j!
CAS3K

JUH MARCH 14, 2003 * VOLUME 54 * NUMBERS 1-8

©

PROCEEDINGS

OF THE
CALIFORNIA ACADEMY OF SCIENCES

®ee*
wer CALIFORNIA
& ACADEMY
ANNIVERSARY Of SCIENCES
CELEBRATION 5 53

SAN FRANCISCO, CALIFORNIA

Copyright © 2003 by the CalifommigAcademy of Sciences
All rights reserved. No aon Apablis: tia, Rey be reproduced or transmitted in any form or by any
means, electronic or nights at ne phot scopying, recording, or any information storage or retrieval
system, without pefission in eae ae the ee

f { Sint

ch
Ne

AY
Oe Petts pave’ page

zs aides tallcte? =

SCIENTIFIC PUBLICATIONS

Alan E. Leviton, Ph.D., Editor
Gary C. Williams, Ph.D., Associate Editor

COVER IMAGE: The “Crew” of the Academy’s Galapagos Expedition Schooner Academy,
taken just before departure from San Francisco, 28 June 1905. (Standing, left to right): Frederick
T. Nelson, mate; Alban Stewart, botanist; Ernest S. King, assistant herpetologist; Rollo H. Beck,
expedition leader and ornithologist; Joseph S. Hunter, ornithologist; Joseph R. Slevin, herpetol-
ogist; Edward W. Gifford, ornithologist; Washington H. Ochsner, geologist and conchologist;
Francis X. Williams, entomologist. (Sitting): Captain J.J. Parker, navigator. California Academy
of Sciences Archives.

ISSN 0068-574X

Published by the California Academy of Sciences
Golden Gate Park, San Francisco, California 94118 U.S.A.

Printed in the United States of America by
Allen Press Inc., Lawrence, Kansas 66044

A New Calotes (Squamata: Nene, ae the

Indo-Burman Range of Western Myanmar (Burma)

Jens V. Vindum!, Htun Win’, Thin Thin”, Kyi Soe Lwin’, Awan Khwi Shein’, Hla Tun?
‘Department of Herpetology, California Academy of Sciences, Golden Gate Park, San Francisco, California

5
94118; email: jvindum@calacademy.org. “Nature and Wildlife Conservation Division, Forest Department,
Ministry of Forestry, Bayintnaung Road, West Gyogone, Insein, Yangon, Myanmar

A new species of Calotes is described from the Indo-Burman Range, western
Myanmar. It was found between elevations of 737 m to 1940 m in areas of shifting
cultivation and secondary forest. It differs from other species of Calotes from
Myanmar by a combination of its large size, SVL up to 142.9 mm, relatively small
body scales, 59-74 midbody scale rows, the tail in adult males being swollen posteri-
or to base and the presence of head spines. The new species increases the diversity of
Calotes species in Myanmar to six. A key to the Calotes of Myanmar is provided.

The genus Calotes Cuvier, 1817 currently contains 20 species (Hallermann 2000), distributed
from eastern Iran east to Sumatra, Indonesia (Welch 1994; Manthey and Grossmann 1997) includ-
ing two isolated species, one in Sabah, Malaysia (C. kinabaluensis de Grijs, 1937) and the other on
Ambon Island, Moluccas Islands, Indonesia (C. nigriplicatus Hallermann, 2000). Five species are
known to occur within Myanmar: Calotes emma, C. jerdoni, C. kingdonwardi, C. mystaceus and
C. versicolor (Smith 1935; Welch 1994). Calotes mystaceus and C. versicolor are widespread.
Calotes emma although it may occur throughout much of Myanmar, to date the Myanmar
Herpetological Survey team has not found it west of the Ayeyarwady River. However, it is known
to occur in Bangladesh (CAS 94323) and northeastern India (Smith 1935; Pawar and Birand 2001).
Calotes jerdoni is known from the Chin Hills of the Indo-Burman Range and along the southern
flanks of the eastern Himalayas. Calotes kingdonwardi is restricted to the eastern Himalayas
(Smith 1935).

MATERIALS AND METHODS

All specimens were hand collected, euthanized, fixed in 10% buffered formalin and later trans-
ferred to 70% ethanol. Latitude, longitude and elevation were recorded using a Garmin 12 GPS
receiver (datum WGS 84). Museum symbolic codes follow Leviton et al. (1985).

Data for the following meristic characters and measurements were recorded for adult speci-
mens: supralabials (SupL), number of enlarged scales bordering left and right margin of upper lip
(not including rostral scale); infralabials (InfL), number of enlarged scales bordering left and right
margin of lower lip (not including mental scale); number of scales around midbody (MidB); num-
ber of nuchal and dorsal crest spines (DC), enlarged mid-dorsal crest scales from posterior portion
of the head to just above anterior margin of vent; number of subdigital lamellae on the right fourth
toe (SDL 4" toe), including the last scale on claw base, snout-vent length (SVL); tail length (TailL):

D) PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 1

head length (HeadL), distance from tip of snout to rear border of right angle of jaw; head width
(HeadW), widest point in the temporal region, anterior to the tympanum; length of third and fourth
toes on right foot (3™ Toe and 4" Toe, respectively); width of tail at widest point of swelling in
males and the homologous area in females (TailD) and ratios of TailL/SVL, HeadW/HeadL,
HeadL/SVL, HeadW/SVL.

Scale counts and observations of external morphology were made using a dissecting micro-
scope. Measurements, except for TailL, were taken with digital calipers and rounded to the nearest
0.1 mm, TailL were measured using a measuring tape with a precision of 1 mm.

Measurements and meristic characters for males and females were compared using the two-
tailed independent samples Student’s t-test. SPSS (version 10.0 for Macintosh) was used for all sta-
tistical analysis.

SPECIES DESCRIPTION

Calotes chincollium Vindum, sp. nov.
Figs. 1-5

?Calotes emma, Shreve 1940, Proc. New England Zool. Club 18:24.

DIAGNOSIS AND COMPARISONS.— A species of Calotes morphologically similar to C. mys-
taceus, head and body robust, slightly compressed dorso-ventrally, snout-vent length to 142.9 mm;
body scales relatively small, homogeneous, feebly keeled, arranged in regular rows, upper dorso-
lateral scales pointing backwards and upwards and lower flank scales pointing backwards, 59-74
midbody scale rows; dorsal and dorso-lateral scales nearly equal in size to ventrals. Head shape tri-
angular, forehead concave; one enlarged temporal spine on either side of the upper head between
the occiput, orbit and tympanum. Nine to 12 supralabial scales and eight to 11 infralabials. Gular
scales feebly keeled, males with large gular pouch and cheek pouches. Nuchal crest composed of
erect compressed scales, partly overlapping, lanceolate, slightly falcate, directed posteriorly; dor-
sal crest follows the nuchal crest without a gap, 42-54 nuchal and dorsal crest spines. Distinct
oblique fold of skin covered with small granular dark brown scales in front of shoulder. Limbs
moderate, fourth toe longer than third, fourth right toe with 23—28 subdigital lamellae. Tail in adult
males swollen posterior to base.

Calotes chincollium can be differentiated from all other Calotes by a combination of the fol-
lowing characters: its large size at maturity, relatively small body scales, the presence of an oblique
fold in front of shoulder, the presence of head spines and the males having a swollen tail posterior
to base. C. chincollium differs from other Myanmar and northeast India congeners as follows: from
the males of all other species, except C. kingdonwardi, by males having a swollen tail base; from
C. maria and C. jerdoni by lacking two parallel rows of compressed scales above tympanum and
oy its larger size (SVL to 120 mm in C. maria [Smith 1935]) and to 100 mm in C. jerdoni (adult
CS specimens); from C. mystaceus by the larger number of midbody scale rows (C. mystaceus
havi.» 47-57 [CAS specimens]; 48-56 [Hallermann 2000]) and by the gradual reduction in size of
the cres: scales from the nuchal crest scales to the dorsal crest scales (crest scales in C. mystaceus
are sho. ~ in the nape area, between the nuchal and dorsal crests); from C. versicolor by having
an oblique —!d in front of shoulder (absent in C. versicolor) and larger number of midbody scale
rows (C. ve. »icolor has 40-50 [Smith 1935]); from C. emma by lacking postorbital spines and by
having small su-:2ocular scales (C. emma has large rectangular supraoculars); and from C. king-
donwardi by hay = more midbody scale rows (45 in C. kingdonwardi (Smith 1935) and KIZ spec-
imens), and having ‘es on the side of the body pointing upwards and backwards and the lower

VINDUM ET AL.: A NEW CALOTES FROM WESTERN MYANMAR 3

flank scales pointing backwards (scales in C. kingdonwardi pointing backwards and downwards
except for upper two to three rows, which point slightly upwards (Smith 1935) or straight back-
wards [KIZ specimens]) and the presence of head spines and a dorsal crest which are lacking and
greatly reduced, respectively, in C. kingdonwardi (Smith 1935 and KIZ specimens).

HoLotyPEe.— CAS 220009 (Figs. 1-3), adult 3, from 21°23’11.2” N, 93°58’15.9” E, 1174 m
elevation, Min Dat Township, Min Dat District, Chin State, Myanmar, collected 20 March 2002 by
Htun Win, Kyi Soe Lwin and Awan Khwi Shein.

DESCRIPTION OF HOLOTYPE.— Adult 3 with a SVL of 142.9 mm; TailL 253 mm (incomplete
tail); HeadL 48.6 mm; HeadW 30.6 mm; and 68 scales around midbody and 46 nuchal and dorsal
crest scales.

Canthus rostralis sharp and straight; upper head scales unequal, smooth; rostral scale low,
same height as touching supralabials, about twice as wide as high, bordered behind by two supral-
abials and four postrostrals; ten supralabials on both sides; a series of four scales run along the mid-
line posterior to the postrostrals, the third and fourth largest and slightly elevated; posterior to the
fourth scale, are two scales on either side directed diagonally towards the orbit; the posterior six
scales form a weak inverted “Y’-shaped pattern; the inner border of the supraocular region with a
semicircular series of enlarged, feebly keeled scales; at closest point three head scales separate the
left and right series; scales within the semicircular series smaller and feebly keeled; a minute tuber-
cle at posterior end of supraciliary ridge; one enlarged spine on either side of the upper head
between the occiput, orbit and tympanum; tympanum exposed, 6.8 mm horizontal diameter, about
2/3 the size of the orbit (10.6 mm horizontal diameter); a transverse series of four enlarged scales
form an elevated ridge between the posterior margin of the orbit and the anterior margin of the tym-
panum, middle two scales being the largest and posterior scale most strongly keeled; a transverse
series of four scales above the tympanum, anterior scale keeled, second and fourth pointed and the
third is an enlarged spine.

Mental triangular becoming narrow posteriorly, as wide as long, slightly wider than rostral;
mental followed by an infralabial on either side and two irregular shaped postmentals; postmentals
higher than long, in contact with the first infralabials and the length of the mental except for the
posterior tip of the mental where the postmentals are separated by two small scales medially; pos-
terior to the postmentals are two chin shields on each side that are longer than wide and parallel to
the infralabials, separated from the infralabials by one scale row. The first and second chin shields
are scarcely larger than the chin shields following; ten infralabials on both sides; large gular pouch,
gular scales strongly imbricate; lateral gular scales form a gradation from almost rounded scales,
each weakly mucronate, most smooth or with a slight rise medially to more lanceolate and
mucronate medial scales, outer scales wider than long, medial scales longer than wide; lateral gular
scales pointing posteriorly and medially, medial scales pointing straight back; most lateral gular
scales with one hair receptor at tip or in notch next to tip; cheek pouches large, extending beyond
the plane of the temporal region (horizontally from the base of the tympanum) by 6.1 mm, and
beyond the tympanum by 9.5 mm on either side.

Large oblique curved fold in front of shoulder, covered with small granular scales; seven scale
rows separate the dorsal border of fold from the base of upright scale series adjacent to the nuchal
crest; nuchal crest composed of erect compressed scales, partly overlapping, lanceolate, slightly
falcate, directed posteriorly, with the 7'+ to 10 scales being the largest, and a continuous reduction
in size anteriorly and posteriorly; dorsal crest follows the nuchal crest without a gap, the gradation
of decreasing scale length is continuous posteriorly and does not change from nuchal crest scales
to dorsal crest scales; longest nuchal scale extends 15.3 mm from the base of the scale bordering
the crest scale; dorsal crest scales do not overlap and become more triangular and gradually lower

4 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 1

es
het So Sees
RN LE MIKDS net
i log a
ae
f Wee
TA
| me $ !
: A ERY

FiGures 1—3. (1) Dorsal view of the holotype of Calotes chincollium, sp. nov. (CAS 220009), male, SVL 142.9 mm,
from 21°23’11.2” N, 93°58’15.9” E: 1174 m elevation, Min Dat Township, Min Dat District, Chin State, Myanmar; (2)
Lateral view of the holotype of Calotes chincollium, sp. nov.; (3) Ventral view of anterior portion of the holotype of Calotes
chincollium, sp. noy. Photographs by Dong Lin, California Academy of Sciences.

