MARINE BIOLOGICAL LABORATORY.

Received

Accession No. cpC-.'^ V /^

Given hy /J^a.^^iL<.<<<^<Ci>^

Place, /'?^L<i--^r^>'rr^ ^&/.:(^...

*^'^o book Of pamphlet is to be removed tfocn the Iiab-
opatopy uiithout the petonission of the Trustees.

Committee on Publication

Marcus Baker
Chairman and Editor
John Hyde O. H. Tittmann

C. Hart Merriam Bailey Willis

PROCEEDINGS

OF THE

Washington Academy of Sciences

Vol. Ill

1901

WASHINGTON
March-December 1901

3
AFFILIATED SOCIETIES

The Anthropological Society of Washington

The Biological Society of Washington

The Chemical Society of Washington

The Columbia Historical Society

The Entomological Society of Washington

The Geological Society of Washington

The Medical Society of the District of Columbia

The National Geographic Society

The Philosophical Society of Washington

5 6 7 /

PRESS OF

THE NEW EflA PRINTING COMPANY,

LANCASTER, PA.

CONTENTS.

PAGE.

Haniman Alaska Expedition.— XX. Nemerteans ; by Wesley
R. Coe . . I

Mammals collected by Dr. W. L. Abbott on the Natuna Islands ;
by Gerrit S. Miller, Jr., m

Branner Agassiz Expedition to Brazil. — V. MoUusks from the
Vicinity of Pernambuco ; by W. H. Dall 139

New birds of the families Tanagridae and Icteridae ; by Robert

Ridgway i49

Harriman Alaska Expedition.— XXI. Hydroids ; by C. C. Nutting 157

Harriman Alaska Expedition.— XXII. Odonata ; by R. P. Currie 217

Harriman Alaska Expedition. — XXIII. Ascidians ; by W. E.
Ritter 225

A new species of Olenellus from the lower Cambrian of York Co.,
Penn. ; by Atreus Wanner 267

Synopsis of the Rice Rats (genus Oryzomys) of the United .States
and Mexico ; by C. Hart Merriam 273

Harriman Alaska Expedition.— XXIV. The Willows ; by F.
V. Coville -297

Hopkins Stanford Galapagos Expedition.— I. Hemiptera ; by OttO
Heidemann 363

Hopkins Stanford Galapagos Expedition. — II. Diptera ; by D.
W. Coquillett 37^

Hopkins Stanford Galapagos Expedition. — III. Odonata ; by R.
P. Currie 3S1

Harriman Alaska Expedition. — XXV. Algae ; by De Alton
Saunders 391

Hopkins Stanford Galapagos Expedition. — IV. Orthoptera ; by

Jerome McNeill 4S7

Synonomy of the Fish Skeleton ; by Edwin Chapin Starks . 507

Hopkins Stanford Galapagos Expedition. — V. Thysanura and
Termitidae ; by Nathan Banks 541

Descriptions of 23 new Harvest Mice ; by C. Hart Merriam . 547

Seven new mammals from Mexico, including a new genus of
Rodents; by C. Hart Merriam 559

Hopkins Stanford Galapagos Expedition. — VI. Isopods ; by
Harriet Richardson 565

Haniman Alaska Expedition. — XXVI. Harrimanella, a new
genus of Heathers ; by Frederick V. Coville . 569

Preliminary Revision of the Pumas (Felis concolor group) ; by

C. Hart Merriam 577

Obituary. Charles Anthony Schott . . .601

Index ........... 603

ILLUSTRATIONS.

PLATES.

FACING PAGE.

I-XIIL Alaska Nemerteans. no

XIV-XXVL Alaska Hydroids 216

XXVII-XXX. Alaska Ascidians 266

XXXI-XXXn. Olejielhis {Hohnia^ Walcottamis 272

XXXin. Salix sitchejisis 362

XXXIV. Salix alaxensis 362

XXXV. Salix a7nplifolia 362

XXXVI. Salix bai'clayi 362

XXXVII. Salix comtmitata 362

XXXVIII. Salix ptilchra 362

XXXIX. Salix glauca 36 2

XL. Salix arctica 362

XLI. Salix stolonifera 362

XLI. Salix leiocarpa 362

XLII. Salix reticulata 362

XLIII-LXII. AlaskaAlga^ 486

LXIII. Roccus lineatiis — Cranium of 540

LXI V. Roccus lineatiis — Face bones, etc 540

LXV. RoccKS lineatiis — Vertebral column, etc 540

TEXT FIGURES.

PAGE

Carinoma s;rifiini 21

16. Linens viridis 65

17. Salix bebbiana 307

iS. Salix niittallii 310

19. Salix richardsoni 315

20. Salix 7iipJioclada 322

sina 29 21. Salix arbiisculoides 324

Zygo n em ertes alb ida ... 31 22. Sa lix myrtillifolia 325

23-
24.

Carinoma griffini 21

Emplectonema gracile. 24

Emplectonema bi'irgeri. 26
Zygonemertes tJialas-

7. Paranemertes

pere-

gf'i^^a 34

S. Paranetnertes carnea... 38

9. Paranemertes carnea... 38

10. Atnp/iiporiis angiilatiis 41

1 1 . Amphiporiis bimacu-

latus 45

12. AmpJiiporiis ncbiilosus. 49

13. Amphiporiis leiiciodiis. 52

14. Amphiporiis exilis 54

15'

Salix chamissonis 326

Salix glacialis 329

25. Salix fuscescens 330

26. Salix ovalif alia 331

27. Salix polaris 335

28 . Salix phlebophylla 336

29. ^Eschna galapagoensis 383

30. ^'Eschna galapagoensis 384

31. Cannacria fumipennis 3S7

32. Cannacria fumipennis 3S7

7 ctrastemma abcrrans. 58 33. Cannacria fiiitiipennis 388

vii

34* Cannacria fuinipe7inis 3SS

3 5 . Teinnopteryx snodgras-

sii 493

36. Teinnopteryx snodgras-

sii 493

37. Tem?iopteryx siiodgras-

sii 493

38. Liparoscelis cooksoni... 497

39. Liparoscelis cooksoni

ensifer 49S

40. Liparoscelis paludicola 499
4 1 • Liparoscelis pallidus. . . 500

42. Xiphidiuju exitiosu7n... 502

43. Nemobius speculi. 504

44. Nemobius speculi 5^4

45 . Roccus lineatus 5^^

46. Lctiobus bubalus 523

47. Lepisvia galapagoensis 542

48. Lepisma galapagoensis 543

49. Lepisma galapagoefzsis 543

50. Lepisma galapagoensis 542

51-

L episi7i a insula ris

543

52-

Lepisma insular is

543

53-

Machilis mutica

544

54-

Ma chilis mtitica

544

55-

Caloter?nes galapago-

efzsis

545

56.

Calotermes galapago-

ensis

545

57-

Calotermes pacijicus. . .

545

58.

Tanais stanfordi

566

59-

Tana is stanfordi

^66

60.

Tafiais stanfordi

567

61.

^L£gat/ioa excisa

567

62.

Harrimanella stelleri-

ana.

570

571

63-

Cassiope mertensiana..

64.

Cassiope tetragona

572

(^s-

Cassiope lycopodioides. .

572

66.

Harrimanella l^'P'

tioides

576

WASHINGTON ACADEMY OF SCIENCES
OFFICERS FOR 1901

President
Charles D. Walcott

Vice-Presiden ts

From the Anthropological Society W. H. Holmes

Biological Society F. A. Lucas

Chemical Society V. K. Chesnut

Columbia Historical Society W J McGee

Entomological Society H . G. Dyar

Geological Society G. K. Gilbert

Medical Society W. W. Johnston

National Geographic Society A. Graham Bell

Philosophical Society J. Howard Gore

Secretary
Frank Baker

Treasurer
Bernard R. Green

Managers

Class of jgo2 Class of igoj Class of jgo4

L. O. Howard F. W. Clarke Marcus Baker

J. W. Powell Whitman Cross George M. Kober

Carroll D. Wright C. Hart Merriam George M. Sternberg

Note.— For list of members, rules, etc., see volume IV.

PROCEEDINQS

OF THE

WASHINGTON ACADEMY OF SCIENCES

Vol. Ill, pp. i-iio PLS. i-xiii. March 26, 1901,

PAPERS FROM THE HARRIMAN ALASKA
EXPEDITION.

XX.

THE NEMERTEANS.

By Wesley R. Coe, Ph.D.

Yale University.

CONTENTS :

Introduction i

Species previously recorded 3

Species here recognized 5

Keys to groups and species 7

Systematic discussion of genera and species .11

Index 84

The Harriman Alaska Expedition, during the months of
June and July, 1899, afforded exceptional advantages for the
investigation and collection of the marine invertebrate fauna
of the Alaska coast south of Bering Sea.

During the course of the expedition collections of Nemer-
teans were made at Victoria (Vancouver Island) and Lowe In-
let in British Columbia ; at New Metlakahtla (Annette Island),
Cape Fox, Wrangell, Farragut Bay, Taku Harbor, Juneau,
Skagway, Glacier Bay, Sitka, and Hot Springs on the islands
and mainland of southeastern Alaska ; at Russell Fiord and
Yakutat Bay ; at Orca and Virgin Bay in Prince William
Sound ; at Kadiak ; at Popof Island, Shumagin group ; at
Kukak Bay on the Alaska Peninsula ; and at Dutch Harbor,
Unalaska.

Proc. Wash. Aacd. Sci., March, 1901. (i)

2 COE

Stops were made also on St. Paul, St. Matthew, and Hall
Islands in Bering Sea ; at Plover Bay, Siberia, and at Port
Clarence, Alaska, just south of Bering Strait. At these latter
localities, however, the shores are surrounded in winter by
pack ice, which prevents the existence of any considerable
amount of life between tides. No Nemerteans were found
here between tides in the limited time at our command, though
it is not unlikely that in deeper water an abundance of these
worms could have been obtained by the dredge. South of
Bering Sea, on the other hand, the shores are densely covered
nearly to high water mark with a luxuriant growth of nearly
all classes of invertebrates.

All along that part of the coast which borders the North Pa-
cific Ocean — from Dixon Entrance to the Aleutian Islands —
with the exception of areas subject to local disturbing elements,
such as the proximity of rivers and glaciers, invertebrates oc-
cur in remarkable abundance. Not only is there an enormous
number of individuals, but also a great variety of species.
And, furthermore, the unusually large size which many of the
species attain is quite as striking as the wealth of species and
individuals. The purity of the water, its freedom from con-
tamination from muddy streams, and the great rise and fall of
the tides with their consequent swift currents, which continu-
ously provide a rapidly changing supply of pure water, furnish
conditions especially favorable to the growth of marine inverte-
brate animals.

Exceptions to these favorable conditions were found at Ju-
neau, where the water is of slightly less than normal density,
and contains a considerable amount of sediment. At Skagway
these unfavorable conditions are still more pronounced, and
here very few Nemerteans could be found between tides. In
Russell Fiord the water was decidedly brackish, there was lit-
tle tide, and almost no shore collecting could be done. Local-
ities in the vicinity of glacier fronts were extremely barren of
marine life, as was to be expected.

Perhaps the locality which proved the most fruitful was
Dutch Harbor, Unalaska, although the stations in Prince
William Sound were but little inferior, and Sitka was found to
possess an extremely rich littoral fauna.

ALASKA NEMERTEANS 3

Of the abundance of Nemerteans collected, colored drawings
were made of most of the species, and many specimens were
carefully preserved for anatomical study, and have proved most
interesting. Serial sections of most of the species have been
prepared, and details of their anatomy will be found on subse-
quent pages.

Two other members of the expedition. Prof. William E.
Ritter, of the University of California, and Prof. Trevor Kin-
caid, of the University of Washington, have generously sub-
mitted to me the Nemerteans collected by them.

SPECIES PREVIOUSLY RECORDED.

The Nemertean fauna of the northwest coast of North America
up to the present time has been almost entirely neglected.
Several small collections have been made, but the only litera-
ture relating to the group in this region seems to be a paper by
Stimpson. This author' gives brief Latin diagnoses of the
following species from the west coast of North America :

JEmpIectonema viride Stimpson [= Neniertes gracilis John-
ston ^= Eniflectonenia gracilis Verrill]. Found by Stimpson
under stones between tides in San Francisco harbor. This spe-
cies occurs abundantly along the whole Pacific coast of Alaska.

Cosmocephala beringiana Stimpson [= Amphiporus angti-
lattis~\. Found in Bering Strait in five fathoms. This species
also is abundant along the whole Pacific coast of Alaska.

Cerehratulns impressus Stimpson [ = Micriira impressa'\ .
Found in twenty fathoms in Bering Strait. This species was
not met with by the Harriman Expedition.

The investigation of the systematic position and anatomy of
this group of worms has been rendered much less difficult since
the publication of Burger's excellent monograph on the Nemer-
teans of the Gulf of Naples.^ Biirger's classification is found to
be admirable, and his nomenclature has been closely followed in
the present work. It will be impossible, however, to retain the
generic names Eunemertes and Eupolia which Biirger adopts,
for the reason that they have been antedated by Stimpson. In

> Proc. Acad. Nat. Sci. Philadelphia, pp. 159-165, 1S57.

2 Fauna und Flora des Golfes von Neapel. Monogr., 22, 1895.

4 COE

1857 Stimpson^ published brief, though accurate, diagnoses of
these same genera under the names Em^lectonema and TcBui-
osoma. Moreover, Stimpson gives as the type of Emplectonejna,
E. camillea \_= Borlasia camillea Quatr. (1846) = Amphiporiis
neesii Oersted (1844) = Eunemei'tes neesii Vaillant (1890)].
Not only this species, but also Stimpson's other species, Em-
flectonema viride \= Nemertes gracilis Johnston, 1837], is a
typical member of the genus named by Vaillant more than
thirty years later. With these facts in mind it is obvious, as
has been already pointed out by Verrill,^ that Stimpson's name
must be retained.

The status in regard to Eupolia Hubrecht (1887) is similar.
Stimpson named Borlasia quinqtcelincata Quoy et Gaimard
(1833) as belonging to his new genus Tceniosoma^ and described
as new species T. seftemlineatwn and T. ceqtiale, which is
probably identical with quinquelineatwn Quoy et Gaimard.
But both of these species are typical of Etipolia Hubrecht.
There can be no doubt, therefore, about the identity of the two
genera^ and, as in the case of Emflectonema, Stimpson's name,
TcBuiosoma, must hold.

The brevity of some of Stimpson's generic diagnoses, as men-
tioned above, is justified by the citation of well known typical
species of those genera, and in such cases their validity cannot
be questioned. Of his specific descriptions, on the other hand,
it is more than likely that some will prove insufficient for an
indisputable determination of the species.

In the preservation of the Nemerteans, I found a 2- to 5"P^'^"
cent solution of formalin in sea-water to yield most satisfactory
results. Most forms die well-extended if a few drops of for-
malin are added to the water in the vessel in which the living
worms are contained. The formalin solution preserves the
general anatomical and some of the histological features excel-
lently, and sharp nuclear stains are easily secured, especially
with the epithelial structures. This solution, however, works

iProc. Acad. Nat. Sci. Philadelphia, 1857.

* Trans. Connecticut Acad., viii, p. 413, 1892 ; ibid., ix, p. 146, 1S95.

3 Burger states distinctly in his great Monograph (p. 26) that " Tcsniosoma
deckt sich mit Eupolia" but, for some reason, does not consider Stimpson's
diagnoses valid.

ALASKA NEMERTEANS 5

havoc with the nervous elements and with the connective tissues.
Specimens killed in strong alcohol proved valuable adjuncts
to the formalin preparations. In most cases the formalin speci-
mens have been eventually transferred to alcohol for permanent
preservation. Corrosive sublimate-acetic, Gilson's fluid, and,
for nervous elements, Flemming's solution are always to be
recommended.

For serial sections no stains have given more happy results
than Delafield's hasmatoxylin, followed by Orange G.

In the limited time at our disposal it was naturally impossible
to make an exhaustive collection or study of the Nemertean
fauna. Some thirty-two species were collected, however, and of
these twenty-seven proved to be new to science, and only two of
the remaining species have been recorded from the Pacific.

LIST OF SPECIES HERE RECOGNIZED.

Following is a list of the species recognized, with brief notes
on their general distribution so far as our collections go :

PROTONEMERTEA.

I. Carinella speciosa sp. nov. Along the whole Pacific coast of
Alaska. Vancouver Island (Shearer). Not uncommon.

3. C. dinema sp. nov. Victoria, B.C.; Sitka. Uncommon.

3. C. capistrata sp. nov. Prince William Sound. Common.

MESONEMERTEA.

4. Cephalothrix linearis {^di\)[^€)OQx%\.&^. Pacific coast. Abun-
dant.

5. Carinoma griffini sp. nov. Collected by Mr. Creswell Shearer
at Albert Head, Vancouver Island.

METANEMERTEA.

6. Emplectonema gracile (Johnston) Verrlll. Pacific coast.
Abundant. San Francisco (Stimpson).

7. £^. burger i s-p. noY. Glacier Bay ; Sitka. Not common.

8. Zygoizemertes thalassina sp. nov. Sitka. Not uncommon.

9. Z. albida sp. nov. Victoria, B. C. Uncommon.

ID. Paranemertes peregrina gen. et sp. nov. Pacific coast.
Vancouver Island (Shearer). Abundant.

II. P. pallida sp. nov. Yakutat Bay ; Popof Island. Uncommon.

6 COE

13. P. car7iea ST^.now. Pacific coast. Vancouver Island (Shearer).
Common.

13. Amphiporus aiigulatiis (Fabrlcius) Verrill. Pacific coast.
Abundant.

14. A. bifnaczclatzis sp. nov. Southeast coast. Uncommon.
Puget Sound (T. Kincaid).

15. A. tigrinus sp. nov. Farragut Bay. Uncommon.

16. A. tiebulostis sp. nov. Kukak Bay, Alaska Peninsula. Un-
common.

17. A. leuciodiis sp. nov. Southeast coast. Common.

18. A. exilis sp. nov. Pacific coast. Abundant.

19. Tetrastem77ia bicolor sp. nov. Kadiak. Not common.

20. T. aberrans sp. nov. Glacier Bay ; Prince William Sound.
Uncommon.

21. T. cceczcjji sp. nov. Dredged by Dr. Ritter at Kadiak. Common.

HETERONEMERTEA.

22. Tcenlosoma princeps %\i. no\. Southeast coast. Uncommon.
33. Linens viridis (Fabricius) Verrill. Annette Island. Com-
mon.

24. L. torquatiis sp. nov. Prince William Sound. Common.

25. Micrura verrilli sp. nov. Prince William Sound. Not un-
common.

26. jSI. alaskensis sp. nov. Southeast coast. Common.

37. Cerebratulus herculeus sp. nov. Sitka. Not uncommon.

28. C. marginattis Renier. Sitka. Not uncommon.

29. C. occidentalis sp. nov. Yakutat ; Prince William Sound.
Vancouver Island (Shearer). Abundant.

30. C. longiceps sp. nov. Yakutat. Uncommon.

31. C. mofttgomeiyi sp. nov. Puget Sound to Unalaska Island.
Common.

32. C. albifrons sp. nov. Near Sitka. One specimen.

It must be remembered that the above list represents but a
few weeks' collecting, and sometimes with only a few hours
at a locality ; too much confidence therefore should not be
placed on the distribution or comparative abundance of the
various species. Further research will undoubtedly add greatly
to the number of species, and judging from the number found
in so short a time it seems not unlikely that the list may eventu-
ally be more than doubled.

ALASKA NEMERTEANS 7

General considerations in regard to the anatomical peculiar-
ities of the Nemerteans described below are reserved for a
future paper. My reason for postponing this important duty in
connection with the study of the collection is that since this
article was written, extensive collections of Nemerteans from
the west and northwest coasts of America and the Pacific Ocean
have been placed in my hands. These collections will doubt-
less include other forms than those collected on the Harriman
Expedition, and it seems desirable that as many as possible be
studied before drawing any general conclusions in regard to the
anatomical peculiarities presented by the Alaska forms.

KEYS TO GROUPS AND SPECIES.

For convenience in determination, the species at present
known from the northwest coast of America may be arranged
in the following analytical table, based mainly on superficial
and easily distinguishable characters.

A. Proboscis without stylets. Mouth posterior to brain. Intestinal
cfficum absent.
a. Lateral nerves external to musculature of body walls.

Protonemertea.
aa. Lateral nerves imbedded in or between muscular layers of body

w^alls b.

b. Musculature of body walls in two main layers, to which an
inner circular layer is sometimes added in the esophagal

region. Cerebral sense-organs absent Mesonemertea.

bb. Musculature of body walls in three main layers, of which the
inner is longitudinal. Cerebral sense-organs present.

Heteronemertea.
AA. Proboscis with stylets. Mouth in front of brain, usually opening
with proboscis in a single terminal or subterminal pore. In-
testinal cKcum present. Lateral nerves internal to muscular
layers of body walls Metanemertea.

Order PROTONEMERTEA.

Body remarkably long, soft, and fragile. Head distinctly marked
off from body, usually broader than neck, often flattened dorso-ven-
trally Carinella^ p. 1 1 .

I. Body very soft, attaining a length of upwards of 2 meters, deep red or
bright vermilion C.speciosa, p. n

8 COE

2. Color brown or chocolate with numerous narrow, transverse white rings

and five parallel, longitudinal white lines, of which one is in the dorso-
median line, and two on each lateral surface C dinema, p. 15.

3. Color brown with numerous narrow transverse rings and three parallel, lon-

gitudinal white lines of which one is in the dorso-median line and one just
below each lateral margin C. capistrata, p. 16.

Order MESONEMERTEA.

A. Body musculature consists of a thin outer circular layer and a
strong inner longitudinal layer — an inner circular layer is
usually wanting entirely, or, if present, is much reduced. Body
long and filiform, often coiled in a close spiral. Head sharply
pointed. Mouth placed very far behind the brain. Neph-
ridia not discovered Cephalothrix^ p. 19.

I. Slender, filiform, pale yellow, sometimes with reddish, grayish or greenish
tinge. No ocelli in adult C. linearis, p. 19.

AA. Body musculature consists of a thin outer circular layer, a
diagonal layer, a longitudinal laj'er, and, in the esophagal re-
gion, an additional, enormously developed inner circular layer.
Head broader than neck. Nephridia well developed.

Carinoma^ p. 20.

I. Body rather stout, flattened posteriorly. Mouth close behind brain. Milk-
white with grayish or brownish mottlings, and darker intestinal lobes.

C griffini, p. 20.

Order METANEMERTEA.

A. Body long and slender. Proboscis sheath usually less than half
the length of body. Proboscis small ; central stylet usually
small Emplectonema^ p. 23.

1. Stylets slender, curved. Dark green or pale green above, and whitish or

very pale greenish below E. gracile,^. 23.

2. Stylets straight and short, with swollen bases. Basis of central stylet

swollen and rounded posteriorly. Dark brown above, flesh-color be-
neath... E. biirgeri, p. 25.

AA. Body moderately elongated. Proboscis sheath about ^ to ^ the
length of body. Central stylet well developed.

Paranemertes, p. 32.

1. Two or 4 pouches of accessory stylets. Color purplish brown, dark-brown,

or orange-brown above, on sides, and on lateral margins of ventral sur-
face; rest of ventral surface (often only the median third) white or
yellowish white P. peregrina, p. 33.

2. Four pouches of accessory stylets. Opaque white P. pallida, p. 36.

ALASKA NExMERTEANS.

9

3. Six to 12 pouches of accessory stylets. Whitish, pinkish or flesh-
color P. carnea, p. 37.

AAA. Body commonly rather short and thick. Proboscis sheath us-
ually reaches nearly or quite to posterior end of body. Pro-
boscis large; central stylet well developed a

a. Ocelli numerous, extending along the lateral nerves beyond the
brain. Basis of central stylet massive, w^ith truncate or con-
cave posterior end Zygojiemertes^ p. 28.

1. Each of the 2 pouches of accessory stylets contains about 5 stubby sty-

lets. Olive green above and below Z. thaCasstna, p. 29.

2. Each pouch of accessory stylets contains 2 or 3 moderately slender sty-

lets. White Z. albida, p. 31.

aa. Ocelli do not extend posteriorly beyond the brain. Basis of

central stylet usually rounded posteriorly b.

b. Body not very small. Ocelli usually numerous.

Amfhiporus^ p. 40.

1. Short and broad. Usually two pouches of accessory stylets with 5 to

7 rather slender stylets each. Dark purplish or chocolate-brown
above, with a triangular white spot on each side of the head ; ventral
surface pinkish or flesh-color A.angulatus, p. 41.

2. Basis of central stylet but half as long as the slender stylet itself.

Usually 4 pouches of accessory stylets. Dorsal surface deep brown-
ish orange with 2 conspicuous oval or dark brown spots on head ;
ventral surface pale orange or flesh-color A. bitnaculaius, p. 44.

3. Body rounded. Color of females yellowish orange, obscured in in-

testinal region (in breeding season) by dark-green ova. Males
yellowish, with white flecks A. iigrt'nus, p. 46.

4. Body rather broad and flat. Whitish, thickly mottled with dark brown

blotches and dots; ventral surface whitish A. nebulosus, p. 48.

5. Body slender, rather small. Ocelli usually 15-40. Usually 3 pouches

of accessory stylets, with 2 or 3 stylets each. White.

A. leuciodus, p. 51.

6. Body extremely elongated for the genus. Ocelli usually 60-250, or

more. Usually 6 to 12 (most commonly 8) pouches of accessory
stylets, with i or 2 stylets each A.exilis, p. 54.

bb. Body very small. Ocelli f ew^ ; usually 4 well-developed
ocelli arranged in a rectangle ; but sometimes these 4 eyes
are double or are fragmented into 4 groups of ocelli ; occa-
sionally ocelli are wanting Tetrastemma^ P- 57-

1. Ocelli 4. Brownish-red above, with median, white stripe; ventral

surface white T. bicolor, p. 57.

2. Ocelli of 4 groups, each consisting of several pigment spots of irreg-

ular size. Pale yellow T. aberrans, p. 58.

3. Ocelli wanting. Hermaphroditic. Whitish T. cczcufn, p. ^g.

lO COE

Order HETERONEMERTEA.

A. Head without deep lateral, longitudinal furrows. Proboscis mus-
culature of two layers Tceniosojna^ p. 6 1 .

I. Of very large size, reaching a length of 2 meters and a diameter of iS mm.
Deep yellow, thickly sprinkled with small irregular, dark red spots.

T. princcps, p. 62.

AA. Head with deep, horizontal furrows. Proboscis usually with
three muscular layers and muscular crossings,
a. Caudal cirrus absent. Body long and slender, roiuided or flat-
tened; very contractile Linens^ p. 64.

1. A single row of 4-6 (or sometimes 8) ocelli on each side of head. Dusky,

or brownish green, dark brown, or reddish brown; commonly paler
beneath L. virtdis, p. 65.

2. No ocelli in adult. Dark, reddish brown or purple, with a narrow trans-

verse white band connecting posterior ends of cephalic furrows across
dorsal surface of neck Z,. torqiiatiis, p. 66.

aa. Caudal cirrus present. Body rather firm ; incapable of swim-
ming Allcriira^ p. dZ.

1. Dorsal surface deep purple or wine-color with 15-40 sharp, pure white,

transverse lines ; tip of snout with triangular, bright orange spot ;
ventral surface of body pure white M. verrilli, p. 68.

2. Body flattened, grayish brown above with narrow, colorless, transverse

lines ; with a small, flesh-colored, sub-truncate head, much narrower
than body, and with minute gray spots near antero-lateral margins.

M. impressa,^ ?• 70.

3. Salmon or flesh-color (more rarely light brownish); lighter, with tinges

of brighter red, or nearly colorless anteriorly. Intestinal lobes more
deeply colored, sometimes chestnut-brown. A cream-colored stripe is
conspicuous in the ventral median line , M. alaskensis, p. 71.

aaa. Caudal cirrus present in perfect specimens, but often missing.
Body long and flat, with thin lateral margins. Swims read-
ily. Dorso-ventral and diagonal muscles well developed
Eyes usually absent Cerebratulus., p. 74

1. Very large and stout, becoming 2 meters or more in length, and 25 mm

in width. Dark, reddish brown C. herculeus, ^. 1^

2. Large and ribbon-like. Slaty brown to grayish green, with thin, white

lateral inargins C. margiiiatiis, p. 75

3. Rather short. Chestnut-brown or reddish brown in esophagal region

chocolate-brown in intestinal region above ; brownish flesh-color to
dark brown beneath, with a median longitudinal ochre stripe along
ventral surface C. occidentalis, p. 76.

^ Known only from Stimpson's original description — Proc. Acad. Nat. Sci.,
Philadelphia, p. 160, 1857.

ALASKA NEMERTEANS II

4. Head and anterior portions of body narrow and slender, with remarkably

long and deep cephalic furrows. Dark brown or purplish, paler
on borders of cephalic slits and tip of snout C. longiceps, p. 77.

5. Very long and ribbon-like, becoming 2 meters or more in length. Deep

blood-red except tip of snout, which is yellowish white both above and
below C. 77iontgomery{, p. 80.

6. Reddish purple except head which is white on both dorsal and ventral

surfaces back to anterior border of mouth C albtfyo7is, p. 82.

SYSTEMATIC DISCUSSION OF GENERA AND SPECIES.

CARINELLA Johnston.
Mag. Nat. Hist. London, vi, p. 232, 1833.

The species belonging to this genus are characterized by a slender,
soft, rounded body, capable of extending and contracting to a remark-
able degree. Head distinctly marked off from body, usually much
broader than parts immediately following, often flattened dorso-ven-
trally, and disk-like.

On each side of body a transverse furrow separates the head from
the esophagal region. Proboscis opening subterminal ; mouth a small
round opening on ventral surface just back of lateral transverse furrows.
The lateral nerves lie outside the muscular layers of the body, and
just beneath the basement layer of the integument. The body walls are
made up of a very thick outer epittielium with abundant glands, a
basement layer, a circular muscular layer, and a longitudinal muscular
layer in the order named from without inwards. In addition to these
there is, in the esophagal region, an inner circular muscular layer
which often forms a dorsal, and sometimes a ventral crossing with the
fibers of the outer circular muscular layer.

Proboscis sheath usually not more than 1/3 the length of bodv.
Proboscis small and short. Ocelli wanting. Cerebral sense organs
usually represented simply by a pair of sensory pits in the epithelium,
although some species (cf. C. speciosa) possess a pair of well-devel-
oped sense organs with ciliated canal, ganglion cells and glands. A
pair of peculiar lateral sense organs is usually situated well back in
the esophagal region. Some of the species show elaborate markings
of fine white lines on a brownish body, others are homogeneous in
color. Most species are colored in some shade of brown or red ;
some are bright vermilion.

I. CARINELLA SPECIOSA sp. nov.

p1. Ill, fig. 6; p1. IX, figs. 1-3; pi. X, figs, i, 2.
Body large, very long, flattened below, rounded above, and remark-
ably soft and pliable. This beautiful Nemertean closely resembles C.

12 COE

polymorpha in external appearance, but is more brightly colored, and
shows conspicuous peculiarities in its internal anatomy. The body is
even softer than that of C. polymoi-pha^ and more distensible ; and the
species is easily distinguished by the presence of enormously developed
cephalic glands, by having very highly specialized and sharply defined
cerebral sense organs (with ciliated canal, lined with glandular and
sensory cells), by the peculiarities in the nephridia, and by many other
details of structure.

Head rounded in front, broader than parts immediately following,
from which it is marked off by lateral, transverse furrows. Proboscis
opening and mouth as in other species of the genus. Body variable in
thickness in its different portions according to state of contraction of
the particular part, but usually flattened below and rounded above,
with well-rounded lateral faces ; posterior portion of body much more
slender than anterior.

The worms are very sluggish. When coiled together the turns of
the body are irregular and angular.

Color. — This is the most brilliantly colored and unquestionably the
most striking Nemertean encountered on the expedition. The color of
the whole body was a deep, rich vermilion, becoming somewhat paler
and more yellowish towards the posterior end. In formalin or alcohol
the natural color quickly disappears, although a characteristic marking
still remains. This feature is peculiar to C. polymorpha as well.
About 25 to 40 mm. back from the tip of the head (in a large speci-
men) is a broad band of dark brown color reaching entirely around
the body. This band is commonly 10 to 30 mm. in width, is very
sharply marked off anteriorly, but fades out gradually behind. All
the rest of the body, both in front and behind, is yellowish or grayish.
This marking remains after imbedding in parafhn, and even after
sectioning and staining in haematoxylin there is a sharp line of distinc-
tion at the anterior end of the dark band. A single section, cut
obliquely, shows the line of demarkation perfectly. The difference
lies wholly in the integument and does not affect any of the layers
beneath. In the integument the dark band is characterized not only
by the presence of an abundance of minute pigment granules, but by a
decided change in the staining qualities of the closely packed gland
cells.

Size. — The body is very long, one specimen measuring — when sus-
pended by its middle portion, and therefore well extended — fully three
meters in length, and about 5 mm. in width. Most specimens, how-
ever, were not more than y^ as long.

ALASKA NEIMERTEANS I3

The epithelium of the body is very thick, and filled with closely
packed gland cells. The two circular muscular layers in the esopha-
gal region show almost no indication of a mutual crossing of fibers
either above or below, although, as indicated in p1. ix, fig. 3, a few
fibers of connective tissue and fine nerve fibers pass at intervals
from the region of the median dorsal nerve to the internal circular
muscles.

Proboscis. — Attached to tissues of head just opposite the mouth,
and therefore a considerable distance back of the brain commisures.
Its posterior attachment is in the region of the efferent nephridial ducts.
The muscular layers are as in other species of the genus. Two large
nerves enter the latero-ventral portion of the proboscis at its attach-
ment, and pass backwards on opposite sides just internal to the cir-
cular muscular layer.

The proboscis sheath shows a homogeneous basement layer beneath
its internal lining of flattened epithelium. The musculature consists
wholly of circular fibers, except that there is a distinct layer of longi-
tudinal muscles between the circular muscles of the proboscis sheath
and the epithelium of the esophagus. Lying on the muscles of the
proboscis sheath is a small median dorsal nerve which runs parallel
with the median nerve outside the circular body-muscles and corre-
sponds to the inner median nerve of other species.

The esophagus is provided with a pair of nerves continuous with
the unusually large buccal nerves. The cerebral ganglia are situated
wide apart, as in other species. The anterior portion of the head
receives a considerable number of remarkably large nerves (p1. ix,
fig. i).

Cerebral sense organs. — These organs deserve special attention
because of their high degree of development. They are each pro-
vided with a narrow but distinct canal leading outward to the lateral
surface of the integument. The sense organs lie immediately external
to the dorsal ganglia, from which they are separated only by a few
fibers of connective tissue which here represent the integumental base-
ment membrane (p1. x, fig. 3). In size, position and structure these
organs are closely similar to those of C. annulaia, as figured by
Burger. ^ Each consists of a large, rounded mass of nerve cells and
connective tissue surrounding a narrow, central canal lined with
ciliated and specialized sensory epithelium (rl. ix, fig. 2). Periph-
erally there is an abundance of peculiar glandular cells, and the
whole is separated from the surrounding integumental cells by a thin

1 Fauna und Flora des Golfes von Neapel, Monogr. 22, p1. xii, fig. 5.

14

COE

sheath of connective tissue. The central canal leads ventrally and
outwards to open through the epithelium of the lateral margin of the
head. The sense organs are innervated by several large nerves from
the adjacent dorsal ganglion (p1. x, fig. 2).

Cephalic glands. — Enormously developed, making up a large por-
tion of tissues of snout (p1. ix, fig. i). These glands are closely
packed throughout the tissues internal to the integument, and form
thick layers around the cephalic blood lacunae and the rhynchodaeum.
Nearer the brain region they become more scattered, and disappear
just in front of the ganglia, those about the rhynchodaeum reaching
back a little farther than those situated external to the blood lacunae.
The secretion from the glands stains most intensely with ha^matoxylin.
Most of the anterior glands pour out their secretion on the tip of the
snout ; those farther back open along the whole circumference of the
body, but most abundantly near the lateral margins (p1. ix, fig. i).
Still farther back all the glands open laterally.

Nephridia. — Situated mainly in, and a little behind, middle third of
esophagal region. There is a single large longitudinal canal on each
side, lying in close contact with the dorsal walls of the large, lateral
blood lacunjE (p1. ix, fig. 3). The nephridial canals reach forward
about as far as the anterior end of the broad band mentioned above as
becoming darkly colored in alcohol. Near its posterior end each
canal exhibits a peculiar relation with the outer integument. At sev-
eral points in each canal the dorsal wall of the canal becomes discon-
tinued and the inner ends of the integumental cells reach inward into
the lumen of the canal (p1. x, fig. i). Sometimes this infolding of
the integumental cells occupies a large part of the lumen of the ne-
phridial canal. Superficially these areas always show one or more
openings in the outer layers of the integument, but such openings do
not penetrate the nephridial canal. At the posterior end of each ne-
phridial canal there is one, or sometimes two, efferent ducts (p1. ix,
fig. 3) which lead to the dorso-lateral surface of the body, as in other
species.

Habitat. — The species was found at Hot Springs (near Sitka),
Virgin Bay in Prince William Sound, and at Dutch Harbor, Unalaska.'
Individuals were occasionally seen crawling about on the bottom be-
neath the water, and in such instances were extremely conspicuous,
because of their large size and brilliant vermilion color. They were
more commonly found under stones near low-water mark, but were
nowhere abundant.

' Collected also bj Mr. Shearer at Vancouver Island, B. C.

ALASKA NEMERTEANS I 5

2. CARINELLA DINEMA sp. nov.

p1. I, figs. 2, 3.

This species somewhat closely resembles C. superba in general color
of body and the thread-like markings thereon. It differs from this
and allied species in many details, and especially in having two lateral
longitudinal white lines on each side — hence the specific name.

Body long and slender, largest in esophagal region, rounded an-
teriorly, somewhat flattened on ventral surface posteriorly. Head
broad, flat, often emarginate in front ; lateral, transverse furrows back
of head deep and conspicuous. Proboscis-pore subterminal. Mouth
rather large for the genus, elongated and situated slightly in front of
second white ring.

Color. — General color dark brown with a more or less pronounced
yellowish tone, becoming more nearly yellow posteriorly. The mark-
ings of fine longitudinal and transverse white lines on this ground
color are very characteristic and stable. Five parallel, longitudinal
lines run nearly the whole of the length of the body, and there are a
great number of transverse lines. Of the five longitudinal lines, one
lies in the middle of the dorsal surface and extends from the most an-
terior transverse ring to the posterior end of the body. The other four
lie symmetrically on the sides — two almost on the lateral margins ; the
other two much nearer the ventral surface. Seen in section, the distance
between the dorsal median line and the upper of the two lateral lines
on each side is 90°. Two lateral lines on either side are separated by
about 45°, while the remaining 90° lies between the two lower lateral
lines. All the four lateral lines extend from the second transverse line
to about ^ the distance towards the posterior end of the body. Here
they become interrupted, and back of this they are usually indicated
only by short segments and scattered dots, though their course may be
traced nearly to the end of the body. The five longitudinal lines are
all very fine and sometimes consist of rows of fine white dots closely
placed together. In addition to these five longitudinal lines some in-
dividuals show an indication of a median ventral longitudinal line in
the form of a row of isolated fine white dots extending from behind
the mouth nearly to the fifth white ring.

The first transverse white marking lies near the tip of the snout. It
is somewhat wavy in outline and does not reach below the lateral
margin. The second white marking is broader than the first, but is
likewise limited to the dorsal surface. It is indicated, however, on
the lateral surface by two narrow and inconspicuous spots. The third

l6 COE

marking is not represented on the lateral surfaces, but is sharp on the
dorsal surface, and is indicated on the ventral surface by a few scat-
tered white dots. The fourth marking is ver}^ sharp and extends as a
ring completely around the body. The fifth is broader than any of the
other rings. The sixth is usually merely indicated by a series of fine
dots. Posterior to the sixth, there are commonly as many as 150 more
or less distinct and perfect rings, arranged at fairly regular intervals
throughout the length of the body. They are much more conspicuous
on the dorsal than on the ventral surface. In fact a large number of
them are interrupted on the ventral surface and represented only by
isolated dots. When the rings are well developed each consists of two
very fine white rings lying side by side, with a fine brown ring between
them. Over the greater portion of the body are alternating wider
and narrower white rings, or double and single rings, but there are
always exceptions to the regularity of arrangement. The white lines,
both longitudinal and transverse, appear as if formed by a coating of
fine white particles on the surface of the body.

The region of the proboscis-pore is very pale. Extending outward
and backward on each side from a point just above the proboscis-pore
is a shallow, horizontal groove. Each of the grooves is marked by a
black line, and each extends backward as far as the first transverse
white line. They do not quite meet in front however. The borders
of the mouth are pale in color. The sides of the brain region are dark
reddish.

In alcohol the body is grayish as far back as the fourth white ring.
Here it abruptly changes to black, which fades into dark brown through-
out the remainder of the body. The markings are well preserved.

The ''side organs'* are indicated as a pair of rounded pits lying just
anterior to the fifth white ring and immediately below the dorsal of
the two white, lateral, longitudinal lines on each side.

Habitat. — This species was found on the piles of a wharf at Vic-
toria, B. C, and was also found among hydroids, etc. at Sitka. The
worms were from 300 to 500 mm. long in extension, with a diameter
of about 2 to 3 mm. They inhabited grayish, fragile, parchment-
like tubes, which were commonly much twisted and coiled. But few
specimens were met with.

3. CARINELLA CAPISTRATA sp. nov.
p1. I, fig. I.
This, like the preceding species, resembles C. superba (Kolliker)
Burger somewhat closely in color and general appearance. A careful

ALASKA NEMERTEANS 1 7

examination of a number of individuals, however, shows that the
markings on the body present such constant differences that the tvv^o
species must be considered distinct. C. capistrata is likewise differ-
ent from any of the other related and described species. It may at
once be distinguished from C superba by lacking all indications of
a median ventral white line. The markings on the head and the ar-
rangement of the anterior transverse white lines distinguish it easily
froin C. a/inulata and C. nothus Burger ; from C. dniema it may be
separated by the presence of but one pair of lateral white lines, in-
stead of the two pairs found in C dinema.

C. capistrata is a very large species, attaining a length of more
than a meter with a diameter of about 5 mm. near the anterior end,
and of 2 to 3 mm. farther back. The worms are therefore extremely
long and slender, and the diameter in the esophagal region is twice as
great as it is farther back. The body is strongly rounded on the dorsal
surface throughout its whole length ; the posterior end is very slender
and easily broken.

Head broad, flattened dorso-ventrally, rounded or emarginate in
front, sharply marked off from succeeding portions by a pair of deep,
latei'al, transverse constrictions, or furrows. Proboscis-pore minute,
subterminal, bounded on each side by a shallow horizontal groove.
Mouth small, slightly elongated, situated immediately behind the
lateral furrows.

Color. — General color of body rich, deep brown, sometimes vary-
ing towards grayish black ; posteriorly becoming lighter, and of a
yellowish brown. On this ground color is a series of conspicuous, but
fine, longitudinal and transverse white lines (p1. i, fig. i). The lon-
gitudinal lines are three in number and parallel — one in the middle of
the dorsal surface, the others just ventral to the lateral margins. The
dorsal line extends without interruption from near the tip of the head
to the posterior end of the body. The lateral lines, on the other hand,
are much broken up in the anterior portions of the body, and in the
esophagal region are only indicated at intervals, though their course
can be followed nearly to the head ; throughout all the rest of the
length of the body they are sharp and distinct.

The transverse white lines, excepting the most anterior one, com-
pletely encircle the body. This first transverse line appears as a sharp
V-shaped marking a little back of the head. Its angle is directed
backward and its extremities reach but little below the lateral margins.
The median dorsal line passes through the angle of this V-shaped
marking, and at the point of intersection the white area is somewhat
Proc. Wash. Acad. Sci., March, 1901.

l8 COE

enlarged to form a small oval spot. The second transverse marking is
situated some 20 to 30 mm. (in large individuals) from the first, and
completely encircles the body. This ring is always sharp and con-
spicuous, and in alcoholic specimens is still more distinct in that it
separates a paler anterior portion from the much darker succeeding
parts. The third transverse ring is also complete, and is commonly
separated from the second by about half the distance that this is separated
from the first. The fourth and fifth rings are sometimes interrupted
below, but are more usually complete, and are much nearer together
than the second and third. Back of the fourth or fifth ring we find a
regular succession of complete rings to the end of the body. It is
common to find as many as 200 such rings arranged at fairly regular
intervals and separated from each other, when the worm is contracted,
by an average distance of about the diameter of the body. Some of
the rings are wider than others, but nearly all are as narrow as fine
threads ; a few are more or less interrupted.

The anterior and lateral margins of the head are bordered by a nar-
row band of light color, as in several other species, but this is mainly
visible from the ventral surface.

The ' side orgatts ' are indicated by a pair of small pits situated in
the third transverse white ring and just dorsal to the position of the
lateral lines. Their position is about the same, therefore, as in C.
S7iperba.

In alcoholic specimens the arrangement of the white longitudinal
and circular lines is still distinguishable, except near the anterior
end of the body, which, back to the second transverse marking
(first complete ring), is pale or grayish brown. Back of this the
color abruptly changes to very dark brown or black, which reaches
posteriorly 1^0 to 75 mm. (in large individuals) and then gradually
fades out into light brown, which continues to the end of the body.
The color is usually paler on the ventral than on the dorsal sur-
face.

Habitat. — This species was found in abundance under stones near
low-water mark at Orca and Virgin Bay in Prince William Sound, but
was not met with elsewhere. The worms lived in long, fragile, gray-
ish, paper-like tubes of about the same diameter as the body. These
tubes were usually twisted about horizontally beneath the stones under
which the animals live. They were also frequently occupied by a
species of polychaetous annelid (^Nereis), which was often associated
with the Nemertean. The worms are sluggish in their movements,
and break up posteriorly if roughly handled.

ALASKA NEMERTEANS

19

CEPHALOTHRIX Oersted.

Entwurf der Plattwurmer, Kopenhagen, 1844.

This genus includes very long, slender, filiform species which show
a tendency to coil in a close spiral. Brain situated well behind tip of
snout ; mouth several times as far posteriorly. Head sharply pointed
in extension ; proboscis-pore on ventral side a little back of its extrem-
ity. Inner circular muscular layer very much reduced, or (commonly)
entirely absent, the body musculature consisting of a thin outer circu-
lar muscular layer and a strong inner longitudinal layer ; lateral nerves
situated in the longitudinal muscular layer ; cerebral sense organs and
cephalic furrows wanting. These worms resemble in external appear-
ance some of the slender Nematodes.

4. CEPHALOTHRIX LINEARIS (Rathke) Oersted.

Planaria linearis Rathke, Skrivter af Naturhist. Selsk. Kjobenhavn, v, p.

84, 1799.
Cephalothrix linearis Q-EKST^n, Entwurf der Plattwiirmer, p. 82, Kopenhagen,

1844.

A very slender thread-like species capable of great extension and
contraction. Individuals may be extended till they resemble a very
fine thread, but when disturbed commonly coil themselves into a closely
wound spiral. Body commonly rather thicker in the middle and taper-
ing toward both extremities. Head very long, acutely pointed when
extended. Proboscis-pore situated ventrally, some little distance back
from tip of snout. Mouth very far back ; commonly distant from tip
of snout 10 to 12 times the diameter of body; or it is perhaps 4 to 5
times as far back as is the brain.

The length is subject to the greatest variation. Alaska specimens
commonly extended themselves 3 to 6 inches or more, but could con-
tract to a small fraction of this length.

Ocelli. — Wanting in adults although the embryos are provided with
a single pair soon after leaving the egg.

Color. — Usually pale yellow throughout, but some specimens had a
distinct reddish tinge, and some were gray, greenish, or pale green.
A median paler line, due to the proboscis sheath, appears on the dorsal
surface in the esophagal region.

Habitat. — The species is very common well up toward high-
water mark under stones in muddy places, among decaying mussels,
etc. The worms were commonly found where the mud was black,
slimy, and very foul. Scores of specimens were sometimes found
under a single stone. They were often associated with slender red-

20 COE

dish Nematodes. Very abundant at New Metlakahtla, Glacier Bay,
Sitka, Orca and at other places. The species is also common on the
coast of New England, and is likewise found along the shores of
northern Europe and in the Mediterranean.

CARINOMA Oudemans.

Circulatory and Nephridial Apparatus of the Nemertea, Quart. Journ. Micr.
Sci., XXV, Suppl., pp. 1-80, 1885.

Body usually slender, often thickened and rounded anteriorly, flat-
tened in intestinal region ; head usually wider than parts immediately
following ; mouth situated immediately behind the brain ; proboscis
pore subterminal. Lateral slits, cephalic groves, and cerebral sense
organs wanting. Intestine with paired, lateral diverticula.

Body musculature composed of two muscular layers throughout
length of body, and of localized supplementary layers. These consist
of a thick internal longitudinal layer and a thin external circular layer,
but in the esophagal region a second circular layer lies internal to the
longitudinal muscles, and just in front of the nephridial region be-
comes enormously thickened. In the anterior portions of the esoph-
agal region a double set of distinct diagonal muscles lies just internal
to the outer circular muscular layer.

The latei^al nerves are situated within the longitudinal muscular
layer.

In the anterior portions of the esophagal region are three pairs of
longitudinal blood vessels, of which one pair represents the main lat-
eral vessels and lies beside the esophagus, a second pair lies beside the
proboscis sheath, and the third pair is situated internal to the ventral
wall of the proboscis sheath and projects freely into the rhynchocccl.

5. CARINOMA GRIFFINI sp. nov.

Two^ species of this interesting genus are already known from
other parts of the world — C. arinandi Oudemans, which is found

' Miss C. B. Thompson has very recently added a third species, C tremaphoros
(Zool. Anz., Vol. XXIII, No. 631, pp. 627-630, Dec, 1900, from a single speci-
men collected at Woods Hole, Mass. I have found this species rather abundantly
in a large pond at Falmouth, Mass., connected with Vineyard Sound by a very
narrow outlet, and consequently but little affected by the tides. The species
must be unusually hardy, for the worms lived just on the edge of the pond in
sand much blackened by decaying organic matter. They heve moreover to en-
dure great changes in the salinity of the water due to irregularity in rainfall and
evaporation. Further notes in regard to the anatomy of this species will be
published later, together with colored figures of the living worms.

ALASKA NEMERTEANS

21

occasionally on the shores of England, and C. fatagonica Burger, o£
which a single specimen has been collected from the Straits of
Magellan.

This new species was found by Mr. Creswell Shearer at Albert
Head on Vancouver Island. It attains a length of upwards of 600
mm. and a diameter of 3-5 mm. Anterior portions of body cylin-
drical ; intestinal region much flattened. Head broader than neck,
and pointed, rounded or emarginate in front according to the state of
contraction (figs, i and 2).

No ocelli were found.

Fig. I.

Fig. 2.
Anterior part of ventral surface.

Figs, i and 2. Carinoma griffim %'^. nov
I. Head extended and pointed. 2. Head contracted and emarginate. X 8.

Color. — General color milk-white anteriorly, with brownish mot-
tlings faither back ; intestinal lobes darker. After preservation the
color is completely lost, and the body becomes nearly cylindrical
throughout.

Body walls. — In internal organization this species agrees closely
with the descriptions which Burger gives of C. armandi^ and of C
patagonica^ but pi'esents a number of peculiarities, which may be
stated briefly as follows : Outer integument divided into a superficial
and a deeper layer of epithelial cells, separated by a network of con-
nective tissue fibers. Basement layer in esophagal region about equal
in thickness to the integument itself, but is much thinner farther back.
Beneath basement layer in esophagal region is a loose sheet of circular
muscles, and beneath this a double set of diagonal muscular fibers.

^ Fauna u. Flora des Golfes von Neapel, Monogr. 22, Nemertinen, 1895.
*Zeits. f. wiss. Zool., Vol. Lxi, pp. 19-20, pi. 3, figs. 1-9, 1S96.

22 COE

The main longitudinal muscular layer in most regions of the body
equals in thickness that of all the other muscular layers combined.
The lateral nerve cords lie imbedded in this layer. Internal to the
longitudinal muscles lies a small amount of gelatinous tissue or paren-
chyma of the body cavity. The inner circular muscular layer extends
from the mouth to the posterior end of the esophagal region. Its
fibers are continuous with those of the proboscis sheath. Anteriorly,
it is even thinner than the outer circular muscular layer, but towards
the posterior end of the esophagal region it increases so greatly in
massiveness that for a short distance it exceeds in thickness the other
muscular layers combined. Having reached its maximum develop-
ment (near the efferent nephridial ducts), it suddenly disappears en-
tirely, its dorsal portion remaining for a few sections as a semicircular
arch above the proboscis sheath.

Proboscis sheath. — This organ is more strongly developed than in
the other species of the genus, and possesses a single muscular layer
composed of circular fibers interlaced with longitudinal fibers in small
groups.

Proboscis. — The anterior end of the proboscis is attached to the tis-
tues of the head at about the middle of the brain region. It walls con-
sist of a thin outer fibrous layer, on which rest the superficial flattened
epithelial cells bathed in the fluid of the rhynchoco^l. Beneath is the
thick layer of longitudinal muscles comprising nearly the whole of the
musculature. Internal to these are a few scattered circular fibers, then
a thin basement membrane, and, finally, the internal layer of columnar
epithelial cells. This inner epithelium is composed of a simple layer
of very long and closely pressed cells, a portion of which contain rod-
like masses of secretion. Their nuclei are closely packed together
peripherally, and comprise several layers as is usual in much crowded
columnar cells.

A pair of rather large nerves extend throughout the length of the
proboscis just internal to the circular muscular layer. These nerves
originate from the ventral commissure of the brain very much as in
Carinella. They arise from the anterior border of the commissure
near its origin from the ganglia, and pass dorsally to the point where
the proboscis is attached to the tissues of the head. They then enter
the proboscis, and take up lateral positions in its walls corresponding to
those of the lateral nerves in the walls of the body.

Nephridia. — The condition of the nephridial system is in many
respects intermediate between that of C. patag'onica and C. ar7nandi.,
the mass of tubules constituting the so-called nephridial glands not ex-

ALASKA NEMERTEANS 23

tending into the cavity of the blood space nearly so far as in the former,
but are much more profusely branched than in the latter species. The
single pair of efferent ducts lies in the same region as the posterior
ends of the series of ' nephridial glands.' Posteriorly to this point a
single unbranched duct of large size runs backward on each side for
a considerable distance, and then bends sharply and runs forward to
the efferent duct, always lying close beside, and just external to that
limb of the duct which is passing backward. The efferent ducts open
on the dorso-lateral surfaces of the body at the posterior end of the
enormously thickened internal circular muscular layer.

The blood vascular system, brain, buccal nerves, lateral nerves, and
dorsal and ventral median nerves are much as in C patagonica.

Specimens collected in August had just discharged their sexual
products. In one individual a very few genital sacks still retained
their mature ova.

EMPLECTONEMA Stimpson.

Emplectonetna Stimpson, Proc. Philadelphia Acad., p. 163, 1857.
Nemertes McIntosh, British Annelids, Part I, Nemerteans, Ray Society,

1872-1873.
Eunemertes Vaillant, Hist. Nat. des Anneles, Tome 3, Paris, 1890.

Body very long and slender, varying greatly in thickness according
to state of contraction, but most commonly considerably flattened ;
often sharply bent and folded into an irregular mass ; integument pro-
vided with an unusual abundance of mucous secretion ; proboscis and
mouth opening together on subterminal portion of snout ; proboscis
sheath limited to anterior third of body ; proboscis slender and very
short, often not more than yi the length of body ; a large number of
minute eyes usually present ; cerebral sense organs situated well in front
of brain, and usually very small. The species are of sluggish move-
ment, and many of them are found very near high water mark among
mussels, barnacles, rockweeds, etc.

Of the genus Emplectonema two species, of which only one (^E. gra-
cile) had previously been described, were found on the Harriman ex-
pedition.

6. EMPLECTONEMA GRACILE (Johnston) Verrill.

p1. viii, fig. 3.

Nemertes gracilis ]on^S'Y:oii, Mag. Zool. and Bot. London, vol. i, 1837-38.
Emplectofiema viride Sti^fso'S, Proc. Philadelphia Acad., p. 163, 1857.
Eunemertes gracilis Vaillant, Hist. Nat. des Anneles, Tome 3. Paris, 1890.
Eunemertes gracilis Burger, Fauna u. Flora des Golfes von Neapel,

Monogr. 22, Nemertinen, p. 543, 1895.
Emplectonema gracilis W^KKl'LL,, Trans. Connecticut Acad., vili, p. 413, 1892;

ix, p. 146, 1895.

24 COE

Body very long and slender, somewhat flattened below, irregular and
ungraceful in form, and sluggish in movement. Head slightly broader
than rest of body ; snout blunt, posterior extremity of body tapering
gradually to a point.

The common opening of the mouth and proboscis lies ventrally, a
little back from the tip of the snout. The cerebral sense organs lie
far in front of the brain, and each communicates with the exterior by
a canal which opens on the ventro-lateral margin near the tip of the
snout.

Color, — Uniform dark green above and nearly white below ; some-
times grayish or yellowish green above, and very pale yellowish green
below. The color of Alaska specimens is as a rule paler and more
grayish green than in Mediterranean specimens. The head is bordered
with a narrow band of white, uniform with that of the lower surface.
Back of the head is an inconspicuous transverse band paler than the
rest of the dorsal surface. The pigment resides wholly in the integu-
ment, the muscular layers and other organs of the body being prac-
tically colorless.

Ocelli. — The eyes (fig. 3) are characteristically arranged in two
groups on each side of the head. The anterior group on each side
usually consists of 8 to 10 pigment-cups ar-
ranged in a single row near the antero-lateral
border of the head. These lie deeply imbedded
in the tissues of the head, and are much more
conspicuous from the ventral surface owing to
the presence of much pigment on the dorsal
surface above them in ordinary states of con-
traction. Each posterior group lies nearly
Fig. 3. Emplectotiema above the brain and consists of 10 to 20 ocelli
gractle. Dorsal view of j^ an irregular cluster. The eyes of the pos-

head to show arrange- . . • -ui 1 r i.u j 1

, ^ ^ tenor groups are visible only from the dorsal
ment of ocelli. X 12. i ,, . .

surface, and are smaller in size than those

of the anterior groups. The integument may be scraped off so as to

expose the eyes clearly.

Proboscis. — The stylet apparatus is especially characteristic of the

species. The basis of.- the central stylet is very long and slender, and

is twice to three times as long as the stylet itself. The posterior

end is sharply swollen into a flattened knob (p1. viii, fig. 3). In

front of the knob the basis decreases in diameter evenly towards the

anterior end, except for a slight constriction which occurs at about Y^

the distance from the posterior end. The anterior portion is slightly

ALASKA NEMERTEANS 25

curved. The central stylet is slender, extremely sharply pointed, and
gracefully curved like a sabre or scythe. There are two accessory
stylet pouches, and each commonly contains 5 to 7 slender stylets
curved like the central one, and of approximately the same size (p1.

VHi, fig. 3)-

Habitat. — This species was found in the greatest abundance at nearly
all the collecting stations between Victoria, B. C, and Dutch Harbor,
Unalaska. It occurred everywhere along the shore, and was most
plentiful near high water mark, crawling over the thick growth of
mussels and seaweeds. Often a number of individuals were found
coiled together in a single slimy mass, and on being disturbed would
crawl apart and move sluggishly about, but usually made no effort at
concealment. Similar masses were met with under stones in very
muddy localities, and often where the water was very brackish. This
is probably the most abundant species of Nemertean on the Alaska
coast, and is found nearer high water mark and in more brackish water
than almost any other species.

The species has previously been recorded from the coasts of Eng-
land, the northern shores of Germany and France, the Mediterranean,
and Madeira. It was also found by Stimpson under stones between
tides in San Francisco harbor and described as E. viride.

An excellent and detailed account of the minute anatomy of this
species is given in Burger's Monograph of the Nemerteans of the Gulf
of Naples.^

Specimens collected in Alaska in June and July were filled with
nearly mature genital products, in closely packed, but irregularly
arranged pouches. The genital glands were often noticeable from the
exterior because of their pale color.

7. EMPLECTONEMA BURGERI sp. nov. '

pi. II, figs. I, 2 ; p1. viii, fig. I ; p1. xii, fig. 3.

Body long, flattened both above and below, ribbon-like, as thick
near the edges as in the median line ; head usually narrower than the
parts immediately following ; posterior extremity slender. The rhyn-
chod^eum opens on the ventral side of the tip of the snout ; a pair of
shallow lateral slits occur just in front of the mouth.

The worms attain a length of more than a meter, and are 5 mm.
or more in width. Like other species of the genus the integument is
furnished with a vast amount of mucus.

^ Fauna u. Flora des Golfes von Neapel. Monogr. 22, 1895.

26 COE

Color. — The color varies considerably, as will be seen from the two
color varieties on pi. ii, figs, i and 2. The commonest form is
dark velvety-brown above, with a paler median line, much paler and
slightly yellowish in front, and flesh-colored or creamy white below.
Other individuals have the whole dorsal and lateral surfaces of a mot-
tled reddish brown color, with a tinge of purple, while the ventral
surface is pale yellow or flesh-colored. Sometimes the head is nearly
colorless. The dorsal surface almost always has an appearance sug-
gestive of velvet. This color is superficial, and is easily removed by
rough handling.

Proboscis. — Proboscis small, very short and delicate, sometimes
scarcely more than yL the length of body. The proboscis sheath
reaches well toward middle of body, but is very small toward its pos-
terior end ; in esophagal region it is well developed, with a thick outer
layer of circular muscles, and a thin, inner, longitudinal muscular
layer. The rhynchodaeum passes backward a considerable distance
before the intestinal canal is separated from the proboscis opening.

The armature of the proboscis consists of a weak central stylet, and
a pair of pouches of accessory stylets (p1. viii, fig. i). The basis of
the central stylet is rather slender in front, slightly contracted near its
posterior third, from which point it swells out
suddenly into a large spherical bulb. The cen-
tral stylet is slightly shorter than its basis. Each
of the reserve pouches usually contains three
small stylets with swollen bases. The posterior
chamber of the proboscis is narrow. The pro-
boscis is provided with ii distinct nerves.

Ocelli. — The ocelli are very numerous. On
Fig. 4. Empledo- each side of the tip of the snout is an elongated
nema burgeri. Dor- cluster of minute eyes, all situated near the dor-
sal view of head to , , 1 -^l . 1 •. r

^ sal surface, and without regularity of arrange-
snow arrangement of . .

ocelli. X 8. ment. Their number is commonly 60 or more

on each side (fig. 4). Back of these are a few
other eyes, likewise very minute, situated deep in the tissues of the
body, and seen only with difficulty. They are scattered irregularly
from near the lateral borders in front of the brain inward towards the
median line (fig. 4).

Cerebral sense organs. — Unusually large. They lie lateral to the
rhynchodasum, and slightly in front of the brain. The ducts by which
they communicate with the exterior pass forward to open latero-ven-
trally.

ALASKA NEMERTEANS 27

The pair of blood lacuncz in the head are rather small, and pass
backward in numerous branches. The dorsal vessel in the proboscis
sheath has several communications vv^ith the lateral vessels in the
esophagal i-egion.

In the region of the brain a great abundance of suh-mtiscular glands
appears, occupying the whole lateral aspects of the body and reaching
far backward. As usual, they are multicellular. Their ducts pierce
the muscles and integument of the body wall to open on the sides of
the body in enormous numbers. Farther back they are restricted to a
narrow region in each section just lateral to the nerve cords, and open
somewhat ventrally from the lateral edge. They continue in dimin-
ished numbers, but of large size, backward beyond the point where
the esophagus opens into the intestine (p1. xii, fig. 3).

Aliitientary canal, — A pair of remarkably narrow intestinal caeca
reach forward well toward the brain. Their diameter for a long dis-
tance back is insignificant compared with that of the esophagus.
They occupy positions ventro-laterally to the proboscis sheath and
above the esophagus. Occasional pouches are sent off laterally into
the tissues above the nerve cords. Farther back they become larger
and extend laterally beyond the ventrally placed nerve cords. Towards
the middle of the esophagal region the pouches become paired with
considerable regularity. A well developed network of muscular fibers
and connective tissue reaches between the pouches from the muscular
layer above to that below the alimentary canal. The pouches lie close
together still farther back, and near the posterior end of the esophagal
region extend on each side below the esophagus and open together
from opposite sides. The two intestinal caeca are thus connected to-
gether. From this point backward the esophagus decreases rapidly in
size, the caecum becomes large and is divided into numerous pouches
by fibrous partitions extending between the muscular layers above and
below, and eventually the esophagus opens into the intestine by a
narrow slit in its dorsal wall (p1. xii, fig. 3).

Re-productive organs. — Far in front of the opening of the esoph-
agus into the intestine the reproductive glands make their first appear-
ance. The anterior pouches are scattered, and lie above the intestinal
caeca well towards the sides of the body. These open directly on the
dorso-lateral surfaces. Farther back similar ones appear below the in-
testinal canal, and these open ventrally (p1. xii, fig. 3). In the intes-
tinal region the glands are very numerous and are scattered just inside
the muscular layers all over the body — dorsally, ventrally and laterally.
As many as 20 to 30 glands in a male are met with in a single section.

28 COE

Their ducts lead directly to the surface of the body, and consequently
open at any point instead of in certain definite regions as in many
species.

Both lateral nerves and blood vessels join above the hind gut as in
most other Nemerteans — the union of the nerves lying ventral to that
of the blood vessels.

This species is named in honor of Prof. Dr. Otto Burger, of Got-
tingen, w^hose monograph on the Nemerteans of the Gulf of Naples
forms by far the most important contribution which has yet appeared
relating to this group of worms.

Habitat. — Several individuals are often found knotted together in a
seemingly inextricable mass. The body lies coiled in a mass, and is
bent and folded in sharp angles.

The species was found under mussels on rocks between tides at
Glacier Bay (W. E. Ritter) and at Sitka.

ZYGONEMERTES Montgomery.
Zool. Jahrb., x, p. 2, 1897.

A species of Nemertean was found at Sitka which agrees very
closely with Verrill's description of Amphiporus virescens} The
detailed anatomical description given by Montgomery,* however, shows
at once that the Alaska Nemertean is a distinct species. Mont-
gomery has created a new genus for A. virescens Verrill based on its
anatomical peculiarities, especially the structure of the proboscis and
proboscis sheath. This genus he named Zygone?nertcs^ with the
following characters as its chief peculiarities :

(i) The proboscis sheath reaches to the end of the body, while (2)
the thickened proboscis is but half as long; (3) basis of central stylet
large, elongated ; flattened or slightly concave posteriorly ; (4) central
stylet massive, not half the length of its basis : (5) ten or eleven pro-
boscidial nerves; (6) body contractile, shape like A^iiphiporus^ not
as elongate as in Emplectonema ; (7) ocelli numerous and small, ex-
tending along the nerve cords posterior to the brain.

While I am of the opinion that these characters are mainly of spe-
cific rather than of generic rank, yet I am convinced that because of
the large number of species already in the genus Atnphiporus the
establishment of this new genus will be of great practical convenience.
The most tangible characters of the group are : ( i ) the eyes extend
posteriorly beyond the brain along the lateral nerve cords, and (2) the

'Trans. Connecticut Acad., viii, p. 20, 1892.
2 Zool. Jahrb., x, p. 2 to 4, 12, 1897.

ALASKA NEMERTEANS 29

basis of the central stylet is massive and has a concave or a truncated
posterior end, while the stylet itself is comparatively weak. Of course
the number of nerves in the proboscis cannot be considered a generic
character. There are two Alaska species, then, which may be placed
in the genus — Z. thalassma^ which is closely allied to the type species,
and Z. albida, which is a minute, white form.

8. ZYGONEMERTES THALASSINA sp. nov.

p1. II, fig. 5 ; p1. VII, fig. I ; p1. xiii, fig. 2.

Zygojiefnertes thalassina differs from Z . virescens chiefly in the
following peculiarities : Color of former species much darker ; eyes
more numerous ; shape of basis and central stylet of proboscis differ-
ent ; usually 5 stubby stylets in each lateral pouch, and 1 2 nerves in
proboscis. There ai^e minor differences in other anatomical details.

Zygonemertes thalassina has a slender, somewhat flattened body ;
head broad, not sharply marked off from body ; one or two pairs of
very inconspicuous oblique furrows on sides
of head. The worms are active, and are rest-
less in confinement.

Ocelli. — Ocelli very numerous ; arranged
in two or three longitudinal rows along sides
of head, and extending backward along lat-
eral nerve cords far behind brain. They are
smaller posteriorly than in front and are more
widely scattered. In front of the brain there
are sometimes 40 or more ocelli arranged in
two or three irregular rows (fig. 5) which
follow the general outline of the lateral mar-
gin of the head. Immediately in front of the
brain about a half dozen much smaller eyes
are seen, and lateral to the brain are com-
monly 10 to 12 ocelli of moderate size. Be-
hind these are usually 8 to 15 small ocelli
scattered along the nerve cords at irregular
intervals. These commonly reach nearly as
far behind the brain as the distance from the
brain to the tip of the snout in moderate ex-
tension. In sections (p1. xiii, fig. 2) they
are found to lie quite internal to the body musculature and almost di-
rectly upon the lateral nerves.

Size and color. — The specimens obtained were from 30 to 60 mm.

Fig. 5. Zyg07iemertes
thalassitia. Dorsal view
of head showing outline
of brain and lateral nerves,
and arrangement of ocelli.
X12.

30 COE

in length in moderate extension, and ratner slender. The color was
olive-green both above and below. A coating of brown particles was
sometimes scattered over the dorsal surface. The proboscis is pale.

Proboscis. — The proboscis sheath extends to posterior end of body ;
the proboscis extends only about half way to posterior extremity. The
central stylet is remarkably short and blunt. Its basis (p1. vii, fig. i)
is at least twice, and often three times, as long as the stylet itself, and
is massive in proportions. Its diameter is nearly constant throughout
its length, although it is narrowed in front and constricted slightly at
about three-fourths the distance towards its posterior end. The pos-
terior extremity is sharply truncate or concave and often shows serrated
edges (p1. vii, fig. i). There are two lateral pouches of accessory
stylets, and each usually contains five stylets similar in size and shape
to the central stylet. The character of these remarkably stubby stylets
is shown in p1. vii, figures \a and \b.

The lateral stylet pouches are imbedded in a thickened glandular
wreath, yellowish in color, which lies around the circumference of the
proboscis in front of the stylet. The mass of muscular tissue surround-
ing the basis of the central stylet is unusually thickened, and this ne-
cessitates an unusually long canal leading from the posterior chamber
to the stylet region. In each of the two specimens sectioned the pro-
boscis was provided with 12 nerves.

Cerebral sense organs. — Situated immediately in front of brain,
but in the ventral portion of the head. The canals leading to the ex-
terior pass obliquely forward and downward, and open on the antero-
ventral surface near the tip of the snout.

Nephridia. — The nephridia extend from a point immediately in
front of the brain backward throughout nearly the whole length of the
esophagal region. There is a single pair of efferent ducts situated in
the region of the brain, and opening to the exterior on the ventro-
lateral aspects of the body. In the specimens examined one of these
ducts lay as far forward as the ventral commissure of the brain, while
the other was situated opposite the posterior end of the ventral brain-
lobe.

Montgomery^ mentions for Z. virescens that the superficial epithelium
of the body contains numerous yellowish, sickle-shaped bodies situated
among the epithelial cells. These also occur in Z. thalassma, but
here they are of various sizes, and differ greatly in shape. Some are
sickle-shaped, others are rod-like or irregular in form, and have every
appearance of having been formed in the gland cells of the epithelium.

'Zool. Jahrb., x, p. 2, 1897.

ALASKA NEMERTEANS

31

Somewhat similar bodies may be found in various other Metanemerteans,
notably in Amphiporus bimaculaius and A. leuciodus. Here too they
appear to originate as secretions in the glandular cells of the epithelium.

The lateral nei've cords unite above the posterior end of the alimen-
tary canal as usual.

The tissues of the head in front of the brain contain large numbers
of sub-muscular glands which open mainly on the anterior extremity,
although a portion of them open directly outwards. These sub-mus-
cular glands do not extend back into the esophagal region as they do
in many other species.

The intestinal caecum is paired from its origin. Its two branches ex-
tend forward on each side of the esophagus until they reach nearly
to the brain. Here they bend dorsally (p1. xiii, fig. 2) and terminate by
abutting closely against the posterior ends of the dorsal brain-lobes.

The ova are large and opaque. They appeared to be fully mature
in June.

Habitat. — This species was found only at Sitka, where it was not
uncommon among hydroids, broken shells, etc., in clear water.

9. ZYGONEMERTES ALBIDA sp. nov.

pi. Ill, fig. 2 ; p1. viii, fig. 5.

This is a small, moderately slender spe-
cies, very active in habits. It was met with
only at Victoria, B. C, on the piles of a
wharf. Sexually mature individuals were
not more than 25 mm. in length. It was
associated with Amphiporus leuciodus.,
which it somewhat resembles.

Color. — There are no markings on the
body, the color being white with a tinge
of yellow both above and below.

Ocelli. — This species may be easily rec-
ognized from other described forms by the
arrangement of the eyes (fig. 6). These
are scattered somewhat irregularly on the
head in front of the brain, and extend back-
ward as a single row on each side along the
lateral nerves for about two-fifths the length
of the esophagal region. The ocelli on the
head are roughly arranged in two irregular
rows on each side. There is an outer row

Fig. 6. Zygotiemertes al-
bida. Outline of anterior
portion of body to show
arrangement of ocelli, br,
brain ; «, lateral nerve ; ps,
proboscis sheath. X 12.

32 COE

of 9 to 1 3 eyes near each lateral border of the snout, and another row
inside and somewhat posterior to the latter, containing about 6 small
ocelli on each side, and of these 3 commonly lie nearly above the
brain. Back of the brain 7 to 10 ocelli occur at widely separated
intervals above the lateral nerve on each side.

Proboscis. — Proboscis comparatively large ; central stylet moderately
slender ; basis dark, moderately elongated, of nearly uniform diameter
throughout, and sharply truncated posteriorly (p1. viii, fig. 5). Ac-
cessory stylet pouches two in number, each commonly containing two
or three moderately slender stylets.

Cerebral sense organs. — Large, and situated immediately in front
of brain.

The eggs of these worms are few in number, but are fully Yi the
diameter of the body in size ; consequently there can be but a single
row on each side. The sexual products are mature m June.

PARANEMERTES gen. nov.

Body of large size, rather stout, usually much rounded in the eso-
phagal region but flattened posteriorly. Head not marked off from
body, of variable form, in some states of contraction often emarginate
in front. There is commonly a pair of inconspicuous oblique furrows
back of head. The nerve cords and blood vessels join on the dorsal
side of the posterior end of the intestine.

The mouth opens into the rhynchodaeum. The proboscis sheath
commonly reaches but little beyond the middle of the body — in P.
peregrina to ^ the distance towards the posterior extremity. The
proboscis is small (^P. pallida)., of medium size {P. peregrina)., or
large (/'. carnea). There is a single central stylet in the proboscis,
and usually four or more pouches of accessory stylets. Some indi-
viduals of P. peregrina., however, have but two. Ocelli are numer-
ous and minute. The cerebral sense organs are rather small and lie in
front of the brain. Sub-muscular glands are usually well developed.

The species of this new genus show considerable resemblance to
those of Emplectonema Stimpson. They differ, however, in general
shape and appearance of body, never being very long or slender, and
individuals do not coil their bodies into a mass as those of Emplecto-
nema are so prone to do. The proboscis is much larger and the
central stylet is always well developed. The proboscis sheath is also
much longer. In many respects the genus resembles Amphiporus.
The body is much longer, however, and not nearly so contractile, the
proboscis is not nearly so large, and the proboscis sheath is not so long.

ALASKA NEMERTEANS 33

The armature of the proboscis resembles that in some species of Am-
phiporus.

Paranemertes is represented on the coast of Alaska by at least
three species.

10. PARANEMERTES PEREGRINA sp. nov.

p1. II, fig. 6; p1. Ill, fig. 5; p1. vii, fig. 7.

Body moderately elongated, flattened below, rounded on dorsal sur-
face ; anterior portion slightly more slender than middle region ; pos-
terior tapering gradually to extremity.

Head very variable in shape, commonly wider than the portion of
body immediately following ; flattened, sometimes sharply demarcated
by lateral constrictions. Tip of snout pointed, rounded, or emargi-
nate according to the state of contraction. On each side of the head
is an inconspicuous V-shaped furrow, with the ends pointing ob-
liquely forward above and below. The upper limb of the furrow
reaches into the dark color of the dorsal surface, where it is sometimes
conspicuous from its light color.

Color. — The color varies considerably as may be seen from a com-
parison of p1. II, fig. 6, and p1. hi, fig. 5, but is commonly homo-
geneous dark brown, orange brown, or purplish brown above and on
the sides, while the ventral surface is opaque white or whitish yellow.
In most specimens the dark purple of the dorsal surface encroaches
considerably on the ventral surface, shading gradually into whitish or
yellowish. Seen from the ventral surface therefore the worms appear
dull white or yellowish white, with a wide border of dark purple.
Oftentimes the whitish color occupies scarcely more than the median
third of the ventral surface. Anteriorly the whitish color covers the
whole ventral surface, and on the head it covers also the sides and
front. The head is dark purplish brown above, bordered in front and
laterally by the light color of the ventral surface. At the posterior
border of the head is a small angular spot on each side corresponding
in color with that of the ventral surface. Behind the head is a nar-
row, V-shaped, dorsal marking, usually of lighter color, with its ends
pointing forward and outward. In paler individuals the pinkish
color of the brain lobes can be distinguished. The natural color of
the body is well retained in formalin or alcohol.

Size. — Individuals of all sizes from 20 to 400 mm. were met with,
but the most common size was about 150 mm. in extension. The
width was commonly about 5 mm.
Proc. Wash. Acad. Sci., March, 1901.

34

COE

Fig. 7. Par-
anemertes per-
egrina Out-
line of anterior
portion of body
to show the ar-
rangement of
ocelli. Dorsal
surface. X 6.

Ocelli. — Numerous minute eyes are arranged in two groups on each
side (fig. 7)* Of these, an anterior group of 13 or more small pig-
ment spots are scattered along each side of the antero-lateral margin,
while about as many more occupy an irregular cluster on
each side just in front of the brain. These latter ocelli
are usually well separated from the anterior, or mar-
ginal groups.

Proboscis. — Proboscis of medium size (as in AinpJii-
portis)., slightly yellowish, and usually everted when the
animal is killed. The extruded proboscis is short and
thick with an unusually slender posterior chamber. The
armature (p1. vii, fig. 7) consists of a small, slender,
sharply pointed central stylet, and with either 2 or 4
pouches of reserve stylets. The basis of the central
stylet is very small and slightly enlarged posteriorly ;
the reserve stylets are slender and sharp like the central
one, and commonly number from 6 to 10 in each pouch.
When four pouches are present the number of stylets in
each is as great as when there are only two pouches.

The mouth and proboscis open together, but the
rhynchodiEum is short. The proboscis sheath reaches
about three-fourths the length of the body, or sometimes more than
three-fourths. In each of four specimens sectioned there were 14 con-
spicuous nerves in the proboscis. The proboscis has a remarkably nar-
row ring of gland cells on the periphery near the posterior end of the
basis of the central stylet.

A crowded mass of multicellular glands occupies the anterior por-
tion of the head. The anterior ones open on the tip of the snout. In
the brain region they open inostly on the lateral surfaces of the body,
vvrhile farther back, and in the esophagal region, they assume the
character of sub-muscular glands. No difference in appearance is no-
ticeable between the cephalic glands and those in the esophagal region.
They are present in the esophagal region only about as far back as
the openings of the nephridia. Throughout their course they open to
the exterior (by innumerable ducts which pass through the muscular
and basement layers) on the ventro-lateral aspects of the body.

Alijnentary canal. — A pair of slender branches of the intestinal
csecum reach forward nearly to the brain commissures and lie well
above the lateral nerve cords. At about the point of the nephridial
openings these branches pass ventrally and occupy a position beneath
the esophagus. Another pair, coming forward from behind, take

ALASKA NEMERTEANS 35

their places. The ventral branches join to form the main median
csecum, from which short branches pass obliquely forward and dor-
sally at intervals. In cross section two pairs of branches are usually seen
besides the main czecum. One pair of these lies above and one be-
low the lateral nerves. They are disposed in such a way that one pair
ends at about the point where the second pair anteriorly originates.
The branches are only irregularly arranged in pairs. The esophagus
opens into the dorsal wall of the main csecum.

Nephridia. — The nephridia occupy the anterior yi of the esoph-
agal region. They are large, with numerous branches lying above the
lateral nerves, and frequently passing internally to the nerves and be-
neath the esophagus. They reach forward nearly to the brain. At
about Yi of their distance posteriorly the main nephridial tubes, lying
above the lateral nerves increase greatly in size and a pair of remark-
ably large efferent ducts pass externally to the lateral nerves to open on
the lateral aspects of the body slightly below the lateral inargins. The
main duct reaching posteriorly from this point is larger than that in
front.

Blood vessels. — There is a pair of large blood lacunae in the head as
usual ; they join anteriorly by a broad anastomosis. The three longitu-
dinal vessels are well developed to the end of the body where they an-
astomose above the anus. Sometimes the dorsal vessel lies beside or
even above the proboscis sheath throughout a portion of its course,
instead of below the sheath as usual.

Nervous system and sense organs. — The cerebral sense organs lie
well in front of the brain, and external to the blood lacunae. They
open into a slight furrow on the lateral aspects of the head a little an-
terior to their own position. The lateral nerves form a commissure
above the anus as usual.

Reproductive organs. — The sexual products were nearly mature in
June and July. They are formed in numerous pouches which sur-
round the intestinal canal on all sides. In a male as many as twenty
sexual pouches were seen in a single section. They open directly to
the exterior, whatever be their position.

Habitat. — This is a restless Nemertean, and on cloudy days was fre-
quently met with crawling about over the stones on the beach between
tides — which peculiarity has suggested its specific name. It was found
abundantly at nearly all the collecting stations from Victoria, B. C, to
Unalaska Island, and is one of the commonest Nemerteans of the coast.
It occurs from low tide well up toward high water mark in every
variety of location — under stones, among seaweeds, barnacles, mus-

36 COE

sels, etc. The individuals are very voracious feeders, and w^ere taken
not infrequently with partially swallowed Chaetopods. Their tenacity
of life is remarkable — they will live for days in a small quantity of
filthy water.

II. PARANEMERTES PALLIDA sp. nov.
p1. VII, fig. 3; p1. XII, fig. I.

Body rather large, stout, rounded, and almost cylindrical anteriorly,
somewhat flattened behind when extended ; head variable in shape,
not sharply marked off from portions immediately following, at
certain states of contraction emarginate in front. A pair of incon-
spicuous oblique furrows back of head. When contracted the worms
are nearly cylindrical and of about the same diameter throughout, ex-
cept at the extremities, both of which are pointed.

Color. — The whole body, both above and below, is commonly uni-
form opaque white, sometimes showing traces of yellowish or reddish
tints, especially in the anterior portions.

Ocelli. — Ocelli minute and numerous. In ordinary states of con-
traction they are arranged in a pair of elongated, irregular clusters on
the antero-lateral margins of the head. The number of such ocelli is
sometimes 30 or more in each of the two clusters.

Proboscis. — Small, short, and unusually slender (p1. xii, fig. i). Its
armature consists of a moderately slender central stylet and usually 4
pouches of accessory stylets. The basis of the central stylet is mod-
erately slender, slightly constricted near its middle portion, rounded
behind, and of approximately equal length with the stylet (p1. vii, fig.
3). There are commonly two accessory stylets in each of the 4
pouches. The chambers posterior to the stylet apparatus are remark-
ably narrow.

The mouth opens into the rhynchodagum. The proboscis sheath
extends but little beyond the middle of the body, and sometimes not so
far as the middle. One specimen had 9 nerves in the proboscis ; an-
other had 10. These nerves do not all enter the proboscis from the
ventral side, as they do in Amphiporus angulatus., but those supply-
ing the dorsal portion enter direct from that side.

Closely packed sub-muscular glands are present on the right and
left sides of the body, and extend well inward towards the median line.
Their ducts pierce the musculature and other layers of the body walls
mainly on the latero-ventral aspects of the body. Twenty or more
are frequently met with in a single section. These glands occupy also
the region in front of the brain, and extend backward in decreasing

ALASKA NEMERTEANS 37

numbers to the commencement of the intestinal region. In front of
the brain is a large and irregular cluster of glands, which open ante-
riorly on the tip of the snout.

Cerebral sense organs. — Situated in front of brain, and near latero-
ventral margins of head. The ducts which place them in communi-
cation with the exterior run obliquely forward and downward, and
open immediately on the surface,

Nephridia. — The nephridial canals extend from near the brain
throughout almost the entire esophagal region. The main trunks lie
above the lateral nerve cords, and send off numerous branches both
ventral and dorsal to the nerves. There is a single pair of remarkably
large efferent ducts opening just below the lateral margins of the
body, and slightly anterior to the middle of the esophagal region. In
one series of sections these ducts are so precisely paired that both ap-
pear in a single section (pi. xii, fig. i). The efferent ducts pass
from above obliquely downwards and external to the nerve cords.

The blood-vascular system consists of cephalic lacunag, and anasto-
mosing longitudinal vessels, as in related genera.

The intestinal cascum does not reach forward nearly to the brain.
There are comparatively few lateral diverticula (p1. xii, fig. i).

The lateral nerve cords unite above the posterior end of the intestine.

Parane7nertes pallida was found only at Yakutat and at Sand Point
on Popof Island, and few specimens were obtained. These were
about 150 mm. to 250 mm. in length, and perhaps 5 mm. in width.
They were found between tides under stones covered with algse.

12. PARANEMERTES CARNEA sp. no v.

pi. Ill, figs. 3, 4; p1. VII, fig. 4 ; p1. VIII, fig. 7.

This species was found to be extremely variable in size, shape of
body, and head, and especially in the armature of the proboscis. As
shown in pi. iii, figs. 3 and 4, the body is rather stout, rounded in
the esophagal region, flattened both above and below posteriorly, and
ending rather abruptly behind. The head is very variable in shape,
being pointed, rounded, broadened, or emarginate in front, according
to its state of contraction. It is most commonly a little broader than
the parts immediately following, and is not distinctly marked off,
although a slight oblique furrow on each side is sometimes seen behind
the brain. From the dorsal surface the furrows of the two sides give
the appearance of a very faint V-shaped marking with the angle pro-
jecting backward in the median line. A little farther forward, as
seen in p1. viii, fig. 7, a pair of shallow furrows occupies the sides of

38

COE

the head. These are also V-shaped with the angle projecting back-
ward on the lateral margins. The ventral limb of each V-shaped fur-
row reaches nearly to the opening of the rhynchodseum ; on the dorsal
surface the ends of the dorsal limbs are separated by about J/S the
diameter of the body in ordinary states of contraction (fig. 8).

In microscopic sections the V-shaped grooves on each side of the
head are conspicuous, because of their differentiated epithelium. In
these shallow depressions the epithelial cells are of smaller size, more
slender in form, and seem to partake more of the nature of sensory
cells. In these grooves the ordinary glandular cells are wanting. In
the ventral limb of each V-shaped groove opens the tube which places
the cerebral sense organs in communication with the exterior.

Ocelli. — There are commonly 20 to 32 eyes on the head arranged
in 4 more or less distinct clusters, though their relative positions change
with the contraction of the head. The two anterior clusters contain
about 4 to 6 eyes each, and occupy the antero-lateral margins of the
head (figs. 8, 9). The posterior clusters lie directly back of these,

Fig. 8. Fig. 9.

Figs. 8 and 9. Paranetnertes carnea. 8. Outline of head from dorsal sur-
face showing arrangement of ocelli. The two pairs of lateral indentations indi-
cate the lateral and dorsal furrows. X S. 9. Outline of the head when extended.
Dorsal surface, showing arrangement of ocelli. X 6.

and immediately in front of the brain. Each of these clusters contains
S to 12 scattered ocelli. An individual from Taku Harbor had 6
ocelli in each anterior cluster, and 10 in each posterior one. Another
specimen had only 3 or 4 eyes in each of the four clusters, but in this
case the ocelli were larger. A specimen from Popof Island had 3
large eyes in each anterior, and 6 to 8 in each posterior cluster ; one
from Virgin Bay had 6 in each anterior, and 7 in each of the others,

ALASKA NEMERTEANS 39

and had 3 single eyes between the anterior and posterior clusters,
while one from Yakutat had 3 large ocelli in a row on the anterior
margin of the head, and a pair of scattered clusters of 8 to I3 eyes
each in front of the brain. When the head is contracted so that it is
emarginate in front, the arrangement of the eyes into clusters disap-
pears, and they are then all scattered irregularly on the antero-lateral
margins. Figs. 8 and 9 show the general arrangement of the eyes.

Color. — The whole body, both above and below, is a homogeneous,
very pale red or flesh-color (p1. hi, figs. 3, 4), and is entirely with-
out markings except for a longitudinal median line of deeper red,
showing the position of the proboscis sheath, and the deeper color of
the intestinal lobes. In some specimens the reddish color was more
pronounced anteriorly, and some were very pale. The intestinal lobes
are seen through the clear tissues of the body walls as narrow, trans-
verse markings, slightly darker than the rest of the body. The reddish
color of the brain and lateral nerves is often conspicuous from the
dorsal surface (p1. hi, fig. 3).

Size. — Individuals were seen which were 500 mm. long in greatest
extension, although the majority were less than half this length.

Proboscis. — Mouth and proboscis open together through a subter-
minal pore. The proboscis is large (p1. viii, fig. 7) and of moderate
length. The short proboscis sheath does not reach far beyond the
middle of the body. The armature of the proboscis shows marked
variations. The size and shape of the central stylet and its basis,
however, remain fairly constant. The central stylet is of the regular
Amphiporus type, is moderately slender, and rests on a moderately
slender basis. The basis is slightly narrower in front and is rounded
behind (p1. vii, fig. 4). The number of pouches of accessory stylets
is commonly from 6 to 12. One specimen had 12 of these pouches,
each with one or two slender stylets ; each of two others had six
pouches with two stylets in each pouch. Three specimens had each
twelve nerves in the proboscis, while a fourth specimen had but
eleven.

Cerebral sense-organs. — Unusually small and situated some distance
in front of brain-lobes. They occupy positions very close to the ven-
tro-lateral borders of the head, and beneath the cephalic blood lacunas.

Nephridia profusely branched, and extending throughout the greater
portion of esophagal region, though they do not reach the brain.
Their numerous branches ramify both above and below the lateral
nerve-cords, and several efferent ducts of small size lead to the exterior
from both the dorsal and ventral branches. There may be about five

40 COE

pairs of efferent ducts, all of which open in the immediate vicinity of
the lateral nerves. In one series of sections a small efferent duct from
one of the branches above the lateral nerve was followed only two sec-
tions farther back by a similar, though larger, duct from a branch below
the lateral nerve on the same side. The nephridia end posteriorly near
the point where the esophagus opens into the intestine.

Sub-muscular glands of limited number lie between the muscula-
ture and the intestine and proboscis sheath. These glands are closely
packed together in the head in front of the brain ; back of this point
they become widely scattered, although they do not cease entirely until
back of the esophagal region.

A short intestinal caecum with a few wide lateral diverticula extends
forward beneath the esophagus. This caecum is shorter than in most
species of the genus, and does not reach nearly to the brain. The
esophagus also is short and opens directly into the dorsal wall of the
intestine.

A pair of large blood lacunae occupies the anterior portion of the head
as usual.

The lateral nerves and longitudinal blood vessels join above the pos-
terior end of the alimentary canal, as in most species.

Reproductive glands in both male and female are very numerous,
and are situated both above and below the intestine. They open di-
rectly to the surface, as could be determined from their rudimentary
ducts, although the sexual products were very immature in June and July.
Habitat. — This species is conspicuous because of its clear, rosy or
flesh-like color, which is all the more striking in contrast with the black
mud in which it is usually found. It occurs between tides in muddy
locations over a large portion of the southern Alaska coast. It was
collected at Taku Harbor, Sitka, Yakutat, Prince William Sound, and
Popof Island, although only a few were found at each locality. Usually
but one or two specimens were found in several hours' digging.

AMPHIPORUS Ehrenberg.
Symbolae Physicae, Berlin, 1831.

This is by far the most common genus on the Alaska coast, and to it
belong six of the thirty species of Nemerteans collected.

The genus Afnphiporus includes mostly rather stout, solid, often
flattened forms, usually of considerable size, which are capable of an
almost incredible amount of extension and contraction. A few forms,
however, are long and cylindrical, even when contracted, but others

ALASKA NEMERTEANS

41

can contract until the body becomes almost barrel-shaped. The worms
can neither swim nor roll up spirally.

Proboscis. — Provided with a single, well-developed central stylet,
with a cartridge-shaped basis, and with two or more pouches of acces-
sory stylets. The proboscis sheath usually reaches nearly or quite to
the end of the body.

Ocelli. — Usually present in very considerable numbers. A few
forms are without eyes, and a few others have but a single pair — there
are never 4. The eyes do not extend far behind the brain.

Cerebral sense organs. — Usually well developed. Their position
is most commonly in front of the brain, but they are sometimes beside
or even behind the ganglia.

13. AMPHIPORUS ANGULATUS (Fabr.)Verrill.

pi. VI, fig. 4 ; p1. VII, figs. 2, 2a ; pi. xi, fig. 2 ; p1. xiii, fig. 3.

Fasciola angiilata O. Fabricius, Muller's Verm. Terrest. et Fluv., i, p. 58,

1774.

Omatoplca stiinpsottu G\KAV.r>, in Stimpson, Invert, of Grand Manan, Smith-
sonian Contributions to Knowledge, p. 28, 1853.

Nareda siiperba (?) Girard, loc. cit.

Cosmoc ph da beringiana Stimpson, Proc. Acad. Nat. Sci. Philadelphia, p.
165, 1857.

Amphiporus angulatits (Fabr.) Verrill, Marine Nemerteans of New Eng-
land, Trans. Conn. Acad., p, 10, 1892.

"This large and conspicuous species is gen-
erally easily recognized by its clear dark purp-
lish or chocolate-brown color above, with pale
margins and a trapezoidal or triangular white
sjjot on each side of the head and usually with
a narrow white line across the neck ; and by the
pinkish or flesh-colored lower surface. Ocelli
in two or more rows in an elongated groove on
each antero-lateral margin of the head, and a
pair of small sub-dorsal clusters on the transverse
white nuchal band." (Verrill, loc. cit.) The
arrangement of the eyes and markings on the
head of the Alaska specimens are shown in fig.
10 and in p1. vi, fig. 4. In ordinary state of
contraction the body is rather short and stout.
When disturbed it can become so greatly thick-
ened anteriorly that its transverse diameter is fully J/3 as great as its
length. In extension the body is but moderately elongated, and is
relatively broad and flat. It contracts very much as does a leech.

Fig. 10. Ampki-
portis angulatus. Dor-
sal view of anterior por-
tion of body showing
markings on the head
and the arrangement of
ocelli. X 8.

42 COE

The Alaska specimens are commonly larger than have been recorded
elsewhere, often measuring 200 mm. or more in length and 10 mm. in
width.

Proboscis. — The proboscis is large, thick, and pale reddish or
salmon in color. The smallest specimen collected had 17 nerves in
the proboscis; four other specimens examined had each 18 probos-
cidial nerves, one had 19 nerves, and two others had 20 each.^ This
shows more strikingly than has previously been pointed out that the
number of nerves in the proboscis is variable to a very considerable ex-
tent. Burger '^ has shown that Drepanoporics crassus may have 19 or
20 nerves, and D. spectabilis 24 or 26. Nevertheless in the other
Alaska species the number has been found surprisingly constant. The
number of nerves in any particular proboscis remains perfectly con-
stant so far as I have observed from the anterior end back as far as the
stylet region. Here they break up into a plexus and lose their iden-
tity. The nerves in A. angidatns enter the proboscis at its anterior
attachment and in its ventral portion (p1. xi, fig. 2). They then
divide into their definite number of branches (usually 18) which pass
obliquely dorsally and arrange themselves symmetrically on the periph-
ery. The proboscis sheath extends within a few sections of the pos-
terior end of the body.

The armature of the proboscis is made up of a moderately slender
central stylet, and (usually) two pouches of accessory stylets. The
basis of the central stylet is about as long as the stylet itself. It is
moderately slender, constricted near its middle (p1. vii, figs. 2, 2a),
enlarged and rounded posteriorly. Each reserve pouch commonly
contains 5 to 7 rather slender stylets, similar in size and shape to the
central stylet.

Ocelli. — Numerous and characteristic in arrangement. The dark
pigment on the head, however, often renders them difficult of accurate
determination. Girard^ states for Otnatoplea stimpsonii that there are
six or more minute eyes " situated in an oblique, simple row, on either
side of the head anteriorly." The same author (loc. cit.) describes

>0£ eight specimens of this species from Eastport, Maine, one had but 17
nerves in the proboscis, six had 18 each, and one had 19 or 20. Other anatom-
ical details in the eastern form agree perfectly with those of specimens from
Alaska. I have recently examined a number from the original locality of Stimp-
son's C. beringiana (Bering Strait), and have no doubt as to the specific identity
of this form with that from southern Alaska and from Eastport, Maine.

"Fauna u. Flora des Golfes von Neapel. Monogr. 22, Nemertinen, p. 372,
1895.

3 Marine Invert. Grand Manan, Smithsonian Contr. to Knowledge, p. 28, 1853.

ALASKA NEMERTEANS 43

Nareda superba as having but a single pair of rounded ocelli situated
wide apart on the transverse white band of the neck. It seems highly
probable, as Verrill suggests/ that both of the species are identica)";
in the one case only the marginal eyes were seen, while in the other
the cerebral clusters were supposed to represent single eyes and the
marginal ones were overlooked. Verrill ' describes the eyes correctly,
and his diagnosis of the species is so full and accurate that it is neces-
sary to describe here the internal anatomy only. A pair of elongated
clusters of ocelli lies on the antero-lateral margins of the head, and an-
other smaller cluster on, or near, the angular white spot on each side
of the head. As shown in fig. 10, each of the anterior clusters may
contain upwards of 20 ocelli arranged in two or more irregular rows
nearly parallel with the antero-lateral margin of the head, while the
posterior groups may consist of 8 to 15 similar ocelli. The posterior
groups are situated deep in the tissues of the head. Of course the
number of ocelli varies greatly in different individuals.

Cerebral sense organs. — Well developed. They lie a little in front
of the brain, beside the esophagus, and below the cephalic blood
lacunae. Each sense organ has a wide canal which leads a short dis-
tance anteriorly and opens to the exterior on the latero-ventral aspect
of the body. The brain itself is of large size, with a thick ventral
and narrow dorsal commissure (p1. xi, fig. 2).

Nephridia. — The nephridia extend from near the brain (p1. xi,
fig. 2) well backward in the esophagal region. In one specimen there
were two pairs of efferent ducts opening on the latero-ventral aspect
of the body ; in another only one pair.

Cephalic glands. — The cephalic glands open on the tip of the snout
and are well developed. Sub-muscular glands, likewise, are remark-
ably abundant. They reach from the brain region well back towards
the end of the esophagus. They are multicellular, each one being
composed of upwards of a score of large, vacuolated cells with small
nuclei situated on the side farthest from the lumen. Each gland has
a twisted duct leading through the muscular layers, basement membrane,
and integument, and opening to the exterior on the ventro-lateral aspects
of the body (p1. xi, fig. 2).

Beneath the esophagus a broad caecal appendage of the intestine
stretches forward well toward the brain region. This caecum consists
of a large median canal with pouch-like diverticula extending dorsally
above the lateral nerve cords.

There is the usual anastomosis of the three longitudinal vessels, and

^Marine Nemerteans of New England, Trans. Conn. Acad., viii, p. 12, 1892.

44

COE

of the pair of lateral nerve cords (p1. xiii, fig. 3), above the hind gut
and slightly in front of the anus. As seen from the figure, the union
of the blood vessels is directly dorsal to that of the nerve cords.

The reproductive glands occur both above and below^ the alimentary
canal. Sexual products were not nearly mature in June and July.

Habitat. — The species is extremely abundant along the whole
Alaska coast as far west as Unalaska Island, and Stimpson records it
from Bering Strait. It is found under stones between tides in all sorts
of situations. Stimpson's specimens came from a depth of five
fathoms. The species is found on the Atlantic coast of North America
from Massachusetts Bay to Greenland (Verrill, loc. cit.^.

14. AMPHIPORUS BIMACULATUS.

pi. I, fig. 4; pi. V, fig. 10; p1. viii, fig. 2 ; p1. XII, fig. 2.

Body rather short, broad, and flattened both above and below. Head
narrower than parts immediately following. Body of about the same
width and thickness throughout esophagal and intestinal regions.
Posterior extremity tapers rather abruptly to the pointed or rounded end.
Opening of rhynchodaBum situated on subterminal portion of snout.
From near this opening a pair of slits pass obliquely backward and
upward behind the eyes to the brain region.

Color. — The color of this species is very striking. The whole dor-
sal surface back of the head is deep brownish orange, somewhat paler
behind. The head is without color, or of a very much paler color
than the rest of the dorsal surface, and in the center of this pale area
two oval, black or very dark brown spots lie side by side. These are
very characteristic, and are conspicuous even in alcoholic specimens.
The black spots sometimes occupy a considerable portion of the pale
area, and are sometimes sharply angular in front (p1. i, fig. 4). In the
median line of the body the color is slightly paler than elsewhere and
in the center of this paler stripe is a dark, but inconspicuous, longi-
tudinal line. The pale stripe and dark line both fade out at a point
about 3^ the distance towards the posterior end of the body. The
brain lobes appear as pinkish bodies just posterior to the black cephalic
spots. The whole ventral surface is of a homogeneous, pale orange
or flesh color, with the exception of pinkish spots marking the position
of the brain, and a slightly paler stripe below the anterior portion of
the proboscis sheath.

Ocelli. — The eyes are rather large, and number 25 to 30 or up-
wards on each side (fig. 11). The majority lie in an irregular margi-
nal row beside and in front of each of the black cephalic spots. At

ALASKA NEMERTEANS 45

the posterior end of each marginal cluster the ocelli are more closely
and more irregularly placed, and often occupy several rows. In ad-
dition to these marginal clusters a closely set group of about a half
dozen smaller ocelli is situated in the light area lateral to the posterior
end of each of the dark cephalic spots. These ocelli lie deeper in the
tissues of the head, and near the brain (p1. i, fig. 4).

Proboscis. — The proboscis is remarkably large, and its constituent
layers are very sharply defined (p1. xii, fig. 2). The proboscis
sheath has a correspondingly massive development,
and reaches to the very extremity of the body. The
stylet apparatus of the proboscis is very characteris-
tic of the species, because of the extreme minuteness
of the basis of the central stylet. The central stylet
itself is very long and slender, while the length of
its basis is but half as great. The basis is con-
stricted in the middle, and is yi as wide as long y\g. ii. Afnphi-
(p1. VIII, fig. 2). There are usually four pouches poms bimaculatus.
of accessory stylets. These pouches are not evenly Outline of head to

distributed on the circumference, for two lie close ^'^°'^ position of

, • 1 r 1 1 • 1 1 markings and ar-

tosrether on one side of the proboscis, the other two . . ,,.

*=> >- ' rangementot ocelli.

on the opposite side. There are usually five to v 8.
seven slender stylets in each of the four pouches.
Most of the stylets are much smaller than the central stylet. Measure-
ments of the stylets of one individual about 100 mm. long are : central
stylet .12 mm. long, .015 mm. wide near base; basis of central stylet,
.06 mm. long, .04 mm. wide ; largest accessory stylet, less than . i mm.
long. The proboscis is provided with 16 large nerves (p1. xii, fig. 2).

The mouth and proboscis open together. There are three large
communicating blood lacunas in the head, one on the right, one on the
left, and one dorsal to the rhynchodseum.

Cerebral sense organs. — Remarkable for their large size, being
fully as large as either of the brain lobes. They lie lateral to the
brain, slightly behind the commissures, and in the angle between the
dorsal and ventral lobes. A large process from the posterior end of
the dorsal lobe furnishes the sense organs of the same side with an
abundant innervation. Their posterior ends extend backwards beyond
the dorsal brain lobes, against the posterior faces of which they are
closely pressed. Behind the dorsal brain lobe the sense organs lie
directly dorsal to the lateral nerve-cords and are bathed on their
internal borders by large blood lacunas. A section through this point
is not very different from a corresponding section of a Heteronemer-

46 COE

tean. Of the Alaska Metanemerteans here recorded this is the only
one in which the cerebral sense organs lie posterior to the brain com-
missures. The canal by which each sense organ communicates with
the exterior is of large size, runs anteriorly in front of the brain, and
opens on the ventro-lateral aspect of the head.

Nephridia. — The nephridia reach forward close to the posterior ends
of the cerebral sense organs. Anteriorly there is a tangle of small
vessels, but farther back these unite into a single large, branched
canal which runs close beside the blood vessels above the lateral nerve
on each side. From these canals a pair of large efferent ducts pass
above the lateral nerve cord, and open to the exterior of the body just
below the lateral margin.

Sub-musctilar glands. — Present along anterior portion of esophagal
region, but not very abundant.

A broad and profusely branched intestinal ccecu?n runs forward
from the intestine well toward the brain region. The caecum lies well
beneath the esophagus and sends off numerous pouch-like branches
dorsally above the lateral nerve cords.

Reproductive glands. — These occur both above and below the intes-
tinal canal. Although the sexual products were very immature in one
of the specimens sectioned, yet the efferent ducts of the glands were
formed as far outward as the basement layer of the cutis. Here each
duct ended in a swollen chamber lined with cylindrical epithelial cells.

Size. — The individuals of this species varied from 40 to 150 mm. in
length. The largest were about 6 mm. wide and 3 mm. thick.

Habitat. — The species was collected at Victoria, B. C, on the
piles of the wharf ; at Sitka among hydroids, etc., near low water (W.
E. Ritter), and a finely preserved specimen from Puget Sound, State
of Washington, was given me by Prof. Trevor Kincaid.

15. AMPHIPORUS TIGRINUS sp. nov.
p1. IV, figs. 5-8; p1. viii, fig. 4; p1. X, figs. 3, 4.

Body moderately slender, rounded throughout, head not marked off
from parts immediately following, rather narrow and pointed in front ;
posterior extremity of body narrow. On each side of the head is a
shallow, inconspicuous, oblique groove.

Color. — In June, at the time the specimens were collected, the
sexual products were fully mature, and the species showed marked
sexual color varieties. The prevailing color of the females was yel-
lowish orange both above and below, but except in the esophagal re-
gion, this color was to a great extent obscured by the dark olive green

ALASKA NEMERTEANS 47

color of the mature ova. These ova developed in large pouches on
each side of the body, and each pouch w^ith its contents appeared as a
dark green spot. In many cases several adjacent pouches lie nearly
in contact, giving the external appearance of dark green blotches.
Seen directly from the dorsal surface the green spots appear on
each side in more or less regular pairs, those of the tw^o sides being
separated by an interrupted, narrow, longitudinal, median band of
yellow. From the sides the ovaries appear as irregular transverse
stripes of dark green alternating with the yellow color of the body —
hence the specific name, tigrinus. The males are much less deeply
colored. They are pale yellowish with a slight tinge of green, and
the spermaries appear as innumerable cream-colored specks. Both
males and females have a narrow, longitudinal, median band of brown-
ish on the anterior dorsal portion of the body. After preservation in
alcohol both sexes assume a greenish color, which is retained even after
mounting in balsam.

Ocelli. — The eyes are numerous, and are arranged in two irregular,
and scarcely separated, clusters on each side of the head in front of
the brain. The individual ocelli are so irregular in shape, so variable
in size, and so closely massed together, that it is difficult to determine
their precise number. Many of them appear as ragged pigment
masses. Commonly, however, there are a dozen or more of such
ocelli in each of the anterior clusters, and perhaps 8 to lo in each of
the posterior ones. The ocelli of the anterior clusters are scattered
through the tissues of the head from the dorsal to the ventral surface.
Some of the ocelli are three times as large as are others. In contraction all
the eyes of the same side form a single confused cluster. Because of
their variability of position no drawing is given of their arrangement.
Proboscis. — The long and well developed proboscis is provided
with a remarkably weak armature. This consists of a small central
stylet and two pouches of accessory stylets. The central stylet is
small and short, but is acutely pointed. Its massive basis, double the
length of the central stylet itself, is short, thickened, and rounded be-
hind (p1. VIII, fig. 4). In a worm 75 min. long, the central stylet
measured about .075 mm. in length; the basis was .15 mm. long and
.075 mm. in average diameter. The accessory stylets are, like the
central stylet, short, broad at the base, but sharply pointed. They
usually number about five to each pouch. The glandular wreath
about the stylet is well developed, and in the specimens examined is
deep green in color even after mounting in balsam. The proboscis
sheath reaches nearly to the posterior end of the body.

48 COE

Cerebral sense organs. — Smaller than in most species of the genus.
They are situated slightly in front of the brain, but are well separated
from it because they lie near the ventro-lateral border of the head.
The canals placing them In communication with the exterior are, con-
sequently, extremely short. Each canal opens into the shallow,
oblique furrow on the side of the head. The posterior ends of the
sense organs lie beneath the anterior borders of the ganglia, but much
nearer the ventral surface. A pair of large nerves given off from the
dorsal ganglia opposite their commissure connect with the sense organs.
The lateral nerve cords unite above the posterior end of the intestine
as usual.

The body cavity in the esophagal region is filled with an unusually
large amount of gelatinous tissue, which occupies the considerable
space between the muscular layers and the esophagus and proboscis
sheath (p1. x, fig. 4). In this gelatinous tissue the lateral nerves are
situated, and through it a complex system of blood vessels and nephri-
dial canals ramifies.

The intestinal caecum is very broad and has but short lateral diver-
ticula. It lies wholly beneath the esophagus, and the branches do
not extend above the lateral nerves. The ctecum ends anteriorly far
behind the brain region. The esophagus becomes very small before
it empties into the dorsal wall of the broad intestine.

The attachment of the proboscis to the tissues of the head, the posi-
tion of the rhynchodaeum and its openings into the esophagus and pro-
boscidial cavity, the position of the dorsal and ventral brain commis-
sures, the cephalic glands and other organs are shown in p1. x, fig. 4.

The sexual products are mature in June. The whole body becomes
distended with the pouches of sexual elements, and the cavity of the
alimentary canal is much reduced in consequence. The ova are large
and deep olive-green.

The length of the specimens obtained, both males and females, was
about 75 to 100 mm. in extension.

Habitat. — This species was met with only at Farragut Bay, where
it occurred under stones in muddy locations at about half tide.

16. AMPHIPORUS NEBULOSUS sp. nov.

pi. IV, fig. I ; p1. viii, fig. 6; pi. xi, fig. i.
Body short, rather broad, and much flattened ; narrower anteriorly
than in the intestinal region, and tapering gradually posteriorly.
Mouth sub-terminal ; head pointed or expanded in front, according to

ALASKA NEMERTEANS 49

state of contraction. A V-shaped furrow is present on each side of
the head near the tip ; the angles of these furrows point obliquely
forward above and below.

Color. — Dull white or pale yellowish on dorsal surface ; very
thickly mottled with confluent dark brown blotches and dots which
largely obscure the ground color. Margins of the head without spots.
There are faint indications of a pair of transverse
lines without color — one near the tip of the snout
and the other near the brain region, the latter some-
times becoming a shallow, irregular, V-shaped fur-
row. Ventral surface dull white or yellowish,
without markings other than deeper yellow spots
which indicate the positions of the genital sacs, Y\g. 12 Amikiio-
and the darker color of the intestinal canal. rns nebulosus. Out-

OcellL — On each side of the head are from 18 line of the head to
to 25 ocelli, arranged in three irregular groups show arrangement of
(fig. 13). Close to the anterior border of the , ' ^
snout are 4 or 5 large cup-shaped ocelli on each

side. Behind these and bordering each lateral margin are about 8
much smaller eyes in an irregular group, while 3 or 4 small eyes are
scattered between these and the anterior group. Behind each lateral
group, and not far In front of the brain, about 7 to 10 small ocelli lie
in an irregular cluster deeper in the substance of the head, and are
therefore less easily visible.

Size. — The specimens obtained measured 100 to 150 mm. in length,
and 5 mm. in width. The esophagal region is short, rounded above,
flattened below, and thicker than the intestinal region.

Proboscis. — The proboscis sheath reaches nearly to the extreme end
of the body. Proboscis thick, fairly large, and white. It is provided
with 17 nerves. Basis of central stylet very much broadened poste-
riorly (p1. VIII, fig. 6), flat or even emarginate behind, narrow in front.
Central stylet as long as the basis, slender, acutely pointed. Accessory
stylets in two pouches ; similar to central stylet, but sometimes very
slightly curved ; commonly 3 in each pouch. The pouches lie well
behind the central stylet in ordinary extension (p1. viii, fig. 6) . Wreath
of gland-cells surrounding basis of central stylet moderately broad.

In the brain region (p1. xi, fig. i) and for some distance posteriorly
an abundance of large multicellular glands are thickly placed in the
ventro-lateral aspects of the body, and are mostly situated among the
fibers of the longitudinal muscular layer. Behind the brain these sub-
muscular glands become so closely packed together that the muscular

Proc. Wash. Acad. Sci., March, 1901.

50 COE

layer is divided into an outer and an inner portion in the region where
the glands are situated. The ducts from the glands (p1. xi, fig. i,
smg) pass through the muscular layers and basement membrane to
open to the exterior among the epithelial cells of the integument.
These glands become smaller and more scattered near the region of the
nephridial openings, but do not disappear entirely until near the end
of the esophagal region.

Alimentary canal. — The mouth and proboscis open together into
the rather long rhynchodaeum. The esophagus is as usual in the genus.
A single pair of small intestinal caeca reach forward well toward the
brain region. They lie immediately below the esophagus, and near
the middle line. Farther back they join a median, broad, unpaired
caecum. This has wide, paired, lateral outgrowths which, still farther
back, alternate with clusters of reproductive glands. These lateral ap-
pendages of the c£ecum lie mostly above the reproductive glands, as
well as alternate with them. Much farther back the esophagus de-
creases greatly in size and opens into the intestine by a longitudinal
slit in the dorsal wall of the latter. The intestinal pouches are rather
deep ; the anus is subterminal.

Nephridia. — The nephridial canals reach anteriorly nearly to the
region of the brain, where there is a single branched longitudinal ves-
sel on each side. Throughout the greater part of its length, this main
canal lies above the lateral nerve cord, but sometimes lies internal to
it or above it. In the anterior third of the esophagal region the main
canal becomes very large ; here an efferent duct branches off, passes
outside the lateral nerve, and bends downward to open on the exterior
of the body on the ventro-lateral aspect. There is but one efferent
duct on each side, and the two are sometimes exactly paired. Back of
the efferent ducts, the nephridial canals decrease rapidly in size, and
disappear far in front of the anterior end of the intestinal region.

A pair of large blood lacunae lie in the anterior portion of the head
as usual. In the brain region they divide into numerous smaller ves-
sels (pi. XI, fig. i). The lateral vessels form a broad anastomosis with
the dorsal vessel above the anus.

Nervous systei7i and se?ise orgaits. — The dorsal ganglia are closely
fused with the ventral. They are without distinct posterior lobes, and
sink gradually into the ventral ganglia (p1. xi, fig. i). The cerebral
sense organs are well developed. They lie in front of the brain,
some distance ventrally from the dorsal ganglia, and communicate with
the exterior by a canal which opens ventro-laterally. Each sense
organ is lobulated posteriorly and provided with a large nerve {son^

ALASKA NEMERTEANS 5 1

p1. XI, fig. i) which arises from the ventral side of the dorsal ganglion
near the ventral commissure. The union of the lateral nerve cords
above the anus lies in the same section as the anastomosis of the three
longitudinal blood vessels. A pair of small nerves from the lateral
cords continues backward beyond the commissure to the end of the
body.

Reproductive organs. — The sexual glands first make their appear-
ance in the esophagal region at the point where the unpaired intesti-
nal caecum receives its pair of anterior branches. Those sexual
glands which are situated most anteriorly lie below the alimentary
canal, internal to the lateral nerves, and open on the ventral surface of
the body. Farther back are commonly four or five reproductive
pouches in a single section. These lie mainly below the intestine,
but no matter what their position they all open to the surface of the
body below the lateral margins. Those lying farthest from the mid-
dle line and above the lateral nerves open ventrally between the nerve
cords and the lateral margins. Sexual products appear to be fully
mature in July.

Habitat. — Beneath stones near low water at Kukak Bay, Alaska
Peninsula (T. Kincaid).

17. AMPHIPORUS LEUCIODUS sp. nov.
p1. VII, fig. 6.

Body usually not more than 50 to 75 mm. in length, slender, elon-
gated, flattened posteriorly, not capable of great contraction. Head
narrower than parts immediately following. An inconspicuous V-
shaped furrow on dorsal surface back of head, seen only under favor-
able conditions.

This species resembles young individuals of A. exilis^ with which
it is often associated. It is likewise similar in many respects to A.
lactijloreus (Johnston) Mcintosh, from which it differs widely in the
armature of the proboscis and in many other anatomical features.

Color. — Opaque white, sometimes with a pale reddish or yellowish
tinge. This color is commonly uniform throughout, though it is some-
what influenced by the internal organs which show through the body
walls. The brain is pinkish, the intestine often brownish.

Ocelli. — There are four irregular, but usually distinct groups of
minute ocelli on the anterior portion of the head. Bordering each
antero-lateral margin of the tip of the snout is an elongated cluster of
about 8 to 12 ocelli (fig. 13). Posterior to these marginal clusters,

52

COE

Fig. 13. Amphi-
p o rti s leuciodus.
Outline of anterior
portion of the body
showing brain and
arrangement of the
ocelli. X 12.

and somewhat nearer the median line is a pair of clusters, each of
which likewise contains 8 to 12 ocelli. These posterior groups lie
immediately above the brain. Smaller and evi-
dently younger individuals have but 4 to 6 eyes in
each of the four groups. There is considerable
variation in the size of the ocelli.

Proboscis. — The proboscis sheath reaches nearly
to the posterior end of the body. The proboscis is
rather slender but ma}^ be contracted so that its di-
ameter is equal to more than half that of the body
itself. The proboscis is commonly attached to the
proboscis sheath at a point situated from ^ to 54^
the distance towards the posterior end of the body.
The armature is weaker than in A. exilis. The
central stylet is moderately slender and acutely
pointed. Its basis is somewhat conical in shape,
contracted slightly toward its middle portion (p1.
VII, fig. 6), and is rounded at its larger, posterior
end. It is slightly longer than the stylet itself. The number of
pouches of accessory stylets is commonly three, although there are
sometimes only two, and occasionally a specimen is found which has
four. There are usually two or three stylets in each pouch. In a few
instances, however, four and five were observed. Where three pouches
are present, as is usual, they are situated at nearly equal distances on
the circumference of the proboscis (p1. vii, fig, 6).

Cerebral sense orgatis. — Moderately small and situated well in front
of brain and on ventral side of head. They communicate with the
exterior by means of a pair of canals whkh open on the latero-ven-
tral margins of the tip of the head.

Nephridia. — The nephridia reach forward to the brain region.
There are several pairs of efferent ducts, some of which open on the
ventral, and some on the dorsal surface of the body. In one of the
specimens sectioned the first pair of efferent ducts extended from the
internal side of the lateral nerve cords and opened directly to the
latero-ventral aspect of the body after passing on the ventral side of
the nerve cords. A little farther back in the esophagal region were
two efferent ducts on one side and one on the other which passed above,
and externally to the nerve cords to open likewise below the lateral
margins. In the remainder of the esophagal region were three more
efferent ducts on each side. With one exception all of these passed
above the lateral nerves and opened on the dorso-lateral surfaces of the

ALASKA NEMERTEANS 53

body as in A. exilis. Another specimen had 7 efferent ducts on the
left side and 8 on the right. Of those opening on the left side the
four anterior ones passed dorsally to the nerve cord and then bent ven-
trally to open on the ventro-lateral aspect of the body ; the fifth one
opened very near the lateral margin, and the last two opened on the
dorso-lateral surface. On the right side the four anterior ducts opened
ventro-laterally, and the four posterior ones opened on the dorso-lateral
surface. The nephridia extend backw^ard beyond the first few pairs
of reproductive glands. Here, then, we find the connecting links be-
tween the typical AmpJiiportis nephridium (which passes above and
external to the lateral nerve and then bends downward to open on the
ventro-lateral aspect of the body) , and the type of nephridium which
is characteristic of A. exilis and the Heteronemerteans, and which
opens directly on the dorso-lateral aspect of the body.

Sub-tnuscular glands are closely packed together in front of the
brain and in the brain region. They are not found farther posteriorly,
and in this respect the species differs markedly from A. exilis.

The intestinal ccecuni reaches forward well toward the anterior end
of the esophagal region. The main caecal cavity, which lies directly
beneath the esophagus, sends off numerous lateral pouches above the
lateral nerves, and at its anterior end branches into lateral diver-
ticula which extend forward on each side as far as the brain re-
gion.

Reproductive glands. — The genital products mature in June in the
region of Victoria, B. C. The ova develop in sacs vvhich are regu-
larly arranged, and extend from the posterior third of the esophagal
region to the posterior end of the body. The ovaries, in all cases
noticed, were situated immediately above the lateral nerves. In the
intestinal region they alternate with the intestinal lobes with a great
deal of regularity. Their efferent ducts occupy positions on the latero-
dorsal aspects of the body exactly corresponding to those of the poste-
rior efferent nephridial ducts. The ducts from the ovaries, however,
pierced only the longitudinal muscular layer, and did not penetrate the
circular muscular layer of the body wall. These rudimentary genital
ducts are further distinguished from the nephridial ducts by lacking a
conspicuous epithelial lining.

Habitat. — These slender whitish worms were found in great abun-
dance beneath barnacles and other growths on the piles of the wharf
at Victoria, B. C. They were found less abundantly under stones be-
tween tides at New Metlakahtla and in Glacier Bay, but were not
noticed farther northwest.

54

COE

i8. AMPHIPORUS EXILIS sp. nov.

p1. Ill, fig. I ; p1. VII, fig. 5 ; pi. xi, fig. 3.

Body extremely elongated for the genus, rounded throughout ; not
capable of the great contraction which characterizes many species of
the genus ; only moderately flattened ; posterior extremity slender.
Head usually narrower than esophagal region. The general shape of
the body resembles that of E?nplectonema. In shape of body, as
well as in color and habits, this species, like the
last, recalls A. lactifloreus (Johnston) Mcintosh.
Its anatomical structures are, however, very differ-
ent, as will be seen from the following description.
Ocelli. — Exceedingly numerous and minute.
They are arranged on the head in four elongated
clusters (fig. 14)' Two of these clusters lie on each
antero-lateral margin of the head, while the two
other groups lie more posteriorly (just in front of
the brain), and extend from near the middle line
obliquely outward and backward. Sometimes the
two posterior clusters are united in front into a
continuous V-shaped group. The number and dis-
tribution of these eyes is indicated by the following table, which shows
the numbers found in ten individuals :

Fig. 14. Amphi-
forus exilis. Dor-
sal view of head to
show arrangement
of ocelli. X 8.

No. of ocelli in

anterior clusters.

No. in posterior

clusters.

Right.

Left.

Right.

Left.

I.

13

II

15

17

2.

16

17

26

28

3-

18

19

33

30

4-

23

23

32

31

5-

25

24

40

42

6.

25

26

52

46

7-

28

26

45

46

8.

35

37

54

45

9-

35

35

51

53

10.

60

56

71

73

In the specimens examined, therefore, the number of ocelli in the
front clusters varied from 11 to 60, that in the posterior clusters from
15 to 73. The average in the 10 individuals is about 28 in each an-
terior cluster, and 41 in each posterior group.

All the ocelli are minute, but nevertheless very irregular in size, some
being several times as large as others. From the ventral surface the
anterior marginal clusters only are seen.

ALASKA NEMERTEANS 55

Proboscis. — The proboscis sheath is long and slender. Even in this
elongated species it reaches within a few millimeters of the posterior end
of the body. The proboscis also is slender, though of moderately large
size. It reaches well backward in the body. The armature of the?
proboscis is especially remarkable. The central stylet is moderately
slender, rather small, and rests on a moderately heavy basis. The
basis is somewhat conical in form, and rounded posteriorly (p1. vii, fig.
5). In addition to the central stylet there are usually 8 pouches of ac-
cessory stylets (p1. vii, fig. 5), though the number varies from 6 to
12. In each pouch are one or two slender stylets. Often there is ^
single fully developed stylet, and a second, immature stylet in most of
the pouches. The proboscis is usually extruded when the animal is
killed.

Cerebral sense orgatis. — Situated far in front of brain — nearly at
end of snout when the head is contracted — and fairly well developed.
Their canals open on antero-lateral borders of tip of snout.

The blood-vascular system resembles that in other species of the
genus.

Nephridia. — The nephridial system shows peculiar deviations from
the arrangement usually found in the Metanemerteans. A pair of
main canals with numerous branches runs longitudinally above the
lateral nerve cords, as in other species. These reach forward welj
toward the brain and extend posteriorly far into the intestinal region.
Their branches ramify both above and below the lateral nerve cordsi
The number and position of the efferent ducts is remarkable — there are
commonly 20 or more on each side. The first is near the anterior end
of the main nephridial canal, and sometimes opens on the ventro-lateral
aspect of the body, as in other species of the genus. Back of this,-
however, were counted nine other efferent ducts on each side in the
esophagal region, and at least as many more were present on each side
in the intestinal region. These ducts were mostly small and opened
on the dorsal aspect of the body, as in many Heteronemerteans. The
positions of the efferent ducts were sometimes immediately above the
lateral nerves, and soinetimes but a little laterally from the proboscis,
sheath. Most commonly, however, the ducts occupied positions be-
tween these two extremes, so that the majority of the nephridlopores
were situated on the dorsal surface about half way between the lateral
margin and the median line (p1. xi, fig. 3). As noted on page 52, am
approach to this condition is found in A. leuciodus. These appear tc>
be the only species of the genus, and indeed the only Metanemerteans^
in which the nephridlopores are situated on the dorsal surface of the
body.

56 COE

The intestinal ccecum is enormously developed. Its diverticula
reach forward on each side even to the anterior end of the brain. In
the brain region each of the caecal diverticula appears as a rounded lobe
on either side directly above, and closely approximating to, the dorsal
brain-lobe. Back of the brain there are several rather slender lobes
on each side. These lie mainly above the lateral nerves, but send off
branches below the nerves at frequent intervals. Somewhat farther
back in the esophagal region these lateral lobes join the main, un-
paired caecum, which lies immediately below the esophagus. This
caecum, throughout its course to the intestine proper, gives off nu-
merous lateral diverticula on each side, and these branch upward
above the lateral nerve cords (p1. xi, fig. 3).

Sub-muscular glands. — These occur abundantly in the brain region,
and are still more closely packed together in the anterior portion of
the esophagal region. They occupy positions, as in other species, in
the connective tissues beneath the musculature in the latero-ventral
regions of the body. Each gland is composed of a number of cells,
and each has a duct leading through the layers of the body wall to the
exterior on the latero-ventral aspect. A much smaller number open
on the dorsal surface. In the posterior end of the esophagal region
these glands have almost entirely disappeared, although a few are met
with in the anterior portion of the intestinal region.

Color. — The color of the individuals of this species is commonly a
homogeneous, opaque white, very pale flesh color, or pale yellowish-
white. This color is continuous throughout the length of the body,
both above and below, except where the internal organs show through.
The smaller specimens are pale, while the larger ones almost always
exhibit a brownish color, which indicates the position of the intestine.
Sometimes minute reddish-brown specks are distributed over the dorsal
surface. Occasionally a worm of this species is met with in which
the intestinal lobes are pale orange. The brain is plainly distinguish-
able in the living worm because of its pinkish coloration.

Habitat. — Amphiporus exilis is one of the most common, as well
as the most widely distributed species of nemerteans met with on the
expedition. It occurred abundantly at nearly all the collecting stations
from Victoria, B. C, to Dutch Harbor, Unalaska. The worms are
restless and are often seen crawling over stones between tides. They
live among barnacles, mussels, etc., from low water well up to high
water mark, and are found abundantly under stones in almost all sorts
of locations. The species is especially hardy.

ALASKA NEMERTEANS 57

TETRASTEMMA Ehrenberg.
Symbolae Physicae, Berlin, 1831,

This genus includes a group of very small, slender worms, seldom
more than 20 to 30 mm. long, with slightly flattened body, and usually
with four well-developed ocelli, which form a quadrangle on the head.
In a few species (cf. T. aberrans) these ocelli are each replaced by a
group of two or three smaller ones, and in other species ocelli are
wanting entirely. The anatomical structures are very similar to those
of AmpJiiporus^ and the distinctions between the two genera are not
clearly defined. The mouth and proboscis open together. The cere-
bral sense organs lie close in front of the brain. The proboscis sheath
extends to the posterior end of the body, and the proboscis is well de-
veloped, armed with central stylet and pouches with accessory stylets,
and usually provided with ten nerves.

Only three species of the genus were met with on the expedition,
although it seems probable that a number of other forms of these minute
worms will be found later.

19. TETRASTEMMA BICOLOR sp. nov.

p1. I, fig. 6.

Body moderately slender, rounded both above and below; much
larger and longer than most species of the genus, sometimes becoming
50 to 60 mm. in length in extension.

Color. — This species is bright brownish-red or orange the whole
length of the dorsal surface ; the whole ventral surface is pale gray or
whitish. The anterior border and lateral margins of the head, as well
as the lateral margins of the body for a short distance back of the head,
have the same whitish color as the ventral surface. A narrow, median,
white stripe, sharply marked off from the reddish color of the dorsal
surface, extends from the white, anterior border of the head nearly to
the posterior end of the body. Posteriorly the stripe becomes more
irregular and is usually lost near the posterior extremity.

Ocelli. — Four, rather large, rounded, arranged nearly in the form
of a square.

Proboscis. — Proboscis sheath and proboscis as in typical species of
the genus. Proboscis provided with a moderately heavy central stylet
about .075 mm. in length. Basis of central stylet somewhat conical
in shape, swollen behind, and about ij^ times as long as the stylet it-
self. There are two pouches of accessory stylets, with usually three
or four stylets in each pouch.

58

COE

The blood is dark red, and the blood vessels may be traced in the
living worm the whole length of the body.

Habitat. — The species was found only at Kadiak, where it was
dredged in about three fathoms. It slightly resembles some varieties
of T. vermiculus Quatrefages, but the longitudinal bands of dark
pigment between the two ocelli of the same side are lacking, and the
median white line is sharply demarkated, so that the two species are
undoubtedly specifically distinct.

20. TETRASTEMMA ABERRANS sp. nov.

This is a minute Nemertean, the specimens found not exceeding 12
mm. in length in greatest extension. Body moderately slender, slightly
flattened. A pair of slight vertical slits on sides of head.
Color. — Pale yellow throughout, both above and below.
Ocelli. — Of moderate size, or rather small, arranged in four groups
which form a rectangle, as do the single eyes of typical species of the
genus. Each of the four groups is composed of three to five ocelli of
variable size. The two anterior groups lie well toward the tip of the
snout, while the two posterior groups lie above or slightly in front of
the brain (fig. 15). The appearance of the eyes
is such as to give at once an impression that the
multiple nature of each group has arisen from the
fragmentation of single ocelli. This I consider to
be the case. At least one other species of Tetra-
ste77ima ( T. cruciatjim Burger ) is known in
which the four ocelli are double, and I have
often noticed other species of the genus — notably
the fresh water T. rjibrinn (Leidy) — which
showed almost conclusively that one or more of
the six or seven eyes present had been derived
from a splitting of the primary ocelli. In one
instance the fragmentation had been carried so far
that no fewer than 30 to 30 pigment spots were
present.

Proboscis sheath afid proboscis as in other
species of the genus. Mouth and proboscis open
together ; proboscis sheath reaches the posterior end of the body. Pro-
boscis provided with a rather slender central stylet and basis ; the two
accessory stylet pouches each with two or three st)-lets of typical form.
The proboscis possesses twelve nerves, and in this respect again shows
a departure from the typical Tetrastemma, where there are usually
but ten proboscidial nerves.

Fig. 15. Tetrasiem-
ma aberrans. Dorsal
view of anterior por-
tion of body, showing
outline of brain and ar-
rangement of the four
groups of ocelli. X 25.

ALASKA NEMERTEANS 59

Nephridia. — The nephridial canals are short, and reach forward to
the brain. Anteriorly each nephridium lies above the lateral nerve,
but farther back the branches ramify both above and below the nerve.
Near the anterior end of each of the main canals a large efferent duct
passes outside of the lateral nerve, and bends downward and outward
to open to the exterior on the lateral margin of the body. In one in-
stance a double efferent canal was observed.

The head is provided with large cephalic glands which open on the
tip of the snout. The cerebral sense organs are large. They lie
mainly in front of the brain, although their posterior ends extend
backward beside and lateral to the brain lobes. Each connects with
the exterior by a canal which runs anteriorly to open on the lateral
margin of the head.

A broad intestinal ccecujn reaches into the anterior fourth of the
esophagal region. It lies below the esophagus, and sends off a few
wide lobes on each side, but these reach dorsally only a little above the
lateral nerve. The esophagus opens into the caecum far behind the
anterior sexual glands, and nearly as far back as the middle of the
body. The reproductive pouches lie both above and below the intes-
tine.

As will be seen from the above description, this species agrees
closely with the typical species of Tetrastemma in size, general appear-
ance, and in the details of the internal anatomy. It differs from
known species of the genus only in the fragmented nature of its eyes
and in the number of nerves in the proboscis. It seems extremely
probable, however, that when more of the described species have been
examined in this regard, some of them will be found to contain more
or less than ten proboscidial nerves. The eyes certainly resemble
those of Tetrastemma more closely than they do those of any de-
scribed species of Amphiporus.

Habitat. — Found among hydroids in abovit four fathoms in Glacier
Bay, and between tides at Orca, Prince William Sound. Not com-
mon.

21. TETRASTEMMA C^CUM sp. nov.

A small species which I shall refer provisionally to this genus was
found by Ritter in considerable numbers at Kadiak. The species is
especially remarkable, and quite aberrant from most other species of
letrastemma.! both in lacking ocelli and in being hermaphroditic.
The body is rounded and of almost even diameter throughout. The
head has a pair of inconspicuous, oblique, lateral furrows.

6o COE

Color, — The worms are very pale, whitish or pale yellowish in color,
with dark intestinal lobes.

Size. — Very small, not usually more than 5 to 10 mm. long and
0.5 to I mm. in diameter when sexually mature.

Ocelli. — Wanting.

Proboscis. — The proboscis sheath reaches nearly to the end of the
body. The proboscis is remarkable for its enormous size as com-
pared with the size of the body — when everted its diameter is practically
equal to that of the body itself, although it is then comparatively short.
Its armature consists of a central stylet with rather slender basis, and
of two pouches of accessory stylets. The basis of the central stylet
is slightly swollen and somewhat sharply truncated posteriorly. Each
pouch contains 2 or 3 long, slender and delicate accessory stylets.

