
throughout the text are made in the first person, especially when
based upon material we had not gone over together. The determi-
nation of all forms of Achatinetta believed to be new to science
was also the joint work of both authors, as indicated in the text.
Indebtedness to Dr. Cooke for various helpful suggestions rela-
tive to Partulina and other genera is acknowledged under the
respective species. I may also be permitted to recall the com-
radeship which made the months spent in the Islands among
the happiest of my life.

Information used in compiling the maps on pages 183, 277,
and 341 was largely supplied by Dr. Cooke and Mr. Spalding.

Professor Hyatt's MS. descriptive of tree snails was not utilized
in the preparation of this volume for the reason that his treat-
ment of species could not be reconciled with the views herein
expressed. Moreover, it is believed that had Hyatt visited the
Islands and seen the actual association of forms, his ideas must
have been very materially changed. Professor Hyatt's general
results in phylogeny and zoogeography, so far as worked out at
the time of his death, are given in an appendix ( pp. 370-399),
for comparison with the views developed in course of this study.

In order to keep this volume within reasonable size, an ac-
count of the soft anatomy of Achatinelloid snails and an appendix
to volume XXI (Amastrida) will form the first part of Vol.
XXIII. That volume will also treat of Auriculclla, Tomatellina
and their allies, and contain a general inaex of all Achatinelloid
snails.

The color nomenclature follows Ridgway's Color Standards and
Nomenclature, except in the first two numbers, which were pre-
pared before the publication of that work.

PREFACE. Vll

There remains the pleasant duty of thanking many friends
who assisted in the preparation of this book. Acknowledg-
ments are due the President and Trustees of the Bernice Pauahi
Bishop Museum, who aided the author's work in the Islands by
a generous grant. The Director and scientific staff of the
Museum furthered the investigation at every point.

Messrs. Irwin Spalding of Honolulu and D. Thaanum of
Hilo, Hawaii, placed their knowledge of Hawaiian shells and
their splendid collections at the disposal of the author with such
generosity as one meets but rarely in a lifetime.

To Professor Wm. Alanson Bryan, Messrs. R. A. Cooke, J. S.
Emerson, Hon. L. A. Thurston, Mr. W. D. Wilder and others
mentioned in the text, the author is deeply indebted for speci-
mens used in this work, and for other favors which aided inves-
tigations.

The open-hearted hospitality of Hon. George C. Cooke of the
Molokai Ranch enabled us to see far more of the Molokai fauna
in a limited time than would have been possible under less
ideal conditions.

Mr. E. A. Smith assisted with various notes on types in the
British Museum, many of them figured for the MANUAL, under
his critical eye.

Mr. H. W. Henshaw prepared an account of his observations
on Hawaiian Partulinas (pp. 91-105) containing much impor-
tant material.

Mr. C. W. Johnson lent a series of Gulick's type-specimens
from the Boston Society of Natural History. Like favors were
granted by Mr. Samuel Henshaw of the Museum of Compara-
tive Zoology, where Pease's types are preserved, and by Pro-
fessor G. D. Harris of Cornell University, custodian of the
Newcomb collection.

Finally it should be said that the production of this work
would not have been possible without the encouragement and
support of the President of the Academy of Natural Sciences,
Dr. Samuel G. Dixon, and of the Publication Committee, with
Mr. S. Raymond Roberts, having the work in charge.

My warmest thanks to all of these friends and colleagues.

H. A. P.

PHILADELPHIA, April, 1914.

INTRODUCTION.

Family ACHATINELLIDJ3 Tryon.

Achatinellida Tryon, 1884 (exclusive of Tornatellinida and

Amastrida).

Orthurethra with the kidney longer than the pericardium,
oblong and narrow. Surface of the lung plain. Genitalia
complicated by the presence of a long appendix provided with
a branch from the penial retractor muscle. Spermatheca em-
bedded in the prostrate gland, its duct long and simple ; albu-
men gland minute; prostate gland enormously developed,
composed of vermiform caeca. Jaw wanting or extremely thin,
indistinctly plaited vertically. Radula broad and short.
Teeth excessively numerous, arranged in V-shaped transverse
rows ; rake-like, composed of a narrow basal plate and a broad
recurved portion bearing numerous unequal cusps.

The shell is more or less glossy, dextral or sinistral, oblong
or ovate, minutely perforate or closed, composed of five to
seven whorls, the last convex ; aperture ovate ; outer lip thick-
ened or simple, sometimes expanding a little; columella usu-
ally bearing a strong, obliquely entering lamella in the last
whorl; internal axis sinuous. No entering lamella or tooth
on the parietal wall. Animal externally as in Helicidce. Foot
moderately broad, usually shorter than the shell. Viviparous..

Distribution, the Hawaiian Islands (except Kauai, Niihau
and perhaps Kahoolawe). Living on trees and other plants.

Studies of the soft anatomy with abundant material have
demonstrated, in my opinion, that the Achatinellidcz and
Amastrida are strongly distinct families. None of the many
species of both, dissected by other naturalists or by myself,
show the slightest approach in their differential characters.

The Achatinellidce in essential structure stand nearest to
the Amastridcr, Enidce and Ferussacida, and are somewhat

(ix)

X TEETH OF ACHATINELLID^E.

more remotely related to Partulidce. The structure of the
pallial organs and kidney is very similar in the first three of
these families. The male reproductive organs also, while dif-
fering in details, are much alike, having features widely pre-
valent in the lowest land snails, also in Onchidium, etc.

Partula, by lacking an appendix on the penis, and also by
the simple prostate gland and the somewhat different form
of kidney, stands a little apart.

By the structure of the female reproductive system, the
AchatinelUdce are most nearly akin to the Amastridce, but are
more evolved by the reduction of the albumen gland to a
minute rudiment, the prostate being at the same time enorm-
ously enlarged. The difference between the two families in
this character is one of degree rather than of kind.

From all other families mentioned above, the Achatinellidcz
differ strikingly in dentition. Those families have teeth of
normal shape, with cusps resembling and homologous with
these of the Helicidce, Lymnccidce, and most other Pulmonate
snails. The teeth of Achatineltidce and fornatellinidce differ
strikingly by having basal-plates of a different shape ;
and their cusps are not recognizably homologous with those
of other land snails. Moreover, there is no division into
lateral and marginal areas of the radula with different forms
of teeth.

Morphology of Achatinellid teeth.

It is well known that arboreal snails usually have the teeth
specially modified, more or less unlike those of their nearest
relatives among ground snails. Liguus and Polymita are not-
able examples, but in many others, such as Drymaus, Papuina,
etc., a process of change is evidently in progress. In all of
these genera, the cusps, though variously changed, are still
clearly homologous with those of normal teeth. Without com-
mitting ourselves to any special hypothesis of how such
changes come about, we cannot avoid the conclusion that in
some way the nature of the food or the conditions of grazing
lead to adaptive remodelling of the teeth in arboreal snails.
So long as the modification does not affect the homologies of

TEETH OP ACHATINELLID^. XI

the cusps, it is not necessarily of much systematic importance.
The amount of change in various groups is apparently some
indication of the length of time since arboreal habits were
assumed. In Achatinellidce and Tornatellinidce the modifica-
tion of the teeth is wholly unlike that usual in arboreal
snails of other families, in which the cusps are always broad-
ened. The Achatinellid type is really a further development
of the multicuspid teeth of small ground snails, and it would
seem, not an adaptation induced by arboreal life, though it
has obviously proved efficient in that station, in the absence
of any competitors.

The multicuspid cutting edges of the teeth, and the absence
of differentiation into laterals and marginals, cause me to
view the Achatinellid radula as one from which central and
lateral teeth have been eliminated, leaving only marginals.
The marginal teeth of Amastrida and Enidce, as of many
other snails, stand in somewhat oblique transverse series. If
the median field was eliminated, the transverse rows would
be broadly V-shaped. Pachnodus in the Enidcz, as figured
by F. Wiegmann, has somewhat such a radula as that we may
suppose the ancestors of Achatinella had. The teeth of
Pachnodus are very numerous, 375 to 393 in a row, the mar-
ginal fields have the rows strongly oblique, and the marginal
teeth are multicuspid. The teeth of Achatinella differ from
the marginals of Pachnodus, or of Leptachatina, chiefly by
the long and narrow basal-plate, whereas in most marginal
teeth this plate is very short and broad.

By a similar reduction, some species of Mesomphix have
practically eliminated the lateral teeth, while others retain a
few of them. The closely related Omphalina has the usual
development of laterals. The Agnatha and Agnathomorph
snails also have lost all laterals.

I believe, therefore, that all the teeth of AchatinelUdce have
been derived from the marginals of some unspecialized group
of ground snails having multicuspid marginal teeth. This
ancestral group was evidently also ancestral to the Amastrida,
in which Leptachatina still retains multicuspid marginal
teeth; but even the most primitive existing Amastrida have

Xll FOSSIL ACHATINELLHXE.

so far specialized the dentition that it seems to me quite im-
possible that Achatinella could have been derived from any
Amastrine genus now living.

The teeth of Physidce may be a similar evolution from the
marginal teeth of some Basommatophorous stock. They have
a remarkable superficial resemblance to those of the Acha-
tinellidce. The Athoracophoridce have also evolved teeth won-
derfully similar to those of Achatinella. This is evidently
a case of convergent evolution, as the relationship must be
remote.

Notwithstanding its strong differential characters, the
family is a very compact one, with remarkably little struc-
tural variety for a group so prolific in species. Most genera
of Helices show much greater specific diversity in characters
of the soft parts than I have been able to find in the whole
family of Achatinellidcz. There has been no adaptive radia-
tion, and with the possible exception of Newcombia, there
are no aged or phylogerontic branches.

This contrasts with the Amastridce, which have been adapted
to a variety of stations, humid, semiarid and arboreal, and in
which the shell varies from cylindric or turrited to discoidal,
with wide diversity in sculpture, color and solidity.

It appears that the AchatineUidcs are a still youthful group
in the full flower of their evolution, probably derived from
some plain terrestrial form which was rejuvenated by the dis-
covery of a new station the leaves and branches of trees.

Fossil Achatinellidcz known up to this time are all of
Holocene, or perhaps in part of Pleistocene age. They are
far less numerous than fossil Amastridcz, probably because the
deposits wherein land shells are preserved lie mainly close
to the sea, while Achatinellidc? are shells of the mountain
forests. The known fossil species follow.

Achatinella phaeozona Gul. Kailua (p. 187).

Achatinella cccsia littoralis P. & C. Kahuku (p. 266).

Partulina montagui Pils. Manoa Valley (p. 66).

Partulina divighti occidentalis P. & C. Molokai (p. 361).

Newcombia philippiana Pfr. Molokai (p. 356).

Partulina confusa Sykes. Hawaii (p. 105).

CLASSIFICATION.

Partulina montagui is a very distinct species, but not in the
least primitive in character. Two other extinct . forms are
distinguishable races of living species, while the other three
found fossil do not seem to differ from modern shells.

Further extinct species are to be expected in comparatively
high beds, such as the Manoa and Palihoukapapa forest de-
posits. Many such must exist, and they may afford a good
deal of light on such obscure questions as the part taken by
Partulina in the ancient Oahuan fauna.

CLASSIFICATION OF ACHATINELLID^E.

It has been stated above that the genera and subordinate
groups of Achatinellidee are based entirely upon characters
of the shell. The soft anatomy, so far as known, is practically
alike in all. The genera are not groups of much systematic im-
portance not more I should say, than the "sections" estab-
lished in my monograph on Partula; but they are obviously
natural groups, which have been found useful in dealing
with long series of species.

The family divides primarily into two groups, Partulina
and Achatinella. By their coloration and sculpture, New-
combia, Perdicella and Baldwinia are obviously derivatives
of Partulina. The Partulina group is more numerous in
species and far more varied in structure than Achatinella,
which consists of three very closely related sections. Ebur-
nella is a group of uncertain affinities, but apparently linked
to Partulina by certain species of Maui and Lanai. The ap-
proximate relationships of the groups are represented in the
following diagram.

Newcombia

Baldwinia Perdicella

| Bulimella

Partulinella

Partulina s. sir.

Eburnella Achatinellastrum

(Partulina) (Achatinella} Achatinella 8. sir'

Ancestral
Achatinellid

XIV CLASSIFICATION.

Key to Genera and Sections of Achatinellida.

a. Columella straight or only weakly folded.

6. Shell slender, turrite, usually sculptured; 3% em-
bryonic whorls, having coarse spirals, the last obli-
quely striped. NEWCOMBIA, p. 1.
6 1 . Shell oblong or ovate.

c. Shell striped or tessellated, sometimes banded,
small, usually less than 17 mm. long.

Perdicella, p. 15.

c. 1 Shell zigzag striped, banded or white, usually
well over 17 mm. long, or if so small, the last
whorl is unusually large. Baldivinia, p. 90.
a 1 . Columella strongly folded.

6. Shell dull or polished, usually with distinct spiral
striation ; the embryonic shell biconic, often striped ;
usually two or several embryos in the uterus at one
time. PARTULINA, p. 14.

c. Shell small, tessellated or striped, columellar
fold thin. Perdicella, p. 15.

c 1 . Shell larger, spirally striate, columellar fold
stout. Partulina, s. str., p. 23.

c 2 . Shell polished, often banded, columellar fold
stout. Eburnella, p. 67.

6 1 . Shell polished; spiral striation weak or almost want-
ing; embryonic shell short and broad, its last whorl
never striped obliquely; only one well-developed
embryo in the uterus at a time. Oahu :

ACHATINELLA, p. 117..

c. Outer lip thickened by an internal callous
ridge ; shape oblong-conic or ovate, summit
obtuse. Bulimella, p. 118.

c 1 . Outer lip but little or not thickened within,
not expanded; summit conic, the embryonic
whorls not conspicuously flattened.

Achatinellastrum, p. 180.

c a . Outer lip thickened within ; shape globose-
conic ; embryonic whorls almost flat.

Achatinella, s. sir., p. 274.

CLASSIFICATION. XV

Systematic List of species and subspecies of Achatinellidce.

Numbers following the species refer to pages where they
are described. In Partulina, where there are several col-
lateral phyla, no linear arrangement can be made which will
show the affinities with Newcombia and Achatinella, and at
the same time preserve a logical sequence of groups and
species within the genus.

Genus NEWCOMBIA Pfeiffer.

(Molokai) (Molokai)

N. plicata High., Pfr. 2. N. pfeifferi Newc. 13, 355.

N. p. gemma Pfr. 3. N. p. ualapuensis Pils. 12.

N. sulcata Pfr. 5. N. p. cinnamomea Pfr. 10.

N. canaliculata Bald. 6. N. p. decorata Pils. 12.

N. c. wailauensis Pils. 7. N. p. honomuniensis Pils. 12.
N. philippiana Pfr. 8, 356. (Maui)

(+ perkinsi Sykes) N. cumingi Newc. 10.

Genus PARTULINA Pfeiffer.

Section Perdicella Pease.

(Molokai) P. mauiensis Pfr. 20.

P. helena Newc. 16, 356. P- zebrina Pfr. 20.

P. fulgurans Sykes 21.
(Maui) p. carinella Bald. 7, 358.

P. ornata Newc. 18. P. thwingi P. & C. 357.

P. zebra Newc. 19. P. kuhnsi Pils. 22.

Section Baldwinia Ancey.
(Hawaii)

P. confusa Sykes 105. P. h. fuscozonata P. & C. 365..

P. horneri Bald. 107, 365. P. physa Newc. 109.

P. h. Candida P. & C. 365. P. p. errans Pils. 111.

P. h. fuscospira P. & C. 365. P. p. konana P. & C. 365.

(Maui)
P. grisea Newc. 111. P. thaanumiana Pils. 112.

XVI CLASSIFICATION.

(Oahu)
P. dubia Newc. 113.

Section Partulina Pfr., s. sir. (Molokai).

P. virgulata High. 25, 358. P. v. kaluaahacola P. & C. 359.
P. v. halawaensis Borch. 27, P. subpolita Hy. & Pils. 359.
359.

Section Partulinella Hyatt (p. 392).
(Molokai)

P. tessellata Newc. 28, 360. P. theodorei Bald. 33, 360.

P. t. meyeri Borch. 29. P. dwightii Newc. 35, 360.

P. rufa Newc. 29, 360. tP. d. occidentalis P. & C. 361.

P. proxima Pse. 32, 360. P. d. mucida Bald. 34, 361.

P. p. schauinslandi Borch. 33. P. redfieldii Newc. 38, 362.

P. p. multistrigata Pils. 34, P. r. kamaloensis P. & C. 362.
360.

(Oahu)
t P. montagui Pils. 66.

(Lanai)
P. crassa Newc. 40, 362.

(Haul)

P. kaaeana Bald. 41. P. tappaniana C. B. Ad. 54.

P ustulata Gul. 47, 362. P. t. ampulla Gul. 57.

P. marmorata Gld. 42. P. t. eburnea Gul. 57.

P. plumbea Gul. 43. P. t. carnicolor Bald. 58.

P. winniei Bald. 44. P. nivea Bald. 59, 363.

P. perdix Rve. 45, 363. P. dolei Bald. 60, 364.

P. induta Gul. 48. P. lemmoni Bald. 61.

P. radiata Gld. 49, 363. P. terebra Newc. 61, 364.

* * * P. t. attenuata Pfr. 63.
P. splendida Newc. 51, 363. P. t. longior Pils. 63.
P. s. baileyana Gul. 52. P. t. lignaria Gul. 63.
P. gouldii Newc. 52. P. fusoidea Newc. 64.

* * P. crocea Gul. 65.

CLASSIFICATION.

XV11

Section Eburnella Pease.

(Maui)

P. mutabilis Bald. 68.
P. porcellana Newc. 69.
P. p. flemingi Bald. 71.
P. p. wailuaensis Sykes 72.

P. variabilis Newc. 83.
P. v. lactea Gul. 86, 364.

P. mighelsiana Pfr. 77.
P. m. bella Eve. 79.

P. p. fulvicans Bald. 73.
P. nattii Bald. & Hartm. 73.
P. anceyana Bald. 75.
P. germana Newc. 76.

(Lanai)

P. semicarinata Newc. 86.
P. s. hayseldeni Bald.

(Molokai)

P. m. polita Newe. 80.

Genus ACHATINELLA Swainson.

Section Bulimella Pfeiffer (Oahu).

A. bulimoides spadicea P.&C.

A. abbreviata Rve. 123.

A. viridans Migh. 125.

A. tasniolata Pfr. 130.

A. byronii Wood 133.

A. b. rugosa Newc. 135.

A. b. waimanoensis P.&C. 137.

A. b. nigricans P. & C. 138.

A. lila Pils. 139.

A. pulcherrima Sw. 140.

A. p. nympha Gul. 144.

A. decipiens Newc. 145.

A. d. planospira Pfr. 147.

A. d. kaliuwaaensis P.&C. 150.

A. d. swainsoni Pfr. 150.

* * *

A. rosea Swains. 151.
A. bulimoides Sw. 154.
A. b. mistura P. & C. 156.

157.

A. b. obliqua Gul. 158.
A. b. ovata Newc. 160.
A. b. rotunda Gul. 163.
A. b. glabra Newc. 164.
A. elegans Newc. 166.
A. e. wheatleyana P. & C. 168.

A. fuscobasis Sm. 170.
A. f . lyonsiana Bald. 172.
A. f . wilderi Pils. 173.
A. pupukanioe P. & C. 174.
A. sowerbyana Pfr. 175.
A. s. thurstoni P. & C. 177.
A. s. laiensis P. & C. 178.
A. s. dextroversa P. & C. 179
A. s. roseoplica P. & C. 180.

XV111

CLASSIFICATION.

Section Achatinellastrum Pfr.
(Main range of Oahu)

A. phseozona Gul. 184.
A. buddii Newc. 187.
A. fulgens Newc. 190.
A. f . versipellis Gul. 196.
A. f . ampla Newc. 198.
A. solitaria Newc. 204.
A. stewartii Green 204.
A. s. producta Rve. 207.
A. vulpina Fer. 212.
A. v. colorata Rve. 224.
A. v. tricolor Smith 226.
A. bellula Smith 230.
A. b. multizonata Bald. 231.
A. casta Newc. 235.

A. casta margaretae P.&C. 240.

A. juncea Gul. 241.

A. papyracea Gul. 243.

A. juddii Bald. 244.

A. livida Swains. 246.

A. 1. emersoni Nc. 247.

A. 1. recta Nc. 248.

A. 1. herbacea Gul. 251.

A. curta Newc. 252.

A. dimorpha Gul. 258.

A. cassia Gul. 263.

t A. c. littoralis P. & C. 266.

A. c. cervina Gul. 267.

A. c. cognata Gul. 267.

(Waianae range)
A. spaldingi P. & C. 271. A. 1. gulickiana P. & C. 273.

A. lehuiensis Smith 271.

A. thaanumi P. & C. 273.

Section Achatinella Swains., s. sir.
(Main range of Oahu)

A. lorata Fer. 278.

A. 1. nobilis Rve. 283.

A. 1. pulchella Pfr. 284.

A. cestus Newc. 286.

A. vittata Rve. 289.

A. v. cinerea Sykes 291.

A. v. simulans Rve. 292.

A. turgida Newc. 294.

A. t. perplexa P. & C. 296.

A. t. ovum Pfr. 297.

A. t. simulacrum P. & C. 299.

A. t. cookei Bald. 300.

A. leucorraphe Gul. 301.

A. 1. irwini P. & C. 302.

A. swiftii Newc. 306.

A. s. chromatacrne P.&C. 316.

A. s. dolium Pfr. 316.

A. apexfulva Dixon 317.

A. a. vespertina Bald. 322.

A. a. alba Sykes 324.

A. a. apicata Newc. 324.

A. a. beata P. & C. 329.

A. a. aloha P. & C. 330.

A. decora Fer. 331.

A. valida Pfr. 334.

A. v. leucophaea Gul. 336.

A. v. cinerosa Pfr. 336.

A. v. kahukuensis P. & C. 338.

DISTRIBUTION, COLOR PATTERNS.

XIX

(Waianae range)

A. mustelina Migh. 342. A. concavospira Pfr. 351.

A. m. sordida Newc. 349. A. c. turbiniformis Gul. 353.

A. m. lymaniana Bald. 350.

Incertce sedis.
A. aptycha Pfr. 54, 145, 363.

DISTRIBUTION OF GENERA, SUBGENERA AND SPECIES.

PARTULINA

ACHATINELLA

cc

T3

C

ISLANDS

!

hH

a

c6

^
'M

M

g

cs

c^

.S

ce

c3

3

1 i

<p

>^

o

'cu

fl

.rH

3

I-H

'o

iD

4*3

^H

&

O
*

o

rH

r W

1 1

C3

~Ui

T3

rt

*H

S

a

c8

v

^

-M
CJ

J3

te
'ct

-4-i

W

<D

eg

ei

^0

^

o

O

K

PH

PH

PQ

w

pq

<^

<5

EH

"NJiihau

Kauai ... ...

Oahu . . .

1

1

13

17

11

43

Molokai

5

1

8

1

15

I f\ i~ O -|

1

2

3

JVIaui

1

8

14

2

6

31

Kahoolawe .....

Hawaii

3

3

Total no. of species.

6

9

24

6

9

13

17

11

95

N. B. No species is common to two islands.

COLOR PATTERNS OF ACHATINELLID^:.

The primitive color pattern of arboreal Achatinellidtz may
have consisted of dark streaks in the direction of growth-
lines, on a light ground. This pattern is common to many
species of various genera of the family, and to some species
in practically all the minor groups. It is one of the common
patterns of coloration of snails in many families to have the
pigment thus deposited periodically instead of continuously.
In many Partulinas the streaks have become distinctly denned,
and lose their primitive relation to the growth-lines, becoming
zigzag or oblique, or they may be interrupted to form a tessel-

XX COLOR PATTERNS.

lated, marbled or mottled pattern. This is a line of pattern
evolution peculiar to Partulina, at its acme in such species as
P. proximo, and the Perdicellas, decadent in many Partulinas
and Newcombias.

In Achatinella the streaks, when present, are not sharply
defined and always run in harmony with growth-lines.

The spiral bands of Achatinellidcu belong to two band-
systems, entirely separate in inheritance. The genesis of
the irregularly placed bands is readily traceable in existing
species, but the origin of those of the four-band pattern is
unknown.

Irregularly placed bands. Many species which have spiral
bands in the adult stage, are streaked when young. The
several stages from the streaked to the banded pattern char-
acterize various races or species ; so that it appears likely
that the continuous-banded pattern was never produced by
a single mutation from the streaked, but by a series of pro-
gressive or orthogenetic mutations. The sequence seems to
be somewhat thus :

(a) longitudinally streaked (pi. 60, figs. 4a-4c).

(b) streaks cut by light spiral lines or bands, leaving
streaked zones or spotted spiral lines (pi. 60, f. 6; pi.
56, f. 56).

(c) color of the dark zones or bands intensified or diffused,
losing the streaked or spotted pattern (pi. 56, f. 5e, 5/).

Many individuals are intermediate between a and b, or b
and c. Often instead of several or many pale or white bands
appearing, the upper half of each whorl loses the streaked pat-
tern, leaving it on the portion below the periphery, producing
a bicolored pattern of cuticle, as in pi. 57, figs. 2&, 15. This
is merely a special modification of stages & or c.

Judging from its sporadic occurrence, albinism may ap-
pear as a mutation at any stage of pattern-evolution; melan-
ism is perhaps more frequent in the later stages.

All stages from streaked to banded may be seen in some
species, such as Achatinella fulgens and Partulina redfieldi.
May species remain in the primitive stage, while in others
only the terminal pattern is seen.

COLOR PATTERNS.

The zoned or many-banded patterns are a higher (more
evolved) stage; the streaked are lower in the scale; but from
the co-existence of all stages in many species, it appears that
the evolution of one from the other is a rapid process.

The transition from longitudinal streaks to spiral bands in
these shells is entirely analogous to the transition from longi-
tudinal to transverse stripes in mammals. It is a common
phenomenon in mollusks, and examples can readily be found
in most faunas, exactly parallel to the process in Ackatinella.
Cf. Drymceus, Amphidromus, Helicostyla.

Spiral bands or lines, resulting from the evolution of stripes,
as described above, are indefinite and variable in number and
position.

The four-band pattern. - - The black-brown bands, which
stain the prismatic layer of the shell and often appear as spots
on the lip, have perfectly definite positions in Achatinellas-
trum, Bulimella, and a few Partulinas, when they occur at
all, and are evidently homologous throughout these groups.
See plate 38. These bands are four in number: 1, sub-
sutural ; 2 and 3, above and below the periphery ; and 4
around the columella. All are developed in pi. 38, figs. 4a,
6, 6a, 14. Often only bands 1 and 4 are developed, figs. 1,
la, 2-4, etc. Other forms may have bands 2 and 3, or various
other combinations. The bands of the four-banded or ' ' tetra-
tseniate ' ' system may be expressed by a formula, such as is
used for the five-banded or so-called pentataeniate Helices.
Thus pi. 38, fig. 6 has the formula 1234. Fig. 7, 0000. Fig.
7c, 0230. Fig. 18, 1(234). Fig. 19c, 1(23)0. In Acha-
tinellastrum a particularly common formula is 1000. As in the
pentatasiiiate Helices of Europe, the bands vary through the
several possible combinations in presence or absence, and in
width when present; they are sometimes split, or confluent;
but their positions are invariable.

In inheritance, bands of this system are entirely independ-
ent of the variable or indefinitely-placed bands. They may
exist in combination with any stage of the other pattern,
from plain or streaked to many-banded; and even in melan-
istic shells they may be apparent as bands of slightly different
luster, visible only in a certain light.

XX11 COLOR PATTERNS.

I have elsewhere noted that in Liguus some forms show
an extremely interesting change in the pattern, whereby the
dark and the light markings exchange places, as in a photo-
graphic negative, without other noticeable change in the shell.
An entirely similar exchange is seen in some individuals of
certain AcTiatinellidcc . Thus on plate 19, it will be seen that
fig. 10 has the normal, tetratamiate pattern; fig. 5 is a "nega-
tive ' ' of it, the black bands occupying the places of the white
ones in the other shell. A. crassidentata, pi. 30, fig. 23, is a
similar ' ' negative ' pattern. In Partulma, P. porcellana has
the normal four-banded pattern, while P. nattii has the
"negative' pattern.

The typical group of Achatinella (Apex) never has the
four-banded pattern. This is one of the most important dif-
ferences between this group on the one hand, and Bulimella
and Achatinella strum on the other.

Parallel and convergent evolution of patterns. Since
many species have forms with primitive streaked pattern and
others with distinct bands, it appears that bands have many
times been evolved independently. Occasionally this has
resulted in the production of strikingly similar varieties as
terminal evolution products of several species. Thus, A. vit-
tata simulans, A. turgida simulacrum and A. leucorraphe
irwini are superficially similar forms, which as their geo-
graphic distribution and the less evolved stages of some of
them show, have descended from less similar streaked stocks.

In Achatinellastrum there are some forms of A. ccusia (be-
longing to the series of A. livida), which resemble forms of
the series of A. vulpina so closely that some authors have
actually united them specifically. Yet they inhabit opposite
ends of the Main Range of Oahu, and their similarity is
merely phenotypic.

Albino or melanic forms, belonging to different species or
races, are sometimes practically identical phenotypically, when
the differentiation in shape is slight or variable.

By the study of many colonies of a species, and of many
species, in various stages of evolution, it appears that in all
the genera of Achatincllidce the evolution of color-pattern

COLOR PATTERNS. XX111

has been orthogeiietic. Partulince and Achatinella of many
groups have followed the same path, beginning with streaked,
leading to banded patterns. While the mutations of any one
species may appear to be multifarious, it is clear that progress
has been in the direction indicated. It is significant that the
early neanic and (in Partulina) the late embryonic stages of
banded forms often show the earlier streaked pattern.

Some forms of Partulina have taken another course, em-
phasizing the vertical markings; but in some of these (such
as P. Helena, P. crassa, Hawaiian Baldwinias etc.), there are
forms which are evolving or have reached the banded stage.

The many-banded stage is the acme of pattern-evolution
in Achatinellidtf. So far as we know, it is capable of no
further modification except by decadence of pattern.

We may perhaps infer that the four-band pattern has
been inherited from the ancestral stock of AchaUnella. It
is now present in many species of Achatinellastrum and
Bulimella, and in some Mauian Partulinas. Sometimes it is
represented by the "negative' or complementary pattern
noticed on p. xxii, and very frequently it is imperfect, only
part of the bands present, or all wanting. It seems thus to
be decadent or in process of disappearing from the modern
forms. There seems some ground for the hypothesis that
in the present Achatinellid fauna we have evidence of two
successive cycles of color-pattern.

(1) In the earlier cycle there were four dark color bands
in definite positions. Of this pattern we see only the final
stage, frequently in decadence, or perhaps remaining only as
bands on the embryonic whorls. (2) The second cycle is now
in the mid-stage of evolution from a streaked or axially striped
pattern to one of many bands without definite arrangement.
In the ' ' Apex ' ' group and some others, all traces of the four-
band pattern has been lost; but in some Achatinellas and
Partulinas it lingers on with a later pattern usually super-
posed upon it.

There seems to be 110 evidence that the axially striped
Partulinas and their derivative groups, Perdicella, Newcombia
and Baldwinia, ever had the four-banded pattern.

XXIV STRUCTURE OF THE SHELL.

SHAPE AND STRUCTURE OF THE SHELL.

The shape of the shell has been varied much less in Acha-
tinellidcr than in Amastridce. Except in Newcombia, where
the spiral has been conspicuously lengthened, the ovate or
ovate-conic contour has been modified only in minor features.
In sculpture the diversity is of degree. Nearly all species
are polished, showing weak growth-lines and minute spirals,
often descending obliquely, and frequently faint or absent on
the later whorls. The later whorls of the embryonic shell are
always minutely striate spirally. In Partulina the spiral
stria? are often rather well developed ; and in Newcombia they
become prominent.

The axis of the shell is somewhat sinuous, and in the last
wiiorl it bears a spiral lamella except in Newcombia, Perdi-
cella and Baldwinia, where it is more or less degenerate.
This lamella is progressively absorbed as the shell grows.
In the embryonic stage it is represented only by a sinuation
of the columella. In both Achatinellidce and Amastridcc the
forms lacking a columellar lamella are obviously not directly
related, being terminal members of several evolutionary series.
Thus Perdicella, Newcombia and Baldwinia have more direct
affinity to Partulina than to one another. That the ancestral
stock of AchatinelUda, Amastridce and Tornatellinidce had a
columellar lamella seems a fair inference.

Professor Hyatt's idea that Partulina dubia is a representa-
tive of the ancestral stock of Partulina seems to me inad-
missible, as its weak columella is far more likely to be a recent
degenerative character than a primitive heritage, and it has
no other feature more primitive than other Partulinas.

The direction of coil of the shell and the dextral or sinistral
disposition of the organs are unusually variable in Achatinel-
lidcc ; far less so in Amastridcu, where sinistrality is unusual,
and in all cases a specific or group character. Many species
of Achatinellidce are invariably sinistral. In Newcombia all
are sinistral. Other species are as invariably dextral. In
such species a reversed specimen is as rare as among Ameri-
can land snails. A few species of Partulina and many of
Achatinella are indifferently dextral and sinistral, either in

THE SPECIES CONCEPTION. XXV

the same colony or less frequently in separate colonies. The
direction of coil is hereditary as a general rule. I have never
found a sinistral embryo in the uterus of a dextral mother.
Mr. Thaanum, whose experience has been very wide, states
that he has never taken both dextral and sinistral embryos
from one mother, in the few species of Partulina which are
indifferent in coil.

Frequently in adjacent colonies of the same race, sinistral
shells may predominate in one, dextral in another. In other
species, the prevalent direction of coil differs in different
parts of its area. See Partulina virgulata, p. 358, Acha-
tinella vulpina, and many other cases noted in the systematic
part of this work. It seems likely that, as Mr. Gulick thought,
dextral and sinistral snails find some mechanical difficulty in
mating; but experimental evidence is needed.

THE SPECIES CONCEPTION IN ACHATINELLA.

The conception of " species ' : applied to Achatinella is in-
volved in more than the usual difficulty. We have to do with
evolution products of several grades, possibly of more than
one kind.

Grade I. Races having certain seemingly permanent as-
sociations of characters (though usually with other variable
characters), and co-existing with allied races without form-
ing hybrid colonies.

Grade II, Geographic races (i. e. with a definite and con-
sistent distribution), having moderately coherent associations
of characters, but blending with other like races through
hybrid or undifferentiated colonies where the geographic
ranges meet.

Grade III. Forms characterized by the possession of one
or more special characters (usually of color), but which occur
for the most part in hybrid colonies with other diverse forms ;
the characters seeming to be freely interchangeable, and oc-
curring in different combinations, as in Mendelian hybrids.

The races included under (I) are species in the usual sense.
It may be supposed that the specific isolation is physiologic.
Thus, Achatinella bellula and A. vulpina are related species

XXVI THE SPECIES CONCEPTION.

occupying the same area without intergrading or hybrid
forms.

Races included under (II) may be illustrated by the group
composed of Achatinella phceozona, fulgens, stewartii and
vulpina. By reference to the map on p. 183, it will be seen
that the areas of these overlap slightly, but are in the main
separate. At the points of contact there are transitional
colonies which have every appearance of being hybrid groups,
composed of individuals having various combinations of the
characters of the adjacent races. It will readily be seen that
the "species' of this grade do not have the same value as
those of grade I, and they might be more logically considered
subspecies. Such treatment of the section Achatinellastrum
would reduce the species from 17 to 10, as explained on p. 181.
In the genus Achatinella it has been thought more practical
to recognize as "species' a certain number of races which
admittedly intergrade at their limits, than to make the species
conception so broad that no definite idea is conveyed. More-
over, the term subspecies is needed for subdivisions of lower
grade. Thus the Koolau forms of A. fulgens show slight
racial divergence, which is recognized by the term A. fulgens
versipellis.

In Achatinellastrum and Bulimella, then, it is frankly ad-
mitted that the species of this work are groups of two grades,
certainly differing widely in degree, quite likely differing in
kind. Related forms which inhabit the same district without
evidence of hybridizing may safely be put down as species
of the first grade. In other forms, which may appear equally
as distinct phenotypically, we find abundant evidence of
hybridizing where their areas overlap ; and these we rank as
species of the second grade. How fundamental this distinc-
tion is in animals generally we do not know, and it has not
yet been satisfactorily worked out from the experimental side.
Practically the grade of any given form is to be decided in
each case by field observations and abundant collections. On
pages 119 and 181 the real or first grade species of Bulimella
and Achatinellastrum are indicated. In the section Acha-
tinella s. str. ("Apex"), the conditions are very complex.

THE SPECIES CONCEPTION. XXvli

The species of the mustelina-decora group, and those of the
apexfulva-leucorraphc group (see p. 275 and diagram on
p. 278) are certainly very closely related while A. lorata
stands isolated. In Partulina most forms admitted as species
seem to belong to the first grade.

Species and subspecies of all grades are usually composed
of numerous color-forms of Grade III that is, elementary
patterns, which may rarely occur in pure colonies, but almost
always in hybrid colonies comprising two or more patterns,
in an almost endless variety of combinations.

What may be considered "fluctuating variations", that is,
variations in the potency with which a pattern may be ex-
pressed, among individuals of a pure colony, are of great
amplitude in AchatineUidce. Thus, plate 60, figs. 17 to 17c
may be presumed to be individuals having the same gametic
constitution, but differing in the degree of potency of the
factor controlling the deposition of pigment. Other examples
are pi. 56, figs. 14 to 14d ; 13 to 13& ; figs. 1 to Ic. Most of
the pure colonies show such variation in greater or less degree.
Most shells having highly developed color-patterns show
similar fluctuation. Cf. Neritina, Liguus, etc.

In a hybrid colony there is often almost perfect segregation
of the color-patterns composing it, as in pi. 39, figs. 8 to 8e.
This is usually the case where only two or three patterns are
involved. Where the number is greater, the combinations be-
come numerous. We are getting into the realm of trihybrids
and multihybrids. Although we see only the phenotypes of
the gigantic Mendelian experiment, it is our belief that with
a good knowledge of the forms of any limited region, a trust-
worthy estimate of the elementary patterns present in a colony
not too complex may be formed, though of course it would be
absurd to say what the genotypic constitution of any individ-
ual of the colony might be.

The good segregation which prevails in hybrid colonies
leads collectors to assort their shells. The strikingly diverse
patterns were not unnaturally described as distinct species
by European conchologists, who were ignorant of the facts of
association. Moreover, before the Mendelian era, it seemed
improbable that such diverse patterns could be hybrids.

XXV111 THE SPECIES CONCEPTION.

From the results visible, it can hardly be doubted that the-
various components of color-pattern act as unit characters in
inheritance.

In species covering an extended area, there is usually dif-
ferentiation correllated with distribution. Thus in A. fulgens
the plumata pattern, streaked in delicate tints, is a con-
spicuous form in Niu, the eastern limit of the species where
green forms are not found (pi. 36, figs. 4 to 6e). Also in
the next valley, Wailupe (pi. 36, figs. 3, 3a). Further west,
in Waialae, the plumata pattern has become rare, though still
present in various combinations, and green forms are com-
mon (pi. 37, figs. 1 to 6&). In Palolo, still westward, the
green and yellow patterns are ascendant, and the plumata
marking is hardl} r represented except in the varia combina-
tion. The ancient pattern of black bands in definite positions
pervades the whole species, though the factor necessary for
the appearance of these bands is often absent. Also albinos,,
doubtless genotypically various, as usual, appear sporadically
throughout the range of the species.

This topic may be further illustrated by a diagram of the
conditions found in A. vulpina. If we represent the various,
color-patterns (or "elementary species", at least in part) by
their initials, the forms of the hybrid colonies of successive
districts, from the east westward, may be represented thus

Pauoa E. Nuuanu W. Nuuanu Kahauiki Halawa Waimano Manana

avco vceo ovl ovsa\ oval or v

(a, adust a v, vulpina- c, ca-stanea; e, ernestina; o, olivacea;
v, virens-, s, suturdtis; a, analoga; 1, longispira).

See also p. 215. The formulas could be made more complex
by introducing various other characters. Thus, dextral in-
dividuals occur in the West, but are absent eastward. More-
over, several colonies represented by the same formula almost
always differ in the relative numbers of individuals of the
component patterns ; and adjacent colonies, sometimes on a
single tree or group of trees, may differ, some lacking pat-
terns common in other colonies of the district.

Many of the "species' of Gulick are based upon colonies

THE SPECIES CONCEPTION.

in which a certain pattern or patterns predominated, though
not to the entire exclusion of others. The same patterns may
be present though uncommon in colonies a few miles removed,
where other patterns take the leading role. Thus Gulick says
of the special pattern he called Apex tuberans, "the metro-
polis of this species is Kalaikoa ; it is also found in Ahonui,
and single specimens have been brought from Wahiawa and
Helemano. ' Any mutation having a recognizable pattern
(races of the third grade) could be treated as a "species'
by this method ; and in fact there are numerous unnamed
mutations neither more or less distinct than many of those
which have been named and denned. It is often a convenience
to have names for patterns in which the association of char-
acters is moderately stable. We may speak of the ernestina
pattern, the dunkeri pattern, etc., without committing our-
selves for or against the theory that these names stand for
races of systematic significance, or are in fact anything more
than character combinations which from their abundance
would seem to have a certain degree of cohesion.

A philosophic method of dealing with intraspecific differ-
entiation is one of the greatest present needs of systematic
zoology. Systematists can 110 more deal with species and sub-
species as their ultimate units than physicists with molecules
or even atoms. It is senseless to oppose the facts which our
vaster collections and more critical methods are forcing upon
us. The terms species and subspecies, though elastic, have a cer-
tain status recognized by those accustomed to deal with these
matters, even though the categories may not be rigidly formu-
lated. To enlarge the species conception to cover every group
of individuals having a recognizable special character, as some
naturalists have recently advocated, would be to reduce the
whole subject to chaos. Some English naturalists give a
special name to every form having a special character, such
as albinism, xanthism, a special pattern, a particularity of
shape, or the like. This is one method, doubtless far better
than none, of making intelligible record of such matters. We
are not prepared to name every individual of polymorphic
species, but in reasonable limits it seems useful to have names

XXX THE SPECIES CONCEPTION.

for groups of individuals having moderately coherent com-
binations of characters, and yet not properly to be ranked
as subspecies. In this sense we have used such terms
as ' * olivacea pattern " ; ' dunkeri pattern. ' In other cases
we have avoided naming the forms, using numbers in-
stead. The ideal (if impracticable) method would be to have
a name or symbol for each unit character, and express the
several combinations like a chemical formula; but although
almost all the world is breeding animals or plants at present,
it is not likely that we will ever have enough knowledge of
unit characters to apply this method generally in nomen-
clature !

Color-pattern is usually independent of direction of coil of
the shell, but in some cases, noted under various species, these
characters are linked in inheritance. See under Partulina
virgulata, p. 358, etc.

The facts indicate that mutations may appear anywhere
in the area of a species, and spread from the initial point as
far as time and conditions permit, mingling with the pre-
existing forms of the species. The population is thus becom-
ing constantly more complex. It occasionally happens that
one comes upon what appears to be a mutation in an early
stage of its career. Thus, near the foot of a small ridge in
Popouwela, there is a large and prolific colony of Acha-
tinella mustelina (pi. 63, figs. 10 to lOc). In the midst of
this colony, in an area of two or three rods square, we found
20 or 30 pure white individuals, most of them immature. Not
one was found in other parts of the colony, which three of
us looked over pretty thoroughly, or in any other colony in
the same district. The inference seems fair that the albino
form had very recently arisen in that place.

Many cases of special color-forms known from single or
very few trees are recorded in the descriptive part of this
work. Some of these are doubtless remains of races once
more widely spread, especially when they occur in decadent
forest; but others we can hardly doubt, are new mutations.
The collections made to-day very often show color-forms dif-
fering from those of the same valleys made sixty years ago;

THE SPECIES CONCEPTION. XXXI

but the comparison is vitiated by the uncertainty whether the
same colonies were found. The older collections were not
closely enough localized to afford evidence of progressive
change. In many cases it is certain that they were from
districts now deforested. The accurate record of colonies now
kept by the best collectors will permit of such comparisons in
the years to come, and if the snails are not exterminated, in-
valuable evidence for evolution will be available for Hawaiian
students of fifty years hence, or even less. Answers may be
found to such questions as the appearance of mutations, their
influence on the colony and their fate, and the permanency or
changes of hybrid and pure colonies under natural conditions,
etcetera. In this connection, suggestive observations were
made by Mr. Thaanum on a colony of Partulina proximo,, re-
visited after an interval of twelve years. In the interval, an
apparently new mutation shells with pale greenish ground
had dispossessed the normal white-ground shells which alone
were found twelve years before (see p. 360).

While there are good reasons for the belief that color
mutations of considerable amplitude are common in Acha-
tinella, it must be remembered that there is no actual evidence
bearing on the production of species of the first grade, in other
words, real species, by mutation.

Observations on the snails of many regions has led me to
believe that the well-defined species are often, perhaps usually,
rather complex populations of individuals differing in many
minor but heritable characters, such as size, minutiae of shape
and sculpture, number of teeth on the radula etcetera. In
Achatinella, as in Liguus, Polymita, Amphidromus etc., the
versatile character is the very conspicuous one of color, and
therefore has attracted far more notice than equally im-
portant but less noticeable characters in plainer snails.

One condition favoring intense speciation in the Hawaiian
Islands is that the snails breed and grow all the year round.
Hence, in a given time, their evolution should be at least
twice as fast as that of snails in temperate latitudes, in which
the organism is quiescent more than half of the time. This of
course applies to humid regions in the tropics in general.

XXX11 THE SPECIES CONCEPTION.

To the naturalist knowing these shells in the field, it would
seem absurd to claim that the colors are protective in any
ordinary sense, or that their evolution is controlled or directed
by any sort of selection. It is perfectly obvious that one
color or pattern is as good as another. A pattern predomin-
ating in one colony may be rare in the next, living under con-
ditions apparently the same, and perhaps not half a mile dis-
tant. White shells, the most conspicuous of all to my eye,
are as abundant as any others; though I would not give the
impression that any Achatinella is very conspicuous. Mr.
Perkins found no evidence that Achatinellidcz are ever eaten
by birds, or in fact by any animal but the imported rats
(Fauna Hawaiiensis I, vi, p. ccxxvi).

Weissmann's treatment of the subject of specific differen-
tiation of Hawaiian snails does not appear to be based upon a
sufficient knowledge of the facts. He considers the pheno-
mena due to adaptive variation of species in a new environ-
ment. 1

I have elsewhere expressed the opinion that the opaque
coloring everywhere characterizing land shells living in ex-
posed situations is an adaptation to exclude light. The
corneous yellow, or light brown tints usual in ground
snails are more translucent. The particular patterns evolved
since arboreal life was assumed do not seem to affect the
existence of the species. In common with Dr. Cooke and
other observers, I have been unable to trace any constant
relation between coloration and physical conditions (except
as noted below), and there is certainly no connection between
color and the species of tree inhabited.

In the typical group of Achatinella (Apex], a certain cor-
relation seems to exist between the pattern and elevation,
banded forms occupying higher situations than streaked forms
of the same species, at least in many cases. I believe that
the relation is an indirect one. Banded patterns have been
evolved from streaked, as stated on p. xx; evolution in more
humid stations would naturally be more rapid than in lower,

ir Phe Evolution Theory, translation by J. A. Thomson and Margaret K.
Thomson, London, 1904, II, 293-296.

SPECIFIC AREAS. XXxiii

less abundantly watered places, since resting periods would
be shorter, and the succession of generations would be faster.
I believe, therefore, that such forms as A. turgida simulacrum,
A. vittata simulans, A. leucorraphe irwini, are accelerated
races of species which at lower levels have lagged behind in
pattern-evolution.

While many conehologists may consider the treatment of
Achatinella in this work an extreme example of "lumping",
(since we recognize but 43 species in place of 171 described),
it really belongs to the splitting school. Both authors hold
that a considerable further reduction would have to be made
to make the species of equal value with most Oahuan species
of Leptachatina or Amastra ; that is, to make them all species
of the first grade.

The immense color-differentiation of Achatinella has not
been exaggerated by former naturalists, but its systematic
value has been overrated. While it is believed that the ' ' sub-
species" of this work are in large part comparable to groups
of equal rank elsewhere, the authors are prepared to find that
some of them may turn out to be undeserving of a trinomial
title. In other words, they may turn out to be forms of the
third grade, like many of the species of Gulick.

Extent of specific areas.

An idea of the size of the specific areas of Achatinella may
be gained from the maps on pp. 183 and 277 and the table on
p. 276. The colonies of one species may be spread over from
a dozen to fifty square miles, but actually occupy only a small
per cent of the territory. Some color-races are known from
very small areas even down to one or two trees, as recorded
in the text. In general, the specific areas in Achatinella are
smaller than in Amastra, though there are exceptions. Some
Amastras of the high peaks of the main range are as yet known
from very limited areas. The species or races of the extremely
humid region on top are generally of very limited distribu-
tion, but of course the summit of the main ridge is only
known in spots, as there is no ridge trail.

XXXIV HABITS AND STATIONS.

HABITS AND STATIONS OP ACHATINELLID.E.

Achatinellas are nocturnal as a rule. By day they sit
dormant, usually attached by dried mucus, except in quite
wet places. They are not firmly attached, like Oxystyla in
dry weather, but securely enough to resist any but very violent
shaking. Forms which frequent the trunk and limbs of trees
often gather in crotches, in knotholes, or hide under loose
edges of bark. Others stick on the leaves, particularly in
the concave of a curled leaf. Often two or three roost in
company; and when one is found alone, there is usually a
companion not far away. Sometimes many gather in a com-
modious knothole.

Pubescent plants are avoided. Otherwise there seems to be
little selection, any species living on a considerable number
of plants. Dr. Cooke whose observations are recorded under
A. bellula multizonata, was unable to trace any connection
between color of snail and species of tree, and other observers
have reached the same conclusion. Various introduced plants,
particularly guava and lantana, are frequented by the snails.

In the axils of ieie one finds Achatinella in company with
Philonesia, Succinea and Auriculella. It is a favorite station
for A. abbreviata and A. lila, but is also frequented by many
other species. This plant reminds one of the tropical Ameri-
can bromeliads, as in the humid zone water collects in the axils
of the terminal bunches, deluging one as they are bent down.

The native bananas are good shell plants, both for Acha-
tinella and Laminella. They grow in small clumps in humid
ravines.

The kukui or candle-nut tree (Aleurites moluccana), so
often mentioned by Mr. Gulick, occupies a broad zone on the
lower mountain slopes. It is conspicuous from a distance by
the very light green foliage. The tree is commonly rather
rounded, with widely spreading, open limbs, branching much
like our Floridan gumbolimbo.

While Achatinella ascends thirty feet or more, most collect-
ing is from quite small trees and shrubs. The snails are hard
to see on such copiously leaved shrubs as mokihana and the

HABITS AND STATIONS. XXXV

like, and it takes some experience to find them. The eye
must be trained. I have had good collectors follow me ami
say that the best collecting was on the trees I had looked over !

I have never seen an Achatinella crawling on the ground,
though occasionally on returning over a trail, one finds shells
which had been dislodged some time before. Mr. Spalding
tells me that his experience is the same. See also pp. 99-102
for observations by Mr. Henshaw. These snails are certainly
more inert than our American Helices, which if dropped usu-
ally make off promptly ; yet it seems likely that by night they
wake up. It seemed to me that the young shells wandered
more widely than the adults. Certainly with us young snails
walk more in the day time and wander more than old ones.
In the head of Kamalo, Molokai, there was a forest of native
trees, now all dead and prostrate, doubtless from the destruc-
tion of the undergrowth by cattle or deer. It is now covered
with low lantana, standing in patches, groups or singly, and
bearing a copious population of Partulina redfieldi kama-
loensis. They are also on the dead trees. It does not seem
likely that they could have spread so universally from the
trees to the immigrant lantana without crawling on the
ground for short distances. Observations with a lantern
should be made in some such prolific colony. However this
may be, it is evident that migration is slow, and probably
for the greater part by way of the interlocking branches of
adjacent trees. Where the vegetation is dense, there is often
a pretty continuous highway; but the colonies of shells on
isolated trees or small groups of shrubs must often be segre-
gated from their fellows for many generations.

Mr. Perkins has mentioned a fact which I find is known to
all Island collectors, that a species may be found on one tree
or shrub, year after year, without spreading to neighboring
shrubs. This is true also of some arboreal Pupillidcc and
Endodontidce. It strikes the outsider as uncanny to be taken
to one certain tree to collect specimens of some snail which
either has never been found anywhere else, or nowhere else
in the neighborhood or district. To find that trees just as
good all around are barren of specimens is always a surprise.

XXXVI HABITS AND STATIONS.

The presence of special color-patterns on single trees or
small groups, in many instances, is an evidence of the slow-
ness of migration, as well as of the facility attending the
origin of varieties.

Food.- -The frequent presence of living snails on trees long
dead indicates that their food is cryptoganic. Moreover, the
leaves show no traces of grazing. The feces of Partulina
confusa examined by Mr. Henshaw were composed chiefly of
remains of fungi or algae (p. 103). The contents of stomachs
of Achatinella mustelina and several other species, examined
by Dr. A. Brown and myself, were recognized as fungi, often
with shreds or fibres probably of bark, which remained un-
digested in the intestine.

AchatinellidcF breed at all times of year. In any catch one
finds part of the individuals carrying embryos. The length
of life of the individual is not known.

The native Hawaiians claim that the tree shells have a
song, which they have fancifully supplied with words. Dr.
Newcomb (P. Z. S. 1853, p. 129) and Dr. N. B. Emerson
(Sacred Songs of the Hula, p. 121), and others have given
versions of this song. Mr. Perkins believes it to be the chirp-
ing of crickets.

Brief notes on the topography and conditions of collecting
may be useful to conchologists who have not visited these
islands. Those interested in the subject should consult the
Introduction of Fauna Hawaiiensis by Mr. B. C. L. Perkins.

The Achatinellge are not shells of the valleys but of the
ridges and upper ravines. The bottoms of the larger valleys
often lie below the zone of requisite humidity. This is doubt-
less much more generally true now than before the valleys
were so extensively deforested. Certainly at the present time
the ridges and their slopes are the chief natural highways of
the tree-snails, and the valleys are barriers, in varying de-
gree, to the spread of species along the ranges.

Since it is the valleys and not the ridges which figure in
topographic nomenclature, it follows that the usual locality
records by valleys are often less definite than we would de-
sire. For instance the locality "Nuuanu valley" might mean

HABITS AND STATIONS. XXXVli

either the northern or the southern slope, a matter of some
importance, since the Nuuanu-Kalihi ridge is a faunal unit
quite noticeably different from the Nuuanu-Pauoa ridge. The
locality records of Gulick are usually by valleys. Newcomb
generally named only the district, a far less definite indication.

Mr. Gulick has related to me that in making his rounds
from one mission station to another, he would engage Kanakas
to collect for him. His own collecting was mainly 011 the
lower slopes, reached from the valleys, in the kukui tree belt.
Much of this territory is now barren, by recession of the forest.

The relatively small number of land snails on the wind-
ward (Koolau or northern) side of the main range of Oahu
is not due to its ' ' rougher climate ? ' as supposed by Professor
Hyatt, but to the fact that the slopes are largely too pre-
cipitous to support forests, often so steep that they are
practically barren. From the summit of the main ridge,
reached by a long climb up the southern slope, a gigantic
pali, ribbed by erosion, drops beneath your feet nearly to sea
level. Such vegetation as finds root-hold often cannot be
reached further than a few rods down from the summit, where
one ventures clinging to bushes for support. At the foot of
the pali there is a talus-slope, often with kukui and other
trees, but at this low level there are few tree-shells or none,
under present conditions.

Where long butresses extend out upon the Koolau side,
as in the Kailua region, and from Kahana northwestward,
these conditions are modified in many places ; and here the
forests were, or still are, rich in tree-snails, though the wood-
land limits have retreated far within their old boundaries.
Once forests with AchatinelUdcE and Endodontidce shaded the
plains far seaward from the lovely peak of Kaneohe, where
now dead shells may be picked up in plowed fields, or gathered
out of 'pockets' in the rocks. It has been the same in the
northwest. Forest-snails are found in the sand-dunes of
Kahuku, now far from where living tree-shells exist,

The poverty of the ocean side of the range is therefore
due chiefly to the small area of forested slopes and ridges,
owing to the colossal erosion of this side, and also to the prac-

XXXV111 HABITS AND STATIONS.

tically complete destruction of primaeval forest on the lower
levels.

The chief area of Achatinella is from the ridge down to
about the 1000 ft. contour on the southern or Kona slope.
The best published maps of Oahu give a wholly false idea of
the topography of the western half of the Main Range. The
fact is that the precipitation along the main mountain axis
has formed great amphitheatres along the lee side. The
ridges running westward or southwest from it are in part
broken down, lowered, and worn to knife-like crests. Several
miles further from the main axis, where erosion has been
less potent, the ridges broaden out and branch, so that to-
wards their lower ends there are more and wider high ridges
than further up. Far down, finally, they may be cultivated
on top. A map of this region, supplied by Mr. Spalding
from his note books, is given on page 277, in order that an
intelligent idea may be formed of the localities mentioned in
the text for the western species.

In the Waianae range (see p. 341) the conditions are
similar. Though it rises to a greater elevation than the Main
range, it has also been more extensively eroded. Broad val-
leys, separated by knife-like or interrupted butresses, have
been carved out on the southern (ocean) side, heading for the
greater part in precipitous slopes, mostly inaccessible. There
is a narrow fringe of forest, with tree-snails in places, around
the heads of these valleys, but the major part of the slopes
themselves are almost or quite barren, and too steep to sup-
port forest. The valleys of the inland side of the range
head up so near those of the Ocean side in some places that
only a narrow, w r all-like arete remains, as above the heads of
the Popouwela gulches. Standing on the wild-goat trail at
the summit, a stone dislodged on either side might bound
downward a thousand feet or more, so narrow is the ridge. At
Waianae Pass the range has been almost gnawed through.
Immediately west of this pass Mt. Kaala rises, the highest
point on Oahu huge, flat-topped, its summit drenched by
daily cloud or rain. West of Kaala the ridge narrows again.
At the northAvestern end of the range, where the foothills ap-

HABITS AND STATIONS. KXX1X

proach the ocean at and west of Kawaihapai, there is now
no forest on top, but Amastra, Leptachatina etc. may be found
on the wooded slopes. The chief forests of the range stand
on the inland slopes and butresses from the top down to per-
haps 1000 or 1500 feet. They are often prolific in Acha-
tinellas of the Mustelina group. Several species of Acha-
tinellastrum have also been found, but they are extremely
local, and among the rarest Oahuan snails. Amastra and
small shells are abundant.

The western half of Molokai is grass land, or barren where
the country is broken, and there are no living land shells.
The eastern half has the general structure of the main range
of Oahu. Western Maui is similar, but here the gulches radi-
ate from a central point. The Kohala region of Hawaii,
like West Maui, is deeply cut by magnificent gorges, as seen
from the sea. I regard it as one of the most important re-
gions to be explored.

Lanai, which I did not visit, has a very limited amount of
forest remaining, and this has been thoroughly worked by
Mr. Thaanum. Important work is still to be done in the
deposits of fossil shells.

While considerable climbing is involved in any serious col-
lecting in the Hawaiian group, the altitudes are not great.
In the Rocky Mountains our basal camps are usually above
the tops of the Oahuan mountains, and the shell country
goes up to over double the elevation. Yet the open, rocky
slopes of our western mountains render them vastly more
accessible than the Hawaiian mountains, where one does not
think of making any considerable ascent unless there is a trail.
I have gone through scrub oak thickets in Arizona as strenuous
as any fern or ieie patches in Oahu, but they are exceptional,
and can usually be avoided. In general, the slopes are more
precipitous in the Islands, many more of them are practically
inaccessible, and owing to the humid atmosphere and dense
jungle, far less ground can be covered. Cross country tramps
or short cuts are hardly thought of unless in a thoroughly
known district.

To the collector who has worked in the West Indies, the

xl HABITS AND STATIONS.

larger shells seem scarce and hard to find. Only small or
minute shells are really abundant, and these only in spots.
I would say that in Cuba, the same amount of work would
result in at least ten times as many large shells as in Oahu,
while the number of small ones would be about the same. In
southern Arizona, one could not expect more "live" shells in
a day's work than in Oahu or Molokai, if so many; and the
number of species would be smaller. Helicina, Succinea,
Philonesia, Tornatellinida: and Leptackatina are usually
rather abundant. Pupillidce and Endodonta often abundant,
but very much more local.

It is easy to understand the genesis of arboreal snails
when one sees the conditions in such humid forests as those
of Hawaii, where the rainfall totals one to two hundred
inches. Snails of all kind HeUcinidce, Zonitidcc, Endo-
donts, Amastras, Pupillidcc wander freely over the herb-
age and tree-trunks. There is no need for them to stay close
to the damp earth and under cover, as in dryer climates.

Achatinellas are usually cleaned by "squirting' without
cooking. A small nozzle attached to a faucet gives a good
stream for the purpose, or a syringe may be used. Hot water
changes the green color to a dirty olive or brown. The chemi-
cal relations of green and brown must be very intimate.
White shells or those with dull colors, like Partulina, Amas-
tra and many others, may be boiled without detriment to
the colors. As a rule, AchatinelUdce, Amastridcc and Auri-
culclla pull easily. Living Achatinellas are daintily clean and
bright; and of course no oil or anything else should be put
on the shells to improve or preserve the colors.

The numerical preponderance of arboreal Achatinellidce
in the Island faunas has been much exaggerated. The actual
number of species known is less than one hundred. The
genera Amastra and Leptachatina both outnumber the tree
shells. Leptachatina will eventually prove to be far the most
prolific genus in the fauna, in the opinion of Dr. Cooke, who
has given it special attention. Kauai and Maui will doubt-
less turn out many new species of Amastridcc, and in Maui,
Achatinellida also. Hawaii - - especially the Kona side is

NOTES ON ZOOGEOGRAPHY. xli

yet very imperfectly known. The spectacular gorges of the
Kohala mountains appeal to me as a promising fielcl. More-
over, it is an important point from the zoogeographic stand-
point. On all the islands, the fossil beds promise good enter-
tainment.