VINDUM ET AL.: A NEW CALOTES FROM WESTERN MYANMAR 5)

posteriorly; 46 mid-dorsal crest scales from the occiput to above the anterior margin of the vent;
scales are reduced to a low crest over the sacrum, forming a keeled ridge along the anterior portion
of the tail. The nuchal and dorsal crests are bordered on either side by a row of lower erect (point-
ing dorsally and posteriorly), keeled, mucronate, scales, becoming smaller posteriorly, ending at
the sacrum; dorsal and lateral body scales feebly keeled, imbricate, dorsal scales being slightly
larger, pointing slightly dorso-posteriorly, lower scales on flanks pointing backwards; dorsal and
lateral scales equal in size with ventral scales, the slightly larger dorsal-most scales on sides are
about equal to the ventral scales in the chest area, and the lower lateral scales are equal to ventral
scales on the abdomen; ventral scales keeled, imbricate, weakly mucronate, pointing posteriorly,
20-22 rows; left hemipenis partially everted.

Forelimbs covered with imbricate, feebly keeled scales, some slightly mucronate; third and
fourth, second and fifth, fingers equal in length, first finger smallest; scales under fingers bicari-
nate, right fourth finger with 23 lamellae (including last scale on claw base); dorsal surface of thigh
and dorsal and ventral surfaces of lower leg with imbricate, feebly keeled scales, some slightly
mucronate; ventral surface of thigh with smooth rectangular, slightly imbricate scales; scales on
posterior portion of thigh imbricate, keeled and mucronate; fourth toe longest (about 1/3 longer
than third) followed by third, fifth, second and first; scales under toes bicarinate, right fourth toe
with 25 lamellae (including last scale on claw base).

Tail slightly oval in cross-section, swollen posterior to base, with large, strongly keeled,
mucronate, scales, arranged in longitudinal rows; tail tip autonomized. Scales of tail larger than
dorsal or ventral body scales.

COLORATION IN ALCOHOL.— Upper head grayish, darker on snout; supraciliaries with two par-
allel light brown bars perpendicular to head axis; irregular brown patch from anterior border of
orbit to dorso-anterior border of tympanum, extending dorsally to the supraciliary ridge, slightly
bifurcating posteriorly, lower branch ending at the tympanum and the upper extending to the nuchal
crest at occiput. Light gray on upper lip extending dorsally to orbit and lower temporal area (below
dorsal borders of tympanum), tympanum and cheek pouches. Body grayish with four brownish
irregular saddles extending across the back. Anterior-most saddle weak, directly behind occiput,
most scales gray with brown at base; second and third saddles brown from base of dorsal crest
scales to fourth scale row below dorsal crest (counting crest scale as first). Fourth saddle extends
down six scale rows, dorsal portion of saddles the widest, each saddle separated by one or two
scales. Tail uniform light olive-brown with dark and light barring toward the tip, darker bars about
four transverse scale rows wide, lighter bars two to three scale rows. Dorsal surface of forelimbs
grayish with irregular brown bars, gray bars one scale row wide, brown bars two scale rows wide;
barring extends on to foot (barring is more prominent on left forelimb). Upper surface of thigh
grayish at base with increasing brown proximally, lower leg and foot with gray and brown barring,
with brown bars being wider. Gular pinkish from scales posterior to mental and postmentals and
extending to ventral portion of cheeks. Inner scales of oblique fold in front of shoulder black.
Venter and ventral surface of limbs gray. Ventral side of tail light brownish-gray.

VARIATION.— Body measurements and meristic characters are shown in Tables | and 2,
respectively, for specimens with a SVL more than 96 mm.

Specimens ranged in size from the smallest juvenile with a SVL of 33.8 mm (CAS 222349) to
the largest male SVL 142.9 mm (CAS 220009); and the largest female SVL 123.5 mm (CAS
220120).

Adult males are significantly larger than adult females, having longer snout-vent lengths, tail
lengths, head lengths, head widths, and fewer dorsal crest scales (Table 3). The heads of adult
males are absolutely and proportionally longer and wider than females of the same snout-vent

6 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 1

TABLE |. Sex and measurements (in mm) for Calotes chincollium
(- indicates incomplete tail and * indicates measurements from left foot)

Catalog SEX SVL TailL TailL/ HeadL HeadW HeadW/ Headl/ HeadW/ 3rd Toe 4th Toe TailD
Number SVL HeadL SVL SVL

CAS 219972 M 133.3 247 1.85 44.6 DAES 0.617 0.335 0.206 14.3 18.7 7/3)
CAS 219973 M 114.3 235 2.06 37.7 22.5 0.597 0.330 0.197 14.0 19.6 13.8
CAS 219974 M 119.0 198- 38.2 22.8 0.596 0.321 0.192 13.4 20.1 15.0
CAS 219976 M 134.7 271 2.01 47.4 28.6 0.603 0.352 0.212 14.0* PAN OF 18.9
CAS 219977 M 120.2 242 2.01 41.2 25.8 0.626 0.343 0.215 14.3 18.5 IES
CAS 219978 M 112.0 235 2.10 37.5 21.0 0.560 0.335 0.188 14.3 19.6 13.8
CAS 219990 M 117.5 238 2.03 41.0 25.0 0.610 0.349 0.213 13.8 20.7 16.3
CAS 219996 M 120.2 267 2.22 40.3 25.6 0.635 0.335 0.213 13.6 20.5 16.1
CAS 219997 M 101.6 214 2.11 33.6 21.0 0.625 0.331 0.207 14.2 19.2 37
CAS 220009 M 142.9 253- 48.6 30.6 0.630 0.340 0.214 17.0 25.0 19.2
CAS 220012 M 134.1 269 2.01 45.4 28.3 0.623 0.339 0.211 IBD e 19.8% Wes)
CAS 220027 M 127.2 269 Pel 44.3 26.3 0.594 0.348 0.207 15.3 dep) 7/3)
CAS 220028 M 101.3 201- 33.0 LOFT 0.597 0.356 0.194 14.3 19.8 13.3
CAS 220029 M 120.5 288 2.39 44.1 27.2 0.617 0.366 0.226 14.9 22D. ES)
CAS 220034 M 100.1 211 2.11 33.1 19.9 0.601 0.331 0.199 12.8 17.6 IDES
CAS 220035 M 115.5 238 2.06 38.5 22.9 0.595 0.333 0.198 132 17.8 16.3
CAS 220039 M 136.3 293 Dali 49.4 28.8 0.583 0.362 0.211 16.9 23.0 19.8
CAS 220046 M 130.8 281 Das 42.7 26.3 0.616 0.326 0.201 16.7 24.0 17.2
CAS 220049 M 139.2 306 2.20 49.0 29.4 0.600 0.352 0.211 16.4 22.8 18.1
CAS 220117 M 132.6 287 2.16 48.5 29.8 0.614 0.366 0.225 S7/ 21.5 18.8
CAS 220121 M 130.2 290 2.23 46.6 29.6 0.635 0.358 0.227 16.3 23.0 20.1
CAS 220125 M 137.9 297 DAS 45.9 29.8 0.649 0.333 0.216 14.6 20.1 18.2
CAS 220577 M 114.2 265 232 37.7 22.2 0.589 0.330 0.194 16.7 LOM 1S
CAS 220580 M 131.3 266 2.03 46.5 30.8 0.662 0.354 0.235 14.5 19.9 18.6
CAS 220581 M 108.2 244 2.26 37.4 21.4 0.572 0.346 0.198 14.7 20.6 14.1
CAS 220582 M 118.5 248 2.09 38.5 23.7 0.616 0.325 0.2 13.6 18.7 15.0
CAS 220583 M 125.5 265 2.11 44.5 27.4 0.616 0.355 0.218 15.6 223 17.9
CAS 222354 M 104.0 216 2.08 34.9 20.7 0.593 0.336 0.199 13:3 17.9 13.6
CAS 222370 M 126.4 261 2.06 43.3 26.6 0.614 0.343 0.21 16.2 PINT) 15.4
MCZ 44727 M 113.4 252 2.22 38.9 22.7 0.584 0.343 0.2 15.0 20.6 14.7
MCZ 44728 M 114.7 255 2.22 40.6 22.2 0.547 0.354 0.194 14.7 INT 14.8
MCZ 44729 M 104.2 222 2.13 35.5 DNS 0.600 0.341 0.204 14.8 18.5 15.4
USNM 547926 M 117.6 232 1.97 40.4 24.3 0.601 0.344 0.207 16.0 DED, 15.4
USNM 547927 M 137.1 299 2.18 49.5 30.8 0.622 0.361 0.225 16.7 24.3 19.8
CAS 219971 F 96.8 206 218 30.8 19.0 0.617 0.318 0.196 12.5 16.8 8.8
CAS 219975 I@ 119.5 234 1.96 35.9 Ppay2 0.618 0.300 0.186 1355 18.6 11.9
CAS 220014 BE 113.7 251 2.21 36.1 22.0 0.609 0.318 0.193 ley 20.4 11.6
CAS 220022 F 117.8 220 1.87 35.6 21.6 0.607 0.302 0.183 16.9 24.1 11.0
CAS 220120 F 123.5 251 2.03 38.9 20.9 0.537 0.315 0.169 14.3 19.9 11.0
CAS 220578 B 97.7 204 2.09 30.8 18.4 0.547 0.315 0.188 13.1 18.5 9.6
CAS 220579 F 99.4 224 2.25 3255 18.7 0.597 0.327 0.188 13.2 19.0 YS)
CAS 220584 F 114.7 229 2.00 38.6 22.0 0.570 0.337 0.192 14.0 PAVE) 11.8
CAS 220587 E 108.2 221 2.04 34.1 20.0 0.587 0.315 0.185 14.2 18.2 11.0
CAS 222351 E Lt 230 2.26 32.9 20.2 0.614 0.324 0.199 13.3 19.7 Wee.

VINDUM ET AL.: A NEW CALOTES FROM WESTERN MYANMAR

TABLE 2. Meristic characters for Calotes chincollium
(* indicates scale counts from left foot)

Catalog SEXGae idee DC: SupL InfL SDL 4th Toe

Number B (L/R) (L/R)

CAS 219972 M 68 47 11/11 10/10 25
CAS 219973 M 63 46 11/10 10/11 26
CAS 219974 M 59 45 12/11 10/10 25
CAS 219976 M 61 45 9/9 8/11 26*
CAS 219977 M 70 45 11/10 9/9 23
CAS 219978 M 65 45 10/11 9/9 23
CAS 219990 M 65 46 11/10 9/9 26
CAS 219996 M 66 45 11/10 10/9 26
CAS 219997 M 61 44 10/10 9/9 25
CAS 220009 M 68 46 10/10 10/10 25
CAS 220012 M 67 48 10/9 10/10 25
CAS 220027 M 69 45 9/10, 10/10 25
CAS 220028 M 63 46 11/10 10/10 27
CAS 220029 M 74 54 12/11 11/11 26
CAS 220034 M 62 45 10/11 9/10 23
CAS 220035 M 66 45 11/10 9/9 26
CAS 220039 M 68 45 11/11 11/9 25
CAS 220046 M 71 45 11/10 10/10 27
CAS 220049 M 69 42 10/10 10/10 26
CAS 220117 M 70 45 10/10 10/11 26
CAS 220121 M 69 45 9/10 9/10 26
CAS 220125 M 67 45 9/11 9/10 25
CAS 220577 M 62 42 11/11 10/10 25
CAS 220580 M 66 48 10/9 10/9 25
CAS 220581 M 62 44 10/9 9/10 27
CAS 220582 M 69 47 10/10 10/11 25
CAS 220583 M 68 46 10/10 9/10 25
CAS 222354 M 70 46 11/11 11/11 26
CAS 222370 M 68 46 10/11 10/9 24
MCZ 44727 M 68 44 10/10 9/9 26
MCZ 44728 M 71 46 10/10 9/10 25
MCZ 44729 M 65 49 9/9 9/9 26
USNM 547926 M 70 47 10/11 11/9 25
USNM 547927 M 67 45 11/10 10/11 28
CAS 219971 F 65 45 11/11 10/10 24
CAS 219975 FE 67 47 10/11 9/9 25
CAS 220014 FE 68 47 10/11 9/9 26
CAS 220022 F 65 47 9/9 9/9 28
CAS 220120 FE 70 48 10/10 10/10 23
CAS 220578 FE 66 50 10/10 9/9 28
CAS 220579 se 70 47 11/11 10/10 26
CAS 220584 F 71 49 10/11 8/10 24
CAS 220587 F 65 46 10/10 11/10 23
CAS 222351 F 70 49 10/10 9/9 23

8 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 1

length. Additional sexual characteristics include males having a swollen tail posterior to its base
(Table 3) and larger cheek pouches. The dorsal coloration differs between the sexes with the dark
saddles in males being lighter or completely faded posteriorly.