In cross section of the proboscis the inner and outer circular muscu-
lar layers appear as usual. The intervening layer of longitudinal mus-
cles, however, is divided into two secondary layers separated by a thick
sheet of nerves and connective tissue. The nerves appear to be more
or less confluent, and not separated into a definite number of longitudi-
nal cords as in most other species of Tetrastemma. This appearance
may be partially due, however, to the action of the formalin in which
the worms were preserved.

Sense Organs. — The cerebral sense organs are very well developed,
and unusually voluminous as compared with the other organs of the
head. They lie in front of the brain, and extend posteriorly on the
ventral side of the brain lobes as far as the ventral commissure. Each
sense organ communicates with the exterior by a small canal passing
obliquely forward to open on the lateral margin of the head.

The brain is of small diameter, but its extent antero-posteriorly is
comparatively great.

Reproductive organs. — The individuals are hermaphroditic, and
probably to some extent protandric although there is considerable varia-
tion in this respect. One of the individuals sectioned was filled with
ripe spermaries only, but all the others possessed enormous ova, with
the spermaries disposed irregularly. The mature ova were fully two-
thirds the diameter of the body, and hence were arranged at irregular
intervals in a single row. Where the ova were mature the spermaries
were smaller and contained much fewer spermatozoa than in those in-
dividuals which were without large ova. Many of the spermaries had
ducts which pierced the muscular layers to reach the dorso-lateral sur-
faces of the body. In some cases where these ducts were fully formed,
and the spermatozoa therefore mature, the ov^ in the same individual
were but half grown. The sexual products were mature in July.

ALASKA NEMERTEANS 6l

T^NIOSOMA Stimpson.

Tam'osoma STlMFSOH.'Froc. Acad. Nat. Sci.» Philadelphia, p. 162, 1857.
Po/i'a Delle Chiaje, Mem. suUa storia e notomia degli anamaii senza

vertebre del regno di Napoli, Naples, 1823-28.
Eupolia HuBRECHT, Report of Challengei Exped. Z06I., xix, 1887.
Eupolia Burger, Fauna u. Flora von Neapel, Monogr. 22, p. 598, 1895.

For the reasons given on page 4, it seems absolutely necessary to
adopt for this genus the name given by Stimpson in 1S57 rather than
accept that of Hubrecht of 30 years later, even though most European
writers have ignored Stimpson's brief, but careful, diagnoses.

The species belonging to this genus show a remarkable specific vari-
ation in the general shape and size of the body. Some are character-
ized by extremely long, slender, flattened, and much twisted bodies,
while others are short, thick, and cylindrical. In all the species, how-
ever, the head in life is rounded in front and is sharply marked off
from the parts immediately following by lateral constrictions. Hori-
zontal furrows are w^anting, but small, oblique or transverse grooves
may be present on the head. In strong contraction the esophagal
region becomes greatly swollen, the head is drawn in (p1. ii, fig. 4),
so that the anterior end of the body is large and shortly truncated.

Proboscis sheath and proboscis short, seldom reaching more than
one-third the length of body. Proboscis opening subterminal, minute.
Mouth a small round opening on the ventral surface immediately be-
hind the ganglia.

Muscular layers of body composed of a thick outer longitudinal, a
circular, and a less thickened inner longitudinal layer. Outside the
muscular layers is a well developed cutis, composed of a thick inner
layer of connective tissue, and an outer layer of glandular tissue. The
external epithelium is thin, as compared with the other layers of the
body, though the fibrous layer separating it from the cutis is well de-
veloped. The musculature of the proboscis consists of an inner longi-
tudinal, and an outer circular muscular layer. Consequently there can
be no muscular crosses.

The cephalic glands are enormously developed. They stretch back-
ward on all sides beyond the brain, and even reach some distance into
the esophagal region.

The lateral nerves lie immediately outside the circular muscular
layer. There are three longitudinal blood vessels.

Ocelli are usually present in great numbers, though very small.

The worms are sluggish in their habits, are unable to swim, and
usually show great irregularities in the diameter of the body. They

62 COE

are prone to twist themselves in sharp coils, or in knots, and often lie
in lumps. They are usually capable of contracting and extending their
bodies to a remarkable degree.

22. T^NIOSOMA PRINCEPS sp. nov.

p1. II, figs. 3, 4.

Body of very large size, long, thick, largest in the esophagal region,
cylindrical in anterior portion, flattened on ventral surface posteriorly ;
in contraction nearly cylindrical throughout. Head sharply marked off
from body in extension, rounded in front, flattened dorso-ventrally ; in
contraction drawn almost entirely into the parts immediately following,
so that the anterior portion of the body is greatly swollen and sharply
truncated in front. In contraction the anterior end of the body is
thrown into massive folds, and the whole body is remarkably short
and thick (pi. 11, fig. 4). A pair of inconspicuous, oblique grooves
lies on the antero-ventral surfaces of the head ; into these the canals
leading from the cerebral sense organs open. The esophagal region is
scarcely more than one-twelfth the length of the body. In alcoholic
specimens there is usually a median ridge on the dorsal surface run-
ning the length of the body, except in the head and esophagal regions.

Color. — The dorsal surface is deep ochre yellow, sometimes inclin-
ing to orange, and sometimes to brownish, and thickly strewn with
minute irregular dark red spots. The reddish markings are most con-
spicuous near the anterior end of the body, and in the dorsal, median
line. In many places a large number of the reddish dots become con-
fluent, and form an irregular patch of deeper color. These patches
commonly occur as broken longitudinal lines. Such lines are most
abundant on the middle of the dorsal surface where they form a median
longitudinal band of reddish-brown. The ventral surface is paler and
the reddish markings are wanting. Along the median line the color is
brighter yellow than elsewhere on the ventral surface. This is appar-
ently due to the absence in this position of the opaque intestinal lobes.
The ventral surface often exhibits a greenish tinge to the yellow ground
color. The posterior extremity is pointed and much paler than the
rest of the body.

Size. — T. princeps grows to a size greater than has previously
been described for any species of the genus. The individuals found
were from half a meter to two meters in length when extended ; when
contracted, but a small fraction of this length, and proportionately
thick (p1. 11, fig. 4). After long standing in alcohol a section of the

ALASKA NEMERTEANS 63

body of one specimen still measured 15 x 18 mm. It is one of the
largest Nemerteans of the coast.

Ocelli. — There are many minute ocelli arranged in an irregular
group on each side of the tip of the head. The number of such ocelli
is commonly 40 or more in each of the two groups.

The mouth is a small rounded pore, and is situated several milli-
meters back from the tip of the head in large individuals. The pro-
boscis-pore is also minute, and is situated subterminally as usual.

Proboscis. — The proboscis sheath reaches some distance into the
esophagal region, but is very short vv^hen compared vv^ith the length of
the body. The proboscis is short and vv^eak. Its muscular and epi-
thelial layers are as in other species of the genus.

The cephalic glands are enormously developed. They occupy a
large proportion of the area in the anterior portion of the head, sur-
round the brain on all sides, and extend still further backwards into
the anterior end of the esophagal region. Here they lie in the outer
longitudinal muscular layer around the whole circumference of the body.

Body walls. — The outer longitudinal muscular layer about equals in
thickness the other two muscular layers combined. In the intestinal
region the inner longitudinal muscular layer becomes extremely thin on
the lateral aspects of the body, and is much reduced dorsally. It is only
on the ventral side that this layer retains its comparative thickness.
The cutis is thinner than in most species of the genus, and especially
is this true of its inner, fibrous layer. This layer is, throughout most
of the body, reduced nearly to the condition of a thin membrane. It
is commonly not much thicker than the muscular layer beneath the
body epithelium. The epithelium itself is thin in comparison with the
massive muscular layers of the body.

The blood lacunce in the head lie directly above the brain, as in other
species of the genus. They are, however, remarkably large, and
are crossed in various directions by numerous bundles of muscular
tissue, which tend to subdivide the lacunae into numerous smaller
spaces. The longitudinal blood vessels are as usual. The dorsal vessel
passes out of the proboscis sheath early in its course.

Nephridia. — Situated in anterior and middle portions of esophagal
region. Several efferent ducts on each side. These are small in
diameter and pass above the nerve cords, opening on the lateral aspects
of the body dorsal to the lateral margins.

Cerebral sense organs. — The dorsal lobes of the brain greatly exceed
the ventral lobes in size, and lie somewhat lateral as well as above
them. The cerebral sense organs are voluminous. They extend for-

64 COE

ward on each side, external and ventral to the dorsal brain lobes, nearly
as far as the ventral commissure. Here, at the anterior extremity of
each sense organ, a canal passes obliquely downward and outward to
open into a shallow oblique furrow on the ventro-lateral aspect of the
head. The sense organs are closely united with the posterior ends of the
dorsal brain lobes. In the middle region of the brain, the anterior ends
of sense organs are triangular in section, and lie external and between
the dorsal and ventral ganglia. Farther back a glandular appendage
of the sense organ pushes itself in between the internal faces of the
brain lobes. This appendage fuses with the ventral portion of the
sense organ more posteriorly. The brain lies deeply buried in the
tissues of the head and nearly in the median line, but the lateral nerves
while still in the region of the cephalic sense organs bend sharply
outward, and occupy throughout the remainder of their course posi-
tions immediately external to the circular muscular layer of the body
wall.

The genital products were nearly mature in July. The oviducts
were in many instances preformed, and opened on the dorso-lateral
aspects of the body.

Habitat. — Only four individuals of this species came under my ob-
servation. One was collected at Cape Fox (Kincaid), two at
Yakutat, and the fourth at Orca in Prince William Sound. All were
found under stones in rather hard mud at low water. The individuals
contract strongly when handled, throwing the surface of the body into
wrinkles, and often coil the posterior portion of the body into a close
spiral, much as does Cephalothrlx.

LINEUS Sowerby.

The British Miscellany, London, p. 15, 1806,

Representatives of this genus are characterized by a slender, some-
times thread-like body, usually rounded throughout. The body is
commonly twisted and coiled into an irregular mass. The movements
are sluggish. The animals creep over objects and readily move about
on the surface of the water, but they are unable to swim. The body
is extremely contractile ; the head is often slightly wider than the body,
of oval shape, and is usually provided with numerous minute ocelli,
often arranged in a single row on each side of the head. A caudal
papilla or cirrus, a diagonal muscular layer, and neurochord cells are
all wanting. The proboscis sheath is often short in comparison with
the length of the body.

ALASKA NEMERTEANS

65

23. LINEUS VIRIDIS (Fabr.) Johnston.

Planarta viridis O. Fabricius in O. F. Miiller, Zool. Dan. Prod., 1776 ; Fauna

Groenlandica, p. 324, 1780.
Planaria gesserensis Muller, Zool. Danica, 11, p. 32, 1788.
Nemertes obscura Desor, Boston Journ. Nat. Hist., vi, pp. i to 12, 1848.
Linens viridis Johnston, Catalogue British Non-parasitical Worms, pp. 27,

296, London, 1865.

As stated by Verrill,^ there seems little doubt that the description of
this species sent by Fabricius to Muller and published by him in the
Zoologica Danica, is entitled to retain priority in nomenclature instead
of 'hJlXxViOx''?, gesserensis of later date, which has been adopted by most
recent European writers-
Characteristic individuals of this species were found under stones at
low water at New Metlakahtla, on Annette Island.

Body moderately slender, rounded throughout, but slightly flattened
posteriorly ; head slightly wider than the parts immediately following ;
cephalic slits long and deep, with pale margins above and below,
reaching anteriorly close to the proboscis pore. The anterior end of
the mouth does not reach quite so far forward as the posterior end of
the cephalic slits. Length usually 100 to 200 mm.

On each side of the head in front of the brain and
close to the lateral borders is a single row of minute
ocelli (fig. 16). The number of these is commonly
from four to six on each side, though some individuals
have as many as eight, and very young specimens but
a single pair.

Color. — The Alaska specimens were dusky or brown-
ish green, becoming dark brown anteriorly, and com-
monly paler on the ventral surface, especially posteriorly.
The head is very pale on lateral margins and in front.
The brain is large, reddish, and shows distinctly
through the pigment of the body. Cerebral sense or-
gans paler but easily distinguished in life, with con-
spicuous canals leading to the posterior ends of the of Linens viri-
cephalic slits (fig. 16). ^« showing ar-

Habitat.—Thh species, besides being found in Alas- '"angementof the
1 • . 1 1 1- -1 1 • 1 ^ , ocelh, and posi-

ka, IS widely distributed in northern waters. On the ^.j^j^ ^^ brain

east coast of America it occurs from Long Island Sound phalic sense or-
to Greenland. It is found on nearly all the coasts of gans and mouth,
northern Europe. It has also been found in the Medi- X^.
'Trans. Connecticut Acad., viii, p. 421, 1892.
Proc. Wash. Acad. Sci., March, 1901.

Fig. 16. Out-
line of anterior-
portion of body

66 coE

terranean, though it is there comparatively rare and small. It is usu-
ally found between tides under stones in muddy localities.

24. LINEUS TORQUATUS sp. nov.
p1. V, figs. 8, 9.

Body rather thick and stout for the genus, somewhat flattened
throughout, but especially posteriorly and on the ventral surface.
Head short, pointed in front, somewhat narrowed behind, slightly flat-
tened. A slight annular constriction commonly marks off the head
region from that immediately following. Sometimes this constriction
is very conspicuous, and the head much narrower behind. Esophagal
region rounded above, flattened below. Intestinal region commonly
well flattened but without narrow margins. Posterior end slender.

Cephalic furrows rather short. In alcohol or formalin they join
the terminal proboscis pore in front, but in life they are separated from it.

Ocelli. — Absent, at least in mature individuals. Mouth a minute
pore or a large slit, according to state of contraction ; situated a little
behind posterior end of cephalic furrows.

Color. — The color is usually dark, reddish-brown, chocolate, or pur-
ple above ; paler and commonly more reddish beneath. The dorsal
surface is often flecked with irregular minute, inconspicuous whitish
specks. A narrow transverse white band passes across the dorsal sur-
face at the posterior ends of the cephalic furrows. This characteristic
marking reaches only to the lateral edges, and is not seen from the ven-
tral surface. The cephalic furrows are sometimes, but not always,
bordered above and below by a narrow band of white. Furthermore
a minute white spot occupies the region of the proboscis pore. The
white borders of the cephalic furrows commonly connect this white
spot with the white transverse dorsal band on the posterior portion of
the head. On the extreme tip of the snout — in the white area around
the proboscis pore — a pair of small pigment spots is sometimes pres-
ent, one on each side of the proboscis pore.

Size. — Length usually 200 to 400 mm. in extension, width about
5 mm.

Proboscis. — The proboscis is moderately slender and of medium
size. It is without color. The inner longitudinal muscular layer is
almost entirely wanting ; the crosses between the circular layers are
distinct, and the nervous plexus is exceptionally well developed. A
single pair of large nerves enters the proboscis at its point of attach-
ment at its anterior end. They originate from the ventral commissure
of the brain near the ventral ganglia, enter the proboscis from the ven-

ALASKA NEMERTEANS 67

tral side, pass backward for some distance on the right and left sides
respectively, and later spread out into a cylindrical plexvis immediately
internal to the muscular layer, and separated from the inner epithelium
only by a few longitudinal muscular and connective tissue fibers.

Glands. — The cephalic glands are very well developed, and occupy
a large portion of the region of the head in front of the brain, both
above and below the rhynchodaeum. They do not extend posteriorly
quite so far as the brain.

Nerves. — The nervous system is far more easily made out than in
any of the related species with which I am acquainted. The individ-
ual nerves are large and are sharply defined in all cases. The nerves
extending from the brain toward the tip of the snout are numerous
and are all of large size. The esophagal nerves are also surprisingly
large and quite conspicuous. They have several transverse connecting
branches after their origin from the ventral ganglia. The most pos-
terior of these connecting branches occurs just in front of the mouth.
In the mouth region they communicate in several instances with the
lateral nerves by means of branches which pass through the nervous
plexus outside the circular muscular layer. These branches from the
lateral nerves follow the nervous plexus to the vicinity of the esophagal
nerves, which they join by passing directly through the circular mus-
cular layer. I do not know that such an anastomosis between the
lateral nerve cords and the esophagal nerves has been previously noted
in any species.

The median dorsal nerve, situated just outside the circular muscular
layer, is unusually conspicuous, and throughout a considerable portion of
the intestinal region is supplemented by a second median nerve lying
directly beneath the first, but in the midst of the internal longitudinal
muscular layer. Branches connect these two nerves at frequent inter-
vals.

Cerebral sense organs. — Voluminous. The canals leading to the
exterior open on the summit of a broad papilla situated at the posterior,
widened end of each of the cephalic slits.

JVephridia. — The nephridial canals are of much greater diameter
than I have observed in any other Nemertean. The canal on either
side is, throughout a portion of its length, equal to the lateral nerve
cord in cross section. The nephridia extend through the anterior half
of the esophagal region. The main tubule has a few very large
branches, and these lie in the walls of the blood spaces about the esoph-
agus. Anteriorly the branches lie mainly dorsal to the lateral
nerves, but towards the posterior ends of the nephridia the branches

68 coE

ramify more towards the ventral side of the esophagus. There is a sin-
gle pair of efferent ducts which are of enormous size as compared with
those of other species. These ducts are situated at about two-thirds the
distance towards the posterior ends of the nephridia, and open on the
dorso-lateral aspects of the body as usual.

Habitat. — This species is common in mud and under stones in
muddy localities at Orca and Virgin Bay in Prince William Sound,
but was not met with elsewhere on the expedition. The individuals
are hardy and of sluggish movements. They do not break up nor
contract excessively when thrown into killing fluid, and the proboscis
is not usually everted when the animal is killed. Some, especially
the smaller ones, when preserved are nearly cylindrical, but most indi-
viduals are flattened ventrally. The color is fairly permanent in for-
malin, and even in alcohol for some months.

MICRURA Ehrenberg.
Symbolae Physicae, Berlin, 1831.

This genus includes mostly moderately small, slender forms, gener-
ally less rounded posteriorly, and of rather more active habits than
Lineus. Its most marked distinction from the latter genus is that the
posterior extremity of the body is provided with a slender, usually
colorless, muscular caudal cirrus. This is formed of a continuation
of the muscular tissues and integument beyond the posterior end of the
alimentary canal.

The species of Micrura are generally, though not always, more
brightly colored and have more distinct markings than those of
Lineus. The vast majority of the species are provided with numerous
ocelli, though some are blind. The head is slender, and not distinctly
separated from the rest of the body. The lateral faces of the body are
not provided with thin edges as in Cerebratuhis^ the intestinal region
is not so much flattened, neurochord cells are not present in those
species which have been studied, and none of the species are able to
swim as do all species of Cerebratulus. The mouth is usually smaller
than in Cerebratulus^ and the intestinal lobes are not so deep.

The proboscis is usually slender and comparatively weak ; the pro-
boscis sheath is sometimes considerably shorter than the body.

25. MICRURA VERRILLI sp. nov.
p1. v, figs. I, 2, 3.
Body moderately elongated, widest anteriorly, tapering to an acute
anterior extremity ; much more slender posteriorly. Ventral surface

ALASKA NEMERTEANS 69

flattened ; dorsal surface rounded. Head narrow in front. Cephalic
furrows long and deep ; at their posterior ends each is met by an ob-
lique, shallow depression above and below. Proboscis-pore exactly
terminal ; anterior ends of cephalic furrows well separated from pro-
boscis pore. Mouth small, situated as far back as posterior ends of
cephalic furrows. Caudal cirrus small, slender, easily broken off.

Color. — This is one of the handsomest and most striking of all the
described species of Nemerteans. The color of the dorsal surface is
deep purple or wine-color ; that of the ventral surface is purest white.
On the dorsal surface are usually 15 to 40 very sharp, pure white,
transverse bands or lines connecting with the white color of the ven-
tral surface. These transverse bands are nearly as narrow as pencil
marks, and are situated at fairly regular intervals throughout the length
of the body. They are sometimes more or less interrupted, but
always sharp and distinct. On the dorsal surface of the tip of the
snout is a small triangular marking, always very conspicuous because
of its bright orange color. Following behind this is a narrow trans-
verse white band, followed by broad purple and narrow white bands
in succession. The cephalic furrows lie within the white color of the
ventral surface, and are bordered above with a very narrow margin of
white. They reach posteriorly as far as the second white band.

After preservation in formalin or alcohol the worms are strongly
rounded below as well as above, and the color of the dorsal surface
changes from purple to red, while the orange spot on the tip of the
snout disappears.

Ocelli. — None were found.

Proboscis. — Colorless, and shorter than in many species of the
genus. Its microscopic structure shows distinctly the three muscular
layers and the muscular crosses characteristic of the family. The
inner circular muscular layer is very thin. The nervous plexus inside
the circular layer is visually well developed.

Nervous system. — The dorsal ganglion of the brain is divided poste-
riorly into two distinct lobes, of which the smaller, dorsal lobe ends
shortly, while the larger, ventral lobe continues into the cerebral sense
organ. Each of these sense organs is of about the same size as one of
the ventral ganglia, but considerably smaller than either dorsal ganglion.
The duct lying on the external border of each of the pair of cerebral
sense organs communicates, as usual, with the enlarged posterior end of
the cephalic furrow by means of a wide duct opening directly outwards.
The dorsal median nerve, lying just outside the circular muscular layer,
is remarkably conspicuous throughout the whole length of the body.

yo coE

A section through the esophagal region shows that the outer longi-
tudinal muscular layer of the body wall is particularly thick, and that
the esophagus is unusually small. The other layers are as in most
species of the genus.

The lateral blood lacunas break up in the esophagal region into
numerous thin- walled vessels, or spaces, which surround the lateral and
ventral walls of the esophagus.

Nephridia. — The nephridia extend through the anterior ^ of the
esophagal region, as a single profusely branched tubule on each side.
All the branches of the main canal are small, and lie in close connec-
tion with the walls of the blood spaces about the esophagus. The
efferent ducts are very numerous, but minute. They open on the
dorso-lateral aspects of the body, usually not far above the lateral mar-
gins. In a single instance one of the efferent ducts was found to
open beneath the lateral margin, but this must be looked upon as ab-
normal. In several instances two efferent ducts lay close together on
the same side, but in such cases one originated much nearer the ven-
tral side of the esophagus than did the other. The actual number of
efferent ducts counted in one medium-sized individual was 17 on one
side, and 24 on the other.

Reproductive organs. — The pouches containing the sexual products
when mature become so voluminous as to occupy more than half the
entire space within the body walls. In July, when the genital prod-
ucts are fully ripe, the ducts leading to the exterior are completely
formed some little time before the elements are discharged. Each
pouch has a single duct opening into a funnel-shaped depression through
the epithelium on the dorso-lateral aspect of the body. The opening
into the cavity of the ovary is on the surface of a broad papilla formed
of long columnar cells, a portion of which bend inward to guard the
opening. The ducts often have a somewhat tortuous course, and are
likely to broaden out considerably in passing through the circular mus-
cular layer.

Size. — The length of the largest individual observed was about 300
mm., and its width 6 mm.

The species is named in honor of Prof. A. E. Verrill, of Yale
University, who, more than any one else, has helped to bring the
American species of Nemerteans into orderly arrangement.

Stimpson ^ briefly describes from Bering Strait a species of Cere-
bratulus ( C ifnpressus = Alicrura impressa) which bears a superficial
resemblance to the above. Stimpson's species, however, was flattened,

'Proc. Philadelphia Acad. Nat. Sci., p. 160, 1857.

ALASKA NEMERTEANS 7I

of a dusky gray color above, with narrow, colorless, transverse lines,
and with a flesh-colored subtruncate head, much narrower than the body.
The same author describes a somewhat similar species ( C bellus =
Alicrura bella) from Yezo Island, Japan. This had 10 bluish-
white, narrow transverse lines across the ashy-gray dorsal surface, and
a white ventral surface. The head was short, and of a vermilion color.
Both these species should be referred to the genus Micrura as here
defined.

Habitat. — This species was found only at Virgin Bay, in Prince
William Sound. Here it was not uncommon under stones at low
water, and was frequently met with in the parchment like tubes of
Carinella captstrata^ with which it was associated.

26. MICRURA ALASKENSIS sp. nov.
pi. IV, fig. 2 ; p1. XIII, fig. I.

Body long and slender, rounded in esophagal region, flattened
throughout the whole intestinal region. Head remarkably slender,
elongated, and sharply pointed ; cephalic furrows correspondingly
long, but their anterior ends do not reach the proboscis pore. Mouth
small, and well back from tip of snout ; its anterior end reaches about
as far forward as posterior ends of cephalic furrows.

No ocelli are present. The brain is reddish in color, and shows
conspicuously through the tissues external to it.

The esophagal region is well rounded, and narrower than the suc-
ceeding portion of the body. The intestinal region is much flattened
both above and below ; its lateral margins are rounded, however.
After preservation in alcohol the intestinal region is flattened or even
hollowed ventrally, while the dorsal surface is very convex. Posteri-
orly the body tapers gradually, and at its pointed extremity an unusu-
ally long caudal cirrus is present. This is quite colorless, and contracts
after preservation to but a fraction of its original size and length.

Color. — Two color varieties were met with. Most commonly the
general color was a salmon or flesh-color. The esophagal region was
pale salmon with tinges of brighter red, becoming lighter anteriorly ;
the head pale or nearly colorless ; the brain region distinctly red ; the
intestinal region pale salmon, with much more deeply colored intestinal
lobes. Running the whole length of the ventral side of the body —
from near the mouth to the caudal cirrus — is a characteristic, narrow,
cream or flesh-colored stripe in the median line. This stripe is con-
spicuous only in the intestinal region, though it may be traced forward
to the mouth, as stated. In alcoholic specimens it can still be detected

72 COE

in the intestinal region, though the natural color of the body has mostly
disappeared. A similar, though much less conspicuously marked, me-
dian stripe occurs on the dorsal surface.

The second color variety had chestnut brown intestinal lobes, brown-
ish esophagal region, slightly paler below ; snout and margins of head
colorless ; brain red. The paler, ventral, median stripe was even more
conspicuous than in the other variety.

Serial sections show that, while the cephalic furrows are unusually
long, yet they are not really as deep as in many related species. The
brain is very large ; the posterior end of each dorsal ganglion is bi-
lobed, the dorsal lobe ending freely, while the ventral lobe continues
directly into the cerebral sense organ. The canals from the sense
organs open into the very posterior ends of the cephalic furrows. The
buccal, or esophagal, nerves are larger than in most species.

Accessory buccal glands. — On each side of the mouth, and extend-
ing a short distance into the esophagal region, is a series of peculiar
glands — accessory buccal glands, they may be called. These glands
are similar in their nature to, and apparently supplement, the ordinary
buccal glands which line the mouth cavity. In the present instance,
however, these accessoiy glands lie imbedded in the outer longitudinal
muscular layer ventral to the lateral nerves. Here they greatly en-
croach upon the domain of the muscles, and occupy a large portion of
the space between the circular muscular layer and the cutis (p1. xiii,
fig. i). The gland cells are large, and are distended with a clear,
granular secretion. These accessory glands appear to have originated
from ordinary buccal glands which have passed outward and taken up
a position outside the two inner muscular layers (p1. xiii, fig. i).
They are arranged in clusters, and discharge their contents by irregular
ducts leading through the two inner muscular layers to the epithelium
of the buccal cavity and adjacent esophagal wall. A short distance
back of the mouth the two lateral series of these accessory buccal
glands unite beneath the esophagus, and are not found further poste-
riorly. The ordinary buccal glands are present as in related species.

Alimentary canal. — The esophagus is large in comparison with the
thickness of the body wall. The histological difference between the
epithelium lining its anterior portion and that of its posterior half is
much more marked than in most related species. The delicate layer
of circular and longitudinal muscular fibers which surrounds the epi-
thelial lining of the esophagus in most of the Heteronemerteans be-
comes remarkably developed in this species. At the very posterior
end of the esophagus — just anterior to the first intestinal pouches —

ALASKA NEMERTEANS 73

the circular muscles of the esophagus increase so greatly in number
that they form a most conspicuous layer. In the region of its maximum
development this layer becomes nearly half as thick as the circular
layer of the body walls in the same section. In no other species of
the Lineidce has this muscle been found of even approximately this
thickness. Its fibei-s connect in part v^ith the circular layer of the body
walls, and to a lesser degree with the circular muscles of the proboscis
sheath. But few fibers lie on the dorsal wall of the esophagus, so
that this organ is largely bound up with the proboscis sheath in a con-
tinuous layer of muscles, and one cannot fail to see the striking resem-
blance between this circular layer and the inner circular muscles that are
so highly developed in precisely the same region in Carinoma (p. 22).

Bcfdy "Walls. — The external longitudinal muscular layer of the body
walls is especially weak when compared with the same layer in related
species. This is partly shown by the fact that the cutis glands extend
Y^ or more of the distance from the exterior to the circular muscular
layer. Nearer the intestinal region the muscular layers are thicker,
and the esophagus occupies correspondingly less space.

The three longitudinal blood vessels are very conspicuous through-
out their length. The dorsal vessel leaves the proboscis sheath near
the posterior ends of the nephridia, or at about ^ the distance towards
the posterior end of the esophagal region.

Nephridia. — The nephridial system consists of a pair of unusually
• large, longitudinal canals, which lie in the dorsal walls of the lateral
blood lacunae beside the esophagus. The nephridial canals send off
very few branches, except near their anterior ends, where they divide
into smaller branches. The main canals extend through about the
middle third of the esophagal region. Each of the pair of longitudinal
canals terminates posteriorly in a single, remarkably large efferent
duct which opens on the dorso-lateral aspect of the body. The ducts
sometimes lie exactly opposite, but in other specimens one lies some
distance farther back than the other. In such cases, of course, one of
the nephridia extends a corresponding distance farther posteriorly than
the other. Each of the efferent ducts spreads out as it passes through
the circular muscular layer into a broad sieve-like or filter-like struc-
ture with scores of small, nucleated cells. From this point a straight
and comparatively narrow duct leads directly to the surface.

Reproductive glands. — The sexual elements are fully mature in the
month of June, and the genital ducts at this time are fully formed.
In the females each duct communicates directly with the cavity of the
ovary by a funnel-shaped opening, and a similar funnel-shaped open-

74

COE

ing lies at the outer end of the oviduct on the dorso-lateral aspect of
the body. The bodies of many of these worms were so distended with
sexual products that they were very easily ruptured and could conse-
quently be preserved intact only with great care.

Size. — The length of this species is commonly about 150 mm. in
extension, although several individuals were found which were more
than 300 mm. long.

Habitat. — The species was found under stones between tides In
rather muddy localities at New Metlakahtla on Annette Island, at
Glacier Bay, Sitka, Yakutat, and at Orca and Virgin Bay in Prince
William Sound. At the four last-named localities the worms were
abundant. Nearly 50 individuals, some of them nearly 300 mm. long,
w^ere found at Sitka massed together in a single cavity in coarse
gravel mixed with mud. All were filled with fully mature sexual
products, and the bodies of nearly all were in contact in a tangled
mass.

CEREBRATULUS Renier.
Prospetto della Classe dei Vermi (t. Burger), 1804.

The species of this genus are distinguished by long, flattened bodies,
the lateral margins of which are thin, adapted for swimming. Most
species are very active, swim readily with undulatory motion, can roll
up spirally and become twisted, but are only moderately conti'actile in
length, and do not draw together into a tangled mass. The dorso-
ventral musculature is highly developed, as are likewise the longi-
tudinal and oblique muscles.

Head usually pointed anteriorly, but very changeable in shape ;
lateral slits long and deep ; proboscis pore terminal ; mouth large, and
situated behind the ganglia ; esophagal region mostly rounded ; intes-
tinal region broad and flat with thin lateral margins ; the posterior end
extremely flattened and provided with a delicate caudal cirrus, which
extends beyond the opening of the intestine, and in most species is easily
broken off and lost. Eyes usually wanting ; body commonly of a nearly
homogeneous color without distinct markings (such as longitudinal and
circular bands and rings) .

Proboscis sheath reaches to posterior end of body ; proboscis very
long and strong. Intestinal pouches deep and mostly forked peripher-
ally ; central intestinal canal narrow. Neurochord cells probably com-
monly present in brain and lateral nerve cords, though they have as yet
been found in but few species.

ALASKA NEMERTEANS 75

27. CEREBRATULUS HERCULEUS sp. nov.

pl. 1, fig. 5.

Body very large, stout, thick, broad, attaining a length of 2 meters
or more, and a breadth of more than 25 mm. ; remarkably broad when
contracted, the body somewhat resembling that of a leech. One speci-
men, preserved in alcohol and strongly contracted, now measures 300
mm. in length, 25 mm. in width in intestinal region; 18 mm, in width
and 15 mm. in thickness in esophagal region. The length of the
esophagal region in this specimen is only about 55 mm. after preser-
vation.

Head very short, thick, and bluntly pointed. Cephalic slits short,
separated in front, reaching back in the alcoholic specimeii but 7 mm.
Mouth large, situated as far back as the posterior end of the cephalic
furrows.

Esophagal region short and thick ; intestinal region very broad and
flat, with thin, wavy margins in life. Posterior extremity provided
with a pale caudal cirrus of comparatively small size. Proboscis
large, colorless, fully equal in size to that of a large specimen of C.
marginatiis. No ocelli.

Color. — Color very dark brown or reddish brown throughout the
whole dorsal surface, rather paler and less bright ventrally.

This gigantic Nemertean is not clumsy in its movements, but is
active, and swims rapidly and gracefully. It was met with only at
Sitka, where it occurred in considerable numbers in mud at low water
mark, associated with C. jyiarginatiis. The worms are fragile, and
break spontaneously unless carefully handled.

28. CEREBRATULUS MARGINATUS Renier.

Cerebratidus margmatus Renier, Prospetto della Classe del Vermi (t. Burger),

1804.
C. angidatus McIntosh, British Annelids ; Pt. i, Nemerteans, p. 195, Ray

Society, 1872-73.
C _/"«^«^^ Verrill, Trans. Connecticut Acad., viii, p. 438, 1892.

This large and active Nemertean was found in several localities
about Sitka, but not in great numbers. The individuals there found
possess the typical color of the species, with which they agree exter-
nally in every detail except the shape of the body. All the Alaska
specimens are much shorter and comparatively broader than those I
have seen in Naples, although careful study of microscopic sections
revealed no essential anatomical differences. It is therefore probable
that this is but a local peculiarity of a single widely distributed species.

76 coE

The worms were found near low water mark living in soft black
mud beneath a considerable growth of 'eel grass.' The mud contained
a great quantity of decomposed vegetable matter, and was saturated
with sulphuretted hydrogen. The worms are excellent swimmers, and
are very rapid in their movements in their undei'ground burrows.

The specimens obtained measured up to 500 mm. or more in length,
and about 15 mm. in width.

The color above was slaty-brown, dark gray, or greenish-gray, with
white or colorless margins.

The species has been previously recorded from the Mediterranean,
the coasts of Great Britain, Madeira, the northeastern coast of Amer-
ica, Greenland, and from other localities.

29. CEREBRATULUS OCCIDENTALIS sp. nov.
rl. VI, fig. 3.

Body 300 mm. or more in length in moderate extension ; rather slen-
der, rounded in front, flattened behind, and with very thin margins in
the intestinal region, as in other species of the genus. Head change-
able in shape, either rounded or acutely pointed, according to state of
contraction. Cephalic furrows rather short.

Color. — Individuals vary considerably in the general color of body,
but are most commonly chestnut brown or dull reddish anteriorly, and
light chocolate brown in the intestinal region. The ventral surface is
brownish flesh-colored, with a median, ochre ventral stripe. Some-
times the ventral surface is nearly chocolate throughout its entire length.
A darker median dorsal stripe is indicated.

Proboscis. — Remarkably small and slender, being many times
smaller than in most related species of equal size. It is colorless, and
is usually everted when the animal is killed.

In internal anatomy but few points need special mention. The
cephalic glands are unusually well developed, and continue backward
as the cutis glands. The brain is remarkably voluminous, the dorsal
ganglion bilobed, the lower lobe continuing into the cerebral sense
organ.

The mouth is very large, and is provided with an unusual abundance
of buccal glands. There are many diagonal fibers between the pro-
boscis sheath and the body musculature.

NepJiridia. — The nephridia are of moderate length, and occupy
the middle third of the esophagal region. The main tubules lie in the
angles between the esophagus and the proboscis sheath. They are
profusely branched towards their anterior ends, each branch lying in

ALASKA NEMERTEANS 77

the wall of one of the esophagal blood lacunee. Farther back the
branches unite into a single tubule on each side. Their main canal
continues backward for a considerable distance in the wall of a lateral
blood lacuna, and is without branches. At its posterior extremity the
efferent duct bends to the dorso-lateral surface of the body, as usual.

The sexual products are fully mature in July.

Habitat. — This is a very active species. It was found in abundance
at Wrangell, Yakutat, Orca, and Virgin Bay. It inhabits the shore at
half tide and below, in muddy places and under stones. The small
size of the proboscis will distinguish it from related species.

30. CEREBRATULUS LONGICEPS sp. nov.

p1. V, figs. 4, 5, 6, 7 .

Body much flattened throughout its whole length ; anterior portion
remarkably narrow and slender, becoming wider posteriorly. Head
much narrower, more slender, and longer than in most species of the
genus, acutely pointed in front, much flattened dorso-ventrally, and with
the tip of the snout often slightly curved upward. A section through
the head is often concave both above and below, showing that the head
is thicker laterally than in the median line. The head is directly con-
tinuous with succeeding portions of the body. Frontal sense organs
very highly developed.

Cephalic furrows very long, and remarkably deep and wide. An-
teriorly they do not reach quite to tip of snout. They are, conse-
quently, well separated from the proboscis pore, which is situated sub-
terminally. The mouth is situated as far back as the posterior end of
the cephalic furrows.

From the narrow, pointed head the esophagal region gradually
widens as it passes backward, but it remains unusually flat throughout ;
the intestinal region is not sharply marked off from the esophagal
region, but widens gradually toward the posterior third of the body,
and narrows toward the posterior extremity. The intestinal region
is even flatter than the more anterior portion of the body. The caudal
cirrus was not observed.

Proboscis. — The proboscis sheath is reduced to a frail tubule in the
posterior third of the intestinal region. The proboscis is slender and
colorless ; it has the three muscular layers and the muscular crosses,
as usual in the genus.

Color. — Dorsal surface dark brownish-black or purplish, much paler
on tip of snout, and on borders of cephalic furrows. This paler border
is wider on the ventral than on the dorsal borders of the furrows. It

78 coE

is sometimes so much increased that it covers the whole ventral surface
in the region of the mouth, and gradually becomes darker tow^ard the
tip of the snout, as well as posteriorly. The color of the ventral surface
is similar to that of the dorsal surface, but is commonly slightly paler,
especially anteriorly, and has a grayish tinge.

No eyes were found.

Size. — Length of largest specimens collected about 300 mm. ; width
in intestinal region 6 mm.

The body is fragile, and the specimens are often broken in killing.
After preservation in formalin or alcohol the body tapers gradually
toward the narrow head, is widest in the posterior third or near the
posterior end, and greatly flattened throughout. The head retains its
long, pointed appearance, the tip is recurved, and the cephalic furrows
in most specimens are deep and widely open.

Frontal sense organs. — On the exact tip of the head are three
rather deep and wide pits, easily overlooked in the entire animal, but
very conspicuous in sections. These are undoubtedly sensory in their
nature, and are lined with slender rod-like cells with especially large
cilia. Burger has described in detail such sense organs in Micrura
and Cerebratuhis, and somewhat similar ones in other genera. These
' frontal organs,' as they are called, lie above the proboscis pore ; one
of them is situated in the median line and the other two are placed
symmetrically on the antero-lateral margins.

The brain is remarkably large, and the dorsal ganglia are much
larger than the ventral. The posterior ends of the long lateral furrows
are greatly expanded.

Cerebral sense organs. — These also are very voluminous and are
continuous with the posterior ends of the dorsal ganglia. The canals
leading to the exterior are large, and open into the posterior ends of
the cephalic furrows. Internally each canal passes directly beneath
the dorsal ganglion to its internal ventral border, and then bends dorso-
laterally to the external border of the sense organ. These canals leave
the cephalic furrows far back of the cerebral commissure, and at about
the point where the ventral ganglia merge into the lateral cords.

Nepkridia. — The nephridial system presents remarkable deviations
from the type usually found in the genus. The nephridial canals ex-
tend throughout the whole length of the esophagal region, and com-
municate with the exterior by upwards of sixty efferent ducts on each
side. In the region of the mouth, or directly behind it, are one or two
nephridial tubules on each side, quite independent of the rest of the
system. Each of these tubules consists of a coil of fine canals lying

ALASKA NEMERTEANS 79

on the dorsal side of the lateral blood lacunae, and projecting freely
into the cavity of the lacuna. From each coiled tubule an efferent
duct leads directly to the dorsal surface of the body. Back of these
anterior, isolated nephridia a continuous richly branched and much
coiled canal extends posteriorly on each side as far as the end of the
esophagal region. The branches of these canals are limited to the
dorsal and lateral surfaces of the lateral blood spaces, which occupy
the angles between the esophagus and the proboscis sheath. The
coiled tubules project freely into the blood spaces, and encroach con-
siderably upon their area. The very numerous efferent ducts, which
are given off from these tubules, pass directly outwards to the surface
of the body. Those which pass out from the most lateral of the ne-
phridial tubules, open on the dorsal surface near the lateral margins of
the body, while those which leave the tubules nearest to the proboscis
sheath open on the dorsal surface not far from the median line. The
greater portion of the nephridiopores, however, lie rather nearer to
the lateral margins than to the median dorsal line. Although there are
practically the same number of efferent ducts on each side, this does
not signify that these ducts are paired, but rather that the average
number on each side is identical. In several instances an unusually
large efferent duct on one side was accompanied by one of similar
size nearly opposite. These ducts vary greatly in size, some being
several times as large as others. Great irregularity likewise exists as
to their distribution along the course of the nephridial canal. In the
anterior third of the esophagal region they are well separated, but
more posteriorly, portions of two or three sometimes appear on the
same side in a single section.

Sexual glands. — The reproductive glands appear immediately be-
hind the nephridia, and in both sexes alternate with the intestinal
pouches. The sexual products were fully mature in July, and in both
sexes the genital ducts were fully formed. They were lined with a
special flattened epithelium, and opened on the dorso-lateral aspects of
the body in both sexes. The ova when immature are attached by a
narrow stalk to the wall of the ovary, and later break off and fall into
the ovarian cavity. The spermatozoa have short, oval heads instead
of the slender, pointed ones so common in the genus.

Habitat. — Yakutat ; under stones at low water ; not common.

8o COE

31. CEREBRATULUS MONTGOMERYI sp. nov.

p1. VI, figs. I and 2.

Body large, very long and ribbon-like when fully grown, and much
flattened, except in esophagal region. Head variable in shape, ac-
cording to state of contraction ; sometimes rounded and obtuse, at
other times much elongated, pear-shaped, and flattened dorso-ventrally.
The anterior portion of the body, back as far as the intestinal region,
is narrower and thicker than the posterior portions, and has rounded
margins ; in the intestinal region the body is very flat, and the lateral
margins are extended to very thin edges. In partial contraction the
dorso-ventral muscles, situated a little distance from the oviter edge of
the margins, contract so strongly as to produce a lateral ridge on each
side both above and below the lateral edge. A transverse section con-
sequently shows that the lateral margin is thicker distally than it is
nearer the median line. These lateral ridges are often seen in speci-
mens preserved in alcohol or formalin. Posteriorly the body becomes
still thinner, and is narrower toward the extremity. The posterior end
is either obtuse or sharply pointed, according to the state of contrac-
tion. The posterior opening of the intestine is subterminal. In none
of the many specimens obtained was the caudal cirrus present, though
such an appendage is probably present in uninjured worms.

The cephalic slits are moderately elongated, and extend backward
as far as the anterior end of the mouth. Their posterior ends are
wide and deep, and in life their margins are thin and widely separated.
Anteriorly they do not reach quite to the proboscis pore, except when
the animal is strongly contracted.

Color. — In life the whole body, both above and below, except the
tip of the head, is bright blood red. The anterior extremity is tipped
with a narrow band of white or yellowish-white. This white tip ex-
tends both above and below, and is not usually more than a few milli-
meters wide. The proboscis is light red in color.

Size. — Individuals were most commonly from one to two meters in
length, and 8 to 10 mm. in width in the intestinal region. A few of
the specimens obtained measured as much as two and a half meters
when extended.

The individuals are generally hardy, and are not so prone to break
spontaneously into fragments as are those of many other species of the
genus. They contract much less violently than most other Nemerteans
when killed, and may consequently be easily preserved entire by kill-
ing in a dilute solution of formalin in sea water.

ALASKA NEMERTEANS 8l

Proboscis. — The proboscis is pale red, and of moderate size for the
genus. Its musculature consists of an inner circular, a longitudinal,
and a very thin outer circular muscular layer. It is, therefore, differ-
ent from those of nearly all the species of the family Lineidce.^ in that
the inner longitudinal muscular layer is wanting. There are fibers
passing between the inner circular layer and the thin outer circular
layer, but these fibers do not form such distinct crosses as are char-
acteristic of most of the members of the family. The internal epi-
thelium is thick and highly glandular. The nerve plexus beneath this
layer is unusually conspicuous, and lies directly internal to the circular
muscular layer. The proboscis is very often retained in place after the
animal has been killed. The muscular layers of the proboscis sheath
are rather thick, and the circular muscular fibers often cross into the
circular muscles of the body in the median line.

In the anterior portion of the head a rich growth of glands pene-
trates the other tissue nearly to the central proboscis sheath. These
glands mostly open directly outward on the whole circumference of
the head, although a few open on the tip of the snout. Back of the
lateral slits the glands are restricted to the cutis. In the region of the
mouth the cutis glands are separated from the outer epithelium hy a
layer of interlaced fibers of connective tissue, forming a basement layer
of double the thickness of the epithelium. In the intestinal region the
cutis glands are much reduced, and scattered. The outer muscular
layer of the body is about as thick as the two inner muscular layers
combined.

The mouth is much elongated, and its anterior end reaches forward
as far as the posterior ends of the cephalic slits. The intestinal pouches
are deep and narrow, and are forked distally.

Nephridia. — The nephridial canals are profusely branched, and lie
in contact with the blood spaces around the esophagus in front of the
middle esophagal region. Their extent longitudinally, however, is
short. The main nephridial canals lie in the angles between the
esophagus and the proboscis. Near their posterior extremities a large
efferent duct passes obliquely upward on each side and opens on the
latero-dorsal aspect of the body. As described by Biirger^ in C.
marglnatus., one of the efferent ducts often lies far behind the other.
Posterior to the efferent ducts the nephridial canals are smaller, and
their branches end a short distance farther back. The blood lacunze
in the head in the specimens sectioned were very much reduced in size.

'Fauna u. Flora Neapel, Monogr. 22, p. 622.
Proc. Wash. Acad. Sci., March 1901.

82 COE

The brain lobes are large, and the ventral commissure enormously
thick. The well developed cerebral sense organs abut closely against
the posterior ends of the dorsal ganglia. Their ciliated canals pass
obliquely inward from the posterior ends of the cephalic slits. In their
course these canals pass external and ventral to the dorsal ganglia. The
median dorsal nerve, lying just outside the circular muscular layer of
the body, is large and conspicuous throughout nearly the whole length
of tlie animal.

Ocelli are wanting in the adult worm.

In specimens collected in June and July the genital products had
evidently been recently discharged.

This species is named in honor of Dr. T. H. Montgomery, Jr., of
the University of Pennsylvania, whose studies on the Nemerteans form
most valuable contributions to the knowledge of the group.

Habitat. — This magnificent Nemertean was found abundantly under
stones in muddy places near low water mark at most of the collecting
stations between Sitka and Unalaska. At Orca and Virgin Bay, in
Prince William Sound, and at Dutch Harbor, Unalaska, it proved to
be one of the commonest species. Professor Kincaid has sent me a
large specimen from Puget Sound, in the State of Washington, where
• it is said to be not uncommon. This would indicate that the species
may be found locally along the whole northwestern coast of North
America south of Bering Sea.

32. CEREBRATULUS ALBIFRONS sp. nov.
p1. IV, figs. 3, 4.

Body elongated, ribbon-like, flattened behind, rounded in front as in
typical species of the genus. Margins of intestinal region pale and
thin. Cephalic slits unusually long and deep, reaching well beyond
posterior end of mouth.

Color. — General color of body brownish purple. Anterior end of
head white, including both dorsal and ventral surfaces, and extending
backward about as far as anterior end of mouth, and sometimes reach-
ing along borders of cephalic slits to their ends. The esophagal re-
gion is dark smoky purple on dorsal surface ; the^ ventral surface is
similar but paler ; the intestinal region inclines more to reddish. A
. darker line runs along the middle of the dorsal surface and the in-
testinal lobes appear more opaque.

Habitat. — Only a single specimen of this pretty species was se-
cured, and this, unfortunately, had its posterior extremity missing.
The specimen was about 150 mm. in length and of moderate proper-

ALASKA NEMERTEANS

83

tions. It was restless in confinement, and an active swimmer. It was
found under a stone near low water mark at Hot Springs, near Sitka.

EXPLANATION OF PLATES.

The colored plates (I to VI) are reduced from colored sketches
made, with a few exceptions, from the living animal. They were com-
pleted after the return of the expedition and have been retouched, and
in some instances entirely redrawn, by Mr. A. H. Verrill. Every
effort has been made to make the colors and the form of the body as
natural as possible.

In the figures on the plates the

ap, attachment of proboscis.

bg, buccal glands.

bg' , accessory buccal glands.

bl, blood lacuna.

ble, epithelium of same.

bm, basement layer.

br, brain.

bv, blood vessel.

bva, anastomosis of lateral v^essels.

cc, ciliated canal of cerebral sense or-
gan.

cgl, cephalic glands.

c;w, circular muscular layer.

cso, cerebral sense organ.

ct, connective tissue.

cugl, cutis glands.

dc, dorsal commissure of brain.

dg, dorsal ganglion.

dgl, secretion of cephalic glands.

dn^ median dorsal nerve.

e, esophagus.

eep^ epithelium of same.

f, fibrous layer of cutis.

gc, glandular cells of cerebral sense
organ.

gl, integumental glands.

gp, genital pouch.

/, integument.

ic, intestinal caecum.

ictn, inner circular muscular layer.

following reference letters are used

ijl, inner fibrous layer.
ilm. inner longitudinal muscles.
in, intestine.

Im, longitudinal muscles.
/», lateral nerve-cord.
m, mouth.

mef, epithelium of same.
«c, nerve commissure.
nd, efferent nephridial duct.
nep, nephridial canal.
np, nerve plexus.
nv, nerve.
oc, ocellus.

ocm, outer circular muscles.
oep, outer epithelium of proboscis.
olm, outer longitudinal muscles.
otf, ova.

pcm, circular muscles of proboscis.
plm, longitudinal muscles of proboscis.
pn, proboscis nerve.
ps, proboscis.

psk, proboscis sheath or its cavity.
rk, rhynchodaeum.
ro, opening of same.
svig, submuscular glands.
sn, nerve to cerebral sense organ.
sop, opening of ciliated canal of sense

organ.
vc, ventral commissure of brain.
vg, ventral ganglion.

84

COE

INDEX TO GENERA AND SPECIES.
New genera and species are in black face type ; synonyms are in italics ; pages
on which generic or specific decriptions occur are in black face type.

Amphiporus 40, 9, 28, 32, 33, 34

angulatus 41, 3, 6, 9, 36

bimaculatus 44, 6, 9, 31

exilis 54, 6, 9, 51, 52, 53

lactifloreus 51, 54

leuciodus 5". 6, 9, 31, 55

nebulosus 48, 6, 9

neesii 4

tigrinus 46, 6, 9

virescens 28
Borlasia camillea 4

quinqiielineata 4
Carinella 11, 7

annulata, 13, 17

capistrata 16, 5, S, 71

dinema 15, 5, 8, 17

nothus 17

polymorpha 12

speciosa 11, 5, 7

superba 15, 16, 17
Carinoma 20, 8, 73

armandi 20, 21, 22

griffini 20, 5, 8

patagonica 21, 22, 23

tremaphoros 20
Cephalothrix 19, 8

linearis 19, 5, 8
Cerebratulus 74, 10, 68

albifrons 82, 6, 11

angulatus 75

bellus 71

fiiscus 75

herculeus 75, 6, 10

imprcssus 70, 3

longiceps 77, 6, 11

marginatus 75, 6, 10

montgomeryi 80, 6, 1 1

occidentalis 76, 6, 10
Cosmoccphala heringiana 41, 3, 42
Emplectonema 23, 4, 8, 28, 32, 54

biirgeri 25, 5, 8

camillea 4
gracile 23, 3, 5, 8

Emplectonema viride 23, 3, 4, 25
Eunemertes 23, 3

gracilis 23

tieesii 4
Eu folia 61, 3, 4
Fasciola angulata 41
Lineus 64, 10, 68

torquatus 66, 6, 10

viridis 65, 6, 10
Micrura 68, lo

alaskensis 71, 6, 10

bella 71

impressa 70, 3, 10

verrilli 68, 6, 10
Nareda superba 41, 43
Nemertes 23

gracilis 23, 3, 4

obscura 65
Omatoplea stimpsonii a^Xi ^2
Paranemeites 32, 8

carnea 37, 6, 9, 32

pallida 36, 5, S, 32

peregrina33, 5, 8, 32
Planar ia gcssere/isis 65

linearis 19

viridis 65
PoZ/Vz 61
Tjeniosoma 61, 4, 10

CBquale 4

princeps 62, 6, 10

quinquelineatum 4

septemlineatum 4
Tetrastemma 57, 9

aberrans 58, 6, 9, 57

bicolor 57, 6, 9

caecum 59, 6, 9

cruciatum 58

rubrum 58

vermiculus 58
Zygonemertes 28, 9

albida 31, 5, 9, 29

thalassina 29, 5, 9
virescens 29, 30

PLATE I.

Fig. I. Carinella cafistrata sp. nov. A large individual" about natural size.
Virgin Bay, Prince William Sound.

2. Carinella dinema sp. nov. Victoria, B. C. Enlarged 1% times.

3. Head of same species. Side view. Enlarged 4 times.

4. Amphiporus bimaculatus sp. nov. Dorsal view of head. Glacier Bay.

Enlarged 12 times.

5. Cerebratulus kerculeus sp. nov. A large individual. Sitka. Half nat-

ural size.

6. Tetrastemma bicolor sp. nov. Kadiak. Three times natural size.

(86)

Proc Wash Acad. Sci., Vol.11

Plate

ALASKA NEMERTEANS.

PLATE II.

Fig. I. Emplectonema burgeri si^. nov. Anterior portion of an individual of
the pale variety. Glacier Bay. J^ natural size.

2. E. burgeri. A large individual of the dark variety. Glacier Bay. %

natural size.

3. Tceniosoma princeps sp. nov. Large individual. Yakutat. % natural

size.

4. T. princeps. A contracted specimen in formalin. Cape Fox. Nat-

ural size.

5. Zygonemertes thalassina sp. nov. Sitka. Twice natural size.

6. Paranctnertes peregrina sp. nov. Brown variety. Victoria, B. C. A

small specimen. Natural size. Compare pi. iii, fig. 5.

(88)

Proc. Wash. Acad. Scl, Vol.111.

Plate

ALASKA NEMERTEANS.

PLATE III.

Fig. I. Ampkiporus exilis s,'^. nov. Glacier Bay. Slightly enlarged.

2. Zyg07iemertes albida sp. nov. Victoria, B. C. Enlarged \%, times.

3. Paranemertes carnea sp. nov. Taku Harbor. Anterior portion of

body with emarginate head. Natural size.

4. P. carnea. Sitka. % natural size.

5. Paranemertes peregrina sp. nov. Virgin Bay, Prince William Sound.

% natural size.

6. Carinella speciosa sp. nov. Hot Springs, near Sitka. ^ natural size.

(90)

Proc. Wash Acad. Scl, Vol.111

Plate hi

2

11

V X

'■-. ''W

\ K^

^_- 'i-^

1

ALASKA NEMERTELANS.

PLATE IV.

Fig. I. Amphiporus nebulosus s'^.nov. Kukak Bay, Alaska Peninsula. Natural
size.

2. Micrura alaskensis sp. nov. Sitka. Enlarged ij4 times.

3. Cerebratulus albifrons sp. nov. Near Sitka. Twice natural size.

4. Side view of head of same individual.

5. Amphiporiis tigrinus sp. nov. Farragut Bay. Mature male specimen,

slightly enlarged.

6. A. tigrinus. Head of male ; dorsal view. Twice natural size.

7. A. tigrinus. Farragut Bay. Female with ripe ova. Slightly en-

larged.

8. Head of female of same species. Dorsal view. Twice natural size.

(92)

Proc. Wash Acad. Sci., Vol.111.

Plate iv.

ALASKA NEMERTEANS.

PLATE V.

Fig. I. Micrura verrilli sp. nov. Virgin Bay, Prince William Sound. Natu-
ral size.

2. Side view of head of same species.

3. The same ; contracted.

4. Cerebratulus longiceps %-^. v\.ov. Yakutat. Enlarged i^ times.

5. 6, 7. Heads of same species, from lateral, dorsal, and ventral aspects re-

spectively.

8. Lineus torquatus sp. nov. Orca, Prince William Sound. Natural size.

9. L. torquatus. Side view of head.

ID. Amphiporus bimaculatus sp. nov. Victoria, B. C Natural size.

(94)

Proc, Wash Acad. Sci., Vol, II

Plate v.

ALASKA NEMERTEANS.

PLATE VI.

Fig. I. Cerebratulus monigomeryi^^. no\. Dutch Harbor, Unalaska. Natural
size.

2. Head of same species from ventral surface.

3. Cerebratnlus occidentalis sp. nov. Yakutat. Natural size.

4. Amphiforus angulatus [Y2ihx.)\Qrr\\\. Kadiak. Natural size of large
specimen.

(96)

Proc Wash. Acad. Sci , Vol

Plate vi

BMei5el!i»Bosl0P

ALASKA NEMERTEANS.

PLATE VII .

Fig. I. Zygonetnertes thalassina. Outline of stylet apparatus of proboscis
( X 50)- i^) i^> several accessory stylets from two individuals (X 225).
Sitka.

2. Amfhifonis angulatus. Central portion of proboscis (X40). Taku
Harbor, za, central stylet and basis of another specimen (X So).

3. Paranemertes pallida. Outline of middle region of proboscis ( X^o).

4. P. carnea. Extremity of everted proboscis, with six pouches of acces-
sory stylets, of which but four are shown ( X 35)' Sitka.

5. Atnphiporus exilis. Middle portion of proboscis, indicating, besides
central stylet and basis, eight pouches of accessory stylets (X4o)-
Yakutat.

6. Amphiporus leuciodus. Stylet apparatus of proboscis showing the three
accessory stylet pouches ( X^o)- Victoria, B. C.

7. Paranetnertes pcregrina. Outline of middle portion of proboscis
(Xso)- Virgin Bay.

(9S)

Proc. Wash. Acad. Sci., Vol.111

Plate vil

AL/\SKA NEMERTEANS.

PLATE VIII.

Fig. I. Emplectonema biirgeri. Outline of stylet apparatus (X35)- Glacier
Bay.

2. Ainphtporus bimaculatus. Middle portion of proboscis, with four
pouches of accessory stylets ( X 35 )• Glacier Bay.

3. Emplectoitema gracile. Stylet apparatus of proboscis (Xso)- Popof
Island.

4. Amphiporus tigrinus. Middle region of proboscis (X^o). Farragut
Bay. The deeply stained wreath of glands is indicated only in section.

5. Zygonemertes albida. Middle portion of proboscis ( X 40). Victoria,
B. C.

6. Atnphiporus nebulosus. Outline of stylet apparatus of proboscis (X 5o)-
Kukak Bay.

7. Paranemertes carnea. Side view of head, with partially everted probos-
cis. The position of the lateral oblique furrows, and the arrangement
of the ocelli are indicated ( X S)- Taku Harbor.

(100)

Proc. Wash Acad. Sci., Vol 11

Plate vi

ALASKA NEMERTEANS.

PLATE IX.

Fig. I. Carinella sfeciosa. Transverse section through head in front of brain
showing, particularly, position of enormously developed cephalic glands
(^cgl') and arrangement of cephalic nerves («f ). The cephalic glands
are thickly placed both above and below the cephalic blood lacunae (3/),
as well as around the rhynchodaeum (^A). The secretion from many of
these glands is discharged near the lateral margins as indicated {dgl).
Between the longitudinal and oblique muscles (/?«), which lie in the
deeper parts of the head, and the circular fibrous layer {cm), composed of
muscles and connective tissue fibers underlying the integument («"), are
numerous and very massive nerves {nv) which supply the head regions
in front of the brain. The broad indentation on the ventral margin has
no relation to the mouth, which lies much farther back. Other reference
letters are explained above (p. 83). A small individual. ( X45)-

2. Carinella sfeciosa. Transverse, but somewhat oblique, section through
head in region of brain. On the right of the section the brain {dg and 7;^)
lies directly beneath the circular fibers {cni). External to the brain, and
lying among the bases of the integumental cells, is the highly specialized
cerebral sense organ (c5o), connected with the dorsal ganglion {dg) by
several small nerves (5«). The cerebral sense organ shows a conspicu-
ous central canal. The rhynchodaeum {rh) has lost its fringe of gland
cells, these being limited to the regions anterior to it. The outer por-
tion of the circular layer (cw) constitutes the basement membrane of the
integument. In the region of the brain the circular layer splits into two
sheets, one passing external to the brain and the other (//?) internal.
Other reference letters as above. A large specimen. ( X25).

3. Carinella speciosa. Transverse section through nephridial region show-
ing positions, and one of the openings, of the nephridial canals. The
lateral blood lacuna ((5f ) and the nephridial canal {nep) lie embedded in
a band of loose connective tissue on each side. The opening of one of
the efferent nephridial ducts {nd) is seen on the right of the drawing.
Other reference letters as indicated above. (X25).

(102)

Proc. Wash Acad. Sci , Vol. Ml

Plate ix.

//!iy/ . _

//A( -

I 'I, 111 ilL^^

ALASKA NEMERTEANS.

PLATE X.

Fig. I. Carinella spectosa. Portion of transverse section through nephridial
region. The ciliated nephridial canal {nep) shows the peculiar in-
folding of the integumental cells on its dorsal border. These integu-
mental cells are here loosely arranged, and show several deep infoldings
{op) which, under certain circumstances, may possibly have a more or less
distinct communication with the nephridial canal. But a small portion
of the closely packed gland cells in the integument are indicated. Ref-
erence letters are explained on p. S3. (X 75)-

2. Carinella spectosa. Portion of a transverse section through the brain
region. The relation of the dorsal ganglion {dg) with the highly spec-
ialized cerebral sense organ [cso) is indicated. The ciliated canal {cc)
of the sense organ connects directly with the exterior by a narrow tube
{sop) — shown in dotted lines in the drawing, because it lies mainly in
another section — opening on the lateral margin of the head. Several
nerves {sn) are seen to pass from the dorsal ganglion to the sense organ.
Other reference letters as above. Only a small portion of the integu-
mental gland cells are shown. (X 75)-

3. Amphiporics tigriiitcs. Dorsal view of anterior portion of body, cleared
in cedar oil. In front of the brain the arrangement of the ocelli is
shown, and farther back the position of the ovaries {ov). The intesti-
nal cceca are not indicated. Farragut Bay. ( X S).

4. AmpJiiporus tigrinus. Median sagittal section through the anterior por-
tion of the body. The cephalic glands ( cgl ) lie above the opening of the
rhyncodaeum {ro). The mouth (■»/) separates from the proboscis
opening a little way back. The attachment («/) of the proboscis to the
tissues of the head is seen to be well in front of the brain commissures
(<fc and vc). The section shows the comparative size and arrangement of
the proboscis, blood vessels, esophagus, and other organs. Reference
letters as above. ( X 30) •

(104)

Proc. Wash Acad. Scl, Vol.

Plate x.

Coe del

ALASKA NEMERTEAN5.

BMeBcliilhiosta

PLATE XI.

Fig. I. Amphiporus nebulosus. Transverse section through posterior portion of
ventral commissure of brain. The submuscular glands i^smg) fill up
a large portion of the tissues of the ventral half of the head. The
pair of nerves leading forward to the cerebral sense organs is seen to
originate {son) from the ventral side of the dorsal ganglia. Other
reference letters are explained on p. 83. (X 26).

2. Amphiporus angulatus. Transverse section through ventral commis-
sure of brain. The dorsal attachment of the proboscis to the tissues of
the head is shown. The proboscis nerves {pn) enter the ventral side of
the proboscis, and divide into a definite number of branches (usually
18), which farther back arrange themselves symmetrically about the
periphery. The roots of the dorsal brain commissure {rdc) are indi-
cated. In the right half of the section the anterior ends of the nephri-
dial tubules {»ep) are seen. Submuscular glands {smg) are not as
numerous as in ^. nebulosus. Other reference letters as above. (X 26).

3. Amphiporus exilis. Transverse section through nephridial region to show
efferent nephridial ducts {nep) opening on dorso-lateral aspects of body.
This condition is extremely rare in INIetanemerteans. Several diverti-
cula of the intestinal caecum {inc) are represented. Other reference
letters as above. (X32).

(106)

Proc Wash Acad, Sci , Vol.111

Plate xi.

ALASKA NEMERTEANS.

PLATE XII.

Fig. I. Paranemeries ^pallida. Transverse section through nephridial region.
In this case the single pair of efferent nephridial ducts («c^) was so
sj'mmetricallv placed that both were cut in a single section. The very
small size of the proboscis {ps) in the large proboscis sheath (ps/i) is
remarkable. The intestinal csecum (/c) shows lateral diverticula above
the lateral nerve cords. (X iS).

2. AinpJiiporus bimaculatus. Transverse section of proboscis. The i6
proboscis nerves i^pn) are very sharply defined. The cylindrical plexus
(«^) of nerve fibers and connective tissue serves to connect the nerves,
and divides the thick longitudinal muscular band into an inner {plm)
and an outer [plm') layer. The inner longitudinal muscular layer,
found in most related species, is wanting — the basement layer {bju) of
the internal epithelium {p^p) lying immediately beneath the circular
muscular layer. ( X 66) .

3. Emplecto7iema biirgeri. Transverse section through body, showing
manner in which the esophagus (e) opens into dorsal wall of intestine
(/■«). This section also shows the genital pouches {gp) lying both
above and below the intestinal lobes, and indicating that they open
respectively on the dorsal and ventral surfaces of the body. Of the
submuscular glands (smg), which extend throughout the esophagal re-
gion in great abundance but a few remain as far back as the position of
the section figured. (X20),

(108)

Proc. Wash. Acad. Sci , Vol.

Plate X!I

IMfflMiMSE^^^"^ Vm. --In.

JBMcisellllhBosron.

ALASKA NEMERTEANS.

PLATE XIII.

Fig. 1. Micrura alaskensis. Transverse section through mouth region. Tlie
peculiar accessory buccal glands {^bg') are seen to lie outside the cir-
cular muscular layer (c;«), and amongst the fibers of the outer longi-
tudinal muscular layer {ohn). Their secretion passes through the layer of
circular muscles, and mixes with that of the true buccal glands {bg),
Avhich lie immediately beneath the epithelium of the mouth {mep). The
buccal nerves {bn) lie on the lateral borders of the mouth as usual.
Other reference letters are explained on p. S3. (X 30).

2. Zygonemertes thalassina. Transverse section immediately back of the
brain. The intestinal caeca (/c) reach forward to abut against the dor-
sal ganglia. The ocelli (oc) extend back of the brain, and occupy posi-
tions, as shown, immediately above and external to the lateral nerve
cords (/«). (X 100).

3. Amfhiforus angulatus. Transverse section through posterior end of
body. The drawing is slightly diagrammatic, as it contains portions of
the two adjacent sections. The anastomosis of the three longitudinal
blood vessels {bva^ is seen to lie immediately dorsal to the commissure
(wc) of the lateral nerves (/«) ; the basement layer (3w) is remarkably
thick in this region. The posterior end of the intestine (t«) opens
ventrally a few sections farther back. Other reference letters are ex-
plained above. (X80).

(no)

Proc Wash^ Acad. Sci., Vol.

Plate xil

ALASKA NEMERTEANS

PROCEEDINGS

OF THE

WASHINGTON ACADEMY OF SCIENCES

Vol. Ill, pp. 111-13S. March 26, 1901

MAMMALS COLLECTED BY DR. W. L. ABBOTT

ON THE NATUNA ISLANDS.

By Gerrit S. Miller, Jr.

About three months during the spring and summer of 1900
were spent by Dr. W. L. Abbott in exploring the Natuna
Islands in the South China Sea.^ Specimens were collected at
the following localities : Pulo Midei, or Low Island (May 23-
26), Pulo Seraia (May 29), Sirhassen Island (June i-io), Pulo
Subi (June 12-13), Pulo Lingung (June 17-19), Bunguran, or
Great Natuna Island (June 24-July 31) and Pulo Laut, or
North Natuna Island (August 5-13). About 265 mammals
were obtained, all of which have been presented to the United
States National Museum. This paper contains an account of
these, and is published here by permission of the Secretary of
the Smithsonian Institution.

Two extensive collections of mammals had been made on
the Natuna Islands previous to Dr. Abbott's visit, the first by
Mr. A. Everett during September and October, 1893, the sec-
ond by Mr. Ernest Hose during July, August, September and
October, 1894. These have formed, either wholly or in part, the
basis of several papers,^ which constitute the literature relating to

^ For location of the Natuna Islands see Proc. Washington Acad. Sci., 11, p.
204. August 20, 1900.

^Thomas (O.) and Hartert (E.). List of the first collection of mammals from
the Natuna Islands. Novitates Zoologicae, i, pp. 652-660. September, 1S94.

Thomas (O.). Revised determinations of three of the Natuna rodents.
Novitates Zoologies, 11, pp. 26-28. Februarj', 1895.

Proc. Wash. Acad. Sci., March, 1901. (m)

112 MILLER

the mammals of the islands.^ Twenty-eight land mammals have
been recorded as actually represented by specimens, though
several others are mentioned which the collectors ascertained to
occur. Dr. Abbott secured forty-four species, but failed to ob-
tain seven ^ of those previously taken. The total number of
mammals collected on the islands thus becomes fifty-one. This
increase is due, in part to the recognition of a larger number of
insular forms than has been admitted by previous writers, but
also to a considerable extent to the actual addition of species
not hitherto taken. Species new in the latter sense are distin-
guished in the present paper by absence of reference to pre-
vious records.

In regard to the faunal relationships of the Natunas, whether
predominantly Bornean or Peninsular, about which much has
been written,^ it may be said that this collection, together with
much of the other work recently done by Dr. Abbott, tends to
show that there is greater general uniformity in the mammalian
fauna of Borneo, the Malay Peninsula, and the intervening
islands than has been hitherto supposed. It seems unprofitable
therefore to offer conjectures as to the probability of greater
nearness of the Natuna mammals as a whole to those of Borneo
or to those of the Malay Peninsula.

Thomas (O.) and Hartert (E.). On a second collection of mammals from the
Natuna Islands. Novitates Zoologicae, ii, pp. 4S9-492. December, 1895.

Bonhote (J. Lewis). On the squirrels of the Ratufa (Sciurus) bicolor group.
Ann. and Mag. Nat. Hist., 7th ser., v, pp. 490-499. June, 1900.

Thoinas (O.). The red living squirrel of the Natuna Islands. Novitates
Zoologicse, VII, p. 592. December S, 1900.

Bonhote (J.Lewis). On the Squirrels of the Sciurus Prevostii Group. Ann.
and Mag. Nat. Hist., 7th ser., vii, pp. 167-177. February, 1901.

^ Gray's " Notice of a species of Tupaia from Borneo, in the collection of the
British_^Museum " in the Proceedings of the Zoological Society of London for
1S65 (p. 322) may be added to the bibliography of Natuna mammals, as the ani-
mal described, though supposed to have been taken in Borneo, is apparently con-
fined to Bunguran Island, the largest of the Natunas.

'^ Megaderma spasma, Myotis 7nuricola, Taphozoics mclanofogon, Mydaics
meliceps, Paradoxtirus hermap/iroditus, Lutra sutnatrana and Mus ep/iippium.

"See papers already cited, also Novitates Zoologicae, i, p. 46S (letter from
Mr. Everett) ; ibid., i, p. 483 (note on land shells by Mr. E. Smith), ibid., 11, p. 478
(Birds) ; ibid., u,'j^. 499 (Reptiles).

MAMMALS OF THE NATUNA ISLANDS II3

MANIS JAVANICA Desmarest.

1895. Manis java7iica Thomas and Hartert, Novitates Zoologicae, ll, p.
492. December, 1895 (Bunguran).

An adult male was taken on Bunguran, June 24, 1900. Total
length 914 ; head and body 508 ; tail 406. ,

TRAGULUS BUNGURANENSIS sp. nov.

Type. — Adult male (skin and skull) No. 104604 U. S. National
Museum. Collected on Bunguran Island, North Natunas, July 9,
1900. Original number, 547.

Characters. — Color pattern essentially as in Tragiihis jiigricaits
Thomas, from Balabac. Size equal to that of T. canescens from the
Malay Peninsula, therefore much greater than in the Balabac animal.

Color. — Back vmiform ochraceous, fading to buff on sides, the hairs
everywhere gray at base. Both back and sides everywhere darkened
by black hair tips, but these never sufficiently abundant to produce
a dark shading in excess of the ochraceous. The relative proportion
of the dark wash to the light under color is precisely the same as in
Tragichcs canescens and T. napu (from Linga Island) but the black
is less conspicuous than in the Bornean form of T. napu. Legs, ex-
cept white area on inner side, like back but slightly brighter and less
shaded with black. Entire dorsal and lateral surface of neck clear
black to base of hairs, a few ochraceous specks visible on close scru-
tiny, particularly at sides near throat markings. On shoulders this
black area fades abruptly into color of back ; on head it passes for-
ward between ears and eyes nearly to muzzle. Cheek, region between
eye and ear, and line extending forward over eye to muzzle and sepa-
rating black median stripe from naked loral space, ochraceous, essen-
tially like that of legs. Throat markings as in Tragulus tiigricans,
but white stripes apparently even more restricted. Region occupied
by posterior white stripes black, continuous with that of neck, but dis-
tinctly speckled with ochraceous. Region occupied by anterior
stripes ochraceous, continuous with that of cheeks and somewhat less
pure and more speckled with black. White stripes as follows : («)
One on each side of naked chin area. These are about 50 mm.
in length and never more than 10 mm. in breadth, but occasionally so
narrow as to break up into tw^o or more spots. They are separated
from naked chin patch by an ochraceous stripe slightly broader than
the white. Chin area narrowly and discontinuously bordered with
white, especially in front, {b) Two posterior lateral stripes varying

Proc. Wash. Acad" Sci., March, iqoi.

114

MILLER

from 50 mm. to 80 mm. in length, and never more than 12 mm. wide.
They are strongly convergent anteriorly, and sometimes nearly joined
too-ether in front by a median spot. These white stripes are always
separated from the anterior stripes by an ochraceous median area
varying from 10 mm. to 25 mm. in width. (c) A median stripe
lying between the posterior lateral stripes. Posteriorly this stripe is
as wide as the lateral stripes, but it quickly narrows and sometimes
disappears at middle of latter, though usually represented again by
the median spot already referred to. In none of the specimens is this
stripe broad and continuous anteriorly to level of front of lateral
stripes as in Nehring's figure of the throat markings of T. nigricans. ^
Collar narrow, ochraceous grizzled with black. It is seldom more
than 25 mm. in width ; therefore much narrower than indicated by
Nehring's figure. Behind the collar is a whitish gray median area
continuous laterally with narrow light stripe down inner side of fore
legs. This light area is sometimes divided by a dark median line
joining collar with buff of belly. Belly and chest buff, essentially
like that of sides, with which it forms no contrast in color. As on
the sides the buff is clouded by black hair tips, but the hairs are
scarcely if at all gray at base. On chest the dark hair tips tend to
form a median stripe, which is sometimes sharply defined and continu-
ous with the ochraceous line occasionally dividing \vhite of breast.
A clear whitish area slightly larger and better defined than that of
breast occupies region between hind legs. It is continuous with white
stripe down inner side of hind legs. This stripe is usually divided on
thigh by encroachment of the surrounding ochraceous. Tail silky
white below and at tip, essentially like back above.

Skull. — The skull of Tragulus bii7tgura7iensis fully equals that of
7. canescens in size, and distinctly exceeds that of the Bornean form
of T. napti. It is much larger than that of T. nigricans., which
proves to be a medium sized species like T. rtifuhis. In general form
the skull agrees so closely with that of Tragulus canesce?ts that it is
only to be distinguished by its slightly greater relative breadth and
smaller, less inflated audital bullae. As compared with the skull of
Tragulus nigricajis,- that of T. bungurane7isis is much larger (dis-
tance from back of occiput to front of canine 103 instead of 92, zygo-

' Sitz.-Berich. der Gesellsch. Naturforschender Freunde zu Berlin, 1893, p. 224.

2 For the opportunity of examining the skull of an adult male from Balabac I
am indebted to the courtesy of Mr. D. G. Elliot. A photograph (slightly reduced)
of this specimen was published by Mr. Elliot in 1S96 (Field Columbian Mu-
seum, Publication 11, Zoological Series, i, No. 3, pi. xi, May, 1896).

MAMMALS OF THE NATUNA ISLANDS II5

matic breadth 53 instead of 45), and the braincase is more conspicu-
ously ridged for muscular attachment. That part of the braincase im-
mediately above posterior root of zygoma is more conspicuously inflated.
Otherwise I can detect no salient differences in the skulls of the two
animals.

Teeth. — The teeth are uniformly larger than those of Tragtdus 7zi-
gricaiis., but in form they present no charactei^s of importance. As
compared with T. canescens the premolars both above and below are
conspicuously more robust, a character in which the Bunguran animal
agrees with the Bornean form of Tragtdus napu.

Measurements . — External measurements of type : total length 647 ;
head and body 571 ; tail vertebrae 76 ; hind foot 146 ; hind foot with-
out hoofs 128. Average and extremes of five adults from the type
locality : total length 643 (628-673) '•> ^'^^^^ '"^nd body 566 (558-584) ;
tail vertebra 77 (70-S9) ; hind foot 142 (140-146) ; hind foot without
hoofs 126 (124-128).

Cranial measurements of type: greatest length 114; basal length
107; basilar length 100; occipito-nasal length 106; length of nasals
32 ; diastema 13 (9) ;^ zygomatic breath 52 (46) ; least interorbital
breadth 33 (28) ; greatest breadth of braincase above base of zygomata
38 (33) ; mandible 91 (78) ; maxillary toothi-ow (alveoli) 38 (34) ;
mandibular toothrow (alveoli) 44 (39) ; anterior upper premolar
7x3.8 (6.4x3) ; middle lower premolar 7.2x3 (5.8x2.4).

Weight. — Weight of type 3.8 kg. ; of two other males 3.6 kg. each.
Two adult females weigh respectively 3.6 kg. and 4.2 kg.

Speci?nens examined. — Six, all from the type locality.

Remarks. — Trao-ulus bunoruranensis is so distinct from the other
known species as to require no detailed comparisons.

TRAGULUS sp.

Two specimens from Sirhassen Island are too immature for deter-
mination. Appai'ently they represent a member of the napu group,
allied to that occurring in Borneo. The throat markings show no ap-
proach to those of Tragulus bungtiraneJisis.

TRAGULUS JAVANICUS (Gmelin).

1894. Tragidus Javanicus Thomas and Hartert, Novitates Zoologicse, i,
p. 660. September, 1864 (Bunguran).

1895. Tragulus jai/aniciis TnomAS and Hartert, Novitates Zoologicae, 11,
p. 492. December, 1895 (part, specimens from Bunguran).

Six specimens from Bunguran.

'Measurements in parentheses are those of an adult male topotjpe of Tragulus
nizricans.

Il6 MILLER

TRAGULUS PALLIDUS sp. nov.

1895. Tragulus Javanicus TnoukS and Hartert, Novitates Zoologicas, 11,
p. 492. December, 1895 (part, specimen from Pulo Laut).

Type. — Adult female (skin and skull) No. 104616 U. S. National
Museum. Collected on Pulo Laut, North Natuna Islands, August
II, 1900. Original number 625.

Characters. — Smaller than Traguhis javanicus from Borneo or
Bunguran and very pale in color. Black clouding of upper parts in-
conspicuous, but dark nape band well defined.

Color. — Back and sides light ochraceous-buff everywhere clouded
oy the blackish hair-tips, but these never in excess, except perhaps
along middle of back and across lumbar region. Flanks, shoulders,
neck, outer surface of legs and narrow line dividing color of sides
from that of belly pale ochraceous. Nape band clear black, sharply
defined from color of sides but quickly fading into that of shoulders.
Top of head dull dark brown. A faint pale stripe over and in front of
eye. Throat markings normal, the dai-k bands like neck. Collar very
narrow. Under parts and inner surface of legs white. A faint yel-
lowish shade along middle of belly. Tail white beneath and at the
tip, ochraceous faintly shaded with brown above.

Skull. — The skull of the type, though fully adult and with all the
teeth distinctly worn, is smaller than in Bunguran specimens so young
that the posterior molars are still below the rim of the alveoli. In
form, however, it shows no marked peculiarities, though in general it
appears to be somewhat broader in proportion to its length than that
of the Bunguran animal.

Teeth. — Teeth as in specimens of Tragiilus javanicus ixor^ Bungu-
ran except that the premolars, both above and below, are shorter and
broader, a difference which may prove to be an individual peculiarity
only.

Measurements. — External measurements of type : Total length
539; head and body 444; tail vertebrce 95; hind foot 107; hind
foot without hoofs 95.

Cranial measurements of type: Greatest length 90 (94^); basal
length 83 (87) ; basilar length 78 (82) ; occipito-nasal length 83
(89) ; length of nasals 25 (29.6) ; diastema 9.2 (9.8) ; zygomatic
breadth 41.4 (40) ; least interorbital breadth 26.4 (25) ; breadth of
braincase over roots of zygomata 29.4 (28.4) ; mandible 72 (75) ;

^Measurements in parentheses are those of a less mature specimen from
Bunguran.

MAMMALS OF THE NATUNA ISLANDS II7

maxillary toothrow (alveoli) 31.6 (34); first upper premolar 6.4
X 2.8 (7 X 2.6) ; mandibular toothrow (alveoli), 35.8 (38).

Spechnens exa?nined. — One, the type.

Remarks. — This is a pallid form of Tragiihis Javanictis^ a species
which apparently shows very little tendency to become differentiated
into local races. The characters of the Pulo Laut animal were pointed
out by Thomas and Hartert in 1895.

SUS NATUNENSIS sp. nov.

1894. Sus sp. Thomas and Hartert, Novitates Zoologicae, i, p. 660. Sep-
tember, 1894 (Bunguran).

1895. Sus sp. Thomas and Hartert, Novitates Zoologicae, 11, p. 492. De-
cember, 1895 (Bunguran).

Type. — Adult female (skin and skull) No. 104856 U. S. National
Museum. Collected on Pulo Laut, North Natuna Islands, August 6,
1900. Original number 609.

Characters. — Externally much like the Tenasserim form of Sus
cristatus., but smaller ; body brownish in marked contrast with black
legs and face ; skull conspicuously shorter and broader.

Fur. — The fur throughout consists of bristles with no admixture of
softer hairs. The bristles are everywhere less stiff than in the
Tenasserim pig, but the difference is most noticeable in the mane,
which, though well developed (about So mm. in length), is composed
of bristles very slightly coarser than those of the surrounding parts,
and of not more than half the diameter of the corresponding hairs in
females of S. cristatus. Muzzle, chest, belly and ears nearly bare.

Color. — General color black, clear and unmixed with brown on
legs, throat, and face, but elsewhere heavily overlaid with brownish
buff, particularly on back and sides. The brownish wash ceases ab-
ruptly just in fi'ont of ears, leaving the face and cheeks clear black.
A conspicuous dull buff streak 100 mm. long and about half as wide
at middle extends back from angle of mouth to level of posterior can-
thus of eye. It is sharply outlined above by black of cheeks, and be-
low by that of chin. A faint buffy mark beneath eye. Tail like
back.

Skull. — The skull while much shorter than that of Sus cristatus
from Tenasserim is actually broader. As a result the width across
postorbital processes is contained only about three times in occipito-
nasal length, as opposed to nearly four times in the related species.
Similarly the zygomatic breadth slightly exceeds one half of the basi-
lar length, while in Sus cristatus it is less than half. Width of pal-

Il8 MILLER

ate between middle molars almost exactly one sixth distance from pos-
terior edge of palate to front of premaxillaries (measured along me-
dian line). In Szis cristatus the palatal width is contained nearly
seven times in the same distance. Dorsal profile of skull slightly con-
cave near base of nasals. Zygomata heavier and deeper than in Sus
cristatus. Audital bullae noticeably smaller and less inflated than in
the Tenasserim pig. Mandible shorter and much more robust than
that of Sus cristatus, the outward bulge of the ramus a little behind
middle of toothrow greatly accentuated.

Teeth. — As the teeth of the two specimens of Sus natunensis are
much worn, while those of the only skulls of Sus cristatus at hand are
not fully grown, it is impossible to make any accurate comparisons.
The smaller size of the Natuna pig's teeth is, however, evident for the
length of the entire upper toothrow does not equal that of S. cristatus
without the posterior molar. The crown of the middle upper molar
appears to be more nearly square in outline than that of the Tenas-
serim pig, but in the very different condition of the specimens it
would be unsafe to assume that this character is constant.

Measurements, — External measurements of type ; total length
1294; head and body 11 17; tail vertebras 177; height at shoulder
558 ; hind foot 220 (170) ; ear from meatus 100 ; w^dth of ear 75.

Cranial measurement of type: greatest length 295 (332'); oc-
cipito-nasal length 282 (316) ; basal length 245 (275) ; basilar length
235 (263) ; length of nasals 135 (157) ; width of both nasals together
posteriorly34 (33) ; median length of bony palate 168 (183) ; width of
bony palate at middle of second molar 30 (29) ; breadth between tips of
postorbital processes 87 (87) ; least interorbital breadth 64 (65) ; zygo-
matic breadth 130 (133) ; occipital breadth 58 (62) ; occipital depth
100 (103) ; least depth of rostrum between canine and incisor 33 (39) ;
mandible 225 (232) ; depth of mandible through coronoid process 104
(no); depth of ramus at front of first molar 40 (41); maxillary
toothrow to front of canine (alveoli) 113 (131^); mandibular tooth-
row to front of canine (alveoli) 120 (138) ; ci^own of first upper molar
12 X 13 (18 X 16) ; crown of second upper molar 18 X 18 (22 x 16).

Weight. — Weight of type, 40 kg. ; weight of adult female from
Pulo Lingung, 35 kg.

Specimetzs exaf?iined. — Two, one from Pulo Laut, the other from
Pulo Lingung.

^ Measurements in parentheses ai-e those of a Tenasserim specimen (female) of
Sus cristatus so young that the posterior molar is not fully in place.
*Last molar not fully grown.

MAMMALS OF THE NATUNA ISLANDS II9

Remarks. — While the two specimens agree in all essential characters
they differ in numerous minor details. The skin from Pulo Lingung
is somewhat darker than the type, but the difference is due to the shade
of the brown wash, not to any extension of the black. The skull of
this specimen is more rounded posteriorly than that of the type, and the
rostrum is shorter. Both specimens show conclusively that their rela-
tionships are with the Sus cristatus of the Malay Peninsula and not
with the S. longirostris of Borneo, a case which finds an exact par-
allel in the giant squirrels.

MUS INTEGER sp. nov.

7>/^.— Adult male (skin and skull) No. 104837 U. S. National Mu-
seum. Collected on Sirhassen Island, South Natunas, June 7, 1900.
Original number 455.

Characters. — A large robust species with coarse but not spinous
fur. Relationships with Mus validus Miller, from Trong, Lower
Siam, and Mus mulleri Jentink from Sumatra. Differs from the
former in smaller size and in the absence of the anterior outer tubercle
of the last upper molar, and from the latter in larger size, and yellowish
brown (not white) underparts.

Color. — Back and sides a fine grizzle of black and dull ochraceous
(the exact shade intermediate between the ochraceous and ochraceous-
buff of Ridgway), the two colors nearly equally mixed on back, but the
ochraceous in excess on sides. Underparts and inner surface of legs
buff. An ill defined drab-gray median line from throat to pubic region.
Head darker and more glossy than back, the cheeks distinctly washed
with gray. Lips and chin drab-gray. Feet an indefinite brown,
darker on metapodials. Ears essentially naked, dark brown. Tail
dark brown throughout. Underfur gray (Ridgway, pi. 11, No. 8), be-
coming paler on under parts where it fades irregularly into the general
buff.

jTur. — The fur is exactly as in Mus validus^ that is the grooved
bristles are so slender that their true nature is not apparent without use
of lens. On middle of back the mass of the fur is about 17 mm. in
length, the long terete hairs scattered through it reaching about 30 mm.
On rump the fur is longer but not conspicuously so, and there is no
noticeable increase in length or abundance of the terete black hairs.

Tail., feet and mamfuce. — Tail slightly more coarsely scaled than in
Mus validus; 9 rings to the centimeter at middle. Hairs scarcely
noticeable except toward tip, where they somewhat exceed the breadth
of the rings.

I20 MILLER

Feet heavy and robust. Thumb short, with a flat bhuit nail. Soles
and palms naked, the former with six well developed tubercles, the
latter with five.

Mammae, p. 2 — 2, i 3 — 2 = 8.

SktilL — In general appearance the skull of Miis mteger resembles
that of Mus valldus} It is shorter (greatest length about 51 instead
of £55) and the rostrum is relatively broader and deeper. Audital
bullae similar in form to those of JMiis validus^ but the surface less
irregular. Region between anterior bases of zygomata broader than
in Mus validus so that the arches are more nearly parallel.

Teeth, — The teeth are relatively as well as actually smaller than in
Miis validus and the enamel pattern is normal, that is, the posterior
vipper molar consists of two transverse folds, and an anterior internal
tubercle. There is no trace of the supplementary outer tubercles of
the corresponding tooth of ]\Ius validus.

Measurements. — External measurements of type : total length 463 ;
head and body 235 ' tail vertebrje 228^; hind foot 48 (45) ; ear from
meatus 19 ; ear from crown 15 ; width of ear 15. In adult male topo-
type : total length 462 ; head and body 234 ; ^ tail vertebrce 228 ; ^ hind foot
46 (44) ; ear from meatus 21 ; ear from crown 16; width of ear i6.

Cranial measurements of type: greatest length 52 (55);^ basal
length 45 (48. 6) ; basilar length 41.6 (45.6) ; palatal length 23 (26) ;
least width of palate between anterior molars 5 (5) ; diastema 14
(14.6) ; * length of incisive foramen 8 (9) combined breadth of incisive
foramina 3 (3.6) ; length of nasals 21 (22.6) ; combined breadth of
nasals 6 (6.2) ; zygomatic breadth 25 (28) ; interorbital breadth 8 (8) ;
mastoid breadth 19 (19) ; breadth of braincase above roots of zygomata
18.8 (20) ; depth of braincase at anterior border of basi-occipital 12.8
(15) ; frontopalatal depth at posterior extremity of nasals 12.8 (13.4) ;
least dejoth of rostrum immediately behind incisors 10 (10) ; maxillary
toothrow (alveoli) 9.6 (11) ; width of front upper molar 3 (3) ; man-
dible 30 (31) ; mandibular toothrow (alveoli) 9 (10).

Speci?nens examined. — Four, three from the type locality, and
one from Pulo Lingung.

Remarks. — This rat is probably a near relative of the Bornean
Mus 7}iiilleri of Thomas.^ The specimen from Pulo Lingung does
not differ appreciably from the others.

' See Proc. Biol. Soc. Washington, xiii, pi. iii and iv.

* Collector's measurement.

^Measurements in parentheses are those of the type of Mus rmlidtis.

*In the type of Mus jniilleri ihe diastema is 12 mm.

*Ann. and Mag. Nat. Hist., 6th ser., xiv, p. 450. December, 1894.

MAMMALS OF THE NATUNA ISLANDS 121

MUS SABANUS Thomas.

1887. Mus sabanus Thomas, Ann. and Mag. Nat .Hist., 5th sen, xx, p.

270. October, 1887 (Mt. Kina Balu, Borneo).
1894. Mils sabanus Thomas and Hartert, Novitates Zoologicae, i, p. 658.

September, 1894 (Bunguran).

Thirteen skins and one extra skull, all from Bunguran. There is

little probability that this rat is the same as the true il/?c5 sabanus of

Borneo.

MUS RAJAH Thomas.

1894. Mus hellwaldi Thomas and Hartert, Novitates Zoologicae, i, p. 658.
September, 1894 (Bunguran).

1894. Mus rajah Thomas, Ann. and Mag. Nat. Hist., 6th ser., xiv, p. 451.
December, 1894 (Mount Batu Song, Borneo).

1895. Mus rajah Thomas, Novitates Zoologicae, 11, p. 26. February, 1895
(Revised determination of Bunguran specimens).

Six specimens (one in alcohol) from Bunguran, two from Pulo Lin-
gung, one from Pulo Laut, four (one in alcohol) from Sirhassen, and
one (in alcohol) from Pulo Midei. It is doubtful whether these series
are referable to one species or whether any of them are the true Bor-
nean Mus rajah. The material is not wholly satisfactory, and I have
been unable to examine specimens from Borneo.

MUS NEGLECTUS Jentink.

1894. Mus rattus var. Thomas and Hartert, Novitates Zoologicae, i, p.
658. September, 1894 (Bunguran).

1895. Mus 7ieglechis Thomas and Hartert, Novitates Zoologicae, 11, p.
492. December, 1895 (Bunguran).

Five specimens from Pulo Lingung, one from Pulo Midei, and nin^

from Sirhassen. In the absence of Bornean material, I follow Thomas

and Hartert in referring the Natuna rats of the ' alexandrt?ms ' type to

]\Ius neglectus.

SCIUROPTERUS EVERETTI Thomas.

1894. Sciicropterus phayrei Thomas and Hartert, Novitates Zoologicae,

I, p. 660. September, 1894 (Bunguran).

1895. Sciuropterus evereiti Thomas, Novitates Zoologicae, 11, p. 27. Feb-
ruary, 1895 (Revised determination of Bunguran specimens).

1895. Sciuropterus everctti Thomas and Hartert, Novitates Zoologicae,

II, p. 490. December, 1895 (Bunguran).

Two specimens, both from Bunguran ; an immature male taken
July 4, and an adult female taken July 21, 1900.

PETAURISTA NITIDULA Thomas.

1894.. Pteroinys nitidus Thomas and Hartert, Novitates Zoologicae, i, p.
660. September, 1894 (Bunguran).

122 MILLER

1895. Pteroinys nitidtis Thomas and Hartert, Novitates Zoologicae, 11, p.
490. December, 1895 (Bunguran).

1900. Petaiirista tiitidula Thomas, Novitates Zoologicas, vii, p. 592. De-
cember 8, 1900 (Bunguran).

Seven specimens from Bunguran.

SCIURUS PROCERUS sp. nov.

1894. Sciurus femds Thomas and Hartert, Novitates Zoological, i, p. 659.
September, 1894 (Bunguran).

1895. Sciurus tenuis Thomas and Hartert, Novitates Zoologicss, 11, p. 492.
• December, 1895 (Bunguran).

Type.—KAxxW. male (skin and skull) No. 10469S U. S. National Mu-
seum. Collected on Bunguran Island, North Natunas, July 18, 1900.
Original number 574.

Characters. — Externally similar to Sciurus tenuis though some-
what smaller. Skull very much smaller and relatively broader than in
the related species.

Color. — The color is exactly like that of Sciurus tefiuis from Sing-
apore.

Skull atid teeth. — Except that it appears to be broader throughout,
relatively to its length, the skull of Sciurus procerus is essentially a
miniature of that of S. tetzuis, as the braincase shows none of the ten-
dency to increased depth characteristic of the Bornean animal. Ratio
of rostral depth to distance between middle of interparietal and lower
rim of audital bulla, 50. This ratio is 49 in S. tenuis.

Measurefnents. — External measurements of type : total length 235 ;
head and body 140; tail vertebrje 95 ; hind foot 35 (33). Average
and extremes of four specimens from the type locality : total length
239-5 (235-247); head and body 140; tail vertebrae 99.5 (95-107);
hind foot 35.2 (34-36.5) ; hind foot without claws 32.9 (31.8-34).

Cranial measurements of type : greatest length 34 (38) ;^ basal
length 28.6 (32) ; basilar length 26 (29) ; palatal length 14.6 (16) ;
diastema, 7.6 (8.8) ; length of nasals 10.4 (11.4) ; greatest breadth of
nasals 4.8 (5.6); interorbital breadth 12 (12.6); zygomatic breadth
20.8 (21) ; greatest breadth of braincase 17 (17.6) ; cranial depth from
middle of interparietal to lower rim of audital bulla 14 (15); least
depth of rostrum 7 (7.2) ; mandible, 20 (21) ; maxillary toothrow
(alveoli) 6 (7) ; mandibular toothrow (alveoli), 6 (7).

Specimens examined. — Six, all from the type locality.

Remarks. — This species is immediately distinguishable from its

^ Measurements in parentheses are those of an adult male topotjpe of Sciurus
tenuis.

MAMMALS OF THE NATUNA ISLANDS 1 23

allies by its small skull, scarcely larger than that of Funambidus
macclellaiidi.

SCIURUS NATUNENSIS (Thomas).

1894. Schirus /otvz Thomas and Hartert, Novitates Zoologicae, i, p. 659.
September, 1894 (Sirhassen),

1895. Sciurus lowi natitnensis Thomas, Novitates Zoologicae, 11, p. 26. Feb-
ruary, 1895 (Revised determination of Sirhassen specimen).

1895. ? Sciurus lowinatunensis Thomas and Hartert, Novitates Zoologicae,
II, p. 491. (Bunguran and Pulo Laut.)

Four specimens from Sirhassen. The average and extreme meas-
urements are as follows: total length 222 (215-229) ; head and body
135 (133-140) ; tail vertebrae %6 (82-89) 5 hind foot 33.6 (33-35) ;
hind foot without claw 31.5 (30.5-32).

SCIURUS LINGUNGENSIS sp. nov.

1895. ? Sciurus lowi 7iatunetisis Thomas and Hartert, Novitates Zoologicae,
II, p. 491. (Bunguran and Pulo Laut.)

Type. — Adult male (skin and skull) No. 104693 U. S. National
Musuem. Collected on Pulo Lingung off southern extremity of Bun-
guran, North Natuna Islands, June 19, 1900. Original number 494.

Characters. — Externally similar to Sciurus natunensis (Thomas),
but slightly larger (hind foot with claws 36 Instead of 33.6). Skull
larger than that of S. natmze?zsis^ the audital bullae much broader
anteriorly.

Color. — The color is precisely as in Sciurus natunensis., and there-
fore requires no detailed description.

Skull. — Skull larger than that of Sciurus natunensis (see measure-
ments) but not different in general form. The audital bullae are,
however, readily distinguishable by the much greater development of
the anterior inner lobe. In Sciurus tzatu7iensis this lobe is so small
as scarcely to form any part of the general contour of the bulla. In
S. lingungensis it is nearly equal to the anterior outer lobe, together
with which it imparts a distinctly triangular outline to the ventral as-
pect of the bulla.

Aleasuretiieitts. — External measurements of type : total length 229 ;
head and body 140; tail vertebrae 89; hind foot 36 (33.7) ; ear from
meatus 12 ; ear from crown 7. A second specimen from the type lo-
cality gives precisely the same measurements.

Cranial measurements of type : greatest length 38 (36) ;^ basal length
33 (3^)5 basilar length 30 (29); palatal length 17 (16); greatest

' Measurements in parentheses are those of an older specimen of Sciurus natu-
nettsis from Sirhassen.

124

MILLER

length of nasals ii (lo) ; greatest width of both nasals together 5 (5) ;
interorbital breadth 12 (11.4) ; zygomatic breadth 33.4 (20) ; mastoid
breadth 17 (16.6) ; depth of braincase at anterior edge of basi-occipital
13.6 (13) ; mandible 23 (33) ; maxillary toothrow (alveoli) 6.4 (7) ;
mandibular toothrow (alveoli) 7 (7).

Speciviens examined. — Two, both from the type locality.

Remarks. — While Sciiirus lingunge7isis is scarcely distinguishable
from S. natiutensis by external characters alone, size of the skull and
form of the audital bullae are clearly diagnostic. Both species from
the Natunas are separated from the Bornean 6*. loxvi Thomas by their
well developed ears, and shorter broader rostral portion of skull.

SCIURUS LUTESCENS sp. nov.

1894. Sciurtis notatus Thomas and Hartert, Novitates Zoologicae, i, p. 659.
September, 1894 (part, specimens from Sirhassen).

Type. — Adult male (skin and skull) No. 104668 U. S. National
Museum. Collected on Sirhassen Island, South Natunas, June 3,
1900. Original number 429.

Characters. — Allied to Sciurtis noiatus., but considerably smaller
than the Bornean representative of the species. Colors very pale, the
under parts buff or cream-buff (Ridgway, pi. v, nos. 13 and 11) ir-
regularly tinged with gray.

Color. — Entire dorsal surface of body and tail a fine grizzle of black
and cream-buff, the individual hairs black with two or three cream
buff rings. On tail the grizzle is less fine than on back, and it shows a
faint tendency to resolve itself into obscure cross bands. On sides of
body and on head the cream-buff brightens to buff. Cheeks and
muzzle buff, scarcely grizzled. Feet slightly yellower than sides,
under parts and inner surface of legs pale buff, palest anteriorly and
laterally (where it about matches the cream-buff of Ridgway) bright-
est along median line. Under side of tail dull ochraceous-buff slightly
grizzled with black. Pencil not different from rest of tail. Between
the colors of sides and belly are the usual longitudinal stripes. The
outer of these is about 5 mm. in width, and cream-buff in color.
The inner is about twice as wide, and black, but much obscured by a
thick sprinkling of bluish gray hairs. Outer surface of ears concolor
with neck, inner surface like cheeks. The sprinkling of bluish gray
hairs on sides of belly extends irregularly forward to axilla and inner
side of front leg, occasionally to throat and chin.

Skull. — As compared with the Bornean form of Sciurus notatus^
the skull of 6". lutescens is much smaller (greatest length about 45 in-

MAMMALS OF THE NATUNA ISLANDS I25

stead of 50) the rostrum is relatively shorter and broader, and the
audital bullae are less elongate antero-posteriorly. Teeth as in Sciurus
notatus except that they are uniformly smaller.

Measurements. — External measurements of type : total length 355 ;
head and body 177 ; tail vertebrae, 177 ; hind foot 45 (41). Average
and extremes of six specimens fi'om the type locality : total length 356
(329-375) ; head and body 186 (177-196); tail vertebrte 170 (152-
178) ; hind foot 43.8 (41-45) ; hind foot without claws 40.7 (39-
42).

Cranial measurements of type: greatest length 45.4 (50.4)^; basal
length 39 (43) ; basilar length 36.4 (41) ; palatal length 20 (23) ;
palatal width between middle molars G (6) \ greatest length of nasals
13 (14.8) ; greatest width of both nasals together 6.6 (7) ; interor-
bital breadth 15.4 (17) ; mastoid bi'eadth 21 (21) ; zygomatic breadth
26 (29) ; depth of braincase at anterior edge of basi-occipital 16 (16.8) ;
mandible 28 (30) ; maxillary toothrow (alveoli) 8 (9) ; mandibular
toothrow (alveoli) 8 (9).

Specimens examined. — Seven (one in alcohol), all from the type
locality.

Rejnarks. — This squirrel is recognizable among the members of the
S. notatus group by its light colors, and particularly by the pallor of the
under parts. In the latter characteristic it is approached by the form
inhabiting Pulo Laut, but with this exception it is unique among the
fulvous bellied species. The six specimens show no variation worthy
of note.

SCIURUS SERAIS sp. nov.

Type. — Adult male (skin and skull) No. 104660 U. S. National
Museum. Collected on Pulo Seraia, South Natuna Islands, May 29,
1900. Original number 415.

Characters. — Most nearly related to the small, pallid, Sciurus
hitescens from Sirhassen Island, but upper parts slightly less pale,
and under parts and pale side stripe buff -yellow, the former without
admixture of gray.

Color. — Upper parts as in Sciurus hitescens except that the pale
bands on the hairs are more nearly buff than cream-buff. Tail essen-
tially as in kS. lutescens but a shade less pale. Under parts buff -yellow
darkening in-egularly to dull orange-buff. Dark side stripe broad and
well defined.

Skicll. — The skull closely agrees with that of Sciurus lutescens in

' Measurements in parentheses are those of an adult Sciurus notatus from
Borneo.

126 MILLER

both size and form, though it is perhaps even broader in proportion to
its length. Teeth as in S. lutescens.

jMeasuremeiits. — External measurements of type : total length 368 ;
head and body 197; tail vertebrse 171 ; hind foot 44 (40). Avei^age
and extremes of four specimens from the type locality : total length
347 (323-368) ; head and body 1S4 (171-197) ; tail vertebrae 163
(152-171) ; hind foot 43.7 (43-45); hind foot without claws 40.1

(39-5-40 •

Cranial measurements of type : greatest length 45 ; basal length
38.6 ; basilar length 36 ; zygomatic breadth 26.4 ; least interorbital
breadth 17; mandible 28; maxillary toothrow (alveoli) 8.6; mandib-
ular toothrow (alveoli) '$>.6.

Specimens examined. — Four, all from the type locality.

Remarks. — As might be expected from the geographic position of
the island it inhabits, Sciurzis seraice differs from the Bornean 6".
7iotaius in much the same way as the Sirhassen representative of the
group. It is readily distinguishable from the Sirhassen animal by the
different color of the under parts. In color Sciurus seraice closely
resembles S. abbottii oi the Tambelan Islands. The latter is, however,
a much larger animal, with a longer and relatively narrower skull.

SCIURUS RUTILIVENTRIS sp. nov.

Type. — Adult male (skin and skull) No. 104658 U. S. National
Museum. Collected on Pulo Midei (Low Island), South Natuna
Islands, May 24, 1900. Original number 405.

Characters. — Size slightly greater than that of Sciurus lutescens
and S. seraice.^ but not equal to that of the Bornean or Bunguran rep-
resentatives of 6*. fiotatus. Color above as in S. seraice. Under
parts bright clear orange-rufous.

Color. — Color exactly as in Sciurus seraice except that the pale
side stripe is light cream-buff and the under parts are bright orange
rufous. Tail without trace of red suffusion.

Skull and teeth. — The skull and teeth are a trifle larger than in
Sciurus lutescens and 6". seraice., but the difference is scarcely a
tangible one.

Measurem^ents . — External measurements of type : Total length
368; head and body 190; tail vertebrae 178; hind foot 45 (41).
Average and extremes of seven specimens from the type locality :
total length 356 (330-368) ; head and body 186 (178-190) ; tail
vertebrae 173 (165-184); hind foot 45.5 (43-48); hind foot without
claws 42.2 (39.5-45).

MAMMALS OF THE NATUNA ISLANDS 1 27

Specimens examined. — Seven, all from the type locality.

Rei7iarks. — This squirrel is remarkable among the Natuna members
of the S. notatus group for the brilliant color of its vmder parts. In
this respect it surpasses all of the related forms with which I am ac-
quainted. The red color is, however, strictly confined to the body,
showing no tendency to spread to the tail as in S. miniatiis of the Ma-
lay Peninsula.

SCIURUS RUBIDIVENTRIS sp. nov.

1894. Sciurus noiatus Thomas and Hartert, Novitates Zoologicas, i, p. 659.
September, 1894 (part, specimens from Bunguran).

1895. Sciurus tiotatitsTnouxs and Hartert, Novitates Zoologlcae, 11, p. 491.
December, 1895 (part, specimens from Bunguran).

Type. — Adult female (skin and skull) No. 10467 1 U. S. National
Museum. Collected on Bunguran Island, North Natunas, June 22,
1900. Original number 49S.

Characters. — Size and general appearance both above and belo\v as
in the Bornean form of Sciurus notatus., but red of under parts brighter,
and cheeks and chin distinctly less fulvous than surrounding parts.
Skull with broader, deeper braincase than in the Bornean animal.

Color. — The color so closely resembles that of the Bornean Scizcrus
notatus that no detailed description is necessary. Under parts ochra-
ceous-rufous, fading to tawny on throat, everywhere lighter and more
tinged with red than in the Bornean animal. In the latter the color
of the under parts extends forward to lips and also strongly suffuses the
cheeks and sides of head which are only a shade browner than the
throat and conspicuously more fulvous than top of head and sides of
neck. In Sciurus rubidiventris the cheeks and lips are noticeably
suffused with gray so that they form a distinct contrast with both throat,
top of head and sides of neck.

Skull. — The skull agrees in general size with that of the Bornean
animal, and is therefore much larger than in thethi"ee species from the
South Natunas. It is distinguishable by greater general breadth and
by the depth of the braincase, which perceptibly exceeds that of S.
fiotatus.

Measure?ne7tts. — External measurements of type : total length 380 ;
head and body 209 ; tail vertebrae 171 ; hind foot 49 (44.5). Averages
and extremes of seven specimens from the type locality : total length
37S (3^S~393) ' head and body 208 (203-223) ; tail vertebrae 173
(165-184) ; hind foot 49.3 (48-50) , hind foot without claws 45.7

(44-5-47) •

Cranial measurements of type: greatest length 52.4 (50.4);^ basal
1 Measurements in parentheses are those of an adult Bornean Sciurus notatus.

128 MILLER

length 44 (43) ; basilar length 41 (41) ; palatal length 23 (23) ; palatal
width between middle molars 6 {6) ; greatest length of nasals 15
(14.8) ; greatest width of both nasals together 7.2 (7) ; interorbital
breadth 18.2 (17) ; mastoid breadth 23 (21) ; breadth of braincase
above roots of zygomata 24 (22) ; zygomatic breadth 30.4 ( 29) ;
depth of braincase at anterior edge of basi-occipital 17.8 (16. 8) ; man-
dible 29 (30) ; maxillary toothrow (alveoli) 9 (9) ; mandibular tooth-
row (alveoli) 9 (9).

Specimens examined. — Seven, all from the type locality.

Remarks. — In both size and general color this squirrel more closely
resembles the Bornean representative of the group than it does either
of the three forms from the South Natunas. Its relationships, how-
ever, appear to be rather with the race inhabiting Singapore Island
than with any of its near geographic allies, Sciurus latitensis excepted.

SCIURUS LAUTENSIS sp. nov.

1895. Sciurus notalus Thomas and Hartert, Novitates Zoologicae, 11, p.
491. December, 1895 (part, specimens from Pulo Laut).

Type. — Adult female (skin and skull) No. 104683 U. S. National
Museum. Collected on Pulo Laut, North Natuna Islands, August 6,
1900. Original number 612.

Characters. — Size slightly less than that of Sciurus rubidiventris
and color conspicuously pallid. Upper parts as in S. lutescens ; lower
parts nearly as in S. seraice but rather less dull ; pale side stripe much
less yellow than belly. Skull as in Sciurus rubidiventris.

Color. — Upper parts and tail as in Sciurus lutescens. Cheeks
faintly washed with ochraceous-buff. Under parts and inner surface
of legs bright ochraceous-buff (distinctly more yellow than Ridgway's
pi. V, No. 10). Lateral stripes as in 6". hitescens (not distinctly yel-
lowish as in S. seraice)., but black band usually less sprinkled with gray.
Scarcely a trace of gray in axillary region or on sides of neck.

Skull. — The skull in all respects closely resembles that of S. rubidi-
ventris except that it is slightly smaller. Its large size and the corre-
spondingly large teeth readily distinguish it from that of the South
Natuna species.

Measurements. — External measurements of type : total length 375 ;
head and body 195 ; tail vertebrae 180; hind foot 44 (41). Average
and extremes of nine specimens from the type locality ; total length
3^3 (355~379) ; head and body 1S9 (171-196); tail vertebree 170
(165-183) ; hind foot 45 (44-46) ; hind foot without claws 42

(41-43)-

MAMMALS OF THE NATUNA ISLANDS 1 29

Specimens examined. — Ten (one in alcohol), all from the type
locality.

Remarks. — Though suggesting two of the small South Natuna
squirrels in color, Sciurus lautensis is obviously related to the dark
colored Bunguran form, with which it more nearly agrees in size.

SCIURUS NAVIGATOR (Bonhote).

1894. Sciurus prevostii Thomas and Hartert, Novitates Zoologicse, I, p. 656.

September, 1894 (Sirhassen).
1901. Sciurus prevostii 7iavigator BotinoiK, Ann. and Mag. Nat. Hist., 7th

ser., VII, p. 171. February, 1901 (Sirhassen).

Nine specimens, three from Sirhassen Island and six from Pulo
Subi.

Those from Pulo Subi, while agreeing with the topotypes in color,
appear to average a trifle smaller, though the series is hardly exten-
sive enough to prove that this is constant.

RATUFA SIRHASSENENSIS (Bonhote).

1894. Sciurus bicolor albiceps Thomas and Hartert, Novitates Zoologicae,

I, p. 659. September, 1894 (Sirhassen).
1900. Ratufa ephippium sirhassenensis Bonhote, Ann. and Mag. Nat.

Hist., 7th ser., v, p. 498. June, 1900 (Sirhassen).

Two specimens, Sirhassen, June 8, 1900.

This species, though related to Ratufa ephippium.^ with which it
agrees in color-scheme, is sharply differentiated by its small size and
cranial peculiarities. It is in no way closely allied to Ratufa
bunguranensis and R. nanogigas.

As compared with that of Ratufa ephippium sandakanensis Bon-
hote, the skull in addition to its small size (greatest length 57 instead
of 65) differs in general narrowness, in the relatively greater breadth
of the nasal branches of the premaxillaries, and in the form of the
audital bullae. When the skull is held upside down and viewed from
behind the bullae are seen to be narrower than in the Bornean animal
and to rise to a much greater height above the surface of the basi-
occipital.

RATUFA BUNGURANENSIS (Thomas and Hartert) .

1894. Sciurus bicolor bu7tguranensis Thomas and Hartert, Novitates Zoo-
logicse, I, p. 658. September, 1894 (Bunguran).

1895. Sciurus bicolor bunguratiensis Thomas and Hartert, Novitates Zoo-
logicae, II, p. 491. December, 1895 (Bunguran).

1900. Ratufa ephippium bungt(ra7iensis Bonhote, Ann. and Mag. Nat. Hist.,
7th ser., V, p. 497. June, 1900.

Thirteen specimens from Bunguran, all in various stages of the

Proc. Wash. Acad. Sci. April, 1901. (129)

130 MILLER

change from the bleached winter coat to the summer pelage. In the
latter there is some color variation, mostly due to the gi'eater or less
distinctness of the drab wash overlying the Prouts-brown or ' choco-
late' of the upper parts. Not only does the drab vary in amount in
different individuals, but on every specimen it is more noticeable when
the animal is viewed from in front. The drab wash is of the same
character as that in Ratufa ajffinis^ though less conspicuous.

As Mr. Thomas has pointed out to me, after examining a specimen
of the latter, Ratufa bungiwanensis is closely allied to R. pyrsonota.
Indeed its relationship to the Siamese species is much closer than to
the R. epkt'ppzum of Borneo. Together with R. pyrsonota the Bun-
guran giant squirrel differs conspicuously from that of Borneo in its
narrow skull, lengthened audital bullce, dark feet, dark median line on
under surface of tail, and entirely brown back. From R. pyrsonota,
however, it is readily separable by its darker, less ochraceous color
both above and below, drab washed back, and by the much less dis-
tinct annulation of the hairs of the dorsal surface.

RATUFA NANOGIGAS (Thomas and Hartert).

1895. Sciurus bicolor nanogigas Thomas and Hartert, Novitates Zoologicae,

II, p. 491. December, 1895 (Pulo Laut).
1900. Ratufa ephippiuni fianogigas Bonhote, Ann. and Mag. Nat. Hist.,

7th ser., V, p. 498. June, 1900 (Pulo Laut).

Four specimens, all from Pulo Laut, the type locality.

This strongly characterized dwarf species is allied to Ratufa Pyr-
sonota and R. bunguranensis with which it agrees in color scheme.
It is in no way closely related to the large Bornean R. ephif>pium.

RATUFA ANGUSTICEPS sp. nov.

lype. — Adult male (skin and skull) No. 104646 U. S. National
Museum. Collected on Pulo Lingung, off south coast of Bunguran,
June 17, 1900. Original number 481.

Characters. — Externally like Ratufa a?zambce and R. ?nelanopepla.
Skull about equal to that of latter in length, but conspicuously narrower.

Color. — As the color is precisely like that of Rattfa ana7nbcB and
R. melanopepla it requires no description.

Skull and teeth. — The skull is immediately recognizable by its gen-
eral narrowness, but particularly in the region of the anterior zygomatic
roots. Ratio of lachrymal breadth to greatest length, 39. In the other
black backed species it is about 42. Audital bulls narrower and
more elongate than in R. melanopepla., and more elevated above level
of basi-dccipital (when skull is held upside down). Lateral processes
of basi-occipital obsolete.

MAMMALS OF THE NATUNA ISLANDS I3I

Teeth'as in the i-elated species.

AIeasureme7its. — External measurements of type : total length 748 ;
head and body 342 ; tail vertebrae 406 ; hind foot 79 (74).

Cranial measurements of type : greatest length 48. 6 (70) ;^ basal
length 57 (59) ; basilar length 52 (53) ; diastema 15.6 (16) ; length
of nasals 22 (23.4) ; breadth of nasals anteriorly 12 (13) ; breadth of
nasals posteriorly 6 (7) ; interorbital breadth 27 (28) ; lachrymal
breadth 28.4 (31) ; breadth between tips of postorbital processes 38
(41) ; zygomatic breadth 41 (44) ; mastoid breadth 31 (32.6) ; man-
dible 40 (41.6) ; maxillary toothrow (alveoli) 14 (14); mandibular
toothrow (alveoli) 14.6 (14.4).

Speci77iens examined. — One, the type.

Remarks. — While this squirrel exactly resembles the other black

backed species with untufted ears, so far as external characters are

concerned, it seems to be well differentiated in cranial peculiarities.

No black backed Ratufa has hitherto been recorded from the Na-

tunas.

RHINOSCIURUS sp.

An immature long-nosed squirrel was taken on Sirhassen Island,
June 4, 1900. In the absence of material for comparison I am unable
to determine the species. The genus is new to the islands.

ARCTOGALIDIA INORNATA sp. nov.

Type. — Adult- male (skin and skull) No. 104859 U. S. National
Museum. Collected on Bunguran Island, North Natunas, June 23,
1900. Original nuinber 502.

Characters. — Much smaller than Arctogalidia leucotis from the
Malay Peninsula or A. stigmatica from Borneo (greatest length of
skull about 100 instead of 115) and in color paler than either, the dark
dorsal stripes obsolete in adult.

Color. — General color of back and sides light silvery gray irregu-
larly suffused with buff and slightly darkened by blackish hair-tips and
by appearance at surface of hair-brown basal portion of fur. The
buff suffusion is least noticeable on back, slightly more apparent on
sides and flanks, and most evident on sides of neck, where it usually
brightens almost to buff-yellow in distinct contrast with surrounding
parts. On middle of back there is a trace of the middle dark stripe
of the three normally present in members of the genus. Head essen-
tially like back though somewhat more gray. Muzzle and ill-defined

' Measurements in parentheses are those of the type of Ratufa 7nelanopepia.
2 Teeth very much worn and many of them absent.

132 MILLER

eye ring blackish. Cheeks and short median stripe on forehead dull
whitish gray. Under parts essentially like back, but buff tinge more
diffuse. Feet and ears dark brown. Tail like back but darken-
ing to uniform brown beyond middle.

Newly born young are clear bluish gray, with scarcely a tinge of
buff. The three black dorsal stripes are clearly defined and normal in
extent.

Skull. — In addition to its smaller size the skull differs from that of
the Bornean Arctogalidia stigmatica in the relatively larger brain-
case, and less prominent audital bullae. The braincase is nearly as
broad as in the Bornean species, but the zygomatic width is distinctly
less. Audital bullae less raised above level of basi-occipital when skull
is held upside down and viewed from behind. The sagittal crest,
though of normal development in very old individuals, is absent at an
age when it is well grown in the larger species. In Arctogalidia leucotis
and A. stigmatica^ even in animals so young that the teeth are un-
worn and all the sutures of the rostrum plainly visible, the sagittal
crest is a knife-like ridge extending from proencephalon to lambdoid
suture, and rising to a height of about 4 mm. over middle of brain-
case. In much older individuals of A. inorjtata^ with worn teeth and
nearly obliterated rostral sutures, the crest is represented by a low
ridge about 5 mm. wide over middle of braincase and flat or grooved
on top. At this stage it rises very inconspicuously above level of the
adjacent surface, from which it is distinguished more by the texture of
the bone than by actual form.

Teeth. — The teeth are uniformly inuch smaller than in Arctogalidia
leucotis and A. stigmatica.^ but I can detect no important differences
in form.

Measurements. — External measurements of type : total length
1027 ; head and body 469 ; tail vertebrae 558 ; hind foot 78 (73.) Ex-
ternal measurements of an adult female: total length 911 ; head and
body 431 ; tail vertebras 480; hind foot 77 (72).

Cranial measurements of type: greatest length 102 (115);^ basal
length 96 (io6) ; basilar length 92 (103) ; median palatal length 53
(60) ; palatal breadth between anterior molars 13 (15.4) ; zygomatic
breadth 55 (60) ; breadth between tips of postorbital processes 41
(39) ; constriction in front of postorbital processes 19 (18) ; constriction
behind postorbital processes 13 (12) ; breadth of braincase above
roots of zygomata 32 (33) ; mastoid breadth 36 (38) ; mandible 76

^Measurements in pa:-entheses are those of a young adult A. stigmatica from
British North Borneo.

MAMMALS OF THE NATUNA ISLANDS I33

(86); maxillary toothrow (exclusive of incisors) 34^ (41)) mandib-
ular toothrow (exclusive of incisors) 39 (44) ; crown of first upper
molar 5.4 x 5 (5.4 X5.6) ; crown of second upper molar 4x5
(5.4 X 6.4) ; crown of second lower molar 7 x 4.2 (8.4 x 5.4).

Specimens examined. — Seven (two young in alcohol and one skull
without skin) , all from the type locality.

Remarks. — Arctogalidia inornata is so distinct from the previously
described species as to require no special comparisons. It is common
on Bunguran where it frequents the cocoanut trees, living for the most
part in the tops among the leaf stalks.

VIVERRA TANGALUNGA Gray.

1895. Viverra tangalunga Thomas and Hartert, Novitates Zoologicse, 11,
p. 490. December, 1895 (Bunguran).

Nine specimens from Bunguran. These agree in all respects with

the Bornean animal.

TUPAIA SPLENDIDULA Gray.

1894. Tupaia splendidida Thomas and Hartert, Novitates Zoologicae, I,
p. 656. September, 1894 (Bunguran).

1893. Tupaia sple^ididula typica Thomas and Hartert, Novitates Zoologi-
cee, II, p. 489. December, 1895 (Bunguran).

Two specimens from Bunguran.

TUPAIA LUCIDA (Thomas and Hartert).

1895. Tupaia splendtdula lucida Thomas and Hartert, Novitates Zoo-
logics, II, p. 490. December, 1895 (Pulo Laut).

Seven specimens (two in alcohol) from Pulo Laut.

TUPAIA SIRHASSENENSIS sp. nov.

1894. Tupaia tana Thomas and Hartert, Novitates Zoologicse, i, p. 657,
September, 1894 (Sirhassen).

lype. — Adult male (skin and skull) No. 1047 12 U. S. National
Museum. Collected on Sirhassen Island, South Natunas, June 5,
1900. Original number 442.

Characters. — In general similar to Bornean specimens of lupaia
tana., but smaller (hind foot 47 instead of 52, greatest length of skull
55 instead of 60), gray markings on head and shoulders less distinct,
and red of tail brighter. Rostral portion of skull less attenuate than
in lupaia tana.

Color. — The color so exactly resembles that of the common Bornean

Tupaia tana as to need no detailed description. Gray of head darker

^ Tooth measurements are from a younger specimen (male) with perfect denti-
tion.

134 MILLER

than in the Bornean animal and light shoulder markings less distinct
and sharply defined. Under side of tail light orange-rufous, darkening
to ferruginous toward edge. (In T. tatia these colors are replaced by
dull ferruginous and hazel respectively.)

Sktill and teeth. — The skull is throughout much smaller than in
specimens of Tupaia tana from Borneo. In form it differs from
that of 1 . tana in less slender and elongate rostrum, narrower brain-
case and slightly shorter audital bullae. Suborbital vacuity much
broader than in T. tana. Teeth as in the Bornean animal.

Measurements. — External measurements of type : Total length 355 ;
head and body 203 ; tail vertebrae 152 ; hind foot 46.4 (44). Average
and extremes of four adults from the type locality : total length 367
(365-371) ; head and body 203; tail vertebrae 163 (162-16S) ; hind
foot 45.4 (44-46.6); hind foot without claws 42.5 (41-44).

Cranial measurements of type: greatest length 54.6 (61);' basal
length 49 (54) ; basilar length 46.4 (51) ; median palatal length 48
(53) ; distance from lachrymal notch to tip of premaxillary 27.6 (31) ;
least interorbital breadth 14.4 (16) ; zygomatic breadth 25 (28.4) ;
mandible 38 (41) ; maxillary toothrow (behind diastema) 20 (21.4) ;
mandibular toothrow (behind diastema) 17 (18).

Specimens examined. — Five, all from the type locality.

GALEOPITHECUS VOLANS (Linnaeus).

1894. Galeopiihecus 7wlans Thomas and Hartert, Novitates Zoologicae, i,
p. 657. September, 1894 (Bunguran and Sirhassen).

Two specimens from Sirhassen and two (one yovmg in alcohol),
from Bunguran. Also foetus of one of the Sirhassen specimens.

EMBALLONURA ANAMBENSIS Miller.

Four specimens from Bunguran. These agree essentially with the
Anamba animal, but show some slight cranial peculiarities.

PIPISTRELLUS SUBULIDENS sp. nov.

Type. — Adult female (in alcohol) No. 104758 U. S. National
Museum. Collected on Sirhassen Island, South Natunas, June 3,
1900.

Characters. — Similar to Pipistrellus pipistrellus (Schreber) in
size, color and external form, but skull with broader rostrum, and inner
upper incisor without supplemental cusp.

'Measurements in parentheses are those of an adult male Bornean Ttipaia
ta7ia.

MAMMALS OF THE NATUNA ISLANDS I35

Skull. — The skull is of the same size as that of Pipistrellus pipis'
trellus., but the braincase is narrower and more elongate, and the ros-
trum is very markedly shorter and broader. The great breadth of the
anterior portion of the skull involves also the palate and interpterygoid
space, both of which are noticeably wider than in Pipistrellus pipis-
trellus. Audital bullae slightly smaller than in the European species.

7 eeth. — The teeth are essentially as in Pipistrellus pipistrellus.,
except that the inner upper incisor lacks the small supplemental cusp.
Mandibular teeth wider than those of P. pipistrellus.

Measurements. — External measurements of type : total length 76 ;
head and body 41 ; tail 33; tibia 14; foot 6; calcar 10; forearm
32.4; thumb 6; second digit 30; third digit 60; fourth digit 53;
fifth digit 43 ; ear from meatus 1 1 ; ear from crown 9 ; width of ear
9.6 ; tragus (measured in front) 4.

Ci'anial measurements of type: greatest length 12.4 (12);^ basal
length 1 1.8 (11. 6) ; basilar length 9 (9) ; zygomatic breadth 8.4 (8) ;
least interorbital breadth 3.2 (3.2) ; greatest length of braincase 8
(7.6) ; greatest breadth of braincase above roots of zygomata 6.6
{^^.^) ; mandible 8.8 (8.4) ; maxillary toothrow (exclusive of incisors)
4.2 (4.2) ; mandibular toothrow (exclusive of incisors) 4.8 (4.8).

Specimens examined. — Six (in alcohol), all from the type locality.

Remarks. — I am unable to identify this bat with any described
species. Externally it is practically identical with Pipistrellus
fipistrellus except that the color, so far as can be judged from speci-
mens preserved in alcohol, is more blackish. Internally it is readily
distinguished by the characters of the skull and teeth. From Pipis-
trellus abramus it differs externally in smaller size, narrower ears,
and in the absence of any unusual development of the penis. The
incisors differ from those of P. abramus in the same manner as from
those of P. pipistrellus.

HIPPOSIDEROS LARVATUS (Horsfield).

Two specimens (one in alcohol) were collected on Sirhassen Island,
June 6 and 7, 1900.

RHINOLOPHUS AFFINIS (Horsfield).

One badly damaged specimen from Bunguran appears to be refer-
able to typical Rhinolophus aff,ms. The forearm cannot be meas-
ured, but the third finger is 75 mm. in length. Tibia 21, foot 10.4,

1 Measurements in parentheses are those of an adult skull of Pipistrellus
pipistrellus from Switzerland.

136 MILLER

ear from meatus 21. Ridge on muzzle beneath edge of nose leaf low,
broad and hairy, not in the least suggesting a supplementary leaflet.

RHINOLOPHUS SPADIX sp. nov.

1894. Rhinolophus affinis Thomas and Hartert, Novitates Zoologicae, 11,
p. 656. December, 1895 (Sirhassen).

Type. — Adult female (in alcohol) No. 104752 U. S. National
Museum. Collected on Sirhassen Island, South Natunas, June, 1900.

Characters. — In general like Rkifiolophtis ajffinis but much smaller.
Color uniform tawny brown. Muzzle with distinct supplemental
leaflets.

Muzzle. — Muzzle and noseleaf precisely as in Rhinolophus affinis^
except that the ridge on muzzle beneath edge of horseshoe is de-
veloped into a distinct supplemental leaflet resembling those present
in Hipposideros. In this respect Rhinolophus spadix resembles the
animal from Burmah referred by Thomas to Rhitiolophus rouxii ; ^
but the terminal erect portion of the noseleaf is not shortened or in any
way peculiar in form.

Ears. — The ears resemble those of Rhinolophus affinis^ except
that they are not as large.

Color. — Fur everywhere russet, slightly paler on ventral surface,
darker and somewhat tinged with hazel above. Ears and membranes
dark brown.

Skull and teeth. — The skull and teeth exactly resemble those of
mainland specimens of Rhinolophus affinis except for their uniformly
smaller size.

Measurements. — External measurements of type : total length, 70
(85''') ; tail 21 (23) ; tibia 17.6 (24) ; foot 8 (10) ; calcar 12 (13) ;
forearm 43 (51) ; thumb 8 (8.6) ; second digit 32 (40) ; third digit
64 (77) ; fourth digit 53 (61) ; fifth digit 54 (63) ; ear from meatus
17 (20) ; ear from crown 14 (17) ; length of noseleaf from lip 13
(16) ; greatest width of noseleaf 8 (9).

Cranial measurements of type : greatest length 18 ( 23 ) ; basal
length 16 (20.4) ; basilar length 14.6 (18) ; zygomatic breadth 9 (ii) ;
least interorbital breadth 2.4 (2.4) ; greatest length of braincase 10.4
(13) ; greatest breadth of braincase above roots of zygomata 8 (9.4) ;
frontopalatal depth (at middle of molar series) 4 (4.8) ; depth of
braincase 6 (7) ; mandible 11.8 (15) ; maxillary toothrow (exclusive

'Ann. Mus. Civ. di Storia Nat. di Geneva, Ser. 2, x, p. 923, p1. xi, 1892.
^ Measurements in parentheses are those of an adult female Rhitiolophus
affinis from Trong, Lower Siam.

MAMMALS OF THE NATUNA ISLANDS I37

of incisor) 6.8 (9) ; mandibular toothrow (exclusive of incisors) 7

(9-8). ^

Specimens examined. — Three (one skin), all from the type locality.

Remarks. — Rhinolophus spadix is so readily distinguished from
its relatives of the R. affinis group that it needs no special comparisons.
It is a much smaller animal than the species from the Anambas that I
recently referred to R. rouxii.^ In color the latter is a dull brown not
in the least resembling the russet of R. spadix.

CYNOPTERUS MONTANOI Robin.

1 894. Cynopterus marginatus Thomas and Hartert, Novitates Zoologicae, i ,
p. 655. September, 1894 (Sirhassen and Bunguran).

1899. Cynopterus montanoi Matschie, Die Fledermause des Berliner Mu-
seums fiir Naturkunde, p. 75. August, 1899. (Natuna record of C.
marginatus placed in synonymy of C. montanoi.')

Five specimens (three skins) from Sirhassen. These agree so
closely with a skin and two bleached alcoholic specimens from Sing-
apore, which I suppose to be the same as the Malaccan Cynopterus
montaftoi., that without more material it is impossible to distinguish
the Natuna animal from that of the southern extremity of the Malay
Peninsula. Cynopterus montanoi as thus understood differs from C.
angulatus Miller^ of Lower Siam in its more slender skull and in the
absence of the white border of the ear, and from C. titthcecheilus
(Temminck) of Sumatra and Java in its conspicuously smaller size.

PTEROPUS VAMPYRUS (Linnjeus).

1894. Pteropus va7npyrus Thomas and Hartert, Novitates Zoologicas, i,
p. 655. September, 1894 (Bunguran).

1895. Pteropus vatHpynis Thomas and Hartert, Novitates Zoologicce, 11,
p. 489. December, 1895 (Bunguran).

Six skins from Bunguran.

.? PTEROPUS HYPOMELANUS Temminck.

1894. Pteropus hyponielatius Thomas and Hartert, Novitates Zoologicae,

I, p. 655. September, 1894 (Sirhassen).

1895. Pteropus hypomelayius Thomas and Hartert, Novitates Zoologicae,

II, p. 489. December, 1895 (Pulo Pandak, Pulo Panjang and Pulo
Laut).

Eight (one in alcohol) from Sirhassen and seven (one in alcohol)
Pulo Laut. It is highly probable that these specimens represent a
species distinct from the true Pteropus hypo7nelanus of Ternate.

^ Proc. Washington Acad. Sci., 11, p. 234. August 20, 1900.
2 Free. Acad. Nat. Sci. Philadelphia, 1898, p. 316. July, 1898.

138 MILLER

NYCTICEBUS TARDIGRADUS (Linnsus).

1894. Nycticebus tardigradus Thomas and Hartert, Novitates Zoologicae,

I, p. 655. September, 1894 (Bunguran).

1895. Nycticebus tardigradus Thomas and Hartert, Novitates Zoologicse,

II, p. 489 (Bunguran).

One specimen from Bunguran.

MACACUS 'CYNOMOLGUS' Auct.

1894. Macacus cynomolgus Thomas and Hartert, Novitates Zoologicae, i,
p. 654. September, 1894 (Bunguran).

1895. Macactis cyiiotnolgus Thomas and Hartert, Novitates Zoologicae,
II, p. 489. December, 1895 (Bunguran).

A specimen from each of the following islands : Sirhassen, Pulo
Lingung and Pulo Laut.