The snail fauna, outside of Achatinelloid forms, has been
only imperfectly worked up, and many undescribed species
are now known. We cannot doubt that a large number re-
mains to be discovered. On the whole, it appears that the
work to be done in faunistics alone is much more than one
generation can accomplish.

Much in the preceding pages may seem trite to Hawaiian
naturalists, accustomed from boyhood to scale the rugged,
jungle-clad mountains of their exquisite islands in the quest
of shells, or wild goats. I wxmld not presume to write for
them upon many topics which have been touched on here, in
order that conchologists elsewhere may better understand the
Hawaiian faunas.

NOTES ON THE ZOOGEOGRAPHY OF ACHATINELLOID SNAILS.

(Supplemental to Vol. XXI, pp. xi-xix.)

Arguing from the distribution of the living snail fauna,
the theory was advocated in Vol. XXI that the islands were
formerly united, allowing land and fresh-water snails and
other land animals and plants to pass from one to another.
The exploitation of the deposits of fossil land shells lends ef-
fective support to this theory.

In the recent fauna, the species of Achatinellidge and
Amastridae are special to each island, though in a few cases
the difference between some species of two islands is small.
In the Pleistocene fauna the relationship was closer. Oahu
had a Partulina of Molokaian type (A. montagui). The
Amastra umbilicata group is found to be common to every-
one of the islands. A. umbilicata of Oahu and Molokai is
barely distinguishable from A. morticina of Maui and
Kahoolawe, and A. ultima of Hawaii is very closely related.

In Leptachatina, the group of forms having a posterior
tubercle and groove in the aperture, formerly known only by

xlii NOTES ON ZOOGEOGRAPHY.

one specimen from Kauai (L. fossilis, Vol. XXI, p. 61), is no;v
known from many beds on the north, west and south shores
of Oahu, from Molokai, Maui, Kahoolawe and Hawaii. This
group of very closely related forms, now extinct, was commo'i
to at least six of the islands in the Pleistocene. One species,
L. subcylindracea, was common to Oahu, Molokai and
Kahoolawe, and the forms of Maui and Hawaii are barely
distinguishable. The extinct forms of the L. oryza group
from Oahu to Hawaii are so very similar that we have doubts
of their specific distinction.

Upon the hypothesis that the snails had been distributed
by ocean drift, birds, etc., it would be expected that the num-
ber of species common to two or more islands would be in-
creasing by the occasional action of such agencies. The fact
is, that species common to two islands were more numerous
in the Pleistocene than they are today, even though the ex-
ploitation of the fossil deposits has only begun, and the recent
fauna has been studied for years. Such a condition is
exactly what we would expect if the islands had been united
in the late Tertiary, and their common faunal elements had
been distributed by known and normal modes.

The hypothesis that Achatinellidce (or rather Amastridce)
first come to Kauai, and thence to the other islands, as advo-
cated by Professor Hyatt (p. 371 et seq.) no doubt had its
inception in the orthodox view of the relative geologic age of
the islands. The biological evidence relied upon by Hyatt
has proved, now that the fauna is more fully known, to
indicate no such succession of faunas. Taking the fossil
species now known into the account, it becomes obvious that
all the islands have, or in the Pleistocene had, very closely
related faunas of the primitive groups: Leptachatina (cer-
tainly the most primitive genus of the family), Amastrella,
Cyclamastra. Between these closely related species it would
be absurd to claim that those of Kauai are more primitive
than those of Hawaii, or any other island. The same evolu-
tionary stage is equally represented by species on all the
islands. Hence, it seems likely that these common faunal
elements remain from a former time when the islands were

NOTES ON ZOOGEOGRAPHY. xliU

united. In addition to these primitive Amastridce, almost
every non-Achatinellid genus of land and fresh water shell
found in the archipelago occurs on all the islands, excepting
the low and arid Niihau and Kahoolawe, where part are want-
ing. It appears highly improbable that so many genera
should have become so generally diffused by any ' ' accidental '
means of over-sea transportation. That Kauai has ever con-
tributed faunally to the other islands by means of drifting
trees or the like is the more improbable because such drift
would be against what current there is, and directly across
the course of the trades. The snail fauna of Kauai is
"primitive" only in lacking representatives of the Achatinel-
lid<z, and the arboreal genus Auriculella. The fauna of
Amastridcc and other families is on a par with the faunas of
the other islands. The absence of Achatinellidcc from Kauai
seems to be a rather strong argument, albeit negative, against
the view that the fauna of the entire group emanated from
Kauai.

Mr. Perkins, in discussing the dispersal of Achatinellidcc
states that ' * once, on Molokai, a young living Achatinella was
found attached to the feathers of the Drepanid bird Chloro-
drepanis. Frequently they become adherent to one's clothes
when passing through the brush. Doubtless in high winds
very young shells are sometimes carried to a distance in curled
up leaves in which they hide' (Fauna Hawaiiensis, Intro-
duction, p. Ixvi). While such means of travel must have
had some part in the distribution of Achatinella, we are in-
clined to believe it a small part, for the reason that
anomalies in distribution are very rare. If birds had carried
the snails about, there would be colonies of "erratics,' like
granite boulders in a glaciated limestone region, instead of
the thoroughly consistent distribution which is the rule on
every island.

It is now known that the family Achatinellidce differs in
important structural characters from the Amastridce. While
the two are allied, their separation must, if we judge by the
analogy of other groups of known age, date back to the early
Tertiary at least. Indeed I believe that no existing genus of

Xliv NOTES ON ZOOGEOGRAPHY.

Amastridce could have been ancestral to the AchatinelUdte.

The points of agreement between Kauaia and Achatinella
ph&ozona, noted by Hyatt (pp. 398-9), pertain to shell char-
acters of little or no phylogenetic significance, and leave the
important anatomical differences absolutely untouched. In
fact I have failed to find any characters in A. phccozona which
indicate it as an ancestor even of the species immediately
related far less of the whole Oahuan genus Achatinella !

The absence of arboreal AchaUnettidcs from Kauai has
provoked comment because the forests seem well adapted to
tree shells. It is a grave difficulty from any point of view.
I am informed that Achatinella introduced there thrives.
Several hypotheses may be formulated. (1) It is possible
that the AchatinelUda reached the region later than Amas-
tridcc, and after Kauai had been isolated from the remainder.
This seems the simplest explanation. (2) Achatinellidcs
may have existed on Kauai, but become extinct from some
unknown cause, as the horse family in America became ex-
tinct. Or (3), it is possible that the events attending the
submergence of the old Hawaiian land did not allow tree
snails an opportunity to gain access to the new forests on the
Kauaian volcanos. However this may be, speculation upon
a negative character of this kind seems rather unprofitable
emplojonent in the present condition of our knowledge.

In my opinion there is no evidence whatever indicating
southeastern Oahu as the point of divergence of the snails
of that island. The geological features and the distribution
of species are both too ' ' mature ' to indicate one place more
than another in the Main Range as the initial point. In
Hawaii both the geological structure and the distribution of
Achatinellidce point to the Kohala mountains as the area from
which land snails have radiated. Species and varieties pro-
gressively diminish from this region down the east and west
coasts.

From studies on Amastridcc (Vol. XXI) I concluded that
the Waianae fauna had been independent of that of Koolau
from the early period when all the islands were connected
until late Pliocene or more likely Pleistocene. The tree shells

NOTES ON ZOOGEOGRAPHY.

fully support this conclusion. There are five species of
Achatinella, one Partuliiia and two Laminellas. Three of
these are referable to Koolau species, and the others are so
clearly allied to several diverse species, belonging to groups
widely spread in the Koolau range, that they must be deriva-
tives from them. There is no endemic Waianaean strain of
Achatinellida', and from all indications it seems that this
family was lacking there until Pleistocene time. The old
Kaala massif, like Kauai, had a fauna of Amastrida. In the
diagram no. 2, in Vol. XXI, p. xx, Kaala might better have
been isolated from the Oahu-Hawaii island. These conclu-
sions have support from the distribution of plants. Mr. Charles
N. Forbes states of Kaala and its range that "its flora is as
distinct from the main range of Oahu as is the flora of any
separate island of the group' (Oec. Pap. B. P. B. Mus. V,
no. 4, p. 13, 1913).

In the case of Maui, the two mountain masses are now
separated by a considerable width of low land, impassible
for land snails. It seems certain that the connection was
formerly more intimate. The number of identical or very
closely related species is quite considerable, and judging from
Mr. Thaanum's last expedition, it is likely to increase. The
isthmus must have been densely wooded, and probably it stood
at a decidedly greater elevation, to permit free exchange of
mountain snails. A rather recent subsidence of the whole
island is indicated by the absence of such an extensive pene-
plain as would result from the tremendous erosion of West
Maui, if that had been accomplished at the present level of
the island.

The Kauaian satellite island Niihau will doubtless prove to
have had all of the widely spread genera of Kauai such as
Leptachatina, Succinea, Helicina, Tornatellina, Pupillidce,
Endodontidce etc. when its Pleistocene deposits are ex-
amined. A general account of the island and list of the plants
has been given by Mr. C. N. Forbes from notes made by Mr.
J. F. G. Stokes (Occ. Pap. B. P. B. Mus. V, no. 3 with map).

Theories relating to the origin of the Hawaiian fauna have
been unduly influenced by the inferences of geologists con-

Xlvi NOTES ON ZOOGEOGRAPHY.

cerning the growth and history of the islands, especially as
to their relative age, and the idea that they were built up
from the sea bottom solely from materials ejected by the
volcanoes. This hypothesis presumes an enormous output
and a duration very long for the active life of a volcano.
Nowhere on land, where the geologic structure is accessible,
has any mass even remotely comparable in magnitude to the
Hawaiian ridge been piled up by purely volcanic agency.
While it is true that most volcanoes rise from low levels, it
is also noteworthy that those whose summits stand high
(18,000 to 30,000 feet above the general level of the
supporting sea bottom, in the case of the Hawaiian
ridge) are borne on the backs of great folded mountain
uplifts, as the Andes, or the Mexican plateau. Since
none of the known volcanic deposits are believed to be older
than tertiary the hypothesis also limits the time available for
the evolution of the peculiar Hawaiian fauna to a period far
shorter than experience has shown is probable. Elsewhere
we know that many groups of generic rank in the land shells
go back to the Oligocene and some to the Eocene. The dif-
ferentiation of the modern families must therefore have taken
place largely in the Mesozoic. It is not likely that so strongly
characterized a family as the Achatinellidce is much later.
The case would be different if Achatinellidce were known to
have existed on any of the continents, for then it might be
claimed that they had drifted to the islands. Personally, I
think the drift hypothesis, applied to these remote islands, is
about as credible as the idea that life was first brought to
earth on a meteorite. I can conceive that wood-boring beetles
might travel in the logs from America which I have seen
thrown upon the windward coast of Molokai logs stripped
of bark and worn deep into the solid wood by the buffeting
of the waves but my faith is too weak to believe that any
snail could get deep enough in the wood to make the journey
of two thousand miles dry, or that if it did, it would ever
get out when the log finally dried on a tropical beach. Only
arm-chair zoogeographers can hold to the hypothesis that the
land snails, Lymn&idce and Melaniidce of these islands were

NOTES ON ZOOGEOGRAPHY.

brought there by drifting logs, or sticking to the feet or
feathers of birds.

We have to account for the presence of representatives of
eight families of land and three of fresh-water snails. 1 There
must have been at least sixteen, probably more, original
immigrants. The affinities of practically all are with Poly-
nesia, and much more remotely with the western border of
the Pacific, in no case with America. Under present condi-
tions, drift material can come only from America. This has
certainly been the case during the duration of the present
islands, as their coast topography shows. In all probability
the trades have blown and the current flowed as they do now,
since early geologic times.

The idea that the mollusks were carried by wind, as some
insects and seeds may have been, is hardly to be entertained..
The distance is too great.

The hypothesis that the Hawaiian volcanoes rise from
a preexisting mid-Pacific ridge, now lost by subsidence,,
gives room in time and space for the development of the
peculiar fauna. The tertiary history of the islands, on this
hypothesis, has been mainly one of subsidence along with
volcanic upbuilding. This subsidence has continued down to
comparatively recent times. Such topographic forms as Pearl
Harbor, Kaneohc Bay, Kahana Bay, can hardly be explained
except as valleys of subaerial erosion. On the northern coast
of Kahoolawe it may be noticed that the ravines and ridges
extend out into the sea in a succession of little bays and points,
although the present erosive action tends to wear away the
points and silt up the bays.

There is also, as many authors have noted, conclusive evi-
dence that on Oahu there has been a small Pleistocene eleva-
tion. This certainly amounted to 20 ft., as beds containing
recent marine shells indicate. According to Professor Hitch-
cock, the bluffs at Kahuku are of marine origin up to 60 ft.,

1 The following groups, now well established, are regarded as introduced
in the human period : Limacidce, PhilomytidcB, Helicidce, Opeas, Gcecilioides,
Planorbis, Viviparus and probably Ancylus and Musculium.

xlviii NOTES ON ZOOGEOGRAPHY.

above which they are consolidated dune formation. Some of
these elevated beds have been referred to by Doctor Dall as
Pliocene. This determination was not based upon actual
determinations of fossils, but, I gather, from general con-
siderations. My collections of marine fossils around Pearl
Harbor and east of Diamond Head show only recent species,
and lend no support to the idea that the raised beds are older
than Pleistocene.

The "raised reefs' I had opportunity to examine in west-
ern Molokai are consolidated calcareous sand. They are dune
deposits, as the irregular bedding clearly shows, although
they often contain sea shells, blown up from the shore. Not
much dependence is to be placed upon Hawaiian geology seen
from a steamer's deck. Upon the whole, it appears that in
Oahu there has been a very recent elevation, following a
subsidence of far greater amplitude.

A good summary of the physiography and geology may be
found in the earlier chapters of Hawaii and its Volcanoes,
by Dr. Charles H. Hitchcock, Honolulu, 1909.

Age of the land-shell beds.

In the Islands the distinction between Pleistocene and
Holocene deposits is a real and necessary one, since human oc-
cupation brought in factors profoundly affecting the physiog-
raphy of the whole lower zone, or on some islands the entire
area. It would also be inexact to call the Holocene beds "re-
cent", though some of them are apparently of no great antiq-
uity. If the terms are used loosely* in this book it is because
not enough work has been done to fix the age of the various
dune deposits definitely. The Manoa, Kailua and Kahuku
bluff deposits, and the beds of Mana, Hawaii, seem to be un-
doubtedly Holocene. The dune-covered beds of northwestern
Oahu, those of Moomomi, Molokai, and those of the neck of
Maui (which I have not seen), are doubtless older than the
first group, but may possibly belong also to the early human
period. The human remains in the Moomomi dunes, however,
are probably intrusive. The land shell breccia and tuff of
Diamond Head and other tuff cones of the Kona side of
Oahu, are undoubtedly Pleistocene.

NOTES ON ZOOGEOGRAPHY.

As yet, only the Acliatinellidce and Amastridcz of all these
beds have been thoroughly studied, and these are the groups
which everywhere show the greatest local speciation, there-
fore less useful for comparative studies of the beds than the
widely distributed small shells of other families.

The Diamond Head breccia contains a considerable num-
ber of extinct species. Dune sand, interbedded with land-
shell breccia, where the section has been fully exposed in the
sand quarry along the road, contains only recent marine shells.
I conclude therefore that the Diamond Head and Punch-
bowl land-shell deposits are Pleistocene, not Pliocene. A cer-
tain proportion of special species would naturally be ex-
pected in a region of such intense local endemicity as Oahu.
They are not of themselves an indication of age.

Whether any fossiliferous deposits older than Pleistocene
will be found on Oahu is doubtful. There is a possibility
of inland deposits, but if the movement of the island has
been mainly downward, as I believe, there is little probability
that marine Tertiary beds will be found near the surface.

Recent Climatic change.

That there has been a change from more humid to dryer
climate in many districts of all the islands, if not known by
historic evidences would be demonstrated by the restriction
of most land snails to higher levels than formerly occupied.
Sixty years ago the Achatiuellas were found in abundance
at half the elevations now inhabited by them. Still earlier
forest shells lived within a few feet of the present sea level, as
the Kailua and Kahuku deposits show. Forest shells,
Amastra and others, are found in many of the low deposits,
and it is obvious that the Pleistocene forests extended nearly
or quite to the sea on the northern and western coasts of
both Oahu and Molokai. The changes within the last cen-
tury are held to be due to deforestation by cattle, which by
destroying the underbrush cause the dessication of the forest
humus, and prevent reproduction of the native trees. In
Laiiai the wild goats have almost, and in Kahoolawe have
totally, completed the destruction of native forests. With

1 SYNONYMY OF ACHATINELLID/E.

loss of the forest there is less precipitation on the heated soil,
and what water falls is not conserved. Whether the earlier
destruction of low-lying forest was due to human agency is
not knOAvn, but in the absence of any other known cause
that seems likely.

The change from humid to arid conditions in the lower
zone seems to have been too rapid to admit of the evolution
of many arid country species. There is one Succinea which
lives under volcanic rocks in the most arid places; a species
of Bifidaria is often found in dry country, and a few Lept-
achatinas exist in similar places. In general however, one
finds little or nothing to encourage search on the slopes of the
lower zone.

The same tale of increasing aridity is heard from the tuff
cones of the Kona coast, where deposits of fossil forest-snails
occur on Koko Head, Diamond Head, Punchbowl, and others.
These desolate cones, furrowed with gullies, chill the traveller
approaching Honolulu with their austerity. Of herbage there
is now little, save for a thin line of dusty algarobas (Prosopis)
bordering the shore and straggling up the gulleys, and tufts
of dry grass which may harbor the dry country Succinea and
Bifidaria. In Pleistocene times, between periods of activity
and after the eruptions ceased, these cones were heavily
wooded, with a copious snail fauna. Amastra, Leptachatina,
Tornatellinidce, Lyropupa and Nesopupa, Endodonta and
Nesophila, Zonitidcc, etc. found congenial environment, if we
may judge by the abundance of individuals.

NOTES ON THE SYNONYMY OF ACHATINELLID^E.

The specific synonymy of Achatinellidcc has been discussed
by Pfeiffer in the various volumes of the Monographia Heli-
ceorum, by Newcomb (1858), Dr. W. D. Hartman (1888),
D. D. Baldwin (1893) and Mr. E. R. Sykes (1900), all of
whom have given synonymic lists of the species known to
them. Messrs. Gulick and Thwing have also made sugges-
tions. While the synonymy of the present work differs
widely from previous arrangements, it has not been thought
desirable to occupy space with criticism of the views of other

SYNONYMY OF ACHATINELLID^E. H

authors, except when occasionally some felicitous grouping
by one or another student inspired approving comment.

It would be quite impossible for an investigator knowing
these shells only by small museum series, already assorted,
and often imperfectly localized, to arrive at right conclusions.
Several factors combine to render the recognition of species
and races unusually arduous in Achatinella. The absence
of differentiation in structural characters is the chief hard-
ship. In any allied series the shape is about the same, and
the size is remarkably uniform throughout the genus. There
are no such extremes of size as in Partulina. Both shape
and size vary so much individually and with locality that they
cannot often be depended upon to separate allied species,
except in the average. There is very rarely any specific
sculptural modification ; and finally, the color and pattern are
kalaidoscopic in many races, and a special pattern is often
produced orthogenetically in several species.

Any polychromatic species is likely to produce individuals
of strange or unusual patterns, sometimes possibly as new
mutations, more often as a result of some unusual combina-
tion of color- factors in complex hybrid colonies. Such unique
or exceptional individuals may readily enough be referred to
their proper species if kept with their associates in life ; but
separated from their colonies, and without definite locality,
it is sometimes almost or quite impossible to tell what racial
stock they belong to. A number of the species described by
Dr. Pfeiffer from specimens sent by Mr. Frick were evi-
dently such exceptional, or as they say, 'freak' shells.
Any large collection contains shells which, if isolated and
described without locality, could be referred to their proper
species only with, the greatest difficulty if at all.

A further source of perplexity is the tendency of some
species to form melanistic or albinistic mutations. In the
absence of structural characters, such forms, when not ac-
companied by their associates, and especially if without exact
locality, may be difficult or even impossible to refer to their
proper species or race. This is particularly true of the Apex
or typical group of Achatinella.

Hi SYNONYMY OP ACHATINELLID^E.

Dissections of a considerable number of species indicate
that the soft anatomy shows even less specific differentiation
than the shells. Further work along this line should be done ;
but my impression is that very little assistance is to be ex-
pected from the soft parts in distinguishing closely related
species of Achatinella. I did not have time to test the value
of body or mantle color as a specific character. It is not
U\ely to be more constant than shell color.

Finally, we have to do with artifacts. Newcomb, Gulick
and others have commented upon the manufacture of "new
species ' ' of Achatinella, practised by some persons. These in-
dividuals were not actuated by a desire to advance science.
They were moved by cupidity, or a perverted sense of humor.
Perhaps all of the old collections contain some of these altered
or ' ' improved ' ' shells. The commonest alteration is a change
of hue by the use of hot water. Other patterns were altered
by scraping the very thin colored cuticle, sometimes adding
bands with colored ink, as in pi. 30, figs. 48, 48a, 51, 52. An
ingenious method practised years ago, which Dr. Cooke ob-
tained from an old resident of Honolulu, was to glue thread
around the suture or elsewhere, and hold the shell in the
smoke of a whale oil lamp. When the thread was soaked off,
some weird color-effects Avere produced.

The synonymy of Partulina is less involved than that of
Achatinella. The species differ much more in shape and size,
and are less variable in color and pattern. In other words,
there has been much more speciation. Moreover, the shells
have been collected less. Probably when Maui is more fully
explored some of the accepted species will be found to inter-
grade. The Amastridcu present 110 special problems in the
distinction of species. They are not more or less difficult
than ground-shells of other parts of the world.

HISTORICAL NOTES ON THE LITERATURE.

I.

The first Achatinellid shells brought to Europe, so far as
we know, were obtained by Captain George Dixon, who visited
the Hawaiian Islands in 1786 and 1787. They were strung

HISTORY OP ACHATINELLID^. 1m

on a lei or necklace, which seeins to have been made entirely
of Achatinella apexfulva and A. decora. It appears that
four specific names were based upon these specimens. The
shell described by Chemnitz as Turbo lugubris was doubtless
from the same lei. It was purchased in London by Spengler.
While the source of Lamarck's Monodonta seminigra is not
positively known (Delessert ascribing it to Captain Cook),
yet very likely it also was from Captain Dixon's lei. There
is no satisfactory evidence that Captain Cook's expedition
brought back any Achatinellidce.

The French corvettes Uranie and Physicienne visited the
Hawaiian group in 1819. They obtained 14 species of land
shells, which were worked up by Ferussac, at that time the
foremost authority on land shells. These shells were from at
least two sources: A. decora, lugubris and spirizona were
probably from a lei made in the Kawailoa-Helemano district,
in western Oahu. A. luteola has not been rediscovered. I
formerly thought it might be from the western slope of Hawaii
(Vol. XXI, pp. 321-2), but the little I could see of that
arid coast from the steamer's deck does not favor such a
theory. A. vulpina, gravida, lor at a, turrit ella, ventulus,
textilis, tristis, auricula were collected near Honolulu; ac-
cording to Dr. C. M. Cooke, probably in Pauoa valley, where
the same association still exists, by members of the expedi-
tion, perhaps in course of an ascent of Mt. Tantalus. En-
dodonta lamellosa and contorta were probably from the same
neighborhood. The latter is said to have been found on ferns,
a somewhat unusual station ; neither has yet been quite satis-
factorily identified.

The first American work on the genus was published by
Jacob Green in 1827, when Achatina stewartii was described
and figured. It was collected by C. S. Stewart, an American
missionary who spent the years 1823-5 in Oahu.

The voyage of Captain Byron in H. M. S. Blonde, 1824-5
supplied the progressive English naturalist William Swain-
son with several fine species from the Kawailoa-Helemano dis-
trict, where the stringing of shell leis or necklaces seems to
have been carried on extensively. From the study of these

liv HISTORY OF ACHATINELLID^.

shells Swainson in 1828 erected the genus Achatinella, with
the following species : A. pica, perversa, livida, bidimoides,
rosea, pulcherrima and acuta. The type of Helix ~byronii
Wood was also from the same source.

In 1845 Dr. J. W. Mighels described twelve species of
Achatinellidtz and many other Hawaiian shells, and in the
same year Dr. A. A. Gould published seven species, collected
by the Wilkes Exploring Expedition.

In the Proceedings of the Zoological Society of London
for 1845, p. 89, Doctor L. Pfeiffer offered some remarks on
Achatinella, which he here for the first time adopted as a
genus. He enumerated 18 species, described by Chemnitz,
Ferussac, Swainson and himself. A few additional species
were described in the following year, all from the Cuming
collection.

The monograph in Conchologia Iconica by Lovell Reeve,
published in April-May, 1850, brought the subject fairly up
to date. It was marred by erroneous identifications of some
of the earlier species, which were a source of error for many
years. 45 species are admitted and figured, 15 being de-
scribed as new from the Cuming collection; about half of
which are still considered valid. Some of the Cumingian
shells were received from Mighels ; others were doubtless from
Thomas Nuttall, who collected plants and shells in the Islands,
and who is known to have contributed Hawaiian marine shells
to Cuming 's collection. The figures of Reeve's monograph
are beautifully drawn and lithographed and well colored
in the original edition, but in separate copies of the mono-
graph I have seen the lithographs are coarse (from improper
removal of the varnish put on the stones to preserve the
drawings), and the coloring is crude.

II.

In the decade 1850-1860 the subject passed largely into the
hands of Hawaiian Americans. It was the era of the dis-
covery and definition of species. The quantity of work along
these lines accomplished in five or six years is marvellous.

HISTORY OF ACHATINELLID^E. lv

Dr. Wesley Newcomb, a physician-naturalist from New
York, collected extensively in Oahu about 1850-55 ; also in
Molokai and Lanai. His two chief papers, published in
1853-4, more than doubled the number of species known.
Newcomb was an excellent conchologist of the old school.
Most of the forms defined by him are still held to be valid
species or subspecies. He gave but little attention to small
or critical species, and his localities are given by districts or
islands.

In 1854 Doctor Pfeiffer enumerated the species then known,
122 in number, and classified them in seven sections. The
systematic classification of the group began with this sketch.
Numerous species were described by Pfeiffer from this time
to 1859, partly from specimens sent by Newcomb, but mainly
from shells collected by Mr. Frick. Many of these species
have proved difficult to recognize, and a large proportion of
them are synonymous with the earlier species of Newcomb
and others. This was no doubt due to the lack of adequate
material. Mr. Frick evidently assorted his specimens
shrewdly, to get the greatest number of "species", and
many of the forms described by Dr. Pfeiffer were based upon
one or two specimens of unusual color or shape. He gave
no definite localities.

Frick himself published only one paper 'Notes on
Hawaiian terrestrial Conchology (Sandwich Islands' Monthly
Magazine, I, May, 1856, pp. 137-140). He mentioned only
one Achatinellid shell by name, Achatinella gigantea-, but
numerous MS. names were given by him, some of which were
printed years later in Paetel's Catalogue, while others linger
only in the traditions of Island collectors. Among concho-
logists, Frick 's name is chiefly remembered for the difficul-
ties ensuing from his ignorant and unscientific methods. He
seems to have been an industrious collector.

Mr. G. S. Emerson began collecting Achatinellae in the
period of Newcomb and Gulick, and some of his finds were
recorded by both of these authors. A variey of A. livida was
named for him. Some time later, Mr. J. S. and Rev. 0. P.
Emerson carried on the work begun by their father. W. T.

M HISTORY OP ACHATINELLIIXE.

Alexander, better known for his work in other directions,
also contributed to Mr. Gulick's material.

John T. Gulick began collecting Achatinellid<z in 1850, at
the age of 18. Most of his collecting was done between that
time and 1853. His work differed from that preceding by
the careful attention given to locality and food-plants of the
snails. He was also the first to collect the small ground-
shells, especially Leptachatina, extensively, as he was the first
to appreciate their characters. Two descriptive papers were
based upon this most comprehensive and valuable of all the
early collections. The first was published in 1856 ; the second,
in collaboration with Mr. E. A. Smith, not until 1873, though
dealing with material collected in the early fifties.

Mr. Gulick's descriptive work, like his field work, was on
a higher plane than that of his contemporaries. He sought
to record the marvelous variation and differentiation of the
island fauna. The suggestive geographic relations of allied
forms made strong appeal to his imagination. One sees the
same trend of thought in C. B. Adams' Jamaican papers of
about the same date; and as everyone knows, Darwin had
been deeply impressed by similar phenomena observed in the
Galapagos.

Many of the species of Gulick are now given another inter-
pretation or value ; many have been confirmed by subsequent
investigations. His classification of the group (P. Z. S. 1873)
was a long stride forward, improving the arrangements of
Pfeiffer and von Martens in important details. Though this
work is concerned chiefly with systematics, allusion should
be made to Gulick's greatest intellectual service, his theory
of segregation in its several forms, as a necessary condition
of the evolution of species, originally suggested by his studies
on Oahuan Achatinellidse. First stated in 1872, it was fully
discussed in his ''Evolution, Racial and Habitudinal, Wash-
ington, 1905.

W. H. Pease, whose occupation as a surveyor took him
over Kauai and other islands, published several new species,
and in 1869 a classification and catalogue of AchatineUid(.

From 1855 on there seems to have been little scientific col-

HISTORY OF ACHATINELLIME. Ivii

lecting for about twenty-five years, and scarcely anything
was published on AchatinelUda, if we except the two papers
by Gulick (1873), which belong properly with his earlier
work.

III.

Several papers on the soft anatomy of Achatinellidce by
Wm. G. Binney and T. Bland, published in 1873-6, were im-
portant as showing that there are two radically diverse types
of dentition in the group.

Mr. D. D. Baldwin and Mr. Thwing began their extensive
collections about this time, and in the later eighties and first
half of the next decade a number of younger men were en-
listed, most of whom contributed new material for Mr. Bald-
win's papers. Dr. C. M. Cooke's collection, which was studied
by Professor Hyatt, and extensively used in this work, was
begun at this time.

Dr. W. D. Hartman, whose material came from Newcomb,
Pease and Baldwin, published a catalogue of the group in
1888. It was marred by many errors of all kinds.

Mr. D. D. Baldwin's papers, published from 1886 to 1908
were the most important expression of the work of this period,
and stimulated the study of the fauna by local naturalists.
His Catalogue of 1893 though modest in form, has been widely
used and quoted on account of its reliability in classification
and geographic distribution. His descriptive papers are
lucid and well considered.

The Mollusca of Fauna Hawaiiensis, by Mr. E. R. Sykes,
(1900), contains a useful synopsis of the group, and a bibliog-
raphy of Hawaiian conchology.

The "Reprint of the original descriptions of the genus
Achatinella" by Mr. E. W. Thwing (Occasional Papers
Bishop Museum, 1907) has proved very useful to Hawaiian
naturalists. The grouping of the species and the suggestions
relative to synonymy involved a good deal of original work
on the part of the compiler.

Herr Fr. Borcherding has published two profusely il-
lustrated papers, AchatineUen-Fauna der Sandwich-Insel Mol-

HISTORY OF ACHATINELLID^E.

okai, Zoologica, vol. 19, 1906, and Monographic der auf der
Sandwichinsel Kauai lebenden Molluskengattung Carelia,
Abh. Senck. Naturforscli. Ges. vol. 32, 1910. The first was
based chiefly upon material collected by Mr. Meyer and his
sons, purchased by Professor H. Schauinsland. It contains
a lengthy history of the literature of Achatinellidae and an
excellent bibliography.

IV.

In the last six or eight years there has been a revival of
interest among Hawaiian naturalists, and a number of large
collections have been formed. The exact data preserved, the
large series obtained from a great number of localities and
the new localities exploited, give these collections great value,
whether for systematic or for biometric investigations. Sev-
eral papers by Dr. Cooke have been based upon this material,
and it has also been utilized in these volumes. The few
months spent by the senior author in the Islands did not
permit anything more than a superficial study of these collec-
tions. Much of what value this work possesses is due to
information derived from material brought together by the
competent and enthusiastic Hawaiian naturalists of the pres-
ent time.

Genus NEWCOMBIA Pfeiffer.

Newcombia PFR., Malakozoologische Blatter, i, 1854, p. 117,
for A. Helena, physa, ornata, cwningi, plicata, newcombiana, cere-
alis, elongata. MARTENS, Die Heliceen, 1860, p. 249 (type A.
plicata Migh.) PEASE, P. Z. S., 1869, p. 649 (exclusive of
Sect. 2). SYKES, Fauna Hawaiiensis, p. 331.--BORCHERDING,
Zoologica, xix, Heft 48 ', p. 94. Newcombiana THWING, Occas.
Papers B. P. B. Museum, iii, no. 1, p. 138, 1907.

Shell oblong-turrited, usually coarsely sculptured, the embry-
onic shell of 3-J- whorls having comparatively coarse spirals (8
to 12 spirals on the second whorl), the last embryonic whorl
longitudinally striped. Outer lip slightly expanded or not;
columella straight or weakly folded.

Type N. plicata Migh. Distribution, Molokai and Maui.

Newcombia, by its distinct, minute, decurrent, spiral striation,
and boldly striped last embryonic whorl, is clearly a specialized
group of Partulinse. It is here accepted as of generic value only
by reason of its considerable secondary modification in sculpture
and shape of the columella. Moreover those concerned with
Hawaiian faunas now generally use the group as a genus. It
is obviously not on a par, systematically, with the genera of
Amastrinee.

N. carinella is the only dextral species known.

Pfeiffer's list under Newcombia was heterogeneous, but con-
tained several species (cumingi, plicata and newcombiana) , now
included in the genus. Professor Von Martens, in 1860, selected
A. plicata as the type. His second section of the group contains
species of Laminella. Pease, in his classification of 1869, asso-
ciated species of Heteramastra with his Newcombias. These dis-
crepant forms have been eliminated by Hartman (1888), Bald-
win (1893), and subsequent authors.

The distribution of closely related forms on Molokai and
Maui shows that the genus was developed in nearly its modern
condition while these two islands formed a single area. It is
somewhat peculiar that no Newcombia has yet been found on
Lanai, and I anticipate the discovery of the genus there, either
recent or as a fossil.

NEWCOMBIA.

Key to Species of Newcombia.

a. Parietal wall making an angle with the columella, the latter

having a low fold or twist.

l>. Surface smooth to the eye, finely striate spirally under
a lens. Molokai. N. perkinsi, no. 6.

N. philippiana, no. 5.
/A Surface spirally ribbed or coarsely striate, sinistral.

Molokai.
c. Pale, the later whorls cream-colored; length 15 to

25 mm.

d. Spirals acute. N. plicata, no. 1.

d *. Spirals rounded or subobsolete.

N. p. gemma j no. la.
c 1 . Dark colored; length 12 to 18 mm.

d. Spirals acute. X. camilicu'lata, no. 2.

d 1 . Spirals obtuse, base darker. N. sukata, no. 3.
//'. Surface with fine beaded spirals, a small carina at peri-
phery; zigzag-striped, dark-colored, dextral. Maui.

N. carmeHa, no. 4.
c/ 1 . Parietal wall continuous with the columella, which is not in

the least folded. Sinistral.

b. Longitudinal folds coarse and strong, rude spirals run-
ning over them. Molokai. .Y. ifciffrri. no. 9.
/A Longitudinal folds weak or wanting.

r. Very weak folds crossed by spiral threads, the inter-
vals of which are minutely wave-striolate spirally.

N. cinninyi, no. 7.

c 1 . Smoother; spiral threads weak or effaced, the minute
wavy striolation well developed. Molokai.

N. cinnamomea, no. 8.

c 2 . Wrinkled or puckered over spiral cords; no minute
striolation. Molokai. JV. c. ualapuensis, no. Sn.

Series of N. pi Ira (a.
1. N. PLICATA ('Mighels' Pfr.). PL 1, figs. 1, 2, 3, 6.

Shell sinistral, turrited, rather thin; sculptured with acute,
elevated, spiral liree and very fine longitudinal lines; brown-

NEWCOMBIA.

whitish, spire elongate, the apex rather obtuse. Whorls 6, a
little convex, those of the summit marbled with brown. Colu-
mellar fold obsolete. Aperture semioval ; peristome unex-
panded, acute. Length 15, diam. 6 mm.; aperture 6x3 mm.

Molokai (Mighels in Cuming coll.): Kalae (Meyer); Moanui
(Thaanum).

Achatinella plicata Mighels MS., PFR., Monographia Hel.
Viv., ii, p. 235 (1848). REEVE, Conch. Icon., pi. 6, f. 44a, b.
(1850). NEWCOMB, Ann. Lye. Nat. Hist, of N. Y., vi, 1858,
p. 312. THWING, Reprint Orig. Descript. Achat., 1907, pi. 3,
f. 11. Newcombia plicata Migh., BORCHERDING Zoologica, p. 94,
pi. 9, f. 1, la.Bulimus liratus PFR., P. Z. S., 1851, p. 261,
under B. newcombianus Pfr. (1853), no description; new naiuo
for plicata ' Migh.' Pfr.

The shell is perforate. The apex is obtuse, rounded, smooth
in the first half-whorl; then spirals begin above the suture at
the end of 1 J whorls; then about 6 to 8 smooth, subequal spiral
cords. On subsequent whorls the cords become more acute,
and some interstitial threads appear. The growth-stride become
strongly developed and irregular. The third whorl is longi-
tudinally striped with brown or pale olive, the rest of the
whorls being uniform cream color. The peristome in fully
adult shells is narrowly expanded and thickened within. Colu-
niella weakly sinuous, its edge dilated in a triangular white
callus above.

Length 18, diam, 8.2 mm.; whorls 6.

Length 25, diam. 9 mm.; whorls 6J.

Newcomb described the animal as "rather shorter than the
shell; tentacles at their origin closely approximating, short and
strongly clubbed; bottom of foot, mantle, and posterior part
light grey, entirely mottled with fine bluish dots. Motions at
first sluggish and timid but soon becoming bold, rapid and
gliding.'

la. N. PLICATA GEMMA (Pfeiffer). PI. 1, figs. 7, 8, 10, 11.

Shell subimperforate, sinistral, oblong-turrite, rather solid,
striatulate and spirally lirate (the lirse rather flattened, closely

4 NEWCOMBIA.

sulcate), white; spire turrited, the apex rather acute; suture sub-
marginate. Whorls 7, the upper ones flat, obsoletely varie-
gated with brown; following whorls rather convex, the last
about two-fifths the total length, somewhat carinated in the
middle by a more acute cord. Columella lightly plicate.
Aperture a little oblique, obauriform; peristome nearly simple;
columellar margin subreflexed, a little expanded outwardly.
Length 17, diam. 6.5, aperture 7x3 mm. Sandwich Is., Frick
in Cuming coll. (Pfr.).

Molokai: Pohakupili (Baldwin); Halawa (Borcherding, for
N. costata).

Achatinella gemma PFR. , Malak. Blatter iv, 1857, p. 230;
Monogrp.phia iv, 560; P. Z. S., 1858, p. 22. Newcombia gemma
Pfr., SYKES, Fauna Hawaiiensis, p. 331.

Achatinella (Newcombia) sulcata Pfr., BALDWIN, Catalogue,
1893, p. 8. Newcombia sulcata Pfr., GWATKIN; Proc. A. N. S.,
Phila., 1895, p. 238 (dentition). Newcombia costata BORCHER-
DING, Nachrbl. d. d. Malak. Gesell, 1901, p. 57; Zoologica, p.
95, pi. 9, f. 2, 2a.

The sculpture of the last whorl or two is more or less obsolete.
This form occurs associated with more strongly lirate shells, in
which the lirse are wider than in plicata, rounded, or more or
less marked with spiral striae. The gemma form is, in fact, only
an extreme phase, with obtuse lirse, of this race, which has been
called " sulcata Pfr. ' by Baldwin, and costata by Borcherding.
The description of the latter follows:

Newcombia costata [pi. 1, figs. 4, 5, copied from Borcherding].

Shell subperforate, sinistral, turrited, rather solid; encircled
with close, elevated line, more slender in the upper whorls,
then more rounded, and sculptured with very fine longitudinal
lines; last whorl with 7 rounded, flattened ribs above the mid-
dle, the ribs suddenly w r eaker below the middle, and becoming
stronger again around the umbilicus. Color yellowish-white,
glossy. Spire regularly tapering; apex rather obtuse; suture
linear. Whorls 6, flatly convex, the upper flamed with yellow,
the last forming half the total length, rounded. Aperture in-
versely ear-shaped, white within. Columella slightly folded,

iNE \VCOMBIA. >

white. Peristome expanded, with a white lip within; colu-
mellar margin recurved. Length 18, diam. 8, aperture 8x6
mm. Halawa, eastern Molokai (Borcherding*) .

In N. p. gemma, as in typical plicata, the Iira3 are variable in
arrangement and number, no two specimens being quite alike.
Figs. 8, 10, 11 represent typical gemma, while figs. 4, 5, 7 are
the " costata' ' type, intermediate between plicata and gemma.

2. N. SULCATA (Pfeiffer). PL 2, figs. 9, 10; pi. 14, figs 9,
10, 11.

Shell subperforate, sinistral, oblong-turrited, rather solid
striatulate, and encircled with close Iira3, which are compressed
in the upper whorls, then rounded; chestnut colored, somewhat
shining. Spire regularly tapering, the apex rather acute; suture
nearly simple. Whorls nearly 6, rather flat, the upper flamed
with white, the last about two- fifths the length, the base saccate,
of a deeper chestnut shade. Columella very lightly folded.
Aperture oblique, acuminate-oval; peristome thin, the colu-
mellar margin dilated above, reflexed, outer margin a little ex-
panded. Length 12J, diam. 5f mm. Aperture 5f mm. long,
3J wide. Sandwich Is., Frick in Cuming coll. (Pfr. ).

Molokai: Ahaina (D. Thaanum).

Achatinella sulcata PFR., Malak. BL, iv, 1857, p. 231; Mono-
graphia, iv, 560; P. Z. S., 1858, p. 22. Newcombia sulcata Pfr.,
SYKES, Fauna Hawaiiensis, p. 332, BORCHERDING, Zoologica,
p. 97, pi. 9, f. 4, 4a. Not of Baldwin, Catalogue p. 8, or of
Gwatkin, Proc. A. N. S.. Phila., 1895, p. 238 (radula).

" This species can be confused with no other. The whorls
are regularly, obsoletely, transversely striate, increasing in
strength to the last whorls and disappear on the lower half of
the last whorl. The color is red-brown, becoming more intense
with the increase of the w r horls, and on the last whorl it is quite
shining dark red-brown. The same analogy which exists be-
tween plicata and costata is here found between canaliculata and
sulcata. 1 (Borcherding, from specimens in the Hartman collec-
tion, probably received from Newcomb. )

Collectors of the present generation seem not to have found
this species until it was taken by Mr. Thaanum at Ahaina.

6 NEWCOMBIA.

The sculpture is most like that of N. plicata gemma, which
differs by its much larger size and lighter color, N. sulcata hav-
ing the last whorl rich chestnut colored, the next earlier whorl
chestnut or reddish, and the preceding one flamed with white.
The spiral cords of the last whorl vary in number and prom-
inence. Specimens from Ahaina are drawn on pi. 14, figs. 9,
10, 11. Length 11. diam. 5 mm. The figures on plate 2 are
copied from Borcherding.

3. N. CANALICULATA (Baldwin). PI. 2, figs. 5, 6. 7.

"Shell sinistral, very minutely perforated, somewhat solid,
acuminately turreted, apex subacute; surface sculptured
throughout with numerous acute, spiral keels, which become
blunter as they approach the apex, the interstices between the
keels exhibiting under a lens very delicate growth striae. Color
brown, upper whorls tessellated with brown and white. Whorls
6, slightly convex, lower one somewhat flattened at the base;
suture lightly impressed. Aperture oblique, oval, livid white
or light brown within; peristome acute, very lightly thickened
within, expanded, columellar margin reflexed over the small
perforation, margined with light brown on both face and the
reverse; columella very slightly developed, plain and smooth.
Length 14, diam. 6^ mm.

"Animal w r hen extended in motion as long as the shell.
Mantle slate color, margined with brown. Foot light slate,
studded on the sides and head above with spots of deeper shade.
Tentacles short and slender, dark slate.' (Baldwin.)

Molokai: Halawa (Baldwin); Punkolekole (Meyer).

Achatinellacanaliciilata'BALVWis, Proc. A. N. S., Phila., 1905,
p. 226, pi. 10, f. 28, 29. Newcombia c., Bald., BORCHERDING,
Zoologica, p. 96, pi. 9, f. 3, 3a.

Smaller and much darker than N. plicata, and with more
strongly developed, more acute spiral ribs than N. sulcata. The
spirals are at first equal and rounded, but become unequal and
acute on the later whorls, with some smaller threads in most of
the intervals, which are marked with growth-striaB. The color
is red-brown, indistinctly streaked along growth-lines with
whitish, the spirals also being whitish or pale at their summits.

NEWCOMBIA. 7

The third whorl, or the second, third and fourth whorls are
flamed with white. The figures represent cotypes. The largest
shell in the type lot of 12 measures length 14, diarn. 7.2 mm.
The smallest is 13.3 mm. long.

3. N. c. wailauensis n. subsp. PI. 14, fig. 5.

Larger, more robust than canaliculata, lighter in color, the
intervals between the spirals flesh-pink; spirals narrower.
Whorls 6.

Length 17 to 17.5, diam. 8.7 mm.

Length 16.5, diam. 8 mm. (smallest).

Molokai: Wailau (D. Thaanum). Cotypes in coll. Bishop
Museum and A. N. S. P.

4. N. CARINELLA Baldwin. PI. 2, figs. 1, 2, 3.

The shell is dextral, minutely perforate, oblong-turrited,
moderately solid; surface lusterless, sculptured with rather
coarse and unequal folds and wrinkles along growth-lines, and
numerous, minutely beaded spiral threads, 10 to 12 on the
penult, whorl; on the last whorl the threads are smaller and
closer below the acute peripheral keel; second and third whorls
with many even, close, nearly smooth spiral threads. Marbled
and zigzag-flamed with dark red-brown and white, the pattern
becoming more confused on the later whorls; the smooth apex
yellowish or fleshy. Aperture whitish or dark within. Peri-
stome blunt, yellow at the slightly expanded edge, usually
with a reddish submargin. Columella weakly folded, white,
reflexed.

Length 14, diam. 7 mm. ; 5J whorls.

Length 14, diam. 6.3 mm.; 6 whorls.

"Length 15, diam. 6.5 mm.; 6 whorls' (Baldwin).

East Mani: Nahiku (Baldwin).

Newcombia carinella BALDWIN, Nautilus, xix, April, 1906, p.
136.

A beautiful species, somewhat variable in contour and mark-
ing, but otherwise constant in a lot of 21 received from Mr.
Baldwin. With the sculpture of N. cumingi it unites the shape
of the plicata group, and is thus a connecting or synthetic form.

8 NEWCOMBIA.

It differs from all other Newcombias by being dextral. Figured
from cotypes.

Series of N. philippiana.

5. N. PHILIPPIANA (Pfeiffer).

Shell sinistral, subimperforate, fusiform-turrited, rather thin,
longitudinally striate, decussated with somewhat wavy close
spiral strise; brown; spire slender, the apex rather acute; suture
margined. Whorls 6, the upper flat, streaked with corneous
and white, the last whorl more convex, about two-fifths the
total length. Columella nearly simple, not folded. Aperture
oblique, angularly semi-oval; peristome simple, unexpanded,
the margins joined by a very thin callus, the outer margin
bordered with brownish, columellar margin dilated above, sub-
adnate. Length 15 to 15J-, diam. 6 mm. ; aperture 7 mm. long,
3 wide (P/r.).

Var. b. Smaller, paler, a little more distinctly striated
spirally (Pfr. ).

Sandwich Islands (Frick, Cuming coll.).

Achatinella philippiana PFR., Malak. BL, iv, 1857, p. 89;
Monographia, iv, 559. Newcombia p. , BORCHERDING, Zoologica,
p. 103.

This species is known by the original examples only.
Borcherding believes it to be an immature stage of N. perkinsi
Sykes, basing this opinion on the smaller size and thin shell as
described by Pfeiffer. This view, which seems quite probable,
was held by Baldwin and Thwing; but until young perkinsi can
be compared with the type of philippiana, and their identity
demonstrated, both may be let stand.

6. N. PERKINSI Sykes. PI. 7, figs. 18, 19.

" Shell sinistral, narrowly perforate, elongate-fusiform, solid,
ashy-brown, or elegantly marked with zigzag streaks of brown.
Spire slender, the apex obtuse, smooth. Whorls 6, rather flat,
the last about half the length of the shell; suture margined.
Columella subplicate. Aperture a little oblique, semioval, sub-
angular at the base; peristome simple, the columellar margin

KEWCOiMBIA. y

dilated above, adnate. Length 25, diam. 7.5 mm.; alt. aper-
ture 7.5, width 4 mm ' (Sykes).

Molokai (Perkins, Hutchison): Makakupaia (Baldwin).

Newcombia perkinsi SYKES, Proc. Malac. Soc. , London, ii,
October, 1896, p. 130; Fauna Hawaiiensis, p. 332, pi. 11, f. 36,
1900. BORCHERDING, Zoologica, p. 102, pi. 9, f. 6. Achatinella
(Newcombia) philippiana Pfr. , BALDWIN, Catalogue, p. 8.
THWING, Reprint Grig. Descript. Achat., pi. 3, f. 10.

"Specimens are to be found in some collections under the
name N. philippiana Pfr. The present species however is larger,
much more solid, the whorls are flatter, the coloring is lighter
and different, the columellar lip is more reflexed and the per-
foration more conspicuous ' (Sy Ices').

In unworn shells there is a faint longitudinal ribbing after
the smooth initial half whorl. Spiral striation sets in about the
middle of the second whorl, and color-flames appear on the
third whorl. The whole embryo consists of 3J whorls, the last
one usually profusely marked with zigzag brown streaks, and
more finely striate spirally than the other species. The post-
embryonic stages have no major spirals, but under a strong lens
are seen to be covered with a minute, irregular striolation, such
as has been described for N. cinnamomea, etc. This sculpture
is more or less effaced on the last whorl, often not discernible
there, especially in individuals having weak longitudinal folds.
The later whorls are mottled, marbled or streaked with white
and flesh-color, and the last whorl often has a peripheral angle
in front. The aperture is white within, with a brown lip-
border. The columella has a long, low fold or twist, and makes
an angle with the parietal wall, as in N. plicata; columellar re-
flection usually brown-tinted. Length 21, diam. 7.5 to 8 mm.,
6J whorls.

This species is apparently the most primitive existing New-
combia, having no trace of coarse spirals on the whorls. The
embryonic spirals also are finer than in other species, more as
in Perdicella. The coloration too is rather that of Perdicella.
Cf. N. philippiana Pfr. , which is probably an immature stage of
perkinsi.

10 NEWCOMBIA.

Series of N. cmningi.

7. N. CUMINGI (New comb). PL 3, figs. 1 to 6.

Shell sinistral, acuminate, turrited, with strongly marked
transversely oblique stria? and with longitudinal incremental
striae more or less developed. Color brown with undulations of
white at the upper whorls. Whorls 5, flatly convex; suture
moderately impressed, margined. Aperture oblong-ovate; colu-
mella slightly callous; outer lip thin, elliptical. Length .7,
diam. .2 inch (Newcomb).

East Maui: Haleakala (Newc. ). Makawao (Gulick, Bald-
win). W. Maui: Wailuku (Gulick).

Achatinella cmningi NEWC., Ann. Ly.c. N. H. of N. Y. vi, p.
25 (May, 1853); P. Z. S., 1853, p. 150, pi. 24, fig. 59 (1854).
PFR., Monographia iv. 559. THWING, Reprint Orig. Descript.
Achat., 1907, pi. 3, f. 12.

N. cumingi is imperforate. The embryonic shell of 3^ whorls
is spirally striate and flammulate as usual (pi. 2, fig. 13). The
following whorls have nearly obsolete longitudinal folds crossed
by crenulated spiral threads, usually coarsest in the peripheral
region of the last whorl, finer on the base; the intervals minutely
marked with spirally descending, wavy stria? (pi. 3, fig. 5, de-
tail from back of last whorl, Wailuku specimen). Very often
a peripheral thread is stronger, forming a low carina on the
front of the last whorl. The surface is lusterless, dull brown.
Aperture nearly w r hite with a brown border within the thin,
acute lip. The columella is long, straight, heavily white-
calloused, and continuous in direction with the parietal wall.

Length 21, diam. 7.2 mm.; 6 whorls. From Newcomb.

Length 20, diam. 7 mm. ; 6 whorls. Wailuku.

Length 19, diam. 6 mm. ; 6 whorls. Wailuku.

The types were from East Maui, but I see no difference be-
tween those before me from Makawao (pi. 3, fig. 4) and the
West-Mauian shells (Wailuku, pi. 3, figs. 1-3, 5).

8. N. CINNAMOMEA (Pfeiffer). PI. 3, figs. 7 to 11; pi. 14, figs.

1 to 4.

Shell imperforate, sinistral, fusiform-turrited, rather solid,

NEWCOMBIA. 11

opaque, longitudinally plicatulate, spirally sublirate and very
lightly granulated, cinnamon colored. Spire long, somewhat
rectilinear, the apex rather acute; suture nearly simple. Whorls
6, slightly convex, the upper marbled with brown and white,
last whorl about two-fifths the total length, tapering and chest-
nut-colored below the middle. Columella simple, receding.
Aperture slightly oblique, semioval, subangular at base; peris-
tome simple, unexpanded, acute. Length 19, diam. 5, aperture
7 x 3 mm. Sandwich Is., Frick in Cuming coll. (Pfr.).

Molokai : Mapulehu (Baldwin, Thaanum) ; Makakupaia
(Perkins, Meyer); Honomuni, Moanui, Kupeke and Ahaino
(Thaanum).

Achatinella cinnamomea PFR., Malak. BL, iv, 1857, p. 230;
Monographia, iv, 559; P. Z. S., 1858, p. 22. Newcombia
cinnamomea Pfr., BALDWIN, Catalogue, p. 8. SYKES, Fauna
Hawaiiensis, p. 331. Newcombia cumingii Newc. (in part),
BORCHERDING, Zoologica, p. 99, pi. 9, f. 7 (Makakupaia).

This form resembles N. cumingi closely in shape and micro-
scopic sculpture, but it differs by the weakness of the spiral
cords, which vary from weak to almost effaced. As this differ-
ential feature seems to hold in considerable series of both forms,
I think it best to retain the name cinnamomea for the Molokai
race, although the relationship to the Mauaian N. cumingi is so
close that perhaps cinnamomea might better be ranked as a sub-
species.

On account of the weakness of the coarse sculpture, the
minute, wavy, crenulated or granular striation (pi. 3, fig. 7) is
more distinct in this race than in Mauian cumingi. The axis is
often perforate. Immature shells are more strongly lirate than
the adults, and often could not be distinguished from N. cumingi.
N. cinnamomea is therefore to be viewed as a derivative of cumingi.

The fully formed embryo of 3J whorls, length 4.7 mm., has
a very heavily calloused columellar fold, but no lamella (pi. 1,
fig. 12). Younger embryos have the callus weaker, hardl}' 1
noticeable in those of two whorls (pi. 1, fig. 9). All are per-
forate.

The types were supplied by Frick who gave no locality other
than the " Sandwich Islands.' In specimens collected by New-

12 KEWLOMBIA.

comb at about the same time almost sixty years ago the color
of the last whorl is a ricn cinnamon, between raw sienna and
ochraceous (of Ridgway's " Nomenclature of Colors''), fading
upwards, and darker at the base (pi. 3, figs. 8, 9). This was
evidently the coloration of Pfeiffer's type. Some of the
Mapulehu shells are of this color, others being darker, as in
those from Moanui.

At Moanui (pi. 14, fig. 1) the base of the last whorl is burnt
umber, the upper part drab or olive, the transition often abrupt
at the periphery. Frequently there are numerous umber bands
occupjdng half the space between periphery and suture. The
size is greater than at other localities up to length 23J, diam.
8J, and length 24, diam. 8 mm.

At Ahaino and Kupeke (pi. 14, figs. 2, 3, 4) the shells are
small, the microscopic granulo-striatkm well developed, other
sculpture obsolete. The last whorl is ochraceous with several
umber bands and basal tract, or olive with an umber base.
The largest shells are 18 mm. long, 6 wide, most being between
16 and 17 mm. long. This local race may be called var.
decorata.

Specimens from Honomuni (pi. 14, figs. 6, 7) are very much
lengthened, ochraceous, base a little darker, spire worn, dull
violet or purple. Columellar reflection very small, adnate; a
"false umbilicus' visible in basal view. Length 19, diam. 6
mm., whorls 6. This form may be called var. honomuniensis.

8ft. N. c. UALAPUENSIS n. subsp. PL 2, figs. 11, 12.

Similar to N. cumingi in having well-developed spiral cords,
but the surface between and over them is minutely but strongly
puckered or wrinkled longitudinally (fig. 12), and not spirally
striolate as in cumingi and cinnamomea.

Molokai: Ualapue (D. Thaanum), no. 94490 A. N. S. P.;
Kahaanui (Thaanum).

The upper whorls are considerably eroded in the type lot, but
the younger individuals show the striped pattern of cumingi.
It is a little smaller than N. cumingi, length of figured type 20,
diam. 6J mm.

This race seems about equally related to cumingi and ciniia-
moraea, with incipient characters of pfeifferi.

NEWCOMBIA. 13

9. N. PFEIFFERI (Newcomb). PL 2, figs. 4, 8, 14.

"Shell sinistral, acuminate, turrited, with the surface irregu-
larly cut up into furrows, ridges and tubercles by deep longi-
tudinal sulcations crossed by strongly developed transverse
striae. Color brown, with white longitudinal lines on the upper
whorls. Whorls 6, flatly convex; suture deep. Aperture
oblong-ovate. Columella plain and smooth; outer lip simple.
Length .65, diam. .25 inch' (Newc.).

Molokai (Newcomb): Puukolekole (Meyer); Kaluaaha
(Baldwin).

Achatinella pfeifferi NEWCOMB, Ann. Lye. Nat. Hist, of New
York vi, p. 25 (May, 1853); P. Z. S., 1853, p. 150, pi. 24, f.
58 (1854). Newcombia pfeifferi Newc., SYKES, Fauna Hawaii-
ensis, p. 332. Bulimus newcombianus PFR. , Monographia iii, p.
414 (1853); P. Z. S., 1851, p. 261 (December, 1853). New-
combia newcombiana PFR., Malak. Bl. i, 1854, p. 119. PEASE,
P. Z. S., 1869, p. 649. A. newcombiana THWING, Reprint Orig.
Descript. Achat. , 1907, pi. 3, f. 9. N. neivcombia Pfr., HART-
MAN, Proc. A. N. S. Phila., 1888, p. 41. BORCHERDIXG,
Zoologica, p. 98, pi. 9, f. 5, 56.

Very distinct by the coarse longitudinal folds and rude spirals
of the last two whorls.

The initial half whorl is smooth; the next whorl has about 8
close, smooth spirals; on the following whorl these become
more widely spaced, and a color pattern of broad brown and
white longitudinal stripes appears, continuing to the end of the
embryonic stage, which comprises 3J- whorls. Weak folds then
begin on the neanic stage, gradually becoming stronger, the
color being indistinctly marbled with white on a dark brown
ground. The last two whorls are brown with the high points
of the ribs yellowish; the sculpture increasing in strength.
There is a minute sculpture of irregular, crenulated striae over
the coarser ribs and cords. The whorls are strongly convex.

Length 15, diam. 6 mm.; 5^ whorls.

Length 17, diam. 6 mm.; 6J whorls.

It varies in details of sculpture and in the degree of elongation.

14 PARTULINA.

Genus PARTULINA Pfeiffer.

Partulina PFR. , Malakozoologische Bliitter, i, p. 114, 1854
(for A. virgulata, tessellata, splendida, perdlx, goiddi, dubia).
VON MARTENS, Die Heliceen, 1860, p. 243, type A. virgulota
Migh. GULICK, P. Z. S., 1873, p. 90, same type. SYKES,
Fauna Hawaiiensis, p. 311, same type.

Includes as sections, Perdicella, Eburnella and Bcddunnia.

Shell ovate-conic, usually perforate, and having distinct
minute sculpture of spirally descending rippled strife. Embryonic
shell after the first whorl spirally striate. Color pattern of
oblique stripes or spiral bands, the last embryonic whorl gen-
erally striped. Lip expanded more or less; columella folded or
almost simple.

Type Partulina virgulata (Migh.). Distribution, Molokai,
Lanai, Maui and Hawaii, one species on Oahu.

The Achatinellina of the islands from Molokai southeastward
form a group of common ancestry, characterized by their spiral
and decurrent sculpture, perforate axis, oblique stripes and
spiral striae on the early whorls, etc. Some or all of these
characters may fail in particular species, but their general prev-
alence leads us to believe them the heritage from a common
ancestral stock. The Oahuan Achatinellina do not possess the
sculpture and color-pattern noted above as characteristic of the
other islands, and evidently are more related inter se than to
any Molokai- Hawaii snails. Although the difference between
Oahuan and Molokai forms are sometimes elusive and not to be
laid down in formal diagnoses, yet it may be allowable to ex-
press, by a generic separation, the idea that we have to do with
two collateral stocks of arboreal snails, the one on Oahu, the
other on the islands eastward.

Newcombia has the color and sculpture pattern of typical
Partulina, but with other special modifications which make it
advisable to treat that group as a genus.

Several subdivisions or sections of Partulina have been named.
They are not of much systematic importance, and there are
some intermediate species; yet as they seem to be natural groups
we admit them here.

PARTULINA, SECTION PKRDICELLA. 15

Sections of Partulma.

a. Columellar fold prominent, heavily calloused.

/>. Spiral sculpture distinct, at least at the early whorls;
shell not brilliantly glossy.