COLOR IN LIFE (Fig. 4; based on color transparency of CAS 222354 [3 ]): Upper head gray,
side of head with distinct dark brown mask from loreal area through orbit and temporal region;

TABLE 3. Body measurements (mm) and meristic characters for male and female
Calotes chincollium (X + SD, followed by ranges in parentheses, N=34 for males (except
for TailL where N=31) and N=10 for females). Statistically significant results at
0.05 => P> 0.01 are marked with one asterick (*), results at 0.01 => P> 0.001 are marked
with two astericks (**) and results at P < 0.001 are marked with three astericks (GA),

Males Females
SVIEx* 121.7 + 12.1 (100.1—142.9) 109.3 + 9.9 (96.8-123.5)
Pallas 258.2 + 26.6 (211-306) 227.0 + 15.9 (204—251)
HeadL*** 41.7 + 5.0 (33.049.5) 34.6 + 2.9 (30.8-38.9)
Head W*** 25.4 + 3.5 (19.7-30.8) 20.5 + 1.5 (18.4-22.2)
3rd Toe 14.9 + 1.2 (12.8-17.0) 14.0 + 1.3 (12.5-16.9)
4th Toe 20.7 + 1.9 (17.6—25.0) 19.7 + 2.1 (16.8—24.1)
TailD*** 16.4 + 2.2 (12.5—20.1) 10.8 £1.0 (8.8-11.9)
MidB 66.7 + 3.5 (59-74) 67.7 + 2.4 (65-71)
DC* 45.7 2.1 (42-54) 47.5 + 1.5 (45-50)
SDL 4th Toe 25.4 + 1.1 (23-28) 25.0 + 1.9 (23-28)
TailL/SVL 2.12 + 0.108 (1.85—2.39) 2.08 + 0.129 (1.87—2.26)
HeadL/SVL*** 0.343 + 0.0122 (0.321—0.366) 0.317 + 0.0109 (0.300-0.337)

HeadW/SVL*** 0.208 + 0.0114 (0.188—0.235) 0.188 + 0.00833 (0.169-0.199)

white along the upper lip to the orbit and upper margin of the tympanum and along lower jaw
extending onto cheek pouch; nuchal crest bright yellow; gular orange; anterior half of body creamy
yellow with flanks spotted with darker yellow, first two dorsal saddles brown; posterior portion of
body with light brown and gray mottling; forelimbs banded dark brown and white.

COLORATION IN ALCOHOL (Fig. 5; based on CAS 220022 [ 2 ]): Upper head speckled tan, gray
and cream; supraciliaries with two parallel brown bars perpendicular to head; brown patch from
anterior border of orbit to dorso-anterior border of tympanum extending dorsally to spines above
tympanum; lighter brown (but darker than upper head coloration) between spines over tympanum
and occiput spines extending from the posterior supraciliary ridge to the back of the head. Light
gray on upper lip extending dorsally to orbit and extending to the lower temporal area including
the tympanum. Dorsal coloration cream becoming gray on flanks. Brown triangular patch extend-
ing along the anterior margin of the fold in front of the shoulder to the posterior side of the angle
of the jaw, extending horizontally between the back of the head (at the level of the upper margin
of the tympanum) back to the shoulder fold. From nape to area dorsal to the anterior margin of the
vent are seven dark brown irregular bands or saddles, with the coloration extending onto the nuchal
and dorsal crest scales. The first four saddles are the most distinct. The anterior most saddle on the
nape extends from the nuchal crest onto the back of the head, forming a V-shaped pattern; poste-
rior to the nape the bands are widest along the dorsal crest becoming constricted on the lower por-
tion of the dorsum and then becoming less distinct, irregular and bifurcated on the flank. The sec-

VINDUM ET AL.: A NEW CALOTES FROM WESTERN MYANMAR 9

FIGURE 4. Paratype of Calotes chincollium, sp. nov. (CAS
222354) a male, SVL 104.0 mm, from 21°11°53.3” N, 94°04’00.5”
E, Kanpetlet Township, Min Dat District, Chin State, Myanmar.
Photograph by Hla Tun, Forest Department, Ministry of Forestry,
Myanmar.

ear 2
FIGURE 5. Paratype of Calotes chincollium, sp. nov. (CAS 220022) a female, SVL 117.8 mm, from 21°22’11.5” N,
93°46'01.7” E, 1,732 m elevation, Min Dat Township, Min Dat District, Chin State, Myanmar. Photograph by Dong Lin,
California Academy of Sciences.

ond and third are separated from the brown flank pattern by one scale row giving a faint hint of a
lateral stripe. When viewed from above the dorsal bands appear triangular in shape, with the tip of
the triangle pointing backwards. The tail is banded throughout with light and dark brown banding,
the darker bands being wider. Forelimbs and hindlimbs barred like in holotype. Gular scales pink-
ish on anterior two-thirds and along the sides; medial posterior third with scales having small dark
gray spotting giving an overall grayish appearance. Venter including upper forearms and legs with
light gray and dark gray mottling, more so on abdomen. Lower arms and legs with uniform light
gray. Underside of tail light and dark gray banding corresponding to the dorsal bands.

CAS 219975, 2, SVL 119.5 mm, tail length 234 mm, is similar in pattern to CAS 220022 but
darker and mottled throughout. Only nape and following two saddles distinct.

10 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 1

COLORATION IN ALCOHOL OF SMALLEST JUVENILE (CAS 222349, SVL 33.8 mm): Upper head

gray-brown; supraciliaries with two faint parallel brown bars perpendicular to head; orbit with nine
dark brown stripes radiating outwards from eye; brown patch from anterior border of orbit to dorso-
anterior border of tympanum extending dorsally to spines above tympanum; lighter brown (but
darker than upper head coloration) between spines over tympanum and occiput spines extending
from the posterior supraciliary ridge to the back of the head. Light gray on upper lip extending dor-
sally to orbit and extending to the lower temporal area including tympanum. Dorsal coloration gray.
Faint dark brown horizontal stripe extending from the dorsal margin of the fold in front of the
shoulder to the back of the head. From nape to area dorsal to the anterior margin of the vent are
five brown irregular saddles. The anterior most saddle on the nape extends from the nuchal crest
laterally two scale rows. When viewed from above the dorsal bands appear triangular in shape, with
the tip of the triangle pointing backwards. The tail is banded throughout with light and brown band-
ing, the lighter bands being wider. Forelimbs and hindlimbs bared. Venter including upper forearms
and legs light gray. Lower arms and legs uniform light gray.
r ETYMOLOGY.— The specific epithet refers
to the type locality being in the Chin Hills; chin-
aed 3 collium is a combination of the Anglicized ver-
tes oe sion of a Burmese inscription word for the Chin
people and collium, Latin in the genitive plural
of collis, meaning hill.

DISTRIBUTION AND NATURAL HISTORY.—
Calotes chincollium is known from the Chin
Hills, Chin State, and from the Ponnyadaung
Range, Sagaing Division, Myanmar (Fig. 6),
between elevations of 737 m to 1940 m. The
Species is most common in areas of shifting cul-
tivation and extends into secondary forest. The
species is not found in primary forest. It is pri-
marily found on the ground but climbs onto the
base of trees. Htun Win and Awan Khwi Shein
observed individuals in recently planted corn
fields digging up newly sprouting corn and eat-
ing the endosperm tissue of the seedlings.

The holotype was found in sympatry with
+ = > tae Calotes jerdoni, C. mystaceus and C. versicolor.
Kuonerers SNS Other agamids in the vicinity of Calotes chincol-

FIGURE 6. Map of western Myanmar illustrating the — ]jz4 include Ptvctolaemus gularis, Draco mac-
distribution of Calotes chincollium. Map prepared by ; 3

Michelle S. Koo, California Academy of Sciences.

la

MYANMAR) [7

io

ulatus and Japalura planidorsata.

DISCUSSION

The new species superficially resembles the description given by Taylor (1963) for Calotes
emma alticristatus Schmidt 1925, except that C. emma alticristatus has enlarged, elongate scales
in the supraocular region; two enlarged chin shields behind each of the postmentals; five postros-
tral scales; a grayish chin and gular area with black interstitial skin; and tail base is not swollen
(Taylor did not give the sex of his specimen, and we have not been able to locate it).

The type locality of C. emma alticristatus is “Yunnanfu, Yunnan” (Schmidt 1925), and Taylor

VINDUM ET AL.: A NEW CALOTES FROM WESTERN MYANMAR i]

(1963) reported the distribution extending into Chiang Mai Province, Thailand. However, Wermuth
(1967), Welch (1994) and Cox et al. (1998) list the distribution of C. emma alticristatus as extend-
ing northwest into Assam, India. We have found no evidence (either specimens or records in the
literature) of C. emma alticristatus occurring west of Thailand (namely in Burma, India or
Bangladesh). Specimens of C. emma from Mon State (CAS 222213), Shan State (CAS 215260)
and Mandalay Division (CAS 216395), Myanmar, and Bangladesh (CAS 94323) all have enlarged
postorbital spines characteristic of C. emma emma.

Recent collections from the lowlands surrounding and the mountains of the Indo-Burman
Range, specifically the Rakhine Yoma and Chin Hills have yielded a number of new species:
Lycodon zawi (Slowinski et al. 2001) from the Rakhine Yoma, Rakhine State, Ponnyadaung Range,
Sagaing Division, and the Khasi Hills, India; Calotes chincollium from the Chin Hills and the
Ponnyadaung Range; Bufo sp. (Wogan et al. 2003) and Chirixalus sp. (Wilkinson et al. 2003) from
the western lowlands of the Rakhine Yoma; and at least three additional new frog species yet to be
described from the Rakhine Yoma (pers. commun. G.O.U. Wogan 2002). All new species seem to
be endemic to the Indo-Burman Range or limited to the Rakhine Yoma or Chin Hills within the
Indo-Burman Range, with the exception of Lycodon zawi, which extends into the Khasi Hills,
India.

The occurrence of new species being discovered in the Indo-Burman Range is not surprising
considering the geological history of the area and the paucity of collections. The Indo-Burman
Range (since the late Miocene) was formed by the subduction of the Indian plate causing the
obduction of an accretionary prism upon the Indian continental margin (Ni et al. 1989). The result-
ing mountain ranges consist of deep canyons, ridges, and high peaks, the latter reaching elevations
of 1,989 m in Rakhine State, 3109 m in Chin State, and 3,826 m in Sagaing Division. The vegeta-
tion types within the Indo-Burman Range includes lowland tropical rain forest along the Bay of
Bengal, subtropical lowland forest, subtropical mountain forest and temperate mountain forest at
higher elevations (Davis 1964). The formation of the Indo-Burman Range clearly impacted local
climate, especially rainfall, and vegetation. With the progressive increase in topographic relief,
occasioned by collision tectonics since late-Miocene, vicariant speciation, a consequence of the
disruption of gene flow among the closely allied but increasingly fragmented populations of rep-
tiles and amphibians, resulted in a high degree of local species endemism.

As noted above, relatively few collections have been made in the Indo-Burman Range. For
instance, other than incidental collections, the only serious collecting forays were made by F.E.W.
Venning, who collected in the Hakha area from 1908 to 1910 (Venning 1910a, 1910b), and Gerd
Heinrich, who collected in the Mount Victoria area of the Chin Hills in 1938 (Shreve 1940). Thus,
the work of the Myanmar Herpetological team, which will return to the Indo-Burman Range in
2003, can be expected to generate exciting new information about the diversity of amphibians and
reptiles in this pivitol region of southeast Asia.

KEY TO THE SPECIES OF THE GENUS CALOTES OF MYANMAR

1. Two parallel rows of compressed spines above tympanum present .............. C. jerdoni
Two parallel rows of compressed spines above tympanum absent....................5. 2
PARAS AMIS PIES ADSCNU Were Pn ert let ne oe cate ents Co TOR OTR C. kingdonwardi
EIAs PINES yPLCSELID renee en ee eres ess teen UP aire Rei DRT IR tO CE coee OT a RESO 3
PauOMiiE tron ot shouldemabsentwts sax yoh5 dell. sehen dul ni eseeene BeE ES C. versicolor

Bolin trouroisnouldenmpresent,: fate treks. site) han creates ck oticese Led tathastoeo en ~

i PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 1

Aaa Ane epOStOLDitaluspINexpPreSEM ts 5 shes ets ae cee serene ctor eer teeta C. emma
arseipostorbital’ spine absent’. 4). ed ER SU lets 5
5. 47-57 midbody scale rows, tail not swollen posterior to base in males ........ C. mystaceus
59-74 midbody scale rows, tail swollen posterior to base in males .......... C. chincollium

MATERIAL EXAMINED

HOLOTYPE:— CAS 220009 (3), from 21°23’11.2” N, 93°58’15.9” E, 1174 m elevation, Min
Dat Township, Min Dat District, Chin State, Myanmar, collected 20 March 2002 by Htun Win, Kyi
Soe Lwin and Awan Khwi Shein.