SEMNOPITHECUS CRISTATUS (Raffles).
Two monkeys from Sirhassen appear to be referable to this species.

SEMNOPITHECUS NATUN^ Thomas and Hartert.

1894. Semnopithecus natuncB Thomas and Hartert, Novitates Zoologicae,

I, p. 652. September, 1894 (Bunguran).

1895. Semnopithectts natuncc Thomas and Hartert, Novitates Zoologicae,

II, p. 489. (Bunguran.)

Ten specimens from Bunguran.

PROCEEDINQS

OF THE

WASHINGTON ACADEMY OF SCIENCES

Vol. Ill, pp. 139-147. April 15, 1901.

RESULTS OF THE BRANNER-AGASSIZ
EXPEDITION TO BRAZIL.

V.

MOLLUSKS FROM THE VICINITY OF
PERNAMBUCO.

By William Healey Dall.
Honorary Curator, Department of Mollusks, U. S. National Museum.

During Dr. J. C. Branner's expedition for the study of the
coast reefs of Northeastern Brazil collections of various inver-
tebrates were made. As this region is seldom visited by natu-
ralists and is important in the study of the geographical distri-
bution of the mollusks of the Western Atlantic, Dr. Branner
requested me to report on that part of the collection.

Both Dr. Pilsbry and myself in listing collections of shells
from the Brazilian and Uruguayan coasts have been struck
with the predominant Antillean element among the mollusks,
and it seems as if the present distribution of littoral species
must have been largely brought about before the discharge of
the Amazonian watershed attained its present volume, since it is
difficult to imagine exclusively shore species passing the barrier
of several hundred miles of fresh water which intervenes be-
tween the present northern and southern strips of coast un-
affected by this mighty flood. The present collection confirms
the view previously taken of the very large Antillean element
in the fauna. Ninety-one species in all were obtained ; deduct-

Proc. Wash. Acad. Sci., April, igoi. (139)

140

DALL

ing from them four confined to land or fresh water, there re-
main eighty-seven, of which thirteen alone are peculiar to the
eastern coast of South America south of the West Indies, and
seventy-four are common Antillean shells. Two appeared to
be undescribed, and one, a very young Octopus, was too imma-
ture for identification. This would leave eight-ninths of the
species in common with the fauna of the Antilles. A large
proportion of these extend at least as far southward as Monte-
video, and the southern limit of the Antillean fauna may be re-
garded as situated in that vicinity.

Most of Dr. Branner's shells were obtained from the vicinity
of Pernambuco, collections were made from the existing coral
reefs, the limestone reefs of doubtful age which exist along
this coast, and to some extent from the sand beaches behind the
reefs. The collection was not large enough to indicate whether
the fauna of the stone reefs and that of the coral reefs differ,
but in all probability they do not.

The localities from which mollusks were obtained are as fol-
lows :

Pernambuco and Tacuara near by.

Maceio ; on granite bowlders, sand beach and reefs, latitude
9° 40' S.

Boa Viagem ; stone reef.

Cotonello Bay.

Managuas and Mangosoules on the Rio Parahyba do Norte,
Cabedello ; mangrove swamps.

Mamanguape; stone reef, latitude 6° 58' S.

Rio Goyanna ; stone reef off the mouth of the river, latitude
6°33'S.

Among the local species Valuta hehrcea and Ttirbinella
ovoidea appeared to be common, and Fasciolaria aurantiaca
abundant. No peculiarity of texture or aspect which could be
regarded as general differentiates the Antillean species of
Brazil from the individuals of the same species found in the
West Indies, the only approximation to such a feature being the
gray color of the specimens of Trivia pedictilus which I have
not observed in any of our very numerous series of that species
from the West Indies and Florida.

MOLLUSKS FROM THE VICINITY OF PERNAMBUCO I4I

LIST OF THE SPECIES OBTAINED.

Ostrea puelchana d'Orbigny.

Mangosoules, Managuas and Macelo.
Ostrea equestris Say.

Managuas ; attached to the shells of Cerithium algicola.
Spondylus echinatus Marty n.

Mamanguape ; a single young valve.
Lima inflata Lamarck.

Maceio.
Margaritifera radiata Leach.

Goyanna ; a single young valve.
Mytilus exustus Linne.

Maceio ; young shell from oysters.
Mytilus exiguus Dunker.

Mamanguape ; young shells.
Modiolus (Gregariella) coralliophagus Gmelin.

A single specimen at Cotonello Bay and another from a rock reef at
Goyanna. This form is remarkable for the series of four or five sharp
conical teeth on the hinge on each side of the largely internal ligament.
These teeth are below the line of the marginal serrations. The exterior
of the shell is covered with a concretionary layer which completely
hides the posterior part of the valves and must act as a sort of oper-
culum closing the entrance of its burrow against enemies and protect-
ing the mollusk. This layer, like that on Diberus^ must be partly due
to the animal itself.
Lithophaga (Diberus) antillarum d'Orbigny.

Mamanguape, Maceio, Goyanna and Pernambuco ; in coral.
Lithophaga nigra d'Orbigny.

Boa Viagem ; one specimen.
Area umbonata Lamarck.

Common on the reefs at Goyanna, Mamanguape, Pernambuco,
Maceio and Boa Viagem.
Area Adamsi (Shuttleworth) Smith.

Maceio, Mamanguape and Goyanna.
Area (Seapharca) Deshayesii Hanley.

Pernambuco ; one valve.
Area (Cunearea) brasiliana Lamarck.

Mamanguape ; one valve.
Phaeoides pectinatus Gmelin.

Mangosoules.

142

DALL

Codakia orbicularis Linne.

Maceio.
Divaricella quadrisulcata d'Orbigny.

Tacuara, Pernambuco.
Cardium muricatum Linne.

Pernambuco.
Venus (Chione) cancellata Linne.

Goyanna.
Venus (Anomalocardia) flexuosa Linne.

Mangosoules.
Tivela mactroides Born.

Mamanguape.
Petricola typica Jonas.

Maceio, Pernambuco and Boa Viagem ; in coral.
Sanguinolaria sanguinolenta Gmelin.

Mamanguape.
Tellina lineata Turton.

Mangosoules and Maceio.
Semele proficua Pulteney.

Managuas.
Mactrella alata Spengler.

Mamanguape.
Mulinia Branneri Dall, n. sp.

Mamanguape.
Gastxochaena ovata Sowerby.

Goyanna ; burrowing in stone.
Martesia clavata Lamarck.

Goyanna.
Discinisca antillarum d'Orbigny.

Goyanna ; on the reef.
Strophocheilus pudicus Miiller.

Mamanguape.
Oxystyla pulchella Spix.

Pernambuco.
Planorbis helopilus d'Orbigny.

Pernambuco.
Bulla striata Bruguiere.

Mangosoules, Managuas and Maceio.
Micromelo undata Biuguiere.

Goyanna ; on the reef.
Terebra cinerea Born.

Goyanna, Maceio, and Boa Viagem.

MOLLUSKS FROM THE VICINITY OF PERNAMBUCO 1 43

Drillia Greeleyi Dall, n. sp.

Goyanna ; on the reef.
Voluta hebraea Linne.

Pernambuco and Maceio ; common.
Turbinella ovoidea Kiener.

Pernambuco, Managuas and Maceio.
Fasciolaria aurantiaca Lamarck.

Maceio, Pernambuco and Managuas.
Melongena mono Linne.

Mangosoules, Managuas and Maceio.
Latinis (Leucozonia) cingulifera Lamarck.

Pernambuco, Maceio and Goyanna ; common.
Latirus (Leucozonia) ocellata Lamarck.

Pernambuco, Maceio and Goyanna ; common.
Tritonidea auritula Link.

Pernambuco, Maceio, Mamanguape and Goyanna.
Pisania janeirensis d'Orbigny.

Maceio ; on the reef.
Nassa vibex Say.

Rio Parahyba and Boa Viagem.
Anachis lyrata Sowerby.

Pernambuco, Boa Viagem and Cotonello Bay.
Columbella mercatoria Lamarck.

Maceio, and on the reef at Boa Viagem.
Murex brevifrons Lamarck.

Mangosoules and Managuas ; young shells.
Purpura deltoidea Lamarck.

Pernambuco and Maceio.
Purpura (var. ?) trinitatensis Guppy.

Mamanguape, Mangosoules, Managuas, Maceio, Goyanna and Per-
nambuco ; common.
Coralliophila galea Gmelin.

Maceio ; very variable.
Janthina communis Lamarck.

Mamanguape.
Distortrix reticulata Link.

Mangosoules.
Gyrineum ponderosum Reeve.

Maceio. This may prove but one of the varieties of Ranella affinis
Broderip, but the subdivisions by sculpture are convenient.
Lampusia pilearis Lamarck.

Goyanna.

144

DALL

Lampusia chlorostoma Lamarck.

Maceio ; young, on the reef.
Cassis tuberosa Linne.

Maceio.
Cassis flammea Linne.

Maceio.
Cassis inflata Shaw.

Mamanguape.
Dolium perdix Lamarck.

Maceio.
Dolium olearium Lamarck.

Maceio.
Cypraea exanthema Linne.

Maceio and Mamanguape.
Cypraea spurca Linne.

Mamanguape.
Trivia pediculus Linne.

Maceio and Boa Viagem. A dark gi-ay form with the usual dark
spots, but no suggestion of pink coloration.
Cerithium algicola C. B. Adams.

Mangosoules, Managuas, Mamanguape, and Maceio ; common.
Cerithium thomasiae Sowerby.

Maceio and Goyanna. Probably a variety of algicola with de-
pauperate sculpture.
Melaraphe nebulosa Lamarck.

Rio Parahyba, Mangosoules and Managuas, on shrubs near the
water ; Maceio, on granite bowlders on the shore.
Melaraphe columellaris d'Orbigny.

Maceio ; on granite bowlders.
Melaraphe columellaris \sa-.Jlava Broderip.

Rio Parahyba.
Ampullaria zonata Spix.

Managuas, and Traicao on the Mamanguape River.
Amalthea antiquata Linne.

Maceio.
Amalthea subrufa Carpenter.

Boa Viagem.
Polynices mamillaris Linn6.

Mamanguape and Maceio.
Polynices porcellana d'Orbigny.

Maceio.

MOLLUSKS FROM THE VICINITY OF PERNAMBUCO I45

Natica canrena Lamarck.

Managuas.
Natica livida Pfeiffer.

Rio Parahyba, Mangosoules and Maceio ; on sand beaches.
Natica marochiensis Gmelin.

Mamanguape, Mangosoules and Managuas.
Acmaea onychina Gould.

Mamanguape and Goyanna.
Turbo Spenglerianus Gmelin.

Managuas.
Astralium latispina Philippi.

Mangosoules, Rio Parahyba and Managuas.
Astralium imbricatum Gmelin.

Mamanguape and Maceio ; common.
Astralium armatum Philippi.

Goyanna.
Omphalius viridulus Gmelin.

Mamanguape, Maceio, Goyanna and Pernambuco ; common.
Neritina virginea Lamarck.

Rio Parahyba, Managuas, Mangosoules, and Goyanna. The| shells
are usually smaller than those found in the Antilles but pass through
the same multitude of color variations.
Subemarginula octoradiata Gmelin.

Maceio.
Fissurella rosea Gmelin.

Mamanguape, Maceio and Goyanna.
Fissuridea alternata Say.

Maceio and Goyanna.
Fissuridea Listeri d'Orbigny.

Maceio ; one young shell.
Ischnochiton squamulosus C. B, Adams.

Maceio ; one young specimen.
Octopus sp. indet.

One very yovmg specimen from the reefs.

DESCRIPTION OF THE NEW SPECIES.

Mulinia Branneri Dall.
Shell small, solid, elevated., short, subtriangular, white under a pale
brownish periostracum ; beaks high, full, subcarinate behind ; an ob-
scure impressed line in front proceeding from the beaks delimits a
Proc. Wash. Acad. Sci., April, 1901.

I/j.6 DALL

lanceolate area simulating a lunule ; a similar line on the opposite slope
within the carina marks a smaller area which may be compared to an
escutcheon ; the space between this line and the carina which extends
from the beaks to the lower posterior angle of the valves, has the
periostracum exceptionally rugose ; and on the carina itself this feature
is elevated into a low keel which disappears in drying ; the periostra-
cum is concentrically wrinkled and even lamellose toward the margin,
with irregular small radial wrinkling; the valves are polished beneath
the periostracum, practically smooth, and nearly equilateral, rounded
in front, arcuate below and bluntly pointed behind ; the hinge is normal
and strong, the pallial impression broad and indented by a narrow
linguiform sinus reaching nearly to the vertical of the beaks. Lon.
of shell 22.5, alt. 18.5, diam. 13.5 mm.

This neat little species is perhaps the same as the " Mulinia near
lateralis Say " mentioned in Pilsbry's list of the shells obtained by Dr.
Rush at Maldonado Bay, Uruguay (Nautilus, xi, p. 8, May, 1897).
It is larger and less rostrate than any specimens of AI. lateralis I have
seen. The nearest relative I have noticed in the American series of
this group is Alulinia portoricensis Shuttleworth, which is pretty close
to the present species, though more produced at the ends, and much
more rostrate than J/. Branneri. The latter is named in honor of
Dr. J. C. Branner of Stanford University to whom the formation of
the collection is due.

Drillia Greeleyi Dall.

Shell small, stout, solid, acute with eight or nine whorls ; nucleus
small, smooth, dark reddish purple with two whorls followed by a third
with strong arcuate ribs concave anteriorly, and which at the end of the
whorl are replaced by transverse nodules separated from the suture
behind by a revolving ridge ; the last whorl has twelve or thirteen
narrow transverse ribs, extending forward from the anal fasciole ; be-
hind the fasciole a stout ridge revolves a little in advance of the ap-
pressed edge of the whorl, the ridge is nodulous where it rides over the
ribs of the preceding whorl ; in front of the fasciole the ribs are crossed
by two adjacent and four rather distant stout revolving threads, beside
which there are four or five smaller threads on the canal, and in the
interstices and on the fasciole extremely fine sharp revolving threads ;
all the large threads form nodules where they cross the ribs and these
nodules are yellow, the shell elsewhere being dark reddish, nearly black ;
the last whorl is more than half the shell ; the aperture narrow and
dark. Lon. 10.5, max. diam. 4.5 mm.

MOLLUSKS FROM THE VICINITY OF PERNAMBUCO I47

This neat shell is not quite mature and its aperture in the adult state
cannot be described, but is undoubtedly normal. The species belongs
in the section Crassispira near such species as D. albinodata Reeve,
but differs in details of sculpture as described from all those with which
I have compared it. It is named in honor of Mr. A. W. Greeley who,
under the direction of Dr. Branner, collected it and nearly all the speci-
mens mentioned in this paper.

PROCEEDINGS

OF THE

WASHINGTON ACADEMY OF SCIENCES

Vol. Ill, pp. 149-155. April 15, 1901.

NEW BIRDS OF THE FAMILIES TANAGRID^
AND ICTERID^.

BY

Robert Ridgway.

Curator Division of Birds, U. S. National Museum.

This is one of a series of papers comprising advance de-
scriptions of new birds, prepared by the author in connection
with his forthcoming work on the Birds of North and Middle
America. Six papers of this series describing new forms in
the families Fringillidas and Corvidae, have appeared in the
Auk, Vols. XV (1898) to XVII (1900).

TANAGRID^.

Calospiza florida arcaei. Arce's Calliste.

Similar to Calospiza Jlorida Jlorida but smaller, the general color
more yellowish green ; adult male without a distinct, if any, occipital
patch of yellow.

Veragua.

Type., no. 146386, U. S. National Museum, $ adult, Veragua;
E. Arce.

Piranga roseo-gularis cozumelae. Cozumel Tanager.

Similar to Piranga roseo-gularis roseo-gularis but red of pileum
duller, wing averaging shorter and tail longer.

Island of Cozumel, Yucatan.

Type., no. 102689, U. S. National Museum, $ ad., Cozumel I.,
Yucatan, Jan. 29, 1885 ; J. E. Benedict and T. Lee.

150 RIDGWAY

Ramphocelus dimidiatus isthmicus. McLeannan's Tanager.

Similar to Ra?nphocelus dimidiatus di?nidiatiis, but colors duller,
the abdomen brownish black or dark brown instead of deep black and
tail longer.

Isthmus of Panama.

Type, no. 16883, U. S. National Museum, $ ad., Frijole Station,
Panama R. R. ; J. McLeannan.

Phoenicothraupis salvini peninsularis. Yucatan Tanager.

Similar to Phoenicothraupis salvini insularis^ but darker and less
grayish, though much paler and grayer than P. salvi?zi salvifii.

Yucatan (mainland).

Type, no. 146607, U. S. National Museum, $ ad., Izalam, Yuca-
tan, 1880; Geo. F, Gaumer.

Phcenicothraupis salvini discolor. Escondido Tanager.

Similar to Phcenicothraupis salvini salvini, but adult male much
darker and less brightly colored, the back, etc., dull reddish brown, the
under parts of body strongly tinged with grayish. (Nearly intermedi-
ate in coloration between P. salvini salvini and P. fuscicauda, but
female and young not at all like those of the latter.)

Southern Honduras to eastern Nicaragua.

Type, no. 126942, U. S. National Museum, $ ad., Rio Escondido,
eastern Nicaragua, May 12, 1892; Chas. W. Richmond.

Chlorospingus sumichrasti. Sumichrast's Chlorospingus.

Similar to C. ophthabnicus, but pileum lighter and clearer brown,
back more brownish olive-green, and chest paler and duller yellowish
olive.

Mountains of Vera Cruz, southeastern Mexico.

Type, no. 3751 1, U. S. National Museum, Mt. Azul, near Orizaba,
Vera Cruz, October, 1864, Professor F. Sumichrast.

Genus Iridophanes. (Type, Dacnis pulcherrima Sclater.)
Similar to Tatiagrella, Swainson, but nasal fossae mostly unfeath-
ered, bill more slender (though relatively broader basally), and four
outermost primaries (sixth to ninth) more nearly equal in length.

Although this form has usually been considered a member of a
Coerebine genus {Dacnis) I am inclined to believe that its relationships
are in reality Tanagrine, as has already been suggested by DuBus, ac-
cording to Dr. Sclater.^

^Cat. Birds Brit. Mus., Vol. xi, 1886, p. 25.

BIRDS OF THE FAMILIES TANAGRID^ AND ICTERID^ I5I

ICTERIDiE.

Zarhynchus wagleri mexicanus. Mexican Oropendola,

Similar to Zarhyfichus wagleri wagleri^ but slightly smaller, es-
pecially the bill, with frontal shield narrower and less arched ; chest-
nut of head and neck slightly, that of rump and flanks decidedly,
darker ; black of breast and abdomen more restricted, duller, less
glossy.

Southern Mexico to Guatemala; British Honduras? Honduras?

Type, no. 14456 1, U. S. National Museum (Biol. Surv. Coll. no.
1755), S ad., Motzorongo, Vera Cruz, Mexico, Feb. 26, 1894 ; Nelson
and Goldman.

Holoquiscalus martinicensis. Martinique Grackle.

Similar to Holoquiscalus inflexirostris, of Santa Lucia, but
smaller, and with stouter bill ; adult female with head and under parts
paler, the chin and throat nearly white. (The adult female similar
in coloration to that of Holoquiscalus guadeloupensis, but smaller
and with more slender bill.)

Island of Martinique, Lesser Antilles.

Type, no. 75156, U. S. National Museum, ^ad., Martinique; F.
A. Ober.

The forms belonging to this group, as I have been able to make
them out, are as follows :

(i) Holoquiscalus gundlachii (Cassin) ; Cuba.

(2) Holoquiscalus jamaicensis (Daudin) ; Jamaica.

(3) Holoquiscalus niger (Boddaert) ; Haiti.

(4) Holoquiscalus brachypterus (Cassin) ; Puerto Rico.

(5) Holoquiscalus caymanensis (Cory) ; Grand Cayman.

(6) Holoquiscalus luminosus (Lawrence) ; Grenada.

(7) Holoquiscalus inflexirostris (Swainson) ; Santa Lucia.

(8) Holoquiscalus guadeloupensis (Lawrence) ; Guadeloupe.

(9) Holoquiscalus martinicensis (Ridgway) ; Martinique.

(10) Holoquiscalus insularis (Richmond) ; Margarita I., Venezuela.

(11) Holoquiscalus lugubris (Swainson); Trinidad, Tobago, and
adjacent mainland.

(12) Holoquiscalus rectirostris (Cassin) ; (habitat unknown).

(13) Holoquiscalus fortirostris (Lawrence) ; Barbados.

Scaphidunis major nelsoni. Sonoran Boat-tail.'

'^ Scaphidurus Swainsox, Philos. Mag., new series, i, 1827, p. 437. Type, S
palustris Swainson .

152 RIDGWAY

Similar to Scaphidurus major graysoni but slightly smaller, with
slightly longer bill ; adult female much paler, the under parts (in winter
plumage) light wood brown, passing into brownish cream-buff on throat
and chin, into grayish brown on flanks, thighs and anal region, the
under tail-covei'ts dark grayish brown, pileum and hindneck light
sepia-brown, the scapulars and interscapulars margined with the same.

Coast district of Sonora, northwestern Mexico.

Type, no. 164596, U. S. National Museum (Biol. Surv. Coll., no.
6140), 9 ad., Alamos, Sonora, Jan. 37, 1S99; E. A. Goldman.

The known forms belonging to this genus are, according to my
recent revision of the group, as follows :

(i) Scaphidurus major major (Vieillot) ; South Atlantic and Gulf
Coast United States.

(2) Scaphidurus major macrourus (Swainson) ; southern Texas
to northern Columbia.

(3) Scaphidurus major obscurus (Nelson) ; southwestern Mexico.

(4) Scaphidurus major graysoni (Sclater) ; western Mexico.

(5) Scaphidurus major nelsoni Ridgway ; northwestern Mexico.

(6) Scaphidurus major assimilis (Sclater) ; central Colombia.

(7) Scaphidurus tenuirostris (Swainson) ; central Mexico.

(8) Scaphidurus nicaraguensis (Salvin & Godman) ; Nicaragua.

Icterus cucullatus sennetti. Sennett's Oriole.

Similar to Icterus cucullatus czicullatus (Swainson) but paler in
color ; adult male less decidedly orange, the color of pileum, chest,
etc., deep cadmium yellow, never cadmium orange; adult female
much paler than that of /. c. cticullatus, the yellow of under parts
dull or pale gamboge, instead of saffron or ochreous, the back and
scapulars lighter grayish.

Lower Rio Grande Valley, in Texas and Tamaulipas ; south in
winter to Morelos.

Type, no. 73654, U. S. National Museum, $ ad., Brownsville,
Texas, April 3, 1877; Geo. B. Sennett.

Icterus gularis tamaulipensis. Alta Mira Oriole.

Similar to Icteriis gularis gularis (Wagler) but decidedly smaller
and coloration more intense, the orange-yellow more decidedly orange
(usually rich cadmium orange) ; black at anterior extremity of malar
region broader ; bill shorter and deeper at base. Similar to Icterus
gularis yucatane?tsis, Berlepsch, but larger, usually less intense in
color, and with the maxilla relatively much narrower.

BIRDS OF THE FAMILIES TANAGRID^ AND ICTERID^ I53

Eastern Mexico, in states of Tamaulipas, Vera Cruz, San Luis
Potosi, and Puebla.

Type^ no. 135 168, U. S. National Museum, $ ad., Alta Mira,
Tamaulipas, E. Mexico, Oct. 17, 1S94; F. B. Armstrong.

Icterus mesomelas taczanowskii. Taczanowski's Oriole.

Similar to Icterus inesomelas mesomelas (Wagler), but slightly
smaller and with black at base of inner webs of lateral rectrices more
restricted, less intense, and less sharply defined.

Western Ecuador; Central Peru? Geographic range widely
separated from that of /. m. mesomelas by the intervening much
larger black-winged Central American form. Icterus meso?nelas
salvinii (Cassin).

lype^ no. 101265, U. S. National Museum, adult (^ ?), Guaya-
quil, Ecuador, 1884; Dr. Wm. H. Jones, U. S. N.

Agelaius phoeniceus fortis. Thick-billed Red-wing.

Similar to Agelaius phoeniceus phoeniceus^ but decidedly larger,
with bill relatively much shorter and thicker ; adult females, adult
male in winter, and immature males similar in coloration to the same
oi A. p. sonoriensis, but distinguished by very different measurements.

Breeding range unknown, but evidently somewhere in the central
portion of North America ; during migrations occurring in Manitoba,
Minnesota, Nebraska, Indian Territory, Illinois (i^arely or casually),
and westward to and including the Rocky Mountains and southward
to Arizona, noithern Chihuahua, New Mexico, and western Texas.

Type^ no. 88093, U. S. National Museum, 9 ad,, Omaha, Nebraska,
March 9, 1878; Dr. R. W. Shufeldt.

Agelaius phoeniceus neutralis. San Diego Red-wing.

Similar to A. p. sonoriensis but smaller ; adult female much darker,
with streaks less strongly contrasted above, those on under paits rather
broader and grayer, the upper parts with little if any of rusty, even in
winter.

Great Basin district of United States northward to eastern British
Columbia, southwestward to southern California (San Diego district)
and noithern Lower California.

Type, no. 134297, U. S. National Museum, 9 ad., Jacumba, San
Diego Co., California, May 36, 1894; Dr. Edgar A. Mearns, U. S. A.

Agelaius phoeniceus caurinus. Northwestern Red-wing.

Similar to A. p. phcvniceus but wings and bill longer, the latter more
slender ; adult male with buff of middle wing-coverts deeper (deep

154 RIDGWAY

ochraceous-buff or ochraceous in winter) ; adult female more heavily
streaked with black beneath, and, in winter plumage, with upper parts
much more conspicuously marked with rusty.

Northwest coast district, from British Columbia (Vancouver Island
and coast of mainland) south through western Washington and Ore-
gon to northern California (Mendocino County, May 3o) .

Type, no. 153275, U. S. National Museum, 9 ad.. Cedar Hill, Van-
couver Island, April 19, 1895 ; John Fannin.

The North American Agelaii, including those of Alexico and the
West Indies, as they appear from my recent revision of the group, are
as follows :

(i) Agelaius tricolor (i\udubon) ; California and Oregon, west of
Sierra Nevada and Cascade Mts.

(2) Agelaius gubernator gubernator (Wagler) ; southwestern part
of Mexican plateau.

(3) Agelaius gubernator californicus, Nelson. ; California (coast
district) and western Oregon.

(4) Agelaius gubernator grandis (Nelson) ; southeastern part of
Mexican plateau.

(5) Agelaius phoeniceus phocniceus (Linnaius) ; eastern North
America, to base of Rocky Mts.

(6) Agelaius phocniceus floridanus, Maynard ; Florida and Gulf
coast to Galveston, Texas.

(7) Agelaius phoeniceus bryanti, Ridgway ; Bahamas and south-
eastern Florida.

(8) Agelaius phccniceus richmondi. Nelson; southern Texas south
through eastern Mexico to Nicaragua.

(9) Agelaius phoeniceus sonoriensis, Ridgway ; southern Arizona
and Colorado Valley in California, south to Tepic, western Mexico.

(10) Agelaius phoeniceus neutralis, Ridgway ; west slope of Rocky
Mts., to California and northern Lower California.

(ii) Agelaius phoenicius caurinus, Ridgway; northwest coast, in
Washington and British Columbia ; northern California in winter.

(12) Agelaius phoenicius foitis, Ridgway; Rocky Mts., Great
Plains, etc., during migration; breeding range unknown.

(13) Agelaius assimilis, Gundlach ; Cuba.

(14) Agelaius humeralis (Vigors) ; Cuba.

(15) Agelaius xanthomus, Sclater ; Puerto Rico

BIRDS OF THE FAMILIES TANAGRID^ AND ICTERID^ I55

Genus Pseudagelaeus. (Type, Agelceus imthurni Sclater.)
Allied to Scaphidurus, but bill more slender, more attenuated and
less sti'ongly decurved terminally ; tail less graduated, the graduation
less than one fourth the length of middle rectrices (instead of between
one fourth and one third as much) ; lateral toes longer, their claws
reaching to base of middle claw.

Type^ Agelceus imthurni^ Sclatek. (Monotypic.)

Genus Xanthopsar. (Type, Oriolus Jlavus Gmelin.)
Similar to Agelaius^ but bill relatively much longer and more slen-
der, equaling or exceeding the head in length ; ninth primary longest,
or equal to the longest (the ninth, eighth and seventh about equal in
length) ; tarsus relatively shorter (but little longer than culmen and but
little more than one fourth as long as wing) ; middle toe, with claw,
longer than tarsus ; claw of inner toe falling short of base of middle
claw, the outer toe still shorter ; hallux decidedly shorter than lateral
toes, its claw very long (exceeding the digit in length) and narrow, the
anterior claws relatively very small.

PROCEKDINQS

OF THE

WASHINGTON ACADEMY OF SCIENCES

Vol. Ill, pp. 157-216. PLS. xiv-xxvi. May ii, 1901

PAPERS FROM THE HARRIMAN ALASKA
EXPEDITION.

XXI.

THE HYDROIDS.

By C. C. Nutting.
University of Iowa.

CONTENTS :

Introduction 157

Geographic distribution 158

Systematic discussion 163

INTRODUCTION.

The collection of Hydroida secured by the Harriman Ex-
pedition is of exceptional interest, and proves to be one of the
most important and most extensive collections of these beautiful
forms of marine life thus far made in Alaskan waters. Our pre-
vious knowledge of the Hydroid fauna of this region rested almost
exclusively on the collection made by Dr. W. H. Dall and his
associates during the years 1871-1874 and reported on by Dr. S.
F. Clark. ^ The number of species listed in Clark's report is 41,
in which was included Co^finia arcta, now known to be merely
the gonosome of JLafoea. Of these 40 species, 15 are well known
British forms, and only one was then known from the Atlantic
coast of the United States. The remaining 24 species were new.

^ Report on the Hjdroids collected on the coast of Alaska and the Aleutian
Islands by W. H. Dall, U. S. Coast Survey, and party, from 1871 to 1874.
inclusive. Proc. Acad. Nat. Sciences, Philadelphia, 1876.

Proc. Wash. Acad. Sci., May, 1901. (i57)

1^8 NUTTING

In 1878 C. Mereschkowsky ^ added a single species to the
Alaskan fauna, bringing the total up to 41.

No other additions were made until 1899, when I added eight,
of which three were well known British species and five were
new.^ That made a total of 49 species reported prior to the
Harriman Expedition.

The collection here treated of comprises 53 species, 24 of
which had been previously reported. Of the remainder, 9 were
previously recorded from other localities, and 20 are new.
Thus the Harriman Expedition has added about 60 percent to
the number of species hitherto known from Alaskan waters.
More than half of the species secured are new to Alaska and
nearly 40 percent are new to science.

The whole number of species of hydroids now known from
Alaska is 78. Considering the small amount of collecting that
has been done in that region, compared with the extensive ex-
plorations of the coasts of Europe and the Atlantic coast of the
United States, one may confidently expect that the waters of
the far Northwest will prove to be very rich in hydroid life.

GEOGRAPHIC DISTRIBUTION.

A table is here given to indicate, first, the localities at which
each species was collected by the Harriman Expedition, and
second, the extent to which Hydroids have been distributed south-
ward along meridional lines from what appears to have been
a polar center of distribution. No attempt has been made to
represent the complete distribution of the species.

A glance at the part of the table showing the distribution as
represented in the collection secured by the expedition, shows
an apparent poverty of the Hydroid fauna of the western, as
compared with the eastern, portion of the territory explored.
For convenience in such comparison the stations are arranged
consecutively from east to west. The largest series were ob-
tained at Berg Inlet in Glacier Bay ; Yakutat Bay ; and at Orca
in Prince William Sound. These localities are all in deep bays,
sheltered from storms and surrounded by rocky shores. On

1 New Hjdroida from Ochotsk, Kamtschatka, and other parts of the North
Pacific Ocean. Annals and Mag. Nat. Hist., Dec, 1878.

^Hydroida from Alaska and Puget Sound. Proc. U. S. Nat. Museum, Vol.
XXI. (No. 1171.)

GEOGRAPHIC DISTRIBUTION OF THE HYDROIDS COLLECTED BY
THE HARRIMAN EXPEDITION.

Name.*

Distribution of Specimens in the Harri-
man Collection.

General Distribution.

i

a
a
1— I

■3

i

s

3

><

6

i

•d
1

0.
0

Ph

1
a

a

1

X

1

4)
d.
0

1

a

0

■s.

1

03

4-*

1

0 0
||

"a
0

<

Syncoryne eximia.
*Coryne brachiata.

+

+

+

+

+

+

+

+

+

+
+

Eudendrium vaginatum.

+

+

+

+

+

^Campanularia ritteri.

Campanularia denticulata.

Campanularia verticillata.

Campanularia lineata.

Campanularia speciosa.

Campanularia urceolata.
*Campanularia reduplicata.
*Campanularia regia.

Clytia caliculata.

Clytia compressa.

Obelia plicata.

Obelia dichotoma.

+

+

+

+

+

+

+

+

+
+

+

+

+

+
+

+

+

+

+

+

+

+

+

+

+
+
+

+

+

+

+

+ "

+

+

+

*Obelia dubia.

+

+

Hebella pocillum.
*GonothyrEea inornata.
*Campanulina rugosa.

Calycella syringa.

Lafcea dumosa.

+

+

+

+

+

+

+

+

+

+

+

+
+
+
+

+
+
+
+

+

+
+

+

Lafoea gracillima.

+
+

+
+

+
+

+

*Lafoea adhserens.

+

*Grammaria immersa.

+

Filellum serpens
Halecium halecinum.

+

+

+
+
+

+
+
+

+
+
+

+

+

+

+

-j-

*Halecium reversum.

+

+

*Halecium robustum

+
+

4-

+

*IIalecium speciosum.

Sertularella tricuspidata.

Sertularella polyzonias.
*.Sertularella saccata

+
+

+

+

+

•f

+

+
+
+
+

+
+

+
+

+

+

Thuiaria argentea.
Thuiaria similis.

+

+

+
+
+

+

Thuiaria variabilis.

+

Thuiaria cupressoides.
*Thuiaria coei

+

+

+

+

Thuiaria fabricii.

+
+

+

+

Thuiaria turgida.
Thuiaria gigantea.
Thuiaria thuiarioides.

+

+'

+
+

+

4-

+

+

+

+

+

+

+

Plumularia lagenifera.
Plumularia palmeri.
Totals.

+

+

+
+

20

lO

4

12

15

13

8

3

I

4

19

18

14

25

1 Species marked by a * are new.

l6o NUTTING

account of the presence of perpetual ice in the form of glacier
fronts and bergs, the water must be very cold the year around.
Such a combination of conditions is particularly favorable to
Hydroid life and accounts for the remarkably rich collections
made at these places and also for the presence of so many arc-
tic species. In this connection it is interesting to note the fol-
lowing paragraph written about twenty five years ago by Dr.
Dall : "The material derived from the northwestern coasts of
America, from Cook's Inlet south and east, indicates a series of
Arctic colonies in favored localities, the future exploration of
which offers a labor of the highest interest. These colonies are
situated where the depth of water, the drippings of glaciers, and
the high and adjacent shores of the Great Archipelago combine
to reduce the temperature of the water below its apparently
normal isotherm. Cook's Inlet affords one of them, one exists
in the Gulf of Georgia, and others only await further explora-
tion."^ It should be noted, however, that nearly all of the
Arctic species are well known forms belonging to the ' Holarctic
Province ' of authors, and that these species are of practically
continuous distribution on all coasts in northern regions so far
as explored.

In the same paper. Dr. Dall divides the coasts of America
from Monterey, California, north and west, into three faunal
areas, as follows : («) the Oregoniaii, extending from Monterey
to the Shumagin Islands ; {b) the Aleutian, extending from the
Shumagin Islands to the end of the Aleutian chain, and north-
ward to the winter line of floating ice in Bering Sea ; (c) the
Arctic, limited on the shore line to the winter line of floating
ice and passing southward indefinitely in deep water.

This paper deals chiefly with what Dr. Dall would call the
Oregonian Fauna, only seven species having been secured to
the westward of the Shumagin Islands. Of these seven species
five are also found in his Oregonian Fauna, and the other two
are new and known, thus far, from only one locality.

Dr. Clark, in reporting on the collection made by Dr. Dall,
enumerates 25 species that occur west and north of the Shu-
magin Islands. Of these 25 species we now know that sixteen

^ Proc. Acad. Nat. Sciences, Philadelphia, p. 206, 1S76.

THE HYDROIDS l6l

also occur to the eastward of the Islands, while five have not
been reported from any locality other than the ones where they
were originally discovered. Our present knowledge therefore
does not support the validity of Dr. Dall's division of faun£e at
the Shumagin Islands. It rather indicates a continuity of fauna
from southern Alaska to the end of the Aleutian chain. Hydroid
life appears to decrease as we go westward, but this may be only
apparent and due to the more extensive exploration of the shores
east of the Aleutian Islands.

Dr. Dall extends his Oregonian fauna down to Monterey,
California. Reasoning again merely from the known distribu-
tion of hydroids, it would seem that Puget Sound is a natural
region of demarcation between faunae, although the region from
Puget Sound to San Francisco has been very little explored.
In 1876 Dr. Clark published a paper on ' The Hydroids of the
Pacific Coast of the United States south of Vancouver Island,' ^
in which he gives a list of twenty-four species ; of these only
two, Lafoea dimiosa and Sertidaria argeiitca, have as yet been
reported north of Puget Sound. The same author, in reporting
on Dr. Dall's collections from Alaska, notes as one of the main
points of interest, the " small number of species that are com-
mon to the Alaskan coast and the western shores of the United
States from Vancouver Island southward.^ In 1899 the present
writer published a paper on ' Hydroida from Alaska and Puget
Sound ' ^ in which it appears that out of twenty-two species from
Puget Sound, only four have been reported farther south, while
fifteen are now known to occur in Alaska. In the same year
Mr. G. N. Calkins published a paper entitled ' Some Hydroids
from Puget Sound,' ^ in which some thirty species are noted,
only two of which are known to occur south of Puget Sound.

From this study of the distribution of the Hydroids of the

northwest coast of America, therefore, I am strongly persuaded

that the region south of Puget Sound constitutes one distinct

faunal area, and that the region from Puget Sound north and

west to the end of the Aleutian chain constitutes another un-

1 Trans. Conn. Acad. Sci., Vol. iii, pp. 250-251, 1S76.
^Proc. Acad. Nat. Sciences, Philadelphia, p. 212, 1S76.
3Proc. U. S. Nat. Museum, Vol. xxi, No. 1171, 1S99.
*Proc. Bost. Soc. Nat. Hist., Vol. xxviii. No. 13, 1899.
Proc. Wash. Acad. Sci., Ma}', 1901.

1 62 NUTTING

broken faunal area that might properly be designated as Alaskan.
From the number of arctic species included in this area it is not
improbable that it extends northward along the shores of Bering
Sea.

Dr. Clark agrees with Dr. Dall that there is a distinct faunal
difference between the region east of the Shumagin Islands and
that west of them. The material added since the publication of
his paper, however, seems to prove that this difference is only
apparent and due solely to lack of exploration.

The most important thing to be noted in that part of the table
devoted to general distribution is the Holarctic distribution of a
number of species. Of the eighteen species known to occur in
the Arctic region, no less than fifteen also occur on the European
coast, fourteen on the Atlantic coast of the United States, and
thirteen on the Pacific coast as far south as Puget Sound. An
examination of the table shows further, that the Hydroid fauna
of Alaska, as represented by the Harriman collection, includes
fifty-three species in all, of which eighteen are Arctic in fact,
having been secured in Arctic waters ; four others are in all
probability Arctic, being found both in European and American
waters ; four are, so far as is known, confined to the Alaskan
and Pacific coast south to Puget Sound ; twenty-five are thus far
known from Alaska only, and two are Californian. If we rec-
ognize the Alaskan faunal region as extending to Puget Sound,
and include those species actually known to be Arctic, together
with those in all probability Arctic, in a group which may justly be
called Arctic, the following significant analysis of the faunal rela-
tions of the collection may be made : Alaskan species, twenty-
nine ; Arctic species, twenty-two ; Californian species, two. This
shows that fifty-five percent of the hydroid fauna as a whole is
peculiar to Alaska, but that there has been a strong invasion from
the Arctic regions of the Holarctic species constituting about
forty-one percent of the collection, and that only two species,
or less than four percent, are Californian. If all the species
known to occur in Alaska were included in the computation the
result would be a larger percentage of Alaskan species, a cor-
responding decrease of the Arctic species, and the addition of
one or two Californian species.

THE HYDROIDS 163

SYSTEMATIC DISCUSSION.

The writer deems it unnecessary to attempt a complete syn-
onymy of the well known European species contained in the
Harriman collection and considers it sufficient to g\vQ,Ji?'st, and
in all cases, the original reference to the species ; second, all
obtainable references to the occurrence of the species on the
Pacific Coast of America, and, third, a reference to verify the
' General Distribution ' as given in the table just discussed. In
this latter case only one reference will be given to verify the
occurrence of a given species in each of the regions included
in the right hand portion of the table.

As to the classification employed in this report, it seems best,
on the whole, to pursue a conservative course, following pretty
closely the lines laid down by the able British naturalists,
Hincks and AUman. While it is true that the classification is
in an unsatisfactory state, the writer does not feel that a thor-
ough revision of the entire group of Hydroida should be at-
tempted here, and frankly confesses his conviction that recent
attempts in that direction have not been successful, though each
contains valuable suggestions. Levinsen, for example, has
made a notable contribution to our knowledge of the Campanu-
linidse in his able and careful exposition of the differences in
the opercula of various species, but his genera founded solely on
these structures appear to be artificial, as usually happens when
a single character is made the basis of classification.^ In his
terse characterization of the genera of Sertularidas this author
has been most fortunate, as well as in his masterly clearing up
of the mystery concerning the gonosome of Lafoea.

Schneider,^ also, has attempted to rearrange the Hydroida on
a logical basis. Instead of multiplying groups, as has been the
tendency of late, he has, in my opinion, gone far to the other
extreme, uniting families that almost any other student acquainted
with the group would regard as surely distinct. It seems un-
likely that he will be followed in uniting such groups as the
Tubularidas and Pennaridae in a single family, although one

' Meduser, Ctenopherer og Hjdroider fra Grdnlands Vestkvst, Copenhagen,
1893.

^Hjdropolypen von Rovigno, nebst ubersicht iiber das system der hjdro-
polypen in allgemeinen. Zool. Jahrb., Sjst. Abth., Vol. x, 1S97.

164 NUTTING

writer, Calkins/ has followed his classification quite closely, and
includes representatives of what would ordinarily be regarded as
at least eleven families in five families as defined by Schneider.
It is not likely that classifications will ever represent anything
but individual opinion, and it is probable that there will always
be two sets of extremists who on the one hand will be too ready
to multiply groups, and on the other will be too conservative to
recognize real progress. The Hydroida offer unusual difficul-
ties and consequently students of that group find agreement,
even along the most general lines, practically impossible. The
objective point of systematic discussion has been the attainment
of a system of classification by which genera could be distin-
guished by means of the trophosome alone. This end, although
in theory greatly to be desired, appears to be unattainable. The
writer has chased this will-o-the-wisp for years, and is ready to
abandon its pursuit as unprofitable. Abler men have had the
same experience, and it appears to be pretty well established
that in practice we must base generic distinctions on the gono-
some alone, although the judgment of students will inevitably
differ as to the extent to which this can be profitably done.
Botanists have encountered the same difficulty in their study of
the lower plants, such as the fungi, and have come to the same
conclusion. In neither case has nature been working for the
convenience of naturalists, and the fact should be accepted
without a bootless chase after the unattainable.

G TMNOBL A S TEA .
Hydroida in which well differentiated hydrothecce and gonangia are
not present. What might be called ' pseudo-hydrothecae ' are found in
some species as in Eudendriuin vaginatum (see description of that
species on pages 167—168).

Family CORTNID^M.

1 rophosome. — Hydranth with a terete body and proboscis and scat-
tered capitate tentacles only.

Gonosome. — Fixed sporosacs, or free medusEe with a very long manu-
brium, four marginal tentacles and four sense-bulbs with eye-spots.

'Some Hydroids from Puget Sound, Proc. Best. Soc. Nat. Hist., Vol.
xxvni, No. 13, 1899.

THE HYDROIDS 165

CORYNE.

Trophosome. — Characters of the family.

Gonosome. — Reproductive elements produced in fixed sporosacs
growing on the hydranth body.

CORYNE BRACHIATA sp. nov.
(Plate XIV, figs. I, 2.)

Trophosome. — Colony forming a dense tuft of irregularly branching
stems, sometimes attaining a height of about ^ inch. Stems and
branches profusely and regularly annulated throughout, fairly stout ex-
cept at the proximal ends where they taper gradually to their point of
origin ; distal ends of many of the branches bear a more or less regular
whorl, or radiating cluster, of annulated branchlets just below the
hydranth body, reminding one of the whorls of cirri around the stems
of the stalked crinoids. Hydranths large, with long, slender body and
proboscis and numerous (20-35) capitate tentacles arranged in a
scattered or sub-verticillate manner over nearly the whole surface.

Gonoso7?ie. — Gonophores very numerous, borne among the tentacles
on the hydranth bodies, globular in outline and showing no traces
of radial canals or other medusoid structures. The specimens
secured were females and the gonophores were packed full of develop-
ing ova.

Distribution. — All the specimens were secured in Yakutat Bay,
Alaska, by Dr. W. R. Coe of the Harriman Expedition.

This interesting species seems to be nearest to C. pusilla Gaertner, if
Allman has properly identified that species. It differs from other
members of the genus in the curious whorl of short branchlets which
bear neither hydranths nor gonophores and are situated a short distance
below the terminal hydranth of the stem or branch to which they are
attached. Another character not shown in the figures of this genus
published by Hincks and Allman, is the narrowing at the proximal
ends of the stems and branches. The specimens were found immei'sed
in sponge so far that only the hydranths extended above the surface of
the sponge.

SYNCORYNE.

Trophosome. — Characters the same as those given for the family.

Gonoso7ne. — Reproductive elements produced in free medusce with
a long manubrium and four marginal tentacles, each having a sense
bulb with an eye-spot at its base.

l66 NUTTING

SYNCORYNE EXIMIA Allman.
(Plate XIV, figs. 3, 4.)

Coryne eximia Allman, Annals and Mag. Nat. Hist,, 3d Series, Vol. iv, p.

141. Aug-., 1859.
Syncoryne eximia Allman, Annals and Mag. Nat. Hist., 3d Series, Vol. xiii,

p. 357. May, 1864.

This appears to be the first record of the occurrence of this species
in American waters. There are many specimens in the collection, but
all are from the same locality,

Distributio7i. — Juneau, Alaska (Harriman Expedition) ; Great
Britain (Allman and Hincks) ; Lofoten Islands, Norway (Sars).

Fam i ly BIMERID^^.
Trophoso7ne. — Hydranths with a conical or dome-shaped proboscis,
around the base of which is a whorl of filiform tentacles.
Gonosome. — Sexual products developed in fixed sporosacs.

GARVEIA.

Trofhosojue. — Colony branched ; perisarc conspicuous.

Gonosome. — Gonophores borne on distinct branchlets which have a

chitinous investment ending in a cup-like expansion just below the

gonophore.

GARVEIA NUTANS Wright.

Garveia nutans Wright, Edinburgh New Phil. Jour., p. 109. July, 1859.
Eudendrium bacciferum Allman, Annals and Mag. Nat. Hist., 3d Series,
Vol. IV, p, 52, July, 1859.

This is another species that has not before been reported from
American waters. The specimens were fragmentary, making the
identification somewhat uncertain, although I have little doubt of its
correctness,

Disiribzition. — Berg Inlet, Glacier Bay, Alaska, (Harriman Ex-
pedition,) Originally described from the British Coast.

GARVEIA ANNULATA sp. nov.
(Plate XV, figs. 1,2.)
Trophosome. — Colony attaining a height of i^ inches, consisting of
a number of closely aggregated and sparingly and irregularly branched
stems. Stems strongly and evenly annulated throughout, the perisarc
expanding distally into thin chitinous pseudo-hydrothecce which cover
the hydranth body nearly to the level of the tentacles. Hydranths with

THE HYDROIDS 167

a conical, or rather conoid, proboscis and about sixteen tentacles all of
which appear to be held more or less erect.

Gonosojne. — Gonophores borne either on the stem or hydrorhiza,
more frequently the latter, oval in shape, borne on pedicels enveloped
in a chitinous perisarc which ends in a slightly expanded collar a little
below the gonophore. The specimens collected were female and the
gonophoi^es were packed with apparently mature ova.

Color. — The label accompanying the specimens bore the following
statement : " Bright orange throughout, heads, stems and all."

Distribution. — Yakutat and Sitka, Alaska. Collected by the Har-
riman Expedition in considerable quantities.

This species can be sharply distinguished from its British relative
by the very distinct and beautiful annulation which covers the entire
stem and branches. It is much less extensively branched than the British
species, and the gonophores are more generally borne on the roots.

The structure that I have designated above as a ' pseudo-hydrotheca '
is of considerable morphological interest, for it may throw light on the
origin of the hydrotheca. The extension of the chitinous perisarc of
the stem over the body of the hydranth appears to be attached to the
latter. A true hydrotheca would be formed if the perisarc around the
hydranth body should become thicker and detached.

Family EUDENDRID^^.

Trophosonie. — Colony branching. Hydranths with a single whorl
of filiform tentacles and a trumpet-shaped or hemispherical proboscis
which is expanded distally and contracted proximally, thus being
sharply distinguished from the hydranth body.

Gonosome. — Reproductive elements developed in fixed sporosacs at-
tached to a usually more or less degenerated hydranth body below the
tentacles.

This family contains but one genus, Eudendriuni, which needs no
further definition.

, EUDENDRIUM VAGINATUM Allman.

(Plate XV, figs. 3-6.)

Eudendrium vaginatum Allman, Annals and Mag, Nat. Hist., Third Se-
ries, Vol. XI, p. 10, Jan., 1863.

As the gonosome of this species has not heretofore been described,
the following is inserted here :

Gonosofne. — Gonophores (female) in dense clusters around the
bodies of hydranths that are usually devoid of tentacles. Each gono-

l68 NUTTING

phore is borne on a pedicel which resembles those of Garveia, having
a distinct expanded collar a short distance below the hydranth.

DisU-ibtction. — Sitka Harbor and Yakutat, Alaska, abundant (Harri-
man Expedition) ; Shetland Island, Scotland (Allman).

This beautiful species beai's considerable resemblance to Euden-
drium anmdatum Allman, especially in its gonosome which Allman
describes as follows : " The gonophores ai^e grouped in clusters, con-
sisting of from eight to twenty egg-shaped bodies attached around the
axis of gonoblastidea, which are of moderate length."^

Were it not for a peculiar character of the trophosome, i. e., the ex-
panded pseudo-hydrotheca investing the body below the tentacles much
as in Garveia^ there might be some suspicion that E. vaghiatiun and
E. annulatum are synonyms, particularly in view of the fact that both
were described from the Shetland Islands.

It also seems not improbable that Eudendrhan pygmceuin Clark -
may be another synonym of E. vaginattan^ as Clark's description of
the gonosome agrees well with the gonosome described above. If
this be true, it is also likely that the dried stems described by Clark
from Santa Cruz, California, will be found to belong to this same
species.

Family TUBULARID^E.

Trophosome. — Hydranths large, with a basal whorl of filiform ten-
tacles and a distal set of closely crowded shorter filiform tentacles.

Gonosome. — Reproductive elements developed in sessile medusae
borne in clusters just above the basal tentacles and producing actinules
instead of planuliB.

TUBULARIA.

The only genus included in the family in the sense here used.

TUBULARIA HARRIMANI sp. nov.
(Plate XVI.)

Trophosoj77e. — Stem usually unbranched, attaining a height of ij4
inches, irregularly and sparingly annulated and increasing in size from
the proximal to the distal end, but more rapidly on the basal portion ;
a marked constriction some distance below the hydranth body ; stem
canaliculated between the constriction and the hydranth. Hydranth
with forty to fifty basal tentacles and about twenty in the distal set.

Gonosome. — Gonophores borne in about twelve very long and
densely crowded racemes, which are supported by long, tentacle-like

'Ann. and Mag. Nat. Hist., 3d Series, Vol. xiii, p. 83, Jan., 1S64.
^Proc. Acad. Nat. Sciences, Philadelphia, p. 232, 1S76.

THE HYDROIDS 1 69

pedicels arising above the proximal row of tentacles. Gonophores (fe-
male) with three to five long tentacular processes which are sometimes
half the length of the gonophore. The actinule at birth is without a
distal row of tentacles, and the gonophore has no indication of radial
canals.

Distribution. — Orca, Prince William Sound, -Alaska (W. E. Ritter).

Exclusive of this species there have now been described four species
of Tubiilaria from the Pacific Coast of North America ; ParypJia
microcephala (A. Agassiz),^ which differs from the present species in
having flattened, instead of round, tentacular processes to the gono-
phores ; Tubularia elegans Clark, ^ which has mere tubercles instead
of the filiform processes to the gonophores ; Tubularia borealis
Clark, ^ which differs in having laterally compressed processes to the
gonophores ; and Tubularia larynx Ellis and Solander * (repoited by
Gary S. Calkins), v^^hich differs from T. harri?nani in the number of
both proximal and distal sets of tentacles, as well as in several other
characters.

The species seems to be abundant at Orca, as numerous specimens
were found.

CAL TPTEROBLASTEA.

Hydroida in which hydrothecte are developed for the protection of
the hydranths and gonangia for the protection of the gonophores.

Family CAMPANULARIDyE.

Trophosome. — HydrotheCcB well developed, non-operculate, never
adnate nor immersed in the stem and always with a septum partially
dividing the hydrothecal cavity from the stem cavity. Hydranths
usually with conical proboscis and a single whorl of filiform tentacles.

Gonosotne. — Gonophores producing planulie or free medusae.

It would be hard to find two authorities who would agree as to the
genera of this exceedingly perplexing family. The arrangement here
adopted is substantially the same as that used by the writer in another
work now in press. ^ It is not offered as a final solution of the diflS-
culty, but as a convenience in discussing the group in the present
connection,

' North American Acalephae, p. 195.
^Transactions Conn. Acad. Sci., Vol. in, p. 253, 1876.
3Proc. Acad. Nat. Sci. Philadelphia, p. 231, 1876.
^Proc. Boston Soc. Nat. Hist., Vol. xxviii. No. 13, p. 335.
^Handbook of the Hjdroids of the Woods Hole Region. To be published by
the U. S. Fish Commission.

lyo

NUTTING

CLYTIA.

Irofhosome. — Stem not regularly branched. Hydrothecae with
toothed margins, or with excessively thick walls and with long
pedicels.

Gonosome. — Reproduction by means of free medusje.

CLYTIA CALICULATA (Hincks).

(Plate XVII, figs. I, 2.)

Campanularia caliculata Hincks, Annals and Mag. Nat. Hist., 2nd ser., Vol.
XI, p. 178, March, 1853. — Verrill, Preliminary check-list of Marine
Invertebrates of Atlantic Coast, etc., p. 16, 1879. — Marktanner-Turn-
ERETSCHER, Hydroidcn von Ost-Spitzbergen, Zool. Jahrb., Vol. viil, p.
406. 1895. — Calkins, Some Hydroids from Puget Sound, Proc. Boston
Soc. Nat. Hist., Vol. xxviii. No. 13, p. 351, 1899.

Some authors, as Levinsen, regard this species as identical with
C. Integra Macgillivray. The mode of reproduction is so different,
however, that the two would go into different genera in the classifica-
tion here adopted.

Distribution. — Yakutat, Alaska (Harriman Exped.) ; British Coast
(Hincks) ; Spitzbergen (Marktanner-Turneretscher) ; New England
Coast (Verrill) ; Puget Sound (Calkins).

CLYTIA COMPRESSA (Clark).
(Plate XVII, figs. 3, 4.)

Canjpamitarja compressa Clark, Proc. Acad. Nat. Sciences, Philadelphia,

p. 214, 1876.
Eticopella campanutaria (VoN Lendenfeld)? Uber Coelenteraten der Sudsee,

IV, Mitth. Zeitsch. Wiss. Zool., xxxviii, p. 497, 1883.

Distribution. — Orca, Alaska (Harriman Exped.) ; Shumagin
Islands (Clark). The figures given of this species well illustrate the
great variation in thickness of the hydrothecal walls. All the speci-
mens thus far discovered were found attached to Laminaria, over
which they creep in great profusion.

Von Lendenfeld makes his Eucopella campattularia the subject of
one of his masterly monographic papers and it appears to agree in
every particular with the species under discussion. If I am correct in
supposing the two species identical, the name Eucopella compantdaria
will become a synonym and a very exceptional distribution will be re-
corded for Clytia co?hpressa. It is interesting to note that von
Lendenfeld's species \vas also found growing on Laminaria.

The present writer does not agree with Calkins in his suggestion
that C. compressa is a synonym of C caliculata .

THE HYDROIDS I7I

CAMPANULARIA.

Trophosome. — Colony unbranched or regularly branched ; stem
simple or fascicled ; hydrothecae campanulate, never completely ses-
sile nor with operculum.

Gonosome. — Gonophores producing planulae without the interven-
tion of meduste,

CAMPANULARIA VERTICILLATA (Linn.)

Serhilaria verticillata Linn., Syst. Nat., loth ed., p. 811, 1758.
Campanularia verticillata Sars, Bidrag til Kundskaben om Norges Hydroider,

p. 46, 1873. — Verrill, Preliminary check-list of Marine Invertebrates

of Atlantic Coast, p. 16, 1879.

Distribution. — Kadiak, Alaska (Harriman Exped.) ; North Cape,
Norway (Sars) ; New England Coast (Verrill).

CAMPANULARIA DENTICULATA Clark.

Campanularia denticiilata Clark, Proc. Acad. Nat. Sciences, Philadelphia,
p. 213, 1876.

Distribution. — Orca, Alaska (Harriman Exped.); Port Etches,

Alaska (Clark).

CAMPANULARIA LINEATA Nutting.

Campanularia lineata Nutting, Hydroida from Alaska and Puget Sound,
Proc. U. S. Nat. Mus., Vol. xxi, p. 744, 1899.

Distribution. — Berg Inlet, Glacier Bay, Alaska (Harriman Ex-
ped.) ; Puget Sound (Nutting).

CAMPANULARIA RITTERI sp. nov. ,
(Plate XVII, fig. 5.)

Trophosome. — Colony usually consisting of unbranched pedicels
growing directly from a creeping rootstock. Pedicels long and
slender, usually witli a single distinct annulation at the distal end just
below the hydrotheca and about three less distinct ones at the proximal
end, the middle portion not being annulated. Hydrothecse cylindrical,
large, delicate in structure and with a perfectly even rim.

Go7ioso9ne. — Unknown.

Distribution. — Juneau, Alaska, 20 fathoms. Collected by Prof.
Wm. E. Ritter to whose efforts the fine series of Hydroids here dis-
cussed is so largely due, and for whom this species is named.

CAMPANULARIA SPECIOSA Clark.
(Plate xviii, fig. 3, Plate xix, fig. 3.)
Campanularia speciosa Clark, Proc. Acad. Nat. Sciences, Philadelphia, p.
214, 1876. — Levinsen, Meduser, Ctenophorer og Hydroider fra Gron-
lands Vestkyst, p. 25, 1893.

172

NUTTING

Camp aft ularia crenata Allman, Diagnoses of new Genera and Species of
Hydroida, Linnaean Society Journal, Zoology, Vol. xi, p. 258, 1876.

Distribution. — Orca, Alaska (Harriman Exped.) ; Yukon Harbor,
Big Koniuji, Shumagin Islands, Alaska (Clark) ; Japan Coast (All-
man) ; Greenland (Levinsen) .

CAMPANULARIA URCEOLATA Clark.
(Plate XVIII, fig. 2.)
Catnpanularia tirceolata Clark, Proc. Acad. Nat. Sciences, Philadelphia,
p. 215, 1876.

Distribution. — Yakutat Bay, Alaska (Harriman Exped.) ; Lituya
Bay, Alaska (Clark).

The specimens collected by the Harriman Expedition were growing
profusely over the stems and branches of Thuiaria costata in com-
pany with another parasitic species.

CAMPANULARIA REDUPLICATA sp. nov.

(Plate xviii, fig. I.)

Trophoso7ne. — Colony consisting of unbranched stems or pedicels
springing from a creeping rootstock. Pedicels one to three times as
long as the hydrothecai, and strongly annulated throughout. Hydro-
thec£e deeply campanidate, thick-walled ; margins armed ^vith twelve
to fourteen rather pointed teeth, and reduplicated once or twice, giving
a striking and unusually ornate appearance.

Gonosome. — Gonangia roughly ovate, irregular in outline, with a
short neck, small terminal aperture and a very short pedicel. They
Avere empty in the specimens examined, so that it was impossible to
ascertain whether they produced planula^ or meduscE.

Distribution. — Yakutat, Alaska (Harriman Exped.). All the
specimens were found growing in a parasitic manner over colonies of
Thuiaria costata., in company \vith Ca?npanular)ia urceolata. The
two species were often so intimately interwoven as they crept over the
stems and branches of the sertularian that I, at first, thought them
dimorphic forms of one species. However, in all cases careful dissection
showed that they were entirely' separate colonies. The reduplication
of the hydrothecal margins seems to be a constant feature and adds
peculiar beauty to this striking little campanularian.

CAMPANULARIA REGIA sp. nov.
(Plate XIX, figs. 1,2.)
7 rophosome. — Colony consisting of a creeping rootstock without
annulations, giving forth strong pedicels that are sometimes longer

THE HYDROIDS I73

than the hydrothecae and sometimes considerably shorter, without a
definite swelling below the hydrothecaj. HydrotheciE immense, in
one case nearly -^ of an inch in height, long, tubular-urceolate, ex-
panded distall}', with slightly everted and broadly sinuous margin.
The margin is reduplicated in one specimen. Hydranth with about
twenty tentacles.

Gonosome. — Not known.

Distribution. — Orca, Prince William Sound, Alaska (W. R. Coe,
Harriman Exped.).

This species is closely allied to C. grandis Allman,^ and may be
identical with it although it differs from Allman's description in the
character of the pedicels which he describes as having distinct node-
like enlargements immediately below the hydrothecae. The hydro-
thecae are larger than those of any other campanularian known to me.
But one small colony was found and it was creeping over another
hydroid.

OBELIA.

Trofhosome. — Colony regularly branching ; stem simple or com-
pound. Hydrothecas campanulate, thin, never with greatly thickened
walls.

Gonosome, — Reproduction by means of medusa; which are disk-
shaped, with four radial canals, more than eight marginal tentacles,
eight interradial lithocysts and a short manubrium without mouth ten-
tacles.

OBELIA DICHOTOMA (Linn.).

Sertularia dicJiotoma Linn., Systema Naturae, Ed. x, p. 812. 1758.

Obelia dichotoma Schulze, Nordsee Expedition, Hydroida, p. 129, 1872. —
Verrill, Preliminary check-list Marine Invertebrates of Atlantic coast,
p. 16, 1879. — Calkins, Some Hydroids from Puget Sound, Proc. Boston
See. Nat. Hist., Vol. xxviii. No. 13, p. 256, 1899.

Distribution. — Sitka, Berg Inlet, and Orca, Alaska (Harriinan

Exped.) ; British Coast (Hincks) ; Helgoland (Schulze) ; Puget

Sound (Calkins) .

OBELIA PLICATA Hincks.

Obelia plicata Hincks, British Hydroid Zoophytes, p. 159, 1868. — Nutting,
Hydroida from Alaska and Puget Sound, Proc. U. S. Nat. Mus., Vol.
XXI, No. 13, p. 741, 1899.

Distribution. — Orca, Alaska (Plarriman Exped.) ; Puget Sound

(Nutting) ; British Coasts (Hincks) .

1 Diagnoses of new Genera and Species of Hydroida; Linntean Soc. Jour.
Zoology, Vol. XI, p. 259.

174

NUTTING

OBELIA BOREALIS sp. nov.
(Plate XIX, figs. 4, 6.)

Trophosome. — Colony sometimes attaining a height of eighteen
inches, but usually much shorter ; stem not truly fascicled, although
several stems may be interw^oven, exceedingly long and slender, sinu-
ous, giving off lateral branches in pairs on proximal portion and more
often singly on distal portion ; branches vs^ith a strong tendency to ver-
ticillate arrangement, forming oblique angles with the stem and
divided into nuinerous branchlets in a flabellate manner. Pedicels
short and completely annulated, or long and annulated only at ends,
set on broad shoulders of the stem. Hydrothecae funnel-shaped, the
sides usually straight, aperture with an even rim. Hydranths not well
preserved in specimens examined.

Gonosofne. — Gonangia borne in axils of branches and branchlets,
oblong-ovate, truncated above, having a collar in mature specimens ;
aperture apparently very large, pedicels strongly annulated. The
gonangia of the specimens examined were filled with developing
medusae of the regular Obelia type.

Distribution. — Yakutat, Alaska (Harriman Exped.).

This fine species is related to O. Jiabe/lata^ but the hydrothccas are
much deeper than in O. Jlabellata, in which they are sub-triangular in
outline. It also bears some resemblance to O. comtnisstcralis, which,
however, is a much more delicate species, with smaller and more
campanulate hydrothecai.

OBELIA DUBIA sp. nov.
(Plate XX, fig. I.)

Trophosome. — Colony attaining a height of about ^ inch ; stem
sparingly branched, the main stem and larger branches sinuous or
slightly geniculate, giving forth pedicels singly or in opposite pairs at
the bends. Pedicels rather long and annulated throughout, the stem
also more extensively annulatedj than in most species of the genus.
Hydrothecae very large, deep, tubular, with very shallow undulations
around the margin, from between which lines run down for a short
distance on the surface of the hydrothecae.

Gonosome. — Unknown.

Distribution. — Orca, Alaska (Harriman Exped.).

This species bears some resemblance to O. bidentata Clark, found
on the New England Coast, but differs in the nature of the hydrothe-
cal teeth w^hich are inere sinuosities, instead of being mucronate with
two denticles each as in the latter species.

THE HYDROIDS l75

HEBELLA.

Trophosome. — Pedicels arising from a creeping rootstock, very
short. Hydrotheca tubular, with entire margins, without opercula,
and having their cavities separated from those of the stems by a partial
septum. Hydranths with a conical proboscis.

This genus was originally instituted by Allman.^ As here defined
it includes several species heretofore included in the genus Lafcea.

HEBELLA POCILLUM (Hincks).

Lafoea pocilhim Hincks, British Hydroid Zoophytes, p. 204, 1868. — Clark,
Proc. Acad. Nat. Sciences, Philadelphia, p. 215, 1876. — Verrill, Pre-
liminary check-list Marine Vertebrates of Atlantic coast, p. 17, 1879. —
Bergh, Goplepolyper (Hydroider) fra Kara-Havet, p. 333, 1887.

Distribution. — Kadiak, Alaska (Harriman Exped.) ; Nunivak
Island, Alaska (Clark) ; Kara Sea (Bergh) ; British Coasts (Hincks) ;
New England Coast (Verrill) .

GONOTHYRffiA.

Trophosome. — Much as in Obelia.

Gonosome. — Planulae produced in sessile medusaform gonophores
which remain attached to the top of the gonangia until the spermatozoa
or planulae are discharged.

GONOTHYR^A INORNATA sp. nov.
(Plate XX, figs. 2-4.)

Trophosome. — Colony attaining a height of about two inches and
consisting of a main stem which almost immediately breaks up into a
number of very slender, erect, almost straight branches which are
ornamented with about three annulations immediately above the origins
of the pedicels. Pedicels alternate, erect, much broader below than
above and with seven to ten annulations. Hydrothecte funnel-shaped,
with entire margins.

Gonoso7ne. — Gonangia borne in the axils of the pedicels, slender,
obconic, with a tendency to annulation. Each gonangium contains a
single sporosac which when mature rests upon the svunmit of the go-
nangium and has little indication of radial canals or tentacles.

Distribution. — Y'akutat Bay, Alaska (Harriman Exped.).

This species differs from others of the genus in having but one
sporosac to each gonangium and in the obliteration of most of the
medusoid characters of the sporosac. The entire margin of the hydro-
theca is also an exceptional character.

• Allman, Challenger Report, Hydroida, -Second Part, p. 29, 1S8S.

iy6 NUTTING

Family CAMPANULINID^^.

Trophosome. — Colonies branched oi* unbranched. Hydrothecce
borne on pedicels, tubular, ending in an operculum composed of
several converging segments or triangular flaps. Hydranth with a
conical proboscis.

Gonosome. — Gonangia producing free medusie or planulas.

CAMPANULINA.

Trophosome. — Hydrothecre thin-walled, the upper portion cleft so
as to produce very long and slender teeth which form an operculum by
the convergence of their free ends.

Gonosome. — Gonangia producing bell-shaped medusa, with four
radial canals, two or four marginal tentacles, and eight lithocysts.

CAMPANULINA RUGOSA sp. nov.
(Plate XXII, figs. I, 2.)

Trophosome. — Colony attaining a height of about ^ inch. Stem
irregularly branched ; branches tending to an alternate arrangement,
straggling, geniculate ; stem and branches strongly and regularly annu-
lated throughout. Pedicels very short, with three to six annulations.
Hydrothecae ovoid-oblong, the distal third being composed of the oper-
culum consisting of ten or twelve segments. The hydranths have about
sixteen tentacles.

Gonosome. — Gonangia in axils of the pedicels and branches, some-
times aggregated on certain branches to the exclusion of hydrothec^.
They are oblong-ovoid in shape and somewhat flattened on their distal
ends. Each gonangium contains a single medusa when matm-e.

Distribution. — Juneau, Alaska (Harriman Exped.). The speci-
mens were found growing on Obelia.

CALYCELLA.

Trophosome. — Stem a creeping rootstock sending forth short an-
nulated pedicels. Hydrothecae tubular, thick-walled, with opercula
that are distinct from the hydrothecal teeth, and composed of several tri-
angular segments.

Gofiosome. — Gonangia borne on the rootstock, and, when mature,
bearing acrocysts.

CALYCELLA SYRINGA (Linn.).

Sertularia syringa LiNN., Systema Naturae, Ed. xii, Tom. i, Pars ii, p.
1311, 1767.

THE HYDROIDS I77

Calycella syringa Clark, Proc. Acad. Nat. Sciences, Philadelphia, p. 210,
1876. — Verrill, Preliminary Check-list Marine Invertebrata Atlantic
Coast, p. 17, 1879. — Bergh, Goplepolyper (Hydroider) fra Kara-Havet,
p. 335, 1887. — Nutting, Hydroida from Alaska and Puget Sound, Proc.
U. S. Nat. Mus., Vol. xxi, p. 741, 1899.

Distribution, — Berg Inlet and Kadiak, Alaska (Harriman Exped.) ;

Coal Harbor, Shumagin Islands, Alaska (Clark) ; Kara Sea (Bergh) ;

British Coasts (Hincks) ; New England Coast (Verrill) ; Puget

Sound (Nutting).

Family LAFCEID^^.

Trophosome. — HydrotheCce tubular, margins without teeth or oper-
cula, the hydrothecal cavity not divided from the stem cavity by a par-
tial septum.

Gonosome. — Gonangia forming a ' Coppinia ' mass.

LAFCEA.

Trophosome. — Colony with a fascicled stem, and with hydrothecas
either free or partially immersed in the stem, the distal portion not be-
ing abruptly turned upward.

Gonosome. — A ' Coppinia' mass.

LAFCEA DUMOSA (Fleming).

Sertularia dumosa Fleming, Edinburgh Phil. Jour., 11, p. 83, 1828.

Lafcsa dujnosa Sars, Bidrag til Kundskaben om Norges Hydroider, p. 45,
1873. — Clark, Proc. Acad. Nat. Sciences, Philadelphia, p. 210, 1876. —
Verrill, Preliminary Check-list Marine Invertebrates of Atlantic Coast,
p. 17, 1879. — Nutting, Hydroida from Alaska and Puget Sound, Proc.
U. S. Nat. Mus., Vol. xxi, p. 741, 1899.

Distribution. — Dutch Harbor, Unalaska (Harriman Exped.) ; Port

Etches, Alaska (Clark) ; North Cape, Norway (Sars) ; British Coast

(Hincks) ; New England Coast (Verrill) ; Puget Sound (Nutting) ;

California Coast (Clark).

LAFCEA GRACILLIMA (Alder).

Campanularia gracillima Alder, Catalogue Zoophytes of Northumberland
and Durham, Trans. Tyneside Naturalists' Field Club, p. 39, 1857.

Lafcea gracillima Clark, Proc. Acad. Nat. Sciences, Philadelphia, p. 216,
1876. — Verrill, Preliminary Check-list Marine Invertebrates of Atlantic
Coast, p. 17, 1879. — Marktanner-Turneretscher, Hydroiden von
Ost-Spitzbergen, Zoolog. Jahrbuch., Vol. viii, p. 410, 1895. — Nutting,
Hydroida from Alaska and Puget Sound, Proc. U. S. Nat. Mus., Vol. xxi,
p. 741, 1899.

Distribution. — Juneau, Berg Inlet and Orca, Alaska (Harriman

Exped.) ; Shumagin Islands, Alaska (Clark) ; British Coast (Alder) ;

Proc. Wash. Acad. Sci., May, 1901.

178 NUTTING

Spitzbergen (Marktanner-Turneretscher) ; New England Coast (Ver-
rill) ; Puget Sound (Nutting).

LAFCEA FRUTICOSA M. Sars.

Lafcea frtiticosa M. Sars, Bemzerkninger over 4 norske Hydroider Vid. Selsk.
Forh., 1862. — G. O. Sars, Bidragtil Kundskaben om Norges Hydroider,
p. 26, 1873. — Clark, Proc. Acad. Nat. Sciences, Phila., p. 216, 1876.

Distribution. — Juneau, Berg Inlet and Kadiak, Alaska (Harriman

Exped.) ; Kiska Harbor, Shumagin Islands, Alaska (Clark) ; British

Coasts (Hincks) ; Lofoten, Norway (G. O. Sars) ; New England

Coast (Verrill) ; Puget Sound (Nutting, MSS.).

LAFCEA ADHERENS sp. nov.
(Plate XXI, figs. 3, 4.)

Trophosome. — Colony forming an encrusting mass of adherent root-
stocks disposed both longitudinally and transversely over colonies of
other Hydroids, the tubes of the rootstock interwoven much like the
threads of a fabric. HydrotheciE sessile, tubular, often more or less
curved, aperture facing upward, entire ; margin slightly expanded.
The hydrothecae are very irregularly disposed, being much more
crowded in some places than in others.

Gonosome. — The ' Coppinia ' mass is much like that of Lafoea du-
fnosa., being composed of closely packed gonangia interspersed with
long, tubular, variously curved modified hydrothecae. The gonangia
are flask-shaped, with a tubular neck and small aperture. Each
gonangium apparently contains a single ovum.

Distribtition. — Kadiak Harbor, Alaska. Growing over stems of
Thuiaria turgida (Harriman Exped.).

This interesting species is so different in appearance from the others of
the genus that I was at first inclined to make it the type of a new genus.

GRAMMARIA.

Trophosome. — Stem fascicled, composed of an axial tube from
which the hydrothec^ spring and to which they are partly adnate,
completely enclosed by a definite number of peripheral nonhydro-
thecate tubes.

Gonosome. — A ' Coppinia ' mass.

GRAMMARIA IMMERSA sp. nov.
(Plate xxi, figs. 5, 6.)
Trophosome. — Stem rigid, erect, giving forth scattered, stiff and
straight, altei-nate branches forming nearly a right angle with the stem.

THE HYDROIDS 1 79

Height of a fragmentary specimen about ^ inch. Stem and branches
sharply constricted proximally, composed of an axial tube which gives
off the eight or nine series of hydrothecae, and a number of peripheral
tubes enclosing the axial tube completely, thus burying all the hydro-
thecse nearly to their distal ends, Hydrothecae arranged in about eight
or nine longitudinal series, forming spirals. The distal ends of the
hydrothecae are abruptly bent outward, so that the round, even aper-
ture is vertical. When the peripheral tubes are removed the hydro-
thecae are seen to be long, tubular, doubly curved, narrowing prox-
imally, but without true pedicels, and all springing from the axial tube.

Gonoso?ne. — Unknown.

Distribution. — St. Paul harbor, Kadiak (Harriman Exped.).

FILELLUM.

Trophosome. — Stem a creej^ing, slender rootstock, parasitic on
other hydroids, often forming a reticulate structure. Hydrothecae
curved, decumbent, and partly adherent ; margin entire, without
operculum.

Gonosome. — A 'Coppinia' mass.

FILELLUM SERPENS (Hassell).

Campanularia serpens Hassell, Zoologist, No. 69, p. 2223.

Filelhim serpens Sars, Bidrag til Kundskaben om Norges Hydroider, p. 29,

1873. — Verrill, Preliminary Check-list Marine Invertebrates of Atlantic

Coast, p. 17, 1879.

Distribution. — Juneau, Alaska (Harriman Exped.) ; British Coast
(Hassell) ; Lofoten, Norway (Sars).

Family HALECID^^.
Trophosome. — Hydrothecae reduced to the form of saucer-shaped or
collar-like hydrophores, usually borne on broad tubular pedicels ;
margins even, often reduplicated. Hydranths large, incapable of re-
tracting within the hydrophores, and with a conical or dome-shaped
proboscis.

Gonosome. — Gonangia producing planulae, and usually differing in
the two sexes.

HALECIUM.
Trophosome. — No specialized defensive ' persons ' developed.
Gonosome. — Female gonangia often surmounted by a pair of
hydranths.

HALECIUM HALECINUM (Linn.).
Sertularia halecina Linn., Systema Naturae, Ed. x, p. 809, 1758.

l8o NUTTING

Halecium halecmiwi Verrill, Preliminary Catalogue Marine Invertebrates
Atlantic Coast, p. 17, 1879. — Marktanner-Turneretscher, Hydroiden
von Ost-Spitzbergen, p. 428, 1895. — Nutting, Hydroida from Alaska
and Puget Sound, Proc. U. S. Nat. Mus., Vol. xxi, p. 741, 1899.

Distribution. — Kadiak, Alaska (Harriman Exped.) ; British Coast
(Hincks) ; Greenland (Marktanner-Turneretscher) ; New England
Coast (Verrill) ; Puget Sound (Nutting).

HALECIUM MURICATUM (Ellis and Solander).

Sertularia muricaia Ellis and Solander, Nat. Hist. Zoophytes, p. 59, 1786.

Haleciiwi muricatutn Verrill, Preliminary Check-list Marine Invertebrates
Atlantic Coast, etc., p. 17, 1879. — Levinsen, Meduser, Ctenophorer og
Hydroider fra Gronlands Vestkyst, p. 61, 1893. — Clark, Proc. Acad.
Nat. Sci., Philadelphia, p. 217, 1876.

Distribution. — Orca, Alaska (Harriman Expd.) ; British Coasts
(Ellis and Solander) ; Greenland (Levinsen) ; New England Coast
(Verrill).

HALECIUM SCUTUM Clark.

Halecium scutum Clark, Proc. Acad. Nat. Sci., Philadelphia, p. 218, 1876.
— Bonnevie, Norwegian N. Atlan. Exped., p. 57, 1899.

Distributio7i. — Berg Inlet and Yakutat, Alaska (Harriman Exped.) ;
Unalaska (beach) and Shumagin Islands, Alaska (Clark) ; North
Cape, Norway (Bonnevie).

HALECIUM REVERSUM sp. nov.
(Plate XXIII, figs. I, 2.)

Trophosome. — Colony attaining a height of about one inch. Main
stem fascicled, branches simple and alternate, making a flabellate
colony. Nonfascicled part of the stem and branches divided into inter-
nodes, each of which bears one or two pedicels springing from its
proxit7ial portion. Pedicels long, of even diameter throughout, often
rugose on proximal portion. Hydrophores with large everted margins
and a distinct row of dots. Reduplication of margins distant, when
present. Hydranth small for this genus, with about twenty tentacles.

Gonosome. — Not present in type specimen.

Distribution. — Juneau, Alaska (Harriman Exped.).

This species possesses the very exceptional, if not unique, character
of having the pedicels spring from the proximal part of the internode
instead of the distal portion, as in all other species of Halecium with
which I am acquainted. This character appears to be constant. The
specimen was dredged from a depth of twenty fathoms.

THE HYDROIDS l8l

HALECIUM ORNATUM sp. nov.
(Plate XXII, figs. 3,4.)

7 ropkosome. — Colony parasitic, branching irregularly. Stems not
fascicled, the stem and branches sparsely and irregularly annulated.
Pedicels long, of equal diameter throughout. Hydrophores with broad,
everted inargins, occasionally reduplicated. Hydranth large, with
twenty-four to thirty tentacles.

Gonosome. — A single apparently young gonangium was borne on a
pedicel just below the hydrophore. It was in form a truncated and
deeply annulated cone. Probably the mature gonangium would
resemble that found in the next species.

Distribution. — Berg Inlet, Glacier Bay, Alaska. Growing on
Lafcea gracillima (Harriman Exped.).

HALECIUM SPECIOSUM sp. nov.
(Plate XXII, figs. I, 2.)

Trophosome. — Colony small, attaining a height of about Yi inch.
Stem not fascicled, the main stem and branches apparently formed of
series of stout pedicels, each giving origin to another pedicel just below
the hydrophore. The pedicels thus take the place of stem joints,
bending alternately to the right and left, giving a geniculate appearance
to the series. Pedicels broad, corrugated proximally and smooth
distally. Hydrophores large, with broadly expanded but not everted
margins, and a well marked row of dots. There appears to be no re-
duplication of the margins. Hydranths very large with twenty-four to
thirty short tentacles, a broad oral disk occupied, by the low dome-
shaped proboscis.

Gonosome. — Gonangia borne on rather long annulated pedicels be-
low the hydrophores, particularly on the upper part of the colony ;
regularly ovoid in outline, and evenly and beautifully annulated
throughout.

Distribution. — Yakutat Bay, Alaska (Harriman Exped.).

This is the most strikingly ornamented species of Halecium known
to the writer. Its manner of growth is exceptional, although not un-
known among its allies, and the hydranth is more like the polyp of
some actinozoon than of a hydroid. The gonangia are beautiful struc-
tures. They seemed to be filled with a granular, ovoid mass, probably
a single large sporosac.

l82 NUTTING

HALECIUM ROBUSTUM sp. nov.
(Plate XXIII, figs. 3, 4, 5.)

Trophosome. — Stem very thick and fascicled, consisting of an im-
mense number of wavy tubes. In the single specimen collected, the
main stem divides near its base into three heavy fascicled branches,
which themselves branch and subdivide extensively, the fasciculation
continuing nearly to the tips of the branches. The entire height of
the colony is about three inches. The branching is so profuse that the
arrangement of the internodes is obscure. The ultimate branches give
off short pedicels and sessile hydrophores in what appear to be clusters
or whorls. Pedicels short, tubular, ending in an exceedingly shallow
hydrophore. Hydrophores reduced to a mere narrow rim, marked by
the internal diaphragm and circlet of dots, the only distinction between
pedicel and hydrophore, as the margin of the latter is not appreciably
everted. Most of the hydrophores are sessile, being set on mere
shoulders of the branch from which they grow. The circlet of dots
can only be made out with great difficulty and the use of high magnifi-
cation. Hydranths exceedingly numerous and large, covering the
branches so as to almost entirely conceal them from view. Tentacles
about twenty, surrounding a low conoid proboscis.

Go?iosome. — Unknown.

Distributio7i. — Berg Inlet, Glacier Bay, Alaska (Harriman Ex-
ped.).

This species bears some relation to Halecijim densum Calkins, ' but
differs from that species in the hydrophores, which are not reduplicated
and have straight, not everted, margins. The hydranths are so crowded
that a branch resembles an expanded colony of Alcyonai-ia, and appears
to be made up entirely of hydranths.

Family SERTULARID.^.

Trophosoine. — Hydrothecae sessile, more or less adnate by their side
to the stem and branches upon which they grow ; always in more
than one longitudinal row on each branch, the arrangement usually
biserial. Hydranth with a conical proboscis.

Gonosome. — Reproduction always by means of planulte, which are
developed within the gonangia. No medusae.

SERTULARELLA.

Trophosome. — Stem and branches divided into regular internodes,
each bearing one or two hydrothecae ; nodes oblique. Hydrothecae
iProc. Boston Soc. Nat. Hist., Vol. 28, no. 13, p. 343, 1S99.

THE HYDROIDS 183

alternate, borne on opposite sides of the stem and branches. Margin
usually more or less toothed, aperture usually provided with an oper-
culum consisting of more than one segment.

SERTULARELLA TRICUSPIDATA (Alder).

Sertidaria tricusptdata Alder, Catalogue of the Zoophytes of Northumber-
land and Durham, Trans. Tyneside Naturalists' Field Club, p. 21, 1857.

Cotuiina tricnspidata A. Agassiz, North American Acalephae, p. 146, 1864.

Serttdarella tricnspidata Clark, Proc. Acad. Nat. Sciences Phila. , p. 224,
1876. — Marktanner-Turneretscher, Hydroiden von Ost-Spitzbergen,
p. 425, 1895. — Nutting, Hydroida from Alaska and Puget Sound, Proc.
U. S. Nat. Mus., Vol. xxi, p. 741, 1899.

Distribtttion. — Juneau, Berg Inlet, and Yakutat, Alaska (Harri-
man Exped.) ; Shumagin Islands, Semidi Islands, Unalaska, Port
Etches, and Kiska Harbor, Alaska (Clark) ; British Coast (Alder) ;
New England Coast (A. Agassiz) ; Greenland (Marktanner-Turner-
etscher) ; Puget Sound (Nutting).

SERTULARELLA POLYZONIAS (Linn.).

Sertularia polyzonias Linn., Systema NaturcC, Ed. x, p. 813, 1758.
Serttilarella polyzonias Gray, List of the Specimens of British Animals in the

Bridsh Museum, part i, Radiata, London, 1847. — Sars, Bidrag til Kunds-

kaben om Norges Hydroider, p. 44, 1873. — Clark, Proc. Acad. Nat. Sci.

Phila., p. 224, 1876.
Cotuiina polyzonias Agassiz, Contributions to the Natural History of the United

States, Vol. iv, p. 356, 1864.
Serttdarella conica Calkins, Some Hydroids from Puget Sound, Proc. Boston

Soc. Nat. Hist., Vol. xxviii, p. 359, 1899.

Distributioiz. — Orca, Alaska (Harriman Exped.) ; Port Etches and
Nunivak Island, Alaska (Clark) ; British Coasts (Gray) ; North
Cape, Norway (Sars) ; New England Coast (Agassiz) ; Puget Sound
(Calkins).

Calkins, in the reference cited above, identifies a small specimen
otherwise identical with S. polyzonias^ as the species S. conica All-
man, and says : "The only character, and that a small one, by which
to distinguish it from the very wide-spread S. polyzonias is the well-
marked wrinkling on the adcauline side of the hydrotheca." Speci-
mens from Alaska in the Harriman Collection, agreeing with Calkins's
description and figures, have the characteristic gonosome of 6". poly-
zonias^ and I therefore consider that I am justified in regarding his
specimen as belonging to that species.

SERTULARELLA SACCATA sp. nov.
(Plate XXIV, figs. 1-3.)
Serttdarella mimosa Clark, Proc. Acad. Nat. Sci. Phila,, p. 224, 1876.

184 NUTTING

Sertularella gefiicitlata Marktanner-Turneretscher, Die Hydroiden Des
k. k. naturhistorischen Hofmuseums, p. 222, 1890.

Trophosofne. — Colony consisting of a sparingly branched, non-
fascicled stem attaining a height of about one inch. Stem annulated
and nonhydrothecate proximally, otherwise bearing alternate hydro-
thecEe, one to each internode ; stems and branches erect. Hydrothecae
roughly ovate in general outline, with a laterally inclined, broad,
smooth, rovmd neck or collar surmounted by a four-toothed aperture
and a four-flapped operculum ; below the neck the body of the hydro-
theca has three or four broad annular corrugations. Hydra nths with
about sixteen tentacles.

Gonosome. — Gonangia large, oblong-ovate in general outline, with
several, seven to nine, broad annular corrugations. The mature ova
are enclosed in an ovoid acrocyst resting on the top of the gonangium.

Distrtbiiiion. — Popof Island, Alaska (HaiTiman Exped.) ; Un-
alaska, Shumagin Islands, St. Paul Island and Nunivak Island, Alaska
(Clark) ; Jan Mayen (Marktanner-Turneretscher).

The specimens of this species secured by the Harriman Expedition
agree well with the figure given by Clark of specimens that he identi-
fied as S. rugosa. The very conspicuous neck, however, would seem
to be sufficient to distinguish S. saccata from S. rugosa^ and the same
difference exists between the gonangia of the two forms. The figure
given by Marktanner-Turneretscher for S. geniculata Hincks differs
greatly from Hincks's original description and figure, and agrees well
with the present species, except that the hydrotheca) are more closely
approximated in the latter.

THUIARIA.

Trophosome. — Colony branched, the branches divided into unequal
internodes, each bearing several pairs of opposite or subopposite hy-
drothecae. HydrotheCcB usually deeply immersed in the stem or branch
to which they are attached. The branches are alternate, and each
springs from an unpaired hydrotheca.

Gonosome. — Much as in Sertularia.

THUIARIA ARGENTEA (Ellis and Solander).

Serin/aria argcntea Ellis and Solander, The Natural History of many curious
and uncommon Zoophytes, etc., p. 38, 1786. — Clark, Hydroids of the
Pacific Coast of the United States south of Vancouver Island, Trans.
Conn. Acad., Vol. iii, p. 251, 1876. — Verrill, Preliminary Check-list
Marine Invertebrates Atlantic Coast, etc., p. 18, 1879. — Bergh, Gople-
polyper (Hydroider) fra Kara-Havet, p. 335, 1887. — Nutting, Hydroida
from Alaska and Puget Sound, Proc. U. S. Nat. Mus., Vol. xxi, p. 741,
1899.

THE HYDROIDS 185

Distribution. — Yakutat, Alaska (Harriman Exped.) ; British Coast
(Ellis and Solander) ; Kara Sea (Bergh) ; New England Coast (Ver-
rill) ; Puget Sound (Nutting) , California (Clark).

THUIARIA SIMILIS (Clark).
Sertularta similis Qi^AKK, Proc. Acad. Nat. Sci. Phila., p. 219, 1876,

Distribution. — Berg Inlet, Glacier Bay, Alaska (Harriman Exped.) ;
Hagmeister Island, Alaska (Clark) ; Puget Sound (Nutting, MSS.).

THUIARIA VARIABILIS (Clark).
Sertularia variabilis Clark, Proc. Acad. Nat. Sci. Phila., p. 221, 1876.

Distribution. — Orca, Alaska (Harriman Exped.) ; numerous sta-
tions in Alaska (Clark) ; San Miguel Island, California (Clark) ; Puget
Sound (Nutting, MSS.).

THUIARIA CUPRESSOIDES (Clark).
Sertularia cupressoides Clark, Proc. Acad. Nat. Sci. Phila., p. 220, 1876.

Distribution. — Yakutat, Alaska (Harriman Exped.) ; Shumagin
Islands and Port Moller, Alaska (Clark).

THUIARIA COEI sp. nov.
(Plate XXVI, figs. 1-3.)

Trophosorne. — Colony consisting of a single flexuous stem giving
forth regularly alternate branches. Stem three inches high, and divided
into irregular internodes each bearing a branch and two hydrothecee on
one side and one hydrotheca on the other. Branches divided into
irregular internodes, each usually bearing several pairs of hydrothecae.
HydrothecEe subopposite, turgid basally, narrowing distally to a hori-
zontal aperture which is pointed on its outer side. An upward pro-
jecting point of chitine at the bottom of each hydrotheca.

Gonosome. — Gonangia top-shaped, or obconical, with a pronounced
round collar and rather broad aperture. Proximal portion broadly cor-
rugated, and narrowing basally to a short curved pedicel.

Distribution. — Dutch Harbor, Alaska (W. R. Coe, Harriman
Exped.).

This is a very distinct species, and the top-shaped gonangia are
quite different from any others of the genus that I have seen.

THUIARIA FABRICII (Levinsen).

(Plate xxiv, figs. 4, 5.)

Sertularia fabricii Levinsen, Meduser, Ctenophorer og Hydroider fra Gron-
lands Vestkyst, p. 48, 1893. — Calkins, Some Hydroida from Alaska and
Puget Sound, Proc. Boston Soc. Nat. Hist., Vol. xxviii, p. 361, 1899.

l86 NUTTING

Distribution. — Dutch Harbor, Alaska (Harriman Exped.) ; Green-
land (Levinsen) ; Puget Sound (Calkins).

A specimen in the Harriman collection agrees perfectly with the
descriptions and figures of both Levinsen and Calkins. I have directly
compared it with specimens of Serttdaria argentea from England,
and find that the two species are evidently distinct, T. fabricii differ-
ing from 6". argentea in the following particulars. The colony is much
more bushy in appearance, and more compactly branched. The hydro-
thecas are more nearly in pairs, and much inore closely approximated,
are more densely corneous and have a more delicate and less clearly
defined aperture. The gonangia are considerably larger and of thinner
texture, and are only occasionally armed with lateral spines.

It should be explained that Levinsen regards his Sertularia fabricii
as identical with the Serttilaria argentea of authors, and gives it the
name S. fabricii. My opinion is that the species is distinct, and I
recognize the name given by him because he has correctly described
and figured the species, although not classing it as distinct. It should,
however, be placed in the genus Thuiaria^ for it comes well within
that genus as here defined.

THUIARIA TURGIDA Clark.

(Plate XXV, figs. 4-6.)

Thuiaria turgida Clark, Proc. Acad. Nat. Sci. Phila., p. 229, 1876. — Nut-
ting, Hydroida from Alaska and Puget Sound, Proc. U. S. Nat. Mus.,
Vol. XXI, p. 741, 1899.

Distributio7t. — Sitka, Orca, Popof Island, and Dutch Harbor,
Alaska (Harriman Exped.) ; Port Etches, Shumagin Islands, Semidi
Islands, and many other points in Alaska (Clark).

THUIARIA GIGANTEA Clark.

Thuiaria gigantea Clark, Proc. Acad. Nat. Sci. Phila., p. 230, 1876 — Nut-
ting, Hydroida from Alaska and Puget Sound, Proc. U. S. Nat. Mus.,
Vol. XXI, p. 741, 1899.

Distribution. — Kadiak and Popof Island, Alaska (Harriman
Exped.) ; St. Paul Island, Hagmeister Island, Unalaska and Kiska
Harbor, Alaska (Clark) ; St. Paul Island (Nutting).

THUIARIA THUIARIOIDES (Clark).
Sertularia thidarioides Clark, Proc. Acad. Nat. Sci. Phila., p. 223, 1876.

Distribution. — Yakutat, Alaska (Harriman Exped.) ; Nunivak
Island and Chignik Bay, Alaska (Clark).

THE HYDROIDS 1 87

THUIARIA ELEGANS sp. nov.
(Plate XXV, figs. 1-3.)

Trofhosome. — Colony consisting of a central nonfascicled stem
branching in a plumose manner and attaining a height of about six
inches. Stem bearing hydrothecae throughout its length, divided by
oblique nodes into long and irregular internodes, each of which usually
bears three or four branches ; branches alternate, pinnately arranged,
unbranched proximally, and distally dividing into a number of branch-
lets ; internodes of stem irregular, but each bearing normally more
than one pair of hydrothecre. Hydrothecae subalternate, short, pitcher-
shaped, with a double curve in front and an even aperture much like
the top of a pitcher ; operculum composed of a single flap.

Gonosome. — Gonophores borne on the upper sides of the distal ends
of the branches in a closely set double row. The individual gonangium
is slender, oblong-oval, with a tiamcated top, an internal distal plug
which appears as a dark collar, and an internal mass of developing sex
elements.

Distribution. — Berg Inlet, Glacier Bay, 20 fathoms ; Dutch Har-
bor, Alaska (Harriman Exped.).

THUIARIA COSTATA sp. nov.

(Plate xxvi, figs. 4-9.)

Trophosome. — Colony usually a single stem giving forth alternate
branches, the whole having a plumose appearance, stem simple,
straight, the lower part composed of regular internodes, each bearing a
pair of subopposite hydrothecae, the upper part divided into regular
internodes, each bearing a branch and three hydrothecae ; nodes
oblique. Branches alternate and themselves branching dichotomously,
divided into unequal internodes, each bearing more than two suboppo-
site hydrothecae. Hydrothecae turgid below and narrowing above into
a short neck which ends in a round aperture facing upward. A chlti-
nous spine projects downward from the lower inner side of each hy-
drotheca.

Gonosome. — Gonangia borne profusely on both faces of the stem
and often on the proximal ends of the branches ; oblong-ovate, with a
short, small neck and round terminal aperture, the gonangia are orna-
mented with about five compressed longitudinal ridges, the crests of
which are colored black. General color of the gonangia orange browm.

Distribution. — Yakutat, Alaska. Abundant (Harriman Exped.).

This species resembles Sertularia itzconsta?ts Clark, but differs
considerably, particularly in its gonosome, which is very strongly
marked.

l88 NUTTING

Family PL UMULARID^^.

Trophosome. — Hydrothecae cup-shaped, usually more or less adnate
to the stem or bi'anches, and always arranged on one side only of the
hydrocladia, or branches, on which they grow. Nematophores present.

Gonosome. — Reproduction by means of planulae. No medusae.

PLUMULARIA.

Trofhosome. — Hydrocladia unbranched alternate, nematophores on
slender pedicels; hydrothecae without marginal teeth. Stem not canal-
iculated.

Gonosome. — Gonangia oval, without corbulae or protective struc-
tures of any kind.

PLUMULARIA LAGENIFERA Allman.

Plwnularia lagenifera P^XAMKH, Jour. Linn. Soc. Zool., xix, p. 157, 1885. —
Nutting, American Hydroida, Part i, The Plumularidae, p. 65, 1900.

Plumularia califorjiica Marktann'er Turneretscher, Annalen des k. k.
Naturhist. Hofmuseums, v, No. 2, p. 255, 1890. — Nutting, Hydroida
from Alaska and Puget Sound, Proc. U. S. Nat. Mus., Vol. xxi, p. 741,
1899.

Distribution. — Berg Inlet, Popof Island, Alaska (Harriman Exped.) ;

Puget Sound (Nutting) ; Coast of California (Nutting).

PLUMULARIA PALMERI Nutting.

Plioniilaria palmcriYi,Xi'X:'X\^G, American Hydroida, Part i, The PlumularidiE,
p. 65, 1900.

Distribution. — Victoria, B. C. (Harriman Exped.) ; San Diego.
California (Nutting) .

This is the only species in the collection that was not from Alaska.
It seemed best to include it in the list, particularlv as a new locality is
thereby noted.

THE HYDROIDS 189

BIBLIOGRAPHY.

The following list includes all the publications referred to in this Report on
Hydroida. It does not include papers that were consulted and found to contain
nothing pertinent to the immediate subject of the report.
Agassiz, Alexander.

1865. North American Acalephse. Illustrated Catalogue, Museum Compara-
tive Zoology, No. II.

Agassiz, Louis.

1862. Contributions to the Natural History of the United States, IV.

Alder, Joshua.

1857. A Catalogue of the Zoophytes of Northumberland and'Durham. Trans.
Tyneside Naturalists' Field Club.

AUman, G. J.

1859. Notes on the Hydroid Zoophytes. Annals and Mag. Nat. Hist., Third

Series, Vol. IV, pp. 52 and 141.
1864. On the Constuction and Limitation of Genera among the Hydroida.

Annals and Mag. Nat. Hist., Third Series, Vol. XIII, No. 77."
1870-1872. Monograph of the Gymnoblastic or Tubularian Hydroids, Ray

Society.
1888. Report on the Hydroids dredged by H. M. S. Challenger, Part 2.
1885. Description of Australian, Cape and other Hydroids, mostly new, from

the collection of Miss H. Gatty. Journ. Linnsean Society, Vol. XIX.

Bergh, R. S.

1887. Goplepolyper (Hydroider) fra Kara-Havet.

Bonnevie, Kristine.

1898. Zur Systematik der Hydroiden. Zeitschr. Wissen. Zoologie, Jahrb. 63

1899. Hydroida of the Norwegian North Atlantic Expedition.

Calkins, G. M.

1899. Some Hydroids from Puget Sound. Proc. Boston Soc. Nat. Hist., Vol.
XXVIII, No. 13.

Clark, S. F.

1876. Report on the Hydroids collected on the Coast of Alaska and the Aleu-
tian Islands, by W, H. Dall. Proc. Acad. Nat. Sciences, ',Philadelphia.

1876. Description of New and Rare Species of Hydroids from the New Eng-
land Coast. Trans. Conn. Acad. Sci., III.

1876. The Hydroids of the Pacific Coast of the United States, South of Van-
couver Island. Trans. Conn. Acad. Sci., III.

Ellis and Solander.

1786. The Natural History of many Curious and Uncommon Zoophytes.
Fleming, J.

1828. A History of British Animals. Edinburgh Philos. Journ., XIII.

Hartlaub, Clemens.

Die Hydromedusen Helgolands. Wissench. Meeresuntersuchungen. II
Band.

1 9©

NUTTING

Hincks, Rev. Thomas.

1868 British Hydroid Zoophytes

1880 On New Hydroida and Polyzoa from Berent's Sea. Annals and Mag.
Nat. Hist,, Fifth Series, VI.
Kirchenpauer, G. H.

1884. Nordische Gattungen und Arten von Sertulariden. Abhand. a. d. Geb.
d. Naturwiss., herausgegeb. vom Naturwiss., Verein, VIII, Pt. 3.

Lamarck, J. B. P. A. de.

1836. Histoire Naturelle des Animaux sans Vertebres, 2d Edition.

Lendenfeld, R. von.

1883 iJber Coelenteraten der Sudsee, V. Mittheilung. Zeitschrift fur Wis-
senscli. Zoologie, XXXVIII.

Levinsen, G. M. R.

1886 Annvilata, Hydroida, Anthozoa, Porifera, in : Vid. Udbytte af Kanon-
baaden, Haucks Togter i.

1892 Om Fornyelsen af Ernaeringsindividerne hos Hydroiderne. Siertryk af
Vidensk. Meddel. fra den naturhist. Fornening i Kbhvn.

1893 Meduser, Ctenophorer, og Hydroider fra Gronlands Vestkyst, Copen-
liagen.

Linnaeus, C.

1758 Systema Naturse, loth Edition.

1767 Systema Naturae, 12th Edition, Pt. II.
Marktanner-Turneretscher , Gottlieb .

1890 Die Hydroiden des k. k. Naturhist. Hofmuseums, V.

1895 Hydroiden von Ost-Spitzbergen. Zool. Jahrb., VIII. Abth.

Mereschkowsky, C.

1878 New Hydroida from Ochotsk, Kamtschatka, and other parts of the
North Pacific Ocean. Annals and Mag. Nat. Hist., Dec.

Nutting, C. C.

1899 Hydroida from Alaska and Puget Sound. Proc. U. S. Nat. Museum,
Vol. XXI, No. 1 171.

1900 American Hydroids, Part I., The Plumularidie. U. S. Nat. Museum,
Special Bulletin No. 4.

Sars, M. (?)

1862 Bemserkninger over fire norske Hydroider; Vid. Selsk. Forh.

1873 Bidrag til Kundskaben om norges Hydroider.
Schultz, Franz Eilhard.

1872 Nordsee Expedition, III, Coelenteraten.

Schneider, K. C.

1897 Hydropolypen von Rovigno, nebst iibersicht iiber das system der Hy-
dropolypen in allgemeinen. Zool. Jahrb., Syst. abth. bd., 10.

Segerstedt, M.

1899 Bidrag til Kannedomen om Hydroid-Faunen ved Sveriges Vestkust,
Stockholm.

Verrill, A. E.

1873 Brief Contributions to Zoology from the Museum of Yale College, No.
XXIII, American Journal of Science.

THE HYDROIDS

191

Verrill, A. E.

1874 Invertebrate Animals of Vineyard Sound.

1879 Preliminary Check List of the Marine Invertebrates of the Atlantic
Coast, from Cape Cod to the Gulf of St. Lawrence.
Winther, Georg.

1879 Fortegnelse over de i Danmark og dets Nordlige Bilande f undue Hv-
droide Zoophyter.

Wright, J. Strethill.

1859 Observations on British Zoophytes. Edinburgh New Philos. Journ.,
New series, Vol. VII.

PLATE XIV.

Fig. I. Coryne brachiata Nutting. Part of colon j.

2. Single gonophore (enlarged).

3. Syncorync cximia AUman. Part of colony.

4. Single hjdranth with budding medusa.

(192)

Proc. Wash. Acad. Sci. Vol. Ill

Plate XIV

ALASKA HYDROIDS

PLATE XV.

Fig. I. Garveia annulata Nutting. Part of colony, showing hvdranths and
gonophores.

2. Single hydranth, showing pseudo-hydrotheca (enlarged).

3. Eudendrium vaginatum Allnian. Part of colony, showing hydranths
and gonophores.

4. Single hydranth, showing pseudo-hydrotheca (enlarged).

5. Cluster of female gonophores.

6. Single gonophore, with expanded chitinous collar (enlarged).

(194)

PROC. Wash. Acad. Sci. Vol. Ill

Plate XV

ALASKA HYDROIDS

PLATE XVI.

Fig. I. Tubularia /tarrtmant"Nutting. Single hydranth with gonophores.

2. Three gonophores, showing apical tentacles and developing actinule
(enlarged).

(196)

Proc. Wash. Acad. Sci. Vol. Ill

Plate XVI

ALASKA HYDROIDS

PLATE XVII.

Fig. I. Clytia caliculata (Ilincks). Single hydrotheca (enlarged).

2. Another hydrotheca (enlarged).

3. Clytia compressa (Clark). Hydrothecae (enlarged) showing different
degrees of thickening of the hydrothecal walls.

4. Single gonangium (enlarged).

5. Campatiulafia ritteri Nutting. Part of colony (enlarged).

(19S)

Proc. Wash. Acad. Sci. Vol. Ill

Plate XVII

ALASKA HYDROIOS

PLATE XVIII.

Fig. I. Campanularia reduplicata Nutting. Part of colony, showing iijdro-
thecae and gonangium (enlarged).

2. Campanularia urceolata Clark. Part of colonv, showing hydrothecae
and gonangia (enlarged).

3. Campanularia speciosa Clark. Single hvdrotheca (enlarged).

(200)

Proc. Wash. Acad. Sci. Vol. Ill

Plate XVlli

ALASKA HYDROIDS

PLATE XIX.

Fig. I. Campatiularia regia Nutting. Single hydrotheca, showing hjdranth
(much less enlarged than the other figures).

2. Another hydrotheca, showing reduplication of margin.

3. Hydrotheca of C. spectosa Clark (drawn to the same scale as fig. 1, to
show comparative size).

4. Obelia borealis Nutting. Part of colony, showing hydrothecae and
gonangium (enlarged).

5 and 6. Gonangia (greatly enlarged).

(202)

Proc. Wash. Acad. Sci. Vol. II

Plate XIX

ALASKA HYDROIDS

PLATE XX.

Fig. I. Ohelia dubia Nutting. Part of colony (enlarged).

2. GonothyrcEa inornata Nutting. Part of colony, showing hydrothecse and

gonangium (enlarged).
3 and 4. Gonangia (enlarged).

(204)

Proc. Wash. Acad. Sci. Vol. Ill

Plate XX

ALASKA HYDROIOS

PLATE XXI.

Fig. I. Campa7iulina rugosa Nutting. Part of colony, showing hydrothecse and
gonangia (enlarged).

2. Gonangium, showing developing medusa (enlarged).

3. Lafoea ad/icerens Nutting. Part of trophosome (enlarged).

4. Part of transverse section of gonosome (enlarged).

5. Grammaria inimeisa Nutting. Part of main stem (enlarged).

6. Portion of branch with the peripheral tubes removed to show connection
of hjdrotheciE with axial tube (more enlarged than fig. 5).

(206)

Proc. Wash. Acad. Sci. Vol. Ill

Plate XXI

ALASKA HYDROIDS

PLATE XXII.

Fig. I. Halccium spcciosum Nutting. Part of colony, showing hjdrophores and
gonangia (enlarged).

2. Single hydranth (greatly enlarged).

3. Halecium ornatum Nutting. Part of colony, showing hydrophores (en-
larged ) .

4. Single immature gonangium (enlarged).

(208)

Proc. Wash. Acad. Sci. Vol. Ill

Plate XXII

ALASKA HYDROIDS

PLATE XXm.

Fig. I. Halecium reversum Nutting. Part of colony, showing internodes and
hydrophores (enlarged).

2. Single hydranth with hydrophore (enlarged).

3. Halecium robusium Nutting. Tip of branch, showing expanded
hydranths (enlarged).

4. Part of branch with soft parts removed to show the hydrophores (en-
larged).

5. Single hydranth with its hydrophore (greatly enlarged).

(210)

Proc. Wash. Acad. Sci. Vol. Ill

Plate XXII

"^.T:

ALASKA HYDROIDS

PLATE XXIV.

Fig. I. Sertularella saccata Nutting. Part of colony, showing hjdrothecae and
gonangia with acrocysts (enlarged).

2. Three hjdrothecae (still more enlarged).

3. Single hydrotheca, showing operculum (drawn to same scale as Fig. 2).

4. T/iutaria fabrtcii {'Levinsen). Part of branch showing hydrothecse and
gonangia (enlarged).

5. Front view of part of branch (enlarged).

(212)

Proc. Wash. Acad. Sci. Vol. Ill

Plate XXIV

ALASKA HYDROIDS

PLATE XXV.

Fig. I. Thuiaria elegans Nutting.

2. Front view of branch (much enlarged).

3. Gonangium (much enlarged).

4. Thuiaria turgida Clark.

5. Two pairs of hydrothecse (much enlarged).

6. Front view of branch (enlarged).

(214)

pRoc. Wash. Acad. Sci. Vol. Ill

Plate XXV

STH B CjAPiKOW. ;

ALASKA HYDROIDS

PLATE XXVI.

Fig. I. Thuiaria coei Nutting. Part of branch, showing hjdrothecje and
gonangia (enlarged).

2. Lateral view of hydrotheca (enlarged).

3. Single gonangium (enlarged).

4. Thuiaria costata Nutting. Part of main stem, showing origin of
branches (enlarged).

5. Front view of terminal branches (enlarged).

6. Two pairs of hjdrothecae (greatly enlarged).

7. 8, 9. Gonangia (enlarged).

(216)

Proc. Wash. Acad. Sci. Vol. Ill

Plate XXV I

ALASKA HYDROIDS

PROCEEDINOS

OF THE

WASHINGTON ACADEMY OF SCIENCES

Vol. III., pp. 217-223. July 13, 1901.

PAPERS FROM THE HARRIMAN ALASKA
EXPEDITION.

XXII.

ENTOMOLOGICAL RESULTS (14):

THE ODONATA.

By Rolla p. Currie.

The Odonata collected by Professor Trevor Kincaid during
the Harriman Expedition comprise eight species, represented by
eighty-nine specimens. None of these appear to be new ; yet
the collection is interesting in that it serves to extend the known
range of certain forms and to add to our knowledge of the
Odonata fauna of Alaska.

No paper on the dragonflies of this territory has been
hitherto published; and only three species, additional to the
eight collected by the Harriman Expedition, have been pre-
viously reported from Alaska. These are JEnallagma annexum
(Hagen), yEschna sitchensis (Hagen) and ^schna clepsydra
(Say), thus making eleven species.

Family A GRIONID^^.

Subfamily AGRIONIN^.

LESTES species (?).

One female belonging to some species of this genus was taken at

Fox Point, July 26-28. The species of Lestes are separated, mainly,

by the characters of the male abdominal appendages, and females taken

Proc. Wash. Acad. Sci., July, 1901. (217)

2l8 CURRIE

singly cai^not always be determined with certainty. The following
description of the specimen is given for future reference :

Length 34 mm.; abdomen 26; hind wing 20.7; pterostigma 1.7.
Blackish-brown. The following parts are pale yellowish, viz : under
parts of head, lips, genas (except a few spots), rlijnarium, tips of two
basal joints of antennjB, mid-dorsal thoracic carina, humeral stripe
(which is wider below), sides of thorax inferiorly (except 2d lateral
suture above and some spots near coxte), and venter of thorax (except
middle line v^^hich is wider behind) . Abdomen yellowish, marked above
with dark metallic green as follows : on segment i , a narrow longitudinal
median line, widening behind, and divided into two large divaricate
spots ; on 2, a mid-dorsal band, divided by a median line which is widened
in front, the band widened and orbicular posteriori}' (enclosing a pale
spot on either side) and connected with the dark apical ring by a short
stalk ; 3-5 similar, but the dividing line uniformly narrow, the band
forming a point anteriorly and barely reaching the base ; 6 without
posterior pale spots, the bands otherwise as in 3-^ ; on 7 and 8 the
band is wider, not separated, not stalked behind; on 9 and 10 they
narrow posteriorly; the bands on 6-10 are blackish-brown, hardly
metallic. Sides of abdomen with indistinct dark markings, these
principally apical. Venter dark. Tenth segment with its apical
margin triangularly notched above, scarcely elevated. Anal append-
ages two-thirds the length of 10, conical, pale, black on apical half.
Genital valves pale, their edges dark apically and minutely denticulate ;
valvular processes dark, about as long as appendages. Legs pale, the
femora without, tibiae within, and the tarsi, black. Pterostigma
brown.

ENALLAGMA CALVERTI Morse.

Enallagma calverti Morse, Psyche, vii, No. 227, p. 208, March, 1895.
(Male described.) — Williamsom, Ent. News, xi, No. 5, p. 455, text fig,
b, PI. IX, figs. 5, 9 and 10, May, 1900. (Female described and male ap-
pendages figured.)

Kukak Bay, June 29- July 5 ; Cook Inlet, July 2 1 ; Kadiak, July
20 ; Juneau, July 25 : thirty-three males and nineteen females.

These specimens average larger than those I have seen from the
United States: Length, <? 30.7 mm. -35 mm., 9 3I-5-35-6; abdomen
S 24-27.3, 9 24.8-27.3; hind wing $ 18. 7-21.8, 9 20-22.2. In the
markings of head, thorax, and abdomen they agree fairly well with JS.
cyathiget'um (Charpentier) as described by Selys^ and exhibit similar
variations in the shape and extent of the dorsal abdominal markings,
as follows :

iBuU. Acad. Belg. (2), xli, p. 505, 1S76.

THE ODONATA 219

Males: On segment i, basal spot usually quadrate, narrowed and
more or less rounded on sides behind, its hind margin sometimes
excised but usually straight or nearly so and generally with a small
median triangular projection ; on segment 2, transverse posterior spot :
171 f7'07it^ convex and more or less rounded, sometimes pointed medially ;
behind^ convex, or straight, or sinuated and pointed ; sometimes
united with apical margin by a short stalk which varies greatly
in length, width and shape ; sometimes connected with an inferior
lateral stripe present in some specimens ; on segments 3 to 6, spots
more or less pointed anteriorly, widened and usually rounded pos-
teriorly, broadly united with apical ring ; spot on segment 3 covering
posterior ■§■ to i ; on 4 covering i to f; on 5, f to |-; on 6, f to f.
The spot on segment 7 covers apical |- or more. Dorsum of seg-
ment 10 entirely black. Other irregularly placed spots are some-
times present.

JPemales : Spot on segment i similar to that of the males ; on 3 quite
variable in shape and size, the narrowed portion not always reaching
the anterior margin, sometimes bifid ; posterior widened portion
rounded or angulate laterally ; connected directly, or by a short wide
stalk, with the apical ring. Spots on segments 3 to 7 similar to those
on male but longer, covering apical |^ or more on all of them. Spot
on segment 8 exceedingly variable in shape, broadly or narrowly
pointed anteriorly, widened behind, connected with posterior margin
directly or by a stalk, in one specimen not reaching posterior margin ;
widened portion sometimes connected with a lateral spot or stripe.
Spots on segments 9 and 10 covering entire dorsum, narrowing behind.

The postocular spots (in both sexes) are large and in most specimens
denticulate behind as in E. anizexiii7i (Hagen), the denticulations
sometimes connected with the blue of rear of head. The male appen-
dages agree well with those of calverti as described by Morse and fig-
ured by Williamson.

These variations seem to be individual rather than geographical,
although the males from Juneau all have very long dorsal spots on the
abdominal segments.

This species, described from Nevada, has also been reported from
Massachusetts, Maine, Wyoming and Washington.

Family ^ESCHNID^^.
Subfamily ^SCHNIN^E.

^SCHNA JUNCEA (Linn.).

Libellula pincea Linn., Syst. Nat., i, p. 544, n. 10, 1758.
^schna juncea Kirby, Synonymic Cat., p. 87, 1890.

220 CURRIE

One male from Kadiak, July 20 ; one female, Unga Island, July 17,
collected by Prof. W. E, Ritter. The National Museum also contains
one male and three females from Nushagak River, Alaska, collected
by McKay (No. 97).

This species occurs throughout the northern portions of Europe,
Asia, and North America.

The pterostigma is brown, paler in the female specimen.

^SCHNA CONSTRICTA (Say).

^shna constricta Say, Journ. Ac. Phila., viii, p. 11, 1839.

^schna constricia Kirby, Synonymic Cat., p. 88, 1890. — Calvert, Trans.

Am. Ent. Soc. , xx, p. 249, 1893. — Kellicott, Bull. Ohio State Univ.

(4), No. 5, p. 83, March, 1899. — Williamson, 24th Ann. Rep. Dept.

Geol. and Natural Resources, Ind., 1889, p. 305, PL iv, fig, 10, PI. vii,

figs. 14 & 15, 1900.

Two males, Kadiak, July 20, and Cook Inlet, July 21 ; two females,
Kadiak, July 20.

This is a common species in the United States and has also been re-
corded from British Columbia, Kamchatka, and Siberia.

The pterostigma in the males is black ; in the females yellowish-
brown.

Family LIBELLULID.^.

Subfamily CORDULIN^.

CORDULIA SHURTLEFFI Scudder.

Cordnlia shurt/effi Scvudeh, Proc. Best. Soc. Nat. Hist., x, p. 217, 1866. —
Selys, Bull. Acad. Belg. (2), xxxi, p. 265, 1871. — Kirby, Synonymic
Cat., p. 52, 1890.

One female, Kukak Bay, June 29-July 5 ; one female. Fox Point,
July 26, A female, labeled ' Alaska ' and determined as this species,
by Dr. P. P. Calvert, is in the collection of the U. S. National Mu-
seum; it was i-eceived from the U. S. Coast and Geodetic Survey,
through Professor T. C. Mendenhall, Superintendent.

This species has also been reported from Nova Scotia, Great Slave
Lake (Fort Resolution), New Hampshire, Maine and Pennsylvania.^

The wings are subfumose in the specimen from Fox Point.

SOMATOCHLORA ALBICINCTA (Burmeister).

Epophtkalmiaalbicincta'Qv'RM-E.lST'E.^, Handb. Ent., 11, p. 847, 1839.

CorduUa albicincta Hagen, Syn. Naur. N. Am., p. 138, 1861.

Cordulia eremita ScuDDER, Proc. Best. Soc. Nat. Hist., x, p. 215, 1866. —

Scudder. Proc. Best. Soc. Nat. Hist., xi, p. 300, 1867.
Epitheca albicincta Selys, Bull. Acad. Belg. (2), xxxi, p. 303, 1871.
Somatochlora albicincta Kirby, Synonymic Cat., p. 49, 1890.

^Ent. News, ix, p. 184, September, 1898.

THE ODONATA 221

Eight males and one female, Kadiak, July 20; one female (L. J.
Cole, collector), Kadiak, July 4; one male and two females, Kukak
Bay, June 39-July 5.

Length, $ 47 mm. -49 mm., 9 49—50; abdomen with appendages,
^ 34-36-8, 9 35-36.4; hind wing, $ 29.4-31.6, 9 31-32.2; ptero-
stigma, $ (^ 2.5-3.5; appendages, ^ 3, 9 3.4. The wings in the
males are almost clear hyaline ; in the females they vary from a clear
hyaline to a distinct subfumose. Triangles on anterior wings free or
crossed by one nervule ; on posterior wings, free (crossed by one ner-
vule in one male and one female) ; sometimes crossed on right wing,
free on left and vice versa. The luteous lateral thoracic spots are
sometimes very small or entirely absent.

This species was described from Labrador by Burmeister, while
Scudder's eremita was from Hermit Lake, in the White Mountains of
New Hampshire. Kirby, in his synonymic catalogue, gives it also
as occurring in Alaska.

Subfamily LIBELLULIN^E.
LEPTETRUM QUADRIMACULATUM (Linn.)

Libelhda qicadrimaculata Linn., Syst. Nat., i, p. 543, 1758. — Calvert,
Trans. Am. Ent. Soc, xx, p. 258, 1893. — Kellicott, Bull. Ohio State
Univ. (4), No. 5, p. 100, March, 1899. — Williamson, 24th Ann. Rep.
Dept. Gaol, and Natural Resources, Ind., 1899, p. 331, 1900.

Leptetrum quadrimaculatum Kirby, Synonymic Cat., p. 27, 1890.

Two males from Fox Point, July 26-3S. This species is found
throughout the Northern Hemisphere.

The wings in these two specimens are not clear hyaline, where
unmarked, but have a subfumose tinge.

LEUCORHINIA HUDSONICA (Selys).

Libellula hudsonica Selys, Revue des Odonates ou Libellules d' Europe, Mem.

Soc. Roy. Sci, Liege, vi, p. 53, 1850.
Diplax hndso7iica Hagen, Syn. Neur. N. Am., p. 180, 1861.
Leucorhinia hudsotiica Hagen, Proc. Bost. Soc. Nat. Hist., xviil, p. 78,

1875. — Kirby, Synonymic Cat., p. 12, 1890. — Hagen, Trans. Am, Ent.

Soc, XVII, p. 233, PI. X, figs. 13 and i2, a Sc b, 1890.
Leucorhinia Aao-eni CAhVEKT, Trans. Am. Ent. Soc, xvii, p. 36, PI. v, figs.

2-4, 1890.

Kukak Bay, June 29-July 5, two males; Virgin Bay, Prince Wil-
liam Sound, June 25-26, six males and five females.

These specimens agree very well with Calvert's description (Z.
hageni) . The color of thoracic dorsum and sides varies in both sexes
from yellow to reddish-brown, this variation being, probably, partly

222 CURRIE

due to difference in age ; the majority of the males are reddish-brown
and of the females, yellow. There is considerable variation, also, in
the extent of the basal spots of anterior and posterior wings ; in two
of the females the triangular spot of hind wings extends to the inner
angle of the triangle ; in two other specimens, also females, the first
(anterior) spot of front wings does not reach the first antecubital ; in
the males the first (anterior) spot of hind wings reaches not more than
half way to the first antecubital, except in two specimens and in
neither of these does it quite reach it. The dorsal abdominal spots
are pointed behind, not always reaching the hind margin of the seg-
ments. Several specimens have a small basal spot on segment 8.
The venation is quite irregular; 6-8 antecubitals in front wings of
males (regularly 7), 7 in females; 6-9 postcubitals. On posterior
wings, 5-6 antecubitals, 6-10 postcubitals. Two or three rows of
discoidal areolets on front wings. Discoidal triangles of both wings
free or crossed by one vein. Internal triangles free or composed of
two or three cells.

This species has been recorded from Lake Winnipeg, Saskatchewan
River and Fort Resolution, British America, from Massachusetts, and
from Pictou, Nova Scotia.

The following are notes on four species of dragon flies previously
reported from Alaska but of which the Harriman Expedition collected
no specimens.

ENALLAGMA ANNEXUM (Hagen).

Agrio7i annexum Hagen, Syn. Neur. N. Am., p. 87, 1861.

Enallagma cyathigerum race atinexiim Selys, Bull. Acad. Belg. (2), xli, p.

506, 1876. — Wadsworth, Ent. News, in, No. i, p. 8, January, 1892.
Enallagma annexum Williamson, Ent. News, xi, No. 5, p. 454, text fig. a,

PI. IX, figs. 3 and 7, May, 1900.

Distribution. — Sitka, Alaska (Hagen) ; Red River and Saskat-
chewan, British America; (.^) Irkutsk, northern Asia; Vancouver
Island; Hermit Lake, White Mountains, New Hampshire; Massa-
chusetts (Selys) ; Manchester, Maine (Miss Wadsworth) ; Wyoming,
California, and Washington (Williamson).

^SCHNA SITCHENSIS Hagen.

^schna sitchensis Hagen, Syn. Neur. N. Am., p. 119, 1861. — Hagen, Proc.
Bost. Soc. Nat. Hist., xviii, p. 33, 1875.— Kirby, Synonymic Cat., p. 87,
1890. — Hagen, Psyche, v, No. 170, p. 353, June, 1890.

Distribution. — Described from Sitka, Alaska, and afterwards re-
ported from Saskatchewan, British America.

THE ODONATA 223

^SCHNA CLEPSYDRA (Say).

^shna clepsydra Say, Journ. Ac. Phila.,viii, p. 12, 1839.

yEschna clepsydra Hagen, Syn. Neur. N. Am., p. 122, 1861. — Kirby,
Synonymic Cat., p. 89, 1890. — Calvert, Trans. Am. Ent. Soc, xx,
p. 248, 1893. — Kellicott, Bull. Ohio State Univ. (4), No. 5, p. 84,
March, 1899. — Williamson, 24th Ann, Rep. Dept. Geol. and Natural
Resources, Ind., 1899, p. 305, PI. vii, figs. 12 and 13, 1900.

yEschna ereinita Scudder, Proc. Bost. Soc. Nat. Hist., x, p. 213, 1866.

Distribution. — Occurs commonly throughout eastern North Amer-
ica ; has also been reported from Dakota and the Saskatchewan in the
Great Plains region ; and from Irkutsk and the Wilvii River, Siberia,
and from Finland.

ANAX JUNIUS (Drury).

Libellula jimia Drury, 111. Nat. Hist., i, pi. 47, fig. 5 ; App. Vol. 11, 1773.

yEshiia JHiiia Westwood, 111., Exot. Ent., by Drury, Westwood's Ed., p.
116, pi. 47, fig. 5, 1837.

yEshna Junius Say, Journ. Acad. Nat. Sci. Phila., viii, p. 10, 1839.

yEschna junta Burmeister, Handb. Ent, 11, pt. 11, 2d half, p. 841, 1839. —
RAxMBUR, Hist. Nat. Ins., Neur., p. 196, 1842.

Anax spintferus Kawbvk, Hist. Nat. Ins., Neur., p. 186, pi. i, fig. 14, 1842.

Anax Junta Selys, Revue des Odonates ou Libellules d' Europe, Mem. Soc.
Roy. Sci. Liege, vi, p. 328, 1850.

Anax Junius Hagen, Syn. Neur. N. Am., p. 118, 1861. — Hagen, Verhandl.
Zool.-Bot. Gesells. in Wien, xvii, p. 33, 1867. — Brauer, Reise der
Oesterr. Fregatte Novara um die Erde, Zool. Theil 11, i Abth., A,
Neuropt., p. 62, 1868. — Calvert, Trans. Am. Ent. Soc, xx, p. 249,
1893. — Kellicott, Bull. Ohio State Univ. (4), No. 5, p. 77, March,
1899. — Williamson, 24th Ann. Rep. Dept. Geol. and Natural Resources,
Ind., 1899, p. 306, 1900. — Holland, Ent. News, xi. No. 3, p. 382,
March, 1900.

Distribtction. — North Am^erica, Cuba, Kamchatka, China and
Hawaiian Islands. Dr. Holland, in the last citation, records one
young male from the mountains between Mission and Fortymile
creeks, Alaska, collected July 35, 1899, by Rev. S. Hall Young, and
one male from Eagle, Alaska, taken August 3, 1S99, by J. Murray
Presnall.

PROCEEDINQS

OF THE

WASHINGTON ACADEMY OF SCIENCES

Vol. Ill, pp. 235-266. PLS. xxvii-xxx July 13, 1901.

PAPERS FROM THE HARRIMAN ALASKA
EXPEDITION.

XXIII.

THE ASCIDIANS.

By Wm. E. Ritter.
University of California.

In view of the wealth of the Ascidian fauna of the Pacific
Coast from Puget Sound southward, and of the number of
specimens received in chance collections from Alaskan waters, it
was something of a surprise that these animals were not found
in greater abundance at the points visited by the Harriman Ex-
pedition. The most diligent search was made in localities
where the conditions appear to be as favorable as at places on
the California Coast where they flourish in great luxuriance,
both as to number of species and of individuals, and yet only
the fifteen species embodied in this report were secured, and
the number of individuals of each species was usually small.

The localities at which Ascidians were obtained and the
species from each locality are as follows :

Sitka. — Alolgula graphica ; Psain7naplidiitm peduticulaUnn.

Glacier Bay. — Ascidia adhcereiis.

Yakutat Bay. — Ascidia adhce?-e?ts ; Cyjitkia villosa ; Styela yaku-
tatensis ; S.truncata; Distoma ptclchra.

Prince William Sound. — Rhizoinolgida arenaria ; A?naroticium
t ra nslu cidu m .

(225)

226 RITTER

Kadiak Island. — Distaplla occidejitalis ; D. confusa; Didemmnn
straizgulatuni ; AmarouciuJii coei.

Shumagin Islands. — Ascidia adhcerens ; A^naroucium californi-
cum; Botryllus magjttis.

Seven species were obtained only by dredging ; and as the
dredge was used but little and not in depths exceeding twenty-
five or thirty fathoms, it seems certain that this method of col-
lecting, when more thoroughly applied in these waters, will
yield richer returns.

The largest number of species obtained at any one place was
five, at Yakutat Bay, where the group seems to be better repre-
sented than at any other collecting station. It must be said,
however, that the shore collecting was carried on there rather
more extensively than elsewhere, and under better conditions,
the tides being particularly favorable at the time.

The distribution of marine littoral animals, always a difficult
problem, is peculiarly intricate in the case of the Ascidians.
The conditions upon which their existence depends appear to
be more complex and delicate than for most other groups of
animals.

In the course of collecting along the Pacific Coast from
southern California to western Alaska during the last ten
years, I have been repeatedly surprised at finding Ascidians
almost wholly absent from places apparently differing in no re-
spect from others where they occur abundantly. Observations in
Alaskan waters have, however, shown that the influence of the
glaciers, discharging into the sea, are there an important factor.
Few Ascidians were found at any point where considerable
glacial debris is present in the water, and where sedimentation
is in rapid progress. In all probability the diminished salinity
of the water, due to fresh water derived from melting ice, is
a condition unfavorable to their existence.

Grateful acknowledgment is here made to two of my as-
sociates on the expedition, Dr. Wesley R. Coe and Professor
Trevor Kincaid, for their courteous help in collecting Ascidians.
Of the fifteen species obtained, eleven are here treated as new
to science, one of them being the type of a new genus. The
complete list is as follows :

THE ASCIDIANS 227

Ascidia adhterens sp. nov.

Molgula graphica sp. nov.

Rhizomolgula arenaria gen. et sp. nov.

Cynthia villosa Stimpson.

Styela yakutatensis sp. nov.

S. truncata sp. nov.

Distoma pulchra sp. nov.

Distaplia confusa sp. nov.

D. occidentalis Ritter.

Didemnum strangulatum sp. nov.

Amaroucium translucidum sp. nov.

A. coei sp. nov.

A. californicum Ritter.

Psammaplidium pedunculatum Herdman.

Botryllus magnus sp. nov.
The material, including slides for microscopic study and dis-
sected preparations, is in the museum of the University of Cali-
fornia ; but, where the number in the collection would permit,
specimens of all the species have been placed in the National
Museum at Washington.

ASCIDIA ADHERENS sp. nov.
(PI. XXVII, figs. 1-5.)

General characters. — iVttached so closely by the whole left side
that the test on this side is much reduced in quantity and the detach-
ment of the animal is made difficult, a peculiarity from which the spe-
cies is named. Much flattened, elliptical in outline, generally regular,
surface for the most part even and smooth. A considerable number
of individuals, usually more or less run together edge to edge, the
whole forming a thick encrusting layer. Siphons distinct, particularly
in larger individuals, though never long. Lobes of siphons incon-
spicuous, though discernible, when orifices are fully expanded, by
small darker specks on the margin. Lobes best seen in smaller indi-
viduals. Usually seven on the branchial siphon, and six on the atrial ;
an occasional individual however, with six and five respectively.

Branchial siphon situated at the anterior end and on the edge ; the
atrial not remote from the branchial on the right upper surface of the
animal, somewhat removed from the edge (p1. xxvii, fig. 2). Color
entirely white (opalescent), except a tinge of yellow on the siphons,
and some deeper yellow in stomach distinctly visible through test.

228

RITTER

MEASUREMENTS OF SOME OF THE LARGER SPECIMENS.

No.

I^ength.

Width.

Thickness.

Distance
Between Siphons.

mm.

mm.

mm.

mm.

I

28.5

18.5

II

13-5

2

26

19

II

7

3

30

22

16

12

4

27

19

H

8

5

25

16

9

II

6

24

16

9

. 8

7

19

13

7

8

8

17

12

7

7

9

17

II

5

5

Test of moderate thickness, containing a network of vessels and
many large vesicular cells. Mantle nowhere highly developed, but
particularly thin on side of attachment where the muscle fibers are al-
most entirely wanting. Muscle fibers of mantle not disposed in any
definite bundles, but spread out in a uniform sheet and running in all
directions.

Branchial apparatus. — (Siphons described above.) Branchial
tentacles variable in number, even in the same cluster of individuals ;
thirty-five or more in the larger specimens.

Hypophysis mouth rather large, horseshoe-shaped, about as broad
as long, limbs incurved somewhat but not produced into spirals, open-
ing nearly directly forward. Ganglion and gland close together,
forming an elongated mass, lying not far behind the hypophysis (p1.
XXVII, fig. 4).

Dorsal lamina a rather broad membrane provided with transverse
ribs which project somewhat beyond the free edge of the membrane.
No serrations of the edge in addition to these projecting ribs (p1. xxvii,
fig- 4) • Lamina extending for some distance behind the esophagal
opening, on the right side.

Folding of the branchial sac not deep, but distinct. Internal longi-
tudmal vessels bearing prominent, though simple, papillae placed at
the points of crossing of the transverse vessels ; intermediate papillae
also frequently present, /. c, papillge in the intervals between the
transverse vessels. Usually about two folds of the sac and about ten
stigmata between each two longitudinal vessels.

Digestive apparatus. — Mouth of the esophagus far back, though
anterior to the posterior end of the dorsal lamina. Stomach rather
prominent, cask-shaped, little longer than broad ; wall with about six
distmct, regular, longitudinal folds on its outer half. Loop of intes-
tme long, reaching more than two-thirds of the way to the anterior end

THE ASCIDIANS 229

of the endostyle, with which it is nearly parallel. Anus hardly as far
forward as the anterior end of the loop (p1. xxvii, fig. 3). The clear
' renal vesicles ' very abundant along the entire length of the intestine,
and extending on the mantle to some extent, but not covering up and
obscvxring the intestine.