Section Partulma s. str., species 8-35.

6 1 . Spiral sculpture weak or obsolete; shell smooth and
glossy, light colored, uniform or banded.

Section Eburnella Pse. , species 36-47.
a 1 . Columellar fold wanting or weak, not calloused.

b. Spire slenderly conic; rather capacious forms; outer lip
narrowly expanded.

Section Baldwinia Ancey, species 48-53.
6 1 . Shell small, oblong or oblong-conic; outer lip not ex-
panded.

Section Perdicella Pse., species 1-7.

Section PERDICELLA Pease.

Perdicella PEASE, P. Z. S., London, 1869, p. 648. SYKES,
Fauna Hawaiiensis, Mollusca, p. 329, type A. helena Newc.

Small Partulintz having protractive stripes on the last embry-
onic whorl, later whorls striped or with a peripheral band;
sculpture Partulinoid; col umella nearly straight, or if developed
thecolumella fold is moderate, convex, scarcely or not truncate;
outer lip thin, not expanded. Type P. helena Nc.

Perdicella differs from Partulma by the unexpanded outer lip,
the weakness of the columella fold and the small size of the
shell. In coloration and sculpture it is altogether Partulinoid.
This affinity was recognized by C4ulick, who in his classification
of 1873 ranked Perdicella as a section of Pa-rial nia. Such species
as fulgurans, theodorei and winniei are directly intermediate be-
tween Perdicella and Partulina. Perdicella is not closely related
to Baldwinia, although in both the columellar fold has degener-
ated. It is a branch from the tessellata or marmorata series of
Partulinse.

Key to Species of Perdicella.

a. Molokai. Shell sinistral, usually with a light or dark band.

P. helena, no. 1.

16 PARTULINA, SECTION PERDICELLA.

a 1 . Maui.

b. Shell sinistral, with long, straight-sided spire, finely
zigzag-streaked or speckled, with a dark peripheral
band. P. ornata, no. 2.

b l . Shell dextral, oblong or ovate-conic.
c. Columellar fold distinct, convex.

d. Pattern of longitudinal chestnut lines, base
umber with a spiral umber band; col. fold
prominent; length 14 mm. P. zebra, no. 3.
d l . Boldly striped or zigzag-striped, without a spiral
band; columella white or pink, with a rather
strong fold.
e. Length about 13 mm. ; form narrow.

P. zebrina, no. 5.
e\ Length about 15-16 mm. ; form wider.

P. falgurans, no. 6.
c 1 . Columellar fold very weak.

d. Lip and columella brown; diam. not much over
half the length. P. mauiensis, no. 4.

d 1 . Columella white or tinged with purple-brown;
diam. nearly two-thirds the length.

P. kuhnsi, no. 7.

1. P. HELENA (Newcomb). PI. 4, figs. 1 to 7.

Shell sinistral, ovate-conical, with finely decussating stria? ;
rufous, alternating with broad longitudinal zigzag white lines
covering the entire shell ; the last whorl often encircled by a
white band. Whorls 5, rounded ; suture deep, simple. Aper-
ture ovate, columella slightly callous. Length 0.5, breadth 0.22
inch (Newcomb).

Molokai, on the Ti tree (Newcomb). Kamalo to Kalae
(Baldwin); Kalae and Makakupaia (Perkins); Kealia (Meyer).

Achatinella helena NEWC., Annals Lyceum of Nat. Hist, of N.
Y. vi, p. 27, April, 1853; P. Z. S., 1853, p. 151, pi. 24, f. 63;
1854, p. 311. PFR., Monogr., iv, 561. A. (Partulina) h.,
BALDWIN, Catalogue p. 6. Perdicella helena Nc. , PEASE, P. Z. S.
1869, p. 648. SYKES, Fauna Hawaiiensis, p. 330. BORCHERD-
ING, Zoologica, p. 75, pi. 4, f. 17 (Kealia), and 18 (Maka-

PARTULINA, SECTION PERDICELLA. 17

kupaia). THWING, Reprint Orig. Descript. Achat. p. 134, pi.
3, fig. 5.

Newcomb states that A. helena "is extremely limited in its
locality, which has been twice carefully searched by myself
without discovering larger specimens, dead or alive, that ap-
proached it in form. Several of the specimens contained young
in the oviducts.' It was taken " within the coil of the Ti tree
leaf, as it starts from the trunk. '

The shell is sinistral, perforate; summit obtuse. There are
fully 2J convex embryonic whorls, the initial half-whorl smooth,
the next two evenly, rather strongly striate spirally; last em-
bryonic whorl is variegated with broad, forwardly descending,
brown and white flames. The neanic and last whorls have a
much finer and very close spiral sculpture of minutely waved
or crinkled striae, and a variable pattern of zigzag streaks, inter-
rupted by a cream-white peripheral belt; the axis in a light
area. The outer lip is thin, columella very short, with a rather
strong callous fold, and broadly but shortly dilated.

Length 11.3, diam. 6.7, aperture 5.3 mm.; 5J- whorls.

Length 12.2, diam. 6.5, aperture 6 mm.; 5^ whorls.

Fig. 3 represents a typical specimen. Five specimens from
Newcomb, and others from Baldwin, the University of Wiscon-
sin, Cooke collection, etcetera, examined. There is considerable
variation in details of color-pattern, as in all related species.
Sometimes the stripes on the later whorls are smeared or partly
defaced, reduced to indistinct streaks. Occasionally the whitish
subperipheral belt does not appear until near the end of the last
whorl (fig. 4, Cooke coll.), or it may be entirely absent (fig. 2).
The color of the peripheral band, like the ground-tint of the
shell, varies from whitish to yellow.

Color var. balteata n. v. has a single brown belt at the peri-
phery, continuous or indistinctly interrupted, on a pale fleshy
or brown-tinted ground, the flames of the last embryonic whorl
faint (pi. 4, fig. 7).

A pretty color-form in the Cooke collection has the periphery
occupied by a series of dark spots, oblique in one example,
chevron-shaped in another, the flammules elsewhere reduced
to weak streaks or spots (fig. 6).

18 PARTULINA, SECTION PERDICELLA.

Var. minuscula Pfeiffer. PL 5, fig. 6. Shell subim perforate,
sinistral, ovate-turrite, rather thin, under a lens minutely decus-
sate, scarcely shining, brown-whitish; spire turrite-conic, the
apex slightly obtuse; suture simple. Whorls 5, very slightly
convex, the intermediate ones variegated with brown, last whorl
a little shorter than the spire, encircled with a brown band and
with a brown area on the rotund base. Columella scarcelv

V

plicate, semioval; peristome simple, acute, the columellar mar-
gin dilated above, reflexed. Length 10, diam. 5 mm. ; aper-
ture 4-| mm. long, 2f wide (Pfr. ).

Molokai Mountains at 4000 ft. (Perkins). Sandwich Islands
(Frick in Cuming coll., type loc. ).

Achatinella minuscula PFR., Malak. BL, iv, 1857, p. 231; P.
Z. S., 1858, p. 22, pi. 40, f. 10; Monographia, iv, 562. Per-
dicella m., PEASE, P. Z. S., 1869, p. 648. SYKES, Fauna
Hawaiiensis, p. 330. Cf. BORCHERDING, Zoologica, p. 76.

Pfeiffer' s figure, which we copy, shows a color-pattern like
that of some immature specimens of Helena. Borcherding con-
siders it a synonym of that species, an opinion which I think
well-founded. It seems to be a wider shell than the dextral
P. zebrina Pfr., which it resembles closely in coloration. Mr.
Baldwin reported minuscula from Lahaina, Maui, evidently con-
fusing it with the quite different P. kufmsi.

2. P. ORNATA (Newcomb). PI. 4, figs. 8, 9, 10.

" Shell sinistral, acutely pyramidal, shining; whorls 6, plano-
convex, margined above; suture well-impressed; aperture sub-
quadrate in adult, subovate in immature shells; lip slightly
thickened at the edge; columella broad and flattened; surface
of shell covered with alternating undulations or zigzag markings
of white and black arranged longitudinally, with a subcentral
transverse black band, sometimes margined with a white one
below on the last whorl. Length 10, width 4 twentieths of an
in ch ' ( Newcomb ) .

East Maui: in a deep ravine back of Lahaina (Newcomb,
type loc.). West Maui: Mt. Helu, 4000 ft. (W. F. Kaae).

Achatinella ornata NEWC. , P. Z. S., 1853, p. 149, pi. 24, f.
55. Pfr., Malak. Bl. 1854, p. 118; Monogr. iv, p. 561.

PARTULINA, SECTION PEUDICELLA. 19

THWING, Reprint Orig. Descript. Achat. , p. 136, pi. 3, f. 7.-
Perdicella ornata Nc. , BALDWIN, Nautilus, xix, p. 113.

A rare species, of which twenty specimens from the Newcomb
and Gulick collections, taken 50 to 60 years ago, are before us,
all "dead" shells, and a smaller series of "live" shells from
Baldwin. The straightly pyramidal spire and subangular peri-
phery give the shell an aspect of its own. The suture is very
distinctly margined by a band defined by an impressed line in
some shells, but in most of those seen there is no trace of such
margination.

First 1J or 2 whorls are flesh or whitish flesh-colored; next
half whorl has broad flexuous or irregular flesh-brown and
white stripes; after which the brown markings become narrow,
zigzag, and on the last whorl they are often dislocated, spotted
or mottled, though sometimes distinctly striped as in Newcomb's
type figure. On the last whorl there is a subperipheral brown
belt in all the specimens seen. It is bordered below by a pale or
cream-colored belt, more or less distinct, The columella is
dilated above in a small triangle, calloused, and convex, ob-
liquely trancated far above the base.

Length 14.8, diam. 6.5, aperture 6 mm.; 6 whorls.

Length 14.8, diam. 6, aperture 5.8 mm.; 6J whorls.

3. P. ZEBRA (Newcomb).

"Shell dextral, conically elongate, shining, with microscopic
decussating striae; whorls 5J round, narrowly margined above,
suture well marked; aperture ovate; lip thin; columella short,
abruptly terminating in a large prominent plait; color of epi-
dermis yellowish white, alternating with longitudinal chestnut
lines; base of an umber color, with a revolving line of the same
color above. Length 11, width 5 twentieths of an inch' [I3f
X 6J mm. (Newc.)

East Maui (Newcomb).

Achatinella zebra NEWC. , Annals of the N. Y. Lyceum of
Natural History, vi, p. 142, Oct., 1855. PFR., Monogr. ix, 537.

Described from a single specimen, and not figured. It seems
to resemble P. ornata in color-pattern, differing from zebrina Pfr.
by the basal band.

20 PARTULINA, SECTION PERDICELLA.

4. P. MAUIENSIS (Pfeiffer). PL 4, figs. 11, 12, 13, 14.

Shell subperf orate, ovate-oblong, rather thin, longitudinally,
irregularly striate, decussated with very close spiral striae visible
under a lens, hardly shining; whitish, variegated with serrated
brownish-corneous streaks. Spire regularly conic, apex obtuse,
suture shallow, slightly margined. Whorls 5J, rather flat, the
last nearly two-fifths the total length, somewhat tapering at the
base. Aperture little oblique, oblong. Columellar fold obso-
lete. Peristome simple, unexpanded, brown-bordered, the
columellar margin reflexed, subadnate. Length 13. diam. 6,
aperture 6x3 mm. (Pfr.).

West Maui (Newcomb); Makawao to Huelo (Baldwin).

Achatinella maniensis Newc. [typogr. err. for mauiensis]
PFEIFFER, P. Z. S., 1855, p. 207 (1856). A. mauiensis Newc.,
PFR. , Monographia, iv, p. 563. NEWCOMB, Amer. Jonrn. of
Conch., ii, 1866, p. 217, pi. 13, fig. 16. THWING, Reprint
Orig. Descript. Achat. , p. 134, pi. 3, f. 6.

The shape and coloration are about the same as in P. zebrina
Pfr. , but the aperture is dark brown within (sometimes over-
laid with whitish), and the brown columella has only a low
fold situated high, and not heavily calloused, much as in some
Bulimuli. The whorls of the spire are sometimes distinctly
margined below the suture. Figured from shells received from
Baldwin. Length 12.2, diam. 6.2, aperture 6 mm.; 5 J whorls.

5. P. ZEBRINA (Pfeiffer). PI. 5, figs. 5, 8 to 12.

Shell imperforate, dextral, rather thin, smooth, very minutely
decussate under a lens; glossy; whitish, very elegantly marked
with blackish-chestnut zigzag streaks. Spire somewhat turrited,
apex obtuse, brown; suture with a thread-like margin. Whorls
5, moderately convex, the last nearly three-sevenths the length.
Aperture slightly oblique, truncate-oblong, lilac within; colu-
mellar fold above, twisted, callus, slightly prominent. Peri-
stome simple, unexpanded, brown-bordered. Length 12J,
diam. 6, aperture 6x3 mm. Sandwich Is., Frick in Cuming
coll. (Pfr.).

East Maui: Honomanu (Baldwin).

PARTULINA, SECTION PERDICELLA. 21

Achatinella zebrina PFR., P. Z. S., 1855, p. 202 (1856); Mono-
graphia, iv, 561. A. zebra No., BALDWIN, Catalogue, p. 7.

''Related to A. ornata Nevvc. , but dextral and quite distinct
in the characters of the aperture ' (Pfr.).

This species is known to us by specimens collected by Mr.
Baldwin. About If whorls at first are uniform light reddish
brown. The next half-whorl has wide alternating, more or less
forwardly-descending stripes of white and dark brown. After
that, the stripes become narrower, more sharply defined, and
more or less irregular or dislocated, the light and dark stripes
being about equal in width. The ground remains white nearly
to the base in some shells, in others becoming coffee-tinted,
darkest at the base. The spiral striation of the embryonic shell
is extremely delicate. The interior is purplish in fully adult
shells. The columella has a rather heavy callous fold, situated
high, or sometimes heaviest at the lower end, but there is no
lamella properly speaking. The axis is imperforate. Length
13.2, diam. 7, aperture 6 mm. 5J whorls.

P. zebrina differs from P. mauiensis chiefly by the much
more prominently folded and calloused white or flesh-tinted
columella.

6. P. FULGURANS (Sykes). PL 5, figs. 1 to 4, 7.

Shell subperforate, dextral, ovate-turrited, glossy, closely
sculptured with spiral lines under a lens, whitish, very elegantly
marked with zigzag chestnut stripes. Suture moderately im-
pressed, the apex rather obtuse. Whorls 5^, plano-convex,
the last two-thirds the length of the shell. Aperture ovate-
piriform, lilac within; right margin of the peristome simple,
columellar margin somewhat reflexed; columellar fold twisted,
somewhat projecting, moderate, rapidly ascending. Length 16,
diam. 8, length aperture 8.1 mm. (Sykes).

East Maui: Makawao to Huelo (Baldwin).

Achatinella zebrina Pfr. , BALDWIN, Catalogue p. 8, not of
Pfeiffer. Perdicella zebrina Pfr., BORCHERDING, Zoologica xix,
p. 77, pi. 4, f. 19. Perdicella fidgurans SYKES, Fauna Haw-
aiiensis, Mollusca, p. 329, pi. 11, f. 5, 1900.

"This very pretty shell is akin to P. zebrina Pfr., but may

22 PARTULIXA, SECTION PERDICELLA.

be readily separated from it by its greater size, by being much
broader in proportion to the length, and by the color-pattern
being finer in design and more zigzag' (Sykes).

The embryonic shell is like that of P. zebr'ma, and the
sequence of patterns exactly similar. An almost fully formed
embryo is 4.2 mm. long, imperforate, with a straight, very
slightly calloused columella, 2^ whorls.

The columella resembles that of P. zebrina, and is much
more strongly twisted than in the East Mauian P. mauiensis.
The color-design is variable, but a majority of the shells before
us are similar to fig. 7. Mr. Sykes's original figure is copied,
fig. 4.

Length 15, diam. 8, aperture 7.6 mm.; 5i whorls.

Length 13, diam. 7.8, aperture 6.7 mm.; (not quite mature).

7. P. KUHNSI Pilsbry, n. sp. PI. 14, figs. 8, 12-15.

The shell is dextral, perforate, conic, thin; outlines of spire
straight; whorls moderately convex. Embryonic whorls white
or brownish, uniform or more frequently having festooned axial
stripes or two bands of lunate spots on the last half-whorl.
Subsequent whorls maculate with brown; the last whorl streaked
and maculate with brown, and encircled with white bands, of
which one on the upper surface, another below the periphery
and an umbilical area are the most constant. Other specimens
have the last embryonic whorl broadly striped with brown, last
whorl copiously zigzag-striped with chestnut on a white ground;
this pattern is sometimes interrupted by one or two white spiral
bands. Surface glossy, densely and minutely striate spirally,
the striation weak on the last whorl, Growth-striae weak.
Aperture white within; peristome thin, unexpanded; columella
white, or having a brown or purplish stain in the lower part,
nearly straight, but in oblique view a low fold is visible within;
sometimes the fold is obvious in a front view; outer edge tri-
angularly reflected.

Length 13.2, diam. 8 mm.; whorls 5-J.

Length 12, diam. 7.2 mm.

West Maui: Honokohua, type loc. ; Honokowai; Mt. Lihau:
Mt. Helu; Mauna Hoomaha; ridges above Lahaina; Hailaau;

PARTULINA. 23

Kapuna; Moomuku and Honolua (Thaanum). Cotypes in
coll. Bishop Museum and A. N. S. P.

Achatinella minuscula Pfr. , Thwing, Reprint Orig. Descript.
Achat., 1907, pi. 3, f. 8. ? Achatinella minuscula Pfr., BALD-
WIN, Catalogue p. 6, 1893. Not A. minuscula Pfr.

"Not common anywhere, but a widely distributed species.
It is undoubtedly the Part, minuscula Pfr. of Baldwin's Cata-
logue' (Thaanum').

It differs from P. zebrina Pfr. and fulgurans Sykes by the
much more slender conic spire, which resembles that of P.
helena.

This is a polymorphic species in which several mutations
have been established in the same colony. The original pattern
was probably zigzag-striped, the derivatives differing by the
development of white zones and coalescence of the interrupted
stripes in a spiral direction. Figs. 12 to 15 represent shells
from the type locality, 12 and 14 being the prevalent color-
forms.

Shells from Honokowai (pi. 14, fig. 8) are snow-white above
the periphery, yellow below it, the periphery marked by a band
of chestnut spots. There is also a very narrow dark crescent
behind the columella and some indistinct dark streaks and
spiral lines on the base. Length 14.5, diam. 8.8 mm. I have
not seen specimens from the other localities given by Mr.
Thaanum.

This species is named in honor of Mr. D. B. Kuhns, to whose
ability in the field we owe many new and interesting Achatinel-

Section PARTULINA (restricted).

Larger, more capacious shells than the Perdicdlce, and having
the columellar fold well calloused, and the outer lip usually
well expanded, thickened within. Sculpture and color-patterns
about the same. They live on Molokai (species no. 8 to 15),
Lanai (species no. 16), and Maui (species 17 to 35).

24 PARTULINA, MOLOKAI.

SPECIES OF MOLOKAI.

Key to Achatinellinae of Molokai.

NOTE. Yellow forms plain or with zigzag black stripes should
be looked for in Laminella, Vol. XXI, p. 345.

A. Shell elongate, dull, with distinct or strong spiral sculpture,
the columellar fold weak or wanting.

Newcombia, p. 1.

A 1 . Shell very glossy and smooth, porcelain-like, coloration
light and in bands, never longitudinal stripes or flames;
columellar fold strong.
a. Shell dextral.

b. Spire long; shell white, uniform or with gray or

light brown streaks, generally having a subperi-
pheral dark band or several bands; length 16-19
mm. mighelsiana, no. 45.

ft 1 . Spire shorter, concavely conic, the last whorl bulg-
ing; yellow, usually with a chocolate subsutural
line or band; length 13-18 mm. polita, no. 44.
6 2 . Spire straightly conic, last whorl not bulging, color
various. bella, no. 42.

a 1 . Shell sinistral, obesely ovate-conic, yellow, the embry-
onic whorls w r hite with a dark band.

subpolita, no. 43.

A*. Shell ovate-conic or pyramidal, showing moderate sculpture
under a lens; last embryonic whorl usually variegated;
columellar fold well developed.
6. Embryonic whorls having a dark spiral band.

virgulata, no. 8.

6'. Embryonic whorls having vertical or protractive stripes,
or sometimes plain.

c. Ovate-conic, rather large, length usually 23 to

26 mm.

d. Surface dull, fine spiral sculpture being well
developed; banded or streaked.

rw/rt, no. 10; tessellata, no. 9.
d l . Surface glossy; striped or plain.

proxima, no. 11.

PARTULINA, MOLOKAI. 25

c . Narrower, pyramidal-conic, sinistral.

d. Small, length 11 to 13 mm. Helena, no. 1.

d l . Larger, length over 17 mm.

e. Longitudinally boldly striped throughout;

length 18 mm. theodorei, no. 12.

e l . Suture bordered below by a dark band,

others at periphery and columella; lip

narrow, white; length 20 to 22 mm.

mucida, no. 13.

e 2 . Suture bordered by a white band or line;
lip and columella brown; length 21 to 25
mm. redfieldij no. 15.

e s . No conspicuous sutural border; variously
streaked, often with a subperipheral white
band, rarely others; lip wide, both lip
and columella white; length 22 to 26 mm.

divightii, no. 14.

Series of P. virgulata.

Embryonic whorls bicolored, having an upper white and
lower dark spiral zone; later whorls varying from many-banded
to plain. Shell either dextral or sinistral.

While this group has the characteristic sculpture and form of
the other Partulinse, it is remarkably aberrant in the color-
pattern of the embryonic whorls.

8. P. VIRGULATA (Mighels). PL 6, figs. 1 to 6, 9, 10.

' ' Shell ovate-conic, light fawn color, beautifully adorned with
dark brown bands, more or less numerous; iniperf orate; whorls
5, convex; incremental strise delicate. Aperture oblong; lip
reflected, slightly inflected. Length 1 inch, diam. f inch '
(Mighels).

Embryonic whorls nearly 3, the first purple-brown, the next
bicolored, having a wide snow-white band below the suture.
A narrow, more or less distinct cream-white subsutural band
continues to the end. Spiral bands appear about the middle of
the first neanic whorl, are often pale at first, becoming dark
chestnut on the last whorl, which is variously marked with

26 PARTULINA, MOLOKAI.

lines and bands. Under a glass the spiral striation is dense
and deeply cut, the striae rippled and descending. The axis is
either closed or slightly open. Lip slightly expanded and well
thickened within. The colnmella is reflexed in a short, wide
triangle, and the columellar fold varies from distinct to weak.

Length 25 to 26, diam. 15.2, length aperture 14 mm.

Length 22.5, diam. 15, length aperture 13 mm.

Length 23, diam. 14, length aperture, 11.6 mm.

Length 21, diam. 12.3, length aperture 10.5 mm.

Molokai, in the eastern part, Ualapue, Kaluaaha, Mapulehn,
Waialua, Halawa and Pelekunu (Meyer).

Partula virgulata MIGHELS, Proc. Boston Soc. of Nat. History,
II, 1845, p. 20. PFR., Monogr., iii, p. 454. Achatinella virgu-
lata High., REEVE, Conch. Icon., pi. 1, figs. 3, 4, 5, 5b. NEW-
COMB, P. Z. S., London, 1854, p. 311; Lye. N. H. of N. Y., vi,
1858, p. 312 (description of animal). THWING, Reprints Orig.
Descript. Achat., pi. 2, f. 8. Partulina virgulata Migh., BOR-
CHERDING, Zoologica, xix, Heft. 48, p. 50, pi. 1, figs. 1-20.
Achatinella rohri PFR., P. Z. S., 1846, p. 38. Bulimus rohri
PFR., Zeitschrift fur Malak., 1846, p. 115; Monographia, ii, p.
74. Bulimus insignis Mighels on labels, REEVE, Conch. Icon.,
Achatinella, species no. 3. Achatinella virgulata var. halawaensis
Baldwin in coll., BORCHERDING, Zoologica, xix, p. 52, pi. 1, f.
13-16, 1906.

This species is well distinguished by the broad band above
the suture on the embryonic whorls, usually purple-brown but
sometimes yellowish and quite pale, and the total absence of
flame markings. It is indifferently dextral or sinistral.

The typical form (pi. 6, figs. 1, 3), such as occurs at Mapu-
lehu, while varied a good deal, is rarely so dark as the forms
from the neighboring valley Kaluaaha, in which the light ground
is often reduced to narrow lines; or in Ualupue, the next valley
westward, the light ground is altogether extinguished (figs. 5, 6).
These dark shells are very glossy, with much fainter spiral striae
than the typical form, which has little or no gloss. Fig. 9 rep-
resents a form from Ualupue, fig. 10 from Pelekunu, after Bor-
cherding. Uniform purplish-flesh tinted examples, without
bands on the post-embryonic whorls, perhaps occur in all the

PARTULINA, MOLOKAI. 27

colonies. They are present in the lots of the typical form which
I have seen.

8a. Var. halawaensis Baldwin (pi. 6, figs. 7, 8, 11), which
occurs further east at Halawa, is white or pale straw-colored,
becoming darker towards the outer lip, without post-embryonic
bands, or with them narrow and few; lip brown-edged within.
Dark band on the embryonic shell normal, or rarely much
reduced. Contour long. Length 28, diam. 15.2, aperture
15 mm.

A peculiar form (pi. 6, fig. 2) from the University of Wis-
consin collection has the embryo white with a narrow chocolate
band, the last 2 or 3 whorls fawn w r ith many narrow longi-
tudinal darker streaks and only faint traces of spiral bands.
The shell is very glossy and small, length 19.7 to 20.5 mm.
Exact locality of these specimens unknown, but Herr Borcher-
ding, who has given a magnificent plate of 40 figures to the forms
of virgulata, figures this race from Pelekunu, on the northern
water-shed, together with small, slender, nearly white examples.

It is evident that a good deal of local differentiation has
taken place among the colonies of virgulata.

A. rohri was acknowledged by Pfeiffer to be a synonym of
virgulata. It was thus described: "Shell perforate, ovate-
conic, rather solid, longitudinally striatulate, decussated by
very close spiral striae; whitish fulvous, variously ornamented
with narrow chestnut bands; spire conic, rather acute. Whorls
6, but slightly convex, the last about as long as the spire, com-
pressed in the middle. Columella twisted, callous. Aperture
subtetragonal, glossy white within; peristome slightly expanded,
labiate within, the margins subparallel, the right margin shortly
arcuate above, columellar margin reflexed, nearly closing the
perforation. Length 24, diam. 13, aperture 13 mm. long, 6f
mm. wide inside.'

Group of P. tessellata.

Embryonic whorls unicolored or having obliquely axial
stripes.

28 PARTULINA, MOLOKAI.

9. P. TESSELLATA (Newcomb). PL 6, figs. 12 to 21.

"Shell sinistral, ovate-oblong, solid, with minute decussating
striae, color white or faw r n-colored, variously striped or not with
black and chesnut bands, upper whorls always tessellated with
black and white; whorls convex, the last somewhat inflated;
aperture white or roseate, ovate, effuse below; columella short
and broadly callous; columellar lip broad and slightly reflected.
Length 1 to 1.1 inch; breadth 0.6 inch.

" Body light gray, mantle slate color ' ' (iVewc.).

Molokai (Newc. ): Kalae, Kealia, Kalawao, Kahanui, Maka-
kupaia, and Pelekunu valley (Meyer).

Achatinella tessellata NEWC., Ann. Lye. N. H. of N. Y. , vi, p.
19, May, 1853; P. Z. 8., 1853, p. 139, pi. 23, fig. 26; 1854, p.
311. PFR., Monogr. , iv, p. 516. SYKES ; Fauna Hawaiiensis,
p. 319. THWING, Reprint Orig. Descript. Achat, pi. 2, f. 6.
Partulina tessellata Newc., BORCHERDING, Zoologica, p. 52, pi.
2, figs. 1 to 16, 1906.

Out of 57 specimens before me, 53 are sinistral. The figure
is stouter than in the following species, and it differs from
virgulata constantly by the obliquely striped last whorl of the
embryo. The first 1 to 1J whorls are some shade of brown;
then broad, slightly retractive stripes of opaque white and dark
brown alternate to the end of the embryonic shell. A clouded
or zigzag-mottled pattern ensues on the first neanic whorl. The
umbilicus is narrowly perforate, and the surface has little gloss
or none, the dense, wavy, spiral lineolation being well developed.

Length 26, diam. 17.3, aperture 14 mm.

Length 23, diam. 15.5, aperture 12 mm.

The typical form probably came from Makakupaia (pi. 6 figs.
16, 18, and fig. 12 received from Newcomb). The ground-color
of the later whorls is whitish, more or less streaked with fleshy,
with a varying arrangement of dark spiral bands and lines; lip
light liver-brown. Rarely the bands are absent, the later whorls
being profusely streaked (pi. 6, fig. 17).

At Kahanui most specimens have the post-embryonic whorls
uniform reddish-chocolate, by coalescence of the bands, or they
are banded with that color (figs. 13, 14, 15, 20), but some

PARTULINA, MOLOKAI. 29

specimens resemble the typical form in pattern. They occur
both sinistral and dextral, the latter rare.

From Kealia Borcherding figures shells with very weak or light
chestnut bands (fig. 19).

In the C. M. Cooke collection there is a set of nearly white
shells (fig. 21).

9a. P. T. MEYERI Borcherding. PL 7, figs. 1, 2.

'"Shell narrowly umbilicate, dextral, rather solid ovate-
conic, under a lens decussated regularly and very finely, glossy;
pale gray or grayish-brown, ornamented with light and darker
brown, or with gray and blackish streaks; spire conic, the apex
rather obtuse; suture crenulated, more or less margined, whorls
6, convex, the last \ the total length; upper whorls tessellated
with reddish-gray and yellowish-gray, the last w^horl rounded,
uncolored, or under the middle having narrow horn-colored
bands here and there. Aperture inversely ear-shaped, glossy
bluish within; peristome narrowly expanding and thickened,
with a reddish-brown internal lip. Columella twisted, calloused;
columellar margin reflexed, nearly covering the umbilicus.
Length 22 to 24, diam. 13, aperture 12 X 8 mm." (Borch-
erding. )

Molokai: Pelekunu (Meyer).

Partulina meyeri BORCHERDING, Nachbl. d. d. Malak. Ges.
1901, p. 55; Zoologica xix, p. 54, pi. 2, f. 17, 18.

Found in Pelekunu valley, where a large, dextral form of
tessellata reminding one of virgulata also occurs.

10. P. RUFA Newcomb. PL 7, figs. 3 to 11.

"Shell sinistral, conically ovate, solid, with decussating
strise; ground color externally and internally brownish red,
covered with an epidemis of a mottled brown and white, the
latter arranged on the central whorls in fine zigzag markings,
which are lost on the last whorl in a uniform grayish umber;
lower half of this whorl encircled by a broad white band,
whorls 6, flatly convex; suture plain, moderately impressed.
Columella strongly callous ; umbilicus open, small. Lip ex-
panded, reflected below. Length 0.9, breadth 0.5 inch'
(Newcomb).

30 PARTULINA, MOLOKAI.

Molokai (Newcomb): Kalae, Kaweeku, Kalamaula, Kahanui
and Makakupaia (Meyer), in the central part of the island.

Achatinella rufa NEWC., Ann. Lye. N. H. of N. Y. vi, p. 21,
May 1853; P. Z. S. 1853, p. 130, pi. 22, fig. 3; Ann. Lye. vi,
p. 324 (description of animal). PFR., Monographia iv, 537.
SYKES, Fauna Hawaiiensis p. 318. Partulina rufa Nc.,
BORCHERDING, Zoologica xix, p. 54, pi. 3, f. 1-9. Partulina
ides BORCHERDING, Nachrbl. d. d. Malak. Ges. 1901, p. 52;
Zoologica p. 56.

(1) Typical P. rufa (pi. 7, fig. 3) is a dull shell with the
growth-lines usually strongly marked, unequal, crossed by low,
usually wide and more or less granulose spirals mingled with
finer waved strife, the latter often predominating, especially be-
low the periphery. The sculpture is extremely variable, the
granulose spirals being very strong in some forms, such as that
described as idae, and in others much reduced, the finer decur-
rent striation then dominating. The summit is yellowish-
brown; last 1^ or 1 embryonic whorls are marked with protractive
yellowish-brown and white stripes; following neanic whorls have
a dense, confused zigzag pattern, often indistinct or illegible.
This may continue on the upper half of the last whorl, or it
may be replaced by a uniform or clouded tawny color, inter-
rupted by a white or yellowish belt just below the periphery.
The columellar fold is strong, calloused, brown tinted, the lip
of similar tint ; interior pink or dull brownish- violaceous.
Length 24, diam. 13.5, aperture 11.7 mm.

The original localit}^ on Molokai is uncertain. Borcherding
figures various forms from localities in the mountains south of
the northern peninsula, none of them just like the typical form,
represented by specimens before me received from Newcomb.
Borcherding tentatively suggests the view that P. rufa is "not
a pure species, but a bastard-form between P. tessellata Newc.
on the one side, and P. proximo, Pse. on the other.' While
this seems rather doubtful, and probably could be proved only
by breeding experiments, rvfa certainly unites characters of the
two species. The embryo has protractive stripes as in P.
proxima. Some very rare color-forms closely resemble proximo
in the later stages; but it differs from that species by the dull

PAETULINA, MOLOKAI. 31

surface, deeply sculptured spirally; proxima being glossy, with
weak spirals and very slight trace of the minute, wavy, spiral
lineolation of the dull Partulinas. PI. 7, figs. 6, 7 represent
two forms of rufa from Kaweeku, copied from Borcherding.

(2) There is a dull brown (reddish to olive brown) form, with
a few scattered light spots, much as in a common form of /'.
warmorata (pi. 7, fig. 4). Sometimes this form has a suit-
peripheral white zone (fig. 5); the sculpture either as in viir.
idie or without major spirals, clothed throughout with minute,
crinkled, spirally descending striae.

(3) Another form has zebra-stripes of white and brown, ex-
actly as in one form of proximo,, the surface having fine deeply
cut Partulina sculpture. Length 28.5, diam. 16.5, aperture 15
mm.; whorls 7 (fig. 8).

(4) Similar to var. ida, except that the spiral engraving is
close, nearly even, without granules. Last W 7 horl of the embryo
having very obliquely protractive brown stripes; following
whorls pale yellow, darker towards the base, or showing some
faint tawny stripes on the spire. Aperture pinkish white, colu-
mella and narrow lip pale flesh-pink. PI. 7, fig. 9, 10, 11.

Length 23, diam. 14.2, aperture 12 mm.
Length 17, diam. 10.2, aperture 8.2 mm.

(5) Partulina rufa var. idee Borcherding (pi. 7, figs. 13, 14,
15), from Kalae and Kealia, may easily be recognized, accord-
ing to Borcherding, by its compressed and ventricose shape and
the regularly granulose spiral sculpture (fig. 13). The last
whorl is more or less distinctly uniform yellowish-brown, or has
a yellowish-white belt below the middle, the upper whorls
being zigzag marked. In a series of 40 specimens before me I
find the characters so variable that I cannot look upon P. idee as
anything but a local race. The sculpture in particular inter-
grades perfectly with typical P. rufa.

In many specimens of idte the color above the periphery is
pale or even whitish, and on the base it becomes yellow, rich
brownish-yellow or olive, the change being gradual in some,
abrupt in other examples. In many of the pale forms the
stripes of the last embryonic whorl are faint or even not dis-
cernible.

32 PARTULINA, MOLOKAI.

11. P. PROXIMA (Pease). PL 9, figs. 1 to 9.

Shell sinistral, im perforate, oblong-ovate, ventricose: rather
solid; finely striated transversel} 7 , striae somewhat flexuous,
granulose and interrupted, more regular and conspicuous be-
neath the sutures. Whorls 6, convex, marginated, the last
being somewhat produced obliquely and flattened on the middle,
so as to give in some specimens a subangulated appearance at
the base. Aperture somewhat oblique, of an oblong-ovate
form; columellar fold strong. Color chestnut-brown, striped
and mottled irregularly with darker brown and white; columella
and inner edge of aperture purplish red (Pease}.

Molokai (Pse. ): Maunahui, Kahanui, Waikolu, Makakupaia,
Pelekunu, Makolelau and Kamalo (Meyer).

Helicter proximus PSE. , P. Z. S , 1862, p. 6. Achatinella prox-
imo, Pse., PFR. , Monographia, vi, 1868, p. 168; viii, p. 217.
HARTMAN, Proc. Acad. Nat. Sci., Phila., 1888, p. 27, pi. 1, f.
1, 2. Partulina proxima Pse., BORCHERDING, Zoologica, xix, p.
63, pi. 4, f. 1-12.

"The above species from the island of Molokai appears to
represent the H. marmoratus and its varieties of the island of
Maui. It is, however, larger and heavier; the last whorl has
also a peculiar shape, in common with that of several of the
larger species found on Molokai. All the specimens I have
seen are sinistral, and the columella and edge of the aperture of
a deep red ' (Pease).

The first 1^ whorls are white; then very obliquely protract! ve
brown stripes appear, continuing to the end of the embryonic
stage, which comprises 3J whorls. The succeeding neanic stage
has stripes in general parallel to the growth-lines, but more or
less interrupted or irregular. This pattern may continue to and
upon the last whorl (figs. 1, 5) ; or the stripes may become
nebulous or reduced on the last two whorls. The ground-color
varies from glossy white to strongly brown tinted, or the last
whorl may be brown tinted towards the base, white above. In
some shells the stripes are very obliquely retractive on the later
whorls (fig. 2).

In another variety the stripes are tawny yellow on a white or
slightly coffee-tinted ground. The sutural margination is dis-

PARTULINA, MOLOKAI. 33

tinct (figs. 7, 8). According to Borcherding, this pattern comes
from Pelekunu.

Another form has close hair-like lines, of brown, cream, and
pale lilac, in harmony with growth-lines, over the whole post-
embryonic shell. It is glossy, with typical embryonic colora-
tion (fig. 9). Borcherding figures a shell of this pattern from
Makakupaia.

In fully adult shells the outer and basal lips expand slightly.
The lip and columella are brown in nearly all specimens, but
in some light ones it is a vinous brown. The interior varies
from nearly white to lilac or violaceous of various tints. The
surface is glossy, and the spiral strife weak or subobsolete.

Length 26.5, diam. 15.5, aperture 14 mm.; 6f whorls.

Length 24, diam. 14, aperture 13 mm. ; 6^ whorls.

Color- var. schauinslandi Borcherding. PL 9, figs. 10, 11.

The shell is very glossy, white or yellowish white, uniform
or with faint, pale buff lines along growth lines on the last one
or two whorls. Interior white or pink, the narrow lip roseate-
brownish or nearly white, collumellar fold fleshy brown. The
suture has a distinct narrow margin defined by an impressed line.
Length 24, diam. 13, aperture 12 mm. ; 6J whorls.

Young shells are sometimes white above, yellow below the
angular periphery. In some specimens there are faint stripes
on the last embryonic whorl. These are transitions from
schauinslandi to the paler forms of proximo, such as that from
Pelekunu valley, clearly showing its genesis from the latter,
the only difference being loss of color. This has been fully set
forth by Herr Borcherding, and is well shown by the large
series I have examined.

Molokai: Kaluahauoni and Waileia (Meyer).

12. P. THEODOREI (Baldwin). PI. 9, figs. 14, 15.

Shell sinistral, minutely perforate, rather thin, narrowly
pyramidal, glossy, with sculpture of rather rude growth lines
and minute, rather weakly incised, spiral stria?. Embryonic
whorls whitish, the last one usually marked with broad, protrac-
tive brown stripes. Following whorls have irregular purple-

34 PARTULINA, MOLCKAI.

brown flames or streaks on a white ground, in general harmony
with growth-lines, widening downwards, scarcely reaching to the
suture above. On the last whorl the streaks are often some-
what dislocated. Whorls nearly 7, but slightly convex, the
suture rather indistinctly margined below. Aperture small,
lilac within; lip narrow, not expanded or noticeably thickened,
brown-bordered within, columellar fold very low, brown.

Length 18.6, diam. 8.8, aperture 7.7 mm.

Length 18, diam. 8, aperture 7.3 mm.

Molokai: Kawela (Baldwin).

AchatineUa theodorii BALDWIN, Catalogue etc., 1893, p. 7,
(name only). AchatineUa theodorei BALDWIN, Proc. A. N. S.,
Phila., 1895, p. 226, pi. 10, f. 27. SYKES, Fauna Ha-
waiiensis p. 330. Partulina theodorei Bald., BORCHERDING,
Zoologica xix, p. 65, pi. 4, f. 14, and f. 15, var.

Figured and described from the types. While closely related
to Partulina proximo, by the sculpture and coloration, this species
is constantly much narrower, in a considerable series seen.
The columellar fold is also less developed. P. proximo, does
not occur in Kawela valley, the type-locality of theodorei.

Mr. Sykes placed this species in Perdicella, a position ap-
proved by Professor Hyatt. Herr Borcherding groups it with
proxima Pse., in Partulina. It is one of several species connect-
ing Perdicella with Partulina.

Var. midtistrigata n.v. (pi. 9, figs. 12, 13, 16) differs by its
broader shape and much more numerous, narrow stripes. The
types are without exact locality but Borcherding figures a sim-
ilar shell from Makakupaia. Length 20.5, diam. 11, aperture
9.8 mm.

13. P. MUCIDA (Baldwin). PL 9. fig. 17.

Shell sinistral, minutely perforate, solid, ovate-pyramidal,
the spire straight-sided, ra.ther acute at apex. First 2 to 4
whorls tawny brown, the last two or three whorls slightly con-
vex, whitish, mottled and banded with flesh color, and having
purple brown bands speckled or mottled with white above and
below the suture, at the periphery and around the axial perfor-
ation. The surface is glossy, with low, rude growth-lines and

PARTULINA, MOLOKAI. 35

faint spiral striation. Aperture small, bluish white within, lip
slightly expanded, nearly white and well thickened within.
Columellar lamella strong, white with a light brown patch at
and above its insertion. Length 21 to 21.2, diam. 11.8, aper-
ture 9.6 to 10 mm. ; 6f to 7 whorls.

Molokai: Makakupaia (Baldwin).

Achatinella mucida BALDWIN, Proc. A. N. S., Phila., 1895, p.
222, pi. 10, f. 23. SYKES, Fauna, p. 315. Partulina muc/d
BORCHERDING, Zoologica, xix, p. 73, pi. 6, f. 7, 8.

This small species, described and figured from the types, has
a peculiar mildewed or mouldy appearance from having the
dark markings overlaid with white. It is very closely related
to A. dwightiij but differs by its smaller size and details of col-
oration. No form of dwightii has a dark band along the suture
below. The figure represents the same shell figured by Bald-
win.

13a. Color-var. macrodon Borcherding. PI. 9, figs. 18, 19, 20.

Last embryonic whorl zigzag striped; following whorls white
with a faint flesh tint, having distinct purple-brown bands,
minutely flecked with white, at suture, periphery and umbili-
cus. Columellar lamella strongly calloused, projecting hori-
zontally. Length 20 to 22, diam. 11.5 to 12, aperture 9.5 to
10 mm. ; whorls 6f to 7.

Makakupaia (Meyer and others).

Partulina macrodon BORCHERDING, Nachrbl. d. d. Malak. Ges. ,
1901, p. 56; Zoologica, xix, p. 74, pi. 6, f. 9, 10.

14. P. DWIGHTI (Newcomb). PL 8, fig. 1 to 5.

"Shell oblong conical, sinistral, solid, surface covered with
longitudinal strire, cut across by very numerous, fine, slightly
undulating cross-stria?. Whorls 6, flatly convex, crenulated at
the sutures; suture simple, well marked. Aperture ovate; colu-
mella broad, short, and slightly twisted; outer lip expanded,
subreflected below. Subumbilicated, color a grayish white,
with numerous blotches and zigzag markings of brown, more
obscure on the last whorl. Aperture and lip of a dingy white.
Length 19, width 9^ twentieths of an inch [about 24 X 12 mm.]
(Newc. )

36 PARTULINA, MOLOKAI.

Molokai (Newcomb): Makakupaia, Kawela, (fig. 2), Mako-
lelau, Kamalo and Punkaeha, fig. 1, (Meyer); Kamalo
(Baldwin).

Achatinella dwightii NEWCOMB, Ann. Lye. N. H. of N. Y. vi,
p. 145, October, 1855; P. Z. S. 1855, p. 207; American Journal
of Conchology ii, p. 213, pi. 19. f. 9. PFR., Monogr., iv, p.
517. SYKES, Fauna Haw. p. 313.- -TnwiNG, Reprint Orig.
Descr. A chat. pi. 2. f. 7. Partulina d. , BORCHERDISG, Zoologica
xix, p. 68, pi. 5, f. 1-8.

P. dwightii is characterized by the solidity, narrow form with
a high, straight-sided, acute spire, and the streaked, mildewed
appearance. Typically the last embryonic whorl has protractive
stripes and the following whorls of the spire show fleshy-brown
mottling or confused zigzag patterns on a pinkish or yellowish-
white ground. Last whorl streaked, usually paler below the
suture. The surface has rather rude, irregular but low growth -
lines, and on the last whorl there are regular, distinct, waved
spiral stria3 in shells from Newcomb (fig. 5). In most other
lots seen the spirals are quite weak or faint. The aperture is
small, and the rib is placed rather far within, so that there is an
unusually wide whitish lip. The outer edge often expands
noticeably. In fully adult shells the columellar lamella is
strong and white or nearly so. Some specimens are white with
only faint brown streaks and stains, the aperture delicately
lilac-tinted, lip white. In others the brown streaks are dark,
in part blackish, and nearly cover the last whorl.

Length 26, diam. 15, aperture 13 mm.

Length 25, diam. 13.5, aperture 12 mm.

Length 22, diam. 13, aperture 10.8 mm.

The shape is more straightly pyramidal than in A. redfieldi,
which also differs by its brown columellar fold and the narrower
and brown lip. Its range of patterns is also different.

14a. P. dwightii compta (Pease). PI. 8, figs. 6, 7, 8.

Shell sinistral, narrowly perforate, rather solid, oblong-conic;
spire obsoletely transversely delicately rugose-striate, the
wrinkles disappearing on the last whorl, and longitudinally im-
pressed with delicate, irregular striae. Spire turrited or oblong-

PARTULINA, MOLOKAI. 37

conic; suture impressed, whorls 6, convex, sometimes sub-
angular above, the last equal to half the length of the shell.
Aperture vertical, ovate; columellar fold superior, strong, white;
lip slightly expanded, callous within; leaden-gray, encircled
with lines or bands of reddish-brown, the last whorl ornamented
with a white band at the base. Length 25, diam. 13 mm.
(Pease).

Molokai (Pease); Kawela and Makakupaia (Meyer).

Partulina compta PEASE, Journ. de Conchyl. xvii, 1869, p.
175; P. Z. S. 1869, p. 647. BORCHERDING, Zoologica xix, p.
57, pi. 3, f. 14-16, 18-20.

"This species, which comes from the island of Molokai, is
very closely related to P. splendida Newc., which lives on Maui.
It is more contracted at the base and sometimes has a peculiar
angulation of the upper part of the whorls. In coloration it is
nearer P. grisea Newc." (Pse.).

The ground-color is white or nearly so in the specimens fig-
ured by Borcherding and before me in the Academy, Cooke and
University of Wisconsin collections. Otherwise they agree well
with Pease's description. It would probably be more correct
to describe the shell as brown, pale or dark, cut by white bands
and lines. It differs from A. redfieldi by its white coiumella
and lip. The following race is almost fully connected with
compta by intermediate specimens, and may perhaps be found
to be superfluous. I retain it here because both Borcherding
and Hyatt have recognized it, though under different names.

146. P. dwightii concomitans Hyatt, n. var. PI. 8, figs. 9 to 13.

Similar to dwightii in the acutely pyramidal contour, solidity,
wide white lip, etc. ; but it differs by the predominance of chest-
nut-brown (rarely olive-brown) which forms nearly a solid
color on the last two whorls except that there is a subperipheral
white belt and sometimes a white line on the upper surface,
continuing on the penultimate whorl; the suture has a narrow
white edge.

Molokai (A. N. S. P. and Cooke collections): Makakupaia
(figs. 9, 10, 12), Kawela (fig. 11) and Makolelau (Meyer).

Partulina grisea Nc., BORCHERDING, Zoologica xix, p. 69, pi.

38 PARTULLXA, MOLOKAI.

5, f. 9-16. Not Achatinella grisea Newcomb. ? Achatinella
redfieldi var. </, SYKES, Fauna Hawaiiensis, p. 318, pi. 11, f. 16
(Makakupaia and Kamalo, Perkins).

Borcherding figures a long series of this form, varying from
pale brown and gra} T to deep chestnut, all having a subperi-
pheral white belt. I have not seen gray examples and copy his
figure of one (pi. 8, fig. 9). Fig. 13 is the type, no. 106061,
A. N. S. P.

While dwighti and concomitans stand near grisea Nc. , they
differ from it decidedly by the pattern of the early whorls and
the stronger columellar lamella. The shell is also larger and
more solid than that of grisea, which is certainly distinct. We
cannot agree with Herr Borcherding that P. grisea occurs on
Molokai.

15. P. REDFIELDII (Newcomb.) PL 7, fig. 12, 16, 17; pi. 8,
figs. 14 to 20.

"Shell sinistral, elongate, conical, striated longitudinally,
color light fawn passing into deep chestnut, paler above, plain
or with transverse chestnut bands with obscure undulation upon
the third whorl only; a white band also traverses the suture,
whorls 6; suture well impressed, slight!} 7 margined. Aperture
subovate; columella brown, flat and twisted; lip slightly re-
flected, the color of the columella. Length 1, breadth 45 inch
(New., 1853).

Animal light flesh color, mantle dark slate (Newc., 1854).

" Animal as long as the shell, grayish above, sprinkled slightly
with brown, tentacles of same color, mantle slate, bottom of
foot of a greenish gray ' (Newc. , 1858).

Molokai: Mapulehu (Baldwin); Ualapue (Thaanum); Maka-
kupaia and above Kamalo (Perkins); Kamoku, Kawela, Mako-
lelau, Kaluaaha and Ualapue (Meyer).

Achatinella redfieldii NEWC., Ann. Lye. N. H. of N. Y. vi, p.
22, April, 1853; p. 325; P. Z. S. 1853, p. 131, pi. 22, f. 5;.
1854, p. 311. PFR., Monogr. iv, 523. SYKES, Fauna Hawaii-
ensis p. 317, pi. 11, f. 15. Partulina redfieldii Newc., BOR-
CHERDING, Zoologica xix, p. 71, pi. 6, f. 1-6.

The outlines of the spire are less straightly pyramidal than in

PARTULINA, MOLOKAI. 39

P. dwightii, which is usually more solid, often larger, with a
white columella and lip. Newcomb's description of the color-
pattern is not quite clear, owing to the failure to place a comma
after the word "bands' in his original description, reprinted
above. This was rectified by him in the P. Z. S., where in
both Latin and English versions he makes it plain that it is
only the "undulations' (or axial stripes) which are restricted
to the third whorl. Sixteen specimens in coll. A. N. S. P., in
two lots received from Newcomb, agree well with his account.
Some of them are figured on plate 8, figs. 15 to 18. It will be
seen that the color-pattern varies widely. In some specimens
the last two whorls are rich brown, uniform or varied with
spiral bands and lines, the spire paler above, suture white-
edged, lip and columellar fold brown. Other shells have a
light cream ground on which there are few or many brown
spiral lines and bands, or in place of them there may be dark
narrow streaks. The last embryonic whorl may have oblique
or zigzag stripes, but they are often lacking (pi. 8, figs. 15-18,
from shells received from Newcomb). Newcomb's type figure
is copied, pi. 8, fig. 14.

Length 25, diam. 14, aperture 13 mm.

Length 21, diam. 13, aperture 10.5 mm.; 6i whorls.

Length 24.5, diam. 13, aperture 11.3 mm.; 6J whorls.

There is a continuous intergradation between the dark brown,
rich orange-brown and longitudinally lineolate forms, all hav-
ing a snowy and usually rather broad subsutural band. Often
the tawny, longitudinally lineolate form has white or pale
brown spiral bands (figs. 15, 16), producing, when the bands
become numerous, the pattern shown in Newcomb's figure
(copied in pi. 8, fig. 14), which I take to be the typical pattern,
though Mr. Sykes has figured one of the dark, bandless shells
as typical. It is no great step from some shells of this pattern
to P. dwightii concomitans.

In the Cooke collection there are white specimens with the
lip very pale brown, apical whorls fleshy (pi. 7, fig. 17), with
others showing traces of the brown bands on the base and latter
part of the last whorl. A peculiar shell has narrow brown lines
near suture and on periphery, and a patch at the axis, on a

40 PARTULINA, LANAI.

pale brownish ground, the spire white, becoming fleshy above
(pi. 7, fig. 12).

Another form (no. 2068, Cooke coll. ) has a snow-white spire
and coffee-colored last whorl, aperture pink within, lip brown
(pi. 7, fig. 16).

In some examples there is a sutural margin banded by an
impressed line, as in P. proximo, schauinslandi. This is probably
a local race.

The color-forms of A. redfteldii may be arranged thus:
A. Uniform dark or light brown with a white sutural border.
Ab. The same, with darker spiral bands.

a. Yellowish-brown with narrow streaks of darker

brown along growth lines.

ab. The same cut by white or whitish spiral bands.
b. White predominating, leaving only traces of dark

bands or none.

To what extent these patterns mingle in the same colony is
not known to me.

SPECIES OF LANAI.
16. P. CRASSA (Newcomb). PI. 18, figs. 1, 2.

The shell is usually dextral, narrowly umbilicate, acutely
ovate-conic, solid, chestnut (the shade varying) with zigzag
lines, dots and streaks of a whitish tint, and a white zone below
the periphery; spire typically a little attenuated above, the whorls
scarcely convex. Surface dull or slightly glossy, closely and
rather deeply engraved with spiral lines. Suture scarcely im-
pressed. Aperture ovate, the lip expanded, sometimes almost
reflected, white, very strongly thickened within, the callous
usually irregular. Columellar lamella obtuse but rather strong.
Columellar lip built forward, reflected.

Length 20, diam. 14 mm.; 6J whorls (typical).

Length 18, diam. 13 mm.

Length 19, diam. 12 mm.

Length 22, diam. 14 mm.

Lanai (Newcomb); near Koela (Perkins).

Achatinetta crassa NEWC., P. Z. S. 1853, p. 155, pi. 24, f. 71

PARTULINA, MAUL 41

(1854). PFR., Monogr. iv, 527. SYKES, Fauna Ifawaiiensis
p. 312.--THWING, Reprint Orig. Descript. Achat. pi. 2, 1 10.

This very distinct species is more closely related to P. dwightii
concomitans of Molokai than to any other existing form. In the
series of 70 specimens in the collection of the Academy only one
is sinistral. There is a good deal of variation in the extent of
light markings, but the subperipheral band seems rather con-
stant, being absent in only one example. The dark color varies
from chestnut to chocolate, and is more or less streaked with
lighter and darker. One specimen (pi. 18, fig. 2) has the
whole post-embryonic shell banded as in P. tessellata a notable
advance in pattern, foreshadowed by a few others having several
basal bands.

The embryonic shell, 5 mm. long, has the first two whorls
light fleshy brown, whitish below the suture; some broad, indis-
tinct darker clouds then appear, changing on the last whorl to
zigzag axial stripes of dark brown and cream- white ground.
The axis is perforate, columella a little convex (pi. 18, figs. 3, 4).

SPECIES OF MAUL

The numerous species and varieties fall into three series: that
of P. marmorata, stout, ovate forms with streaked or striped
pattern; series of P. splendida, thinner shells with profuse spiral
bands and lines; and the series of P. tappaniana, mainly light
colored, rather narrow forms, uniform, or with spiral bands on
a white ground.

Series of P. marnwrata.

17. P. KAAEANA Baldwin. PI. 10, figs. 5, 6, 7.

"Shell sinistral, subperforated, solid, globose with a conical
spire, apex subacute; surface rather lusterless, covered with
rather coarse wavy growth lines, and under a lens exhibiting
close and delicate decussating spiral lines; nuclear whorls faintly
decussated. Color ashy brown, with a light brown band just
below the periphery which enters the aperture; whole surface
covered with minute longitudinal white flecks or streaks; apex
tessellated white and brown. Whorls 6, not margined above,
flatly convex; suture lightly impressed. Aperture a little

42 PARTULINA, MAUI.

oblique, oval, livid white, showing the outside coloring within.
Peristome acute, slightly thickened within, expanded, colu-
mellar margin reflexed, light brown on both face and the re-
verse. Columella white, terminating in a strong, plaited, pro-
jecting tooth.' (Baldwin 1 ).

Length 21, diam. 13.8 mm.; whorls 6^.

Length 19.8, diam. 13 mm.

"Animal extended in motion longer than the shell. Mantle
brownish-black with outer edge bordered with a narrow white
line. Foot below and sides light slate color. Head above and
tentacles dark slate and granulated' (Baldwin).

West Maui; Mt. Helu, 4000 ft. elevation.

Partulina kaaeana BALDWIN, Nautilus xix, p. 113, February,
1906.

A dull brown shell, yellows-banded below the periphery,
speckled with whitish flecks, or with the last whorl having only
some indistinct flecks and streaks, dark on each side of the light
band. Another specimen is reddish brown w r ith many irregular
streaks and spots of white, but without the subperipheral light
belt. The last embryonic whorl is striped white and brown,
as usual. While related to marmorata, this is a quite distinct
species. Figured from the type lot.

"This shell was found on a mountain peak quite isolated
from the main mountain mass of West Maui. We dedicate it
to Mr. W. F. Kaae who seems to have been the only one in
quest of shells who has ventured to climb this lonely peak. He
found the shell in company with Perdicetta ornata Nc., a species
supposed long since to have become extinct,' 1 (Baldwin').

18. P. MARMORATA (Gould). PL 10, figs. 1, 2, 3, 4.

"Shell dextral, rather solid, of an elongated, acutely conical
form, composed of 6 convex whorls which are somewhat
shouldered superiorly; the last large in proportion, and ven-
tricose. Surface delicately striated with lines of growth. Color
a dusky chestnut, marbled with white irregular and angular
markings, generally arranged somewhat in longitudinal stripes,
the whole seeming to be coated with a very thin, ash-colored
epidermis. Aperture moderately large, ovate, the lip slightly

PARTULINA, MAUI. 43

expanded; the columellar fold white, compressed, nearly trans-
verse, and standing out very prominently. Length f , breadth
Mnch" (Old.)

East Maui: foot of Mt. Haleakala (U. S. Exped.); Makawao
(Baldwin); Kula (Gulick).

Achatinella marmorata GLD., Proc. Bost. Soc. N. H. ii, p.
200, March, 1847; Expedition shells p. 35; U. S. Exploring
Exped., Mollusca, p. 85, pi. 7, figs. 94, 94a.--PFR., Monogr.
iii, p. 455. NEWCOMB, Ann. Lye. N. H. of N. Y. vi, p. 311.
-TnwiNG, Reprint Orig. Descript. Achat. pi. 2, f. 13. Achati-
nella adamsi NEWCOMB, Ann. Lye. vi, p. 19 (May, 1853).
A. adamsii NEWCOMB, P. Z. S. 1853, p. 137, pi. 22, f. 20.

Perforate, dextral, rather solid, gray-brown, with scattered
white spots or irregular stripes dark-bordered on the left side.
Last embryonic W 7 horl having wide, strongly protractive white
and brown stripes. Surface dull, without gloss, by reason of
the dense, microscopic, crinkled spiral striae. Gould's figure is
copied, fig. 1.

Length 25, diam. 14.3, aperture 13 mm.; 6^ whorls.

Length 21.2, diam. 13.3, aperture 11.3 mm.; 6 whorls.

Length 21, diam. 12.2, aperture 11 mm.; 6-J whorls.

Newcomb described the animal as "light pea green, strongly
granulated, one-third longer than the shell, tentacles light slate;
mantle and base of foot same color as the body. ' He included
as synomous: A. adamsi Newc., A. induta Gul., A. plumbea
Gul., A. ustulata Gul. Mr. Thwing states that "in over one
thousand examples only one sinistral specimen was found.'

li A. adamsii. Shell dextral, conically acuminate, whorls 6,
finely striated transversely, roughly striated longitudinally;
suture deeply impressed, lip expanded, columella terminating
in a strong, abruptly twisted plait. Interior of shell of a leaden
blue color, exterior dark chestnut, irregularly mottled and
striped with white. Length 18, diam. 9-twentieths of an inch.'
A. adamsii is terrestrial in its habits.

19. P. PLUMBEA (Gulick). PI. 10, figs. 8 to 12.

"Shell dextral, sometimes perforate, ovate-conic, solid, shin-
ing, irregularly striated, very finely decussated; lead-colored

44 PARTULINA, MAUL

with a white band on the periphery of the last whorl; apex sub-
acute; spire conical; suture marginate, well impressed. Whorls
6-J, very convex. Columellar fold central, white, strong. Aper-
ture slightly oblique, sinuately oval, bluish- white within; peri-
stome brown, thickened within, with dextral margin arcuate,
scarcely reflected; columellar margin dilated, white, sometimes
slightly detached; parietal margin wanting. Length 23, breadth
13 mm.; length of body-whorl 16J mm. Average weight 6
grains; least w r eight 4.3 grains.

Var. b. Globose, conic, spire shortened, with concave out-
lines, body-whorl inflated.

"Var. c. Without the white band on the body-whorl'

East Maui : Kula, on the trunks of the aiea and kukui.
(E. Bailey).

Achatinella plwnbea GUL., Ann. Lye. N. H. of N. Y. ? iv, p.
213, pi. 7, f. 39, 1856.

Very closely related to mannorata, perhaps not always dis-
tinguishable, but polished, typically belted with white, the
band rarely wanting, and without the special marking of mar-
morata. The typical form from Gulick (pi. 10, figs. 9, 10) is
usually speckled on the spire with white on a fawn-brown
ground. A lot from Baldwin, also from Kula (pi. 10, figs. 11,
12) is drab with darker and lighter streaks, the shell smaller.
Another lot from Baldwin, locality not stated, is fawn with

*/ j

narrow brown streaking, with or without a belt, sometimes with
some scattered whitish dots (fig. 8).