PARATYPES (43 specimens, all from Chin State, Myanmar and, except as noted, collected by
combinations of Htun Win, Thin Thin, Kyi Soe Lwin, Awan Khwi Shein and Hla Tun).— CAS
219971 (2), 21°23’26.6” N, 94°03’31.2” E, Htin Chaun Village, Min Dat Township, Min Dat
District, collected 18 March 2001; CAS 219972-—219973 (32), from 21°23’26.6” N, 94°03’31.2”
E, Htin Chaun Village, Min Dat Township, Min Dat District, collected 18 March 2001; CAS
219974 (3), CAS 219975 (2), from 21°22’20.1” N, 93°58’34.6” E, 1482 m elevation, Baw Khue
Plantation, Htin Chaun Village, Min Dat Township, Min Dat District, collected 18 March 2001;
CAS 219976 (3), from 21°22’52.7” N, 93°53’43.8” E, Baw Khue Plantation, Min Dat Township,
Min Dat District, collected 19 March 2001; CAS 219977 (3), CAS 219978 (2), from 21°22’20.1”
N, 93°58'34.6” E, 1482 m elevation, Baw Khue Plantation, Htin Chaun Village, Min Dat Township,
Min Dat District, collected 19 March 2001; CAS 219990 (3), from 21°24’08.2” N, 93°52’45.0” E,
1920 m elevation, Baw Khue Plantation, Min Dat Township, Min Dat District, collected 19 March
2001; CAS 219996 (3), from 21°22’15.5” N, 93°59’13.6” E, 1418 m elevation, Min Dat Township,
Min Dat District, collected 20 March 2001; CAS 219997 (3), from 21°23’20.9” N, 93°52’29.0” E,
1940 m elevation, Baw Khue Plantation, Min Dat Township, Min Dat District, collected 19 March
2001; CAS 220012, USNM 547926 (33), from 21°22’50.1” N, 93°58’20.6” E, 1297 m elevation,
Min Dat Township, Min Dat District, collected 20 March 2001; CAS 220014 (2), from
21°22’50.1” N, 93°58’20.6” E, 1297 m elevation, Min Dat Township, Min Dat District, collected
20 March 2001; CAS 220022 (¢), from 21°22’11.5” N, 93°46’01.7” E, 1732 m elevation, Min Dat
Township, Min Dat District, collected 22 March 2001; CAS 220027-—220028 (3c), from
21°22'18.3” N, 93°49’00.6” E, 1787 m elevation, Hee Laung Village, Min Dat Township, Min Dat
District, collected 24 March 2001; CAS 220029 (3), from 21°23’16.1” N, 93°58’14.9” E, 1138 m
elevation, Min Dat Township, Min Dat District, collected 20 March 2001; CAS 220034 (2), from
21°22'07.6” N, 93°49’04.0” E, 1624 m elevation, Hee Laung Village, Min Dat Township, Min Dat
District, collected 25 March 2001; CAS 220035 (¢), from 21°21’33.7” N, 93°49'13.6” E, Hee
Laung Village, Min Dat Township, Min Dat District, collected 25 March 2001; CAS 220039 (2),
from 21°20'13.3” N, 93°55’22.1” E, 1046 m elevation, Che stream, Min Dat Township, Min Dat
District, collected 29 March 2001; CAS 220046 (2), from 21°19’42.6” N, 93°55’25.2” E, Che
stream, Min Dat Township, Min Dat District, collected 29 March 2001; CAS 220049 (2), from
21°2114.9” N, 93°56’08.3” E, Che stream, Min Dat Township, Min Dat District, collected 29
March 2001; CAS 220117 (¢), from 21°22’14.2” N, 93°48'14.1% E, Che stream, Min Dat
Township, Min Dat District, collected 30 March 2001; CAS 220120 (2), CAS 220121, USNM
547927 (oc), from 21°20’53.8 N, 93°59’56.3 E, 1112 m elevation, Che stream, Min Dat
Township, Min Dat District, collected 31 March 2001; CAS 220125 (¢), from 21°21’02.3” N,
93°56'00.2” E, 783 m elevation, Che stream, Min Dat Township, Min Dat District, collected 2 April
2001; CAS 220577 (3), from 21°11’44.3” N, 94°04’53.3” E, Kanpetlet Township, Min Dat

VINDUM ET AL.: A NEW CALOTES FROM WESTERN MYANMAR i

District, collected 25 February 2001; CAS 220578 (2), from 21°11’23.0” N, 94°01’49.3” E,
Kanpetlet Township, Min Dat District, collected 26 February 2001; CAS 220579 (2), from
21°11’24.2” N, 94°04’11.2” E, Kanpetlet Township, Min Dat District, collected 27 February 2001;
CAS 220580 (3), from 21°13’19.7” N, 93°57’52.5” E, Kanpetlet Township, Min Dat District, col-
lected 11 March 2001; CAS 220581 (c), from 21°11°55.2” N, 94°03’57.8” E, Kanpetlet Township,
Min Dat District, collected 28 February 2001; CAS 220582 (), from 21°11’53.5” N, 94°04’00.5”
E, Kanpetlet Township, Min Dat District, collected 28 February 2001; CAS 220583 (3), CAS
220584 (2), from 21°11’37.1” N, 94°02’58.2” E; 1572 m elevation, Kanpetlet Township, Min Dat
District, collected 12 March 2001; CAS 220587 (2), from 21°23’25.4” N, 94°03’06.9” E, Htin
Chaun Village, Min Dat Township, Min Dat District, collected 17 March 2001; CAS 222351 (2),
from 21°11°31.9” N, 94°03’00.8” E, Kanpetlet Township, Min Dat District, collected 26 February
2001; CAS 222354 (3), from 21°11’53.3” N, 94°04’00.5” E, Kanpetlet Township, Min Dat
District, collected 28 February 2001; CAS 222370 (3), from 21°21°17.3” N, 93°56’11.0” E, 751 m
elevation, Che stream, Min Dat Township, Min Dat District, collected 2 April 2001; MCZ
R44727-44729 (dds), from Mt. Victoria, Chin State, Myanmar, collected 31 March to 2 July
1938, by Gerd Heinrich.

ADDITIONAL MATERIAL EXAMINED

Calotes chincollium — CAS 222349, 21°11'27.5” N, 94°04’56.9” E, Kanpetlet Township, Min
Dat District, Chin State, Myanmar; CAS 222350, 21°11’23.0” N, 94°01’49.3” E, Kanpetlet
Township, Min Dat District, Chin State, Myanmar; CAS 222352-222353, 21°11'44.2” N,
94°04’47.5” E, Kanpetlet Township, Min Dat District, Chin State, Myanmar; CAS
222355—222356, 21°15’43.8” N, 93°59’20.3” E, 1734 m elevation, Thui Shwn Village, Kanpetlet
Township, Min Dat District, Chin State, Myanmar; CAS 222357, 21°11’45.5” N, 94°03’48.2” E,
1372 m elevation, Kanpetlet Township, Min Dat District, Chin State, Myanmar; CAS 222358,
21°1132.3” N, 94°05’11.2” E, Kanpetlet Township, Min Dat District, Chin State, Myanmar; CAS
222359222360, 21°22’00.4” N, 94°00’40.2” E, Htin Chaun Village, Kanpetlet Township, Min Dat
District, Chin State, Myanmar; CAS 222361, 21°22’20.1” N, 93°58’34.6” E, 1482 m elevation,
Htin Chaun Village, Kanpetlet Township, Min Dat District, Chin State, Myanmar; CAS 222362,
21°23’26.6” N, 94°03/31.2” E, Htin Chaun Village, Kanpetlet Township, Min Dat District, Chin
State, Myanmar; CAS 222363, 21°22’20.1” N, 93°58’34.6” E, Baw Khue Plantation, Min Dat
Township, Min Dat District, Chin State, Myanmar; CAS 222364, 21°22’18.3” N, 93°49’00.6” E,
1787 m elevation, Hee Laung Village, Min Dat Township, Min Dat District, Chin State, Myanmar;
CAS 222365-66, 21°20'13.3” N, 93°55’22.1” E, 1047 m elevation, Che stream, Min Dat Township,
Min Dat District, Chin State, Myanmar; CAS 222367, 21°18’52.4” N, 93°54’48.7” E, 1308 m ele-
vation, Che stream, Min Dat Township, Min Dat District, Chin State, Myanmar; CAS 222368,
21°22'14.2” N, 93°48’14.1” E, Che stream, Min Dat Township, Min Dat District, Chin State,
Myanmar; CAS 222369, 21°21'14.9” N, 93°56’08.3” E, Che stream, Min Dat Township, Min Dat
District, Chin State, Myanmar; CAS 222371, 21°21'13.1” N, 93°56’03.3” E, 738 m elevation, Che
stream, Min Dat Township, Min Dat District, Chin State, Myanmar; MCZ 44730, Mt. Victoria,
Chin State, Myanmar; CAS 215505—215507, 22°15’18.0” N, 94°16’46.5” E, Lesha Chaung camp,
Alaungdaw Kathapa National Park, Sagaing Division, Myanmar.

Calotes emma — CAS 94323, Lawachera Forest, Srimangal, Bangladesh; CAS 172718, ca.
1200 m elevation, park headquarters complex, Doi Suthep National Park, Chiang Mai Province,
Thailand; CAS 172764, ca 1500 m elevation, ca 3.5 km above upper ranger station, on trail to peak,
Doi Inthanon National Park, Chiang Mai Province, Thailand; CAS 215260, 20°41’47.5” N,
96°30'17.5” E, Wat Phu Ye camp, Kalaw Township, Shan State, Myanmar; CAS 216395,

14 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 1

22°57'16.8” N, 96°14’26.5” E, Ondan Village, Shwe U Daung Wildlife Sanctuary, Moe Kok
Township, Pyin Oo Lwin District, Mandalay Division, Myanmar; CAS 222213, 17°31/23.8” N,
97°03’00.9” E, near Kyauk Phyar Village, Kyaik-Hti-Yo Wildlife Sanctuary, Kyaik-Hti-Yo
Township, Mon State, Myanmar.

Calotes jerdoni — CAS 94324, 11.6 km S of Pynursla, Assam, India; CAS 219992-219993,
21°23’01.5” N, 93°53’55.9” E, 1788 m elevation, Baw Khue Plantation, Min Dat Township, Min
Dat District, Chin State, Myanmar; CAS 220020, 21°11’08.3” N, 93°45’33.8” E, 1938 m elevation,
Min Dat Township, Min Dat District, Chin State, Myanmar; CAS 220026, 21°26'04.6” N,
93°49’29.6” E, 1663 m elevation, Min Dat Township, Min Dat District, Chin State, Myanmar; CAS
221514, 27°26'28.4” N, 97°55’07.5” E, Rabaw, Naung Mon Township, Putao District, Kachin
State, Myanmar; CAS 221551, 27°17’23.8” N, 97°51’30.5” E, road between Ahtonga and Rabaw,
Machanbaw Township, Putao District, Kachin State, Myanmar.

Calotes kingdonwardi — KIZ 730010 (holotype of C. kingdonwardi bapoensis), KIZ 730036,
Bapo, Gongshan Xian, Yunnan Province, China.