Reproductive organs. — In the intestinal loop, not very voluminous.
Vas deferens large and distinct, running along the dorsal side of the
rectum.

This Ascidia has so many characters in common with Ascidia
koreana Traustedt that in a former paper (1900) I identified, with con-
siderable hesitation, a few specimens from Puget Sound with Trau-
stedt's species ; but now, with ample material at hand, it is clear that
the American and Asiatic forms are distinct, though closely related.

Traustedt says of the general form of A. koreana : " Corpus ut conus
inversus formatum est, infra subacuminatum et supra pi"aecisum." ^
This is so different from that of A. adkcerens and the differences in the
intestinal tract and the dorsal lamina are so great that it is impossible
to regard them as belonging to the same species. The intestinal loop
of koreana is neither so long relatively, nor so nearly parallel with the
endostylic edge of the sac as that of adhcerens ; and apparently since
Traustedt makes no mention of the point, the folding of the stomach-
wall does not occur in his species. The irregular serration of the edge
of the dorsal lamina in A. koreana does not occur in the American
species, as the description and figure show.

A word should be said with regard to the intermediate papilla of
the branchial sac; i. e., papilla on the internal longitudinal vessels in
the intervals between the crossings of the transverse vessels. These
apparently exist, as a rule, but they may or may not be present even in
different portions of the same sac. It may be, however, that their
presence, as actual intermediate papillce, is sometimes apparent rather
than real, and that when present, this may be due to the suppression of
transverse vessels corresponding to them. In regions where they are
present (p1. xxvii, fig. 5) the alternate transverse vessels are much
smaller than the others, and one and the same vessel maybe interrupted
for a considerable distance while the papills corresponding to it remain.
On the other hand by carefully examining an area in which the inter-
mediate papillae are present, fragments of vessels corresponding to them
are found here and there (pi. xxvii, fig. 5), so that it may be doubted
whether the papillae are ever present except where transverse vessels
actually do cross the longitudinal ones, or where they have existed, or
are going to develop.

^Vidensk. Meddel. fra den naturh. Foren, p. 49, 1884.

230 RITTER

Taken in considerable abnndance at Popof Island, Shumagin group,
July S, 1S99 ; also in small numbers at Yakutat Bay and Glacier Bay ;
found on the under side of stones at low tide. Seventy-five or a hun-
dred specimens in the collection. Occurring also in Puget Sound.

MOLGULA GRAPHICA sp. nov.
(PI. xxvii, figs. 6-9.)

General characters. — Regular in outline, nearly spherical, though
somewhat higher than wide and considerably compressed laterally.
Attached by small area at posterior end. Siphons not prominent,
about 15 mm. apart. Surface of test somewhat uneven, though with-
out pronounced interruptions of any sort, no foreign bodies adhering.
Qiiite transparent, but the numerous, fine, much-branched, almost jet
black vessels showing through distinctly in all parts, giving the animal
the very characteristic appearance indicated by its specific name (p1.
XXVII, fig. 6).

Measurements of the only specimen secured : length 3S mm., width
30 mm., thickness 22 mm.

Test. — Thick and firmly gelatinous. Thickness in middle region
of body 6 mm. ; somewhat more than this at anterior end, and less
posteriorly. The black vessels within it confined mostly to the outer-
most layers ; the test cleaves away from the mantle very easily ; its
inner surface with a thin stratum, distinctly more opaque. Mantle
rather thin, of nearly uniform thickness throughout ; musculature of
the usual molguloid kind.

Branchial apparatus. — Siphons (with animal removed from test) :
branchial short and broad with six small almost filiform lobes ; atrial
two or three times as long, considerably narrower, tapering at the end,
with fovn- lobes. Branchial tentacles, about eight, very large, bushy
ones, alternating with about as many more of similar size (p1. xx\'ii,

%• 9)-

Dorsal tubercle prominent, hypophysis mouth horseshoe-shaped,
opening turned to the right, horns rolled in with about two spirals
(p1. xxvii, fig. 8). Endostyle not large; dorsal lamina a plain, rather
broad membrane, though with suggestions of a few remote processes
on the edge. Branchial sac with seven well-developed and nearly
equal folds on each side, from twelve to fourteen longitudinal bars on
each fold, no bars between the folds. Transverse vessels very irregular ;
the largest and most constant of them placed between the deep infun-
dibulse of the sac which reach into the folds ; many finer vessels, ir-

THE ASCIDIANS 23I

regular in direction and length, between the larger ones, on the surface
of the infundibulte ; as a rule, the larger vessels only crossing from
one fold to the next.

Stigmata long but irregular, rarely disposed in spiral form, so large
and numerous as to leave almost no expanse of branchial membrane
unperforated, its whole substance being composed of the narrower
bars between them. Considerable dark pigment along the edges of
the branchial folds and on the tentacles.

Digestive apparatus. — Situated on left side, esophagus very short,
stomach relatively small ; intestinal loop narrow and long, extending
entirely across the posterior end of the animal : intestine of nearly
uniform diameter throughout its length ; anus somewhat trumpet-
shaped, adherent.

Reproductive organs. — Situated on both sides of the body, ovary
on each side a single elongated cylindrical mass extending crosswise
of the animal ; in the single specimen at hand, testis apparently
wanting.

This species is closely related to M. gregaria Lesson, but differs
from it in possessing the black, much-branched blood vessels in the
test, and also in the structure of the branchial sac. Herdman,^ who
has fully redescribed M. gregaria., states that the folds are narrow,
the internal longitudinal bars few, and that broad horizontal membranes
attached to the internal longitudinal bars run along the chief transverse
vessels. Moreover, the stigmata are rather small and are arranged in
spii'als. In all these particulars the two species differ distinctly.

The single specimen was dredged at Sitka in about ten fathoms, on
June 15, 1899.

RHIZOMOLGULA gen. nov.

Pedunculate, the peduncle long and slender, bearing many
rootlets. Branchial sac with six folds on each side. Gonads
on the left side only, situated within the intestinal loop.

RHIZOMOLGULA ARENARIA sp. nov.

(PI. xxviii, figs. 10-192.)

General characters. — Peduncle long and slender, giving off many

filiform branches, or root hairs, which sei^ve as hold-fasts to anchor

the animal to the sand. Body nearly spherical in form, though hav-

1 Herdman, Report on the Tunicata of the Challenger Expedition, Part i,
1882.

232

RITTER

ing the antero-posterior diameter somewhat longest. Very regular in
outline and unbroken in surface. Test thin, entirely covered by a thin
uniform layer of closely adhering sand grains, clinging to the test sub-
stance itself and not to processes growing from it (p1. xxviii, fig. lo).
Frequently several individuals cling together by their peduncles, but
this is apparently due to the tangling of the slender peduncles with
one another, and not to actual fusion. Peduncle frequently broken off
at or near body.

MEASUREMENTS.

No

Height.

Greatest

lycngth of

Trans. Diam.

Peduncle.

mm.

mm.

mm.

I

13

21

2

17

12

15

3

9-5

II

4

13

20.5

5

17-5

12.5

16

6

13

12.5

7

17

13

9

Usually somewhat laterally compressed.

Color uniform gray from the sand covering the surface ; test, except
for the sand, very transparent. A small area of chocolate color at and
around the branchial orifices.

Mantle very thin, separating readily from the test. A large num-
ber of small ectodermal processes projecting from its outer surface
toward the test, not, however, reaching into the test so far as ob-
served (p1. xxviii, figs. 19^ and 19^).

Musculature not highly developed. Longitudinal muscle fibers
grouped into very definite, regularly spaced bundles radiating from
the two siphons. Between the siphons a band of closely-placed par-
allel bundles reaching entirely across from one side of the animal to
the other. Circular fibers also disposed in bundles^ smaller and less
regular, however, than the longitudinal bundles. Circular fibers out-
side the longitudinal ones ; a thin but distinct layer of mesenchymatous
tissue between the two.

Branchial apparatus. — Siphons very short and inconspicuous, al-
most invisible to surface views, situated at the anterior end, nearly as
far apart as the width of the animal; branchial with six (occasionally
fewer) small lobes ; atrial with four lobes. Branchial tentacles
twelve to fourteen in number, branched, but not deeply or profusely.

Hypophysis mouth horseshoe-shaped, broader than long, opening
forward, somewhat to the left, horns not coiled (p1. xxviii, fig. 19).

THE ASCIDIANS 233

Ganglion very close to the hypophysis opening. Branchial sac with six
well developed, nearly equal folds on each side ; six or seven longitudinal
vessels on each fold ; none between the folds. No transverse vessels
present besides the imperfect septa between the infundibulae. Infun-
dibulae deep and rather narrow, the stigmata reaching entirely around
them; quite regular in size and arrangement (p1. xxviii, fig. 18).
Endostyle narrow. Dorsal lamina a plain broad membrane terminating
at the left of the esophagal opening.

Digestive apparatus. — Esophagal opening at the extreme postero-
dorsal angle of the branchial sac. Intestinal loop reaching entirely
across the animal ; rather broad at its ventral end, bvit closed together
dorsally where the rectal bend comes in contact with the esophagus.
Dorsal limb of rectum rather short, running close along the dorsal edge
of the animal.

Esophagus narrow, stomach nearly cylindrical, somewhat longer
than thick, set off from the esophagus by an abrupt shoulder, its wall
thrown into about ten or twelve distinct, closely approximated, some-
what tortuous, longitudinal ridges, of a light yellow color. Intestine
from stomach to anus of nearly equal diameter throvighout. Anus
unarmed (p1. xxviii, fig. 11).

Reproductive orgatts. — Present on left side only, and both ovary
and testis placed tvithin the intestinal loop, which they completely
fill. Ovary occupying the middle; testis in the form of a great num-
ber of nearly spherical masses surrounding the ovary and in close
contact with the intestine. Ovidvict very short, running along the
dorsal side of the rectum (p1. xxviii, fig. 11).

Renal organs. — Consisting of two small, somewhat elongated, but
irregular bodies situated one on each side of the body far back and
clinging to the inner surface of the mantle.

I have not observed the ectodermal processes or papillae on the outer
surface of the mantle (p1. xxviii, figs, 19^ and 19^) in any other Asci-
dian, nor have I found anything like them described in any other
species, though they are undoubtedly homologous with the ectodermal
portion of the processes occurring on many species of Molgula.

They are scattered over the entire surface of the mantle, but are
somewhat more abvmdant at the posterior end. Each one consists of
a wall of distinct epithelial cells, and a lumen which communicates
freely with the body space, and into which the blood passes. The
ends are somewhat bulbous owing to the larger size of their epithe-
lial cells.

The peduncle, with its system of rootlets and holdfasts, is a remark-

Proc. Wash. Acad. Sci., July, 1901.

234

RITTER

able structure. Its resemblance to the branching ' tap-root ' of some
plants Is striking enough. The peduncle proper is usually as long as
the body of the animal or longer. It is always thread-like, or lash-
like. At its junction with the body it is considerably larger in diam-
eter than it is elsewhere, but it diminishes abruptly to the small size
which it holds for most of its length.

The root-hairs are not scattered vmiformly throughout the length of
the peduncle, but are in the main grouped in tufts with a few single
ones given off here and there. Sand grains cling tenaciously to the
hair tufts and increase the resemblance to a plant root (p1. xxviii, figs.
lO and 13).

The minute structure of the peduncle and rootlets is peculiar, and
quite unexpected in view of the structure of similar growths in other
Ascidians. The smallest of the root-hairs, examined with a high power,
have a striking resemblance to a medullated nerve fiber. Each con-
tains a central fiber, homogeneous in appearance and of regular out-
line that would correspond to the axis cylinder ; and a transparent,
structureless outer portion forming the greater bulk of the whole, that
would correspond to the medullary substance (p1. xxviii, fig. 15).
Each of these rootlets, if traced distally, is found to terminate in a more
or less bulbous enlargement, the main structural element of which is
the thickened and closely tortuous end of the central fiber. The outer
transparent substance of the bulb does not differ essentially, either in
structure or relative amount, from that of the intermediate portions of
the rootlet. In some instances the bulbs are merely slight enlarge-
ments of the rootlet, the length being considerably greater than the
thickness. In others again they are thicker than long and are set off
from the rootlet by a pronounced shoulder. In these latter cases the
central fiber is often converted into a considerable mass by being itself
somewhat enlarged, and by being thrown into close windings.

By tracing the rootlets, proximally to their origins in other secon-
dary roots (p1. xxviii, fig. 15) or in the main stem (p1. xxviii, fig. 14)
one finds that the central fibers enter the main stem, either directly or
by continuing to it through a larger rootlet, and then run on, preserv-
ing their individuality, up through the stem to the body of the animal
(p1. xxviii, fig. 14). Thus it follows that the peduncle, at its junction
with the body, contains a very large number of these fibers — as many
as the entire number of rootlets.

The nature of the axial fibers is not obvious at first sight. Atten-
tive examination, however, proves them to be muscle fibers. The
chief difficulty in the way of immediately recognizing their true nature

THE ASCIDIANS 235

is the fiict that in many places they are contracted into a more or less
regular spiral, the coils of which are so close that, seen as they always
are through the material in which they are imbedded, they appear to
be made up of discrete bodies which may easily be mistaken for cells.
By staining them in various ways and clearing them in glycerine and
clove oil, and, especially, by following the individual fibers far enough
(for often they may be traced all the way into the body of the animal)
their real character can be determined beyond question. One point
concerning them, however, I have not been able to make out with cer-
tainty. In the terminal bulbs are frequently seen what appear to be
definite loops formed by the central fibers, and in places between the
bulb and the point of branching the fiber sometimes appears to be dou-
ble. Both these facts suggest the possibility and support the belief
that each fiber is double ; that is, extends down the rootlet to the end
and there turns back upon itself. These fibers are, without much
doubt, fibers from the mantle that have grown into the peduncle first of
all, and then out into the rootlets ; and it would seem probable that in
this process they have looped in rather than grown in at a free end.
One may imagine that the thing was accomplished by a process similar
to the looping- out of the mantle fibers on the atrial languet in some
species of Amaroucium or Distaplia for example, or over the incuba-
tory pouch of Colella. The evidence for this interpretation, however,
is not conclusive. In many places in the rootlets the fibers are cer-
tainly not double when fully mature ; whether the two strands may
have become fused into one here, or one of them become wholly degen-
erated, I am unable to say.

The most puzzling thing in the structure of the rootlets is the fact
that the muscle fibers are actually ejubedded in the test, and are not sit-
uated in a cavity with an epithelial wall derived from the ectoderm
of the mantle. From all analogies one would expect the rootlets to be
in reality ectodermal vessels, each with a thin layer of test. I have,
however, searched in vain for epithelial cells of any sort either in the
rootlets or the larger portions of the peduncle. Even after staining
with htematoxylin, which can always be relied on to bring out the
nuclei in epithelial tissues of the Ascidians, not even a suggestion of
nuclei has appeared in these structures. All the outside transparent
material is apparently test and nothing else. Moreover, by splitting
the peduncle at its larger, proximal end next to the body in the plane
of the fiber bundles, one obtains conclusive evidence that the muscle
fibers are fully surrounded by the test material. In such a place as
that shown in p1. xxviii, fig. 16, for example, one or two fibers are

236 RITTER

separated out distinctly from the main bundle, and the test matter with
its scattered test cells can be clearly seen, fully enveloping the fibers.
It can be confidently asserted that, in most cases, no lumen is present
in the ultimate rootlets in which blood may flow.'

In all probability these are all developed at first as ectodermal ves-
sels, and the ectodermal layer and lumen of the vessel become sec-
ondarily obliterated. It is difficult to see by what other method the
conditions could be reached. The only direct confirmatory evidence
that I have been able to get, however, is the fact that in one rather
small specimen from which most of the peduncle was wanting, because
of having been broken off, as I at first supposed, the rather short re-
maining piece was found to contain an outpocketing of the ectoderm
of the mantle, with the epithelial layer showing very distinctly. No
muscle fibers were present, however, in this instance, and this fact to-
gether w4th the failure to find, on closer examination, evidence that
the peduncle had been broken, leads me to believe that this was an in-
cipient peduncle into which the muscle fibers had not yet penetrated,
and from which the ectodermal layer had not begun to degenerate.
Before leaving this subject, so well deserving further study on young
animals, I would draw particular attention to the appai-c7tt gradual
transition of the test cells themselves, into the muscle fibers, as shown
in p1. XXVIII, fig. 17. This figure is drawn from a peduncle near the
body and might be duplicated from any specimen. It is, of course,
impossible to afiirm that there has been a production of muscles fibers
from test cells, as the adult structure here suggests. In view, however,
of the fact that the test cells are now known to be of mesenchymatous
origin it is not difficult to suppose that the development has taken
place as the structure actually found indicates. Muscle fibers might
very well be formed, probably are formed, from mesenchyme cells in
the vessels while these still exist. But the mesenchyme cells migrate
through the ectoderm to become test cells. With the disappearance
of the ectoderm, then, mesenchyme and test would be in direct con-
tact, and what would be essentially a transition from test cells to mus-
cle cells would then be possible.

^On consulting the classical work of Lacaze-Duthier,i874, on the Molgulidae,!
find the following concerning the villosities of the test. One may make out the
nuclei of the walls of the canals, the author says, " mais on ne distingue jamais
aussi bien qu'apres avoir employe les imbibitions ou les autres manipulations
histologiques, ces cellules regulierement poljedriques." It is possible therefore
that further study of the root hairs of Rhizomolgula on fresh and specially pre-
pared material will bring out the epithelium, but from my experience it seems
hardly probable.

THE ASCIDIANS 237

The already rather large number of genera of the Molgulidcs^ sev-
ei*al of which are not very clearly defined and contain only a few
species, has made me very reluctant to add another one-species genus.
From this course, however, there seems to be no escape. Of the pre-
viously established genera, Etigyriopsis^ proposed by Roule in 1SS5,
and modified by Pizon in 1S9S, would receive this Alaskan species
with less forcing, perhaps, than would any other genus. This group,
regai'ded by Roule as a subgenus of Molgula^ was established for
species with a single sexual gland situated on the left side of the body,
outside and i?t fro7it of the intestinal loop. The last part of the
definition, it will be observed, would stand in the way of the admission
of the species under discussion. But Pizon has very I'ecently (1S9S)
concluded, after a consideration of the classification of the Molgulid(^
based upon the examination of a considerable number of genera and
species, that Etigyriopsis should be regarded as a genus including all
species with gonads on the left side only ; and with curved stigmata,
inore or less regular, and more or less elongated, but not arranged in
the great infundibulse characteristic of Eugyra, The situation of the
gonads within or without the intestinal loop, he considers a matter not
of generic importance, and in this view I am disposed to coincide.
The gonads are so inconstant in this family, particularly as to position,
that the small difference here considered, at any rate when taken by
itself, seems insufficient to establish generic rank. If, however, it is
linked with the other characters, in which R. are7zaria differs from
Eugyriopsis^ viz. ; first, the possession of a pedvmcle, and second, the
presence of six folds in the branchial sac instead of seven, the coupling
of the two in the same genus is made obviously impossible.

Another genus which has considerable claim on our species is Bos-
trichobranchiis Traustedt. Here the single gonad is situated on the
left side of the body and within the intestinal loop ; but the absence of
the peduncle and of true folds in the branchial sac, decides against the
admission of R. arejiaria to this genus.

The only pedunculated genus of the family is Ascopera Herdman,
1882; but the peduncle here is an entirely different thing from that
of Rhizomolgula being nothing more than a somewhat produced pos-
terior end of the body. Furthermore, the infundibulcE in the bran-
chial sac are lacking, and the gonads are present on both sides of the
body, so that Rhizomolgula is far removed from this genus.

The collection contains about thirty specimens all from Orca, Prince
William Sound, where they were taken on fishing lines by Dr. Coe
and Mr. Cole, on June 28, 1S99, ^^"^ about twenty fathoms.

238 RITTER

CYNTHIA VILLOSA Stimp.

Cynthia villosa Stimpson, Proc. Acad. Nat. Sci., Phila., Vol. 16, p. 160,
1864. — Herdman, Trans. Liverpool Biol. Soc, Vol. xii, p. 258, figs. 7-
II, pi. XII, 1898.

Herdman has recently described in detail this species, which Stimp-
son characterized by its external features only. The only serious
difficulty in identifying the Alaskan specimens now under consideration,
with the Puget Sound forms is in the number of folds of the branchial
sac. According to Herdnian, the latter possess eight on each side, and
his statement is confirmed by the examination of some large specimens
from Puget Sound, undoubtedly belonging to the same species, and
which were sent me by the Young Naturalists' Club of Seattle. The
Alaskan specimens, on the other hand, possess only six, with rudi-
ments of a seventh. These individuals are all small however, the largest
being only I3 mm. high by 12 mm. deep, and 8 mm. thick; and one
specimen of about the same size among those from Puget Sound has,
so far as I am able to determine, only six on each side. I deem it
probable therefore that this difference in the number of folds is due
to difference in age. There is some difference in the spines also,
these in the Alaska specimens being somewhat shorter, and more in-
clined to an arrangement in groups on papillae of the test. As a rule,
however, they are neither in groups nor on papillae ; so this differ-
ence is not constant. Herdman has fully discussed the relationships
between the echinated species of the genus that were known at the
time of his writing. Since then, however, Hartmeyer, 1899, has de-
scribed a new species, C. arctica^ to which he refers a form of C.
echlnata from Kara Sea, described by Traustedt. In the two partic-
ulars in which the Alaskan specimens differ from the typical villosa
they incline toward C. arcttca, for in this species the spines are situ-
ated in groups of eight or ten on prominent papillae of the test, and the
folds of the sac are seven, or occasionally eight, in number.

In the number of tentacles, however, the Alaskan form of villosa
agrees with the type and not with arctica ; and since, as shown above,
neither of its arctica tendencies is pronounced, it must be considered
most closely allied to villosa. Perhaps further study of more material
will show the necessity of establishing a new species for it ; but as the
matter now stands this course is not justifiable.

There are about fifteen specimens in the collection, all from Yaku-
tat Bay, taken on rocks at extreme low tide, June 20-22, 1899.

THE ASCIDIANS

239

STYELA YAKUTATENSIS sp. nov.

(PI. XXVII, figs. 20 and 21.)
General characters. — A pedunculate species with the following
measurements :

No.

Length.

Depth,
greatest.

Thickness,

Total.

Of body.

Of pedunc.

greatest.

mm.

mm.

ram.

mm.

mm.

I

57

38

20

15

12

2

3

4S
38

24

25

1.5
13

10
II

8
8

4

40

28

14

II

9

5
6

3i

21
14

II
8

9

7

7
5

7

38

22

17

10

7

8
9

5?

20

18

II
II

7
6

5
6

10

30

iS

12

9

8

11

20

13

7

8

7

12
13

21

II

15

5
10

5
6

%

14

21

12

8

7

6

15

22

14

9

5

4

16

22

15

7

7

6

17

29

20

9

ID

9

18

37

26

II

10

10

19
20

19

12
14

7
8

8
8

7
7

21

14

10

4

5

4

22

14

9

4

7

5-5

Mean

27.7

18.1

10. 1

8.2

6.8

These averages give :

Length of body = i.S X length of peduncle.
Depth of body = 1.3 x thickness of body.

Outlines are shown in p1. xxvii, figs. 30 and 3i.

Transition from body to peduncle usually, though not always, rather
abrupt. Surface but little corrugated or roughened, usually free from
foreign bodies. Siphons rather small and short, the branchial directed
forward, the atrial turned ventral ward.

General color light with a distinct tinge of yellow ; the •siphons and
a small area around them deep salmon, the color, however, entirely
disappearing in alcoholic specimens.

Test rather more delicate and transparent than usual for species of
this genus, the intestine showing through it even in nearly full grown
specimens.

Branchial apparatus. — Form and proportions of siphons described

240

RITTER

above. Both obscurely four-lobed, the lobes in preserved specimens
often imperceptible. Branchial tentacles, in the largest specimens at
hand, not much exceeding forty and not differing greatly in size.

In a specimen 15 mm. in total length (one of the smallest at hand)
the tentacles are as follows : Twelve of good size and nearly equal,
twelve considerably smaller, alternating with these, and, alternating with
these twenty-four, the merest buds of as many more. From a com-
parison of this condition with that found in the largest individuals it
appears that, typically, new tentacles are added, first twelve and next
twenty-four, until the individual is well grown, and that then the
smaller, later tentacles continue to grow until all of the circle are in old
specimens, practically, of the same size. Dorsal tubercle prominent.

Branchial sac with four folds on each side. Formula of internal
longitudinal vessels as follows :

Endostyle.

First.

Second.

Third.

Fourth.

Fifth.

Dorsal
I/amina.

l,ength of
specimen.

space.

fold.

space.

fold.

space.

fold.

space.

fold.

space.

(Count

55 mm.

57 "
?

3
3

5
5
4

4
6

5

10
6

8

4

5
5

6

10

9

5
7
4

H

3
3

5

just be-
hind ten-
tacles.)

The frequent uncertaint}' as to \\hether vessels at the bases of the
folds should be counted as belonging to fold or space, and the frequent
change in direction of the vessels with reference to the folds at differ-
ent levels of the same sac, render it impossible to rely on this charac-
ter, except within rather wide limits. This remark applies especially
to small folds.

Endostyle narrow and tortuous. Dorsal lamina a narrow, plain
membrane.

Digestive tract. — Of the typical Styela type ; folds of the stomach
wall crowded, about thirty in number ; anus encircled by a fringe of
from ten to twenty blunt, irregular processes.

Gonads, — Ovaries two cylindrical bodies on each side of the sac,
the ventral one on the left side suppressed in the region of the stomach.
Testis in the usual form of scattered nodules along the ovaries. Both
ova and sperm nearly, if not quite, ripe.

This species is allied to S. mo?itereye?isis Dall, and S. greeleyi Rit-
ter, but the material now available for study furnishes no transitional
forms toward either. The most distinctive character is the propor-
tionate length of peduncle and body. In this respect the three species

THE ASCIDIANS

241

form a series, S. greeleyi having the longest peduncle ; ^S". Jiionterey-
ensis the next and 6'. yakutatensis the shortest. S. yakutatensis dif-
fers from kS. montereye?isis, also in possessing a maximum of less than
fifty tentacles ; whereas montereyensis reaches as high as one hundred.
The difference between yakutatensis and ntontereyensis in the struc-
ture of the sac cannot be regarded as of specific importance.

In large specimens of greeleyi the first and fourth folds of the sac
possess a considerably greater number of bars than they ever appear to
have in yakutatensis. Furthermore, the bilobed condition of the anus
in greeleyi does not occur in yakutatensis.

Further study of specimens from localities between those from which
these three species come, may be expected to throw light on the
relations of this interesting series. The collection contains about
seventy specimens of all sizes, all taken from a reef near Khantaak
Island, Yakutat Bay, at extreme low tide, on June 21, 1899.

STYELA TRUNCATA sp. nov.

(PI. xxvii, figs. 22 and 23.)
Ge?ieral characters. — Irregularly cone-shaped, attached by the base,
which is cleanly truncated by the surface which serves to attach it ; the
base distinctly broader than any other transverse section of the body.
Several of the largest individuals inclined somewhat to one side.

MEASUREMENTS .

Specimens.

Height.

Diameter of base.

Greatest.

I^east.

mm.

20

18

8

mm.
17
13
ID

mm.

13
10

One of the smaller

8

Siphons small and near together ; in larger specimens almost en-
tirely hidden by the large irregular tubercles of the test which beset a
small area of the anterior end. The remaining portions of the body
nearly even and smooth, though in some specimens more or less cov-
ered with an encrusting bryozoan. Color of whole anterior end a
reddish brown, siphons not differing from surrounding portions. Test
rather thin, but firm ; mantle thin, the musculature being rather feebly
developed.

^ The measurements of the smallest one in the table are exceptional ; in general,
the proportions of the small specimens more nearly agree with the largest one in
the table.

242

RITTER

Braitchial apparatus. — Siphons short and small, in larger indi-
viduals almost entirely hidden by the tuberculation of the test, ob-
scurely four-lobed. Branchial tentacles about twenty, subequal in
length.

Hypophysis mouth horseshoe-shaped, horns not rolled ; opening to
the left and toward the branchial siphon. Branchial sac with four folds
on each side, all of about equal prominence excepting No. 4, which is
somewhat larger. Formula for internal longitudinal vessels :

Endostyle.

First. Second.

Third.

Fourth.

Fifth.

L,ength
of specimen.

space.

fold.

space.

fold.

space.

fold.

space.

fold.

space.

18 mm.

I
I

6

8

I
I

8
8

I
I

6

8

I
I

ID

II

I
I

Transverse vessels of three sizes. From five to seven stigmata to
the mesh. Endostyle narrow ; dorsal lamina a plain, low membrane.

Digestive apparatus. — Stomach long and narrow, about twice as
long as thick ; somewhat curved, the concave side looking upward ;
having about fifteen longitudinal folds in its wall. Anus surrounded
"by a circle of about ten or twelve short thick knobs ; the terminal por-
tion of the rectum in the specimens examined narrowed down to a
thin neck.

Gonads. — Two ovaries on each side, each a long, simple, tortuous,
cylindrical mass. Testis pouches scattered along these in the usual
way. Both sets of glands nearly or qviite ripe, but no tadpoles seen.

Externally this species is very similar to S. joannce Herdman, from
Puget Sound (Herdman, 1899), but differs from it in having only two
in place of seven or eight ovaries on each side, and also in having only
one, in place of four, internal longitudinal vessels between the folds
of the branchial sac. Also it rather closely resembles S. pccssleri
Michaelson ^ from the region of Cape Horn from which it differs, how-
ever, in possessing fewer branchial tentacles, ^^55-/erz having sixty, and
in the prominence of some of the folds of the branchial sac ; in pcessleri
the first, or both first and second, are more or less rudimentary.

Found on rocks at extreme low tide, Khantaak Island, Yakutat Bay,
June 31, 1S99. There are about fifty specimens of different sizes in
the collection.

' Zool. Anzeiger, No. 560, 189S.

THE ASCIDIANS 243

DISTOMA PULCHRA sp. nov.
(PI. XXX, figs, 24 and 25.)

General character of the colony. — Composed of sheaf-like bunches
united by their basal ends only, each bunch containing from ten to
twenty large and long zooids, some of which are free at the anterior
end for rather more than the length of the thorax, the remainder being
embedded in the common test, which is present, however, in small
quantity between the bodies of the zooids. Height of the bunches
about 30 to 25 mm. ; thickness at the upper end, /. <?., in the portion
containing the thoraces, about I3 mm.; in the next thicker portion,
i. ^., that containing the abdomens, about 10 mm. ; of the constricted
portion immediately behind the thoraces, about 6 mm. Bunches quite
regular both in form and size, having much the appearance of sheaves
of ' club ' wheat, the thoraces corresponding to the heads of the wheat,
the constricted portion to the position of the band of the sheaf, and the
enlarged abdominal region to the butt of the sheaf.

Color. — Anterior end transparent, wholly without pigment ; poste-
rior end pinkish yellow from the great amount of food matter contained
in the abdomens of the zooids, which show distinctly through the thin
layer of colorless test.

Test. — Small in quantity, indeed in the posterior half of the bunches,
scarcely visible at first sight ; somewhat more abundant anteriorly,
especially where a few thoraces are united. Examination of ample
material might reveal some evidence that closely adhering, flattened,
stolonic masses of test are present. Ordinary test cells very abundant,
but no bladder cells, spicules, or nodules present.

Zooids. — Maximum number in a bunch about twenty. Very closely
packed in the bunches, particularly at their posterior ends ; anterior
ends more or less free, most of them for only a small portion of the
thorax, but a few for considerably more than the thorax. Frequently
a small number — two or three — in a bunch united to nearly the end,
and together free from the rest of the bunch for a part or all of the
extent of the thorax.

Entire length of body distinctly visible through the test. Thorax
not withdrawn from the test in preservation. Thorax and posterior
end of abdomen strikingly thicker than middle portion of body. Total
length 20 to 25 mm., equaling height of the bunches. Thorax about
7 mm. ; narrow neck-like portion of body, about S mm. ; thick portion
of abdomen about 10 mm., maximum thickness of thorax, including
test, about 4 mm. ; of abdomen about 3.5 mm.

244

RITTER

Mantle. — Very thin throughout ; circvilar muscle fibers absent except
in a narrow, weak sphincter for each siphon. Eight or ten weak lon-
gitudinal muscle bands on each side, running obliquely across the
thorax, antero-ventrally to postero-dorsally ; anastomosing sparingly
anteriorly, and becoming spread out into a thin layer apparently
increasing somewhat in number posteriorly, in the abdominal region.

Branchial apparatus. — Siphons long and prominent, both deeply
and regularly six-lobed, the lobes expanded. Tentacles about twelve,
all nearly equal in length and rather long ; placed in a close cluster
near the siphonal opening. Peripharyngeal field broad (p1. xxx, fig.

25)-

Eight series of stigmata, about thirty-five in each half-series ; stig-
mata exceedingly regular in form, rather long and narrow. Internal
membranes, on the circular vessels, broad. Dorsal languets about as
long as branchial tentacles, broad at base, shading off into the internal
membranes of the branchial sac to which they correspond in number.
Endostyle rather narrow. Postero-ventral corner of the sac produced
backward, the esophagus issuing from the postero-dorsal angle. The
whole branchial apparatus quite transparent, and usually very little
contracted in preservation.

Digestive tract. — Esophagus long and narrow, the length more
than twice that of the branchial sac. Stomach long and narrow,
about 4 mm. long, by 1.5 mm. broad; cylindrical, the wall with about
six regular, longitudinal folds. Loop of the intestine extending but a
short distance behind the stomach ; very narrow. Rectal portion of
intestine soon joining the esophagus and running close along with it
through the narrow portion of the body. Whole posterior end of in-
testinal loop so embedded in the food material of this part of the abdo-
men as to make study of it difiicult.

Reproductive organs. — Situated in the intestinal loop ; ovary
composed of numerous cylindrical, often anastomosing, branches.
No specimens in reproductive activity at hand.

The question of the affinities of this striking species is very puzzling.
For a time I was much inclined to consider it a Stereoclavella rather
closely related to S. australis Herdman, the general character of the
colony and especially the freedom of many of the thoraces apparently
justifying this view ; and, although I now regard it as more closely
allied to Distotna than to any other genus, its kinship with Stereocla-
vella is undoubted. The combination of Distomid and Clavelinid
characters in this species furnishes additional evidence, of which so
much has been forthcoming during recent years, of the close relation-
ship between the Distomidae and the Clavelinidae.

THE ASCIDIANS 245

The two considerations of most weight in assigning the species to
Distoma are, first, the deep and broad siphonal lobes, and, second,
the fact that while many of the thoraces in the colony are distinct most
of them are not at all free. Further study of more material, particu-
larly of young colonies and the process of budding, may reverse my
conclusion on this point ; for a few young blastozooids are present on
the specimens at hand, and these are at least as much isolated as the full
grown ones. Moreover, the large quantity of mesenchymatous food
material stored in the abdomen is probably rather more Distomid than
Clavelinid.

Within the genus Distoma the general character of the colony re-
sembles considerably that of D. laysani^ recently described by
Sluiter,' from Laysan, South Pacific. The resemblance goes no far-
ther, however, for the zooids are very different in the two, D. lay-
sani being much smaller, having a very inconspicuous, smooth-
walled stomach and only three rows of stigmata. On the whole D.
pulchra is probably more closely related to D. adriaticum v. Drasche
than to any other known species, though the two really have very
little in coinmon. The colonies, though pedunculated in both, are very
different. Adriaticum has fifty branchial tentacles and twenty-four
series of stigmata.

According to Lahille D. cristallium Renier is the only other
species known in which the stomach wall possesses parallel folds.
This, however, is not a pedunculated form.

Found at Yakutat Bay, on rocks at extreme low tide, June 20, 1S99.
One large and one small cluster of bunches in the collection.

DISTAPLIA OCCIDENTALIS Ritter.

Distaplia occidentalis Banxroft, Bull. Mus. Comp. Zool., Vol. xxxv, No. 4,
p. 59, 1899. — Ritter, Annals N. Y. Acad. Sci., Vol. xii, No. 14, p.
609, 1900.

Although the identification of the specimens at hand with this wide-
ranging Pacific Coast Distaplia^ is doubtful for several reasons yet the
species is so exceedingly variable and the quantity of material of the
Alaska form now available for examination so small that I do not feel
justified in regarding the differences as of more than individual value.

The two most important differences that I note are the denser, more
muscular thorax of the northern specimens and the greater distinct-
ness of the folds of the stomach wall.

A single small colony was dredged in Kadiak harbor, July 3, 1899.

^Zool. Jahrb., Bd. xiii, 1S90.

246 RITTER

DISTAPLIA CONFUSA sp. nov.
(PI. XXIX, figs. 26'and 27.)

General characters of the colo?ty. — Thin and encrusting, about
3 or 4 mm. tliick, and quite uniform ; largest piece at hand 40
mm. in length by about the same width. Irregular in outline, very
soft. Of a dark, purplish ashen color, due to pigment in the test.
Test in large quantity as compared with the number of zooids. A
few branched vessels, some bearing ampullae \vithin the test.

Zooids. — Few in number, scattered in the most irregular way, and
occupying the most varied positions as regards direction of axes rela-
tive to surface of colony. Apparently no systems present. Of a
nearly uniform sooty color, but little darker, however, than the test,
and, therefore, showing indistinctly through it.

Many abdomens without thoraces present ; of the characteristic dis-
tomid dumbbell shape, the handle long, straight and slender, the out-
line of the ends irregular, owing to prominent interruptions caused
by the few voluminous ova projecting from the posterior end of the
abdonien.

Length about 7 mm. (greater than thickness of the colony, this be-
ing possible from the extremely oblique position of the zooids).
Thickness of thorax about 1.5 mm., abdomen about the same.

Mantle. — Extremely delicate ; circular muscle fibers apparently
wholly absent ; longitudinal few, very fine, and but imperfectly
grouped in bundles.

Branchial apparatus. — No visible orifices on the surface of the
colony ; branchial siphon very broad and short, wide open in correla-
tion with the absence of sphincter muscle fibers ; lobes wide, low, and
irregular. Atrial orifice large, no siphon, but a wide and, in some
zooids, rather long languet.

Four series of long, narrow stigmata with about sixteen stigmata in
each half-series. Tentacles and dorsal languets not seen owing to the
contracted state of the sac.

Posterior ventral corner of sac not extended back behind esophagal
opening, but esophagus issuing from middle of posterior end of sac
(p1. XXIX, fig. 26).

Digestive tract. — Nearly straight, loop narrow, post-gastric intestine
returning abruptly back upon the esophagus and meeting it at the en-
trance of the latter into the stomach ; from here the straight rectal por-
tion closely applied to the esophagus, and accompanying it forward to
the thorax. The general form of the tract thus made to resemble closely

THE ASCIDIANS 247

the distomid type. Narrow esophagal-rectal portion about equalling
thorax in length. Stomach of moderate size, somewhat shorter than
branchial sac ; longer than broad, the longer axis forming an angle of
about 45 degrees with main axis of animal ; external surface of wall
smooth, but inner surface marked by a close, rather regular reticula-
tion showing through, due to low narrow folds in the wall. Reticula-
tion more pronounced in fully grown zooids (p1. xxix, figs. 26 and 27).

Reproductive organs. — Ovary situated in part on the right side of
the intestinal loop but chiefly behind it. Ova few but very large.
Testis not seen. No einbryos in the specimens at hand, consequently
presence or absence of brood pouch not determined, facts probably in-
dicative of immaturity of specimens. The presence of well-developed
ectodermal appendages of the mantle in most of the zooids indicative
of same conclusion, according to the statement of Bancroft,^ that in
D. magnilarva and D. occidentalis these structures are found only in
the young.

The close general resemblance of the zooids of D. co7ifusa to the typical
Disiotna zooid is striking indeed. So far as I am aware, no other
Distaplia possesses the long straight esophagus found here.

For a time I was much disposed to identify the species with D. liv-
ida Huitfeldt-Kaas (Sars), of the Norwegian coast, and there can be
no doubt about the close relationship of the two forms. It is quite
possible that further study of more specimens from both Alaska and
Norway, will prove them to be the same. At present, it seems clear
that the shortness of the zooids of D. livida (4 mm. according to
Huitfeldt-Kaas) indicates that the peculiar, long, straight esophageal-
rectal stretch of the intestine is not present in that species. Further-
more, Huitfeld-Kaas ^ speaks of the colonies of D. livida as being
" composed of very soft, thick, irregular, lobed lamellae with a maxi-
mum area of 2.5 cm. and a thickness of about 7 mm.

Dredged in about twenty fathoms, Kadiak harbor, July 3, 1S99.
Encrusting on lamellibranch shells. Two colonies in the collection.

DIDEMNUM STRANGULATUM sp. nov.

(PI. XXIX, fig. 28.)

General characters of the colony. — Flat, incrusting, thin. Largest
colony in the collection 4 mm. long by 2 mm. broad ; 5 mm. thick in
thickest part, i mm. in thinnest part. Color ashen, zooids showing
distinctly through the test. Apparently no systems present ; zooids
distributed very uniformly, rather numerous.

'Bull. Mus. Comp. Zool., xxxv, No. 4, 1899.

2Norw. N. Atl. Exped., 1876-1879, Zool., Christiania, 1896.

248

RITTER

Test. — Containing many cells of various kinds, but no bladder cells.
The granular bodies (cells?) numerous. Calcareous stellate bodies
present, though not abundant ; not found at all in some colonies. Rays
short and blunt.

Zooids. — Small, about 2 mm. long ; uniformly distributed. Divi-
sion of body into thorax and abdomen distinct ; abdomen considerably
larger than thorax, and broader than long. The narrow peduncle con-
necting thorax and abdomen always (in specimens at hand) with a
deep, narrow constriction at which severance occurs with great ease.
Many abdomens free in the test from normal rupture at this point (p1.
XXIX, fig. 28). Mantle sending off one or more clfivate ectodermal
appendages from the middle region of the body. Longitudinal muscle
fibers moderately developed ; a few circular ones also present even in
middle portion of thorax.

Branchial apparatus. — Branchial siphon of moderate length ; six-
lobed, the lobes usually distinct though never prominent, sometimes
nearly absent. Atrial siphon shorter, though present ; placed far back,
and inclined somewhat backward ; lobes as prominent as those of
branchial siphon. Branchial sac so much contracted as to make de-
termination of internal structure difficult. Tentacles not clearly seen.
Three series of rather long stigmata, a considerable area between the
peribranchial band and the first series. A delicate muscle band
present (?) in the interserial spaces. Endostyle strongly curved from
the contraction of the sac, but not tortuous.

Digestive apparatus. — Loop of intestine very wide, its transverse
diameter frequently greater than its longitudinal ; esophagus exceed-
ingly narrow, issuing from middle of posterior end of branchial sac.
Stomach of liberal size, smooth-walled, nearly globular. Intestine
thickest at pyloric end, and gradually tapering to the rectum which
seems to terminate at the deep recto-esophageal constriction though
found by careful examination to extend to atrium (p1. xxix, fig. 28) .

Reproductive apparatus. — A single large ovum appears to consti-
tute the ovary ; situated on the left-posterior surface of the intestinal
loop. The single spherical lobe constituting the testis nearly as large
as the ovum, and wholly on left side of the intestinal loop. About
four (?) coils to the vas deferens. Embryos not found, though both
ova and sperm apparently near maturity.

This species is very closely related to D. inarviatum v. Drasche
{Didemnoides inarmatum Lahille) ; but differs from it in possessing
calcareous spicules in the test and in lacking the black pigment patterns
characteristic of that species and also in possessing the recto-esophagal

THE ASCIDIANS 249

constriction mentioned in the diagnosis. The absence of this latter
character in D. inartnattan may be inferred from Lahille's statement
that the digestive tube does not differ here from the ordinary Didemmi?7i
form. One might doubt the constancy of this character in D. stran-
gulatinn^ but for the fact that it is present in zooids not yet fully grown.
From all that can be learned from the material available, it seems that
the abdomen, cut aw^ay from the thorax at this point, becomes freed
from the test in some vs^ay, and so liberates the embryos into the water.
I have been able to pick out a number of such severed abdomens from
the test, where they lay merely embedded in shallow pits in the sur-
face layer.

D. strangidatum is also rather closely related to D. cereui7i Giard,
but is readily distinguished from this species by its narrower branchial
and longer atrial siphon and by its peculiar recto-esophageal neck.

The specific name chosen has reference to the cutting off of the isth-
mus that connects thoi'ax and abdomen.

Kadiak Island, Alaska, July 3, 1899. Dredged in St. Paul harbor
in about twenty fathoms. Four colonies in the collection, the smallest
about 6 mm. in diameter, all on fragments of molluscan shells.

AMAROUCIUM TRANSLUCIDUM sp. nov.

(PI. XXX, figs. 29 and 30.)

General characters of the colony. — In unusually regular conical
masses, the attachment being by the apex of the cone, which generally
constitutes a short peduncle. Frequently, several of. these masses are
united by their peduncles. Upper surface, forming the base of the
cone, gently arched. A longer and a shorter diameter to the base of
the cone at right angles to each other.

Size of largest mass in collection : height 15 mm., greatest trans-
verse diameter 20 mm., least 15 mm. Height of all the masses nearly
the same, and the other dimensions much more uniform for all the
specimens than is usual among a series of specimens of Compound
Ascidians.

Color. — Light red, not confined to the zooids but diffused through-
out substance of test. Test unusually transparent for this genus, and
the zooids very plainly seen through it.

Zooids. — Nutnerous, rather large, closely crowded, and regularly
placed ; systems hardly recognizable except in young colonies. Com-
mon cloacal orifices few, apparently becoming obliterated in the older
colonies, rather large and distinct in the smaller ones. About 10 mm.
long as measured in the test of preserved colonies, very straight and

Proc. Wash. Acad. Sci., July, 1901.

250

RITTER

regular, not contracted away from the upper surface of the colony ;
the three divisions of the body, as seen in their natural position in
the colony conspicuously set off from one another, the post-abdomen
by its light reddish yellow color and by its abrupt constriction from
the abdomen, the abdomen by its pale dirty green and by its shoitness
and thickness, and the thorax by its ashy color and prominent endo-
style. As a rule the thorax -and abdomen combined seem to con-
stitute nearly one-half the length of the entire body, as seen with the
zooids in place in the colony; in reality, however, the post-abdomen
is considerably longer than these two together (p1. xxx, fig. 30).

Test. — Not great in quantity, because the zooids are so closely
crowded ; no bladder cells or vessels present ; cells small, numerous,
often containing pigment, particularly in surface layers.

Mantle. — Very delicate ; longitudinal muscle fibers arranged in a
few definite bands ; circular muscles almost wholly wanting.

Branchial apparatus. — The six lobes of the branchial siphon
sometimes short and broad, sometimes rather long and slender ; con-
stant in number. Atrial siphon short; its languet usually long and
pointed, frequently trifid. Branchial tentacles counted with difficulty ;
about ten or twelve in number, of unequal sizes, situated very near the
branchial siphon. Branchial sac about 3 mm. long and 2 mm. deep.
Usually ten (rarely nine or eleven) series of stigmata ; in one zooid
(immature?), only seven; the individual orifices long and narrow and
very regular. About eighteen or twenty stigmata in each half series.
The interserial vessels narrow ; a rather delicate muscle band in each.

Digestive tract. — Intestinal loop short (never as long as the bran-
chial sac) and wide. Esophagus very short, in preserved specimens
usually hardly recognizable; loop of the intestine behind the stomach
also very short. Distal portion of the ascending limb of the intestine
lying at the left side of the stomach and esophagus, rather than dorsal
to them, thence turning almost with a right angle to reach, in its rec-
tal portion, the left dorsal aspect of the branchial sac. The form of
the intestinal loop thus approximates the twisted condition of this
organ as found in Polyclinum (p1. xxx, fig. 30).

Stomach large and prominent, nearly globular, but generally some-
what broader than long, its position slightly transverse to the general
long axis of the animal. Folds in the wall numerous (twenty or more) ,
many of them very regular and extending without interruption the
entire length of the organ. In some cases, they are broken up into
islands, some elongated, others nearly round ; in others the folds branch
and anastomose with one onother. On the dorsal-sinistral side a nar-

THE ASCIDIANS 25 1

row, elliptical area with no folds ; the folds in the immediate vicinity
of this disposed at an angle rather than parallel to it.

Reproductive organs. — Many embryos present in the atrial cham-
bers of many of the zooids, but the sexual glands unusually inconspic-
uous ; in many zooids not recognizable at all without particular atten-
tion. Ovary situated far behind the intestinal loop, though in some
cases, apparently only a short distance behind it.

This Amaroiichan bears a strong superficial resemblance to an
undescribed species of Synoictiin from the coast of California, and
was at first supposed to be this or a closely allied form, but more care-
ful study removed all doubt about the genus to which it belongs.
As mentioned in the diagnosis, and shown by the figures, there appears
at first sight to be no grouping of the zooids into systems ; and the true
state of things was ascertained only after this point had received special
attention. It would seem that the common cloacal orifice becomes
wholly obliterated in most of the colonies, the zooids then becoming
as vmiformly distributed as they are in genera like Disto7tia. in which
each atrial siphon opens by itself. What the method of exit of the
excreta is in these conditions, I am not sure, as I have not been able to
detect any single orifices. The long atrial languets may be seen
embedded in the test, and apparently the siphons to which they belong
have, in most cases, no openings to the surface of the colony. In a few
colonies, particularly those of small size, an occasional rather large and
conspicuous common cloacal orifice may be seen.

The species is more closely related to Amarouci'um proliferufn M.
Edw. than to any other, particularly if this species is held, as Lahille
believes it should be, to include several forms that have been described
by other authors as separate species. From all the varieties of prolif-
erum the present form differs, however, in the absence of composite or
ramified colonies characteristic of proliferzan^ in the smaller size of
the masses, the less distinct systems of the zooids, and the greater
regularity of the folds of the stomach wall. The specific name refers
to the unusual transparency of the test.

Found at Orca, Prince William Sound, Alaska, on reefs at low tide.
Collected June 26, 1899. About one hundred colonies in the collec-
tion.

AMAROUCIUM COEI sp. nov.

(PI. XXIX, figs. 31 and 32.)

General characters of the colony. — In large pear-shaped masses,

many of them loosely joined together by their narrow bases. The

masses, as a rule, very regular in form, the expanded end nearly cir-

252 RITTER

cular in outline, the upper part domed and oversetting somewhat, par-
ticularly in large specimens, the regularly narrowing smaller, or
peduncular part. The upper surface somewhat uneven by the projec-
tion of the systems which are distinct in life, though coiisiderably ob-
scured in the preserved specimens ; six or seven zooids in each system.

Color. — Light salmon, nearly uniform throughout, zooids more
highly colored than test. In preserved condition color uniform ashen
with tinge of green ; test somewhat transparent so that the zooids show
through rather indistinctly, at the sides, and scarcely at all on the up-
per surface.

Size. — Of largest mass: Height 58 mm., of peduncular portion 31
mm., of dome 33 mm., thickness in thickest part 43 mm. (cross sec-
tion a circle). Of a medium sized mass: Height 33 mm., of pedun-
cular part 20 mm., of dome 10 mm., thickness in thickest part 20 mm.
by 24 mm.

Test. — Cells very numerous and relatively large, but no bladder
cells.

Zooids. — Extremely long and slender, thorax very small in pro-
portion to entire zooid ; total length about 25 mm., thickness in
thickest place about 2 mm. Of this length (measurements on pre-
served specimens, though contractions not great), thorax 4 mm., ab-
domen 5 to 7 mm., post-abdomen 21 mm. Not very closely crowded,
rather regularly distributed, placed nearly parallel with one another ;
few (not more than six or seven) in a system. Not much contracted
except as to thorax ; easily removed from test.

Ma7itle. — Musculature feeble, a few small, irregular, longitudinal
bands.

Branchial apparatus. — Branchial orifices (in life) moderately
large, not visibly lobed or with other characteristic marks ; common
cloacal orifices distinct, not lobed. Branchial orifices (in preservation)
not recognizable, cloacal much obscured. Both siphons rather long
and narrow and much alike ; branchial six-lobed, the lobes always
broad and low ; atrial also obscurely and irregularly lobed, usually the
dorsal side extended more or less to produce an imperfect atrial lan-
guet, though frequently showing no intimation of this structure.

Branchial tentacles around the branchial orifice about seventeen, of
several sizes, some very short ; in a close bunch, as seen in dissections.

Branchial sac very small for the size of the zooids. Series of stig-
mata varying from eleven to fifteen, the individual orifices small and
regular ; muscle bands in interserial spaces of moderate development.
Dorsal languets at least as numerous as the series of stigmata ; rather
long.

THE ASCIDIANS 253

Digestive apparatus. — Enteric loop as a whole rather long (twice
as long as branchial sac in most of the zooids with latter somewhat
contracted) , narrow, straight ; esophagus rather longer than stomach ;
stomach distinct but not prominent, usually somewhat longer than
broad ; folds of the wall well defined, extending throughout its length,
usually regular but sometimes broken and anastomosing. Loop of in-
testine very narrow ; ascending limb taking the course usual for the
genus, but owing to the narrowness and length of the loop and the
small size of the branchial sac having inconspicuous windings. Loop
as a whole very straight.

Post-abdomoi and reproductive apparatus. — Former very long
and slender and of uniform diameter for most of its length ; not dense
with food material and sexual products. Ovary a short distance be-
hind intestinal loop, rather small ; testis beginning immediately behind
the ovary and extending through almost the entire remaining portion
of post-abdomen ; masses regular in size and form, nearly spherical,
not crowded. Vas deferens distinct.

Although this species is devoid of characters that differentiate it
strikingly from a number of other species, it is still impossible to iden-
tify it with any of them, and several of the differentia are important.
In the form and size of the colony it is strikingly like A. Jigarium
Ritter (MS.) from the coast of California; but the surface of the test
of this latter species is firmly incrusted with sand, while that of A. coei
is peculiarly devoid of sand or other foreign substances. The most
striking difference between the two is found in the zooids. The length
of the entire animal is about the same in both, but the branchial sac of
Jigariu?7i is three or four times as large as that of coei. In coei the
folds of the stomach wall are much more numerous and distinct than
'\\\ Jigariutn . Various other trivial but constant differences might be
mentioned. Another species with which it has much in common is
A. blochinanjti Heiden.^ From this it differs, however, in its less
distinct peduncle, larger size of the masses, smaller number of stig-
matic series, length and character of the atrial languet, and in the
greater distinctness and regularity of the folds of the stomach. It
gives me pleasure to name this species after my friend and efficient
fellow-collector on this expedition. Dr. W. R. Coe.

Taken with the dredge, near the wharf in St. Paul harbor, Kadiak,
on July 3, 1899, in a few fathoms, rocky bottom, with coarse pebbles
and shells. Apparently very abundant here, as every haul brought up
a number of specimens.

^ Zool. Jahrb., Band vii, 1894.

254

RITTER

AMAROUCIUM CALIFORNICUM (?) Ritter.

Amarouciiim califor7iicum Ritter, Ann. N. Y. Acad. Sci., Vol. xii, No. 14,
p. 608, 1900.

A few small colonies of an Amaroucium found at Popof Island are
doubtfully identified with this species. The chief difficulty in the way
of identification is in the small size of the zooids and the fungiform
shape of the colonies in the specimens at hand. It is highly probable,
however, that none of the colonies are really typical. None of the
zooids are sexually mature, and in one colony, the lai-gest, most of
them are in a broken-up, degenerate condition, indicating that the gen-
eral form is flat and expanded, and that the fungiform pieces are in
reality only incidental. Such knobs are very common in A. call-
fornicum.

The zooids are all so contracted that it is impossible to make out
much about their internal structure, but so far as one can see there is
nothing here in the way of the identification.

Taken on shore rocks at low tide, Popof Island, Shumagin Islands,
July 8, 1899.

PSA]MMAPLIDIUM PEDUNCULATUM ( .?) Herdman.

Psammaplidium pedimciilaium Herdmax, Linn. Soc. Journal, Zool., Vol.
XXIII, p. 620, 1891 ; and Catalogue of Tunicata in Australian Museum,
No. XVII, p. 88, and Pi, vi, figs. 7-9, 1899.

The identification of the specimens with this species is very doubt-
ful. This assignment is provisional only, pending an opportunity to
examine more material.

Meantime the following points may be noted, mostly supplementary
to Herdman's description, but to some extent obviously at variance
with it :

Form of the club-shaped masses constituting the colonies essentially
the same as in the Australian specimens, though probably the size is
slightly larger in the Alaskan species.

Color. — Cherry red in life, this, however, disappearing in pre-
served condition except a slight tinge particularly noticeable around
the siphons.

Zooids. — Arranged in very distinct systems of about a dozen each ;
one or two systems only to each lobe. Common cloacal opening large
and distinct. Zooids much contracted and very opaque. A distinct,
though not long, single pointed atrial languet. About twelve series of
branchial stigmata. Stomach rather large, though not clearly set off
from the intestine, which is also large. A single narrow longitudinal

THE ASCIDIANS 255

fold on the left side of the stomach triangular in form with very nar-
row base directed toward the posterior end of stomach, the apex reach-
ing nearly, though not quite, to the anterior end.

Post-abdomen very dense and opaque ; gonads not recognized.

Comparison of the points here noted with Herdman's description,
will show that the color of the colony, the distinct systems, and the
character of the stomach offer the chief difficulties to the identification
of the two forms as one species.

Dredged at Sitka in about ten fathoms, June 15, 1899. A single
colony with two lobes.

BOTRYLLUS MAGNUS sp. nov.

(PI. XXX, figs. 33-37-)

General characters of the colony. — Flat and incrusting, though
with a tendency to lobulation in some parts. Found on the under side
of rocks, to which it clings very closely. Irregular in outline, several
mm. (five or six) in diameter, 3 or 4 mm. thick, with pronounced
lobes here and there. Dusky purple in color, uniform in all parts,
and showing practically no variation from colony to colony. Systems
distinct, circular, regular. Five to eleven zooids in each system.
Common atrial orifices large, lobed, having as many lobes as zooids.
Each branchial orifice standing somewhat above the general surface.
The ectodermal vessels and ampullae in the test numerous ; the am-
pullae at tjie margins of the colony much longer than thick ; those in
other portions shorter. A mass of brown pigmented cells in the am-
pullae (p1. XXX, fig. 37).

Zooids. — Large, 3.5 or more mm. long; 2.5 mm. broad. General
form that of a cylinder slightly curved with the convexity dorsal and
the ends nearly truncate. Full grown individuals rather opaque from
the considerable quantity of mesenchymatous material {J) and blood
in the mantle. Musculature of mantle exceedingly weak ; a, few scat-
tered fibers in the anterior end only, confined chiefly to the siphons.

Branchial apparatus. — Branchial orifices distinct on surface of
colony; circular, not lobed, projecting slightly above the general level.
Six branchial tentacles projecting into the siphon in the living colonies.
Branchial siphon short and wide. Atrial siphon wide and fi'equently
of considerable length ; provided with a prominent ventral lip and
with a wide single-pointed doi'sal languet^ of very variable length,
wanting in young zooids (p1. xxx, figs. 35 and 36).

Tentacles about sixteen, eight moderately long, and eight slightly
shorter alternating with these. Eleven to thirteen series of stigmata

256 RITTER

with about twenty stigmata in each half series ; the individual stigmata
when fully grown about four times as long as broad ; border cells dis-
tinct. Interserial spaces narrow and without muscle fibers. Three
internal longitudinal bars on each side. The dorsal lamina, as seen
through the animal, a plain narrow membrane, resembling one of the
internal longitudinal vessels.

Digestive apparatus. — Situated on the left side of and projecting
but very slightly behind the branchial sac at the posterior end. Esoph-
agus issuing from the postero-dorsal angle of the sac; nearly as long
as the stomach. Stomach situated nearly transversely to the sac, and
extending somewhat more than half way across it, having its pyloric
and cardiac ends about equally remote from the dorsal lamina and
endostyle respectively. Stomach about one-third longer than broad ;
slightly curved, the convex side looking backward. Wall with about
eight or ten double folds which extend the entire length of the organ.
The intestine behind the stomach in the form of an open S, with the
rectal limb considerably longer than the pyloric and quite straight.
The rectum reaching forward in the cloaca anterior to the middle of
the sac. Anus usually showing more or less of a lip. The post-gas-
tric intestine nearly uniform in diameter throughout.

Reproductive apparatus. — Gonads on both sides of the body ; im-
mature and very small in all the material at hand. Budding zooids
abundant ; two buds, one on each side, on each budding individual.

The specific character particularly relied upon for this form, viz,
the large size of the zooids, is a strong point in favor of regarding it as
belonging not to Botryllus at all, but to Polycychis., and I am much
inclined to believe that its closest affinities are in reality with P. Sa-
vignyi Herdman. According to prevailing views regarding the gen-
eric characters of the Botryllida^, however, the comparative thinness
of the colony and the softness of the test require that it be placed in
the genus Botryllus. The large size of the zooids, the uniformity of
color of the colonies, and the relative nearness of the atrial to the
branchial siphon make it a rather distinct species in this genus.

A single colony dredged in about 20 fathoms at Kadiak Island ;
about a half dozen colonies found on rocks at low tide at Popof Island,
Shumagin Islands, July S, 1S99.

THE ASCIDIANS 257

BIBLIOGRAPHY.
Bancroft, F. W.

1899 Ovogenesis in Distaplia occidenfalis Ritter (MS.), with remarks on other
species. Bull. Mus. Comp. Zool. at Harvard University, Vol. XXXV,
No. 4, pp. 59-112.

Drasche, Richard von.

1883 Die Svnascidien der Bucht von Rovigno, Wien, pp. 1-4 1.

Edwards, Milne.

1842 Observations sur les ascidies composees des c6tes de la Manche. Mem.
Acad. Roy. Sci. de I'lnst. de France, Tome XVIII, pp. 217-326.

Giard, Alfred.

1872 Recherches sur les Ascidies composees ou Synascidies. Theses. Paris,
pp. 1-204.

Hartmeyer, Robert.

1899 Die Monascidien der Bremer Expedition nach Ostspitzbergen im Jahre
1SS9. Zoologische Jahrbiicher, XII Bd., pp. 453-520.

Heiden, Heinrich.

1894 Ascidiffi aggregate und Ascidise compositse von der Insel Menorca.
Zoologische Jahrbiicher, VII Bd., pp. 341-364.

Herdman, W. A.

1882 Repoi-t on the Tunicata of the Challenger Expedition, Part I, Ascidiae

Simplices. Challenger Reports, Zoology, Vol. VI.
1891 A Revised Classification of the Tunicata, etc. Linn. Soc. Journ. Zool.,

Vol. XXIII, pp 55S-652.

1898 Description of some simple Ascidians collected in Puget Sound, Pa-
cific Coast. Trans. Liverpool Biol. Soc, Vol. XII, pp. 248-267.

1899 Descriptive Catalogue of the Tunicata in the Australian Museum,
Sydney, N. S. W. Catalogue No. XVII, pp. 1-139.

Huitfeldt-Kaas, H.

1896 Synascidiae, pp. 1-27. Norwegian North Atlantic Expedition, 1876-
1879, Zoology, Christiania.

Lahille, Fernand.

1890 Contributions a I'dtude anatomique et taxonomique des tuniciers.
Theses, pp. 1-32S. Toulouse.

Metcalf, M. M.

1900 Notes on the Morphology of the Tunicata. Zoologische Jahrbiicher,
XIII Bd., pp. 495-602.

Uichaelsen, W.

1898 Vorlaufige Mittheilung liber einige Tunicaten aus dem Magalhaensi-
schen Gebiet, sowie von SUd-Georgien. Zoologische Anzeiger, No.
560, pp. 363-371.

Pizon, Antoine.

1898 Etude anatomique et systematique des Molgulidees appartenant aux col-
lections du Museum de Paris. Annals des Sci. Naturelles, Zoologie, 8
serie, T. VII, pp. 305-391.

258 RITTER

Ritter, Wm. E.

1900 Some Ascidians from Puget Sound, Collections of 1S96. Annals N. Y.
Acad. Sci., Vol. XII, No. 14, pp. 5S9-616.

Roule, Louis.

1885 Recherches sur les Ascidies simples des C6tes de Provence. Annales
des Sciences Naturelles, 6 serie, T. XX, pp. 1-229.

Sluiter, Ph.

1900 Tunicaten aus dem stillen Ocean. Ergebnisse einer Reise nach dem
Pacific (Schauinsland, 1896-1S97). Zoologische Jahrbiicher, XIII Bd.
PP- 1-35-
Stimpson, Wm.

1864 Description of new species of Marine Invertebrates from Puget Sound
collected bj the naturalists of the North-West Boundary Commission,
A. H. Campbell, Esq., Commissioner. Proc. Acad. Nat. Sci., Phila.,
Vol. 16, pp. 153-161.

Traustedt, M. P. A.

1884 Ascidiae simplices fra det stille Ocean. Vidensk. Meddel. fra den
naturh. Foren., pp. 1-60.

THE ASCIDIANS 259

ABBREVIATIONS USED IN THE FIGURES.

a. s. — atrial siphon.

d. s. — branchial siphon.

d. I. — dorsal lamina.

d. l'g''s. — dorsal languets.

ec. ap. — ectodermal appendages.

f-f-^f-Y-^ — folds of branchial sac.

g: g. — ganglion-gland.

gon. — gonads.

liy. — hypophysis.

m.f. — muscle fibers.

ov. — ovary.

P, — peduncle.

p. i. — intermediate papillae of branchial sac.

;'. b. — terminal bulb of rootlets.

r. c. — rectum.

f. /.' — ultimate rootlets.

st. — stomach.

^5/.— test.

tes. — testis.

ist. c. — test cells.

PLATE XXVII .

Figs, i to 5. Ascidia ad/tcErcjis.

I and 2 are habit sketches, i showing particularly the way in which the
animals aggregate together to form a continuous flattened layer. The
position of the atrial siphon somewhat on the upper surface is shown.
Natural size.

3. Left side of animal, dissected from the test.

4. Portion of peripharyngeal field showing a portion of the tenacles. the
hypophysis inouth, the gland-ganglion mass, .?. 5., and the dorsal lamina,
d. /., with its few, simple marginal teeth.

5. Small portion of branchial sac drawn with special reference to the irreg-
ularity in distribution of the intermediate papillae,^, i.; i. e., their pres-
ence or absence in the intervals between the transverse vessels.

Figs. 6 to 9. Molgula graph ica.

6. Habit sketch, seen from right side. Enlarged 2 times.

7. Left side of animal, dissected from test.

8. Portion of peripharyngeal field; s.s., the gland ganglion mass.

9. One of the much branched tentacles. The extent of the branching, and
the rolling up of the ultimate branches is even more pronounced in the
specimen than is here shown.

Figs. 20 and 21. Styela yahutatensis. Natural size.

Figs. 22 and 23. Styela tnmcata. Natural size. Fig. 22 suggests the aggrega-
tion of the individuals and 23 shows the general conical form and the
truncation of the basal end.

(260)

Proc. Wash. Acad. Sci. Vol. Ill

Plate XXVI

ALASKA ASCIDIANS

PLATE XXVIII .

Figs, io to 19^. Rkizomolgula arenaria.

10. Habit sketch. Natural size. The method of drawing makes it impos-
sible to successful!}' show the sand on the surface.

11. Left side of animal dissected from the test. The stomach (5/".) with
its well-defined ribs is shown, as is also the peculiar coiled-up condition
of the material within the intestine.

12. View of orifices of anterior end with test removed.

13. Portion of peduncle (-P. ) with numerous rootlets, some showing the
bulbous terminations (r. 3.). (X 15)'

14. Magnified view of main stem of peduncle (Z'. ) with a single rootlet
{r. A) arising from it. This shows its single muscle fiber {m. f. ) con-
tinuing up into the peduncle.

15. Branching rootlet showing how the single muscle fiber of the ultimate
rootlet (r. t.) reaches into the larger portion of the branch.

16. Optical section of portion of a peduncle ; its outer edge marked by the
test {tst.); bundle of muscle fibers {m. b.). A single muscle fiber [m.
f.) is separated from the general mass and is clearly seen imbedded in
the test with test cells {tst. c.) entirely surrounding it.

17. Similar to last but more clearly showing the tendency of the test cells to
become elongated.

18. Portion of branchial sac showing four folds ( f-f-^f-'^f-^)- The folds
f.^ and/!^ a^e spread apart to bring out the infundibula; and the inter-
nal longitudinal vessels confined to the folds.

19. Portion of peripharyngeal field.

19^. Small piece of mantle showing the papillae.

19^. One of the papillae, magnified, showing its wall and lumen.

(262)

Proc. Wash. Acad. Sci. Vol. II

Plate XXVIll

ALASKA ASCIDIANS

PLATE XXIX.

Figs. 26 and 27. Distaplia confusa. (X 17-)

26. A mature zooid. An approach to the dumbbell shape, characteristic of
many species of Dtstoma is shown, though imperfectly, in this figure.

27. An immature zooid. The difference in the character of the stomach
walls of 26 and 27 is noteworthy.

28. Didemtiiim strangulatum. (X I7-) Atrial siphon (rt. 5.); end piece of
the rectum {^r. c.) nearly cut off along with the esophagus, by the con-
striction.

Figs. 31 and 32. Atnarouctiim coei.

31. Natural size.

32. Zooid, much enlarged.

(264)

Proc. Wash. Acad. Sci. Vol. Ill

Plate XXIX

ALASKA ASCIDIANS

Proc. Wash. Acad. Sci., July, 1901.

PLATE XXX.

Figs. 24 and 25. Distoma pulchra.

24. Several bunches. Natural size.

25. Thorax of immature zooid. The specimen was dissected from the test
and cut lengthwise nearly in the sagittal plane. The figure represents
the left side. The cut passed to the left of the endostvle and to the right
of the dorsal languets.

Figs. 29 and 30. Amaroiict'urn translticidum.

29. Slightly less than natural^ size. Usually more masses than are here
shown are joined together.

30. Zooid much enlarged.
Figs. 33 to 37. Botryllus magnus.

33. Part of a colony. Natural size.

34. A few of the systems.

35. Outline of adult zooid. (X 16.)

36. Young zooid. Atrial languet not yet developed.

37. Piece from the margin of a colony, showing the ectodermal vessels with
their ampulla.

(266)

Proc. Wash. Acad. Sci. Vol. Ill

Plate XXX

ALASKA ASCIOIANS

PROCEEDINGS

OF THE

WASHINGTON ACADEMY OF SCIENCES

Vol. Ill, pp. 267-373. PLS. xxxi-xxxii July 13, 1901,

A NEW SPECIES OF OLENELLUS FROM THE

LOWER CAMBRIAN OF YORK COUNTY,

PENNSYLVANIA.

By Atreus Wanner.

Reference is made by Mr. Charles D. Walcott, in Bulletin No.
134, United States Geological Survey, to the Olenellus fauna
of the York county, Pennsylvania, Cambrian. At that time
fragmentary impressions of trilobites had been collected, but no
locality had yielded specimens that could then be referred with
certainty to any described species of Olenellus. I have re-
cently found complete specimens of several species, one series
illustrating different stages of growth and presenting variable
features of Olenellus thompsoni. One new species of the sub-
genus Holmia is beautifully preserved and I take pleasure in
naming it after Mr. Walcott in recognition of his work on the
Cambrian faunas.

OLENELLUS (HOLMIA) WALCOTTANUS sp. nov.

(PI. XXXI, figs. I, 2; PI. XXXII, figs. I, 2, 3, 4.)

General shape of entire specimen (PL xxxi, fig. i) ovate; width
about two-thirds of length.

Transverse diameter of cephalic shield about twice the longitudinal ;
marginal furrow shallow ; rim of moderate and nearly uniform width,
being apparently narrowest in front of the glabella and terminating in
two postero-lateral spines. In all the specimens found, the anterior
part of the rim is obscured and evidently narrowed by the flattening

(267)

268 WANNER

and forward thrust of the large frontal segment of the glabella. A low
narrow ridge extends along the posterior margin of the cephalic shield.

The anterior marginal fold is striated by closely placed parallel
lines, conforming to the curvature of the perimeter and occasionally
inoscvilating.

There are prominent and moderately long genal spines and the base
of a meso-occipital spine.

The anterior part of the glabella, evidentlv originally hemispherical
and very prominent, is flattened and pushed forward in all the speci-
mens found. The glabella is widest across the middle of the promi-
nent frontal lobe, uniformly narrowing to the posterior lobe. The
posterior part is divided into four transverse ridges by the occipital
furrows and three additional pairs extending nearly to the top.

Eyelobes crescentiform and prominent, with a raised rim around
the outer margin, cin-ving outward from the posterior part of the frontal
lobe of the glabella and terminating in front of the occipital furrows.

Thoracic segments seventeen ; niedian lobe strongly convex, the
base of a spine extending across the middle of each segment.

Pleura slightly curving part way and then strongly falcate ; chan-
nelled with a broad groove, extending to half the length of the pleura
and where the grooves terminate, presenting a prominent line of de-
marcation, beyond which the flattened and broad extremities curve
more sharply downwards. Posterior pairs of pleura strongly incurv-
ing and inclosing the pvgidium.

Beyond the termination of the pleural groove, closely placed trans-
verse folds, overlapping each other at short intervals, finely fret the
surface with short, curved, inosculating lines.

The strongly falcate ends of the pleura increase in relative length
towards the pygidium, attaining their greatest extension shortly ante-
rior to the ultimate pair of pleura.

With the exception of the falcate ends of the pleura and part of the
pygidium a network of incised lines (PI. xxxi, fig. 2) ornaments the
surface dividing it into minute hexagonal, or pentagonal, plates. In
all casts of the surface the inosculating lines are raised above the en-
closed spaces.

Pygidium small, circular ; extending but slightly further longitudi-
nally than its transverse width along the line of attachment. The
posterior margin (PI. xxxii, fig. 4) parted in the middle by a trian-
gular incision penetrating about one third of the distance across. In
this specimen, the triangular incision is open and the pygidium, to-
gether with the thoracic segments, somewhat fore-shortened ; in PI.

A NEW SPECIES OF OLENELLUS 269

XXXI, fig. I, the margins of the incision are pushed together and the
pygidium narrowed. In the one case, pressure on the matrix was
exerted longitudinally, and in the other, laterally. Thus in the two
specimens are presented opposite effects of like pressure, results that
must be considered in forming a conception of the shape of a specimen
free from all distortion due to pressure.

For about half the distance to the posterior end, the pygidium
presents a slightly raised surface. The curved lines, on each side of
the thoracic segment, marking the ends of the pleural grooves, extended
to meet, follow the margin of this overlapping fold and unite at its
middle point.

The anterior half of the pygidium is marked by transverse lines
similar to those on the falcate ends of the pleura.

Associated with the entire specimen (PL xxxi, fig. i) is a doublure
with hypostoma attached. The doublure (PI. xxxii, fig. 3) is
partly overlapped by the cephalic shield of which it was evidently
once a part. Its inner margin is beaded by a row of closely placed
and deep indentations which are still more strikingly and clearly pre-
sented in part of another doublure (PI. xxxii, fig. i).

The attached hypostoma, whilst distorted, presents enough of its
serrated lateral margin to identify it with the entire specimen (PL

XXXII, fig. 3) from the same locality.

The immediate anterior margin is entire, but its lateral extension, on
each side, is serrated, bearing six teeth. The number of teeth, as de-
termined by all the hypostomas I have thus far found, seems to be
constant. Moreover the fact that, in the same localities, specimens of
Olenellus thompsoni are found in greater abundance than those of
Olenellus ivalcottanus^ and that the type (PL xxxii, fig. 3) illus-
trates the only form of hypostoma observed, justifies the conclusion
that the hypostomas of both species are essentially the same in charac-
teristic features.

Formation and Localities. — Lower Cambrian, argillaceous shales,
about three miles north and northwest of York, Pennsylvania, near the
contact line of Cambrian and Trias.

PLATE XXXI.

Olenellus {Hobnia) walcottanus.

Fig. I. A nearly perfect specimen crushed flat in the shale and with the exterior
shell largely exfoliated.
2. A segment of the thorax greatly enlarged, showing the ornamentation of
the exterior surface, and on the falcate portion the inosculating lines
marking the surface of this portion of the segment.

(270)

Proc. Wash. Acad. Sci, Vol.111

Plate XXXI

/

''/

Olenellus (Holmiaj Walcottanus

PLATE XXXII .

Olenellus (^Holmia) walcottanus.

Fig. I. Cast of the under side of the doublure, showing the pits formed bj the
semi-spinose inner margin.

2. A hypostoma attached to the doublure.

3. An unusually well preserved hypostoma, showing the six spines on each
side.

4. An enlargement (X 2) of posterior segments of the thorax and the
pygidium, showing the peculiarities of the surface ornamentation.

(272)

Proc. Wash. Acad. Sci. Vol III

Plate XXXI

!&•-

Fig. 2

'^^*^

Olenelius (Holmia) Walcottanus

PLATE MADE BY PROCESS PATENTED DEC. 26, 1899

PROCEEDINQS

OF THE

WASHINGTON ACADEMY OF SCIENCES

Vol. Ill, pp. 273-295. July 26, 1901.

SYNOPSIS OF THE RICE RATS (GENUS ORY-

ZOMYS) OF THE UNITED STATES

AND MEXICO.

By C. Hart Merriam.

The genus Oryzomys, comprising the native American Rice
Rats, is of tropical origin and attains its highest development
in South and Central America. In Mexico it is generally dis-
tributed except in the most arid parts, and is represented by a
considerable number of species and subspecies. In the United
States it is restricted to the southeast coast region, reaching
northward only as far as southern New Jersey.

Nearly all the species are inhabitants of the Tropical and
Austral life regions, but in the mountainous parts of southern
Mexico and Central America, where boreal types are scarce
and often replaced by modified tropical types, a few forms have
pushed up into the Boreal zone and developed warm woolly
coats and other peculiarities which distinguish them as special-
ized local species. The boreal mountain forms here described
are crinitus from mountains bordering the Valley of Mexico,
and rhabdofs and angusiiceps from mountains in eastern
Chiapas and western Guatemala.

Up to the present time 10 species have been described from
Mexico and Guatemala (including those from the peninsula of
Lower California and the Tres Marias Islands). To these are
here added 20 species and subspecies, bringing the total num-

Proc. Wash. Acad. Sci., July, 1901. (273)

!74

MERRIAM

ber up to 30. In addition to these, 4 are known from the
United States.

The material on which this paper is based comprises 600
good skins with skulls, 350 of which were collected in Mexico
by E. W. Nelson and E. A. Goldman during their field work
under the auspices of the Biological Survey of the U. S. De-
partment of Agriculture. All the specimens are accompanied
by field notes and flesh measurements.

Thirty-five species and subspecies are here recognized from
the United States, Mexico, Guatemala and Nicaragua; these,
with their type localities, are :

Oryzomas palustris (Harlan) Southern New Jersey

natator Chapman Gainesville, Florida
jiatator Jioridanus subs^. nov. Everglade, Monroe Co., Fla.

aquaticus Allen Brownsville, Texas

fenmsidce Thomas Santa Anita, Lower California

nelsoni'hl.Q.xxxQ.vci Maria Madre, Tres Marias Ids.

albiventer sp. nov. Ameca, Jalisco, Mexico

cozumelce sp. nov. Cozumel Id., Yucatan, Mexico

fulgens Thomas ? Valley of Mexico

crinitus sp. nov. Tlalpam, Federal District, Mexico

crinitus aztectis subsp. nov. Yautepec, Morelos, Mexico

mexicanus Allen Tonila, Jalisco, Mexico
?nexicanus peragrus subsp.

nov. Rio Verde, San Luis Potosi, Mex.

richmo7idi sp. nov. Escondido River, Nicaragua.

jalapa; Allen Jalapa, Vera Cruz, Mexico

ialapce rnjimis subsp. nov. Catemaco, Vera Cruz, Mexico

zygomatictis sp. nov. Nenton, Guatemala

teape?isis sp. nov. Teapa, Tabasco, Mexico

buUeri Allen Valle de Banderas, Jalisco, Mexico

rufus sp. nov. Santiago, Tepic, Mexico

couesi (Alston) Coban, Guatemala

goldmani sp. nov. Coatzacoalcos, Vera Cruz, Mexico

chap77ia7ii Thomas Jalapa, Vera Cruz, Mexico
chapmani saturatior subsp.

nov. Tumbala, Chiapas, Mexico
ckap??iani d Hut tor suhsp. nov. Huauchinango, Puebla, Mexico
chap77iani caudattis subsp.

nov. Comaltepec, Oaxaca, Mexico

palatimis sp. nov. Teapa, Tabasco, Mexico

hylocetes sp. nov. Chicharras, Chiapas, Mexico

rhabdops sp. nov. Calel, Guatemala

angusticeps sp. nov. Volcan Santa Maria, Guatemala

melanotis Thomas San Sebastian, Jalisco, Mexico

RICE RATS 275

rostratus sp. nov. Metlaltoyuca, Puebia, Mexico
rostratus 7)iegadoii subsp.

nov. Teapa, Tabasco, Mexico

y7icata7tensis sp. nov. Chichenitza, Yucatan

jfulvescetzs (De Saussure) Orizaba, Vera Cruz, Mexico

These species fall naturally into four groups, which, from
their typical species, may be named the pahistrts-mexicamis ,
the chafmani, the melanotis, and the fulvescens groups. Two
of these, the mexicmms 2indi chapmani ^ro^x^^^ are partly bridged
by intermediate forms.

1. The ■pahistris-7nexicanns group is characterized in the
typical forms by rather large size, short ears, large and massive
skulls, long incisive foramina, and strongly developed super-
ciliary beads. All of the known large species, and a few of
medium size, belong to this group. The members may be ar-
ranged in two series according to the color of the underparts,
as follows :

{a) White-bellied forms : -pahistris^ natator, ^entnsul<^, albi-
venter, cozmnelcB (sometimes pale fulvous), nelsoni, aztectis^
inexicamis (belly sometimes buffy), btilleri^ rtifiis.

{U) Fulvous bellied forms : aquatictts, cozunielce (usually
\^\)SX.€), fidgens ^ crmittis, j'alapce, teapensis, zygomatiais, gold-
mam, coziest', mexicmius (belly usually white), mexicamis ^ara-
grus, richniondi.

2. The cha^Tiiani group is characterized by rather small size,
dark coloration, and small mouse-like skulls, with only faint
superciliary beads. The species are chap?naiii, palatmus,
hylocetes, rhabdoj)s, angtistice^s .

3. The melanotis group is characterized by medium size,
large dark ears, and relatively short incisive foramina. The
species are melanotis, rostratus, yucatanensis.

4. T^x^ fulvescens group is characterized by extremely small
size, yellowish fulvous coloration, tiny thin and papery, mouse-
like skulls, without supraorbital beads, resembling the smaller
species of Reithrodontomys . Only one species has been de-
scribed. It is smaller than a house mouse, has a yellowish
fulvous body, long and slender tail, and rather large black
ears. If of larger size it would doubtless be separated into sub-
species.

276 MERRIAM

In studying the species of Oryzotnys it is very necessary to
use skulls of fully adult and even old individuals, as in many
forms the specific characters do not develop until rather late in
life. The males are larger than the females and usually have
the characters more accentuated. The cranial characters that
have proved most useful are the form and thickness of the zygo-
matic arches, length and breadth of rostrum, form of nasals,
degree of development of supraorbital (or superciliary) beads,
size and form of incisive foramina (or palatal slits), and size of
molar teeth.

ORYZOMYS PALUSTRIS (Harlan).

Mus palustris Harlan, Am. Journ. Sci. and Arts, xxxi. No. 2, pp. 385-386,

1837-
Hesperomys {Oryzoviys) palustris Baird, Mammals N. Am., 482, 1857.
Oryzotnys palustris texensis Allen, Bull. Am. Mus. Nat. Hist., Vol. vi,

177-179, May 31, 1894.1

Type locality. — Southern New Jersey (near Salem, Delaware
Bay).

Characters. — Size medium or rather large (about equaling a half-
grown house rat, which the animal greatly resembles) ; ears rather
small; tail rather short; color grayish; belly white.

Color. — Upperparts grizzled grayish brown ; sides somewhat buffy ;
back copiously mixed with black hairs ; underparts white, the slaty
underfur showing through; tail dusky above, whitish below; feet
whitish.

Cranial characters. — Skull of medium size, somewhat flattened,
with moderately developed superciliary beads; rostrum elongate;
nasals long and tapering, wedgeshaped, ending about on plane with
premaxillse ; zygomata rather narrow, their outer sides nearly parallel ;
palatal slits exceedingly long. Skull very like that of mexicatius
but flatter, nasals longer and less decurved ; palatal slits longer.

Measurements. — Average of 9 specimens from Raleigh, N. C. :
total length 231 mm. ; tail vertebrce 117; hind foot 29.

Note. — In the absence of specimens from the type locality (south-
ern New Jersey) the above description has been drawn from speci-
mens from Dismal Swamp, Va., and Raleigh, N. C.

' I am unable to distinguish Allen's subspecies texensis, either externally or by
the skulls, from O. palustris from Raleigh, N. C, and Dismal Swamp, Va.
The braincase may average a trifle narrower, but the difference is very slight.

RICE RATS 277

ORYZOMYS NATATOR Chapman.

Oryzomys palustris ftatator Chapman, Bull. Am. Mus. Nat. Hist., Vol. v,
pp. 44-46, March 17, 1893.

Type from Gainesville, Florida.

Characters. — Size very large ; ears short ; tail long ; pelage long
and full ; color fulvous brown with an extensive dark dorsal patch ;
underparts white ; skull large, long and massive ; general appearance
of upperparts suggesting the muskrat.

Color. — General color of upperparts grayish bister, becoming buffy
gray on sides, and buffy fulvous on rump and posterior back ; head
and back darkened by copious admixture of long brownish black hairs
which are so profusely abundant as to form a broad, ill-defined patch
covering the greater part of back and rump ; underparts white, the
slaty underfur showing through ; tail brownish black above, soiled
whitish below.

Cranial characters. — Skull large and massive, elongate, flattened
on top, with strongly projecting everted superciliary ridges, which
greatly increase the breadth across the orbits posteriorly ; zygomata
moderately spreading ; palatal slits large and open. Compared with
palustris the skull is much larger and heavier, with broadly expanded
supraorbital ridges and much larger and more broadly open incisive
foramina.

Measurements. — Average of 6 males from Lake Kissimmee, Fla. :
total length 295 mm. ; tail vertebrae 151 ; hind foot 35.

Remarks. — Specimens from Cape Canaveral agree in color with
natator but are decidedly smaller. Their skulls are intermediate be-
tween those of natator and palustris.

ORYZOMYS NATATOR FLORIDANUS subsp. nov.

Type. — From Everglade, Monroe County, Florida. No. 71349, $
ad. U. S. National Museum, Biological Survey Coll. March 29,
1895. J. Alden Loring. Original number 2819.

Characters. — Similar to natator but much smaller, with cheeks
and sides more inclining to fulvous; skull decidedly shorter, with
more broadly spreading zygomata (outer sides nearly parallel) .

Measurements. — Average of 6 males from type locality: total
length 367 mm. ; tail vertebrae 137; hind foot 33.

ORYZOMYS AQUATICUS Allen.
cus Allen, Bull. Am. Mus.

Type locality. — Brownsville, Texas.

Oryzomys aquaticus Allen, Bull. Am. Mus. Nat. Hist., New York, iii, 289,
June, 1 89 1.

278 MERRIAM

Characters. — Size large ; ears small ; color light fulvous ; belly
yellowish fulvous.

Color. — Upperparts yellowish or buffy fulvous, moderately lined
vi^ith black hairs; underparts buffy yellow or buffy fulvous; ears
brown, only slightly darkened toward margins; tail brown above,
yellowish below.

Cranial characters. — Skull large and heavy, with marked super-
ciliary beads which in old skulls extend posteriorly over sides of brain-
case and are strongly tJiciirved before reaching occiput.

Remarks. — The only species resembling aquatictis externally is
albiventer^ which may be distinguished at a glance by its white
belly.

Measurements. — Average of 4 adults from type locality: total
length 2S8 mm. ; tail vertebree 149; hind foot 32.5.

ORYZOMYS PENINSULA Thomas.

Orysomys peninsula: Thomas, Ann. & Mag. Nat. Hist., 6th Ser., xx, pp.
548-549, Dec, 1897.

Type locality. — Santa Anita, Peninsula of Lower California,
Mexico.

Characters. — Size rather large ; ears rather small ; tail short ; color
grayish ; belly whitish ; skull broad and massive.

Color. — Upperparts grayish brown, becoming pale fulvous on rump ;
back only moderately mixed with black hairs; underparts soiled
white, the slaty basal fur showing through ; feet whitish ; tail dusky
or brownish above, soiled whitish below.

Cranial characters. — Skull broad and massive with broadly and
rather squarely spreading zygomata ; slender tapering nasals ; long
premaxillae reaching far behind nasals; slight superciliary beads;
large and broadly open palatal slits, and rather large molars. The
skull is so unlike any known species that no comparisons are necessary.
It is about the size of mexicanus but has a widely different aspect
owing to the broadly and squarely spreading zygomata. The frontal
region is lower and narrower than in mexicanus., nasals more slender,
premaxillae longer, incisive foramina larger ; molars about the same size.

Measurements. — Topotype( 9 ad): total length 257 mm.; tail
vertebra 138; hind foot 33.

ORYZOMYS NELSONI Merriam.

Oryzomys nelsoni Merriam, Proc. Biol. Soc. Washington, Vol. xii, p. 15,
January 27, 1898.

Type locality. — Maria Madre, Tres Marias Ids., Mexico.

RICE RATS 279

Characters. — Size very large ; ears medium or rather small ; hind
feet and tail long; color pale fulvous; skull unique : large, massive;
with greatly swollen and strongly decurved rostrum.

Color. — Upperparts ochraceous or buffy fulvous, most intense on
rump, becoming buffy grayish anteriorly, especially on face ; under-
parts white, the slaty underfur showing through ; tail brown, paler
below. Ears brown ; feet whitish.

Cranial characters. — Skull large, massive, somewhat flattened
above, with strongly marked and nearly straight superciliary beads,
rather heavy zygomata, bowed outward in middle ; acute decurved
nasals, ending essentially on plane of premaxillae ; enormously swol-
len and strongly decurved rostrum; moderate palatal slits; rather
large molars (not large for size of skull), and large, broad, and
shortly curved upper incisors.

Measuretnents. — Average of 2 adult males from type locality : total
length 343 mm.; tail vertebrse 190; hind foot 38.5. An adult
female: total length 320; tail vertebrae 185; hind foot 37.

ORYZOMYS ALBIVENTER sp. nov.

Type from Ameca, Jalisco, Mexico. No. 82236 $ ad. U. S.
National Museum, Biological Survey Coll. February 6, 1897. E.
W. Nelson and E. A. Goldman. Original number 10478.

Characters. — Size large ; ears short ; color light golden fulvous ;
underparts ivhite. Skull and teeth large and heavy.

Color. — Upperparts bright ochraceous, brightest on rump, more
grayish anteriorly ; back strongly lined with black hairs ; underparts
clear white without buffy or salmon suffusion ; tail bicolor, dusky
above, whitish below.

Cranial characters. — Skull large and heavy ; zygomata rather
broadly spreading ; nasals large and broad, slightly exceeding pre-
maxillae; supercilary ridges well developed, nearly straight; in-
cisive foramina long, but not so broadly open as in crinitus ; molars
larger and heavier than in any other known species, the series
measuring over 5 mm.

Remarks. — The only near relative of albiventer seems to be
aquaticus from the lower Rio Grande region in Texas. Externally it
resembles aquaticus closely as seen from above, although the rump is
more intense fulvous, but the underparts are white, while in aquaticus
they are buffy fulvous. The skull differs from aquaticus in having
much less pronounced superciliary beads, less marked orbital angle
of frontal [frontals less swollen anteriorly] ; larger incisive foramina,
and decidedly larger molars.

28o MERRIAM

Measurements. — Type specimen ( (? ad.): total length 2SS mm.;
tail vertebra 153; hind foot 37.5. Average of 7 adult males from
type locality: total length 295; tail vertebriE 163; hind foot 37.3.
Average of 2 adult females from type locality: total length 287;
tail 155; hind foot 33.7.

ORYZOMYS COZUMEL^ sp. nov.

Type from Cozumel Island, Yucatan, Mexico. No. 10S462 $ ad.
U. S. National Museum, Biological Survey Coll. April S, 1901. E.
W. Nelson and E. A. Goldman. Original number 14666.

Characters. — Size large; similar to O. aquaticus.ihut darker; ears
and hind feet larger ; tail much longer and much darker.

Color. — Upperparts dark grayish bister with pale fulvous suffusion
on sides and rump ; in old pelage back (es^Decially rump) rusty red ;
underparts varying from soiled whitish to pale buffy salmon; ears
dark brown, darkest on outer half; tail dusky, paler below.

Cranial characters. — Skull large and heavy, with strongly marked
superciliary beads, long rostrum, and long incisive foramina. Closely
related to aquaticzis^ from which it differs in the following characters :
anterior roots of zygomata more depressed and less spreading; orbital
angle of frontal less marked ; posterior part of braincase broader, car-
rying the lateral beads outward posteriorly, so that they form almost
a straight line from side of occiput to angle of orbit ; incisive foramina
longer and more open.

Re7narks. — Compared with aqiiaticus., the only species to which
it bears any near resemblance, the color in fresh pelage is grayer and
darker (less golden fulvous) ; in worn pelage redder — the rump and
hinder part of back more rusty ; ears and face darker, the face strongly
grizzled with black hairs. The underparts are never buffy yellow as
in aquaticus.

AIeasure?fients. — Type specimen ( $ ad.) : total length 333 mm. ;
tail vertebrae 182; hind foot 35. Average of 5 adults from type lo-
cality: lotal length 315 ; tail vertebrae 1 76 ; hind foot 34.5.

ORYZOMYS FULGENS Thomas.

Oryzomys ftilgens Thomas, Ann. & Mag. Nat. Hist. (6), xi, 403-404, May,
1893.

Type locality. — Mexico, exact locality unknown. Collected by A.
Boucard in southern Mexico, probably in or near the Valley of
Mexico.

RICE RATS 251

Characters. — "Size large. Fur very thick, coarse and wooly.
General colour above bright fulvous, brighter than in any other Cen-
tral-American species; anterior half of the body, including the head,
rather paler and duller than the posterior half. Ears decidedly small,
broadly rounded, thinly hairy, their hairs practically the same colour
as those of the head in general, so that they are not distinguishable by
colour at a distance. Lips, chin, throat and inguinal region whitish,
belly with a strong suffusion of fawn, which reaches a maximum on
the breast between the forelegs, passage of upper colour into lower
quite gradual. Outer sides of limbs like back, inner sides whitish;
upper surfaces of hands and feet thinly clothed with pale silvery-fawn
hairs. Tail long, thinly haired, the scales not hidden by the hairs;
above blackish, below yellowish, darkening towards the tip."

Cra7iial characters. — " Skull readily distinguishable from all allied
species by its great breadth, the bold expansion of the zygomata, and
especially by the evenly incurved outline of the supraorbital edges ; in
all other species these edges form two approximately straight lines
diverging from the narrowest interorbital point, but in O . fulgens the
whole inner wall of the orbit forms one even curve, the breadth at the
posterior end of the olfactory chamber being scarcely greater than at
the anterior end. Nasals broad and flattened. Frontal premaxillary
processes very narrow and barely attaining to the same level as the
back of the nasals. Anterior palatine foramina large, widely open,
their posterior margin just level with the front of m^."

Remarks. — In the absence of an authentic specimen of this species
I have been obliged to quote Thomas' original description as above.
Since it was written, several species have been discovered which share
part of the characters mentioned.

Measure?ne?zts. — Type specimen (male adult) measured from the
skin by Thomas : head and body i6o mm. ; tail 151 ; hind foot 37.5 ;
ear from notch 13.3.

ORYZOMYS CRINITUS sp. nov.

Type. — from Tlalpam, Federal District, Mexico. No. 501 82 $
ad. U. S. National Museum, Biological Survey Coll. Nov. 30, 1892.
E. W. Nelson and E. A. Goldman. Original number 3905.

Characters. — Size large, pelage long, full and soft; hind feet long
and large ; tail bicolor ; ears rather short.

Color. — Upperparts buffy fulvous, most intense on rump ; back
copiously lined with and obscured by black hairs ; underparts buffy
(except on chin, throat, and forelegs, which are whitish) ; tail dusky
above, yellowish below.

282 MERRIAM

Cranial chai-acters. — Skull large and broad ; anterior roots of
zygomata spreading, the outer sides as seen from above more nearly
parallel than usual (instead of strongly approximating anteriorly),
but dipping down below level of posterior root; nasals broad and
blunt posteriorly ; superciliary beads moderate ; incisive foramina
very large and broadly open; molars large (series 5 mm.). The
skull is very close to that of aztecus from Yautepec, Morelos, from
which it differs in slightly larger and higher braincase, heavier zygo-
mata, slightly larger bullae, palatal slits larger and more broadly
open (posteriorly) ; nasals broader posteriorly ; premaxillae slightly
shorter; molars broader anteriorly.

Remarks. — This strongly marked species appears to be closely re-
lated to O. fulgens of Thomas, the type locality of which unfortu-
nately is unknown. Externally it differs ixox-a fulgens^ as described
by Thomas, in being much less bright fulvous. Thomas states that
fulgens is brighter than any of the species known to him, and in the
same paper describes O. melanotis^ which is very much moi^e highly
colored than crinitiis. The skull of crinitus ( $ ad.) is about 2 mm.
shorter than Thomas' measurements of the type of fulgetis; it is
also narrower across the zygomata and has shorter nasals. It is not
unlikely that crinitus will prove only a subspecies oi fulgens. Crani-
ally it is related to aztecus^ as shown above, but externally the resem-
blance is less close. It is larger than azteczcs, has smaller ears, larger
feet, longer tail, longer and softer pelage, and the belly is buffy-ful-
vous instead of white.

Measurements. — Type specimen ( $ ad.) : total length 307 mm. ;
tail vertebrae 161 ; hind foot 37.

ORYZOMYS CRINITUS AZTECUS subsp. nov.

Type from Yautepec, Morelos, Mexico. No. 51 173 $ ad. U. S.
National Museum, Biological Survey Coll. Jan. 16, 1S93. E. W.
Nelson and E. A. Goldman. Original number 4290.

Characters. — Closely related to crinitus^ from which it differs in
slightly smaller size and paler coloration; sides and rump pale buffy
fulvous; underparts 7f/^//e instead buffy fulvous. Old specimens in
worn pelage resemble albiventer.

Color. — Upperparts buffy fulvous or ochraceous, redder in old
pelage, moderately lined with black hairs ; underparts white ; tail
brownish above, paler below.

Cranial characters. — Skull similar to that of crinitus but nasals
and zygomata more slender; bullae smaller; incisive foramina not
quite so large.

RICE RATS 283

Meastiretncnts. — Type specimen ( Z ad.) : total length 290 mm. ;
tail vertebrae 154; hind foot 35.

Remarks. — A specimen from Tehuantepec (No. 734^7 ^ ) ^gi'ees
externally with an old specimen from Piaxtla, Puebla, and in cranial
characters agrees in the main with aztecus^ having the same downward
bowed zygomata and so on, but differs in having the palatal slits more
broadly open posteriorly.

ORYZOMYS MEXICANUS Allen.

Oryzoinys mexicatius Allen, Bull. Am, Mus. Nat. Hist., ix, pp. 52-53,
March 15, 1897.

Type from Hacienda San Marcos, Tonila, Jalisco, Mexico.

Characters. — Size medium; hind feet large; ears rather small
(smaller than \x\ jalap en) \ color grizzled fulvous brown; underparts
white or buffy ; skull massive.

Color. — Upperparts ochraceous or pale fulvous, deepest on rump,
strongly lined with black hairs: underparts white, the belly some-
times suffused with buffy ; ears not conspicuously darker than sur-
rounding parts ; tail dark brown or dusky, paler below.

Cranial characters. — Skull broad and massive, with strong zygo-
mata, well defined superciliary ridges, long palatine slits, and rather
heavy molars. Compared with jalapce the skull is broader and more
massive, the zygomata much heavier, especially anteriorly ; supercil-
iary beads more pronounced ; molars heavier.

Measurements. — Average of 2 specimens from Armeria, Colima :
total length 275 mm.; tail vertebrae 139; hind foot 33.

Remarks. — Oryzotnys mexicanus has a wide range in western
Mexico. The Biological Survey Collection contains specimens col-
lected by Nelson and Goldman from Rosario ni southern Sinaloa,
across the states of Tepic, Jalisco, Colima, Guerrero, and Oaxaca, to
the isthmus of Tehuantepec.

ORYZOMYS MEXICANUS PERAGRUS subsp. nov.

Type from Rio Verde, San Luis Potosi, Mexico. No. 821 11 $
yg.-ad. U. S. National Museum, Biological Survey Coll. January 8,
1897. E. W. Nelson and E. A. Goldman. Original number 10398.

Characters. — Similar to mexicaiius but tail decidedly longer and
rather sharply bicolor ; color very much grayer (less fulvous) ; back
more copiously and evenly mixed with black hairs ; underparts deeper
buffy-fulvous ; ears darker; ascending arms of premaxillae a little
heavier posteriorly ; bulla larger ; incisive foramina larger ; molars
largfer and heavier.

284 MERRIAM

Remarks. — Externally peragrus resembles jalapce more closely
than it does 7nexicanus., but the skull is nearer mexicanus, being larger
and broader \.ha.n Jalapce^ with heavier zygomata and much more pro-
nounced superciliary ridges. The bullae and molars are decidedly
larger than in mexicanus.

Measurements. — Type specimen (^ yg.-ad.): total length 294
mm. ; tail vertebrae 167; hind foot 35.

ORYZOMYS RICHMONDI sp. nov.

Type from Escondido River (50 miles above Bluefields), Nicar-
agua. No. ffrrf ^ ^^' U" ^' National Museum, Biological Survey
Coll. June 31, 1S92. Chas. W. Richmond. Original number 63.

General characters. — Size rather large ; ears rather short ; pelage
rather coarse ; color grizzled fulvous. Animal closely related to
mex ica?ius ., but much darker.

Color. — Upperparts grizzled yellow'ish fulvous, everywhere (but
especially on top of head) conspicuously lined and darkened by coarse
black hairs ; black hairs least abundant on rump which in some speci-
mens (as in the type) is intense fulvous or even rusty; underparts
ochraceous fulvous, chin and lips paler; tail dark, paler below but not
distinctly bicolor.

Cranial characters. — Skull rather large and heavy with well marked
superciliary beads, moderately spreading zygomata, and rather long
and relatively slender rostrum. Compared with mexicanus, with
which the skull agrees in size, the rostrum averages decidedly longer
and more slender, with longer nasals (which usually end behind pre-
m axillae).

Measurements. — Type specimen (^ ad.): total length 295; tail
vertebrae 150; hind foot 33.5. Average of 10 ^ 's from type locality :
total length 275 ; tail vertebras 137; hind foot 31.3.

ORYZOMYS JALAPS Allen.

Oryzomys jalapcB Allen, Bull. Am. Mus. Nat. Hist. New York, ix, 206-207,
June 16, 1897.

Type from Jalapa, Vera Cruz, Mexico.

Characters. — Size medium, about equalling mexicanus ; hind foot
smaller than in mexicanus; ears larger and darker; general color
grizzled buffy fulvous; skull rather slender.

Color. — Upperparts buffy fulvous or ochraceous, conspicuously
lined with black hairs ; underparts strongly ochraceous except chin
and throat, which are whitish; ears and upper side of tail dusky,
underside of tail yellowish.

RICE RATS 285

Cranial characters. — Skull similar in general to that of inexicatius
but narrower and lighter, with much slenderer zygomata and less dis-
tinct superciliary beads; molars decidedly smaller.

Remarks. — Externally jalapce resembles mexicanus rather closely
in size and color, but it is easily distinguished by its much larger ears
and smaller hind feet, and also by the strong fulvous underparts, the
underparts in mexicanus being white or buffy. The cranial differ-
ences have been already pointed out. O. jalapce is common in most
parts of the State of Vera Cruz and in parts of northwestern Puebla.

MeasureTiients. — Average of 6 specimens from Orizaba, Oaxaca :
total length 266 mm. ; tail vertebras 140.5 ; hind foot 31.

ORYZOMYS JALAPS RUFINUS subsp. nov.

Type from Catemaco, Vera Cruz, Mexico. No. 65499 ? ^^* U.
S. National Museum, Biological Survey Coll. April 27, 1894. E.
W. Nelson and E. A. Goldman. Original number 61 12.

Characters. — Similartoya/a/^ but back, and especially rump^ much
redder^ the color intense fulvous ; tail dusky all round., the underside
even to very base being dark, and the white hairs smaller and less
abundant; incisors heavier than \xi jalapce ; molars in female (type
specimen) as large as in the largest male oi jalapce.

Measure77ients. — Type specimen ($ ad.): total length 270 mm.;
tail 139; hind foot 32.

ORYZOMYS ZYGOMATICUS sp. nov.

Type from Nenton, Guatemala. No. 76794 $ ad. U. S. Na-
tional Museum, Biological Survey Coll. Dec. 15, 1895. E. W.
Nelson and E. A. Goldman. Original number 879S.

Characters. — Size and ears medium ; color dull fulvous ; size and
external appearance much as in jalapce but hind feet larger ; ears
slightly smaller; underparts much paler; skull characters distinctive.

Color. — Upperparts dull pale fulvous, conspicuously lined on middle
of back with black hairs ; underparts in type specimen white with
slight buffy suffusion, strongest between forelegs ; in younger speci-
mens suffusion deeper.

Cranial characters. — Skull similar in size and general form to
jalapce^ but superciliary ridges much more stro7igly developed and
everted; nasals narrower on middle and posterior thirds ; zygomata
strongly and evenly bozued 02itzuard and dowmvard., sinking far be-
low level of posterior root — instead of approximating and sloping
forward and upward from posterior root as usual.

286 MERRIAM

Remarks. — Externally this species resembles viexicamis as closely
(perhaps more closely) than it does Jalapce, and its relationship to
mexicanus appears to be close. But the skull differs from that of
mexicanus in the curious form of the zygomata, which are strongly
bowed outward and downward ; in the narrower nasals, narrower
rostrum, smaller braincase, and slightly smaller molars. In the form
of the zygomata and some other respects it suggests aztectis.

Measurements. — Type specimen ( $ ad.) : total length 390 mm. ;
tail 152; hind foot 33.

ORYZOMYS TEAPENSIS sp. nov.

Type from Teapa, Tabasco, Mexico. No. 99973 $ yg.-ad. U. S.
National Museum, Biological Surv^ey Coll. April 4, 1900. E. W.
Nelson and E. A. Goldman. Original number 14100.

Characters. — Similar in general to O. jalapcB but ears decidedly
smaller; hind feet (and tail.^) shorter; pelage shorter and coarser;
color much darker; rostrum much smaller.

Color. — Upperparts very dark grizzled bister, becoming yellowish
on sides, and profusely mixed with black hairs; underparts, except
chin, throat and forelegs, suffused with buffy fulvous ; tail dusky, be-
coming yellowish at base below.

Cranial characters. — Skull small, with poorly developed super-
ciliary beads, and small deciirved rostrum; incisive foramina broadly
open behind premaxillary sutures. It differs markedly ixovs\ jalapce
in the small rostrum and broadly open (posteriorly) incisive foramina.

jMeasure?nents. — Type specimen ( $ yg.-ad.) : total length 259
mm.; tail 129; hind foot 32.5.

ORYZOMYS BULLERI Allen.

Oryzomys bullcri Allen, Bull. Am. Mus. Nat. Hist., ix, 53-54, March 15,
1897.

Type from Valle de Banderas, Jalisco, Mexico.

Characters. — Size medium or rather small ; ears rather small ; color
buffy fulvous.

Color. — Upperparts buffy fulvous, deepest on rump, palest
anteriorly, lined on back with black hairs ; underparts white ; nose
dusky; tail brownish above, yellowish below.

Cranial characters. — Skull similar to that of jalapce but slightly
smaller, with smaller rostrum, more acute nasals, vanishing super-
ciliary beads, and slightly longer molar series.

RICE RATS 287

Rctnarks. — Oryzomys bulleri is most nearly related to Jalapce,
but externally it resembles mexicanus most closely. In fact, judged
from the skin alone, it is merely a small edition of mexicamis — slightly
paler above, white bellied, and with very much smaller feet. But its
light and delicate skull requires no comparison with the larger and
more massive skull of mexicantis.

Measjiremejtts. — Average of 6 specimens from Armeria, Colima:
total length 251 mm. ; tail vertebras 134: hind foot 31.

ORYZOMYS RUFUS sp. nov.

Type from Santiago, Tepic, Mexico. No. 91404 9 old. U. S.
National Museum, Biological Survey Coll. June 20, 1S97. E. W.
Nelson and E. A. Goldman. Original number 11 232.

Characters. — Similar in general to bulleri but smaller and redder ;
pelage softer; skull very short; ears rather small.

Color. — Upperparts deep dull fulvous, the fulvous reaching forward
to head and not obscured on back by black hairs (the black hairs small
and scattering) ; underparts soiled white, the plumbous underfur show-
ing through except on chin and in inguinal region ; tail dusky, paler
below.

Cranial characters. — Skull short and heavy ; superciliary bead
only slightly developed; incisive foramina remarkably long; rostrum
small and short. Compared With, bulleri the skull is decidedly shorter
and more massive; superciliary ridges slightly more pronounced;
rostrum much smaller and shorter ; zygomata more spreading and
more strongly elbowed anterioi^ly ; palatine slits very much longer
and larger ; teeth smaller.

Measurements. — Type specimen ( 9 old) : total length 250 mm. ;
tail 136; hind foot 28.

ORYZOMYS COUESI (Alston).

Hesperomys couesi Alston, Proc. Zool. Soc. London, p. 756, 1876.
Oryzotnys couesi Thomas, Ann. and Mag. Nat. Hist. (6), Vol. XL, p. 403,
May, 1893.

Type locality. — Coban, Guatemala.

Characters. — Size and ears medium ; pelage harsh ; fulvous brown ;
skull light and thin (apparently intermediate between the Jalapce and
chapmajii groups) ; molar series very long.

Color. — Upperparts varying from buffy fulvous to deep fulvous,
conspicuously grizzled with black hairs ; underparts varying from
yellowish fulvous to salmon fulvous; ears and upper side of tail
dusky; tail below whitish ; hind feet whitish.

288 MERRIAM

Cranial characters. — Skull thin, light, rather long and narrow,
with slender zygomata and faint superciliary beads; incisive fora-
mina very long and open (remarkably large for size of skull) ; molar
series long (5 mm.) but not massive.

Remarks. — The above description is based on specimens from Tum-
bala, Chiapas, which are believed to be fairly typical of the species
as redefined and restricted by Thomas (Ann. and Mag. Nat. Hist,, 6
ser., vol. XL, no. 65, p. 403, May, 1S93).

The species is very distinct, and appears to have no near relative.
Compared with jalapce., which it most resembles externally, it differs
materially in cranial characters. The skull is lighter, the rostrum
smaller, zygomata more slender, superciliary beads more nearly obso-
lete, palatal slits much longer and more open; molar series decidedly
longer. Compared with bullert, to which it bears less resemblance
externally, the cranial characters are closer. The skull agrees with
bulleri in size and general characters, but is somewhat lighter and has
larger palatal slits and longer molar series.

Meas7irements. — Average of 5 specimens from Tumbala, Chiapas:
total length 252 mm. ; tail vertebrae 131 ; hind foot 30.5.

ORYZOMYS GOLDMANI sp. nov.

Type from Coatzacoalcos, Vera Cruz, Mexico. No. 78110 9 ad.
U. S. National Museum, Biological Survey Coll. April 11, 1S96.
E. W. Nelson and E. A. Goldman. Original number 95 n.

Characters. — Size small ; ears rather large ; tail long ; color dull
fulvous brown ; skull narrow, smoothly rounded, with faint trace of
superciliary bead.

Color. — Upperparts finely mixed fulvous and black, the black hairs
very abundant; underparts soiled buffy (in one specimen almost sal-
mon) ; tail dusky above, pale below.

Cranial characters. — Skull small and light, narrow, smoothly
rounded, with hardly a trace of superciliary bead; nasals broad,
slightly exceeding premaxillce ; incisive foramina rather long; bullee
rather large and more globular than usual ; molars large and broad
(disproportionally large for size of skull), equalling those of the much
larger ya/a/^.

Remarks. — This species differs so widely from its allies that close
comparison is unnecessary. Its nearest known relative appears to be
couesi from Coban, Guatemala, from which it differs not only in much
smaller size (both externally and cranially), but also in possessing
equally large and heavier (broader) molar teeth. In some respects it

RICE RATS 289

seems to be intermediate between the smaller chapmani group and
the \?^x^QX jalap ce group, with bulleri as the nearest relative in the
direction oi jalap c^. But the gaps in both directions are wide.

Measurements. — Type specimen ($ ad.): total length 233 mm.;
tail vertebras 124; hind foot 30.

ORYZOMYS CHAPMANI Thomas.

Oryzomys chapmani Tnouxs, Ann. & Mag. Nat. Hist., 7th ser., i, 179-180,
February i, 1898.

T'ype locality. — Jalapa, Vera Cruz, Mexico.

Characters. — Size small ; pelage short or rather coarse ; coloration
dark; ears rather large and dark; tail small and slender; skull small,
smooth and mouselike.

Color. — Upperparts finely grizzled fulvous and black, the black hairs
most numerous on top of head and middle of back ; underparts whitish,
the slaty underfur showing through ; ears and tail blackish ; under side
of tail yellowish, except on terminal part, which is dark all around.

Cranial characters. — Skull resembling that of an ordinary white-
footed mouse, Peroi7tysctis\ small, light, smoothly rounded, without
conspicuous superciliary beads ; zygomata rather strongly spreading
anteriorly, the outer sides nearly parallel but slightly nearer together
anteriorly than posteriorly; rostrum rather long; nasals blunt pos-
teriorly, ending about on a plane with premaxillag ; incisive foramina
rather small.

Measurements. — [From Chapman] Average of 1 1 males from
type locality: total length 222 mm.; tail vertebrae 119; hind foot
25.2. Average of 5 females from type locality: total length 219;
tail vertebrae 116; hind foot 24.6.

Average of 2 adults ( $ and $ ) from Jico, Vera Cruz [measured
by Nelson and Goldman]: total length 224; tail vertebras 11 7.5;
hind foot 26.

ORYZOMYS CHAPMANI CAUDATUS subsp. nov.

Type from Comaltepec, Oaxaca, Mexico. No. 68641 $ ad. U. S.
National Museum, Biological Survey Coll. Ji-'^y 31, 1894. E. W.
Nelson and E. A. Goldman. Original number 6516.

Characters. — Shmilav to chapmani hut larger, tail much longer:
color slightly darker ; skull larger and heavier with larger rostrum
and broader nasals.

Measurements. — Type specimen ( ^ ad.) : total length 257 mm. ;
tail vertebrae 141 ; hind foot 30.
Proc. Wash. Acad. Sci., July, 1901.

290

MERRIAM

ORYZOMYS CHAPMANI SATURATIOR subsp. nov.

Type from Tumbala, Chiapas. No. 76183 9 ad. U. S. National
Museum, Biological Survey Coll. October 23, 1895. E. W. Nelson
and E. A. Goldman. Original number 8574

Characters. — Similar to chapmani but decidedly darker, particu-
larly on top of head and middle of back, which are blackish, slightly
' peppered ' with fine points of fulvous ; underparts soiled buffy, in
some specimens salmon ; cheeks fulvous ; ears, hind feet, and tail
blackish.

Measurements. — Type specimen (9 ad.): total length 218 mm.;
tail vertebrae 120; hind foot 25.5.

ORYZOMYS CHAPMANI DILUTIOR subsp. nov.

Typeixova. Huauchinango, Puebla, Mexico. No. 93124 $ ad. U.
S. National Museum, Biological Survey Coll. January 10, 1898. E.
W. Nelson and E. A. Goldman. Original number 12040.

Characters. — Similar to chap7nani but slightly paler ; hind foot
longer ; rostrum and nasals broader ; I'ostrum more swollen at base ;
anterior root of zygoma heavier.

Measurements. — Type specimen (^ ad.): total length 223 mm.;
tail vertebrae 117; hind foot 28.

ORYZOMYS PALATINUS sp. nov.

Type from Teapa, Tabasco, Mexico. No. 99977 $ ad. U. S. Na-
tional Museum, Biological Survey Coll. April i, 1900. E. W. Nel-
son and E. A. Goldman. Original number 14080.

Characters. — Size small ; color dark with intense red-fulvous suf-
fusion. Similar to chap7nani but back redder ; cranial characters dis-
tinctive.

Color. — General effect of upperparts nearly burnt umber, from fine
admixture of intense rufous and black hairs ; sides, to and including
cheeks, pale fulvous; underparts whitish, the slaty underfur showing
through; hind feet whitish; ears and tail dark, the tail paler under-
neath basally.

Cranial characters. — Skull small, light, and slender, with only
faintest trace of supraorbital bead ; zygomata narrow, their outer sides
parallel ; incisive foramina short, broadly spreading posteriorly and
separated by a very broad septum.

Remarks. — Compared with chap7na7ii., its only known near relative,
the general color is redder ; the cheeks decidedly more fulvous ; the

RICE RATS 291

skull narrower, with narrower parallel zygomata, smaller bulla, and
widely different incisive foramina, as above described.

Measurements. — Type specimen ( $ ad.) : total length 209 mm. ;
tail vertebrag 106; hind foot 25.

ORYZOMYS HYLOCETES sp. nov.

Type from Chicharras, Chiapas, Mexico. No. 77605 $ old. U.
S. National Museum, Biological Survey Coll. Feb. 14, 1896. E. W.
Nelson and E. A. Goldman. Original number 9306.

Characters, — Size small ; ears large ; color dark with ruddy suf-
fusion. Similar to palathius from Teapa but skull different.

Color. — Upperparts very dark rufous brown; underparts soiled
whitish, the dark slate basal fur showing through; hind feet dark;
ears and tail blackish.

Cranial characters. — Skull small, light, and narrow, with spread-
ing zygomata, and flat nasals which are shortly truncate anteriorly.
Similar in size and general characters to chapmatzi and palatinus^
with braincase narrow as in palatinus, but zygomata more spreading
than either, and nasals peculiar, being rather flat and shortly and
squarely truncate anteriorly.

Measurements. — Type specimen ( $ ad.) : total length 217 mm. ;
tail vertebrae iiS; hind foot 27.

ORYZOMYS RHABDOPS sp. nov.

Type from Calel, Guatemala (alt. loooo ft.). No. 76813 $ ad. U.
S. National Museum, Biological Survey Coll. Jan. 15, 1S96. E.
Nelson and E. A. Goldman. Original number 9135.

Characters. — Size medium ; ears thin, rather large ; nose and
streak from nose to eye black ; pelage long and woolly; color dark
grizzled fulvous ; underparts soiled white. Skull with broadly spread-
ing squarish zygomata and short incisive foramina.

Color. — Upperparts fulvous, on head and back grizzled and ob-
scured by black hairs, on sides purer and reaching forward to cheeks ;
underparts soiled whitish ; nose and streak from nose to eye blackish ;
sides of nose below black streak pale fulvous ; ears blackish ; tail
dusky brown above, yellowish below. The young are darker and re-
semble angusticeps.

Cranial characters. — Braincase large and rounded ; rostrum long
and slender; zygomata very broadly and squarely spreading, their
anterior roots in adults reaching far ourward, their outer sides parallel
or even slightly broadest anteriorly (in this respect unlike any other
known species) ; incisive foramina short and rather broad.

292

MERRIAM

Remarks. — This very distinct species and its narrow skulled, small
toothed relative, angusticeps., belong to the chapmani-hylocetes group,
of which they are high mountain representatives. They differ from
the other members of the group in having long woolly coats and dis-
tinct face markings.

A specimen from Pinabete, Chiapas ( $ yg.-ad. No. 77604), is
darker than adults of rhabdops and has shorter rostrum and nasals,
and less spreading zygomata. A series from San Cristobal, Chiapas
(all immature), are in the darker pelage of the young rhabdops. They
have long rostrums and narrow zygomata but are too young to warrant
separation.

Measure7nents. — Type specimen (^ ad.): total length 355 mm.;
tail vertebras 141 ; hind foot 29.5. Average of 3 adults from type
locality: total length 250; tail vertebriE 136; hind foot 29.

ORYZOMYS ANGUSTICEPS sp. nov.

Type from Volcan Santa Maria, Guatemala (alt. 9000 ft.). No.
76S16 t ad. U. S. National Museum, Biological Survey Coll. Jan. 32,
1S96. E. W. Nelson and E. A. Goldman. Original number 9190.

Characters. — Size medium, ears large; color very dark fulvous
brown. Externally similar to rhabdops from the high mountains at
Calel, Guatemala, but ears larger, color decidedly darker (upperparts
more finely mixed fulvous and black) ; cranial characters distinctive.

Color. — Upperparts and sides (encroaching on sides of belly) finely
mixed dark fulvous and black, the black hairs most numerous on top
of head and middle strip of back, but not forming a defined darker
area; underparts narrowly buffy, the dark slate underfur showing
through ; ears and upper side of tail blackish ; underside of tail paler
on basal half.

Cranial characters. — Skull light and smoothly rounded, long and
narrow throughout (including braincase, zygomata, and rostrum) ;
nasals very long; supraorbital beads obsolete, incisive foramina rather
large for the group, broadest in middle ; teeth small and light.

Measurements. — Type specimen ( ^ ad.) : total length 345 mm. ;
tail vertebrae 134; hind foot 39.

ORYZOMYS MELANOTIS Thomas.

Oryzomys melanotis Thouxs, Ann. & Mag. Nat. Hist., 6th ser., xi, 404-
405, May, 1893.

Type locality. — San Sebastian, Jalisco, Mexico.

Characters. — Size medium ; ears large and dark ; color orange

fulvous; underparts whitish.

RICE RATS 293

Color. — Upperparts bright intense ochraceous fulvous, moderately
lined with black hairs ; underparts buffy whitish, the slaty underfur
showing through ; tail dusky above, yellowish below.

Cranial characters. — Skull of medium size and strength ; zygomata
moderately spreading; superciliary beads onl}-- slightly developed;
rostrum rather long and straight (not decurved) ; nasals rather broadly
(but not squarely) truncate posteriorly ; incisive foramina rather short
and broadly open posteriorly — differing markedly from the long
slender slits of the ?nexicanus group.

Measurements. — Average of 5 from type locality : total length
235 mm. ; tail vertebrae 128; hind foot 28. 3.

ORYZOMYS ROSTRATUS sp. nov.

Zy/e from Metlaltoyuca, Puebla, Mexico. No. 931 12 $ old. U.
S. National Museum, Biological Survey Coll. Feb. 5, 1898. E.
W. Nelson and E. A. Goldman. Original number 12130.

Characters. — Size medium or rather large; ears large. Similar to
melanotis but larger; pelage coarser (but less harsh than in subspecies
megado7i from the coast region of Tabasco and Campeche) ; cranial
characters distinctive.

Color. — Upperparts ochraceous fulvous abundantly lined with black
hairs ; underparts buffy white, the slaty basal fur showing through ;
ears and tail dark brown; underside of tail irregularly yellowish.

Cranial characters. — Skull large, rather massive, long and flat-
tened; rostrum long and somewhat swollen basal ly by projecting
anterior lip of antorbital vacuity ; nasals large, broad, and flat, ending
about on plane with premaxilloe ; superciliary beads moderate ;
zygomata rather narrow anteriorly and posteriorly but bowed out in
middle; zygomatic or outer lip of antorbital vacuity strongly pro-
duced; incisive foramina medium. Compared with melanotis the
skull is larger ; rostrum much longer and more swollen ; nasals less
squarely truncate posteriorly; zygomata bowed strongly outward,
widest a little behind middle.

Rettiarks. — Oryzomys melanotis is a West Coast species (type from
San Sebastian, Jalisco). Its representatives on the East Coast are
slightly larger, with coarser pelage, larger, flatter and more angular
skulls, with decidedly larger rostrum. They range on the east from
Alta Mira, Tamaulipas, southerly and easterly to Apazote, Campeche,
and perhaps to Yucatan. The southern form (subspecies megadon.,
from Tabasco and Campeche) is redder in color, the pelage is more
hispid, and the molar teeth are heavier than in the northern form.

294

MERRIAM

Specimens of the northern form from Alta Mira have smaller rostrum
and nasals and are not typical.

Measurements. — Type specimen ( $ old) : total length 277 mm. ;
tail vertebrse 141 ; hind foot 32.5. Average of 10 from type locality :
total length 251 ; tail vertebrae 135 ; hind foot 32. The average of 5
topotypes (all young adults) of melanotis\%'. total length 235; tail
vertebrae 128; hind foot 28. 2.

ORYZOMYS ROSTRATUS MEGADON subsp. nov.

Type. — From Teapa, Tabasco, Mexico. No. 9997S $ old. U.
S. National Museum, Biological Survey Coll. March 24, 1900. E.
W. Nelson and E. A. Goldman. Original number 14062.

Characters. — Similar to rostratus\i\i'(. somewhat smaller; pelage
more hispid; color decidedly redder; skull smaller with narrower
rostrum; molars larger and heavier. Also similar to wc/a«c//5, but
hind foot larger; pelage harsher; color redder [averaging warm in-
tense fulvous instead of yellowish or golden fulvous] ; darker from
more bountiful admixture of black hairs ; rostrum and nasals longer
and larger ; nasals less squarely truncate posteriorly ; molars heavier
(broader).

Remarks. — In ranging westward this animal undoubtedly grades
into rostratus^ the type locality of which is Metlaltoyuca, Puebla.

Measurements. — Type specimen ( $ ad.) : total length 272 mm. ;
tail vertebra; 140; hind foot 32.5. A topotype female measures:
total length 237; tail vertebrae 122; hind foot 31.

ORYZOMYS YUCATANENSIS sp. nov. .

Type from Chichenitza, Yucatan, Mexico. No. 108139 $ yg.-ad.
U. S. National Museum, Biological Survey Coll. Feb. 9, 1901. E.
W. Nelson and E. A. Goldman. Original number 14527.

Characters. — Very similar to melanotis but fulvous more extended
anteriorly, covering sides of face to nose \melanotis has a whitish
cheek patch between eye and mouth] ; ears fulvous brown instead of
blackish [blackish in young] ; underparts white ; underside of tail
yellowish throughout; head and back moderately lined with black
hairs.

Cranial characters. — Skull rather small, with distinct superciliary
beads, short incisive foramina, and very small molars. Contrasted
with melanotis the skull is smaller (?), superciliary bead more
prominent ; nasals less squarely truncate posteriorly ; incisive fora-
mina shorter; molars smaller.

RICE RATS 295

Measurements. — Type specimen ( $ yg.-ad.) : total length 235
mm.; tail 119; hind foot 32.

ORYZOMYS FULVESCENS (DeSaussure).

Hesperovtys fulvescens DeSaussure, Rev. et Mag. Zool., xii, pp. 102-103,

March, i860.
Oryzo77iys fulvescens Allen, Bull. Am. Mus. Nat. Hist., N. Y., ix, p. 204,

June, 1897.

Type locality. — Unknown. Collected by Sumichrast, probably in
the vicinity of Orizaba, Vera Cruz, which is here fixed as the type
locality.

Characters. — Size smallest of the known species, with much the
appearance of the smaller species of Reithrodontomys ; color, includ-
ing underparts, yellowish fulvous; skull small, thin, and papery,
about the size of that of the common house mouse, and suggesting the
smallest species of Reithrodontomys.

Color. — Upperparts yellowish fulvous, copiously lined with black
hairs; underparts bright and rather intense yellowish fulvous; chin
and throat paler. In immature specimens the belly is sometimes
whitish.

Cranial characters. — Skull small and thin, without trace of supra-
orbital bead ; braincase swollen and smoothly rounded ; zygomata
slender, almost threadlike, weak anteriorly; incisive foramina short
and broadly open ; nasals rather small, ending about on plane with
premaxillae.

Remarks. — This tiny species has a very extensive distribution,
ranging over considerable parts of Vera Cruz, Oaxaca, Guerrero,
Chiapas, Yucatan, and Guatemala.

PROCEEDINGS

OF THE

WASHINGTON ACADEMY OF SCIENCES

Vol. Ill, pp. 297-362. August 23, 1901.

PAPERS FROM THE HARRIMAN ALASKA
EXPEDITION.

XXIV.

THE WILLOWS OF ALASKA.

[Plates XXXIII-XLII. Text Figures 17-2S.]
By Frederick V. Coville.

The Harriman Alaska Expedition of 1899 furnished an op-
portunity to observe and collect specimens of most of the willows
of the Alaskan coast south of Bering Strait. The species were
distinguished in the tield without much difficulty, but the labor re-
quired to find the correct names of the species was greater, since
it involved a careful review of all the literature on the subject
and an equally careful examination of all accessible Alaskan
collections. It has not been possible to examine the willows
collected in that territory by the expeditions of various European
nations, and in view of that fact I desire to express here the same
hope as did Trautvetter in his admirable treatise De Salicibus
Frigidis published in 1832 :

Errare quidem humanum est, sed discrimen statuimus inter errores,
qui excusari possunt et qui non possunt. . . . Solatio mihi est spes,
vos, benevolos lectores, errores meos in iis numeraturos esse, qui ex-
cusari possint.

Twenty-three species of willows are enumerated in the present
paper. Two of these, Salix arbusculoidcs and Salix myrttllt-
Jblia, were collected only on the Canadian side of the Alaskan

Proc. Wash. Acad. Sci., August, 1901. (297)

298 COVILLE

boundary but so near it as to warrant the assumption that they
occur on the other side also. Several species reported from
Alaska, in local lists, have been omitted, chiefly because the
obvious errors of some identifications throw doubt on all. It is
probable, however, that Alaska contains species additional to
those herein admitted. These are to be expected along the
whole eastern boundary of the territory, where least systematic
collecting has been done and where the chance of intrusion of
species from the interior continental flora is greatest.

Of the twenty-three Alaskan willows, five are trees, 10 to 30
feet high, seven are erect bushes, commonly 2 to 8 feet high,
and eleven are prostrate. The species in these groups are as
follows :

Tree Species.
^S". alaxensis S. behbiana

S. avipli/olia S. nuttallii

S. sitchensis.

Bush Species.
S. arbusculoidcs S. giauca

S. bai'clayi S. niphoclada

S. commutata S. -pulchra

S. richardsoni.

Prostrate Species.
S. arctica S. myrtillifolia

S. chamissonis S. ovalifolia

S.fusccscens S. fhlchuphylla

S. glacialis S. polaris

S. lewcarj)a S. reticulata

S. stolonifcra.

Salix myi-tiUifolia is a rare species of whose habit there is no
precise collector's record. Our specimens indicate that it is
normally a prostrate plant but that when growing in sphagnum
moss it tends to assume a somewhat upright form, a character
observed in Salix fusccsccns when growing in sphagnum bogs
to the south of its more typical range on the northern tundra.
Those bushy willows which range upward to timber line or
northward to the tundra, as for example Salix ftilchra, often

THE WILLOWS OF ALASKA 299

become prostrate from the severe climatic conditions to which
they are exposed, particularly by their long enforced prostra-
tion beneath the snow and the repeated killing of upward shoots
by cold winds. When these adverse conditions are removed
such species assume their normal form. When a tree species
flowers young, as is often true of Salix sitchcnsis, it may easily
be mistaken for a bush willow ; but an observation of the aver-
age height of mature plants will prevent error from such a
source. Under very adverse surroundings even a tree willow
may become prostrate, a condition observed in Sali'x alaxensis
on the gravels at the terminal moraine of Muir Glacier.

Only one of the willows enumerated, mn^lifolia, is confined
to Alaska, and on the other hand only two, retiailata and
glauca, are strictly circumpolar. Between these two extremes of
very restricted and very wide range, the species show three
definite geographic relationships, with Siberia, with the Pacific
Northwest Coast, and with the Interior of British America.
Excluding the strictly circumpolar species, the willows common
to Siberia and Alaska are

S. alaxensis S. Ieioca?'^a

S. arctica S. ovalifolia

S. chamissonis S. -phlebo^hylla

S. fuscescens S. -polai'is

S. ^ulchra.

One of these Siberia- Alaska species, alaxensis, is a tree,
another, ^?^/c//rflr, a bush, and the remainder are prostrate. All
the prostrate species belong to the true arctic or arctic alpine
flora. Some of them overlap into the Hudsonian flora, while
-pulchra and alaxensis are Hudsonian species which abut against
and even intrude into the arctic flora. The Alaskan willows of
Siberian relationship may be considered, therefore, prevailingly
arctic.

The species showing a relationship to the flora of the interior
of the North American continent are :

S. arbusculoides S. ni^hoclada

S. bebbiana S. nuttallii

S. myrtillifolia S. richardsoni.

300 COVILLE

Four of these species, arbusculoidcs, myj'tillifolia, nirphoclada^
and rickardsoni^ are believed to belong to the Hudsonian flora,
and two, behhiana and mUtallii, to the next southerly flora, the
one known technically as the Canadian. Thus, while there is
considerable diversity in the zonal relationship of the species of
this group, they may be considered prevailingly Hudsonian, or
one stage more southerly than the Siberian species.

The remaining species, four in number, belong to the Pacific
coast flora. They are :

S. bai'clayi S. sitchcnsis

S. commtitata S. stolonifcra.

Stolonifera is a local alpine species occupying a restricted
area in the coast range of mountains from the vicinity of Juneau
to Yakutat Bay, and the others are characteristic species of the
so-called Sitkan flora, a modified Canadian flora with some
Hudsonian elements, all adapted to the conditions of great
humidity that prevail along a narrow strip of the Pacific coast
from Oregon to southern Alaska. The average zonal position
of these species is about the same as that of the preceding group.

To facilitate the identification of specimens, two keys have
been prepared, a shorter one intended for use with typical
specimens, and a longer one intended to bring out more pre-
cisely the characters of the various species and to cover the
normal variations. The specimens cited cover primarily the
collections of the Harriman expedition. Secondarily they cover
the collections of Alaskan willows in the National Herbarium,
and in the herbaria of the Missouri Botanical Garden, the New
York Botanical Garden, and Columbia University. The last
three collections, through the courtesy of the directors of these
institutions, were loaned for use in preparing this paper.

As a further aid in recognizing species, each is illustrated by
drawings made by Mr. F. A. Walpole. The drawings for the
plates were made in the field from living plants ; those for the
text figures were made from herbarium specimens.

Key to Alaska Willows, Based on Typical Complete

Specimens.
Ovaries smooth.

Catkins sessile on old wood 6". richardsoni^ p. 315.

THE WILLOWS OF ALASKA 3OI

Catkins terminal on leaf-bearing, though sometimes very short and
small-leaved, branches.
Plant an erect bush or tree.

Plant a tree, the leaves without stipules. .^S. amplifolia^ p. 314.
Plant a bush, the leaves with toothed stipules.

Leaves glaucous beneath, the catkin scales black.

6^. bar clay i^ p. 316.
Leaves green beneath, the catkin scales yellowish.

S. com?nutata^ p. 317-
Plant prostrate.

Leaves crenate about the whole margin.