Large specimens from Gulick measure 25 mm. long, 14.5 wide.
The smaller drab form is 22.5 x 12 mm.

20. P. WINNIEI Baldwin. PI. 13, figs. 1, 2, 3.

Shell sinistral, subperforate, rather thin but moderately
strong, long-conic, somewhat glossy. First 1| or 2 whorls of
the embryonic shell flesh-colored, next half whorl boldly
marked with broad brown and white axial stripes. Following
neanic and adult whorls cream-white, copiously striped with
dark chestnut, the stripes more or less irregular and unequal.
Growth-lines and spiral sculpture rather weak. A sutural mar-

PARTULINA, MAUI. 45

gin is defined by an impressed line. Aperture flesh-colored
within, the lip narrowly thickened, a trifle expanded, white.
Columellar fold rather strong, white and oblique.

Length 16.5, diam. 8, aperture 7.8 mm.; whorls 6J.

Length 15, diam. 7.8 mm.; whorls 6.

Length 16, diam. 8 mm.

West Maui: Kahakuloa.

Partulina winniei BALDWIN, Nautilus, xxii, p. 67, November,
1908.

"Named in honor of Miss Winnie of Wailuku, Maui.'
Figured and described from cotypes sent by Mr. Baldwin, who
states that it is the counterpart of Partulina theodorei Bald, of
Molokai. There is a strong resemblance, but I am inclined to
think winniei a diminutive relative of P. perdix. It approaches
Perdicella in structure of the columella.

21. P. PERDIX (Reeve). PL 10, figs. 14, 15, 16; pi. 13, figs.
10 to 13.

"Shell ovate, ventricose, rather thin, dextral, spire exserted,
somewhat obtuse at the apex, whorls 6, convex, margined
around the upper part, obliquely sculptured with flexuous im-
pressed lines, columella plicately twisted, rather callous,
whitish, painted and varigated throughout with olive-brown,
sutures and columella white' (Reeve).

W. Maui: Lahaina (Baldwin, here selected as type loc. );
Olinda at 4000 ft. (Perkins); Honokowai (D. Thaanum).

Achatinella perdix REEVE, Conch. Icon, vi, pi. 6, f. 43 a, b
(May, 1850). SYKES, Fauna Hawaiiensis p. 315. --BALDWIN,
Catalogue p. 7.--THWING, Reprint Orig. Descript. Achat. pi.
2, f. 20.

Achatinella undosa GULICK, Ann. Lye. N. H. of N. Y., vi
p. 20/>. pi. 7, f. 33, 1856.

The apex (1 to 1^ whorls) is brownish white; the next one
or 1 J w r horls have broad protractive alternating stripes of snow-
w r hite and dark brown. The neanic whorls following have
narrow stripes of the same colors, running with the growth-
lines. This pattern may continue upon the last whorl, or the
stripes may become wider and bisinuate or angular there, with

46 PARTULINA, MAUL

a streak of darker brown on the left side of each white stripe.
In some shells the white is reduced to irregular spots on the last
whorl. These patterns are figured. The surface is rather
glossy, showing fine spirals under the lens; suture margined
with a white cord. Axis imperforate. Length 22, diam. 12^,
aperture 10 mm. ; 6f whorls.

The shade of brown varies a good deal in different lots.
Lahaina specimens are figured, pi. 10, figs. 14 to 16, and pi.
13, fig. 13. In pi. 13, figs. 10, 11, 12, we figure three sinistral
specimens from Honokowai, collected by Mr. Thaanum.

Var. undosa Gulick (pi. 10, figs. 17, 18, 19). Usually more
slender than perdix, the spire longer, stripes generally narrow,
sometimes pale and linear; distinctly perforate.

Length 22, diam. 11.8, aperture 10.6 mm.; 6J whorls.

Length 20, diam. 11.9, aperture 10.9 mm.

Length 21, diam. 11, aperture 9.6 mm.

Mountain ridges of Waihee, on the ilima (Sida.}, ki (Cordy-
II lie terminalis) and other low bushes (E. Bailey).

21a. P. perdix pyramidalis (Gulick). PL 10, fig. 20.

"Shell dextral, very rarely sinistral, imperforate, pyramidal,
rather solid, shining; chestnut or ash brown, with a white or
yellowish band encircling the base, with oblique white and
brown markings on the second and third w r horls, bluish-white
within the aperture. Apex subacute; spire regularly conical,
sometimes decollated. Whorls 6J-, convex, margined with white;
finely striated transversely, and microscopically decussated with
faintly impressed wrinkled spiral lines; the last regularly
rounded, equal to 65 hundredths of the length. Columellar fold
well developed, white, oblique. Aperture rather oblique, semi-
oval; peristome thickened within, very slightly reflected; with
columellar margin dilated, adnate or slightly detached; parietal
margin wanting.

"Length 21.6, breadth 11.4 mm., aperture 9.66 mm. The
dimensions of a large specimen are as follows: Length 23,
breadth 12, aperture 10 mm. Average weight 5.7 grains.'
(Gulick).

West Maui: Lahaina, on trees (S. T. Alexander); Huelo
(Baldwin); Waihee (Perkins).

PARTULINA, MAUL 4<

Achatinelia pyramidata GUL., Ann. Lye. vi, p. 204, pi. 7. f.
32, 1856. --BALDWIN, Catalogue p. 7. SYKES, Fauna Hawaii-
ensis p. 317.

Figured from a typical specimen from the Gulick collection,
no. 92704 A. N. S. P.

" This shell differs from the A. splendida Newc. in its smaller
size and pyramidal form, with spire less convex, body-whorl
less ventricose, aperture smaller and less oblique, lip less ex-
panded and reflected. It also differs in being without perfora-
tion, and is very rarely sinistral.

" The A. perdix Reeve differs from this in its broader and
more ventricose form, its more convex spire, with whorls more
swollen, with aperture broader, and frequently subangulated,
the arrangement of colors is also different. Looking only at the
type of this species, it would be placed in the same group with
A. splendida, yet it is so closely connected by intermediate varie-
ties with A. perdix that it has hitherto been considered a variety
of that species.

" Var. b. With black basal band. A small speciman has the
following dimensions: length .8, breadth .43, length of aper-
ture .35 inch, weight 4 grains.

"Var. c. Without band. I have a small specimen of the
brown variety of A. splendida which closely resembles this.

"Var. d. With apex chestnut-brown, not tessellated; very
rare.

"Var. e. With irregular white spots. This variety passes
into A . perdix. ' ( Gulick).

22. P. USTULATA (Gulick). PI. 13, figs. 5, 6.

"Shell sinistral, perforate, ovate-conic, obliquely produced at
the base, solid, scarcely shining, striated and microscopically
distinctly decussated, brown with a whitish band encircling the
base. Apex subacute, tessellated with white and brown; spire
conical, somewhat curvilinear; suture simple, lightly impressed,
marked with a narrow white line; whorls 6^, moderately con-
vex, the last large, equal to 72 hundredths of the length. Col-
umellar fold strong, white, oblique. Aperture very oblique,
sinuously oval; peristome thickened within, reflected anteriorly,

48 PARTULINA, MAUL

*

with columellar margin dilated, white, unattached; parietal
margin wanting. Length 25.4, breadth 14.5, length of aperture
13.2 mm. Weight 10 grains" (Gulick).

Maui: Beautiful Valley, on the mamaki (S. T. Alexander).

Achatinella nstulata GUL., Ann. Lye. N. H. of N. Y. , vi, p.
211, pi. 7, f. 37, 1856.

"This species is more ventricose than A. perdix Rve. and
differs in having the whorls more convex, the suture simple,
the aperture more oblique and the sculpturing more distinct.
I have received but two examples for which I am indebted to
the brothers James and Samuel Alexander ' (Gulick).

Also reported from Lahaina (Baldwin). I have copied
Gulick's figure (pi. 13, fig. 6), and add another (fig. 5) repre-
senting a shell from the Gulick collection from Kahoma, W.
Maui. It has two light bands on the last whorl, and the spiral
sculpture is very weak.

23. P. INDUTA (Gulick). PL 13, fig. 7.

Shell dextral, perforate, ovate-conic, solid, unpolished, micro-
scopically very minutely and obsoletely decussated, brown.
Apex subacute, tessellated with white and chestnut; spire con-
vexly conical; suture lightly impressed, not margined. Whorls
6, slightly convex, the last equal to 66 to 72 hundredths of the
whole length. Columellar fold strong, white, nearly transverse.
Aperture slightly oblique, sinuously oval, bluish-white within;
peristome thickened within, with anterior margin slightly re-
flected. Columellar margin dilated, white, unattached; parietal
margin very thin or wanting. Length 21, breadth 12|, length
of aperture 10 mm. Average weight 6.7 grains' (Gulick).

West Maui: mountain ridges of Wailuku, on the ilima (Sida)
and other bushes (E. Bailey).

Achatinella induta GULICK, Ann. Lye., vi, p. 207, pi. 7, f.
34a, 34e, 1856. --BALDWIN, Catalogue, p. 6.

"This species is characterized by its unpolished brown ex-
terior' (Gulick). It has a close resemblance to the East
Mauian marmorata Gld., and is very near ustulata Gul. A.
perdix and undosa differ by their polish and more definite color-
patterns. Figured from specimens of the Gulick collection, no.

PARTULINA, MAUI. 49

92712 A. N. S. P. Another lot from Waihee is similar, but
contains some individuals marked somewhat like marmorata.

1 ' Var. 6. Ash-brown with whitish streaks and spots. This
variety passes into A. undosa.

"Var. c. With a white band on the periphery of the last
whorl; very rare.

"Var. d. Small, ovate-conic. The dimensions of one of the
smallest specimens are as follows: length 18, breadth 10, length
of aperture 8f mm. Weight 4 grains.

"Var. e. Lip considerably expanded and reflected, slightly
thickened within; columella broadly reflected. This variety
presents a remarkable feature; appearing as if the body-whorl
had been enlarged and the lip expanded by a second growth
after the shell has been perfected and the lip thickened. The
second growth forms a broad lip sometimes an eighth of an inch
or more in width from the interior callous ridge to the edge.
This lip seem to be of the same material as the thickening
within, the exterior being of an unpolished ivory color, not
covered with the brown coat which clothes the rest of the shell.
The interior of both the lip and columella is of polished ivory
white; aperture within the lip bluish white. Specimens pre-
senting more or less of the features of this variety are not rare '
( Gulick) .

24. P. RADIATA (Gould). PL 13, figs. 8, 9.

Shell solid, ovate-conic, longitudinally painted with chestnut,
yellow and whitish lines, sometimes girdled with a median or
basal white band; subumbilicate; whorls 6, convex, the suture
margined, whitish; aperture ovate, peristome reflexed, yellow;
fold broad. Length f , diam. 9-twentieths inch. In size and
markings it resembles Bui. radiatus, but the lines are finer and
more numerous (6Wc?.).

Sandwich Islands (Old.).

Achatinella radiata OLD., Proc. Boston Soc. N. H., 1845, p.
27; Otia Conch., p. 195. PFR., Monogr., iii, 454; iv, 517; vi,
162. Bulimus gouldi PFR., Zeitschr. f. Malak., 1846, p. 116;
Monogr., ii, p. 74. Partula densilineata REEVE, Conch. Icon.,
vi, pi. 2, fig. 9, April, 1850.

50 PARTULINA, MAUI.

The locality of this species is not known; radiata and gouldi
were stated to be from the Sandwich Islands, and for densilineata
no habitat was known. It is not even certain that the three
names pertain to one species, though I think that they do.
Gould's white-banded specimens were probably different. The
original descriptions are given below.

I believe that Newcomb correctly identified radiatus but un-
fortunately the specimens from him in our collection bear no
locality. Two of them are illustrated in pi. 13, figs. 8, 9. The
shell is distinctly perforate, ovate, with rather straight-sided
spire. Spiral striation weak, very faint on the last whorl,
whorls weakly convex, the last convex; the suture usually
marked with a white line. Embryonic whorls uniform pale
brown but darker at the tip; later whorls closely streaked with
narrow r streaks or lines of brown, cream and whitish. Lip
slightly expanded, more so below, whitish, calloused within.
vColumellar fold rather sharp but not prominent.

Length 18, diam. 11.3, aperture 9.8 mm.; 6 whorls.

Length 18, diam. 10.5 mm.

One of the lot has the brown replaced by olivaceous, and an-
other has faint traces of several spiral lines on the base.

While the absence of flammules on the last embryonic whorl
is anomalous, yet the species has the appearance of a Maui

shell.

Bulimus gouldi Pfr. Shell narrowly umbilicate, ovate-conic,
rather solid, very delicately decussate-striatulate, glossy, ele-
gantly radiated with ashen, brown and white. Spire conic,
rather acute. Whorls 6, a little convex, the last about three-
sevenths the total length. Columella twisted plicate, white.
Aperture oblong, white within. Peristome narrowly expanded,
the margins subparallel, columellar margin broadly reflexed,
not covering the umbilicus. Length 19, diam. 10, aperture 9
mm. long, 5 wide within. Sandwich Islands, Mus. Cuming

* f

Partula densilineata Reeve (pi. 13, fig. 4). 'Shell acumi-
nately conical, rather thin, umbilicated, sutures of the spire
impressed, whorls rather flattened and finely margined around
the upper part, rounded and slightly angled at the lower, colu-

PARTULINA, MAUI. 51

mella flatly reflected, obsoletely plicate, aperture ovate, lip but
little reflected; whitish, very thickly painted with oblique rust
and ash lines. Hab. ? The lip of this shell, the only ex-
ample of the species known to me, is scarcely mature. The
linear painting is discontinued before reaching the umbilicus '
(Rve.)

Series of P. splendida.
25. P. SPLENDIDA (Newcomb). PL 11, figs. 1 to 10.

"Shell sinistral, solid, ovate-acuminate, finely decussately
striated; suture moderately impressed, margined. Whorls 6;
body whorl somewhat inflated. Aperture ovate; lip expanded;
columella short, broad and twisted. The upper two whorls tes-
sellated w r ith chestnut and white, lower whorls with numerous
chestnut-colored transverse lines and fillets traced on a polished
white ground, markings correctly lined superiorly and irregu-
larly serrated inferiorly. Length 1, diam. .55 inch' (Newc.).

West Maui : Wailuku, on tutui trees (Newcomb); Mauna
Hoomaha (Thaanum); Lahaina, Wailuku and Wailee (Gulick).

Achatinella splendida NEWC., Ann. Lye. N. H. of N. Y. vi, p.
20, April, 1853; P. Z. S., London, 1853, p. 131, pi. 22, f. 4;
Ann. Lye. N. H. of N. Y. vi, p. 324. PFR., Monogr. iv, 516.
-TnwiNG, Reprint Orig. Descript. Achat., pi. 2, f. 17. Acha-
tinella solida Gulick, MS. according to PFEIFFER, Monographia
iv, 516. Achatinella baileyana GULICK, Ann. Lye. N. H. vi, p.
202, pi. 7, f. 31a, b (Dec. 1856).

The embryonic shell consists of 3 to 3^ whorls, the first 1 or
1^ flesh-tinted, wide, indistinct dark stripes then appearing.
The last embryonic whorl has alternately snow-white and brown
stripes, very obliquely descending forwardly. The banded pat-
tern then begins. The lines and bands of brown are usually a
little serrate. The surface is glossy, with dense, weakly devel-
oped decurrent striae. Banding variable, as shown in the fig-
ures, in all the localities mentioned above.

In a lot of 42 from Wailuku, 24 are sinistral. A lot of 32
from Wailee has 7 sinistral, all with broad bands. From La-
haina, in a lot of 60 shells, 34 are sinistral, patterns the same
as in the dextral shells. Specimens from Wailuku are figured,

52 PARTULINA, MAUI.

figs. 1 to 9. Fig. 10 is from Lahaina. Wailuku specimens
measure:

Length 25J, diam. 15 aperture 12^ mm.

Length 25J, diam. 13.8 aperture 11.2 mm.

Length 24, diam. 15 aperture 12 mm.

Length 22, diam. 12.5 aperture 10 mm.

Var. baileyana (Gulick) . PI. 11, figs. 11, 12, 13. The shell
is sinistral. perforate, white with few or many narrow deep
brown spiral lines, sometimes with a broad belt of pale brown
under the lines; otherwise as in splendida. Length 21.6, diam.
12.4, aperture 10 mm. weight 7 grains, or larger, length 25.4,
diam. 13.5, aperture 11.4 mm. Wailuku mountain, on trees,
Ed. Bailey Jr. Gulick describes three varieties as follows:

"Var. b. Regularly conical, rather broad at the base, with
spire shortened. Whorls nearly 7, convex, not margined.
Aperture very oblique, semiorbicular, pale blue within. Length
17, breadth 13, length of aperture 8f mm. Weight 6 grains.

"Var. c. Ovate-conic, marble gray, bands usually wanting;
within the aperture pale lilac; whorls somewhat swollen beneath
the sutures; columella not so short. This variety passes into
the w r hite variety of A. splendida Newc.

' ' Var. d. Lightly tinged with pink both within and without. '

26. P. GOULDII (Newcomb). PI. 11, figs. 14, 15, 16.

Shell dextral, conically ovate, longitudinally striate; suture
moderate, not margined or but slightly so; whorls 6, flatly con-
vex, color yellowish-white with zigzag lines of brown on the
third whorl, and brown lineations more or less numerous en-
circling the three lower whorls. Aperture rounded ovate,
yellowish; columella lightly callous; lip expanded and below
reflected. Length .85, breadth .45 inch (Newc.).

West Maui: Wailuku valley, on tutui trees (Newcomb).

Achatinella gouldii NEWC., Ann. Lye. N. H. of N. Y., vi, p.
21, April, 1853; P. Z. S., 1853, p. 129, pi. 22, f. 1, 1854.
PFR., Monogr., iv, 517. A. myrrhea Gul., according to Pfeiffer,
/. c., undescribed. Achatinella talpina GULICK, Ann. Lye., vi,
p. 212, pi. 7, f. 38, 1856.

This species resembles P. splendida Nc. in color-pattern, but

PARTULINA, MAUL 5->

it is smaller, with less convex whorls, the last one shorter. So
far as we know it is always dextral. Perforation small, some-
times closed. In the typical form the bands, of a light brown
color, are more or less distinctly spotted with darker; and when
the bands spread to cover most of the surface, they are marked
with dark streaks. The last embryonic whorl has oblique
stripes as in related species, and the first neanic whorl is irregu-
larly clouded and streaked. The columellar fold is low and
oblique, white. Lip distinctly but narrowly thickened within,
white, outwardly a little expanded.

Length 21, diam. 12, aperture 10 mm.; 6J whorls.

Length 20, diam. 11 mm.

A. talpina Gul. has always been considered synonymous with
gouldii. It is here figured (pi. 11, fig. 17) from a specimen
from Gulick. Although Gulick mentions only one band, his
figure and specimens show numerous lines also. The colu-
mellar fold is slightly stronger and less oblique than in typical
gouldii. The original account follows:

" Achatinetta talpina, shell dextral, slightly perforate, elongate
ovate-conic, rather thin, subcorneous, not shining, finely
striated, beneath the lens scarcely decussated; brown with a
whitish band encircling the base, covered with a thin, brown
ash, velvety epidermis. Apex somewhat obtuse, tessellated
with white and chestnut; spire regularly conical; suture simple,
moderately impressed, whorls nearly 7, convex, the last regu-
larly rounded, equal to 62 to 76 hundredths of the length.
Columella strongly plaited above. Aperture semi-orbicular,
brownish-white within; peristome thickened within, scarcely
reflected anteriorly, with columellar margin white and reflected
over the deep perforation; parietal margin wanting. Length
21.6, breadth 12, length of aperture 9 mm.; weight 5.5 grains.
"Var. b. With apex corneous, without tessellations; very
rare.

"Var. c. Ventricose; spire short, curvilinear. A small speci-
men has the following dimensions: length 17, breadth lOf,
length of aperture 7f mm., Weight 4.5 grains. r

Maui: Wailuku, on the kukui, Aleuritis triloba, rare, E.
Bailey Jr.

54 PARTULINA, MAUI.

"This species passes into A. gouldiiRewc. Much handling
gradually removes the epidermis from the shell, leaving the
surface more or less polished' (Gulick').

26a. P. gouldi perfecta n. var. PL 11, figs. 18 to 21.

The bands are very dark chestnut on a cream-white ground.
The lip has a stronger callous rib within and is slightly flesh-
tinted, with faintly traced bands. The columellar fold is
twisted, very strongly projecting, subhorizontal; axis perforate
or closed. Color-pattern various, as figured.

Length 20, diam. 12, aperture 9.8 mm.; 6f whorls.

Length 19.5, diam. 11 mm.

Wailuku (D. D. Baldwin). A small form or race from the
same valley (pi. 11, figs. 20, 21) measures, length 17, diam. 10
mm. ; 6J whorls.

27. P. APTYCHA (Pfeiffer). PL 18, fig. 5.

Shell imperf orate, dextral, oblong-conic, rather thin, striatu-
late, a little shining; white, ornamented with dotted chestnut
girdles. Spire long-conic, the apex rather obtuse, suture nearly
simple. Whorls 6J, a little convex, the last nearly equal to
two-fifths the total length, rotund at base. Aperture oblique,
truncate-oval, milk-white within. Columellar fold obsolete,
very slightly twisted. Peristome simple, very lightly thick-
ened within, the right margin unexpanded, columellar margin
narrow, adnate. Length 21-J, diam. 10 mm., aperture 9^ x
5iy mm.

fA

Sandwich Islands (Frick in Cuming coll.).

AchatineUa aptycha PFR., P. Z. S., March, 1855, p. 1, pi. 30,
f. 1; Monographia, iv, 562; Nomenclator Hel. Viv., 1881, p.
315 (Perdicella) .

A "lost' species, unknown to Hawaiian collectors. It
seems as near a young gouldii or baileyana as anything. Sykes
remarks ' ' probably from Maui. '

Series of P. tappaniana.

28. P. TAPPANIANA (C. B. Adams). PL 12, figs. 1, 2, 3.
"Shell reversed, elongate ovate-conic; pure white, with one

PARTULINA, MAUI. 55

narrow brown spiral band on the periphery of the last whorl;
with very unequal and irregular transverse stria3 and numerous
excessively minute, wrinkled spiral impressed lines. Apex sub-
acute; spire elongate, with the outlines a little curvilinear.
Whorls nearly 7, moderately convex and more or less sub-
angular, margined above, with a well-impressed suture. Aper-
ture ovate; lip well thickened within the margin, expanded
much anteriorly, but not above; columellar fold strong. Mean
divergence 43 degrees. Length 1.06, breadth .55 inch; length
of aperture .44 inch. Length 26 J, diam. 13f, aperture 11 mm.

" Var. dubiosa differs in being a little more ventricose; its
whorls are not margined above, and the brown stripe is replaced
by two fine paler brown lines, below which one or two other
yet finer lines may be seen.

" Sandwich Islands.

"This beautiful species is named in honor of Hon. Benjamin
Tappan, of Ohio." (C. B. Ad.")

W. Maui: Wailuku (Gulick); Lahaina (Gulick, Baldwin).

Achatinella tappaniana C. B. Ad., Ann. Lye. N. H. of N. Y.,
v, p. 42, 1851; Contrib. to Conch., no. 8, p. 126.--PFR.,
Monogr., iii, p. 462. SYKES, Fauna Hawaiiensis, p. 318.-
BALDWIN, Catalogue, p. 7, 1893. THWING, Reprint Orig. De-
script. Achat. , pi. 2, f. 19.

This well-known shell has a long, noticeably concave-sided
spire, margined suture, and one narrow brown band just below
the periphery, so that it enters the aperture below the insertion
of the outer lip. There are 3J embryonic whorls, of a whitish
tint, the last half whorl in young or very fresh examples usually
showing protractive yellowish stripes. Young shells are covered
with a yellow cuticle which is deciduous, disappearing in the
adult stage. The surface is rather dull, under the lens showing
spiral impressed lines and weak decurrent striation. The young
and half grown shells are rather acutely angular at the peri-
phery in front of the aperture. Wailuku is here selected as the
type locality (pi. 12, fig. 1). Length 24, diam. 12 mm.; 7
whorls.

Kahoma specimens are somewhat stouter, length 25, diam.
14 mm. ; 6J whorls. They have one band or none, and one

56 PARTULINA, MAUI.

specimen shows faint traces near the lip of numerous other
bands, thus approaching var. fasciata.

In a series from Lahaina the shell is pure white at all stages
of growth, or there may be some faintly sketched yellowish
spiral lines below the periphery, which is acutely angular in
front in the young. The tine descending striae are more distinct
in most specimens (pi. 12, figs. 2, 3).

28a. P. tappaniana fasciata (Gulick). PI. 12, figs. 5, 6, 7.

Shell sinistral, sometimes perforate, ovate-conic, solid, shin-
ing, finely striated, microscopically very minutely and faintly
decussated, white, with brown bands; apex subacute, white,
with a brown line above the suture; spire regularly conical;
whorls 6 or 7, convex, not margined; the last rounded, equal to
_6_^. O f the length; columellar fold white, superior, slightly
developed, moderately twisted; aperture somewhat oblique,
roundly oval, white within; the bands sometimes appear in pale
brownish stripes in the inner surface; peristome white, slightly
thickened, subreflected anteriorly; with columellar margin re-
flected, scarcely adnate; parietal margin wanting. Average
length 22; greatest length 25; broadest 11^; length of aperture
10mm. Average weight 6.8 grains; greatest weight 9 grains;
least weight 6 grains. ( Gulick. )

Maui: Honukawai, on the wiliwili (Erythrina monosperma'),
S. T. Alexander.

Var. 6. Thicker and more elongate, with dark brown or black
bands; lip well thickened within, and deeply colored on the edge
opposite the bands; columella slightly toothed.

Var. c. Ventricose, conical; one specimen has the following
dimensions: Length 0.86; breadth 0.54; length of aperture 0.40
inch.

Remarks. Has been confounded with A. tappaniana and A.
splendida Newc., but is smaller than the former, with fainter
sculpturing and more regularly conical spire; it also differs in
its dark bands, rounded body whorl, less reflected lip, and
slight columellar fold. It resembles A. splendida Newc. in its
brown bands, but is otherwise quite distinct. A nuclear char-
acter which distinguishes it from either of the above is the spiral
line on the first whorls. (Gulick.)

PARTULINA, MAUI. 57

Achatinella fasciata GUL., Ann. Lye. vi, p. 201, pi. 7, f. 30,
1856.

Figured from Honokawai shells of the Gulick collection.
Hardly distinguishable from var. ampulla; I think them ex-
tremes of one variety. The figures are from topotypes from
Gulick. The same form is in the Gulick collection from
Wailuku.

286. P. tappaniana ampulla (Gulick). PI. 12, fig. 4.

"Shell sinistral, sometimes slightly perforate, elongate, sub-
pyriform, rather thin, shining, finely striate, scarcely decussate
beneath the lens, white, with a broad chestnut band encircling
the base, and revolving within the shell beneath the suture,
sometimes with fine spiral lines above; apex subacute; spire
concavely conical, sometimes decollated; whorls 6-J-, convex,
margined above; the last inflated, equal to -f/-^ of the length;
columella white, plaited near the whorl, strongly twisted, not
tuberculate; aperture rounded; peristome expanded, reflected,
very slightly thickened; with columellar margin reflected, adnate
or slightly detached; parietal margin wanting. Length 23;
breadth 13; length of aperture 10 mm. Weight 6 grains"
(Gulick).

Maui: Honukawai, on trees, S. T. Alexander.

Achatinella ampulla GULICK, Ann. Lye. vi, 200, pi. 7, f. 29.

"This species differs from A. fasciata Nob., to which it is
closely allied, in the concave outlines of the spire, in its inflated
body whorl, and more expanded and reflected lip' (Gulick).

28c. P. tappaniana eburnea (Gulick). PI. 12, figs. 9, 10, 11, 12.

"Shell sinistral, usually perforate, ovate-conic, somewhat
solid, shining, striated, very minutely decussated, ivory-white;
apex rather acute; spire conca^ely conical; suture slightly mar-
gined, lightly impressed; whorls 6, slightly convex; the last
large, obliquely produced, and angulated above the aperture;
columellar fold central, white, strong, nearly transverse; aper-
ture oblique, truncately auriform, white within; peristome
thickened; with external margin reflected, anteriorly arcuate;
columellar margin dilated, slightly detached; parietal margin
wanting. Length 24; breadth 14; length of body whorl 17 mm.
Average weight 9 grains; greatest weight 13 grains ' (Gulick).

58 PARTULINA, MAUI.

East Maui: Honuaula, on the trunks of trees, quite exposed,
E. Bailey.

"Var. b. Brownish-yellow, with a white girdle on the peri-
phery of the last whorl, waved with white and yellow on the
third whorl.

1 1 Var. c. With body whorl rounded, not angulated. This spe-
cies is the analogue of A. tappaniana C. B. Ad. found on W.
Maui, but differs from it in its more ventricose form, its angu-
lated body whorl, and more regularly curved lip, besides the
difference in color" (Gulick).

AchatineUa eburnea GULICK. Ann. Lye. N. H. of N. Y. vi, p.
199, pi. 6, f. 28a, b. 1856. BALDWIN, Catalogue, p. 6
(Makawao). THWLNG, Reprint, pi. 2, f. 14.

I cannot distinguish the typical form of eburnea (pi. 12, figs.
9, 10) from the pure white West-Mauian form of tappaniana.
The Var. b (pi. 12, figs. 11, 12) however is unlike any West-
Mauian form I have seen, and approaches very close to carnicolor
Baldwin. I doubt whether a specific line can be drawn between
them.

29. P. CARNICOLOR Baldwin. PI. 12, figs. 13, 14.

"Shell dextral, minutely perforated, solid, acuminately
ovate conic, apex subacute; surface lusterless, marked with
delicate incremental stride, and under a lens exhibiting a fine
pattern of decussating spiral striae; nuclear whorls faintly decus-
sated. Of a uniform brown color, with a narrow white line
below the periphery which enters the aperture, and a very
narrow white line traversing the suture. Whorls 6, slightly
marginate above, a little convex; suture lightly impressed.
Aperture oblique; oval, livid white, showing the exterior color-
ing within. Peristome acute, expanded, columellar margin
broadly reflexed. Columella white, terminating in a well
developed, flexuous fold. Length 25, d.iam. 14 mm.' : (Bfcld-
win) .

East Maui: Nahiku.

Partulina carnicolor BALDWIN, Nautilus xix, February, 1906,
p. 112.

In sculpture this form agrees with eburnea Gulick. It is

PARTULINA, MAUL 59

fleshy brown, darkest at the base, the aperture flesh-colored
within. The band below the periphery is sometimes very
narrow less than 1 mm. wide. I think it doubtfully distinct
from the var. b of eburnea. Figured from cotypes.

30. P. NIVEA (Baldwin). PI. 12, fig. 8.

The shell is sinistral, subperforate, acutely conic, thin but
moderately strong, glossy, snow-white with a brownish-yellow
band below the periphery. Growth-striae distinct but fine and
irregular; spiral striation almost effaced on the last whorl, more
distinct above. The spire is a little attenuated near the acute
apex; whorls but slightly convex, the suture not deep, margined
below. Last whorl a little swollen just below the suture, then
flattened, often having a peripheral angle weakly sketched in
front of the aperture; periphery very convex; base not very con-
vex. The aperture is oblique, white within, with a rather
strong rib within the lip, which is a little expanded. Columellar
lamella white, rather strong and oblique.

Length 24, diam. 14 mm. ; 6-J whorls.

Length 23, diam. 12 mm.; 6^ whorls.

''Animal when extended in motion longer than the shell.
Mantle densely black, with minute brown flecks and a broad
yellowish band encircling the outer edge. Tentacles and supe-
rior portion of foot brown with a slight slate tinge. Under por-
tion of foot light brown. Front above covered with light granu-
lation ' ( Baldwin ) .

East Maui: Makawao to Huelo (Baldwin).

Achatinella nivea BALDWIN, Proc. A. N. S. Phila., 1895, p. 222,
pi. 10, f. 19. THWING, Reprint Orig. Descript. Achat., pi. 2,
f. 16.

"I have before me over 500 examples of this species, which
are without variety in color or shape' (Baldwin'). It differs
from tappaniana by the more acute apex, less attenuated spire,
flatter whorls, the last having its greatest convexity lower down,
and by the more glossy surface. In A. dolei the upper part of
the last whorl is much more convex, and the color differs. De-
scribed from the type lot, no. 65691 A. N. S. P. Several other
lots seen agree exactly with this one.

60 PARTULINA, MAUI.

30a. Var. kaupakaluana n. v., pi. 13, fig. 16, has the last 2J
whorls closely streaked with ochraceous or olive-buff on a
whitish ground, a white belt on the base. It is known to us
by two examples, not quite mature, from Kaupakalua, no. 2101
coll. C. M. Cooke.

31. P. DOLEI (Baldwin). PL 12, figs. 15 to 18.

Shell sinistral, perforate, acutely pyramidal, thin but strong,
glossy, white with a dark chestnut band below the periphery,
and usually several lines and obliquely-streaked bands of light
brown in the peripheral region. Surface polished, delicately
marked with growth-strise and almost effaced spiral lines. Spire
straight! ) T pyramidal, rather acute, the apex white. Post-
embryonic whorls decidedly convex, the last swollen below the
suture, then somewhat compressed; suture deep, narrowly mar-
gined. The aperture is oblique, white within, with a rib within
the lip, which expands somewhat. Columellar fold strong and
oblique.

Length 24J, diam. 14 mm. ; 6f whorls.
Length 25, diam. 15 mm.
Length 25J, diam. 14 mm.

"Animal when extended in motion longer than the shell.
Mantle intensely black with a broad outer margin of yellow and
a conspicuous orange spot in the center of the margin. Foot
tapering behind, above and below light yellow. Tentacles long
and slender, light slate. Head above lightly granulated'
(Baldwin).

East Maui: Honomanu, very abundant in humid forests of
the exterior slope of Haleakala crater, 7000 feet above sea level
(Baldwin).

AchatinelladoleiBA-Lvwix, Proc. A. N. S. Phila., 1895, p. 221,
pi. 10, f. 17, 18.

The dark band below the periphery is a constant feature, and
rarely no other markings are present. When lighter bands are
developed they may ascend the spire above the suture, or they
may be confined to the last whorl. The embryonic shell (pi.
15, fig. 18) is acutely angular and has a subperipheral band.
The columella is convex. It is closely related to P. nivea but

PARTULINA, MAUI. 61

differs by the swollen upper part of the last whorl and the color-
ation. P. splendida Newc. of West Maui has a differently
shaped last whorl. Figures 15-18 are from cotypes. Named
in honor of the President of the Republic of Hawaii, Hon. S. B.
Dole.

32. P. LEMMONI Baldwin. PI. 12, figs. 19, 20.

The shell is sinistral, imperforate, pyramidal, ivory-white
variously marked with dark chestnut bands, which are darker,
chocolate, at the base, broad band at and below the periphery,
another around the columella constant; suture bordered with a
white line, dark bands above and below r it, the embryonic shell
white. Surface glossy, finely striate with growth-lines and fine,
weak spirals. Outlines of spire nearly straight, last two whorls
convex. Aperture oblique, blue-banded within on a white
ground. Lip a trifle expanded, thickened within, spotted by
the bands. Columellar fold rather strong.

Length 23, diam. 12.5, length aperture 10 mm. 6J- whorls.

Length 20, diam. 11.7 mm., whorls 6J.

"Animal in motion as long as the shell. Mantle brownish
black, mottled with white streaks, with a broad yellowish brown
border somewhat interrupted. Foot above and below very light
brown. Tentacles slate color' (Baldwin}.

East Maui: Nahiku (N. E. Lemmon).

Partulina lemmoni BALDWIN, Nautilus xix, p. 112.

This quite distinct species is related to eburnea, dolei and the
smaller flemingi, in which however the bands extend upon the
embryonic shell. Figures and description from cotypes.

33. P. TEREBRA (Newcomb). PL 15, figs. 1 to 9.

"Shell turrited; whorls 6, rounded, last one inflated, mar-
gined above; suture well impressed. Aperture elongate-ovate;
lip slightly reflected, thickened within. Columella short, ter-
minating in a prominent, twisted plait. Stria3 decussating.
Color light yellow, with wave-like brown-colored markings,
nearly obsolete on the upper whorls; lip, columella and aper-
ture white. Length if, diam. -$ inch ' (Newc.).

West Maui (Newcomb); Honokowai (Baldwin).

62 PARTULINA, MAUI.

Achatinella terebra NEWC.. P. Z. S., 1853, p. 144, pi. 23, f.
40. PFR., Monographia iv, 525; vi, 168.--TnwiNG, Reprint,
pi. 2, f. 18. A. corusca Gulick and A. perforata Gulick, unde-
scribed, PFR., Monogr. iv, 525 (names only). Achatinella
attenuata PFR., P. Z. S., 1855, p. 4, pi. 30, f. 12. Achatinella
lignaria GULICK, Ann. Lye., p. 209, pi. 7, f. 35, 1856.

Newcomb's original figure is copied, pi. 15, fig. 2. A speci-
men received from Dr. Newcomb, typical in markings though
somewhat smaller, is drawn in pi. 15, fig. 1. The last embry-
onic whorl has rather wide stripes, the following whorls narrow,
more or less angulated in places, and of a tawny color on a
white ground. The surface is glossy and the spiral and decur-
rent striation weak. Length 18, diam. 9.5 mm., of fully 6
whorls.

Besides this typical form, Newcomb sent out lignaria, corusca
and longior as terebra. It will probably be possible for a natur-
alist in the field to arrive at some trustworthy conclusion as to
the rank or value of the several named forms of this group.
Without such knowledge, I must be satisfied to give the data
relative to the various forms.

In a series from Waiehu (pi. 15, figs. 4 to 8) from Baldwin
the shell is stouter than PfeifTer's figure of attenuata, and the
stripes are less flexuous. Surface glossy. The}'' vary from pure
white, through faintly striped to copiously striped forms, the
latter usually having the ground of a pale brown tint. The
shape also varies. Length 16 to 18 mm. The white examples
are not distinguishable from the following variety.

Var. corusca Gulick. PI. 15, fig. 9. Gulick referred speci-
mens from Wailuku (which intergrade with lignaria) and from
Waihee, to attenuata. His undescribed A. corusca, according to
unlocalized specimens before me, agree well with the Waihee lot,
and were probably from that place. The name was afterwards
abandoned by Gulick, who adopted attenuata in its place. The
shells are white, rather glossy, often with some faint bluish
stripes on the last embryonic whorl. The sutural margination
is conspicuous. Spiral sculpture visible throughout but rather
weak. Length 16.3, diam. 8.5 mm. to length 18, diam. 9 mm.

The Wailuku lot consists partly of shells like those from

PARTULINA, MAUI. 63

Waihee, partly of more swollen shells, with more or less straw
or brownish tint. One of the white shells has dark bands at
periphery and columella. On another there are very faint
tawny stripes.

33 P. t. attenuata (Pfeiffer). PI. 15, fig. 3.

Shell subperforate, ovate-turrited, rather thin, striatulate and
under a lens granulate, white, painted with corneous streaks
which are often angular. Spire concavely turrited, the apex
rather obtuse, suture distinctly thread-margined, whorls 5J-, the
upper flat, those following convex, last whorl about three-
sevenths the total length, rounded, base subcompressed. Aper-
ture oblique, subtetragonal-auriform ; columellar fold strong,
compressed, tongue-shaped; peristome acute, narrowly ex-
panded, labiate within, basal margin forming an angle within
with the dilated columellar margin. Length 16, diam. 7-J mm.,
aperture with peristome 7J- mm. long, 4 wide. Maui, Frick

(flW-

The type figure is copied, pi. 15, fig. 3. The smaller size
and narrower form distinguish it from terebra and lignaria.
Type locality unknown.

335. Var. longior n. v. (pi. 15, fig. 12). Straightly pyramidal,
whitish with a pale yellow tint, the last embryonic whorl
generally having faint stripes, and under a lens very faint yellow
streaks may often be seen on the last whorl. Surface with but
little gloss, and usually deeper spiral sculpture than in the
smaller var. corusca.

Length 22, diam. 10J mm. 6f whorls.

Length 22, diam. 10 mm. 7J- whorls.

Wailuku (Gulick); also Waiehu (Baldwin).

33c. P. t. lignaria (Gulick). PI. 15, figs. 13 to 17.

"Shell dextral, perforate, ovate-conic, solid, more or less
shining, microscopically minutely decussated, yellowish ivory-
white, sometimes streaked; apex subacute, white, obliquely
marked with brownish-yellow; spire regularly conical, suture
moderately impressed, scarcely margined; whorls 6J, somewhat

64 PARTULINA, MAUI.

convex; the last equal to -^g- of the length; columellar fold
strong, white; aperture somewhat oblique, sinuously oval; peri-
stome thickened, sometimes slightly expanded and reflected
anteriorly; columellar margin dilated, white, unattached; pari-
etal margin wanting. Length 20J; breadth 10-f ; length of aper-
ture 9 mm. Average weight 5.5 grains ' (Gulick*).

West Maui: Wailuku, on the alii and other low trees, in
damp, elevated regions. E. Bailey.

" This species differs from A. induta Nob. in its smaller size,
more acuminate form, lighter color, and in the polished exterior
of some of its varieties.

" Var. b. More ventricose; aperture nearly semiorbicular.
This variety passes into A. induta.

"Var. c. More elongate. Length 19 J; breadth 9; length of
aperture 8 mm. ; average weight 3.50 grains. This variety ap-
proaches and passes into A. terebra Newc. The two species are,
however, distinctly characterized; the latter being more strongly
sculptured, having the aperture more elongately oval, and the
suture more distinctly margined.

' ' Var. d. With black or brown spiral bands. This passes
into A. splendida Newc., but is distinguished by its smaller size
and stronger columellar fold. Rather rare.

u Var. e. White. This passes into A. attenuataPL

li Var. /. Pale ash-color. Passes into A. undosa Nob.

" Var. g. Pale yellowish brown. Passes into A. crocea Nob.'
(Gulick).

Specimens of Gulick's series (no. 92717 A. N. S. P.) are
figured. The un described A. perforata Gulick, according to
specimens (no. 57783 A. N. S. P.) is identical with lignaria.

34. P. FUSOIDEA (Newcomb). PI. 15, figs. 10, 11.

11 Shell sinistral, rather solid, elongately produced; whorls 6
to 7, scarcely rounded (excepting the last), margined above;
suture w r ell impressed. Aperture roundly ovate, small; colu-
mella very short, robust and twisted; lip simple above, slightly
expanded below. Color straw yellow. Length 15, diam. 5J
twentieths of an inch ' (Newc. ) .

East Maui (Newcomb); Haleakala (Thwing).

Aehatinellidse

PLATE 1

. !

. .

'

6

-

~ - " /'

~ - - ,

. - '-

7

'

3

-

}

5

.

VT

8

9

/

.

/ -

10

-V *
-. ^ ', ..' Ki-i

i-i. ^- ., "*?*

^5

'

- f. ,- -i

''

Achatinellidse

PLATE 2

..

:*--.
-*

P. /

'

. :-W

-

8

'-.

" ' . , '

%

- *

-T- V

v' f >ftir. f !t'i',*gi7iGii!ti. '' <' i"" r *?^v :i .-. /

H ,.

. '..'AKr^fi, ' >

11

3

14

Achatinellidse

PLATE 3

!

;

. i -.:
\ I \ &

J ? . :

-^^

"-

t -

9

10

Achatinellidse

PLATE

/J- ~~'

M\

-.

Achatinellidae

PLATE S

1

9

11

Achatinellidae

PLATE 6

10

Achatinellidae

PLATE 7

Achatinellidae

PLATE 8

13

Achatinellidae

PLATE

9

13

6

10

,

8

17

18

19

20

Achatinellidae

PLATE 1O

9

11

17

19

Achatinellidae

PLATE 11

-.. .

2

5

6

8

9

11

12

13

.

17

19

Achatinellidae

PLATE 12

1 -V '

'

9

6

10

11

o

17

18

8

12

16

PARTULINA, MAUI. 65

Achatinella fusoidea NEWC., Ann. Lye. N. H. of N. Y., vi,
1856, p. 144; Amer. Journ. of Conch., ii, 1866, p. 213, pi. 13,
f. 8. PFR., Monogr. iv, 539. THWING, Orig. Descript, p.
112.

" This is an exceedingly rare species, but two specimens
having come under my observation, one of which serves for
the illustration, the other in the cabinet of the late Hugh
Cuming, in London ' (Newc.).

This species seems close to A. terebra var. longior of "West
Matii, a dextral shell.

Shells from Ukumehame, West Maul (pi. 15, fig. 11), sent
by Mr. D. Thaanuni, agree well with Newcomb's figure of the
type of fusoidea. They are buff with faintly darker streaks,
having wide light-brown stripes on the last embryonic whorl.
Under the lens the surface is weakly plicate, as in P. terebra
longior. Length 17, diam. 8 mm.

35. P. CROCEA (Gulick). PI. 13, figs. 14, 15.

" Shell dextral, perforate, conic-ovate, solid, scarcely shin-
ing, lightly striate and microscopically distinctly decussated
with crowded undulating spiral lines ; orange yellow ; apex
subacute, obliquely marked with white and yellow; spire
conical; suture simple, slightly impressed; whorls 6, some-
what convex, the last slightly flattened in the middle, equal
to 71 hundredths of the length; columellar fold moderately
developed, white; aperture slightly oblique, sinuously oval,
snowy white within ; peristome not expanded, moderately
thickened within ; with columellar margin dilated, white, un-
attached ; parietal margin wanting. Length 18 ; breadth 10 ;
length of aperture 9^5 mm. Weight 4 grains.' (Gulick.)

Maui : Waihee, rare. E. Bailey, Jr. Honokowai (D. Thaa-
num).

Achatinella crocea GUL., Ann. Lye. N. H. of N. Y., p. 210,
pi. 7, f. 36, 1856.

The sculpture of impressed spirals and dense, fine, pro-
tractive wrinkles or ripples is very clearly developed. The
color is yellowish-green (tawny-olive of Ridgway's Nomen-
clature of Colors) , but in the young there is often more of a

66 PARTULINA, OAHU.

gamboge tint. There is often a pale line on the periphery.
Embryo pale brown, the last half whorl striped with brown.
The lip is thickened within, white or flesh-tinted. Columellar
fold often very strong and more or less tinted with brown,
but white at the end. Figured from cotypes. Probably a
variety of A. terebra lignaria, yet I have seen no actually in-
tergrading specimens. It seems well characterized by the
color and short contour.

In a series from Honokowai, West Maui, sent bv Mr. Thaa-

/

num, the color of the last whorl varies from pale buff to dull
gamboge, being darkest on the base. The spire is generally
more drawn out than in the figures on plate 13, and is a little
attenuated, its outlines distinctly concave or sometimes almost
straight. The flames on the last embryonic whorl are occa-
sionally lost by wear.

SPECIES OP OAHU.
36. P. MONTAGUI Pilsbry. PI. 24, figs. 7, 8.

The shell is sinistral, perforate, thick and solid, ovate-conic,
the spire acuminate, its outlines straight or somewhat con-
cave above. Embryonic whorls distinctly striate spirally;
last three whorls rudely sculptured with close, irregular
wrinkles, and on the last whorl there are spirally and ob-
liquely descending ridges producing flattened facets (as in
many large Lymnaeas). The minute spirals disappear on
the later whorls. The upper whorls are nearly flat, the last
three slightly convex, last whorl more convex, the base rather
rotund. Suture superficial, sometimes having a margin de-
fined by an impressed line on the last whorl, but in other ex-
amples this is wanting. Aperture slightly oblique. Outer
and basal margins of the peristome are expanded, thick, and
have a heavy callous thickening within ; columellar margin
raised, thick; columellar fold thick and moderately project-

ing.

(a) Length 25, diam. 14, length aperture 12 mm.; l
whorls.

(b) Length 24.7, diam. 12.5, length aperture 11 mm.;
whorls.

PARTULINA, OAHU; EBURNELLA. 67

(c) Length. 26.9, diam. 13.1, length aperture 12.1 mm.

Oahu: Manoa road at its junction with the upper road,
back of Rocky Hill, cotypes (a) no. 108181 A. N. S. P.; (6)
no. Ill coll. Irwin Spalding; and (c) no. 33581 B. P. Bishop
Mus. Also found on Rocky Hill about ^4 m ^ e from the type
locality.

Partulina montagui PILSBRY, Nautilus, xxvii, p. 40, August
1st, 1913.

This species differs strikingly from all others by its sculp-
ture and small apical whorls. It seems to be nearest to P.
divightii of Molokai. No specimens yet found show any trace
of color. The periphery is angular in the young, the angle
persisting as far as the last whorl in some individuals.

It was first found by Dr. C. Montague Cooke. The shells
occur in the humus near the surface, along the roadside, and
are very rare, apparently lying in " pockets ' which have
been filled and covered by the wash down the slope. Probably
the forest disappeared from where the shells are found not
much more than a hundred years ago.

Section Eburnella Pease.

Eburnella PEASE, P. Z. S., 1869, p. 647. GULICK, P. Z. S.,
1873, p. 91, as section of Partulina ; A. variabilis selected as
an example.

Very smooth, glossy shells, uniform white, yellow or brown,
or variously banded, not striped longitudinally; outer lip
simple or scarcely expanded; columellar fold strong. Type

P. variabilis (Newc.).

The typical group consists of Lanaian species in which the
peripheral keel persists throughout the young stages, and
often upon the first half of the last whorl. In the allied
groups of Molokai and Maui the keel is lost much earlier ; the
later stages of youth and the mature stage being round at
the periphery. The group is not strongly individualized,
some species, as P. mutabilis, showing transition to the pattern
of Perdicella, others approaching Partulinse of the striped
group of Maui.

Eburnella was originally a heterogeneous group comprising

68 PARTULINA, MAUL

the following species : A. casta Nc., curta Nc., livida Sw., por-
cellana Nc., recto Nc., saccata Pfr., undulata Nc., semicari-
nata Nc., and variabilis Nc. Gulick, in his classification of
1873 restricted it to Lanaian species, giving Partulina varia-
bilis Newc. as an example. I think it may with advantage be
enlarged to include species of Molokai and Maui which seem
related.

Series of P. porcellana. MAUI.

Glossy shells with weak spiral s-triation ; uniform or with
color in bands, usually few in number and well contrasted;
the embryonic shell is banded above or below the suture or
is plain, and has no longitudinal or oblique markings (except
in P. mutabilis) .

These medium-sized or small shells have much the appear-
ance of Achatinellastrum, but various indications the char-
acteristic decurrent sculpture more or less visible on all the
forms, and the vanishing zebrine stage in P. mutabilis lead
me to consider them modified Partulinae. This is also the
view taken by Mr. Baldwin, and suggested in Hyatt's Ms.
Other authors have grouped part of them in the Oahuan sec-
tion Achatinellastrum.

37. P. MUTABILIS Baldwin. PL 20, figs. 1 to 7.

Shell dextral or sinistral, minutely perforate, narrowly
conical, the sides slightly concave, apex rather obtuse; solid,
rather glossy, striatulate and under the lens showing weak,
rippled spiral striation. " Color varying from pure white
to dark fulvous, often variously striped with brown lines and
bands ' ' ; usually pale buff-yellow or whitish with numerous
chestnut spiral lines and bands, or snow-white throughout;,
the last embryonic whorl frequently striped longitudinally.
Suture generally margined below. The aperture is small,
white, lip very narrowly thickened within ; columellar lamella
white, rather strong, thick and oblique.

Length 17, diam. 8.8, aperture 7.5 mm. ; 6^4 whorls.

Length 16.3, diam. 9.3, aperture 7.8 mm. ; 6 whorls.

West Maui : Waiehu valley.

PARTULINA, MAUI. 69

Partulina mutabilis BALDWIN, Nautilus, xxii, p. 68, No-
vember, 1908.

i

Distinct by its narrow contour, color in bands when pres-
ent, and the frequent presence of a brief zebra-striped stage on
the last embryonic whorl. It has a superficial resemblance
to the Lanaian A. variabilis, which belongs however to a
somewhat different series. Figured from specimens of the
type lot received from Mr. Baldwin.

P. mutabilis is an important species in that it forms a con-
nection between the Partulinas having zebra striping and
those having spiral bands only. The embryonic whorls are
flattened, the first flesh-tinted. In most banded individuals
some indistinct protractive brown streaks appear near the
end of the second whorl, becoming stronger and angularly
bent on the next fourth to half whorl, gradually giving place
to spiral bands which arise about the middle of the third
whorl (pi. 20, figs. 6, 7). This pattern is not essentially un-
like that of Partulina splendida. In some individuals, how-
ever, the longitudinally striped stage is omitted, and the spiral
bands appear (at first very faintly) on the latter part of the
second whorl, 'arising from the uniform fleshy tint of the pre-
ceding part of the embryonic shell (pi. 20, fig. 5). Such
shells are comparable to P. porcellana and its allies, and may
be regarded as more accelerated than those having an ob-
liquely striped stage.

In a series of 18 sent by Mr. Baldwin, 10 are dextral. The
same color patterns are found in dextral and sinistral indi-
viduals. The variations are as follows:

(1) Pure white throughout.

(2) Pale buff, last embryonic whorl faintly striped longi-
tudinally.

(3) White to buff-yelloAV with chestnut bands and lines.

38. P. PORCELLANA (Newcomb) . PI. 19, figs. 1 to 4.

" Shell rather solid, conical, polished, shining. Whorls
5, rounded, margined above. Aperture ovate; lip expanded,
thin at the margin, thickened within. Columella short, tuber-
cular. Color of a porcelain whiteness with a light fawn-

70 PARTULINA, MAUI.

colored 'band encircling each whorl, the last whorl having
two. Length 12, diam. six-twentieths of an inch ' (Newc.).

East Maui (Neivcomb) : Huelo, Makawao to Nahiku
(Thiving) ; Nahiku (Baldwin).

Achatinella porcellana NEWC., P. Z. S., 1853, p. 146, pi. 23,
f. 47. PFR., Monogr. iv, 525. THWING, Orig. Descriptions,
p. 105, pi. 2, f. 12.

In a specimen from Newcomb the first whorl is white, a
yellow band beginning on the second below the suture; but
in others seen the apex is brown or blackish, a band of the
same color \y 2 whorls long running from it above the suture,
fading into white at its upper edge, the first \ l /2 whorls being
therefore bicolored. Then a yellow band appears faintly
below the middle of the whorl, finally becoming the supra-
peripheral band on the last whorl. In some specimens a
band bordering the suture below also is found. The last
whorl has two to four bands: (1) the subsutural, often want-
ing; (2) one above the periphery; (3) one below it, entering
the aperture at its upper angle, and sometimes there is (4) a
narrow crescent around the columella. Occasionally a few
spiral lines are developd in addition to these bands. These
bands and lines vary from ochraceous to dark brown (mummy
brown of Ridgway). The ground color is white, with a yellow
tint around the columella or over the whole base. Surface
glossy, with only very weak spiral striation. The lip is nar-
rowly thickened within in adult shells, its face tinted or
spotted at the bands. Columella white or brown. Specimens
from Nahiku (pi. 19, figs. 1-3) measure:

Length 18, diam. 10 mm. ; 6 whorls.

Length 18, diam. 10.5 mm. ; 5 l / 2 whorls.

Length 15.5, diam. 9 mm.

All the shells I have seen are dextral. A. flemingi and
wailuaensis are apparently varieties of porcellana, as Mr.
Thwing has intimated. Very likely A. fulvicans, possibly
A. nattii also, will eventually be included in porcellana.

Fig. 4 is a copy of Newcomb 's figure of porcellana.

PARTULINA, MAUL 71

38a. P. porcellana flemingi Baldwin. PI. 19, figs. 5 to 11.

Shell dextral, minutely perforate, rather solid, high-conic,
glossy, with very weak traces of spiral striation 'or almost
none ; variously colored ; white, Landless or with one to four
dark brown bands, or light fleshy-brown with bands. Lip a
little thickened within; columellar fold moderately strong.

Length 20, diam. 10 mm. ; 6^ whorls.

Length 18, diam. 9.2 mm.

Length 18.2, diam. 9 mm.

Length 18.3, diam. 10 mm.

Length 17.2, diam. 9.5 mm.

" Animal when extended in motion as long as the shell.
Mantle intense black with a narrow white band encircling the
outer edge. Foot above and below very light-brown, granu-
lated above. Tentacles long and slender, slate color.' (Bald-
win.)

East Maui: Nahiku (D. T. Fleming).

Partulina flemingi BALDWIN, Nautilus, xix, p. Ill, Febru-
ary, 1906.

A large set from the type lot shows great variation in color-
ation.

(1) White, uniform (fig. 8), or yellow tinted at base and
above the suture of the embryo.

(2) White with four dark brown bands, situated below the
suture, above and below the periphery and around the colu-
mella. Embryonic shell white with a yellow band below the
suture or a dark band above the suture, fading upwards;
columella and lip more or less purplish brown. This is the
most abundant color pattern.

(3) White, with bands above and below periphery, apex
lip and columella white.

(4) White, with narrow dark bands at the periphery, in
the middle of the base, and a distance below the suture (fig.
5) ; or having a wide band above the narrow basal band (fig.

ID.

(5) Ground color brownish, bands as in no. 3 (fig. 9).
This subspecies differs from porcellana chiefly by its nar-

72 PARTULINA, MAUI.

rower contour. The system of banding (i. e., positions of the
bands) is the same as in porcellana.

386. P. porcellana wailuaensis (Sykes). PI. 19, figs. 12,
13, 14.

Shell dextral, subperforate, glossy-turrited, rather solid,
very lightly striate, white painted with chestnut lines, marked
at the periphery with a white zone and in the suture a black-
ish-chestnut line. Whorls 5 to 5J^, regularly increasing, con-
vex. Aperture auriform; columellar margin having a mod-
erate brown fold, the right margin acute, parietal callus very
thin. Length 15.5, diam. 8.4 mm. (Sykes).

East Maui : Wailua (Perkins). Hana (Cooper).

Achatinella (Achatinellastrum) ivailuaensis SYKES, Fauna
Hawaiiensis, ii, Mollusca, p. 328, pi. 11, f. 19 (and var., f.
20) , 1900. Partulina cooperi BALDWIN, Nautilus, xix, p. 135,
April, 1906.

" A very pretty little shell of the group of A. bella Reeve,
of Molokai. A variety also occurred (pi. 19, fig. 14) in which
the banding is almost obsolete save in the suture of the earliest
whorls and in one strong dark band below the periphery '
(Sykes). Fig. 13 is copied from Mr. Sykes 's type figure.

The arrangement of bands is essentially the same as in
porcellana, but the shell is somewhat wider in contour, and
the bands are typically paler and more diffuse. I have not
Been specimens from Wailua.

Mr. Baldwin's P. cooperi from Hana (pi. 19, fig. 12, co-
type) seems to belong to the same race. The shell is pale
buff with a white peripheral belt bordered on both sides by
ochraceous bands; base and upper surface have some diffuse
paler ochrey lines or bands, and there is a darker line along
the suture. Embryonic shell white or whitish with a purple
spiral band above the suture and a yellow line below the
suture. The columella and its fold are purplish-brown, and
the narrow callus within the outer lip either white or purplish
brown. A half-dozen specimens of the type lot, sent by Mr.
Baldwin, show but slight variation.

Length 17, diam. 9.5 mm. (Baldwin).

Length 17.3, diam. 10 mm.

PARTULINA, MAUI. 73

39. P. PORCELLANA FULVICANS Baldwin. PL 19, figs. 15, 16.

" Shell dextral or sinistral, very minutely perforated,
rather thin, acuminately ovate conic, apex subacute; surface
shining, marked with delicate incremental striae, under a lens
exhibiting extremely close, minute decussating spiral striae;
nuclear whorls faintly cross-lined. Color very light yellow,
or sometimes white with one or two light yellowish transverse
lines; a conspicuous dark band near the apex. Whorls 6,
somewhat convex, narrowly margined above; suture well im>-
pressed. Aperture oblique, oval, white within. Peristome
acute, slightly thickened within, columellar margin narrowly
reflexed. Columella terminating in a strong white nexuous
fold. Length 18^, diam. 10^ mm.' (Baldwin).

East Maui: Kipahula Valley, Hana (G. 0. Cooper; D.
Thaanum).

Partulina fulvicans BALDWIN, Nautilus, xix, p. 135, April,
1906.

Topotypes from Mr. Baldwin are figured. Fig. 16 is green-
ish-yellow with the embryo nearly white, without a dark band.
Fig. 15 is white above the periphery, light olive below, the
embryo bicolored, being white, broadly banded with flesh
color above the suture, the band extending to the apex. Other
shells are greenish-yellow, fading upwards to white at the
suture and with a faint light peripheral band, the embryo
having a rather dark olive or brown band. The shell is
glossy, with the Partulina spiral sculpture but faintly devel-
oped. It differs from Partulina crocea by the gloss, faint
sculpture, and differently colored embryonic shell. It is very
closely related to A. porcellana and ivailuaensis from which
it differs only in coloration, the dark bands being absent, and
the aperture white throughout. I do not know whether it
occurs in a pure colony, but presume this is the case. Mr.
Baldwin remarks: " The sinistral variety may prove to be a
distinct species. The molluscan life of this rather secluded
valley has only recently been explored. '

40. P. NATTII (Baldwin et Hartman). PL 20, figs. 8, 9, 10, 11.
" Shell dextral, turbinate, spire half the length; whorls 5,

74 PARTULINA, MAUL

polished, the two last rapidly enlarged and inflated. Suture
impressed, columella yellow, stout and twisted. Color bright
gamboge yellow, with one white and three wide chestnut bands
beneath the suture, the latter visible from within the aper-
ture ; aperture round ovate, white, labium white, slightly
thickened within. Length 16, diam. 10, length aperture 8,
width 5 mm.' (Hartm.)