Calotes mystaceus — CAS 208441, 21°38'36.3” N, 96°00’28.8” E, fields west of Kyauk Se,
Shan Ywa Village, Kyauk Se Township, Mandalay Division, Myanmar; CAS 208446, 21°34’05.4”
N, 96°09’43.7” E, fields and hills just east of Kyauk Se Village, Kyauk Se Township, Mandalay
Division, Myanmar; CAS 213960, 20°57'14.4” N, 95°11’23.6” E, Popa Mountain Park, Kyauk Pan
Tawn Township, Mandalay Division, Myanmar; CAS 214092, 20°52’43.7” N, 95°09’56.3” E, Popa
Mountain Park, Kyauk Pan Tawn Township, Mandalay Division, Myanmar; CAS 214097,
20°53’01.8” N, 95°10/29.9” E, Popa Mountain Park, Kyauk Pan Tawn Township, Mandalay
Division, Myanmar; CAS 214117, 20°58’06.7” N, 95°14’32.4” E, Popa Mountain Park, Kyauk Pan
Tawn Township, Mandalay Division, Myanmar; CAS 214168, 21°07’21.3” N, 94°51’29.6” E,
Nyaung Oo Township, Lawka Nanda Park, Mandalay Division, Myanmar; CAS 215321,
215323-215326, 20°46’07.3” N, 96°20'05.5” E, Forest Department office, Yin Mar Bin Village,
Thazi Township, Mandalay Division, Myanmar; CAS 215930, 21°24’53.2” N, 95°46’55.0” E,
Shout Taw Yoe camp, Minsontaung Wildlife Sanctuary, Na Htoe Gyi Township, Mandalay
Division, Myanmar; CAS 215970, 21°24’09.8” N, 95°48’06.0” E, Shout Taw Yoe camp,
Minsontaung Wildlife Sanctuary, Na Htoe Gyi Township, Mandalay Division, Myanmar; CAS
216014, 21°22’44.4” N, 95°48’05.8” E, Kat Lan Dam, Minsontaung Wildlife Sanctuary, Mandalay
Division, Myanmar; CAS 216031, 21°24’43.4” N, 95°49’54.1” E, Htan Taw Village, Minsontaung
Wildlife Sanctuary, Na Htoe Gyi Township, Mandalay Division, Myanmar; CAS 215766,
22°14'36.1” N, 94°38’59.1” E, Pwint Kyi camp, Alaungdaw Kathapa National Park, Sagaing
Division, Myanmar; CAS 210705, 21°33’43.5” N, 95°12’43.9” E, near Pakokku, Magwe Division,
Myanmar; CAS 215895, 21°35’47.5” N, 95°07’18.1” E, Yar Gyi Gone Village, Shin Ma Taung
Reserve, Ye Sa Gyo Township, Pakokku District, Magwe Division, Myanmar; CAS
220574-220576, 21°14’45.5” N, 94°09’53.6” E, Forest Department, Saw Town, Saw Township,
Gan Gaw District, Magwe Division, Myanmar; CAS 220016, 21°23’16.1” N, 93°58’14.9” E, 1138
m elevation, Min Dat Township, Min Dat District, Chin State, Myanmar; CAS 220586,
21°23'26.6” N, 94°03’31.2” E, Htin Chaun Village, Kanpetlet Township, Min Dat District, Chin
State, Myanmar.

ACKNOWLEDGMENTS

We wish to thank U Shwe Kyaw, Director General, Forest Department, Ministry of
Forestry and U Khin Maung Zaw, Director, Nature and Wildlife Conservation Division, Forest
Department, Ministry of Forestry, for their continued support of the Myanmar Herpetological
Survey program. We especially wish to thank Alan E. Leviton for taking over the responsibilities

VINDUM ET AL.: ANEW CALOTES FROM WESTERN MYANMAR 3)

of the project following Joseph B. Slowinski’s untimely death. The senior author thanks Carol
Prince for her constant support. The senior author also thanks the past and present members of
Myanmar Herpetological Survey field team for their dedication and enthusiasm, which has been
invaluable to the success of this project. José P. Rosado and James Hanken, Museum of
Comparative Zoology, Harvard University, provided specimens in their care. Rao Dingqi, Division
of Herpetology, Kunming Institute of Zoology, kindly allowed the examination of specimens in the
KIZ collection. Ronald I. Crombie, National Museum of Natural History, Washington, D.C., pro-
vided information for USNM specimens. Financal support for this project was provided by funds
from NSF grant DEB-9971861 to Joseph B. Slowinski and George R. Zug, the California Academy
of Sciences and the Department of Herpetology, California Academy of Sciences. Additional fund-
ing was made possible through generous donations to the Department of Herpetology by Malcolm
and Jeanne Miller, David L. Jameson, Emily Date and the late Richard Dumke. We thank Michelle
Koo for preparing figure 6 and Dong Lin for figures 1—3 and 5. Wojciech J. Pulawski and E. N.
Arnold provided advice in Latin derivations and grammar. We thank Jeffery A. Wilkinson,
Michelle S. Koo, Alan E. Leviton, and Carl J. Ferraris Jr. for reviewing various drafts of the man-
uscript and Guinevere O. U. Wogan for assistance with the statistics. We thank Jakob Hallermann
and George R. Zug for reviewing and providing helpful comments on the manuscript. We thank the
late Joseph B. Slowinski for initiating the Myanmar Herpetological Survey project.

LITERATURE CITED

Cox, M.J., P.P. VAN Duk, J. NABHITABHATA AND K. THIRAKHUPT. 1998. A Photographic Guide to Snakes and
Other Reptiles of Thailand and Southeast Asia. Asia Books Co., Ltd. Bangkok, Thailand. 144 pp.

Davis, J.H. 1964. The forests of Burma. Sarracenia 8:1-41.

HALLERMANN, J. 2000. A new species of Calotes from the Moluccas (Indonesia), with notes on the biogeog-
raphy of the genus (Sauria: Agamidae). Bonn. Zoologische. Beitrage 49:155—-163.

LevITON, A.E., R.H. Grpss, JR., E. HEAL, AND C.E. DAWSON. 1985. Standards in herpetology and ichthyolo-
gy: part 1. Standard symbolic codes for institutional resource collections in herpetology and ichthyology.
Copeia 1985:802-832.

MANTHEY, U. AND W. GROSSMANN. 1997. Amphibien & Reptilien Stidostasiens. Natur und Tier — Verlag,
Matthias Schmidt, Miinster, Germany. 512 pp.

Ni, J.F., M. GUZMAN-SPEZIALE, M. Bevis, W.E. HoLt, T.C. WALLACE AND W.R. SEAGER. 1989. Accretionary
tectonics of Burma and the three-dimensional geometry of the Burma subduction zone. Geology 17:68—71.

PAWAR, S. AND A. BIRAND. 2001. A Survey of Amphibians, Reptiles, and Birds in Northeast India. CERC
Technical Report #6, Centre for Ecological Research and Conservation, Mysore, India. 118 pp.

SCHMIDT, K.P. 1925. New Chinese amphibians and reptiles. American Museum Novitates 175:1-3.

SHREVE, B. 1940. Reptiles and amphibians from Burma with descriptions of three new skincs [sic].
Proceedings of the New England Zoological Club 18:17—26.

SLOWINSKI, J.B., S.S. PAWAR, HTUN WIN, THIN THIN, SAI WUNNA KyI, SAN LWIN OO AND HLA Tun. 2001. A
new Lycodon (Serpentes: Colubridae) from Northeast India and Myanmar (Burma). Proceedings of the
California Academy of Sciences 52:397-405.

SuitH, M.A. 1935. The Fauna of British India, Including Ceylon and Burma. Reptilia and Amphibia, Vol II
Sauria. Taylor and Francis, London, UK. 440 pp.

TAYLor, E.H. 1963. The lizards of Thailand. University of Kansas Science Bulletin 44: 687-1077.

VENNING, F.E.W. 1910a. A collection of the ophidia from the Chin Hills. Journal of the Bombay Natural
History Society 20:331-344.

VENNING, F.E.W. 1910b. Further notes on snakes from the Chin Hills. Journal of the Bombay Natural History
Society 20:770-775.

WELCH, K.R.G. 1994. Lizards of the World: A Checklist. 5. Agamidae, Chamaeleonidae, Cordylidae and
Gerrhosauridae. KCM Books, Somerset, UK. 97 pp.

16 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 1

WERMUTH, H. 1967. Liste der rezenten Amphibien und Reptilien: Agamidae. Das Tierreich 86: 1-127.

WILKINSON, J.A., HTUN WIN, THIN THIN, Kyl SOE Lwin, AWAN KHwI SHEIN, AND HLA TuN. 2003. A new
species of Chirixalus (Anura: Rhacophoridae) from western Myanmar (Burma). Proceedings of the
California Academy of Sciences 54(2):17—26.

WoaGan, G.O.U., HTUN WIN, THIN THIN, Kyl SOE LWIn, AWAN KHWI SHEIN, SAI WUNNA KyI, AND HLA TUN.
2003. A new species of Bufo (Anura: Bufonidae) from Myanmar (Burma), and a redescription of the little
known species Bufo stuarti Smith, 1929. Proceedings of the California Academy of Sciences
54(7):141-154.

Copyright © 2003 by the California Academy of Sciences
San Francisco, California, U.S.A.

PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES

Volume 54, No. 2, pp. 17-26, 7 figs., 1 table March 14, 2003

A New Species of Chirixalus (Anura: Rhacophoridae) from
Western Myanmar (Burma)

Jeffery A. Wilkinson!*, Htun Win’, Thin Thin’, Kyi Soe Lwin’,
Awan Khwi Shein’, Hla Tun?

‘Department of Herpetology, California Academy of Sciences, Golden Gate Park, San Francisco,
California 94118; Nature and Wildlife Conservation Division, Forest Department, Ministry of Forestry,
Bayintnaung Road, West Gyogone, Insein, Yangon, Myanmar; SHOT Harvey & Associates,

3150 Almaden Expressway, Suite 215, San Jose, CA 95118

A new species of the rhacophorid genus Chirixalus is described from western
Myanmar. As with other members of Chirixalus, this species possesses a hand in
which the two outer fingers oppose the inner fingers. This species differs from other
members of this genus by a dorsal pattern of many dark brown spots on a lighter
background of the head, trunk, and legs.

Chirixalus is a relatively small genus of 13 species from Asia (Frost 2002). Members of this
genus closely resemble species of Philautus, except that they possess opposable fingers (Boulenger
1893). Of the 13 species, three are known from Myanmar (Chirixalus doriae, C. nongkhorensis,
and C. vittatus). Here, we report a fourth species collected in June of 2001 in western Myanmar
near the Bay of Bengal (Fig 1).

Specimens are housed in the collection of the Department of Herpetology, California Academy
of Sciences (CAS), the Myanmar Biodiversity Museum (MBM), and the Division of Amphibians
and Reptiles, National Museum of Natural History, Smithsonian Institution (USNM). Museum
acronyms follow Leviton et al. (1985). Tissues were removed from some specimens, then all were
fixed in 10% buffered formalin before preserving in 70% ethanol. Latitude and longitude were
recorded with a Garmin 12 GPS, set to datum WGS84.

The preserved specimens were examined, measured, and compared with available specimens
and all published descriptions of currently recognized (Frost 2002) species of Chirixalus
(Boulenger 1887, 1893; Boettger 1895; Annandale 1915; Smith 1924; Cochran 1927; Bourret
1942; Smith 1953; Roonwal and Kripalani 1961; Kuramoto and Wang 1987; Chanda and Ghosh
1989; Ray 1992). Measurements were taken using dial calipers to the nearest 0.1 mm as follows:
snout-vent length (SVL); head length (HL); head width (HW); internarial distance (IND); interor-
bital distance (IOD); snout length (SL); distance from nostril to eye (DNE); forelimb length (FLL);
hand length (HLT); thigh length (THL); tibia length (TIL); foot length (FL); width of disk of third
finger (3FDW); and width of disk of fourth toe (TDW).

M7

18 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 2

Rakhine Yoma
Elephant Wildlife
Sanctuary
Headquarters

Califomia Academy of Sciences
Department of Herpetology
lj ; . =“
———— x

90°0 95°0
30°0 95°0

FiGurE |. Distribution of Chirixalus punctatus in Myanmar with type locality indicated by a star.

SPECIES DESCRIPTION

Chirixalus punctatus Wilkinson, sp. nov.
Myanmar spotted treefrog

DIAGNOsIS.— Because Chirixalus punctatus is a relatively small frog that possesses expand-
ed discs on the fingers and toes, an intercalary element between the penultimate and terminal pha-
langes, a flange on the distal end of the third metacarpal, and opposable fingers, it has been placed
within Chirixalus. It can be distinguished from all other species of Chirixalus by a relatively uni-
form pattern of dark brown spots on a lighter ground color of the dorsal aspect of the head, trunk,

WILKINSON ET AL.: A NEW SPECIES OF CHIRIXALUS FROM MYANMAR 19

and legs. In addition, C. punctatus differs from all other Myanmar species of Chirixalus as follows:
from C. doriae by an indistinct tympanum, a more pointed and elevated snout (Fig. 3), an external
vocal sac, webbing only at the base of the third and fourth fingers and absence of webbing on the
remainder, and the absence of dark stripes along the head and dorsum (Fig. 4); from C.
nongkhorensis by the absence of a prominence at the end of the snout (Fig. 3), the absence of small
tubercles on the dorsum, the absence of striping on the dorsal aspect of the thighs (Fig. 5), an indis-
tinct canthus rostralis, and an indistinct tympanum; and from C. vittatus by a stockier body, a
broader head, and larger toe pads (Fig. 6).