S. 7nyrtillifolia^ p. 325.
Leaves entire or sparingly toothed toward the base.
Leaves bright green and shining on both surfaces.

vS. leiocarpa^ p. 338.
Leaves glaucous beneath.

Style about 1.5 mm. long S. stolonifera^ p. 333.

Style about .5 mm. or less in length.

Pedicels about three times as long as the nectaries,

style almost wanting S. fuscescens^ p. 329.

Pedicels not longer than the nectaries, style about .5

mm. long S. ovalifolia^ p. 331.

Ovaries hairy.

Catkin scales several times longer than their hairs, filaments hairy

below S. reticulata,, p. 340.

Catkin scales little or not at all longer than their hairs, filaments
smooth throughout.
Catkin scales yellowish throughout.

Pedicels several times the length of the nectaries, in fruit ex-
ceeding the catkin scales S. bebblana^ p. 306.

Pedicels at most three times the length of the nectaries, never
exceeding the scale.
Leaves acute at the base, on petioles several millimeters in

length S. glauca, p. 321.

Leaves rounded at the base, on petioles 2 mm. or less in

length 6'. niphoclada, p. 322.

Catkin scales black, at least at the apex.

Leaves finely and evenly serrate or crenate about the whole
margin.
Plant erect, the leaves oblanceolate, sericeous beneath.

S. arbusculoides ^ p. 323.

302 COVILLE

Plant prostrate, the leaves obovate, smooth beneath.

.S. chamissonis, p. 325.
Leaves entire or remotely denticulate.
Plant prostrate.

Leaves bright green on both sides.

^S". phlebophylla^ p. 336.
Leaves pale or glaucous beneath.

Twigs barely i mm. in diameter, rooting in the second

year S. polaris^ p. 335.

Twigs thicker, seldom rooting.

Catkin ovoid-globose, style almost none.

S. glacialis^ p. 329.
Catkin cylindrical, style well developed.

S. arctica^ p. 326.
Plant an erect bush or tree.

Leaf smooth on both surfaces S. pulchra^ p. 319.

Leaf with some permanent pubescence beneath.

Leaf with a white tomentum of curled hairs beneath.

S. alaxensis^ p. 311.

Leaf with dense or scanty short appressed hairs beneath.

Catkins appearing before the leaves, the permanent

pubescence of the latter beneath very scanty, stamens

2 in each flower 6". mittallii^ p. 310.

Catkins appearing with the leaves, the permanent pu-
bescence of the latter beneath dense and satiny,
stamens i in each flower S. sitchensis^ p. 307.

Key to Alaska Willows, Based on Pistillate Leaf-
bearing Specimens.
Leaves obovate, usually wedge-shaped at the base, glabrous, usually
glaucous beneath, minutely and closely glandular-serrate, not crenate,
about the whole margin ; plant prostrate.

6". chamissonis^ p. 325.
Leaves not with all the characters given above, particularly with refer-
ence to the serration ; plant often erect.
Leaves not exceeding 2 cm. in length, without stipules, bright green
and shining on both surfaces, not glaucous beneath, entire,
glabrous, except for the few long hairs usually present on the
margin, a few of the principal veins standing out very conspic-
uously on the fresh leaves and often forming a skeleton after
a few years from the weathering of the softer matter of the
persistent dead leaves.

THE WILLOWS OF ALASKA 303

Leaves broadly obovate or orbicular, rounded or retuse at the apex,
commonly 4 to 6 mm., and rarely exceeding 10 mm., in length ;
pistillate catkins globose to oblong, few- (usually 2- to 8-)

flowered; ovary smooth S. leiocarpa^ p. 33S.

Leaves oblanceolate to obovate, acute to obtuse at the apex, taper-
ing to the base, commonly 10 to 15 mm. long; pistillate cat-
kins cylindrical, many-flowered ; ovary usually hairy.

6". pJilebophylla^ p. 336.
Leaves not with all the characters given above.

Leaves orbicular to oval, glaucous and veiy prominently reticu-
lated beneath, rugose above ; catkins equaled or nearly equaled
by their peduncles ; scales of the catkins reddish purple, usually
smooth on the outside, several times longer than the hairs of the
margin and inner surface ; capsules hairy ; stigma sessile or

nearly so S. reticulata^ p. 340.

Leaves or other parts not as described above.

Plant prostrate, cespitose ; branches very slender, a millimeter
or less in diameter, becoming firmly rooted ; leaves oval to
orbicvilar or nearly so ; catkins oblong to globose ; ovaries
hairy or sometimes nearly smooth, but not glaucous ; style
plainly developed but less than a millimeter in length.

S. polaris, p. 335.
Plant not with all the characters given above.

Leaves very smooth on both surfaces, except for a few long
hairs (early deciduous) on the margins and midrib when
young, bright green and shining above, glaucovis beneath,
rhombic-oblanceolate or sometimes rhombic-obovate, nar-
rowly wedge-shaped at base and usually at the apex also ;
stipules narrowly linear-lanceolate, pei'sistent, serrate ; cat-
kins sessile or nearly so on old w^ood, the scales black, ob-
lanceolate ; stalk of the hairy capsule about as long as the
nectary; style about i to 1.5 mm. long. S. pulchra^ p. 319.
Leaves or other parts not as described above.

Ovary and capsule smooth, at least over most of its surface.

Plant prostrate or nearly so, less than a foot in height.

Leaves finely and evenly crenate about the wdiole

margin ; catkins almost sessile, but the very short

peduncle bearing a few small leaves.

S. myrtillifoUa^ p. 325.

' Leaves entire, or sometimes with a few scattered teeth

' below, catkins borne on a well-defined leafy peduncle

304 COVILLE

or branch of the season half a centimeter to two
or three centimeters long.

Leaves rhombic-obovate, narrowed toward the
base, broadest above the middle ; pedicel of the
ovary about three times the length of the nec-
tary ; capsule shining, not glaucous ; style almost
none, barely .25 mm. in length.

S. ftiscescens^ p. 329.
Leaves oval to orbicular, sometimes obovate ; ped-
icel of the ovary little or not at all exceeding
the nectary ; style evident.
Capsule glaucous; style about .5 mm. long.

5". ovalifolia^ p. 331.
Capsule not glaucous; style about 1.5 mm. long.

S. stolo7iifera^ p. 333.
Plant an erect bush or tree.

Plant a tree with oval leaves when fully developed 3
to 5 cm. wide, and devoid of stij^ules ; catkins
about i.^ cm. in diameter in flower and often 2 cm.

in fruit 6". ampUfolia^ p. 314.

Plant a shrub, commonly i to 2 meters high ; leaves
usually smaller, commonly oblong-lanceolate
(though sometimes as broad and large as the
last), and with ovate to lanceolate, glandular-den-
ticulate stipules.
Stems densely hirsute ; stipules usually persistent
for several years on the stout stems ; catkins ses-
sile on old wood, all leaves and green bracts ap-
pearing later and from separate buds.

S. rickardsoni, p. 315.
Stems often smooth ; stipules seldom persistent on
old wood ; catkins on short leafy-bracted pe-
duncles appearing with the leaves.
Leaves crenate-serrate (the somewhat gland-
tipped teeth incurved), glaucous beneath,
though this character sometimes obscure in
young leaves ; catkin scales usually black.

S. bar clay I ^ p. 316.

Leaves finely glandular-denticulate (the teeth not

incurved) throughout or only toward the base, or

entire, not glaucous beneath, of nearly the same

THE WILLOWS OF ALASKA 305

color on both surfaces ; catkin scales usually
yellowish or straw-colored.

vS. cotnmutata^ p. 317.
Ovary densely hairy, the capsule by expansion more spar-
ingly so.
Leaves provided beneath with a dense, white, opaque
covering of curled and tangled hairs.

S. alaxefisis, p. 311.
Leaves not as described.

Leaves rather narrowly elliptical-lanceolate, acute at
both ends, minutely and rather closely crenate-den-
ticulate about the whole margin, glaucous beneath,
the lowermost usually smooth on both surfaces, the
upper clothed beneath with fine appressed hairs ;
stalk of the ovary equalling or twice as long as the
nectary ; style less than i mm. long.

S. arbusculoldes, p. 323.
Leaves or other parts not as described.

Catkins appearing before the leaves, the usually
very short peduncle devoid of leaves, though
sometimes with very rudimentary leaves or
bracts ; leaves with permanent appressed pubes-
cence of short hairs beneath, dense when young,
sparse when old, the upper surface with some
similar pubescence above when young, oblong-
lanceolate to obovate-lanceolate, usually acute at
either end ; catkin scales black ; style short, 5

mm. or less in length S. nuttallii,, p. 310.

Catkins appearing w^ith or after the leaves, the fresh
growth on which the catkin is borne always
having either leaves or green leaf-like bracts ;
other organs not all as described above.
Leaves oblanceolate to obovate, tapering to the
base, usually rounded at the apex, entire or re-
motely glandular-denticulate, permanently sat-
iny beneath with a dense covering of very
short, straight, appressed hairs.

S. sitc/ietisis^ p. 307.
Leaves not as described.

Stalk of the ovary few to several times the
length of the nectary, very slender, in fruit

3o6 COVILLE

exceeding the scale and sometimes becoming
twice or three times as long.

S. bebbiana^ p. 306.
Stalk of the ovary not more than twice or
sometimes three times the length of the
nectary but always shorter than the scale.
Plant an erect shrub ; leaves permanently
clothed on the lower surface, and some-
times also the upper, with a usually sparse
covering of long appressed hairs ; scales
of the catkinlightbrownto straw-colored.
Leaves wedge-shaped at the base, on peti-
oles several millimeters in length.

S. glauca^ p. 321.
Leaves rounded or cordate at the base, on
very short petioles, those on our speci-
mens 2 mm. or less in length.

S. iiipJiocIada^ p. 322.
Plant prostrate ; leaves often with some long
silky hairs when young but glabrous at
maturity, or with hairs only on the
margins or midrib; scales of the catkin
almost always black.
Leaf-blade 1.2 cm. or less in length; cat-
kin short, ovoid-globose ; style reduced

to a mere neck S. glacialis^ p. 329.

Leaf -blade usually much larger ; catkin
elongated, cylindrical ; style well de-
veloped S. arctica^ p. 326.

I. SALIX BEBBIANA Sargent. Bebb Willow.

Sali.x rostrata Richardson in Franklin, Journ. Pol. Sea 753. 1823 ; not

Thuill. 1799.
Salix bebbiana Sargent, Gard. & For. 8 : 463. 1895.

A small tree with leaves elliptical-lanceolate or ovate-lanceolate,
either rounded or acute at the base, acute or acuminate at the apex,
sparingly pubescent when young, becoming nearly smooth in age,
usually crenate-denticulate, but sometimes entire, commonly 3 to 5 cm.
long. The pale brown or straw-colored catkin scales, the pubescence
of the slender ovaries, their very long slender pedicels, and the almost
complete absence of a style are especially characteristic of the species.
It is widely distributed, extending almost entirely across British

THE WILLOWS OF ALASKA

307

America. It occurs in most of the northern portions of the United

States, reaching southward in the east to Pennsylvania, in the Rocky

Mountain region to New Mexico and Arizona, and on the Northwest

Coast to Oregon. The original specimens were

collected on Sir John Franklin's first expedition,

in the interior of British America. In Alaska

the species has a very restricted range from the

upper waters of Cook Inlet eastward to the

upper Yukon Valley. Near Homer, in Cook

Inlet, on the brushy portion of the point, it is

reported by Mr. Evans as a shrub or small tree

4 to 15 feet (i.2to>4.5 meters) high, at Kasilof '

10 to 25 feet (3 to 7.5 meters) high and 8 inches

(20 cm.) in diameter.

The following specimens have been examined :

Cook Inlet. — At Homer, Sunrise, and Kasi-
lof, Walter H. Evans, iS97(Nos. 470, 49i),i89S
(No. 693). At Kenai, F. A. Walpole, 1900
(No. 1 143). Between Cook Inlet and the
Tanana River, probably from the lower
Sushitna, E. F. Glenn, 1899.

Tukon Valley. — Near the junction of Forty-
mile Creek and the Yukon, Frederick Fun-
ston, 1S93 (Nos. 39, 42). At Fort Yukon, F. C. Schrader, 1899.

Dr. P. A. Rydberg has published recently,^ while this paper was
in press, a Salix perrostrata., based primarily on specimens collected
by himself near Hermosa, in the Black Hills of South Dakota, and to
that species he has referred a specimen collected by Mr. R. S. Wil-
liams at Dawson, Yukon Territory, besides assigning the species a
range northwestward to Alaska. While his bibliographical references
indicate that the species is a segregate from bebbiana., the author gives
no comparison of distinguishing characters. I am unable to find in
his description anything to distinguish our Alaskan specimens from
what I take to be typical bebbiana.^ whatever may prove to be the re-
lation of that species to the Black Hills willow.

Fig. 17. Salix beb-
biaiia Sargent: a, fruit-
ing catkin, natural
size ; b, capsule, with
pedicel, scale, and
nectary, enlarged two
diameters.

2. SALIX SITCHENSIS Sanson. Satin Willow.

Plate XXXIII.

Salix sitchensis Sanson ; Bongard, Mem. Acad. Petersb. VI. Math. & Nat.
2 : 162. 1831.

^ Variously spelled Kassiloff, Kassilof , Kussilof , Kussilow, etc.
2 Bull. N. Y. Bot. Gard. 2 : 163. 1901.

308 COVILLE

Salix sitc/iettsis is ordinarily a large shrub or small tree, individual
specimens often reaching lo to 15 or rarely 30 feet (3 to 9 meters) in
height, with a trunk four to six inches or occasionally even a foot (10
to 30 cm.) in diameter. In exposed situations it sometimes fiow^ers
and fruits as a small shi'ub a few feet in height, and on the bare
morainal gravels at Muir Glacier it even joined the other smaller
willows in becoming almost prostrate. In addition to the attractive-
ness of a rather graceful form, this willow bears a foliage of singular
beauty. On close inspection the leaves present a satiny sheen of
daintiest delicacy and softness, varying in intensity with the uneven-
ness of the leaf surface. This appearance, largely lost in the dried
specimens, is due to a dense covering, on the louver surface of the
leaves, of closely appressed short hairs. No other Pacific Coast
willow has the same character, and when one has known it in the
field he can, by its foliage alone, readily distinguish this satin willow
from all the other species. The leaves are oblanceolate to obovate,
broadly acute to obtuse at the apex, narrowed at the base, entire or
sometimes remotely denticulate, and when fully developed about 4 to
6 cm. long, except on vigorous shoots, where they often reach 10
cm. The flowers of the male catkins have but a single stamen each.
All other Alaskan willows have two stamens.

Salix sitchensis is a characteristic plant of the Sitkan floral district,
extending from the southernmost limit of Alaska northward and west-
ward along the coast to Cook Inlet and the eastern end of Kadiak
Island. It was observed at as high an elevation as 1,400 feet. Where
the forests of Sitka spruce were dense this willow grew along the
streams and beaches and in similar sunshin}^ situations, but where the
forest was open or wanting it might occur anywhere on the solid up-
land at the proper elevation. South of Alaska, Salix sitchensis ex-
tends along the coast to California, and along the mountains through
British Columbia into the Cascade Range of Washington and Oregon.

The wood of the satin willow is sometimes used by the Indians of
southern Alaska in drying salmon, since its smoke does not give the
fish a bad taste as do some other kinds of fuel. Its pounded bark is
sometimes applied directly to a cut or wound, to assist in healing.
The Tlinkit tribes, which extend from Yakutat Bav nearly or quite to
Dixon Entrance, call it chaktl, or more precisely, c/i^a'-tlk. ^ They
give the same name to other and probably to all the species of willow
in that region. This shows how little the willows enter into their use-
ful arts, for wherever in the United States the willow is employed in

^ The diacritic marks used are those of the Century Dictionary.

THE WILLOWS OF ALASKA 3O9

basket making or other arts, the widely differing qualities of the
various species have caused the Indians to give them discriminating
names. Among the natives of Kadiak the word for various other
species of willow, 7iimuyok (^ni-mo-ydk')^ is doubtless applied to the
satin willow also.

Specimens of siichensis have been examined as follows :

Behm Canal. — At " Shehshooh ' Lake," M. W. Gorman, 1895
(No. 38).

Wrangell. — On the hillside back of the town, Coville and Kearney
(No. 426), Trelease and Saunders (Nos. 3352, 3363), Walter H.
Evans, 1897 (No. 78).

Sitka. — Along Indian River, Coville and Kearney (No. 838), A.
Kellogg, 1867 (175, in part), Walter H. Evans, 1897 (No. 22S). It
was here, undoubtedly, that Henry Mertens, the botanist of Liitke's
Expedition, in an excursion from Sitka to the summit of the neighbor-
ing Mount Verstovia, in the year 1827, discovered this willow. And
here the writer on June 16, 1899, on a similar excursion found it still
growing. The dense forests of spruce about Sitka do not afford suit-
able conditions for the tree, and it was not observed at any other point
in the vicinity. Exactly similar was the situation in Mertens' time. He
says, " Here alone [at the crossing of a ' wild mountain current '] is
seen the solitary species of Salix which the environs of Sitcha afford." ^

Stephens Passage. — At Taku Harbor, Coville and Kearney (No.
481).

Juneau. — In and above Silver Bow Basin, Coville and Kearney
(Nos. 566, 594, 2534), Walter H. Evans, 1897 (No. 155)-

Skagway. — At Glacier on the White Pass railroad, F. A. Walpole,
1900 (No. 1264).

Glacier Bay. — Abundant at various points, Coville and Kearney
(Nos. 624, 633, 634, 698a, 701), Trelease and Saunders (Nos. 3354,
3355? 3377? 3378)? Brewer and Coe (Nos. 38, 39), Kincaid.

La Perouse Glacier. — Near the beach, west of the glacier, Fernow.

Takutat Bay. — At many points, Coville and Kearney (Nos. 998 to
1000, 1082a, 1 121, 1 154), Trelease and Saunders (Nos. 3357? 335^,

3360,3361,3369)- ^ _ .

Kadiak Island. — At English Bay, about 8 miles south of Kadiak*

village, Coville and Kearney (No. 1440).

^ Supposed to be the lake at the head of Yes Baj^

2 See Hooker, Bot. Misc. 3: iS. 1833.

''^This is Woman's Bay of some charts, having been named babia {-voiiian's)
by the Russians. W. H. Osgood and F. A. Walpole report that it is known at
Kadiak by the name English Bay only.

3IO

COVILLE

3. SALIX NUTTALLII Sargent. Nuttall Willow.

Salix flavescens NuTT. Sylv. i : 65. 1842, not Host, 1828.
Salix ituiialluSA.KG^^T, Gard. «S: For. 8: 463. 1895.

A small tree 3 to 5 meters (about 10 to 16 feet) in height and com-
monly 7.5 to 10 cm. (3 to 4 inches) in diameter. The leaves are ob-
ovate to oblanceolate, tapering at the base, acute or sometimes obtuse

at the apex, entire or occa-
sionally somewhat crenate-
denticulate, the lower surface
with some scattered appressed
hairs or, on vigorous shoots,
a somewhat velvety pubes-
cence, but without the satiny
lustre of sitchensis. When in
flower this willow is conspic-
uous for its leafless twigs and
the prominence of the black,
sparsely hairy catkin scales.
The original specimens were
collected by Nuttall "in the
range of the Rocky Mountains "
at some point near the present
route of the Union Pacific
railway from Nebraska to
Oregon.

Our Alaska material is as
follows :

Skag^vay to Lake Ben-
nett.— From Skagway to
Glacier on the White Pass
railway, Coville and Kearney (Nos. 493, 503), F. A. Walpole, 1900
(No. 1076). From Bennett to Log Cabin, F. A. Walpole, 1900
(Nos. 10S3, 1 271). AtLakeBennett, J. B. Tarleton, 1899 (No. 21).
Cook Inlet. — At Halibut Cove in Kachemak Bay, Coville and
Kearney (No. 2416), the specimens in leaf only.

The name Salix nuttallii is ordinarily used to designate a species
(or perhaps a group of species) finding its eastern limit in the Rocky
Mountains and its western limit at the shore of the Pacific. It occurs
as far south as Utah and Arizona, and in California extends south-
ward along the coast to the bay of Monterey and along the Sierra Ne-
vada to the San Bernardino range. Northward from California it is

Fig. iS. Salix 7tHttallii Sargent: «,
pistillate catkin, natural size; b, pistillate
flower, enlarged six diameters ; c, d, leaves,
natural size.

THE WILLOWS OF ALASKA 3II

abundant along the coast and in the mountains to British Columbia.
The form that occurs along the Pacific Coast from California to Brit-
ish Columbia differs in several respects from the Rocky Mountain
plant and has been designated, either as a species or a variety, by the
names brachystachys^ capreoides^ and scolder iana. My attention has
recently been called by Professor C. V. Piper to the fact that this last
name, scouleriana ^ is a valid name, and the oldest name, for the
Pacific Coast plant, and is older than the name nuttallii. It ap-
pears that Andersson in making a critical identification of the type
specimens of scouleriana^- found that the leaf specimen belonged to
sitchensis^ while the flowering specimen, which therefore rei^resented
the real scouleriana., he referred to Salix Jlavescens Nutt. Professor
Sargent, finding Jlavesceits to be a homonym and considering the
Rocky mountain plant and the Pacific Coast plant to belong to the
same species, followed a course which was inadmissible under the
circumstances and gave the species a new name, nuttallii^ instead of
taking up scolder iana. Until a critical revision of these willows has
been made it will probably be most convenient and least confusing to
treat scouleriana and Jiidtallii as distinct species.

Salix nidtaUii., like Salix bebbiana^ evidently reaches Alaska by an
extension of its range in the interior of British America, through the
humid Sitkan flora, to the coast at Skagway and Cook Inlet. Salix
scouleriana apparently does not extend as far north as Alaska. Its
only claim to admission into that flora rests on some imperfect
young specimens from Wrangell, without flowers or fruit, which may
be referable to sitchensis. Kellogg's specimens of scouleria^ia., re-
ported as collected at Kadiak and Sitka, in fact came from Van-
couver, British Columbia.

4. SALIX ALAXENSIS (Anders.) Coville. Feltleaf Willow.

Plate xxxiv.

Salix speciosa Hook. & Arn. Bot. Beech. Voy. 130. 1832, not Host, 1828.
Salix speciosa alaxensis Anders, in DC. Prodr. iS"*: 275. 1868.
Salix alaxensis Coville, Proc. Wash. Acad. Sci. 2 : 280. 1900.
Salix longistylis Rydberg, Bull. N. Y. Bot. Gard. 2: 163. 1901.

This willow differs notably in its foliage from all the other Pacific
Coast species, so much so, indeed, that Hooker and Arnott, the con-
servative authors of the ' Botany of Captain Beechey's Voyage in the
Blossom,' described the plant from specimens devoid of flowers or
fruit. The under surface of the leaves is covered with a dense white

iBarratt; Hook. Fl. Bor. Am. 2 : 145. 1838-9.
2 See Bebb, Gard. & For. 8 : 373. 1S95.

312 COVILLE

wool or felt made up of curled and tangled hairs, presenting an appear-
ance quite different from the velvety, or satiny, or cobwebby pubes-
cence of other species. The leaves are narrowly to broadly elliptical-
lanceolate, acute at the apex and base, smooth or nearly so on the
upper surface, of an apparently thick texture, due largely to the dense
woolly covering of the lo^ver surface, often corrugate-reticulate, the
margins of the very young leaves often minutely glandular-denticulate,
at maturity usually entire and sometimes revolute, the stipules usually
present, narrowly linear-lanceolate to filiform, and entire. The fruit-
ing catkins are very long and thick, some of those collected reaching
a length of 14 cm, and a diameter of 1.8 cm. The stout twigs are
either smooth or densely hairy, and sometimes they have a decidedly
blue color from the presence of a conspicuous bloom.

Salix alaxensis extends from the northern part of Alexander
Archipelago westward along the Alaskan coast to the peninsula,
northward along the eastern side of Bering Sea through Bering Strait
to Cape Lisburne, and through the interior of Alaska to the Mackenzie
River in British America. East of Kadiak it is associated with other
species of tree willows, but west of that point it is the only willow
that presents the form and dimensions of a tree. From the Shumagin
Islands eastward full-grown specimens are ordinarily about 20 to 30
feet (6 to 9 meters) in height, with a trunk four to six inches ( 10 to
15 cm.) in diameter. Under suitable conditions it doubtless reaches a
still larger size. On the wind-swept Aleutian Islands, like all other
arboreal vegetation, it appears to be wanting, but on the mainland
to the north it again appears, on Buckland River, at the eastern end
of Kotzebue Sound, reaching a height of 16 to 20 feet (according to
Seemann in the 'Botany of the Herald'), farther north in the sound
only eight feet, and at its northern limit, Cape Lisburne, being reduced
to a shrub only two feet high. Like Salix sitchensis^ it becomes
almost prostrate on the naked gravels at Muir Glacier, while only a
few miles away, on older glacial deposits which have been reclothed
with an abundant vegetation of shrubs, it develops into a handsome
small tree, a difference of habit illustrating the marked effect of
different local conditions.

The original specimens of Hooker and Arnott's speciosa came from
Kotzebue Sovuid, those of Andersson's speciosa alaxensis from "Alaxa
Americae occidentali-borealis."

The feltleaf willow holds an important economic relation to the
mining industry and to human existence generally in northern Alaska,
for while spruce timber apparently does not extend north of the divide

THE WILLOWS OF ALASKA 313

separating the waters of the Yukon and the rivers flowing into Kotzebue
Sound from those of the Arctic coast, this tree-willow does extend be-
yond that divide to the headwaters of the Arctic rivers, and in many
places furnishes the only wood available as fuel to Indians, prospec-
tors, and explorers.

Mr. Rydberg's Salix longistylis^ based on a specimen collected at
the mouth of Klondike Creek, Yukon Territory, although compared
by him only with Salix sitchensis^ is clearly identical with the present
species.

Specimens of Salix alaxensis have been examined as follows :

Skagxvay and White Pass. — At Glacier on the White Pass railroad,
and at Skagway, F. A. Walpole, 1900 (Nos. 1071, 1077, 1106).

Lake Bennett. — On the shore and uplands, J. B. Tarleton, 1S99
(No. 12, in part).

Glacier Bay. — At various points about the bay, Coville and Kearney
(Nos. 622, 639, 663, 696, 698, 699, 702), Trelease and Saunders
(Nos. 3344103348, 3364), Brewer and Coe (No. 57), Walter H.
Evans, 1897 (No. 156).

Takutat Bay. — Abundant on the west side of the bay, and at its
northern end and arms, Coville and Kearney (Nos. 996, 105S, 11 20,
1 149), Trelease and Saunders (Nos. 3349, 3359? 347°, in part).

Prince William Sound. — At Port Wells, Coville and Kearney
(No. 1270), Trelease and Saunders (No. 3372), Brewer and Coe
(No. 136).

Cook hilet. — At several points between the head of Cook Inlet and
the Tanana River, E. F. Glenn, 1S99. Sparingly on the delta of a
glacier at Halibut Cove, Kachemak Bay, Coville and Kearney, No.
2419).

Kadiak Islatid. — Sparingly on the lower mountain slopes in Eng-
lish Bay, Coville and Kearney (No. 1441). Near Kadiak village,
Trelease and Saunders (No. 3351), Brewer and Coe (No. 223).

Shu77iagin Islands. — Sparingly at Sand Point and in other localities
on Popof and Unga, Coville and Kearney (No. 1622), Trelease and
Saunders (No. 3343), M. W. Harrington, 1 87 1-2. At Mist Har-
bor, Nagai Island, C. H. Townsend, 1S93.

Alaska Peninsula. — At Kukak Bay, Coville and Kearney (No.
1622), Trelease and Saunders (No. 3350) ; in Stepovak Bay, Palache ;
and at Nushagak, C. L. McKay, 1881.

Yukon Valley. — Headwaters of Koyukuk River, F. C. Schrader,
1899. At Rampart House, on the Porcupine River, Frederick Funston,
1S94 (No. 175). At Dawson, Yukon Territory, R. S. Williams, 1899.

Proc. Wash. Acad. Sci., August, 1901.

314

COVILLE

Kotzebue Sotind. — At Camp Retreat, on the delta of the Kowak or
Putnam River, G. M. Stoney, iSS6.

5. SALIX AMPLIFOLIA Coville. Yakutat Willow.
Plate XXXV.
Salix amplifolia Coville, Proc. Wash. Acad. Sci, 2 : 282. pi. /j. 1900.

A shrub or small tree, ordinarily about 10 to 16 feet (3 to 5 meters)
high, with a trunk three to four inches (7.5 to 10 cm.) in diameter,
but sometimes attaining a height of 25 feet (7.5 meters) with a trunk
a foot (30 cm.) thick. The young twigs are stout and densely villous-
pubescent. The leaves are large, oval to broadly obovate, 5 to S cm.
long when fully developed, entire, or denticulate on the margins below,
somewhat villous when young, smooth or nearly so at maturity, with-
out stipules. The catkins, appearing with the leaves on leafy-bracted
peduncles, are very large and thick, about 1.5 cm. in diameter and 4
to 6 cm. in length, the fruiting catkins often becoming 2 cm. in
diameter. The ovary and cai^sule are smooth, the style commonly 3
to 4 mm. long. Occasionally the ovaries are slightly hairy. The
species is known only from Yakutat Bay, Alaska, where it was dis-
covered by the Harriman Expedition. The tree was first observed on
the west shore of the bay growing on and near the sand dunes that lie
back of the beach, and was afterward collected in Disenchantment
Bay at Egg Island, Hubbard Glacier, and Haenke Island, and later
at the Indian village at the head of Yakutat Bay, east shore. Speci-
mens were collected by Coville and Kearney (Nos. 1013a, 1061, 1062,
1074, 1089, 1 1 22, 1 1 23, 1 153, 1 158), Trelease and Saunders (Nos.
3340 to 3342, 3465), Brewer and Coe (Nos. 89, 90), and Fernow.
The type specimen. No. 1153, was collected at the Indian village, not
as stated in the original description on the west shore of the bay.

Salix amplifolia bears considerable resemblance to richardsoni and
barclayi., but the species from which it differs least is Salix hookeriana
Barratt, which occurs on the sea coast from Oregon to British Colum-
bia. Like affiplifolia, hookeriana lacks the stipules so conspicuous
in richardsoni and in barclayi^ but hookeriana is readily distinguish-
able from the Yakutat tree by its shorter styles, about i mm. in length,
the permanent though sometimes thin tomentum on the backs of the
leaves, the earlier appearance of the catkins, which precede the leaves,
and the absence or rudimentary character of the green bracts on their
peduncles.

THE WILLOWS OF ALASKA

315

6. SALIX RICHARDSONI Hook. Richardson Willow.
Salix richardsoni YioOYi. Fl. Bor. Am. 2: I47- t- ^^2. 1838-9.

An erect bushy willow, from a meter or sometimes less to 2.5
meters in height, the branches stout and hairy,, and retaining the dead
stipules of the preceding year's growth. These stipules are lanceolate,
glandular-serrate, and very large, commonly i to 3 cm. in length.
The catkins, which appear be-
fore the leaves, are sessile on one-
year-old wood, stout, and closely
flowered, with densely long-
hairy scales but smooth long-
styled ovaries and capsules.
The leaves are bright green,
smooth on both sides (the scant
cobwebby hairs present in the
unfolding leaf-buds very early
deciduous), paler and at ma-
turity perhaps glaucous beneath,
oblong-ovate to obovate-lanceo-
late, apparently 3 to 6 cm. long
at maturity, sparingly glandular-
denticulate, particularly toward
the base, on short woolly peti-
oles. This willow was orig-
inally collected by Richardson,
naturalist of Sir John Franklin's
expeditions, at Fort Franklin
on the Mackenzie River in the
interior of northern British
America. We now have speci-
mens from the vicinity of White

Pass, Alaska, and from Port Clarence, indicating an extensive range in
the interior of Alaska and the interior of northwestern British America.

Specimens have been examined as follows :

Mackenzie. — At Fort Franklin, on Mackenzie River, Richardson.
One of the labels reads, " 7 feet high, erect and spreading."

Yukon, — On the shore of Lake Bennett, J. B. Tarleton, 1S99 (Nos.
12, in part, 14). Marked " 5 to 8 feet high."

Port Clarence. — In a brushy area along a stream in the tundra,
Coville and Kearney (N0.1S74).

Fig. 19. Salix richardso7ii Hook. : a,
pistillate catkin, showing at the base two
persistent stipules of the preceding sea-
son's development, natural size ; b, pis-
tillate flower, enlarged six diameters ; c,
leaf, natural size.

3l6 COVILLE

Seemann in his Botany of the Herald, page 40, reports Sa//x 7-ich-
ardsoni as occurring on the Arctic sea coast of Alaska from Kotzebue
Sound to Cape Lisburne, but his description of the ovaries as "cov-
ered with a slight down " throws doubt on the identification and sug-
gests that they may belong to Salix pulchra. The willow from Kadiak
reported by Kellogg as richardsoni pi'oves to he pulchra.

In Hooker's original description the leaves of Salix richardsoni are
described as " integerrimis." One of the duplicate type specimens
from Fort Franklin, in the Columbia University Herbarium, has some
partially developed leaves and these show a sparse denticulation pre-
ciselv as in the Lake Bennett and Fort Clarence specimens. The more
nearly mature leaves of the Port Clarence plant indicate also that the
base is not always as narrow as is shown in the Hooker plate.

7. SALIX BARCLAYI Anders. Barclay Willow.

Plate XXXVI.

Salix ba'clayi Anders. Oefv. Vet. Akad. Foerh, 15 : 125. 1858.
Salix conjimcta Bebb, Bot. Gaz. 13: iii. 1888.

The commonest bush willow of the Alaskan coast. It is an erect
shrub, ordinarily one to two, sometimes three, meters in height, with
leaves oblong-obovate, acute or somewhat acuminate at the apex,
rounded or wedge-shaped or even cordate at the base, commonly 4 to
7 cm. long, glaucous on the back, smooth or with scant villous or
tomentose pvibescence which is mostly early deciduous, the stipules
ovate or ovate-lanceolate, toothed, persistent, and the ovaries smooth
throughout. The characters given in the key distinguish it from re-
lated Alaskan species. It ranges throughout the coast region of south-
ern Alaska, from Dixon Entrance northward and westward as far as
Unalaska. Southward from Alaska it extends at least as far as the state of
Washington. The original specimens of Salix barclayiwexe collected
at a point near Cape Greville on Kadiak Island, Alaska, by George
Barclay, the botanical collector of the British ship Sulphur, in 1S39.

Alaskan specimens of this species have been examined as follows :
Wra?7gell. — On a hillside back of the town, Coville and Kearney
(No. 444), Fernow.

Juneau. — In and near Silver Bow Basin, Coville and Kearney (No.
565), Grace E. Cooley, 1S91, F. A. Walpole, 1900 (No. 1053).

White Pass and vicinity. — At various points about Skagway and
along the White Pass railroad to Lake Bennett, F. A. Walpole, 1900
(Nos. 1055 to 1057, 1065, 1069, 1261, 1276).

Glacier Bay. — Common on almost all the shores of the bay, Coville

THE WILLOWS OF ALASKA 317

and Kearney (Nos. 623, 641, 652, 660, 700), Trelease and Saunders
^Nos. 3406, in part, 3459 to 3461), Trevor Kincaid, 1S97.

Takutat Bay. — Common all about the bay, Coville and Kearney
(Nos. 1034, 1046, 1060, II 19, 1 139, 1 152), Trelease and Saunders
(Nos. 3415, 3462, 3466 to 346S, 3471, 3473), Frederick Funston,
1S93 (No. 6).

Prifice Willia?n Sound. — In Fort Wells and in Columbia Fiord,
Coville and Kearney (Nos. 1260, 1361).

Cook Inlet. — At Kenai, F. A. Walpole, 1900 (No. 1144), and an
abnormal form at Halibut Cove, Coville and Kearney (No. 2416a).
Also between Cook Inlet and the Tanana River, E. F. Glenn, 1S99.

Kadiak Island. — At various points near Kadiak village, Coville
and Kearney (Nos. 1436, 2329), Trelease and Saunders (Nos. 3476,
3478, 3479), Fernow, Cole, A. Kellogg, 1867 (Nos. 175, in part,
221, in part), Walter H. Evans, 1S97 (Nos. 313, 314), F. A. Wal-
pole, 1900 (Nos. 1 149 to 1 153, 115S, 1 159).

Alaska Peninsula. — At Kukak Bay, Trelease and Saunders (Nos.
34S0, 34S1).

Shumagln Isla?ids. — On Popof and Unga, Coville and Kearney
(No. 1796), Trelease and Saunders (Nos. 3453a, 3455), Kincaid, M.
W. Harrington, 1 87 1-2.

Unalaska. — Near Iliuliuk, Coville and Kearney (No. 1785),
Fernow, C. Hart Merriam, 1891, J. M. Macoun, 1891, 1896, B. W.
Evermann, 1892 (No. 19), Walter H. Evans, 1897 (No. 537).

For remarks on the relationship of this species with Salix cominu-
iata., see the notes under that species.

8. SALIX COMMUTATA Bebb. Greenbacked Willow.
Plate xxxvii.
Salix comnmtata Bebb, Bot. Gaz. 19: no. 1888.

An erect bushy willow commonly .5 to 1.5 meters in height, similar
to Salix barclayi but distinguishable by the characters given in the
key. The most prominent difference lies in the color of the backs of
the mature leaves, which in barclayi are covered with a conspicuous
bloom but in com?nutaia have no bloom, but nearly the same green
color as the upper surface. The differences in the toothing of the
leaf margins of the two species are indicated in the plates. The
upper leaves on the leaf shoots of cojnmutata often present a grayish
appearance, due to a thin but permanent tomentum on both surfaces.
The species extends from the coast region of eastern Alaska southward
through British Columbia to Oregon, in Alaska not extending west of

3l8 COVILLE

Kadiak Island. Mr. Bebb's original specimens were all from the
mountains of Oregon and Washington.

The Alaska specimens examined are as follows :

White Pass.— At Log Cabin, Vista, and White Pass, F. A. Wal-
pole, 1900 (Nos. 1066, 1270, 1279).

Glacier Bay. — On the gravels at Muir Glacier, at Hooniah village,
Coville and Kearney (Nos. 638, 654), and in Berg Inlet, Kincaid.

Yakutat Bay. — At several points in Disenchantment Bay and Rus-
sell Fiord, Coville and Kearney (No. 997), Trelease and Saunders
(Nos. 3463, 3464), and on the west side of Yakutat Bay, Coville and
Kearney (No. mS), Fernow.

Prince William Sound. — Incomplete specimens from Port Wells
are doubtfully referred here, Coville and Kearney (No. 1274), Tre-
lease and Saunders (No. 3474).

Kadiak Island. — Along a small stream near the summit of the
mountain back of Kadiak village, F. A. Walpole, 1900 (No. 11S2).
Specimens of this willow were brought from a mountain back of
English Bay by Miss Harriman, but the specimens unfortunately were
not preserved.

Alaska Penitisula. — At Kukak Bay, Coville and Kearney (Nos.
1618, 1680), Trelease and Saunders (No. 34S2).

The history of this species, briefly stated, is as follows : In the year
1 888 Mr. M. S. Bebb published two closelv related new species of
willows, commutata ^ and conjuncta,'^ which he considered interme-
diate hetw een g-lauca and cor data. Later, in the light of further col-
lections, he recognized and alluded to ^ the close relationship of these
species with the older Salix Barclay i of Andersson and his own Salix
calif ornica.^ Still later he described^ Barclay i as a species exhibit-
ing " a degree of variation remarkable even among willows," and was
inclined to refer to it as synonymous his coffi??iutata and co7zjuncta, an
inclination to which he afterward yielded as evidenced by his identifi-
cations of willows sent to him from the National Herbarium. From
the large series of specimens now available and from field observation
of these willows in the Cascade Mountains and in Alaska, I do not
hesitate to express the opinion that Salix barclayi and Salix commu-
tata are two valid species, and that conjuncta is a synonym of bar-
clayi. No botanist with mature leaf specimens of the two plants be-

^Bot. Gaz. 13: 110. 18SS.

^Idem, III.

'Idem, 16: 106-7. 1891.

^Bebb in Brewer & Wats. Bot. Cal. 2 : 89. 1S80.

sContr. Nat. Herb. 4: 198. 1893.

THE WILLOWS OF ALASKA 3I9

fore him would unite them, but the young leaves of barclayi do not
always show the glaucous character of their lower surface, and even
the characteristic toothing of the margin is then often obscured also.
Salix califor7iica may be distinguished from commjitata by its hairy
capsules.

9. SALIX PULCHRA Cham. Diamondleaf Willow.

Plate XXXVIII.

Salix pule hr a QnkM. Linnaea 6 : 543. 1831.

Salix phylicoides AyiD^KS. Oefv. Vet. Akad. Foerh. 15 : 123. 1858.

Salix fulcrata Anders. Kongl. Vet. Akad. Handl. IV. 6 : 139. 1867.

An erect and bushy, though in arctic or alpine situations often pros-
trate, willow with diamond-shaped leaves bright green and shining
above, glaucous beneath, entirely smooth on both surfaces except when
very young, usually entire though sometimes sparingly crenate-denticu-
late near the middle, the stipules lanceolate, glandular-denticulate, and
persistent. It is a trim, handsome species, well deserving Chamisso's
name pulchra. The original description was based on specimens
collected by the Kotzebue expedition at Cape Espenberg in Kotzebue
Sound and on St. Lawrence Island in Bering Sea. The species ranges
from the coast of eastern Siberia through the islands of Bering Sea to
Point Barrow on the Arctic coast of Alaska, to Kadiak Island on the
south coast, and to the upper Yukon valley in the interior.

The specimens examined are as follows :

Siberia. — On the coast of Bering Sea, between St. Lawrence and
Mechigme bays, James T. White, 1S94.

Arctic Seacoast of Alaska. — At Point Barrow, John Murdoch,
1SS3-3.

Kotzebue Sound. — On the Kowakor Putnam River, G. M. Stoney,
1SS6.

Port Clarence. — On the tundra, often prostrate, sometimes a ineter
and a half in height, reaching at least 5 cm. in diameter of trunk,
Coville and Kearney (No. 1876), Trelease and Saunders (Nos. 3374
to 3376), Brewer and Coe (Nos. 377, 37S), Cole.

St. Lawrence Island. — At Northeast Cape, Coville and Kearney
(No. 3001), Trelease and Saunders (No. 3445).

St. Matthew Island. — On St. Matthew, Coville and Kearney (No.
2088) Brewer and Coe (No. 466).

Pribilof Islajtds. — On St. Paul Island, C. Hart Alerriam, 1891.

Shumagin Islands. — Common at sea-level on both Popof andUnga,
Coville and Kearney (No. 1796a), Trelease and Saunders (Nos. 3453,
3454» 3454a, 345S), C. H. Townsend, 1893.

320 COVILLE

Kadiak Island. — A common species, Coville and Kearney (No.
1434), Trelease and Saunders (Nos. 3366, 3475), Brewer and Coe
(No. 220), Cole, A. Kellogg, 1867 (No. 221, in part), F. A. Wal-
pole, 1900 (No. 1 156).

Alaska Peninsula. — At Kukak Bay, Coville and Kearney (Nos.
1633, 1640). On the Nushagak River, C. L. McKay, iSSi.

Yukon Valley. — On Fortymile Creek, near its junction with the
Yukon, Frederick Funston, 1S93 (^o* 40 )• Junction of Allen and
Koyukuk rivers, F. C. Schrader, 1S99.

Skagivay to Lake Bennett. — Collected at various points, F. A.
Walpole, 1900 (Nos. 1060, 1061, 106S, 1084, 10S5, 1269).

Adelbert von Chamisso, in reporting on the Arctic willows collected
by himself and Eschscholtz on the Kotzebue Expedition, describes a
willow as follows :

Salix pluries nobis obviam venit, an nova species, pjilchra., fruticu-
losa, depressa, foliis breviter petiolatis pollicaribus majoribusque
lanceolatis ellipticisve utrinque acutis integerrimis glabris, subtus glau-
cescentibus, venis pennatis obliquis tenuibus utrinscens [utrinsecus]
5—8 notatis, stipulis lineari-filiformibus serratis, amentis lateralibus
sessilibus villosis, femineis maturis crassis circiter bipollicaribus, cap-
sulis lanatis subsessilibus. Habitus fere Salicis arctlcae Pall. — In Pro-
montorio Espenbergii Americae transbeeringiensis, inque insula Sti.
Laurentii. — An eadem, quam sterilem prope Tigil Kamtschatcae occi-
dentalis legit Erman? cui folia majora, latiora, bipollicaria, subinde
' obovata, venis pluribus primariis utrinsecus S-io-ornata.^

This is an excellent description of a certain willow common in west-
ern Alaska, and the description applies to no other species of that
region. It is therefore assumed, even in the absence of the type speci-
mens, that Chamisso's Salix ptdchra is identifiable. The species to
which the description applies is the Salix phylicoides of Andersson,'
which was afterward divided by its author so as to make an additional
?,\>Q.c\&'a., fulcra t a. ^ Mr. M. S. Bebb critically investigated the two in
1 888 and pointed out their identity, figuring a series of leaves and
fruits.* Chamisso's name appears to have been lost sight of, as it is
not included in the Index Kewensis, nor is it cited as a published name
in any of the works on willows that I have consulted. Andersson al-
luded to the plant in 1858, when he referred it '" dubiously to the S. lap-
p07ium of Linnteus, but likened its leaves to those of the European
species phylicifolia .^ a willow so nearly resembling our Alaskan plant

^LinnaeaC: 543. 1S31.

^Oefv. Vet. Akad. Foerh. 15 : 123. 1S5S.

3Kongl. Vet. Akad. Handl. IV. 6 : 139. 1867.

♦Bot. Gaz. 13: iS6--j.pi.io. iSSS.

sQefv. Vet. Akad. Foerh. 15: 120. 1858.

THE WILLOWS OF ALASKA 321

that Andersson himself recognized that resemblance in his name S.
fhylicoides.

There is some doubt as to the real type locality of Andersson's Salix
phylicoides. He said " Hab. in arctica America occidentali (Awatcha
Bay : Seeman[n])." Awatcha Bay, however, is not in Alaska but in
southern Kamchatka. It is the bay on which the Russian port of
Petropavlovsk is situated. The British ship Herald, on which See-
mann visited the North Pacific, worked also, however, extensively in
Kotzebue Sound and neighboring parts of Arctic Alaska, and it is
probable that the type specimens came from that coast.

lo. SALIX GLAUCA L. Grayleaf Willow.

Plate xxxrx.

Salix glauca L. Sp. PL 2 : 1019. 1753.

Salix seeinxnnii'Rx'Q^^^G, Bull. N. Y. Bot. Gard. 2: 164. 1901.

An erect willow, commonly .5 to 1.5 meters high, with entire leaves
glaucous on the lower surface and permanently clothed beneath, and
the upper ones usually also above, with a rather thin covering of long
appressed hairs, the ovaries white-hair}-, and the catkin-scales light
brown to straw-colored. The plant was originally described from
noith European material but the species has a circumpolar distribution.
In Alaska it ranges from the shores of Bering Sea eastward to the
upper Yukon.

Specimens have been examined as follows :

White Pass. — At Bennett and Vista, F. A. Walpole, 1900 (Nos.
loSi, 1275).

Glacier Bay. — On the gravels at the Muir Glacier, very scarce,
Coville and Kearney (No. 641a).

Kadiak Island. — One of the commonest willows about Kadiak vil-
lage, Coville and Kearney (Nos. 143S, 3248, 2328, 2331), Trelease
and Saunders (Nos. 3362, 3367, 3477), Cole, F. A. Walpole, 1900
(Nos. 1 155, 1 157). At Sturgeon Bay, near Cape Karluk, Trelease
and Saunders (No. 3368), Brewer and Coe (No. 482).

Shujtiagin Islands. — Abundant at Sand Point, Popof Island, and on
Unga Island, Coville and Kearney (No. iSoo), Trelease and Saun-
ders (Nos. 3369 to 3371), Kincaid, M. W. Harrington, 1 87 1-2, C.
H. Townsend, 1893.

Bristol Bay. — Near Nushagak, C. L. McKay, 18S1.

St. Michael. — On the mainland opposite the island, Frederick Fun-
ston, 1894 (No. 236).

Port Clarence. — Along a stream valley in the tundra, Coville and

322

COVILLE

Kearney (No. 1S75), Trelease and Saunders (No. 3373), Brewer and
Coe (No. 3S0).

Kotzebue Sotind. — On Chamisso Island, Berthold Seemann, 184S-
1S50.

In addition to these localities, the plant is reported from the Upper
Yukon valley.

There is a tendency among American willow students to exclude
Salix glaiica from the North American flora, but our Alaskan speci-
mens show so close an agreement with some European material of this
species that I am unwilling to separate them. A Salix seemaiinii has
been described very recently,^ the type specimen collected at Dawson,
Yukon Territory, by R. S. Williams, June 11, 1899. The species is
credited also to Kotzebue and Norton sounds, Alaska, on the basis of
collections made by Berthold Seemann. No comparison of the differ-
ences between glaiica and sceinannii is made by the author, nor am
I able to find in the description a record of any characters that serve to
distinguish the specimens assigned to the latter species from forms of
glauca found in America and Europe.

II. SALIX NIPHOCLADA Rydberg. Mouseleaf Willow.
Salix niphoclada Rydberg, Bull. N. Y. Bot. Gard. i : 272. 1899.

An erect bushy willow, at least half
a meter in height, the twigs and lower
surfaces of the leaves in dried specimens
very gray, almost mouse-colored, from their
glaucousness and appressed hairs, and the
darkening of the tissues in drying. The
leaves are oblong-lanceolate to ovate-
lanceolate, acute at the apex, rounded or
cordate at the base, on petioles 3 mm. or
less in length, entire or with a few ob-
solescent teeth toward the base, commonly
1.5 to 3.5 mm. long, the lower surface
very glaucous and, in all except some of
the lowermost rudimentary leaves, with a
rather sparse, permanent, rather appressed
pubescence of somewhat curly hairs, the
upper surface green and with a sparse (except in the upper leaves of
vigorous twigs) pubescence of longer appressed hairs like those on
the twigs. The slender catkins, about 4 to 6 mm. in diameter, are
> Rydberg, Bull. N. Y. Bot. Gard. 2 : 164. 1901.

Fig. 20. Salix niphoclada
Rydberg : a, pistillate flower-
ing twig, natural size; b, pis-
tillate flower, enlarged six
diameters ; c, mature leaf,
natural size.

THE WILLOWS OF ALASKA 323

borne on branches of the season which are sometimes 2.5 cm. long
and commonly bear 3 to 5 leaves. The catkin scales are pale brown
to straw-colored, the ovaries white-toinentose, and the styles from .5
mm. long to almost nothing. The plant is a rare one in herbaria,
having been collected, so far as known, only near the mouth of the
Mackenzie River in British America (the type locality), on the
Porcupine River in Alaska, and on the northwestern shore of Hud-
son Bay.

The specimens examined are as follows :

Mackenzie River. — At a point 30 miles north of the Arctic Circle
Miss E. Taylor, 1S93.

Porctipi7ze River. — At its junction with the Yukon, growing along
the margins of woods, Frederick Funston, 1894 (No. 185).

Hudson Bay. — Near the mouth of Seal River, 40 miles (64 km.)
northwest of Fort Churchill, E. A. and A. E. Preble, 1900 (No. 26).
Identified by P. A. Rydberg.

Dr. Rydberg based his description on a single fruiting specimen
collected by Miss E. Taylor, as above cited, and deposited in the
herbarium of the Geological and Natural History Survey of Canada.
Mr. Funston's material is in flower and full leaf, and includes both
pistillate and staminate specimens. This additional material necessi-
tates a change in the original description in only minor details, all of
which are included in the general characterization given above. The
staminate catkins in our specimens are about i cm. in length, and ex-
cept the exserted portions of the stamens, about 3 mm. in diameter,
with 2 stamens to each flower, the filaments smooth. The nearest
relative of the species among Alaskan willows is Salix glauca.

12. SALIX ARBUSCULOIDES Anders. Little-tree Willow.

Salix arbusculoides Anders. Kongl. Vet. Akad. Handl. IV. 6: 147. /. 8. f.

81. 1867.
Saiix humillima Anders, in DC. Prod. 16°: 248. 1868.

An erect shrub with reddish stems and narrowly elliptical -lanceolate
leaf-blades acute at both ends, smooth and bright green above, glaucous
beneath, the margins finely and closely denticulate throughout, the
lower surface with a permanent pubescence of straight, closely appressed
rather short hairs. The catkins, which appear with the leaves on short
usually leafy-bracted peduncles, are cylindrical and slender, the ovaries
pubescent. The species was described from specimens collected in
eastern British America at Prince Albert Sound and Rae River ; the
Labrador locality, also assigned the plant in the original publication,
is probably an error. None of the types has been accessible to the

324

COVILLE

writer, and the present reference of our specimens is based on an iden-
tification by Herr Rittmeister von Seemen of the Berlin Herbarium.
Our Yukon Valley specimens are as follows :

F'ortymile Creek. — Near the junction of Fortymile Creek with the
Yukon River, Frederick Funston, 1893 (No. 42a).

Dazvson. — "Island near footbridge," R. '^. Williams, 1899.
Fiftyinile River. — Collected by Arthur L. Bolton, 1S99.
Salix arbusctdoides most resembles, among Alaskan willows, Salix
sitchensis., but the toothing and less distinctly oblanceolate character
of the leaves, the lack of the permanent satiny
lustre of the lower leaf surface, and the presence of
two stamens instead of a single stamen in the male
flowers, distinguish arbusculoides. Salix sericea
Marsh., of the eastern United States, is another
species resembling arbiiscidoides but its leaves
are slightly hairy on the upper surface, at least on
the midrib, and the apex is acuminate, while the
catkins are more densely flowered than those of
arbusculoides., with the scales usually black, and
rather densely villous. In arbuscidoides the leaves
are entirely smooth above, even to the midrib, and
the apex is merely acute, the point sometimes
blunt ; and the catkins are more loosely flowered,
he scales light to dark brown, scarcely black,
sparingly villous. The willow described by von
Seemen in 1S95 as Salix saskatchavana^ see.\ViS
to be closely related to Salix arbusctdoides but
the statement regarding the leaves, " untere Seite
in der Jugend fein seidig behaart, im ausgewach-
senen Zustande bcide Seiten kaJd" furnishes the
presumption of a specific distinction.

In addition to the flowering specimens collected
by Mr. Funston mature leaves and fruit were
collected at Dawson, Yukon Territory, June 19, 1899, by Mr. R. S.
Williams, and similar mature specimens at Fiftymile River, Yukon
Territory, in the same year, by Mr. Arthur L. Bolton. From the upper
Yukon the species extends eastward nearly or quite across the northern
part of the continent. None of these collectors described the habit of
the plant, but it certainly is not a prostrate species, and doubtless is
not a tree.

Fig. 21.
arbusculoides
ders. : a, pistillate
flowering twig,
natural size ; b,
pistillate flower, en-
1 a r g e d six dia-
meters; c, scale,
same enlargement;
d, mature leaf,
natural size.

'Seemen, Engl. Bot. Jahrb. Beibl. 52 : 7. 1S95.

THE WILLOWS OF ALASKA 325

13. SALIX MYRTILLIFOLIA Anders. Myrtle Willow.

Sali.x inyrtillifoUa Anders. Oefv. Kongl. Vet. Akad. Foerh. 15: 132. 185S.
Also Anders. Proc. Am. Acad. 4 : 74. 1858.

A small, trailing, rather than prostrate, willow, the branches often
standing erect for a decimeter or two above the ground. The leaf-
blades in our specimens are oblong, i to 3 cm. in length, broadly acute
or obtuse at base and apex, entirely smooth, evenly crenate about the
whole margin, on smooth petioles about i .5 to 3 mm. in length. The
catkins appear with the leaves from lateral buds on old wood, the
short pubescent peduncles, i to 3 mm. in length, bearing usually
3 to 3 small leaves. The pistillate catkins are i to 2 cm. in length,
and about 4 to 5 mm. in diameter, the
scales short, black-tipped, and sparingly
hairy. The ovaries are smooth, lance-
olate, tajDering into a short style .5 mm.
or less in length, the pedicel about twice
or three times the length of the nectary
and only a little shorter than the catkin
scales. The species is known only from YiG.22. SalixmyrtilUfoUa

the interior of British America, the type Anders. : «, pistillate fiower-
T . 1 J ti r> 1 AT insT twisf, natural size ; ^, pis-

specimens benig marked " Kocky Moun- .,f „ , , . ,.

■^ ^ •' tillate flower, enlarged six di-

tains, east side, low situations." It has ameters ; c, leaf , natural size,
been found so close, however, to the

Alaskan boundary, namely on the shores of Lake Bennett, Yukon,
J. B. Tarleton, 1899 (Nos. 4a, 4b), and below White Horse Rapids,
R. S. Williams, 1S99, that its occurrence in Alaska may be expected.

From lack of specimens it has been impossible to give this species
the critical study it needs, particularly in the direction of establishing
its relationship to Salix pse?idomyrsi?zites and Salix curtijlora.
These were published as species by Andersson in the same paper in which
myrtillifolia was published, but he afterward brought all three together
as varieties of his Salix novaeangliae^ a treatment which has not been
followed by recent American students of Salices.

14. SALIX CHAMISSONIS Anders. Chamisso Willow.
Salix c/iamissonis Anders, in DC. Prod. 16': 290 1868.

A prostrate willow with obovate smooth leaves minutely and closely
glandular-serrate about the entire margin. It is the only normally pros-
trate willow of Alaska with glandular-serrate stipules. Its creeping
branches and bright green leaves form rather dense patches in moist

326

COVILLE

places in the tundra. The plant is confined to the trvily Arctic zone
on the islands and shores of the northern part of Bering Sea, the orig-
inal specimens having been collected by Chamisso and Eschscholtz,
botanists of the Kotzebue Expedition, in 1816, at St. Lawrence Bay,
on the Siberian side of Bering Strait, about latitude
65° 40'.

Specimens have been examined as follows :
jPor^ Clarence. — Abundant on the tundra of the
mainland at the landing place of the expedition,
Coville and Kearney (Nos. 1877, 1878, 1879), Tre-
lease and Saunders (No. 3472), Cole.

St. Lawrence Island. — Found sparingly on the
tundra at Northeast Cape, Coville and Kearney (No.
2003). The plant was also collected on this island
by Chamisso and Eschscholtz, according to Ledebour,
and by Kjellmann, of the Vega Expedition.

Siberian Coast. — Collected at Plover Bay by W. H.
Dall, of the Western Union Telegraph Expedition, in
1865-6.
This species is easily distinguishable from the other prostrate willows
of Alaska by the form and serration of its leaf -blades. These are
commonly 1.5 to 3 cm. in length, narrowly obovate to nearly or-
bicular, apiculate, broadly acute or rounded at the apex, and either
glaucous or shining beneath. The catkins, which stand erect on leafy
peduncles, are a centimeter or a little more in diameter and commonly
3 to 6 times as long. In most of the specimens the capsules are dis-
tinctly but not densely pubescent, but in some they are smooth. The
stvle is about a millimeter in length.

Fig. 23. Sa-
lix chamissonis
Anders., a
mature leaf,
natural size.

15. SALIX ARCTIC A Pallas. Arctic Willow.

Plate XL.

Salix arctica Pall. F1. Ross, i" : 86. 1788.

Salix diplodictya Trautv. Nouv. Mem. Soc. Nat. Mosc. 2 : 307. t. 14.

1832.
Salix crassijulis Trev. ; Trautv. Nouv. Mem. Soc. Nat. Mosc. 2: 308. /.

15. 1832.
Salix pallasii A^T>%^s. in DC. Prod. i6' : 285. 1868.

The commonest and the largest of the prostrate willows of Alaska.
Its short half buried trunk is sometimes 2 to 3 cm. in diameter, and sends
out stout horizontal creeping branches sometimes a centimeter thick at
the base. Its leaf blades are normally about 2.^ to 4 cm. long, obovate,
obtuse at the apex, tapering at the base, glaucous beneath, entire, or

THE WILLOWS OF ALASKA 327

very sparingly denticulate toward the base, smooth on both sides at
maturity, but usually with some long soft appressed hairs on the mar-
gins and lower surface when young. Sometimes, however, the leaf-
blades are almost orbicular with a scarcely tapering base, sometimes
the apex is acute, sometimes retuse, and in robust specimens they may
reach a length of 7.5cm., often then developing ovate-lanceolate entire
stipules. It is the only prostrate willow of Alaska, except chamissonis^
that produces stipules, and in that species the stipvdes are glandular-
serrate. Occasionally the upper surface of the leaves bears some cob-
webby hairs, even till maturity. A conspicuous feature of the plant
when growing is the long, erect, stout, hairy catkins standing above
the leaves, which look like little cat tails. The catkin scales are
usually black, obtuse, and long-hairy, the ovaries densely white-woolly,
becoming more thinly tomentose as thev expand and mature, and the
styles are commonly i to 3 mm. in length. The plant is an arctic-
alpine species occurring near timber line, or in cold situations at
lower elevations, along the southern coast of Alaska from Juneau west-
ward to the Siberian coast, and northward to the coast of the Arctic
Ocean. The original specimens of arctica were collected by Sujef
in Siberia along the Gulf of Obi, of diplodictya on St. Lawrence
Island, Bering Sea, and of crassijulis in St. Lawrence Bay, Siberia.

A large number of specimens have been examined, as follows :
Jtmeau. — Above Silver Bow Basin, Grace E. Cooley, 1891.

White Pass. — Near the summit, altitude about 3000 feet, Coville
and Kearney (No. 518), Trelease and Saunders (No. 3399), R. S
Williams, 1S99, F. A. Walpole, 1900 (Nos. 1104, 1109, 1348, 1378)

Glacier Bay. — At various points, Coville and Kearney (Nos. 631
in part, 635, 655), Trelease and Saunders (Nos. 3400, 3401, 3401a
3403 to 3405, 3407 to 3409, 3433, in part), Brewer and Coe (Nos
37, 41), Kincaid, Walter H. Evans, 1897 (No. 157). Most of these
specimens were collected on the cold glacial gravels near sea level, the
timber line belt being deeply covered with snow at the time of our
visit.

Takutat Bay. — At various localities, Coville and Kearney (Nos.
looi to 1003, 1013, 1040, 1076, 10S3, 1099, 1 140), Trelease and
Saunders (Nos. 3410, 341 1, 3414, 3441? 3472), Frederick Funston,
1893 (No. 117). Most of these specimens were collected on glacial
gravels near sea level.

Prince Willia?n Sound. — At several localities, Coville and Kearney
(Nos. 1195, 1371, 1363), Trelease and Saunders (Nos. 3416 to 3418).
Some of these grew at sea level, others at 3500 feet elevation.

328 COVILLE

Kadiak Island. — At various points, Coville and Kearney (No.
1473), Trelease and Saunders (Nos. 3419 to 3421), Brewer and Coe
(Nos. 206, 232), Cole, Walter H. Evans, 1S97 (Nos. 437 to 439), F.
A. Walpole, 1900 (Nos. 116S, 1169).

SJmmagin Islands. — At several localities on Popof and Unga, Tre-
lease and Saunders (Nos. 3423. 3424, 3456, 3457), Brewer and Coe
(No. 2S0), Kincaid, M. W. Harrington, 1S71-2. On Nagai Island,
C. II. Townsend, 1893.

Alaska Peninsula. — At Kukak Bay, Coville and Kearney (No.
1670), Trelease and Saunders (No. 3422), Kincaid. At Stepovak
Bav, Palache. At Bailey Harbor, Belkofski Bav, C. H. Townsend,
1S93. At Nushagak, without collector's name, 1SS2.

Aleutian Islands. — On Akun Island, C. H. Townsend, 1S93. On
Unalaska, Coville and Kearney (Nos. 17863, 2200), Trelease and
Saunders (No. 3426), Brewer and Coe (No. 296), Cole, H. G. von
Langsdorff, 1805-6, A. Kellogg, 1S67 (No. 275), M. W. Harrington,
1S71-2, L. M. Turner, 1880, C. Hart Merriam, 1891, J. M. Macoun,
1S91, B. W. Evermann, 1S92, Walter H. Evans, 1897 (No. 537, in
part). On Attn Island, J. M. JMacoun, 1891, W. V. E. Jacobs, 1894.

Pribilof Islands. — On St. Paul, Robert White. 1879, William
Palmer, 1S90 (Nos. 310, in part, 425, 724), C. Hart Merriam, 1891,
J. M. Macoun, 1892, 1897, C. H. Townsend, 1893, Trevor Kincaid,
1897.

St. Matthc-ju Islands. — On St. Matthew, Coville and Kearney
(No. 2086), Trelease and Saunders (Nos. 3394, 3430, 3431, 3447)-
On Hall Island, Coville and Kearney (No. 20S2), Trelease and
Saunders (Nos. 3428, 3429, 3446), Brewer and Coe (No. 410), J.
M. Macoun, 1891.

Arctic Seacoast. — Collected at Point Hope by some person on the
Revenue Cutter Corwin in 1883.

Siberia. — At Plover Bay, Coville and Kearney (No. 1865), Cole,
W. H. Ball, 1865-6. On Bering Island, L. Stejneger, 1S82-3.

Salix arctica is exceedinglv variable in its leaf characters, and this,
together with certain nomenclatorial accidents, has caused the species
to be known under several different names. Mr. M. S. Bebb pub-
lished a critical statement on the subject in 1889,^ based in part on
correspondence with Trautvetter, which I have here followed. The
large number of specimens examined tends to confirm the idea that
the extreme variation in the leaves is chieflv an individual characteristic
and does not mark recognizable incipient species. The nearest ap-

^ Bot. Gaz. 14: 1 15-17. 1SS9.

THE WILLOWS OF Ai;^ASKA

329

proach I have found to a subspecific differentiation is in some of the
specimens from the Pribilof and St. Matthew Islands in Bering Sea,
and the Shumagin Islands. In these specimens the leaves are orbicular
or nearly so and only about 2 to 3 cm. in diameter, w^hile the catkins
are shorter than usual, about 1.5 to 3.5 cm. in length. Occasionally
specimens are found which lack the glaucousness of the lower leaf
surface, a character on which Trautvetter based chiefly his separation
of diplodictya.

16. SALIX GLACIALIS Anders. Icy Willow.
Salix glacialis h.-^M^'^s. Oefv. Vet. Akad. Foerh. 15: 131. 1858.

A prostrate willow with smooth stems and buds,
small oval obtuse leaves 12 mm. or less in length,
smooth, or when young slightly appressed-villous
on the lower surface, glaucous on the back, entire
or with a few teeth toward the base, the catkins
ovoid to globose, several-flowered, the scales black
and nearly smooth on the outside, the ovaries
densely white hairy, and the style wanting. Its
foliage resembles that of ovalifolia^ though
smaller, but in the pistillate flowers it is very
different from any species of the ovalifolia group.

The species is known only from the type
specimen collected by Lieutenant W. J. S. Pullen,
in 1S49, on the Arctic seacoast between Point
Barrow and the Mackenzie River, and from speci-
mens collected at Point Barrow by John Murdoch, in 1SS2-3
nate specimens and mature leaves are unknown.

Fig. 24. Salix
glacialts Anders.: a,
pistillate flowering
twig, natural size: b,
pistillate flower, en-
larged six diameters.

Stami-

17. SALIX FUSCESCENS Anders. Bog Willow.
Sa/zx /Ksc^scens Anders. Kongl. Vet. Akad. Handl. IV. 6: 97. 1867.

A prostrate or trailing willow, with small smooth leaves glossy
above, glaucous beneath, of a rhombic-obovate form, the broadest
part distinctly above the middle, and usually with a few teeth toward
the narrowxd base, the ovaries smooth, styles less than half a milli-
meter in length, and the pedicels two or more times the length of the
nectaries. It is a species of western Alaska, occurring from the peat
bogs of Kadiak Island and Cook Inlet westward to the tundra of the
shores and islands of Bering Sea and the Arctic Ocean, and is found
also in eastern Siberia.

Proc. Wash. Acad. Sci., August, 1901.

330

COVILLE

Specimens have been examined as follows :

Cook Inlet. — In bogs at Kasilof, Walter H. Evans, 1S98 (Nos.
682, 726).

Kadiak Islafzd. — In a bog at the head of a lake about a mile and
a half from the village of Kadiak, F. A. Walpole, 1900.

Alaska Peninsula. — At Kukak Bay, Coville and Kearney (No.
1660).

Norton Sound. — At Unalaklik, Frederick Funston, 1S94 (No.
232).

St. Lawrence Island. — On the marshy tundra at Northeast Cape,
Coville and Kearney (No. 2002), Trelease and Saunders (Nos. 3443,
in part, 3444), Cole.

St. Matthew Island. — On the tundra, Coville and Kearney (No.
2087a).

Siberiafi Coast. — At Plover Bay, W. H. Dall, 1S65-6, Coville and
Kearney (No. 1S56), Cole.

In his original description Andersson gave the range of this species
as "in Kamtschatka (Beechy et Mertens), ad Alaxa (Kostalski), ad
ostium fl. Uda, ad sinus Manga, Nichta, Ujakon, et in insula Schan-

tar (Middend.)," and he found it also
among the Kotzebue plants from Esch-
scholtz Bay and Shishmaref Bay or Inlet,
in Kotzebue Sound. It has been impos-
sible to consult any of these specimens,
and our identifications of the species rests
on descriptions alone. In Hooker and
Arnott's Botany of Captain Beechey's
Voyage and various subsequent publi-
cations the species was misidentified as 6*.
rhamnifolia Pallas. Among the Alaskan
species fuscescens most nearly resembles
ovalifolia^ but may readily be distin-
guished by the characters given in the
key. Its nearest relative among North American vv'illows is Salix
7nyrtilloides L., a circumpolar species occurring on our continent
chiefly on the eastern side, in bogs, and apparently not extending
northwestward to Alaska. The mxa^ fuscescens was given to the plant
to indicate the tendency of the young leaves to turn blackish in drj-ing,
a phenomenon characteristic, however, of several other species of
willows.

Fig. 25. Salix fuscescens
Anders. : a, pistillate flower-
ing twig, natural size; b,
pistillate flower, enlarged six
diameters ; c, mature leaf,
natural size.

THE WILLOWS OF ALASKA

331

iS. SALIX OVALIFOLIA Trautv. Oval Willow.

Salix ovalifolia'TKXVTY. Nouv. Mem, Soc. Nat. Mosc. 2: 306. /. /J. 1832.
Salix tinalaschcensis Cham.; Anders. Oefv. Kongl. Vet. Akad. Foerh. 15 :

130. 1858.
Salix rotimdata Rydberg ; Jordan, Fur Seal Islands 3: 571. 1899, not

Salix rotundata Forbes 1829.
Saiix cyclophylla Rydberg, Bull. N. Y. Bot. Gard. i : 274. 1899.

A common and characteristic prostrate willow of western Alaska
and eastern Siberia, extending in Alaska along the coast from the
Shumagin Islands westward to the Aleutian Islands, northward through
the islands and shores of Bering Sea,
and northeastward along the Arctic
Ocean to Point Barrow. The type
specimen of ovalifolia came from Cape
Espenberg, Kotzebue Sound, of tin-
alaschcensis from Unalaska, of rotun-
data and cyclophylla from St. Paul
Island, of the Pribilof group. It is
distinguishable from the other prostrate
willows of Alaska by its oval to orbi-
cular, nearly or quite entire leaves
usually 10 to 20, or sometimes even
25, mm. in length, glaucous beneath,
and at maturity thick, tough, and promi-
nently marked beneath by a net-work
of fine veins, its slender shoots of
the season running over the surface
of the ground, its smooth and glau-
cous capsules on stalks little or not
at all exceeding the nectaries, and its styles about .5 mm. in length.
It forms large carpet-like patches in the grass, extending to sea level,
apparently, in the southernmost parts of its range and reaching there-
fore below the limits of most strictly arctic plants.

Specimens have been examined as follows :

Takutat Bay. — On the glacial gravels along the west shore of the
bay, only a single patch seen, Coville and Kearney (No. 1141).
Although this locality is far to the east of the main range of the species,
the specimens have the characteristic long creeping shoots and glaucous
capsules. The fact that only a single patch was observed, close to the
bank of a stream, suggests that a stray seed may have been brought to
the spot by the wind or some water bird. It is quite possible, too,
that Salix ovalifolia occurs at several points east of Kadiak, for the

Fig. 26. Salix ovalifolia
Trautv. : a, pistillate flowering
twig, natural size ; b, pistillate
flower, enlarged six diameters ; c,
d, mature leaves, natural size.

332 COVILLE

limited character of the collections thus far made in Cook Inlet and
Prince William Sound do not preclude the possibility of its occurrence
in those places.

Kadiak Island. — Mountain back of Kadiak village, F. A. Wal-
pole, 1900 (No. 1237).

Alaska Peninsula. — Collected in Stepovak Bay, where it was
rather abundant, Palache.

Shumagin Islands. — At Sand Point, Popof Island, Coville and
Kearney (No. 1799), and elsewhere on the same island, Trelease and
Saunders (No. 3425).

Pribilof Islands. — Abundant and forming large mats near the vil-
lage of St. Paul, St. Paul Island, Coville and Kearney (No. 1S37),
Trelease and Saunders (No. 3442), Kincaid, Cole, Brewer and Coe
(Nos. 335, 329), Mrs. Bryant, 1875, William Palmer, 1890 (Nos. 353,
424, 556), C. H. Townsend, 1893, J. M. Macoun, 1S97.

St. Matthexu Islands. — On St. Matthew Island, Coville and Kear-
ney (No. 20S7), Trelease and Saunders (No. 3448), Brewer and Coe
(No. 469). On Hall Island, Coville and Kearney (Nos. 2084, 2085).

Si. lazvrence Island. — At Northeast Cape, Trelease and Saunders
(No. 3443), Cole.

Point Barrozv. — Collected by John Murdoch, 18S2-3 (No. 65).

Siberian Coast. — At Kayne, or Arakamtchechene, Island, collected
on the Ringgold and Rodgers U. S. North Pacific Exploring Expedi-
tion, C. Wright, 1853-6.

Various other collections of the plant have been made about Bering
Sea and north of Bering Strait along the Arctic coast of Alaska, the
easternmost record, between Point Barrow and the Mackenzie River,
resting on a specimen collected by Lieutenant W. J. S. Pullen, in 1849.

The name Salix unalaschcensis which is often cited as published
by Chamisso in 1831,^ can not properly, it seems to the writer, be
cited as of that date, the word unalaschcensis being merely the first
word in a Latin note on a willow to which Chamisso did not give a
name. In another precisely similar case he began "Salix ad portum
Sancti Francisci." It evidently did not occur to Ledebour that the
word had any nomenclatorial intent or value for he did not cite it in
his Flora Rossica. The first use of Salix unalaschcensis in such a
way as to give it standing in nomenclature appears to be Andersson's
citation of it in 1858, as given above. He at that time referred it
doubtfully to the plant we now know as Salix fuscescefts Anders.
Later, in 1868,^ he referred it to Salix ovalifolia., the species, and the

iCham. Linnaea 6: 541. 1831. ^Anders, in DC. Prod. 16' : 291. 1S6S.

THE WILLOWS OF ALASKA 333

only species, with which Chamisso's description agrees. Dr. Ryd-
berg's recent application of Chamisso's name unalaschcensis to a
prostrate willow of the Juneau district of eastern Alaska is discussed
under Salix stolonifera.

Dr. Rydberg's Salix rotundata^ a homonym for which he after-
ward substituted the name Salix cyclophylla^ as cited above, I am un-
able to distinguish from ovalifolia except as a common, and perhaps
indeed the prevalent, form with orbicular instead of oval leaves. Both
forms of leaves occvu- sometimes on the same plant, and all the other
characters that distinguish the oval-leaved ovalifolia from other wil-
lows are found also in the orbicular-leaved plant.

19. SALIX STOLONIFERA Coville. Sprouting Willow;
Plate XLi, Figure i.

Salix stolonifera sp. no v.

Plant prostrate ; branches of the season smooth, i to 2 mm. in
diameter, the underground portions of old stems frequently develop-
ing slender subterranean ascending branches or stolons, usually less
than a millimeter in diameter; leaf blades commonly 1.5 to 3 cm.
long, oval to obovate, rounded or sometimes broadly acute at the apex,
either wedge-shaped or rounded at the base, sparingly provided when
young with a few early deciduous long straight hairs on the margins
and back, entire or with a few teeth near the petiole, glaucous beneath
and the reticulations not very conspicuous ; catkins terminating leafy
branches of the season, oblong to cylindrical ; scales black, obovate,
rounded at the apex, rather sparingly provided on both surfaces with
long straight white hairs inclined to be deciduous ; stamens two to
each scale, the filaments smooth ; ovary smooth or with some traces
of pubescence toward the apex, lanceolate, its stalk shorter than the
nectary, the style 1.5 to 3 mm. long, each of the stigmas bifid into two
linear segments ; capsule not glaucous.

Type specimens in the United States National Herbarium, collected
June 10, 1900, at the railway station of Glacier, between Skagway
and White Pass, Alaska, by F. A. Walpole (No. 1075).

The nearest relative of Salix stolonifera in Alaska is Salix ovali-
folia. The characteristic of the production of slender, leafless, sub-
terranean branches or stolons, which the specimens of stolonifera
usually show, I have never detected in ovalifolia. The length of the
styles is perhaps the easiest diagnostic character for one not familiar
with the two species. The leaves of stolonifera while similar in form

334

COVILLE

and size to tnose of ovalifolia have the reticulations of the veins much
less conspicuous on the back at maturity, and the capsules of stoloni-
fera are not glaucous. Salix ovalifolia is a characteristic species of
the Bering Sea and Alaska Peninsula region, though found sparingly
to the eastward as far as Yakutat Bay, while stolonifera is a species
of eastern Alaska, in the glacier region from Yakutat Bay to Glacier
Bay and Lynn Canal.

Stolonifera is evidently the species to which Dr. Rydberg has re-
cently applied, mistakenly as I believe, the name itnalaschceizsis of
Chamisso.^ From the original description of Chamisso ^ the following
characters of zoialaschcensis are abstracted : plant prostrate, with
branches a foot long ; leaves obovate, subglaucous beneath, entire
(this last character drawn by inference fi'om Chamisso's comparison
of his plant with Salix arctica Pall.) ; ovaries smooth. This would
identify the plant with ovalifolia^ or ftcscescens^ or stolonifera. The
additional character "stylo elongato" excludes _/«5ce5ce/?5, and it is
doubtless this same phrase that led Dr. Rydberg to identify Chamisso's
plant with the very long-styled stolonifera. But as stolonifera is not
known within more than a thousand miles of Unalaska, as the long
creeping branches are characteristic of ovalifolia., while they are not
characteristic of stolotiifera., and as "elongated" is a relative term,
it is reasonable to conclude that Chamisso had in hand not stolonifera
but ovalifolia. For further remarks on the name Salix unalasck-
censis^ see the account of Salix ovalifolia.

Specimens of stolonifera have been examined as follows :
Jiineaic. — From the mountains above Silver Bow Basin, Grace E.
Cooley, 1S91. From the mountains east of Juneau, at an altitude of
3500 to 3000 feet, F. A. Walpole, 1900 (Xos. 1232, 1242).

White Pass. — From the summit of White Pass, about 3000 feet
altitude, R. S. Williams, 1S99, F. A. Walpole, 1900 (No. 1249),
and at the railway station of Glacier, between Skagway and White
Pass (No. 1075).

Glacier Bay. — On the gravel deposits near the foot of Muir Glacier,
Coville and Kearney (Nos. 621a, 627), Trelease and Saunders (Nos.
3433' in part, 3434, 3435, in part, 3437, 3439, 3449 to 3452), Brewer
and Coe (No. 51), Weaker H. Evans, 1S97 (No. 15S).

Yakutat Bay. — Near Hidden Glacier, Trelease and Saunders (Nos.
3412, 3413). At Hubbard Glacier, Coville and Kearney (No. 1079).

1 Rydberg, Bull. N. Y. Bot. Card, i : 275. 1S99.
^Cham. Linnaea 6 : 541. 1S31.

THE WILLOWS OF ALASKA 335

20. SALIX POLARIS Wahl. Polar Willow.

Sa/Lv pohiris VJ AHi.. Fl. Lap. 261. t. ij. 181 2.

A very small prostrate willow with stems imbedded in the ground,
oval to orbicular-obovate, entire or nearly entire, smooth, deciduous
leaves bright green on both surfaces or somewhat glaucous beneath,
commonly i to 1.5 cm. long, the oblong
to globose pistillate catkins few-flowered,
with pubescent or partly smooth ovaries
and capsules. It resembles in size and
general appearance both leiocarpa and
■phlcbophylla^ but it has shorter catkins
than the latter and lacks its persistent
skeletonizing leaves. These organs too
are less tapering at the base in polar is Fig. 27. Salixpolaris\V^\i\.:

J ,1 1 n 1 • 1 J- . 1 «) pistillate catkin, natural

and seem to be wholly devoid of the . ^ , . .„

•' size ; 0, pistillate flower, en-

long hairs often present on the margins 1^,^^^ six diameters ; c, leafy
in pJilebophylla. In its slender naked branch, natural size,
stoloniform stems, imbedded in the

ground and rooting, it resembles leiocarpa^ but that species has
smooth capsules and its leaves are smaller and more uniformly oi'bic-
ular, never glaucous on the back, and inclined to be persistent. In
the Alaskan region the species, so far as known, is confined to the
Arctic zone, about the shores of the northern part of Bering Sea. It
was described originally from Lapland and afterward found in northern
Asia.

The following specimens, in addition to European ones, have been
examined :

Port Clarence. — On the tvuidra, Trelease and Saunders (No. 33S7,
3385, in part). Reported by Lundstrom also as collected here by the
Vega Expedition.

Siberia7i Coast. — On Kayne, or Arakamtchechene, Island, collected
on the Ringgold and Rodgers U. S. Exploring Expedition, C. Wright,
1853-6.

Cape Vajtconver. — Collected by J. M. Macoun, 1891.

Our specimens of this plant are few and not sufficiently ample to
furnish a comprehensive idea of the form of the leaves and the char-
acter of the fruit in Alaskan polar is. This species was reported from
Alaska by several of the early expeditions but it is probable that most
if not 'all these early collections were either leiocarpa or pJilebophylla.
Mr. M. S. Bebb in preparing the list of North American willows for

336

COVILLE

Patterson's Checklist in 1892 omitted Salix polaris. Our material,
however, though fragmentary and unsatisfactory, makes it probable
that the species should again be considered a member of the North
American flora. Dr. Rydberg has already taken this view by identi-
fying Mr. Macoun's Cape Vancouver specimen as Salix polaris.

21. SALIX PHLEBOPHYLLA Anders

Skeleton Willow.

^01. /. 10.

Salix buxifolia Trev. ; Trautv. Nouv. Mem. Soc. Nat. Mosc. 2

1832, not ScHLEiCH. 1815.
Salix arctica mitior Ledeb. F1. Ross. 3: 619. 1849-51, not Salix arbitscula

minor Laest. 1845.
Salix p/ileiop/iflla ADDERS. Oefv. Kongl. Vet. Akad. Foerh. 15 : 131. 1858.
Salix palaeonejira Rydberg, Bull. N. Y. Bot. Gard. i : 267. 1899.

A handsome little willow growing in mats on the ground, well dis-
tinguished from other Alaskan willows by the characters given in the
key. As in leiocarpa, the bright green leaves are covered on both

surfaces with minute light-colored dots,
visible under a lens, each of which marks
the position of a breathing j^ore. The
ovaries and capsules are usually pubescent,
though sometimes specimens are found in
which nearly the whole surface is entirely
smooth. The venation of the leaves is strik-
ing. On each side of the midrib are from
3 to 6 conspicuous nerves nearly parallel
with the margin but running into it toward
catkin, natural size; *, pis- the apex. These principal nerves are con-
tillate flower, enlarged six ^ected with each other by several cross nerves,
diameters; c, leaf, natural \ c^ .^ • 11 1. 1

, , , , r , , , After the growmg season the leaves die but
size ; a, old leaf, skeleton- .

ized, natural size. remain attached to the stems, the tissue be-

tween the nerves of the leaf becoming trans-
lucent, and by the third season waste away, leaving the skeletonized
framework to clothe and protect the stems for several years. In
leiocarpa the nerves are equally conspicuous but the leaves are not
very persistent after the second season and they do not so regularly
become skeletonized. The stems of pJilebophylla are stouter than
those of leiocarpa., are less completely imbedded in the soil, and do not
so often form stolons. The plant grows on the tundra, and occurs
along probably the whole Arctic coast of Alaska from the British
boundary to Kotzebue and Norton sounds.

Specimens examined.

Siberian coast. — On Kayne, or Arakamtchechene, Island, collected

Fig. 28. Salix phlebo-
phylla Anders. : a, fruiting

THE WILLOWS OF ALASKA *337

by the Ringgold and Rodgers U. S. North Pacific Exploring Expedi-
tion, C. Wright, 1853-6.

Port Clarence. — On the tundra of the mainland at sea level, Coville
and Kearney (No. 1S73), Trelease and Saunders (Nos. 3385, in part,
3386, 33S8, 33S9), Brewer and Coe (No. 379).

Point Barro-iV. — At sea level, John Murdoch, 1S82-3 (Nos. 6, 97).

Porcupine River. — ^Without special locality, J. H. Turner, 1891.

Under the names Salix retusa L. and Salix angloriim Cham, this
willow has been reported from St. Lawrence Island in Bering Sea,
St. Lawrence Bay on the Siberian side of Bering Strait, from Chamisso
Island and other points in Kotzebue Sound, and from Felly Isle at the
mouth of the Mackenzie.

S. fhlebophylla was published by Andersson in 1S58 through the
citation of an earlier published description, namely, that given under
the name Salix retusa by Hooker, Fl. Bor. Am. 2 : 153. 183S-9.
Hooker cited two specimens, one collected in Kotzebue Sound,
Alaska, by the botanists of Captain Beechey's expedition, the other
on the Arctic seashore of British America by Richardson. A mis-
understanding of the species phlebophylla on the part of some
authors, has been occasioned by the fact that Andersson referred to
phlebophylla other specimens which did not in reality belong to the
plant described by Hooker as " a very distinct and beautiful species,
with glossy, strongly nerved, perennial leaves, the skeleton nerves re-
maining in the lower parts of the stem." Andersson further compli-
cated the matter when he separated his complex phlebophylla into
three forms, by describing each of them as having " capsulae glaber-
rimae." Both these mistakes, however, were corrected by Andersson
ten years later, when he described phlebophylla as having " capsulis
. . . tenuiter puberulis " and " foliis coriaceis nitidis sub lente punc-
ticulatis . . . utrinque nervoso-costatis, tertio anno in reticulum can-
cellatum ob parenchyma evanescens solutis." ^ By these means An-
dersson clearlv restricted the use of the name phlebophylla to the
plant to which I have here applied it. Dr. F. A. Rydberg has re-
cently ^ referred to phlebophylla several specimens of willow with the
lower surface of the leaves glaucous. These do not belong to the
skeleton-leaved species with which we are now dealing, and this latter
plant being therefore apparently left without a name Dr. Rydberg de-
scribed it as Salix palaeoneura^ basing his description on specimens

1 Anders, in DC. Prod. 16^: 290-91. 1868.

2 Rydberg, Bull. N. Y. Bot. Gard. i : 275. 1S99.

3 Rydberg, Bull. N. Y. Bot. Gard. 1 : 267. 1S99.

338 COVILLE

collected at Point Barrow by Murdoch. This disposal of the name
pklebophylla and the consequent redescription of the skeleton-leaved
plant under the name palaeo7teura is clearly inadmissible for the rea-
sons given above, and furthermore because the specimen in the Co-
lumbia University Herbarium marked " No. 96 He;-b. H[ooker],"
which Dr. Rydberg has cited as the type of pklebophylla^ although a
mere fragment consisting of a fruiting catkin and a single leaf, is
identical with our skeleton-leaved plant.

It is of interest to note that Trautvetter in his original description of
this plant under the name buxifolia^ the type of which came from St.
Lawrence Island, Bering Sea, calls attention particularly to the skele-
tonizing of the old leaves: "Folia . . . anno primo baud articulis
solvuntur, sed marcescunt, fusca evadunt, quo statu per secundum an-
num plantae adhaerent, tertio vero foliorum parenchyma evanescit re-
teque venosum cancellatum foliorum plures adhuc annos persistit . . .
Species foliis edurantibus notabilis." ^

The name Salix a?iglorum C\\am.'^ has sometimes been used for this
species because the specimens of Chamisso and Eschscholtz identified
as Salix anglonwi by Chamisso are identical with Andersson's phlcb-
ophylla. Dr. Rydberg has pointed out,^ however, that Chamisso did
not describe the species but cited Robert Brown's Salix arctica^ a
species of eastern Arctic America to which the name anglorum must
therefore attach.

A plant which has sometimes been confused with pklebophylla^ but
which is readily distinguishable by its promptly deciduous leaves, often
glaucous beneath, is referred to in the present paper under the name
polar is.

22. SALIX LEIOCARPA (Cham.) Coville. Least Willow.

Plate XLi, Figure 2.

Salix polaris leiocarpa Cham. Linnaea 6 : 542. 1 831.

Salix rotundifolia Trautv. Nouv. Mem, Soc. Nat. Mosc. 2 : 304, /. //.
1832.

This charming little plant is the smallest of the Alaskan willows.
It forms close tufts or cushions, the very slender stoloniform stems im-
bedded in the soil, and the leaves, orbicular, smooth, entire, bright
green on both surfaces, from a few millimeters to a centimeter in
diameter, rising a centimeter or two above the surface. The tyj)e

^ Trautv. Nouv. Mem. Soc. Nat. Mosc. 2 : 302-303. 1S32.

2 Cham. Linnaea 6 : 541. 1S31.

3 Rydberg, Bull. N. Y. Bot. Gard. i : 266. 1S99.

THE WILLOWS OF ALASKA 339

specimens of leiocarpa were collected on the island of Unalaska and
at St. Lawrence Bay, Siberia, about latitude 65° 40'. It grows on the
islands and both shores of Bering Sea and the Arctic Ocean, and above
timber line on the Pacific coast of Alaska eastward to Prince William
Sound, preferring not a wet soil, like most willows, but a well drained
humus intermixed with finely broken rock. The little pistillate cat-
kins, barely projecting from the tuft of leaves at the ends of the stems,
bear commonly 2 to 4, or occasionally 6 to S flowers, the ovaries and
capsules entirely smooth.

The specimens examined are as follows :

Prince William Sound. — Above timber line on the large, partly
forested nunatak of the Columbia Glacier, Coville and Kearney (No.

1364)-

Kadiak Island. — Above the alder line on a movmtain back of En-
glish Bay, Coville and Kearney (No. 1475), F. A. Walpole, 1900
(No. 1226).

SJiumagin Islatids. — Popof Island, from Soo to 1500 feet elevation,
Trelease and Saunders (No. 33S0), Kincaid, and at similar elevations
on Unga Island, Trelease and Saunders (No. 3379).

Aleutian Islands. — At Unalaska, J. M. Macoun, 1S91. Part of
Chamisso's type material was also collected here. Doubtless all the
Unalaskan specimens were collected at a considerable elevation above
sea level. On Atka Island, L. M. Turner, iSSo, where it is a com-
mon plant among the patches of crowberry {Einfetruyn nigruni).

St. Matthew Island. — On well drained slopes of rock-weathered
soil close to the level of the sea, Coville and Kearney (No. 2154),
Trelease and Saunders (Nos. 3383, 33S4), Brewer and Coe (No.
476). In similar situations on Hall Island, Coville and Kearney (No.
20S3), Trelease and Saunders (Nos. 30S1, 30S2), Brewer and Coe
(No. 438).

Point Barrow. — Near sea level, John Murdoch, 1S82-3 (No. 24).

Siberian Coast. — Collected at Kayne, or Arakamtchechene, Island
by the Ringgold and Rodgers U. S. North Pacific Exploring Expedi-
tion, C. Wright, 1853-6.

On the Asiatic side of Bering Sea the plant has been found at St.
Lawrence Bay as already noted, and was collected on the Russian is-
land of Karaginski by Mertens, the botanist of Lutke's voyage, in the
years 1S26-9. Ledebour in his Flora Rossica reports several localities
in the inland parts of eastern Siberia. It is probable also that some
of the specimens heretofore reported from Bering Sea and the adjacent
coast of the Arctic Ocean under the name Salix polaris^shl. belong

340

COVILLE

to this species. The type locality of Trautvetter's rotutidifolia is
given as St. Lawrence Island.

The nearest Alaskan relative of Salix leiocarpa^ as indicated in the
key, is Salix fhlebophylla to which the student is referred for further
comments on the peculiarities of this little group of species. Salix
folaris also is a near relative. Salix hehringica Seemen, recently
described ^ from a specimen collected at " Luetke-Hafen," ^ near Bering
Strait, agrees in most respects with this species. Its leaves, however,
are larger, lo by 14 mm., and it is described as having an erect bushy
growth. The specimen on which the description was based was a
pistillate twig 17 cm. in length.

23. SALIX RETICULATA L. Netleaf Willow.

Plate xLii.

Salix reticulata L. Sp. PI. 2 : 1018. 1753.

Salix orbicularis A-iiUKKs. in DC. Prod. 16': 300. 1S68.

A prostrate willow very distinct in its characters, as given in the
key, from any other Alaskan willow. Its filaments, in the staminate
catkins, have their basal portion densely pubescent ; in all other
Alaskan willows the filaments are glabrous throughout. The scales
of the pistillate catkins are broadly deltoid-obovate, dark, or sometimes
pale, red-purple in color, smooth on the outside, or slightly hairy at
the base, pubescent with short curled hairs on the inside, these hairs
projecting only slightly beyond the end of the scale. In most willows
the hairs on the catkin scales are long, straight and silky, commonly
equaling the scale in length. Usually the leaves at maturity are en-
tirely devoid of hairs, but sometimes, notably in the specimens from
the St. Matthew Islands, the long, straight silky hairs frequently pres-
ent on young leaves persist on both surfaces till the leaves are full
grown, usually disappearing in age. The species is widely distributed
in North America in arctic and arctic-alpine situations and occurs also
in Asia and Europe. In Alaska it stretches over the Arctic zone, ex-
cept in the extreme north, and southward it occurs at timber line on
the mountains from the Juneau region to Kadiak Island, and westward
to the Aleutian Islands.

Alaskan specimens have been examined as follows :
Juneatc.—^Qa.r the summit of the mountains east of Juneau, at an
elevation of 3000 feet, F. A. Walpole, 1900 (No. 1233).

1 Seemen, Engl. Bot. Jahrb. Beibl. 52 : 6. 1S95.

2Lutke Harbor, I am informed by the Rev. Sheldon Jackson, general agent
of education in Alaska, is the anchoring place for vessels behind the sandspit in
• St. Lawrence Bay, Siberia.

THE WILLOWS OF ALASKA 34 1

White Pass. — At an altitude of about 3,000 feet, F. A. Walpole,

1900 (No. 1 1 10). At Lake Bennett, J. B. Tarleton, 1899 (No. 15).

Tiikon River. — At Coal Creek Hill, Frederick Funston, 1893 (No.

H3)-

Glacier Bay. — At various points near sea level, Coville and
Kearney (No. 669), Trelease and Saunders (Nos. 3395, 3435, in part,
3436' 343S, 3440)^ Brewer and Coe (No. 49).

7'aktitat Bay. — Near Hidden Glacier, Trelease and Saunders (No.
3441, in part).

Prince William Sound. — At timber line at Orca, elevation about
3500 feet, Coville and Kearney (No. 1192).

Cook hilet. — Between Cook Inlet and the Tanana River, E. F.
Glenn, 1899.

Kadiak Island. — On the mountain back of Kadiak village, Tre-
lease and Saunders (No. 3396), Coville and Kearney (No. 2312),
Cole, F. A. Walpole, 1900 (No. 1181). Mountains back of English
Bay, Coville and Kearney (No. 1455). At Sturgeon River, Coville
and Kearney (No. 2280). Without special locality, C. H. Townsend,
1888.

Alaska Peninsicla. — At Kukak Bay, Coville and Kearney (Nos.
1490, 1537), Trelease and Saunders (No. 3390), Kincaid.

Shuinagi7i Islands. — On Fopof Island, Kincaid. Also specimens
of M. W. Harrington, 187 1-3, probably collected on these islands.

Aleutian Islands. — Unalaska, Coville and Kearney (No. 2199),
Trelease and Saunders (No. 3397), B. W. Evermann, 1892 (No.
156), Attn Island, J. M. Macoun, 1891.

Pribilof Islands. — On St. Paul Island, William Palmer, 1890
(No. 310, in part), C. Hart Merriam, 1891, Kincaid, 1897.

St. Alattheiv Islands. — On St. Matthew, Coville and Kearney
(Nos. 2089, 2147, 2177), Trelease and Saunders (No. 3393), Brewer
and Coe (No. 465). On Hall Island, Coville and Kearney (No.
2081), Trelease and Saunders (Nos. 3391, 3392).

Cape Vancouver. — Collected by J. M. Macoun, 1891.

Port Clarence. — On the tundra, Coville and Kearney (No. 1872),
Trelease and Saunders (No. 3398), Brewer and Coe (Nos. 377a,
396). At the Reindeer Station, James T. White, 1894.

Reported at various points on the Arctic seacoast of Alaska but not
beyond Cape Lisburne.

Andersson in 1868, as cited in the synonymy, gave to the east Si-
berian and Alaskan representative of the circumpolar arctic and
arctic-alpine Salix reticzilata the distinguishing name Salix orbicu-

342 . COVILLE

laris^ and Dr. Rydberg in taking up this name ^ extends the range of
the species across the continent, and states that he has seen no typical
specimens of Salix 7'eticulata from North America. The distinguish-
ing characters assigned by Andersson are the yellowish rose-colored
catkin scales and elliptic-rotund, oval-rotund, elliptic-obovate, or ob-
long leaves of 7'eticulata and the fuscous scales and suborbicular, or
obovate-orbicular leaves of orbicularis. He says also of the latter
species, " Truncus longissime serpit ramulos subvillosos radiciformes
undique emittens," and, " A formis vulgaribus S. reticulatae habitu
abunde diversa." Of these leaf characters it may be said that w^e have
specimens w^ith oblong leaves from Alaska, and specimens with orbic-
ular leaves from northern Europe, and Dr. Rydberg, although main-
taining the species to be distinct, says, "The two cannot be distin-
guished by the leaves." Our Alaska collection contains specimens
with pale rose-colored scales, and we have European specimens with
dark purple scales. As for habit character, we have specimens
from Alaska, collected in drier situations than usvial, with branches
just as short and rootless and congested as any of those from Europe.
An equal lack of constancy is found in the distinguishing characters
assigned by Dr. Rydberg. Our specimens of European reticulata
have bracts not " oblong" but broadly obovate, and instead of being
' ' densely woolly " they are rarely hairy over the entire surface and are
often smooth throughout. The size of the capsule and the density of
its hairy covering appear to be correlated not with the continental
origin of specimens but with the degree of maturity of the capsule
itself. While European as well as American specimens show much
variation among themselves, I am unable, from the material now avail-
able, to see in our American plant a species distinct from the Euro-
pean.

Andersson in iS68 published a Salix venusta"^ from Sitka, the de-
scription of which suggests that the plant may prove to be a form of
Salix reticulata grown in a shaded situation. It is described as dif-
fering from reticulata in its long styles and its leaves bright green on
both sides though somewhat joaler beneath. It is very desirable that
the plant be again collected. It should be looked for on Mount Vers-
tovia.

1 Rydberg, Bull. N. Y. Bot. Gard. i : 260. 1S99.

2 Anders, in DC. Prod. 162; 28S. 1868, not Host. 1S2S.

THE WILLOWS OF ALASKA

343

INDEX TO THE NAMES OF SPECIES AND SUBSPECIES

PAGE

alaxensis 311

amplifolia 314

anglorum 338

arbusculoides 323

arctica of Robert Brown 338

arctica of Pallas 326

barclayi 316

bebbiana 306

behringica 340

brachjstachys 311

buxifolia 336

calif ornica 319

capreoides 311

chamissonis 325

commutata 317

conjuncta 318

crassijulis 326

curtiflora 325

cyclop h vlla 333

diplodictya 326

flavescens 310

fulcrata 319

fuscescens 329

glacialis.. 329

glauca 321

hookeriana 314

humillima 323

leiocarpa 33S

longistylis 313

minor 336

PAGE

mjrtillifolia 325

mjrtilloides 330

niphoclada 322

novaeangliae 325

nuttallii 310

orbicularis 342

ovalifolia 331

palaeoneura 337

pallasii 326

perrostrata 307

phlebophjUa 336

phylicoides 319

polaris 335

pseudomyrsinites 325

pulchra 319

reticulata 340

rhamnifolia 330

richardsoni 315

rostrata 306

rotundata 333

rotundifolia 338

saskatchavana 324

scouleriana 311

seemannii 322

sericea 324

sitchensis 307

speciosa 311

stolonifera 333

unalaschcensis 334

venusta 342

PLATE XXXIII.
Sali'x sitcheiisis Sanson.

Fig. a. Staminate flowering branch, natural size.

b. Staminate flower, showing the nectary, scale, and the single stamen,

enlarged six diameters.

c. Pistillate flowering branch, natural size.

d. Pistillate flower viewed from above, enlarged six diameters.

e. Same, lateral view.

f. Fruiting branch, with mature foliage.

g. Capsule, viewed from the side, showing dried scale and nectary, en-

larged two diameters.
h. Capsule, split open in dehiscence, enlarged two diameters.

(344)

Proc. Wash. Acad. Sci. Vol. Ill

Plate XXXI

HEUOTYPE PRINTING <».

salix sitchensis

Proc. Wash. Acad. Sci., August, 1901.

PLATE XXXIV.
Salix alaxensis (Anders.) Coville.

Fig. a. Staminate flowering twig, natural size.

^. Staminate flower, enlarged six diameters.

c. Pistillate flowering twig, natural size.
d. Pistillate flower, viewed from above, enlarged six diameters.

e. Branch with nearly mature catkins, natural size.
f. Capsule, two views, enlarged two diameters.
g: Leafy branch, natural size.

(346)

Proc. Wash. Acad. Sci. Vol. Ill

Plate XXXIV

HEUOTYPE PRiirrmG 00.

SALIX ALAXENSIS

PLATE XXXV.

Salix ajnplifolia Covillc.

Fig. a. Pistillate flowering twig, seven-eighths natural size.

b. Pistillate flower, enlarged five diameters.

c. Staminate flowering twig, seven-eighths natural size.

d. Staminate flower, enlarged five diameters.

e. Branch with fruit and nearly mature leaves, seven-eighths natural size.

(34S)

Proc. Wash. Acad. Sci. Vol. Ill

Plate XXXV

HEUOTYPE miNTlNG 03.

SALIX AMPLI FOLIA

PLATE XXXVI.
Salix iarc/ayz Anders.

Fig. a. Staminate flowering twig, natural size.

b. Staminate flower, enlarged six diameters.

c. Pistillate flowering twig, natural size.

d. Pistillate flower, viewed from above, enlarged six diameters

e. Same, lateral view.

,/". Fruiting branch, natural size.
g. Capsule, two views, enlarged two diameters.
//. Outlines of leaf margins, enlarged three diameters.

(350)

Proc. Wash. Acad. Sci. Vol. Ill

Plate XXXVI

HOJOTYPE PRIKTIMG (B.

salix barclayi

PLATE XXXVII .
Salix commutata Bebb

Fig. a. Staminate flowering twig, natural size.

b. Staminate flower, enlarged six diameters.

c. Pistillate flowering branch, natural size.

d. Pistillate flower, enlarged six diameters.

e. Fruiting branch, natural size.

f. Capsule, two views, enlarged two diameters.

g. Outline of typical medium-sized leaf, natural size.
h. Outline of leaf margin, enlarged three diameters.

(352)

Proc. Wash. Acad. Sci. Vol. Ill

Plate X XXVII

HEUOTTPE PRINTING OX

SALIX COMMUTATA

PLATE XXXVIII.

Salix pulchra Cham.

Fig. a. Staminate flowering twig, natural size.

b. Staminate flower, enlarged six diameters.

c. Pistillate flowering twig, natural size.

d. Pistillate flower, viewed from above, enlarged six diameters.

e. Same, lateral view.

f. Another pistillate flower, viewed from above, same enlargement.

g. Twig with mature fruit, natural size.

h. Portion of twig and fruiting catkin, showing a peduncle, natural size

/. Capsule, two views, enlarged two diameters.

/. Leafy branch, natural size.

k. Typical leaf, showing stipules, natural size.

(354)

Proc. Wash. Acad. Sci. Vol. Ill

Plate XXXVll

f 1/

SALIX PULCHRA

MEUOTYPE PRIHTING CO.

PLATE XXXIX.
Salix glauca L.

Fig. a. Staminate flowering twig, natural size.

b. Staminate flower, enlarged six diameters.

c. Another staminate flower, same enlargement.

d. Pistillate flowering twig, natural size.

e. Pistillate flower, enlarged six diameters.

/. Another pistillate flower, same enlargement.

g. Branch with leaves and mature fruit, natural size.

//. Capsule, two views, enlarged two diameters.

(356)

Proc. Wash. Acad. Sci. Vol. Ill

Plate XXXIX

HEUOTYPE PRINTINS CO.

salix GLAUCA

PLATE XL.
Salix arctica Pall.

Fig. a. Staminate flowering branch, natural size.

b. Staminate flower, enlarged six diameters.

c. Pistillate flowering twig, natural size.

d. Pistillate flower, enlarged six diameters.

e. Branch with leaves and mature fruit, natural size.

f. Capsule, two views, enlarged two diameters.

< 35S)

Proc. Wash. Acad. Sci. Vol. Ill

Plate XL

HEUOTYPE PBINTING 00.

SALIX ARCTICA

PLATE XLI.

Figure i .
Salix stolonifcra Coville.

a. Staminate flowering branch, natural size.

b. Staminate flower, enlarged six diameters.

c. Pistillate flowering twig, natural size.

d. Pistillate flower, enlarged six diameters.

e. Fruiting branch, natural size.

f. Capsule, two views, enlarged two diameters.

g. Portion of underground stem, with stolon, natural size.

Figure 2.

Salix Iciocarpa (Cham.) Coville.

a. Staminate flowering branch, natural size.

b. Staminate flower, enlarged six diameters.

c. Pistillate flowering twig, natural size.

d. Pistillate flower, enlarged six diameters.

e. Fruiting branch, natural size.

/; Capsule, two views, enlarged two diameters.

(3^0)

Proc. Wash. Acad. Sci. Vol. Ill

Plate XLI

MEUOTYPE PRINTING 00.

SALIX STOLONIFERA
SALIX LEIOCARPA

PLATE XLII.

Salix reticulata L.

Fig. a. Staminate flowering branch, natural size.
I). Staminate flower, enlarged six diameters.

c. Pistillate flowering twig, natural size.

d. Pistillate flower, viewed from above, enlarged six diameters.

e. Same, lateral view.

f. Fruiting branch, natural size.

ir. Capsule, two views, enlarged two diameters.

(362)

Proc. Wash. Acad. Sci. Vol. Ill

Plate XLII

mw b

W^ We

SALIX RETICULATA

HaJOTYPE PRINTING 00.

PROCEEDINGS

OF THE

WASHINGTON ACADEMY OF SCIENCES

Vol. Ill, pp. 363-370. August 23, 1901,

PAPERS FROM THE HOPKINS STANFORD GALA-
PAGOS EXPEDITION, 1898-1899.

I.

INTRODUCTION.

In the fall of 1898, Messrs. G. W. Kneass and W. Johnson
fitted out, in San Francisco, a sealing schooner, the Julia E.
W/ialeu, for a cruise to the Galapagos Islands. Thanks to the
generous cooperation and liberality of Mr. Timothy Hopkins,
Stanford University was enabled to transform this proposed
sealing voyage, in part at least, into a naturalists' voyage. It
was arranged that Capt. W. P. Noyes, who commanded the
schooner, should take on board two representatives of Stanford
University, together with their outfit, should land them on the
various islands in the Galapagos group as also upon Cocos and
Clipperton Islands, should afford them opportunities to make
collections of plants and animals, and should then bring them
and their collections back to San Francisco. Dr. C. H. Gil-
bert, head of the zoological department in Stanford University
selected as collectors on this voyage, Robert E. Snodgrass and
Edmund Heller, advanced students in his department, who had
had experience as collectors. Everything was carried out as
planned. Snodgrass and Heller with their outfit on board,
sailed from San Francisco with Captain Noyes on Oct. 25, 1898,
touched at Guadalupe Island, November 5, were at Clipperton
Island, November 23d and 24th and reached the Galapagos
Islands, December 8, 1898. Here they remained about six

Proc. Wash. Acad. Sci., Aug., 1901. 363

364 HEIDEMANN

months collecting in the various islands till June 23, 1899, and
then sailed to Cocos Island where they remained five days, June
29th to July 3d. Thence they sailed homeward and after
touching at Clarion Island, in the Revillagigedo group, on
August 2, they arrived in San Francisco on August 15, 1899.
Their collections include mammals, birds, reptiles, fishes,
insects, mollusks, arachnida, Crustacea, echinoderms and plants
in all of which departments they strove to make full collections.
Most of the material will be permanently retained in Stanford
University. The spiders and insects have been sent to Dr. L.

0. Howard, in Washington, for distribution to and study by the
government specialists in the various orders. All the collec-
tions are to be studied and reported upon by specialists and the
results published in these proceedings under the general heading
Papers from the Hopkins Stanford Galapagos Expedition^
i8g8-j8gg.

ENTOMOLOGICAL RESULTS (1):

HEMIPTERA.

By Otto Heidemann.
U. S. Department of Agriculture.

In the collection of hemipterous insects brought back from
the Galapagos Islands by Mr. R. E. Snodgrass, in 1899, I find
two species new to science. These are described in this paper.
Several others, though previously known, had not been previously
reported from or collected in the Galapagos.

Family PENTATOMIDiE.

1. Podisus sordidus Stal.

Podisus sordidus Stal, Freg. Eug. Resa, Ins., p. 221, 1859.

Charles, May 10, 1899. Indefatigable, April 28, 1899. Chatham,
May 25, 1899. Albemarle, March 11 and 23, 1899.

Six specimens and five specimens in larval stage ; three males and
females. Stal's type came from the Galapagos Islands.

2. Mutica grandis Dallas.

Canthecona grandis Dallas, List i, Hemipt., p. 91, 1852.

HEMIPTERA 365

Charles, June 35, 1899.

One specimen, a female. This species is found in the West Indies,
Mexico and Brazil. It is also recorded from Florida.

3. Nezara viridans Stal.

Nezara viridans Stal, Freg. Eug. Resa, Ins., p. 128, 1859.

Albemarle, February i, 1S99.

A single specimen, a male. The species also occurs in Peru and
Panama. It had previously been recorded from Chatham Island in
the Galapagos archipelago by Mr. P. R. Uhler (Scien. Res. Expl.
Fish Comm. Steamer Albatross).

4. Thyanta perditor Fabricius.

Cimex perditor Fabricius, Ent. Syst., iv, p. 102, 1794.

Albemarle, June 9, 1899. Chatham, May 25, 1899.

Two specimens ; male and female. It belongs to the group which
has the spines on the lateral margin of the thorax shorter.

Family COREID^.

5. Anasa obscura Dallas.

Anasa obscura Dallas, List 11, Hemipt., p. 505, 1852.

Chatham, May 25, 1899.

Two specimens, male and female. Originally described from the
Galapagos Islands.

6. Jadera sanguinolenta Fabricius.

Cimex sanguiiiolenta Fabricius, Syst. Ent., p. 721, 18, 1775.

Wenman, December 18, 1889.

Four specimens, males and females. Two are of the brachypterous
form. This species occurs in the West Indies, in Brazil, and is also
found in Paraguay.

7. Stenocephalus insularis Dallas (.?).

Stetiocephalus insularis Dallas, List 11, Hemipt., p, 482, 1852.

Indefatigable, April 28, 1899.

A single specimen. It seems to agree with the description by
Dallas, but the specimen is too soiled to allow a definite determination.

The type specimen came from the Galapagos Islands.

8. Harmostes serratus Fabricius.

Coreus scrratus Fabricius, Ent. Syst. iv, p. 133, 1794.

Albemarle, February 25, 1899.

Two specimens, male and female. The specimens have the prin-
cipal characters of Harmostes serratus Fabr., but differ somewhat in
having a shorter rostrum ; they belong to the pale variety.

366

HEIDEMANN

Family BERYTID^.

9. Jalysus (Metacanthus) tenellus Stal.

Jalysus tenellus Stal, Freg. Eug. Resa, Ins., p. 236, 1859. Enum. Hem., iv,
p. 128, 1874.

Chatham, May 27, 1S99.

One specimen, which agrees very well with Stal's description, differ-
ing only in the length of the fourth joint of antennae, which is much
shorter than the second.

Family LYGiEID^.

10. Nysius marginalis Dallas.

Nysius margifialis Dallas, List 11, Hemipt., p. 556, 1852. — Stal, Freg. Eug.
Resa, Ins., p. 252, 1859. — Butler, Proc. Zool. Soc. London, p. 85,
1877.
Two specimens ; male and female. This species has never been re-
ported from any place outside of the Galapagos Islands.

Family CAPSID^.

11. Miris lineata Butler.

Miris lineata Butler, Proc. Zool. Soc. London, p. 89, 1877.

Albemarle, June 12, 1899.

Two specimens, one is a larva. The adult specimen agrees very
well with Mr. A. G. Butler's description of a Miris from Charles
Island (Galapagos Islands) except that it is somewhat larger.

Family REDUVIID^.
(Subfamily Harpactoridae.)

12. Cosmoclopius (Harpactor.) sp. .^

Albemarle, June 12, 1S99,

One specimen in the larval condition. Probably Harpactor nigro-
annulatus Stal.

(Subfamily Nabidae.)

13. Nabis punctipennis Blanchard.

Nabis punctipen7iis Blan'CHARD : Gay, Hist, de Chili, Zool., vil, p. 16 f,
1852. — Sign, Ann. Soc. Ent. de France, Ser. 4, iii, p. 577.

Charles, May 10, 1899. Albemarle, March 23, 1899.
Two specimens; male and female. It very closely resembles the
common species Nabis ferns Linne.

HEMIPTERA 367

(Subfamily Emesidae.)
14. Ghilianella galapagensis sp. nov.

A female. — Body slendei', color brown, abdomen mottled with
darker brown. Head nearly thi'ee-fourths the length of prothorax,
sparsely granulated, eyes moderate ; the usual porrect fi'ontal spine is
reduced to a blunt tooth. Antennaj light brown, first and second
joints equal ; the terminal joints are wanting in the specimen. Front
legs ochraceous, comparatively short and stout, the large spine near
base of femora tipped with black ; tibiae faintly annulated with brown,
front tarsi claw-like, equal in length with the tibiae ; femora of the
middle legs about as long as abdomen, but the femora of the hind
legs longer; the tibiae annulated, at base pale orchraceous. Meso-
thorax and metathorax equal in length ; prothorax a little longer and
ornamented with two oval-shaped spots, and a longitudinal line in the
middle, pale ochraceous, also a streak on the sides near apex; above
densely covered with small brown granules, less so below. Meso-
thorax and metathorax have a flat surface, the sides obsoletely granu-
lated and sharply edged, and a longitudinal, raised line in the middle ;
a well defined carina on the underside of the body runs from the base
of mesosternum down to the apex of sixth segment of abdomen.
The abdomen is gradually inflated from the end of second segment to
the fifth and then narrowing slightly towards apex ; the lateral mar-
gins are a little raised, more so on the last segments ; the first genital
segment is slightly sloping, and the apex rounded off, the second
abruptly declivous ; dorsal part of abdomen flat, imder side much
rounded ; first and second segments, seen from above, equally long ;
the third, fourth and fifth gradually become shorter, the sixth segment
is shortest ; at sides of second segment a little behind its base are two
small erect spines, also small tubercles at apex in the middle of nearly
all the segments, the one on the fourth segment is more prominent.
The under side of abdomen is beset with small, stiff, golden hairs,
which are irregularly arranged. Length 13 mm.

Type. — No. 4931, U. S. Nat. Mus.

One specimen, collected on Hood Island, May 8, 1S99, and one
larva collected on Albemarle Island, March 15, 1S99. This insect
seems so distinct from the other species of the genus, that I have ven-
tured to describe a new species from but one specimen. It somewhat
resembles G. gibbiventris Champ, and also G. Jiliventris Spin., but
differs from both in length of meso- and metathorax, which are
equally long, and in the structure of the abdominal segments. The

368 HEIDEMANN

species can be easily distinguished by the reduced frontal spine and
by the pale ornamental spots on the prothorax.

Family SALDID^.
15. Salda rubromaculata sp. no v.

A male. — Body blackish, pubescent, with fuscous and yellow
markings. Head stout, black between the eyes, in front yellow; two
short, black lines running sideways to the cheeks, which are yellow ;
tylus testaceous, polished, prominent, and much curved downwards ;
underside of head yellowish, on the middle of the throat a black mark;
the occiput carries behind the ocelli two orange-yellow, round dots,
which are quite conspicuous. Antenuie yellowish, the ultimate joints
more fuscous. Rostrum yellow, at tip blackish, reaching the inter-
mediate coxae. Pronotum twice as wide as long, slightly narrowing
anteriorly to nearly the same width as the head; the callosities, form-
ing the anterior lobe, convex, not reaching the sides, rounded by an
impressed, transverse line, which is deeper in the middle; the anterior
margin straight, narrow, yellow, behind it a round, sunken line; pos-
terior margin very concave; the lateral margins straight, narrow,
flattened, somewhat reflexed, anteriorly rounded; the humeral angles
rectangular, with an oblong tubercle near the sides ; color of the
thorax yellow, except the anterior lobe, which is black, a little shining,
but shows on front of the sides a small forked yellow streak. Scu-
tellum large, twice the length of the thorax, hollowed out a little, be-
fore the middle, margins and apex yellow. Hemelytra flat, sericeous ;
clavus yellow with a black line in the middle, and a brown dot near
the apical part; corium yellow, the inner side towards apex dark
fuscous, maculated with reddish-brown ; near costal margin a darker
streak with a black spot behind the middle, also one at apex of corium
and base of membrane; the discal and apical areoles more pale
yellow; sutures and nervures blackish; costal margins narrow and
entirely yellow. Membrane pale, with five long areoles ; veins are
brownish. Legs yellow, the femora above a little infuscated, coxae
fuscous. Sternum black, with broad yellowish-white margins. Ab-
domen dark brown, densely covered with fine hairs ; the segments
edged with yellow, the genital segment yellowish at apex.

Length, 3.5 mm. ; width, 1.5 mm. Type — No. 4930,1!. S. Nat.
Mus.

One specimen from Albemarle Island, January 23, 1899. This new
species belongs to the group, having membrane consisting of five cells.
Judging from the descriptions of 6". ventra/is Stal', and S. argentina

HEMIPTERA 369

Berg, it seems to approach these. In size and color, it more nearly
resembles the latter, but differs in the arrangement of the spots on the
corium and membrane, and in having costal margins entirely yellow.

Family GERRID^.

16. Halobates Wuellersdorffi, Frauenfeld.

Halobates Wuellersdorffi, YKKV^-^-p^\.Ti, Verb. Zool. Ges., v, 17, p. 418. — B.
White, Challenger Exp. Zool., v, vii, p. 40.

Seven specimens, three males and four females, were taken between
Clarion and Clipperton Islands, November 2, 189S. Other specimens
were previously found near James Island, Galapagos Islands (Proc.
U. S. Nat. Museum, v, xni, p. 194), and in the North Pacific near
the California coast. In this lot of Halobates v^iis one female, which
carried a few eggs on her under side, attached to the last segments of
the abdomen. This bug must have been captured in the very act of
egg-laying, as one egg is protruding from the ovipositor. The fol-
lowing notes on the subject of egg-laying are from Professor Buchanan
White in his report on the Pelagic Hemiptera (Challenger Exp.
Zool., V, VII, p. 71). He says: " No observations have been made
as to when and where the eggs are deposited. The statement, that
the female carries them about, attached to the abdomen, after they
have been extruded, Pi'ofessor Moseley informs me is a mistake."

Dr. E. Witlaczil in his treatise on Halobates (Wiener Ent. Zeit.,
Vol. V, p. 333, 1856) mentions, that, during the voyage of the Pisani
a feather of a bird was fished out from the ocean, off the southwest
coast of the Galapagos Islands, entirely covered with eggs of a reddish
color. Doctor Witlaczil prepared them for microscopical examina-
tion, and could distinctly observe the embryos of a Halobates.

17. Halobates sp..?
Albemarle, January 3, 1899.

Five specimens, four females and one male. Doubtless a new
species. The material, however, is in such condition that it can not
be satisfactorily described. The middle tarsi of all the specimens are
more or less damaged. This oceanic bug closely resembles Halobates
sericeus Eschscholtz in shape of body and in color; also to Halobates
hayanus B. White in form of antennae and front tarsi. But it re-
sembles more closely Halobates gennanus B. White in the structure
and color of the abdominal and genital segments, and differs only in
the terminal joints of antennae, of which the third and fourth are
nearly equal, and the second a little longer than the fourth, whilst in

370 HEIDEMANN

that species the fourth joint is longer, and the third comparatively
smaller. In the front tarsi the second joint is shorter, being half the
length of the first joint. The insect also shows some affinities to
Halobates regalis Carpenter, and Halobates incanus Witlaczil.

SUBORDER HOMOPTERA.
Family FULGORIDiE.
i8. Mycterodus productus Stal.

Mycterodus productus Stal, Freg. Eug. Resa, Ins., p. 278, 1859.
Albemarle, January 23, 1899.
Nine specimens, varying in color.

19. Jassus (Deltocephalus) sp. ?
Albemarle, January 23, 1899.

One single specimen, not quite mature.

20. Jassus sp. ?
Albemarle, January i, 1S99.

One specimen; mutilated beyond recognition.

21. Cicada sp.?

A single specimen, in the first or second pupal stage, found on Cocos
Island, February 2, 1899. It burrows in the ground and may be found
in that stage of development about two feet beneath the surface

PROCEEDINGS

OF THE

WASHINGTON ACADEMY OF SCIENCES

Vol. Ill, pp. 371-379. November 7, 1901.

PAPERS FROM THE HOPKINS STANFORD GALA-
PAGOS EXPEDITION, 1898-1899.

II.

ENTOMOLOGICAL RESULTS (2) :

DIPTERA.

By D. W. CoqiTiLLETT,

Custodian of Diptera, U. S. National Museum.

The earliest recorded capture of Diptera on the Galapagos
Islands occurred during the cruise of the British Frigates Ad-
venture and Beagle in the years 1826 to 1836, made famous by
Darwin's celebrated Narrative, published in 1839. "^^^ ^^^
Diptera then collected were described by Walker in the well-
known British Museum Lists.

During the years 185 1 to 1853 the Swedish frigate Eugenics
also visited these islands, and collected Diptera ; these were de-
scribed by C. G. Thomson in 1868.

The British steamship Petrel visited them in 1875, and ob-
tained two species of Diptera ; these were reported upon by F.
Smith in the Proceedings of the Zoological Society of London
for 1877, page 84.

Finally, the U. S. steamer Albatross, during her cruise in
1887 and 1888 also visited these islands and collected several
specimens of an undetermined species of Culex, reported upon
by Dr. S. W. Williston in the report published by Dr. L. O.
Howard in the Proceedings of the U. S. National Museum,
Vol. XII, 1889.

Proc. Wash. Acad. Sci., Nov., 1901. 371

372 COQUILLETT

During the autumn of 1899 ^^'- Howard received, in ad-
dition to insects belonging to nearly all other orders, a series of
Diptera collected in the Galapagos Islands by Mr. Snodgrass,
of the Hopkins Stanford Galapagos Expedition in the early
part of that year. This collection, submitted to the National
Museum through Professor Vernon L. Kellogg, contains 413
specimens, representing 35 species, distributed in 26 genera and
17 families. One genus and nine species are believed to be
new to science ; eleven species were originally described from
specimens collected on these islands, while the remaining fif-
teen species have hitherto been reported from South or Central
America, the West Indies, or the warmer portion of North
America.

Family CHIRONOMIDiE.

CER.\TOPOGOX GALAPAGENSIS sp. nov.

Head black, antenncE yellow, the first joint and apical half dark
brown, its hairs dark brown, many towards the apex tipped with yel-
lowish white, proboscis and palpi yellowish brown, the latter slender
and almost linear ; body dark brown, its hairs yellow, the humeri and
a vitta on upper part of pleura yellow, mesonotum opaque ; legs, in-
cluding the coxie, wholly light yellow, femora not swollen, destitute
of spinous bristles, first joint of hind tarsi slightly over one-half as
long as the second, last joint subequal to the fourth, not spinose below,
the claws small and of an equal size ; wings hyaline, thickly covered
with hairs, third vein ending in the costa near middle of the latter,
united to the first vein except at each end, second posterior cell rather
long petiolate ; halteres yellowish white; length 1.25 mm. Tagus
Cove, Albemarle. Three males, collected January 22, 1899.

Type. — Cat. No. 4714, U. S. National Museum.

Family CULICID^.

CULEX T^NIORHYNCHUS Wied.

Culex tceniorhynchus Wiedemann, Diptera Exot., p. 43, 182 1.

Albemarle, January 28 and February i and 4. Eight specimens.
Occurs over the warmer portion of this continent. The female has
dentate front tarsal claws, and Arribalzaga's Tceniorhyiichus tcenio-
rkynchzis ^'led.^ with simple claws, must therefore belong to some
other species.

DIPTERA 373

Family TIPULIDiE.

GERANOMYIA STIGMATICA PhiHppi.

Plettusa stigmatica Philippi, Verhand. Kais. Kon. Zool.-bot. Gesell. Wien,
XV, p. 598, 1865.

Albemarle, June 9. One specimen. Described from Chile.

Family STRATIOMYID^.

PELAGOMYIA ALBITALUS Williston.

Pelagomyia albitalus Williston, Trans. Ent. Soc. London, p. 299, Sept.,
1896.

Albemarle, January i . One specimen. Type locality, St. Vincent,
W. Ind.

NEMOTELUS ALBIVENTRIS Thomson.
Nemotelus albiveniris Tno-^isoyi , Fregatt. Eugen. Resa, p. 462, 1868.

Albemarle, January i. A single specimen. The locality,
"Manilla," given by Thomson, is probably erroneous. The frigate
Eugenics also made collections at the Galapagos Islands.

NEMOTELUS ACUTIROSTRIS Loew.
Nemotelns aciitirostris Loew, Berl. Ent. Zeitsch., vii, p. 8, 1863.

Albemarle, January i. A single specimen. This species was
originally described from Cuba.

Family TABANIDiE.
TABANUS VITTIGER Thomson.
Tadanus vtWger Thousot^, Fregatt. Eugen. Resa, p. 451, 1868.

James, April 21 and 22; Indefatigable, April 28, and Albemarle,
January 23. Six specimens. Originally described from the Gala-
pagos Islands.

Family BOMBYLID.E.

ANTHRAX NUDIUSCULA.? Thomson.

Anthrax lateralis Thomso-h, Fregatt. Eugen. Resa, p, 482, 1868. Nee Say,

1823.
Anthrax nudiuscula ? Thomson, I.e.

Albemarle, January i, 17, iS and 23, and March 21 ; Charles, May
10. Ten specimens. A. lateralis \\a.% described from the Galapagos
Islands ; A. nudiicscula from Panama.

ANTHRAX BRACHIALIS Thomson.
Aftthrax brachialis ThOiMSON, loc. cit., p. 484.

374 COQUILLETT

Albemarle, January i . A single specimen. Described from the

Galapagos Islands.

Family ASILID^.

OMMATIUS MARGINELLUS Fabr.

Asihis 7nargifiellus Yabkicwjs, Spec. Ins., ii, p. 464, 1781.

James, April 21. A single specimen. Originally described from

the West Indies, it has also been reported from Brazil.

Family DOLICHOPODID^.
ANCHINEURA TIBIALIS Thomson.
Anchineura tibialis Thomson, loc. cit., p. 507.

Albemarle, January 18 and 23; Narboro, January 13. Five speci-
mens. Originally described from the Galapagos Islands.

PARACLIUS PUSILLUS.? Macq.
Dolichopus pusillus? Macquart, Dipt. Exot., Sup. I, p. 121, 1844.

Albemarle, January 23; Narboro, January 13. Fourteen speci-
mens. Macquart's specimens o\ pusillus came from Brazil.

Family SYRPHID.E.
BACCHA CLAVATA Fabr.

Syrphus clavatiis Y KSs-KiQivs,, Ent. Syst., iv, p. 298, 1794.
Baccha facialis Thomson, Fregatt. Eug. Resa, p. 504, 1868.

James, April 22 ; Charles, May 10. Two specimens. Thomson
described it from the Galapagos Islands. The synonymy is by Dr.
Williston.

SPH^ROFHORIA SPLENDENS Thomson.
Syrphtis spletidens Thomson, Fregatt. Eug. Resa, p. 501, 1868.

James, April 21 ; Charles, May 10. Three specimens. Thomson's
specimens also came from the Galapagos Islands.

MESOGRAMMA DUPLICATA Wied.
Syrphus duplicatusWi'E'DK^iX'iiyi, Ausser. Zweif. Ins., 11, p. 142, 1830.

James, April 22. Two specimens. This species has been reported
from nearly all j^arts of South America.

Family SARCOPHAGID^.
SARCOPHAGA OBTUSIFRONS Thomson.
Sarcophaga obtusifronsTnouso'H, Fregatt. Eugen, Resa, p. 536, 1868.

Albemarle, January I, and Charles, May 10. 103 specimens. This

DIPTERA 375

and the two following species were originally described from the
Galapagos Islands.

SARCOPHAGA INOA Walker.
Sarcophaga inoa Walker, List Dipt. Ins. Brit. Mus., iv, p. 832, 1849.

Albemarle, January i and 16; Narboro, January 13 and 26;
Hood, May 15, and Culpepper, December 10. 48 specimens.

SARCOPHAGA VIOLENTA Walker.
Sarcophaga violenta Walker, loc. cit., p. 826.
Albemarle, January iS, Seven specimens.

CHRYSOMYIA QUADRISIGNATA Thomson.
Lucilia quadrisignata Thomson, Fregatt. Eug. Resa, p. 544, 1868.

Albemarle, January i, and Charles, May 10. Eight specimens.
Originally described from the Galapagos Islands.

Family MUSCID^.

SYNTHESIOMYIA BRASILIANA Br. and Bergen.

Synthesio7nyia brasiliana Brauer and Bergenstamm, Denks. Math, Nat.
Kais. Akad. Wiss., LX, p. 96, 1893.

Albemarle, January i and 16. Two specimens. Originally de-
scribed from Brazil; the U. S. National Museum contains specimens
collected in Georgia and Florida.

Family ANTHOMYID.E.

LEUCOMELINA PICA Macquart.
Leticotnclina pica Macquart, Dipt. Exot., Sup. iv, p. 235, 1849.

Cocos Island, in July. Twenty-five specimens. Described from
Brazil, and also reported from Mexico.

OPHYRA SETIA Walker.
Anthomyia setia Walker, List Dipt. Ins. Brit. Mus., iv, p. 956, 1849.

Albemarle, January i and 16; Narboro, January 13. Twelve
specimens. Originally described from the Galapagos Islands.

HOMALOMYIA CANICULARIS Linne.
Musca canicularis Linne, Fauna Suec, p. 1841, 1761.

Albemarle, January i . Three specimens. A. nearly cosmopolitan
species.

376 COQUILLETT

Family ORTALID.S:.
EUXESTA NOTATA Wied.

Ortalis notata Wiedemann, Ausser. Zweif. Ins., 11, p. 462, 1830.

Charles, May 10. A single specimen. A common species over the
greater part of the United States.

EUXESTA NITIDIVENTRIS Loew.

Euxesta nitidiventris Loew, Mon. Dipt. N. Amer. , iii, p. 157, 1873.

Albemarle, January i, and Charles, May 10. Thirteen specimens.
Originally described from Texas ; it also occurs in Georgia and Florida.

PAREUXESTA gen. nov.

Near Euxesta^ but the first vein covered with bristles on its apical
third. Front not punctured, sparsely covered with bristly hairs except
around the ocelli, slightly tapering to the lower end, where it is almost
as wide as either eye, antennie two-thirds as long as the face, the third
joint ellipsoidal, one-half longer than wide, twice as long as the
second, which is scarcely longer than wide, arista bare, face in profile
concave, not carinate nor foveolate, projecting slightly further forward
at oral margin than at base of antennie, clypeus prominent, proboscis
short and swollen, palpi well developed, of nearly an equal width,
cheeks less than one-fourth as wide as the eye-height, occiput slightly
concave on its upper part, thorax bearing one pair of acrostichal
bristles, two dorsocentral, three supra-alar, one humeral, two post-
humeral, one mesopleural and one sternopleural, scutellum bearing
four bristles, abdomen ovate, legs short and robust, calypteres small,
wings slightly tapering to their apices, fourth vein distinctly converg-
ing toward the third, lower outer angle of anal cell drawn out in a
long lobe, small crossvein perpendicular, the hind one nearly so, stigma
short, distance between apices of auxiliary and first veins much less
than lensrth of hind crossvein.

Type^ the following species :

PAREUXESTA LATIFASCIATA sp. nov.
Head reddish brown, opaque, the occiput (extending as ocellar and
orbital triangles upon the front), also usually the face, clypeus and
lower edge of cheeks, bluish ; narrow frontal orbits, face, upper
and lower border of cheeks and occiput, whitish pruinose ; antennas
and palpi yellowish brown ; proboscis and apices of aristae black ; body
bluish, subopaque, the abdomen somewhat polished ; an opaque black-
ish streak in front of each wing, extending along the posterior side of

DIPTERA 377

the mesonotal suture ; last two abdominal segments of female yellow,
her ovipositor broad, flattened, scarcely tapering at the apex, bluish,
the base yellow ; legs blackish brown, apices of femora yellow ; wings
whitish hyaline, marked with four rather broad black cro^sbands, the
first just beyond humeral crossvein, second filling stigma and extend-
ing in a curve to slightly below fifth vein beyond apex of anal cell ;
small crossvein situated beyond this band, third band extends from
beyond apex of first vein to lower end of hind crossvein, the fourth
borders apex of wing from nearly midway between apices of first and
second veins to slightly below apex of the fourth, and is sometimes
narrowly connected with the third band along the costa ; at the third
vein the apical brown band is from two-thirds as wide to as wide as
the preceding hyaline interval ; length, 3 to 4 mm. Fifty-nine males
and eleven females. Tower, June 23.

Type. — Cat. No. 5557, U. S. National Museum.

PAREUXESTA OBSCURA sp. nov.

Same as latifasciata except that the entire abdomen and ovipositor
of the female are bluish, and the black bands of the wings, especially
the last two, are narrower, the apical band at the third vein less than
one-fourth as wide as the preceding hyaline interval; length 3 to 3.5
mm. Seventeen males and four females. Albemarle, January i.

Type. — Cat. No. 5558, U. S. National Museum.

PAREUXESTA INTERMEDIA sp. nov.