East Maui: Makawao (Baldwin) ; Makawao to Honomanu
(Baldwin).

A.[chatmella] nattii Baldw. Nobis (Achatinellastrum) ,
HARTMAN, Proc. A. N. S. Phila., 1888, p. 34, pi. 1, f. 3
(Achatinellastrum nealii in explanation of plate, p. 56) .
THWING, Orig. Descriptions, p. 106.

" This with the preceding, A. porcellana Newc., and A.
flemingi Bald. MS., and also A. wailuaensis Sykes, form a
group very similar in appearance and which run together '
(Thwing).

P. nattii, of which we figure specimens from Baldwin,
differs from porcellana by its stouter contour and the posi-
tions of the bands, which are quite definitely located in 27
specimens examined. There is one brown band a short dis-
tance below the suture, another at the periphery, and a third
about midway between periphery and columella, this one usu-
ally being the widest. The peripheral band shows above th
suture on the spire, but there is no dark band or margin
along the suture. The bands vary in color from chestnut to
ochraceous. The ground may be white throughout, but it is
usually snow-white between the suture and upper band, and
elsewhere gamboge yellow. The embryonic shell is rarely
white, usually some shade of brownish-yellow with a white
line or band below the suture. About the end of the second
whorl this solid color splits into two bands which become the
upper and peripheral.

The embryo and early whorls are quite distinctly engraved
spirally, but on the last whorl this sculpture is very weak.
The surface is glossy. The lip is white except where colored
by the bands, and the strong, oblique columellar fold is in-
variably white. Sutural margin usually set off by an im-
pressed line.

PARTULINA, MAUI. 75

Length 16.8, diam. 11, aperture 9 mm. ; 6^/3 whorls.

Length 18, diam. 11, aperture 9 mm. ; 5^ whorls.

While nattii is very closely related to porcelldna, I have
seen no intergrading specimens and therefore allow it to
stand as a species.

A variety in the C. M. Cooke collection from Honomanu
(Baldwin) has light ochraceous bands on a gamboge ground
and no snowy sutural band.

41. P. ANCEYANA (Baldwin). PI. 20, figs. 12, 13, 14.

Shell dextral, imperforate, rather thin, oblong-conic, the
summit obtuse. Embryonic shell chestnut, fading to, or banded
with white above ; following whorls chestnut with a white line
revolving a short distance below the suture, and on the penult,
whorl another above the suture; last whorl rich yellowish-
chestnut with a white line near the suture, another at the
periphery, the latter bordered below with a yellow band.
Another yellow band, usually wider, occupies the base, leaving
a small dark patch around the axis. The surface is glossy,
with sculpture of fine crenulated spirals throughout. Aper-
ture slightly oblique, brown-banded within ; lip not expanded,
narrowly thickened within. Columellar fold brown or brown
and white.

Length 14.5, diam. 8.5 mm. ; 5 whorls.

Length 14, diam. 9 mm.

Length 15.5, diam. 8.3 mm.

" Animal when extended in motion longer than the shell.
Mantle light-brown mottled with black, outer edge encircled
with an interrupted orange band. Foot above and below a
very light brown, superior portion with a slate tinge, posterior
portion tapering and sometimes with a yellowish tinge. Ten-
tacles long, light-slate. Head above minutely granulated '
(Baldwin).

East Maui : Makawao, in forests of the exterior slope of the
extinct Haleakala crater at an altitude of 4,000 ft., living on
foliage near the tops of the trees (Baldivin). Type no. 65707
A. N. S. P.

Achatinella anceyana Baldwin, Proc. A. N. S. Phila., 1895,

76 PARTULINA, HAUL

p. 223, pi. 10, f. 16. THWING, Reprint Orig. Descript, Achat,
pi. 2, f. 15.

This species is related to P. porcellana wailuaensis and per-
haps to A. germana, of which it is thought by Dr. Cooke to be
a small race. " It is remarkable for the constancy of its deli-
cate color lines; over 200 examples before me show only
trifling divergence in this respect : (Baldw.). In one speci-
men of the lot sent by Mr. Baldwin, the lower yellow band is
reduced to a line. In some others the ground-color is pale
fleshy-brown above the periphery. There is some variation
in shape, as shown in the figures. It was named for the late
C. F. Ancey. Lives on bark of the guava,

42. P. GERMANA (Newcomb) . PI. 20, fig. 15.

" Shell dextral, ovately conical; whorls 6, flatly convex.
Aperture ovate; lip acute, thickened within. Columella
brown, short and tubercular. Color chestnut, with fine, ob-
solete, darker, transverse lines; a white band traverses the
last sutural whorl and cuts the body- whorl centrally ; around
the columella is a broad white band, losing itself in the aper-
ture. Length sixteen, diam. eight-twentieths inch ' (Newc.).

East Maui: Makawao (Newc.).

Achatinella germana NEWC., P. Z. S., 1853, p. 151, pi. 24,
f. 61, 1854. PFR., Monogr., iv, 533. THWING, Orig. Descript.
p. 109. A. (Achatinellastrum) germana Nc., SYKES, Fauna
Hawaiiensis, p. 324.

" Closely resembles the A. solitaria, and but for the widely-
distant localities might -be taken for varieties of the same
species. The markings and columellaB are however quite dis-
tinct ' (Newc.).

Known to us by the original account only. I can only
suggest its probable relationship to P. anceyana and P. wail-
uaensis, both of which differ in being of a more oblong shape
and smaller size, P. germana measuring length 20, diam. 10
mm. according to Newcomb.

PARTULINA, MOLOKAI. 77

Series of P. mighelsiana. MOLOKAI.

Rather small shells with a high polish, weak or almost
wanting spiral stria tion, and bright coloring yellow often
with white or chestnut bands, or white or chestnut through-
out; never striped longitudinally; lip unexpanded. Embryo
spirally banded or plain.

These shells are related to the variabilis group of Lanai
and the porcellana group of Maui. Whether they belong to
the Oahuan group Achatinellastrum, or represent a branch of
Partulina convergent towards that Oahuan group, remains
uncertain. In Hyatt's earlier writings he placed part of these
species in Achatinellastrum, but later he doubted this group-
ing, and was disposed to view them as modified Partulinee.
As there are Maui-Molokai types in Oahu, such as Laminella
and Partulina dubia, there seems no good reason why some
Achatinellas of Oahuan type should not occur in the more
eastern islands.

For the present I leave the mighelsiana group in Partulina
for the reason that among the related Mauian species it would
be very difficult to draw the line between Partulina and Acha-
tinellastrum. The coloration of the embryonic shell is more
like that of P. virgulata than like the ordinary Partulinas.
Dr. Oooke suggested that in view of the numerous forms of
intermediate character between the three conventional species
of the mighelsiana group, it would be more logical to rank
bella and polita as subspecies of mighelsiana.

43. P. MIGHELSIANA (Pfeiffer) . PI. 18, figs. 10 to 17.

" Shell ovate-coniform, smooth, opaque, glossy; snow-white
variegated with ashy streaks. Spire conic, the apex rather
acute, suture somewhat margined. Whorls 5^, convex, the
last encircled with a blackish line (sometimes doubled) below
the middle. Columellar fold strong, tooth-like, chestnut col-
ored at the base. Aperture semi-oval, brown-margined ; peris-
tome simple, acute. Length 17, diam. 8 mm., aperture 8 T /2
mm. long, 4 wide (Pfr.).

Molokai (Cuming coll.) : Kalae (Baldwin, Thaanum) ; Kilo-

78 PARTULINA, MOLOKAI.

hana (Cooke and Pilsbry) ; Kalae, Puunea, Iloli, Waileia,
Maunahui and Kaanaola (Borcherding, coll. Meyer).

Achatinella mighelsiana PFR., P. Z. S., 1847, p. 231; Mon-
ogr. ii, 238, 1848. REEVED Conch. Icon., vi, pi. 5, f. 40, April,
1850. NEWCOMB, Ann. Lye. N. H., vi, 319 (animal).
SYKES, Fauna Hawaiiensis, p. 314. BALDWIN, Catalogue, p.
6. GWATKIN, Proc. A. N. S. Phila., 1895, p. 238 (radula).
THWING, Orig. Descript., p. 97. Achatinellastrum miglielsi-
ana Pfr., BORCHERDING, Zoologica, xix, p. 79, pi. 7, f. 1-13,
15, 17, 19 ; with var. martensi, p. 80, pi. 7, f . 14, 16, 18, 20.

The shell is dextral, perforate, very smooth and glossy, por-
celain-like, showing weak spiral striae under a lens. The
typical form is white with narrow gray streaks and a black-
brown belt just below the periphery, and a narrow blackish
submargin in the lip. The embryonic whorls are white (fig.
13). This pattern varies to forms having the ground white
without streaks, or buff, with or without yellow streaks. The
subperipheral belt may be split, or rarely another band de-
fines an umbilical area. Probably Kalae is the type locality.

(6) White, without bands or streaks (fig. 15) Kalae.

(c) White, with or without gray streaks, and variously
banded and lineate (figs. 10-12). This is var. martensi of
Borcherding.

(d) Pale yellow, with several bands, embryonic shell brown
with a white band above.

(e) White, with yellowish-brown streaks interrupted by
spiral white bands (fig. 16).

(f) Similar to the type, but the base is yellow (fig. 17)
Kalae.

Well-grown specimens measure length 19, diam. 10 mm.

Borcherding describes a var. martensi as differing from the
type by: white color, well-impressed suture, the whoris more
convex, encircled by lines either wide or narrow, the last
whorl having the typical broad brown peripheral band, colu-
mella light reddish, the lip never bordered within with brown,
interior fleshy-white. PI. 18, figs. 10-12 represent this color-
form. It is from Kawela. The rounded whorls and deep
suture are equally well marked in some specimens of other

PARTULINA, MOLOKAI. 79

color-patterns. The same color-pattern occurs in Kilohana.
There is a similarly marked form of polita.

43a. P. MIGHELSIANA BELLA (Reeve). PL 23, figs. 2 to 12; pi.
26, f. 13 to I5e.

Shell conical, dextral, whorls rather ventricose, columella
with a twisted plait. Olive-yellow, sutures black-brown, then
white-banded, last whorl brown and white banded. The dark
sutural band is sometimes absent in the last whorl, but the
white band appears constant ' (Reeve).

i Animal of a uniform yellowish-white, with slate colored
tentacles ' (Neiucoml)).

Molokai (Cuming coll.), Puunea, on leaves of small trees
(Cooke and Pilsbry), Kalae, Kealia, Kaupelua, Waileia,
Kaunakakai, Maunahui, Hanakalilolilo, Kawela, Ualapue and
Kaluaaha (Borcherding, Meyer coll.).

Achatinella bella REEVE, Conch. Icon., vi, pi. 3, f. 17, April,
1850. PFEIFER, Monogr. iii, 461. SYKES, Fauna Hawaiien-
sis, p. 321. NEWCOMB, Ann. Lye. N. H., vi, p. 316 (animal).
BALDWIN, Catalogue, p. 6. THWING, Reprint Orig. De-
script. Achat., pi. 2, f. 5. Achatinellastrum bella Rve., BOR-
CHERDING, Zoologica, xix, p. 77, pi. 6, f. 11-20.

The shell is dextral in the large number seen. The early
whorls are closely striate spirally, but the last whorl has only
faint traces of spirals on the glossy surface. Columellar fold
very strong. Axis imperforate. The spire is shorter and
stouter than in mighelsiana, but it must be admitted that
some forms, such as color-var. e, approach that species. P.
polita differs chiefly <by its bulging last whorl, narrower spire
and diverse coloration wanting white bands.

In my opinion, supported by Dr. Cooke, the forms figured
for polita by Borcherding, on his plate 8, figures 1 to 7, per-
tain to bella.

The color-forms are as follows: (a) Typical form; em-
bryonic whorls white or black-tipped; last whorl yellow with
a chestnut band bordering the suture followed by a white
band, another just below the periphery with a white band
above it; columella fleshy-brown, the tooth and aperture

80 PARTULINA, MOLOKAI.

fleshy-pink Length 16, diam. 9 to 10 mm. See pi. 23, fig. 12
(copy from Reeve). Embryos of this form show very pala
wax-yellow 'bands proceeding from a yellow apex, which ap-
parently fade out in the adult stage. Waikolo and Puunea
specimens correspond exactly with Reeve's figure. The orig-
inal locality was probably Kalae, or perhaps Puunea, within
easy walking distance from Kalae. A series collected by
Cooke and Pilsbry in Puunea is figured, pi. 26, figs. 13 to 13c.
A little further north, between Puunea and Kilohana, there
are transition forms to mighelsiana, pi. 26, fig. 14, both color
and shape being intermediate in various examples. This
place is not far from the northern cliff.

(6) Similar to the above, but lacking the peripheral chest-
nut band. Occurs with the typical pattern.

(c) Both sutural and peripheral bands wanting. Colu-
mella and aperture white or pink.

(d) Variously banded with chestnut, the bands often oc-
cupying most of the surface. Kalae (Meyer).

(e) White, variously banded with chestnut or with a sutural
line only, or no bands whatever, the umbilical region alone
yellow tinted, or sometimes also white.

(/) Bicolored, white above, greenish-yellow or olive below
the periphery. Aperture and columella white or pink. May
be a derivative of forms g or c.

(g) Same as /, but having a dark sutural line, sometimes a
subperipheral line also, and occasionally an olive line in the
middle of the upper surface. Fig. 9. This form is evidently
a derivative of the typical bella pattern.

"Whether the more eastern of Borcherding's localities really
have bella forms I do not know, but the typical bella coun-
try is in the region southward of the Leper Settlement pen-
insula. A series from Ualapue, collected by Mr. Thaanum
(pi. 26, figs. 15 to 15e) seems to carry ~bella into the polita
area. I do not know the locality of pi. 23, figs. 2-8, collected
by Mr. Meyer.

43&. P. MIGHELSIANA POLITA (Newcomb). PI. 24, figs. 1 to 6.
Shell dextral, conic-ovate, polished, light-yellow, a portion

PARTULINA, MOLOKAI. 81

of the upper whorls marked with umber, suture with a broad
black band, and a narrow line of the same color -continued
around the body-whorl. Whorls 5, convex; suture well de-
fined, margined. Aperture oblong-ovate, dark purple ; colu-
mella with a large tubercle tipped with purple ; lip simple.
Length .55, breadth .45 inch (Neiuc.).

Animal of a uniform yellowish- white, including tentacles;
tentacular sheath slightly tinged with brown; mantle black
(Neiucomb) .

Molokai (Newcomb), Kaluaaha to Halawa (Baldwin;
Bishop Museum), Kawela, Kaamola and Ualapue (Borcher-
ding).

Achatinella polita NEWC., Annals of the Lyceum of Nat.
Hist, of N. Y., vi, p. 24, May, 1853; t. c. p. 328 (description
of animal) ; P. Z. S., 1853, p. 142, pi. 23, f. 37. PFR., Mon-
ogr., iv, 536. BALDWIN, Catalogue, p. 81. THWING, Reprints
Orig. Descript. Achat., pi. 2, f. 9. Achatinellastrum polita
Nc., BORCHERDING, Zoologica, xix, p. 81, pi. 8, f. 8, 9.

The typical polita has the upper 3 whorls bicolored, dark
yellow, chestnut or olive with a white band below the suture ;
following whorls paler, Naples yellow to wax yellow or olive
ocher, with a chocolate band below the suture. This is the
form figured by Newcomb as typical. One of his specimens is
drawn in pi. 24, fig. 2. In his description he mentions a line
of the same color (black) upon the body- whorl. Figs. 3, 4
represent specimens from Newcomb so marked. The aper-
ture is purplish within, varying in shade. Length 13.7, diam.
9 mm., 5 l / 2 whorls, to length 15, diam. 9.2 mm.

Ne\vcomb 's type locality was doubtless Kaluaaha or the ad-
joining valley east, Mapulehu, where specimens exactly like
the types are found. It is likely that Newcomb stayed with
Mr. Hitchcock, a missionary who lived at Kaluaaha, and had
a place at Kalae. Mapulehu shells are further illustrated in
pi. 24, figs. 5, 6, collected by Mr. Thaanum.

Another form is larger, up to length 18, diam. 10 mm., with
the embryonic 3 whorls white with a minute dusky tip, fol-
lowing whorls yellow (often paler on the last whorl), suture
edged below with a dark line or none. The aperture is of a
purplish tint, darker or dark-streaked near the lip.

82 PARTULINA, MOLOKAI.

P. polita differs from P. ~bella by its bulging last whorl,
which gives the sides of the spire a concave outline. In hello,
the general outlines of the spire are straight. The color-
patterns are also different. P. polita is never white and never
has white bands on the last whorl. It is also more distinctly
striated spirally, the striae visible though often weak on the
last whorl.

Color-form latizona Borcherding. PL 23, fig. 13. Shell
dextral, conic-ova.te, rather solid, glossy, very delicately lon-
gitudinally striate; epidermis pale-yellow, the whorls en-
circled in the middle with a broad chestnut-brown band ; spire
conic; suture margined, impressed. Whorls 5^, weakly con-
vex, the last scarcely broader than the spire. Columella
having a small, short, white tubercle. Aperture oblique, in
versely auriform, pale violet ; peristome simple, unexpanded,
red-lipped within. Length 14, diam. 8 mm. Kaamola, Meyer
(Borcherding) .

Acliatinellastrum latizona BORCH., Zoologica, xix, Heft 48 1 ,
p. 82, pi. 8, f. 10, 1906.

Doctor Cooke collected this form in Kaluaaha and Mapu-
lehu. It is a further development of the ordinary banded
polita.

Color-form dixoni Borcherding. PL 23, figs. 14, 15, 16.

Shell dextral, ovate-conic, rather solid, glossy, deep chest-
nut-brown, the upper whorls and a patch around the umbili-
cus yellow; very delicately striated longitudinally. Spire
conic; suture margined, impressed; whorls 5^, weakly con-
vex, the last rounded, scarcely half the total length of the
shell. Columella having a small short, flesh-red or dirty-
purple tooth. Aperture oblique, half-oval, gray- violet; peri-
stome simple, unexpanded. Length 14, diam. 8 to 9 mm. Ka-
wela and Kaamola, Meyer (Borcherding).

Acliatinellastrum dixoni BORCH., Zoologica, xix, p. 83, pi.
8, f. 11, 12, 13, 14, 1906.

" This species occurs in the above-mentioned localities in
a darker and a lighter coloring. Fig. 14 comes from Kawela,
has a dark columellar tooth and no yellow patch around the
umbilicus. Others from Kawela and fig. 15 from Kaamola

PARTULINA, LANAI. 83*

are normally colored. Fig. 16 from Kawela shows the lighter
coloration, in which there are usually two yellowish bands,
one at the periphery, the other below the suture ' (Borcher-
ding).

Merely the culmination of the dark phase of polita. Simi-
lar shells have been taken by Doctor Cooke on Mapulehu
ridge.

Color-form hepatica Borcherding. PI. 23, figs. 17, 18.

Shell dextral, long ovate, rather strong, glossy, with lon-
gitudinal and spiral lines, visible under the lens only, very
fine and not decussating. Surface yellow-greenish, liver-
brown, veined with darker streaks. Spire turrited; suture
distinctly margined, impressed. Whorls 5^, convex, the last
a little wider than spire, rounded. Columella weakly!
twisted ; columellar fold white. Aperture oblique, half-ovate,
whitish within; peristome simple, unexpanded. Length 17,
diam. 9 mm. (Borch.).

Molokai: Kawela (fig. 17), Waileia (fig. 18).

Achatinellastrum hepaticum BORCHERDING, Zoologica, xix,
p. 83, pi. 8, f. 15, 16.

The largest in Dr. Cooke 's collection measures length 17,
diam. 11 mm. Others taken by Father Dutton are smaller.
The interior is white or pale purple.

Series of P. variabilis. LANAI.

44. P. VARIABILIS (Newcomb). PI. 21, fig. 14; pi. 22, figs. 1
to 12.

" Shell sinistral or dextral, conically acuminate, polished,
shining; whorls 6, rounded, lightly margined above. Aper-
ture roundly ovate; columella short, armed with a strong
tubercle of a roseate color ; lip expanded, acute. Color of
shell white, yellow, black and other shades, plain or with
from one to six bands of various colors on the last whorl.
Length 8, width 14-twentieths inch ' (Newc.).

Lanai (Newcomb, Thaanum and others) ; Windward side,
on ridges facing Maui, above Waiapaa, behind Koele, and
Lanaihale (Perkins).

Achatinella veriabilis NEWC., P. Z. S., 1853, p. 154, pi. 24,

84 PARTULINA, LANAI.

f. 70 (1854) ; Ann. Lye., vi, p. 333. PFR., Monogr., iv, 256;
vi, 169. SYKES, Fauna Hawaiiensis, p. 319. THWING, Re-
print Orig. Descript. Achat., pi. 2, f. 11. Achatinella fulva
Newcomb, PFR., P. Z. S., 1855, p. 208 (1856). Achatinella
lactea GULICK, Ann. Lye. vi, p. 198, 1856.

Dr. Newcomb included several color-forms in his original
description, but the dextral shell with several blackish- chest-
nut bands and an umbilical patch, depicted in his figure, may
be taken as the typical form. It is copied in my pi. 21, fig.
14. In the lot given by him to the Academy there are two
dextral and nine sinistral shells, three having a dark um-
bilical patch, the others white there. Fully adult shells are
rather pot-bellied, the base being notably convex (especially
as viewed dorsally), but in some shells with thickened lip
the juvenile character of an acute peripheral angle is re-
tained for a short distance in front of the aperture. Three of
the Newcomb lot are figured, pi. 22, figs. 1, 2, 3, to show the
variation of shape and pattern.

A lot of 48 shells, from Gulick, all sinistral, shows substan-
tially the same forms and banding (pi. 22, figs. 4 to 9) . The
bands are either deep chocolate or ochraceous, and vary from
rather wide to linear. As in the Newcomb set, some shells
are quite narrow, the dimensions being about as follows :

Length 18, diam. 10 mm., average specimen.

Length 15, diam. 10 mm., rare broad form.

Length 15.3, diam. 8.5 mm., narrow form.

Four in the lot are brownish- tawny with extremely weak
bands or none; the rest are banded with chocolate and with
ochraceous in about equal numbers.

Another lot from " Large Valley, Lanai," Gulick, contains
11 shells all dextral, color white, with ochraceous or dark
chestnut bands. A lot of 7 from Mr. D. Thaanum is also en-
tirely dextral, beautifully polished and chocolate-banded.

Another lot of 58 shells from Mr. Thaanum is polymorphic,
(a) Three-banded with dark chestnut or ochraceous, the um-
bilical region always white, sometimes with a few accessory
lines, 25 specimens, all sinistral, fig. 10. (&) bands reduced
to lines or split, the upper one often wanting, 10 specimens,

PARTULINA, LANAI. 85

all sinistral. (c) White, with a faint yellow or brown stain
within the lip- callus, 7 shells, one of them sinistral. Resem-
bles lactea except in the paler aperture. Fig. 11. (d) Cream
colored, darker towards the base, to ochraceous with darker
streaks, bandless, the peripheral angle usually conspicuous,
but wanting in 7 shells; 16 specimens, six of them sinistral,
pi. 22, fig. 12. Of 4 embryos taken from a banded shell,
fig. 10, three are banded as in pi. 21, fig. 16, the fourth,
fig. 17, being ochraceous with a white band below the periph-
ery, which is more distinctly carinate than in the others.
Of 5 embryos from a pure white shell, fig. 11, three are dis-
tinctly ochraceous, one having a white band below the
periphery ; the others being white.

It appears that some colonies are exclusively sinistral,
others wholly dextral, the same color-patterns running
through both. Other colonies contain both sinistral and dex-
tral forms, and a variety of color-patterns, which are unlike
in the proportions of sinistral and dextral individuals. In
mixed colonies, left-handed coil and banding seem to be
dominant, right-handedness and plain color recessive. The
diversity of young taken from one parent indicates hybridity
between the color-forms, but data are wanting to show
whether sinistral and dextral forms hybridize.

The three-banded type, with bands placed as in the Mauian
P. nattii f is probably the most fundamental (or nearest to the
ancestral) pattern.

A. fulva (' Newcomb ' Pfr.), pi. 22, figs. 13, 14, is not dis-
tinguishable from the tawny-ochraceous form occurring in
colonies of variabilis. In sonie specimens the periphery is
acute in front of the aperture, but in others more accelerated
it is rounded. The original description follows : Shell sub-
perforate, turrite-conic, solid, striatulate, and under a lens
very delicately decussate, glossy, buff-tawny; spire sub-
concavely conic, the apex white, acute, suture thread-mar-
gined; whorls l / 2 , the upper fiat, following ones convex, the
last a little over one-third the total length, swollen below the
suture, base sac-like. Aperture oblique, subtetragonal-auri-
form, columellar fold superior, nodiform, orange colored;

86 PARTULINA, LANAI.

peristoine unexpanded, labiate within, the right margin some-
what straightened, columellar margin dilated, vaultingly re-
flexed. Length 18^, diam. 9, aperture 8x83^ mm. Sand-
wich Is. (Pfr.) .

Dr. Newcornb and all more recent authors have agreed in
considering A. fulva a synonym of variabilis. Specimens
measure: Length 17 to 18, diam. 10, aperture 8.3 mm. The
lip-callus is white, or brown with a white lip-edge. The widest
specimens approach semicarinata, but none are so wide as that.
It is sometimes cream-white.

Mr. Thaanum found P. variabilis lower, P. semicarinata^
higher on the main ridge of Lanai.

44a. Var. lactea Gulick) . PI. 22, figs. 15, 16.

" Shell sinistral, perforate, acuminately ovate, solid, shin-
ing, finely striated, and microscopically very finely decus-
sated, color ivory white ; apex acute ; spire conical, with out-
lines slightly convex; suture rnarginate, slightly impressed;
whorls 6y 2 , flatly convex; columellar fold central, brown,
strong ; aperture oblique, sinuately oval, reddish brown
within ; peristome white, thickened within ; with external mar-
gin arcuate, slightly reflected anteriorly; columellar margin
dilated, slightly detached; parietal margin thin. Length 22,
breadth 11% ; length of body- whorl 15 mm. Weight 8
grains ' (Gulick).

Lanai : S. T. Alexander, D. Thaanum.

The shell is rather snow-white than ivory-white, the colu-
rnella brown, tip of the fold paler, lip-callus purple-brown,
interior lilac. The spire is straightly conic, whorls but very
slightly convex. It is minutely but constantly perforate. It
is a very beautiful shell, differing from the white form oc-
curring in some variabilis colonies by its darker aperture. In
a series of about fifty before me, all are sinistral.

Length 18 to 19, diam. 10 mm.

Length 16.8, diam. 9 mm.

45. P. SEMICARINATA (Newcomb). PI. 21, figs. 1 to 4, 13, 15.

" Shell dextrorsal, solid, ovate'ly-conical, longitudinally

substriated ; whorls 5, flattened, margin-ate above the last, an-

PARTULINA, LANAI. 87

gulated in the middle and semicarinated. Aperture ovate;
columella ending in a tortuous, obtuse, dentiform plait ; outer
lip anteriorly produced, strongly marginate within. Color
pale yellow, aperture white internally. A straw-colored, coni-
cal species, with the last whorl partially surrounded with an
elevated keel : (Newcomb). No dimensions given.

Lanai (Newcomb, Thaanum et al.) ; mountains at high ele-
vations (Perkins) ; at the crest of the main ridge, extending a
little way down (Thaanum).

Achatinella semicarinata NEWC., P. Z. S., 1853, p. 156, pi.
24, f. 76, 1864. PFR., Monogr., iv, 532. Achatinella varia-
bilis var. semicarinata SYKES, Fauna Hawaiiensis, p. 320.

This species differs from P. variabilis fulva by its decidedly
broader contour and dextral coil. The glossy surface shows
only weak traces of spiral striation. The aperture, lip-rib
and columella are white, or in the darkest shells, the interior
is faintly pink tinted. The early whorls are the color of the
shell in most specimens, but in some they show distinct dark
bands. Embryo shells usually show bands, though not always
so distinct as in fig. 15, an embryo from a cream-colored
mother. The color of adults varies from white to tawny or
light OTange-rufous. The acute carina may extend half
way around the last whorl, or it may not be developed at all
there, ending on the penultimate whorl. Usually it extends
part way across the front of the shell. The axis is perforate.
In a series of 23 from Thaanum and Baldwin, all are dextral.
Neweomb's type figure is copied in pi. 21, fig. 13. It repre-
sents the most strongly carinate form. The size varies.

Length 18, diam. 12 mm.

Length 17.5, diam. 11 mm.

Length 16, diam. 11 mm.

Length 15, diam. 9.5 mm.

Mr. Sykes ranked semicarinata as a variety of variabilis
on account of the difficulty he experienced in endeavoring to
separate it from fulva. " Mr. Perkins remarks that ' the
broader form with ridge more raised is from higher eleva-
tions.' It appears to be gradually replaced by the form fulva,
at lower altitudes.' I have here retained semicarinata as a

PARTULINA, LANAI.

species because in the large series of both species seen the in-
tergrading specimens are comparatively very rare. Nearest to
Koela the white form occurs, further along the ridge the
darker forms. Either sinistral or dextral forms may pre-
dominate in different colonies. Fig. 13 is copied from New-
comb's type figure. Figs. 1-4, 15 are from specimens taken
by Mr. Thaanum.

45a. Var. hayseldeni Baldwin. PI. 21, figs. 5 to 12.

" Shell sinistral, minutely perforated, rather solid, ovately
conical, apex subacute ; surface shining, marked with delicate
incremental striae, and under a lens exhibiting very close,
minute, decussating spiral lines; embryonic whorls faintly
cross-lined. Color generally of a uniform reddish-brown;
sometimes the coloring of the middle portion of the whorl
shades into white on the apical whorls, and in some examples
a white line revolves below the suture. Whorls 5^, slightly
convex, narrowly margined above, the last carinated or angu-
lated at the periphery, the angle becoming almost obsolete
towards the aperture ; suture distinctly impressed and often
margined above by the continuation of- the peripheral keel.
Aperture oblique, subovate, white within with a pinkish tinge ;
peristome white, rather obtuse, thickened within, the basal
and columellar margins slightly reflexed; columella termin-
ating in a strong, flexuous, white fold. Length Yl l / 2 , diam.
10 mm.

" Animal when extended in motion longer than the shell.
Mantle slate color with a brown band encircling the outer
edge. Foot above and below almost white with a yellowish
tinge. Tentacles white tinged with slate ' (Baldwin).

Lanai (Walter H. Hayselden) ; Lanaiha'le, near highest
point of mountain (Perkins, Thaanum).

Partulina hayseldeni BALDWIN, Nautilus, x, p. 31, July,
1896. Achatinella (Partulina) hayseldeni Baldwin, SYKES,
Fauna Hawaiiensis, p. 313, pi. 11, f. 2.

This species is allied to P. semicarinata Newc., which is
found in another district of the same island. The latter is a
light straw-colored, more conical, and invariably dextral

PARTULINA, LANAI. 89

shell. The animals of the two species are somewhat similar,
but sufficiently different to warrant the separation ' (Bald-
win).

It is a local race of the highest peak of Mt. Lanaihale.
The color-patterns of the type lot are shown in pi. 21, figs. 5,
6, 7. It is uniform orange-rufous with the tip of the apex
pale, aperture pink; the same with a white subsutural band;
or in addition to this there may be a white band or bands mid-
way between the sutures on the spire, gradually deepening
on the penult, whorl, to the ordinary ground-color.

In a series of 29 shells, collected by Mr. D. Thaanum, pi. 21,
figs. 8 to 12, 5 -are sinistral. the rest dextral. The sinistral
shells have the color-patterns described above, the last whorl
being uniform orange-rufous. Seven of the dextral shells are
exactly similar in pattern to the sinistral, and they differ from
P. semicarinata only by the deeper shade of color. Ten adult
dextral shells are pale-yellowish flesh-tinted with several or
numerous orange-rufous bands, varying in intensity among
the specimens.

Embryo. A dextral specimen of the typical orange-rufous
color with white sutural band on the spire only, contained
three embryos, two having a white sutural band, a white
band a short distance below the suture and white columellar
area, the other being multilineate with brown above and
below. A dextral banded snail (pi. 21, fig. 10) contained 3
embryos, two 3-banded with some accessory lines, the other
having numerous weak bands. In both examples the embryos
are dextral.

It seems likely that the ancestral stock of semicarinata was
banded, and typical hayseldeni is a more evolved form in
which the banding has been replaced on the lower whorls or
throughout by the dark color of the bands, while typical pale
semicarinata has lost its bands entirely or on the later whorls.
The whole series is so closely interrelated that I do not think
it divisible into two species. I retain the name hayseldeni in
a varietal sense because it is already in the literature, rather
than from any real distinction, as I believe the intergradation
with semicarinata is practically -complete.

90 PARTULINA, SECTION BALDWINIA.

Section BALDWINIA Ancey.

Baldwinia ANCEY, Proc. Malac. Soc. London, iii, p. 270,
July, 1899.

Shell sinistral (except in P. dubia), perforate, thin, ovate-
acuminate or ovate-conic, with the color and sculpture of
Partulina; lip narrowly expanded, but little thickened ; colu-
mella somewhat twisted but without a callous fold or tooth-
Type A. physa Nc., 1855 = A. confusa Sykes.

Baldwinia differs from Perdicella by its larger size and
enlarged last whorl, giving the shell a conic form. It is no
doubt, like Perdicella, a derivative of Partulina, which has
the same sculpture and color-patterns.

It is quite likely that the Baldwinias of Hawaii, Maui and
Oahu are so many independent derivatives from the Partu-
lina stock, but until this is demonstrated it is simpler to group
them together. In any case the forms of Hawaii and Oahu
are in all probability the descendants of emigrants from the
Maui-Lanai-Molokai center. Whether the degeneration of
the columellar fold was expressed in the original stock is un-
certain, since its degeneration is without doubt a secondary
modification. The presence of Partulinoid forms in Oahu
seems anomalous, but the distribution of Laminella and Ptero-
discus also show that there was a limited amount of inter-
change between the Oahu and the Maui-Lanai-Molokai centers
after the modern groups had become differentiated.

Species of Baldwinia.

1. Hawaiian species.

a. Aperture much more than half the shell's length; shell
white or banded, without axial or oblique color streaks ;
surface smoothish, finely striate, last whorl much in-
flated. P. horneri, no. 47.

a 1 . Aperture slightly more half the length; last whorl
rather roughly striate ; length 'over 20 mm.

P. confusa, no. 46.

a 2 . Aperture about half the length ; shell variously banded
or streaked, smoothish; densely striated spirally
throughout, length less than 20 mm. P. physa, no. 48.

PARTULINA, SECTION BALDWINIA. 91

2. Mauian species.

a. Last embryonic whorl finely zigzag-lineate ; later whorls
moderately convex. P. grisea, no. 49.

a 1 . Last embryonic whorl obliquely striped; later whorls
very convex. P. thaanumiana, no. 50.

3. Caiman species. P. dubia, no. 51.

Mr. H. W. Henshaw, of Washington, D. C., well known as
a trained and acute observer in several departments of sci-
ence, has given the results of his studies upon this group in
the following essay.

OBSERVATIONS ON HAWAIIAN ACHATINELLHXE, BY H. W.

HENSHAW.

"Although much has been written upon the Achatinellidce
of the Hawaiian Islands, comparatively little has appeared
upon the habits of any of the species. During the early part
of the year 1903 (January- April) the writer enjoyed unusual
opportunities for making observations on the three species of
Partulina inhabiting the Island of Hawaii, and the follow-
ing notes pertain thereto.

' * The peculiar interest attaching to these particular species
lies in the fact that they occur only upon the Island of Hawaii,
the youngest of the archipelago, that the three are specifically
quite distinct from each other, that two of the three are, so
far as known, absolutely restricted to limited sections of the
northern end of the island, which is much the oldest, and that
the third is found outside of this district only rarely. Thus,
living in comparatively unoccupied regions and practically
without competition, the three species offer an unusually fav-
orable opportunity for a study of their habits, and especially
of the method of their colonization and dispersal.

" Upon Calm, the metropolis of the Achatinellida proper,
there are so many closely-related species, several often crowded
into the same area, that the facts relating to habits and
means of dispersal are much complicated. An understanding
of the manner in which the species of Achatinella form new
colonies and of the laws of color variation in the group will

92 PARTULINA, SECTION BALDWINIA.

throw a flood of light upon the vexed question of specific re-
lationship. The writer believes that a study of a carefully-
selected series of the one hundred and more species of the
Oahu Achatinellas by a student familiar with the local geo-
graphic conditions would reduce the number by at least one-
half.

Partulina horneri Baldwin.

" This species, described in 1895 from specimens collected
by Mr. J. Lewis Horner, appears to possess a present rather re-
stricted range in the sparsely timbered region above Kukui-
haele, Hamakua, and on the ridges above the Waipio and Wai-
manu valleys, an area of perhaps three or four square miles.
Except for a few specimens obtained above Honakaa (on the
doubtful authority of a Kanaka's statement) and Kukaiau
(introduced or * planted : there by Eugene Horner) ten to
twenty miles distant, this shell has never been found else-
where. This restricted habitat, in a region where the compe-
tition is almost nil, represents all the territory the species has
been able to acquire since the unknown date of its occupancy
of the island. That the period of its occupancy has been con-
siderable appears from the striking unlikeness of the species
to all its congeners.

" Although occasionally living upon ohia (Metrosideros
polymorpha) , kolea (Myrsine lassertiana) and, rarely, upon
kawaao (Bryonia sandwichense) , the species is chiefly con-
fined to a small berry-bearing tree called the ahakea (Bo~bea
elatior) which is rather common at an altitude of 1,800-2,500
feet. This species, and in fact practically all the Achati-
nellas the writer is familiar with, is often found on half-dead
trees, and not; infrequently a thriving colony may inhabit a
tree which boasts of but a handful of foliage. This appears
to indicate that once a colony is established on a tree it per-
sists, as long as it can obtain food, and I have found small
colonies of shells on stubs of trees that apparently had been
dead twenty-five or thirty years.

" This and the following species belong to the sub-genus
Baldwinia of Partulina, and they are much more closely re-

PARTULINA, SECTION BALDWINIA. 93

lated to the shells of Molokai and Maui than to those of the
more distant islands of the group farther north. It is highly
probable, indeed, that the latter island furnished the parent
forms, the channel separating the two islands being less than
twenty miles wide. As species are reckoned in this group,
the Hawaiian forms are distinct enough from any Maui shells,
although the A. physa of Hawaii is rather closely related to
the P. grisea of Maui, both in form and coloration, and the
latter may have been the parent form, or at least descended
from a common ancestor.

" The color of the typical and prevailing form of A. hor-
neri is white with a broad zone of brown around the periph-
ery, thence visible to the apex as a faint line above the suture.
There is also, as noted by Mr. Baldwin, a pure white form of
this species, and a third variety with only the apical whorls
brown banded. To these is to be added a fourth color-variety,
not previously noted, which is white save that the apex, for
a varying distance, is tinged with brown. Of the four color
forms, only the typical shell and the white variety are at all
common ; the other two are rare.

" There is some evidence to prove that formerly, even so
late as ten years ago, the brown-whorled form was much
more abundant than it is now, and many dead shells have been
found on the heights between Waipio and Waimanu valleys
where the living shells are very rare indeed (some living ones
have been found by Mr. Thaanum) .

" It is to be noted particularly that all the above forms of
horneri occur in the same locality and on the same species of
trees, but individual trees are never 'occupied by two forms.
Moreover, with the possible exception of the banded form and
the pure white one, the color varieties appear not to inter-
grade.

' * As regards shape the four forms do not seem to differ ap-
preciably. Roughly speaking there may be said to be a large
and a small form of each of the four, but it is of interest to
note that, while the individuals of each form intergrade in
size, large and small forms do not live upon the same tree.
The adult individuals on a tree are either all large or all

94 PARTULINA, SECTION BALDWINIA.

small. Thus the isolation of a few feet seems sufficient to
permit recognizable changes both of size and color, though
not of form.

" As these four color-forms occur in the same locality and
are only to be distinguished by pattern of the shell, it might
naturally be inferred that the embryonic young of the four
would be of a common type, perhaps reverting in color to an
ancestral form, and in any event not perpetuating the color
peculiarities of their respective parents. The contrary is
true. The young are not of a common type, but are always
distinguishable inter se. Though markedly different in the
adult state, there is less difference between the young of the
white and the banded form than of the other two, but there
is still a recognizable difference. The embryonic young of the
pure white form retain the peripheral brown band of the
typical horneri until about a third grown when only traces
of it are discernible near the aperture and these finally en-
tirely disappear leaving the shell pure white. The embryonic
shells of the other two forms differ more markedly from each
other and from the young of the other forms.

The unlikeness of the respective young of several mol-
luscan forms would appear to be an excellent criterion of
their permanent distinctness, and the writer is inclined to
view these two forms as far better entitled to specific rank
than many named species of the group whose claim to dis-
tinction rests upon equally slight color characters supple-
mented, it is true in some cases, by present geographical
separation of habitats.

" It is an interesting fact that the embryonic young of the
brown-tipped form differ more from the young of the others
than do the parent shells. The adult of the brown-tipped var-
iety is to be distinguished only by the light-brown color of
the apical whorls. The embryo shells have the lower half of
the basal whorl of a deep reddish-brown. Evidently the brown
tends to fade out as the shell matures. It is probable that in
time the deep brown apex will become a permanent character
of the adult of this form as it now is an invariable character
of the young.

PARTULINA, SECTION BALDWINIA. 05

"It is probable enough, too, that in time the four color
sports, if indeed two of them are not already properly to be
ranked a;s species, will become completely isolated geographic-
ally when the present differences will be accented and per-
haps further differentiation occur. No doubt many of the>
recognized species of the group have had origins and life his-
tories essentially similar to the forms here noticed.

' How long these forms of Jiorneri have been in reaching
their present degree of differentiation cannot, of course, be
told. Possibly they are nearly as old as the life of the species
on this island; but in the case of creatures as plastic as the
Acliatinellidce appear to be, probably no very long interval,
nor very complete isolation, is requisite to establish such var-
iations.

Partulina pliysa Newc. (hawaiiensis Baldwin).

" This shell was described by Mr. Baldwin from the same
locality as the preceding, and its vertical range is about the
same. It lives chiefly upon the smooth-leaved variety of the
ohia but is found also of the kolea and casually on the haa
(Antidesma platypJiyllum Mann).

" Though this species varies much in coloration and size,
there cannot be said to be any constant and well-defined color
varieties, all the forms appearing to intergrade without limit.
As in the case of the previous species, however, every colony,
and sometimes the shells from individual trees within the
confines of a colony, possess peculiarities of their own slight
differences of color and size, often sufficient to the eye of a
trained collector to distinguish them from those of ether
'colonies. The tendency to differentiate is the same In kind,
but for some reason or other is has not progressed so far as
in the case of horneri ; possibly isolation has not been so com-
plete as in the other species, although observation does not
confirm this.

The embryos of this species tend to follow the coloration
of the parent form to some extent, especially when the latter
are deep brown, but in most cases their colors do not indicate
their parentage with any certainty.

96 PARTULINA, SECTION BALDWINIA.

" This species has been far more successful in extending
its range than the previous one, and is found in three small
and isolated colonies far beyond the confines of Hamakua,
which probably was its starting point on the island. The
Partulina physa errans from Olaa, Kaiwiki and Puna, from
forty to sixty miles distant and separated by innumerable
gulches and water-courses, is a derivative of this species if
indeed, as the author believes, the variation is not entirely
compatible with specific identity.

" The author has learned of a tree shell which many years
ago lived abundantly on the ohias in South Kona, above and
not far from Kealakekua Bay, and which now appears to be
extinct there. If, as is probable from its description, this
shell was physa (haivaiiensis) , the species has nearly encircled
the island of Hawaii, affording a marked contrast to the re-
stricted range enjoyed by most of the species of this family,
especially upon Oahu.

" It is to be remarked that this species, like the preceding,
'occurs chiefly in sparsely wooded districts and upon isolated
trees. It is difficult to understand the wide dispersal of this
particular species and especially its speedy appearance in
clearings after the cutting of the forests, unless individuals
exist here and there in the deep and virgin forest, where, how-
ever, the shell has never been found. Deep forest tracts ap-
pear to be inimical to the welfare, not only of the Achati-
nellas proper, but to almost all other Hawaiian land shells.

Partulina confusa Sykes (physa of authors).

This species probably was formerly much more wide-
spread in the districts of Kohala and Hamakua than it is
to-day, owing to the present restriction of the forested area.
It occurs at a higher altitude than the two preceding shells
and extends at least as high as the upper edge of the Waimea
Plains (3,500 feet) over most of which area it formerly oc-
curred, although, like the preceding species, it is doubtful if
it ever lived in the deep forest except casually.

' ' During the spring of 1903 the author had the opportunity
of examining an isolated colony of this species on the Waimea

PARTULINA, SECTION BALDWINIA. 97

Plains, which for its extent and peculiar character deserves
more than passing notice.

The P. confusa lives chiefly upon the pua tree (Olea
sandwichensis) although in some localities it has been found
abundant upon the ilima (Sida sp.), a low shrub with yellow
flowers. It occurs also upon the mamani (Sophora chryso-
phylla) . Whether from the fact that the pua is not common
over the island of Hawaii, or for some other reason, I did not
find this shell, which is far more abundant than either of the
other species, outside of the above region (Mr. D. Thaanum
has found it widely spread in the Waimea plains) .

The colony in question occupies perhaps 150 pua trees,
which cover an area of perhaps half a mile square, many
trees being more or less widely separated from their fellows,
The pua is a small tree fifteen or twenty feet high and with a
small spread of branches. A rough estimate of the number
of adult shells inhabiting this area when first visited is more
than 75,000 shells, and it was possible to ride under the trees
and from their trunks, leaves, and branches to pick shells
literally by the handfuls. Cavities in the trunks and branches
were usually packed with shells, mostly immature, from 50 to
75 being often found together. Wherever there were logs,
dead branches or large rocks under the trees, they, too, har-
bored large colonies of shells, and the presence of young in
numbers showed that these places were their permanent
abode.

' ' In six hours, without climbing a tree, the author had no
difficulty in gathering 1,100 adult shells, and his companion,
Mr. William Horner, did the like a statement which will suf-
ficiently attest the abundance of the mollusks. The locality
becoming known to several shell collectors, something like
10,000 adult shells were gathered in the course of threes
months without seriously diminishing the numerical strength
of the colony. It is doubtful if so many tree shells of any
one species were ever found in a like area, even in the Ha-
waiian islands, abundant as the Achatinellas are, or used to
be, in some places.

" The colony in question comprised numerous color varie-

98 PARTULINA, SECTION BALDWINIA.

ties and the shells differ much in shape. Mr. Homer, of Ku-
kuihaele, has succeeded in selecting from several thousand
no fewer than 200 and odd more or less distinct varieties.
For the present purpose of the author it will suffice to enum-
erate a few only of the more marked forms in the author's
own collection.

1. Uniform light gray, with or without faint brown
markings on apical whorls.

' 2. Distinct brownish-gray with more or less distinct mark-
ings on apical whorls. Undoubtedly these two forms, the
second being but a slight variation of the first, are to be con-
sidered as the typical ones of the colony; at a rough guess
one-half of all would come under one or the other.

' 3. Dark brown, almost 'chestnut, with lighter colored
apical whorls; lower whorl sometimes with, sometimes with-
out, gray spots.

" 4. Light gray with broad or narrow light-brown band
across body whorl.

' 5. Light gray with broad white band across body whorl,
bordered with narrow brown band.

: 6. Light gray, more or less streaked and spotted with
brown; a distinct white band on body- whorl bordered by a
chestnut band.

" 7. Light gray, with chestnut body- whorl bordered above
and below with white.

" 8. Many-banded with alternate bands of brown and
white from apex to base, some narrow, some broad.

' 9. Dwarfed and somewhat distorted form, running
through all the above variations, and found only on mamani
trees (Sophora chrysophylla) .

[Plate 16, figs. 2 to 12, represent shells from this colony.]

" The exact size and shape of the above varieties vary
almost interminably and, it is perhaps needless to say, inter-
grade with each other completely.

" The above brief notes afford but little idea of the great
variety and apparent distinctness of some of the color forms.
Many species of the genus AcJiatinella appear in printed lists
which are based upon color characters much less marked than

PARTULINA, SECTION BALDWINIA. 99

those which distinguish this series. The number of forms
might be multiplied almost indefinitely, but the -above will
suffice to indicate the general character of the variations ex-
isting within this single colony.

11 As in the case of P. horneri, a distinct tendency is ob-
servable, though by no means so marked, to the segregation of
the different color varieties upon individual trees, indicating
that isolation has been sufficient to permit differentiation
within certain limits, while the fact that many trees support
a sprinkling of other varities, added to the completeness of
the intergradation, proves that intercommunication has been
to some extent potent in limiting the amount of the diver-
gence.

" The following facts bear upon the question of intercom-
munication between the several parts of the colony. The pua
and mamani trees, upon which the shells live, rarely touch
each other, and are usually separated by a distance varying
from a few feet to a hundred yards. The land is pasture
and the grass under many of the trees not in the open is
deep. There is thus no chance for the shells to pass freely
from tree to tree.

" Upon Oahu and the other islands of the group deep
valleys and high ridges in most localities tend to the more or
less complete isolation of shell colonies, and also present
diverse conditions of foliage and environment favorable to
the differentiation of new forms. The environment of the
present colony on the contrary is exceedingly uniform and
hence presumably unfavorable to the origin of new varieties,
but the isolation of the trees and the open nature of the
ground form barriers to free intercourse quite as effective
as ridges and valleys, and hence tend to the perpetuation of
any chance varieties that may arise.

The author is aware that some observers 'consider the
Achatinellidcc to be active and rather extensive travelers, but
his study of the three species under consideration points to
the opposite conclusion. The fact, as noted above, that logs
and rocks under the trees form the permanent abode of such,
shells as chance to fall from above, seems to show that the

100 PARTULINA, SECTION BALDWINIA.

shells from some cause are unable to find the way back to the
tree on which they were born though but a few feet distant.
A favorite resting place of the Achatinellas is on the under
side of leaves, especially dead leaves, and as these fall or are
blown off by high winds they naturally carry the shells with
them usually to the ground immediately beneath the tree.
In a number of instances shells that had shaken from the
tree into the grass from six to twelve feet away from the
trunk were found where they had fallen several weeks later,
being apparently quite lost. That Achatinellas can travel
twelve or fifteen feet over a tree in a single night there can
be no doubt, since the author has observed them in the act
and has measured the distance traveled; no doubt they arc
able to travel much farther than this, as certainly can the
Amastras and the apparently still weaker Succineas. It
seems likely, therefore, that natural sluggishness or defective
vision, perhaps both, are responsible for the failure of such
shells as fall or are blown from trees to find their way back
to the parent trunk. "Whatever the explanation, the fact that
the three species of Achatinellas here considered are as a
rule very sedentary and rarely or never regain their place on
a particular tree when once dislodged seems to the author in-
disputable.

" The question then arises as to the means of dispersal
from tree to tree and from locality to locality possessed by
shells under 'Conditions similar to those here indicated. The
author is convinced that the dispersal of the three species
here mentioned is effected chiefly through the agency of wind
when the shells are young. In several instances he has found a
single young Partulina, but a few days old, to be the sole oc-
'Cupant of a shrub or tree (so small as to be readily examined
leaf by leaf) separated several hundred yards from the near-
est shell-bearing trees. In such cases no other transporting
agency suggests itself but the wind or birds.

" It would require a considerable gale to carry to a dis-
tance a leaf laden with an adult Acliatinella, but a very mod-
erate wind would suffice to <carry a leaf with a young one
which weighs but a gramme or two for a considerable dis-

PARTULINA, SECTION BALDWINIA. 101

tance. An adult shell once upon an isolated tree, one that is
separated from its fellows even by a few yards,, might live
and die leaving its progeny, if it chanced to 'be pregnant, to
perpetuate its peculiarities, subject only to the modifying
influences of such individuals ais by rare chance might be
blown to the same tree.

Where shell-inhabited trees of the same species interlock
branches a more or less free interchange of inhabitants would
naturally take place, limited only by the natural sluggishness
of the mollusks, and, as a matter of observation, the forms of
P. confusa inhabiting closely adjacent pua trees were found
to be practically indistinguishable.

' As bearing upon the problem of the variation of this
species, a most suggestive fact was noticed in connection with
the colonies upon the mamani trees. This is a small-leaved,
leguminous tree, and in appearance, and no doubt in fact, is
not well adapted to molluscan life. Yet generally where the
mamanis adjoined the pua trees, and invariably where they
touched branches, the mamanis were found to be occupied by
-colonies of confusa. These shells were almost always smaller
than those of the neighboring pua trees, often had diminutive
and distorted spires, and in every external aspect suggested
the effect of insufficient food or of unfavorable conditions r f
life. Whatever the explanation, the change of habitat from
-one species of tree to another closely adjoining has resulted
in establishing a recognizable form, striking evidence of the
plasticity of the Ackatinellidce, and of the slight nature of
what seem to be insurmountable barriers to free intercourse
between separate, but closely adjoining, molluscan colonies.

" An examination of the embryonic shells of confusa dis-
closes that in a few of the more strongly marked forms the
embryo tends to perpetuate the color peculiarities of the
parent. Thus the young of the gray form are like their par-
ents, and are almost destitute of the zigzag markings which
decorate the spires of almost all the forms, but are often
wanting in adults of the gray variety. So, too, the young
of the brown shell, with brown and white body band, are
often but not always marked like their parents. But it would

102 PARTULINA, SECTION BALDWINIA.

be impossible to determine with any degree of certainty the
parentage of the greater number of embryonic shells of this
species from their appearance alone whether they came from,
banded, multi-banded, or unicolored shells. Clearly the color
varieties are not yet fully established.

" It is well known that many of the species of the Achati-
nellidce are indifferently dextral or sinistral. Some, how-
ever, are consistently one or the other, and the P. confusa be-
longs in the latter class. Among upwards of ten thousand
individuals of confusa examined, not a single dextral shell
was found.

' ' Most of the Achatinellidtf proper, as is well-known, dwell
upon trees and the greater number of individuals of the
several species pass most of their time upon the leaves, feed-
ing and moving about only by night and, to a very limited
extent, on dark and rainy days. A certain amount of)
moisture apparently the more the better seems indeed nec-
essary to their mode of progression, if not to their very exist-
ence, and in dry weather they are strictly sedentary.

' ' The three species in question dwell indifferently upon the
trunk, the branches, and the leaves of trees, though in day
time the greater number are attached to leaves. As a rule
they appear to be very sluggish and often remain for weeks,
especially when young, attached to leaves and branches with-
out moving and apparently without eating. Individuals in
captivity have been known to live a couple of months without
food or water, which argues the possession of considerable
vitality. Under such circumstances adult shells are known
often to give birth to apparently healthy young.

' ' Unlike the Arnastras and the Succineas, the Acha,tmelUd&
proper do not appear to feed upon the chlorophyl of
leaves; at least the leaves upon which the shells are found
seem never to show visible signs of their work. The two
former groups consume leaf tissue voraciously, both dead
and living foliage, and often leave them completely skeleton-
ized. Both of these groups live also no doubt to a considerable
extent on fungi, as they are often found upon dead wood.
Indeed the presence of Succinea in numbers upon coffee trees,

PARTULINA, SECTION BALDWINIA. 103

for instance, is a pretty sure sign that the trees are no longer
in a healthy condition.

" With a view to determining the nature of the food of
A. confusa, the writer with the assistance of Dr. Nicholas
Russel, examined microscopically the feces of a number of
individuals. In only two instances were minute particles of
chlorophyl found and these, perhaps, were accidental. The
great mass of the remains appeared to be of minute fungi or
algae.

' ' To what extent the fungus-eating habit prevails through-
out the Achatinellidce remains to be determined by future
observations, but from the fact that many other species are
known to reside upon dead trees, or upon the dead branches
of living trees, it may be presumed to be somewhat general if,
indeed, fungi be not the main reliance of the family. In con-
nection with the food of the Achatinellas, it is of interest to
note that in a number of instances small but apparently thriv-
ing colonies were found on dead stubs which had not borne a
green leaf for years, perhaps twenty or more. It would seem
that under such circumstances fungi must be the sole depend-
ence of the shells. As such stubs are often within a few yards
of living trees of the same species, there would at first sight
appear to be no reason why the colonies did not abandon the
trees after they died and make their way to other trees. Ap-
parently, however, the mollusks either do not care to migrate
even a short distance, or are incapable of doing so. - In any
event, their isolation, whether voluntary or involuntary, is
complete, and successive generations of mollusks live and die
on the same tree where they were born, even after the latter
perishes and until it finally crumbles away.

" The ability of the Achatinellas to maintain life for con-
siderable periods without food, and to found colonies and
live indefinitely upon dead trees, greatly favors their chances
in the struggle for existence.

' Many experiments have been tried from time to time in
transplanting the Achatinellas from one locality to another
and even from island to island, but in no instance known to
the author have they been successful. The writer experi-

104 PARTULINA, SECTION 3ALDWINIA.

men ted in this way with the three species here mentioned,
removing perhaps a hundred individuals, old and young,
from an altitude of between 1,600 and 3,000 feet to about 800
feet above and near the sea, where they were placed upon a
magnolia tree. The chances of the colony surviving such a
complete change of environment seemed very small. Never-
theless for a short time the shells seemed to thrive, but they
were discovered by rats and the colony was soon exterminated.

" Great numbers of Achatinellas, as well as Succineas, in
their natural habitat on the several islands are destroyed by
rats and mice, and the extermination of many of the species
is now being hastened by these agencies.

" An Achatinella which Mr. D. D. Baldwin refers to Par-
tulina physa (=s confusa) was found by the writer abun-
dantly in the semi-fossil deposits closely adjoining the colony
of that shell above described. The great size of this fossil
leads to a first impression of its distinctness from confusa,
but a careful comparison of many specimens constrains the
writer to adopt Mr. Baldwin's view, since, as remarked by
that veteran conchologist, a description of the fossil would
read precisely like that of confusa save in respect of size.

" Tables are appended which indicate the degree of pro-
lificness of the species here treated of. The season when the
shells were collected (January- April) was unusually cold and
stormy, and perhaps in mid-summer different ratios might
be had, although Mr. Thaanum. who has dissected many Oahu
species with reference to the number of young, tells me that
he has never noted much if any seasonal difference in this
respect, the shells bearing young the year round.

' In several of the forms the number of shells examined is
too small to give results of much value. In the case of the
brown- whorled Iwrneri, for instance, a greater number would
doubtless materially reduce the average number of young
per adult shell. ' '

PARTULINA, SECTION BALDWINIA.

105

4-3
S O3

H3 :H
a v

o ~
ft

CO

o

b

X!

a

V

o

fc

6
>>

M

X5

a

o>

r-t

2 embryos.

to
O

b

X3

a

OJ
CO

CO

O

b

a

o>
^

./'
o

b

,0

a

0)

' .

CO

O

t~.

--I

X!

a

0>
CO

co

c r^
c " <u

f^ _^^^

c g-3

<D >-, <U

t* 5r ^

Cj ^-< -t-J

s- fl

g: S^s
<i

P. physa (hawaiiensis) . .

P. horneri, brown banded
variety .

247
16

47
2

148
3

44
3

4
6

4
2

1.32

2.5

P. horneri, white variety.

P. horneri, brown whorled
variety

33
10

J6
1

5

4
1

5
3

3
3

1

i

2.35^-

3.77+

P. confuscL

149

50

59

24

IB

1.5

j

46. P. CONFUSA (Sykes). PI. 16, figs. 1 to 12.

; Shell sinistral, pointed at the summit, strongly inflatad
below, rudely striated obliquely. Whorls 6, first five flatly
convex, the last largely inflated and obsoletely carinated.
Suture simple and deeply impressed. Oolumella short,
slightly callous, broad, and partially covering a deep um-
bilicus. Aperture large, semiovate; lip subreflected below,
thin and simple above. Color of a dingy white, occasionally
marked with yellow flammules, internally of a light-lemon
yellow. Length 18, diam. 12-twentieths of an inch' (Newc.).

Hawaii (Newcomb) ; Kohala (Baldwin) ; near Mana, Ham-
akua (Thaanum) ; Waimea plains; a very large form in fossil
deposits of Mana (Henshaw) ; Waimea side of Kohala Mts.
(L. A. Thurston).

Achatinella physa NEWCOMB, Proc. Boston Soc. Nat. Hist.,
v, p. 218, September, 1855 ; American Journ. Conch., ii, 7).
214, pi. 13, f. 10, 1866. BALDWIN, Catalogue, p. 7 (Kohala) ;
Proc. A. N. S. Phila., 1895, p. 225 (animal). THWING, Re-
print Orig. Descript, Achat., pi. 2, f. 21. HENSHAW, Journal
of Malacology, xi, p. 57, 63, Sept., 1904. Achatinella (Par-
tulina) confusa SYKES, Fauna Hawaiiensis, ii, p. 312, 1900,
based upon A. physa Newcomb, Sept., 1855. Partulina

106 PABTULINA, SECTION BALDWINIA.

physa Newc., var. ph&ostoma ANCEY, Proc. Malac. Soc. Lon-
don, vi, p. 121, June, 1904. Achatinella physa Newc., var.
procera ANCEY, Journ. of Malacol., xi, p. 69 (Palihoukapapa,
in fossil bed).

Doctor Newcomb's second description is reproduced. I
believe with Mr. Sykes that Newcomb's first description
and figure (pi. 17, fig. 6) of A. physa (1854) pertain to the
form subsequently called A. liawaiiensis, which he took to be
the young stage of the larger shell which he afterwards
(1855) obtained and later figured as A. physa. One lot in
the Robert Swift collection (A. N. S. P.) at least 50 years
old, received from Newcomb, contains two ' hawaiiensis }
with others of the large form like Newcomb's figure of 1866.
Dr. C. Montague Cooke and Mr. Thaanum, who have gone
over the evidence and figures with me, agree in this interpre-
tation of Newcomb's A. physa, confirming the opinion of Mr.
Sykes.