Ho.LotyPeE.— CAS 221555 (Fig. 2), an
adult male, collected at the Rakhine Yoma
Elephant Wildlife Sanctuary Headquarters
(17°36’48.8” N, 94°36’50.2” E), Gwa, Gwa
Township, Rakhine State, Myanmar, collected 3
June 2001 by Htun Win, Thin Thin, Kyi Soe
Lwin, and Awan Khwi Shein.

DESCRIPTION OF HOLOTYPE.— Habitus
moderately stocky; 22.9 mm SVL; head as long
as broad, head length 33.8% of SVL; head
width 34.5% of SVL; snout pointed in lateral
view (Fig. 3), slightly longer than the diameter
of the eye, sloping anteroventrally, projecting
beyond the mouth; canthus rostralis is rounded,
not distinct; loreal region is slightly oblique and .
concave; nostrils are slightly protuberant and FIGURE 2. The holotype of Chirixalus punctatus (CAS
nearer to the tip of the snout than the eye; 221555).
internarial space is slightly smaller than interor-
bital space; interorbital space is wider than upper eyelid; eyes are moderately large, diameter of eye
41% of head length; tympanum not distinct, slightly dorsoventrally oval with slight posterior incli-
nation, less than half the diameter of eye; tympanic annulus anteroventrally raised, obscured pos-
terodorsally by slightly raised supratympanic fold.

Vomerine teeth absent; choanae oval partially hidden by edge of jaw when viewed ventrally;
tongue, attached anteriorly, V-shaped notch posteriorly; external vocal sac and vocal slits present;
premaxilla and maxilla with minute straight teeth.

Third finger longest; followed by the fourth finger, second finger, and the first finger is the
shortest; expanded disks on fingers with circummarginal grooves; disk on the third finger largest,
approximately the same size as the tympanum; the webbing between the first and second, and sec-
ond and third fingers is absent, between the third and fourth fingers only at base; the two outer fin-
gers distinctly separated from the two inner fingers (opposable); subarticular tubercles between
penultimate and adjoining proximal phalange oval and well developed.

Hindlimbs relatively short, the tibio-tarsal articulation reaches to between the eye and the tip
of the snout; heels overlapping when thighs and tibia are placed at right angles to body; length of
tibia 51.4% of SVL; foot length 45.1% of SVL; fourth toe longest, fifth toe a little longer than third,
second much shorter than third and slightly longer than first; all toes with disks with circummar-
ginal grooves, a little smaller than those of fingers; the webbing on the toes extends to more than
two-thirds the length of the toe with a webbing pattern of I 11/2-2 If 1-2 II 1-2 IV 2-11/2 V fol-
lowing Myers and Duellman (1982); the subarticular tubercles are moderately developed (oval);
the inner metatarsal tubercle is small indistinct and oval; no outer metatarsal tubercle. The skin on

20 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 2

FicurE 3. Lateral views of the heads of A) Chirixalus punctatus (CAS 221555), B) C. doriae (CAS 210088), C) C. vit-
tatus (CAS 221114), and D) C. nongkhorensis (FMNH 172569).

the dorsum is smooth, on the throat with small granules, on the abdomen and underside of thighs
with larger granules.

COLORATION IN ALCOHOL.— A uniform pattern of minute dark flecks with larger dark (brown)
spotting on a lighter background over the head, dorsum, and legs, spotting merges together to form
a broken line from snout to eye across the canthus rostralis and from the eye dorsolaterally to groin;
a light dorsolateral stripe beginning well behind the eye and ending well before the hindlimbs; there
is no pattern or striping on the legs or groin.

VARIATION.— The females are larger than the males (Table 1) but the tibio-tarsal articulation
reaches only to the posterior end of the eye, whereas, in the males it reaches to the eye or between
the eye and the tip of the snout. The tympanum is more obscure in some individuals than in others.
Also, the amount of spotting varies, being sparser in some individuals, restricted primarily to the
dorsal aspect of the head whereas in others to the entire dorsum. In some individuals, there is no
merging together of the spots to form a broken dorsolateral line from the eye to the groin and no
light dorsolateral stripe after preservation, as in the holotype. In life [based on a color transparen-
cy of a paratype (JBS 9274); Fig. 7], the dorsal color is a yellowish brown and the ventrum is
brownish pink; a pale yellow dorsolateral stripe extends from the snout to just before the vent; the
iris is bicolored, with a light upper color meeting the dorsolateral stripe as if the stripe extends
through the eye; the fingers are a lighter yellowish brown appearing almost translucent. After
preservation in ethanol the dorsal and ventral ground color is white and the dorsolateral stripe
fades.

ETYMOLOGY.— The name punctatus, from the Latin meaning of having spots, because of the
dorsal spotting that distinguishes this species from all other species of Chirixalus.

WILKINSON ET AL.: A NEW SPECIES OF CHIRIXALUS FROM MYANMAR 2]

FIGURES 4-6. Dorsal views of (4) Chirixalus doriae (CAS 210088); (5) Chirixalus nongkhorensis (FMNH 172569); (6)
Chirixalus vittatus (CAS 221114).

DISTRIBUTION AND NATURAL History.— At present, Chirixalus punctatus is known only
from the southwestern foothills of Rakhine Yoma, Gwa Township, Rakhine State, in western
Myanmar (Fig. 1). All individuals were found in disturbed habitat between mountain evergreen
forests and agricultural land, 1 to 2 m off the ground in bushes. Searches of the evergreen forests
failed to find any individuals. The males were calling at the time of collection. The call consisted
of a series of a short “dat” separated by 3-4 second intervals. Amplectant pairs were also observed
in the plants, where the female would select an oviposition site to build a foam nest for egg depo-
sition on leaves overhanging standing water. The foam nests were found only in a particular plant
of the genus Arum in the family Areceae.

Other frogs collected in the vicinity of the localities of Chirixalus punctatus were Amolops sp..
C. nongkhorensis, Occidozyga sp., Philautus sp., Polypedates leucomystax, Rana alticola, R. ery-
thracea, R. lateralis, R. cf. limnocharis, and R. nigrovittata.

2, PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES

TABLE |. Measurements of type series of

Chirixalus punctatus. Mean (in mm) followed by

range (in parenthesis). Please see text for abbreviations.

Males (N=21)

Females (N=6)

Volume 54, No. 2

SVL 22.9 (21.5—25.0) 26.7 (24.7—28.4)
HL 8.0 (7.4-8.8) 9.2 (8.1-10.0)
HW 7.8 (7.19.5) 8.7 (7.59.6)
IND 2.4 (2.0-2.8) 2.7 (2.2-3.0)
IOD 3.1 (2.9-3.4) 3.5 (3.0-3.9)
SN 3.2 (2.8-3.5) 3.9 G.4-4.2)
DNE 1.8 (1.5-2.7) 2.1 (1.6-2.4)
FLL 10.5 (9.6-11.9) 12.2 (10.8-12.9)
HL 6.6 (6.0—7.5) 7.5 (6.5—8.2)
THL 11.1 (9.8—-12.6) 12.1 (11.3-12.6)
Te 11.3 (10.5—12.3) 12.8 (11.9-13.8)
FL 9.6 (8.9-10.6) 10.9 (9.4-11.8)
3FDW 1.3 (0.9-1.7) 1.4 (1.3-1.6)
4TDW 0.9 (0.6—1.1) 1.0 (0.7-1.2)
DISCUSSION

As stated earlier, species of the genus Chirixalus are distinguished from other Asian rha-
cophorids by the presence of opposable fingers. Liem (1970) also suggested that the combination
of an elongated vertebral column (2.4 times the width) and the absence of vomerine teeth, along
with opposable fingers, can distinguish Chirixalus from other rhacophorids. However, many
species of Philautus also lacked vomerine teeth and all possessed an elongated vertebral column
(1.6 to 2.4 times the width) in Liem’s study. The close resemblance of species within Chirixalus to
those within Philautus has prompted some researchers to either move some of the 13 species of
Chirixalus into Philautus or not recognize Chirixalus as a genus separate from Philautus (Cochran
1927; Pope 1931; Bourret 1942; Taylor 1962). For example, C. palpebralis was originally assigned
to Philautus by Smith (1924), even though he noted that the “first two fingers [were] partially
opposed to the others” and though he considered it “most nearly related to C. doriae”. Based on
these statements, Cochran (1927) and Pope (1931) did not recognize Chirixalus as a genus sepa-
rate from Philautus. Taylor (1962) followed suit though he did suggest that Chirixalus may be rec-
ognized as a genus separate from Philautus based on the presence of opposable fingers. However,
Bourret (1942) recognized Chirixalus as a genus separate from Philautus, but he did not place all
current species into Chirixalus (1.e., laevis and vittatus). Finally, C. palpebralis was again assigned
to Chirixalus by Inger et al. (1999), based on the presence of opposable fingers.

Recently, Bossuyt and Dubois (2001) tentatively assigned C. cherrapunjiae (Roonwal and
Kripalani), C. romeri (Smith), and C. shyamrupus (Chanda and Ghosh) to Chirixalus based on the
presence of a tadpole stage in C. cherrapunjiae and C. romeri, since, all species of Philautus are
thought to be direct developers (Dring 1987), and because C. shyamrupus has longitudinal lines on
the body, which they contend are not present in species of Philautus. Upon examining the figures
and reviewing the text from the original descriptions, it is our view that C. shyamrupus is the only
species to possess opposable fingers (Chanda and Ghosh 1989, Fig. 1) and thus should be assign
to Chirixalus. Philautus romeri and P. cherrapunjiae appear to lack opposable fingers (Smith 1953;

WILKINSON ET AL.: A NEW SPECIES OF CHIRIXALUS FROM MYANMAR. 2B

FIGURE 7. Paratype of Chirixalus punctatus (JBS 9274).

Roonwal and Kripalani 1961; Karsen and Lau 1986; Fei 1999). Philautus romeri may be a mem-
ber of the new genus Kurixalus Ye, Fei, and Dubois, whereas P. cherrapunjiae may be a member
of Rhacophorus but a determination will require examinations of the type specimens and observa-
tions of individuals in the field.

In a recent study of rhacophorid phylogeny, Chirixalus was the only paraphyletic genus
(Wilkinson et al. 2002). Of five species of Chirixalus sampled (C. doriae, C. eiffingeri, C. idiooto-
cus, C. palpebralis, and C. vittatus), C. eiffingeri and C. idiootocus formed a strong sister taxon
relationship, separate from C. doriae and C. vittatus. The (C. eiffingeri, C. idiootocus) clade formed
a weaker relationship with members of Philautus, however, these two taxa shared only a few
molecular and morphological characters with Philautus species (Wilkinson et al. 2002). Fei (1999)
recently placed C. eiffingeri into the new genus Kurixalus, and assigned C. idiootocus to Philautus.
The results of the Wilkinson et al. (2002) study and the absence of opposable fingers in C. eiffin-
geri and C. idiootocus led Wilkinson and colleagues to assign C. idiootocus along with C. eiffin-
geri to Kurixalus. Their results also suggest a close relationship of Philautus and Kurixalus.

Chirixalus doriae formed a strong clade with Chiromantis, whereas C. vittatus formed a weak
clade with Polypedates in the Wilkinson et al. (2002) study. Chirixalus palpebralis did not form a
clade with any rhacophorid genus suggesting that it may be a representative of a heretofore
unnamed genus.

It is interesting to note that as with species of Chiromantis, Polypedates, and Rhacophorus,
Chirixalus doriae (the type species for this genus) constructs a foam nest. Chirixalus punctatus also
constructs a foam nest. We therefore place C. punctatus within Chirixalus due to the presence of
opposable fingers, the foam-nesting behavior, and other characters as described above but recog-
nize the need for a systematic study of this genus that includes all known species and further eval-
uates their relationships with other rhacophorid genera.

24 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 2

We provide the following key to distinguish between ten species of Chirixalus (minus P. cher-
rapunjiae, K. eiffingeri, K. idiootocus, and P. romeri) based on specimens we examined and origi-
nal descriptions. We could not distinguish between C. hansenae and C. vittatus. One difference
according to the original descriptions is a distinct tympanum in C. hansenae (Cochran 1927) and
an indistinct (hidden) tympanum in C. vittatus (Boulenger 1887). However, we observed the tym-
panum to be indistinct in half of the specimens of C. hansenae that we examined, and we observed
the tympanum to be distinct in one specimen of C. vittatus (MVZ 222098). We observed the tym-
panum on a paratype of C. hansenae (USNM 70110) to be distinct. Cochran (1927) based her
description of C. hansenae on nine specimens. Of these nine, she reported that the tympanum in
two specimens “is not so distinct because it is more nearly the color of the surrounding skin,
although it can be readily outlined when the frog is placed in a different light’. In addition, she
compared these specimens with specimens of C. palpebralis and C. doriae but not with C. vitta-
tus. Therefore, C. hansanae may be a junior synonym of C. vittatus, and the following key should
be used with that caveat.