Equal to latifasciata except that the penultimate segment of ab-
domen of the female is bluish, like the preceding segments, and the
black bands of the wings are noticeably narrower, the apical band at
the third vein being from one-half to three-fifths as wide as the pre-
ceding hyaline interval ; length, 3 to 4 mm. Two males and a
female. Albemarle, January i .

Type. — Cat. No. 5559, U. S. National Museum.

PAREUXESTA HYALINATA sp. nov.

Like latifasciata except that the penultimate abdominal segment of
the female is bluish, and the wings are hyaline, unmarked ; length,
3.5 mm. Two females. Albemarle, January i.

Type. — Cat. No. 5560, U. S. National Museum.

Family EPHYDRID^.
EPHYDRA GILVIPES sp. nov.
Green, tinged in places with blue ; halteres, femora and tibiae yel-
low ; tarsi brown ; antennae and proboscis black ; iniddle of front

378 COQUILLETT

polished, the narrow sides, occiput, cheeks and face, except middle of
the upper part of the latter, opaque, gray j^ruinose; two pairs of
fronto-orbital bristles, no medio-frontal macrochajta below the ocelli ;
antennal arista bearing a short pubescence on the upper side ; body
polished, thorax bearing five pairs of dorsocentral bristles, scutellum
bearing four bristles ; wings hyaline, spines of costa only slightly
longer than the diameter of the costal vein ; front tarsi of male not
swollen; length, 4 to 5 mm. Albemarle, February 13. One male
and five females.

Type. — Cat. No. 4429, U. S. National Museum,

CANACE SNODGRASSII sp. nov.

Black ; the palpi, halteres and tarsi yellow ; front polished ; four
pairs of fronto-orbital bristles, and a short distance toward middle of
front from these a row of three or four bristles ; occiput, cheeks, face
and clypeus opaque, light gray pruinose ; two stout bristles in the vibris-
sal region and one slightly above and behind the center of each cheek ;
antennal arista almost bare ; eyes slightly longer than high ; cheeks pos-
teriorly as wide as the eye-height ; body subopaque, thinly grayish prui-
nose, the pleura densely light gray pruinose ; five pairs of dorsocentral
bristles ; scutellum bearing four marginal bristles and a smaller discal
pair; wings grayish hyaline ; length, 3 mm. Albemarle, January 23;
and Narboro, January 13 and 26. Many specimens.

Type. — Cat. No. 4430, U. S. National Museum.

Family OSCINID^.
HIPPELATES PUSIO Loew.
HippelatiS pusio Loew, Berl. Ent. Zeitsch., p. 279, 1872.

Tagus Cove, Albemarle, March 23. Three specimens. Also oc-
curs in this country, ranging from Massachusetts southward to Florida
and Texas.

Family AGROMYZIDiE.
RHICNCESSA COSTALIS sp. nov.
Head black, opaque, gray pruinose ; lower half of front yellowish
brown ; face and cheeks yellow ; antennae brown ; pubescence of arista
very short ; proboscis brown ; palpi yellow ; body black, opaque, gray
pruinose, mesonotum marked with two indistinct brown pruinose
vittae ; four pairs of dorsocentral bristles ; legs yellow, the coxae and
front legs dark brown ; wings grayish-hyaline, the costal margin be-
tween apices of first and second veins brown, sometimes encroaching

DIPTERA 379

considerably on the submarginal cell; halteres whitish; length, 1.5
mm. Tagus Cove, Albemarle, March 23. Two specimens.
Type. — Cat. No. 4715, U. S. National Museum.

Family HIPPOBOSCID^.

PSEUDOLFERSIA DIOMEDE^ sp. nov.
Head brown; a transversely-oval, elevated, polished, frontal spot
reaching slightly below the middle of the front ; a transverse parallelo-
grammatic one occupying the lowest median fourth of the front ;
orbits elevated and polished ; remainder of the front depressed, opaque,
gray pruinose ; shorter hairs of antennae yellow, the stronger ones dark
brown, changing into yellow at their apices ; thorax polished brown,
the angles yellow, the sides, transverse suture, a spot towards the middle
of the thorax from each humerus and a second spot a short distance
behind each of these, also two spots in front of the scutellum, opaque,
gray pruinose ; middle of sternum yellow ; scutellum polished dark
brown, destitute of bristles, posteriorly truncate and ciliate with very
short hairs ; abdomen dark brown, opaque, gray pruinose ; wings
hyaline, veins brown, the first vein, except its base, usually yellow,
last section of the fifth vein and the whole of the sixth except its base,
whitish ; apex of first vein noticeably before the small crossvein, apex
of second vein about twice as far from the apex of the first vein as from
tip of the third ; legs brown, the lower side of the femora and the
tibiae except their outer and inner edges, yellow ; length 7 mm. Albe-
marle, March i. Four specimens, taken on an albatross {^Diotnedea
irrorata Salvin).

Type. — Cat. No. 4431, U. S. National Museum.

PSEUDOLFERSIA FOSSULATA Macquart.

Olfersia fossidata Macquart, Dipt. Exot., 11, p. 434, 1842.

Wenman, December 13. A single specimen. Originally described
from Brazil.

Proc. Wash. Acad. Sci., November, 1901.

PROCEEDINQS

OF THE

WASHINGTON ACADEMY OF SCIENCES .

Vol. Ill, pp. 381-389. November 7, 1901.

PAPERS FROM THE HOPKINS STANFORD GALA-
PAGOS EXPEDITION, 1898-1899.

III.

ENTOMOLOGICAL RESULTS (3):

ODONATA.

[Text Figures 29-34.]

By Rolla p. Currie,

Aid, Division of Insects, U. S. National Museum.

The dragonflies contained in the following list were collected
by Mr. R. E. Snodgrass in the period from February to August,
inclusive, of the year 1899. They comprise seven species in-
cluding sixteen specimens.

The only references to Galapagos Odonata heretofore pub-
lished are contained in the following papers :

1. Account of the Zoological Collection made during the visit
of H. M. S. Petrel to the Galapagos Islands. Communicated
by Dr. Albert Giinther * * * Neuroptera. By R. McLachlan.
Proc. Zool. Soc. London, pp. 84-86, 1877. Two species were
recorded, Pantala hymencBa and Traniea species (probably the
species afterwards described as T. darzutni hy Kirby).

2. Scientific results of explorations by the U. S. Fish Com-
mission steamer Albatross. V. Annotated catalogue of the
insects collected in 1887-88. By L. O. Howard. Proc. U.
S. National Museum, XII, No. 771, p. 194, 1889. Four speci-

Proc. Wash. Acad, Sci., Nov., 1901. (381)

382 CURRIE

mens of ySschna and one Tramea mentioned without specific
determination.

3. A revision of the subfamily Libellulinae, with descriptions
of new genera and species. By W. F. Kirby. * * * Trans.
Zool. Soc. London, XII, p. 315, PI. LI, fig. i, August, 1889.
Tramea darwini^ new species, described.

It will therefore be seen that the collection here reported upon
is the most complete thus far made upon these islands.

Family AGRIONID^.

I. ANOMALAGRION HASTATUM (Say).

Agrion hastata Say, Journ. Ac. Phila., viii, p. 38, 1839.

Agrion hastatmn Hagen, Syn. Neur. N. Am., p. 'j'j , 1861.

Anomalagrion hastatum Selys, Bull. Acad. Belg. (2), xli, p. 255, 1876. —
Kirby, Synonymic Cat., p. 140, 1890. — Calvert, Trans. Am. Ent.
Soc, XX, p. 240, 1893. — Kellicott, Bull. Ohio State Univ. (4). No. 5,
p. 49, March, 1899. — Williamson, 24th Ann. Rep. Dept. Geol. and
Natural Resources, Ind., 1899, Blatchley, Indianapolis, p. 280, 1900.

Three males and three black females, Albemarle Island, February 14
(Nos. 47-52, inclusive). At first I regarded the females as belonging
to some species of Ischnura^ but their resemblance to the males, in
size and general appearance, led me to compare them with the sup-
posed black females of A. hastatum contained in the National Mu-
seum collection and also with the females of IscJmiira verticalis Say.
This comparison resulted in revealing a difference in the form of the
median lobe of the prothoracic hind margin. In Anomalagrion this
is obtusely rounded, flattened, and slightly emarginate or grooved in
the middle. In the specimens of Isckfiura the lobe is rounded, but
more acutely, not flattened nor emarginate medially. The emargina-
tion of A. hastatutn seems to be slightly more pronounced in the
Galapagos specimens than in those from Washington, D. C, and
other localities in the eastern United States.

Distribution. — North America, Cuba, Venezuela, Galapagos Is-
lands.

Family iESCHNID^.

2. ^SCHNA GALAFAGOENSIS sp. nov.

Male. — Face pale blue or green, the f ronto-nasal suture black ;

frons with a line separating it from the eyes, and a T-spot above,

black, stem of T narrowing anteriorly : clypeus twice as broad as

long; labrum pale green or blue, a transverse black line on its basal

ODONATA 383

margin, its free edge brown; maxillae and labium pale blue or green,
more or less tinged with brown, especially at their apices; vertex
black at base, the elevated portion emarginate anteriorly, pale yellow
or blue; occiput pale yellow, its hind margin straight; rear of eyes
black.

Thorax pale brownish, each side with two oblique pale blue stripes,
beneath with some black spots; mid-dorsal thoracic carina, humeral
and second lateral sutures blackish. Legs black, coxse and bases of
femora brown, anterior femora pale brown or luteous beneath.

Abdomen with segments i and 2 inflated, 3 constricted before the
middle, remaining segments subequal in width; brown, marked with
blue or green as follows: all of i, except a posteriorly-emarginate
basal spot on the dorsum ; sides of 2, except some dark spots inferiorly ;
apical portion of dorsum of 2, except transverse median carina and
posteriorly a line on each side parallel to it (the basal brownish area
on this segment reaches this carina at its middle) ; sides of 3 anteriorly ;
a transverse band on the apex of 3-9, interrupted at the longitudinal
median carina ; a transverse spot on the middle of 3-8, also interrupted
at the carina, large on 8, small on the other segments ; sides of 3-8
each with three consecutive spots, sometimes indistinct, sometimes
large and more or less coalescent, the posterior spot uniting with the
apical dorsal band; a dorsal spot covering apical two-thirds of 10,
and some apical spots. Sternum of r with an acute, prominent tuber-

FiG. 29. ^dEsc/i/ia galapagoensis. Right side of loth segment and
appendages.

cle which is beset with short spines and long hairs. Auricles of 2
bidentate. A mid-dorsal, basal, keel-like tooth on 10 as in californica
and cornigera^ but with three parallel, smaller ones each side; apical
margin of 10 somewhat denticulated, this segment a little wider than
long.

Superior appendages as long as, or a little longer than, 94-10,
slightly recurved, their bases slender; basal fourth narrow, then

384

CURRIE

widened on the inner side ; apical two-thirds, before the bluntly-
pointed apex, with outer and inner sides nearly parallel; outer side
slightly convex, with a few small, short spines or denticles; inferior

side grooved, with a rather prominent
tooth near the base ; above, a prominent
median longitudinal carina on the
widened portion, this carina sharp and
strongly elevated apically. Inferior ap-
pendage less than one-half as long as the
superiors, subtriangular, strongly grooved
above ; apex narrow, truncate, a little
upcurved.

Wings hyaline, with a uniform slight
fuscous tinge, the posterior ones yellow-
ish at extreme base. Antenodal portion
of costa and some transverse veins in
basal part of wing pale, other veins black.
Pterostigma almost black. Membranule
cinereous, its basal third or fourth whit-
ish. Supra-triangular space with one
transverse vein in hind wings, none in
front wings ; triangle with four cells, two
on proximal side (one on this side in
right hind wing of this specimen) ; in-
ternal triangle i -eel led (2-celled in left
forewing) ; on or more columns of four
cells between subnodal sector and the sup-
plementary sector below it; fore wings
with 13-14 antecubitals in first series (14
in right wing, 13 in left), 10 postcubi-
tals; hind wings with 8-10 antecubitals
(8 in right wing, 10 in left), 11 postcubitals ; ist and 5th antecubitals
thicker than the others (ist and 7th in left hind wing) ; anal triangle
3-celled.

Female. — Differs from the male as follows : brown area on dorsum
of 2 covering all but two apical spots and a transverse stripe, inter-
rupted in the middle just in front of the carina ; dorsal median trans-
verse spot large on both 7 and 8; 10 almost wholly pale, without
basal teeth on dorsum but with more numerous denticles on its apical
margin above and with many beneath. Appendages slender, lanceo-
late, as long as 9 -f 10. Genital valves reaching a little beyond apex

Fig. 30. ^Efchtia galapa
goensis. Dorsal view of loth
segment and appendages.

ODONATA 385

of 9 ; valvular processes rather short. The first segment of the ab-
domen has a sternal tubercle similar to that of the male. Supra-
triangular space without cross-veins, triangle 3-celled, internal triangle
i-celled; in front wings, 14-15 antecubitals (14 in left wing, 15 in
right), 1st and 5th thickest, 10 postcubitals ; in hind wings, 8-9 ante-
cubitals (9 in left wing, 8 in right), the ist and 5th thickest (ist and
6th in left wing), 11-13 postcubitals (13 in left wing, 1 1 in right).

Length with appendages, $ 60 mm., $ 58; abdomen, $ 45, $ 44;
front wing, ^ 41, 9 42.5; hind wing, (? 40, 9 41; pterostigma of
front wings 3.8, of hind wings 3.4; superior appendages, ^ 4.7, 9 4.

Type. — No. 5419, U. S. National Museum. One male and one
female, collected on Chatham Island, May 26 (Nos. 45 and 44). The
specimens were imperfectly dried and the colors, therefore, are obscured
so as to prevent their accurate description.

This species is nearly related to y^schna cornigera Brauer and A.
califor7iica Calvert.^

3. AN AX sp.

One female, Chatham Island, May 26 (No. 40). This may possi-
bly prove to be A. amazili oi Burmeister. It agrees well with the
description of this species in Hagen's Synopsis of the Neuroptera of
North America, except that in this specimen there is no triangular
blue spot on the superior surface of f rons on each side of the triangular
black median spot. The appendages are 5 mm. in length, elliptic-
ovate, flat, with a low median carina above.

Family LIBELLULID^.

4. PANTALA FLAVESCENS (Fabricius).

Libellula flavescens Fabricius, Ent. Syst. Suppl., p. 285, 1798.

Pantala flavescens Hagen, Syn, Naur. N. Am., p. 142, 1861. — Kirbv

Synonymic Cat., p. I, 1890. — Calvert, Trans. Am. Ent. Sec, xx, p.

254, 1893. — Kellicott, Bull. Ohio State Uni. (4), No. 5, p. 93, March,

1899. — Williamson, 24th Ann. Rep. Dept. Geol. and Natural Resources,

Ind., 1899, Blatchley, p. 315, 1900.

Two females, Charles Island, May 10 and 20 (Nos. 42 and 43).

DistributioTi. — This species is nearly cosmopolitan, being found in
the tropical portions of both hemispheres, and in the United States as
far north, occasionally, as New Hampshire and Wisconsin.

5. PANTALA HYMEN^A (Say).

Libellula hymencea Say, Journ. Ac. Phila., viii, p. 18, 1839.

Pantala hytnencBa Hagen, Syn. Neur. N. Am., p. 142, 1861. — Kirbv, Syn-

iProc. Cal. Acad. Sci. (2), iv, pp. 505-508, pi. xv, figs. 19, 20, 23, 24, 31 and
32, Feb. 19, 1895.

386 CURRIE

onymic Cat., p. 2, 1890. — Calvert, Trans. Am. Ent. Soc, xx, p.
254, 1893. — Kellicott, Bull. Ohio State Univ. (4), No. 5, p. 93, March,
1899. — Williamson, 24th Ann. Rep. Dept. Geol. and Natural Resources,
Ind., 1899, Blatchley, p. 315, 1900.

One female, Charles Island, August 2 (No. 39).

Distribution. — United States, Mexico, Cuba, Galapagos Islands.

6. TRAMEA DARWINI Kirby.

Tramea darwtni Kirby, Trans. Zool. Soc. London, xii, p. 315, PL li, fig.
I, August, 1889. — Kirby, Synonymic Cat., p. 3, 1890.

Two males (Nos. 37 and 38) and one female (No. 46) ; the former
were collected on Chatham Island, May 23 and 27; the latter on Al-
bemarle Island, March 23.

Only the female of this species was described by Kirby, and the
single female specimen in this collection agrees well with that descrip-
tion. The males were thought by me to be a different species until
compared with a series of seven males and four females in the U. S.
National Museum from Hood, Chatham, Charles and South Albemarle
Islands, collected by the U. S. Fish Commission. The females of
this series show a remarkable variation in the length of the dark-brown
basal spot of posterior wings. In some of them it is hardly longer
than wide, in others it reaches one-third of the way, one-half of the
way, or even farther, towards the hind margin of the wing. Unfor-
tunately the wings are badly torn and the hind margin, in some speci-
mens, is entirely absent so that it is impossible to know whether or not
the spot does reach this margin in any of them. In the males it reaches
the hind margin or to the torn hind edge of the wing. In both sexes
the superior surface of the frons and the anterior face of the vertex are
more or less extensively, sometimes entirely, metallic purplish, or
violet; the labrum has the black portion quite variable in extent, some-
times almost completely covering it, sometimes a mere margin. As
the two males of the Snodgrass collection are in fairly good condition,
it may be well to describe them here.

TRAMEA DARWINI Kirby.
Male. — Similar to the female. Superior appendages about as long
as segments 9 4- 10 of abdomen, black, rufous at base; inferior ap-
pendage a little more than one-half as long as the superiors, reaching
a little beyond the denticulated portion. Hamule not exceeding the
genital lobe. Wings very slightly fumose apically; the posterior ones
with the basal spot extending outward as far as the median cross-vein
and backward to the hind margin, the anal margin hyaline, tinged with
rufous.

ODONATA

387

Fig. 31. Caniiacria ftiviipentiis. Geni-
talia viewed from the left side : a, ante-
rior lamina; ^, hamule ; c, genital lobe.

7. CANNACRIA FUMIPENNIS sp. nov.

Male. — Vertex prominent, truncate, brownish, the suture between
it and frons black. Frons, nasus and rhinarium brownish. Labrum
brownish, bordered anteriorly
with black. Labium black ; its
palpi brownish, their internal
margins bordered with black.
Mandibles black, their bases
brownish. Occiput dark brown-
ish. Rear of head black, with
some brownish spots near the
eyes.

Frothorax dark brown or
black, the articulating facets
or the head brownish ; posterior
lobe broad, erect, bilobed, edged

with brownish. Thorax reddish-brown, the sterna and lateral sutures
with some dark brown or black markings. Feet black, the anterior

femora at base beneath, and the
coxae, brownish.

Abdomen yellowish- or red-
dish-brown ; apex of dorsum of
3 black; 4-9 with a median
dorsal back band, widening at
the apex of each segment and
on 8 and 9 covering almost the
entire dorsum ; 10 mostly red-
dish-brown, indistinctly darker
above. Abdomen beneath with
a dark band each side, beginn-
ing on 4.

Genitalia similar to C. gra-
vida Calvert, i. c.,the "anterior
lamina stout, deeply bifid, its
apices projecting further down,
wards than any other part, and
armed with short spines. Hamule small, simple, curved; apex acute,
black, directed outwards. Genital lobe short, projecting downwards
as far as the hamule projects, or a little farther, rounded at the apex
which is broader than the base."

Fig. 32, Cannacria fumifennis. Gen-
italia viewed from below : a, anterior
lamina; ^, hamule; c, genital lobe.

388 CURRIE

Superior appendages a little shorter than segments 9 -f- 10, brownish,
with short, sparse hairs. Viewed from above, they are subparallel in
their apical half; basal half with inner margins concave, this con-
cavity caused by the basal constriction of the appendage ; apex ob-
tuse, rounded, with a small terminal spine on the outer side directed
backwards. Viewed from the side, the appendage curves downward
strongly from its base, widening to the middle, where it is widest, and
where there is a prominent tooth, on the basal side of which are sev-

FiG. 33. Cannacrta fiimipennis. FiG. 34. Cannacria fumipennis.

Right side of loth segment and appen- Inferior appendage, viewed from

dages. below.

eral denticles; beyond this tooth the superior and inferior margins are
straight, subparallel up to the rounded apex and horizontal.

Inferior appendage exceeding the tooth of the superiors by more
than one-fourth of the former's length. Viewed from beneath, its
lateral margins are slightly convergent from base to apex ; the apex
broad, rather deeply excised, outline of excision nearly straight in
middle portion and ivithout a tubercle. Viewed from the side, it is
slightly concave, a little widened in the middle ; extreme tip black, up-
curved, forming a minute blunt hook.

Wings yellowish fusco-hyaline, this color darkest between triangle
and front margin (darker on posterior wings than on anterior) ; the
portion of the wings basad of the arculus, from costa to hind margin,
is almost clear hyaline with the exception of the extreme base. Vena-
tion dark brown or black, lighter basally near front margin. Ptero-
stigma long, rather narrow, dark yellowish, veined with black, sur-
mounting one entire cellule and portions of two others. Membranule
small, grayish. lo-ii antecubitals in first series on front wings (10
on left side, 11 on right), 6 on hind wings ; 9 postcubitals on front
wings, lo-ii on hind wings (10 on left wing, 11 on right, in this
specimen), the first three or four postcubitals occurring only in the first
series. Three rows of discoidal areolets on anterior wings ; on pos-
terior wings, two, increasing. Triangle of front wings crossed by a

ODONATA 389

single vein, of hind wings uncrossed. Internal triangle of anterior
wings 3-celled. On posteriors, no internal triangle. No supra-tri-
angular cross-veins. The single basal postcostal cross-vein is nearer
the base than is the first antecubital, in posteriors much nearer.

Length with appendages, 49.7 mm.; abdomen, 34; front wing,
37.6; hind wing, 35.6; pterostigma, 4 ; superior appendages, 2.6.

Type. — No. 5420, U. S. National Museum, one male collected on
Albemarle Island, March 23 (No. 41).

Owing to the condition of the specimen it was impossible to de-
scribe the colors accurately. This species agrees remarkably well
with the original colored figure and description of Ca7inacria batesii
Kirby, but not with that author's figure of the male appendages.^ As
pointed out by Calvert,^ Kirby's figure represents the superior ap-
pendages as without an inferior tooth. This separates C. batesii., not
only from C. furcata, but also from the species here described. Ac-
cording to that figure, too, the apical outline of the inferior appendage
of C. batesii is unlike that of C. fufnipennis.

'Trans. Zool. Soc. London, xii, pp. 300 and 341, PI. liii, fig. i, and PI.
Lvii, fig. 9, August, 1889.

2Proc. Cal. Acad. Sci. (2), iv, pp. 547 and 551, Feb. 19, 1S95.

PROCEEDINQS

OF THE

WASHINGTON ACADEMY OF SCIENCES

Vol. III., pp. 391-486. [Plates xlih-lxil] November 15, 1901.

PAPERS FROM THE HARRIMAN ALASKA
EXPEDITION.

XXV.

THE ALG^.

By De Alton Saunders.

CONTENTS :

Introduction 391

Geographic distribution 39^

Catalogue of species .... 396

INTRODUCTION.

Collections of algse were made by the writer and other
members of the Harriman Alaska Expedition at Victoria, in
British Columbia, and at the following localities in Alaska :
Wrangell, Annette Island, Juneau, Glacier Bay, Sitka, Yakutat
Bay, Prince William Sound (near Orca and in Virgin Bay),
Cook Inlet, Kukak Bay, Kadiak Island, and Popof Island and
Unga Island of the Shumagin group. This paper contains an
enumeration of the species collected at these localities. Except
where otherwise stated, the numbers representing localities are
those of the collections made by the writer.

The list of Chlorophyceee, except the Oedogoniacege which
were determined by Dr. K. E. Hirn, was prepared by Mr. F.

Proc. Wasli. Acad. Sci., November, 1901. (390

392'

SAUNDERS

S. Collins. Most of the Oscillatoriace^ were determined by
Dr. Maurice Gomont, and the Kst of Bacillariaceai is entirely
the work of Professor A. M. Edwards. Acknowledgment is
also due Dr. F. R. Kjellman, who examined many of the
Alarias and the species of JFiicus and named the entire collec-
tion of CorallinacecC ; to Dr. W. A. Setchell for examining
several species of Lamina?-ia andCyanophyce^e ; and to Dr. W.
G. Farlow for determining several species of the Rhodophyceae.
My thanks are also due to Dr. C. E. Bessey for the loan of
valuable books, to Professor Conway MacMillan for librar}^ and
herbarium privileges and for the loan of books, and to Miss
Josephine E. Tilden for the privilege of examining material,
preserved in fluid, of her entire Puget Sound collection.

In this paper are catalogued 380 species of alg«, of which
nine are new to science and 240 new to Alaska.

The number of species, both fresh water and marine, in each
of the six classes of algas is :

Fresh Water.

Marine.

TOTA

Schizophycese

26

2

2S

Conjugat:c

96

0

96

Chlorophjceae

18

26

44

Phaeophyceae

0

70

70

Rhodophyceae

I

68

69

Bacillariaceje

32

41

73

GEOGRAPHIC DISTRIBUTION.

Though the algal flora of the Pacific coast in not well enough
known to enable a map showing the exact distribution of species
to be made, yet, enough collecting has been done to indicate
the general distribution.

The Cyanophyceas, Chlorophyceae, Bacillariacese and the
single freshwater species of the Rhodophyceee {Bai?-ackospcr-
mum vagmini) are so largely cosmopolitan that they have been
disregarded in the table which follows. Of the remaining 126
species, comprising the Phagophyceae and Rhodophycege, only
seven are found on the southern California coast, 37 occur on
the central California coast, 68 range from Puget Sound north-
ward into Alaska waters, 49 are peculiar to the Pacific coast of
North America, and 55 are either circumpolar or inhabitants

THE ALG^ 393

of the north Atlantic. Furthermore, of the 126 species, 39 are
reported from Bering Sea, and of these, 18 are circumpolar or
north Atlantic forms and 21 are peculiar to the Pacific coast of
North America.

At least three distinct regions may be distinguished in the
algal flora of the Pacific coast of North America ; a southern,
a Californian, and a northern.

The southern region extends from Point Conception south-
ward, perhaps to the equator. It is characterized by JVereo-
cystt's giganteus, Sargassiim agardianimi, Taonia Icnneb acker ce^
Zonaria tourncfortii and many other tropical species.

The Californian region extends from Point Conception on
the south, to Puget sound on the north, and is characterized by
forms like Dictyoneiiron., Post eh ia, Laminaria Sinclair ii^ Lam-
inaria andersonu, and Fuciis hai'veyanus. It may more
properly be limited on the north by the mouth of the Columbia
River and the shore line between that point and Puget Sound
may be regarded as a transitional area.

The northern region begins at Puget Sound and extends
northward to and including Bering Sea. It is characterized by
such species as Odonthalia aleutica, Polysiphojiia btpmnata,
Eiithora cristata, Rhody7nenia perticsa, Constanttnea rosa-
marma, Fucus cvanescens megacephaliis ^ Agariun tumeric Cy-
mathere triplicata, Laminaria hongardiana^ Laminaria bul-
lata, Dictyosiphoti Joeniculaceus, and Myelofhycus intestinalis .

394

SAUNDERS

GEOGRAPHIC DISTRIBUTION OF ALASKAN ALG^ ON THE
PACIFIC COAST.

Names.'

S

,o .

= rt

U S
E<

1

t/3

'a
■.^■^

rt 2

us

a
u
U

o

3

^1

£2

.b o

J?

^- s

c o.H
•^ to S
» « E

lis

C3

E

_3

"o

'•J

X.

'n

P3

a

ir>

bo

c
•c

n

Streblonema minutissima

Streblonema pacifica

*
*

Streblonema irregularis

*

*
*
*
*
*

*
*
*

Kctocarpus tonientosus

*

*

*

*
*
*

*

Sphacelaria racemosa arctica

*

Homeostroma undulata

1

*

Homeostroma lobata

*
*

*
*

«

*

Scytosiphou lonientarius

*

*

Scy tosiphon bullosus

*
*

■X-

*

*

*

*

*

Coilodesme linearis

*
*

Myelophycus intestinalis

*

*

*

*

*

*

*

*
*

*

Dictyosiphon feniculaceus

*

*

Desmarestia aculeata

*
*

•St

Desmarestia viridis

Liebmannia sp

*

*

*

*

*

*
*

Mesogloia simplex

Chordaria flagelliformis

*

Chordaria abietina

*

*

Ralfsia deusta

*

*
*
*

*

Ral fsia clavata

*

Chorda filum

*

Alaria cordata

*

*
*
*
*
*
*
*

Alaria fragilis

*
*
*
*

Alaria laticosta

*

Alaria lanceolata

*

Alaria fistulosa

*

Laminaria bullata

*

*

Laminaria bongardiana

*

Laminaria cuneata

*
*

*
*
*

Laminaria solidungula

Laminaria saccharina

1

*
*
*

Hedophyllum sessile ... .

*

*
*

Hedophyllum subsessile

Cymathere triplicata

*

' Black face type indicates new species.

THE ALG^

395

Names.

2

S
.0 .

^ rt
U a

1
en

L

g

u

0

■a

3

§1

Q a

as

S a
£2

So-
il

11

1=1

B

3
"c

n

to

M

a
*C
u

n

*

*
*
*

*

*

*

*
*

*

*

*

*

-X-

*

*
*

*

*

*

*
*

*

*

*

*

*

*
*
*
*

*

*
*

*

*
*

X-

*
*

*

*

*

*

*

*
*

*

*

*
*

*
*

*
*
*

*

*

*
*?

*

*
*

*

*

*

*

■X-

*

*

*

*

*

*

Delesseria b?erii

*

Delesseria alata

Delesseria sinuosa

*

Delesseria crassifolia

*

*
*

*

*

*?

*

Odonthalia aleutica

*

*

*

Odonthalia kamtschatica

*

*

Polysiphonia bipinnata

*

*

Polysiphonia arctica

*

*

Calithamnion floccosum pacificum

Calithamnion plumula

*

[••••*■■■

*

Calithamnion pikeanum

*

*

Ptilota asplenoides

*
*

*

Ptilota plumosa

1 *

1

1

396

SAUNDERS

Names.

S

51
1

.s

'5
— ^

2 o

0

T3

IT. S

l<
S
PL,

"=■2
= 1

X-t.

Q§

E o
o c

£2

11

(2^ «

e

"o
U

en
ho
.S

n

Ptilota serrata

•x-

*
*

Antithamnion boreale

*

Ceramium rub rum

■X-

*

*

*

Ceramium codicola

*
*

Microcladia borealis

Rhodochorton rothii

*

Gloiosiphonia furcata

*

*

*

Gloiosiphonia californica

Cryptonemia obovata

*

*

Dumontia filiforme

*

*

Cryptosiphonia ^ravana

*

*
*

Dilsea californica

*

Dilsea arctica

*

*

*

Constantinea rosaniarina

*

*

*

Hildenbrandtia rosea

*

*

*

Melobesia patens

*

Lithophyllum farlowii

*

*

LitlTophyllum compactum

*
*
*
*

*

*

*
*

Lithothamnion Iceve

Clathromorphum circumscriptum

Lepidomorphum yendori

-X-

*

Amphiroa tuberculosa

*

Amphiroa epiphle^moides

Corallina arbuscula

*?

*

Corallina pilulifera filiformis

*

CATALOGUE OF SPECIES.

SCHIZOPHYCEiE.

Family CiZff OOCOCCA CE.^.
Chroococcus turgidus (Kuetz.) Naegeli.

Distributed through a mass of Microcystis jiiarginata which formed
a slimy coating on a perpendicvdar cliff over which water was trick-
ling. Juneau (75).

Chroococcus rufescens (Breb.) Naegeli.

Forming with Stigonenia a reddish coating on a rock five hundred
feet above sea level. Prince Williain Sound (Trelease 501).

Schizothrix lardacea (Cesati) Gomont.

Forming bright rose-red tufts on rocks exposed to fresh water spray,
near Orca, Prince William Sound (304). Identified by Gomont.

THE ALGyE 397

Schizothrix lacustris A. Braun.

In a freshwater pool with Stigonema near Prince William Sound
(300). Identified by Setchell.

Microcoleus vaginatus (Vaucher) Goniont.

Forming, with Scytotiema tnirabilis^ a thin coating on damp ground
recentl}' covered by snow. Glacier Bay (104).

Family NOSTOCACEyS.

Nostoc commune Vaucher.

Forming thin leathery thalli of indefinite size and shape on damp
ground near Glacier Bay (106) ; Hidden Glacier, Yakutat Bay (502).

Cells spherical or oblong, 12-18 // in diameter; often two to four
cells coalesced. Tegument colorless ; cytoplasm finely granulated.

Aphanothece microspora Naegeli.

Forming with Chroococcus tiirgidiis a slimy coating on a perpendic-
ular cliff over which water was trickling. Juneau (75).

Microcystis marginata Naegeli.

Forming a slimy coating on a perpendicular cliff near Juneau (75).

Dermocarpa prasina Born. & Thur.

Abundant on SpJiacclaria raceniosa arctica and Sphacelaria cir-
rosa. From Puget Sound to the Shumagin Islands.

Dermocarpa fucicola sp. nov. (Plate xlvi, figs. 4 and 5.)
Phycotheca Boreali-Americana, No. 801.

Plant forming dark violet brown patches 2—12 mm. in extent; cells
40-60 II high, 18-25 /^ broad, ovate, clavate or spatulate, much nar-
rowed below ; schizospores abvmdant. On Fucus evanescens mega-
cephala^ Puget Sound (440). Forming orbicular or irregular patches
which become confluent into irregular masses of indefinite extent.

This plant is closely related to Savageau's D. biscayensis^ but dif-
fers from it in the larger size of the patches and the shape and size of
the individual cells. In D. biscayensis the plant forms orbicular
patches i mm. broad and the cells are 25-30 p. broad.

Family OSCILLATORIACE^E.
Oscillatoria amoena (Kuetz.) Gomont.

The plant formed a soft, felt like, dark bluish-green mass 3-10 mm.
thick, of indefinite extent, lining the bottom of the outlet of a hot
spring. The water in the outlet where the plant was abundant ranged
from 80° F. some distance from the spring to 120° F. near the spring.
Near Sitka (158). Identified by Gomont.

398 SAUNDERS

Phormidium autumnale (Ag.) Gomont.

Forming a thin dark blue coating on small rocks in a rapid stream
emptying into Kukak Bay (332). Identified by Gomont.

Phormidium laminosum (Ag.) Gomont.

Forming a thin membranaceous stratum on perpendicular rocks
moistened by spray from a waterfall, Orca (301). Identified by Go-
mont.

Lyngbya aerugineo-coerulea (Kuetz.) Gomont.

In a felt-like mass of filaments of Vaucheria.

Juneau (74) ; with Zygne7na sp. in a small pond on an island in
the Muir Glacier (107).
Nostoc sphaeroides Kuetz.?

Forming a soft bluish green coating on rocks near Juneau (7=>).

This species was submitted to Dr. Setchell, who says of it "The
Nostoc is a minute spherical one very often found but I am luicertain
as to whether it is N. sphceroldes or not."

Desmonema wrangelii (Ag.) Bor. & Fla.

Mixed with Tolypothrix tenuis from a clear brook emptying into
Glacier Bay (103). In a similar locality and associated with the same
species on Fopof Island (404).

Family 6" C TTONEMA CByE.

Scytonema varium Kuetz.

On rocks moistened by spray from a waterfall near Juneau (76).
Identified by Setchell.

Scytonema myochrous (Dillw. ) Ag.

With Tolypothrix tenuis^ forming small tufts on rocks in a brook
emptying into Glacier Bay. Identified by Setchell.

Scytonema mirabile Bornet.

On moist ground near Glacier Bay, with Microcoleus vaginatus
(104). On the perpendicular surface of a rock, with Vaucheria^
moistened by dripj^ing water, Kukak Bay (347).

Scytonema figuratum Ag.

In a freshwater stream emptying into Glaicier Bay (103 ^).

Tolypothrix tenuis Kuetz.

Forming brownish or blue-green tufts attached to rocks in fresh
water. Glacier Bay (300); Popof Island (404). Identified by Set-
chell.

THE ALG^E 399

Family STIG ONE MA TA CE^^.

Hapalosiphon pumilus (Kuetz.) Kirchner.

In a freshwater pond near Seldovia, Cook Inlet (424).

Stigonema minutum (Ag.) Hassall.

Forming a thin brown coating with Chroococctis rjifescens on damp
rocks several hundred feet above sea level, Prince William Sound
(Trelease i^oi).

The threads are about i mm. high, 13—25 <j. broad, irregularly
branched and flexuously curved ; sheath yellowish brown ; internal
tegument much darker ; main filament usually of one layer of cells ;
the branches usually of two or more ; heterocysts abundant, yellowish,
lateral or intercalary.

Stigonema ocellatum (Dillw.) Thur.

On rocks in a rapid stream emptying into Glacier Bay ; floating in a
quiet freshwater pool, Prince William Sound (300, 303).

The plant forms dark brown, loosely CJEspitose tufts 2-7 or S 'mm.
high which are attached at first but finally are floating ; irregularly
branched, branches elongated, patent, primary filaments one or two cells
thick (35—50/^) all bearing hormogones ; trichomes 21— 40 /^ broad ;
sheath broad, indistinctly lamellose, yellowish brown except near the
ends on young branches ; cells 25—35 ,a, shorter tha-n broad, surrounded
by a dark brown tegument ; heterocysts lateral, scarce ; hormogones
13 /u broad, 50-65 ,a long.

Y^xx^xXy RiVULARIACEvE. ■

Calothrix fusca (Kuetz.) Bornet.

Imbedded in the gelatinous coating of Batrachospermtun vagum
from a freshwater pond, Cook Inlet (423) ; Kadiak Island (504).

The plants, which were apparently immature, resemble the form
of C. cBriiginosa of Kuetz. ^ The threads are loosely gregarious,
broad, curved and somewhat enlarged at the base ; sheath broad,
ochraceous with age, produced into long thread-like articulations ;
heterocysts 10 /a broad.

Calothrix scopulorum (Weber & Mohr) Ag.
In salt water, Puget Sound.

Rivularia biasoUetiana Menegh.

On rocks in freshwater streams, Juneau (76) ; Glacier Bay (102).
Identified by Setchell.

1 Kuetz. Tab. Phvc. 2 : PI. 45, Figs. 5 and 6.

400 SAUNDERS

The plant forms minute, hard, dark bkiish green thalli 1-3 mm. in
diameter, which finally become agglutinated into hollow indefinite
masses.

Family HTDR URA CE.^.

Hydrurus penicillatus Ag.

Plant forming dark olive-green filaments attached to smooth round
pebbles in a rapid brook emptying into Kukak Bav (351).

The filaments are 7—20 mm. in length, having long primary branches
bearing many short penicillate ones.

CONJUGATE.
Family DESMIDIACE^^.

Mesotaenium braunii De Bary. (Plate xliii, fig. 39.)

In a freshwater pond near Cook Inlet.

The cells are cylindrical, 2^^ times as long as broad; 16 to iS/^
broad, 38-4:^ ,a long.

Penium interruptum Breb.

In a freshwater pond near Seldovia, Cook Inlet.

Plant 24-40// wide, 130-145// long. Slightly smaller in all its di-
mensions but agrees otherwise with the description.

Penium closterioides Ralfs.

In a freshwater pond on Popof Island.
Diameter 140//, length i^o//.

Penium margaritaceum (Ehrenb.) Br^b.
Occasional in freshwater, Kukak Bay.
Cells 25 // wide, 100 // long.

Penium polymorphum Pcrty.
In quiet water near Kukak Bay.
Cells 22 // wide, 35 // long. Very delicately punctate-striate.

Penium oblongum De Bary.

In a freshwater pond near Virgin Bay, Prince William Sound
(302).

Diameter 21 //., length 64//.

Penium digitus (Ehrenb.) Breb. (Plate xliv, fig. 3.)
In quiet water, Virgin Bay. Rare.
Diameter 54/i, length 1S3/7.. The plant averages almost a third

THE ALG^ 401

smaller in all its parts than the measurements given by various au-
thorities.

Closterium juncidum Ralfs.

In quiet water near Seldovia, Cook Inlet (422).

Diameter 10;/., length 240 .a. The plant is slightly bent, but little
contracted toward the ends, which are obtusely rounded ; tnembranes
smooth, light brown ; vacuoles not apparent.

Closterium angustatum reticulatum Wolle.

Occasional in a freshwater pond near Cook Inlet.

Diameter 16-19, a, length 340 /^t. The plant is dark brown, marked
with delicate longitudinally and spirally arranged stride.

Closterium acerosum (Schrank) Ehrenb. (Plate xliii, fig. 27.)
In a freshwater pond near Seldovia, Cook Inlet.
Diameter 18—29//-, length 200—300//.,
Virgin Bay, Prince William Sound.
Diameter 29/7., length 265—300//.

Closterium striolatum Ehrenb.

In quiet water in Kukak Bay (283).
Diameter 25— 30/7., length 275— 300 /a.

Closterium brebissonii Delp.

In a freshwater pond near Cook Inlet (224).
Diameter iS/z, length 650 //.

Closterium lunula (Muell.) Nitzsch,

Occasional in quiet water near Kukak Bay.
Diameter 70 /z, length 450 //.

Closterium acutum (Lyngb.) Br^b.

In a freshwater pool near Prince William Sound (300) .

Diameter 8-9//, length 85-100 /jt. The plant is small and slightly
bent, ten times as long as broad, tapering from the middle to the
rounded ends ; cytoderm smooth and colorless.

Closterium dianae Ehrenb.

Freshwater pool. Prince William Sound (302) . Only one specimen
seen.

Diameter 16//, length 180//.

Closterium venus Kuetz. (Plate xliii, fig. 15.)
In a pond near Seldovia, Cook Inlet (422).
Diameter 9//., length 85//..

Closterium parvulum Naeg. (Plate xliii, fig. 14.)

402

SAUNDERS

Very abundant in freshwater pond near Seldovia, Cook Inlet (424)
Diameter 8-1 2 /i, length lOO//.

Tetmemorus brebissonii (Menegh.) Ralfs.

In freshwater near Prince William Sound : frequent.
Diameter 23/7., length 78 /i,

Tetmemorus laevis ( Kuetz.) Ralfs.

In a freshwater stream, Virgin Bay (300) ; Yakutat Bay.

Diameter 21-24/i, length 91-100 /^i. The plant is slightly constricted
and not at all punctate.
Pleurotaenium nodosum (Bail.) Lund.

Common in a freshwater pond, Popof Island.

Diameter 40-4^ /i, length 290-325//.

Pleurotaenium truncatum (Breb.) Naeg.

In a freshwater pond near Seldovia, Cook Inlet.
Diameter 45,", length 216/i.

Disphinctium cucurbita (Br6b.) Reinsch.

In a freshwater pond near Prince William Sound (302).
Diameter 27//., length 67 /i.

Disphinctium connatum (Brdb.) DeBary. (Plate xliii, fig. 30.)

In a freshwater pond near Seldovia, Cook Inlet. Only one speci-
men was seen.

Diameter 24//., length 3S/1.

Docidium baculum Breb.

In a freshwater pond, Popof Island.

Diameter 27//, length 228 /i. The plant is slightly wider than the
description given, but is not at all narrowed as are the other species.

Docidium coronulatum Grun.

In a freshwater pool near Kukak Bay.
Diameter 40//, length 432//.

Docidium dilatatum (Cleve.) Lund.

In a freshwater pond, Popof Island.

Diameter 13 a, length 165/-*. The plant has five pearly granules
at each end.

Docidium gracile Wittr.

Triploceras gracile Bail. Micr.
Common in fresh water near Prince William Sound.
Diameter 27//.. The plant has two rows of rather long spines.

THE ALG^ 403

Docidium minutum Ralfs.

PleurotcEniwn? juinuiuin (Ralfs.) Delponte.
In freshwater near Prince William Sound.
Diameter 1 1 //, length 135 M-

Arthrodesmus convergens Ehrenb. (Plate xliv, fig. 14.)

In a freshwater pond near Seldovia, Cook Inlet (422). Rare.
Diameter 32//, length 32/2 without the spines.

Xanthidium antilopaeum (Breb.) Kuetz. (Plate xliii, fig. 38.)

In a freshwater pond near Cook Inlet.

Diameter 55 /M, length 55 // including the spines. The spines near
the apex of the semi-cells are more nearly straight than in the figures
given of this species.

Xanthidium armatum (Breb.) Ralfs. (Plate xliv, fig. 18.)

In a freshwater pond near Virgin Bay, Prince William Sound.
Diameter 81 /x, length 128^. Most of the spines are bifurcated,

with the points divergent.

Cosmarium granatum Breb. (Plate xliii, fig. 8.)
In a freshwater pond near Seldovia, Cook Inlet.
Diameter i8^«, length 33 /a.

Cosmarium constrictum Delp. (Plate xliii, figs. 33, 34.)

In a freshwater pool near Virgin Bay, Prince William Sound (300).
Diameter 40 /i, length 65 //.

Cosmarium bioculatum Breb. (Plate xliii, fig. 28.)

Common in a freshwater pond, Seldovia, Cook Inlet (422).
Diameter 16-1S/2, length 17-18 />t.

Cosmarium hammeri Reinsch. (Plate xliii, fig. 7.) .

In a freshwater pond near Seldovia, Cook Inlet (422).

Diameter 10-14 /y, length 18-22/^. Very close to the variety ^M^aw-
gustatum of Boldt. The angles are a little more acute than in the
type ; the apex of the semicells is very emarginate.

Cosmarium depressum (Naeg.) Lund. (Plate xliii, fig. 17.)
In a freshwater pond near Seldovia, Cook Inlet.
Diameter 16-20//, length 18, a.

Cosmarium holmiense Lund. (Plate xliv, fig. 28.)
In freshwater near Kukak Bay.
Diameter 32//., length 54//.

Cosmarium venustum (Breb.) Arch.

In a freshwater pond near Seldovia, Cook Inlet.

404

SAUNDERS

Diameter iS//, length 37 /x. In its measurements the Alaskan plant
agrees with Nordstcdt's variety indiu-atuw but has the form of the type.

Cosmarium latum Breb. (Plate xliv, fig. 4,)
In a freshwater pool near Kukak Bay.
Diameter 68 /7., length 100, a.

Cosmarium contractum Kirch. (Plate xliv, fig. 16.)
In freshwater near Kukak Bay.
Diameter 22 /a, length 22 //.

Cosmarium sexangulare Lund. (Plate xliii, fig. 39.)
In a freshwater pond near Seldovia, Cook Inlet.
Diameter 27 /i, length 33/2.

Cosmarium tumidum Lund. (Plate xliv, fig. 21.)

In a freshwater pond near Seldovia, Cook Inlet.

Diameter 27-30 //, length 35-38 ij.. Several specimens were found, all
of which belong to Kirchner's vaviety subtile, the cytoderm being deli-
cately punctate over the whole surface.

Cosmarium parvulum Breb. (Plate xliii, fig. 6.)
In freshwater. Virgin Bay, Prince William Sound.
Diameter 17-19//, length 35-38 ,«. Some of the specimens have a

smooth cytoderm and on some it is delicately punctate.

Cosmarium kitchelii Wolle. (Plate xliv, fig. 17.)

In a freshwater pond Seldovia, Cook Inlet.

Diameter 43 ,«, length 51//.. The plant agrees in all three views
with Wolle's description except that the three central rows of granules
are neither longer nor more conspicuous than the marginal ones.

Cosmarium pyramidatum Breb. ? (Plate xliii, fig. 40.)

In a freshwater pool near Juneau ; Kukak Bay.

Diameter 35 //, length 54//, isthmus 11 /^ wide; cell membranes very
delicately punctate.

Cosmarium pachydermum Lund. (Plate xliv, fig. 12.)
In a freshwater pond near Seldovia, Cook Inlet.
Diameter 73 //, length 103 11.

Cosmarium undulatum Corda. (Plate xliv, fig. 9.)
In a freshwater pond near Seldovia, Cook Inlet (432).
Diameter 38 /j., length 59//; cytoderm smooth, edge crenate, ten

crenfe to a semicell ; sinus enlarged outward.

THE ALG^ 405

Cosmarium pseudogranatum Nordst. (Plate xliii, fig. 21.)
In a freshwater pond, Popof Island.
Diameter 3^ //, length 51 p..

Cosmarium botrytis Menegh. (Plate xliii, fig. 11.)
In freshwater, Popof Island.
Diameter 35-43 ,«, length 40-52//.

Cosmarium conspersum Ralfs. (Plate xliii, fig. i.)
In a freshwater pond, Cook Inlet (424).

Cosmarium caelatum Ralfs. (Plate xliii, fig. 2.)

In a freshwater pond, Glacier Bay ; only one specimen seen.
Diameter 40—41^ [m.

Cosmarium portianum nephroideum Wittr. (Plate xliii, figs. 12,

Freshwater pond, Seldovia, Cook Inlet.

Diameter 21 ,u, length 32 /i ; isthmus 9 /jt broad; sinus 5 [j. wide.
The verruciE are short, absent from the sinus.

Cosmarium ochtodes Nordst. (Plate xliii, fig. 10.)
In a freshwater pond near Seldovia, Cook Inlet.
Diameter 64 //, length 97 /-t.

Cosmarium intermedium Delp. (Plate xliv, fig. i.)

In a freshwater pond near Seldovia, Cook Inlet.

Diameter 74 /i, length 66 n, isthmus 1 2 // wide. The plant is cov-
ered all over with large pearly granules.

Cosmarium subcrenatum Hantzsch. (Plate xliii, fig. 20.)
In freshwater ponds, Glacier Bay ; Kukak Bay.
Diameter 26—30 /i, length 35—40 //..

Cosmarium phaseolus Breb. (Plate xliv, fig. S.)
In freshwater near Seldovia, Cook Inlet.
Diameter 24 //, length 27 fi ; sinus 3 fx broad; isthmus 4 /x broad.

Cosmarium costatum Nordst.

In freshwater near Virgin Bay, Prince William Sound.
Diameter 40 /j.

Cosmarium pulcherrimum Nordst. (Plate xliii, figs. 18, 19.)
In freshwater near Kukak Bay.
Diameter 38 //, length 52 /x.

Cosmarium quadrifarium Lund. (Plate xliii, fig. 22.)
In freshwater near Virgin Bay, Prince William Sound.
Diameter 33 ,a, length 40 /j..

406 SAUNDERS

Cosmarium broomei Thwaites. (Plate xliii, figs. 26, 35, 36.)
In freshwater, Popof Island.
Diameter 40 /x, length 40 ij..

Cosmarium ornatum Ralf. (Plate xliii, fig. 3.)

Very abundant in a freshwater pond, Kukak Bay ; Cook Inlet.
Diameter 32-45 p.. length 38-75 /i.

Cosmarium sphalerostichum Nordst. (Plate xliv, fig.. 6.)
In a freshwater pond near Seldovia, Cook Inlet.
Diameter 16 yu, length 16 //.

Cosmarium pseudotaxichondrum Nordst. (Plate xliii, fig. 5.)
In freshwater ponds near Prince William Sound.
Diameter 33 //, length 22 /i.

Pleurotaeniopsis pseudoconnata (Nordst.) Lagerh.

In a freshwater pond near Seldovia, Cook Inlet.

Diameter 27 /i, length 38 11. The sinus is broad and very shallow
and the margins distinctly striate ; cytoderm finely punctate ; the end
view a perfect circle and the chlorophyll body divided into eight equal
arms.

Pleurotaeniopsis debaryi (Archer) Lund. (Plate xliv, fig. 20.)

In a freshwater pond near Seldovia, Cook Inlet,

Diameter 65 ,a, length 100 //, isthmus 40 ix broad. Only one speci-
men w'as found and this agrees with Nordstedt's \^x\€(.y spitsbergensis.

Pleurotaeniopsis ralfsii (Breb.) Lund.

In a freshwater pond near Seldovia, Cook Inlet.
Diameter 54 //, length 70 <)..

Staurastrum dejectum Breb. (Plate xliv, fig. 5.)

In a freshwater pond near Seldovia, Cook Inlet (422).
Diameter 21 ,a, length 21 //.

Staurastrum dejectum mucronatum (Ralfs) Kirchn.
Occurs with the type.
Diameter 27 //, length 27 //..

Staurastrum ravenelii Wood. (Plate xliv, fig. 29.).
In a freshwater pond, Seldovia, Cook Inlet.

Staurastrum pygmaeum Breb.

In a freshwater pond near Seldovia, Cook Inlet.
Diameter 27 .a, length 30 ji.

Staurastrum calyxoides WoUe. (Plate xliii, fig. 31.)
In freshwater near Prince William Sound (300).
Diameter 10 //, length 20 //.

THE ALG^ 407

Staurastrum furcigerum Breb.

In a freshwater pond near Seklovia, Cook Inlet (422).
Diameter 45 //., length 50 //.

Staurastrum arctiscon (Ehrenb.) Lund.

In a freshwater pool, Virgin Bay, Prince William Sound (302).
Common.

Diameter 1 16 ij..

Staurastrum polymorphum Br^b. (Plate xliii, figs. 23, 24.)
In a freshwater pond near Seldovia, Cook Inlet (402).
Diameter 40 //, length 4^^ /z. The end view shows four arms

rounded at the ends.

Staurastrum echinatum (Perty) Rab. (Plate xliii, fig. 16.)
In a freshwater pond near Seldovia, Cook Inlet.

Euastrum verrucosum Ehrenb. (Plate xliii, fig. 9.)
In freshwater, Popof Island.
Diameter 81 //, length 88 fi.

Euastrum gemmatum Breb.

In a freshwater pond near Seldovia, Cook Inlet.

Diameter 40 //, length 67 //. The plant agrees exactly with Br^b-
isson's description except that the terminal lobe is not emarginate.

Euastrum pokornyanum Grun.

In a freshwater pond near Kukak Bay.

Diameter 21 /a, length 43 [j.. The basal lobe of the trilobe semicell is
crenate, the terminal lobe subcuneate, truncate and excised; membrane
smooth.

Euastrum oblongum (Grev.) Ralfs. (Plate xliii, fig. 37.)

In a freshwater pond near Prince William Sound.

Diameter 67 /^, length 130 /x. The semicells are five lobed, all of
the lobes having the broad shallow marginal cavity.

Euastrum crassum (Breb.) Kuetz. (Plate xliii, fig. 4.)
In a freshwater pond near Seldovia, Cook Inlet.
Diameter 68 //, length 148 //.

Euastrum didelta (Turp.) Ralfs. (Plate xliv. Fig. 24.)

In freshwater near Kukak Bay.

Diameter 46-60 ,a, length 86-120 //, sinus 16 // deep, bridge 1 1 /x
wide.

Euastrum ansatum Ralfs.

In a freshwater pond near Yakutat Bay.
Proc. Wash. Acad. Sci., November, 1901.

408 SAUNDERS

Diameter 27 /^, length 54 //.

Euastrum elegans (Breb.) Kuetz. (Plate xliv, figs. 2, 25, 26, 30.)
Common in freshwater near Yakutat Bay ; Prince William Sound ;

Popof Island.

Diameter 13-15 a*, length 20-30 /./..

Euastrum affine Ralfs. (Plate xliii, fig. 32.)
In freshwater material from Glacier Bay.
Diameter 60 11 wide, length 121 /x.

Micrasterias truncata (Corda) Ralfs. (Plate xliv, fig. 7.)
In a freshwater pond near Glacier Bay.
Diameter 92 //, length 97 /i.

Micrasterias oscitans pinnatifida (Kuetz.) Rabenh. (Plate xliv,
fig. 27.)

In a freshwater pond near Seldovia, Cook Inlet.

Diameter 60 /z, length 60 /i. Plant is considerably smaller than the
measurements given by various authors but agrees perfectly as to shape
and markings.

Micrasterias rotata (Grev.) Ralfs. (Plate xliii, fig. 25.)

In a freshwater pond, Prince William Sound.

Diameter 190-220 /j. The plant varies considerably in the length
of the processes on the ultimate division of the semicells.

Micrasterias denticulata (Br^b.) Ralfs. (Plate xliv, fig. 11.)
In a freshwater pond near Kukak Bay.
Diameter 90 //, length iio,«.

Micrasterias fimbriata elephanta Wolle. (Plate xliv, fig. 31.)
In freshwater pond near Seldovia, Cook Inlet.
Diameter 350 /a, length 350 11.

Micrasterias kitchelii Wolle. (Plate xliv, fig. 23.)
In a freshwater pond near Seldovia, Cook Inlet.
Diameter 135 m, length 135 /j'-.

Sphaerozosma excavatum spinulosum (Del Ponte) Hansg.
In a freshwater pond near Seldovia, Cook Inlet.
Diameter 8-10 /x, length 11 11.

Desmidium swartzii Ag.

In a freshwater pond near Seldovia, Cook Inlet.
Diameter 35 ,a.

THE ALG^ 409

Gymnozyga (?) longata (WoUe) Nordst.

In a freshwater pond near Seldovia, Cook Inlet.
Diameter 35 //.

Family Z TGNEMA CE^E . '

Spirogyra varians (Hass.) Kuetz.

In a freshwater pond near Seldovia, Cook Inlet.
Diameter 38-40//.

Spirogyra porticalis (Muell.) Cleve.
In running water, Popof Island.
Diameter of the zygospores 32-42//..

CHLOROPHYCEiE.

Family PLEUR O CO CCA CE^E.
Oocystis solitaria crassa (Wittr.) Hansgirg.

Forming a mucous coating on damp rocks near waterfalls. Juneau
(75;. Identified by Setchell.

YsimWy PROTOCOCCACE^E.

Sphaerella lacustris (Girod.) Wittr.

On snow near Yakutat Bay, June 23.

This is the so-called " red snow " found in standing rain water and
on snow throughout Europe and North America. It was also ob-
served by members of the party on Muir Glacier and on snow above
Orca, Prince William Sound.

Family HTDRODTCTIACEyE.
Pediastrum boryanum (Turp.) Menegh.
In freshwater, Popof Island.

Pediastrum angulosum (Ehrenb.) Menegh.
Abundant in a freshwater pond, Popof Island.
The plant is discoidal, 32-celled, the cells all angled.

Family UL VA CE.^.

Monostroma fuscum (Post & Rupr.) Wittr.

On rocks, Muir Inlet, Glacier Bay (105); Sitka (144); Virgin

Bay, Prince William Sound (277) ; Kukak Bay (315) ; Lowe Inlet

(16).

^ Several sterile specimens of Zygnenia were collected, but no fruiting material
was obtained.

4IO SAUNDERS

Frond 40-60 ij. thick, dark green, becoming brownish or blackish
when dried, not adhering to the paper ; cells in cross-section usually
longest at right angles to the surface of the frond.

In the Kukak Bay specimens the frond is about 40 ,a thick, and the
cells are nearly square in cross section, agreeing with the typical AI.
fuscu?n. In the Lowe Inlet specimen the frond is 30//. thick, the cells
in cross section somewhat rounded, agreeing with the form known as
M. splendens. The color however is duller than usual in that form.

Monostroma vahlii J. Ag.

Kukak Bay (316).

Frond persistently tubular, torn at the top so as to form a flat mem-
brane only at a quite late stage; thickness of membrane 15-25 /z; cells
about 1 2 /J in cross section, somewhat arranged in series when seen
from the surface.

This species has much resemblance in habit to tfie genus Entero-
morpha, at least when young.

Monostroma groenlandicum J. Ag.

Kukak Bay (346).

Frond filiform, opening only at the extreme top, when in fruit ;
thickness of membrane 25-30//; cells roundish-angular seen from sur-
face, radiately elongate in cross-section.

Externally the frond of this species is that of a slender, unbranched
Enterornorpha^ but the structure is that of Monostroma. It has here-
tofore been found only in the Arctic region and along the Atlantic coast,
from Greenland to Nahant, Massachusetts. The specimens from
Kukak Bay have cells little more than half the size of those in Green-
land and New England specimens, otherwise there is no difference.

Ulva lactuca rigida (J. Ag.) Le Jobs.

Frond visually ovate when young, later becoming of indefinite shape,
rather firm.

Common in quiet pools, Virgin Bay, Prince William vSound (278) ;
Shumagin Islands (396).

A very common species, varying in form, texture and dimensions,
distributed all over the world.

Ulva lactuca myriotrema Le Jobs.

On rocks and algte, Sitka (133).

Frond pierced with numerous irregular holes. Rather a form than
a definite variety.
Enteromorpha linza (L.) J. Ag.

Yakutat Bay (232).

THE ALG^ 411

Frond flattened, the membranes united except for a small space at
the margin.

Enteromorpha linza. forma lanceolata J. Ag.

Yakutat Bay (232).
Margins smooth and even.

Enteromorpha linza. forma crispata J. Ag.

Sitka (156).

Margins and often the whole frond crisped and wavy.

This species connects the genera Enterojyiorpha and Ulva and has
perhaps been oftenest placed in the latter ; but its affinities seem to be
more with the present genus.

Enteromorpha intestinalis (L.) Link.

Metlakatla, Annette Island (37).

Frond simple or with a few proliferations at the base, usually en-
larged upward; cells arranged in no definite order, 6-12 ix wide, in
cross section radiate, 16-30 /-t long, the membrane about twice as thick
as the cell length in cross section.

Enteromorpha intestinalis ybr/^a cylindracea J. Ag.

In a protected sandy pool, Kukak Bay with fortna maxima (316) ;
Sand Point, Popof Island (382).

The frond is long, slender, and of nearly uniform diameter through-
out its length.

Enteromorpha intestinalis y£>r;;m maxima J. Ag.

Orca (311) ; Kukak Bay, Wx'Cn. forma cylindracea (316) ; Victoria,
British Columbia, a -form with small cells and thin membrane, ap-
proaching E. Tninima (12).

Frond much inflated, usually contorted.

Enteromorpha micrococca Kuetz.

On cliff, Orca (305) ; Shumagin Islands (398).

Frond simple, tubular, of small size, much contorted ; cells 4-5 /a
in diameter; in membrane 18—20 11 thick.

This species usually grows in dense masses on rocks between tide
marks.

Enteromorpha prolifera (Muell.) J. Ag.

Annette Island (35).

Frond more or less abundantly branched ; branches like main frond,
not much smaller ; cells arranged in longitudinal series in all but the
oldest parts of the frond.

These specimens have few branches.

412 SAUNDERS

Enteromorpha crinita (Roth) J. Ag.

Wrangell (65); Sitka (155); Prince William Sound (309).

Frond abundantly branched, usually with a main stem and virgate
branches, with cells in longitudinal series, beset with short, tapering
ramuli, the smallest of a single series of very short cells.

In specimens from this last locality the habit is that of E. intes-
tinalis forma cylindracea^ but the structure and the branching are
those of E. crinita.

Family UL O THRICHA CE^E.

Ulothrix flacca (Dillw.) Thuret.

Glacier Bay (82). Very abundant, forming a dark green coating
on rocks and pebbles on the shore, extending up to within a quarter
of a mile of the glacier.

Cells one-sixth to two-thirds as long as broad ; filaments 20-40 11 in
diameter.

Hormidium parietinum (Vauch.) Kuetz.

Yakutat Bay ; St. Paul, Kadiak.

Cells one-fourth to one diameter in length, often dividing into two or
more lateral series, which may develop into a flat membrane.

As now understood, U. parietinum includes two forms that were
long considered distinct species, and that in their fully developed con-
dition are of quite different habit, — the filiform Ulothrix parietina
and the membranaceous Prasiola crispa. In the specimens collected
at Yakutat both forms are to be found, each apparently usually occur-
ring by itself, but both sometimes together. The frond of the Pra-
siola may reach several millimeters in width. The species occurs on
damp ground, not like most other algae, submerged.

Family (EDOGONIACE^E?

Oedogonium concatenatum (Ilass.) Wittr.

Popof Island.

This species was reported by Wolle from Pennsylvania and New
Jersey ; I have also observed it in material from Maiden, Mass.

Bulbochaete brebissonii Kuetz.

In a freshwater pond near Seldovia, Cook Inlet.

This species has not been hitherto known to occur in America.
The form reported by Wolle does not belong to this species.

'This family -was determined by Dr. K. E. Him, of The Rojal University of
Finland, whose notes are here given in translation.

THE ALG^ 413

Bulbochaete intermedia De Bar.

In a freshwater pond near Seldovia, Cook Inlet.

This species according to Wolle is generally distributed throughout
the United States. Wittrock records it from north Greenland.

Bulbochaete nordstedtii Wittr.

In a freshwater pond near Seldovia, Cook Inlet.

This species has also been reported by Wittrock from Greenland.
I have also found it from Norwich, Conn.

Bulbochaete nana Wittr.

In a freshwater pond, Shumagin Islands.

Reported from north Greenland.

The form which Wolle refers to this species belongs, it seems to
me, according to his figures, to B. monile.

Bulbochaete insignis Pringsh.

Wolle reports this species as occurring in many lakes in New
Jersey.

Family CLADOPHORACE^^.

Trentepohlia iolithus (L.) Wallr.

Orca, on rocks at 1,000 feet elevation (Trelease).

Forming an orange or brick-red coating on rocks ; filaments di-
chotomously or irregularly branched; cells 14-20^1 thick in the middle,
much constricted at the ends, ij^ to 2 times as long as broad, thick-
walled. This is the Veilchenstein of the Germans. It grows on rocks,
requires little moisture and can withstand prolonged drouth. When
moistened it has a distinct violet odor.

Urospora penicilliformis (Roth) Aresch.
Phycotheca Boreali -Americana, No. 18.

Forming a light green coating on cliffs, Kukak Bay (349).

Frond 20-60 // in diameter ; cells from one-third to three diameters
in length, usually constricted at the nodes.

Chaetomorpha cannabina (Aresch.) Kjellm.

Annette Island (46), filaments light green, more curled than usual,
generally 80-110 // in diameter; Cook Inlet, near Seldovia (417), less
curled, and filaments coarser than in the previous specimens.

Filaments light green, S0-150 11 in diameter; cells i to 4 times as
long as broad.

Chaetomorpha melagonium ybrzwa rupincola (Aresch.) Kjellm.
Yakutat Bay (243).

4H

SAUNDERS

Filaments dark green, attached at base, 300-500 ,a in diameter ; cells
i^ to 3 times as long as broad.

A large and rather coarse species, abundant throughout the whole
Arctic region, and as far south as the cold currents extend in the At-
lantic and the Pacific.

Rhizoclonium riparium implexum (Dillw. ) Rosenvinge.

Yakutat Bay (193), floating in large flocculent masses.

Rather light green in color ; filaments 20-30 /z in diameter ; cells i
to 5 times as long as broad.

In these specimens the filaments reach a diameter of 30 /x ; the cells
are sometimes five times as long as broad. No rhizoidal branches were
noted, thus placing the form as variety implexum. It is difiicult to
distinguish this variety technically from species of ChcEtotnorpha., but
all forms of Rhizoclonium have a certain irregularity in form of the
cells, which is readily recognized when one becomes familiar with
these plants.

Cladophora arctica (Dillw.) Kuetz.

Glacier Bay (91) ; Ocean Cape, entrance to Yakutat Bay, on rocks
(233) ; near Sand Point, Popof Island (381).

Filaments 40-90 ix in diameter, straight and rather stiff ; branches
erect ; basal parts, especially in older plants, emitting numerous slender
rhizoidal descending filaments, by which the whole tuft is matted to-
gether.

Cladophora scopaeformis (Rupr.) Harv.

Yakutat Bay (225) ; Sitka, on exposed rocky points (185) ; Kukak
Bay (320, 327), on rocks exposed to direct washing of the waves.

Filaments 100—200 ii in diameter, straight and stiff ; branches erect,
all but the youngest parts attached to each other by short hooked
branches, forming long, simple or branching, slender tufts, from 2 to
10 mm. in diameter.

This is a characteristic species of the coast from California north-
ward, resembling a larger and coai'ser C. arctica in its later stages.

Cladophora flexuosa (Griff.) Harv.

Filaments pale green, flexuous, sparingly branched, 20—60 // in di-
ameter ; cells 2 to 3 times as long as broad ; ultimate ramuli short,
curved, usually secund.

The determination of these specimens is based on their resemblance
to No. 206. Alg. Am.-Bor. Exsicc. The species is found on both

THE ALG^ 415

sides of the Atlantic, but appears not to have been previously reported
from the Pacific.

The following specimens of Cladophora can be noted by numbers
only, specific determination being impracticable at present.

Annette Island (17, 45, 48); Wrangell (56); on rocks, Glacier
Bay {^^(i^ ; forming masses on rocks, Sitka (157) ; on protected side
of exposed rocks, Virgin Bay, Prince William Sound (295).

Family GOMONTIACB^^.

Gomontia polyrhiza (Lagerh.) Born. & Flah.

Popof Island.

Basal layer growing in the substance of marine shells, erect filaments
extending to the surface, zoospores formed in sporangia.

Family DERBESIA CE.^.

Derbesia vaucheriaeformis (Harv.) J. Ag.

On a sponge in Yakutat Bay (234).

Filaments 30-40 it in diameter, simple below, dichotomous above,
branches patent, 20-30 11 wide, often with a cuboidal cell near the
base; zoosporangia ovoid or pyriform, 140-200 by 50-80 /z, short-pedi-
celled.

As there are no mature spores on this specimen, there is a possible
doubt as to the identification, but the characters agree well with the
species named.

Derbesia marina (Lyng.) Kjellm.

Sitka, in quiet water (149).

Filaments 50-60 /;t in diameter at the thickest, tapering slightly;
sparingly branched laterally, usually two partitions found at the base
of each branch, enclosing a short cell; zoosporangia 150-180 by 90—
120 /i, short pediclled.

Family VA UCHERIA CEyE.

Vaucheria sessilis (Vauch.) DC.

On an overhanging dripping cliff, Juneau (74).

Filaments up to 70 11 in diameter ; oogonia usually two or three to-
gether, sessile, ovate or ovate-oblong, about 60-150 ,u, beaked ; anther-
idia in the vicinity of the oogonia, formed at the end of a short, hooked
or curved ramulus.

With antheridia and oosfonia.

4l6 SAUNDERS

Family CODIACE^^.
Codium adhaerens (Cabr.) Ag.

Dredged at Kadiak, at 15 meters depth (350). This species usually
grows between tide marks and its occurrence at this depth is exceptional.

Forming flat expansions, adherent by the lower surface.

Codium mucronatum californicum J. Ag.

Sitka (170).

Forming a terete, dichotomously branching, fleshy, erect thallus ;
filaments tipped with a short mucron.

Family CHARACE^^}
Chara contraria A. Braun.

In ponds and streams near Glacier Bay, very abundant (300).

Chara fragilis Desv.

In a freshwater pond, Shumagin Islands (400) .

Nitella acuminata subglomerata'A. Braun.

In a fresh water pond near Prince William Sound (300).

Nitella opaca Ag. ?

In a freshwater pond near Kadiak (419).

PHiEOPHYCEiE.
Family E CTO CAR PA CE^E.

Phycocelis baltica (Reinke) Foslie.

Forming minute tufts one mm. or less in diameter on Ralfsia
deusta. Sitka (169a).

The erect filaments are unbranched, 200-300 /x long by 5-7 ij. wide ;
cells 8-10 IX long; plurilocular sporangia 60-80 ii long, 6-9 ix wide,
borne on a 3-5-celled stalk, containing 20-30 uniseriate zoospores.

Streblonema minutissima sp. nov. (Plate xlv, fig. 3.)

Plant composed of penetrating filaments ramifying through the
cortical filaments of the host, from which arise short erect filaments
intermingled with the peripheral filaments of the host, which are
once or twice dichotomously branched bearing above a few short
branches that become transformed into uniseriate plurilocular spor-
angia 20-30 !i long by 3-5 !x wide.

'Determined by Dr. T. F. Allen.

THE ALG^ 417

In the tyfanches of Liebmanjiia sp. from Sitka (i42($).

A very minute plant the erect branches of which might easily be taken
for a part of the host plant. The penetrating filaments are 1-2 //
wide, short, sparingly branched ; cross partitions few and inconspicu-
ous, at intervals closely applied to the host cells ; no hairs or unilocu-
lar sporangia were observed.

Streblonema pacifica sp. nov. (Plate xlv, fig. \a and \b.)

Plant composed of irregular branching horizontal threads, from
which arise mostly unicellular, haustoria-like filaments which pene-
trate into the host plant, and erect filaments arising at right angles to
the horizontal ones ; cells of the horizontal filaments 4-8 /i wide, twice
as long as the diameter ; erect filaments short, unbranched or once di-
chotomous, 30-70 iJ. long, most of them bearing a narrow elliptical plu-
rilocular sporangium 5 /i and 13 /i, which contains about five uniseriate
zoospores. The plant forms circular dark brown patches 2-4 mm. in
diameter on the sporophylls of Alaria. Related to Streblonema niinu-
tulu77i of Heydrich, but larger in all its measurements.
Yakutat Bay (438).

Streblonema irregularis sp. nov. (Plate xlv, fig. 2.)

Plant consisting of irregularly branching surface filaments applied
closely to the host plant, from which arise numerous simple or spar-
ingly branched erect filaments 1-2 mm. high, 9-14 /^ wide, cells as
long to twice as long as broad ; from the surface filaments, filaments
10-14 ^'- wide, with cells a little shorter or longer than broad, penetrate
the substratum irregularly ; plurilocular sporangia linear, lanceolate or
ovate, terminal or lateral on the erect filament, 14-18 ft. wide, 55-70 //.
long ; no unilocular sporangia observed.

Forming small brown patches on the bulbs of Nereocystis friapus^
Sitka (164).

This plant is closely related to Streblonema stilophorea in its gen-
eral appearance and method of branching, but differs from it in the
shape of the sporangia and the chromatophores of the vegetative fila-
ments which are small, round and numerous in this species.

Ectocarpus tomentosus (Huds.) Lyngb.

Abundant on Fucus evanesceizs^ Sitka harbor (166), and Victoria,
British Columbia (49).

The rope-like tufts of the Alaskan specimens are fully as long as
those from the Atlantic ocean, while the specimens from the Califor-
nia coast are rarely over three mm. in length.

41 8 SAUNDERS

Ectocarpus confervoides (Roth) Le Jol.

On rocks in Yakutat Bay (226^).

This plant comes very close to the typical form of the species, it is
10 or more centimeters long, closely intertwined, the branches few,
ateral, and secund, 20-40 /jt broad at the base, the ultimate ones short
and pointed ; plurilocular sporangia ovate, sessile or short stalked,
borne laterally on the main branches, especially abundant on the short
ultimate ones, 20-30 /i, by 40— 8o/ji long.

Ectocarpus confervoides corticulatus Saunders.

Ectocarpus corticulatus SwanKKS, Phyc. Mem. 152.//, 20.

On Des77iarestia aculeata^ Popof Island (36S).

This species is the same as that described from the California coast
though the tufts are longer, and only the main filaments and the lower
part of the long primary filaments are uniformly corticated. After
examining a large amount of material of this and several other varie-
ties of E. confervoides the writer is convinced that E. corticulatus
should be considered as a variety of E. confervoides.

The plant is of a light olive green, forming flocculent tufts a few
mm. to 5 cm. or more in length, the main filament and lower part of
the primary branches densely corticated, 60-100 /x broad, ultimate
branches short, bearing numerous ovate plurilocular sporangia 16-25//
broad, and 40-70 // long.

Ectocarpus confervoides pygmaeus (Aresch.) Kjellm.

Forming a velvety covering or minute tufts on various algag, from
Pugct Sound to the Shumagin Islands. Yakutat Bay (439) ; Shumagin
Islands (386).

The plant is 2-15 mm. high, sparingly branched ; filaments 10-20 /x
broad, 2 to 3 times as long.

Pylaiella littoralis (L.) Kjellm. acuta, form. nov.

Plant 3—10 cm. or more long, loosely disposed ; main filaments 25-
40 ji wide, branches few, alternate or opposite, ultimate filaments
short, pointed, but not pilate ; unilocular sporangia in the ultimate
branches 18-24 /^ broad, 5-15 or more in a chain.

This plant agrees with the variety opposita Kjellm. in its general ap-
pearance but is smaller in all its parts and is not pilate.

Very abundant from Wrangell westward to the Aleutian Islands.
Kukak Bay (322).

The plants form large loose tufts on Fucus evanescens macrocephala
and occasionally on rocks in all quiet coves. In several instances it
was found extending up brooks much beyond the mean tide level. In

THE ALGyE 4I9

fact it seemed to be most abundant and to reach its best development
in quiet waters at the mouths of brooks where the percent of salt
must be perceptibly reduced.

Pylaiella littoralis varia (Kjellm.) Kuck.

Common on rocks, and on Fucus and other algae in quiet water from
Puget Sound to Bering Sea. Victoria, Juneau, Yakutat and Shumagin
Islands.

This form is i-io cm. in length, a light faded brown in color, very
intricate, the ultimate branches short and standing at right angles to the
axis.

Pylaiella littoralis densa Saunders.

The plant forms rope-like masses 2-4 cm. or more long on Fucus
and other alg£e. Victoria, Sitka, Prince William Sound (294), Shuma-
gin Islands (3S6).

Pylaiella littoralis macrocarpa (Foslie) Kjellm.

On fruiting tips of Fucus evanescens tnacrocefhalus ^ Victoria.

The plant is 1-3 mm. or more long, the branches and the upper
part of the main filament, except for a few-celled stalk at the base and
one of the cells at the tip, form plurilocular sporangia. No unilocular
sporangia were observed.

Family SPHA CELARIA CE^^.
Sphacelaria cirrosa (Roth) Ag.

Forming small light olive tufts on Fucus evanescens^ Annette
Island.

Sphacelaria racemosa arctica (Harv.) Reinke.

Wrangell (70) ; Yakutat Bay (195) ; Prince William Sound (283).

Forming dark olive-brown, densely tufted mats sometimes several
cm. wide and i cm. high ; unilocular sporangia abundant on all ma-
terial collected at the various stations.

Family ENCOELIACE^^.

Homeostroma undulatum J. Ag. (Plate xlvi, fig. 3.)

On Zoostera ?narina in a quiet cove near Seldovia, Cook Inlet

(412).

Fruiting plants 3 cm. long, 1-3 mm. wide.

The plurilocular sporangia project little if at all above the surface of

the plant and are massed together more than is indicated in Reinke's

figures. The single hairs are very scarce on all the Pacific coast

material.

420 SAUNDERS

Homeostroma lobatum sp. nov. (Plate xlvi, figs. 6a ^ 6^, 6c.)

Plant broadly linear, lanceolate or ovate, lo or more cm. long, 1-5

cm. wide, narrowed below to a short stipe, dark olive-brown; the

edges deeply and irregularly lobed, the lobes irregularly cut and divided ;

unilocular sporangia scattered over the whole surface of the plant.
Attached to Zostera 7narina. Sitka (114) ; Prince William Sound

(296).
Cross sections of the young plants were two cells thick, those of the

older plants 4 cells thick, the central cells slightly larger than the outer

but not at all elongated. No plurilocular sporangia were observed.

Homeostroma latifolium (Grev.) J. Ag.

Occasional in quiet sandy coves. Annette Island (39) ; Sitka (143) ;
Popof Island (367).

Punctaria plantaginea Grev.

On exposed rocks near entrance to Yakutat Bay (229).

The plant is 4-10 mm. wide, and 5-10 cm. long. Both unilocular
and plurilocular sporangia are abundant. Some of the specimens
approach Foslie's variety linearis?

Myelophycus intestinalis sp. nov. (Plate xlvii.)

Plant dark reddish bi-own, loosely caespitose, cylindrical, hollow
with age, much twisted and intestiniform, narrowed below to a dis-
tinct solid stipe, 5-12 mm. long; the inner layer of tissue composed
of long colorless cells, the intermediate layer composed of 2-4 irregu-
larly arranged rows of thick-walled cuboidal cells giving rise to broad
coarse paraphyses composed of 4-8 thick-walled cells ; sporangia very
abundant, elliptical or obovate, scattered irregularly throughout the
frond, 45-60 /Jt long, 20-30 /x wide, arising like the paraphyses from
the intermediate layer of tissue.

Attached to rocks in the sublittoral zone from Puget Sound to the
Shumagin Islands. More abundant in quiet coves. Glacier Bay
(113); Sitka (192); Yakutat (252); Popof Island (359); Puget
Sound (Gardner 215).

This plant was at first referred tentatively by the writer to Foslie's

Chordaria attenuata. Foslie states '^ however that this i^lant is a form

of Scytosiphon ; Mr. F. S. Collins has kindly examined specimens of

Foslie's plant and agrees that it is a Scyiosipkon and quite distinct

from the Pacific plant. M. intestinalis is less firm in texture and is

lighter colored than Kjellman's M. ccespitosa ; in cross section the

'Foslie, Om Nogle Arctiske Havalger.
^Nja Havalger, Vol. 13 : 97.

THE ALG^ 421

paraphyses are much broader and shorter, and the sporangia broader
than in M. ccespitosa and the central layer of tissue composed of only
2-4 rows of cells while in M. ccespitosa there are 10 or more rows.

The plant so closely resembles Scytosiphon lomentarius in color
and general appearance that it has probably been passed over by col-
lectors. It is however somewhat firmer in texture and not at all con-
stricted. Specimens collected from Puget Sound averaged much
smaller than those collected farther north. From Sitka northward the
plant is more abundant than Scyiosip/ioti lomentarius.

Scytosiphon lomentarius (Lyngb.) J. Ag.

Abundant, Puget Sound; Annette Island (40) ; Glacier Bay; Sitka
(892) ; Yakutat Bay (2290) ; Kukak Bay; Shumagin Islands (360).

Scytosiphon lomentarius complanatus Rosenv.
Juneau; Glacier Bay (98).

Scytosiphon buUosus Saunders.

On rocks in the sublittoral zone, Sitka (145) ; Cook Inlet (408).

Heretofore this species was known only from the type locality,
Monterey Bay. (Am. Alg. 251.)

Phyllitis fascia (Muell.) Kuetz.

Abundant in protected places in the littoral zone from Puget Sound
to the Shumagin Islands. Annette Island (41); Glacier Bay (94);
Yakutat Bay (220) ; Kukak Bay (319) ; Cook Inlet (409, 410).

The specimens from the northern localities are much larger than the
average plant from the California coast and the Atlantic coast of
North America. Those from Cook Inlet are 20-35 cm. long and
2-4 cm. wide.

Colpomenia sinuosa (Roth) Derb. & Sol.

On Cystophyllum lepidum, Prince William Sound (268) ; Yakutat
Bay (420).

This widely distributed and usually common species was seen but
twice during the trip. It is common in quiet coves from Puget Sound
to southern California.

Coilodesme linearis sp. nov. (Plate xlviii.)
Phycotheca Boreali-Americana No. 824.

Plant linear, tubular, olive brown, 8-20 cm. long, narrowed below
to a short stalk (1-2 mm. long) ; tissues thin and delicate, composed
of three to four layers of cells ; unilocular sporangia scattered singly
throughout the plant, 11-14// wide, 15-20/^ loi^g*

On Cystophyllum lepidufn^ Popof Island (399) ; Kukak Bay

42 2 SAUNDERS

The plant is attached to the host in great numbers. It differs from
C. californica in its shape, size and the size of the unilocular spo-
rangia. The tissue is more delicate, being composed of from two to
three layers of colorless cells, a single row of endodermal cells and
one of poorly differentiated epidermal cells.

Coilodesme californica (Rupr.) Kjellm.

Occasional on Systophyllum leptdu?ti from Puget Sound to Yakutat
Bay. Victoria; Annette Island (38) ; Wrangell (58) ; Yakutat Bay
(224).

No fruiting plants were seen, but the shape and structure of the
specimens obtained are the same as in those of the California coast. It
is much less abundant in southern Alaska than on the shores of Califor-
nia and apparently is wholly replaced further north by C. linearis.

Coilodesme bulligera Stroemf .

Abundant on rocks in the littoral zone in quiet coves, from Puget
Sound to the Aleutian Islands. Wrangell (=58) ; Yakutat Bay (222) ;
Prince William Sound (282, 312) ; Kukak Bay (345/^) ; Shumagin
Islands (390a).

Soranthera ulvoides Post & Rupr.

Saunders, Phyc. Mem. 165. PI. 29, fig. 4 and 5.
Abundant in the littoral and sublittoral zone, in rather quiet pro-
tected places, on rocks, Rhodomela larix, and Rhodufnea Jloccosa.
Victoria (2) ; Wrangell (69, 120, 162) ; Yakutat Bay.

Family DESMARESTIA CE.E.
Desmarestia viridis (Muel.) Lamour.

In the ellittoral and sublittoral zones. Glacier Bay (iio) ; Prince
William Sound (274) ; not uncommon but less abundant than the next
species.

Desmarestia aculeata (L.) Lamour.

One of the most abundant of plants in the ellittoral and sublittoral
zones from Puget Sound to the Aleutian Islands. Victoria (8) ; Sitka
(i8o>^); Wrangell; Yakutat Bay(226>^); Kukak Bay (322^^ ) ;
Shumagin Islands (369).

Family DICTTOSIPHONA CE^E.
Dictyosiphon foeniculaceus (Huds.) Grev.

A common plant on rocks, Scytosiphon and other plants from Puget
Sound to Bering Strait. Annette Island (42) ; Wrangell (59) ;

THE ALG^ 423

Juneau (72) ; Prince William Sound (285) ; Shumagin Islands (341) ;
Glacier Bay (96).

Family EL A CHISTA CE.^.
Elachista lubrica Rupr.

Phycotheca Boreali-Americana No. 828.
On Rhodymenia pahnata in the littoral zone. Wrangell (66) ;
Glacier Bay (83) (lOi) ; Prince William Sound (306) ; Yakutat Bay
(242).

Family CHORD ARIACEyE.

Myrionema strangulans Grev. (Plate xlvi, figs. 1,2.)

On Ulva lactuca. Sitka (146). Abundant also on the California
coast.

Eudesme virescens (Carm.) J. Ag.

Not uncommon on rocks and Zostera mariiza in the sublittoral and
littoral zones. Sitka (115-177); Glacier Bay (194); Shumagin
Islands (406^^ ) ; Prince William Sound (284).

The specimens from Prince William Sound are somewhat doubt-
fully referred to this species. The plant is much smaller than the other
specimens, much more branched throughout and of a light yellow color,
resembling in these respects Zanardini's E. Jlavescens . The micro-
scopic structure however is identical with that of E. virescens.

Leathesia difformis (L.) Aresch.

On rocks and algas in the littoral zone from Puget Sound to the
Shumagin Islands. Victoria (2^) ; Annette Island (32) ; Wrangell
(7o>^) ; Sitka (142) ; Yakutat Bay (254).

Mesogloia simplex sp. nov. (Plate l, figs. 3, 4.)

Plant soft mucous, olive-brown, unbranched, tubular or intestini-
form, hollow, rounded above, narrowed below to a short indistinct
stipe ; central filaments few, distinct, irregularly branched ; peripheral
filament short, simple, 2-3-4-celled ; unilocular sporangia ovate or
elliptical, pyriform, 15-25 /;l by 25-35 /j, arising from the subcortical
area.

In structure this plant very closely resembles Gobia^ agreeing closely
with Gobi's figure and description, but lacks the parenchymatous
structure of that genus. It is no doubt one of the Ckordariacece,
which is given this somewhat provisional name until the plurilocular
sporangia are found.

Attached to old worn plants of Chordaria abietina. Sitka (128).

Proc. Wash. Acad. Sci., November, 1901,

424

SAUNDERS

Small immature specimens of this plant are also found on the speci-
mens of Chordaria abietina^ from Puget Sound, distributed by Miss
J. E. Tilden (Am. Alg., No. 348).
Chordaria flagelliformis (Muell.) Ag.

Abundant in protected places in the sublittoral zone, attached to
rocks and to other algse. Sitka (122) ; Glacier Bay (96) ; YakutatBay

(193^ 231).

The plant has frequently been reported from the Atlantic shores of
Europe and America, the Baltic Sea, the Arctic Ocean, and the shores
of Kamchatka.

Chordaria abietina Rupr.

Attached to rocks in exposed places in the littoral zone. Sitka
(126) ? Prince William Sound (291) ; Shumagin Islands (380).

This species is much less abundant than on the California coast and
all specimens collected are smaller than the average plant of the species
from that region.

? Liebmannia sp. (Plate xlix.)

A single specimen of a AIesogloia-Y\\iQ. plant was collected at Sitka
(142a). It bears an abundance of unilocular but no plurilocular
sporangia, hence its exact position is in doubt. The arrangement
and structure of the axial tissue and the peripheral filaments closely re-
semble those of Liebma7inia.

Family RALFSIACE^^.

Ralfsia deusta (Ag.) J. Ag.

Sitka (169); Kukak Bay (324) Orca, Prince William Sound
(267a).

The plants were loosely attached to rocks in the sublittoral zone.
All specimens collected were sterile.

Ralfsia clavata (Carm.) Farl.

Yakutat Bay and Cook Inlet (4133^ ) .

Forming light olive-green patches 5-10 mm. in diameter, which
finally produce an indefinite coating on rocks ; in the sublittoral
region.

Family LAMINARIACEyE.

Chorda filum (L.) Lamour.

Found in the sublittoral region. Prince William Sound (273) ;
Popof Island (383).

At both stations the plant was found in quiet sandy coves in great
abundance.

THE ALGyE 425

Alaria fragilis sp. nov. (Plate liv.)

Plant of small size (4-8 dm. long) ; blade lanceolate or linear,
undulate on the margin, 1-2 dm. broad, substance thin membrana-
ceous, brittle, drying dark ; midrib quadrangular in cross section, pro-
truding equally on each margin ; stipe short, rounded at base, slightly
flattened above, 7-12 cm. long; rachis 6—15 cm. long, slightly flat-
tened but no broader than the stipe; sporophylls few (8-18), distant,
oblong elliptical, often oblique at the base, veiy obtuse above, nar-
rowed below to a stalk (5-15 mm. long) ; sporangia confined to the
basal third of the sjDorophyll.

Dr. Kjellman compai'es this plant to Harvey's specimen labeled
Alaria pylaii Grev.^ from Vancouver Island, but he agrees that
Harvey's specimen is distinct from Alaria pylaii of the Atlantic and
Polar Seas and is an undescribed species. This plant differs essen-
tially from the description of A. pylaii in having a longer stipe and
the sporophylls few and distinct.

In the sublittoral zone. Glacier Bay (So^ ) ; Prince William Sound
(257); KukakBay (333^)-

Alaria fragilis forma buUata form. nov.

With the last, Glacier Bay (79).

With this species were collected several specimens that agree with
it except that the blade is densely covered with small bullations, occa-
sionally a plant being found that had only a very few or almost no bulla-
tions. It may be a distinct species but it seems preferable to regard it
as a bullate form of the last until it can be further studied.

Alaria laticosta Kjellm. (Plate lv.)

In the sublittoral zone, in protected coves, Kukak Bay (333).

Plant of medium size, 1-2 mm. ; stipe short, round, 2-5 mm. long;
rachis long muriculate, slightly broader than the stipe ; lamina broadly
linear, dark brown, drying blackish, undulate on the margin, plicated
and fluted, 10-20 cm. broad, tapering below to the transition point,
midrib 7-12 mm. broad; sporophylls numerous, long and narrow,
rounded at end, gradually narrowed to a short indistinct stalk, fruit
usually confined to the lower two-thirds of the sporophyll, var3dng in
size, 10-30 cm. long, 1-2 cm. wide, and borne on a stalk 4 mm. long.

Dr. Kjellman referred the plants sent him to this species with some
doubt. He says : " The form, color and consistency of the blade, and
the form, width and rigidity of the sporophyll differ somewhat from
this species."

'Harvey's Notes Col. Alg. N. W. Coast, 165.

426 SAUNDERS

Alaria lanceolata Kjellm. (Plate liii.)

In the sublittoral zone. Glacier Bay (iii); Sitka Harbor (178).

The specimens obtained agree well with Kjellman's description, and
specimens submitted to him were pronounced to be this species, which
is easily recognized by the tufts of long cryptostomata which in no other
species are so large and abundant.

Alaria cordata Tilden. (Plate lvi.)

In the sublittoral zone on exposed point of an island opposite the
entrance to Yakutat Bay (230). The plant was growing in great
abundance at this station but was not seen again on the trip.

The writer's specimens are certainly identical with Miss Tilden's
plant, of which, by her kindness, he has seen both herbarium and for-
malin specimens. There is also in the writer's herbarium a young
plant of this species, from Puget Sound, collected by Mr. N. L. Gard-
ner and labelled by Dr. Setchell Alaria esculenta ?.

The stipe is of medium length (15 cm.), round, dark and firm;
rachis short and broad. The blade is oblong, lanceolate (250-450 cm.
long), somewhat undulate, plicate, light olive-green, firm ; midrib pro-
truding equally on both sides ; medulla slightly swollen near the mar-
gins; sporophylls few (7-10 on a side), arising seriately on a short
stalk, broadly linear, ovate, cuneate or somewhat cordate at the base ;
obtuse or occasionally acute above, sporangia covering most of the sur-
face, 25-40 cm. long, 3-6 cm. wide.

Alaria fistulosa Post & Rupr. (Plate Lvii.)

The first specimen of this large and interesting plant was a fragment
of the midrib washed ashore near Wrangell. At Juneau several much
worn specimens were obtained but no plants were found in situ. In
Glacier Bay it was abundant from the lower part of the sublittoral
zone to a quarter of a mile from the shore. Although some immature
plants measured i 2 feet in length the plant does not reach the size nor
is it as abundant as farther north. In Yakutat Bay, Prince William
Sound, and Cook Inlet a few fragments were found washed ashore but
no mature specimens were seen growing. This may be due to the
fact that all landings were made in protected places in the bays while
this plant loves considerable exposure. Near the moutb of Kukak
Bay there are numerous reefs 5-10 fathoms or more below the surface.
These reefs are marked by patches of this species, sometimes an acre
or more in extent. The plant not only reaches the surface but floats
for several meters on the surface.

The plant is of a very dark olive-brown color, the blade being
broadly linear, thin, papery, and smooth, 2-8 or 10 dm. broad; the

THE ALG/E 427

midrib is 15-40 or more mm. broad, the central part inflated, and di-
vided by narrow septa into air vesicles of various lengths ; near the
base of the blade the septa become indistinct and then disappear, the
inflations also disappear some distance above the transition point ; the
stipe is short (5-10 cm, long), black, solid, and rounded at the base,
flattened above and gradually passing into the rachis, which is broad
and short ; sporophylls very numerous, crowded, spatulate, elliptical
or obovate, rounded at the ends and narrowed below to a stipe of con-
siderable length (3-4 cm.) ; fruiting area covering nearly the entire
surface of the sporophyll.

No complete specimens of mature plants were measured but many
fragments were cast ashore having a blade which measured 3—7 meters
in length and 4-10 dm. in width.,

De Toni ' credits this species, on the authority of Dr. Anderson, to
California. Dr. Anderson informed the writer that he had seen no
specimens from the California coast and had no record of its occur-
rence there. He admitted that several of the I.a7ninariaccce credited
in his list to the northern coast of California had been included in the
belief that they might occvir there.

Pleurophycus gen. nov. Setchell & Saunders. (Plate lii.) Plant
attached to the substratum by hapteres, consisting of a single un-
divided blade with one central distinct midrib ; no perforations or
auricles at the base of the blade ; stipe simple ; rauciferous canals
wanting ; fruiting area confined to the midrib ; sporangia and para-
physes as in Laitiinaria,

Pleurophycus gardneri sp. nov. Setchell & Saunders. (Plate lii.)
Tilden'sAm. Alg., No. 346.

Blade broadly linear in outline, tapering below to the transition
point, 7-12 dm. and more long, 12-25 ^•''^- wide, thin and soft, stri-
ate and " lung like," wrinkled or somewhat regularly pleated near
the midrib giving it a buUate appearance which disappears near the
base of the blade; midrib broad (3-7 cm.) and flat, 2 mm. thick,
narrow above and below; stipe dark brown, drying black (3-7 dm.),
firm and solid, round below, much flattened above and gradually pass-
ing into the midrib ; sorus single, covering the upper part of the midrib.

Yakutat Bay (236) ; Puget Sound (450).

On June 26, 1S99, °^ ^^ island opposite the entrance to Yakutat Bay
the writer collected a few fragments of a plant washed up with severa

1 SjU. Alg. 3: 322.

428 SAUNDERS

species of kelp that he at once took to represent a new genus. The
locality was visited for three successive mornings during the lowest tides
and although an abundance of material was washed ashore the plant
was not found in situ. Several other points in the bay were visited
but no sign of the plant was found. If it grows off the shore on
which it was collected, and the condition of the material collected
would indicate that it does, it must grow wxll down in the elittoral
zone, for twice a careful search was made along the whole shore line
at the lowest tide, where one could get out beyond the " kelp line."
The location in which the plant was collected and the frayed and torn
condition of the ends of the blade would indicate that it grows in ex-
posed localities. In all specimens collected the stipe was broken off
apparently just above the holdfast. The only holdfast seen was on an
almost perfect specimen (from which Plate Lii was drawn) collected
by Miss J. E. Tilden in Puget Sound.

Pleurophycus has no midrib in a proper sense, but has a broad shal"
low furrow indented on one surface and prominent on the other, the
surface of which is little thicker than that of the adjacent portion of
blade, except in the region of the sorus.

This plant was first collected by Mr. N. L. Gardner in Puget Sound
in the summer of 1898 and sent to Dr. Setchell for identification.
Dr. Setchell recognized it at once as a new genus and gave it the above
manuscript name. The writer not knowing of Dr. Setchell's name
gave his specimen a provisional name, but on learning from Mr. Gard-
ner of a previous name offered his specimens for comparison to Dr.
Setchell, who at once suggested the joint authorship of the name.

Referring to the distribution of the plant Dr. Setchell writes " while
Pleurophycus may grow in the elittoral zone, all the evidence in
Gardner's and my possession shows that it extends even to the upper
sublittoral, as is the case with so many species credited to the elittoral,
Gardner found them just below low waiter mark, but in places much
exposed to the fury of the waves. Several of Gardner's specimens
have holdfasts which show several whorls of hapteres branched in a
somewhat irregularly dichotomous fashion and several times, the distal
branches being slender.

Pleurophycus stands as the simplest of the subtribe Agarece., forming
something of a transition between that subtribe and the Laminariece.

Laminaria buUata Kjellman.

In the sublittoral zone. Puget Sound ; Sitka (188) ; Prince William
Sound.

All specimens collected were quite young and sterile but agree with
Kjellman's figures and descriptions.

THE ALG^ 429

Laminaria bongardiana P. & R.

Abundant in the sublittoral zone, from Sitka to Shumagin Islands.
Sitka (186) ; Prince William Sound; Kukak Bay (337).

Laminaria solidungula J. Ag.

Occasional in the sublittoral zone from Yakutat Bay northward and
westward. Yakutat Bay (260) ; Kukak Bay (337) ; Popof Island

(387:^)-

All specimens collected were sterile and of small size; the blade

averages 6 dm. long, 3 dm. broad; the stipe is about 17 cm. broad,
firm, thick and abundantly supplied with mucous canals as are the broad
flattened rhizoids which are fused almost to the tips to form a disk-
like attachment.

Laminaria cuneifolia J. Ag.

Two specimens of this species were collected from a small rock on
Popof Island (387). The rock had apparently been washed up from
the sublittoral zone. The blade is very distinctly wedge-shaped, thin,
papyraceous, light olive-green, and very brittle in drying, quite regu-
larly wavy on the margin ; the stipe is black, very firm, rounded be-
low and slightly flattened above, 6-10 cm., the rhizoids are long and
slender, resembling those of L. saccharina.

Laminaria saccharina (L.) Lam.

Abundant in the sublittoral zone from Puget Sound to the Shumagin
Islands. Two forms were collected.

Forma (a). Blade 300-800 cm. long, cuneate at base, 20-40 cm.
wide; stipe 3-5 cm. long, texture thin, papyraceous or membranace-
ous, drying light-green; sori usually small, 1-3 dm., irregular, con-
fined to the upper part of the frond ; muciferous canals small, abun-
dant in blade and stipe.

Glacier Bay (77); Sitka; Wrangell (63); Yakutat Bay; Cook
Inlet ; Shumagin Islands.

jForma (3). Blade 50-150 cm. long, 5-10 cm. wide; stipe 1-3 cm.
long, submembranaceous to coriaceous ; sorus occurring as a band in
the central part of the blade, extending from a few centimeters to two-
thirds the length of the blade ; muciferous canals abundant in blade.

Prince William Sound (259^ ) ; Kukak Bay (337).

Hedophyllum sessile (Ag.) Setchell. — (Plate li.)

Laminaria sessilis Kg., Syst. Alg., p. 270. Tilden, Am. Alg., 344.
This species was collected in the sublittoral zone in Yakutat Bay
the latter part of July.

430

SAUNDERS

The young plants have a short distinct flat stipe i-3 cm. long and
5 or more mm. broad, the stipe soon disappears or becomes attached to
the substratum by its whole length, the old blades are a foot or two
in length, irregularly longitudinally torn above, bullate and folded in
the lower part.

Hedophyllum subsessile (Aresch.) Setchell (Mss.).

Laminaria bongardiana subsessilis Aresch., Obs. Phyc. 4 : 5.

Abundant in exposed places, from Puget Sound northward. Yaku-
tat Bay (218); Prince William Sound (259); Kukak Bay (337>^).

This plant, heretofore considered a form of L. bongai-diatia^ is made
the type of a new genus by Dr. Setchell, on account of the prostrate
Ihizome. The creeping rhizome-like affair is apparent only in old, well
developed plants. The stalk is always short, 1-4 or 5 cm., and in ma-
ture plants much flattened above, in old specimens being as much as
2 cm. or more wide. The blade is dark brown, thick and leathery,
reaching a length of 3 or more meters. It is usually split nearly to the
base into 3 somewhat equal parts each one of which is more or less
irregularly cut and torn.

Cymathere triplicata (Post & Rupr.) J. Ag.

Abundant from Puget Sound to Shumagin Islands. This species is
gregarious and usually found pretty well down in the sublittoral zone
or in the upper part of the elittoral zone. While not of extreme size
this plant forms a large part of the kelp flora in many places.

Agarum gmelini Mert. (Plate lxi.)

In the sublittoral and elittoral zones in exposed localities. This spe-
cies and Des7narestia aczileata form the majority of the elittoral flora
in many places. Prince William Sound ; Yakutat Bay ; Kukak Bay ;
Popof Island.

In mature specimens the outline is rotund or reniform and the mid-
rib broad and flat.

Three specimens from different localities gave the following meas-
urements.

Blade.

Stipe.

Midrib.

Length.

Width.

Length.

Width.

Width.

cm.
60
60
60

cm.
80
60
60

cm.

17

wanting

mm.
10
ID

mm.

12

12

THE ALG^E 43 1

Agarum turneri (Post & Rupr.).

Abundant in the sublittoral zone. Yakutat Bay (200(7) ; Prince
William Sound (261) ; Popof Island.

This species differs from the last chiefly in its smaller size, more
ovate outline, and the much narrower midrib. Its relationship would
probably be better expressed by placing it as a variety rather than a
distinct species.

Costaria turneri Grev.

Abundant in the sublittoral zone from Puget Sound to the Shumagin
Islands. Victoria; Sitka (187) ; Yakutat Bay ; Prince William Sound;
Kukak Bay ; Shumagin Islands.

The plant reaches a much larger size on the Alaska coast than on
the California and Washington coast. Specimens were seen measuring
from 180 to 220 cm. in length and 50 cm. wide at the base.

Eisenia arborea Aresch.

A broken fragment of this plant w^as obtained at Wrangell and
several specimens were obtained in Puget Sound.

Macrocystis pyrifera (Turn.) Ag. (Plate lx.)

In the elittoral zone off rocky points and in unprotected places.
Wrangell; Sitka (171, 189) ; Juneau.

Although a constant watch was kept for this plant it was not seen
north of Sitka.

Nereocystis priapus (Gmelin) Saunders. (Plates lviii, lix.) •

Ulva priapus (^WE.iAV., Hist. Fucorum, 231, 1768.
Nereocystis lutkeanusyiKKT. fil. in Linnaea, p. 48, 1829.

Abundant from Puget Sovmd to the Shumagin Islands. The plant is
seldom found growing in protected places, being confined to the elit-
toral zone at the mouth of bays and in the open ocean from a few yards
to a mile or more from shore. The plant, unlike Macrocystis pyrifera
and Alaria jisttilosa^ does not form floating masses.

Mature plants measure from 50-70 feet long, about two-thirds of
which is the long tube-like stipe with its terminal air bulb, which
floats the dichotomously torn lamina on the surface of the ocean.

Mertens' statement, copied by Harvey, in which in speaking of the
stipes, he says : " They are said to be 45 fathoms long," is probably
not true. The writer has measured many fully-developed plants on
the California, Oregon, and Washington coasts as well as on the
Alaska coast and has never yet found one exceeding the above figures.

Gmelin's figures and description of Ulva priapus leave no doubt
as to the identity of his plant although he had only an imperfect stipe.

432

SAUNDERS

As his description antedates Martens' by some sixty years it seems
best to the writer to reinstate the name given by him.

Family FUCACE^^.

Fucus evanescens fortna macrocephala KjeUm. (Plate lxii, fig. i.)

Puget Sound ; Annette Island ; Wrangell ; Juneau ; Sitka ; Glacier
Bay ; Prince William Sound ; Cook Inlet ; Kukak Bay ; Shumagin
Islands.

This is by far the most abundant seaweed on the northwest coast.
It is found in all quiet bays and protected places from Puget Sound to
Bering Sea, forming the characteristic light brown covering extending
some distance above the average tide level. In many places the plant
is not covered by salt water more than twice a month. It is able to
thrive from the moisture in the atmosphere. This species and Pylaiella
littoralls extend the farthest up the mouths of streams and fresh water
bays.

This form is more commonly evesiculose, but there are often found
indefinite vesicles just below the fruiting tip, either singly or in pairs.
It is a variation of this form from Puget Sound that Dr. Setchell has
referred to F. flaty carpus^ which is quite a distinct species.

Identified by Kjellman.

Fucus evanescens y"(9r/«a cornuta Kjellm. (Plate lxii, fig. 2.)

Juneau; Yakutat Bay (256); Prince William Sound (264a);
Kukak Bay (376) ; Popof Island.

This form was collected in more exposed places than the last and is
not nearly as common. The plant is darker colored and firmer in
texture than the last, with narrower branches and fruiting tip. The
fruiting tip is usually not at all inflated but in one extreme variation it
is inflated 5-7 cm. long.

Cystophyllum lepidum (Rupr.) Harvy.

Victoria; Wrangell; Sitka (121, 190) ; Yakutat Bay; Kukak Bay;
Shumagin Islands.

In slightly protected places in the elittoral zone from Puget Sound
to Bering Sea. Mature plants measured from i to 3 meters in length.

RHODOPHYCEiE.

Family BANGIA CE^E.

Bangia atropurpurea pacifica J. Ag.

Specimens of this species were collected near Victoria in Puget
Sound but it was not obtained in Alaskan waters.

THE ALG^ 433

Porphyra laciniata (Lightf.) Ag.^

On the stems of Fiiciis and on rocks in the littoral and sublittoral
zone, Yakutat Bay.

All specimens obtained were sterile.

Porphyra perforata J. Ag.

On rocks in the littoral and sublittoral zones, Glacier Bay (lOo);
Sitka (130) ; Shumagin Islands (394)-
The most abundant species.

Porphyra amplissima (Kjellm.) Setchell & Hus.

On rocks in the sublittoral zone, Prince William Sound. This large
and beautiful species was found growing in great abundance at this
station but was not collected again.
Porphyra miniata forma cuneiformis Setchell & Hus.

Usually found floating, occasionally attached to stems of Nereo-
cystis and other algae ; Lowe Inlet (20).

Porphyra tenuissima (Stroemf .) Setchell & Hus.

Abundant on rocks and also epiphytic on algte. Sitka (148a, 137) ;
Yakutat Bay (214).

Not previously reported from the west coast of North America.

Family HELMINTHOCLADIACE^^.
Batrachospermum vagum flagelliforme Siridot.

In ponds, pools and streams near Virgin Bay, Prince William
Sound (299) ; Cook Inlet (423) ; Kadiak Island.

Plants collected in June and July at or near sea level bore an abund-
ance of mature carpospores.

Family GELlDIACEyE.

Choreocolax polysiphoniae Reinsch.

On the stems of Polysiphonia sp., Sitka (123a).

Plants collected in June bore only tetraspores. The 'plants are
closely attached to the lower part of the main stem of the Polysi-
phonia^ spherical, dark brown or black, about J^ mm. in diameter.
The tubes of the Polysiphonia are much deformed wherever the
"parasite" is attached. The central tube is much enlarged, the sur-
rounding ones somewhat so, the walls considerably contorted and
thickened. This species has been found in the Atlantic on the Euro-
pean and American shores but has not been previously reported from
the Pacific ocean.

1 All specimens of the genus Porphyra were identified by Dr. Setchell and
Mr. Hus.

434

SAUNDERS

Family GIGARTINACEyS.

Endocladia muricata (P. & R.) J- Ag.

On perpendicular rocks in exposed places at or above the high tide
line. Wrangell (62) ; Prince William Sound (293) ; Popof Island
'370). This species, so common on the California coast, is compara-
tively rare or local on the Alaskan coast.

Iridaea membranacea J. Ag. ?

On rocks in the sublittoral zone. Sitka (125); Yakutat Bay (207) ;
Shumagin Islands (377). This is the plant distributed from Puget
Sound by J. E. Tilden as Iridcca heterocarpa (Am. Algae, No.
329). The plant is abundant and evidently ranges from Puget Sound
to Bering Sea.

To this species Dr. Farlow has ver}' questionably referred a plant
with small, once to twice dichotomously divided plant body. It is
broadly obovate or reniform in outline, tapering below to a short
stalk, entire or crenate on the margin or in the sterile plants with few
or many tooth-like proliferations. Dr. Farlow says "This plant has
made the tour of European algologists * * * Bornet is sure that it is
a form of Iridcea laminarioi'des. It also agrees with some of Bory's
specimens of that species. I must admit that having seen a series of
the two species in foreign herbaria, I found that I could not tell where
one began and the other ended."

The plant is smallei' than any of the specimens of Iridcea lamhtari-
oides seen by the writer, ranging from 5 to 10 cm. in length and the
primary division of the plant being 2-10 cm. broad. It is also some-
what thinner in texture, dark red in color, and lighter on the margins.
The cystocarps are small and evenly distributed over the entire surface
of the blade except the basal part.

Iridaea laminarioides Bory.

This plant is abundant in the sublittoral zone in Puget Sound but
was not seen in Alaskan waters.

Gigartina papillata Ag. forma typica.

Washed ashore, Shumagin Islands (357).

A few small plants 5-8 cm. in length, were collected that are iden-
tical with the forma stibsimplex of Setchell (Phycotheca Boreali-
Americana, No. 425). Dr. Farlow states that they are identical with
typical G. papillata given him by Agardh. Dr. Setchell in a note
on the distributed form says : " Under this name is included the form
figured by C. Agardh as the type."

THE ALG^ 435

Gigartina papillata forma cristata Setchell.

A single plant of this variety was collected in Kukak Bay (331a)
which is apparently identical with Dr. Setchell's cristata. It is
also very closely related to some forms of G. niamillosa.

Gigartina pacifica Kjellm.

On rocks in the sublittoral zone. Yakutat Bay (200) ; Prince
William Sound (30S) ; Shumagin Islands (358, 377).

No. 377 agrees exactly with Kjellman's figures and descriptions, the
others are somewhat intermediate between G. pacifica and G. papil-
lata.

Kalymenia calif ornica Farlow.

In tide washings, Kukak Bay (342).

The plant collected in Kukak Bay is much smaller in all of its parts
than specimens from the California coast. The proliferations are
very numerous, obovate, and only 2-5 cm. long by 1-2 cm. wide.

Ahnfeldtia plicata (Hudson) Fries.

On rocks in the sublittoral zone, Yakutat Bay (350) ; Prince Wil-
liam Sound (270).

The specimens collected at both stations were sterile and had fewer
and shorter branches than the typical forms but agree in structure.

Family RHODOPHTLLIDACE^^.

Turnerella mertensiana (P. & Rupr.) Schmitz.

Washed ashore in exposed places. Kukak Bay (352) ; Shumagin

Islands (453).

The plant is oblong or ovate, fastened to rocks by a basal disk,
entire or lobed on the margin, coriaceous, and of a deep dark blood-red
color ; several specimens were obtained 6 dec. long by 3 dec. wide.

This species was seen at only three stations in Alaska, a large spec-
imen was seen in the herbarium of Mr. N. L. Gardner, collected in
Puget Sound. There is also a small sterile specimen in the writer's
herbarium, from the central Californian coast, Monterey Bay.

Euthora cristata (L.) J. Ag.

Abundant in the sublittoral zone. Seldovia, near the entrance to
Cook Inlet (415) ; Shumagin Islands (356).

One of the most abundant of the red algae. It is also reported from
Puget Sound.

436 SAUNDERS

Family RHODTMENIACE^^.
Rhodymenia pertusa (P. & Rupr.) J. Ag.

On rocks in the lower part of the sublittoral zone, Yakutat Bay
(205) ; Prince William Sound (267 and 271).

No. 271 is a smaller form and not at all pertuse.

Rhodymenia palmata (L.) J. Ag.

On rocks in the upper part of the sublittoral zone. Glacier Bay
(88) ; Yakutat Bay (206) ; Prince William Sound, near Orca (310) ;
Kukak Bay (345) ; Popof Island (378).

The entire form of this species seems to be the more common on the
Pacific Coast, though the palmately divided form and the form with
numerous proliferations are also abundant.

The plant is gathered in large quantities, dried and eaten by the In-
dians of the northwest coast, as is Porphyra pertusa by the China-
men on the California coast.

Plocamium coccineum uncinatum J. Ag.

In the sublittoral zone, Sitka (148).

Halosaccion firmum (P. & R.) J. Ag.

In the sublittoral zone in a quiet muddy cove. Cook Inlet (414).
This species was collected at only the one station, but it was abundant
there.

Halosaccion fucicola (Post & Rupr.) J. Ag.

On rocks, Rhodo77iela larix, and other algae in the littoral zone
from Puget Sound to Bering Sea. Victoria (5) ; Annette Island
(33) ; Sitka (145a) ; Prince William Sound; Popof Island.

Halosaccion ramentaceum (L.) J. Ag.

In the sublittoral and littoral zones. Kukak Bay (331, 317) ; Cook
Inlet (414).

Nos. 317 and 414 agree with the typical form of this variable spe-
cies. No. 331 is very close to the forma densa of Kjellm. It was found
well up in the littoral zone while the other forms are sublittoral. At
each station this species was collected only once.

Halosaccion tilesii (Ag.) Kjellm.

In the littoral zone from Wrangell to Bering Sea, Wrangell (54) ;
Yakutat Bay (235, 248, 249) ; Prince William Sound (306a ); Ku-
kak Bay (453).

Halosaccion microsporum Rupr.

In the littoral zone. Glacier Bay (85, 109); Yakutat Bay (452).

THE ALG^E 437

The three numbers, referred to this species somewhat tentatively,
represent three very variable and diverse forms w^hich however agree
in structure.

Family DELESSERIACEy^.
Nitophyllum ruthenicum (Post& Rupr.) Kjellm.'

A single sterile specimen of this plant was collected in the sublit-
toral zone at Sitka (119).

The plant is 15 cm. high bearing many cuneate branches which are
delicately longitudinally striate, especially near the base. It is quite
distinct from any of the Californian species of Nitophyllum^ but is re-
lated to A^. latissimum. The nerves are much more delicate and not
branched and disappearing above the middle of the lobes ; in this re-
spect it is intermediate between A^. latissimum and N. fryeanum.

Delesseria baerii (Post & Rupr.) J. Ag.

Two small sterile plants of this species were collected in the sublit-
toral zone near Sitka (183).

Delesseria alata (Huds.) J. Ag.

A few sterile plants of this species were collected in Puget Sound
at Victoria (3). It was not seen in Alaskan waters.

Delesseria sinuosa (Good. & Wood.) Lamon.

Abundant in the sublittoral zone. Sitka (151); Prince William
Sound (265) ; Kukak Bay (344, 323) ; Shumagin Islands (371, 389).

A very variable species, some of the forms resembling D. querci-
folia but with a more distinct midrib and opposite nerves. Speci-
mens from Kukak Bay bore an abundance of cystocarps. The
species occurs in the Arctic and North Atlantic oceans and has re-
cently been reported by Kjellman from Bering Sea. From Sitka to
the Shumagin Islands it is the most common Delesseria.

Delesseria crassifolia Rupr.

A specimen of this plant was collected by Prof. Trevor Kincaid in
1898 on the shores of St. Paul Island, It was not collected by the
Harriman Expedition in Bering Sea.

Delesseria decipiens J. Ag.

In the sublittoral zone, Prince William Sound (290).

An abundance of cystocarpic material was collected. The plants
are much smaller (only 8-15 cm. in length) than those from the Cali-
fornia coast, but agree in other respects.

> Kjellman, Om Behringshaf. Algfl., 25.

438 SAUNDERS

Delesseria serrata Post & Rupr,

To this species is referred a plant with the general appearance of a
broad form of D. alata, but the branches are more distant and fewer
and the margins of the upper and younger branches are regularly or
irregularly serrate. It may be an extreme form of D. alata but it
seems best to keep it under a separate name until this is proven.

Family RHODOMELACE^^.

Rhodomela floccosa Ag.

In the littoral and sublittoral zones from Puget Sound to Bering
Sea. Annette Island (30, 56) ; Wrangell (68) ; Sitka (139) ; Yakutat
Bay (191* 203) ; Kukak Bay (321) ; Popof Island (374).

One of the most abundant plants. It occurs also on the Washing-
ton, Oregon, and California coasts.

Rhodomela larix Ag.

Attached to rocks in the upper sublittoral and littoral zones ; abun-
dant from Puget Sound to the Shumagin Islands. Sitka (131) ; Ku-
kak Bay (31S).

Odonthalia aleutica (Mertens) J. Ag.

Attached to rocks in the sublittoral zone, Shumagin Islands (385).

The same species has been collected in Puget Sound by the writer.
The branches are narrower than in O. dentata^ the perithecia are ra-
cemed and very slightly oblong-urceolate, and the plant turns black in
drying.

Odonthalia kamtschatica (Rupr.) J. Ag.

On rocks in the sublittoral zone. Kukak Bay (341) ; Yakutat Bay
(219).

In this species the plant is smaller than the last species and remains
a dark blood-red color in drying : the perithecia are short, racemed and
very strongly urceolate.

Polysiphonia arctica J. Ag.

In the sublittoral zone, Shumagin Islands (364, 366).

Polysiphonia bipinnata Post & Rupr.

Abundant on rocks and on other algae in the sublittoral zone, from
Puget Sound northward. Annette Island (19); Glacier Bay (84);
Sitka (117) ; Yakutat Bay (244).

Several specimens of Polysiphonia are undeterminable and for the
present can be noted only by numbers as follows; 25,64, 123, 150,
154, 212, 363.

THE ALG^ 439

Family CERAMIA CE^S.

Calithamnion floccosum pacificum Harv.

On Nereocystis friapus and other algs, and on rocks. Puget
Sound; Lowe Inlet (15) ; Yakutat Bay (210, 244, 199).

The most abundant species of Caliihamnion on the Northwest
coast.

Calithamnion plumula Lyngb.

This species was collected several times in Puget Sound but was not
seen in Alaska waters. It is a comparatively common plant in Puget

Sound, while on the central California coast it is very rarely found.

,•

Ceratothamnion pekeanum (Harv.) J. Ag.

On rocks in exposed localities in the littoral and upper sublittoral
zones. Sitka (152) ; Yakutat Bay (211) ; Shumagin Islands (373).

The northern plant is smaller and with shorter branches than the
same species from the California coast.

Ptilota asplenoides Ag.

Abundant on rocks in the sublittoral zone. Yakutat Bay (198, 219) ;
Prince William Sound (269).

Ptilota plumosa Ag.

Attached to rocks and large algae in the sublittoral zone. Sitka
(140); Shumagin Islands (397).

A much finer and more delicately branched plant than the last, with
opposite similar branches.

Ptilota serrata Kuetz.

In the sublittoral zone. Shumagin Islands (355).

Antithamnion boreale (Gobi) Kjellm.

Occasional in the sublittoral region. Wrangell (55) ; Sitka (135).

In its general appearance the plant agrees withyl. boreale corallina
but the branches near the apex are not elongated as in that form.

Ceramium rubrum Ag.

In the sublittoral zone. Sitka (127) ; Prince William Sound (275) .

The Prince William Sound material, collected in June, bears an
abundance of mature carpospores.

Ceramium codicola J. Ag.

Attached to Codium niucronatum calif ornicum^ Sitka (170).

The plants are small and sterile, but agree with specimens of C.
codicola collected by the writer on the California coast, and with speci-
mens distributed in the Phycotheca Boreali- Americana.
Proc. Wash. Acad. Sci., November, 1901.

440

SAUNDERS

Microcladia borealis Rupr.

In the littoral zone, attached to rocks in exposed places, Paget
Sound (4).

The plant \vas not collected in Alaskan waters.

Rhodochorton rothii (Turton) Naegeli.

Forming a dense coating on rocks, especially in caverns in the upper
sublittoral or littoral zones. Prince William Sound (292).

The writer has collected this plant also on the central California
coast and in Puget Sound.

Family GLOIOSIPHONIACEy^.

Gloiopeltis furcata (P. & R.) J. Ag.

Attached to rocks in the sublittoral zone from Puget Sound to the
Shumagin Islands. Annette Island (26) ; Sitka (124) ; Vakutat Bay
(201) ; Shumagin Islands (372).

Not common at any of the stations.

Gloiosiphonia calif ornica (Farl.) J. Ag.

In the sublittoral region in a protected cove, Prince William Sound

(277)-

This species was originally described by Dr. Farlow from the cen-
tral California coast. He placed it somewhat questionably in the genus
Nemasto77ia on account of having only dry material for examination.
J. Agardh, in Till Algernes Systematik, transferred the species to
Gloiosiphonia. Dr. Farlow recently stated that since the publication
of the species he had examined both living and alcoholic material and
believed that Agardh was right in placing it in Gloiosiphonia.

Family GRA TL O UPIA CE^E.

Cryptonemia obovata ? J. Ag.

Washed ashore, Prince William Sound (276).

Two sterile specimens of a Cryptonemia are referred somewhat
questionably to this species. They agree in shape and structure with
fruiting specimens collected in Puget Sound which undoubtedly be-
long to this species.

Family D UMONTIA CE^^.

Dumontia filiformis (Lyngb.) Ag.

Abundant in the sublittoral zone in protected places from Prince
William Sound northward and westward. Prince William Sound
(307, 280); Cook Inlet (411); Kukak Bay (340); Shumagin
Islands (365).

THE ALG^ 441

The specimens fi'om Cook Inlet bear an abundance of cystocarps ;
the others have tetraspores.

Cryptosiphonia grayana J. Ag.

In the sublittoral zone. Wrangell (162); Sitka (132); Yakutat
Bay (208) ; Kukak Bay (330).

The Yakutat and Kvikak Bay specimens bear cystocarps, the Sitka
and Wrangell material tetraspores.

Dilsea calif ornica (J. Ag.) Schmitz.

In the sublittoral zone, Orca, Prince William Sound (313).

The large sterile plants (3-5 dm. long by 1-2 dm. broad) were
submitted to Dr. Setchell and were referred by him to this species.

Dilsea arctica (Kjellm.)

Attached to rocks in the sublittoral zone, Shumagin Islands (353).

Young plants are broadly ovate or reniform, and of a light rose-red
color, older specimens are once to twice longitudinally divided nearly
to the base and of a light faded red color.

Constantinea rosa-marina (Gmel.) J. Ag.

In the upper part of the elittoral zone. Prince William Sound
(298).

It was observed at several other stations, including the Shumagin
Islands, but no collections were made. It occurs in both protected and
exposed situations, but seems to be more abundant in exposed locali-
ties. Its occurrence in Puget Sound would indicate a general distri-
bution from that locality to Bering Sea.

Family S^ UAMARIA CE^^.
Hildenbrandtia rosea Kuetz.

Forming a very delicate rose-red crust on rocks in the littoral and
sublittoral zones, from Puget Sound to the Shumagin Islands.

The same species is reported by Kjellman from Bering Sea. The
writer has also found it common on the central California coast. All
specimens were sterile.

Family CORALLINACE^^}
Melobesia patens.

On A??7phiroa epiphlegtnoides^ Prince William Sound (423a).
Dr. Foslie is somewhat in doubt as to the species, labeling speci-
mens submitted to him, " M. patens or n. sp."

' Specimens of all Corallinaceae enumerated were sent to Dr. Kjellman who
requested Dr. Foslie to work them over. The determinations are entirely those
of Dr. Foslie.

442 SAUNDERS

Lithophyllum farlowii Foslie.

On rocks in the littoral zone, Sitka (13S).
• The young plants were somewhat doubtfully referred to the species
by Dr. Foslie.

Lithothamnion compactum Kjellm.

Forming an indefinite coating on rocks in the sublittoral zone,
Prince William Sound (431).

Lithothamnion glaciale Kjellm.

In the upper elittoral and lower sublittoral zones. Prince William
Sound (430) ; Kukak Bay (427) ; Shumagin Islands (426). Abund-
ant in exposed localities.

Lithothamnion laeve (Stroemf.) Foslie.

Abundant on rocks in exposed localities in the sublittoral zone,
Kukak Bay (425a), with Clathromorphuni circumscriptum.

Clathromorphum circumscriptum (Stroemf.) Foslie.

On rocks in the sublittoral and elittoral zones. Kukak Bay (425) ;
Sitka (441).

Lepidomorphum yendori Foslie.

On rocks in the sublittoral zone, Sitka (438).

Amphiroa tuberculosa Rupr.

On rocks in the lower sublittoral zone, Sitka (432).

Amphiroa epiphlegmoides J. Ag.

On rocks and other algae. Sitka (433) ; Prince William Sound (423) .

Amphiroa planiuscula (Kuetz.) Foslie.

' On mussel shells, rocks, etc., in the lower sublittoral zone. Yakutat
Bay (255) ; Prince William Sound (429).

Corallina arbuscula Rupr.

On rocks in the sublittoral zone. Shumagin Islands (438).

Corallina pilulifera filiformis Rupr.

On rocks in the sublittoral zone. Prince William Sound (435).

BACILLARIACEiE.^
Melosira granulata C. G. E.

In a freshwater pond, Popof Island, near Kukak Bay ; Kadiak
Island ; Cook Inlet.

^ Determined bj Prof. A. M. Edwards.

THE ALG^ 443

Melosira nummuloides B. de st. V.

In salt water near Juneau ; Glacier Bay.

Melosira sol C. G. E.

Attached to seaweeds from Annette Island.

Melosira sulcata C. G. E.

Attached to seaweeds from Annette Island.

Trochiscia moniliformis F. C. M.

Attached to seaweeds from Annette Island ; Glacier Bay.

Coscinodiscus argus C. G. E.

In salt water material from Annette Island ; Yakutat Bay.

Coscinodiscus lineatus C. G. E.
In salt water from Glacier Bay.

Thalassosira cleve H. H. G.

In salt water from Glacier Bay.

Triceratium wilkesii J. W. B.
In salt water near Wrangell.

Chaetoceros hispidum C. G. E.

One specimen w^as found in material from Annette Island.

Biddulphia aurita L. W. D.

Attached to seaweeds from Annette Island ; Popof|Island.

Isthmia obliquata J. E. S.

In salt water from Yakutat Bay ; Victoria, British Columbia.

Rhabdonema biquadratum J. B.

Material from Annette Island.

Rhabdonema fauriae P. P.

Material from Annette Island.

Rhabdonema japonica T. & B.

Material from Annette Island.

Rhabdonema elegans J. B.
Material from Annette Island.

Rhabdonema striatulum J. E. S.

In salt water material from Annette Island ; Glacier Bay ; Popof
Island ; Yakutat Bay.

Rhabdonema arcuata F. T. K.
In salt water, Yakutat Bay.

444 SAUNDERS

Tabellaria flocculosa F. T. K.

In a freshwater pond, Popof Island ; Kadiak Island ; Kukak Bay.

Grammatophora marina (Lyngb.) Kuetz.
In salt water from Yakutat Bay.

Licmophora granulata V.

In salt water from Glacier Bay.

Licmophora cuneata F. C. L.

Attached to seaweeds, Glacier Ba}'.

Licmophora pennatula V.

In salt water near Juneau ; Annette Island ; Yakutat Bay ; Wran-
gell ; Popof Island.

Meridion circulare C. A. A.

In a freshwater pond, Popof Island.

Diatoma hyemale H. C. L.

In a freshwater pond, Popof Island.

Diatoma pectinale O. F. M.

In a freshwater pond, Popof Island.

Fragilaria construens C. G. E.

In a freshwater pond, Popof Island.

Fragilaria virescens J. R.

In a freshwater pond, Popof Island.

Fragilaria striatula H. C. L.

In material from salt water. Sitka ; near Juneau ; Popof Island.

Fragilaria exilis A. G.

In salt water, Yakutat Bay.

Synedra crotonensis A. M. E.

In a freshwater pond, Popof Island ; Cook Inlet.

Synedra ulna C. L. N.

In a freshwater pond, Popof Island ;]^Kadiak Island.

Synedra prolongata A. S.

In a fresh water pond, Cook Inlet.

Synedra fasciculata C. A. A.

Attached to seaweeds from Annette Island ; Glacier Bay ; Yakutat
Bay.

THE ALG^ 445

Eunotia robusta J. R.

In a freshwater pond, Kadiak Island ; Cook Inlet ; Kukak Bay.

Eunotia gracilis C. G. E.

In a freshwater pond, Popof Island.

Eunotia lunaris C. G. E.

In a freshwater pond, Popof Island ; Kadiak Island.

Cocconeis placentula C. G. E.

In a freshwater pond, Kadiak Island.

Cocconeis scutellum C. G. E.

In salt water from Victoria, British Columbia ; attached to seaweeds
from Annette Island ; freshwater pond, Kadiak j salt water, Yakutat
Bay ; Juneau ; Popof Island.

Achnanthes subsessilis C. G. E.

In salt water from Glacier Bay.

Achnantlies lanceolata A. B.

In a freshwater pond, Popof Island ; near Kukak Bay.

Achnanthes glabrata A. G.

In salt water, Yakutat Bay.

Cocconema lanceolatum C. G. E.

Material from salt water, Annette Island ; in a freshwater pond,
Popof Island ; Kadiak Island ; Cook Inlet.

A freshwater species, the dead shells of which had been carried
into the ocean by streams or glaciers.

Navicula silicula C. G. E.

In a freshwater pond, Popof Island.

Navicula cyprinus C. G. E.

In a freshwater pond, Popof Island.

Navicula elliptica F. T. K.

In a freshwater pond, Popof Island ; attached to seaweeds from
Popof Island.

Navicula major F. T. K.
In a freshwater pond, Popof Island ; Annette Island ; Kukak Bay.

Navicula radiosa F. T. K.

In a freshwater pond, Popof Island.

446

SAUNDERS

Navicula legumen C. G. E.

In salt water material from Annette Island.

A freshwater species, the dead shells of which had been carried
into the ocean by streams or glaciers.

Navicula pupula C. A. A.

In salt water material from Annette Island.

A freshwater species, the dead shells of which had been carried
into the ocean by streams or glaciers.

Navicula apis C. G. E.

Material from Annette Island ; Popof Island.

Vanheurckia rhomboides A. B.

In a freshwater pond near Kukak Bay.

Pleurosigma angulata W. S.
In salt water, Yakutat Bay.

Pleurosigma attenuata F. T. K.

In a freshwater pond, Kadiak Island.
Pleurosigma fasciola C. G. E.

In salt water near Wrangell.

Nitzschia diadema F. T. K.
In salt water, Yakutat Ba3\

Nitzschia angularis W. S.

In salt water, Yakutat Bay.

Nitzschia closterium C. G. E.

In salt water material, Annette Island.

Nitzschia sigmoidea C. L. N.

In salt water material from Annette Island.

A freshwater species, the dead shells of which had been carried into
the ocean by streams or glaciers.

Nitzschia vermicularis F. T. K.

In a freshwater pond, Popof Island.
Surirella elegans C. G. E.

In a freshwater pond, Popof Island.

Staureoneis phcenicenteron C. L. N.

Material from Annette Island; freshwater pond, Kadiak Island;
Kukak Bay.

A freshwater species, the dead shells of which had been carried into
the ocean by streams or glaciers.

THE ALG^. 447

Amphora ovalis C. G. E.

In salt water from Glacier Bay.

Amphora elliptica C. A. A.
In salt water, Yakutat Bay.

Gomphonema geminatum C. G. E.

In a freshwater pond, Kadiak Island.

Gomphonema affine F. T. K.

In a freshwater pond, Kadiak Island.

Gomphonema subtile C. G. E.

In a freshwater pond, Popof Island.

Rhoicosphenia curvata F. T. K.

In salt water material from Annette Island ; Yakutat Bay ; Popof
Island.

Cymbella inaequalis C. G. E.

From salt "water material from Annette Island.

A freshwater species, the dead shells of which had been carried
into the ocean by streams or glaciers.

Cymbella ehrenbergii F. T. K.

In a freshwater pond, Popof Island ; Kadiak Island.

Epithemia gibba C. G. E.

In a freshwater pond, Popof Island ; Kadiak Island.

Epithemia jurgensii C. A. A.

In a freshwater pond, Popof Island; Cook Inlet; Annette Island.

Epithemia westermannii C. G. E.

In a freshwater pond, Popof Island.

PLATE XLIII.^

Fig. I. Cosmarium conspersum Ralfs.

2. " ccelatum Ralfs.

3. " ornatum Ralfs.

4. Euastriim crassum (Bieb.) Kuetz.

5. Cosmarium fseudotaxichondricm Nordst.

6. " parvulnm Breb.
7- " hammeri Reinsch.

8. " fi'ranatuvt Br^b.

9. Euastrum verrticosum Ehrenb.

10. Cosmarium ochtodes Nordst.

11. " (5t>//-_y//5 Menegh.

12. " portianum nephroideum Wittr. Front view.

13. " " " " End view.

14. Closterium parvulum Naeg.

15. " venus Kuetz.

16. Staurastrum echinatum Pec ten (Perty) Rab.

17. Cosmarium dcpressum (Naeg.) Lund.

18. " pulckerrimum Nordst.

19. " " " From Kukak Bay.

20. " subcrcnatum Ilantzsch.

21. " pseudogranatum Nordst.

22. " quadrifarium Lund.

23. Staurastrujn polytnorphutn Breb. End view.

24. " " " Front view.

25. Micrasterias rotata (Grev.) Ralfs.

26. Cosmarium broomei \iir. Thwaites.

27. Closterium acerosum (Schrank), Ehrenb.

28. Cosmarium bioculatum Breb.

29. MesotiV)iium braunii De Bary.

30. Disphinctium connatum (Breb.) De Bary.

31. Staurastrum calyxoides Wolle.

32. Euastrum affine Ralfs.

33. Cosmarium constrictum Delp. Side view.

34. " " " Front view.

35. " (^roo;;??/ Thwaites, End view.

36. " " " Front view.