The first whorl of the embryo is smooth ; subsequent whorls
are finely striate spirally. The first two whorls are creamy-
brownish, after which there are zigzag brown flames (fig. 12).
These may continue as far as the middle of the fifth whorl,
but they usually disappear earlier. In other embryonic shells
there is a brown band just below the periphery, followed by
a white band (fig. 11). These persist in the adult shell to the
end. In a number of examples with uterine young sent by
Mr. Thaanum, the young shells of one parent are not diverse,
but in every case agree with the mother in pattern. These
observations have been confirmed by the examination of large
series in Mr. Thaanum 's collection. The periphery is acutely
carinate and the columella is sinuous in embryo shells at all
stages examined.

On the last whorl of the adult shell the crowded spiral lines
of the early stages become weak, often almost obsolete, and
wide apart. The color varies as shown in the figures and
described below. The columella is brown or white, nearly
straight or somewhat sinuous. Umbilicus half open or more.

Length 26, diam. 17, aperture 14 mm.; 6^ whorls.

Length 26.2, diam. 15.5, aperture 12.8 mm. ; 6 T / 2 whorls.

PARTULINA, SECTION BALDWINIA. 107

Length 26.3, diam. 16, aperture 14 mm. (near Mana).

Length 23, diam. 15.2, aperture 12 mm. (near Mana).

Length 23, diam. 13.8, aperture 11.2 mm. (near 'Mana).

Length 27.7, diam. 17.5 mm. (Mana, Bishop Mus., no.
22357).

Var. phcuostoma Ancey. ' ' Shell glossy, pale coffee colored,
the apex white, posterior part tessellated, marked with close
spiral lines and rough growth-striae. Aperture and colu-
mella brown, throat glittering, brown, the edge a little paler.
Length 23, diam. 14.5, oblique length of aperture 11.5 mm.
Hawaii, Thaanum ' (Ancey). The type of phceostoma, now
in coll. Bishop Museum, shows no racial characters whatever
to differentiate it from confusa, and I do not think it a valid
variety or race.

Var. procera Ancey : ' was never denned in any way, but
one may infer that Mr. Ancey 's intention was to name the
large fossil form of the Palihoukapapa bed.

Hon. L. A. Thurston collected a series of confusa on the
Waimea side of the Kohala Mts., including some shells darker
than any from localities farther east.

47. P. HORNERI (Baldwin). PI. 17, figs. 1 to 5.

" Shell sinistral, minutely perforated, thin, globose with a
short acutely conical spire, apex acute ; surface shining, stri-
ated with fine incremental lines, and under a lens exhibiting
very close and delicate decussating spiral striae; embryonic
whorls faintly cross-lined. Color dull white, encircled at the
periphery with a faint brown zone which is continued on the
suture, also with a very small patch of same color around the
umbilicus. Whorls 6, the upper five slightly convex, the last
very much inflated, forming the greater part of the shell;
suture distinctly impressed. Aperture oblique, sub-rotund,
very large, white within, distinctly showing the external
peripheral band ; peristome margined with light brown, rather
thin, very slightly thickened within, expanded, basal and
columellar margins narrowly reflexed, extremities slightly
converging and united by a thin callus; columella light
brown, very slightly developed, plain and smooth. Length
24, diam. 18 mm.

108 PARTULINA, SECTION BALDWINIA.

1 1

Animal in motion longer than the shell. Mantle black,
margined with gray. Foot above and below gray, the superior
portions lightly granulated. Tentacles light gray ' (Bald-
win).

Hawaii : Above Kukuihaele, Hamakua, type loc. ; ridges
above the Waipio and Waimanu valleys ; above Honakaa and
Kukaiau (see p. 92) . Fossil in the Mana deposits.

Achatinella horneri BALDWIN, Proc. A. N. S. Phila,, 1895,
p. 224, pi. 10, figs. 20, 21, 22. HENSHAW, Journ. of Malac.,
xi, p. 63.

Distinguished from P. confusa by its shorter spire, much
more inflated, broader last whorl, and the less roughened,
more even and shining surface. Moreover, the embryo has no
zigzag-striped stage. Named in honor of Mr. J. Lewis
Horner. The following color-races are found.

(a) PI. 17, fig. 1. White, with a brown band at the periph-
ery, ascending the spire in a narrow line above the suture.
Size large, length 23.5, diam, 18, aperture 14.6 mm. ; 6 whorls.

(ft) PI. 17, figs. 2, 5. Brown, with a narrow white band
below the suture, one above, another below the periphery.
The brown fades on the penultimate or on the last whorl,
which in large part or entirely is white. 23 x 16.4 mm.

(c) PI. 17, figs, 3, 4. White, with a pale brown band im-
mediately below the periphery, which fades out on the fifth
whorl, being visible therefore only in immature shells, the
adult stage pure white throughout. Embryo ivory-white
with a distinct brown subperipheral ba.nd. 22 x 15.5 mm.
Above Kukuihaele, Hamakua,

(d) Young stage light brown, darkest just below the
periphery; adult white with pale brown spire. Embryo bi-
colored, pale brown above, darker brown below the periphery.
20.5 x 15.5 mm. Above Kukuihaele.

These several color-forms are evidently incipient races, as
they O'Ccur in pure colonies, so far as known. Races c and d,
Mr. Thaanum informs me, occur three or four miles apart.
The embryonic forms are much more distinct than the adults,
and are constant in the colonies.

By the loss of zigzag stripes on the embryo and young

PARTULINA, SECTION BALDWINIA. 109

stages, this species is more evolved than other Hawaiian Bald-
winias.

48. P. PHYSA (Newcomb). PI. 17, figs. 6 to 13.

" Shell sinistral, acutely conical, thin, inflated. Whorls 5,
rounded; suture well marked. Aperture widely ovate; colu-
mella thin, slightly twisted ; lip simple. Color light yellowish
brown with white longitudinal flammules, with or without a
subcentral revolving white line on the body-whorl. Length
11, diam. five-twentieths of an inch ' (Newc.).

Hawaii: Mauna Kea (Newcomb); Hamakua (J. Lewis
Homer) ; Waimea Plains, Hamakua. Chiefly on dead trees
but sometimes on living lehua (D. Thaanum).

Achatinella physa NEWC., P. Z. S., 1853, p. 152, pi. 24, f.
64, 1854, SYKES, Fauna Hawaiiensis, Mollusca, p. 316.
Achatinella kawaiiensis BALDWIN, Proc. A. N. S. Phila., 1895,
p. 225, pi. 10, f. 24-26 (Hamakua). GWATKIN, t. c., p. 238
(radula). THWING, Reprint Orig. Descript. Achat., 1907,
pi. 2, f. 22.

Smaller and less roughened than P. confusa, and differing
in coloration. Some shells have very much the pattern of the
young stage of confusa. Newcomb 's original description and
figure (copied in pi. 17, fig. 6) are given above. A. hawaiien-
sis Baldwin is figured from the type lot sent by Mr. Baldwin,
pi. 17, figs. 7, 8, 13. It is believed by Mr. Sykes to be iden-
tical with Newcomb 's original A. physa, an opinion in which
I agree, having gone over the matter carefully with Dr. Cooke
and Mr. Thaanum.

A. physa varies widely in color-pattern, as follows:

(a) White, with brown oblique or zigzag streaks above a
narrow peripheral brown band, and having a broad brown
zone on the base (pi. 17, figs. 7, 8).

(&) White, with more or less brown on the spire, the
peripheral brown band and basal zone more or less inter-
rupted (fig. 13).

(c) Brown, indistinctly mottled and streaked with cream-
color, with a subperipheral band of the same or of white (fig.
9).

110 PARTULINA, SECTION BALDWINIA.

(d) Brown, with subperipheral white 'band, and zigzag
white stripes more or less developed (figs. 10 to 12).

Specimens more or less intermediate in pattern connect
these forms, which (figs. 9 to 12, Waimea Plains) live to-
gether in the same colonies. Pattern d changes to a streaked
pattern on the latter part of the last whorl in adult shells
(fig. 12). The embryonic shell is like that of P. confusa in
form except that the columella is more strongly convex. It
is marked with broad vertical stripes of cream-white and
brown, their edges irregular or zigzag.

Length 19.7, diam. 12, aperture 9.8 mm.; whorls 6%.

Length 18, diam. 10.2, aperture 9 mm. ; whorls 6.

Length 17, diam. 10.5, aperture 9 mm. ; whorls 6.

The original description of A. hawaiiensis follows.
' Achatinella hawaiiensis. Shell sinistral, minutely per-
forated, very thin, acutely conical, apex acute ; surface rather
lusterless, covered with, fine lines of growth, and under a lens
showing extremely close and delicate decussating spiral lines ;
nuclear whorls faintly decussated. Color very variable, plain
brown or dingy white, sometimes irregularly striped or
mottled with brown and white, the base generally uniform
brown, but sometimes with undulating markings of brown
and white ; the only constant characters being a brown, some-
times interrupted, line at the periphery, bordered below with
a broader white line. "Whorls 6, slightly convex, the last in-
flated. Suture lightly impressed. Aperture oblique, oval,
brown, the peripheral -brown and white bands distinctly
marked within. Peristome acute, not thickened within, ex-
ternal margin straight, basal expanded, the expanded portion
being very thin and fragile, the columella margin reflexed
over the minute perforation ; color white on both face and the
reverse. Columella. white, very slightly developed, plain ani
smooth. Length 18, diam. 10^ mm.

" Animal when extended in motion longer than the shell.
Mantle almost white, margin of a darker shade. Foot above
and below dingy white, superior portion sometimes flecked
with gray. Tentacles of darker shade. The dentition is the
same as that of the arboreal Achatinellas generally. A cen-

PARTULINA, HAWAII AND MAUI. Ill

tral tooth is present ; and the formula of dentition is
125-1-125 X 120 = 30,120. ' ' (Baldwin) .

48a-. P. physa errans, n. var. PL 17, figs. 14, 15, 16.

The shell is thin, sinistral, more slender than usual in P.
physa (haiuaiiensis) , minutely perforate or almost imperf or-
ate, smooth and slightly shining, under a lens minutely
marked with growth-lines and closely, distinctly striate spir-
ally throughout, the striae rippled. Embryonic shell marked
with broad axial brown stripes as in P. physa ; neanic whorls
marked with brown zigzag stripes on a white or cream
ground ; on the last whorl the stripes are usually interrupted
by a light band below the periphery, and are irregular on the
base. Sometimes the brown markings are very faint. The
whorls are rather strongly convex. Aperture varying from
purplish-brown to faint pink within ; lip thin, very narrowly
expanded, white-edged within. Columella reflexed as in P.
physa.

Length 18, diam. 10.8, aperture 9.2 mm. ; whorls 6.

Length 18, diam. 11, aperture 9.8 mm.

Length 16, diam. 9, aperture 8 mm.

Near Pahoa, Puna, type loc., fig. 14, and Kaiwiki in the
district of South Hilo, figs. 15, 16 (Thaanum).

This very beautiful form was thought by the late Mr. Bald-
win to be distinct from his A. haivaiiensis, but Mr. Thaanum
agrees with us that it may more justly be ranked as a sub-
species. It is separated by innumerable gulches and water
courses from the range of P. pliysa, some forty to sixty miles
distant, but the intervening territory has not yet been fully
searched ; still there can be no doubt that the range of errans
is discontinuous with that of physa. In Puna it has reached
the extreme southeastern range of Achatinellidce.

In Olaa Mr. Thaanum found P. p. errans in ieie heads,

49. P. GRISEA (Newcomb). PI. 18, figs. 8, 9.

" Shell sinistral, inflated below, pointed at the summit;
whorls 6, rounded, not margined. Aperture ovate ; columella
short, flat and but slightly twisted; lip expanded, thickened

112 PARTULINA, MAUL

within. Suture well impressed ; umbilicus open. Color above
grayish-white mottled with light brown, below ashy-gray ; the
body-whorl encircled by a narrow white band. Length 16,
diam. nine-twentieths of an inch ' (Neivc.).

Maui: Makawao (Newcomb, type loc.).

Achatinella grisea NEWC., P. Z. S., 1853, p. 153, pi. 24, L
66. PFR., Monogr., iv, 518. A. diibia var. & PFR., Malak.
Bl., 1854, p. 117.

The conic spire rather slender above, and the weak colu-
mellar fold give this shell somewhat the contour of A. physa
Nc. of Hawaii. After the smooth first whorl, two are very
closely and deeply striate spirally. The spirals become weaker
on the last whorl. \]/2 apical whorls are pale brownish white.
Then broad, irregularly zigzag brown stripes usually appear,
and on the last embryonic whorl a dense pattern of narrow
zigzag lines (pi. 18, fig. 8). After 3^ embryonic whorls the
neanic pattern of oblique streaks sets in. These become zig-
zag or irregular on the later whorls. The last whorl is per-
forate, narrowly banded with white, dark fawn color below
the band. In two specimens I have seen there is no band.
The aperture resembles that of P. divightii except that the lip
is far less thickened and the columella moderately twisted
with no callous fold superposed upon the convexity. Figures
are from Makawao shells received from Newcomb.

Length 20, diam. 11.5 mm.

Length 21.2, diam. 12.5 mm. ; whorls 6^3.

P. grisea is not closely related to any other M>auian snail,
but it has much in common with P. dwightii of Molokai, the
shape of the spire, pattern and color being nearly alike in
the two species. The pattern of the last embryonic whorl is
different, and P. divightii has a stronger columellar fold. It
is one of the rarest Mauian shells, and as yet entirely un-
known to Hawaiian conchologists of the present day, although
the Makawao district has been thoroughly worked over. It
may be extinct.

50. P. THAANUMIANA Pilsbrv, n. sp. PL 18, figs. 6, 7.
The shell is sinistral, narrowly umbilicate, tliin, ovate-conic,

Achalinellidee

PLATE 1O

11

r

\ |
^-.._ -

J6

Achatinellidse

PLATE

'

12

I?

j>

14

Achatinellidae

PLATE 15

,-,.

Achatinellidre

PLATE 16

1

M

I

\

.

'.

1

' *-'<&>-

o

. ///-/

: / >-,

-
::..

::.. ,

,..,,.., ^

7

1

I

1

i ' ^;T-

- -'/'' V ..'.'ACfefl

' I

--

1 >^
; Pi

;:/:. i

,

10

11

x-ftv : -L '

12

Achatinellidse

PLATE 17

6

10

11

j

3

13

15

Achatinellidse

PLATE 18

4

& - 1

:?\ " - .

IE

f :T Jft ".< ' / i<*ffi

/';,-;*

' ='*

6

10

1L

.

.

.--"

^

N.^-

14

15

.

<

17

Achatinellid]e

PLATE 19

f

9

L3

i

15

16

Achatinellid^e

PLATE 2O

n

Achatinellidae

PLATE 2 1

- - ".- wi- "*'.

. -..-.-;., ;,

15

Achatinellidae

PLATE 2

/ *

/ '

/""" *- ; -0

'

: .
~

T

^

-'

.:

1

;- ,.

/ ''

"-->.

-

... "* J -

/

.'

:

-.

~

''

r . . :

1

. -.,

1

. A

1 ^ JW

>

',.. !

/

13

14-

it
c

Achatinellidae

PLATE 23

I I'

3

>*

5

6

9

10

13

17

Achatinellidae

PLATE

\

10

I ">

Lo

-

I
..

15

-f p
) I

.

-

16

18

\

\

2.0

j

..^r

Achatinellidae

PLATE 25

5a

5b

7b

8

9b

11

lla

12

lie

13a

13b

lie

13c

Achatinellidae

Plate 2 6

3a

5b

5c

4a

5d

8

9a

13

5 f

9b

13 a

.

_^.

13 c

14

\ -i^

'

15

15 b

15c

15d

15 e

PARTULINA DUBIA, OAHU. 113

surface slightly shining, the later whorls densely striatei
spirally, strias waved, descending, embryonic whorls very
densely and distinctly engraved. Embryonic shell of 8^2
whorls, the first 2 whorls brown under a whitish layer; third
whorl marked with oblique, angulated white stripes on a
reddish-brown ground. Following (neanic) whorls mottled
and streaked with white on a reddish-brown ground, which
becomes paler on the last whorl, which is indistinctly streaked
and mottled with creamy or brownish white, and girdled with
a narrow dark band at the periphery. The whorls are
strongly convex. Aperture slightly oblique, purplish-brown,
with light streaks within. Peristome a trifle expanding at
the base, narrowly bordered within with white, but not thick-
ened. Golumella slightly convex, not plicate, the margin
spreading in a triangle, white.

Length 17, diam. 10.5 mm. ; fully 6 whorls.

Length 15.5, diam. 10.2 mm. ; 5^4 whorls.

West Maui: Waiehu Gulch (D. Thaanum). Gotypes in
coll. Bishop Museum and A. N. S. P. Also in Mr. Thaanum 's
collection at Hilo.

This very distinct species is evidently related to the East
Mauian P. grisea Newc., from which it differs by the thinner
shell of somewhat different contour, the more convex whorls,
different pattern of the last embryonic whorl, larger umbili-
cus, dark peripheral band, etc. The aperture, lip and colu-
mella are more like P. physa than like P. grisea. The spire
is less drawn out and not so narrow above. P. aptycha Pfr.,
which has not been found by recent collectors, is obviously
distinct.

51. P. DUBIA (Newcomb). PL 26, figs. 7 to 12.

" Shell dextral, conically elongate, thin, finely decussately
striated, light corneous with radiating zigzag lines and
blotches of a light color. Whorls 6 ; suture simple. Aperture
ovate; 'columella white and bulimoid except in strongly de-
veloped adults where it is callous and obtusely dentated.
Umbilicus small but pervious. Lip thickened. Slightly re-
flected. Length 0.8, breadth 0.55 inch ' (Newc.).

114 PARTULINA DUBIA, OAHU.

Oahu, among stones; Waianae, found on bushes (New-
comb) . Waimano, eastern ravines southwest of forest fence ;
eastern ravines of Waiawa; east side of stream, Kawaiha-
lona, in Waialua; also Makalia and Makua valleys, Waianae
range (Irwin Spalding). In knot holes and crevices of loose
bark of kukul trees. In Waimano on guavas.

Achatinella dubia NEWC., Ann. Lye. N. H. of N. Y., vi, p.
23 (May, 1853) ; P. Z. S., 1853, p. 152, pi. 24, f. 65, 1854.
Achatinella platystyla GULICK, Ann. Lye. N. H. of N. Y., vi,
p. 196, pi. 6, f. 25, 1856. Achatinella pexa GULICK, Ann.
Lye. N. H. of N. Y., vi, p. 197, pi. 6, f . 26, December, 1856.
Achatinella fucosa Frick, LYONS, Hawaiian Almanac and An-
nual for 1892, p. 105, pi. 1, f. 15. A. morbida PPB., P. Z. S.,
1859, p. 30.

P. dubia is a foreigner in Oahu ; the strong spiral striation
and the pattern of the embryonic whorls show it to be a real
Partulina. Like Laminella, it seems to have wandered west-
ward just before the connection with the Molokai-Maui mass
was submerged. It is somewhat related to P. grisea and P.
thaanumiana of Maui, P. radiata Gld. also has a close re-
semblance, but in this the Partulina pattern has been lost
from the embryonic whorls, and the columellar fold is some-
what stronger. P. radiata, according to specimens in Mr.
Emerson's collection, is from West Maui.

In his second account of this species Newcomb states that
it may be either dextral or sinistral, and that old specimens
exhibit " a flat, twisted tooth ". His type figure is copied,
pi. 24, fig. 9. This form has light girdles at periphery and
base, with elsewhere a confused zigzag pattern of purplish-
brown and white or cream. The narrow callus within the
lip is white. Length 19.5, diam. 11, aperture 9.8 mm., whorls
l / 2 . The smoothish and rather shining surface shows engraved
spiral striae throughout. The lower part of the outer lip and
the basal margin are very slightly expanded. Ten of the spe-
cimens of this typical form before me are dextral, one sinis-
tral. The last embryonic whorl has conspicuous zigzag white
stripes on a dark ground. PL 26, fig. 7, is a specimen re-
ceived from Newcomb. Other specimens, pi. 26, fig; 8, have

PABTULINA DUBIA, OAHU. 115

straight light yellow and brown streaks on the post-em-
bryonic whorls. The smallest adult seen is 16 mm,, long.

Probably all of Newcomb's shells, to which the above notes
apply, were from the Waianae range, where it has been found
in Makaha valley (on the southern side) by Mr. Spalding,
and perhaps elsewhere by other collectors. PI. 26, fig. 8, is a
Waianae range shell.

In the main range P. dubia has an extended distribution,
from Waimano to Waialua or Kawailoa, but only in a few
places, so that it is regarded as a rather rare shell. Very fine
series are in the collections of Messrs. Spalding, Emerson and
Wilder. As a general rule, main range shells have a more
thickened lip than those of Waianae, and the apex is often
perceptibly blunter, whorls a trifle more convex ; so far as I
know, no sinistral shells have been found ; but I am not satis-
fied that these small differences hold in all specimens.

In Waiawa (pi. 26, figs. 9, 9a, 9&, coll. by Spalding in the
eastern ravines) the shell is streaked and mottled, with boldly
zigzag-striped last embryonic whorl ; or similar with a whitish
peripheral band and several lines. Others are chestnut with
pale streaks, a white peripheral band and several spiral light
lines, the white zigzags of the embryo very much reduced
(pi. 26, fig. 9 6), or white may predominate, the chestnut
being reduced to bands narrower than the white, one speci-
men. In a few the later whorls are whitish with brown stains
(pi. 26, fig. 9).

An extraordinary series was collected by Mr. Spalding in
Waimano. Most of the colony is of the usual streaked and
mottled form, but in a few the color is very deep, rich brown
(between liver-brown or carob-brown and black, of Ridg-
way's Color Standards). This may be uniform or varied
with light buff bands. The embryonic whorls are brown with
light bands and lines, and on the last embryonic whorl there
is a trace of the light stripes of the typical form. The figures,
pi. 26, figs. 10, 10&,, are from two specimens of Mr. Spalding 's
no. 2181. This color-form is one of the rarest Oahuan trea
snails. A suspicion has been entertained that it might be a
hybrid between clubia, and a rare black form of Adi. turgida-,

116 PABTULINA DUBIA, OAHU.

but having examined all ol* the specimens found, I think the
evidence against that hypothesis.

Achatinella platystyla Gulick, pi. 26, fig. 11, and pi. 50, fig.
17, is a peculiar color-form of dubia, described from a single
shell. The surface is a good deal pitted, but it was evidently
a " live ' shell, and an old one. The surface is entirely
eroded to the middle of the fourth whorl. What cuticle re-
mains on the next two whorls is pecan-brown with creamy
streaks, and fading upwards. The last whorl is light buff in
the upper third, shading downwards into pecan-brown, which
is streaked with a lighter tint and has very indistinct spirals
of the darker shade. The color is almost exactly that of some
specimens of Achatinella glabra. The surface shows faint
spiral striation, which becomes stronger upwards, as usual in
P. dubia. The aperture, lip and perforation are exactly as
in Main Range dubia. The lip has a faint yellow tint, deeper
at the edge. Length 20, diam. 10.3, aperture 8.4 mm. ; 6^2
whorls.

The unique type of platystyla was collected by Mr. Gulick
in Kawailoa. It is no. 25 of 'his type collection, Boston So-
ciety of Natural History. I have given two views of this
specimen.

Achatinella pexa Gulick, pi. 26, fig. 12, is certainly a sinis-
tral specimen of A. dubia. The unique type specimen, no.
26 of Gulick 's type collection, coll. Boston Society of Natural
History, is figured. The shell has the texture and thin sub-
stance of dubia. The ground-tint is between light buff and
white. It appears in spirals and obliquely axial streaks, leav-
ing rather wide interrupted streaks of vinaceous pink, here
and there darker. There is a brown line at the periphery
and a group of three around the umbilical region. The early
whorls are eroded, but traces of alternate light and darker
stripes can be made out on the last embryonic whorl. The
glossy surface is engraved with rather distinct close spirals,
as in P. dubia. The lip expands a little. Aperture, lip and
perforation are as usual in P. dubia; the lip having a pale
yellowish-brown edge and slight thickening. Length 19, diam.
10.7, aperture 9 mm. ; 6^/2 whorls.

ACHATINELLA. 117

Another figure of the type of pexa is given, pi. 50, fig. 18,
to show the form of the columella more distinctly than the
figure on plate 26.

The pattern of color is practically the same as in some
Waiawa diibia, though developed in delicate tints; and the
specimen may have come from that neighborhood; yet it is
sinistral with a thin lip, characters more in harmony with the
Waianae range form. Gulick did not know the locality.

Achatinella morbida Pfeiffer. " Shell subperf orate, sinis-
tral, ovate-turrite, rather solid, striate and under a lens seen
to be decussated with close spiral striae; white, variously
streaked and banded with brown; spire long, slender, the
apex rather acute, suture simple; whorls 6y 2 , very slightly
convex, the last a little shorter than the spire, convex ; colu-
mellar fold white, short, oblique; aperture oblique, inverted
ear-shaped ; peristome thickened, narrowly expanded, the
eolumellar margin much dilated, broadly adnate. Length 19,
diam. 9 mm. Sandwich Islands, Dr. Frick in Mus. Cuming. '

(P/T.).

Achatinella morbida PFB., Proc. Zool. So<c. London, 1859,
p. 30 ; Monographia, vi, 167.

This species has not been figured. The description agrees
well with sinistral P. dubia, such as are found in the Waianae
Mountains; and a figure of the type-specimen which Mr. E.
A. Smith kindly had made for me (pi. 41, fig. 4) confirms this
determination. A. morbida is merely a somewhat unusual
color-pattern of P. dubia.

Genus ACHATINELLA Swainson.

Achatinella SWAINS., Quarterly Journal of Science, Liter-
ature and Art (2 ser.), iii, June, 1828, p. 83 ; type Monodonta
seminigra Lam. Achatinella apexfulva Dixon. GULICK,
P. Z. S., 1873, p. 90, classification. Cochlogena, quatrieme
groupe, Helicteres, FERUSSAC, Tableau Syst. de la Famille des
Limaqons, p. 56, 1821. Helicteres Fer., BECK, Index Mollus-
corum, p. 51, 1837. GRAY, P. Z. S., 1847, p. 178, type Helix
vulpina. HERRMANNSEN, Indicis Gen. Mai., i, p. 515, type
Helix vulpina Fer. Helicter PEASE, P. Z. S., 1862, p. 3.

118 BULIMELLA.

Apex VON MARTENS, Die Heliceen, 1860, p. 248, type Achati-
nella lugubris Chemn.

Shell imperforate or minutely perforate, oblong, ovate or
globose-conic, smooth or longitudinally corrugated, with only
minute and weak traces of spiral sculpture; color in spiral
bands or streaks in the direction of growth-lines, never in
forwardly-descending stripes; lip simple or thickened within,
sometimes slightly expanding ; columella bearing a strong cal-
lous fold.

Type A. apexfulva (Dixon). Distribution, the island of
Oahu. Living on the leaves and limbs of trees and bushes.

Data on the soft anatomy, distribution, etc., of Achatinella
may be found in the Introduction of this volume.

Achatinella was early divided into several subgenera or sec-
tions, but their contents as seen in the lists in Die Heliceen^
Nomenclator Heliceorum Viventium and other works were
rather heterogeneous. Gulick and Baldwin grouped the
species more naturally, practically limiting the groups as in
this work. Three sections or subgenera are recognized. There
are a few species on the border between Bulimella and Acha-
tinellastrum, but otherwise the sections are. quite distinct,
though they can scarcely be rigidly diagnosed.

Sections of Achatinella.

Bulimella Pfr. Outer lip thickened by a strong callous rib
within; shape oblong-conic or ovate; summit obtuse, rounded
or convexly-conic.

Achatinellastrum Pfr. Outer lip thin or a little thickened
within, not expanded ; early whorls not conspicuously flat-
tened, the summit conic, not very obtuse.

Achatinella s. sir. Outer lip well thickened within, scarcely
or not expanded ; shape globose-conic or ovate-conic, the sum-
mit conic, embryonic whorls flattened.

Section BULIMELLA Pfeiffer.

Bulimella PFR., Malakozoologische Blatter, i, 1854, p. 119.
MARTENS, Die Heliceen, 1860, p. 244, type A. rugosa Newc.

Achatinellse with the shell ovate-conic or somewhat pyra-

BULIMELLA. 119

midal, -the 'Outlines of the spire more or less convex, at least
near the rather obtuse summit; embryonic whorls not con-
spicuously flatter than those following; lip distinctly thick-
ened by an internal callous rib (except in A. abbreviata and
A. lila) , the outer edge often slightly expanded.

Type: A. byronii rugosa Nc. Distribution: Entire Koolau
or Main range of Oaliu.

About forty-six names, most of them originally introduced
as "species", have been applied to members of the Bulimella
group up to this time. Mr. Baldwin, in his Catalogue of
1893, enumerated twenty- five species. Mr. E. R. Sykes
(1900) admitted twenty species and three varieties. "In the
present monograph, thirteen species and eighteen subspecies
are recognized ; but this number includes two species and ten
subspecies not before published. Most of the other names are
synonyms, but a few may be retained for local forms thought
to be below the gra.de of "subspecies". A more rigorously
logical treatment might reduce the "good species' to nine,
by the union of viridans and tceniolata; byronii, decipiens and
pulckerrima; and bulimoides and elegans; but the price of
such consistency would, I imagine, be the loss of clear con-
ceptions.

Bulimella is a shell of the ravines as well as of the ridges,
and before the extensive deforestation of the lower slopes,
many fine species were to be found low in the valleys. This is
especially true of the bulimoides group, which formerly occu-
pied much territory now quite -barren. The fuscobasis group
is almost confined to high ridges and peaks.

The greatest development of Bulimella is- in the northwest-
ern half of the Koolau range. There are but four species and
two or three subspecies in the whole southeastern third or
more of the length of the island, leaving nine species and
fifteen subspecies in the western five-eighths of the length.
Moreover, the differentiation has been greater in the west,
and doubtless new forms remain to be found there, whilst the
eastern part of the range has been so fully explored that no
new forms of Bulimella can be expected.

Bulimella contains the most roughly sculptured species of

120 ACHATINELLA, SECTION BULIMELLA.

the genus, some forms being irregularly corrugated in the
direction of the lines of growth. Spiral sculpture is weak or
obsolete in all the species.

There are three collateral groups or series of species in
Bulimella, the first subdivided as a convenient geographic di-
vision.

Series of A. byronii: shell oblong-conic, often -with streaks
or bands of dark green, and often roughly striate or corru-
gated; mainly dextral. A. byronii, pidcherrima, decipiens,
lila.

Series of A. viridans: a subdivision of the above series, fo-r
the exclusively dextral species of the eastern end of the range.
A. abbreviata, viridans, and tceniolata.

Series of A. bidimoides: shell ovate, capacious, never
marked with green; smooth, dextral or sinistral. A. rosea,
bidimoides, elegans.

Series of A. fuscobasis: shell ovate-come, never marked with,
green; smooth; mostly sinistral. A. fuscobasis, sowerbyana,
pupukanioe.

On account of the wide variation in shape and color, it is
not possible to rigidly diagnose these groups, which never-
theless express the natural affinities of the species.

Key to species and most subspecies of Bulimella.

I. Forms of the southeastern third of the main range (east of

Mjoanahia) .

a. Lip acute, but slightly or not thickened within ; shell dex-
tral, very smooth, richly colored. A. abbreviala, no. 1.
a 1 . Lip strengthened by a callous rib.
b. Shell dextral.

c. Streaked with green or chestnut on a lighter or

yellow ground. A. viridans, no. 2.

c 1 . Banded or streaked with chestnut on a white

ground, or white. A. tceniolala, no. 3.

6 1 . Shell sinistral.

c. Shell white with few brown bands or none, about
16 x 10 mm. A. fuscobasis, no. 11.

c 1 . Similar but larger and more capacious.

A. f. lyonsiana, no. lla.

ACHATINELLA, SECTION BULIMELLA. 121

c 2 . Shell richly colored, banded. Lanihuli.

A. f. wilderi, no. 116.

II. Forms of the northwestern two-thirds of the main range.
a. Lip acute, but slightly or not thickened within ; shell sin-
istral, very smooth, richly colored. A. lila, no. 5.
a 1 . Lip distinctly thickened within.
6. Shell oblong-conic.

c. Surface corrugated or somewhat roughened by

growth- wrinkles.

d. Interior drainage ; shell sinistral, later whorls
corrugated, intensely black or banded with
dark green. A. &. nigricans, no. 46.

d 1 . Interior drainage; shell dextral.

e. Roughly striate, streaked with green, chest-
nut or black, usually with sutural chest-
nut band or others; lip edged with
brown. A. byronii, no. 4.

e 1 . More corrugated, occasionally black; sur-
face with a gummy gloss ; whole lip usu-
ally brownish ; length 16-19 mm.

A. 6. rugosa, no. 4&.

e 2 . Larger, broader, green-streaked; lip white

or narrowly dark-edged; length 19-21

mm. Var. capax.

e s . Nearly smooth, whitish to greenish-yellow,

plain or banded, lip white or pale lilac ;

length 15-18 mm. Var. waimanoensis.

d 2 . Northern (Koolauloa) slope; shell sinistral or

dextral, the lip mainly white.

A. decipiens, no. 7.
c 1 . Surface smooth ; shell dextral.

d. Northern slope ; lip white. Various varieties of

A. decipiens, no. 7, la.
d 1 . Interior slope.

e. Lip bordered with blackish brown.

A. pulclierrima, no. 6.
e 1 . Lip white throughout. A. p. nympha, no. 6a.

122 ACHATINELLA, SECTION BULJMELLA.

6 1 . Shell ovate, smooth.

c. Larger, capacious species, diam. 10 mm. or more.
d. Shell sinistra.l, highly polished, usually white
or with rose bands, lip bright rose ; rarely
greenish or dark, with dark brown lip.

A. rosea, no. 8.

d 1 . Dextral or sinistral, variously banded with
chestnut or suffused, lip pale or brown.

A. bulimoides, no. 9.

d 2 . Dextral or sinistral, conspicuously streaked
with brownish, sometimes with white bands
also, lip brown. A. elegans, no. 10.

d 3 . Chiefly sinistral, blue-gray or purplish, some-
times banded. A. e. wheatleyana, no. 10&.
c 1 . Smaller species, usually less than 10 mm. diam.
d. Shell sinistral.

e. Wax-yellow with a brown sutural band,

sometimes one or two other bands, colu-

mella pink. A. sowerbyana, no. 13.

e 1 . Wax-yellow below, white above periphery,

with a subperipheral brown band; colu-

mella pink. A. s. thurstoni, no. 13a.

e 2 . "White, usually with broad chestnut bands;

columellar fold white or brownish.

A. s. laiensis, no. 13&.
d 1 . Shell dextral.

e. Wax-yellow with white sutural band, colu-

mella pink. A. s. roseoplica, no. 13d.

e 1 . Ground white or pale, uniform, or banded

or streaked with chestnut.
A. pupukanioe, no. 12. A. s. dextroversa, no. 13c.

SERIES OP A. VIRIDANS.

(Species of the east end of the Main Range.)

These forms, which are invariably dextral, inhabit the val-
leys from Nuuanu to the southeastern end of the range. Be-
tween Xuuanu and Waimanalu there are a.bout half a dozen
great ridges and valleys where no BuUmella is known except
A. fuscoltasis irilderi on some high points of the main axis.

ACHATINELLA ABBBEVIATA. 123

The viridans group is somewhat intermediate between the
byronii and the bulimoides series, but most nearly related to
the former by the frequent development of bright green
color, which none of the bulimoides series has. Besides the
following forms, one unrelated sinistral species occurs on the
southeastern ridges, A. fuscobasis, no. 11.

1. A. ABBBEVIATA Reeve. PL 31, figs. 1 to 5a.

' ' Shell ovate, somewhat ventricose, dextral ; whorls convex,
margined round the upper part ; spire rather short, obtuse at
the apex ; columella callous, twisted. Olive-yellow, with a
black-brown line at the sutures, lower part of the last whorl
very dark green, apex black.' (Reeve.)

Oahu: Eastern ridge of Nuuanu to Niu. Palolo (Gulick
and others; type loc.) ; around edge of the crater, and Palolo-
Waialae ridge (Spalding) ; northwestern ravine, next to the
Palolo-Manoa ridge (Pilsbry & Cooke). Konahuanui (Bald-
win) ; eastern Nuuanu, perhaps washed down from Kona-
huanui ( Spalding) . W'aialae, Niu, and over the ridge in Kai-
lua ( Gulick) .

Achatinella abbreviata REEVE, Conch. Icon., vi, pi. 3, fig.
19, April, 1850. PFR., Monogr., iii, 460; iv, 532; vi, 172.
BALDWIN, Catalogue, p. 5. SYKES, Fauna Hawaiiensis, p. 305.
THWING, Orig. Descr. Achat,, p. 81, pi. 1, f. 22. GULICK,
Evolution, Racial and Habitudinal, p. 41, pi. 2, f. 25. Acha-
tinella bacca Reeve, Conch. Icon., vi, pi. 6, f. 45, May, 1850.
GULICK, Evolution, Racial and Habitudinal, p. 41, pi. 2, f . 22.
Achatinella nivosa NEWC., P. Z. S., 1853, p. 132, pi. 22, f. 6
(Niu). Achatinella Clementina PFR., P. Z. S., 1855, p. 205;
Monogr., iv, 534.

A. abbreviata differs from the viridans forms by its broad
contour, thin outer lip, smoother surface and different pat-
tern, yet is related to them by color and the faint traces of
corrugation noticeable in some specimens. It is a very dis-
tinct species.

A. abbreviata is a common species around the crater of
Palolo valley, and on its bounding ridges near the heads of the
upper ravines. It lives by preference in the axils of ieie. Hot

124 ACHATINELLA ABBREVIATA.

water, or too prolonged soaking, changes the vivid green color
-to a dirty olive-brown. The western extreme of its range is
(or was) in eastern Nuuanu, near the southeastern edge of
the new dam, where Mr. Spalding found five immature speci-
mens of the profusely-banded form (no. 1817 of his collec-
tion), possibly washed down from Konahuanui.

PI. 31, figs. 1, la, 5, 5a ? Palolo, are from Gulick shells. The
others, figs. 2 to 4<i, are from material collected recently. A
lot from the northwestern ravine of Palolo varies from many-
banded to black (pi. 31, figs. 4, 4a-4d). There was also one
shell all green except the brown summit. Figs. 3&, 4<i, 5 and
5a are rare color-forms.

The shell is obesely-ovate, very glossy, showing faint spiral
striation on the lower, clear but very fine incised lines on the
early whorls. The lip is acute, not thickened within or but
slightly so. All I have seen are dextral.

1. The typical color-pattern as figured by Reeve is yellow
above the periphery, green below it, the colors separated by a
peripheral chocolate band. The suture is bordered above and
below by chocolate bands. Near the summit the yellow ground
gives place to reddish-brown, becoming dusky or purplish-
brown at the apex (pi. 31, fig. 2). This pattern occurs in
Palolo valley together with the following, the inter gradation
being complete throughout the series.

2. Similar to no. 1, but having additional chocolate lines
and bands (pi. 31, fig. 2a, Palolo; figs. 3, 3, Palolo- Waialae
ridge), which sometimes almost cover the yellow and green,
ground.

3. Same as no. 2, but without green (pi. 31, figs. 4, 4&), or
the bands may be confluent, leaving yellow bands at periph-
ery, suture or both (fig. 46).

4. Last whorl black or nearly so, landless, the color becom-
ing lighter upward, dark again at the summit (pi. 31, figs. 4c,
4d) . This is the form called A. bacca by Reeve. A. abbre-
viata and A. bacca are therefore merely selected patterns in a
continuous series of variations occurring in the same colony.
Figs. 4 to 4:d are from a northwestern ravine of Palolo, next
to the Manoa ridge, coll. by Pilsbry.

ACHATINELLA VIRIDANS. 125

5. Shell yellow above (varying to olivaceous or brownish-
yellow) , green, dull olive or even chestnut below the periphery,
usually bandless, sometimes having one or several reddish-
brown bands. The columellar fold is not so strong as in the
preceding. (PI. 31, figs. 1, la, Palolo.) So far as I know,
this form does not occur in colonies of the chocolate-banded
forms. There is a large lot from Palolo in the Gulick collec-
tion, 92572 A. N. S.

Waialae. Several lots collected by Gulick and others have
the patterns of figs. 2a, 3, 3a, 4&, c, d, not distinguishable
from Palolo shells. Another lot agrees with fig. 1. PL 31,
figs. 3, 3a are from the Palolo- Waialae ridge, collected by Mr.
Kuhns.

Wailupe. Occurrence doubtful.

Niu. Many-banded shells, pattern of fig. 3, collected by Gu-
lick.

The locality "head of Kawailoa Gulch, Perkins," given by
Mr. Sykes, must be an error in labeling or determination.

A. nivosa Newc. is universally admitted to be a scraped
abbreviata. The original figure is reproduced, pi. 29, fig. 6.

A. dementina Pf r. has not been figured. It has been placed
in the synonymy of (Mreviata by Pfeiffer himself and by sub-
sequent authors. The description, which is not very good for
dbbreviata, here follows: " A. dementina. Shell imperforate,
dextral, ovate-conic, solid, nearly smooth, glossy; greenish-
brown, banded below the suture with buff. Spire regularly
conic, obtuse, suture chestnut, impressed-marginate, crenu-
lated. Whorls 6, a little swollen above, the last about equal
to two-fifths the total length. Aperture oblique, truncate ear-
shaped, white within; columellar fold above, moderate, white
or flesh-colored. Peristome unexpanded, the margins joined
by a callous, right margin somewhat straightened, thickened-
labiate within; columellar margin dilated, adnate. Length
19, cliam. 10, aperture 9% mm. long, 4% wide. Oahu, Frick
in Cuming coll.' (Pfr.)

2. A. VIRIDANS Mighels. PI. 25, figs. 1 to 4 ; pi. 31, figs. 6, 6a.
11 'Shell dextral, elongate-conic, green with light streaks in-

126 ACHATINELLA VIRIDANS.

termixed, imperforate. Whorls 5, convex, with a revolving,
slightly impressed line .below the suture. Aperture subovate,
stained with a pink color just within the margin ; lip slightly
thickened. Length three-fourths, diam. seven-fifteenths inch' :
(Mighels).

Oaliu (Mighels). Nuuanu to Palolo, various color-forms as
far east as Niu.

Achatinella viridans MIGHELS, Proc. Boston Soc. Nat. Hist.,
ii, p. 20, January, 1845. PFR., Monographia Hel. Viv., ii,
240; iv, 519; vi, 164; Conchyl. Cabinet, Bui, p. 286, pi. 67,
f . 16, 17. NEWCOMB, P. Z. S., 1854, p. 310 (animal) . Acha-
tinella radiata PFR, ? P. Z. S., 1845, p. 89 (January, 1846).
REEVE, Oonch. Icon., pi. 5, f. 35.

Achatinella subvirens NEWC., P. Z. S., 1853, p. 136, pi. 22,
f. 18, 1854. Achatinella rutila NEWC., P. Z. S., 1853, p. 138,
pi. 22, f. 21, 1854; p. 310 (animal) ; Ann. Lye. N. H. of N. Y.,
vi, 326 (animal). Achaiinella macrostoma PFR., P. Z. S.,
1855, p. 2, pi. 30, f. 6; Monogr., iv, 520 (Sandwich Is., Frick
in Cuming coll.).

The type of A. viridans was lost by fire, but the descrip-
tion applies well to the form found in Nuuanu, Manoa and
Palolo valleys, and which has always been considered viridans
by Hawaiian collectors. C. B. Adams, who had seen Mighels '
specimen in his collection, stated that it was identical with A.
radiata Pfr. as figured by Reeve. The traditional identifica-
tion of viridans is therefore confirmed, that figure being the
same as our pi. 25, fig. 1. The shell is glossy, noticeably,
though weakly, corrugated. The first three whorls are whitish,
or 'white when the thin cuticle is lost ; next whorl or two yellow
or olivaceous, more or less streaked; last 1% or 2 whorls
copiously streaked throughout with dark and light green of
several shades. The sutural border is distinctly marked by
an impressed line. The aperture is white within, the lip a
trifle expanded, thickened within, and either white or flesh-
pink. In a large number of specimens seen, not one is sinis-
tral. The size and form vary, but shells 20 mm. long, 11 wide
are about normal. A wide shell from Nuuanu measures,
length 20.2, diam. 13 mm.

ACHATINELLA VIRIDANS. 127

In Nuuanu valley the shells vary from typical green color
to a form in which the dark green is replaced by blackish-
chestnut, light green by yellow (pi. 25, fig. 4). A specimen
of the rather widely-spread form of Upper Nuuanu, pale
yellow with light yellowish olive streaks, is figured on pi. 31,
fig. 6o> coll. by Mr. Richard A. Oooke. A small, thin-shelled
form was found only on a few isolated bushes at the head of
the valley just west of the pali. It is chestnut-brown with
lighter streaks and pale lines below suture and periphery;
lip-callus narrow; length 15.8, diam. 8 mm. PI. 31, fig. 6,
coll. by R. A. Cooke.

In Manoa Gulick collected the typical green form (pi. 25,
figs. 1, la) ; others having black, others olive (pi. 25, fig. 3)
or chestnut streaks on a yellow ground. There is now no
forest low in Manoa valley.

Palolo has the same range of forms shown in figs. 1 to 3.
A black-streaked shell is figured (pi. 25, fig. 2, collected by
Gulick). Really typical viridans does not go eastward of
Palolo, so far as I know. It is a rather homogeneous race,
having the same pattern but in varying shades and colors.
Probably green, olive, chestnut and black are varying stages
of oxidation of the same pigment; or perhaps the dull ground
of some specimens may be due to cleaning with hot water.
In Palolo the forms rutila and subvirens also occur, whether
associated with the typical viridans pattern I do not know.

Doctor Newcomb has given the following description of the
soft parts. A. viridans: ' ' Animal light gray; tentacles and
tentacular sheath dark slate; mantle thick, yellowish-brown.
Tentacles strongly clubbed, short and robust, when extended,
longer than the shell.'

A. rutila: " Animal small in proportion to the shell, of a
uniform yellowish-white, retractile part of upper tentacles of
a light-brown ; tentacles filiform and slightly clubbed ; foot
very broad, long as the shell ; mantle same color as the ani-
mal. ' '

Eastward of Palolo the typical viridans pattern disappears,
and the color-forms rutila and subvirens replace it. These are
often hybridized with more or less blending, so that the ap-

128 ACHATINELLA VIRIDANS.

pearance of a colony suggests rutila X subvirens or rutila X
tceniolata. It might be well to recognize siibvirens as an east-
ern subspecies of viridans, rutila and macrostoma to be syn-
onyms of it.

Color form rutila Newcoonb. PI. 25, figs. 10, 11, lla-e, 12.

' ' Shell ovately-conic ; whorls 6, rounded, the last margined
above; suture well marked; aperture subquadrate; lip ex-
panded, subreflected, strongly thickened within; columella
short, terminating in a strong, twisted plait. Color a light
straw, olive or brown ; lip white or somewhat roseate. Length
17, diam. nine-twentieths inch.' (Newc.)

Niu (Newcomb, type loc., pi. 25, figs. 10, 11 to lie. The
color-form figured by Newcomb as typical is streaked with
light green on a yellow-green ground, with several spiral
brown bands below the periphery (pi. 29, fig. 21, reproduction
of Newcomb 's type figure ; pi. 25, fig. 10, specimen from New-
comb) . In other shells from Niu, collected by Gulick, figs. 11
to lie, the spiral bands may be more numerous or they may
be wanting. The streaks are sometimes chestnut or umber,
and either distinct or blended. The sutural border is almost
invariably tessellated with brown, thus differing from var.
subvirens, in which it is white, or at least not more heavily
marked than the rest of the shell. Wailupe shells are similar
(pi. 25, fig. 12).

In Waialae valley the color is usually light, and specimens
having bands like rutila with the white suture of subvirens
were found by Gulick and kept by him in the same lot with
others having the tessellated suture of rutila. See pi. 25, figs.
5 to 5d. There is no definite break between the two forms,
merely a matter of one or another color-mutation prevailing
in the colony. Even as far west as Palolo there are some
shells with more or less tessellated sutural border.

A. macrostoma Pfr. seems to me to be a form of rutila, as
Newcomb held, specimens of similar pattern to Pfeiffer's fig-
ured type occurring in Waialae Nui (pi. 25, fig. 8, Cooke
coll.). The original figure is reproduced, pi. 30, fig. 6. Mr.
Sykes has thought it a synonym of t&niolata,, and it must

ACHATINELLA VIRIDANS. 129

be admitted that some of the yellow-ground specimens from
colonies presumably t&niolata X subvirens have a great re-
semblance to Pfeiffer's figure. The original description fol-
lows. ' A. macrostoma Pfr. Shell dextral, imperforate,
conic-ovate, rather solid, very lightly striated, glossy ; fulvous,
variegated with some pale and brown bands. Spire conic,
lather obtuse; suture submarginate; whorls 5%, moderately
convex, the last equal to the spire, swollen above, 'contracted
in the middle, rounded at base. Aperture slightly oblique,
ample, reversed auriform, white within; peristome white,
labiate within, the right margin expanded, strongly curved
above ; columellar margin dilated, adnate. Length 211/2, diam.
11 mm. ; 'aperture 12 mm. long, 5 wide inside. Inhabits the
Sandwich Islands, Frick; Mus. Cuming (Pfr.).

Color-form snbvirens Newc. PI. 25, figs. 5 to lit, 9 to 95.

" Shell 'conically ovate. Whorls 6, rounded and margined
above ; suture distinct and lined with white. Aperture ovate,
expanded below; lip thickened; coluniella short, flat and ob-
liquely truncated ; color of epidermis light green, interspersed
with a lighter shade 'arranged longitudinally; columella, lip
and aperture white. Length 15, diam. seven-twentieths of an
inch. Niu, Oahu.

" Var. a. Pure white.

" Var. 5. Brown or chestnut replacing the green color.

" Through var. a- this species approaches a variety of A.
rutila, and through var. ~b, A. decipiens. It is readily dis-
tinguished from the first by its smaller size, greater solidity,
stronger striae and more elongate form ; from the latter by its
less acuminate form, white suture, less solidity, and their
widely separated localities ' (Newcomb).

Newcomb's figure of the type from Niu is reproduced, pi.
29, fig. 18. The lip is white or yellowish, the green color is
paler than in viridans, surface less wrinkled, and typically
the sutural margin is white, though very often the dark lines
extend over it. It differs from viridans in the average, but
one could not pronounce on every specimen, and if Gulick's
lots are reliable, it occurs in the same colonies with rutila.

130 ACHATINELLA T^NIOLATA.

The same variety occurs in Wailupe and Waialae valleys. A
series is figured, pi. 25, figs. 5 to 5d, Waialae ; also pi. 25, fig.
6, a clear green shell occurring with the brown-streaked form,
Wailupe. Some shells from Waialae have spiral bands as in
var. rutila (pi. 25, figs. 7 to 76).

Gulick found a few subvirens also in Palolo, the white,
green-streaked and brown-streaked forms. Also specimens
connecting with form rutila, the sutural border yellow in
some, white in other examples. Color yellow with indistinct
olive lines, bandless or with white or chestnut bands (pi. 25,
25, figs. 9, 9a, 96, Palolo) .

3. A. T^CNIOLATA Pf eiffer. PI. 24, figs. 14 to 19 ; pi. 25, figs.
13 to 13c.

" Shell ovate-oblong, solid, striatulate, glossy; white orna-
mented with varying brown bands, more obsolete above. Spire
conic, rather acute. Whorls 6, slightly convex ; the last about
four-ninths the length. Columella white, strongly tooth-
folded above. Aperture irregularly semi oval, white within,
glossy; peristome very narrowly thickened outside, strongly
lipped within, the columellar margin dilated, reflexed, ap-
pressed. Length 20, diam. in the middle 11 mm. ; aperture
10x4!/2 mm. Sandwich Is., Mus. Cuming.' (P/V.)

Oahu : Palolo to Keawaawa, and northward across the range
in Maunawili.

Achatinella t&niolata PFR., P. Z. S., 1846, p. 38; Mono-
graphia, ii, 239 ; iv, 519 ; vi, 164. REEVE, Conch. Icon., pi. 1,
f. 7. NEWCOMB, Ann. Lye. N. H. of N. Y., vi, p. 320. SYKES,
Fauna Hawaiiensis, p. 310. Achatinella rubiginosa NEW-
COMB, P. Z. S., 1853, p. 154, pi. 24, f. 69 (Palolo).

A. taniolata typically has a white ground, variously banded
or streaked with rufous or ochraceous-orange, or without
markings. The contour is usually somewhat more capacious
and the surface smoother than in the forms subvirens and
rutila of A. viridans; but both contours and sculpture inter-
grade in some shells. The coloration is characteristic in typi-
cal examples, but here again there are colonies from Palolo,
AVaialae, Keawaawa and elsewhere, which could be placed,

ACHATINELLA T^NIOLATA. 131

with .almost equal propriety in tceniolata, viridans rutila or
viridans subvirens, or 'assorted into these seve'ral strains.
The fact seems to 'be that several forms, well differentiated
in pure colonies, have interbred in other places, forming hy-
brid colonies. Strictly speaking, tceniolata is a subspecies of
A. viridans, because there is complete intergradation between
them; but as the race is usually quite recognizable, it is here
admitted as a species. In dealing with Achatinella it is im-
possible to be logical and at the same time preserve a practi-
cally useful classification.

The color in Palolo shells varies from white to few-banded
or many-banded with light reddish-brown, apex and sutural
border white, o>r rarely the apex is purplish-brown, perhaps
stained from within (pi. 24, figs. 15, 16, Pialolo, Gulick coll.).

Length 20, diam. 10.2 to 11.2 mm.

Length 18, diam. 11.5 mm. (exceptionally obese).

Length 21, diam. 10.8 mm.

Length 21, diam. 11.8 mm.

As these shells agree well with Pfeiffer's description and
Reeve's figure, Palolo may be taken as type locality.

In some colonies the color-form nibiginosa Newc. (pi. 24,
fig. 19, Palolo, and pi. 25, figs. 13, 13&, Wailupe, all from
Gulick coll.) is prevalent, together with banded shells. In
this form the shell is suffused and streaked with rufous, a
band or line below the suture and a columellar patch usually
white. Typical t&niolata might be described as rubiginosa in
which white bands traverse the shell spirally, cutting the
dark color into bands. It has thus a more advanced pattern
than that of rubiginosa, and the white shells are still further
evolved, though the pattern is degenerate.

According to Newcomb, riibiginosa has a light flesh-colored,
and tceniolata a brown or dark slate-colored animal and
mantle. In his later paper he admits the specific identity of
rubiginosa and tceniolata.

In Waialae (pi. 24, fig. 14) and Wailupe some large lots
taken by Gulick are white or with but few spiral lines or none,
while in others the profusely banded and the rubiginosa type
of coloring prevails. PL 25, figs. 13 to 13c are Wailupe shells

132 ACHATINELLA TJSNIOLATA.

collected by Gulick. A recently collected Waialae lot from
Thaanum consists of (a) white shells with brown spire, (b)
white shells with numerous brown bands, as in Palolo tanio-
lata, and (c) greenish-yellow shells with chestnut-olive streaks,
sometimes cut by white bands. There are various blends be-
tween a and c patterns. This colony is probably a siibvirens
X t&niolata hybrid.

A small lot from Niu (Gulick) consists of well-colored
banded shells (t&niolata) , or streaked (rubiginosa) , with
white ground.

From Kuliouou (Thaanum) the specimens have the pattern
of Keawaawa form (b), see below.

In a set of 9 from Keawaawa, coll. by Gulick, some have a
few faint bands, others being white, peristome and parietal
wall yellowish. The shape varies, as in Palolo shells : length
19, diam. 10 mm., to length 17, diam. 10^2 mm. A lot from
the Thaanum collection, recently collected, there are (a)
shells variously banded with chestnut on a yellow ground; (b)
the same with many oblique streaks or lines added, and (c)
others of rubiginosa pattern, the streaks 'blended. The yellow
ground shows probable hybridism with rutila or siibvirens.

On the northern side of the main range Gulick obtained a
few specimens in Kailua rather an indefinite locality. They
have streaked rubiginosa coloring, with paler spiral bands.
One has a greenish tone. They are transitional between rubi-
ginosa, t&niolata and siibvirens. Size rather small, length 17
mm. It occurs also in Maunawili, on the northeastern slope
of Mt. Olympus, a place exploited by Messrs. Kuhns and
Wilder. Here the shells are white with dark or light-brown 1
upper whorls, paler (sometimes white) embryonic whorls, the
last whorl either banded or with rubiginosa pattern cut by one
or more white bands. PL 24, figs. 17, 18.

(Species of the northwestern two-thirds of the Main Range.)

SERIES OF A. BYRONII.

What forms of the byronii group may occur upon the crest
of the main axis of the range 'between the known a>reas of
byronii and decipiens are not known to me. For this reason,

ACHATINELLA BYRONII. 133

and because there are differences in 'the patterns and colors,
I am recognizing the forms from the northward- side, collec-
tively, as a species (A. decipiens} distinct from those on the
south side of the range. The decipiens forms 'all appear to
be more closely related to one another than any of them are
to forms from the other side of the range, They seem to form
a parallel series to the latter. In both series there are smooth
forms in the west, corrugation becoming progressively more
emphatic eastward. In both series the western forms are in-
variably dextral, and sinistral forms occur in the east. This
correspondence would suggest communication across the
range, but the coloration is against that explanation of the
parallelism. From northwest to southeast the forms are ar-
ranged thus: smooth, dextral forms above (westward), rough,
sinistral forms below (eastward).

kaliuwaaensis

pulcherrima

byronii decipiens

rugosa corrugata, torrida

nigricans

Dr. C. Montague Cooke considers decipiens with its varieties
to be a subspecies of 'byronii. In retaining A. pulcherrima as
a species, I am following Dr. Cooke 's counsel, although I have
some doubt whether it should be given higher rank than a sub-
species of A. 'byronii. It is a case where there are numerous
local forms the variations of which overlap more or less, and
whether they 'are ranked as forms of one species, or are more
or less arbitrarily -assorted into several, is a matter of conveni-
ence in referring to the forms. The series shows such diver-
sity of differentiation that it would be rather meaningless to
lump all of the races under the one name, A. byronii.

Practically all of the Newcomb 'and Grulick shells were taken
at low levels, 'chiefly in places where there are now no forests.
The modern collector in this part of Oahu rarely finds tree-
snails below the forest fence, which follows the 1,000 ft. con-
tour from Waimea to Moanalua.

4. A. BYRONII (Wood). PL 27, figs. 1 to le, 3.

The shell is dextral, imperf orate ; pyramidal-conic with ob-

134 ACHATINELLA BYRONII.

tuse summit, solid, glossy. Color variable, but typically green
and light greenish-yellow in oblique streaks on the last two
whorls, having a faint green peripheral band and a dark
chestnut band bordering the suture below ; next earlier whorl
yellow with a chestnut band, nearly three embryonic whorls
pinkish gray; aperture white, the lip bordered with dark
brown. Sculpture of faint spiral strias on the embryonic
whorls, later whorls irregularly wrinkled in the direction of
growth-lines, under the lens showing very faint traces of a
fine oblique malleation. Whorls 6%, somewhat convex, the
last often very obtusely angular at the periphery. Aperture
strongly oblique, the lip thickened within by a strong rib near
the margin. Columellar fold moderate, white or tinted.

Length 20, diam. 11 mm. (typical size).

Length 18, diam. 11 mm.

Length 17, diam. 9% mm.

Oahu: Kalaikoa, Ahonui (Gulick).

Helix byronii WOOD, Supplement to the Index Testaceo-
logicus, or a Catalogue of Shells, British and Foreign, 1828,
p. 22, no. 30, pi. 7, f. 30. Acnatinella byronii Wood, NEW-
COMB, Ann. Lye. N. H. of N. Y., vi, p. 318. PFR., Monogr.,
vi, 165. SYKES, Fauna Hawaiiensis, p. 306. Achatinella me-
Unostoma NEWCOMB, P. Z. S., 1853, p. 132, pi. 22, f. 7 (1854).
PFR., Monogr., iv, 521. Achatinella limb at a GULICK, Ann.
Lye., vi, 252, pi. 8, TOa, 705 (Feb., 1858). Helicteres byron-
ensis (Gray), BECK, Index Mjolluscorum, 1837, p. 51.

A. byronii, in its southeastern forms, is closely related to
A. viridans and A. decipiens. All have more or less corru-
gated varieties, and others almost smooth.

The type of A. ~byronii was denned only by a figure, but
this is of such excellence that no doubt of its identity can be
entertained. The shell was presumably brought to England
by Lord Byron, whose mission to Oahu in 1824 has been noted
in Vol. XXI, p. 157. All of his AcliatineUidcc were from near
the west end of the main range, Wahiawa to Kawailoa. The
present species does not occur in Kawailo'a district, but it is
to be found at Kalaikoa and Ahonui, places in Wahiawa dis-
trict, a few miles southward. I select Ahonui as the type

ACHATINELLA BYRONII. 135

locality, pi. 27, figs. 1, la representing typical shells. The
chestnut subsutural band is invariable in Ahonuj. and Kalai-
koa shells.

In a lot of 68 from Ahonui, Gulick coll., the colors are as
follows :

1. Typical, green-streaked (pi. 27, fig's. 1, 1), 16 shells.

2. Yellowish or olivaceous tawny, clear or more or less
roughened and streaked -with black (figs. 1&, Ic), 18 shells.

3. Like 1 or 2, but having spiral bands, at periphery and
midway between sutures, sometimes not extending to the last
whorl (fig. Id), 33 shells.

4. One shell (fig. le) has a peripheral band only.

The above division is somewhat arbitrary, as the patterns
and colors blend in some specimens.

In Kalaikoa a large series from Gulick contains no bright-
green shells. Olive-green, with pattern no. 1 (above), and
pattern no. 2 and no. 3 predominate. There are a few dark
chestnut shells (pi. 27, fig. 3) . The shells are decidedly rough.