KEY TO THE SPECIES OF CHIRIXALUS

I> ongitudinal stripes on dorsum present... 2... 3.5...5.22 - -05 1 ee eee 2
Eongitudinal stripes on dorsum) absent... . 522... ~ 82-22. 1. yo eee 9
2 Dorsolateral stripes presents... 25. ee ee SS oe cares oe nye Renee ee 3
Several dark longitudinal stripes present, distinct dorsolateral stripes absent.............. 8
S-Lympanum Gistnct 2. 5. o.ce ee s as gress ov + cites oh Se aa Re 2 SPICED See eee 4
ym panumaNGiShMct s,s acs. aie sic. us, seteve a! aisle Ayns wien @ so ee eke eee 7
As innermetatarsal tubercle present 252. 6.44529 2.-% + ere ete te Gk eee 5
Inner metatarsal tubercle absent, found in Arunachal Pradesh, India............ shyamrupus
5. Third and fourth fingers 1/4 webbed, found in Uttar Pradesh, India............ dudhwaensis
Noor very slight webbing on fingers... «22.7. 222). 26 42% on = «ees a ee 6
6. Disk on third finger as large as tympanum, 1/2 to 2/3 webbing on toes, indistinct
metatarsal tubercle, found in east-central Thailand). 2.22 464.45 So. a ee hansenae
Disk on third finger smaller than tympanum, 1/3 webbing on toes, distinct oval
metatarsal tubercle; found in Annam; Vietnam..- 22: 52 sai... 4? (soca laevis
7. Dark spots on dorsum present, stocky body, broader head, larger toe disks, found in
westem: Myanmar ..vc).ni sR dene tee bam oer en it tee ee punctatus
Dark spots on dorsum absent, slender body, narrower head, smaller toe disks, found
throuchout Southeast and East Asiavs si. cc8 oe da 2 2G 2s bee ele ee vittatus
8. Snout obtusely pointed, skin smooth, glandular fold between eye and shoulder absent,
found throughout South; Southeast, and East Asia. 2. ... 2.) 2:2 ¢ 292) ee doriae
Snout truncated, skin of head with small round scattered warts, glandular fold between
eye and shoulder present, found in northeasterm India 2.2222.) -41)..4-2 See simus

9. Tympanum distinct, interorbital space broader than upper eyelid, fingers distinctly
opposable, blotched pattern on dorsum, white patch on side of upper jaw absent, found
inssoutheaswA sian te 3 swears. ieee Oe Le Se ee nongkhorensis
Tympanum indistinct, interorbital space as large as upper eyelid, fingers only partially
opposable, hourglass pattern on dorsum, white patch on side of upper jaw present,
found/in:Annam, Vietnam:and Yunnan; China). .:.). i 2. see eee See palpebralis

WILKINSON ET AL.: A NEW SPECIES OF CHIRIXALUS FROM MYANMAR. 25

MATERIAL EXAMINED

HowotyPe.— CAS 221555, adult male, collected at the Rakhine Yoma Elephant Wildlife
Sanctuary Headquarters (17°36’48.8” N, 94°36’50.2” E), Gwa, Gwa Township, Rakhine State,
Myanmar, 3 June 2001, by Htun Win, Thin Thin, Kyi Soe Lwin, and Awan Khwi Shein.

PARATYPES.— CAS 221553-221554, CAS 221556—221559, USNM 547928-547929, Joseph
B. Slowinski (JBS) field numbers to be deposited in the MBM, JBS 8996, 8998-8999, 9201 seven
adult males and five adult females, collected at the same locality and date as the holotype; CAS
221560, an adult male, from the De-Pok Village camp (17°53’35.5” N, 94°30/30.8” E), Gwa
Township, Rakhine State, Myanmar, collected 4 June 2001 by Htun Win, Thin Thin, Kyi Soe Lwin,
and Awan Khwi Shein; CAS 221561, 221964—221965, JBS 9231-9232, 9234, six adult males,
from the De-Pok Village camp (17°53’50.1” N, 94°30’31.7” E), Gwa Township, Rakhine State,
Myanmar, collected 4 June 2001 by Htun Win, Thin Thin, Kyi Soe Lwin, and Awan Khwi Shein;
CAS 221562-221563, 221966, JBS 9237 four adult males, from the De-Pok Village camp
(17°53’44.6” N, 94°30'30.8” E), Gwa Township, Rakhine State, Myanmar, collected 4 June 2001
by Htun Win, Thin Thin, Kyi Soe Lwin, and Awan Khwi Shein, and JBS 9274, 9276, two males,
from the De-Pok Village camp (17°53’5” N, 94°30’23” E), Gwa Township, Rakhine State,
Myanmar, collected 6 June 2001 by Htun Win, Thin Thin, Kyi Soe Lwin, and Awan Khwi Shein.

ADDITIONAL MATERIAL EXAMINED

Chirixalus doriae: China, NMNS 3183; Myanmar, CAS 210078—210083, 210085—210088,
210091—210094; Vietnam, ROM 32762, 32764, 32782-32783, 32884, 32915, 32951-32952,
33006, 33008. Chirixalus eiffingeri: Japan, CAS 211453-211458; Taiwan, CAS 211527-—211531.
Chirixalus hansenae: Thailand, FMNH 182568, 182572—182575, 182577, 182580—182584,
182586—182588, 182590-182591, USNM 70110. Chirixalus idiootocus: Taiwan, CAS
16603 1—166034, 211366, 211524—211526. Chirixalus laevis: Vietnam, ROM 30282. Chirixalus
nongkhorensis: Myanmar, CAS 215915, 221564; Thailand, FMNH_ 172562-172563,
172567—172570, 182596, 182598—-182602, 182607, 182610, 187518, 187519. Chirixalus palpe-
bralis: China, NMNS 3104; Vietnam, MVZ 221980—221982, 221986-221987, 221989, 222008,
222010—222011, 222014. Chirixalus vittatus: China, NMNS 3184; Myanmar, CAS
221114221115, 221212—221213, 221266—221267, 221293, 221565—221566; Vietnam, MVZ
222098, ROM 30338-30339, 30341, 30349, 30352, 30362-30363, 30369, 30381.

ACKNOWLEDGMENTS

We thank U Shwe Kyaw, Director General, Forest Department, Ministry of Forestry, and U
Khin Maung Zaw, Director, Nature and Wildlife Conservation Division, Forest Department,
Ministry of Forestry for their continued support of the Myanmar Herpetological Survey Project.
We also thank the Field Museum of Natural History, Museum of Vertebrate Zoology, University of
California, Berkeley, National Museum of Natural History, Smithsonian Institution, National
Museum of Natural Science, Taiwan, and the Royal Ontario Museum for loaning specimens. Field
work was supported by National Science Foundation Grant (DEB-9971861) to. Joseph B.
Slowinski and George Zug. Michelle S. Koo provided Figure 1. Alan E. Leviton provided Figures
2-4. Jens V. Vindum, Guin Wogan, George Zug, and Darrel R. Frost critically read and provided
valuable comments on the manuscript. The Myanmar Herpetological Survey Project was initiated
by the late Joseph B. Slowinski.

26 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 2

LITERATURE CITED

ANNANDALE, N. 1915. Herpetological notes and descriptions. Records of the Indian Museum 11:341-347.

BOETTGER, O. 1895. Neue Frésche und Schlangen von den Liukiu-Inseln. Zoologeshen Anzeiger 18:266-270.

BossuytT, F. and A. DuBois. 2001. A review of the frog genus Philautus Gistel, 1848 (Amphibia, Anura,
Ranidae, Rhacophorinae). Zeylanica 6:1—112.

BOULENGER, G.A. 1887. An account of the batrachians obtained in Burma by M. L. Fea, of the Genoa Civic
Museum. Annali del Museo Civico di Genova, ser. 2, 5:418—424.

BOULENGER, G.A. 1893. Concluding report on the reptiles and amphibians obtained in Burma by Signor L.
Fea, dealing with the collection made in Pegu and the Karin Hills in 1887-88. Annali del Museo Civico
di Genova, ser. 2,13:304—347. ;

Bourret, R. 1942. Les batracines de l’Indochine. Institut Océanographique, Hanoi. 547 pp.

CHANDA, S.K. and A.K. GHOsH. 1989. A new frog of the genus Philautus Gistel, from the proposed Namdapha
Biosphere Reserve, Arunachal Pradesh, Northeast India. Journal of the Bombay Natural History Society
86:215—217.

COCHRAN, D.M. 1927. New reptiles and batrachians collected by Dr. Hugh M. Smith in Siam. Proceedings
of the Biological Society of Washington 40:179-192.

DRING, J. 1987. Bornean treefrogs of the genus Philautus (Rhacophoridae). Amphibia-Reptilia 8:19-47.

Fel, L. 1999. Atlas of Amphibians of China. Publishing House for Scientific and Technological Literature,
Hunan. 432 pp.

Frost, D.R. 2002. Amphibian Species of the World: an Online Reference. V2.21 (15 July 2002). Electronic
database available at http://research.amnh.org/herpetology/amphibia/index.html.

INGER, R.F., N. ORLOV, and I. DAREvsKy. 1999. Frogs of Vietnam: A report on new collections. Fieldiana
Zoology 92:146.

KARSEN S.J. and M.W.-N. Lau. 1986. Hong Kong Amphibians and Reptiles. Urban Council, Hong Kong.

KurAMoTO, M. and C.-S.WANG. 1987. A new rhacophorid treefrog from Taiwan, with comparisons to
Chirixalus eiffingeri (Anura, Rhacophoridae). Copeia 1987:931—942.

Leviton, A.E., R.H. Gipss, JR., E. HEAL, and C.E. DAwson. 1985. Standards in herpetology and ichthyology:
Part I. Standard symbolic codes for institutional resource collections in herpetology and ichthyology.
Copeia 1985:802-832.

Liem, S.S. 1970. The morphology, systematics, and evolution of the Old World treefrogs (Rhacophoridae and
Hyperoliidae). Fieldiana Zoology 57:1—145.

Myers, C.W., AND W.E. DUELLMAN. A new species of Hyla from Cerro Colorado, and other treefrog records
and geographical notes from western Panama. American Museum Novitates (2752):1—32.

Pope, C.H. 1931. Notes on amphibians from Fukien, Hainan, and other parts of China. Bulletin of the
American Museum of Natural History 61:397-611.

Ray, P. 1992. Description of a new rhacophorid, Chirixalus dudhwaensis (Anura: Rhacophoridae) from
Dudhwa National Park, District Lakhimpur Kheri, Uttar Pradesh, India. Indian Journal of Forestry
15:260-265.

RoonwaL, M.L. and M.B. KRIPALANI. 1961. A new frog, Philautus cherrapunjiae (family Ranidae) from
Assam, India, with field observation on its behaviour and metamorphosis. Records of the Indian Museum
59:325-333.

SmiTH, M.A. 1924. New tree-frogs from Indo-China and the Malay Peninsula. Proceedings of the Zoological
Society of London 1924:225-234.

SmitH, M.A. 1953. Description of a new species of frog of the genus Philautus. Annals and Magazine of
Natural History 6:477-478.

TAYLOR. E.H. 1962. The amphibian fauna of Thailand. University of Kansas Science Bulletin 43:265—599.

WILKINSON, J.A., R.C. DREWEs, and O.C. Tatum. 2002. A molecular phylogenetic analysis of the family
Rhacophoridae with an emphasis on the Asian and African genera. Molecular Phylogenetics and
Evolution 24:265-273.

Copyright © 2003 by the California Academy of Sciences
San Francisco, California, U.S.A.

PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES

Volume 54, No. 3, pp. 27-64, 4 figs., 7 pls., 1 table March 14, 2003

Diatom Biostratigraphy of the Neogene
Milky River Formation, Alaska Peninsula,
Southwestern Alaska

Andrey Yu. Gladenkov

Institute of the Lithosphere of Marginal Seas, Russian Academy of Sciences,
Staromonetny per., 22, Moscow 119180, Russia; E-mail: agladenkov @ilran.ru;
FAX: (7095) 953-5590; Phone: (7095) 951-2723

The Milky River Formation of the Alaska Peninsula, southwestern Alaska, contains
more than 100 marine diatom and three silicoflagellate taxa in more than 50 samples
from the 263-m-thick (863-foot-thick) Sandy Ridge stratigraphic section. Age-diag-
nostic diatoms include Neodenticula kamtschatica, Thalassiosira oestrupii and
Cosmiodiscus insignis, whose co-occurrence characterizes the latest Miocene to early
Pliocene subzone b of the North Pacific Neodenticula kamtschatica diatom Zone of
Barron and Gladenkov (1995). The occurrence of additional biostratigraphically-
important diatoms, including Thalassiosira temperei and Thalassiosira latimarginata,
is the basis for refining the age of different parts of the formation. Diatoms occur
directly within the shells of the bivalve Astarte and other marine mollusks of the
Milky River Formation. The first appearance of Astarte in North Pacific stratigraph-
ic sections signals the earliest opening of Bering Strait, because the genus otherwise
dwelled only in the North Atlantic-Arctic region. Diatoms from the stratigraphically
lowest Astarte-bearing horizon in the Milky River Formation indicate an age range
of 5.5-5.4 Ma for the strait’s earliest opening. Data on the age of Neogene diatom-
bearing sediments resulting from this study also contribute to the stratigraphic
framework of Alaska and to refined regional correlations.