37. Euastrum oblongum (Grev.) Ralfs.

38. Xanthidium antilop(eu7n (Breb.) Kuetz.

39. Cosmarium sexangulare Lund.

40. " pyramidatum Breb.

' Plates xliii-lxii are from drawings by Eva M. Saunders, except for plates
xlvi, figs. I and 2, which were drawn by Edna L. Hyatt.

(448)

Proc. Wash. Acad. Sci. Vol. 111.

Plate XLIII.

HBJOTYPE PRINTING OOi

ALASKA ALGAE.

PLATE XLIV.

Fig. I. Cosmayium intermedium Delp.

2. Euastrum elegans (Breb.) Kuetz.

3. Penium digitus (Ehrenb.) Breb.

4. Cosmarium latum Breb.

5. Staurastrum dejcctum Br^b. a, b and r.

6. Cosmarium sphalerostichum Nordst.

7. Micrastcrias truncata (Corda) Ralfs.

8. Cosmarium phaseolus Breb.

9. Cosmarium undulatum Corda.

11. Micrastcrias denticulata (Brdb.) Ralfs.

12. Cosmarium ftachydcrmum Lund.

13. Tetmcmorus brebissonii {^lenegh.) Ralfs.

14. Arthrodesmus cotivergeiis Ehrenb.

15. Cosmariu77i blyttii Wille.

16. Cos7narium contractum Kirch.

17. " kitcheliiW oWe.

18. Xanthidium armatum (Breb.) Ralfs.

19. Cosmarium inetiegkinii brautiii {'R.e'wxsch.) Ilansg.

20. Pleurotictiiopsis debaryi (Archer) Lund.

21. Cosmarium tumidum Lund.

22. " A-rt//5/V Brdb.

23. Micrasterias kitcheliiV\'o\\c.

24. Euastrutn didelta (Turp.) Ralfs.

25. " e/e^a«5 (Breb.) Kuetz.

26. " " '■ Two forms.

27. Micrasterias oscitans j)innatijida {Y^ueiz.^ Rabenh.

28. Cosmarium holmiense Lund.

29. Staurastrum ravenelii 'Wood, a, front view : ^, end view.

30. Euastrum elegatts, large form.

31. Micrasterias Jimbriata elephanta Wolle.

(450)

Proc. Wash. Acad. Sci. Vol. III.

Plate XLIV

HEUOTTPE PRINTIMG 00.

ALASKA ALGAE.

PLATE XLV.

Fig. I. a and b. S/reblonema pacijica sp. nov.
2. Streblonema irregularis sp. nov.
3- " minutissima sp. nov.

(452)

Proc. Wash. Acad. Sci. Vol. 111.

Plate XLV.

HEUOTYPE PRiraiNG (H

ALASKA ALGAE.

PLATE XL VI.

Fig. I. Myrioticma scrangulans Grc\.] tufts natural size.
-• ■' " " section (X 400).

3. Homeostroma tuidulatum J. Ag.

4. Dermocarpa fucicola %-^. ViO\.\ section (X 400).
5- " " " natural size.
6«. Homeostroma lobatum sp. nov. ; natural size.

^*^- " " " portion of edge (X 30).

6<^- " " " section (X 400).

(454)

Proc. Wash. Acad. Sci. Vol. III.

Plate: XLVI.

HFilOTYPE PRINTING 00.

ALASKA ALGAE.

Proc. Wash. Acad. Sci., Nov., 1901.

PLATE XLVII.

Fig. I. Myelophycus intestinalis sp. nov. ; natural size.

2. '• " " cross section (X 400).

3. " " " longitudinal section (X 400)-

(456)

Proc. Wash. Acad. Sci. Vol. III.

Plate XLVII

HEUOTYPE PRrKTIKG CD.

ALASKA ALGAE.

PLATE XL VIII.

Fig. I. Coilodesme linearis sp. nov. : natural size.

2. " " " cross section (X 400)-

(4.SS)

Proc. Wash. Acad. Sci. Vol. III.

Plate XLVIII.

HBJOTYPE PRINTING CD.

ALASKA ALGAE.

PLATE XLIX.

Fig. I. Liebmannia ?; natural size.

2. " peripheral filaments (X 400)-

(460)

Proc. Wash. Acad. Sci. Vol. Ill

Plate XLIX.

MEUOTYPE PRIKTING Oa

ALASKA ALGAE.

PLATE L.

Fig. I. Laminaria cu7ieifolia J. Ag.

2. Mesogloia simplex sp. no v.

3. " " section (X 100).

4. •' " (X400)-

(462)

Proc. Wash. Acad. Sci. Vol. 111.

Plate L.

HEUOTYPE PRIKTIHG Oa

ALASKA ALGAE.

PLATE LI.

HedopJiyllum sesst'le (Ag.) Setchell.

(464)

Proc. Wash. Acad. Sci. Vol. 111.

Plate li.

HtLiOrrPE "KiMiro to.

ALASV.A ALGAE.

PLATE LII.

Fig. I. Pleurophycus gardiieri %<tx\. <i\. sp. nov. Selchell and Saunders : % nat-
ural size.

2. " " young plant.

3. " " cross section of blade (X400).

4. " " longitudinal section of blade, showing fruit.

5. " " a, section of stipe near the transition point
natural size ; b, section of stipe near base, natural size ; c, portion of
same (X Soj

(466)

Proc. Wash. Acad. Sci. Vol. II'.

Plate lii

HEUOrrPE POINTING oa

ALA5KA ALGAE.

PLATE LIIL

Fig. I. Alaria lanccolata, Kjellm. ; ^2 natural size.

2. " " " section of blade, natural size.

3,4. •' " " young plants.

(46S)

Proc. Wash Acad. Sci. Vol. ill.

Plate lI!

HBJOTYPE PftlNTIKt CT

ALASKA ALGAE.

PLATE LIV.

Alaria fi-agilis sp. nov.

(470)

Proc. Wash. Acad. Sci. Vol. III.

Plate liv.

,H,.>'

iiwy "n

ALASKA ALGAE.

MEUOTYPE PRUOIHS OX

Proc. Wash. Acad. Sci., Nov., 1901.

PLATE LV

Alaria laticosta Kjellman.

:472

Proc. Wash. Acad. Sci. Vol. 111.

P'-ATE LV.

HEUOTYPE PRINTIHG 00.

ALASKA ALGAE.

PLATE LVl.

Alatia cor data Tilden.

(^74

Proc. Wash. Acad. Sci. Vol. 111.

Plate lvi.

MEl-10'nPEP'ilN'l1&'

ALASKA ALGAE.

PLATE LVII.

Fig. I. Alaria fistulosa Post & Riipr. ; % natural size.

2. " " " " section of midrib : natural size.

3,4. •' " " " portions of the midrib (X 80).

(476)

Proc. Wash. Acad. Sci. Vol. III.

Plate lvii.

HELIOtTPr OSINTIN& CO.

ALASKA ALGAE.

PLATE LVIIL

Fig. I. IVereocys/ts pr/a/>us (Gmel.) Saunders ; much reduced.

2. " " " section of blade showing fruit.

3. '• " " longitudinal section ^^houing mucous canals •

(X40)-

4. Nereocystis friapus (Gmel.) Saunders (X4o)'

5. " " " cross section of stipe showing mucous
canals and secreting cells.

(478)

Proc. Wash. Acad. Sci. Vol III

Plate LVlll.

HEUOTYPE PRINTING CD.

ALASKA ALGAE.

PLATE LIX.

Figs. i-8. Nereocystis priapus, j-oung plants in various stages of development.
9. " " An abnormal plant.

(480)

Proc. Wash. Acad. Sci. Vol. 111.

Plate LIX.

MEUOTTPE miNTIKS 00.

ALASKA ALGAE.

PLATE r.X.

P'iG. I. Afacrocystis pyrifera (Tv\vr\Qr) i\g.

f482)

Proc. Wash. Acad. Sci. Vol. III.

Plate LX.

HEUOTYPE PRINTIKG OX

ALASKA ALGAE.

PLATE LXL

Agarum gmeliin Mert.

(484)

Proc. Wash. Acad. Sci. Vol. III.

Plate LXI,

HEUOTYPE PRItaiNG (XX

ALASKA ALGAE.

PLATE LXII.

Fic;. I. Fiicus evartescens Ag. ; forma macrocephala Kjellm.
2. " " forma cornuta Kjellm.

(486;

Proc. Wash. Acad. Sci. Vol. III.

PL.ATE LXII.

HQJOTYPE PRINTING 00.

ALASKA ALGAE.

PROCEEDINQS

WASHINGTON ACADEMY OF SCIENCES

Vol. Ill, pp. 487-506. November 7, 1901,

PAPERS FROM THE HOPKINS STANFORD GALA-
PAGOS EXPEDITION, 1898-1899.

IV.

ENTOMOLOGICAL RESULTS (4):

ORTHOPTERA.

[Text Figures 35-44.]
By Jerome McNeill.

CONTENTS :

Introduction 487

Technical descriptions.

Forficulidse 492

Blattidae 493

Mantidae 494

Acrididse 495

Locustidae 496

Grillidae 502

INTRODUCTION.

In his paper on the Orthoptera of the Galapagos Islands^
Scudder enumerates eight expeditions which have made collec-
tions there. These are Darwin in the Beagle^ 1835 \ the Swed-
ish frigate Etigenie^ 1852 ; Louis Agassiz, in the Hassler^ 1872 ;
Commander Cookson, in the Petrel, 1875 ; Wolf, State Geolo-

>Bull. Mus. Com. Zool., Vol. xxv, No. i, pp. 1-24, PI. i-iii, Cambridge, Sept.,
1893.

Proc. Wash. Acad. Sci., November, 1901. (4S7)

488 McNeill

gist of Ecuador, 1875; \h.e. Albatross, 1888; Alex. Agassiz, in
the Albatross, 1891 ; and G. Baur for Clark University, 1891.

Scudder's paper is based on the specimens obtained by all the
expeditions except the Beagle, Eugenie and Petrel, and he had
seen specimens of all the species enumerated by him, except
one, which is of uncertain occurrence. Butler referred two larval
mantids to the genus Vates and Scudder concludes that it was
impossible to have referred the specimens he knew to that genus.

Twenty species of Orthoptera are given in this paper, these
species representing all the families except Phasmidee. All the
Blattids, however, five in number, are cosmopolitan species. Of
the remaining ffteen two are of such doubtful occurrence that
they may be excluded, leaving but thirteen indigenous species.
Moreover, since the forficulid, Anisolabis atmulipes Lucas, must
be regarded as an introduced species, this number must be still
further reduced to twelve. These twelve are included in eleven
genera, representing four families. The Blattidj£ and Forficu-
lidas are represented only by cosmopolitan species, and Phas-
midas is not known. Excluding the few doubtful cases, species
are referred to thirty-six localities, or excluding the cosmopol-
itan forms to twenty-eight.

The above outline of our previous knowledge of Galapagos
Orthoptera has been given in order to clearly show the additions
made by the Hopkins Stanford Expedition.

Messrs. Snodgrass and Heller spent about six months in the
Islands; usually staying only a few days at each one. Albemarle
was however twice visited and several weeks spent in its ex-
ploration. No special effort was made to collect Orthoptera as
the purpose of the expedition was to make as complete collec-
tions as possible of the entire fauna and flora of the islands.
The insect collecting was done chiefly by Mr. Snodgrass.

The following table exhibits the distribution of the Orthoptera.

It will be seen from this that Orthoptera were obtained from all
islands previously visited except two, Jarvis and Gardner, and
from four new localities, Narboro, Culpepper, Abingdon and
Bindloe. Omitting the cosmopolitan Blattids, which he made
no effort to collect, Mr. Snodgrass obtained all the species
hitherto known, except two, Closteridea batiri Scudd. and

ORTHOPTERA

489

DISTRIBUTION OF ORTHOPTERA IN THE GALAPAGOS ISLANDS.

Note. New species in black face type. I,ocalities previously reported are shown by an *,
not previously reported, by a t-

Species.

Si

1

CO

S

<

t
t

HI

3

>— >

'>
u
cd
1— >

a

s

n
Q

V

n

s
0

to
a
't,

u
c«
M

a

u

*t

0
0

n

u
0

»i

V

to
.a
0

s

0

1

u
<u

a.
a

V

p.

"5
0

1

a
0

bo

a
2
<

6

a

3

I Anissolabis annulipes

2 Temnopteryx snodgrassii

3 Periplaueta americana

*

t

*

*
*

4 Periplaneta australasise

5 Nauphaeta cinerea

6 Nauphceta circumvagans

*

7 L/Cucophaea surinamensis

*t

*

*t

"t

*t

t

9 ClosVeridea bauri

10 Sphiiigonotus fusco-irroratus

t

?

*t

t

t
*t

t

*t

t*
*

II. Schistocerca melanocera

12. Schistocerca literosa

13 Halmenus robustus

*

t*

14. Halmenus cuspidatus

t

15. Halmenus choristopterus

t

16. Cesmopleura concinna

*

t
t

17, Anaulocotnera darwinii

»

18. Liparoscelis cooksoni

t

t

*t

19. Liparoscelis palludicola

20. Liparoscelis pallidus

t

*t

22. Xiphidium exitiosum

t

t

23. Gryllus galapageius

t
t
t

*?

t

24. Nemobius speculi

25. Cycloptilutn erraticum

t

*

t

26. Cycloptilum lepismoide

'

'

Desmopleura concinna Scudd. In addition to this, he obtained
eight new species or two-thirds as many (indigenous species)
as had been previously known, besides several subspecies and
races not included in the table. These species fall in six
genera, three of which had not been previously reported from
the Islands. Of these genera one, Temnopteryx, belongs to
the Blattidae, a family not hitherto represented in the indigenous
fauna. These specimens were obtained in fifty-four localities
or excluding the cosmopolitan species in fifty. Thirty-two of
these localities, or excluding the cosmopolitan species thirty,
are new.

Fifteen of the Galapagos Islands are now known to possess
Orthoptera. This number does not include small islets (like

490

McNeill

South Seymour) lying so near larger ones that they may be
regarded as forming an integral part of them. The number
of known species is twenty-six. This number includes six
species which are almost certainly cosmopolitan, but excludes
the doubtful Vates and Gry litis given in Scudder's List.

The Hopkins Stanford Expedition, as is obvious, has greatly
enlarged our knowledge of the orthopterous fauna of these
Islands. But it has done more. By collecting a large series
of individuals of the several species it has shown how potent a
factor isolation is in the production of species. This example
is especially instructive because the differentiation cannot well
be explained by natural selection, or by the effect of environ-
ment. All the indigenous species are apparently new. Of
these 20 indigenous species twelve are incapable of flight. Of
the remaining eight, one, G. solllaria, has a wingless female ;
another, Gryllus galapageius^ is wingless twice as often as
winged. Thus nearly seventy per cent, of the native orthoptera
are wingless.

Of the native species nine, including the cockroach, mantis,
crickets, and the Liparoscelis were in all probability brought
to the islands on timber or in partially decayed logs, the Man-
tid, at least, in the egg state. Though nothing is known of
Halmentis^ yet judging from some of its allies, this species
may have been introduced in the Q.gg state, the eggs being de-
posited in partly decayed or at least dead timber. The Schis-
tocera and the Sfhingonotus were almost certainly brought
there by their own wings, and it seems probable that the long-
winged Desmopleura was introduced in the same way. As to
the four remaining species there is greater uncertainty. Both
Anaulocomera and Conocephalus have efficient wings but
neither are great flyers and it seems very unlikely that they
should have been able to keep on the wing long enough to
reach the islands. Nor can we readily suppose that they could
have drifted in tree tops, since their wings would have tempted
them to fly, probably thus ruining their only chance for safety.
All things considered, the most reasonable hypothesis is that
they were introduced in the Qgg state, the eggs being attached
to twigs or leaves.

ORTHOPTERA 49I

To account for the presence of the last two species presents
the greatest difficulty of all. Nothing is known, I believe of
the habits of Closteridce. Only one species of the TriixalmcB^
Ch/oealiis conspersa Harr., so far as I know, has been observed
to lay its eggs in wood, and it is possible that either Closteridea
has, or its ancestors had, the same habit. All the Xifhidia^ I
believe, live on the ground and never on trees, so that its occur-
rence here is altogether unlikely and I can form no plausible
theory of the manner of its introduction. The problem is not
simplified by the fact that it is found on two islands, James and
Indefatigable.

Albemarle is represented by the greatest number of species
(twelve), and this is quite to be expected, since it is the largest
of the islands and its greatest length lies across the path of the
currents and trade winds. It is on the leeward side of the
group. Indefatigable is next in size and, with Chatham, it has
the next largest number (seven) of native species. Chatham is
fourth or fifth in size and is, I believe, the only one of the
islands now inhabited. It is the easternmost of the group.
Charles and James have each six representative species.
James is probably fourth in size and has no more than its pro-
portionate number of species. Charles is smaller and its rela-
tively larger number of species is probably due to the fact that
it has been inhabited and presumably more frequently visited
than most of the others. It is difficult to understand why Hood,
which is rather smaller than Charles, and has been much less
frequently visited, should have the same number of species.
The remaining islands, with one exception, are small and three
of them — Abingdon, Bindloe and Tower — lie far from the
others, so that the small number of species credited to each is
natural. The exception referred to is, however, curious.

Narboro ranks second or third in size. It has but three
species credited to it and all these are found on Albemarle. This
condition of things may be partly explained by the fact that the
island is completely cut off from the prevailing currents by
Albemarle, so that it is apparently impossible that drift wood
should ever reach it from the main land.

All the specimens of orthoptera collected by Mr. Snodgrass

492

McNeill

now form a part of the entomological collection of Stanford
University, and the types of the new species here described are
consequently to be found there.

FORFICULID^.

ANISOLABIS ANNULIPES Lucas.

Forficasila annulipes Lucas, Am. Soc. Ent. Fr. Bull., 1847, lxxxiv.
Anisolabis anmilipes 'QoK. , Biol. Centr. Am. Orth., p. 5.
Anisolabis bormansi ScvTiH., Bull. Mus. Comp. Zool., Vol. xxv, p. 5.
Anisolabis tnaritmia? Brun. (jiecl^ox.), Proc. U. S. Nat. Mus., Vol. xii, p.
192.

Careful examination of numerous specimens has convinced me that
Scudder's species is identical with a Pacific coast species, A. annulipes.
The joints of the antennae vary in what appear to be full-grown speci-
mens (those 4 or 5 mm. in length have from six to eight joints) from
thirteen to nineteen. In many cases, no doubt, the terminal joints are
lost. The position of the ring of luteous is quite uncertain. There
may be one, two or three joints beyond it and it .may include a single
joint or as many as three, or it may be absent entirely. A variable
number from one to four or more of the basal joints are also luteous.
Scudder's description was based upon a single female, consequently
his description of the forceps would apply to that sex only. In the
male these appendages are very unsymmetrical, the left one being de-
cidedly but regularly curved throughout its length, while the right is
bent, leather than curved, not far from the middle, and sometimes
the apical half is nearly at right angles to the basal. The femora are
only very obscurely banded with fuscous ; indeed, in many cases this
character is indistinguishable. The specimens from Albemarle are
much larger than the others, the largest measuring 20 mm. in length.
The same species was found at Clipperton Island, perhaps a thou-
sand miles northwest of the Galapagos group. Snodgrass collected
specimens from Albemarle, Chatham and Clipperton. Since all these
islands are inhabited, and as this is a cosmopolitan species found in
both the new and old world, and as no specimens have been found on
any of the uninhabited islands, there can be little doubt that it is an
introduced species.

Clipperton, 22 specimens varying from 4 mm. to full grown ; Chat-
ham, 4 specimens; Albemarle, 13 specimens, 5 males.

ORTHOPTERA

493

BLATTIDiE.

Subfamily BLATTIN^.

TEMNOPTERYX SNODGRASSII sp. nov.

Description. — Dark chestnut, somewhat lighter on the pronotum
and tegmina. Labrum, coxae and femora testaceous; tibiae and feet
chestnut. Pronotum parabolic, the posterior edge barely arcuate,

horny, polished and remotely and shal-
lowly punctate, with the anterior and
lateral margins a little lighter than the
disk. Tegmina short, not quite cover-
ing the first segment of the abdomen,
horny, with the veins exceedingly ob-
scure and the anal sulcus entirely obso-
lete ; sutural and costal margins straight
and parallel ; posterior margin straight
and slightly oblique, with the exterior
angle moderately rounded. Anterior
femora with three spines near the mid-
dle of the anterior lower carina, fol-
lowed by a series of minute spinules on
the apical half, no spines on the pos-
terior lower carina ; terminal spines f .
Middle and posterior femora with three
spines on the anterior lower carina and
four on the posterior lower carina ; ter-
minal spines \. Supra-anal plate triangular and slightly keeled (male),
or transverse and deeply incised (female). Cerci twice as long as the
supra-anal, plate in the median line, somewhat depressed, dark chest-
nut; stylets of the male very small.

Fig. 35. Tctnnofteryx snod-
grassii, female, head, pronotum
and anterior part of abdomen
dorsal view. (X^O

Fig. 36. Tetnnop-
teryx snodgrassii, fe-
male, dorsal view of ^ig. 37. Tetnnop.
end of the abdomen. teryx snodg-rassn,ma.\e,
(^8 \ dorsal view of end of

abdomen. (XS-)
Measurements. — Length of body, 9 8 mm., $ 7 mm.; tegmina, 9
S 3; pronotum, 93, ^ 3.

494

McNeill

This is the only blattid which is probably indigenous. All the
other species in the collection belonging to this family and the three
previously reported are well-known cosmopolitan forms.

This species is pei'haps more closely related to Temnopteryx surni-
chrasti Lucas than to any other, but it is readily distinguished from
it by the plain color and by the divided supra-anal plate of the male.

The specimens upon which the species is based were collected at an
elevation of 2,000 feet, on the south side of Iguana Cove Mountain,
Albemarle Island, on very wet ground covered with an abundant and
luxurious vegetation of ferns, shrubs and vines. The specimens were
found under leaves. As Mr. Snodgrass searched diligently for insects
living under stones and leaves, it is highly probable that the species is
restricted to this single locality. One of the specimens contains,
partly extruded from its body, a species of Gordias (?)•

Albemarle, i male, 3 nymphs.

Subfamily PERIPLANETINiE.
PERIPLANETA AUSTRALASIA Fabr.
Previously recorded by Bruner from Charles Island. No effort was
made to collect blattids, as they were supposed to be cosmopolitan
species entirely.
Chatham : i male.

Subfamily PANCHLORIN.E.
LEUCOPH.^A SURINAMENSIS Linn.
Two immature specimens from Chatham, both females.
Three previous expeditions to the Galapagos have reported this
species from Charles and Chatham.

MANTID.E.

Subfamily MANTIN^.

GALAPAGIA SOLITARIA Scudd.

The collection contains eleven specimens from three localities.
Seven of them are from Albemarle, the island from which came the
single male and female on which Scudder's description is based.
These should, therefore, be typical specimens but they are all dull
brown in color instead of clay-yellow, the females usually having the
pronotum and especially the abdomen with three longitudinal fuscous
stripes on an obscure testaceous ground. The males, however, and
especially the single adult, are grayish-brown with the tegmina of the
same color but with the cross-veins on either side of the main veins

ORTHOPTERA

495

deep fuscous. The specimens from Indefatigable are so different in
size and color that they deserve to rank as a distinct subspecies. The
adults are very light testaceous with the cross-veins of the tegmina
not at all infuscated. Both forms have a very marked depression on
the inner side of the anterior femora for the reception of the tibial
spur, and beyond this impression three large spines. The Albemarle
form, which maybe called typicalis^ has distad to this impression four
spines on the outer and eight on the inner side; the Indefatigable form,
which may be called ?najor^ has five and ten spines respectively.

Length of body, typicalls^ 40 mm., ?najor^ 41 mm. ; tegmina,
typicalis^ 25 mm., fnaj'or, 28 mm. ; hind femora, typicalis^ 12.5 mm.,
major ^ 18 mm.

Albemarle, i male, 2 females, 4 nymphs; Indefatigable (South
Seymour Island), 2 males, i female; Hood, i nymph.

ACRIDIDiE.

If^we exclude the doubtful Va^es and Gry/lus, the only two indige-
nous species previously described from the Galapagos which are not
found in the present collection belong to the family Acrididae. These
species are Closteridea bauri Scudd., a truxalid, and Desmopleura
concinna Scudd. , an acridid. The four other previously known species
were collected in large numbers and Mr. Snodgrass has prepared a
paper dealing with their variations and distribution.

Subfamily (EDIPODIN^.
SPHINGONOTUS FUSCO-IRRORATUS Stal.

Numerous specimens from six islands, viz. : Charles, 11 males, 28
females; Albemarle, 4 males, 9 females; Barrington, i male, 8
females; Hood, 3 males, 7 females; Chatham, 4 males, 14 females;
Indefatigable, 13 males, 31 females.

Subfamily ACRIDINiE.
SCHISTOCERCA MELANOCERA Stil.

Represented by numerous specimens which, as Mr. Snodgrass has
shown, are separated into distinct races which generally but not always
have their range limited by the geographical division of an island.

Specimens were collected on Charles, Albemarle, Barrington, In-
defatigable (South Seymour), James, Narboro, Duncan, Abingdon,
and Bindloe.

49^ McNeill

SCHISTOCERCA LITEROSA Walk.
Taken on Chatham, Tower, and Hood.

HALMENUS ROBUSTUS Scudd.

In addition to this species heretofore recognized in the specimens
from Indefatigable, there are, on James, two other quite distinct
forms which I am inclined to consider of specific value.

Indefatigable, 2 males, 6 females.

HALMENUS CUSPID ATUS Snodgrass (MS.).

This form is represented by so many specimens that there seems
little doubt that the species is a good one.
Albemarle, 2 males, 5 females.

HALMENUS CHORISTOPTERUS Snodgrass (MS.).

This form is unfortunately represented in the collection by only a
single specimen.
Charles, i female.

LOCUSTID^.

Subfamily PHANEROPTERINiE.

ANAULOCOMERA DARWINII Scudd.

The collection contains a single pair which agree well with the
published description, except that the anterior and middle femora have
no trace of spines on the lower surface.

James, i female. Indefatigable, i male.

LIPAROSCELIS Stal.
Nescecia Scudd., Bull. Mus. Comp. Zool., Vol. xxv, 1893, No. i, p. 20.

Thirty-seven specimens of this genus are in the collection, taken
from six of the islands, which are, however, very unequally repre-
sented. Twenty-six specimens are credited to Albemarle while Inde-
fatigable has but a single larva, and Barrington and James a single
adult each. So great is the difference among these specimens that it
is necessary to recognize three species and one subspecies, or prob-
ably four species, and each of the islands represented possesses at
least one distinct variety. All of these forms are, without doubt,

ORTHOPTERA 497

more closely related to each other than to other species of the genus,
and all are easily separated from their congeners. In 1S95, Brunner,
in his Monographic der Pseudophylliden, assigned four species to this
genus, two being from Mexico and one from Colombia. Lifaros-
celis cooksotii is at once distinguished from two of these species by its
smooth face. It is probably much more closely related to the remain-
ing species, L. nlgrispinis Stal, but is readily distinguished by the
fewer spines on the under side of the fore and middle femora and by
the acute unemarginate subgenital plate and straight ovipositor of the
female.

Albemarle, 26; Indefatigable, i larva; Harrington, i; James, i.

LIPAROSCELIS COOKSONI Butler. .

Agrcecia cooksoni^x^ii^., Proc. Zool. Soc. Lond., 1877, pp. 87-88.
Biicrates? cocoannus ? Brun. {nee Boliv,), Proc. U. S. Nat. Mus., Vol. xii,

p. 192.
Nescecia cooksom Scvbd., Bull. Mus. Comp. Zool., Vol. xxv, 1893, No. i, p.

20.
Liparoscelis cooksotii Vi'&Xi^^., Monag. Pseudophyl., p. 176, 1895.

Butler's type was from Charles, so there is no doubt as to the forms
to which his name applies. True, he had some immature specimens
from Albemarle, and some or all of these may have belonged to another
species. Brunner's specimen
was from Charles and its iden-
tification may be considered
certain. As Scudder's speci-
mens (except a single immature
one from Indefatigable) came
from Albemarle, they probably
belong here, though they are
somewhat larger than my speci-
mens from the same locality. Fig. 38. Liparoscelis cooksoniBuiXtv^
and the ovipositor is more than female, ovipositor and end of a bdomen.
proportionally longer; how-
sever, this may be due to the fact that the method of measuring
this organ used by Scudder and myself are different. .He measures
a straight line from the tip to the junction of the upperand lower
laminae where they leave the body; in other words he measured
the chord and I the arc. If the curvature was measured, and es-
pecially, in this case, if the keel of the lower lamina was measured,
the length here given would be increased two or three millimeters.

498 McNeill

Brunner gives Scudder's measurements and they are exactly the same,
except that he adds the length of the four femora, so that his speci-
mens were the same as Scudder's. His description does not fit the
Albemarle specimens in all respects, but on the whole it suits this
species better than any other.

To this species are referred the specimens from Charles Island, a part
of those from Albemarle, and those from James and Hood. The spe-
cimens from Hood, however, form a distinct subspecies and, as before
stated, each of the islands probably has its distinct race which cannot be
properly characterized here for want of material. The Charles race,
however, is decidedly larger and more robust than the others; the teg-
mina overlap plainly in the females ; the spines of the under side of the
femora are 2 or 3, 2, 4, and the spinules on the outer carina of the upper
side of the hind tibiae are 2 or 3. The Albemarle specimens are decid-
edly smaller ; the tegmina of the females scarcely more than touch at the
base on the basal half; the femoral spines are 3 or 2, 2 or 3, 4 or 5 ;
the hind tibiae spinules (outer carinas upper side) 4 or 5 or none be-
yond the middle. The James form is represented by a male only,
and it must be compared with the male of the^Albemarle form as there
is no male from Charles. Compared with this form, the male has
the spinules on the outer carinae of the upper side of the hind tibiae
uncertain in number and so exceedingly minute as to be scarcely dis-
coverable. The musical organ occupies about one-third instead of
one-half the length of the tegmina, and the stylets of the subgenital
plate much exceed (by their whole length and a part of the plate
equal to half their length) the cerci. These forms resemble each
other in having the pronotum strongly tuberculate, the tegmina of the
female decidedly shorter and little if any longer than the pronotum,
the spines of the hind femora being generally four, the spinules of the

outer carinae of the hind
tibiae never more than five
and confined to the distal
half, and the ovipositor com-
paratively short and wide
and quite strongly curved.
Tlie remaining form is of

_, - .^ ,. , . .r r subspecific value and may

riG. 39. l^tfaroscelis cooksoni enstjer,ie.- '■ ''

male, ovipositor. (X3.) ^^ known as Liparoscelis

cooksoni etisifer (Fig. 39)
from its comparatively longand gracefully curved ovipositor. It differs
from L, cooksoni Butler in the following particulars: It is relatively

ORTHOPTERA 499

more slender and more compressed; disk of the pronotum only ob-
scurely tuberculate and moderately rugose; tegmina of the female at
least as long as those of the male, decidedly longer than the pronotum ;
femoral spines, 3, 3, 5 ; spines of the outer carinse of the upper side
of the hind tibiae eight, occupying the distal three-fourths; cerci of
the male short, exceeded by the subgenital plate (exclusive of the
stylets) by at least their own. length ; ovipositor of the female long,
decidedly curved and slender.

Hood, I male, 2 females. May 13 and 14.

Measurements are: Length of body, $ 34 mm., 9 33; pronotum,
$ 6.5, 9 7; tegmina, ^ 9, 9 9 ; hind femora, $ 17, 9 18; length
of ovipositor, 9 14.2; width of ovipositor, 2.

Charles, 2 females, 6 nymphs, May 9.

Aleasuremenis o{ female: Length of body, 37.5 mm.; pronotum,
7 to 8 ; tegmina, 7; hind femora, 18.5 to 19; ovipositor, 11.5 to 12.

Albemarle, i male, 2 females, June 9 and 12.

Measurements. — Length of body, $ 27 to 28 mm. 9 25 to 28;
pronotum, $ 6.2 9 6.5; tegmina, $ 8 to 8.7, 9 5.5; hind femora,
$ 13, 9 14 ; length of ovipositor, 9 10 to 10.5 ; width of ovipositor, 2. 5.

James, i male, i larva.

Measurements of male: Length of body, 28 mm.; pronotum, 6;
tegmina, 8; hind femora, 14.5.

Indefatigable, one larva.

LIPAROSCELIS PALUDICOLA sp. nov.

Description. — Color nearly uniform chestnut brovs'n except the sides
and top of the head, which are grayish and the face an hour-glass-
shape and the disk of the pronotum which are light clay color. The

Fig. 40. Liparosceli's paludico/a, female, ovipositor. (Xs)

500

McNeill

pronotum is obscurely tuberculate on the disk, with a distinct median
sulcus which is colored like the body but fades out on the front edge
of the prozone and on the posterior part of the metazone. The teg-
mina of the male are but slightly longer than those of the female;
in both they are much longer than the pronotum, and oblong in shape,
with the apex rounded. The femoral spines are usually 3-2-4 ; the
spines on the outer carinjE, upper side of the tibiae about three confined
to the distal half. The cerci are not exceeded by the stylets of the
subgenital plate. The ovipositor is very stout and moderately curved.

This species is closely related to L. cooksoni but easily distinguished
by its peculiar coloration as well as by its size and several structural
characters. According to Mr. Snodgrass, it is confined to mangrove
swamps, where it is found under the bark of Avice?ztzia^ a tree which
always occurs with the mangrove. Specimens were numerous at
Tagus Cove and a single one was found in a mangrove swamp at
Elizabeth Bay. The species is sluggish in its habits, and its song was
not heard.

Measurejnents. — Length of body, $ 40 mm., $ 43; pronotum, $
6, 9 6; tegmina, $ 10-12, 9 10; posterior femora, $ 14, 9 19;
length of ovipositor, 16; width of ovipositor, 3.

Albemarle, 12 males, 6 females.

LIPAROSCELIS PALLIDUS sp. nov.

A single specimen, rescued from a mocking-bird, is so different
from all the others that, although it is without a head or forelegs, or

middle legs, I feel obliged to
make it the type of a new
species. The following is
a description :

Description. — Yellowish-
white, body and limbs mot-
tled with purple, becoming
black on the pronotum ; ovi-
positor whitish on the basal,
black on the distal half.
Pronotum evenly and rather
strongly tuberculate, the two
transverse sulci distinct, the median sulcus apparent only at the prin-
cipal transverse sulcus; shoulders somewhat distinct on the metazone,
obsolete on the prozone ; posterior margin straight ; lateral lobes

Fig. 41.
ovipositor.

Liparoscelis pallidus, female,
(X3-)

ORTHOPTERA 5OI

with the posterior margin very slightly sinuate and distinctly less
oblique than the anterior margin. Tegmina ovate, subacuminate,
touching only at the base, with the median and radial veins distinct
and the dorsal angularly separated from the lateral field. Hind
femora armed below on the distal half with seven spines, the first two
minute ; hind tibiae armed above with three spinules on the outer and
nine on the inner carinae. Ovipositor only slightly curved on the
upper, more decidedly on the lower margin; subgenital plate trian-
gular acute.

Measurements. — Length of body (estimated), 37 mm. ; pronotum,
6.5; tegmina, 10; hindfemora, 17.5; length of ovipositor, 14; width
of ovipositor, 2.5.

Barrington, i female.

Subfamily CONOCEPHALIN^.
CONOCEPHALUS INSULANUS Scudd.

It is rather remarkable that this species is known only from Chat-
ham. Mr. Snodgrass reports that it was common in a wet pasture
about 1,000 feet above sea level. It was not found elsewhere. Three
of the males and one female nymph are brown. In this variety the
tegmina are maculate pretty evenly with obscure fuscous. Besides the
fuscous stripe which borders the stridulating organ in both varieties,
the brown specimens have the upper half of the lateral lobes of the
pronotum infuscated, and this color is continued as a stripe on the teg-
mina bordering the radial vein. The lower half of the lateral lobes is
also infuscated more or less, and this region is separated from the
upper part by a moderately distinct stripe of the lighter ground color.

Chatham, 5 males, 3 females, 4 nymphs.

XIPHIDIUM EXITIOSUM sp. nov.

Description. — Color, green with the usual dorsal stripe on the head
and pronotum. Pronotum widens posteriorly and continues to the hind
margin ; it is darkest medianly and on the edges. The vertex is only
slightly pinched at the eyes and is scarcely wider than the basal joint
of the antennas. The posterior border of the lateral lobes of the pro-
notum is scarcely perceptibly sinuate. The posterior femora are nearly
as long as the body, unarmed beneath. The ovipositor is at least three-
fourths the length of the hind femora and is very slightly curved.

502 McNeill

This species is closely related to X. nemorale Scudd., from which
it differs strikingly in color. It is also readily distinguishable by the

Fig. 42. Xiphidiutn exitiosuni, female, lateral view. (X3O

length of the tegmina which are much shorter in the male and much
longer in the female, and by the longer and straighter ovipositor.

Measurements. — Length of body, S 15 mm., 9 17; tegmina, $ 6,
$ 8; hind femora, $ 12, 9 16; length of ovipositor, 12.

Indefatigable, i female, 2 nymphs; James, 2 males, 2 females, i
nymph.

GRILLIDiE.

Subfamily GRILLIN.^:.

GRYLLUS GALAPAGEIUS Scudd.

This species is represented in the collection by a considerable series
taken in seven of the islands. The variation in size is considerable as
indicated by the measurements, and the color variation (especially
among the immature) is quite marked. Less than one-third of the
adults (three males and three females) are winged. The obscure,
minute, reddish spots are always present in the females, but are gen-
erally entirely wanting in the male. On the other hand, while the
females show considerable uniformity in the color of the tegmina, these
are entirely black in males from Albemarle and Narboro, and wholly
testaceous in a single male from Culpepper. The variation in color

ORTHOPTERA 503

among the immature specimens is much more marked, but I believe
they all belong to the same species. Five specimens from Culpepper
and one from Abingdon are testaceous mottled with fuscous. Two
specimens from Tower and one from Hood have the general body
color, which is black, considerabl}- varied with testaceous on the pro-
notum and abdomen. Eight other specimens from Hood and two
from Albemarle are entirely black, except for the hind femora which
are always either plainly or obscurely transversely barred with fuscous
on a testaceous ground, and sometimes, also, a pair of minute reddish
spots on the disk of the pronotum and very obscure testaceous or red-
dish spots on the abdomen and pronotum. Scudder was unable to
distinguish the tympanum on the inner side of the anterior tibiae. It
is present in all these specimens though minute.

The examination of 42 specimens of Grylli of all ages from seven
of the islands compels the conclusion that the second species of this
genus, which Scudder thought to be indicated by three immature
specimens from Charles, probably does not exist; at the same time
the absence of all specimens of this genus from Charles leaves the
matter in doubt.

Measurements. — Length of body, $ 14 to 23 mm., 5 18 to 22;
tegmina, $ 7.5 to 14, 9 10 to 14.5; hind femora, $ 10 to 14, $ 14;
ovipositor, 18 to 19.

Of its song, Mr. Snodgrass says it " consists of a continuous chirp-
ing sound differing from the intermittent sound made by the small
tree crickets ( Cycloptiluiri) found in swamps."

Albemarle, 3 males, 2 nymphs ; Hood, 3 males, 8 females, 9
nymphs; Chatham, 2 males, 2 females, i nymph; Tower, 2 nymphs;
Abingdon, i nymph ; I^arboro, i male ; Culpepper, i male, 6
nymphs ; uncertain, i female.

NEMOBIUS SPECULI sp. nov.

Description. — Dark brown more or less varied with testaceous.
Sometimes the lighter color predominates but usually the darker, ex-
cept in the legs which are always obscure testaceous banded more or
less regularly with brown. The face, occiput and disk of the pronotum
are obscure testaceous maculate with brown ; on the posterior part of
the disk the markings are minute round spots; the lateral lobes of the
pronotum are brown with about seven round light spots, five above
and two below ; the veins of the tegmina are for the most part testa-

Proc. Wash. Acad. Sci., November, 1901.

504

McNeill

ceous. The frontal costa is rather thickly pilose with stiff black hairs.
The antennae are about four times as long as the body. The pronotum
is only slightly narrowed anteriorly, less than one and a half times as
wide as long and usually sparsely pilose with long bristly hairs. Teg-

mina of the male about two-thirds as long
as the abdomen, truncate or very slightly
arcuate posteriorly with the external and
internal posterior angles extended into
rather distinct spines ; speculum very
large, including nearly all of the dorsal
field, the apical field therefore very much
reduced ; wings apparently obsolete. The
anterior and middle femora are strongly
compressed, the latter decidedly arcuate.
The posterior tibiag are armed with three
spines on both the inner and outer angles ;
these spines are situated on the apical half,
the lowermost inner one being half as long
as the tibite and about as long as the
metatarsus ; spurs of the posterior tibial,
three, on both the inner and outer side,
the upper inner one being about as long as
the metatarsus, the latter a little less than
twice as long as the two joints of the
tarsus. Cerci about as long as the head
Ovipositor of the female of uncertain length, not

Fig. 43. Nemobius speculi,
male, dorsal view except
hind legs. (X8.)

hind

and pronotum.
toothed at the tip.

Measurements. — Length of body, 5 mm. ; tegmina,
femora, 4.

The collection contains about
twenty-three specimens of Nemobius.,
all collected at the same time in a
single locality. Unfortunately only
two of them are mature, and both
are males, though a considerable
number of both males and females
are in next to the last moult. Since the tables for determining
species in this large genus (at least forty-five species are described)
are generally based on the females, and since in many cases the males
are not known, I have not been able to satisfactorily determine the
position of this species; but since all species indigenous to the Gala-

FiG. 44. Nemobius speculi, male,
posterior tibia. (X^.)

ORTHOPTERA 505

pagos have proved to be new, the following description is given with-
out much fear of its specific distinctness.

Albemarle, Tagus Cove, 2 males, 20 nymphs.

Subfamily MYRMECOPHILIN^.
CYCLOPTILUM ERRATICUM Scudd.

This species has been known heretofore by a single specimen, a
male from Charles Island. The present collection contains five adults
and several nymphs. The female is so unlike the male in appear-
ance as to make it difficult to recognize it as belonging to the same
species. The best preserved specimens are dull cinnamon brown,
especially anteriorly, becoming chestnut posteriorly, with the legs
sometimes lighter. The surface of the body and legs is smooth or
very sparsely pilose. The female is wingless with the pronotum sub-
quadrate, very moderately widening postei'iorly, and not covering the
mesothorax. The ovipositor is as long as the hind femora. A few
scales are found on the under side of the body and sometimes a single
row bordering some of the abdominal segments. The antennae are
about twice as long as the whole body.

It occurs in mangrove swamps on the under side of leaves and
under the bai"k of dead trees, and is reported to be numerous, but shy
and difficult to catch. Mr. Snodgrass says: " Their singing can be
heard the whole day in the swamp. It consists of an intermittent
chirping sound, the intervals between the series of chirps being short,
however, so that the whole has a continuous sound, but it is readily
distinguished from the louder and continuous chirping of the large
cricket living everywhere beneath the rocks." Two of the specimens
from Albemarle, apparently young males, though the pronotum is
very small, have large tegmina and wing pods.

Albemarle, 3 males, i female, 5 nymphs; Hood, i male.

CYCLOPTILUM LEPISMOIDE sp. nov.

This species is remarkable among Orthoptera for the possession of
scales. Though this character is shared by other species of the genus
Cycloptllum^ it is much less evident. C. lepismoide differs from C.
erraticum in its color, which is very dull brown, becoming blackish
posteriorly. The legs are quite thickly pilose, and these, together
with the under surfaces, are luteous or white varied with black. The

5o6 McNeill

white and black of the under surface and the black of the head is due
to the presence of thickly imbricated scales. These scales are of vari-
ous forms but possess the common characters of a prominent carina,
which ends distally in a point, and a complicated sculpture. This
sculpturing consists of thickened plates which are arranged in rows
parallel, approximately, to the median carina, and also in rows forming
a right angle with the fii'st set. Under a moderate power of the micro-
scope the appearance presented is that of concentric curved depres-
sions running lengthwise of the scale, and less evident depressions
curving radially from the midrib. The legs, and especially the pos-
terior pair, are plainly shorter and more robust. The last abdom-
inal segment of the male is shaped like the, bow of a boat and from its
apex extends a pair of minute forked appendages which are approxi-
mate nearly throughout their length. The anal cerci are quite as long
as and distinctly stouter than in C. erratictan and they are quite
thickly pilose with hairs, the longest of which are more than twice as
long as the diameter of the cerci at the base. On account of these
hairs the cerci appear much stouter.

]\feas7ireme?its. — Length of body, S 6.5 mm., 9 9; pronotum
S 3-5' 5 2; hind femora, ^ 3.5, 9 5.

Except for the omission of any mention of scales, Scudder's descrip-
tion of Cycloptihim erraticum would appear almost equally well to
each of these species. It seems unlikely that he has overlooked so
striking a character as this, however, and the smaller size of the only
adult male of C lepisj7ioide has determined me to consider this one
as the undescribed form.

Found on the leaves of bushes and possibly also on the leaves of
the mangrove.

Albemarle, i male, i nymph ; Narboro, 6 nymphs ; Abingdon, 1
female.

PROCEKDINGS

OF THE

WASHINGTON ACADEMY OF SCIENCES

Vol. Ill, pp. 507-539. November 7, 1901.

SYNONOMY OF THE FISH SKELETON.

[Plates LXIII-LXV. Text Figures 45-46.]

By Edwin Chapin Starks.
Leland Stanford Junior University.

CONTENTS.

Introduction 507

List of synonyms 508

Alphabetical list of terms 527

Bibliography 538

INTRODUCTION.

In works on the fish skeleton much confusion of nomenclature
has arisen from the difficulty of homologizing the bones with
those of the higher vertebrates. Some authors have attempted
to homologize every element in which there is a trace of homol-
ogy while others have only homologized those in which it is most
evident, giving special names to the remaining ones. These
special names have not caused confusion, as each has been used
consistently for the bone to which it was originally assigned ;
but names involving the assertion of homology have frequently
been applied to more than one element, according to the differ-
ent interpretations of different authors. Thus the word cora-
coid (in some of its forms) has been applied to four distinct
bones of the shoulder girdle.

Proc. Wash. Acad. Sci., November, 1901. (507)

508 STARKS

This troublesome confusion is not confined to the older writ-
ings, but still continues. The present paper has been prepared
as a key to the labyrinth.

In the accompanying plates and text figures ^ each element of
the skeleton is numbered, and under corresponding numbers in
the text the synonomy of each term is given. The terms first
given, though perhaps not always the best, are in the most gen-
eral use by English writers and are therefore here adopted.
These terms form a nomenclature for which Owen and W. K.
Parker are chiefly responsible, though Huxley and Gill have
contributed to it.

Personal names indicate the nomenclature adopted. The
first name after a term is generally that of the author who first
so used it.

The nomenclature of Cuvier, Owen, Giinther and Parker is
given in full, that of other anatomists only where they differ
from these authors.

I have had some difficulty in determining the identity of a few
of the elements described by Erdl in his paper on the skeleton
of Gyimiarchus niloticus published in Akademie der Wissen-
schaften, 1847. In Sir Richard Owen's personal, interleaved,
annotated copy of his Lectures on Comparative Anatomy,
which is in my possession, I find, among many other notes, his
interpretation of some of these elements. When it has seemed
advisable I have added his notes, or commented on them.

Synon3^mic lists published by Sir Richard Owen on the
skull ;- by Dr. Albert Giinther on the entire skeleton ;^ and by
Dr. Theo. Gill on the shoulder girdle,^ have been of great
help in preparing this paper.

LIST OF SYNONYMS.
I. Vomer. Cuvier, Owen, Stannius, Giinther, Parker and many
others.
Anteal. Gill,

^Made from drawings by Chloe Lesley Starks.

2 Owen, Lect. on the Comp. Anat. and Phys. of Vert. Anim., Pt. i, Fishes,
London, 1846.

''Giinther, Introd. to the Study of Fishes, *Edinburgh, 18S0.
*Gill, Arrang. of Fam. of Fishes, Wash., D. C, 1872.

SYNONOMY OF THE FISH SKELETON 509

Pfltigschar. Meckel.
PJlugschaarbei7i^ vomer. Erdl.
Rhinosphenal. Geoffrey.

2. Ethmoid. Cuvier, Giinther, Parker. Not eth^noid of Owen,

No. 14; nor etimioldeiim of Erdl, No. 3; nor of Vogt and

Yung, No. 16.
Crista ethtnoidei. Bojanus.

Nasal} Owen, Vogt and Yung. Not nasal oi Cuvier, No. 51.
Nasentheil des Stirnbeines^ 7tasalis ossis frontis. Erdl.
Os eth??20tdeuni. Stannius.
Prosethnioid. Gill.
Riechbei7tk6rper . Meckel, Wagner.
Siebbein. Kostlin.

3. Prefrontal. Owen, Gill.
Ecto-ethtnoid. Parker (W. K.).
EthtJioideum cribrostnn. Bojanus.
Frontal anterieur. Cuvier, Agassiz.
Lateral ethmoid. Parker.

Lachry77ial. Geoffroy. Not lacJiry77ial of Parker, No. 49.

Orbitaljliigel des Keilbei7ts. Miiller.

Os fro7itale anterius. Stannius.

Pareth77ioid. Parker (T. J.).

Riechbein^ os eth7noideu77i. Erdl. Not et/i77ioidetc77? of Kall-
mann, No. 2 ; nor of Vogt and Yung, No. 16 ; nor ethmoid of
Cuvier, No. 2 ; nor of Owen, No. 14.

Seitlichen Riechbein. Meckel, Wagner.

Thrdnenbein.'^ Erdl.

4. Frontal. Owen, Parker, Geoffroy and in general use.

Frontal principal. Cuvier, Agassiz.

Mittlerer Stirnbein., frontalis ossis fro7it is. Erdl.

Os frontale. . Stannius.

Stir7ibei7i. Meckel, Wagner.

5. Sphenotic. Parker.

Ala 77iag7ia ossis sphenoidei. Erdl.
Frontal poster ieur. Cuvier, Agassiz.
Os frontale poster ius. Stannius.
Post fro7ital. Owen, Gill.

' Nasale oder mittleres ethmoideum. Vosrt and Yungr.

2 " Von anderen wiirde diesen Knochen als Thranenbein gedeutet worden
sein." Erdl, op. cit.^ p. 209.

5IO STARKS

Schldfbeinschuppen. Bojanus.

Temporal. Geoffroy. Not temporal of Vogt and Yung, No.
9; nor of Cuvier, No. 17.

6. Parietal. Owen, Parker, Giinther.

Os parietale. Stannius, Erdl.
Scheitelbein. Meckel, Wagner, Erdl.

7. Epiotic. Parker.
Epioticum. Huxley.

Exocclpltal. Geoffroy. l^ot exoccipital oi Owen, No. 11.
Occipital externe. Cuvier, Agassiz.
Os occipital extermi77i. Cuvier, Vogt and Yung.
Os mastoideum. Kallmann .

Zxvischetz-scheitelbein^ os interparietale. Erdl. Not inter-
parietal of Cuvier, No. 8.
Paroccipital. Owen, Giinther.
Seitlichen obern Hinterhauptbeine. Meckel, Wagner.

8. Supraoccipital. Owen, Parker, Giinther.
Hinterhauptschuppe. Meckel, AVagner.

Interparietal. Cuvier. Not os interparietal of Erdl, No. 7.
Occipital superieure. Cuvier.

Os occipitale stiperius. Stannius, Vogt and Yung.
ScJnippentheil des Hinterhaiiptsbeins., occipitalis ossis occipitis.

Erdl.
Szipra-occipitine. Gill.
Squama occipitalis. Hallmann.

9. Pterotic. Parker, Huxley.^
Ecaille du tetnporal. Agassiz.
Exrtipeal. Geoffroy.
Felsentheil. Bojanus.

Gehordeckel^ opercuhun pori acoicstici. Erdl.' "

Mastoid. Owen, Giinther. "Not ?fiastoideu7n of Bojanus, No. 10.
Os extrascapiilare. Stannius.
Os mastoideum. Stannius.
Schuppentheil des Schldfbeins. Kostlin.

1 Prior to the appearance of his Anatomv of Vertebrate Animals, in 1871,
Huxley homologized this bone with the squamosal of the higher vertebrates ;
later he accepted Parker's interpretation.

2 From Erdl's plate and description I cannot be certain as to the identity
of this bone. Owen was evidently puzzled over it as I find in his MS. notes the
following: '■'Appears to be pterotic. Compared to outer auditory bone in
beaver."

SYNONOMY OF THE FISH SKELETON 5II

Squafna tetnporis. Hallmann.

Squatnosal. Huxley.

Tempo?-ale. Vogt and Yung. Not temporal of Geoffrey, No.

5, nor of Cuvier No. 17.
Zitsenbein. Meckel.

10. Opisthotic. Parker, Huxley.
Felsenbein. Oken.

Inter calare. Segemehl.

In-rupeal. Geoffroy.

Mastoideum. Bojanus, Kostlin. Not mastoid of Owen, No. 9.

Oberjldchliche Knochen-lamelle. Stannius.

Os innominatiim. Hallmann.^

Otosteal. Owen.

Petrosal. Owen. Not petrosal oi Segemehl, No. 15.

Rocher. Cuvier.

Rocher rudimentaire. Agassiz.

11. Exoccipital. Owen, Giinther, Parker. Not exoccipital of

Geoffroy, No. 7.
GeleJiktheil. Meckel.
Gelenktheil des Jwiterhauptbems., condyloidea ossis occipitis.

Erdl.
Occipital lateral. Cuvier, Agassiz.
Os occipitale laterale. Stannius, Vogt and Yung.
Suroccipital. Geoffroy.
Seitlichen untern Hinterhauptbeine. Meckel, Wagner.

12. Basioccipital. Owen, Parker, Giinther.
Basilaire. Cuvier, Agassiz.

Basiltheil des hinterhauptbeins basilar is ossis occipitis. Erdl.

Basisphenal. Geoffroy.

Corpus ossis occipitis. Hallmann.

Hinterhauptbeinkorper . Wagner.

Os basilare. Stannius.

Otosphenal. Geoffroy.

13. Parasphenoid.^ Huxley, Parker.

Basisphenoid (the posterior half). Owen,^ Giinther. Not basi-
sphenoid of Huxley, No. 14.

^ Hallmann has also used this term for basisphenoid No. 14.

2 "The bones of fish and amphibians usually denominated vomers must
part with their claims to that title and yield it to the so-called parasphenoid."
Sutton in Proc. of the Zool. Soc. London, 1SS4, p. 570.

'Owen and Cuvier give the opposite ends of the paraspkenoid different
names.

512 STARKS

Corpus ossis sphenoidei. Erdl.
Hyposphenal. Geoffroy.
Keilbeinkorper. Meckel, Wagner, Erdl.
Os sphenoideuni basilare. Stannius.
Progressus ascendens piscium. Hallmann.
Presphenoid (the anterior half). Owen.
Spheno'ide principal. Agassiz.
Sphenoideuyn basilare. Hallmann.
Spheno'ide posterieur (the posterior half) | „
Sphenotde principale (the anterior half) j

14. Basisphenoid. Huxley. Not 3ajzV/^<?«o/fl? of Owen, No. 13.
Entosphenal. Geoffroy.

Ala parva sphenoidei (in carp) . Hallmann.

Dichost. Gill.

Ethmoid. Owen. Not ethmoid of Parker, No. 3 ; nor Cuvier,

No. 2 ; nor Ethmoideum of Vogt and Yung, No. 16,
Ethmo-turbinal. Owen.
Ethmozde crdnien. Agassiz.
Os sphenoideum anterius. Stannius.
Os innominattim (in perch). Hallmann.
Sphenoide anterieur. Cuvier.

15. Prootic. Parker.

Alisphenoid. Owen, Giinther. Not alisphenoid of Huxley,

No. 16.
Ala temporalis. Stannius.
Felsenbein. Meckel, Wagner.
Felsentheil des Schldfettbeifies, petrosa. Erdl.
Grande aile du spheno'ide. Cuvier, Agassiz.
Hinterer Schldfenjliigel. Kostlin.
Petrosum. Hallmann.

Petrosal. Segemehl. l^ot petrosal oi Owen, No. 10.
Ptereal. Geoffroy.
Prooticufn. Huxley.

16. Alisphenoid. Huxley, Parker. Not alisphenoid of Owen,

No. 15.
Aile orbitaire. Cuvier, Agassiz.
Ala orbitalis. Stannius.
Ala magna sphenoidei. Hallmann.
Eth?noideum. Vogt and Yung. Not ethmoidetim of Erdl, No.

3 ; nor of Hallmann, No. 2 ; nor eth??zoid of Owen, No. 14.

SYNONOMY OF THE FISH SKELETON 513

Grosse Keilbeinjlugel. Meckel, Wagner.

Ingrassial. Geoffrey.

Kleine Flugel des Keilbeins. Bojanus.

Kleiner Keilbeinjliigel^ ala minor ossis sphenoidei. Erdl.

Orbitosphenoid. Owen.

Subtectals. Gill.

Vordere Schldfenjliigel. Kostlin.

17. Hyomandibular. Huxley, Parker, Giinther.
Epitympanic. Owen.

Gelenktheil des Schldfensbeins. Kostlin.

Mastotdien. Agassiz.

Obere Gelenkbein. Meckel, Wagner.

Os quadratum S. tympanicum. Kallmann.

Os temporale. Stannius.

Paukering-ktiochen. Bojanus.

Schuppentheil des Schldfensbeines^ squamosa ossis temporum.

Erdl.
Serrial. Geoffroy.
Temporal. Cuvier. Not temporal of Vogt and Yung, No. 9 ;

nor of Geoffroy, No. 5.

18. Symplectic. Parker.

Griffelformig Knoche. Meckel.
Mesotympanic. Owen.
Os symplecticum. Stannius, Hallmann.
Symplectique. Cuvier.
7ympano-malleal. Agassiz.
Uro-serrial. Geoffroy.

19. Quadrate. Huxley, Parker, Giinther.
FlUgelbein. Bojanus.
Hypocotyleal. Geoffroy.
Hypotympanic. Owen.
Jochfortsatz. Kostlin.

Jugal. Cuvier. Not yV/^a/ of Agassiz, No. 50.
Os quadratum. Agassiz.
Os carre. Agassiz.

Os quadratojugale. Stannius, Hallmann.
Unter Gelenkbeifi. Meckel, Wagner.
Vorderes Stiick des os quadrattcm. Erdl.

20. Pterygoid. Owen, Parker, Giinther.
Adgustal. Geoffroy.

514 STARKS

Ectopterygoid. Huxley.

JSxtermim. Stannius.

Gaumenbein ^ os palatlnu?n. Erdl. Not gaumenbein of

Meckel, No. 21 ; nor palat in um of Stannius, No. 21.
KeilbeinJIugel. Kostl i n .

Os transversu7n s. pterygoideu?n externum. Stannius.
Pterygoideum externum. Kallmann.
Transverse. Cuvier, Agassiz.

21. Palatine. Owen, Parker.

Gautnefibetn. Meckel. Not Gaumenbein of Erdl, No. 20.

Os palat inutn. Stannius.

Palatin. Cuvier, Agassiz, Giinther, Geoffrey.

22. Mesopterygoid. Parker.
Entopterygoid. Owen, Giinther, Huxley.

Ergdnzungsstiick oder Fliigelfortsatz^ os pterygoideum. Erdl.
Uerisseal. Geoffroy.

Os pterygoideum. Stannius.
Os transverstim. Kostlin.
Pterygo'idien inter?ze. Cuvier, Agassiz.
Pterygoideum internufn. Hallmann.
Unterc KeilbcinJingeL Meckel.

23. Metapterygoid. Huxley, Parker, Giinther.
Caisse. Agassiz.

Epicotyleal. Geoffroy.
Gaumenjliigel des Keilbein. Bojanus.
Hintere^ Stiick des os quadratum. Erdl.
Os tympanicum. Stannius.
Os tytnpafii. Agassiz.
Ptcrygoideuftz poster ius. Kallmann.
Prety?npanic. Owen.

Scheibenformig Knoche. Meckel, Wagner.
Tytnpatial. Cuvier. Not tympanal of Geoffroy, No. 24.

' Only one element is shown in Erdl's plate where the palatine and ptery-
goid usually are. Its articulation below is typical for the pterygoid, while
above it is described as being attached to the vomer. Either the palatine is
missing or the suture between it and the pterygoid has been overlooked. Erdl
uses the same names as Meckel and Stannius for the palatine. In his MS.
notes Owen has identified it as the pterygoid and it is here so inserted.

2 In his MS. notes Owens refers this to the symplectic {mesotymfanic). The
shape, size and position of the bone in Erdl's plate would seem clearly to indi-
cate the metapterygoid .

SYNONOMY OF THE FISH SKELETON 515

24. Preopercle. ]

Preopercular. V General English usage.
Preoperculum. J

Apophyse styloide du teinporal. Agassiz.
Pre-opercule. Cuvier.

Tyntpanal. Geoffrey. Not tympanal of Cuvier, No. 23.
Vorderdeckel. General German usage.
Vorkietnendeckelstiick. Meckel.

25. Opercle. 1

Opercular. V General English usage.
Operculum. )

Engentlich Kiefnendeckelstuck. Meckel, Wagner.
Kiernendeckel. Erdl.
Operctilaire. Cuvier, Agassiz.
Stapeal. Geoffrey.

26. Subopercle. 1

Subopercular. V General English usage.
Suboperculum. )

Inceal. Geoff roy.
Sous-opercule. Cuvier, Agassiz.
Unterkiemendeckelstiick. Meckel, Erdl.
Unterdeckel. General German usage.

27. Interopercle. 1

Interopercular. \ General English usage.
Interoperculum. j

Inter-opercule. Cuvier, Agassiz.
Malleal. Geoffroy.

Zivischendeckel. General German usage.
Zivischenkiemendeckelstuck. Meckel, Erdl.

28. Articular. Owen, Parker.
Artictdaire. Cuvier, Agassiz.
Articulary. Giinther.

Ast des Unterkiefers^ ratnus ?nandibulce. Erdl.
Gelefikstiick des Unterkiefers. Meckel.
Os articulaire. Stannius, Kallmann.
Os condyloideiim. Franque.
Subtnalleal. Geoffroy.

29. Angular. Owen, Giinther, Parker.
Angiilaire. Cuvier, Agassiz.

5l6 STARKS

Os annulare. Stannius.
Sub-cotyleal. Geoffroy.

30. Dentary. Owen and most modern authors.
Dentaire. Cuvier, Agassiz.

Korper des U?iterkiefers^ corpus mandihula. Erdl.

Os dentate. Stannius, Kallmann.

Subdental. Geoffroy.

Zahnstiick des Unterkiefers. Meckel.

31. Maxillary. Owen, Parker and in general English use.
Addental. Geoffroy,

Maxilla superior. Kallmann, Erdl.
Maxillaire super ieur. Cuvier, Agassiz.
Os maxillare. Stannius.
Oberkieferbein. Meckel, Wagner, Erdl.

32. Premaxillary. Owen, Parker.
Adnasal. Geoffroy.

Inter 7naxilli7ics. Gill.
Intermaxillary . Owen.
Inter ?naxillaire. Cuvier, Agassiz.
Os intermaxillare. Stannius.

Zwischenkieferbein. Meckel, Wagner and most modern Ger-
man authors.
Zivischenkiefer, os incisivum. Erdl.

33. Interhyal. Parker, Gill.

Griffelforsdtz des Schlafensbein. Meckel, Wagner.

Griffelbeift.^ Vogt and Yung.

Os styloideuf?i. Stannius.

Styloide de I'os hyo'ide. Agassiz.

Styloide. Cuvier.

Stylohyal. Owen, Giinther, Geoffroy.

34. Epihyal. (Not differentiated by name from No. 35.) Owen,

Giinther, Parker.
Abgeplattetes Stuck des Zungetibogen. (Not differentiated by

name from No. 35.) Vogt and Yung.
Branche laterale. (Not differentiated by name from No. 35

Agassiz.
Grande piece laterale. Cuvier.

' Vogt and Yung give this term also to the inferior post clavicle No. 56.

SYNONOMY OF THE FISH SKELETON 517

Hinteres Stiick des grossest Zungenbeinhornes^ ossis hyoidei.

Erdl.
Uyo-sterftal. Geoffrey.
Zungen-horn oder Zungcn-bogen. (Not differentiated by name

from No. 35.) Meckel.

35. Ceratohyal. Owen, Giinther, Parker. Not ceratohyal of Geoff-

rey, No. 36.
Abgeplattetes Stiick des Zungenbogen. (Not differentiated by

name from No. 34.) Vogt and Yung.
Branche laterale. (Not differentiated by name from No. 34.)

Agassiz.
Grande piece laterale. (Not differentiated by name from No.

34.) Cuvier.
Uyfosternal. Geoffroy.
Zungen-horn oder Zungen-bogen. (Not differentiated by name

from No. 34.) Meckel.
Vorderes Stiick des grossen Zungenbeinhor?ies, Cornu majus.

Erdl.

36. Basihyals. Owen, Giinther, Parker. Not basihyals of Geoffroy,

No. 40.
Apohyal. Geoffroy.

Ceratohyal. Geoffroy. Not ceratohyal oi Owen, No. 35.
Gelenkstiick des Zungenbogens . Vogt and Yung.
Kleines Zu7igenbeinhor7i^ Cornu mimis ossis hyoidei. Erdl.
Petite piece laterale. Cuvier.
Tete glenoidale. Agassiz.

37. Glossohyal. Owen, Giinther. Not glossohyal of Geoffroy,

No. 39.
Cotes sternales. Geoffroy. Not cotes sternales of Cuvier, No.

69.
Entoglossutn. Vogt and Yung.
Os liizgual. Cuvier.
Os lingual s. entoglossum. Stannius.

38. Urohyal. Owen, Giinther, Parker. Not urohyal of Kallmann,

No. 40.
Basibranchiostegal. Parker.
Corps de Uhyoide. Agassiz.
Episternal. Geoffroy.
^ueue de Vos hyoi'de. Cuvier.
Zungenbei?i. Meckel.

5i8

STARKS

39. Branchiostegal. Owen, Gvinther, Parkei-.
Glosso-hyal. Geoffroy. ^ot glossohyal oi Owen, No. 37.
Kiejtienhautstr allien. Meckel.

Kiemenstrahlen. Erdl.

Rayons branchiosteges. Cuyier, Agassiz.

Radii branchiostegi. Stannius.

40. Basibranchials. Owen, Giinther, Parker.
Axial branchihyals. Cope.

Basi-hyal. Geoffroy. Not basihyal of Owen, No. 36.

Chaine intermediare d'osseletes. Cuvier.

Copula. Stannius, Vogt and Yung.

Ento-hyal. Kallmann.

^ueue de Vos hyoi'de. Agassiz.

Uro-hyal. Kallmann. Not urohyal of Owen, No. 38.

Roccus lineatus. Branchial arches.

41. Hypobranchials. Owen, Giinther, Parker.
Basal branchihyals. Cope.
Kietnenbogen. Meckel.

Piece articulaire. Agassiz.

Piece interne de partie inferieure de Varceau branchiale

Cuvier.
Thy real. Geoffroy.
Unteres Stiick des Kiemenbogens . Vogt and Yung.

42. Ceratobranchials. Owen, Giinther, Parker.
Lateral branchihyals. Cope.

Mittleres Stiick des Kiemetibogens. Vogt and Yund.
Piece branchiale. (Not differentiated by name from No. 43.)
Agassiz.

SYNONOMY OF THE FISH SKELETON 519

Piece externe de la partie inferleure de Varceau branchiale.

Cuvier.
Phireal inferieure. Geoffroy.

43. Epibranchials. Owen, Parker.

Oberes Stiick des Kiemenbogens. Vogt and Yung.
Partie superieure de Varceau branchiale. Cuvier.
Piece branchiale. (Not differentiated by name from No. 43.)

Agassiz.
Plureal superieure. Geoffroy.
Superior branchihyals. Vogt and Yung.

44. Suspensory pharyngeal. Starks.

Stylet de prcjniere arceau branchiale. Cuvier.
Upper epibranchial ofjirst brattchial arch. Owen.
VerbindungstUck. Vogt and Yung.

45. Upper or Superior pharyngeals. Parker, Giinther, Geoff roy, Cope.
Ep ipha ryngea Is. Gill.

Obere Schlundknocken. Meckel, Vogt and Yung.
Os pharyngien superieur. Agassiz.
Os pharyngeufn superius. Stannius.
Pharyngo-branchial. Owen.

46. Lower or Inferior pharyngeals. Owen, Giinther, Parker.
Cerato-branchial of fifth arch. Owen.
Hypopharyngeals. Gill.

Tnfrapharyngeals. Cope.

Ossa pharyngea inferiora. Stannius.

PharyngieJts inferieurs. Cuvier.

Pharyngeal arch. Owen.

Untere Schlundknochen. Vogt and Yung.

47. Gill-rakers. In general English use.
Kie7nenstechel. In general German use.

48. Supraorbital.^ Owen.

Sur-orbitatre. Cuvier, Agassiz.
Oberaugenhohlenbein. Meckel.
Oberajigenbein. In German use.

49. Preorbital. Gill.
Adorbital. Geoffroy.

1 This element not occurring in the species here figured it seems useless to
make a separate drawing from another species as the term is self-explanatory.

520

STARKS

Antorbital. Franque.

Lachrymal. Owen, Cuvier, Parker. Not lachrymal of Geoff-
rey, No. 3.
Thrdnenbein. Meckel.

50. Suborbitals. Owen.

Augenring^ ossa orbitalia (^Os zygomaticum). Erdl.

Infraorbitalring. Owen.

Jochbein. Meckel.

Jugal. Agassiz, Geoff roy. Not jugal of Cuvier, No. 19.

Ossa infraorbitilia. Stannius.

Ossa jugalia. Bojanus.

Sous-orbitaires. Cuvier.

Unteraugenhohlenbein. Wagner.

51. Nasal. Cuvier, Parker, Gill. Not nasal of Owen, No. 2.
Cartilage mobile du nez. Agassiz.

Ethmophysal. Geoff roy.
Nasenbein. Meckel, Erdl.
Os nasi. Erdl.
Os terminale. Stannius.
Turbinal. Owen, Gitnther.

52. Supra- temporal. Owen.
Augenboge}isch2ippe. Bojanus.
Sur-temporal. Cuvier.
Supralinear. Shufeldt.

53. Posttemporal. Parker, Giinther, Gill.
Oberes Schulierblatt^ scapula superior. Erdl.
Omolite. Geoff roy.

Schulterblatt. (Not differentiated by name from No. 54.)

Meckel.
Scaptclare. Vogt and Yung. Not scapula of Owen, No. 54;

nor of Parker, No. 57.
Supracleithra. (Not differentiated by name from No. 54.)

Gegenbaur.
Suprascapula. Owen.

54. Supraclavicle. Parker.

Coracoideu7n. Vogt and Yung. Not coracoid of Owen, No.

55; nor of Agassiz, No. 56; nor of Parker, No. 58.
Oi72oplate. Geoffroy.
Posterotemporal. Gill.

SYNONOMY OF THE FISH SKELETON 52I

Scapula. Owen, Stannius. ^ot scapular e oiYo%t and Yung,

No. 53 ; nor scapula of Parker, No. 57.
Scapulaire. Cuvier, Agassiz.
Supraclavicula. Parker, Giinther.

Schulterblatt (not differentiated by name from No. 53) . Meckel.
Supracleithra (not differentiated by name from No. 53). Gegen-

baur.
Unteres Schulterblatt., scapula i?iferior. Erdl.

55. Clavicle. Parker.

Clavicula. Stannius, Parker, Giinther,

Clavicula anterior. Erdl.

Cleitkru?n. Gegenbaur.

Ccenosteon.^

Coracoid. Owen. Not coracoid of Agassiz, No. 56 ; nor of

Parker, No. 58 ; nor coracoideum of Vogt and Yung, No. 54.
Hufneral. Cuvier. Not humeral of Agassiz, No. 59; nor

Htanerus of Geoffroy, No. 58; nor of Erdl, No. 56.
Proscapula. Gill.

Vordere Schliisselbein. Meckel, Erdl.
Vordere clavicida. Erdl.

56. Postclavicle. Parker.
Accessorisches Stiick. Gegenbaur.

Basal superius (the upper one). Vogt and Yung.

Coracoid. Agassiz. Not coracoid of Owen, No. 55 ; nor of

Parker, No. 58; nor Coracoideum of Vogt and Yung, No. 54.
Epicoracoid. Owen.

Griffelbeiji"^ (the lower one). Vogt and Yung.
Hintere Schliisselbein. Meckel.
Humerus. Erdl. Not hu?7tcrus of Geoffroy, No. 58 ; nor of

Owen, No. 59; nor humeral oi Cuvier, No. 55.
Oberar7nbein. Erdl.
Os coraco'idiefi. Cuvier.
Postclavicula. Parker, Giinther.
Problematischer Skelettheil. Gegenbaur.
Teleotefnporals. Gill.

57. Hypercoracoid. Gill.

Basale medium. Vogt and Yung.

'I get this reference from Owen's Comp. Anat. Lectures (Vertebrates),
p. 118. " By some Ichthjotomists the bone in question has received the special
name of ' coenosteon.' "

2Vogt and Yung also give this term to the interhyal (No. 33).

522 STARKS

Oberes Stiick {^scapular e). Gegenbaur.

Ossa carpi (not differentiated by name from No. 58). Stannlus.

Radial. Cuvier, Agassiz. Not radius of Owen, No. 58; nor
of Geoffroy, No. 60.

Scapula. Parker. Not scapula of Owen, No. 54 ; nor scapu-
lar e of Vogt and Yung, No. 53.

Ulna. Owen,

Vorderartnknochen (not differentiated by name from No. 58).
Meckel.

Vorderarmbein^^ antibrachium. Erdl.

58. Hypocoracoid. Gill.

Basale inferius. Vogt and Yung.

Coracoid. Parker, Giinther. Not coracoid of-Owen, No. 55;

nor of Agassiz, No. 56 ; nor coracoideutti of Vogt and Yung,

No. 54.
Cubital. Cuvier, Agassiz. Not c«^/V?^5 of Geoffroy, No. 60.
Minteres Schliisselbein, clavicula posterior. Erdl. Not Aint ere

Sc/ililsselbein of Meckel, No. 56.
Humerus. Geoffroy. Not humerus of Owen, No. 59; nor

humeral oi Cuvier, No. 55.
Ossa carpi (not differentiated byname from No. 57). Stannius.
Radius. Owen. Not radius of Geoffroy, No. 60; nor radial

of Cuvier, No. 57.
Vorderarjuknochen (not differentiated by name from No. 57).

Meckel.
Vorderes stiick i^Procoracoid'). Gegenbaur.

59. Mesocoracoid. Gill.

Humerus. Owen. Not humerus of Geoffroy, No. 58 ; nor of

Erdl, No. 56.
Humeral. Agassiz. Not humeral oi Cuvier, No. 55.
Precoracoid. Parker.
Spangenstiick. Gegenbaur.
Proisieme os de rava?it bras qui porte la nageoire pectorale.

Cuvier.

60. Actinosts. Gill.
Branchial ossicles. Parker.
Basalia. Huxley, Giinther.
Basalstiicke der Brustjldsse. Gegenbaur.

1 " Durch ein Loch in ihm ist die Abtheilung in eine grossere und kleinere
Portion — ulna und radius — angedeutet." Erdl.

SYNONOMY OF THE FISH SKELETON

523

Basal pectoral radii. Cope.
Carpals. Owen, Giinther.
Carpus. Vogt and Yung.

Fig. 46. Ictiobus bubalus. Inner view of shoulder girdle.

Carpe and Metacarpe. Agassiz.
Mittelhandknochen., metacarpus.
Os du carpe. Cuvier.
Ossa metacarpi. Stannius.
Radius and cubitus. Geoffroy.

nor radial of Cuvier, No. 57

58.

Erdl.

Not radius of Owen, No. 58 ;
nor cubital of Cuvier,'; No.

61. Pectoral rays. In general English use.
Brustjlossenstrahlen. Stannius.
Carpe afid Phalanges. Geoffroy.
Aletacarpo-phalanges. Owen.
Rayons de la pectorale. Cuvier.

62. Pelvic girdle. In general use.
Becken. Stannius.

Pubic cones. Owen.

63. Ventral spine. In general use.
Bauchjlossefi Stachel. Stannius.

64. Abdominal vertebrae. Owen and in general use.
Bauckwirbel. Stannius.

Vertebres abdofninales. Cuvier.

Proc. Wash. Acad. Sci., November, 1901.

524 STARKS

65. Caudal vertebrae. Owen and in general English use.
Schivajizwirbel. Stannius.

Vertebres caudales. Cuvier.

66. Centrum. In general use.
Corps de vertebre. Cuvier.
Cycleal. Geoffroy.
Cyclo-vertebral element. Grant.
Tertiar-wirbel. Carus.
Wirbel-korper. Miiller, Wagner.

67. Neurapophyses. Owen.
Bogenstiicke des Riickenwirbels. Carus.
Deckplatteji. Carus.
Grufzdplatten. Carus.

Obere Wirbelbogen. Wagner.
Neural process or neural arch. Parker.
Parte an7iulaire. Cuvier.
Perial. Geoffroy.
Peri-vertebral elements. Grant.

68. Neural spine. Owen, Parker.
Apophyse epineuse. Cuvier.
Epial. Geoffroy.
Epi-vertcbral elements. Grant.
Oberer Dorn-forsatz. Carus.

69. Haemapophyses. Owen, Vogt and Yung.
Bogenstiicke des Bauchwirbel. Carus.
Cata-vertebral elejnents.^ Grant.
Cataal. Geoffroy.

Cotes ster?iales. Cuvier. Not cotes ster?iales of Geoffroy, No

37-
Hcemal processes or ha:j7ial arch. Parker.

Unter-sternal-theil des Urwirbelbogens. Carus.

70. Haemal spine. Owen, Parker.
Sternal Wirbel Korper. Carus.
Unterer Dorn-forsatz. Carus.
Untere Bogen. Gegenbaur.

71. Zygapophyses. Owen, Parker.
Aphophyse articulaire. Cuvier.
Gelenk-forsatz. Miiller, Wagner.

1 Used also for ribs, No. 73, b}' Grant.

SYNONOMY OF THE FISH SKELETON 525

Schiefer Forsatz. Vogt and Yung.
Seitlicher Tertiar-ivirbel. Cai'us.

72. Parapophyses. Owen, Gill.
Apophyse transverse. Cuvier.
Hcemapophyse.^ Vogt and Yung.
Paraal (also used for rib by this author). Geoffroy.
Para-vertebral elements. Grant.
^uerforsatz. Carus.
Rippe Gelenkstuck. Erdl.
Transverse process. Parker.
Untere querforsatz. Wagner.

73* Rib. In general use.

Cata-vertebral elements (used also for JK^mapophyses., No. 69,

by this author) . Grant.
Costal process. Owen.
Cotes vertebrales. Cuvier.
Pleurals. Gill.
Paraal. Geoffroy.
Pleur apophyses. Owen.
Rippen. Stannius.
Riickentheil tmd Ober-sternal-tkeil des Uriuirbelbogens. Carus.

74. Epipleural spines or Epipleurals. Owen, Gill, Giinther.
Intermusculars. Parker.

Muskel- Grdthen. Stannius.
Stylets. Cuvier.

75. Interneural spines or Interneurals. Owen, Parker.

Flossen-trceger . Stannius.

Plossenstrahltrager . Gegenbaur.

Innere Flossenstrahl. Erdl.

Tnterepineux. Cuvier.

Obere-rippe. Meckel.

Ossa interspinalia s. obere Flossen-trceger. Stannius.

76. Dorsal rays and spines. In general English use.
JDermoJieural spines. Owen.

Epines et rayotis dorsales. Cuvier.
Riicken-Jlossen strahle?i und stechehi. Stannius.

' Vogt and Yung do not differentiate byname the inferior processes of the
abdominal vertebrae (^parapophyses) from those of the caudal vertebriE (^hcema-
pophyses).

526 STARKS

Weichstrahlen und Stechelstrahlen der Ruckenjlosse. Gegen-
baur.

77. Interhaemal spines or Interhaemals. In general English use.
Apophyses epineuses ittferietires. Cuvier.
Unterjlossen trceger. Stannius.

78. Anal rays and spines. In general use.

Weichstrahlen und Stechelstrahlen des Afterjlosse. Gcgen-
baur.

79. Hypural. Huxley, Parker, Giinther.
Aggregated interhcemals. Owen.
Plaque triangulaire et vertlcale. Cuvier.

Verticale Platte. Stannius.

80. Caudal-rays. In general English use.
Schivanzjlossen Strahlen. Stannius.

SYNONOMY OF THE FISH SKELETON

527

ALPHABETICAL LIST OF TERMS.

Synonym of:

Abdominal vertebrae

Abgeplattetes Stiick des Zun
genbogen

Abgeplattetes Stiick des Zuri'
genbogen

Accessorisches Stuck

Actinosts

Addental

Adgustal

Adnasal

Adorbital

Aggregated interhsemals

Aile orbitaire

Ala magna sphenoidei

Ala magna ossis sphenoidei .

Ala minor ossis sphenoidei..

Ala orbitalis

Ala parva sphenoidei

Ala temporalis

Alisphenoid

Alisphenoid

Anal rays and spines

Angulaire

Angular

An teal

Antibrachium

Antorbital

Apohjals

Apophjse articulaire

Apophyses dpineuses

Apophyses epineuses in-
fer ieures

Apophyse styloide du temporal

Apophyse transverse

Appendices

Articulaire

Articular

Articulary

Ast des unterkiefers

Augenbogenschuppe

Augenring

Axial branchihyals

Epihyal.

Ceratohyal . .
Postclavicle

Maxillary

Pterygoid

Premaxillary .

Preorbital

Hypural

Alisphenoid ..
Alisphenoid ..

Sphenotic

Alisphenoid ..
Alisphenoid ..
Basisphenoid.
Prootic

Basal branchihyals

Basale superius

Basale inferius

Basale medium

Basalia

Basal pectoral radii

Basalstiicke der Brustflosse.

Basibranchials

Basibranchiostegal

Basi-hyals

Basihyals

Basilaire

Basilaris ossis occipitis

Prootic...
Angular.

Vomer

Hypercoracoid

Preorbital

Basihyals'

Zygapophyses.
Neural spines...

Interhaemal spines.

Preopercle

Parapophyses

Epipleurals

Articular

Articular

Articular

Supratemporal.

Suborbitals

Basibranchials..

Hypobranchials.

Postclavicle

Hypocoracoid ...
Hypercoracoid. .

Actinosts

Actinosts

Actinosts

Urohyal

Basibranchials

Basioccipital
Basioccipital

64

34

35
56
60

31
20

32
49
79
16
16

5

16
16
14
15
16

15
78
29
29
I
57
49
36

71
68

77

24

72

74
28
28
28
28
52

50
40

41
56
58
57
60
60
60
40
38
40
36
12
12

Page.

523
516

517
521
523
516
513
516

519
526
512
512
509
512
512
512
512
512
512
526
515
515
508
522
519
517
524
524

526
515
525
525
515
515
515
515
520
520
518

518
521
522
522
523
523
523
518
517
518
517
5"
5"

528

STARKS

Terms.

Basiltheil des hinterhauptbeins

Basioccipital

Basisphinal

Basisphenoid

Basisphenoid

Bauchflossen Stachel

Bauchwirbel

Becken

Bogenstiicke des Bauchwirbel
Bogenstiicke des Riicken-

wirbels

Branche laterale

Branche laterale

Branchial ossicles

Branchiostegals

Brustflossenstrahlen

Caisse

Cartilage mobile du nez

Carpals

Carpe and Metacarpe

Carpe and Phalanges

Carpus

Cataal

Cata- vertebral elements

Cata- vertebral elements

Caudal rays

Caudal vertebrae

Centrum

Ceratobranchial

Ceratobranchial of fifth arch..

Ceratohyal

Ceratohyal

Chaineintermediaired'osselets

Clavicle

Clavicle anterior

Clavicula

Clavicula posterior

Cleithrum

Coenosteon

Condjloidea ossis occipitis

Copula

Coracoid

Coracoid

Coracoid

Corocoideum

Cornu majus

Cornu minus ossis hjoidei

Corps de I'hyoide

Corps de vertebre

Corpus mandibulas

Corpus ossis occipitis

Corpus ossis sphenoidei

Costal process

C6tes sternales

C6tes sternales

C6tes vertdbrales

Crista ethmoidei

Synonym of :

Basioccipital
Basioccipital

Parasphenoid

Ventral spine

Abdominal vertebrae .

Pelvic girdle

Hiemapophyses

Neurapophyses

Ceratohyal

Epihyal

Actinosts

Pectoral rays

12
12
12

H
13
63
64
62
69

67
35
34
60

39

61

23
51
60
60
61
60
69

73
69
80

65
66
42
46
35
36
40

, ^^
Clavicle I 55

Metaptervgoid .

Nasal

Actinosts

Actinosts

Pectoral rays....

Actinosts

Haemapophyses

Ribs

Ha;mapophyses ,

Inferior pharyngeals

Basihyal

Basibranchials.

Clavicle
Hypocoracoid ...

Clavicle

Clavicle

Exoccipital

Basibranchials .
Hypocoracoid ...

Clavicle

Post clavicle

Supraclavicle....

Ceratohyal

Basihyal

Urohyal

Centrum

Dentary

Basioccipital

Parasphenoid....

Rib

Glossohyal

H aemapophyses

Ribs

Ethmoid

55
58
55
55
II
40
58
55
5<^
54
35
36
38
66

30

12

13
73
37
69

73

Page.

5"
5"
5"

512

5"
523
523
523
524

524
517
516
523
518
523

514
520

523
523
523
523
524
525
524
526
524
524
5i8
519
517
517
518
521
521
521
522

521
521
5"
518
522
"521
521
520
517
517
517
524
516
5"
5"
525
517
524
525
509

SYNONOMY OF THE FISH SKELETON

529

Terms.

Cubital

Cjcldal

Cjclo-vertebral element.

Deckplatten ,

Dentaire ,

Dentary

Dermoneural spines...

Dichost

Dorsal rays and spines.

Ecaille du temporal

Ecto-ethmoid

Ectopterygoid

EngentlichKiemendeckelstuck

Entoglossum

Ento-hjal

Entopterjgoid

Entosphenal

Epial

Epibranchials

Epicoracoid

Epicotjl^al

Epihyal

Epines et rajons dorsales

Epiotic

Epioticum

Epipharyngeals

Epipleurals

Episternal

Epitjmpanic

Epi-vertebral elements

Erganzungstiick oder Fliigel-

fortsatz

Ethmoid

Ethmoid

Ethmoide cr&nien..

Ethmoideum

Ethmoideum

Ethmoideum cribrosum

Ethmophysal

Ethmo-turbinal

Exoccipital

Exoccipital

Ex-rupdal

Externum

Synonym of :

Hypocoracoid

Centrum

Centrum

Neurapophyses
Dentary

Dorsal rays and spines.
Basisphenoid

Felsenbein

Felsenbein

Felsentheil

Felsentheil des Schlafenbeines

Flossenstrahltrager

Flossen-trager

FlUgelbein

Frontal

Frontal anterieur

Frontalis ossis frontis

Frontal posterieur

Pterotic

Prefrontal

Pterygoid

Opercle

Glossohyal

Basibranchial...
Mesopterygoid.
Basisphenoid...
Neural spine ...

Postclavicle

Metapterygoid.

Dorsal spines and rays.

Epiotic

Superior pharyngeals.

Urohyal

Hyomandibular
Neural spines....

Mesopterygoid.

Basisphenoid.
Basisphenoid.

Ethmoid

Alisphenoid .,
Prefrontal . ..

Nasal

Basisphenoid.

Epiotic

Pterotic ...
Pterygoid

Prootic

Opisthotic

Pterotic

Prootic

Interneural spine ,
Interneural spine .
Quadrate

Prefrontal.

Frontal

Sphenotic,

Page.

.ss

522

66

524

66

524

67

524

30

51b

.30

5 It)

76

525

H

512

76

525

9

510

3

509

20

513

25

515

37

517

40

5i«

22

514

14

512

68

524

4.^

519

56

521

23

515

34

5i<5

76

525

7

510

7

510

45

519

74

525

3«

517

17

513

68

524

22

514

2

509

H

512

14

512

2

509

16

512

3

509

51

520

14

512

11

511

7

510

9

510

20

513

1.5

512

10

511

9

510

15

512

75

525

75

525

19

513

4

509

3

509

4

509

5

509

530

STARKS

Frontal principal.

Gaumenbein

Gaumenbein

Gaumenfliigel des Keilbein

Gehdrdeckel

Gelenk-forsatz

Gelenkstuck des Unterkiefers..
Gelenkstiick des Zungen-

bogens

Gelenktheil

Gelenktheil des Hinterhaupt-

beins

Gelenktheil des Schlafenbeins

Gill-rakers

Glossohyal

Glosso-h jals

Grande aile du sphenoide

Grandes pieces laterales

Grandes pieces laterales

Griffelbein

Grilifelbein

Griffelfdrmig Knoche

Griffelforstitz des Schliifens-

bein

Grosse Keilbeinfliigel

Grundplatten

Synonym of:

Frontal .

Palatine

Pterygoid

Metapterjgoid
Pterotic

Zygapophjses.
Articular

Basihjal ....
Exoccipital.

Exoccipital

Hvomandibular,

Branchiostegals

Prootic

Epihyal

Ceratohjal

Post clavicle

Interhyal

Symplectic

Haemal arch

Hieinal process

Haemal Spine

Hajmapophyse

Haemapophyses

Herissdal

Hintere Schliisselbein

Hintere Schliisselbein

Hinterer Schljifenfliigel

Hinteres Stiick des grossen

Zungenbein homes

Hinterhauptbeinkorper

Hinterhauptschuppe

Hinterstuck des os quadratum

Humeral

Humeral

Humerus

Humerus

Humerus

Hyomandibular

Hyo-sternal

Hypercoracoid

Hypobranchials

Hypocoracoid

Hypocotyleal

Hypopharyngeals

Hyposphenal

Hyposternal

Hypotympanic

Hypural

Interhyal

Alisphenoid

Neurapophj'ses .

Haemapophyses
H aemapophyses

Parapophyses ,

Mesapterygoid.

Postclavicle

Hypocoracoid ..
Prootic.

Epihyal

Basioccipital...,
Supraoccipital.
Metapterygoid.
Mesocoracoid .

Clavicle

Postclavicle

Mesocoracoid .
Hypocoracoid .

Epihyal.

Quadrate

Lower pharyngeals'

Parasphenoid

Ceratohyal

Qiiadrate

21
20
23
9
71
28

36
II

II
17
47
37
39
15
34
35
.';6

33
18

33
16
67

69
69
70

72
69
22
56
5S
15

34
12

23

59

55
56
59
58
17
34
57
41
58
19
46

13
35
19
79

Page.

509

514
513
514
510

524
515

517
5"

5"
513
519
517
518
512
516
517
521
516

513

516
512
524

524
524
524
525
524
514
521
522
512

516
511
510

514
523
521

521
523
522
513
516
522
518
522
513
519
5"
517
513
526

SYNONOMY OF THE FISH SKELETON

531

Terms.

Inc^al

Inferior pharyngeals .

Inf raor bitalri ng

Inf rapharyngeals

Ingrassial

Innere Flossenstrahl.

In-rupdal

Intercalare

Interepineux

Interhaetnals

Interhyal

Intermaxillaire

Intermaxillary

Intermaxillines ,

Intermusculars ,

Interneurals

Interopercle ,

Inter-opercule

Interoperculum

Interparietal

Jochbein

Jochfortsatz.

Jugal

Jugal

Keilbeinflligel

Keilbeinkorper

Kiemenbogen

Kiemendeckel

Kiemenhautstrahlen

Kieinenstrahl

Kiemenstechel

Kleine Fliigel des Keilbeins.

Kleines Zungenbeinhorn

Kdrper des Unterkiefers

Lachrymal

Lachry m al

Lateral branchihyals.
Lateral ethmoid

Malleal

Mastoid

Mastoideum

Masto'idien

Maxillaire superieur.

Maxillary ,

Maxilla superior

Mesocoracoid

Mesopterygoid

Mesotympanic

Synonym of :

Subopercle.

Suborbitals

Lower pharyngeals

Alisphenoid

Interneural

Opisthotic

Opisthotic

Interneurals

Premaxillary
Premaxillary
Premaxillary
Epipleurals...

Interopercle ...
Interopercle . .
Supraoccipital

Suborbitals

Quadrate

Suborbitals
Quadrate....

Pterygoid

Parasphenoid...
Hypobranchial

Opercle

Branchiostegal.
Branchiostegal.

Gill-rakers

Alisphenoid

Basihyal

Dentarj'

Preorbital

Prefrontal

Ceratobranchials.
Prefrontal

Interopercle

Pterotic

Opisthotic

Hyomandibular
Maxillary

Maxillary ,

Symplectic

Metacarpo-phalanges Pectoral rays.

Metacarpus Actinosts

Metapterygoid

Mittelhandknochen Actinosts

Mittlerer Stirnbein Frontal

Mittleres ethmoideum Ethmoid

No.

Page.

26

515

46

519

50

520

46

519

16

512

7.?

525

10

511

10

511

1^

525

77

526

?,?>

516

?>2

516

33

51b

32

51b

74

525

75

52s

27

515

27

515

27

515

8

510

.SO

520

19

513

SO

520

19

513

20

513

13

511

41

5i«

25

515

7,9

5i»

.39

5i«

47

519

16

512

36

517

30

51b

49

519

3

509

42

5i«

3

509

27

51S

9

510

10

511

17

513

31

516

31

516

31

516

59

523

22

514

18

513

61

523

60

523

23

514

60

523

4

509

2

509

532

STARKS

Terms.

Mittleres Stuck des Kiemen-

bogens

Muskel Grathen

Nasal

Nasal

Nasalis ossis frontis ,

Nasenbein

Nasentheil

Nasentheil des Stirnbeines.

Neural-arch

Neural process ,

Neural spines

Neurapophyses

Synonym of:

Ceratobranchials.
Epipleurals

Oberarmbein

Oberaugenbein

Oberaugenholenbein

Obere Gelenkbeln

Oberer Dorn-forsatz

Obere rippe

Obere Schlundknochen

Oberes Schulter blatt

Oberes Stiick

Oberes Stiick des Kiemenbo-

gens

Obere Wirbelbogens

Oberflachliche Knochen-lam-

elle

Oberkieferbein

Occipital externe

Occipitalis ossis occipitis

Occipital laterale

Occipital supcirieure

Omolite

Omoplate ■

Opercle

Operculaire

Operculum ,

Operculum pori acustici

Opisthotic ,

Opisthoticum

Orbitalfliigel des Keilbeins

Orbitosphenoid ,

Os angulare ,

Os articulare

Os basilare ,

Os carre

Os condyloideum

Os coracoidien

Os dentale

Os du carpe

Os ethmoideum

Os ethmoideum

Os extrascapulare

Os f rontale

Os frontale anterius

Os frontale posterius

Ethmoid

Ethmoid ,

Nasal

Ethmoid

Ethmoid

Neurapophyses
Neurapophyses

Postclavicle

Supraorbital

Supraorbital

Hyomandibular

Neural spines

Interneurals

Superior pharyngeals ,

Posttemporal

Hypercoracoid

Epihyals

Neurapophyses

Opisthotic

Maxillary

Epiotic

Supraoccipital

Exoccipital

Supraoccipital
Posttemporal ..
Supraclavicle..

Opercle .
Opercle .
Pterotic.

Pterotic

Prefrontal ...
Alisphenoid .

Angular

Articular

Basioccipital.

Quadrate

Articular

Postclavicle ..

Dentary

Actinosts

Ethmoid

Prefrontal

Pterotic

Frontal

Prefrontal ...
Sphenotic

Page.

42

51S

74

525

51

520

2

509

2

509

51

520

2

509

2

509

67

524

67

524

68

524

67

524

56

521

4«

519

48

519

17

513

68

524

75

525

4.5

519

53

520

57

522

43

519

67

524

10

511

31

516

7

510

8

510

11

511

8

510

53

520

54

520

25

515

25

515

25

515

9

510

10

5"

9

510

3

509

16

512

29

515

28

515

12

511

19

513

28

515

56

521

30

516

60

523

2

509

3

509

9

510

4

509

3

509

5

509

SYNONOMY OF THE FISH SKELETON

533

Terms.

Os incisivum

Os innominatum

Os innominatum

Os intermaxillare

Os interparietale

Os lingual

Os lingual s. entoglossum...

Os mastoideum

Os mastoideum

Os maxillare

Os nasi

Os occipitale laterale

Os occipitale superius

Os occipital externum

Os palatinum

Os palatinum

Os parietale

Os pharjngien supdrieur...
Os pharjngeum superius....

Os pterjgoideum

Os quadratojugale

Os quadratum

Os quadratum s. tjmpanicum

Ossa carpi

Ossa carpi

Ossa infraorbitalia

Ossa interspinalia s. obere

Flossen-trager

Ossa jugalia

Ossa metacarpi

Ossa orbitalia

Ossa pharjngea inferiora

Ossis hvoidei

Os sphenoideum anterius

Os sphenoideum basilare

Os styloideum

Os sjmplecticum

Os temporale

Os terminale

Os transversum

Os transversum s. pterygoi-

deum externum

Os tjmpani

Os tjmpanicum

Os zygomaticum

Otosphenal

Otosteal

Synonym of:

Palatin

Palatine

Paraal

Paraal

Parapophyses

Parasphenoid

Para-vertebral elements.

Parethmoid

Parietal

Paroccipital

Premaxillary

Basisphenoid

Opisthotic

Premaxillary ,

Epiotic

Glossohjal

Glossohyal

Pterotic

Epiotic ,

Maxillary

Nasal

Exoccipital

Supraoccipital

Epiotic

Pterygoid

Palatine

Parietal

Superior pharyngeals
Superior pharyngeals

Mesopterygoid

Qiiadrate

Quadrate

Hyomandibular

Hypercoracoid

Hypocoracoid

Suborbitals

Interneurals

Suborbitals

Actinosts

Suborbitals

Lower pharyngeals

Epihyal

Basisphenoid

Parasphenoid

Interhyal

Symplectic

Hyomandibular

Nasal

Mesopterygoid

Pterygoid

Metapterygoid.,
Metapterygoid.

Suborbitals

^ Basioccipital....
Opisthotic

Palatine

Parapophyses
Ribs

Parapophyses.
Prefrontal

Epiotic.

Page.

32

516

14

512

10

511

32

516

7

510

37

517

37

517

9

510

7
31

51

510
516
516

II

Sii

S

510

7

510

20

513

21
6

510

45

519

45

519

22

514

19

513

19

513

17

513

57
58

1522
522

50

520

75

525

60

520
523

50
46

520
519

34

51b

14

512

13

511

33
18

51b
513

17

.SI 3

51

520

22

514

20

513

23

5H

23

514

50

520

12

5"

10

511

21

514

21

514

72

525

73

525

72

525

13

5JI

72

525

3
6

509
510

7

510

534

STARKS

Terms.

Synonym of:

Parte annulaire

Partie superieure de I'arceau

branchiale

Paukering-Knochen

Pectoral rays

Pelvic girdle

Perial

Peri-vertebral elements

Petite pi^ce laterale

Petrosa

Petrosal

Petrosal

Petrosum

Pflugschar

Pharyngeal arch

Pharyngiens inferieurs

Phary ngo-branchial

Piece articulaire

Piece branchiale

Pifece branch iale

Pifece externe de partie inferi-

eure de I'arceau branchiale..
Pi^ce interne de partie inferi- |

eure de I'arceau branchiale..
Plaque triangulaire et verticale

Pleurals

Pleurapophyses

Plureal inferieure '

Plureal superieure

Postclavicle

Postclavicula

Posterotemporal

Post frontal

Posttemporal

Precoracoid

Prefrontal

Premaxillary

Preopercle

Preopercule

Preoperculum

Preorbital

Presphenoid

Pretym panic

Problematischer Skelettheil ...

Procoracoid

Progressus ascendens piscium

Prootic

Prooticum

Proscapula

Prosethmoid

Ptereal

Pterotic

Pterygoid

Pterygoideum externum

Pterygoideum internum

Pterygoideum posterius

Pterygoidien interne

Pubic bones

Neurapophyses .

Epibranchials ...
Hyomandibular.

Neurapophj'ses ,

Neurapophyses

Basihyal

Prootic

Prootic

Opisthotic

Prootic

Vomer

Lower pharyngeals ...
Lower pharyngeals ...
Superior pharyngeals

Hy pobranchial

Ceratobranchial

Epibranchial

Ceratobranchials.

Ilypobranchials...

Ilvpural

Ribs

Ribs

Ceratobranchials.
Epibranchials

Postclavicle ...
Supraclavicle.
Sphenotic

Mesocoracoid

Preopercle.
Preopercle.

Parasphenoid .
Metapterygoid.

Postclavicle

Hypocoracoid .
Parasphenoid ..

Prootic ..
Clavicle ,
Ethmoid ,
Prootic...

Pterygoid

Mesopterygoid.
Metapterygoid..
Mesopterygoid.
Pelvic girdle

67

Page.
524

43

519

17

513

61

523

62

523

67

524

67

524

36

517

I.";

5'2

I. 5

512

ID

5"

1";

512

I

508

46

519

46

519

45

519

41

5i«

42

5i«

43

519

42

518

41

518

79

526

73

525

73

525

42

518

43

519

.S6

.521

.56

521

54

520

5

5"

53

520

59

523

3

509

32

520

24

515

24

515

24

515

49

519

13

511

23

514

56

521

5«

522

13

5"

15

512

15

512

55

521

2

509

15

512

9

510

20

513

20

513

22

514

23

514

22

514

62

523

SYNONOMY OF THE FISH SKELETON

535

Terms.

Quadrate

Querforsatz

Queue de I'os hyoide.
Queue de I'os hyoide.

Radial

Radii branchiostegi

Radius

Radius and Cubitus

Ramus mandibul«

Rayons branchiosteges.
Rayons de la pectorals .

Rhinosphenal

Rib

Riechbein

Riechbeinkorper ,

Rippe Gelenkstiick ,

Rippen

Rocher

Rocher rudimentaire

Riicken-flossen strahlen u

stecheln

Riickentheil und Ober-sternaL

theil des Urwirbelbogens...

Scapula

Scapula

Scapula inferior

Scapulaire

Scapulare

Scapula superior

Scheibenfdrmig Knoche.........

Scheitelbein

Schiefer Forsatz

Schlafbeinschuppen

Schulterblatt

Schuppentheil des Hinter
hauptbeins

Schuppentheil des Schlafbein

Schuppentheil des Schlafens
beines

Schwanzwirbel

Schwanzflossen Strahlen ,

Seibbein

Seitlichen obern Hinterhaupt
beine

Seitlichen Richtbein

Seitlichen untern Hinter-
haupt beine

Seitlicher Tertiar-wirbel

Serrial

Shulterblatt

Sous-opercule -

Sous-orbitaires

Spangenstiick

Sphenoide anterieur

Sphdnoide principal

Sphenoideum basilare

Synonym of :

Parapophyses..
Basibranchials,
Urohyal

Hypercoracoid .,
Branchiostegals
Hypocoracoid ....

Actinosts

Articular ,

Branchiostegal .,

Pectoral rays

Vomer

Prefrontal

Ethmoid ,

Parapophyses.

Ribs

Opisthotic ,

Opisthotic ,

Dorsal rays and spines.
Rib

Supraclavicle...
Hypercoracoid
Supraclavicle...
Supraclavicle...
Posttemporal...
Posttemporal...
Metapterygoid.

Parietal

Zygapophyses.

Sphenotic

Posttemporal...

Supraoccipital.
Pterotic

Hyomandibular..
Caudal vertebrae

Caudal rays

Ethmoid

Epiotic

Prefrontal.

Exoccipital

Zygapophyses ..
Hyomandibular

Supraclavicle

Subopercle

Suborbitals

Mesocoracoid ...
Basisphenoid....
Parasphenoid ...
Parasphenoid ...

19

72
40
38

57
39
.^S
60
28

39

61

I

73

3

2

72
73
10
10

76
73

Page.

513
525
518
517

522
518
522
523
515
518

523
!;o8

525
509
509

525
.'525
5"
5"

525

525

54

520

57

522

54

520

54

520

53

520

53

520

23
6

514
510

71

524

5

509

53

520

8

520

9

520

17
65
80

513

524
^26

2

509

7

510

3

509

II

511

71

524

17

513

54
26

520

515

50

520

59

523

14

512

13

5"

13

5"

536

STARKS

Terms.

Sphenoid posterieur

Sphenotic

Squama occipitalis

Squama temporis

Squamosal

Squamosa ossis temporum ,

Stapeal

Sternal wirbel Korper

Stirnbein

Stylet de premiere arc(

branchiale

Stjlets

St jloh y al

Stjloide

Styloide de I'os hjoide

Sub-cotyleal

Sub-dental

Submalleal

Subopercle

Suboperculum

Suborbitals

Subtectals

Superior branchihyals

Superior pharyngeals

Supraclavicle

Synonym of:

Parasphenoid

Supraoccipital ..

Pterotic

Pterotic ,

Hyomandibular.

Opercle

Haemal spine

Frontal

Suspensory pharj^ngeal

Epipleurals

Interhyal

Interhyal

Interhyal

Angular

Dentary

Articular

Subopercle.

Alisphenoid ...
Epibranchials

Supraclayicula

Supracleithra

Supracleithra

Supralinear

Supraoccipital

Supra-occipitine

Supraorbital

Suprascapula

Supratemporal

Suroccipital

Sur-orbitaire

Sur-temporal

Suspensory pharyngeal.

Symplectic

Symplectique

Supraclavicle...
Posttemporal...
Supraclavicle...
Supratemporal.

Supraoccipital.
Posttemporal..

Exoccipital

Supraorbital....
Supratemporal.

Symplectic.

Teleotemporals

Temporal

Temporal

Temporale

Tertiar-wirbel

T6te glenoidale

Thranenbein

Thriinenbein

Thyr^al

Transverse

Transverse process

Troisieme os I'avant bras
qui porte I'nageoire pec-
torale

Tur binal

Tympanal

Tympanal

Postclavicle

Sphenotic

Hyomandibular

Pterotic

Centrum

Basihyal

Prefrontal

Preorbital

Hypobranchials

Pterygoid

Parapophyses ...

Mesocoracoid..,

Nasal

Preopercle

Metapterygoid.

Page.

13

5"

5

509

8

510

9

510

9

510

17

513

25

515

70

524

4

509

44

519

74

525

33

516

33

51b

33

51b

29

515

^o

516

28

515

26

515

26

515

50

520

16

512

43

519

45

519

54

520

54

520

53

520

54

520

52

520

8

510

8

510

48

519

53

520

52

520

II

511

48

519

52

520

44

519

18

513

18

513

56

521

5

509

17

513

9

510

66

524

36

517

3

509

49

519

41

5i«

20

513

72

525

59

523

51

520

24

515

23

514

SYNONOMY OF THE FISH SKELETON

537

Terms.

Tjmpano-malleal

Ulna

Unteraugenhohlenbein

Unterdeckel

Untere Bogen

Untere Dorn-forsatz

Untere keilbeinfliigel

Untere querforsatz

Untere Schlundknochen

Un teres Schulter blatt

Unteres Stiick des Kiemenbo-

gens

Unterflossen triiger

Unter Gelenkbein

Unterkiemendeckelstuck

Unter-sternal theil des Urwir-

belbogens

Upper epibranchial of first

branchial arch

Urohyal

Uro-hjal

Uro-serr ial

Synonym of:

Symplectic

Hjpercoracoid

Suborbitals

Subopercle

Hsemal spines

Haemal spines

Mesopterygoid

Parapophjses

Lower Pharyngeals.
Supraclavicle

Hypobranchials.

Interhaemals

Quadrate

Subopercle

Hsemapophyses ....
Suspensory pharyn

Verbindungstiick

Ventral spine

Vertebres abdominales

Vertebres caudales

Verticale Platte

Vomer

Vorderarmknochen

Vorderar mkn ochen

Vorderar mbein

Vorder deckel

Vordere clavicula

Vordere Schlaf enfliigel

Vordere Schliisselbein

Vorderes Stiick

Vorderes Stiick des grossen

Zungenbeinhornes

Vorderes Stiick des os quad-

ratum

Vorkiemendeckelstiick

Weichstrahlen and Stechel-

strahlen der Afterflosse

Weichstrahlen und Stechel-

strahlen der Riickenflosse...

Wir bel-korper

Zahnstiick des Unterkiefers ...

Zitsenbein

Zungenbein

Zungen -bogen

Zungen-bogen

Zungen-horn

Zungen-horn

Zwischendeckel

Zwischenkieferbein

Zvvischenkiemendeckelstiick...

Zwischerischeitelbein

Zygapophyses

Basibranchial

Symplectic

Suspensory pharyngeal

Abdominal vertebrae

Caudal vertebrae

Hypural

Hypocoracoid..
Hypercoracoid
Hypercoracoid

Preopercle

Clavicle

Alisphenoid ....

Clavicle

Hypocoracoid .,

Ceratohyal

Qiiadrate ..
Preopercle.

Anal rays and spines.

Dorsal rays and spines.

Centrum

Dentary ,

Pterotic

Urohyal

Epihyal

Ceratohyal

Epihyal

Ceratohyal

Interopercle

Premaxillary

Interopercle

Epiotic

Page.

i8

513

.S7

522

50

520

26

515

70

524

70

524

22

514

72

525

46

519

54

520

41

518

77

526

19

513

26

515

69

524

44

5 '9

3«

5'7

40

518

18

513

44

519

63

523

64

523

65

524

79

1526

I

508

.5«

522

57

522

57

522

24

515

55

521

16

512

55

521

5«

522

35

517

19

513

24

515

78

526

76

52,S

66

524

30

516

9

510

3«

517

34

51b

35

517

34

51b

35

517

27

515

32

51b

27

515

7

510

71-

524

538 STARKS

BIBLIOGRAPHY.

The follo%ving works contain all the terms here quoted :

Agassiz, Louis.

1833-1845. Histoire des Poissons Fossiles.

Bojanus.

1818. Versuch einer Deutung der Knochen in Kopfe der Fische. In ' Oken's
Isis.'
Carus.

1834. Lehrbuch der Vergleichenden Anatomie.

Cope, Edward D.

1871. Observations on the Systematic Relations of the Fishes. Proc. of the
American Association for the Advancement of Science.

Cuvier, Georges.

1829. Le Regne Animal distribue d'apres son organisation, pour servir de
base a I'histoire naturelle des animaux, et d'introduction a I'anatomie
comparde. Paris.

Cuvier and Valenciennes.

1828-1845. Histoire Naturelle des Poissons.

Erdl, M. P.

1848. Beschreibung des Skeletes des Gymnafchus niloticus. Abhand. Akad.
Wiss., V, 209.

Franque, Henricus.

1847. Amiffi calvK Anatomium Descripsit Tabulaque lUustravit. Berlin.

Gegenbaur, C.

1865. Untersuchungen zur Vergleichenden Anatomie der Wirbelthiere.
Heft II. Schultergurtel der Wirbelthiere. Brustflosse der Fische.
Leipzig.

Gegenbaur, C.

1870. Grundziige der Vergleichenden Anatomie. Leipzig.

Geoffrey, St. Hilaire.

1824. Annales des Sciences Naturelles, III.

Gill, Theodore.

1872. Arrangement of the Families of Fishes, Smithsonian Institution
Washington, D. C

Gill, Theodore.

1888. On the Classification of the Mail-cheeked Fishes. U. S. National
Museum, Washington, D. C
Grant, R.

1835-1841. Outlines of Comparative Anatomy.

Grant, R.

1833-1834. Lectures on Comparative Anatomy. Lancet.

Gunther, Albert.

1880. An Introduction to the Study of Fishes. Edinburgh.

SYNONOMY OF THE FISH SKELETON 539

Kallmann.

1837. Die Vergleichende Osteologie des Schliifenbeins.

Huxley, T. H.

1872. Manual of the Comparative Anatomy of the Vertebrated Animals.
New York.

Kostlin.

1844. Der Bau des Knochernen Kopfes.

Meckel.

1821-1833. System der Vergleichenden Anatomie.

Muller, Johannes.

1843. Eericht liber die Fortscliritte der Vergleichenden Anatomie der Wir-
belthiere. Archiv fiir Anat.

Oken.

1843. Lehrbuch der Natur-Philosophie.

Owen, Richard.

1866. On the Anatomy of Vertebrates.

Owen, Richard.

1846. Lectures on the Comparative Anatomy and Phj'siology of the Verte-
brate animals, Delivered at the Royal College of Surgeons of England
in 1844 and 1846. Part I. Fishes. London.

Parker, W. K.
1868. A Monograph on the Structure and Development of the Shouldei"-
girdle and Sternum in the Vertebrata. London.

Parker, W. K.
1872. On the Structure and Development of the Skull in the Salmon.
Bakerian Lecture II.

Parker, T. Jeffrey.
1893. A Course of Instruction in Zootomy- (Vertebrata). London.

Shufeldt, R. W.

1885. The Osteology of Amia Calva. Report of the U. S. Fish Commission
for 18S3. Washington, D. C.

Stannius.

1839. Symbols ad Anatomiam Piscium.

Stannius, H.

1854. Zootomie der Fische. Berlin.

Starks, E. C.

1899. The Osteological Characters of the Fishes of the Suborder Percesoces.
U. S. National Museum. Washington, D. C.

Vogt and Yung.

1888. Lehrbuch der Praktischen Vergleichenden Anatomie.

Wagner, R.

1843-1844. Lehrbuch der Zootomie.

Proc. Wash. Acad. Sci., November, 1901.

Proc. Wash, Acad. Sci. Vol, hi

Plate lxiii

FIG. 3.

Cranium of Roccus lineatus.

FIG. 1. TOP VIEW
FIG, 2. BOTTOM VIEW

FIG. 3, SIDE VIEW
FIG, 4. END VIEW

Proc. Wash. Acad. Sci. Vol. hi

Plate lxiv

Proc. Wash. Acad. Sci. Vol. hi

Plate lxv

PROCEEDINQS

OF THE

WASHINGTON ACADEMY OF SCIENCES

Vol. Ill, pp. 541-546. [Text Figures 47-57.] November 29, 1901,

PAPERS FROM THE HOPKINS STANFORD GALA-
PAGOS EXPEDITION, 1898-1899.

V.

ENTOMOLOGICAL RESULTS (5) :

THYSANURA AND TERMITID^.

By Nathan Banks,
U. S. National Museum.

Two species of Thysanura were collected on the Galapagos
Islands, and one, in considerable numbers, on Clipperton Island.
All appear to be new, but belong to genera of world-wide dis-
tribution. One of the species from the Galapagos Islands is of
very large size. The species from Clipperton Island is inter-
esting on account of lacking a curious character present in
many, if not all, of the other species of the genus. The Col-
lembola do not appear to have been collected, although doubt-
less fairly abundant on the islands.

Family LEPISMID^.
LEPISMA GALAPAGOENSIS sp. nov.
Length 20 mm., width of prothorax 5 mm. Body above and
below covered with dark steel bhie scales; antennae and cerci an-
nulate; legs pale, yellowish on tarsi. Body of usual shape and
Proc. Wash. Acad. Sci., November, 1901. 54^

542

BANKS

appearance. Head broad, with 14 tufts of hair seen from above
as follows : one in front of each eye, one behind each antenna,
two above base of each antenna, one in front of each antenna, and
two each side on anterior margin ; below on clypeus there are other

Fig. 47. I^e^pi'sma n-a/(iJ>a^oeusis, head. Fig. 4S. Lcpimta galapagoensh, tail.

Fig. 49. Lepiima galapagocnsis,
maxillary palp.

Fig. 50. Lepisnia galapagoensis., la-
bial palp.

tufts. Maxillary palpi with intermediate joint longest, penultimate
nearly twice as long as the last. Labial palpi with last joint foot-
shaped. The antennae: are much longer than the body, the cerci
about length of body; telson slender and acutely pointed over base of
median cercus. Legs with many hairs and spine-like bristles. Thorax
with spine-like bristles along sides. On the meso- and metathorax
there are on each side margin five roundish groujDs of yellowish pores,
on the hind margin about half way from middle to side is an oblique
row of them, and there is a roundish group nearer to side. On the
posterior margin of each abdominal segment are two oblique rows of
pores, one near edge of body, the other about one-third way from
edge to median line ; these rows have their inner ends more forwar''
than the outer ends. On each ventral segment there are each side two
tufts of yellowish hairs or bristles, besides the appendage.

THYSANURA AND TERMITID^

543

Several specimens from Hood Island in May. The largest species
known to me.

LEPISMA INSULARIS sp. nov.

Length 7 mm., width of prothorax 1.4 mm. Body with light
brownish scales, some specimens show the abdomen more or less
banded ; legs pale, with many hairs ;
antennae and cerci pale, but more
or less distinctly annulate with dark.
Head more conical than preceding
species ; there is a crest of hairs over
base of each antenna, and a few scat-
tered groups on anterior margin.
The maxillary palpi have the last
joint a little longer than preceding
joint, the intermediate joint a little
longer than any of the others. The
labial palpi have the last joint ob-
scurely triangu lar. The antennae are
about the length of the body, and
the cerci are plainly shorter than
body. The telson is broadly trian-
gular and rounded at tip. The legs
have the usual stout spur at upper
tip of each tibia. There are no
distinct groups of pores on the
thoracic segments, but on the ab-
dominal ones there is on each side
near the edge of body an obliq'ie row of three pores.

Several specimens from Narboro Island and one from Tagus
Cove, Albemarle Island, all in January.

Fig. 51.
and tail.

Lcpiima tnsiclan's, head

Fig. 52. Lcpisma insularis, max-
illarj palp.

Family MACHILID^E.

MACHILIS MUTICA sp. nov.

Length when extended 7 mm. Body clothed above with dark
scales, some paler patches on. the thorax ; pale beneath ; legs and palpi
pale ; antennas annulate ; cerci dark above, pale beneath ; abdominal
appendages pale. Body curved, the head bent down. Eyes united
for only a short distance, below each eye is a white hour-glass figure,
the anterior end rather larger but shorter than posterior end. Between

544

BANKS

the bases of antennaj is a median brown spot. The last joint of maxil-
lary palpi is plainly shorter than the penultimate, which is about
equal to the one preceding it; the basal joints are marked above with

Fig. 53. Machilis mutica^ head. Fig. 54. Machilis. miitka. end of body.

brown. Antennaj once and one-half length of body; the basal joint
about twice as long as wide; beyond the joints are of the usual curious
arrangement ; first several hard dark rings alternate with pale, then
three dark rings in a group, then four, then six, then eight, then ten
or twelve, and keeping at about ten to sixteen to the tip. Median
cercus as long as body, lateral cerci less than one-third the length of
the median. There does not appear to be any trace of the appendages
to the second and third coxaj, found in other species.

A number of specimens from Clipperton Island in November.

Family TERMITID^.

Of the two species of Termites collected both appear to be
distinct from any previously described. When the species from
the Central American region are more full known it may be
found that they also occur there. The genus Calotermes is
found throughout tropical countries.

CALOTERMES GALAPAGOEXSIS sp. nov.

Length, with wings, 9.5 mm. ; body 5 mm.

Yellowish brown ; mouthparts, antennas, legs and ventral region
paler; wings nearly hyaline, subcostal vein audits branches brown,
costal area slightly infuscate. Head quite broad, eye situate before
the middle, ocelli present. Prothorax much broader than long,
scarcely narrower than the head, rounded behind, with a median pale
line above. Abdomen nearly as long as head and thorax together.
Wings much longer than body, quite broad, pointed at tip ; the sub-

THYSANURA AND TERMITID^E

545

CCE

Fig. 55. Calotcrmes galapagoensts^ wing.

Fig. 56. Calotermes galapa-
goensis, head and pronotum.

costal vein does not branch till beyond the middle of wing, median
vein about half-way between the subcostal and submedian with the
usual branches simple, except those near the apex which are forked.
Several specimens from Wenman Island in December.

CALOTERMES PACIFICUS sp. nov.

Length, with wings, 11 mm. ; body 8 mm.

Pale yellowish brown ; beneath paler, as also on legs and anten-
na ; wings hyaline, costal and subcostal veins with the branches brown.
Head not very large, eyes at about middle, ocelli present. Prothorax
nearly twice as broad as long, concave in front,
fully as broad as head. Meso- and metathorax
much longer than prothorax, and fully as broad.
Abdomen plainly longer than head and thorax
together. Wings not greatly longer than the
body, moderately broad, branches before mid-
dle and before end of costal, the median vein
is about half-way between the subcostal and sub-
median veins, branched near the tip and con-
nected to subcostal, the submedian has most of its
apical branches forked, those toward base are very
oblique.

Soldier — Length 12 mm.

Head castaneous, darker in front than behind; Fig. 57. Calotermes
mandibles black; rest of body, legs and antennse /«"/c«5, soldier and
pale yellowish. Head about once and one-third '"^"'^'b^^-
longer than broad, sides slightly convex, equally broad in front
and behind, somewhat flattened above, near anterior margin slightly
concave above the middle. Mandibles stout and large, with two
small teeth toward tip and a larger one basad, on outer side at
base there is a hump. Antennae slender, about as long as width of
the head, 14-jointed, the third joint plainly longer than any of the
others, also darker. Prothorax plainly broader than long, fully as
large as meso- and metathorax together, the latter but slightly sepa-

546 BANKS

rated from each other, and each with a small projection or tubercle
near each posterior outer corner; abdomen short, in the middle a little
broader than the head.

A number of si^ecimens from Albemarle Island in December and
February, and from Narboro Island in January. Some soldiers are
plainly smaller than others and have their heads a trifle shorter.

PROCKEDINQS

OF THE

WASHINGTON ACADEMY OF SCIENCES

Vol. Ill, pp. 547-55S. November 29, 1901,

DESCRIPTIONS OF 23 NEW HARVEST MICE
(GENUS REITHRODONTOMYS).

By C. Hart Merriam.

The mammal collection of the Biological Survey contains a
large number of specimens of the so-called harvest mice, genus
Reithrodo7itomys. Among these are many not heretofore known
to naturalists. Twenty-three of these are here described, as fol-
lows :

REITHRODONTOMYS TENUIROSTRIS sp. nov.

737^^ from Todos Santos, Guatemala (altitude 10,000 ft.). No.

76919. $ ad. U. S. National Museum, Biological Survey Coll.

Dec. 29, 1895. E. W. Nelson and E. A. Goldman. Orig. no.

8917.

Characters. — Size large ; tail and ears long and dark; fore feet very
large ; fur long, soft and full ; color intense fulvous.

Color. — Upperparts rich deep fulvous, becoming intense ferrugi-
nous on posterior part of back and rump, somewhat more yellowish an-
teriorly; palest on top of head; underparts salmon fulvous; chin and
lips much paler, almost whitish; wrists and middle part of fore feet
dark brown ; outer side of feet and toes white ; ankles and hind feet
nearly to base of toes, dark brown; toes whitish; tail dark brown or
dusky all around ; a small dark ring around eye.

Cranial characters. — Braincase inflated, subglobular, abruptly
contracted anteriorly at posterior corner of orbits ; zygomata squarely
spreading anteriorly, their outer sides nearly parallel; rostrum and
nasals long and slender, slightly upturned from plane of frontals;
nasals acute posteriorly, ending on plane of premaxillae ; incisive for
Proc. Wash. Acad. Sci., Nov., 1901 547

548 MERRIAM

amina long and rather slender ; bulke small, set very obliquely to axis
of skull ; molars rather large.

Measureinents. — Type specimen ( $ ad.) : total length 210 mm. ;
tail vertebrae 124; hind foot 23.

Skull: basal length 21.5; zygomatic breadth posteriorly 13; length
of palate 12.5 ; length of molar series 4.

REITHRODOXTOMYS TENUIROSTRIS AUREUS subsp. nov.

Type from Calel, Guatemala (altitude 10,200 ft.). No. 76939.

9 ad. U. S. National Museum, Biological Survey Coll. Jan. 15,

1896. E. W. Nelson and E. A. Goldman. Orig. no. 9140.

Characters. — Similar in general to A', temcirostris^ but very much
paler and yellower; underparts in adult purer fulvous (less of the sal-
mon tinge) ; underparts in a young specimen whitish. The skull dif-
fers from that of tenuirostris in the following particulars: braincase
less inflated; nasals narrower posteriorly and ending slightly in front
of premaxillaj; zygomata slightly less spreading anteriorly, their sides
not parallel; palate shorter; the incisive foramina long and cutting
plane of first molars ; bullae less oblique.

i\Ieasitrements. — Type specimen (9 ad.): total length 196 mm.;
tail vertebraj 112; hind foot 22.5.

Skull: basal length 20; zygomatic breadth posteriorly 13; length
of palate 11.5 ; length of molar series 4.5.

REITHRODONTOMYS MICRODON sp. nov.

Type from Todos Santos, Guatemala (altitude 10,000 ft.). No.

76923. 9 ad. U. S. National Museum, Biological Survey Coll.

Dec. 31, 1S95. E. W. Nelson and E. A. Goldman. Orig. no.

8977.

Characters. — Size small; tail and ears long, dark, and nearly
naked; fur long, soft and full; general appearance almost exactly as
in the much larger R. tenuirostris., of which externally it is an almost
exact miniature.

Color. — Upperparts deep reddish fulvous, most intense on rump,
palest on head ; sides bright fulvous; underparts salmon fulvous (in
young specimens sometimes white) ; a dark ring around eye; a dark
brown spot on wrists; fore feet and toes white; ankles and hind feet
dark brown; toes white; tail dark all around, but slightly paler
beneath.

Cranial characters. — Skull small, light and papery, with rounded,
inflated braincase and rather squarely spreading zygomata; rostrum
and nasals slender, nasals ending about on plane of premaxillaa. Com-

NEW HARVEST MICE 549

pared with tenuirostris the skull is about two-thirds as large, the
rostrum relatively shorter and less upturned, the teeth very much
smaller.

Measurements. — Type specimen (9 ad.): total length 185 mm.;
tail vertebrae 113; hind foot 31.

Skull : basal length 17.5 ; zygomatic breadth posteriorly 11 ; length
of palate 10; length of molar series 3.

REITHRODONTOMYS MICRODON ALBILABRIS subsp. nov.

Type from Cerro San Felipe, Oaxaca, Mexico (altitude 10,000 ft.).

No. 68393. 9 ad. U. S. National Museum, Biological Survey

Coll. Aug. 35, 1894. E. VV. Nelson and E. A. Goldman. Orig.

no. 6629.

Characters. — Similar in size and general appearance to R. micro-
don., but very much paler, the upperparts fulvous instead of intense
ferruginous; lips, sides of nose, and underparts pure white instead of
salmon. The skull differs slightly from that of microdon : braincase
larger, more inflated ; zygomata broader anteriorly ; incisive foramina
shorter; bullae less inflated and more flattened on outer side; molars
slightly larger and heavier.

Measurements. — T3'pe specimen (9 ad.) : total length 187 mm.;
tail vertebrae 117; hind foot 20.

Skull: basal length 17; zygomatic breadth posteriorly 11.5; length
of palate 10; length of molar series 3.

REITHRODONTOMYS CHRYSOPSIS TOLUC^ subsp. nov.

Type from north slope of Volcan Toluca, State of Mexico (altitude
11,500 ft.). No. 55895. 9 ad. U. S. National Museum, Biological
Survey Coll. Sept. 10, 1893. E. W. Nelson and E. A. Gold-
man. Orig. no. 5454.

Characters. — Similar to chrysopsis^ but top of head and upper
parts generally much darker, more profusely and finely mixed with
black tipped hairs; sides duller and much less fulvous; tail shorter;
underparts whitish, the plumbeous underf ur showing through ; hind
feet brownish to near base of toes. Skull similar to that of chry-
sopsi's, but slightly smaller and narrower, with even more slender
rostrum, slightly smaller bullae, and slightly smaller molars. Ex-
ternally the resemblance to perotensis is much closer than to chry-
sopsis, but the skull is quite different from that of perotensis and very
close to chrysopsis.

Measurements. — Type specimen (9 ad.): total length 180 mm. ;
tail vertebrae 98; hind foot 21.

550

MERRIAM

REITHRODONTOMYS PEROTENSIS sp. nov.

Type from Cofra de Perote, Vera Cruz, Mexico (altitude 9,500 ft.).

No. 54424. $ ad. U. S. National Museum, Biological Survey

Coll. May 31, 1893. E. W. Nelson and E. A. Goldman. Orig.

no. 4912.

Characters. — Pelage long, soft, and woolly; ears large and dark;
tail medium; similar in general to cZ/rj^o/^w but color everywhere
duller; top of head and middle part of back darker; skull and teeth

much smaller.

Color. — Upperparts pale dull yellowish fulvous, tending toward
buffy gray on head and toward fulvous on rump ; top of head and
back profusely lined with black hairs, most abundant along median line
where they form a dark median dorsal area extending to base of tail ;
underparts buffy salmon, ankles and wrists brownish ; fore and hind
feet whitish ; tail dark above, indistinctly paler below.

Cranial characters. — Skull in general similar to that of chrysopsls
but smaller; braincase flatter; zygomata squarely spreading anteriorly,
their outer sides parallel ; rostrum shorter but equally slender ; nasals
ending on plane of premaxilhe ; incisive foramina long and open;
length of upper molar series 3.30 (contrasted with 3. So in chrysopsis).

Measurements. — Type specimen {$ ad.): total length 176 mm.;
tail vertebras 102; hind foot 19.

REITHRODONTOMYS ORIZAB^^ sp. nov.

Type from Mount Orizaba, Puebla, Mexico (altitude 9,500 ft.).

No. 53626. $ ad. U. S. National Museum, Biological Survey Coll.

April 24, 1S93. E. W. Nelson and E. A. Goldman. Orig.

no. 4734.

Characters. — Size medium; ears blackish and well haired; tail
long and rather slender; pelage long, soft and woolly; closely related
to chrysopsis and perotensis.

Color. — Upperparts golden fulvous, becoming fulvous on rump,
slightly darkened on posterior part of back with dark tipped hairs (al-
most exactly as in chrysopsis) \ lips and chin grayish; underparts
buffy salmon; tail dark above, indistinctly paler below; wrists and
ankles dark ; fore and hind feet whitish.

Cranial characters. — Skull similar to that oi perotensis but rostrum
slightly broader; braincase narrower and more rounded; bullie more
transverse ; molar teeth as in perotensis.

Rctnarks. — Externally orizabcc is hardly distinguishable from the
more highly colored specimens of chrysopsis, although the head is
more strongly fulvous, the ears decidedly more hairy, and the tail less

NEW HARVEST MICE 55 1

distinctly bicolor. The skull is nearer that of perotcnsis^ from which
it differs as above pointed out.

Measurements. — Type specimen {$ ad.) : total length 1S2 mm.;
tail vertebrae 105 ; hind foot 20.

REITHRODONTOMYS COLIMyE sp. nov.

Type from near timberline, Sierra Nevada de Colima, Jalisco,

Mexico (altitude 12,000 ft.). No. fffH- ^ ^^' U. S. National

Museum, Biological Survey Coll. April 21, 1S92. E. W. Nelson

and E. A. Goldman. Orig. no. 2447.

Characters. — Size smallest of the chrysopsis group; ears large
and dark ; tail rather short and sharply bicolor ; pelage soft and woolly ;
similar in general to chrysopsis but decidedly smaller and with dis-
tinctive cranial characters.

Color. — Upperparts varying from buffy fulvous to bright golden
fulvous strongly darkened on top of head and back by dark tipped
hairs ; underparts buffy salmon ; wrists and ankles dark ; fore feet
whitish; hind feet brownish, becoming whitish toward toes; tail
dusky above, whitish below, with distinct line of demarcation.

Cranial characters. — Skull similar to that of chrysopsis tolucce but
decidedly smaller, with nasals ending on plane of premaxillae instead
of anterior to this plane ; zygomata slightly more squarely spreading
anteriorly ; rostrum slightly shorter and broader ; incisive foramina
more broadly open ; molar series same size, measuring about 3.6 mm.

Measuremejits. — Type specimen ( $ ad.): total length 165 mm.;
tail vertebrcE 90 ; hind foot 20. A female from type locality : total
length 166; tail vertebrae 90; hind foot 20.

REITHRODONTOMYS COLIM^^ NERTERUS subsp. nov.

Type from foothill region of Sierra Nevada de Colima, Jalisco, Mexico
(altitude 6,500 ft.). No. fffff. $ ad. U. S. National Museum,
Biological Survey Coll. April 21, 1892. E. W. Nelson and E.
A. Goldman. Orig. no. 2514.

Characters. — Similar to colimce^ but tail decidedly longer and less
distinctly bicolor; pelage much shorter and coarser (not soft and
woolly) ; skull slightly smaller, with smaller molars (molar series
measuring 3.3 instead of 3.6),

Remarks. — This animal is particularly interesting in showing the
changes that take place when a member of the chrysopsis group
descends from the high altitudes at which all of the species live. The
specimens of typical colimce collected by Nelson and Goldman were
found at altitudes of 10,000 and 12,000 feet, while those of the

55:

MERRIAM

present form were from four to five thousand feet lower. Externally
subspecies iterterus appears to be slightly larger than the high moun-
tain form, but the difference is mainly due to the greater length of
tail [the skull is actually smaller than that of coU»icc\. The pelage
has lost the soft, woolly quality characteristic of the chrysopsis series
and has become much shorter and coarser, resembling the ordinary
members of the genus.

Measjirc7)ie7tts. — Type specimen ( 9 ad.) : total length 190 mm. ;
tail vertebra; iio; hind foot 20. Average of 3 adults from type lo-
cality: total length 186; tail vertebric 105: hind foot 20.5.

REITHRODONTOMYS COSTARICENSIS JALAP.^
subsp. nov.

Type from alapa, Vera Cruz, Mexico (altitude 4,000 ft.). No.

10S538. $ ad. U. S. National Museum, Biological Survey Coll.

May 10, 1901. E. W. Nelson and E. A. Goldman. Orig. no.

14712.

Characters. — Similar to costaricctisls^ but paler anteriorly and with
greater admixture of black hairs on the back; hind feet grayish
brown throughout, including toes, without distinct white edgings;
hairs at base of claws white; ankles distinctly darker than feet; side of
nose, upper lip (broadly), and underparts white, tlie plumbeous under-
fur showing through except on chin and nose patch ; skull similar to
that of co^/ar/ctf/w/'^, but rostrum slightly shorter; braincase slightly
more inflated; nasals truncate posteriorly about i mm. in front of
ends of premaxilljE.

Meas7irc7ncnts. — Type specimen ( $ ad.) : total length 197 mm. ;
tail 1 19 ; hind foot 21.

REITHRODONTOMYS GOLDMANI sp. nov.

Type from Metlaltoyuca, Puebla (altitude 800 ft.). No. 93096.
9 ad. U. S. National Museum, Biological Survey Coll. Feb-
ruary 12, 1898. E. A. Goldman. Orig. no. 12 153.
Characters. — Size small, similar to costar7'ce7tsis^ but very much
paler; ears smaller and brownish instead of blackish; tail shorter.