Achatinella limb at a Gul., from Ahonui and Kalaikoa is
identical with the typical green and yellow forms of byronii
shown in figs. 1 to Ic, having the same subsutural line, etc.
It is described as * ' striate, sometimes rugose, green or yellow ;
apex rose, frequently faded in mature specimens ; . . . peris-
tome black.' " It differs from A. melanostoma Newc., in
being more rugose, with rose-colored apex and subangulated
body- whorl." Gulick subsequently (P. Z. S., 1873, p. 91)
conceded the identity of limb at a with byronii.

A. BYRONII RUGOSA Newcomb. PI. 29, figs. 22, 22a ; pi. 27,
figs. 8 to 10a.

" Shell dextral, conical, glossy; whorls 6, rounded, mar-
gined above ; suture well impressed. Lip expanded, of a red-
dish-brown at the margin, thickened near the edge. Columella
white, short, twisted, and with a strong callus spread over
the umbilicus. Aperture subrhomboidal. StrisB longitudinal,
numerous, and on the last whorl rugose. Color of epidermis
of a deep green, a light or dark umber, sometimes alternating
and arranged with the striae, with or without pale green trans-

136 ACHATINELLA BYRONII.

verse lines. Length fourteen-twentieth^, diam. eight-twen-
tieths inch ' ' (Newc.).

Oahu: Ewa (Newcomb) ; Waimalu, "Waiawa, Waipio (Gu-
lick).

Achatinella rugosa NEWC., P. Z. S., 1853, p. 138, pi. 22, f.
22, 22a, 1854. PFR., Monogr., iv, 521. SYKES, Fauna Ha-
waiiensis, p. 309. THWING, Orig. Descript., pi. 1, f. 24.
Bulimella rugosa LYONS,, Hawaiian Almanac and Annual for
1892, p. 104, pi. 1, f. 2.

Newcorab's original figures are reproduced, pi. 29, figs. 22,
22a-. They are like the shells collected by Gulick in Waiawa
valley (pi. 27, fig. 8), which is probably what Xewcomb meant
by" Ewa."

This race inhabits valleys and ridges mainly southeast of
those occupied by byronii. It differs from that by having the
last whorl usually more strongly corrugated, the lip-callus
tit inner, the whole Up of a brownish flesh-tint, at least typi-
cally, and the shell lias the gloss of fresh varnish.

There are, of course, specimens which taken by themselves
could hardly be classified correctly; occasional individuals
may be smooth or nearly so. The occurrence of almost black
(brown-black or green-black) shells in most colonies of rugosa
is characteristic, since byronii is rarely if ever so dark. The
embryonic whorls are grayish-buff, flesh-colored or white.

The color-patterns are: (1) greenish-yellow, profusely
streaked with green, or yellow streaked with brownish-olive,
either plain or with two spiral brown bands and a subsutural
band, (2) last whorl black or nearly so. There are also inter-
mediate specimens connecting the darkest and lightest.

In a lot from Waimalu there are no banded shells (pi. 27,
figs. 9, 9a). This place must be at or near the extreme south-
eastern limit of the species. Waiawa shells are very often
banded, and with a small proportion of the blackish form (pi.
27, fig. 8). In a lot from Waipio the blackish- chestnut form
predominates (pi. 27, figs. 10, 10a). All the preceding are
from the Gulick collection, and probably all of them were
taken at rather low elevations.

Length 19.3, diam. 10.2, aperture 9 mm. ; 6% whorls. AVai-
pio.

ACHATINELLA BYRONII. 137

Length 17.5, dia.ni. 10 mm. Waipio.

Length 18.2, diam. 9.4, aperture 8.7 mm. Waiawa.

Length 16.3, diam. 10, aperture 8.8 mm. Waiawa.

In Dr. C. M. Cooke's collection there is a lot -of 9 specimens
(no. 1828-30) agreeing entirely with some of Gulick's Wai-
nialu rugosa (such as pi. 27, fig. 9a), said to have been col-
lected in " a very small and extremely isolated clump of trees
on the extreme S. slope of the Wai ana e range, ' in the region
of Palihua. As the collector was a person knowing noth-
ing of shells, I cannot help suspecting that he got a Koolau
range lot confused with lids Waianae catch. The shells are
greenish-black with the first neanic whorl green or olive
streaked on a yellow ground, embryonic whorls deep livid
purple in some, white in other examples. They have the
gummy gloss of rugosa.

1. Sub variety capax Pils. & Cooke. PL 31, figs. 7, la, 11.
At the head of Waiawa gulch, along the Waiahoie- Waiawa,
ditch trail, 14 mile from top, Mr. Spalding collected large,
robust specimens, less wrinkled than the typical form. The
shell is streaked with olive 'and yellowish-olive on a lime-green
ground, often with chestnut 'bands at periphery and base.
The lip is either pure white or narrowly bordered with purple
or brown; embryonic whorls either flesh-pink or bicolored, a
brown zone above, white below. They were on banana and
ieie. The colony was found to extend as far as the Kipapa
division ridge, and probably goes farther north. There is but
little variation in color and shape, and it is invariably dextral.
The patterns are shown in figs. 1-71}.

Length 21, diam. 12.5, aperture 11 mm. ; 6 whorls.
Length 19, diam. 10.5, aperture 9.2 mm.

2. Subyariety ivaimanoensis P. C., new. PI. 31, figs. 9 to
9f7. On the crest of the Waimano-Manana ridge, a half mile
to a mile 'above the locality of A. t. cookei, 'there is a dextral
pulcherrima-like shell, with ground of almost white, buff, or
greenish yellow, plain or with chestnut bands in the typical
rugosa positions, or variously split into lines ; lip white or pale
lilac. Surface much smoother than in rugosa. Embryo
white, or in the darkest shells, light brown or bicolored. All
seen are dextral.

138 ACHATINELLA BYRONII.

Length 17.8, diam. 10.7, aperture 9 mm.

Length 15.5, diam. 9.5, aperture 7.5 mm.

Through the darkest specimens, this race seems to connect
with the more wrinkled small form of rugosa, no. 3.

3. At the lower edge of the above colony, just above the
locality of A. cookei, there is a diminutive race similar to the
preceding in shape, but more strongly wrinkled, with the
color-patterns of subvar. capax ; also a few very deep chestnut
shells. The narrow lip is purple, and the upper whorls flesh
pink. Length 17, diam. 9.8, aperture 8.3 mm., o 1 /^ whorls.
PI. 31, figs. 8, 8a.

Forms 2 and 3 were collected Feb. 17, 1913, in company
with Messrs. Spalding and Merriam.

4Z). A. BYRONII NIGRICANS U. Subsp. PI. 31, figS. 10, 11, 12.

The shell is sinistml, oblong-conic, rather solid. Embryonic
3 whorls snow-white, the next whorl brown, or brown and
green, with a light band, last ivhorl intensely Hack above,
but on close inspection showing very indistinctly a dusky
greenish-yellow band at periphery and a wider one on the
base. The surface of the last two whorls is rather coarsely
wrinkled, and has a brilliant gloss. The aperture is very
oblique, ovate, white OF with a faint lilac tint within; peris-
tome has a narrow callous rib within of a fine purple color,
changing to violet at the lip-edge. Columellar fold purple
with white tip or crest, Parietal wall deep purplish brown.

Length 17.4, diam. 11, aperture 9 mm. ; 6% whorls.

Length 21, diam. 12.3 mm.

Waimano-Manana ridge at about 1400 ft. elevation, in a
very small area along the summit trail (Spalding, Merriam,
Pilsbry, Wilder).

This form differs from rugosa Xewc. by its capacious form,
coloration and sinistral coil. The last character would not
be of much significance were it not that the whole byronii-
rugosa series is dextral, in hundreds of individuals which have
been examined from many localities. A snow-white embryo
is also occasionally seen in rugosa. A. I), nigricans is doubt-
less a derivative from rugosa.

ACHATINELLA LILA. 139

It lives in a clump of lehua trees alongside of the trail,
about 1000 yards down the ridge from the locality of
A. turgida cookei. The colony is a very small one.

In Mr. Wilder 's collection there are some nigricans from
the type colony distinctly banded with green on the last whorl.

Mr. Spalding also found three specimens on a large moki-
hana bush on a lateral spur of the ridge 'about 100 yards
mauka from the locality of A. t. cookei. One of -these speci-
mens, fig. 10, is larger than those from the lower station. The
upper post-embryonic whorl is green, banded with brown.

It is a rare and handsome shell, possibly deserving specific
rank. There is a black form of A. turgida which has an as-
tonishing resemblance to nigricans, but differs by being
smoother with the real turgida apex, that of nigricans being
quite different,

So far as known, this subspecies is restricted to very few
small tree, in two spots less than a mile apart.

5. A. LILA Pilsbry, n. sp. PL 31, figs. 15 to 15 d.

The shell is sinistral, ovate-conic, thin but strong, nearly
smooth, brilliantly glossy. The embryonic whorls are burnt
sienna brown (weathering to whitish in adult shells) or some-
times there is a light median zone. Last whorl either (fig.
15cZ) uniform blackish chestnut, or (fig. 15) having a chest-
nut peripheral band and baso-columellar patch on a yellow
ground, or (figs. 15a, 6) like the last but with a green band
midway between periphery and suture, or (fig. 15c) with
sutural and peripheral bands and baso-columellar patch of
yellow on a chestnut ground. There are also a few speci-
mens more or less intermediate between the patterns of figs.
15 a, 5, d. The aperture is moderately oblique, white or
faintly lilac within ; peristome acute, very little or not per-
ceptibly thickened within; columellar fold strong, purple or
white.

Length 17, diam. 11, aperture 9 mm. ; 5% whorls.

Length 16.8, diam. 10.2, aperture 8.8 mm.

Length 16, diam. 15.3 mm.

Oahu : crest of the Waimano-Manana ridge at junction with

140 ACHATINELLA PULCHERRIMA.

the main range, running from the summit down about three-
fourths of a mile along the ridge trail; in axils of ieie (Spal-
ding, Merriam and Pilsbry). Types no. 108066 A. N. S. P.
Go type in Bishop Museum.

This charming shell has a great resemblance to A. abbreviata
of the eastern end of the range, from which it is separated by
about a fourth of the length of the island, wherein no re-
lated forms have been found. A. abbreviata is invariably
dextra'l, A. lila sinistral. There are also differences in the
patterns of the banded forms. It is likely that A. lila is more
closely related to A. nigricans, and therefore a derivative of
the byronii group of species, and not directly related to
A. abbreviata. The color-pattern 15c is a "negative" of 15a,
the light and dark areas being reversed. In treating of the
genus Liguus I have noted similar instances.

In the type lot the 'colors are in the following numbers:
Pattern of fig. 15, 1 specimen; figs. 15a-, b, 8; fig. 15c, 5;
fig. 15<r7, 10. Four of the lot have the columellar fold white.

A. lila is rather abundant on its misty pe-ak, but soon be-
comes rarer and disappears as one descends the trail into less
cloud-hung levels. Near the 'summit only the black form is
found, the beautifully green-banded shells occurring lower on
the trail, where there 'are few of the black ones. The lateral
slopes of the ridge drop so precipitously that collecting is
almost restricted to what mav be reached from the trail. On

-

the north side of the peak a pali of a couple of thousand feet
permits 'collecting only close to the summit. Named for L. L. C.

6. A. PULCHERRIMA Swainson. PL 27, figs. 2 (Ahonui), 5-5g,
6 (Wahiawa) ; pi. 31, figs. 13 (Kawaihalona), 14 (Hele-
mano).

"Shell ovate-oblong, subcylindical ; white or yellow with
broad bands of chestnut; margin of the lip brown. Var. a,
golden yellow, suture chestnut. This very elegant species is
about 0.8 inch long, 'and is much more slender than any of the
preceding. It varies somewhat in form, some specimens being
more ventricose than others, and also in the number and color
of its bands. The ground color is a deep and rich chestnut,

ACHATINELbA PULCHEREIMA. 141

with from one to three bands of orange, yellow, fulvous or
white. The marginal groove to the suture is very, close and
distinct in all. The golden yellow variety is without bands.
In all the colors -are remarkably rich and vivid ' : (Swa/inson).

Oahu: Waialua (Newcomb) ; Helemano, Wahiawa, Kalaikoa
and Ahonui (Gulick) ; Helemano, low down, and eastern spurs
of Kawaihalona (Irwin Spalding).

Achatinella pulcherrima SWAINS, The Quarterly Journal of
Science, Literature and Art, The Royal Institution of Great
Britain, January to June, 1828, p. 86 ; Zoological Illustrations,
iii, Second Series, 1832-3, p. 123, pi. 123, f. 2. PFR., Mono-
graphia id, 237 ; iii, 460 ; vi, 172 ; Conchyl. Oab. p. 280, pi. 67,
f. 30, 31. Newcomb, Ann. Lye. vi, 308. THWING, Orig. Des-
cript. pi. 1, f. 18. Bulimella p., LYONS, Hawaiian Almanac
and Annual fo<r 1892, p. 105, pi. 1, f. 6. Achatinella lorata
Fer., REEVE, Conch. Icon. pi. 1, f. 6, 1850. Achatmella
melanostoma NEWC., P. Z. S. 1853, p. 132, pi. 22, f. 7.
Achatmella multicolor PFR., in part, P. Z. S. 1855, p. 4, pi. 30,
fig. 11 (not fig. llfl). Achatinella mahogani GULICK, Ann.
Lye. Nat. Hist, of N. Y. vi, p. 254, pi. 8, fig. 72.

This specie's differs from byro-nii chiefly by its smooth sur-
face. The last whorl is more evenly rounded, -while in byronii
it is often subangular. The pattern of Swainson's left-hand
figure was selected by Pfeiffer as typical. It is shown in our
pi. 27, fig. 5c ; also pi. 31, fig. 13. There are two broad chest-
nut zones on a yellow ground. This happens to be a rather
rare color-form which however occurs in Gulick 's Wahiawa
series (fig. 5c), and has been collected by Mr. Spalding on the
eastern spurs of Kawaihalona (pi. 31, fig. 13). I take the
latter valley as type locality, since it was certainly from shells
taken in that neighborhood that Lord Byron's lei of Achat-
inellas was made. The lip is obtuse, black-brown, with a
whitish rib within. The glossy surface has exactly the char-
acter of that of nymph a, etc. All of the large series examined
are dextral. Color-forms of a lot from Wahiawa (Gulick),
are as follows.

PI. 27, fig. 5c. Two broad chestnut zones on a yellow
ground ; sutural line usually chestnut.

142 ACHATINELLA PULCHERRIMA.

PI. 27, figs. 5d, 5e. Two narrow chestnut bands or lines;
rarely some additional lines. One or both lines may be want-
ing on the last whorl.

PI. 27, fig. 5/\ Yellow, without bands except the sutural
(Swainson's var. a) ; or with a white sutural border, fig. 5g.

PL 27, fig. 5, 5#, 5&. Chestnut, with or without darker
bands.

Length 18.5, dizain. 9.3 mm.

Length 19, diam. 10 mm.

Length 20, diam. 11.2 mm.

Length 17, diam. 10 mm.

In Helemano and Waialua district, the northwestern limits
of pulclierrima^ the same forms occur. Those of Kalaikoa do
not differ noticeably. In Ahonui part of the lots examine 1
are like those of "AVahiawa.' In others there are many whit-
ish shells with a gamboge tint, with or without dark lines.
Also a lot of shells varying from chestnut to sulphur-tinged
yellow, with sutural band only, many of them stouter in form
than typical pulclierrima. Length 18.5, diam. 11.5 mm.
These lots are from the Gulick collection. PI. 31, fig. 14 is
from low in Helenrano, coll. by Spalding,

Judging from the description and figure of A. melanostoma
Xewc., as well as from specimens received from him, it is a
synonym of pulclierrima, being based upon a somewhat dif-
ferent color form from that of Swainson, though occurring
with the typical color- form in AVahiawa district. Newcomb's
locality, Ewa, refers to the district, which takes in the ex-
treme southeastern part of the range of pulcherrima. The
original description follows.

A. melanostoma Newc. (pi. 29, fig. 7, reproduction of ori-
ginal figure). "Shell dextral, polished, solid conical; whorls
6, plano-convex, slightly margined above ; suture moderate,
accompanied with a black revolving band. Aperture sub-
ovate; lip black, thickened within. Columella short, brown-
ish-red, terminating abruptly in a strong plait or tuberosity.
Color yellowish or umber, with or without transverse brown
lines. Length fourteen-twentieths, diam. seven-twentieths
inch. Ewa, Oahu : (Neivcomb).

ACHATINELLA PULCHERRIMA. 143

According to a note from Mr. Sp'alding, melanostoma is
now found only in the northern 'branch of Kipapa,; but I do
not know just what form he found so far east.

A. multicolor Pfr., PI. 30, fig. 11, reproduction of original
figure. " Shell dextral or sinistral, imperf orate, conic-
oblong, solid, striate, and under the lens most minutely de-
cussated, glossy; buff or whitish, variously ornamented with
blackish-chestnut bands, rarely uni colored. Spire long-conic,
towards the apex somewhat attenuated, rather acute; suture
margined. Whorls 6, rather flat, the last about two-fifths the
total length, rounded basally. Aperture oblique, truncate in-
verted ear-shaped; columellar fold above, strong, twisted.
Peristome black-edged, the outer margin somewhat straight-
ened, narrowly expanded; columellar margin dilated, sub-
a dilate. Length 17, diam. 9 mm., aperture 8% mm. long, 4%
wide. Sandwich Is., Frick in Cuming coll. (Pfr.).

In the above description Pfeiffer obviously confused two
species. His two figures, reproduced in my pi. 30, figs. 11,
lla, belong to A. pulcherrima (fig. 11) and A. soiuerbyana
oviformis (fig. lla). Doctor Hartman recognized the latter
and added A. oviformis to the synonymy of multicolor; but
he included Pfeaffer's fig. 11, and therefore did not restrict
the species or realize its dual composition (Proc. A. N. S.
Phila. 1888, p. 30). Mr. Sykes referred Pfeiffer 's fig. lla to
A. oviformis, 'and fig. 11 to A. recta Newc., which he ranks as
a variety of A. byronii. As several of Pfeiffer's phrases, par-
ticularly "perist. nigrolimbatum' apply best to his dextral
shell, I would restrict A. multicolor to fig 11. It becomes
therefore an absolute synonym of A. pulcherrima, having the
typical color-pattern, which seems to have been more common
in the lower forests of three-quarters of a century ago than
it is now. A. recta Newc., which Mr. Sykes and others have
confused with the present group, belongs in the series of A.
livida.

Form mahogani Gulick (pi. 27, figs, 4 to 4<i, Ahonui). This
form has the smooth surface of A. pulcherrima, merely marked
with growth-lines, not corrugated or wrinkled. A subsutural
dark band is often wanting, or not darker than the rest of the

144 ACHATINELLA PULCHERRIMA.

shell, which is nearly free from dark streaks. The lip has a
narrower black edge than well-colored byronii, and has a white
rim within. The figures, from a large set in the Gulick col-
lection, show the range of color from mahogany to yellow,
with or without bands. Except in the usually darker colora-
tion, I can see no reason for separating mahogam from pul-
cherrima. Gulick 's description follows. His figure is like
pi. 27, fig. 4a, which is exactly like the type-specimen.

" Achatinella mahogan-i. Shell dextral, imperf orate, tur-
reted, solid, shining, striate, reddish chestnut; apex obtuse,
rose; spire concavely turreted; suture lightly margined, mod-
erately impressed ; whorls 6, convex, the last inflated, with the
base black-chestnut and regularly rounded; columellar fold
central, white, strong; 'aperture truncately auriform, white
within, with external margin scarcely reflected anteriorly,
arcuate; columellar margin dilated, adnate; parietal margin
wanting. Length 20.6, breadth 11, length of body-whorl 12
mm. Average weight 8 grains.

"On trees, Ahonui and Kalaikoa, J. T. G. It lacks the
black color of lip and suture which characterizes A. melano-
stoma Newc. 'and A. limlata Nob. ? and 'also differs in the gen-
eral form and color ' : (Gulick) .

A form which would be referred to mahogani if it occurred
on the south side of the main range is in the Gulick collection
(no. 125 ; no. 92461 A. N. S. P.) , with the locality Kahana. It
is not unlike Pfeiffer's figure of A. plunospira except that
'the base is not abruptly darker as in that figure. It differs
from Kahana detipiens by having light brown early whorls
instead of white.

6a. A. PULCHERRIMA NYMPHA Gulick. PL 32, figs. 13, 14 ; pi.
27, figs, 7, la, 71).

"Shell dextral, imperf orate, oblong-ovate, solid, shining,
very lightly striated, yellow or chestnut, sometimes lined with
brown." Interior and columellar fold white. Peristome
well thickened within close to the edge, white, or in banded
shells showing marginal spots. Dark chestnut shells have a
narrow dark lip-bo-rder. Length 18.5, diam. 10 mm.

ACHATINELLA DECIPIENS. 1-15

Oahu: Ahoiiui, Kalaikoa, Wahiawia, Helemano, Opaeula,
Kawailoa and Waimea (Gulick) .

<

Achatinella nympha GUL., Ann. Lye. Nat. Hist, of N. Y.,
vi, 251, pi. 8, f. 69, 1858.

Merely a variety of pulcherrima, from which it differs by
the white Up, -which however is dark-edged in some examples
of Gulick's lots. It differs from recta, by the 'Conspicuous
thickening of the lip. Some of the large lot in the G-ulick col-
lection have a narrow dark sutural band, but most of them
have none, and in some the suture has a white border. In a
lot from Wahiawa the patterns are in these numbers: yellow
ground bandless 6, with bands or lines 25 ; chestnut ground,
bandless 20, banded 9 shells.

According to Gulick, Helemano is the metropolis of
nympha. The specimens figured are from Helemano (pi. 32,
fig. 14) and Wahiawa (pi. 27, figs. 7, la, 76; pi. 32, fig. 13).
Fig. 13 is too deep in color. The shell is much lighter, -what
Rldgway calls barium yellow. PL 27, fig. 5#, might be re-
ferred to nympha as well as to pulcherrima.

Dr. Cooke, who examined the type of A. aptycha Pfr. (see
p, 54) , considers it a form of nympha. The figure is repro-
duced in my pi. 30, fig. 1.

7. A. DECIPIENS Newcomb. PL 32, figs. 2 to 6c.

' ' Shell conieaily-elongate, solid ; whorls 6, slightly rounded,
margined above. Aperture elongately-ovate ; lip subrenected ;
columella short, obliquely twisted, with an expanded callus.
Suture slightly (impressed; striae numerous, longitudinal, ob-
lique. Color white with yellow transverse bands, or yellow
with white transverse lines and longitudinal chestnut colored
stripes. Length sixteen-twentieths, diam. seven- twentieths
inch ' (Newcomb).

Oahu: Kahana (Newcomb) ; Hakipuu (Gulick for A. corni-
gata) .

Achatinella decipiens Newc., P. Z. S., 1853, p. 153, pi. 24, f.
68, 1854. Achatinella corrugata GULICK, Ann. Lye. N. H. of
N. Y., vi, p. 248, pi. 8, f . 66, 1858. Achatinella torrida GUL.
Ann. Lye. N. H. of N. Y., vi, p. 250, pi. 8, f. 68, Feb., 1868.

146 ACHATINELLA DECIPIENS.

A. decipiens (including torrida and corru-gata) represents
A. byronii on the northeastern side of the Main Range in the
valleys from Waikane to Kahana. As we do not know what
forms occur on top, and whether there is complete intergrada-
tiou there, the races from the windward side are for the tame
treated as a species distinct from byronii It differs from
byronii by having the lip entirely or mainly white and by
having part of the patterns different.

Xewcomb's typical lot consisted of shells resembling pi. 32,
figs. 2, 2a, from examples received from him; fig. 2a repre-
senting the pattern of his type figure. The shell is yellow,
profusely streaked and lineate with deeper yellow and chest-
nut along growth-lines, having a white zone below the suture
and typically another at the periphery; spire mainly white,
lip and aperture also white. Under a hand-lens the surface
is seen to be quite distinctly sitriate spirally, with a finer de-
scending wrinkling such as is usual in Partyl/ina, but this
varies in distinctness a good deal, among the specimens from
Newcomb himself, on some of which it would not be noticed if
not especially looked for. Length IS, diam. 10 mm.

In other Kahana lots, collected by Gulick, the patterns vary
somewhat, (pi. 32, figs. 3 to 5). Olive-green or yellowish-
green may take the place of chestnut streaking. The surface,
typically almost smooth, may be rather distinctly corrugated
in some shells. These pass directly into the form called cor-
rugata Gulick.

Length 23, dram. 11 mm.

Length 17, diam. 10 mm.

Length 16.3, diam. 9 mm. (rough, stunted shell).

In another Kahana lot the prevalent color is light yellowish-
green, faintly streaked, and fading to whitish near the suture
and on the spire (pi. 32, fig. 35). Other shells are broadly
banded with chestnut, glossy like the preceding; lip tinted
(pi. 32, fig. 4a). Another pattern occurring sparsely in this
and other Kahana lots is dull yellowish, copiously streaked
with chestnut and blackish, and having two spiral brown
bands. It is almost lusterless and somewhat corrugated.

It is evident that Gulick did not collect in Newcomb 's type
colony of decipiens, which may have been of small extent.

ACHATINELLA DECIPIENS. 147

Achatinella corrugata Gulick from Hakipuu (pi. 32, figs.
7, la, 7b) seems to me to be almost identical with decipiens,
from which it differs chiefly by the usually more corrugated
surface, which shows only very indistinct traces of minute
spiral sculpture. The color-patterns resemble the streaked
forms of torrida except that none in the large lot has a white
subsutural band, such as torrida and decipiens often possess.
Gulick 's type is streaked with olive-green on a pale-green
ground (pi. 32, fig. la). Others are streaked with chestnut
on a yellow ground, with or without one or two spiral 'chest-
nut or blackish bands, while still other shells : are brownish -
black with a white summit. None has spiral white bands, and
very few have two chestnut bands. GuLick remarks that sims-
tral examples have been found. I have noticed one in the lot
of several hundred shells from his collection. The original
description of corrugata follows.

" Shell dextral, imperf orate, ovate, solid, shining, striated,
green ; apex subacute, pale ; spire convexly conic ; suture mar-
gined, moderately impressed; whorls 5%, convex, the last
rugose; columellar fold central, white, strong; aperture truii-
cately auriform, white within; perlstome thickened 'within,
white, sometimes margined with brown, with external margin
dilated, adnate ; parietal margin wanting. Length 18%, diam.
10 mm. ; average weight 5.3 grains. Hakipuu, Oahu, on trees.
The color is often nearly black, and sometimes reddish-yel-
low " (Gulick).

A rather large shell measures, length 19.1, diam. 10.8 mm. ;
a very narrow blackish one, length 18, diam. 8.5 mm. (fig. 7&) .

Another colony from Hakipuu was correctly considered by
Gulick to be transitional to decipiens proper. Some speci-
mens have a white sutural band, and occasionally there is a
white line at the periphery (pi. 32, figs. 6 to 6c) .

Variety ( ?) planospira Pfeiffer. PI. 30, fig. 8.

Achatinella planospira Pfr. has been placed in the syn-
onymy of decipiens by Newcomb, who is followed by Hart--
main, Sykes and Thwing. I have seen specimens of form cor-
rugata having the same pattern except that the summit is
light, while Pfeiffer 's figure shows it dark. The status of the

148 ACHATINELLA DECIPIENS.

form is therefore uncertain. It needs comparison with pul-
cherrima and mahogani. There is, however, a lot of maho-
gani-like shells in the Gulick collection (no. 125) from Ka-
hana, having the apex dark, and agreeing fairly well with
planospira.

The original account follows. A. planospira Pfr. (pi. 30,
fig. 8, photographically reproduced from Pfeiffer's plate).
; Shell dextral, imperforate, ovate-pyramidal; solid, closely
striate and under the lens very obsoleteily striate spirally,
glossy; blackish, ornamented with several narrow white or
buff bands. Spire conic, rather acute; suture linear, mar-
gined. Whorls 5!/o, flat, the last a little shorter than the
spire, obsoletely angular, rounded basally. Aperture oblique,
inverted ear-shaped. Columellar fold above, strong, twisted,
white. Peristome thickly white-lipped, the right margin,
spreading, narrowly expanded, columellar margin adnate.
Length 19, diam. 9y 2 mm., aperture 9 mm. long, 4 wide. Var.
Z>, above blackish-chestnut, the last whorl tawny, with brown
lines. Sandwich Is., Frick in Cuming coll : (Pfr.).

Variety torrida Gulick. PL 32, figs. 8 to 12c.

The sinistral form of corrugata was described as A. torrida
by Mr. Gulick. The distinction from corrugata is rather too
vague to rank it as a subspecies, though there is a certain
amount of racial differentiation. Its patterns differ in many
specimens, and then most torrida are sinistral. The original
account follows.

" Shell sinistral, imperforate, ovate-conic, solid, shining,
somewhat rugose, green or fulvous; apex subacute, chestnut;
spire conic ; suture lightly margined, white, moderately im-
pressed; whorls 6, convex; columella with a strong white fold
near the body-whorl. Aperture oblique, truncately auriform,
white within ; peristome white, thickened within, with external
margin scarcely reflected anteriorly, slightly compressed;
columellar margin dilated, adnate ; parietal margin wanting.

Length IT 1 /-}, breadth 9% mm. ; average weight 4.5 grains '
(Gulick).

11 Var. I), with black spiral lines. I am in doubt 'Concerning
the limits of variation of this species ; var. & may be distinct '
(Gulick) .

ACHATINELLA DECIPIENS. 149

Oahu: Kahana, Kaaawa and Waikane (Gulick) ; Waiolu
(J. S. Emerson).

Waiolu may be considered the type locality, as here the
typical form (pi. 32, figs. 11, Ha, 11&) occurred. It is closely,
.streaked with rather dull-green or chestnut on a light ground,
the streaks sometimes confluent. A wide subsutural white>
band (often denuded of cuticle) is frequently present. The
apex is not chestnut, as Gulick states, in any of the lot of over
a hundred from Ms collection, from all the localities men-
tioned above. Six out of a set of 56 from Kahana are dex-
tral, all others seen being sinistral. Specimens with streaked
pattern are almost mirror images of A. decipiens, but they
differ by having the aperture a little more oblique, the lip is
usually a trifle less thickened within, and the surface in the
average torrida is more corrugated. The lip usually has a
fleshy or brownish border. Specimens from Kaaawa (pi. 32,
fig. 9) are similar to those of Waiolu.

A small set from Waikane is strongly corrugated, heavily
streaked, with a dark band below the suture (pi. 32, fig. 8) .

In Kahana the shells are variable (pi. 32, figs. 10 to lOc?) .
Those having blackish streaks on a yellowish or chestnut
"ground resemble forms of other valleys. Others have one or
two black-brown zones and a white sutural band ; and in some
the blackish-chestnut color spreads over all but the earliest
whorls. The surface has very little gloss in this lot, Kahana
specimens in 'coll. C. M. Cooke are similar except that the
ground-'Color is greenish and the surface glossy.

Length 18, diam. 10.7 mm. Kahana.

Length 18.5, diam. 10.5 mm. Kahana.

Length 17, diam. 10 mm. Kahana,

Length 17.5, diam. 10 mm. Waiolu.

Kaaawa-Hakipuu division ridge (pi. 32, figs. 12 to 12c).
In a very handsome series collected by Mr. Spa.ldlng the em-
bryonic whorls are light-brown, differing thus from the large
series of torrida in Gulick 's collection, in which the early
whorls are white or nearly so. The last whorl, fig. 12, is green,
streaked with a much darker shade, or similar, with two black
"bands and a white sutural band. Fig. 12c, black with a yellow

150 ACHATINELLA DECIPIENS.

band and white line at the suture. Fig. 12a, chocolate. The
lip is dark-edged, or in form fig. 12c has two spots at the ter-
minations of the bands. The surface is rather rough. This
colony is mainly sinistral, but there are some dextral shells.

la. A. DECIPIENS KALIUWAAENSIS Pilsbry & Cooke, n. subsp.
PI, 32, figs. 1, la, 16.

The shell is similar to decipiens but smoother, more glossy,
with only obsolete traces of sculpture. Color various: (a)
White with a brownish sutural line, (&) Olive-ocher with a
white zone below the suture and a white line at periphery.
(c) Olive-ocher with a chestnut subsutural line or none, and a
pia'le peripheral line. Lip only narrowly thickened within,
white.

Length 19.2, diam. 10, aperture 9 mm.

Length 17, diam. 9.7, aperture 8 mm.

Length 15.3, diam. 8.9, aperture 7.6 mm.

Eastern ravines of Kaliuwaa, type loe., also central and
western ravines (Irwin Spalding).

This new form has such a relation to decipiens as pulcher-
rirna to byronii. The locality lies some distance northwest of
that of decipiens, in a district where the other Bulimellas dif-
fer from the Kahana forms. The coloration is much like that
of typical decipiens. The white-banded form may be consid-
ered the typical pattern. The cotypes are in coll. A. N. S. P.
and Bishop Museum, collected by Mr. Spalding who has a
long series. It is also in the Thaanum, Thurston 'and some
other recently gathered collections.

75. A. DECIPIENS ( ?) SWAINSONI Pfeiffer. PI. 30, fig. 13.

" Shell sinistral, imperforate, ovate-conic, solid, nearly
smooth, glossy ; whitish, delicately streaked with tawny. Spire
conic, apex fulvous, rather acute ; suture margined. Whorls
5!/2, a little convex, the last nearly equal to two-fifths the
length, swollen below the suture, rounded >at base. Aperture
slightly diagonal, reversed auriform. Oolumellar fold super-
ior, strong, nodiform; peristome bordered with black-brown,
the external margin a little reflected, thickly labiate within j

ACHATINELLA ROSEA. 151

columellar margin thick, flexuous, adnate. Length 20, ddam.
11 mni.

"5. Greenish buff, the last whorl chestnut anteriorly :

(P/r.).

Sandwich Islands, Frick, in coll. II. Cuming (P/r.).

Achatinella sivainsoni P^R., Pro<c. Zool. Soc. Lond., 1855,
p. 4, pi. 30, f . 13.

Dr. Cooke, who examined the type, thought this a siiiistral
individual of A. pulcherrima. " Newcomb suggested that this
might be only a form of A. sordida ; it appears however to be
distinct, being broader, brown in general coloration, and hav-
ing a brown in place of a white lip. It is a little doubtful
from its form if it be correctly placed in this group [Acha-
tinella s. sir.], but the sections are very artificial : (Sykes,
Fauna Hawaiiensis, p. 304). As I have not seen the shell, I
cannot properly express an opinion; but from the figure it
seems as near the torrida form of decipiens from the Kaaawa-
Hakipuu ridge as anything, and I put it here for want of a
better place. It may be noted that several enigmas of Frick-
Pfeiffer origin were from the Koolauloa district. We know
very little about Frick, but from what he got we may infer
that the country between Kaneohe Bay and Kahuku was one
of his collecting grounds.

SERIES OF A. BULIMOIDES.

Rather large, compact, capacious shells, smooth, never
streaked with green, usually banded. This series comprises
numerous forms on both sides of the western half of the Main
Range.

8. A. ROSEA Swainson. H. 34, figs. 1 to 85.

' ' Shell reversed, pale rose-color, with obsolete white bands.
I place this, for the present, as a variety of the last [A. ~buli-
moides] to which, except in being reversed, it bears a 'dose
resemblance in size, form and general habit. It is entirely
of a pale and delicate rose-color, with two obsolete bands of
white on the body-whorl. The margin of the lip and columella
are of a deeper rose- color, and the aperture white. It should

152 ACHATINELLA ROSEA.

be observed tha.t the marginal [subsutural] groove, which is
scarcely perceptible in the last, is in this very distinct '
(Sivainson).

Length 22.3, diam. 13.5 mm. ; G 1 /^ whorls.

Length 23, diam. 13 mm.

Length 19, diam. 12 mm.

Oahu : Helemano and Poainoho to Kaukinehua ; formerly,
according to Gulick, from Kawailoa to Ahonui.

[A. bulinwides] var. ? (rosea-) SWAINSON, The Quarterly
Journal of Sci., Lit. and Art, January to Jiine, 1828, p. 85.
Achatinella rosea SWAINS., Zoological Illustrations, III, 2d
series, p. 123, pi. 123, f. 2. REEVE, Conch. Icon., vi, pi. 4, f.
28. PFR., Monogr., ii, 236; iii, 456; iv, 518; vi, 163.
THWING, Orig. Descript., p. 93.

A. rosea is invariably sinistral. It further differs from A.
biilimoides by being smoother, more glossy. The colour varies
widely, but the older collectors found only the white and pale-
rose forms, which, from the large series in old collections,
must have been abundant in the time of Newcomb and Gul-
ick. Figures la to 2a represent the usual color-forms of the
species, the other forms figured being more or less rare. It
evidently extended lower than the dark-colored forms, in a
zone now barren by the recession of the forests. Gulick's
localities none of them more exactly located than by the val-
ley or district were Wahiawa, Waialua, Helemano (pi. 34,
figs. 1, la-), Opaeula and Kawailoa, all of the many speci-
mens being white or pale-rose or light rosy-brown, with roseate
lip, the colonies apparently homogeneous except for these fluc-
tuations of tint.

According to Mr. Spalding the headquarters of white or
nearly white rosea are in Helemano, but it is found from Ka-
wailoa to Kaukinehua in the Wahiawa district. There are
many gradations of tint between the white shells, the faint
rose, and the rose with white bands. An albino form with
the lip pure white has been taken by Mr. Spalding in a small
ravine in the eastern edge of Helemano, just above the forest
limit, similar to pi. 34, fig. 1, Helemano, Gulick coll.

A. rosea varies a good deal in size and shape. Typically
colored shells from Wahiawa, Gulick, measure:

ACHATINELLA ROSEA. 153

Length 22.3, diam. 13.6, aperture 11.5 mm. ; 6 l / 2 whorls.
Length 22.7, diam. 13, aperture 10.7 mm.

f

Specimens of a lot from ' Waialua, ' Thaanum, measure :

Length 21.7, diam. 12.3, aperture 10.5 mm. ; GI/? whorls.

Length 21.5, diam. 12, aperture 10.5 mm.

Length 18.3, diam. 12.4, aperture 10 mm.

Length 17.3, diam. 10, aperture 8.3 mm. ; 6 whorls.

Specimens from Ahonui in Mr. Thaanum 's collection are
white and pink, the usual coloration.

Messrs. J. S. and Oliver P. Emerson collected quantities of
dark-red rosea in the ravine in which the Waialua plantation
built its Kamoku reservoir, next west of Opaeula. This was
over fifty years ago, and the colony has been extinct these
many years.

On the middle ridge of Poamoho (between the western and
central branches of the stream), a colony about a mile above
the forest limit contains shells with the last whorl white, white
with two rose bauds, or rose-tinted with deeper rose bands
defining a median white zone. The lip and columellar fold
are deep rose color. PI. 34, figs. 2, 2a, coll. by Irwin Spald-
ing. The same form has been given me by Mr. Thaanum, coll.
by Mr. Kuhns. A mile further up, just west of the main
ridge, the shells have the summit dark; color of last whorl
pinard yellow with white suture ; the same with tawny bands
and spire ; yellow ocher with tawny bands and darker spire,
or flesh color changing to coral-pink near the suture. In all,
the lip is more or less blackish (pi. 34, figs. 6 to 6c, coll. by
Irwin Spalding). These shells are small, length 18 to 19.5
mm.

In Opaeula, at about 1,700 ft., Mr. Wilder found very dark
specimens, last whorl raw sienna with chestnut base, together
with pale green-yellow shells, both having the peristome and
summit dark (pi. 34, figs. 5, 5a) . It was probably lower down
that Gulick found the ordinary white and pink rosea from
Opaeula in his collection.

Similar but larger shells were taken by Mr. Spalding in the
eastern branch of the north fork of Kaukinehua (pi. 34, figs.
7). Other specimens from the same or an adjacent col-

154 ACHATINELLA BULIMOIDES.

ony are shown in pi. 34, figs. 8, 8 a, 85, taken by Mr. Wilder,
One measures, length 22, diam. 13.6, aperture 11.3 inm.

In a gulch west of Helemano Mr. Spalding found charming
pale green-yellow shells, with geranium-pink lip and faint
traces of two pink bands (pi. 34, fig. 4, no. 434 Spalding coll.).

In the bottom of Kawaihalona the shells vary from flesh or
salmon color to vinaceous in several tints (pi. 34, figs. 3, 3a,
coll. by Irwin Spalding; fig. 3a being no. 2167 of his collec-
tion).

Achatinella rosea is one of the most exquisite of land shells.
While the light-colored form is not uncommon in collections,
the ' ' green rosea/ the wine-colored and the dark forms are
rare even in Honolulu. They are among the prize specimens
of the finest collections. I did not myself collect in the rosed
territory, which has been thoroughly " shelled : ' in the last
few years, but the generosity of Messrs. Irwin Spalding -and;
W. D. Wilder has made it possible to illustrate the main var-
iations of this charming shell.

9. A. BULIMOIDES Swainson. PI. 33, figs. 1 to lj, 3 to 4.

" Shell ovate-oblong, subventricose, whitish with chestnut
bands ; spire thickened, the tip pale brown. Larger than the
last [A. livida], and of nearly the same form, but the spire is 1
less thickened and more pointed at the apex. The ground-
color, in some specimens, is pale chestnut or ferruginous,,
banded with darker shades and another of pure white. In
others the upper half of each whorl is whitish, and the lower
chestnut, marked by darker bands. The suture is scarcely if
at all margined by a groove. The aperture and pillar white '
(Swainson).

Oahu: Waialee, Waimea, Kawailoa valley (Gulick). Head
of Kaipapau, on the north side of the range (Irwin Spald-
ing).

Achatinella bulimoides SWAINSON, The Quarterly Journal
of Science, Literature and Art, The Royal Institution of Great
Britain, January to June, 1828, p. 85 ; Zoological Illustrations,
iii, 2d Series, 1832-33, pi. 108, f. 1. REEVE, Conch. Icon., vi,
pi. 1, f. 8, 1850. HEYNEMANN, Malak. Bl., xiv, 1867, pi. 1, f.

ACHATINELLA BULIMOIDES. 155

2 (dentition). PFR., Monographia, iii, 460; iv, 518; vi, 143;
viii, 216. THWING, Reprint Orig. Descr., p. 92, pi. 2, f. 1.-
GULICK, Evolution, Racial and Habitudinal, p. 41, pi. 3, f. 1-5.

Swainson's figures agree exactly with, my pi. 33, figs. Ifr,
Id, from " Kawailoa,' which I take to be the type locality.
It ' ' has now become a very rare shell, ' ' owing to the destruc-
tion of forest, but I imagine that it will be found higher up,
in the region between Kawailoa, Waimea and Kaipapau.

A. bulimoides is always dextral, imperf orate or nearly so,
solid; the upper whorls are finely striate spirally and flesh-
tinted ; the lip is well thickened within and usually white or
yellowish, but brown in dark forms. There is never a sub-
sutural dark line as in brown-band forms of A. rosea, but the
suture is often bordered with an impressed line.

Specimens from Kawailoa measure:

Length 20.6, diam. 12.4, aperture 11 mm. ; whorls 5%.

Length 21.3, diam. 11.8, aperture 11 mm. ; whorls 614.

In a lot of 48 from " Kawailoa " (pi. 33, figs. 1 to Ij, Gu-
lick coll., 92489, A. N. S. P), the following color-forms occur:

1. White, the upper whorls pale-pink or flesh-tinted; peri-
stoane yellowish; suture usually margined (fig. 1), 6 speci-
mens, la. Similar, but having a 'brown streak behind the
outer lip, 5 specimens.

2. White with two chestnut bands bordering a peripheral
white belt, upper whorls fleshy, lip white or brownish (figs.
la, 6 ) ; 6 specimens.

2a. Similar but the peripheral belt and base tawny ; a white
umbilical patch or none (fig. le) ; 5 specimens.

3. Entire base purple-brown, a brown, band above the
peripheral white band; spire sometimes dark (fig. 1<Z) ; 9 spe-
cimens.

4. A white zone below the suture, elsewhere chestnut or
purplish-brown, lip often brown (fig. 10) ; 14 specimens.

5. Purplish-brown with fleshy suture and summit (fig. li) ;
2 specimens.

6. Base whitish, upper part purplish-brown (fig. l/i) ; one
specimen.

7. Basal patch and peripheral band dark chestnut, a light
chestnut zone above (fig. Ic).

156 ACHATINELLA BULIMOIDES.

A small lot from Waimea 'oonsasts of forms figs. Id, \g and

In Waialee the color-form fig. Ic occurred, but only dead
shells 'are in the lot taken by Gulick in the early Fifties, so
it was probably extinct there at that time.

Across the range, near the head of Kaipapau (pi. 33, figs.
3 to 3d, coll. by Mr. Spalding), there is a colony having some
patterns identical 'with Kawailoa bulimoides. The embryonic
shell is usually dull purple ; last whorl variously banded with
dark chestnut and white, or yellowish-brown with or some-
times without bands; sutural border white or like the ground
color. The lip is white or sometimes dark-edged; and in a
large series taken, it is invariably dextral. This form has been
called vidua by some collectors, but I believe erroneously. It
is a slightly modified bulimoides, and not, I think, so closely
related to the Bulimellas of Kaliuwaa -as to those across the

mam range.

Mr. Gulick had this form from an adjacent locality. One
dead shell labeled " Hauula/ ' pi. 33, fig. 4, is now no. 92445,
A. N. S. P.

9a. A. BULIMOIDES MISTURA Pilsbry & Cooke. PI. 33, figs. 5-
5c, 6-7.

In Kaliuwaa valley and the ridge eastward along the Castle
trail, several colonies have been found by Messrs. Spalding,
Thurston and others, which 'are not directly referable to ovata,
obliqua, oomorpha or spadicea, while they belong to the same
group of local races. The shell is either sinistral or dextral;
an obesely ovate shape prevails; embryonic whorls are invar-
iably colored, usually chestnut, darker than in ovata and ob-
liqua; the lip-callus is rather narrow and the lip is narrowly
chocolate-bordered outside and within. The color varies in
different colonies, as follows:

PL 33, fig. 7, coll. by Spalding. On the edge of Punaluu
valley, where the Castle trail passes over the crest into Kaliu-
waa, a pure colony lives in a clump of ieie. The color is pale
grayish blue-violet (Ridgway, pi. 35) varying towards lav-
ender or pearl blue in different examples, suture bordered

ACHATINELLA BULIMOIDES. 157

with a white line or a band, another sometimes at the periph-
ery. The spire is usually darker, becoming chestnut or light-
brown at the summit. All but two of a large series in M'r.
Spalding's collection 'are dextral. Length 20, diam. 12.1,
aperture 10.2 mm. ; 6 whorls.

PI. 33, figs. 5 to 5c., coll. by Spalding. In Kaliuwaa valley,
about three-fourths of a mile east from the following colony
there is a patch of bananas upon which the shells vary from
about the color of the preceding colony to ivory yellow or
cream buff with or without two dusky bands; sometimes
white above the upper band. The spire is usually bicolored,
summit indian red or brown. Shells both sinistral and dex-
tral, mainly dextral. Some specimens resemble obliqua and
oomorpha.

PI. 33, figs. 6, 6a, coll. by Spalding. At the bottom of Kaliu-
waa, along the stream back of a cabin, there is 'a colony sim-
ilar to the banded pattern of ovata except that the embryonic
whorls are darker, usually dark viiiaceous. This colony is
mainly sinistral, but contains some dextral shells.

95. A. BULIMOIDES SPADICEA Gulick. PI. 33, figs. 13, 13a.

The shell is sinistral, glossy, embryo mars brown, blackish
at the tip, becoming lighter on succeeding whorls, the last
whorl pale or light orange-yellow above, below the periphery
burnt sienna shading to chestnut downwards, and showing
faint traces of two darker bands. Sutural line white, extend-
ing to the apex. Lip chestnut, bordered outside with the
same; outer edge of the columella raised; columellar fold
white or tinted. The later whorls show only the weakest
traces of spiral sculpture.

Length 17.6, diam. 10.5, aperture 9 mm.; 6 whorls (type).

Length 17.5, diam. 11, aperture 8.5 mm.

Oahu: Kahana, on trees (Gulick, type loc.) ; Hakipuu (Gu-
lick) .

Achatinella spadicea GULICK, Ann. Lye. N. H. of N, Y., vi,
p. 247, Feb., 1858. BALDWIN, Catalogue, p. 6 (" Hauula to
Kahana ").

" The white sutural band is continued up to the very tip

158 ACHATINELLA BULIMOIDES.

of the nuclear whorls in a manner quite different f roin what
is seen in the bands of A. obliqua and other species of this
group ; (Gulick) .

This small form is like mistura in having a dark summit,
but differs by the white sutural line running to the apex and
the color of the shell. I have been told that spadicea occurs
below the falls in Kaliuwaa, but I suspect that the shells from
there are mist lira. Spadicea is known by very few specimens.
There is one, the type, in coll. Boston Society of Natural His-
tory, described above and figured in pi. 33, fig. 13, and two
from it-he Gulick collection in A. N. S. P., one of which is fig-
ured, pi. 33, fig. 13(/. Also two shells from the same source in
coll. Bishop Museum. I do not remember seeing specimens in
other Hawaiian collections, and very likely the race is now
extinct. It has resemblance to some dark forms of A. rosea in
surface gloss and sutural border.

9c. A. BULIMOIDES OBLiQUA Gulick. PI. 28, figs. 8 to 9a.

" Shell sinistral, subperf orate, ovate, obliquely truncated
at the base, solid, shining, striated ; o'f ashy lead-color, with a
broad whitf band beneath the suture. Apex obtuse, of a yel-
lowish white color ; spire convexly conical ; whorls 6, subconvex,
the last large. Columellar fold central, white, strong. Aper-
ture oblique, truncately auriform, nearly white within. Peri-
stome 'white, thickened within; with external margin some-
what reflected, arcuate; columellar margin forming an ob-
tuse, sinuous ridge with a small umbilical cleft behind it ;
parietal margin thin. Length 23, diam. 13 mm. Average
weight 9 grains ' (Gulick).

Kahana, on trees (J. T. Gulick).

Achatinella obliqua GULICK, Ann. Lye. N. H. of N. Y., vi,
p. 245, pi. 8, f . 63, Feb., 1858 ; Evolution Racial and Habitu-
dinal, p. 41, pi. 3, f. 16/i-20fo. Achatinella oomorplia GULICK,
Ann. Lye., vi, p. 246, pi. 8, f. 64.

Is allied to A. ovata Newc., with which it is associated
geographically, but differs from it by its broader form and!
white lip. The dark coloring of the body-whorl abruptly ter-
minates near the external margin of the peristome, leaving a

ACHATINELLA BULIMOIDES. 159

white border aibout an eighth of an inch in width 'around the
outside of the lip, whereas in A. bulimoides and other allied
species the color 'becomes darker on this portion of the shell
though the lip itself 'be white. Dextral 'specimens are very;
rare. The color of the darker portion of the shell varies in
different specimens from a dirty cream to slate color, but the
ashy color given in the description is the most common ' (Gu-
lick).

The color of the darker, lower portion of the shell varies
through several shades resembling purple-drab, viiiaceous
purple and slate-drab of R-idgway's Color Standards. These
tints are produced by films of white laid over a brown layer,
which may be exposed by scraping. The lip is white or nearly
so typically, but in one lot it is fleshy-brown. The white
border behind the lip mentioned by Gulick is by no means con-
stantly present; often the dark color runs to the lip, which
may have a narrow t>rown edge outside, but not so wide or
dark as in ovata. The summit is usually light yellowish-
brown ; sutural margin well defined, bluish- white. Lip some-
what expanded, as in A. ovata, having a moderate callus
within. The parietal callus is generally imperceptible. Be-
tween 3 and 4 per cent of the shells seen ( are dextral. Length
22, diam. 13 mm. ; 6% whorls.

Occasionally the dark color is restricted to a peripheral
band, or it may disappear 'altogether, leaving the shell pure
white. More often some 'brown remains on the spire. In a
few shells the parietal callus is rather thick and brownish at
the edge. The above notes are from the large series collected
by Gulick.

A. 1}. obliqua was described from a colony in which the sin-
istral form very largely predominated. Mr. Gulick described
shells from a colony in which the dextral form prevailed as
A. oomorpha. While not exactly like typical obliqua, it does
not seem sufficiently differentiated to require 'a name. The
original account follows.

A. oomorplm Gulick. PI. 28 ? figs. 10 to lOc. " Shell dex-
tral, perforate, ovate, solid, shining, striated ; ash or ash-brown
with two obscure brown bands, white beneath the suture.

160 ACHATINELLA BULIMOIDES.

Apex rather obtuse, chestnut-brown ; spire convexly conical;
suture margiiiate, moderately impressed; whorls G 1 /^, convex.
Columellar fold central, white, strong. Aperture truncately
auriform, white within; peristome thickened within, with ex-
ternal margin slightly reflected anteriorly, arcuate, white or
'brown ; columellar margin reflected, detached, white ; parietal
margin very thin. Length 20 1 /, breadth 11% mm. Average
weight 7 grains. Kaliana, Oahu, on trees ' (Gulick) .

; Sinistral specimens are sometimes found which resemble
A. obliqua, but are readily distinguished by the darker color-
ing around and upon the lip. I have from Hauula a few spec-
imens which seem to belong to this species. Some of them
are nearly white ' (Gulick).

This dextral form of obliqua has the same peculiar colors,
drab, dull-brown overlaid with lilac, etc., but also sometimes
the chestnut-brown of A. ovata. It has another ovata char-
acter in the two-banded pattern of many shells of the typical
colony. This pattern is not found in the typical colony of
A. obliqua, judging by a lot of about 50 shells seen, though it
occurs in a smaller brown-lipped lot in the Gulick CGJ lection.
The form oomorpha often has exactly the coloration of typical
obliqua, except that the lip is brown both externally and
within. PL 28, fig. 10a is the typical oomorpha pattern ac-
cording to Gulick 's figured type, two j banded over a streaked
ground. Shells also occur having the last whorl white. The
length varies from 19 to 22.5 mm., but most specimens before
me are smaller than obliqua.

9c7. A. BULIMOIDES OVATA Newcomb. PI. 28, figs. 1 to Id, 2 ;
pi. 29, figs. 2, 2a ; pi. 33, fig. 2.

' Shell dextral, elongate-ovate, polished, finely striated lon-
gitudinally. Color light flesh-colored above, last two whorls
white, obscurely banded with light-brown. "Whorls 6, convex;
suture slightly impressed, margined. Aperture subovate;
columella twisted into a plait, slightly callus; lip slightly re-
flected, dark-brown. Length 0.9 inch. Breadth 0.45 inch.
Variety a : shell white, bluish-white above, without coloring or
bands. Habitat, Waiauai, Oahu : (Netvcoml), May, 1853).

ACHATINELLA BULIMOIDES. 161

" Shell dextral or sinistral, elongately ovate; whorls 6,
slightly margined above, rounded; suture moderately im-
pressed. Aperture subovate, entirely margined with black;
columella short, plicate, strong and twisted ; lip thickened and
slightly expanded. Color of shell pure white or yellowish-
white, with or -without obsolete brownish bands above. Length
eighteen-twentieths, diam. ten-twentieths inch. Hab. Kahana,
Koolan, Oahu ' (Newcomb, 1854).

Oahu: Kahana (Newcomb) ; Kahana and Hakipuu (Gu-
lick) .

Achatinella ovata NEWCOMB, Ann. Lye. N. H. of N. Y., vi,
p. 22, May, 1853 ; P. Z. S., 1853, p. 130, pi. 22, f . 2, 2a, 1854.
GULICK, Evolution, Racial and Habitudinal, p. 41, pi. 3, f.
6/1-15/t. Achatinella Candida PFR., P. Z. S., 1855, p. 2, pi. 30,
f. 4; Monogr., iv, 519. Achatinella vidua PFR., P. Z. S., 1855,
p. 3, pi. 30, f. 10; Monographia, iv, 522. Cf. Newcomb, Ann.
Lye. N. H. of N. Y., vi, p. 324. A. ivheatleyi NEWCOMB, Ann.
Lye., vi, pp. 147 (1855) and 324 (1858), not described (see
under A. elegans ivheatleyana). A. fricki var. d, PFR., Mon-
ogr., iv, 521 ; P. Z. S., 1855, pi. 30. f . la.

Dr. Newcomb included both white and banded forms in his
description and figures, which are reproduced on pi. 29, figs.
2, 2a. They occur in the same hybrid colony in Kahana, pi.
28, figs. 1 to \d ; pi. 33, fig. 2. The surface is not very glossy,
often rather distinctly striate. The apex is light colored, yel-
lowish, pale-brown or white. The suture has no dark border
below, and the impressed line defining the margin is usually
rather weak, sometimes wanting. Outside the lip has a dark
border behind, as in pi. 28, fig. 1&, though not always so fully
developed. The lip is distinctly expanded outwardly and es-
pecially below, and has a 'chocolate edge, both outside and
within. The internal callus is rather strong. In all color-
forms 'at Kahana the shell may be either dextral or sinistral.

Length 23, diam. 13 mm. ; 6% whorls.

Length 22.5, diam. 13.5 mm.

Length 21.5, diam. 12 mm.

Length 18, diam. 11 mm.

Banded specimens would probably prove smaller than white-

162 ACHATINELLA BULIMOIDES.

in the average. The following color- forms occur in a lot from
Kahana, Gulick collection.

1. White, initial whorls often yellowish, lip .chocolate (f. 1).
la. Same, but whorls of spire with, a dark band (fig. la).
16. Brownish-cream color, spire as in la (fig. 16).

2. "White or whitish, with two chestnut bands separated by
a peripheral white band; spire with a wide chestnut band
above the suture (fig. Ic).

2a. Similar, but base entirely chestnut, usually lighter than
the bands (fig. Id).

Forms la and 16 are blends between 1 and 2.

In a lot of 49 specimens from Hakipuu, Gulick coll., all are
dextral. Color-forms fig. 1 and no. 2a, chestnut streaked with
darker, with a broad white band below the suture (pi. 28, fig.
2, Hakipuu) predominate, though there are some of the pat-
tern of fig. Ic.

In another lot of 7 dextral shells in the Gulick collection
from Kahana the lip is white, slightly yellowish at the edge.
In three of them there are pale traces of brown banding,
chiefly on the spire ; the others being pure white throughout
(pi. 28, fig. 7).

A. Candida Pfr. is generally admitted to be merely the white
form of ovata. The original figure is reproduced on pi. 30,
fig. 4. The description follows. Achatinella Candida Pfr.
Shell dextral, imperforate, ovate-conic, solid, striatulate, a
little glossy, pure white; spire convexly conic, the apex
minute, black, rather acute; suture light, thread-margined.
Whorls 6i/2, rather flat, the last 'three-sevenths to four-ninths
the length, obsoletely angular, rounded at base. Aperture
oblique, truncate-auriform; columellar fold moderate, super-
ior, oblique, subcompressed ; peristome brown-violaceous,
strongly labiate within, the right margin expanded, columellar
margin dilated, flat, thick, adnate. Length 22, diam. 11 mm.,
aperture 10% x 5 mm. Sandwich Is., Frick (Pfr.).

It is likely that all the patterns of A, fricki Pfr., except pat-
tern a, were based upon ovata. Pfeiffer's figs, la, 76, repro-
duced in my pi. 30, figs. 7a, 76, certainly look to me like ovata.
I do not feel competent to pronounce upon fig. 7, which has

ACHATINELLA BULIMOIDES. 163

been taken as the type of fricki, as I have not seen the shell,
and the figure is not closely matched among the specimens I
have seen.

A. vidua Pfeift'er, of which the original figure is reproduced,
pi. 30, fig. 10, is in my opinion merely a very small or stunted
specimen of ovata. Dr. Neweomb, who examined the type in
Mus. Cuming, decided it ' to 'be a somewhat worn and faded
specimen of this species ' (A. ovata). The coloration is ex-
actly that of some shells in Gulick 's KaJiana series, such as pi.
33, fig. 2, and pi. 28, fig. Id out of the same lot. Adult shells
in this lot of ovata range from 19.3 to 22.8 mm. long. I do
not believe that vidua has any racial status. The description
follows : ' * A. vidua Pf r. Shell dextral, subimperf orate, conic-
ovate, striatulate, under the lens most minutely decussate,
glossy; deep brown, two-banded with white. Spire convexly-
conic, the apex somewhat obtuse; suture nearly simple.
Whorls 6, very slightly convex, the last a little shorter than
the spire, rounded basally. Aperture a little oblique, trun-
cate-oblong; columellar fold obsolete, slightly twisted. Peri-
stome strongly labiate, the right margin narrowly refiexed,
columellar margin dilated, subadnate. Length ISi/o, diam. 10
mm., aperture 9 mm. long, 4% wide inside ' (Pfr.).

The weakness of the columellar fold, noticed by both Pfeif-
fer and Sykes, can be paralleled in the large series of ovata
before me. The smallest ovata I have seen is 17^ mm. long,
dextral, with the pattern of pi. 28, fig. Ic.

A. BULIMOIDES ROTUNDA Gulick. PI. 28, figs. 3 to 6.

Very dark chestnut or chocolate, with a white or pale tawny
zone around the upper part of the whorls ; glossy ; suture very
distinctly margin-ate in the last 21/2 whorls, dark-bordered
above and below, apex dark at the tip. Whorls rather swol-
len, sinistral. Lip moderately thickened within, but less
than in A. ovata, deep brown. Columellar margin gener-
ally less raised outwardly than in ovata, the columellar fold
brown or white ; parietal callus a very thin film.

Length 21, diam. 13 to 13.6 mm. ; 6!/4 whorls.

Kaaawa and Kahana (Gulick), on the dividing ridge.

164 ACHATINELLA BULIMOIDES.

Achatinella rotunda GULICK, Ann. Lye. N. H. of N. Y., vi r
p. 248, pi. 8, L 67, Feb., 1858.