Color. — Upperparts uniform ochraceous fulvous not noticeably
darkened by black hairs, but darkest on crown of head between
ears; palest between eyes; sides of nose, upper lip (broadly), chin and
underparts white ; fore feet whitish ; ankles and hind feet brownish,
the brown fading insensibly into whitish of toes; tail dusky all round.
Cra7iial characters. — Skull similar to that of costarice7isis but in-
cisive foramina shorter and less expanded in middle; molar series
slightly shorter.

NEW HARVEST MICE 553

Measurements. — Type specimen (9 ad.): total length 190mm.
tail 109; hind foot 21.5.

REITHRODONTOMYS HIRSUTUS sp. nov.

Type from Ameca, Jalisco, Mexico (altitude 5,500 ft.). No. 82200.

$ old. U. S. National Museum, Biological Survey Coll. Feb.

15, 1S97. E. \V. Nelson and E. A. Goldman. Orig. no. 10537.

Characters. — Size large (among the largest of the genus) ; tail
long and slender; pelage coarse, almost hispid ; ears rather short and
pale ; color grizzled yellowish fulvous.

Color. — Upperparts uniform yellowish fulvous, heavily lined with
black Jiairs; the fulvous brightest on sides; underparts including
sides of nose, lips (broadly) and fore feet buffy whitish (sometimes
pale buffy fulvous) ; a dark spot on ankle ; hind foot sparingly clothed
with whitish hairs ; tail dusky above, soiled whitish beneath, without
sharp line of demarcation.

Cranial characters. — Skull large and substantial, in general like
\\-\2^\. oi perotnyscus ; braincase somewhat flattened; rostrum medium
or rather large ; zygomata rather squarely spreading anteriorly ; nasals
broad and bluntl}'^ wedge-shaped posteriorly, ending decidedly anterior
to elongated arms of premaxillae ; incisive foramina of moderate
length and rather broadly open; bulla; rather small; upper molar
series 3.7 mm.

Measurejnents. — Type specimen ( $ old) : total length 233 mm. ;
tail vertebraa 143 ; hind foot 22.

REITHRODONTOMYS GRISEOFLAVUS sp. nov.

Type from Ameca, Jalisco, Mexico (altitude 4,000 ft.). No. 82 196.

$ ad. U. S. National Museum, Biological Survey Coll. Feb. 9,

1897. E- ^- Nelson and E. A. Goldman. Orig. no. 10501.

Characters. — Size medium; ears medium or rather short; well
haired; tail short; pelage coarse ; similar to ///ri'«/?/^ from the same
locality, but very much smaller, with shorter tail (relatively as well as
actually) and more hairy ears.

Color. — Upperparts buffy gray anteriorly, becoming buffy yellow-
ish or even buffy fulvo.us posteriorly, everywhere conspicuously and
coarsely lined with black hairs; sides (including cheeks in full pelage)
light golden fulvous; underparts whitish sometimes suffused with buffy
or pale buffy fulvous, the fulvous of side of neck showing in axilla on
inner side of fore leg ; a small buffy or fulvous pectoral spot some-
times present; lips and chin broadly whitish; fore and hind feet
whitish, a dark spot on ankles ; tail brownish above, whitish below,

554

MERRIAM

without sharp line of demarcation; ears brown with a conspicuous
tuft of fulvous hairs at anterior base.

Cranial characters. — Skull small, light and thin, of normal form;
zygomata standing out rather squarely anteriorly, the outer sides
nearly parallel or only slightly narrower anteriorly; nasals relatively
broad and bluntly pointed posteriorly falling short of premaxillae; in-
cisive foramina large and open for so small a skull ; audital bulla)
rather large; molai- series rather small, measuring 3.20 in length.
Compared with tenuis the skull is slightly larger, the braincase more
inflated and higher.

Remarks. — This species is almost a miniature of R. Jiispidus of
the same region and adds another case to those already mentioned
in which two species, almost identical in external appearance, but one
very much larger than the other, inhabit the same locality.

Measurements. — Type specimen {$ ad.): total length 169 mm.;
tail vertebrie 94; hind foot 22. Average of 3 from type locality :
total length 169; tail vertebrtc 97; hind foot 21.4.

REITHRODONTOMYS GRISEOFLAVUS HELVOLUS

subsp. nov.

Type from Oaxaca City, Oaxaca, Mexico. No. 6S3S7. 9 ad. U. S.

National Museum, Biological Survey Coll. Aug. 14, 1S94. E.

W. Nelson and E. A. Goldman. Orig. no. 6576.

Characters. — Similar to griseojlavus but upperparts everywhere
conspicuously redder, the back and sides intense ochraceous or orange
fulvous instead of buffy yellowish ; grizzling of black hairs less con-
spicuous.

Remarks. — In color helvolus resembles levipes from San Sebas-
tian much more closely than it does griseojlavus., but its skull is
smaller and narrower than that of levipes^ with shorter tooth row, in
all of which respects it agrees with griseojlavus.

Measure?nents.—Ty^e: specimen (9 ad.): total length 188 mm.;
tail vertebras iii; hind foot 20. Average of 6 from type locality:
total length 182; tail vertebrae 105.5; ^"''"'^^ ^^'^^ 20.1.

REITHRODONTOMYS LEVIPES sp. nov.

Type from San Sebastian, Jalisco, Mexico (altitude 3,000 ft.). No.

8S057. $ ad. U. S. National Museum, Biological Survey Coll.

March 30, 1897. E. W. Nelson and E. A. Goldman. Orig. no.

10S39.

Characters. — Size medium; ears rather small and thick; tail
rather short; pelage course; color fulvous.

NEW HARVEST MICE - 555

Color. — Upperparts fulvous, moderately lined with black hairs,
brightest on sides, where in some specimens (as in the type) it is
bright orange fulvous ; underparts pale fulvous (in some specimens
buffy) ; a dark spot on ankle ; fore and hind feet usually soiled
whitish.

Cratiial characters. — Skull of medium size; braincase normal;
rostrum small; zygomata squarely spreading; bullae rather small; in-
cisive foramina medium or rather large ; molars rather small. Com-
pared with hirsutus the skull is smaller; rostrum decidedly smaller;
zygomata usually more widely and squarely spreading anteriorly ;
bullae and teeth decidedly smaller. Compared with the %vi\?^W. griseo-
Jlavus from Ameca, Jalisco, the skull averages slightly larger and
broader; zygomata decidedly more widely and squarely spreading
anteriorly; bulla; smaller; incisive foramina usually longer; molars
slightly larger.

Measurements. — Type specimen ( $ old) : total length 188 mm. ;
tail vertebras no; hind foot 21.

REITHRODONTOMYS LEVIPES OTUS subsp. nov.

Type from foothill region of Sierra Nevada de Colima, Jalisco,
Mexico (altitude 6,500 ft.). No. f|f|f. 9 ad. U. S. National
Museum, Biological Survey Coll. April 11, 1892. E. VV. Nel-
son and E. A. Goldman. Orig. no. 2439.
Characters. — Similar to levipes but pelage softer (less coarse), the

dark hairs smaller and less noticeable ; ears decidedly larger ; tail

longer; skull similar, but rostrum slightly longer; bullae averaging

slightly larger.

Measurements. — Type specimen ( 9 ad.): total length 202 mm.;

tail vertebrae 120; hind foot 22.

REITHRODONTOMYS LEVIPES TOLTECUS subsp. nov.

Type from Tlalpam, Valley of Mexico. No. 50746. 9 ad. U. S.

National Museum, Biological Survey Coll. Dec. i, 1892. E. W.

Nelson and E. A. Goldman. Orlg. no. 3935.

Characters. — Similar in genei-al to levipes but somewhat larger;
color more yellowish (less fulvous) and much more heavily lined with
black hairs ; underparts nearly white, instead of fulvous ; fore and
hind feet white. Skull narrower; zygomata less widely spreading
anteriorly ; nasals longer and ending with premaxillae ; bullae slightly
larger.

Remarks. — Specimens from Patzcuaro, Michoacan, and Zimapan,
Hidalgo, while less yellow, appear to be referable to this form.
Proc. Wash. Acad. Sci., Nov., 1901.

;56

MERRIAM

Measurements.— 'Yy^^& specimen (9 ad.): total length 196 mm.;
tail vertebra; loS; hind foot 21. Average of 2 specimens from type
locality: total length 193; tail vertebrae 106; hind foot 21.5.

REITHRODONTOMYS SATURATUS CINEREUS subsp. nov.

Type from Chalchicomula, Puebla, Mexico. No. 53633. $ ad.

U. S. National Museum, Biological Survey Coll. April 13, 1893.

E. W. Nelson and E. A. Goldman. Orig. no. 4659.

Characters. — Similar to saturatus but very much paler and grayer,
particularly the head and shoulders; tail shorter and more sharply
bicolor. Skull similar but slightly smaller; rostrum shorter.

Measurements. — Type specimen ( $ ad.) : total length 149 mm. ;
tail vertebrae 73 ; hind foot 19.

REITHRODONTOMYS SATURATUS ALTICOLUS
subsp. nov.

Type from Cerro San Felipe, near Oaxaca City, Oaxaca, Mexico (al-
titude 10,000 ft.). No. 6S392. $ old, U. S. National Museum,
Biological Survey Coll. Aug. 24, 1S94. E. W. Nelson and E.
A. Goldman. Orig. no. 6623.
Characters. — Similar to saturatus but tail somewhat shorter;

rostrum shorter; bullae larger.

Measurcmc7its. — Type specimen {$ old): total length 153 mm.;
tail vertebrae 75; hind foot 19. Average of 2 males from type lo-
cality; total length 152; tail vertebra; 78.5; hind foot 18.5.

REITHRODONTOMYS DIFFICILIS sp. nov.

Type from Orizaba, Vera Cruz, Mexico (altitude about 4,500 ft.).

No. 63735. $ ad. U. S. National Museum, Biological Survey

Coll. ' Feb. 20, 1S94. E. W. Nelson and E. A. Goldman. Orig.

no. 5868.

Characters. — Size a little larger than saturatus; color fulvous;
tail bicolor; ears medium or rather short; pelage coarse.

Color. — Upperparts fulvous, grizzled with black tipped hairs;
underparts washed with pale fulvous except on chin and throat, which
are whitish; fore and hind feet whitish ; a dark spot on ankle; tail
dusky above, whitish below, with a rather sharp line of demarcation.

Cranial characters. — Skull similar in general to that of satura-
tus^ but incisive foramina more broadly open; molar series slightly
shorter.

Measuretnefits. — Type specimen ( ^ ad.): total length 177 mm.;
tail vertebrae 100; hind foot 19.5.

NEW HARVEST MICE 557

REITHRODONTOMYS DORSALIS sp. nov.

Type from Calel, Guatemala. No. 77009. $ ad. U. S. National

Museum, Biological Survey Coll. Jan. 14, 1S96. E. W. Nelson

and E. A. Goldman. Orig. no. 91 12.

Characters. — Size rather large ; ears and tail medium ; tail bicolor ;
general color grayish above with dusky dorsal area.

Color. — Upperparts grayish washed with buffy, becoming buffy
ochraceous on sides ; top of head and broad dorsal area blackish (from
profuse admixture of black tipped hairs) ; underparts buffy whitish,
in many specimens suffused with buffy fulvous ; lips and chin whitish ;
wrists dusky ; fore and hind feet whitish ; tail dark above, whitish
below, with rather sharp line of demarcation.

Cranial characters. — Skull apparently of the mexicanus type, con-
siderably larger iha.n saturatus., but molar teeth small as \x\. saturatus ;
rostrum rather long; braincase moderately rounded ; basal length of
skull 18.5; zygomatic breadth 1 1.8 ; molar series 3.20.

Measurements. — Type specimen ( <? ad.): total length 171 mm.;
tail vertebrae 92; hind foot 19.

REITHRODONTOMYS MEGALOTIS ZACATECyE subsp.

nov.

73^/^ fi"©"^ Valparaiso Mts., Zacatecas, Mexico. No. 91910. 9 ad.

U. S. National Museum, Biological Survey Coll. Dec. 6, 1897. E.

W. Nelson and E. A. Goldman. Orig. no. 11S59.

Characters. — Similar in general size and proportions to megalotis
but much darker and yellower; ears and tail darker, pelage coarser
and less fluffy. Upperparts pale dull fulvous brown, becoming buffy
ochraceous on sides ; top of head and back conspicuously lined with
black hairs; underparts dark (instead of white), washed with buffy
salmon ; a pale fulvous pectoral patch. Skull like that of megalotis
but braincase somewhat flatter and broader ; nasals ending on plane of
premaxillae instead of more anteriorly.

Remarks. — R. 77i. zacatec(£ resembles subspecies obsctirus from
Sierra Madre, Chihuahua, very much more closely than it does typical
megalotis. It differs from obscurus in having the upperparts less dark
and more yellowish, the sides much deeper buffy ochraceous.

Measuremetits. — Type specimen (9 ad.): total length 156 mm. ;
tail vertebrcB 87 ; hind foot 17.5. Average of 4 specimens from type
locality: total length 152; tail vertebrae 83; hind foot 17.5.

^^8 MERRIAM

REITHRODONTOMYS MEGALOTIS OBSCURUS subsp. nov.

Type from Sierra Madre, near Guadalupe y Calvo, Chihuahua,
Mexico. No. 95277. $ ad. U. S. National Museum, Biological
Survey Coll. Aug. 26, 1898. E. W. Nelson and E. A. Goldman.
Orig. no. 12900.

Characters. — Similar to megalotis but everywhere very much
darker, upperparts conspicuously lined with and darkened by black
hairs; underparts buffy salmon (instead of white), with pale fulvous
pectoral patch. Skull as in megalotis. Compared with saturatus
which is a size larger, the ground color of upperparts is grayer (less
fulvous) the underparts slightly more tinged with buffy salmon, and the
pectoral region is pale fulvous. The skull, like that of tnegalotis^
differs from saturates in smaller size, shorter rostrum, usually shorter
incisive foramina, smaller bullae, and shorter molar series.

Measuremoits. — Type specimen {$ ad.) : total length 167 mm.;
tail vertebra 90; hind foot 19.

PROCEEDINQS

OF THE

WASHINGTON ACADEMY OF SCIENCES

Vol. Ill, pp. 559-563. November 29, 1901

SEVEN NEW MAMMALS FROM MEXICO, IN-
CLUDING A NEW GENUS OF RODENTS.

By C. Hart Merriam.

Among the remaining undescribed mammals collected in
Mexico by E. W. Nelson and E. A. Goldman, while engaged
in field work for the Biological Survey of the U. S. Depart-
ment of Agriculture, the following seven appear to be of more
than ordinary interest.

BLARINA MAYENSIS sp. nov.

Typelrom a Maya ruin at Chichenitza, Yucatan. No. 10S0S7. 9 ^^'
U. S. National Museum, Biological Survey Coll. February 5,
1901. E. W. Nelson and E. A. Goldman. Orig. no. 14495.
Characters. — Similar in general to mexicana but pale plumbeous
or slaty instead of blackish; unicuspidate teeth larger and broader.

Color. — Upperparts plumbeous, darker on nose ; back with fine
'pepper and salt ' appearance ; underparts plumbeous, overlaid with
ash brown, strongest on throat and breast.

Cranial characters. — Skull similar to that of mexicana but slightly
longer and less massive; braincase lower (practically continuing plane
of rostrum) ; molariform teeth large and full, not excavated anteriorly
or posteriorly; large premolar smaller than that of vtexicanus; ist
and 2d unicus^Dids large and swollen, much larger than those of mexi-
canus^ and with inner tubercle obsolete, in this respect resembling
pereffrina.

Measurements. — Type specimen ( $ ad.): total length 102 mm.;
tail vertebrai 29; hind foot 13.

Proc. Wash. Acad. Sci., Nov., 1901. 559

560 MERRIAM

ODOCOILEUS HEMIONUS CANUS subsp. nov.

Type from Sierra en Media, Chihuahua, Mexico. No. 99361. $ ad.

U. S. National Museum, Biological Survey Coll. October 7, 1900.

E. W. Nelson and E. A. Goldman. Orig. no. 13996.

Characters. — Similar in general to O. hemionus but in fresh winter
pelage very much paler and grayer; throat pale gray, faintly vermic-
ulated [in hemionus much darker and more heavily grizzled with
blackish] ; white of chin less sharply defined ; breast less black ; ingui-
nal region white instead of buff ; face, top of head, ears, and outer
side of thighs much paler; dorsal neck stripe narrower, paler, and
grizzled throughout; upper side of tail dark on basal half and in some
specimens all the way to the black tip. The usual nose and tail mark-
ings are present but those on the nose are paler and more diffuse.
The hoofs are larger than in hemiot?iis from northern Arizona and
eastern Colorado.

Cranial characters. — Skull like that of heviioniis but muzzle
broader; nasals broader and flatter (less pinched in) ; antlers similar
but lighter and more slender, the beam rather short and concave an-
teriorly— not long and straight as in erem'icus Mearns from the desert
region about the Gulf of California.

Aleasurenients. — Type specimen ( $ ad.) : total length 1S30 mm. ;
tail vertebrae 230; hind foot 500; height at shoulder 955.

TYLOMYS TUMBALENSIS sp. nov.

Type from Tumbala, Chiapas, Mexico. No. 76059. $ yg-ad. U. S.
National Museum, Biological Survey Coll. Oct. 23, 1895. E. W.
Nelson and E. A. Goldman. Orig. no. S568.

Characters. — Size rather large, apparently larger than midlcaudiis
and with much larger hind feet ; ears large and nearly naked ; tail long
and naked, the posterior half yellow; color brownish gray (becoming
brownish fulvous with age?).

Color. — (Type specimen, yg-ad.) Upperparts dark gray, darkest
on top of head and (broadly blackish) along median part of posterior
half of back to base of tail; sides lightly washed with brownish;
face dark gray ; a broad dusky ring around eye ; chin, pectoral region
and small patch in inguinal region whitish; throat and belly plumbe-
ous, washed with buffy fulvous ; fore and hind feet and toes dark
brownish, with a few white hairs at base of toes; basal half (or a
little more than half) of tail blackish, terminal half yellow.

Cranial characters. — Skull large, long and flat. Differs from
nudicaudus as follows: less massive and more slender; rostrum and

NEW MAMMALS FROM MEXICO 561

nasals decidedly more slender (nasals narrowly wedge-shaped, trun-
cate posteriorly on plane of premaxillaa) ; zygomata weaker and less
broadly spreading (probably due to immaturity) ; interparietal large,
broadly and strongly convex anteriorly, tapering to point on sides ;
incisive foramina broadest posteriorly and ending on plane of front of
molars; bulL-e slightly larger ; upper incisors weaker and more slen-
der ; molar series very large and heavy, decidedly larger and longer
than in midicatidus (measuring 9.5 mm.).

Meastiretnents. — Type specimen ( $ yg-ad., probably not full
grown) : total length 448 mm. ; tail vertebrae 234; hind foot 46.

TYLOMYS BULLARIS sp. nov.

Type from Tuxtla, Chiapas, Mexico. No. 7605S. $ im. U. S.

National Museum, Biological Survey Coll. Sept. 7, 1895. E. W.

Nelson and E. A. Goldman. Orig. no. 8406.

Characters. — Type and only specimen too young to give reliable
external characters. Ears very large, black, and sparsely haired; tail
long and slender, naked, basal third blackish, terminal two- thirds yel-
low. Color paler and grayer than young of corresponding age of
nudicaudtis ; underparts white ; upper lip and patch on side of nose
whitish ; face much paler than in nudicatidus ; fore feet brown to base
of toes; toes and edging on outer side of foot white; hind feet to toes
dark brown ; toes white.

Cranial characters. — Skull similar to that of nudicaudiis of cor-
responding age, but interparietal broader on outer side; zygomata
more strongly spreading posteriorly; bullce widely different, large and
broadly rounded anteriorly, without anterior prolongation, suggesting
the condition in Ototylomys. Teeth of same size as in midicaudiis.

Remarks. — I am reluctant to describe a new mammal from an
immature specimen, but in this instance no other course seems open.
The white upper lip and patch on side of nose, the pale grayish
plumbeous of the upperparts, and the peculiar character of the audital
bullae seem to distinguish the species from its congeners. It is hoped
that adults will be obtained in the near future.

Measurements. — Type specimen ( $ im., not nearly full grown) :
total length 324 mm. ; tail 158; hind foot 37.5.

OTOTYLOMYS gen. nov.

Exter7ial characters. — Appearance rat-like; ears large, thin and
naked; tail long and naked.

Cranial characters. — Skull flattened, with large interparietal and
strongly developed superciliary ridges which form a horizontal shelf

562 MERRIAM

over orbits and reach posteriorly to occiput, as in Tylomvs, but differ-
ing from Tylomys in the following characters:

1. Bulhe rotate-capsular, inflated, as broad anteriorly as posteriorly,
without narrow anterior prolongation ; axes essentially parallel to axis
of skull. Similar to those of Xenoifiys but smaller.

2. Anterior or maxillary root of zygoma strongly notched above (at
base of rostrum) as in Peromysczts and Xenomys.

3. Outer lip of anterior opening of antorbital vacuity vertica/ us in
Neotofua and Pe7-omyscus^ not concave as in Tylotuys.

4. Fossa at base of rostrum less developed.

5. Incisive foramina enormous, as broadly open anteriorly as pos-
teriorly.

6. Angle of jaw excavated posteriorly, leaving a sharp point pro-
jecting backward as in Pcromyscus^ the infracondyloid notch corre-
spondingly long and deep (differing widely from the broadly ex-
panded and rounded angle of Tylomys^ which is convex posteriorly
instead of sharply concave).

7. Coronoid process reduced to a mere point as in Perornysciis ;
postcoronoid notch flat and nearly horizontal.

OTOTYLOMYS PHYLLOTIS sp. nov.

Type from Tunkas, Yucatan. No. 108099. $ old, U. S. National

Museum, Biological Survey Coll. Feb. 17, 1901. E. W. Nelson

and E. A. Goldman. Orig. no. 1455 1.

Characters. — Size of Mus rattus (equaling the smaller species of
Neotoma')\ ears enormous and leaf-like; tail long (as long as head
and body), naked, the annulations and scales broad. General ap-
pearance as in Tyloviys but size smaller, ears much larger, upperside
of tail dark throughout (terminal half not yellow).

Color. — Upperparts varying fiom grayish blown to very dull ful-
vous brown, in some specimens conspicuously mixed with black-tip-
ped hairs; underparts, including inner side of fore and hind legs,
white or buffy white; ankles and wrists dark: fore and hind feet
whitish ; face brownish or grayish, washed on cheeks and around
ej^e with pale fulvous; tail brownish dusky above, irregularly yellow-
ish beneath. Young grayish plumbeous with top of head (from be-
tween eyes posteriorly) and middle part of back darker.

Cra?zial characters (additional to generic characters already given).
— Rostrum long and slender; nasals long and slender; their sides
nearly parallel, truncate posteriorly about on plane of premaxillae ;
zygomata moderately spreading, broadest joosteriorly ; interparietal

NEW MAMMALS FROM MEXICO 563

large, broadly strap shaped or subquadrangular, but narrower on outer
sides than in middle.

Aleastiremejits. — Type specimen ( $ old) : total length 303 mm. ;
tail 148 ; hind foot 28. Average of S adults of both sexes : total
length 297; tail 150; hind foot 27.5.

OTOTYLOMYS PHYLLOTIS PH^^US subsp. nov.

Type from Apazote, near Yohaltun, Campeche. No. 107940. 9 *

ad. U. S. National Museum, Biological Survey Coll. December

28, 1900. E. W. Nelson and E. A. Goldman. Orig. no. 14369.

Characters. — Similar to O. phyllotis but somew^hat smaller; tail

much smaller and shorter ; upperparts much darker, the posterior

median dorsal region dusky, the sides and fore-back grayish brown

with a pale dull fulvous tinge. Skull similar to that of phyllotis but

slightly smaller ; rostrum more slender ; frontals and nasals flatter and

in same plane (nasals not decurved) ; superciliary shelves not so

strongly developed.

Measurements. — Type specimen ( 9 ad.): total length 266 mm, ;
tail 136; hind foot 26.5.

CALLOSPERMOPHILUS MADRENSIS sp. nov.

Type from Sierra Madre, near Guadalupe y Calvo, Chihuahua,
Mexico (altitude 7,000 feet). No. 95363. 9 ad. U. S. National
Museum, Biological Survey Coll. August 27, 1S9S. E. W. Nel-
son and E. A. Goldman. Orig. no. 12923.

Characters. — Similar to Callospermophilus lateralis^ but tail
about half as long.

Color. — (In fall pelage) ground color above grizzled grayish, be-
coming dull grayish fulvous on rump; black stripes nearly obsolete ;
white stripes buffy gray, slightly grizzled with black hairs; fore and
hind feet whitish ; underparts soiled whitish more or less grizzled ;
tail above grizzled dull grayish fulvous and black, below yellowish.
Cranial characters. — Skull similar to that of lateralis., but ros-
trum less swollen.

Measurements. — Type specimen ( 9 ad.) : total length 233 mm. ;
tail vertebrae 58 ; hind foot 39. Average of 8 females from type local-
ity ; total length 228 ; tail vertebrae 62 ; hind foot 39.

PROCEEDINQS

OF THE

WASHINGTON ACADEMY OF SCIENCES

Vol. Ill, pp. 565-5 68. November 29, 1901

PAPERS FROM THE HOPKINS STANFORD GALA-
PAGOS EXPEDITION, 1898-1899.

VI.
THE ISOPODS.

By Harriet Richardson,

Collaborator, Smithsonian Institution.

The Isopods, collected by the Hopkins Stanford Expedition,
comprise four species of which two, new to science, are here
described. One of the new species comes from Clipperton
Island ; the other was collected at sea about 300 miles north of
Galapagos. The two species previously known are for the first
time recorded from the localities cited.

Family TANAID^.

TANAIS STANFORDI sp. nov.

Description of ]\Iale. — Head very broad at base, attenuated anteri-
orly, where it is produced straight forward, forming a sort of neck.
The anterior margin of this projection has a triangular median point,
on either side of which the first pair of antennae are inserted in the
depression formed by the median point and the acute antero-lateral
angles of the projection. The eyes, which are black and distinct, are
situated in these lateral angulations. The first pair of antennae are
composed of three joints each, the first joint very much elongated,
about equal to two-thirds the length of the head; the other two joints

Proc. Wash. Acad. Sci., November, 1901. (565)

\66

RICHARDSON

are short, subequal, both together but
little longer than half the length of the
first joint. The second pair of antennae
extend only to the end of the second
joint of the fust pair. Both pairs of an-
tennae terminate in long hairs.

The first thoracic segment is coalesced
with the head. The second segment is
short, the next two equal in length and
longer than the first. The last three are
subequal and the longest.

There are five abdominal segments,
including the terminal one. The first
three are subequal in length; the fourth
is half as long as any of the preceding
ones; the terminal segment is rounded
posteriorly.

The three pairs of pleopods are densely
setose. The uropods are single-branched,
each branch composed of three joints.

The first gnathopods have the hand
strongly chelate and very large, the finger
and thumb gaping. The finger is long
and curved ; the thumb elongated and
squarely produced ; the upper surface is

provided with two teeth, a slight depression separating them. The

carpus is produced at the upper inner angle in a strong tubercle.
The first pereiopods are very small, feeble and

almost inconspicuous; the second pair longer, but

not quite as long as the other four pairs, which

are equal in length.

Description of Feviale. — Head broad at base

and attenuated anteriorly, but not produced in a

straight process as in the male. The anterior

margin has a triangular median point on either

side of which the first pair of antennae are inserted

in the depression formed by the median point and

the antero-lateral angles. The eyes are situated as

in the male. The first pair of antennae are much

shorter than those of the male. The first joint is longest, equal to one-
third the length of the head, the two following joints are subequal and

Tanais stunfordi.

Fig. 59. Tanais
staiifordi ; hand of
male.

f. THE ISOPODS 5^7

both together as long as the first joint. The second pair of antennas do
not quite reach the end of the second joint of the first pair.

The thoracic segments are similar
to those of the male.

The abdomen and the uropods are
also similar to those of the male.

The gnathopods have the hand
strongly chelate, but very much smaller
than the hand of the male, the finger
and thumb not gaping. The pereio-
pods are all alike and of equal
length.

Color of specimens white with
more or less numerous markings of F^«- ^o- T.^"^is stanfordi. a.

Head of female. /^. Hand of female,
brow^n.

A number of specimens of both males and females were found at
Clipperton Island Lagoon.

Type, U. S. National Museum. Cat. No. 25172.

Family CYMOTHOID^.
^GATHOA EXCISA sp. nov.

Description. — Body narrow, elongate, three and a half times longer
than broad.

Head quadrangular, the front produced over and beyond the basal
joints of the antennae, the anterior margin of which is deeply ex-
cavate, the lateral margins rounded. Eyes large, oc-
cupying the greater part of the area of the head, but not
contiguous. The first pair of antennas extend a little be-
yond the posterior margin of the head, and are composed
of eight joints. The second pair of antennae are equal in
length to the first pair, and are composed of about four-
teen joints.

The first and fifth thoracic segments are somewhat longer
than any of the others. The epimera are short, not reach-
ing beyond the posterior angles of their respective seg-
ments, but all equalling these segments in length.
Fig. 61. All of the first five abdominal segments are distinct and
^^gathoa subequal; terminal segment is about equal in length to the
excisa. ^^g preceding ones, is triangularly shaped, with rounded

extremity, and has a lateral incision on each side near the posterior

568 RICHARDSON

margin. At the base of this segment there is a deep depression. The
uropods do not surpass the terminal segment. The outer branch is
narrow, lanceolate ; the inner branch is broad, truncate ; both branches
are equal in length.

The color of the entire body is a dark bluish gray with the excep-
tion of the terminal segment and the uropods, which are a light
yellow.

One specimen was taken from the fin of a dolphin ( Coryphcena
hippurus) in 5° N. lat., 90° W. long.

Type U. S. National Museum. Cat. No. 25173.

Meinertia gaudichaudii (Milne-Edwards).

Cyntothoa gatidichaiidii Vi\\v\e.-'^<lwAx^'i, Hist. Nat. Crust., iii, p. 271.
Ceratothoa rapax Heller, Reise Novara, Crust., p. 146, pi. xii, f. 17.
Ceratothoa gaudichaudii Schicedte and Meinert, Naturhistorisk Tidsskrift

(3), XIII, 1881-83, P- 335-340, pi. XIII. figs. 11-15, pi. XIV, figs. 1-5.
Meinertia gaudichaudii Stebbing, Hist. Crust., 1893, p. 354.

Black Bight, Albemarle Island, Galapagos Islands; 200 miles north
of Wenman Island, Galapagos Islands.

Family LIGIID^.
Ligia exotica (Roux).

Ligia exotica, Roux, Crust. Medit., p. 3, pi. xiii, fig. 9. — Budde-Lund, Crust.

Isop. Terrestria, pp. 266-268, 1885.
Ligia gaudichaudii, Milne-Edwards, Hist. Nat. des Crust., iii, p. 157. —

Dana, Exp. Ex., p. 741, pi. xlix, figs. 6*"''. — Nicolet in Gay's Hist,

Chile, III, p. 265.
(?) Ligia {Italica) coriacea, Koch, Deutschl. Crust., p. 36.

Guadalupe and Clipperton Islands.

PROCEKDINQS

OF THE

WASHINGTON ACADEMY OF SCIENCES

Vol. Ill, pp. 569-576. [Text Figures 62-66.] DECEMBER II, 1901.

PAPERS FROM THE HARRIMAN ALASKA
EXPEDITION.

XXVI.

HARRIMANELLA, A NEW GENUS OF HEATHERS.

By Frederick V. Coville.

The German naturalist Pallas, for more than forty years the
leader of Russian researches in natural history, published in
1788 a description of a new heather from eastern Siberia under
the name Andromeda stelleriana. Willdenow, in 1799, trans-
ferred the species to the genus Erica. In the year 1834 David
Don, a Scotch botanist, placed it in the genus Bryanthus, while
Sir W. J. Hooker, studying the same plant independentl}^ re-
ferred it back to Andromeda, citing the opinion of another bot-
anist who considered it a Menziesia. In 1839 ^he plant was
referred by De Candolle to Cassiope, and under the name Cas-
stope stelleriana it has since passed.

The genus Cassiope was founded by David Don, in 1834, '
the plants originally included in it consisting of five species
taken out of the older genus Andromeda. Two of these, tetra-
gona and hypnoides, were species described by Linnagus and
occurring in northern Europe ; two others, lycofodioidcs and
ericoides had been described by Pallas from eastern Siberia,
and owti, Jiistigiata, by Wallich from the Himalayas. Cassiope
tetragona {Andromeda tetragona L.) was named by Don as the
type of his new genus. To Cassiope belong, therefore, the
species tetragona and such others as are congeneric with it. A

*D. Don, Edinb. New Phil. Jom-n. 17: 157. 1834.

Proc. Wash. Acad. Sci., December 1901. 569

570

COVILLE

careful examination of the so-called Casstopc stclleriana makes
it clear that this species is not a true Cassiope but should be
made the type of a hitherto undescribed genus, which is charac-
terized as follows :

Harrimanella gen. no v.

Plant belonging to the tribe Andromedeae, of the family Ericaceae ;
leaves empetriform, many-ranked, decurrent ; flowers pentamerous,
nodding, singly terminal on ordinary leafy branches ; calyx ebracteo-
late, the sepals distinct, imbricated in the bud, not becoming fleshy

Fig. 62. Harrimanella stellcriaua. a, flower, with part of calyx and corolla
cut away (Xs) *. *. diagram of flower ; c, ovary and style (X5) "> '^> ^^^^^ (X2) ; <";
longitudinal section of fruit (X2);/. transverse section of fruit (X2); g, seed
fXio) ; h, end of branch, showing leaves and position of flower (Xs); ''i out-
line of transverse section of leaf (Xio).

and not exceeding the capsule in fruit ; corolla campanulate, vertically
pL>ited at the base, the lobes pendent and overlapping; filainents
tumid below, the broad anthers situated in two planes of five each,
biaristate below the apex, opening by large pores ; style short, stout,
ovoid, persistent on the fruit ; capsule globose, many-seeded ; seed
oblong, slightly or not at all winged.

The genus is named for Mr. E. H. Harriman, of New York, through
whose courtesy the opportunities and pleasures of his voyage to Alaska,
in 1899, were shared by a party of scientific investigators. It is hoped
that the Alaska heather, Harrimanella stellcria7ia, one of the most
charming and beautiful little plants of the whole Alaskan flora, will
serve as a happy reminder of the expedition to all its members.

The genus Cassiope, from which Harrimanella is now segregated,
has opposite, cuprcssiform, closely sessile leaves, conspicuously swollen
or spurred at the base and closeh*^ appressed to the stem, the branches

HARRIMANELLA, A NEW GENUS OF HEATHERS

571

thus becoming quadrangular. The flowers of Cassiope occur singly
in the axils of the leaves, the pedicels bearing at the base four mem-
branaceous bracts ; each sepal has a transverse fold at the base as if
the organ had first been bent sharply backward, then at a slightly
higher point bent sharply forward again, a character best brought out
by a longitudinal median section of a sepal ; the corolla is not plaited
at the base and the usvially short lobes are recurved and not overlap-
ping in anthesis ; the anthers lie in a single plane ; and the style is
elongated, slender, and nearly cylindrical. The seeds may furnish a
generic character, as suggested by the conspicuous apical appendage
in Cassiope merteiisiana and the absence of any such appendage in
Harrunariella stelleria7za.

Fig. 63. Cassiope mertensiana. a, flower, with part of calyx and corolla cut
a\vay(X5) ; ^^ diagram of flower; c, ovary and style (X5) ! ^> fruit (X2); e,
longitudinal section of fruit (X^) ; p, transverse section of fruit (X2) ; ^, seed
(Xio); ^< end of branch, showing leaves and position of flowers (X3)) *'.
outline of transverse section of leaf (Xio) > 7i bract from base of pedicel (Xs)-

Of the ten species that have been referred to Cassiope, seven are
still referable to the genus as here restricted, two are referable to
Harrimanella, and one must be excluded from both. The species of
true Cassiope are distinguishable by their vegetative characters alone,
as indicated in the following key :

Key to the Species of Cassiope, based on Leaf Characters.

Leaves with a deep channel on the back.
Leaf margins with a few coarse bristles.

C. ericoides (Pall.) D. Don.
Leaf margins pubescent or fimbriate, but without bristles.

Proc. Wash. Acad. Sci., December, 1901.

572

COVII.LE

Leaves with scattered pubescence on the back and margins, but
without membranaceous or fimbriate margins or a terminal

bristle C. tetragona (L.) D. Don.

Leaves without hairs or fimbriae except on the margins or in the
furrow.
Leaves with persistent white membranaceous fimbriate margins
connecting with a terminal acumination.

C. fastigiata (Wall.) D. Don.
Leaves with the marginal fimbria: early deciduous, a slender
terminal bristle persistent.

C. selaginoides Hook. f. & Thom.
Leaves not channeled on the back.

Leaves not ciliate C. mertensiana (Bong.) D. Don.

Leaves more or less ciliate on the margins or at the apex when young.
Leaves shining, armor-like, without white margins.

C. redoivskii (Ch. & Schl.) G. Don.
Leaves dull, with white margins.

C. lycopodioides (Pall.) D. Don.

Of these species one, tetragona^ is circumpolar, two, fastigiata

f 4 k

Fig. 65. Cas.s.iope lycopodioides.
a, flower, with part of calyx and
corolla cut awayCXs); ^^ fruit
(X-); <"i longitudinal section of
fruit (X-)) ^> end of branch,
showing leaves and position of
flowers (X3) » «) outline of trans-
verse section "^ of leaf (Xio); /<
bract from base of pedicel (Xs)-

and selaginoides^ are Himalayan ; two, ericoides and redoivskii^ be-
long to eastern Siberia ; one, lycopodioides^ extends from Alaska to

Fig. 64. Cassiope tetragona. «, flower,
with part of calyx and corolla cut away
(X5); ^, fruit (X2); c, longitudinal sec-
tion of fruit (X2); d, end of branch,
showing leaves and position of flowers
(Xs); ^1 outline of transverse section of
leaf (Xio) ; y, bract from base of pedicel
(X5)-

HARRIMANELLA, A NEW GENUS OF HEATHERS 573

Japan ; and one, 7nertensiaiia^ is confined to northwestern North
America. Three species of Cassiope, therefore, are found on the
American continent, tetragona^ lycopodioides^ and mertensia7ia.

The species which must be exchided from both Cassiope and Harri-
manella is C. oxycoccoides Gray.* Our whole knowledge of this plant
is based on a single flowering twig 43 mm. long, collected by Dr.
Leonhard Stejneger August 22, 1SS2, on the northern shore of Bering
Island, off the east coast of Siberia, on "a rounded hill, about 300 feet
high, just behind the fishing settlement of Saranna." An examina-
tion of the specimen, which is in the Harvard University Herbarium,
shows that it differs in various ways from both Cassiope and Harri-
manella. Its leaves are situated in whorls of three and very closely
resemble broad ^leaves of Cha7naecistiis froctimbeiis. They are,
however, smooth on the back between the midrib and margin and they
have an apical gland similar to that of Vaccbtiiim vltisidaea. The
three pentamerous flowers are borne on short bibracteolate pedicels in
a tribracteate umbel on a short terminal peduncle. The calyx does
not differ from that of Harrimanella, while the corolla is of the urceo-
late short-lobed form found in Cassiope. The style is cylindrical as in
Cassiope, but the stamens differ from those of both Cassiope and
Harrimanella. In both these genera the pollen pore, the two awns,
and the point of attachment of the filament are close to the apex of
the anther, but in the plant under consideration the filament is attached
to the base of the anther, only the two awns and the pollen pores being
apical. Although the fruit is unknown, the whorled character of the
leaves and the structure of the stamens give reason for a surmise that
the plant belongs to the Ericeae instead of the Andromedeae. Appar-
ently, however, it is not congeneric with any known species of the
Ericeae. Its corolla is not chartaceous ; it is a depressed plant barely
projecting above the mosses and lichens among which it grows ; its
leaves have peculiar apical glands ; and it is geographically isolated
from the rest of the Ericeae, none of which are known from either
northeastern Asia or northwestern America. I venture to propose for
it the generic name Arcterica, the whole name being Arcterica
oxycoccoides (Gray). It is greatly to be desired that future travelers
visiting Bering Island procure a good series of both flowering and
fruiting specimens of this rarest of plants.

The two Cassiopes which are now referred to Harrimanella may be
distinguished as follows :

'Gray, Proc. Am. Acad. 20: 300. 1885 (Jan. 26); GrayMn Stejneger, Proc.
* U. S. Nat. Mus. 7 : 534. 1SS5 (Jan, 27).

574

COVILLE

Key to the Species of Harrimanella.

Leaves divergent, linear to linear oblong, obtuse or broadly acute at
the apex, the margins somewhat erose but not serrulate ; peduncles
in flower barely or not at all exceeding the leaves, in fruit about
twice as long; corolla about 6 mm. long; capsule commonly about
4, sometimes 5, mm. in diameter H. stelleriana.

Leaves loosely appressed, lanceolate-subulate, acute, minutely serrulate
on the margins ; peduncles few to several times as long as the leaves
in both flower and fruit ; corolla commonly 4 to 5 mm. long ; cap-
sule about 3 mm, in diameter H. hypnoides.

Harrimanella stelleriana (Pall.).

SVNOXVMY.

Andromeda stelleriana Pall. Fl. Ross. 12 : 58. 1788.

Erica stelleriana Willd. Sp. PI. 2 : 387. 1799.

Andromeda empetri/olia Martens ; Bong. Mem. Acad. Petersb. vi. Math. &

Nat. 2 : 153. 1831.
Bryanthus stelleri'D. Don, Edinb. New Phil. Journ. 17 : 160. 1834.
Menziesia stelleriana Fisch. ; Hook. Fl. Bor. Am. 2 : 37. 1834.
Cassiope stelleriana DC. Prod. 7: 611. 1839.

Earlier Illustrations.

Pallas, Fl. Ross. t. 74. f. 2. 1788, as Andromeda stelleriana.
Hooker, Fl. Bor. Am. /. iji. 1833-40, as Andromeda stelleriana.
Harriman Alaska Exped. 1 : 32. 1901, as Cassiope stelleriana.

Distribution.
The original specimens of Harrimanella stellcriaiia were collected
in eastern Siberia by Steller, probably on Bering Island. The species
has since been collected from Kamchatka southward to the mountains
of northern Japan, but apparently does not extend on the Asiatic
continent far from the Pacific. In America it occurs from Unalaska
eastward and southward along the whole coast of Alaska through
British Columbia and as far as Mount Rainier in the Cascade Moun-
tains of Washington. Toward the north, as in Prince ^V"illiam Sound,
it sometimes descends to sea level but ordinarily it grows near timber
line or in the upper elevations of the forest within a thousand feet of
timber line. On Mount Rainier it has been collected at an elevation
of 5,500 feet.

General Remarks.

We first saw this plant, on the Harriman Expedition, at the summit
of White Pass, Alaska, where it grew with the crowberry {Etnpe-
trum nigrum^ and reindeer moss ( Cladonia raftgiferina^ that car-
peted the mountain slopes at timber line. On the east side of Muir
Inlet, in Glacier Bay, at a point about six miles below the end of

HARRIMANELLA, A NEW GENUS OF HEATHERS. 575

Muir Glacier, we found it again, growing at an elevation of about
2,000 feet in the rather open forests of black hemlock ( Tsuga iner-
tensiatia). At this time, June 12th, it was not quite in flower. The
first flowering specimens were found June i6th on Mount Verstovia,
near Sitka. The plant occurred in abundance from an elevation of
3,400 feet to the summit, and one patch was seen on an exposed ridge
at about 1,500 feet. Near sea level on the west shore of Yakutat
Bay on the glacial gravels half covered with vegetation was seen a
single yellowed plant with pale whitened flowers, doubtless the prod-
uct of a stray seed from the mountains above. At various points in
Prince William Sound the plant grew in great profusion and was in
full bloom at the time of our visit, June 24th to 29th, occurring on the
mountain slope back of Orca among the dwarfed black hemlocks at
1,300 to 1,600 feet, in Port Wells, and in Columbia Fiord. This is
the plant that lead Mr. Gilbert to adopt the name Heather Island for
the rocky, mossy, scantily timbered island that stands in Columbia
Fiord immediately in front of the Columbia Glacier. On this spot,
chilled by the cold current of air flowing down from the great glacier,
grew this and other arctic-alpine plants in profusion all the way down
to sea level. On a timbered nunatak rising from the glacier about
eight miles from its front the plant occurred to an elevation of about
3,000 feet. On the mountains at the head of English Bay, Kadiak
Island, it was flowering abundantly at 1,500 to 3,000 feet. It was
found on the Alaska Peninsula at Kukak Bay by Mr. Kearney and at
Chichagof Bay by Mr. Palache, at an elevation of about i ,000 feet,
and sparingly on the Shumagin Islands by Mr. Kincaid and Mr.
Saunders at the same altitude, flowering till the middle of July.

Harrimanella hypnoides (L.).

Synonymy.

Andromeda hypnoides L. Sp. PI. i: 393. 1753.

Cassiope hypnoides Y). Don, Edinb. New Phil. Journ. 17: 158. 1834.

Earlier Illustrations.

LiNN^us, Fl. Lapp. t. I. f. j>. 1737, as Andromeda hypnoides.
Oeder, Fl. Dan. /. 10. 1761, as Andromeda hypnoides.
Pallas, Fl. Ross. /. 73./ 2. 1788, as Andromeda hypnoides.
Curtis, Bot. Mag. t. 2qj6. 1829, as Andromeda hypnoides.
Engler & Prantl, Nat. Pfl. 4I : /. 12. A. 1889, as Cassiope hypnoides.

Distribution.
The plant on which Linnaeus based his description came from the
mountains of Lapland, but the species is now known to occur in arctic

576

COVILLE

or arctic-alpine situations from the Ural Mountains westward through
northern Europe, Iceland, and Greenland to the American continent on

Fig. 66. Harrimanella hypnoides. a, flower, with part of calvx and corolla
cut away ( X S) ; *. fruit ( X 2) ; c, longitudinal section of fruit ( X 2) ; d, end
of branch, showing leaves and position of flower ( X 3)-

the western side of Baffins Bay, and thence southward through Labra-
dor to the high mountains of Quebec, New England, and New York.
The species is not known to occur in Siberia, Alaska, or central and
western British America.

PROCEEDINQS

WASHINGTON ACADEMY OF SCIENCES

Vol. Ill, pp. 577-600. December n, 1901.

PRELIMINARY REVISION OF THE PUMAS (FELIS
CONCOLOR GROUP).

By C. Hart Merriam.

The Pumas or Cougars form a strongly marked group, easily
distinguished from the other Cats by the following characters
(taken collectively) : size large ; build slender ; head relatively
small; tail long ; body without markings (except in the very
young) . They are confined to America, where they range from
southern Patagonia northward over nearly the whole of South
and Central America, Mexico, and the United States, and reach
their northern limit in southern Canada (Ontario in the east;
British Columbia in the west). During the past centur}^ their
range in the United States has become greatly restricted and
over large areas they have been exterminated by man. They
are not now known to inhabit New England, with the possible
exception of the Green Mts. of Vermont, though formerly found
In Vermont, Massachusetts, Connecticut, and Rhode Island.
In the State of New York they were killed in the Catskills within
the memory of our fathers, and in the Adirondacks, where until
recently they were fairly common, a few probably still exist.
They still occur in Florida and in the lowlands of Louisiana, but
in other parts of the United States are rather strictly confined to
mountainous regions.^

Pumas w^ere known to the early naturalists, and in 1771 Lin-
neeus named the Brazilian species I^clis concolor. From time to

1 Their distribution by States was carefully worked out by Dr. F. W. True in
1889 — Report U. S. National Museum for 1889, pp. 595-600, 1891.

Proc. Wash. Acad. Sci., December 1901. 577

578 MERRIAM

time other names were proposed, some deliberately intended to
replace Linn^eus's concolo)-, others to distinguish forms belong-
ing to particular regions. But for a long time the name concolo?'
was used in a collective sense for the entire group and this prac-
tice has prevailed until very recently. In 1896 Cory recognized
the Florida Puma as distinct from concolor ; and in 1897 I de-
scribed as new two forms from the west — one from the northern
Rocky Mountains ; the other from the northwest coast region.
Of late the belief has spread that several quite different animals
were concealed under the name concolor^ but the absence of suf-
ficient material for comparison has delayed the revision of the
group.

Recently the Hon. Theodore Roosevelt, President of the
United States, presented to the U. S. Biological Survey a series
of 12 skulls of the Rocky Mountain Cougar or Mountain Lion,
taken from animals killed by him in Colorado in January and
February of the present year (1901).' The examination of
these skulls led to comparisons with skulls from other localities,
and graduall}^ to a study of such specimens from various parts
of North and South America as could be easily brought together.
The more important conclusions resulting from this study are
here recorded.

The Roosevelt series of skulls is of unusual value, not only
because each skull is accompanied by precise data giving the
color, measurement and weight of the animal from which it was
taken, but also and more particularly because the study of so
large a series from a single locality shows the nature of the
differences resulting from sex and age, affords an index to the
kind and quantity of individual variation, and sets a standard for
comparison with other members of the group — thereby furnish-
ing a means of estimating the probable value of particular
cranial or dental peculiarities observed in single skulls from re-
mote regions.

A discussion of the variations presented by the Roosevelt

'Since this paper was written Mr. Roosevelt has published, in Scribners
\tagazine (October and November, 1901), an account of his Colorado Cougar
hunt, in the course of which he discusses the habits of Cougars in general. His
article is by far the most noteworthy contribution thus far made to the life
history of these animals.

REVISION OF THE PUMAS 579

series, with a detailed table of measurements of the skulls, will
be found under the species to which they belong — Felts hippo-
lestes. It may be stated here however that the limits of varia-
tion for adults of each sex fall within surprisingly narrow
bounds. This points toward stability in the characters presented
by other species of the group. In this connection I may be
permitted to state what I believe to be a general law (confirmed
by critical studies in most groups of North American mammals),
namely, that completely differentiated types (specific or sub-
specific) from the areas where the typical forms occur, show
relatively little individual variation, while the intergrades con-
necting such types, and also local forms not yet completely
established, show a wide range of individual variation.

Specific characters in some groups are shared equally by the
two sexes, but this is not the case with the Pumas, for these
animals conform to the rule among the Carnivora that the male
is the larger and more highly developed. The male Puma is
much larger and more powerful than the female, the skull and
teeth are much larger and more massive, and, what is still more
important, the male alone attains complete specific differentia-
tion. By this is meant that the distinctive cranial characters of
the various forms are onl}^ fully developed in the males, skulls
of females resembling each other so closely that in several cases
they are distinguished with difficult}' .

While the skulls of the several members of the group bear a
strong general resemblance to one another, marked differences
exist, chiefly in the proportions of the several parts. The most
conspicuous of these are in the degree of elevation or depres-
sion of the face and frontal region, the extremes of which are
presented by Fclis -patagonica of the Southern Andes and F.
olyinpus of the Puget Sound region. In Fclis patagonica the
skull is long and low, the face slopes strongly backward, the
orbital process of the jugal rises only slighth^ above the body of
the bone, the frontal region is depressed and flattened, and the
highest part of the cranium is behind the postorbital processes.
In F. olyniptis the face rises abruptly, the orbital process of the
jugal rises strongly and suddenly, the frontal region is highly
elevated and swollen, and the highest part of the cranium is in

580 MERRIAM

front of the postorbital processes, the resulting physiognomy be-
ing widely different from that of F. -patagonica.

Other important differences occur. The massive skulls of F.
htppolestcs and olympiis, with their heavy powerful under jaws,
contrast strongly with the relatively thin and light skull of F.
concolor, with its small and slender jaw ; and the huge canines
and carnassials of F. -puma contrast strikingly with the slender
and light corresponding teeth of F. bano-si.

Of the teeth, the carnassials furnish the most valuable char-
acters, and in distinguishing the species the upper is of service
more often than the lower. The significant variations in the
upper carnassial are in the length, breadth, form of outer side,
relative size of the cusps, and degree of development of the inner
tubercle or protocone. The tooth as a whole is largest in F.
puma, smallest in F. bangs/'. In F. concolor the outer side of
the crown (as seen from below) lies in two planes which meet
and form a prominent obtuse angle opposite the interspace be-
tween the middle and posterior cusps (paracone and metacone).
In most species the outer side of the tooth is concave at this
point and the angle is carried much farther back (usually to the
middle of the metacone) and is less prominent. The inner
tubercle or protocone varies somewhat in specimens from the
same region, and shows important differences when specimens
from different localities are brought together. It is largest and
most distinct in a voung male Fclis puma from Santiago, Chile
(No. 36851 U. S. National Museum), while in specimens oi F.
puma pcarsoni from the coast region of Patagonia it is nearly
obsolete, the inner side of the front part of the tooth sloping
down to the inner root very gradually (as in No. 20918, U. S.
National Museum). In F. hippolcstcs it is large and promi-
nent ; in aztccus it is moderately developed ; in olympus it is still
smaller [in some specimens nearly obsolete], and \\\ concolor
from Brazil, bangsi from the northern x\ndes and Costa Rica,
and coryi from Florida, it is only slightly developed.

Most of the Cougars present two color phases, independent of
season. One of these is usually redder than the other. Thus
at any time of year the animal of the Puget Sound region i^F.
olympus) may be ' red ' or ' brown ' — both of these terms, it must

REVISION OF THE PUMAS 58 1

be understood, giving an exaggerated idea of the real color and
of the difference between the two. On Vancouver Island, ac-
cording to John Fannin, Curator of the Provincial Museum at
Victoria, the brown pelage is the common one ; the red pelage
rather rare. On the mainland the red is more common. In the
case of the Rocky Mountain species [hippolestes), according to
Roosevelt, the colors designated as ' red ' and ' blue ' are equally-
divided, 6 of each being recorded among the twelve animals
killed by him in midwinter in Colorado.

A curious semi-pathological condition is common in skulls of
Pumas from widely separated regions, namely, a tendency to
cleft palate, consisting in a complete perforation of the palatine
bones on or near the middle of the roof of the mouth This per-
foration is usually bifurcate anteriorly and often measures lO
mm. or more in length. It is present in 6 out of the 12 skulls
of the Roosevelt series from Colorado, in 5 out of 10 skulls in
the Biological Survey Collection from Mexico, and in odd skulls
from Arizona, Florida, and Vancouver Island.

For the loan of material additional to that in the collection of
the Biological Survey I am indebted to Dr. J. A. Allen, Curator
of Mammals and Birds in the American Museum of Natural
History, New York; Outram Bangs, Curator of Mammals,
Museum of Comparative Zoology, Cambridge, Mass. ; Gerritt
S. Miller, Jr., Assistant Curator of Mammals, U. S. National
Museum, and Dr. Edgar A. Mearns, U. S. A. I wish particu-
larly to acknowledge the generosity of Outram Bangs, who has
sent me for description two new forms obtained by his collector
in tropical America. In addition to these the only specimens of
the Florida Puma I have seen are those in the Bangs collection.

In searching the literature for names that have been given to
members of the group, I have been greatly aided by my assist-
ant, Wilfred H. Osgood.

The number of specimens available at the present time is too
small, and the gaps in their distribution are too great, to admit
of anything like a complete and satisfactory study of the group,
or to permit the ranges of the several forms to be mapped.
The species and subspecies I am led to recognize, with their
type localities, are :

582 MERRIAM

Felis cougitar Kerr. Pennsylvania.

JPelis coryi Bangs. Florida.

Fell's hippolestes Merriam. Wind River Mountains/ Wyoming.

Pelts hippolestes olympiis Merriam. Olympic Mountains, Wash.

Felis hippolestes aztecus nob. Colonia Garcia, Chihuahua, Mexico.

Fclis bangsi nob. Colombia, South America.

Felis bangsi costaricensis nob. Chiriqui, Panama or Costa Rica
(boundary in dispute) .

Felis concolor Linn. Brazil.

Felis p7ima iSIolina. Chile.

Felis putna patagoJiica nob. Base of Andes (latitude 47° 30' S.).
Patagonia.

Felis puma pca7-so?ii Thomas. Coast of Santa Cruz, Patagonia.

FELIS COUGUAR Kerr. Adirondack Cougar ; Panther.

Fills co7tgiiar KV.KV., AnxrmX Kingdom, p. 151, 1792,
Felis pe7insvlvanica Link, Beitrage zur Naturgesch., 11, p. 90, 17^5.
Fclis co7icolor Merriam, Mammals of Adirondacks, pp. 29-39, 1882 (habits).
Felis ore(^07ie7tsis hippolestes Miller, Bull. N. Y. State Museum, No. 38, vol.
VI II, p. 124, 1900.

Type locality. — Pennsylvania.

Geographic dist7'ibiitio7i. — Adirondack Mountains, New York,
Green Mountains, Vermont,^ and until recently higher Alleghenies of
Pennsylvania, Maryland, and West Virginia, where a few may still
occur.

Characters. — Size medium , head (apparently) disproportionately
small for size of body ; color dull fulvous ; skull smallest of the known
species.

Color. — "Body and legs of a uniform fulvous or tawny hue . . .
Ears light-colored within, blackish behind. Belly pale reddish or
reddish white. Face sometimes with a uniform lighter tint than the
general hue of the body." — DeKay.''

Crajiial characters. — Skull smaller and less massive than in any
other North American species ; nasals broader and blunter posteriorly
than in hippolestes and aztecus., but very much smaller and narrower

' Cougars are exceedingly rare in New England where, so far as now known,
thej are restricted to the Green Mountains of Vermont. Professor George H.
Perkins, of Burlington, Vt., writes me that panthers were killed in that state in
1870, 1875, 18S1, and 1894, and that hunters believe that a few still exist in the
Green Mountains. He tells me further that the one killed in 1S81 was shot in
Barnard, Vt., and was a very large animal; it is now in the state collection at
Montpelier. During the present fall (1891) according to a Brattleboro news-
paper, two men, while hunting ruffed grouse in the town of Andover, suddenly
came upon a panther into which they emptied several charges of shot, apparently
without serious result.

^DeKay. Mammals of New York, p. 47, 1842.

REVISION OF THE PUMAS 583

than in coryi from Florida ; bullae smaller ; basioccipital broader ;
teeth smaller and more slender, particularly the large upper premolars
(carnassial and pm 1).

Remarks. — Felis cougiiar Kerr, from Pennsylvania, although not
heretofore recognized, appears to be specifically distinct from its near-
est allies. Unfortunately specimens are so exceedingly rare that I have
not been able to obtain a skin for comparison v^'ith the other forms of
the group, and have examined only two skulls. Both skulls are
adult, and both are from the Adirondack Mountains in northeastern
New York (one from the east side, Essex County: No. 3811,
U. S. National Museum ; the other from the west side, probably
Hamilton County: No. 947, Merriam Collection). These skulls
are not marked for sex, but as one is larger than the other, and as the
upper carnassial teeth present the usual sexual differences in size, I
assume that one is a male (3811), the other a female (947). If this
assumption is correct, the male is considerably smaller than the male
of any other known species, while the female is about the same size as
adult females of concolor and aztectis. The skull appears to be smaller
in proportion to the size of the body than in the case of any other
species.

Cranial measurements. — Two skulls from the Adirondacks, as-
sumed to be male and female (measurements of female in paren-
theses) : basal length 158 (154); zygomatic breadth 135 (128);
occipito-sphenoid length 59 (55.5) ; postpalatal length 83 (80) ; inter-
orbital breadth 36 (36) ; length of upper carnassial on cingulum
22.5 (20.5).

FELIS CORYI Bangs. Florida Cougar.

Felis concolor Jloridana CoKY , ' Hunting and Fishing in Florida,' pp. 109-110,

1896. (Name preoccupied.)
Felis coryi 'Qx'i^GS, Proc. Biol. Soc. Wash., xiii, pp. 15-17, Jan., 1899.

Type locality. — Florida.

Characters. — Size large ; color of back deep intense ferruginous.
Head large; legs and tail long; feet small (Bangs) ; skull large and
rather massive, with enormous nasals.

Color. ^ — Upperparts rich intense ferruginous on back (from top of
head to tail), becoming dull fulvous on sides, and fading gradually to
belly where the hairs are much longer and paler ; inguinal region
soiled buffy-whitish ; chin and lips yellowish buffy, becoming fulvous
on throat and breast ; ears and spot at base of whiskers black, the ear-

^ The present description is based on a skin and 6 skulls from near Sebastian,
Florida, kindly loaned me by Outram Bangs.

584 MERRIAM

spot descending to skin of head at base of ear ; tail dark brown with a
darker streak along dorsal line, becoming black before reaching tip,
which is black all round.

Cranial characters. — Skull large and rather massive ; nasals very
large, exceeding in length and breadth those of all other North
American forms ; sagittal crest long and high ; bullae medium or
rather small ; underjaw large and massive. Compared with P.
couguar from northern New York the skull is very much larger and
heavier (an ad. male having a basal length of 175 and a zygomatic
breadth of 152 contrasted with 158 and 135 in an old male from the
Adirondacks) ; the sagittal crest (obsolete in cotiguar) long and high
and reaching forward to frontal shield; nasals much larger, acutely
pointed posteriorly and with a strongly marked hump at juncture of
upper and middle thirds; frontals more flattened; bulUe slightly
larger ; teeth of the same size or very slightly larger ; underjaw more
massive (especially posterior part of horizontal ramus) and more
bellied below posterior part of tooth row.

Comi:)ared with aziccus, with which it agrees essentially in size,
and which is decidedly its nearest relative, it may be distinguished by
the following characters: skull as a whole less massive; nasals very
much larger and longer, with a distinct angle or hump at junction of
upper and middle thirds ; posterior ends of nasals reaching back
nearly 10 mm. beyond plane of jjosterior endings of maxillae [in aztccus
the nasals and maxilkc end on essentially the same plane] ; sagittal
crest higher ; under jaw slightly more bellied below tooth row ; coro-
noid process longer and curved more strongly backward ; upper car-
nassial larger; middle upper premolar smaller (thinner and weaker) ;
lower carnassial slightly larger; first lower premolar smaller and
weaker.

Re?t2arks. — The Florida Puma differs so markedly from its North
American congeners that a glance at either the skin or the skull is
sufficient to distinguish it, the intense rusty red of the back and the
great size of the nasal bones being characters not shared by any other
species.

Individual skulls vary surprisingly in zygomatic breadth, as shown
in two adult males of nearly equal size from Sebastian, Florida, both
in the Bangs collection. One of these has a zygomatic breadth of 152
mm., and is very broad interorbitally. The other has a zygomatic
breadth of only 135 mm., and is correspondingly narrow interorbitally.
The nasals are largest in the narrowest skull, but are very broad and
large in both.

REVISION OF THE PUMAS 585

In color the Florida Puma comes nearest to the widely remote F.
batigsi costaricensis of Central America, with which it nearly agrees
in the red of the back; but the face, legs and feet are grayer (less
fulvous), and the tail is very different, being dark brown with a
blackish stripe along the median line above ; in bangsi the upper side
of the tail is uniform dark reddish fulvous without trace of a dorsal
stripe.

Measuret7ients. — An adult male [tail and feet measured from dry
skin by O. Bangs]: total length 2057; tail vertebrae 760; hind foot
280. An adult female: total length 1918; tail vertebrae 670; hind
foot 271.

Cranial ineasurements. — Largest of 2 males, and largest of 2
females (females in parentheses) from Sebastian, Florida : Basal
length 175 (158); occipito-nasal length 193 (174); zygomatic
breadth 152 (123); occipito-sphenoid length 66 (59); postpalatal
length 92.5 (82.5) ; interorbital breadth 44 (39) ; length of upper
carnassial on cingulum 23.9 (22.8).

FELIS HIPPOLESTES Merriam. Rocky Mountain Cougar ;
Mountain Lion.

Felis hippolestes Merriam, Proc. Biol. Soc. Washington, Vol. xi, p. 219, July,
1897.

Type locality. — Wind River Mountains, Wyoming.

Characters. — Size largest of the knowni members of the group ; color
dull fulvous ; skull large and massive with highly developed sagittal
crest.

Color. — Upperparts and sides dull pale fulvous brown, dai'kest on
middle of back and tail ; tip of tail black ; face from nose to eyes gray-
ish brown ; a pale patch over each eye ; back of ears blackish ; chin,
lips (except dark patch at base of whiskers), throat, breast, inner side
of forelegs, inguinal region, and hinder part of belly soiled white ;
under side of tail grayish white.

Cranial characters. — Skull largest of the known species ; heavy and
massive ; frontal region elevated but less elevated than in males of
olympus ; sagittal crest highly developed and arched ; interorbital region
broad [interorbital breadth in adult males about 48-50] ; under jaw
long, large and massive, somewhat swollen under premolars. Teeth
large and powerful, the upper carnassial and 2d premolar much heavier
and broader than in aztecus. Skulls of female hippolestes from the
Rocky Mountain region are sometimes difficult to distinguish from
males of aztecus from Chihuahua.

586 MERRIAM

Ret7iarks. — Skulls of Felis hippolestes are not only larger and more
massive than those of other members of the group (save olympus) but
differ in certain cranial and dental characters. A striking character is
the enormous development of the sagittal crest, which in old males is
considerably more than half an inch (13 mm. -j- ) in height and strongly
arched. In the females it is also well marked though by no means so
high. The breadth between the orbits is much greater than in any of
the other known forms except Olympics^ which agrees with it in this
respect. In adult males of hippolestes the interorbital breadth varies
from 47 to 1^0 mm; in adult females from 38 to 41 mm. In Felis
aztecus from Mexico, Texas, southern Arizona, and southern Califor-
nia, the interorbital region is very much narrower, measuring in adult
males fi^om 38 to 41 mm. and in adult females from 34 to 36 mm.
Similarly, the breadth across the postorbital processes, though more
variable, is much greater in hippolestes^ varying from So to 84 in adult
males and from 71 to 76 in adult females. In males of aztecus this
breadth varies from 68 to 75 ; in females 65 to 69. The braincase on
the other hand is as broad in aztecus as in hippolestes.

The upper carnassial and the large premolar in front of it are much
heavier and broader in hippolestes than in aztecus., while the length
of the crowns is about the same in both. The lower molariform teeth
are larger throughout in hippolestes., particularly the lower carnassial,
which is conspicuously larger and more massive than in aztecus.

REMARKS ON THE ROOSEVELT SERIES OF SKULLS FROM

COLORADO.
The 13 Cougars or Mountain Lions killed by Mr. Roosevelt near
Meeker, Colorado, in January and February 1901, with their accom-
panying data as to size, weight, color, and time of breeding, form the
most important series thus far recorded and their skulls afford an ex-
ceptional opportunity for ascertaining the constancy or inconstancy of
particular cranial characters and proportions. The series comprises 3
adult males and 9 females of various ages. All were very fat and had
been feeding on deer. The males are much larger than the females
and have much larger teeth. One male is a giant, and has the largest
skull of any Puma I have seen, surpassing even the type specimen of
Felis hippolestes from the Wind River Mountains of Wyoming. In
the flesh this animal measured 8 feet in total length from nose to tip of
tail, and weighed 227 pounds. Among the females was a young of
the preceding 3'ear, which had not yet shed its milk teeth, and which
was still following its mother. The full data for these specimens, as
furnished me by Mr. Roosevelt, are as follows :

REVISION OF THE PUMAS

587

No.

Sex.

Len
Ft.

gth.'
In.

Weight
lbs.

Color.

I

<?

8

227

red

2

<?

7

8

164

blue

^

^

7

6

160

red

4

9

7

133

blue

5

9

6

9

124

red

6

9

6

7

120

blue

7

9

6

7

118

blue

8

9

6

9

108

red

9

9

6

5

105

blue

10

9

6

4

102

red

II

9

6

80

blue

12

9yg-

4

ii'A

51

red

Feb. 14, 1901
Jan. 27, 1901
Feb. 13, 1901
Feb. 8, 1901

Feb. 5, 1901
Jan. 31, 1901

Jan. 15, 1901

Jan. 18, 1901
Feb. 12, 1901

Jan. 28, 1901
Jan. 14, 1901
Feb. 12, 1901

Remarks.

Contained 3 foetal young;
would have been born in
month or 6 weeks.

Contained 3 fcetal young ;
would have been born in a
few days.

Mother of 3 kittens, still
blind, which were only three
or four days old.

Milk in teats ; accompanied
by No. 12 [and undoubtedly
its mother].

Mating with No. 2.

With No. 9 [and undoubt-
edly its last year's kitten].

CRANIAL MEASUREMENTS OF THE ROOSEVELT SERIES OF
PUMAS FROM COLORADO.

1

B

3

<u
m

s

a

'u

'O

a
"3

C tn
<u a

— D

few

tn 0

CO

05

0

0 S

KM

o|

0 a

0

Foramen mag-
num to plane
of back of last
pre-molar.

"5

'2'S
0 a

1

OS

1'

Depth of skull

(height of fron-

tals above

presphenoid.)

0

0,2

ca
0

108681^ old.

I

196

190

16.^

67

102

122

47

207

68

159

23

108682^ ad.

2

184

178

148

68

96

II2.5

46

191

65

143

24

108683^ "

?,

148

25

1086849 "

4

172

167

1.^4

62

91

105.5

.38.5

183

58

136

20.2

1086859 "

5

166

161

132

60

86

IOI.5

38

176

57

130.5

21.8

1086869 "

6

170

165

1.^8

60

88.5

102.5

41

172

58

134

22.5

108687 9 ygad.

7

163

158

I.S5

57

85

100

38

173

57

129

22

108688 9 old.

8

1595

155

130

56

83

95

37-5

167

57

126

22

1086899 "

9

16

159

134

56

83.5

99

40

174

57

129

22

1086909 "

10

I

158

133

5^

85

98

40

I7J

58

129

21.5

1 08691 9 ad.

II

I

I52..S

130

55-5

82

95

37

167

5(>

126

21

1086929 juv.

12

129-5

125

105

47

64.5

31

134

47

lOI

milk
tooth

SEXUAL VARIATION.

T'/ie skicll. — The above measurements show that the sexual differ-
ence in size is greater than is usually recognized, the basal length of
the largest male (196 mm.) exceeding that of the largest female (172
mm.) by 24 mm. (about an inch). In the oldest male the ratio of

' Length in straight line, between uprights, from nose to tip of tail.
Proc. Wash. Acad. Sci., December, 1901.

588

MERRIAM

zygomatic breadth to basal length is 84 (essentially the same as in the
type of the species, from the Wind River Mountains, Wyoming) ; in
the two oldest females it is about 81.5. The mean interorbital breadth
in the males is 46.^ ; in the females 38.7. The sexual differences are
well shown in the following table :

TABLE OF AVERAGE OR MEAN CRANIAL MEASUREMENTS.

Mean of males

Mean of females...

190
160.4

01 o
08

n

cd>S
o cQ

*J3
CO ho

6 "=

i 155-5 67.5
155-6,130.1; 56-6

99
83.1

3 oj

p a I
Sol

a a (

05

q ti

II7.2 46.5 199 65.5
99.537.8:1685 56.1

P-d

►3 >,

150 24

126.7 21.6

77/^ teeth. — The teeth which show sexual differences to best advan-
tage are the canines and carnassials. The diameter of the upper
canine (measured on the cingulum) varies in the males from 14 to 15
mm. ; in the females from 1 1^ to 12^ . The length of crown of the
upper carnassial varies in the males from 33 to 25 mm. (mean 24) ;
in the females from 30.2 to 22.5 (mean 21,6).

INDIVIDUAL VARIATION.

The skull. — Individual variation in this series is less than was ex-
pected. The differences in size of the two perfect male skulls (amount-
ing to 12 mm. in basal length and 15 mm. in zygomatic breadth) ap-
pear to be attributable to age, the smaller one being a young-adult with
practically unworn teeth ; the other a very old male with much worn
teeth. In the females, individual variation in size is more marked.
The largest and the smallest females are both young-adults with only
slightly worn teeth; one has a basal length of 157 mm.; the other
of 172 mm. — a difference of 15 mm.

The parts of the skull presenting the largest amount of individual
variation are the nasal bones, the breadth of frontal across postorbital
processes, the breadth of postorbital constriction, and the audital bullae.
The nasal bones in the two perfect males measure respectively :
length 58 and 53 mm. ; breadth between tops of premaxillae 30 and 29 ;
breadth between maxIlUe posteriorly 20 and 16. In 4 of the 8 females
the nasals are surprisingly uniform, the length falling beteen 45 and
46 mm.; breadth between points of premaxillte 24 to 26; breadth

REVISION OF THE PUMAS 589

between maxillae posteriorly 15-17. In the two oldest females the
length of nasals is 44 and 41 mm. The breadth across the postorbital
processes in the two males is 84 and 81 — the difference apparently due
to age. In the females it varies from 67 to 75 mm., and the skull
in which it is only 67 is young and the frontals have not attained the
full adult form. In the others it varies from 70 to 75.

The postorbital constriction in 6 of the 8 females falls between 40
and 43 mm. In the other two skulls, both of which are very old, this
part of the skull is broader, measuring 44.5 in one and 45.5 in the
other. In many Carnivores (and other mammals as well) the breadth
of the postorbital constriction decreases with age. In the Pumas it
appears to increase with age. I am not perfectly sure however that
the increased breadth in the two old skulls may not be due to the pres-
ence of parasites in the frontal sinuses, as is often the case in the
Mustelidas.

The bullae are surprisingly constant in size and form throughout the
series except in one skull, in which they are greatly inflated. This
is the skull of a very old female and is the same in which the postor-
bital inflation is greatest, hence there seems to be ground for the sus-
picion that in both cases the increased 'size may be the result of dis-
ease. In this skull the breadth of the basioccipital is materially
decreased by the great size of the bullae, measuring 4 mm. less than
in a skull of approximately equal age with normal bullze.

A curious feature connected with these skulls is the tendency to
cleft palate. One of the males and two of the females have each a
large perforation in the middle of the roof of the mouth. This
perforation or vacuity extends completely through the palatine bones
on the median line, measures 13 mm. in length in two and 10 in
the third, and is bifurcate anteriorly, the anterior ends notching the
posterior margin of the palatal floor of the maxillary bones on each
side of the vomer. Three other females have a complete perforation
a little behind the maxillo-palatine suture, but of considerably smaller
size. Hence six out of the twelve specimens have the palate com-
pletely perforated, and in four of these the perforation is of consider-
able size and bifurcate anteriorly.

The teeth. — Individual variation in the teeth in the 3 male skulls of
the series is small (only i mm. in each direction from the mean) and
mainly attributable to the usual wear with age. In the female it is
somewhat greater, the extremes in length of crown of upper carnas-
sial falling 1.4 mm. below the mean, and reaching 0.9 mm. above the
mean. Part of this also may be attributed to wear.

590 MERRIAM

FELIS HIPPOLESTES OLYMPUS Merriam. Pacific
Coast Cougar.

} Felix oregonensis Rafinesque, Atlantic Journal, Vol. I, No. 2, p. 62, 1832.
Felis hippolesies olympus Merriam, Proc. Biol. Soc. Washington, Vol. xi,

p. 220, July, 1897. [The type specimen proves to be a 9. not a ,? , as

marked by collector.]
Felis oregonensis Stone, Science, NS., Vol. ix, pp. 34-35, Jan. 6, 1889.

Tyfe locality. — Olympic Mountains, State of Washington.

Geographic distribution. — Coast region of British Cohimbia,
Washington, Oregon and California.

Characters. — vSimilar to hippolestes^ but color much darlvcr ; whitish
areas of underparts more restricted and less white ; tail decidedly
darker (dark all round, not grayish white below), the black tip
longer ; nose and face very much darker.

Cranial characters. — Skull large and massive ; frontals enor-
mously swollen, elevated, narrow, and highly arched transversely,
with tendency to the development (especially in young-adults) of a
median longitudinal sulcus ; frontals much swollen over postorbital
processes and anterior part of orbits. Skull similar in size, massive-
ness, and general characters to that of hippolestes, from which it
differs in having the facial and frontal regions more abruptly elevated ;
the frontals narrower, more swollen, and more highly arched trans-
versely, with usually a median longitudinal depression or groove ;
nasals with more decided angle or ' hump ' at junction of upper and
middle thirds (upper third horizontal, then bent abruptly downward) ;
basal part of skull longer, giving greater length to the occipito-sphenoid
and postpalatal measurements; pm. y slightly longer; pni. ^ thinner
(especially posteriorly) ; pm. ^ thinner.

Measuretnents. — Type specimen, 9 ad. (from well made dry skin) :
total length 2095 ; tail 775 ; hind foot 260.

Cranial ^neasuremeftts. — S yg. ad. from Vancouver Id., B. C, and
$ ad. (type specimen) from Olympic Mts., Washington (measure-
ments of 9 in parentheses) : Basal length 192 (161. 5); zygomatic
breadth 153.5 (^27) ; occipito-sphenoid length 70.'' (60) ; postpalatal
length 120.? (97) ; interorbital breadth 45.5 (39) ; upper carnassial

23-5 (22.5)-

Respecting the technical name of the Pacific Coast Cougar, a few

words may not be out of place : Witmer Stone, in Science of Jan.
6, 1899 (Vol. IX, pp. 34-35), announced that he had recently un-
earthed an old name, given by Rafinesque in the Atlantic Journal in
the summer of 1832, to a Cougar from western America. This name
Mr. Stone adopts for the northwest coast animal in place of the name

REVISION OF THE PUMAS 59I

olympus given by me in 1S97. He quotes Rafinesque's original de-
scription as follows: " i. Var. Oregonensis. Dark brown, nearly
black on the back, belly white ; body 6 feet long, 3 high, tail 2 or 3
feet long. A large and ferocious animal of the mountains. Is it not
a peculiar species? Felix oregonensis.^^

Rafinesque knew nothing of the animal he attempted to describe.
He had never seen a specimen, nor had he ever been within two thou-
sand miles of the northwest coast country. Furthermore, his brief de-
scription is grossly incorrect, as the animal is not " nearly black on
the back," and its belly is not white. In this, as in numerous other
cases, his information was obtained from travelers at second or third
hand. Who now-a-days would think of proposing or accepting a
modern name based on hearsay accounts of travelers and not even ac-
companied by the mention of a definite locality ?

In cases where the description of an old author leaves no room for
doubt as to the species meant, or where a modern author adopts or re-
defines an old name, as I have sometimes done myself, its status is of
course fixed and not subject to change. But the deliberate displacement
of a name based on a definite type specimen from a stated locality and
accompanied by an adequate description, in order to replace it by an
older name of exceedingly dubious applicability and unaccompanied
by either definite locality or adequate description, involves, in my
judgment, a fallacious interpretation of our principles of nomenclature.

In the case in point, Mr. Stone candidly admits that "no definite
locality is given under the description of F. oregonensis^" and it is
only by implication that the name can be attributed to the Oregon
country. In this connection it should be borne in mind that Oregon
in those days embraced not only the present states of Oregon and
Washington, but also Idaho, the western part of Montana, and much
of British Columbia. In the same paper Rafinesque names another
species, Felis macroura, said to be not larger than a cat, "with tail
as long as the body, which is from one to two feet long only." Even
Mr. Stone admits that " the source of this information is unreliable
and the probability is that no such animal existed " !

In the introductory paragraph of his paper Rafinesque says (also
quoted by Mr. Stone) : "In addition to the article on our Couguars,
page 19 [in which 5 varieties are characterized from Carolina, Green
Mountains, Kentucky, New Hampshire, and Pennsylvania] , I have to
state that several other varieties of tygers are found in the western
wilds of the Oregon mountains, or east and west of them, which de-
serve to be noticed. I find in my notes that two other varieties of

592

MERRIAM

Couguar have been seen there, and east of the mountains." Hence it
appears that Rafinesque, as long ago as 1S32, recognized at least 8
varieties of cougar as inhabiting the United States from Carolina and
Kentucky northward. I find no evidence to show that he ever saw a
specimen of any of these, his descriptions (if such they may be called)
being based wholly on the accounts of hunters and travelers, assisted
by his fertile imagination.

FELIS HIPPOLESTES AZTECUS subsp. nov.
Mexican Cougar.

Type from Colonia Garcia, Chihuahua, Mexico. No. 99658 $ ad.

U. S. National Museum, Biological Survev Coll. Oct. 17, 1899,

H. A. Cluff. Orig. No. 2401.

Characters. — Animal large and powerful but decidedly smaller
than hippolestes ; general color dull fulvous as in hippolestes^ but tail
darker, browner, with longer black tip and no white underneath
(much as in olympus) ; ears almost wholly black.

Cranial characters. — Skull large and massive, similar in general
to that of hippolestes but decidedly shorter ; braincase of same size;
frontals narrower interorbitally (the interorbital breadth in adult males
averaging about 40 or less instead of 48 or more) ; sagittal crest less
highly developed ; upper carnassial teeth lighter and thinner ; lower
teeth (particularly the carnassial) smaller throughout.

Remarks. — Compared with coryi from Florida the whole animal
is very much paler and grayer ; top of head and upperparts generally
dull grayish fulvous instead of intense ferruginous ; underparts whiter ;
black of ear not extending below actual base of ear ; color of chin and
lips very different, being soiled white instead of intense yellowish buff.

A skin in the collection of the Biological Survey from Fort Bowie,
Arizona, is paler and grayer (less fulvous) than the skins from Chi-
huahua, and has the back much more conspicuously mixed with black
hairs. The same is true in even greater degree of a skin in the
National Museum from Eagle Pass, Texas.

The skull of aztecus is considerably smaller than that of hippolestes
and may be characterized briefly as short., heavy, and massive, with
narrow interorbital region., elevated arched frontals, and well devel-
oped sagittal crest — though never so high as in hippolestes. Three
skulls of adult males from a single locality, Colonia Garcia, Chihuahua,
exhibit a most amazing variation in the size "of the audital bullae, a
degree of variation which in most mammals would characterize per-
fectly distinct species. In one of these skulls (No. 99660) the bullae

REVISION OF THE PUMAS 593

are small, low, and narrow and not of the same size. In another (No.
99659) they are enormously swollen and inflated posteriorly, and rather
sharply pointed anteriorly. It is probable that the medium condition
shown in No. 99658 is the normal, that No. 99659 is exceptionally
inflated, and that No. 99660 has the bullae imperfectly developed.
This is the more probable because of the difference in the bullae on
the two sides in No. 99660, the left being considerably smaller and less
perfect than the right.

Measure?nents. — Type (^ ad.) : total length 2268; tail vertebras
731 ; hind foot (from dry skin) 270 ; height at shoulder 731. Average
of 3 males from type locality : total length 2074 ; tail vertebrae 77^ '
hind foot (from dry skin) 260; height at shoulder 714. An adult
female from type locality: total length 1814; tail vertebrae 630; hind
foot (from dry skin) 230 ; height at shoulder 630.

Cranial measurements. — Largest of 3 males (the type), and
largest of 3 females (female in parentheses) from Colonia Garcia,
Chihuahua, Mexico: basal length 171. 5 (149); zygomatic breadth
142 (121) ; occipito-sphenoid length 60 (53) ; postpalatal length 91
(76.5) ; interorbital breadth 41 (36.5) ; upper carnassial 22 (20).

FELIS CONCOLOR Linn. Brazilian Puma.

Felis concolor Linn^us, Syst. Nat., ed. xii, Addendum to Vol. iii, Mantissa
Plantarum, p. 522. 1771.

Type locality. — Brazil (probably southeastern Brazil).

Characters. — Size medium ; color apparently yellowish fulvous ;
cranial and dental characters distinctive.

Color. — The original diagnosis by Linnaeus is merely " cauda elon-
gata, corpore immaculato fulvo." It is based mainly on Brisson, who
states that the color is reddish yellow. Burmeister (whose description
is apparently of a specimen from Neu-Freiburg, southeastern Brazil)
gives the color as rather pale reddish yellowish gray ; with throat, chest
and inner sides of arms and thighs pure white ; belly yellow ; outside of
ears blackish brown with a yellowish spot ; tip of tail blackish brown.
(Thiere Brasiliens, Saugthiere, pp. 88-89, 1854).

Cranial characters.^ — Skull of medium size, thin and light com-
pared with the North American and Chilean species ; face and frontal

' The present description is based wholly on two skulls from Brazil in the col-
lection of the Biological Survey: an adult male, No. 100118 from Piracicaba,
Sao Paulo ; and an adult female without particular locality. For both I am in-
debted to the courtesy of Prof. Herman von Ihering. The male when young had
suffered some injury to the top of the head, in the posterior frontal region, as if
he had been dealt a blow with a machete. The resulting deformity may have pre-
vented the development of the sagittal crest.

594 MERRIAM

region elevated and set forward ; ascending arms of maxillje much
more nearly vertical than in J^. puma; interorbital region rather
broad ; sagittal crest absent or only slightly developed ; nasals short,
broad and blunt posteriorly ; no pit in frontal over apex of nasals
[present in all other species] ; suture between basi-sphenoid and pre-
sphenoid closed in adults ; presphenoid very broadly exposed, its base
broader than in any other species ; pterygoid wings correspondingly re-
duced ; under javj srnall, light afid straight^ not enlarged or swollen
beneath the teeth ; symphysis shorty truncate, and more anterior than
in the other species ; bullae rather large ; ratio of zygomatic breadth
to basal length in an ad. ^ 80; teeth rather large; upper carnassial
distinctive, having a well-marked salient angle on outer side between
middle (paracone) and posterior (metacone) cusps, which is not pres-
ent in any other species.

Compared with F. batigsi from Colombia and Peru, the skull of
concolor is larger ; f rontals decidedly more elevated ; bulhe larger ; teeth,
particularly carnassial and middle upper premolar, larger ; upper car-
nassial with a salient angle on outer side as already described. Com-
pared with F. puma from Chile the skull as a whole is lighter, frontals
more elevated; 2d premolars, above and below, very much smaller;
crown of lower carnassial much shorter and more swollen. (For more
detailed comparison see under F. puma.)

Compared with the North American members of the group, Felis
concolor irovn Brazil differs markedly in cranial characters. The skull
as a whole is lighter, the braincase thinner and more smoothly rounded,
and in the specimens at hand devoid of sagittal crest. The nasal bones
are more bluntly truncate posteriorly than in any skulls I have seen from
Mexico or the United States, and the pit at point of junction with the
frontal bone is absent. The presphenoid, viewed from below, is much
more broadly exposed, and the pterygoid wings are correspondingly
shortened. The under jaw is straighter and lighter than in any of the
North American forms and lacks the usual swelling beneath the teeth ;
the lower border of the ramus is slender, narrow, and straight ; the
symphysis is shorter, more anterior, and more truncate. In an adult
male from Sao Paulo, Brazil (No. 1001 18) the bullae are enormously
inflated and the inflation is much broader than in any others I have
seen. The only approach to it is in a specimen from Colonia Garcia,
Chihuahua, Mexico (No. 99659), in which the posterior part of the
bulla is greatly swollen, but the anterior part is abruptly narrowed.
The skull in question from Colonia Garcia is abnormal in this respect,
the bullae being strikingly larger than in other specimens from the same

REVISION OF THE PUMAS 595

locality. The large amount of individual variation in the bullae leads
me to distrust their reliability as affording characters. Their great size
in the Sao Paulo specimen may be abnormal.

Cranial measurements. — An adult Z (No. 100118 U. S. N. M.)
from Sao Paulo, Brazil, and a yg-ad. $ from Brazil (exact locality un-
known— the 9 in parentheses): Basal length 171 (141); zygomatic
breadth 139 (112) ; postpalatal length 88. 5 (72) ; interorbital breadth
41 (31) ; upper carnassial 23.5 (22).

FELIS BANGSI sp. nov. North Andean Puma.

Type from Dibulla, Colombia, South America. No. 8413 $ ad.

Bangs Collection. October 8, 1899. W. \V. Brown, Jr.

Characters. — Size medium; pelage (in type specimen) short and
rather coarse ; color I'usty or reddish fulvous ; skull rather light ; teeth
smallest and slenderest of the known species.

Color. — Upperparts from head to tail ferruginous, not noticeably
mixed with black hairs and rather narrowly confined to top of back,
shading to dull fulvous on sides ; the fulvous becoming still paler on
belly, throat, legs, and feet ; inguinal and pectoral regions whitish ;
chin and lips buffy white, the upper lip posteriorly suffused with pale
dull fulvous ; ears blackish except at extreme base ; tail above ferrugi-
nous, becoming dark rusty brown, but without trace of dark median
stripe ; below grayish brown, the dark brown underfur showing
through ; tip black, and in the type specimen indistinctly annulated,
there being an incomplete blackish ring above the tip, preceded and
followed, on the sides, by grayish fulvous intervals.

Cranial characters. — Skull of medium size and rather light ;
sagittal crest in male, high where it ci^osses occipital sulcus, and con-
tinuing in a low ridge to frontals ; frontals not elevated and rather
flattened; nasals (in type specimen) moderately broad and obliquely
truncate posteriorly ; females without trace of sagittal crest except
across occipital sulcus ; bullae and basisphenoid normal. Teeth smaller
than in any other known species ; canines very slender ; upper car-
nassial in adult male measuring only 22 mm. Sexual differences in
size unusually marked, both in skull and teeth. Skull similar in gen-
eral to that of concolor ixovci Brazil, from which it differs in flatter and
lower frontals, presence of distinct sagittal crest (in adult male) ,
smaller bullae ; narrower exposed base of presphenoid ; persistence of
sphenoid suture ; much more slender canines, smaller and narrower
premolars and carnassials, and absence of outer angle on upper car-
nassial (the place of this angle being concave). Compared with F.

596 MERRIAM

puma from Chile, the skull is smaller (particularly shorter') and much
less massive ; anterior base of coronoid process much less broadly ex-
panded ; teeth strikingly smaller throughout — the difference much more
marked than in the comparison with co?tcolor.

Remarks. — Mr. Bangs has sent me two skins with skulls from
Colombia. One (the type) is an adult male collected October 8 at
Dibulla; the other an adult female collected February 15 in the Santa
Marta Mountains. The Santa Marta specimen has narrow ' pinched
in ' nasals (the posterior two-thirds abruptly narrower than the anterior
third), and differs so strikingly in pelage and markings as to suggest
the possibility that it may be subspecifically distinct — a high mountain
form perhaps. The ears appear to be longer. The pelage is long and
full and has a very different appearance from that of the short smooth
coat of the Dibulla specimen. The rusty dorsal area (from head to
tail) is broader and darkened by profuse admixture of black hairs ;
dull fulvous of sides also strongly mixed with black hairs ; ferruginous
of tail continues to black tip ; tip not annulated ; whitish of under-
parts much whiter and much more extensive, completely covering
inner sides of legs and in foreleg reaching continuously to pads on
vmderside of foot ; outer sides of forelegs and feet grayish fulvous in-
distinctly vermiculated with brownish.

Two skulls from Peru, in the American Museum of Natural History,
appear to be young-adult males of this species.

Measurements. — (Female ad. no. 8147 Bangs Coll., from Santa
Marta, Colombia, measured in flesh) : Total length 1600 mm. ; tail
vertebrae 610; hind foot 225 ; ear 80.

Cranial measurements. — An ad. $ (the type) ; and an ad. $ (in
parentheses) from Santa Marta Mountains, Colombia, both in the
Bangs Coll. Basal length 162 (141) ; occipito-nasal length 173 (151) ;
zygomatic breadth 128 (i 15) ; occipito-sphenoid length 60 (50) ; post-
palatal length 85 (72.5) ; interorbital breadth 37 (31.5) ; upper car-
nassial 22 (20).

FELIS BANGSI COSTARICENSIS subsp. nov.
Central American Puma. ;

Type from Boquete, Chiriqui, Panama (or Costa Rica — the province

of Chirqui being claimed by both countries). No. 10118 9 ^d.

Bangs Coll. April 22, 1901. W. W. Brown, Jr.

Characters. — Similar to F. bangsi but color very much darker and

redder, particularly on the belly which is red like the sides, with only

an indistinct narrow whitish line down the middle, barely connecting

the whitish pectoral and inguinal areas ; fur between foot pads black.

REVISION OF THE PUMAS 597

Measurements . — (Type specimen, 9 ? measured in flesh) Total
length 1680 mm. ; tail vertebrae 6S0; hind foot 220; ear 75.

Remarks. — The U. S. National Museum contains two skins from
Costa Rica, presented by the World's Fair Costa Rica Commission in
1893. In both of these the whitish pectoral and inguinal regions are
suffused with dirty yellowish brown, which may be due to immersion
in pickle. The skull of one of these (the female, No. 61 193 U. S.
National Museum) agrees essentially with the skull of the type speci-
men except that the nasals are longer, and with the female bangsi from
the Santa Marta Mountains, except that the nasals are not narrowed as
in that specimen. The skull labeled as belonging to the skin of the
male (No. 61 194) has the middle upper premolar and carnassial greatly
swollen and presents other characters which lead to the suspicion that
it really has nothing to do with this skin, but probably came from some
other region. This view is the more probable since the skull of a
fully adult male from Pacuare, Costa Rica (No. 15967 U. S. Nat.
Mviseum) agrees closely with the type specimen of batigsi (a male
from Dibulla, Colombia) and, allowing for normal sexual differences,
with the female from Costa Rica. In this connection it may be stated
that the lower carnassials in females of bangsi and costaricensis are
surprisingly small.

FELIS PUMA Molina. Andean Puma.

Felis puma Molina, Saggio sulla storia Naturale del Chili, pp. 295-297, 1782.

Type locality. — Chile.

Characters. — Size large ; color gray or grayish instead of fulvous.
According to Molina the animal has a mane. Skull and teeth large
and massive.

Color. — Upperparts " ash-colored with a sprinkling of yellowish
. . . underparts whitish." — Molina. The skin of the head of a speci-
men from Santiago, Chile (No. \%\%\ $ young, U. S. Nat. Museum)
has the nose and face brownish gray, becoming pale dull fulvous on
top of head ; upper eyelid, lips and chin yellowish white ; ears grayish,
mixed with black hairs at base and toward tip but without distinct
dark spot ; a small blackish streak on upper lip just below whiskers ;
occiput and back of neck dull grizzled grayish fulvous.

Cranial characters. — Skull and teeth large and massive; 2d pre-
molar, above and below, much larger than in any other known species
of the group; inner tubercle (protocone) of upper carnassial very
large. The skull of a young male from Santiago, Chile (No. 36851 U.
S. National Museum), contrasted with an adult male of concolor from

598 MERRIAM

Sao Paulo, Brazil (No. 100118U. S. National Museum, Biological
Survey Coll.) presents the following differences : nasals more angular
posteriorly (less bluntly truncate) ; bullae smaller; basioccipital broader
between bullae ; premaxillae more swollen over canines (owing to
larger size of these teeth) ; presphenoid much narrower posteriorly
between wings of pterygoids ; suture between presphenoid and basi-
sphenoid not closed ; under jaw more massive and more swollen under
tooth row ; basal half of anterior border of coronoid process much
more broadly expanded transversely ; upper incisors and canines much
larger and heavier; middle upper premolar and lower premolars
larger ; body of upper carnassial thinner and lacking the external angle
of concolor (the part occupied by the angle in cowco/or being concave) ;
inner tubercle (protocone) very much larger than in any other Puma
1 have seen ; posterior end of crown of middle upper premolar curved
strongly outward ; 2d lower premolar much broader posteriorly ; and
nearly twice as massive as in coitcolor ; lower carnassial with anterior
blade more flattened or ' dished ' on outer side ; upper molar twice
as large as usual (crown 10 mm. in length).

Remarks. — The skull from Santiago, Chile (36851), on which the
present description is based, is not full grown (evidently in second
year) , and the cranial characters are not full}- developed. The dental
characters, however, are shown to unusual advantage, the teeth being
unworn. The most noteworthy tooth characters are the enormous
size of the 2d upper and lower premolars and upper molar, and the
large size of the inner tubercle (protocone) of the upper carnassial.
The upper carnassial measures 24 x 12.S mm. The skin of the head
and neck of this specimen differs from the type of J^. puma patagon-
ica from east base of Andes, Patagonia, in having the face and head
darker and browner (quite different from the clear gray of patagonica) ;
neck decidedly moi^e fulvous ; ears without distinct markings.

FELIS PUMA PATAGONICA subsp. nov.
South Andean Puma.

Type from east base of Andes, Patagonia (lat. 47° 30'). No. 108693

[ (?yg-ad.] U. S. National Museum, Biological Survey Coll. 1899.

J. B. Hatcher.

Characters. — Size large ; color gray ; black of ears nearly obsolete ;
skull and teeth large and massive ; frontals low (depressed contrasted
with concolor and the North American forms) , giving the cranium an
appearance of unusual length ; tooth characters distinctive.

Color. — Upperparts ash gray, everywhere abundantly mixed with

REVISION OF THE PUMAS 599

black hairs, the median dorsal region from top of head to end of tail
suffused with buffy ; belly buffy gray, the hairs much longer than on
the back ; pectoral and inguinal regions soiled whitish ; throat gray ;
chin and lips white, the posterior part of upper lip and cheeks suffused
with buff; patch at base of whiskers gray (instead of black) ; rest of
face gray, except a buffy streak above and below the eye ; backs of
ears gray, with apical third and a faint basal band dusky ; feet buffy
gray ; tail pale buffy fulvous above, ash gray below, with only the ex-
treme tip dusky.

Cranial characters. — Skull massive, elongate; facial and frontal
regions low and narrow ; interorbital region narrow ; nasals rather
broad posteriorly and obliquely truncate ; sagittal crest low, contin-
uous to frontal shield ; bullae rather large ; postpalatal notch broad ; the
palatal margin nearly straight ; presphenoid normal. Canines large
(the upper ones 14.5 mm. in diameter) ; crowns of premolars and
carnassials long ; upper molar small (as usual in the group) ; inner
cusp (protocone) of upper carnassial very small. Compared with the
skull of JF. puma from Santiago, Chile, the following tooth differ-
ences are conspicuous : crown 2d upper premolar larger and thinner ;
crown of upper carnassial longer, with inner tubercle greatly reduced ;
upper molar small (less than half the size of this tooth in puma') ; 2d
lower premolar smaller (decidedly less swollen) ; lower carnassial de-
cidedly larger.

Compared with an old male concolor from Sao Paulo, Brazil, the
facial and frontal regions are very much lower ; frontals flatter ; bullge
smaller ; presphenoid narrower ; sphenoid suture distinct ; postpalatal
notch broader ; sagittal crest present ; under jaw more massive, the sym-
physis much longer ; canines much larger ; crown of second upper pre-
molar much longer and narrower ; crowns of upper and lower carnas-
sials (particularly the lower) decidedly longer ; crowns of lower
premolars longer, the 2d much more swollen posteriorly.

Re?>iarks. — Compared with the skin of a head of I^. puma from
Santiago, Chile, the color of the corresponding parts in patagonica is
lighter and everywhere much grayer (less brownish and without fulvous
cast) ; ears with distinct dark markings at base and tip ; upper lip with
only a trace of the dark mark below whiskers ; lips and chin whiter ;
light band on under eyelid much broader and purer. Compared with
subspecies pearsoni Thomas, the general color is much paler and
grayer, the ear and face markings much more distinct, the tip of tail
dusky.

Measurements, — (From dry skin.) Total length 2015 ; tail verte-
brae 670.

600 MERRIAM

Cranial measureineiits. — Type specimen ( 9 yg-ad.) from base of
Andes : Basal length 168 ; occipito-nasal length 177 ; zygomatic breadth
135 ; postpalatal length 86 ; interorbital breadth 36 ; upper carnassial
24.3.

FELIS PUMA PEARSONI (Thomas). Santa Cruz Coast Puma.

Felis concolor pearsoni Thomas, Annals & Mag. Nat. Hist., 7th Ser., Vol.
VIII., pp. 188-189, September, 1901.

Type locality. — Santa Cruz, Patagonia (about 70 miles from coast).

Characters. — Distinguished from F. puma " not only by its different
general color, but also by its shorter tail, light colored ear-backs, and
the absence of the dark markings round the digital pads." — Thomas.

Color. — " General colour nearest to Ridgway's ' clay-colour,' there-
fore exceedingly different from the nearly ' drab-grey ' of F. c. puvia.
This colour is most vivid along the back, paler laterally on the sides,
but there is nothing that can be called a distinct dorsal dark line.
Under surface whitish fawn, the hairs sandy at their bases, whiter ter-
minally. Face very much like back, darker markings practically ob-
solete ; the usual lighter markings near the eye present but not conspic-
uous. Ears of normal length, their backs imiformly whitish fawn,
without darker markings. Outer sides of limbs like back, inner side
like belly ; ends of fingers and toes whitish, without any darker mark-
ings round the pads. Tail proportionally very short, brownish clay-
colour above, whitish below, the tip not or scarcely darker."

Cranial characters. — In describing this form Mr. Thomas had no
skull. The U. S. National Museum and Biological Survey, however,
have two skulls, evidently male and female, from the coast region of
Santa Cruz, near the type locality of Mr. Thomas' skin. The male
is from Coy Inlet (lat. 51° S., collected by Hatcher) ; the exact locality
of the female is unknown. These skulls agree among themselves, and
differ from F. puma from Chile and F. puma patagonica from the
base of the Andes, in having the frontal region more elevated, the in-
terorbital breadth greater, the nasals very much broader posteriorly and
abruptly truncate instead of angular, and the anterior base of the coro-
noid process of under jaw much less broadly expanded.

Re?narks. — From Thomas' description it appears that the coast
animal (^pearsotzi) is very much yellower than the animal from the
Andes (^patagonica^., and that it lacks the dark ear patches and distinct
face markings of patagonica. The tail appears to be short.

Measurements. — (Type specimen, "tanned and stretched, so that
the measurements are merely approximate"). Head and body 1370
mm. ; tail 530 ; ear 80.

CHARLES ANTHONY SCHOTT 6oi

Charles Anthony Schott.
1826-1901.

Charles Anthony Schott, astronomer, geodesist, magnetician,
mathematician, meteorologist, was born August 7, 1826, in Mann-
heim, Germany, and died July 31, 1901, in Washington, D. C.
Graduating C. E. from the Polytechnic School in Carlsruhe, in
1847, he came to the United States and entered the Coast Sur-
vey in 1848. The rest of his long, industrious and useful life
was spent in that organization and almost exclusively in office
work. For more than half a century he served in its compu-
ting division and for 45 years of that time he was its chief.
The amount of work performed by this diligent, methodical and
industrious scientist was enormous, and his influence upon the
character and work of the Coast Survey probably exceeds that
of any individual ever connected with the organization.

INDEX

Note : — New species in black face type, synonyms in italics.

Abbott, W. L., Mammals collected by,

Natuna islands iii
Achnantlies glabrata 445

lanceolata 445

subsessilis 445
Acrididse 495
^gathoa excisa 567
.?Jschna clepsydra 223

constricta 220

ereviita 223

galapagoensis 382

juncea 219

sitcliensis 222
^sclinidas 219, 382
yEshnajunia 223

Junius 223
Agarum gmelini 430

turneri 431
Agelaeus imtliurni 155
Agelaii, of North America and Mexico
Agelaius phoeniceus caurinus 153

fortis 153

neutralis 153
AgrcEcia cooksoni 497
Agrion annexum 222

hastata 382

hastatum 382
Agrionidse 217, 382
Agromyzidae 378
Ahnfeldtia plicata 435
Alaria cordata 426

fistulosa 426

fragilis 425

fragilis forma bullata 425

lanceolata 426

laticosta 425
Alaska, Harriman Expedition

algae 391

ascidians 225

harrimanella 569

hydroids 157

nemerteans i

odonata 217

willows 297
Algae of Alaska bj' Saunders 391

catalogue of species 396

geographic distribution 392
Amaroucium californicum 254

coei 251

figariuni 253

translucidum 249
Aniphiporus 40

angulatus 41

bimaculatus 44

exilis 54

leuclodus 51

nebulosus 48

neesii 4

on Amphiporus (Continued).

tigrinus 46

virescens 28
Amphiroa epiphlegmoides 442

planiuscula 442

tuberculosa 442
Amphora elliptica 447

ovalis 447
Anasa obscura 365
Anaulocomera darwinii 496
Anax 385

junia 223

Junius 223

spiniferus 223
Ancbineura tibialis 374
Andromeda empetrifolia 574

hypnoides 575

s'telleriana 569, 574
Anisolabis annulipes 492
154 boemansi 492

maritinia f 492
Anomalagrion hastatum 382
Anthomyia setia 375
Anthomyidae 375
Anthrax brachialis 373

lateralis 373

nudiuscula ? 373
Antithamnion boreale 439
Aphanothece microspora 397
Arce's calliste 149
Arcterica oxycoccoides 573
Arctogalidia inornata 131

leucotis 132

stigmatica 132
Arthrodesmus convergens 403
Ascidia adhserens 227, 229

koreana 229
Ascidians from Alaska, by W. E. Ritter
.225

bibliography 257

list of species 226

localities where obtained 225
Ascopera 237
Asilidae 374
Asilus viarginellus Tfj^

Baccha clavata 374

facialis 374
Bacillariaceae 442
Bangia atropurpurea pacifica 432
Bangiaceae 432

Banks, Thysanura and Termitida? 541
Batrachospermum vagum flagelliforme

433
Berytidae 366
Bibliography of ascidians 257

fish skeleton 538

hydroids 189

(603)

6o4

INDEX

Biddulphia aurita 443

Bimeridae 166

Blarina mayensis 559

Blattidce 493

Branner Agassiz Expedition 139

Brazil, Branner Agassiz Expedition ; Mol-

lusk's from Pernambuco 139
Bryanthus s telle ri 574
Boat-tail, Sonoran 151
Bombylidse 373
Borlasia catnillea 4

quinqueliiieata 4
Bostrichobranchus 237
Botryllus magnus 255
Ell crates? cocoa nnus 4,()"
Bulbochsete brebissonii 412

insignis 413

intermedia 413

nana 413

nordstedtii 413

Calithamnion floccosum pacificum 439

plumula 439
Callospermophilus madrensis 563
Calospiza florida arcsei 149
Calotermes galapagoensls 544

pacificus 545
Calothrix fusca 399

scopulorum 399
Calycella 176

syringa 176
Calypteroblastea 169
Cambrian, new species of Olenellus from

267
Campanularia 171

caliculata 170

compressa 170

denticulata 171

gracillima 177

lineata 171

redupUcata 172

regia 172

ritteri 171

serpens 179

speciosa 171

urceolata 172

verticillata 171
Campanularidae 169
Campanulina 176

rugosa 176
Campanulinidse 176
Canace snodgrassii 378
Cannacria baiesii 389

fumipennis 387
Canthecona grandis 364
Capsidce 366
Carinella ii

annnlata 13, 17

capistrata 16

dlnema 15

nothus 17

polymorpha 12

speciosa 11

superba 15
Carinoma 20

armandi 20

grifflni 20

patagonica 21,

tremaphoros 20
Cassiope ericoides 569, 571

Cassiope (Continued).

fastigiata 569, 572

hypnoides 569, 575

lycopodioides 569, 572

mertensiana 572

oxycoccoides 573

redowskii 572

selaginoides 572

stelleriana 569, 574

tetragona 569, 572
Cephalothrix 19

linearis 19
Ce rami a cere 439
Ceramium codicola 439

rubrum 439
Ceratopogon galapagoensls 372
Ceratothaninion pekeanum 439
Ceratothoa gaudichajidii 568

rapax 568
Cerebratulus 74

alblfrons 82

angulatus 75

bell us 71

fuscvs 75

herculeus 75

inipressus 70

longlceps 77

marginatus 75

montgomeryi 80

occidentalis 76
Chictoceros hispidum 443
Cluetomorpha cannabina 413

nielagonium forma rupincola 413
Characeee 416
Chara contraria 416

fragilis 416
Cliironomidte 372
Chlorophycese 409
Chlorospingus sumiclirasti 150
Chorda filum 424
Chordaria abietina 424

flagelliformis 424
Chordariaceae 423
Choreocolax polj'siphoniae 433
Chroococcacese 396
Chroococcus rufescens 396

turgidus 396
Chrysomyia quadrisigfnata 375
Chrysopsis 551
Cicada 370
Cimex perditor 365

sanguinolenta 365
Cladophora arctica 414

flexuosa 414

scopaeformis 414
Cladophoraceae 413
Clathromorphum circumscriptum 442
Closterium acerosum 401

acutum 401

angustatum reticulatum 401

brebissonii 401

dianse 401

juncidum 401

lunula 401

parvulum 401

striolatum 401

venus 401
Clytia caliculata 170

compressa 170
Cocconeis placentula 445

INDEX

605

Cocconeis (Continued).

scutellum 445
Cocconema lanceolatum 445
Codiacese 416
Codium adhaerens 416

mucronatum californicum 416
Coe, W. R., Nemerteans i
Coilodesme bulligera 422

californica 422

linearis 421
Colpomenia siniiosa 421
Conjugatse 400
Conocephalus insulanus 501
Constantinea rosa-marina 441
Coquillett, D. W., Diptera 371
Corallina arbuscula 442

pilulifera filiformis 442
Corallinaceae 441
Cordulia albicincta 220

eremita 220

shurtleffi 220
Coreidae 365
Coreus serratus 365
Coryne bracMata 165

eximia 166
Corynidae 164
Coscinodiscus argus 443

Hneatus 443
Cosmarium bioculatum 403

botrytis 405

broomei 406

caelatum 405

conspersum 405

constrictum 403

contractum 404

costatum 405

depressum 403

gran at um 403

hammeri 403

holmiense 403

intermedium 405

kitchelii 404

latum 404

ochtodes 405

ornatum 406

pachydermum 404

parvulum 404

phaseolus 405

portianum nephroideum 405

pseudogranatum 405

pseudotaxichondrum 406

pulcherrimum 405

pyramidatum 404

quadrifarium 405

sexangulare 404

sphalerostichum 406

subcrenatum 405

tumidum 404

undulatum 404

venustum 403
Cosmocephala beringiana 41, 42
Cosmoclopius (Harpactor) 366
Costaria turneri 431
Cotulina polyzonias 183

tricuspidata 183
Cougar, Adirondack 582

Florida 583

Mexican 592

Pacific coast 590

Rocky Mountain 585

Cougars, cranial measurements of Roose-
velt's 587

table of cranial measurements 588
Coville, F. v., Harrimanella 569

willows of Alaska 297
Cryptonemia obovata 440
Cryptosiphonia grayana 441
Culex tseniorhynchus 372
Culicidae 372
Currie, R. P., Odonata from Alaska 217

from the Galapagos 381
Cycloptilum erraticum 505

lepismoide 505
Cymathere triplicata ^30
Cymbella ehrenbergii 447

insequalis 447
Cyinothoa ga udicha iidii 568
Cymothoidae 567
Cynopterus niarginalus 137

montanoi 137
Cynthia villosa 238
Cystophyllum lepidum 432

Dall, W. H., Mollusks from Pernambuco

139
Delesseria alata 437

brerii 437

crassifolia 437

decipiens 437

sinuosa 437

serrata 438
Delesseriacece 437
Derbesiaceae 415
Derbesia marina 415

vaucheriaeformis 415
Dermocarpa biscayensis 397

prasina 397

fucicola 397
Descriptions of harvest mice by C. Hart

Merriam 547
Desmarestia aculeata 422

viridis 422
Desmarestiaceae 422
Desmidiaceae 400
Desmidium swartzii 408
Desmonema wrangelii 398
Diatoma hyemale 444

pectinale 444
Dictyosiphonaceae 422
Dictyosiphon fceniculaceus 422
Didemnum inarmatum 249

strangulatum 247
Dilsea arctica 441

californica 441
Diplax hudsonica 221
Diptera, by D. W. Coquillett 371
Disphinctium connatum 402

cucurbita 402
Distaplia confusa 246

livida 247

magnilarva 247

occidentalis 245
Distoma adriaticum 245

cristallium 245

laysani 245

pulchra 243
Docidium baculum 402

coronulatum 402

dilatatum 402

minutum 403

6o6

INDEX

Dolichopodidse 374
Dolichopus pustllus 374
Drepanaporus crassus 42

spectabilis 42
Drillia greeleyi 146
Dumontiaceie 440
Dumontia filiforinis 440

Ectocarpaceae 416
Ectocarpus confervoides 418

confervoides corticulatus 418

confervoides pygmaeus 418

corticulatus 418

tomentosus 417
Eisenia arborea 431
Elachistaceae 423
Elachista lubrica 423
Emballonura anambensis 134
Emplectonema 23

biirgeri 25

camillea 4

gracile 23

viride 23
Enallagma annexura 222

calverti 218

cyatliifferttin race annexum 232
EncceliaceEt; 419
Endocladia muricata 434
Enteromorplia crinita 412

intestiualis 411

intestinalisy'orwa cylindracea 411

intestinalisybrw/a maxima 411

linza 410

linza. /or ??t a crispata 411

\inza./ort>ia lanceolata 41 1

micrococca 41 1

prolifera 411
Entomological results of Harriraan
Alaska Expedition, Odonata, 217

Hopkins Stanford Expedition 364
Diptera 371
Hemiptera 363
Odonata 381
Orthoptera 487

Th5'sanura and Termitidse 541
Epitheca albicincta 220
Epitheraia gibba 447

jurgensii 447

westermannii 447
Ephydra gilvipes 377
Ephydridse 377
Epoplithalmia albicincta 220
Erica stelleriana 574
Euastrum affine 408

ansatum 407

crassum 407

didelta 407

elegans 408

gemmatum 407

oblongum 407

pokornyanum 407

verrucosum 407
Eu CO pel I a campanularia 170
Eudendridae 167
Eudendrium bacciferum 166

vaginatum 167
Eudesme virescens 423
Eugyriopsis 237
Eunemerles 23, 3

gracilis 23

Eunevicr tes (Continued).

neesii 4
Eunotia gracilis 445

lunaris 445

robusta 445
Eupolia 61
Euthora cristata 435
Euxesta nitidiveutris 376

notata 376
Expedition, Branner Agassiz, see Brazil

Harriman Alaska, see Alaska

Hopkins Stanford, see Galapagos

Fasciola angulata 41
Fasciolaria aurantiaca 140
Felis aztecus 580, 586

bangs! 580, 595

bangsl costaricensis 595

concolor ^-j-j, 580. 582, 593

concolor Jioridana 583

concolor pearsoni 600

coryi 580, 583

cougar 582

hippolesles 579, 580, 581, 585

liippolestes aztecus 592

liippolestes olympus 590

list of species 582

raacroura 591

oregonensis 590, 591

oregonensis liippolestes 582

olympus 579, 580

patagonica 579, 580

penusylvanica 582

puma 580, 597

puma patagonica 598

puma pearsoni 580, 600
Filellum serpens 179

Fish skeleton, synonom5' of, by E. C.
Starks 507

alphabetical list 527

bibliography 538

list of synonyms 508
Forficasila annulipes 492
Forficulida; 492
Fragilaria construens 444

exilis 444

striatula 444

virescens 444
Fucacese 432
Fucus evanescensybrwia cornuta 432

forma macrocephala 432
Fulgoridic 370
Galapagia solitaria 494
Galapagos, Hopkins Stanford Expedi-
tion, account of, 363

diptera 371

entomological results 364, 371, 381

487. 541 ,

hemiptera 364

isopods 565

odonata 381

orthoptera 487

thysanura and termitidae 541
Galeopithecus volans 134
Garveia annulata 166

nutans 166
Gelidiacese 433
Geranomyia stigmatica 373
Gerridoe 369
Ghilianeila galapagenls 367

INDEX

607

Gigartinacese 434
Gigartina pacifica 435

papi\\a.ta./orma cristata 435

papillata Ag. forma typica 434
Gloiopeltis furcata 440
Gloiosiphonia californica 440
Gloiosiphoniacese 440
Gomontiacea; 415
Gomontia polyrhiza 415
Gomphonema afEne 447

geininatum 447

subtile 447
Gonothyroea inornata 175
Grammaria immersa 178
Grammatophora marina 444
Gratloupiacese 4 '
GrillidcC 502
Gryllus galapageius 502
Gymnoblastea 164
Gymnozyga longata 409

Halecidae 179
Halecium halecinum 179

muricatum 180

ornatum 181

reversum 180

robustum 182

scutum 180

speciosum 18 r
Halmenus choristopterus 496

cuspidatus 496

robustus 496
Halobates sp.? 369

wuellersdorfS 369
Halosaccion firmum 436

fucicola 436

microsporum 436

ramentaceum 436

tilesii 436
Hapalosiphon pumilus 399
Harmostes seriatus 365
Harpactor nigroannulatus 366
Harriman Alaska Expedition, see Alaska
Harrimanella, a new genus of heathers,
by F. V. Coville 569

hypnoides 575

key to species 574

stelleriana 574
Harvest mice, descriptions of, by C. Hart

Merriam 547
Heathers, Harrimanella, a new genus of,

by P. V Coville 569
Hebella pocillum 175
Hedophyllum sessile 429

subsessile 430
Heidemann O., Hemiptera 364
Helminthocladiaceae 433
Hemiptera, by O. Heidemann 364
Hesperomys couesi 287
fulvescens 295

(Oryzomys) palustris 276
Hildenbrandtia rosea 441
Hippelates pusio 378
Hippoboscidae 379
Hipposideros larvatus 135
Homalomyia canicularis 375
Homeostroma latifolium 420

lobatum 420

undulatum 419
Holoquiscalus, kinds of 151

Holoquiscalus (Continued).

martinicensis 151
Homoptera 370
Hopkins Stanford Galapagos Expedition,

account of 363
Hormidium parietinum 412
Hydrodyctiaceoe 409
Hydroids of Alaska, by C. C. Nutting 157

bibliography 189

geographic distribution 157
Hydruraceae 400
Hydrurus penicillatus 400

Icteridse, new species of 151

Icterus cucullatus sennetti 152
gularis tamaulipensis 152
mesomelas taczanowskii 153

Iridaea laminarioides 434
membranacea 4^4

Iridophanes 150

Ischnura 382

Isopods, by Harriet Richardson 565

Isthmia obliquata 443

Jadera sanguinolenta 365

Jalysus (Metacanthus) tenellus 366

Jassus sp.? 370

Jassus (Deltocephalus) sp.? 370

Kalymenia californica 435

lyafoea 177

adhserens 178

dumosa 177

fruticosa 178

gracillima 177

pocillum 175
Lafcfiidae 177
I,aminaria bongardiana 429

bo7igardiana subsessilis 430

bullata 428

cuneifolia 429

saccharina 429

sessilis 429

solidungula 429
Laminariaceoe 424
Leathesia difformis 423
Lepidomorphum yendori 442
Lepisma galapagoensis 541

insularis 543
lyCpismidae 541

I^eptetrum quadrimaculatum 221
Lestes sp. (?) 217
Leucomelina pica 375
Leucophaea surinamensis 494
Leucorhinia hageni 221

hudsonica 221
Libellula flavescens 385

hudsonica 221

hymencea 385

juncea 219

junia 223

quadrimaculata 221
I<ibellulidae 220, 385
Ivicmophora cuneata 444
t-s, granulata 444
'^ pennatula 444
Iviebmannia ? sp. 424
I,igia exotica 568

gattdichaudii 568

68o

INDEX

Ligia (Continued).

(italica) coriacea 568
l,igiidae 568
L,ineus 64

torquatus 66

viridis 65
lyiparoscelis 496

cooksoni 497

pallidus 500

paludicola 499
Lithophyllum farlowii 442
Lithothamnion compactum 442

glaciale 442

Iseve 442
Locustidae 496
Lygaeidae 366
Lyngbya aerugineo-ccerulea 398

Macacus cynomolgus 138

Machilidce 543

Machilis mutica 543

Macrocystis pyrifera 431

Mammals, seven new, from Mexico by

C. Hart Merriam 559
Manis javanica 113
Mantidte 494
Martinique grackle 151
McNeill, J., orthoptera 487
Meinertia gaudichaudii 568
Melobesia patens 441
Melosira granulata 442

nummuloides 443

sol 443

sulcata 443
Menziesia stelleriana 574
Meridion circulare 444
Merriam, C. Hart, harvest mice 547

new mammals from Mexico 559

pumas 577

rice rats 273
Mesogloia simplex 423
Mesogramma duplicata 374
Mesotsenium braunii 400
Mexican oropendola 151
Mexico, new mammals from, by C. Hart

Merriam 559
Micrasterias denticulata 408

fimbriata elephanta 408

kitchelii 408

oscitans pinnatifida 408

rotata 408

truncata 408
Microcladia borealis 440
Microcoleus vaginatus 397
Microcystis marginata 397
Micrura 68

alaskensis 71

bella 71

impressa 70

verrilli 68
Miller, Gerrit S., Jr., mammals from Na-

tuna islands in
Miris lineata 366
Molgula graphica 230

gregaria 231
MoUusks from Pernambuco, by W. H.

Dall 139
Monostroma fuscum 409

groenlandicum 410

vahlii 410

Mountain lion 585
Mulinia branneri 145, 146

portoricensis 146
Mus It el I zeal di 121

integer 119

miilleri 120

neglectus 121

rajah 121

rattus 121

sabanus 121
Musca canicularis 375
MuscidcC 375
Mutica grandis 364
Mycterodus productus 370
Myelophycus intestinalis 420
Myrionema strangulans 423

Nabis punctipennis 366
Nareda stiperba 41, 43
Natuna islands, mammals from in
Navicula apis 446

cyprinus 445

elliptica 445

legumen 446

major 445

pupula 446

radiosa 445

silicula 445
Nemerteans, by W.R. Coe i

discussion of 11

index to 84

keys to groups and species 7

species here recognized 5

species previously collected 3
Nemertes 23

gracilis 23

obsctira 65
Nemobius spectili 503
Nemotelus acutirostns 373

albiventris 373
Nereocystis lutkeanus 431

priapus 431
Nesa'cia 496

cooksoni 497
New birds of families Tanagridae and

Icteridzc, by Robt. Ridgway 149
Nezara viridans 365
Nitella acuminata subglomerata 416

opaca 416
Nitophyllum ruthenicum 437
Nitzschia angularis 446

closterium 446

diadema 446

sigmoidea 446

vermicularis 446
Nostocacese 397
Nostoc commune 397

sphaeroides 398
Nutting, C. C, liydroids of Alaska 157
Nycticebus tardigradus 138
Nysius marginalis 366

Obelia 173

borealis 174

dichotoma 173

dubia 174

plicata 173
Obituary, C. A. Schott 601
Odocoileus hemionus canus 560
Odonata from Alaska, R. P. Currie 217

INDEX

609

Odonata (Continued).

from Galapagos, by R. P. Currie
381

previous collections 381
Odonthalia aleutica 438

kamtschatica 438
CEdogoniacete 412
CEdogonium concatenatuni 412
Olenellus, new species of, from York Co.,

Penn., by A. Wanner 267
Olenellus (liolmia) walcottanus 267

thonipsoni 269
Olfersiafossulata 379
Omaiopiea stimpsonii 41, 42
Ommatius marginellus 374
Oocystis solitaria crassa 409
Ophyra setia 375
Oriole, Alta Mira 152

Sennett's 152

Taczanowski's 153
Oriolus flavus 155
Ortalidse 376
Ortalis notata 376

Orthoptera, from the Galapagos, by J.
McNeill 487

distribution 489
Oryzomys, by C. Hart Merriam 273

albiventer 279

angusticeps 292

aquaticus 277, 279

bulleri 286

chapmani 289

chapmani caudatus 289

chapmani dilutior 290

chapmani saturatior 290

couesi 287

cozumelse 280

crinitus 281

crinitus aztecus 282

fulgens 280, 282

fulvescens 295

goldmani 288

hylocetes 291

jalapae 284

jalapse rufinus 285

melanotis 282, 292, 293, 294

mexicanus 283

mexicanus peragrus 283

natator 277

natator floridanus 277

nelsoni 278

palatinus 290

palustris 276

Palustris natator 277

palustris texensis 276

peninsulae 278

rhabdops 291

richmondi 284

rostratus 293

rostratus megadon 293, 294

rufus 287

teapensis 286

yucatanensis 294

zygomaticus 285
Oscillatoria amoena 397
Oscillatoriaceas 397
Oscinidse, 378
Ototylomys 561

phyllotis 563

phyllotis ptiseus 562

Pantala flavescens 385

liymenaea 385
Panther 582
Paraclius pusillus 374
Paranemertes 32

carnea 37

pallida 36

peregrina 33
Pareuxesta 376

hyalinata 377

intermedia 377

latifasciata 376

ohscura 377
Pediastrum angulosum 409

boryanum 409
Pelagomyia albitalus 373
Penium closterioides 400

digitus 400

interruptum 400

margaritaceum 400

oblongum 400

polymorphum 400
Pentatomidse 364
Periplaneta Australasia 494
Petaurista nitidula 121, 122
Phaeophycese 416
Phoenicothraupis salvini discolor 150

salvini peninsularis 150
Phormidium autumnale 398

laminosum 398
Phycocelis baltica 416
Phyllitis fascia 421
Pipistrellus subulidens 134
Piranga roseo-gularis cozumelse 149
Planaria gesserensis 65

Ihiearis 19

viridis 65
Plettusa stigmatica 373
Pleurococcacese 409
Pleurophycus 427

gardneri 427
Pleurosigma angulata 446

attenuata 446

fasciola 446
Pleurotseniopsis debaryi 406

pseudoconnata 406

ralfsii 406
Pleurotaenium nodosum 402

truncatum 402
Plocamium coccineum uncinatum 436
Plumularia 188

calif ornica 188

lagenifera 188

palmeri 188
Plumularidse 188
Podisus sordidus 364
Polia 61
Polysiphonia arctica 438

bipinnata 438
Porphyra amplissima 433

laciniata 433

miniataybrwa cuneiformis 433

perforata 433

tenuissima 433
Prasiola crispa 412
Preliminary revision of the pumas, by C.

Hart Merriam 577
Protococcaceae 409
Psammaplidium pedunculatum 254
Fseudagelseus 155

6io

INDEX

Pseudolfersia diomedeae 379

fossulata 379
Pteroniys niiidus 121, 122
Pteropus hypomelanus 137

vampyrus 137
Ptilota asplenoides 439

pluniosa 439

serrata 439
Puma Brazilian 593

Santa Cruz 600
Pumas, preliminary revision of, by C.

Hart Merriam 577
Punctaria planta^inea 420
Pylaiella littoralis acuta 418

littoralis densa 419

littoralis macrocarpa 419

littoralis varia 419

Ralfsiaceae 424
Ralfsia clavata 424

deusta 424
Ramphocelus dimidlatus isthmicus 150
Ratufa angusticeps 130

bunguranensis 129

ephippiutn bungiiranetisis 129

ephippium nanogigas 130

ephippiutn sirhassenetisis 129

nanogigas 130

sirhassenensis 129
Reduviidse 366
Red-wing, Northwestern 153

San Diego 153

Thick-billed 153
Reithrodontoniys 275
Reithrodoutomys, descriptions of new-
species of, by C. Hart Merriam 547
Reithrodoutomys chrysopsis tolucse 549

colimae 551

colimse nerterus 551

costaricensis Jalapse 552

difiBcilis 556

dorsalis 557

goldmani 552

griseoflaviis 553

griseoflavus helvolus 554 ^

hirsutus 551

levipes 554

levipes otus 555

levipes toltecus 555

megalotls obscurus 558

megalotis zacatecse 557

microdon 548

microdon albilabris 549

orizabse 550

perotensis 550

saturatus alticolus 556

saturatus cinereus 556

tenuirostris 547

tenuirostris aureus 548
Rhabdonema arcuata 443

biquadratum 443

elegans 443

fauriae 443

Japan ica 443

striatulum 443
Rhicncessa costalls 378
Rhinolophus affinis 135, 136

spadlx 136
Rhinosciurus 131
Rhizoclonium riparium implexum 414

RMzomolgula 231

arenaria 231
Rhodochorton rotbii 440
Rhodomelacese 438
Rhodomela floccosa 438

larix 438
Rhodophycese 432
Rhodophyllidaceae 435
Rhodymeniacetc 436
Rhodymenia palmata 436

pertusa 436
Rhoicosphenia curvata 447
Rice rats of United States and Mexico,
by C. Hart Merriam 273

four groups 275

type localities 274
Richardson, H., isopods 565
Ridgway, R., new birds of families Tan-

agrida; and Icteridae 149
Ritter, W. E. ascidians 225
Rivularia biasolletiana 399
Rivulariaceoe 399

Rodents, a new genus of, from Mexico,
by C. Hart Merriam 559

Salda rubromaculata 368

Saldida:368

Salix alaxensis 311

amplifolia 314

anglorum 337, 338

arbusculoides 323

arctica 326

arctica minor 336

barclayi 316, 317

bebbiana 306

behringica 340

brachystachys 31 1

bnxifolia 336

californica 319

capreoides 311

chamissonis 325

commutata 317

conjuncta 316, 318

crassijiilis 326

curti flora 325

cyclophylla 331

diplodiciya 326

Jlavesceiis 310

fulcrata 319, 320

fuscescens 329

glacialis 329

glauca 321

hookeriana 314

huviillima 323

leiocarpa 335, 336, 338

longistylis 311

mj-rtillifolia 325

myrtilloides 330

niphoclada 322

novseangliae 325

nuttallii 310

orbicularis ,340, 341, 342

ovalifolia33i, 333

palcconeura 336, 337

pallasii 326

perrosirata 307

phlebophylla 335, 336

phylicoides 319, 320, 321

polaris 335, 339, 340

pseudomyrsinites 325

INDEX

6ll

Salix (Continued).

pulchra 319

reticulata 340, 342

retusa 337

richardsoni, 315

rostrata 306

rotundata 331

rotundifolia 338

saskatchavana 324

scouleriana 311

seetnan nil 2,21, 322

sericea 324

sitchensis 307

speciosa 311

speciosa alexensis 311

stolonifera 2,2>2)

unalaschcensis 331, 332

venusta 342
Sarcophaga inoa 375

obtusifrons 374

violenta 375
Sarcophagidae 374
Saunders, D., Alaskan algae 391
Scaphidurus, known forms of 152

major nelsoni 151
Schistocerca literosa 496

melanocera 495
Schizophyceae 396
Schizothrix lacustris 397

lardacea 396
Schott, C. A., obituary of 601
Sciuropterus everetti I2i

phayrei 1 21
Sciurus abbottii 126

bicolor albiceps 129

bicolor bunguranensis 129

bicolor nanogigas 130

lautensis 128

lingungensis 123

loifi 123

lowi natunoisis 123

lutescens 124

natunensis 123

navigator 129

notalus 124, 127, 128

prevostii 129

prevostii navigator 129

procerus 122

rubidiventris 127

rutin ventris 126

seraiae 125

tenuis 122
Scytonemaceae 398
Scytonema figuratum 398

mirabile 398

myochrous 398

variutn 398
Scytosiphon bullosus 421

lomentarius 421

lomentarius complanatus 421
Semnopithecus cristatus 138

natunae 138
Sertularella 182

conica 183

geniculata 184

polyzonias 183

rugosa 183

saccata 183

tricuspidata 183
Sertularia argentea 184

Sertularia (Continued).

cupressoides 185

dichotoma 173

dumosa 177

fabricii 185

halecina 179

muricata 180

polyzonias 183

similis 185

thuiarioides 186

tricuspidata 183

variabilis 185

verticillata 171
Sertularidae 182
Seven new mammals from Mexico, by

C. Hart Herriam 559
Somatochlora albicincta 220
Soranthera ulvoides 422
Sphacelariaceae 419
Sphacelaria cirrosa 419

racemosa arctica 41
Sphaerella lacustris 409
Sphaerophoria splendens 374
Sphaerozosma excavatum spinulosum 408
Sphingonotus fusco-irroratus 495
Spirogyra porticalis 409

varians 409
Squamariaceae 441
Starks, E. C, synonomy of the fish

skeleton 507
Staurastrum arctiscon 407

calyxoides 406

dejectum 406

dejectum mucronatum 406

echinatum 407

furcigerum 407

polymorphum 407

pygmaeum 406

ravenelii 406
Staureoneis phcenicenteron 446
Stenocephalus insularis 365
Stigonema minutum 399

ocellatum 399
Stigonemataceae 399
Stratiomyidae 373
Streblonema irregularis 417

minutissima 416

pacifica 417
Styela greeleyi 241

monterej-ensis 241

truncata 241

yakutatensis 239
Sumichrast's chlorospingus 150
Surirella elegans 446
Sus cristatus 117

natunensis 117
Syncoryne 165

eximia 166
Synedra crotonensis 444

fasciculata 444

prolongata 444

ulna 444
Synonymy of fish skeleton, by E. C.
Starks 507

alphabetical list 527

bibliography 538
Synthesiomyia brasiliana 375
Syrphidae 374
Syrphus clavatus 374

duplicatus 374

6l2

INDEX

Syrphus (Continued)
spiendens 2,7 A

Tabanidae 373
Tabanus vittiger 373
Tabellaria flocculosa 444
Tceniosoma 61

(Eqiiale 4

princeps 62

quinquelineatum 4

septenilineatum 4
Tanager Cozumel 149

Escondido 150

McLeannan's 150

Yucatan 150
Tanagridae, new species 149
Tanaidae 565
Tanais Stanford! 565
Temnopteryx snodgrassii 493
Termitidae 544
Termitidae and Thysanura, by N. Banks

541
Tetmemorus brebissonii 402

laevis 402
Tetrastemma 57

aberrans 58

bicolor 57

caecum 59

cruciatum 58

rubrum 58

vermiculus 58
Thalassosira cleve 443
Thompson, Miss C. B. 20
Thuiaria 184

argentea 184

coei 185

costata 187

cupressoides 185

elegans 187

fabricii 185

gigantea 186

similis 185

thniarioides 186

turgida 186

variabilis 185
Thyanta perditor 365
Thysanura and Tennitidae, by N. Banks

541
Tipulidae 373
Tolypothrix tenuis 398
Tragulus sp. 115

bunguranensis 113

javanicus 115, 116

pallidus 116
Tramea darwini 386
Trentepohlia iolithus 413
Triceratium wilkesii 443
Trivia pediculus 140
Trochiscia moniliformis 443
Tubularia 168

harTimanl 168
Tubularidae 168
Tupaia lucida 133

sirhassenensis 133

splendidula 133

Tupaia (Continued).

splendidula lucida 133

splendidula lypica 133

tana 133
Turbinella ovoidea 140
Tumerella mertensiana 435
Tylomys bullaris 561
Tumbalensis 560

Ulothrichaceae 412
Ulothrix flacca 412

parietina 412
Ulvaceae 409
Ulva lactuca myriotrema 410

lactuca rigida 410

priapus 431
Urospora penicilliformis 413

Vanheurckia rhomboides 446
Vaucheriaceae 415
Vaucheria sessilis 415
Viverra tangalunga 133
Voluta hebraea 140

Wanner, A., new species of olenellus 267
Willow, arctic 326

barclay 316

bebb 306

bog 329

chamisso 325

diamondleaf 319

feltleaf 311

grayleaf 321

greenbacked 317

icy 329

least 338

little-tree 323

mouseleaf 322

myrtle 325

netleaf 340

nuttall 310

oval 331

polar 335

richardson 315

satin 307

skeleton 336

sprouting 333

yakutat 314
Willows of Alaska, by F. V. Coville 297

distribution 299

groups, 298

key to 300, 302

Xanthidium antilopaeum 403

armatum 403
Xanthopsar 155
Xiphidium exitiosum 501

Zarhynchus wagleri mexicanus 151
Zygnema 409
Zygnemaceae 409
Zygonemertes 28

albida 31

thalassina 29

virescens 29, 30