As it occurs 011 the Kaaawa valley side (figs. 3, 4), this form
is so distinct that it might well be given specific rank. Some
Kahana lots however, figs. 5, 5a, 6, show every pattern linking
rotunda with ovata. Gulick remarks that ' ' there is a variety
with two black bands -which resembles certain forms of A.
ovata Newc., but is distinguished by its black suture, and its
thinner and more regularly arcuate lip. Its bands are also
deep black, while those of A. ovata are brown as in A. buli-
moides Swains.' Having before me several large lots from
Kahana labeled rotunda by Gulick, I find that the subsutural
dark border is as often absent as present, and the bands vary
in shade, often being quite as light as in ovata. Most of this
Kahana lot are more lengthened shells than the Kaaawa ro-
tunda, but some, such as fig. 6, are typical in shape. These
colonies are clearly ovata X rotunda hybrids.

A typical rotunda from the type locality, Kaaawa, is shown
in pi. 28, fig. 3, Gulick coll.

9/. A. BULIMOIDES GLABRA Newcomb. PL 29, fig. 25 ; PI. 33,
figs. 8 to 11.

" Shell -conically ovate, glossy, whorls 6, somewhat inflated,
margined above ; suture well marked. Aperture ovate ; lip
slightly subreflected, dark-brown edged with black, thickened
within. Columella short, robust, expanded, 'and terminating
in a twisted plait. Color bluish-slate, much lighter on the
upper portion of the whorls; sometimes transversely banded
with white or -chestnut. Length nine-twentieths, width seven-
teen-twentieths of an inch : ' (Newc.).

Oahu : Kolaupoko (Newc.); Kawailoa and Waialee (Gu-
lick) ; Waimea (Spalding).

Achatinella gldbra NEWC., P. Z. S., 1853, p. 139, pi. 22, f.
25 ; P. Z. S., 1854, p. 310. PFB., Monogr., iv, 520. THWING,
Orig. Descript., p. 93, pi. 2, f . 2. B [ulimella] glabia, Newc.,
HARTMANN, Proc. A. N. S. Phila., 1888, p. 29 ("Kawaiawa").

Newcomb 's type figure (pi. 29, fig. 25) shows a shell with
banded spire and light colored base. He subsequently (1854)
stated that A. glabra is always dextral.

ACHATINELLA BULIMOIDES. 165

Some shells received from Newcomb are whitish, becoming
streaked with purplish-drab on the last whorl, with a pair of
brownish- drab bands at the periphery, lip with a dark-brown
edge (pi. 33, fig. 9).

The specimens before me from Waialee (pi. 33, fig. 11),
Waimea (pi. 33, figs. 8 to 8c) and Kawailoa (pi. 33, figs. 10,
lOo-) show a good deal of variation in color; all collected by
Gulick.

1. Purple-brown, with a belt of light reddish-brown below
the suture, upper whorls reddish or pale, lip fleshy-brown.

2. Base reddish-chestnut, upper third of the last whorl yel-
lowish or white.

3. Base flesh-colored, upper part white, a dark band be-
tween, lip white.

4. Purplish-fleshy with indistinct whitish streaks, lip fleshy.
The preceding are Waimea shells, the same patterns occur-
ring in Kawailoa. In Waialee I note an additional pattern.

5. Similar to no. 3, but whorls of the spire banded; lip
brown. This is the typical pattern.

It will readily be seen that glabra inter grades in some of
its .color-forms with bulimoides, but both glabra and buli-
moides have other color-forms special to each. The area of
glabra overlaps that of bulimoides in part, and extends thence
to the western end of the range.

Newcomb 's type locality, " Koolaupoko, ' is an extensive,
district on the north side of that portion of the Main Range
north and northeast of Honolulu (koolau signifying the wind-
ward or northern side, as kona the lee or southern district, on
any of the islands). Later collectors have found nothing re-
sembling A. glabra in the eastern part of the island ; but Gru-
lick, Spalding and others have taken the species in valleys of
the northwestern end of the range. There can be no doubt
that Newcomb was mislead as to the habitat of A. glabra.

Var. ( ?) fricki Pfeiffer. PI. 30, fig. 7.

" Shell dextral or sinistral, subimperf orate, oblong-ovate,
solid, lightly striate, glossy, of very various colors; spire a
little convexly-conic, the apex acute; suture narrowly mar-
ginate ; whorls 6, a trifle convex, the last about equal to three-
sevenths the total length, rotund at base. Aperture oblique,

16G ACHATINELLA ELEGANS.

reversed ear-shaped; columellar fold above, twisted, strong;
peristome slightly expanded, Obtuse, labiate and bordered
with violaceous or black. Length 20-21, diam. lO 1 /^ mm. ;
aperture 10 mm. long, 5 wide.

" a. Isabelline, subfasciate with pale-brown, suture white.

" b. Bright chestnut, ornamented with darker 'and white
bands, sinistral.

" c. Gray-brown, ornamented with darker bands, white
above.

" d. AYhite or buff, ornamented with two or three black-
brown bands.' (Pfr-}

Sandwich Islands, Frick in coll. Pfr.

Achatinella fricki PFR., Proc. Zool. Soc. Lond., 1855, p. 3,
pi. 30, f. 7, la, lb- Malak. Bl., 1855, p. 5; Monographia, iv,
521. Cf. SYKES, Fauna Hawaiiensis, Mollusca, p. 307, under
A. glabra.

A. fricki has not been recognized by Honolulu concholo-
gists. Pfeiffer described and figured several varieties. Hh
first figure (see pi. 30, fig. 7), which may be ta.ken for his
type, has been considered to be A. glabra Newc. by Mr.
Sykes. I have not been able to match it at all closely among
the glabra I have seen. It differs from pulcherrima by hav-
ing a white sutural band. This band I have never seen in a
dark pulcherrima, and I therefore give fricki temporary place
as a variety of glabra, following Mr. Sykes.

Pfeiffer 's color-form b and those following are probably
different subspecies. His second figure (reproduced in pi. 30,
fig. la) has been referred by Sykes to A. ovata Newc., I be-
lieve correctly, as it agrees fully with some of the old ovata
received from Newcomb. Pfeiffer's var. b (see pi. 30, fig. lb)
is a sinistral shell, not exactly matched by any specimen I
have seen, and probably not the same as typical fricki. The
contiguous dark bands bordering a peripheral white band
show it to belong to the bulimoides group, and it is quite pos-
sible that this figure also is an ovata.

10. A. ELEGANS Newcomb. PI. 28, figs. 12 to 13d; pi. 32, fig.

15.
" Phell conically-elongate, polished, shining, rather solid;

ACHATINELLA ELEGANS. 167

whorls 6, plano-convex, margined above ; suture well im-
pressed. Aperture subovate ; lip white, expanded, subrs-

<

fleeted, somewhat contracted in its center, thickened within;
columella short, flat and lightly toothed. Color light and
dark-brown alternating, longitudinally arranged in lineations
or broad patches; sometimes with a white sutural band and
an additional one on the body-whorl. Length eighteen-twen-
tieths, diam. eight-twentieths inch.' (Newc.)

Oahu: Hauula (Newcomib) ; Hauula, Kaliuwaa and Ka-
huku (Gulick) ; Hauula and Kaipapau (Baldwin).

Achatinella elegans NEWC., P. Z. S., 1853, p. 149, pi. 24,
f. 57, 1854.

This species differs from all forms of bulimoides by its
streaked coloration, a pattern not found in bulimoides. It is
also more slender, in the average, but occasional specimens
are as broad and compact as bulimoides.

It is said to be now extinct, having passed with the lower
forests of the Hauula region, but it was an abundant shell in
1850-55, when Newcomb and Gulick were collecting. Besides
the lots in these collections I have seen a beautiful series in
the collection of Mr. J. S. Emerson.

Large series from. Newcomb and Gulick, taken in Hauula,
show considerable variation in the pattern. Usually the shell
is streaked in the direction of growth lines with brown (wal-
nut-brown to brownish- vinaceous of Ridgway's Color Stand^
ards), the shade variable, but usually appearing overlaid
with white, sometimes dull chestnut-brown. The typical
form as figured by Newcomb has the streaks; cut by white
bands at suture and periphery as in pi. 28, fig. 13a, but often;
one or both of these white bands is lacking. Newcomb 's
description applies to the form shown in pi. 28, fig. 12. Be-
sides these prevalent patterns, there are a few additional
white spirals in some shells, and sometimes the dark streaks
are reduced or blurred. Rarely the streaks are confluent, the
brown color almost evenly diffused. There are also a few
pure white shells (fig. 13c) . In a set collected by Dr. New-
eomb I note rather indistinct dusky bands above and below
the periphery, sometimes parted by a light-brown peripheral

168 ACHATIXELLA ELEGANS.

band. The lip is sometimes white, more often fleshy-brown
or darker. The white band below the suture sometimes runs
to the very apex, as in A. l>. spadicea, but oftener not. The
early whorls are usually dull-brown. Dextral shells largely
prevail, but all of the color-forms are common to both dextral
and sinistral. In one lot from Hauula there are 82 dextral,
23 sinistral shells.

A set of 8 from Kahuku are very solid, coarsely streaked
throughout, without bands (pi. 32, fig. 15, coll. by Gulick).

An unusual color-form, pi. 28, fig. 13cZ, has a pattern re-
sembling some specimens of oomorpha or obliqua somewhat.
The streaking of elegans is only weakly shown in places on
the spire. The sutural white line runs to the apex, as in
spadicea and some specimens of elegans. In its compact
shape this shell is similar to several characteristically streaked
elegans in the same Hauula lot, It is obviously an extreme
form of elegans.

Color- variety inelegans n. var. PI. 33, fig. 12. The shell is
dextral, rather thin; glossy, rather distinctly striated spir-
ally to the last whorl, which is smoother. Embryonic whorls
corneous, following whorls cinnamon with white sutural
border, last whorl pale cinnamon, shading to darker at the
base, and having two orange-cinnamon bands, one above,
the other below the periphery; sutural band pure white, be-
ginning on the third whorl. Peristome a little expanded ait
base, very little thickened within, edged with dark-brown.
Columellar fold weak, whitish. Length 19, diani. 10, aperture
9.5 mm. ; 6 whorls.

Kaliuwaa, J. T. Gulick, no. 92504 A. N. S. P.

This shell was labeled A. spadicea by Mr. Gulick, but it is,
in my opinion, a form of the elegans stock with less modified
coloration than elegans. It does not seem closely related to
the Bulimellas of upper Kaliuwaa, above the falls.

10a. A. ELEGANS WHEATLEYANA n. var. PI. 28 7 figs. 11, lla,
115.

The shell is cinereous, blue-gray or slightly purplish, rarely
purplish-brown, with faint whitish streaks and white or pale-

ACHATINELLA ELEGANS AND FUSCOBASIS 169

brown band above the periphery, sometimes bisected by a
dark line; suture white-bordered; summit light yellowish.
Aperture pale blue within, the lip thickened within, with a
narrowly but distinctly expanded, acute brown edge. It is
also narrowly bordered with brown outside. The strong coiu-
mellar fold is white. Length 21, diam. 12 mm. or smaller,
length 19.2, diam. 11 to 12 mm.

Punaluu (Gulick).

This race approaches A. obliqua somewhat, but seems
nearer elegans. It has the maroon-brown or purplish-brown
layer overlaid with a streaked white film, giving various tints
difficult to name, but near dutch blue, deep madder blue and
slate purple of Ridgway's Color Standards. The whitish or
pale-brown band above the periphery is nearly always pres-
ent, and often more conspicuous than in fig. lla. Sometimes
(fig. Ha) there is a thin yellow cuticle, but usually none.
One lot of 8 from Newcomb, without habitat, consists of sinis-
tral shells. Another lot from Gulick has 3 dextral shells in a
total of 22. It is probably extinct now.

This form was named for Charles M. Wheatley, the well-
known collector of Phoenixville, Pennsylvania.

"A. wheatleyi' was mentioned twice by Newcomb: in
Annals of the Lyceum of Natural History of New York, vi,
p. 147, October, 1855, where he says " A. wheatleyi Nob. is
A. vidua Pfeift'er, ' and again in the same volume, p. 324,
September, 1858, " The A. vidua Pfr., which I had supposed
was my manuscript A. ivheatleyi, I find in Mus. Cuming to be
a somewhat worn and faded specimen of this species ' [A.
ovata] . As no characters were ever assigned to A. ivheatleyi
except by implication in the above remarks, we have no
option but to accept Newcomb 's published statement of what
his A. wheatleyi was, even though he sent out specimens of
what we are now calling ivheatleyana as A. wheatleyi.

SERIES OF A. FUSCOBASIS.

Ovate shells, smaller than those of the bulimoides series;
mainly sinistral, never marked with green, and having some
resemblance to the cast a series of Achatinellastntm. They
are almost entirely shells of the high ridges and peaks.

170 ACHATINELLA FUSCOBASIS.

11. A. FUSCOBASIS (E. A. Smith). PI. 35, figs. 1 to 4.

Shell ovate, sinistral, glossy ; white, the last whorl yellow-
ish, ornamented with a median zone and base of brown.
Whorls 6, a little convex, suture distinctly margined. Aper-
ture white ; peristome thick, brown, columellar fold strong.
Length 16, diam. 10 mm. High up on Mt. Kaala on the Mo-
kuleia side, on the island of Oahu, arboreal. (Smith.)

Oahu: Head of Kuliouou-Niu division ridge to Mt. Olym-
pus, -at the head of the Palolo-Manoa ridge (Spalding. Cooke,
Kuhns and Wilder).

Bulimella fuscoliasis SMITH, P. Z. S., 1873, p. 77, pi. 9, f.
15. Achatinella fuscobasis Sin., THWING, Original Descrip-
tions, etc., p. 83. Achatinella luteostoma BALDWIN, Proc.
Acad. Nat. Sci. Phila., 1895, p. 217, pi. 10, f. 7, 8 (Palolo to
Niu). B. rosea, a small white variety with a yellow lip,
HARTMAN, Pro<c. A. N. S. Phila., 1888, p. 30, pi. 1, f. 4.

A. fuscoliasis has the white ground of A. t&niolata, but it
is invariably sinistral in a large number I have seen in var-
ious collections. It is also smaller, with the lip not so thick,
the columellar margin less raised, and it has a different range
of patterns. The reddish-brown color (mars orange, or burnt
sienna, rarely chestnut or light orange-yellow of Ridgway's
Color Standards) of the peripheral and basal bands, and the
brown peristome are also characteristic of the typical form.
The following color-forms occur in most colonies.

(a) Pure white with brown (liver-brown to dull-chestnut)
peristome. Typical color-form of A. luteostoma (fig. 2).

(&) Peripheral band and basal patch brown, the space be-
tween them white or yellowish-brown. Typical color-form of
A. fuscolasis (pi. 35, fig. 1, type-specimen).

(c) Same as 1), but a basal band in place of the patch (fig.

3).

(d) Same as 5 or c, except that there is a light-brown band
near the suture. In one shell from Mt. Olympus in the
Spalding collection the upper and lower bands are broad,
light-brown, the peripheral band narrow, darker.

(e) Whole base brown, dark or light (fig. 4).

Most of the specimens from Kuliouou (Spalding and

ACHATINELLA FUSCOBASIS. 171

Thaanum collections), fig. 2, are pure white, tout there are a
few of the patterns of figs, 1 and 3.

Mr. Smith's figured type-specimen, no. 110 Gulick coll.,
Boston Soc. N. H., measures length 15.2, diam. 9.3 mm. It
is shown in pi. 35, fig. 1. Other shells measure:

Length 16, diam. 9.2 mm.; 16x9.5 mm. (near top of Mt.
Olympus; pi. 35, figs. 3, 4).

Length 17, diam. 10.5 mm. ; 17.5 x 10 mm. ; 16.2 x 9 mm.
(Kuliouou).

The figured type of A. luteostoma, no. 65704 A. N. S. P.,
measures 16 x 9.7 mm., and one of Baldwin's examples 14.6 x
9 mm.

The several color-forms occur in the same colonies. I do
not know that any of them has been found as a pure race.

A. luteostoma Baldwin was based upon the white form of
fuACobasis, but Baldwin also included the banded shells in his
original account, which follows. " Shell sinistral, imper-
f orate, solid, ovate, spire convexly conical, apex obtuse; sur-
face shining, marked with fine growth-lines, under a strong
lens seen to be decussated by close, extremely minute spiral
stride, apical whorls smooth. Color white, with a reddish-
yellow lip. Whorls 5!/2, margined above, slightly 'Convex.
Suture moderately impressed. Aperture oblique, sdnuately
oval, white within. Lip obtuse, thickened within, columellar
margin very slightly reflexed, extremities united by a very
thin yellowish -callus. Columella reddish-yellow, terminating
in a strong tortuous fold. Length 15% , diam. 9% mm -

' ' In occasional examples the basal portion below the
periphery is light-brown, and sometimes a bright chestnut
spiral band encircles the periphery : (Baldwin) .

The types of both fuscobasis and luteostoma were probably
from somewhere in the head of Palolo valley. The locality
" M't, Kaala " given in the original description of fuscoibasis
was undoubtedly an error, as no such shell has been found in
the Waianae range. A native boy probably brought the
species to Gulick, who did not himself collect high in the
mountains. The two original specimens of fuscobasis and the
figured types of luteostoma are now before me.

172 ACHATINELLA FUSCOBASIS.

A. fuscobasis is a species of the high ridge and peaks of
the main range. Northeast of Mt. Olympus, on the ridge
leading to Konahuanui, it passes into an undifferentiated
form having the brown or 'brown-tinted lip of fuscobasis and
the more capacious shape of lyonsiana (which also rarely
shows some brown on the lip, in the type locality) . An in-
spection of series in the collections of Messrs. Spalding,
Thurston and Wilder convinces me that the Honolulu con-
ch ologists are right in regarding lyonsiana as a variety of
fuscobasis.

lla. A. FUSCOBASIS LYONSIANA Baldwin. PI. 35, figs. 5, 6, 7.

" Shell smistral, imperf orate, solid, ovate; spire convexly
conical, apex obtuse ; surface shining, striated with delicate
growth-lines; apical whorls smooth. Color white, with two
reddish-brown bands, one encircling the base, the other pass-
ing around the periphery and revolving on the spire just
above the suture ; the two bands are sometimes confluent, and
often the shell is uniform white, without bands. Whorls 6,
narrowly margined above, somewhat 'convex; suture moder-
ately impressed. Aperture oblique, white within, sinuately
oval ; peristome w T hite, rather obtuse, thickened within, basal
and columellar margins slightly expanded, extremities united
by a very thin callus ; columella terminating in a strong, tor-
tuous, white fold. Length 17, diam. ll 1 /^ mm -

' ' Animal : Mantle black, sometimes mottled with white,
margined with a narrow line of lighter shade. Superior and
under portion of foot light brown. Tentacles, tentacular
sheath, and front above very dark brown ' (Baldwin).

Oahu: Mt. Konahuanui, at about 3,000 ft. elevation (type
loc.) to the top, spreading on the ridge southeastward, where
it passes into fuscobasis.

Achatinella lyonsiana BALDWIN, Proe. A. N. S. P. Pliila,
1895, p. 218, pi. 10, f. 9, 10, 11 SUTER, t. <c., p. 239, pi. 11,
f. 52 (teeth).

A larger, more capacious shell than A. fuscobasis, wdth the
lip and <corurnella white. The patterns are :

(a} Pure white.

ACHATINELLA FUSCOBASIS. 173

(6) White, with upper, peripheral and basal burnt sienna
or chestnut bands, the upper usually weaker or .sometimes
wanting.

(c) A unique color-form in Mr. Spalding's collection (no.
2027) has the penultimate whorl deep chestnut, last whorl
with two narrow dark bands (ii and iii). The ordinary
brown-banded and white forms occurred in the same colony,
which is about half a mile from the top of Konahuanui, in
the main and typical locality.

The margin below the suture is invariably white, the sur-
face is very glossy, and spiral lines are scarcely visible even
on the early whorls; lip white, or rarely with a faint brown
tint. It differs from tccniolata by being sinistral, with the
outer border of the columellar lip decidedly less elevated.

Length 19.3, diam. 11.4 mm.

Length 18, diam. 12 mm.

Length 17.7, diam. 10.2 mm.

Dedicated to Prof. A. B. Lyons, formerly of Oahu College.
PL 35, figs. 5, 7 are cotypes, 65693-4 A. N. S. P. Fig. 6, from
the Irwin Spalding collection, taken half a mile from the top
of Konahuanui, somewhat approaches the following sub-
species.

. A. FUSCOBASIS WILDER: Pilsbry, n. subsp. PL 41, figs.

The shell is shaped as in A. /. lyo'nsiana. First two whorls
buff-gray; last whorl having a chestnut or blackish- chestnut
peripheral band and basal area, the latter split into bands in
some colonies. Above the peripheral band there is a white
band; below it a wider white or mars-yellow band. The rest
of the upper surface is mars^yellow, mars-orange or burnt-
sienna, usually excepting a sutural band or line of white.
Lip and aperture fleshy in various tints, the lip narrowly
thickened within. Columellar margin but slightly or not
raised ; parietal callus milky or transparent, very thin.

Length 17, diam. 10 mm.

Length 15.2, diam. 10 mm.

Summit of Lanihuli, at head of the Nuuanu-Kalihi ridge

174 ACHATINELLA PUPUKANIOE.

(Irwin Spalding, W. D. Wilder). Also " Mauna Kope " a
peak at head of the Kalihi-Moanalua ridge (Wilder).

This beautiful shell is the most ornate of the fuscobasis
series. It is isolated from lyonsiana and fuscobasis by the
deeply cut Nuuaiiu valley, which even at the pali is lower
than any form of fuscobasis has been found. The Lanihuli
colony is practically a pure race ; the splitting of the basal
patch into bands in one lot from Mr. Wilder, and the varia-
tion in the subsutural white border being fairly referable to
" fluctuating variation.' In the specimens taken by Spald-
ing on the Koolau side of the peak the basal dark area is un-
split (fig. 2).

On the peak at the head of the Kalihi-Moanalua ridge there
are shells with the typical pattern, a few with darker periph-
eral and basal bands on a tawny ground (pi. 40, figs. 3, 3&)
and others with a tawny ground and blackish-chestnut base
(A. castanea pattern), the subsutural border white. Both of
these are rare mutations.

12. A. PUPUKANIOE Pilsbry & Cooke, n. sp. PI. 35, figs. 14
to 17.

The shell is dextral, conic, the greatest diameter near the
base, solid, brilliantly glossy; uniform white, ivory yellow
with white sutural line, or either of these tints- with a burnt
sienna band immediately above, a wider and darker one
below the periphery. Suture margined but without a dark
line. Spire white. Lip not expanded, the edge brownish,
internal rib white, or sometimes the whole lip is pale-pink.
Columellar fold rather strong and abrupt, white.

Length 16.3, diam. 9.7, aperture 7.8 mm. ; 6 whorls.

Length 16, diam. 9.2, aperture 8 mm.

Oahu : Crest of the Waimano-Manana ridge, 1 to 1% miles
from its junction with the main range. Types 108068 A. N.
S. P., Pilsbry, 17-11-1913 ; also head of Waiawa valley, Irwin
Spalding.

This species, which I collected in company with Messrs.
Spalding and Merriam, has a great resemblance to A. fusco-
basis ; but that is invariably sinistral, and separated from tho

ACHATINELLA SOWERBYANA. 175*

habitat of A. pupukanioe by about a fourth of the length of
the island, wherein no similar form has been found. It also
resembles A. s. dextroversa, of the northwestern end of the
range. A. cast a, in some Waimano lots, resembles pupu-
kanioe in coloration, but is more lengthened and sinistral.
The resemblance is so 'dose, however, that I would refer tli3
new form to casta as a variety, were it not that old specimens
(such as pi. 35, fig. 15) have the columellar lip built up to
form a ledge, exactly as in Bulimella, and unlike any of the
very long series of casta I have seen.

The apical whorls are rather deeply eroded in all of the
adult shells from the type locality. In a young one they are
cartridge-buff with a white band above the suture.

In Mr. W. D. AVilder's collection a series from the 'type
locality contains several aberrant individuals : streaked with
burnt-sienna on a yellow ground, and others with the same
pattern interrupted by broad white zones at suture, periph-
ery and base.

A long series of smaller specimens was taken by Mr. Spali-
ing in a very limited area on the southern slope of the peak
at the northern head of Waiawa valley. White shells are in
the majority, but some have a basal band (iii), and in one
before me there are two faint bands (ii and iii), and thy
eroded embryonic whorls are brown. Two measure :

Length 14.3, diam. 8 mm.

Length 16, diam. 8, aperture 7.7 mm.

13. A. SOWERBYANA Pfeiffer. PL 30, figs. 14, 14 ; pi. 34, figs.
9, 10.

Shell sinistral, imperforate, conic-oblong, rather solid,
smoothish, having a gummy gloss; tawny buff, slightly
streaked with a deeper shade. Spire a little convexly-conic,
the apex subacute, suture margined. Whorls 6, very slightly
convex, the last a little shorter than the spire, rounded sack-
like at the base. Aperture oblique, inversely ear-shaped,
white within; columellar fold superior, strong, twisted, rose-
ate ; peristome rose-lipped, the outer margin shortly expanded,
columellar margin dilated, adnate. Length 18, diam. 9 mm. ;

176 ACHATINELLA SOWERBYANA.

aperture 8 mm. long, 4 wide. Sandwich Is., Frick. (Pfr.)

Oiahu : Punaluu to Kaipapau ; various varieties as far
northwest as Kahuku and Pupukea.

Achatinella sowerbyana PFR., P. Z. S. 1855, p. 4, pi. 30, i.
14; Malak. Bl., 1855, p. 65; Monogr., iv, 527. BALDWIN,
Catalogue, p. 6 (Kalihi?). Bulimella sowerbiana Pfr.,
HARTMAN, Proc. A. N. S. Phila., 1888, p. 31.

Achatinella oviformis Neweomb, PFR,, P. Z. S., 1855, o.
208; Monogr., iv, 540. NEWCOMB, Ann. Lye. N. H. of N. Y.,
vi ; p. 147 (as a synonym of A. soiverbyana Pfr.). SYKES,
Fauna Hawaiiensis, p. 309. A. multicolor PFR., in part, P.
Z. S., 1855, pi. 30, f. lla; cf. Sykes, Fauna Hawaiiensis, Mol-
lusca, p. 309.

Pfeiffer described var. 6 as " a little smaller, yellow- whit-
ish, the base chestnut or greenish.' His fig. 14a represents
this color-form, and is reproduced in pi. 30, fig. 14a.

A. soiverbyana is a handsome shell, very little known before
the last few years, but now familiar by the splendid series in
the collections of Hon. L. A. Thurston and Irwin Spalding.
It is invariably sinistral, typically wax yellow or sulphine
yellow, fading upwards, with the sutural margin brown, sum-
mit usually pale or white. The surface has the gloss of var-
nish, and the lip and columella are of a luscious pink tint
which has given it the local name of ' ' watermelon shell. ' It
is smaller and more slender than other smooth Buli-
mellas of the region. The lip-rib is narrow. It is no doubt
related to A. fuscobasis, of the eastern end of the range, but
I agree with all the island conchologists in holding the two
distinct. Like fuscobasis, it lives mainly on the heights.

Frick, who supplied Pfeiffer's type, gave no locality, but it
seems likely that he got the species in some valley of the
Kaipapau-Kaliuwaa region where it reached a lower level
than usual, and thus -came within the zone accessible to the
early collectors.

Plate 30, fig. 14 is reproduced from Pfeiffer's type figure.
PI. 34, fig. 9, from the bottom of the central ravines of Kai-
papau, and fig. 10, Kaliuwaa, near the Castle trail, are prob-
ably from as near where Frick took the original lot as a mod-

Achatinellidae

PLATE 27

la

1 b

1 d

Njjjjjg-

5e

:-

7a

7 b

10 a

Achatinellidae

PLATE 28

- H

lb

lc

9

m

13

7

9a

11

13 a

8

10

11 a

13 b

8a

10 a

i ^

fetttttiM

lib

13

8b

10 b

13d

Achatinellidae

PLATE 29

2a

8

9a

10

10 a

11

12

14

15

17

18

19

20

21

22

22 a

23

25

Achatinellidae

PLATE 30

11

15

23

7a

12

45

48 a

49

50

14 a

48

52

Achatinellidae

3 a

4d

la

^

9 a

:-,

4a

f

7b

\

9b

15

IS a

8

9c

13

15 c

Plate 31

4c

6a

14

15 d

Achatinellidae

Plate 32

la

3a

<

Ib

Jb

6a

^L***

6b

-

2a

4a

10

7b

lOb

lib

12

i

I2b

12c

13

14

15

Achatinellidae

1 f

10

la

g

3a

A

5a

8

lOa

s**

v ^*te^ w>

Ib

1 h

3b

51)

^rtB!

1C

li

3c

.

JW

r

8a

8b

n

12

Plate 33

Id

* '.; 7

3d

8c

13

1 e

6a

-

13a

Achatinellidae

Plate 34

la

3a

H

6 b

8

8 a

8b

9

5a

10

, -.

14

14 a

13

14 b

13 a

15

13 b

15 a

Achatinellidae

PLATE 35

10

8

.

v
16

Achatinellidae

Plate 36

1 f

la

Ib

5a

5b

Ic

1

'

3a

5c

Id

5d

le

4a

5 e

6a

7a

6b

7b

8a

6c

7c

9a

lOa

AchatinelHdae

Plate 37

7a

--^

s

4a

5c

-

8

i

2a

4b

8a

10

11

lla

3f

.

%

4c

h.

^

6a

8b

lib

3g

,

6b

8c

3b

3h

5a

12

Achatmellidae

Plate 38

,

X

9

V J
15

,_>

10

16

V

Ib

4a

7b

11

16a

12

17

2a

2b

13

14

18

18 a

19b

19c

Achatinellidae

Plate 39

8

8b

8c

3a

11

m

9a

lla

9b

lib

12d

9c

lie

13c

13d

13 e

13f

10

12

13 a

14

13b

14 a

ACHATINELLA SOWERBYANA. 177

em collector will ever get. The shells are either plain except
for the sutural band (pi. 34, fig. 9, Kaipapau), or have a
band around the base, as in pi. 34, fig. 10 from Kaliuwaa val-
ley near the Castle trail. This shell measure, length 19.2,
diam. 10, aperture 9 mm. ; 6y 3 whorls.

In one lot from the Kaipapau end of the Castle trail, no.
1866 Spalding 'coll., there are 20 of the plain typical form
with a sutural band only, 3 with a basal band. There is also
one chestnut specimen with a light sutural band. It has
also been taken by Mr. Thurston in the banana patch in Kai-
papau along the Castle trail towards Punaluu.

Color-form oviformis ' Nc.' Pfr. PI. 34, fig. 11; pi. 30, fig.
11&. ' Shell sinistral, imperforate, ovate-conic, solid, striatu-
late, white under a glossy olivaceous epidermis, one to three
banded with brown. Spire convexly- conic, the apex rather
acute, white. Suture deeply margined. Whorls 5, scarcely
convex, the last about three-sevenths the total length. Aper-
ture nearly diagonal, ear-shaped; columellar fold moderate,
tooth-like, reddish ; peristome unexpanded, the margins joined
by a thin callus, the outer margin labiate within ; columellar
margin dilated, adnate. Length 17, diam. 8%, aperture
7% mm. long, 4 wide. Oahu, Newcomb ' (Pfr.). Type in
Pfeiffer's collection.

Mr. Sykes hasr referred Pfeiffer's sinistral form of A. mul-
ticolor to -oviformis, no doubt correctly. Pfeiffer's figure is
reproduced in pi. 30, fig. lla. Specimens from Kaliuwaa are
three-banded with brown (" Hay's russet ") on a wax-yellow
ground, pi. 34, fig. 11, coll. by Spalding. It seems to be
merely a color-form or mutation of typical sowerbyana,
hardly worthy of a name.

13a. A. SOWERBYANA THURSTONi P. & C., n. subsp. PI. 34,
figs. 13 to 14&.

The shell is invariably sinistral; ground-color wax-yellow
below the periphery, white above it, with a band of burnt-
sienna bek)w the periphery. Sutural margin defined by a
deep line, usually with an inconspicuous, scarcely noticeable,
dark line next the suture. Apex generally tipped minutely

178 ACHATINELLA SOWERBYANA.

with dull purple. Lip-rib narrow and whitish or very pale.
Columellar fold moderate, pink.

Length 17.2, diam. 10.5, aperture 8.5 mm. ; whorls 5~y 2 .

Length 16, diam. 9.5, aperture 8 mm. ; whorls 5%.

Length 16, diam. 8.7 mm.

Kahuku, 1,500-1,700 ft. elevation, L. A. Thurston, cotypes
in coll. A. N. S. and Bishop Mus. Also in Thurston collection.
Waimea, overlooking Laie, Irwin Spalding.

Rarely the pigment is deficient, ground white throughout
and the band reduced to a group of pale-brown lines (fig.
136). This decolored form approaches A. s. laiensis. There
may also rarely be traces of faint spiral lines near the lip on
the upper surface. These lines are yellowish with several pink
ones near the peripheral white band. These colors are so
delicate that they can hardly be seen without a lens, and are
mentioned here chiefly because this pattern reminds one of
A. casta. The fact is, A. s. thurstoni stands on the border-
line between Bulimella and the casta group. Looking at
some individuals one is disposed 'to rank it as a variety of
sowerbyana, while others have features which certainly come
as near to casta. Figs. 13 to 136 are cotypes from Kahuku.

Further southeast on the main range, upon -the Waimea-
Laie ridge, Mr. Spalding found a lovely color-form, illus-
trated in pi. 34, figs. 14, 14&, 146. As in the typical color-
pattern, the ground-'color is wax-yellow below the periphery,
white above it (when not covered by another 'Color) ; either
bandless or with four peach-red, coral-red or geranium-pink
bands: sutural, supra- and infra-peripheral and columellar;
or bands i and ii may be concreiscent into a broad zone, as in
fig. 14 ; lip-rib narrow, whitish or with spots at the ends of the
bands ; apex dusky purplish or nearly white. Length 15 to
16 mm.

136. A. SOWERBYANA LAIENSIS Pilsbry & Oooke, n. subsp. PI.
34, figs. 15, 15a.

The shell is sinistral, ovate-conic, moderately solid ; white,
with several brown bands, from burnt-sienna to chestnut in
color; the one below the periphery is widest and most con-

ACHATINELLA SOWERBYANA. 179

stant, one above the periphery is usually present, with some-
times a sutural line and columellar patch also; rarely the
shell is pure white. The apex is whitish or tinged with dull
purple. Surface has a moderate gloss, and) is weakly seriate
under the lens. Suture distinctly margined. The aperture
is not very oblique, white within ; lip a trifle expanded at the
edge, having a rather narrow whitish callus rim within, dull
purplish-'brown towards the edge in well-banded individuals.
Columellar fold white, or often in part brownish.

Length 17, diam. 9.7, aperture 8.3 mm. ; 5~y 2 whorls.

Length 16, diam. 9.5, aperture 7.4 mm.

Oahu : Laie, division ridge above the Gastle cut trail, Irwin
S pal ding. Co types in A. N. S. and Bishop Mus. Also in
Sp aiding coll.

This subspecies does not have the greenish-yellow cuticle or
brilliant gloss of A. sowerbyana. The bands are vertically
streaked with chestnut on a lighter, more yellow tint, the con-
trast more obvious in some specimens than in others. Whether
it will eventually prove separable from A. s. thurstoni cannot
now be decided. Its locality is rather remote from the area
of A. cast a, some of the Waimano patterns of which certainly
resemble laiensis.

There are a few white, bandless specimens in Mr. Spald-
ing's lot from the type locality, no. 3556 of his collection, and
one shell with yellow base and wide subperipheral blackish
band (band iii), establishing a connection with var. thurstoni.

13c. A. SOWERBYANA DEXTROVERSA P. & C., n. SUbsp. PI. 35,

figs. 8 to 13.

Shell dextral, white, sometimes uniform, but typically en-
circled by several bands, which are light-brown, vertically
streaked with chestnut; a narrow chestnut columellar area;
suture margined with a dark line which ascends to the apex,
in banded individuals. Penstome narrowly thickened within,
pale with dark spots at the ends of the bands. Columellar
fold moderate, white or nearly so. Length 18.5, diam. 9.5,
aperture 9 mm. ; 5% whorls.

Pupukea, D. Thaanum. Cotypes in A. N. S. P. and Bishop
Museum. Also in Thaanum coll.

180 ACHATINELLA, SECT. ACHATINELLASTRUM.

This is the western terminal member of the sowerbyana
series. In a considerable lot seen it is always dextral. Ex-
cept in direction of coil and some details of banding it has a
close resemblance to laiensis. Some of the white shells have
the lip white, others having it purple-brown.

Further up, on the Kahuku division ridge, the same sub-
species has been taken by Mr. Spalding, 3558 of his col-
lection, 108129 A. N. S. P. The shells are pure white, or light
brown with dark sutural line and bands ii, iii, and the colu-
mellar region are narrowly dark; band ii being narrow, iii
wide and darker.

This subspecies is quite unlike A. s. roseoplica except that
both are dextral.

13d. A. SOWERBYANA ROSEOPLICA P. & C., n. subsp. PL 34,
fig. 12.

The shell is dextral, oblong-conic, rather solid, white under
a greenish-yellow (sulphine-yellow) cuticle which is slightly
streaked and is deciduous in a band below the suture and on
the spire; apex white or pale^brown. Whorls but slightly
convex. Lip-rib narrow, white or pale pink, the columellar
fold roseate.

Length 18, diam. 9, aperture 8.2 mm.; 6 whorls (A. N. S.).

Length 17.2, diam. 8.7, aperture 7.8 mm. (Bishop Mus.)

Opaeula, above forest- fence line, type loc. ; also on the
northeastern division ridge between Opaeula and Ka/waiha-
lona, Irwin Spalding.

A few of the specimens from the last locality have a pink-
ish-brown line bordering the suture on the last whorl. All of
the specimens from both colonies are dextral. It is widely
separated from all other known 'Colonies of sowerbyana.

Section ACHATINELLASTRUM Pfeiffer.

Achatinellastrum PFR., Malakozoologische Blatter i, 1854,
p. 133. PEASE, P. Z. S. 1869, p. 646 (restricted to " species
allied to A. productum Reeve"). SYKES, Fauna Hawaiiensis,
p. 320, 1900.

The shell is imperforate, ovate-conic or oblong-conic,

ACHATINELLA, SECT. ACHATINELLASTRUM. 181

smooth, with unexpanded lip, which is but slightly or not
thickened within ; columellar margin not raised or thickened
on the face.

Type, A. producta Reeve. Distribution, Oahu, chiefly on
the Main Range.

This is the most generally distributed group of Achatinella
In many places it is found on the northern side of the main
range, and in a doubtless more humid time of the Pleistocene
it lived down nearly to sea level. In the Waianae range there
are a number of colonies, but all excessively small, and situ-
ated on the inland slope.

Achatinellastrum is related to Bulimella through such
species as casta and sowerbyana, where the sectional borders
are debatable. Such forms seem to be the least changed
descendants of the ancestral common stock. No point of con-
tact with the section Achatinella (Apex) is traceable among
the recent species.

Achatinellastrum is more prolific in color-mutations than
any other group of the family. The number of patterns runs
into hundreds. About 72 names have been applied to sup-
posed species. This number was reduced to 49 by Mr. Bald-
win in his Catalogue of 1893. Mr. Sykes, 1900, recognized
35 species and 3 varieties. In the following account 17
species and 14 subspecies are admitted, two species and three
subspecies being new.

In the section Achatinellastrum, it is not likely that any
conservative zoologist having adequate collections and data,
would recognize more than seventeen species; but if the evi-
dence is critically examined, it appears that there are pheno-
typically intermediate forms hybrids or undifferentiated
remnants of the parent stock between many of the conven-
tional species. It would be quite possible to reduce the * * good
species" to nine or ten.

Thus, in the eastern end of the Main Range, we have a
chain of connected forms in (1) A. ph&ozona fulgens ste-
wartii vulpina. A little apart from them stands (2) A.
buddii. This is succeeded by the form-chain of (3) A. bellula
casta juncea. Allied, but not connected are (4) A. juddii

182 ACHATINELLA, SECT. ACHATINELLASTRUM.

and (5) A. papyracea. Westward we find the connected series
(6) A. livida curia dimorpha, with a distinct satellite
species, (7) A. ccesia. The Waianae species are as yet but
little known, and apparently are distinct.

Much remains to be done in the definition of the critical points
where one polymorphic population gives place to another,
especially in the western half of the Main Range. It is quite
likely that further collections and study will modify our
present specific boundaries, or perhaps abolish some of them.

I have not constructed a key to the forms of this group
for the reason that most species of Achatinellastrum vary
widely in color. Such small differences in size and shape as
there may be in the average between allied species, are usually
covered by individual variation. I fear that any key I could
make would surely mislead anyone who attempted to name
single specimens by it. The expert will naturally turn to the
group or * * series ' ' to which his unnamed shell belongs. Those
without special knowledge of Achatinella will doubtless most
easily get from the plates a clue to what they seek.

Series of Achatinellastrum.

Series of A. vulpina. Eastern end of the Main range, west
to Manana. Mainly rather large and moderately strong
shells, conspicuously colored, green, yellow or chestnut, gen-
erally streaked or banded, rarely white. Species no. 14 to 18.

Series of A. cast a. Tantalus to Helemano. Smaller shells,
whitish, yellow or chestnut, usually with bands. Species no.
19 to 21.

Series of A. papyracea. Middle of Main range. Rather
capacious, ovate, thin shells, the embryonic whorls not marked
with an ocher band. Species no. 22.

Waianae Range species, intermediate between the papy-
racea and livida series. Species no. 28 to 30.

Series of A. livida. Western half of the Main Range.
Rather small, stout, ovate or short shells, dull green, yellow
or white, often with a few bands, or sometimes streaked ;
embryo often with an ocher band. Species no. 23 to 27.

MAP OP EASTERN OAHU.

183

Q o 2

*<V r .',
fc\V. *'

-P^^x'-^-y v,
^>xVH A \% V.

^ i q v \\\v.,-=:. x

^^ ^

.. ,.-!i .xv^ ?|- *>\XS"fS. -

' '****tS. 3?t

Series of A. vulpina.

Large, highly colored forms, usually chestnut, yellow or
green, and variegated with streaks or bands ; inhabiting ridges

184 ACHATINELLA PELEOZONA.

and ravines from Manana valley to the eastern end of the
Main Range. No other group of Achatinellastrum is found
east of Manoa valley, but westward the casta group appears
in the area inhabited by vulpina forms.

The distribution is diagrammatically indicated in the ac-
companying map, p. 183. The stations of the respective
species are included between the looped lines and the main
axis of the range, but in reality the colonies actually occupy
only a small fraction of the areas indicated.

14. A. PEUEOZONA Gulick. PL 24, figs. 10 to 13 ; pi. 36, figs.
10, 10a ; pi. 43, figs. 1 to Id.

"Shell sinistral, scarcely perforate, oblong-ovate, solid,
shining, striated; white with from one to six black or chest-
nut bands varying in width; apex subacute; spire convexly
conical; suture marginate, moderately impressed; whorls 7,
moderately convex; columellar fold central, white, strong;
aperture a little oblique, lunately rounded ; peristome acute,
well thickened within, with columellar margin dilated, adnate,
or sometimes slightly detached; parietal margin wanting.
Length 22, diam. 12% mm. ; length of body whorl lGi/4 mm. ;
length of aperture 11 mm. ; an average sized specimen.
Length of a large specimen 25.4 mm. Average weight 10.5
grains.' (Gulick.)

Mr. Gulick enumerates the following color-variants. ' ' Var.
a. With one broad band encircling the base. Var. &. With
two dark bands, one entering the aperture, the other revolv-
ing above the suture. This and var. a are sometimes found
in Kailua, Oahu. Var. c. White, with 3 or 4 bands at the
base. Var. d. Without bands, but more or less streaked
with fawn brown. Var. e. Dark brown, with two white
bands, one sutural, the other on the periphery of the body-
whorl. Var. f. Brown, with one or more black bands. Var.
g. Ash or olive brown, with one or more light bands. Var.
h. Chestnut or olive brown, with fine, black, spiral lines. '

Oahu: Keawaawa, on kukui and ki (Gulick, Spalding).
Formerly on the northern side of the range in Waimanalo,
Kailua and Olomana (Gulick) ; fossil in a coconut plantation

ACHATINELLA PH^EOZONA. 185

about half a mile from the shore, southeast of Kailua Bay, in
humus of plowed fields (Spalding).

Achatinella phaeozona GULICK, Ann. Lye. N. H. of N. Y.,
vi, p. 214, pi. 7, f. 40, December, 1856.

"May be grouped with A. buddii and A. fulgens Newc., but
differs in its more solid structure, its thicker lip and colu-
mellar fold, and in the more convex outline of its spire. It
also lacks the black tip which characterizes A, buddii. The
typical varieties found in Keawaawa are rare, that sterile re-
gion affording but few trees, which occupy the ravines near
the summit of the ridge.' (Gulick.)

Professor Hyatt looked upon A. phceozona as the common
ancestor of Achatinellastrum, Bulimella and Apex (Science,
viii, p. 395). Later he claimed for A. phceozona a relationship
to Kauaia. Our investigations lead to totally different re-
sults, and we believe the affinities claimed by Hyatt to be alto-
gether erroneous.

Some specimens of phceozona are hardly separable from cer-
tain shells of the plumata pattern of A. fulgens ; but other
patterns of both differ widely. A. phcuozona never has green
or yellow varieties.

Although this species is at present restricted to a very
small area, it formerly inhabited an extensive district on the
northern or Koolaupoko side of the island, which was then
wooded down to the shore. About sixty years ago, when Mr.
Gulick discovered the species, it was still to be found in a
few scattered colonies in Waimanalo and Kailua. On his
labels Mr. Gulick indicated that it was almost extinct in these
valleys.

The Keawaawa lot in Gulick 's collection consists wholly of
dead shells. A series is shown in plate 24, figs. 10-13, pi. 36,
f. 10, 10a. The shell is commonly white with chestnut bands
0230 or 0030, but the bands vary in width, sometimes nearly
covering the last whorl. There is almost always a white band
at the periphery, and the sutural margin and summit are
always white. Often the bands are split, giving rise to nu-
merous band-forms, mentioned by Gulick.

Another pattern is closely streaked with fleshy-brown, cut

186 ACHATINELLA PH^EOZONA.

into bands by white spiral lines and zones, and often with
darker bands also. This is a plumata pattern. Albino shells are
also found. Specimens of Gulick 's Keawaawa lot measure :

Length 20.5, diam. 12 mm., 6% whorls.

Length 23.2, diam. 13 mm.

Length 22.5, diam. 14 mm.

Gulick 's collection was no doubt from rather low, where
the forest was already almost gone in his time. Higher up,
in the bottom of the ravine next to the head of Kuliouou,
Mr. Spalding found living shells on dead kukui trees, in some
abundance, in 1908 to 1910. Some of these are figured, pi.
43, figs. 1 to Id. The white, the two-banded, the split-banded
and the streaked patterns are about equally prevalent. The
white shells are more or less tinted behind the lip, and around
the root of the columellar fold. Banded shells often have the
fold wholly white.

Length 25.2, diam. 13.2 mm., 7*4 whorls.

Length 24, diam. 13.5 mm., 7 whorls.

Length 20.6, diam. 12 mm., 6% whorls.

Length 18.8, diam. 12.4 mm., 6 whorls.

Northward, across the range, A. phceozona is probably now
extinct. About sixty years ago Gulick found it in small num-
bers in Waimanalo (similar to pi. 36, fig. 10a), in Kailua and
on Olomana. The Kailua specimens (no. 589 Boston Soc., pi.
36, figs. 9, 9a), are rather small, length 20 to 21 mm., and
mainly slender ; but others of the same lot are typical in con-
tour, similar to fig. 10a. Of the specimens from the Koolau-
poko side Gulick writes: "Smaller, with outlines of spire
less convex; passing into A. plumata. Average weight 5.3
grains. Habitat : vars. i-k in Kailua ; vars. l-o in Olomana.
Var. i. white with numerous chestnut bands on the lower
part of the whorls. Var. j. Light olive brown with dark
bands. Var. k. Dark brown with narrow white bands.'

Olomana is a shapely and elegant peak terminating the
butress thrown out between Kailua and Waimanalo. Here
Mr. Gulick found a few ph&ozona, for the greater part small
and slender, with bands or lines of carob brown below the
periphery, or sometimes above also (pi. 48, figs. 19, 20, Boston

ACHATINELLA BUDDII. 187

Soc. N. H. coll.). Also white, of the ordinary phaozona
shape and size, in contour like pi. 24, fig. 11. No tree snails
are now to be found on this peak. Mr. Gulick notes the fol-
lowing patterns. "Var. 1. White with one or two broad
black bands at the base. Var. m. Elongate, white with two
black bands, one revolving above the suture, the other enter-
ing the aperture, and sometimes a third accompanies the
sutural margin. Var. n. White with from three to five
crowded bands at the base. Var. o. Pure white.'

On the base-leveled plain north of Olomana, about half a
mile from the sea southeast of Kailua Bay, Mr. Spalding
found phaozona of normal size and marking in the humus
turned up by the plow in a grove of young coconuts, west of
the stream. While one would not expect land shells to with-
stand disintegration for many years, it must be a century, and
probably much more, since forest suitable for tree snails
existed in this place. It will be inferred from the planting
of coconuts that the plain lies only a few feet above sea level.
The situation is however a dry or semiarid one for the Koolau
side. East of the stream in this plain I have found rather
rich deposits of fossil land shells, including Amastra and
Leptachatina.

15. A. BUDDII Newcomb. PI. 36, figs. 7 to 8a.

''Shell sinistral, conically ovate, solid; whorls 6, convex,
slightly margined above ; suture moderately impressed, banded
with white ; aperture ovate ; lip acute, thickened within ; colu-
mella short, with a terminal plication. Color yellowish (or
cinnamon) slate or fawn; columella and aperture white.
Length 16/20, diam. 9/20 inch" (Newc.).

Oahu: Waialae (Gulick); Palolo (Newcomb, type loc. ;
Gulick) ; Manoa (Emerson) ; Head of Makiki (Spalding,
Thurston, Cooke, Pilsbry and others).

Achatinella ~buddii NEWC., P. Z. S. 1853, p. 155, pi. 24, f. 73.
PFR., Monogr. iv, 538; vi, 173. GULICK, Evolution, Racial
and Habitudinal, p. 41, pi. 2, f. 16 (Makiki). Achatinella
fuscozona Smith, GULICK and SMITH, P. Z. S. 1873, p. 76,
pi. 9, f. 9.

188 ACHATESTELLA BUDDII.

In shape A. buddii does not differ materially from the wid-
est specimens of A. fulgens. The apex is dark. The colu-
mellar fold is often smaller and generally vinaceous.
The chief difference is one of color; in buddii the shell
in its more primitive pattern is closely streaked with liver
brown to purplish vinaceous, flesh color and whitish or creamy,
or with flesh-tint or yellowish brown alone, on a paler ground,
the sutural margin self-colored or narrowly white. The
streaks are sometimes continuous, but usually the pattern is
varied by darker or lighter spiral zones, or interrupted by
white bands or zones, and there are occasional albino in-
dividuals. Through various stages, there is a passage to the
banded pattern, in which there are spiral bands and lines of
chestnut brown or blackish brown on a white or buff ground,
often shading towards the base to cream-buff, sometimes
streaked with brown. Often there is a very faint brownish
line below the suture, and in the rare mutation described as
A. fuscozona, there is a subsutural chestnut band. The em-
bryonic shell is often brown with a wide white or pale zone
below the suture ; and when white it always has a dark tip,
even in albinos. In A. fulgens the embryo is white as a rule,
but sometimes it has a dark tip. Specimens of buddii from
Waialae, Palolo, pi. 36, f. 7 to le, and Makiki, pi. 36, f. 8, 8a,
do not differ materially.

A. buddii was formerly not uncommon in Palolo, where
large numbers were collected by Newcomb, Gulick and doubt-
less many others. The supposed A. fuscozona recorded by
Messrs. Gulick and Smith from Palolo have no direct connec-
tion with the fuscozona of Makiki, but are an independent
though somewhat similar form of buddii. There is a very
pale sutural band of a light ochraceous-buff tint, on a straw
yellow or nearly white ground, and the apex is that of typical
buddii. The specimens are no. 678 of Gulick 's collection,
and no doubt were selected out of his Palolo lot of buddii.

About 1855 Mr. J. S. Emerson collected an ample series
in the bottom of Manoa valley on the Sugar Loaf side, above
where Dr. Cooke's house now stands, in a grove of kukui
trees then being cut by Chinese to obtain pepeitao-laau, an

ACHATINELLA BUDDII. 189

edible fungus. The shells are of the streaked and also the
banded patterns. These localities have long since been de-
forested, and the species is now to be found only high in
Makiki, where the banded pattern, pi. 36, fig. 8a, prevails.
A few specimens which had escaped shell collectors could still
be found hiding in knot holes and crevices in the bark of cer-
tain old kukui trees, when Doctor Cooke and I visited the
place in 1913. It seems to be a shell of the kukui zone, prob-
ably not found at greater elevations.

An unusual pattern of buddii was selected to form the sup-
posed species fuscozona. The description follows, with notes
on the type and other specimens. It seems to be a mutation
which did not become general in occurrence. There are many
instances where a particular pattern has been found on a few
trees only.

Color var. fuscozona Smith. PL 38, fig. 15. "Shell sinis-
tral, ovate-conic, perforate, slightly shining, striated with
growth lines and (under the lens) very minute transverse
lines; whitish, more or less streaked obliquely with light
brown, and transversely indistinctly lined and zoned ; suture
distinctly margined with brown ; whorls G 1 /^, convex, the last
one ample ; apex blackish ; aperture white, peristome thin,
lightly bordered within; columellar fold strong, reddish
(sometimes whitish). Length 21, diam. 12 mm.

* ' Var. Shell long-conic ; suture girdled by a very wide
brown zone. Length 23%, diam. 11 mm.

' ' Station : on the trunks of trees.

1 i Habitat : Makiki, on Oahu. Two or three specimens have
been found in Palolo. Affinities: It is intermediate between
A. adusta Rv. and A. buddii Nwc.' (Smith).

The type-specimen is figured, pi. 38, fig. 15. It is no. 75 of
Gulick's type series, Boston Society. The original figure
shows the aperture too narrow, and the two lines near the
middle of the last whorl too strong. They are barely visible.
The shell is white, pale buff behind the outer lip and on the
parietal wall of the aperture, where several darker, isabella
colored lines or narrow bands are indistinctly visible. The
first half-whorl is ocher-red, the next whorl violet-plumbeous

190 ACHATINELLA FULGENS.

in the lower half, fading to white above. This dusky tint
fades on the following whorl. The last three whorls have a
chestnut border below the suture about y 2 of. a millimeter
wide. The subsutural furrow bisects this border. The aper-
ture and columella are white, exactly as in A. buddii. There
is a very small dark area below the columellar reflection
which makes it appear perforate, but it is not really so.
Length 21, diam. 12.3 mm., with 6*4 whorls.

In his collection Mr. Gulick selected specimens of buddii
having a brown subsutural band and segregated them as fus-
cozona. Altogether he had about a dozen, found among per-
haps a couple of hundred buddii. They vary in pattern
from that of the type of fuscozona to specimens like pi. 36,
figs. 7a, 7c, except that they have the sutural band. A few
are very small, length 18, diam. 9.5 mm., with 6Vs whorls.

In one of Mr. Gulick 's lots from Makiki, no. 804 Boston
Soc., there is one buddii with a wider sutural band and
slightly purplish-brown columellar fold, and four stewartii
of unusual pattern, two of them figured in pi. 38, figs. 16, 16a.
This is what Mr. Smith described as a long variety of fus-
cozona.

Having examined nearly all the fuscozona ever taken by
Gulick, including the type, I am satisfied that Mr. Sykes was
right in placing it as a synonym under A. buddii. It has
nothing to do with stewartii, except that Mr. Gulick mixed
them in one of his lots.

16. A. FULGENS Newcomb. PL 36, figs. 1 to 6e ; pi. 37, figs.
1 to 9 ; pi. 43, figs. 2 to 4c.

" Shell elongately conic, polished, shining; whorls 6, flatly
convex ; suture slightly impressed ; aperture ovate ; columella
short, tuberculated ; lip simple, ribbed within ; color rich
chestnut-brown, with a broad white sutural fascia cutting the
center of the last whorl ; apex and columella white. Length
eighteen, diam. eight-twentieths inch.' (Newc.)

" Var. a, white with broad chestnut bands.' (Newc.). PL
29, fig. 24.

Oahu: Niu (Newcomb) to the Palolo-Manoa ridge; var.
versipellis over the range in Kailua.

ACHATINELLA FULGENS. 191

Achatinella fulgeus NEWC., Proc. Zool. Soc. London, 1853,
p. 131 ; varieties figured on pi. 22, f . 24, 24a. PFR V Mono-
graphia, iv, p. 537. Achatinella liliacea PFEIFFER, P. Z. S.,
1859, p. 31 ; Monographia, vi, p. 173. A. lilaceum Pfr.,
HARTMAN, Proc. A. N. S., Phila., 1888, p. 34. Achatinella
vulpina Fer., REEVED Conch. Icon,, vi, pi. 4, f. 29a, b, c.
THWING, Orig. Descriptions, etc., pi. 1, f. 12. Not of Ferus-
sac. Achatinella crassidentata PFR., P. Z. S. 1855, p. 6, pi.
30, f. 23; Monographia, iv, p. 539. Achatinella plumata Gu-
lick, Annals of the Lyceum of Natural History of New York,
vi, p. 217, pi. 7, f . 41, December, 1856 ; Evolution, etc., p. 41,
pi. 2, f. 23 (Waialae}. Achatinella diversa GULICK, t. c. p.,
220, pi. 7, f. 42a, 426; Evolution, etc., p. 41, pi. 2, f. 24 (Wai-
alae). THWING, Orig. Descript., pi. 1, f. 11. Achatinella
varia GULICK, t. c. p. 222, pi. 7, f . 43 ; Evolution, Racial and
Habitudinal, p. 41, pi. 2, f. 21 (Palolo). Achatinella tri-
lineata GULICK, t. c., p. 226, pi. 7, f. 43 ; Evolution, etc., p.
41, pi. 2, f. 20 (Palolo). Achatinella augusta SMITH, P. Z.
S., 1873, p. 74, pi. 9, f. 7. Achatinella angusta Sm., PAETEL,
Catalog, p. 105. HARTMAN, Proc. A. N. S. Phila., 1888, p.
33.

A. fulgens comprises particolored, sinistral shells, inhabit-
ing the ridges and ravines between the areas of A. phceozona
on the east, and A. stewartii on the west. It is ordinarily a
more slender shell than A. phceozona, with the spire nearly
straight-sided, and further differs from that by the frequent
presence of a yellow or green cuticle. A. stewartii is usually
more solid, more obtuse above, and differs in color-patterns.
It never has white bands, which are frequent in fulgens.
A. fulgens is sinistral as a rule, but dextral as a very rare
variation. Probably about half a dozen dextral shells are
known. The Kailua race, versipellis, is frequently dextral.
There are some transitional examples on the boundaries be-
tween phceozona, fulgens, and stewartii, where narrow areas
of overlap exist on the western and northern confines of Kuli-
ouou, and in the northwestern ravines of Palolo.

Many strikingly diverse color-mutations have arisen and
become more or less generally spread throughout the fulgens

192 ACHATINELLA FULGENS.

area. Nearly every colony is hybrid, and in some the mixture
is very complex. If any gametically pure colonies exist, they
must be rare. Segregation of the elementary patterns is
often incomplete, and in assorting any large lot, one en-
counters specimens which seem to be blends, often in large
proportion.

In general, one may say that forms obliquely streaked and
lineate with various shades of vinaceous or cinereous (plu-
mata patterns) predominate in the east, and fade out west-
ward, where they finally appear chiefly in blends with other
patterns. Green and yellow cuticle is mainly developed in
the western district. Some other mutations, such as the
augusta,, fulgens and crassidentata patterns, have had their
rise in the intermediate district, and have not spread over the
whole area of the species.

The numerous names proposed by Gulick and others have
no validity in existing taxonomic usage as standing for
species or subspecies, because the forms occur only as con-
stituents of mixed colonies, and not as pure strains; even
though one or two of the patterns may predominate in one
or another colony. These names, however, provide conveni-
ent terms for the designation of particular patterns. The
chief patterns are as follows. How many of them would
turn out to be elementary patterns, if tested by suitable
breeding experiments, is of course quite uncertain.

1. Plumata pattern. Finely streaked with vinaceous, purple

drab or plumbeous gray, pi. 36, figs. 4a, 5, 6, 6a.

2. Varia pattern. White above, yellowish or olivaceous be-

low the periphery; obliquely streaked with cinnamon or
tawny; usually with a dark line below the suture. PL
36, figs. 1, la.

3. Diversa pattern. Base and a band above the periphery

green, olive or yellow, elsewhere white. PL 37, figs. 4a,
4c.

4. Augusta pattern. Green or olive with a yellowish or

white band below the suture. PL 37, figs. 12, 12a.

5. Trilineata pattern. Yellow below, white above the periph-

ACHATINELLA FULGENS. 193

ery, with two black bands, one above, the other below
the periphery. PL 37, fig. la.

6. Fulgens pattern. Chestnut-brown above and below a

peripheral white or yellow band, which ascends the spire
above the suture. PL 37, fig. 1. This is a rare form,
but it is what Newcomb selected as typical of fulgens.

7. Crassidentata pattern. White, with yellow or olivaceous

bands below the suture, at the periphery and around the
axis, the upper one often wanting. PL 30, fig. 23. A
rare pattern.

8. Liliacea pattern. Albino or albinistic forms occur in some

colonies, and may be derived from various patterns.

The typical fulgens color-form described by Newcomb is
shown in pi. 37, fig. 1, this specimen being from Waialae
nui, Cooke collection. Newcomb 's original figures, repro-
duced in pi. 29, figs. 24, 24a, represent other color-forms,
fig. 24 being Newcomb 's var. a, while fig. 24a is what was
subsequently described as A. augusta.

Niu. Plate 36, figs. 4 to 6e. The plumata pattern of fine
oblique, vinaceous cream or slaty-purple lines and streaks
over a white, vinaceous, or white-and-ye