Fossil diatoms occur widely in deep-sea and onshore stratigraphic sequences in the middle to
high latitudes of the North Pacific and are the primary biostratigraphic tool for the precise dating
and correlation of marine sediments in this region. The current high-resolution North Pacific
Neogene diatom zonation includes numerous biohorizons based on datum levels, which have been
directly correlated to the magnetostratigraphy of the late early Miocene to Quaternary (Barron
1980, 1992a; Koizumi and Tanimura 1985; Koizumi 1992; Barron and Gladenkov 1995;
Yanagisawa and Akiba 1998). However, even though Cenozoic marine diatoms of different ages
are well represented in sea floor sediments (Kanaya and Koizumi 1966; Jousé et al. 1969; Sancetta
198la-b, 1982; Sancetta and Silvestri 1986) and deep-sea cores in the eastern part of the Bering
Sea and the Gulf of Alaska (Schrader 1973; Barron and Gladenkov 1995), they are extremely rare
or absent in onshore sequences of Alaska. As a result, it is often difficult to determine the precise
ages and correlations of these sediments. A relatively diverse Neogene diatom flora has previous-
ly been studied only from the Pribilof Islands, southern Bering Sea, Alaska (Hanna 1919, 1970),
and sparse remains of poorly preserved marine diatoms have also been reported from the
Middleton Island section of the Yakataga Formation along the northeastern Gulf of Alaska margin
(Plafker and Addicott 1976). Assigning precise ages to these floras has proved difficult, owing to

Ly

28 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 3

the absence of biostratigraphically significant taxa. The present study of diatoms from the Milky
River Formation, including age-diagnostic species, is the first description of a Cenozoic diatom
assemblage from the Alaska Peninsula, southwestern Alaska. These diatoms are present in strati-
graphically well-controlled samples and are highly significant for age, regional correlations, and
paleoenvironmental reconstruction of the Milky River Formation.

PREVIOUS STUDIES, MATERIALS AND METHODS

Certain diatom species in the Milky River marine diatom flora have been the basis for docu-
menting the earliest opening of Bering Strait (Marincovich and Gladenkov 1997, 1999, 2001;
Gladenkov and Marincovich 1998, 1999). Part of the present diatom flora was first discovered
within five mollusk shells from two stratigraphic levels in the Sandy Ridge stratigraphic section
that were housed in the California Academy of Sciences, San Francisco. These samples had been
collected from Sandy Ridge in the course of petroleum geological field work by Mobil Oil
Company geologists in the 1970s. At that time, this sequence was assigned to the upper part of the
Bear Lake Formation (Detterman et al. 1996). Initial study of the diatoms showed them to be char-
acteristic of Subzone b of the North Pacific Neodenticula kamtschatica diatom Zone of Barron and
Gladenkov (1995), with an age range of 5.5-4.8 Ma based on correlations with the time scale of
Berggren et al. (1995). Even though these diatoms and mollusks (Astarte) were the basis for dat-
ing the earliest opening of Bering Strait (Marincovich and Gladenkov 1999, 2001; Marincovich
2000), the precise location of these fossils in the Sandy Ridge stratigraphic section was unknown.
In order to place these fossils within a stratigraphic, geological and paleontological context, field
work at Sandy Ridge was conducted in August 1998 and July 1999 under the leadership of L.
Marincovich, Jr., and supported by National Science Foundation grant OPP 9806461 (Marincovich
et al. 2002; Gladenkov et al. 2002). This additional sampling at Sandy Ridge was critical to more
precisely dating the first occurrence of Arctic-Atlantic mollusks within the North Pacific diatom
biostratigraphic scheme, especially since diatoms were absent or extremely rare in the enclosing
matrix but relatively abundant in the mollusk shells. Sampling the interiors of mollusk shells was
done because in some cases fossil diatoms are present in fine sediment within shells and are there-
by protected from abrasion and dissolution (Barron and Mahood 1993). Fine grained sediments
containing diatoms might also be protected from winnowing by bottom currents if they are within
molluscan shells.

A total of 56 samples for diatom analysis were collected during field work at the Sandy Ridge
Stratigraphic section (Figure 1). These strata had previously been assigned to the Bear Lake
Formation (Detterman et al. 1996). However, in the course of field work it was discovered that the
main part of these rocks contain, exclusively, late Miocene mollusks, in contrast to the early mid-
dle Miocene mollusks known from the Bear Lake Formation (Marincovich et al. 2002; Gladenkov
et al. 2002). These strata were therefore reassigned to the Milky River Formation, which had been
proposed by Galloway (1974) for volcanic and sedimentary rocks that overlies the Bear Lake
Formation on the Alaska Peninsula. A profound angular unconformity separates these two forma-
tions at Sandy Ridge, with semi-isoclinally folded non-marine conglomerate, sandstone and coal
beds of the lower middle Miocene Bear Lake Formation underlying subhorizontal shallow-marine
sediments of the upper Miocene Milky River Formation. It is now clear that all previously studied
marine diatoms and mollusks from this stratigraphic section came not from the upper part of the
Bear Lake Formation, as supposed by Marincovich and Gladenkov (1997, 1999, 2001), but from
the overlying Milky River Formation. The stratigraphic horizon with the first appearance of
Astarte, which is the signal of an open Bering Strait, is 28 meters above the angular unconformity

GLADENKOV: DIATOM BIOSTRATIGRAHY IN SOUTHWESTERN ALASKA 29

Lp &
| GULF OF
w A
i 1] Oa * ; ALASKA
2 ¥° GULF OF K

ALASEA PACIFIC OCEAN

5 kilometers

CAS 608412
CAS 608388

Quaternary alluvium io) Milky River Formation
(marine facies)
Quaternary volcanic rocks

Bear Lake Formation
Tertiary volcanic rocks

Paleogene rocks

. Milky River Formation :

Ficure 1. Location of the study area on the Alaska Peninsula and geological map of the Sandy Ridge area (modified
after Detterman et al. 1996), with arrow showing the Sandy Ridge stratigraphic section of the Milky River Formation (after
Gladenkov et al. 2002). Localities of the stratigraphically lowest (CAS 608372) and highest (CAS 608412) samples in the
Sandy Ridge stratigraphic section yielding diatoms are shown. CAS 608382 indicates the locality of sample collected from
the stratigraphically lowest Astarte-bearing horizon in the Milky River Formation.

30 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 3

Meters
300 buelegy

and within the lower part of the Milky River
Formation (Marincovich et al. 2002; Gladenkov et
al. 2002). In the course of field work at Sandy
Ridge, 52 samples for diatom analysis were col-
lected from the 263-m-thick (863-ft-thick) section
of the Milky River Formation and four samples
were collected from the uppermost part of the
underlying Bear Lake Formation (Figures 1-2;
Table 1).

Due to the induration of the sediments and the
low concentration of diatoms, the following proce-

dure was used to process the samples. Samples
were preliminary crushed and then placed in acetic 100 -
acid for several days. The acid-treated material was 5 ed
then made pH-neutral by repeatedly filling and eeepc
decanting the beakers with distilled water, allowing 50>. =

3 hr or more for each rinse. The material was then = oe a CAS 608382 (F Astarte)
processed by boiling it in a solution of water and |
sodium pyrophosphate (Na,P,07x10H,O), fol-

lowed by repeated washing with distilled water,

SSS

0-4 2SSeS= —€— CAS 608372
Beane foal Foe |
fS cs cg

FIGURE 2. Generalized stratigraphic column of the

and then a centrifuge (at 2500 rpm for 10 min) was
used with heavy liquid (specific weight = 2.3) to
separate a suspension fraction. To remove the
heavy liquid, both the suspension and residue were
again rinsed in distilled water. Finally, strewn
slides were prepared by placing the material in a
vial, adding distilled water, agitating the vial, and
removing part of the upper suspension with a
pipette. Strewn slides! were prepared by spreading

Milky River Formation at the Sandy Ridge stratigraph-
ic section (after Gladenkov et al. 2002)

The wavy line indicates an angular unconformity
separating the Milky River Formation and the Bear
Lake Formation. Arrows show stratigraphically lowest
(CAS 608372) and highest (CAS 608412) samples lev-
els possessing diatoms, and the sample level (CAS
608382) coinciding with the stratigraphically lowest
Astarte-bearing horizon (F Astarte). Abbreviations
used: fs—fine-grained sandstones, cs—coarse grained
sandstones, cg—conglomerate.

the pipette suspension onto a cover slip (size 24x24
mm), drying on a hot plate, and mounting in Naphrax (index of diffraction = 1.74). Slides prepared
from both fractions obtained after centrifuging were studied as a control. The slides were examined
in their entirety under a Jeneval (Zeiss) light microscope at 400x, with identifications routinely
checked at 1000x. Whenever possible, all diatom taxa were counted up to a maximum of 200 spec-
imens (other than Chaetoceros spores). Chaetoceros spores were tabulated separately while count-
ing the other taxa. After counting, slides were scanned to record the presence of other diatom
species missed among the first 200 specimens. When fewer than 200 diatom valves were encoun-
tered on a slide, all of the taxa were tabulated. Silicoflagellates, if present, were tabulated for the
entire slide. The preservation of diatoms is listed as G (good), M (moderate), and P (poor) depend-
ing on the degree of destruction and dissolution of valves. The relative abundance is evaluated as
A (abundant, more than 1500 diatom valves per slide), C (common, 400-1500 valves), F (few,
250-400 valves), and R (rare, 250-50 valves), and VR (very rare, less than 50 valves).

Numerical ages have been updated herein according to the geochronologic and geomagnetic
polarity scales after Berggren et al. (1995). The correlation of geological epochs and periods also
follows Berggren et al. (1995), with the exception of the Pliocene subepochs. The Pliocene epoch

|The original set of slides are in Gladenkov’s laboratory in Moscow. A duplicate set of slides has been deposited in
the Diatom Collection at the California Academy of Sciences.

GLADENKOV: DIATOM BIOSTRATIGRAHY IN SOUTHWESTERN ALASKA iil

is divided into early, middle, and late subepochs that are equivalent to the European Zanclean,
Piacenzian, and Gelasian stages, respectively. As a result, the former late Pliocene is divided into
middle and late subepochs, as was adopted at the 30th International Geological Congress (Beijing,
China, 1996). The boundary between middle and late Pliocene is at 2.58 Ma coinciding with the
top of Chron C2An. The North Pacific Neogene diatom zonation follows Barron and Gladenkov
(1995) (Figure 3).

RESULTS AND DISCUSSION

OCCURRENCE OF BIOSTRATIGRAPHICALLY SIGNIFICANT TAXA.— Three of 4 samples from the
Bear Lake Formation are barren of siliceous microfossils. Rare diatoms representing only a few
taxa were found only in sample CAS 608372 (1.5m below the unconformity) (Figures 1-2).
Determining a precise age for the poorly preserved assemblage in sample CAS 608372, from the
uppermost part of the Bear Lake Formation at Sandy Ridge, has proven difficult owing to the
absence of co-occurring, biostratigraphically significant taxa characteristic of the Neodenticula
kamtschatica Zone or older diatom zones (Table 1). About 100 species and varieties of diatoms and
three taxa of silicoflagellates were identified in the 52 samples examined from the 263-m-thick sec-
tion of the Milky River Formation exposed above the unconformity at the Sandy Ridge stratigraph-
ic section. Their occurrences are shown in Table 1. In addition, because of its importance, the rich-
est diatom-bearing sample (CAS 60269-s) of five samples reported by us previously (Marincovich
and Gladenkov 1999, 2001) is included in the occurrence chart here. Its stratigraphic position
approximately coincides with that of sample CAS 608382 (about 28 m above the unconformity),
which was collected from the horizon characterized by the first appearance of mollusks Astarte in
the Milky River Formation. The Appendix gives taxonomic citations for the taxa encountered.

In general the diatom flora of the Milky River Formation consists of boreal and subarctic
species typical of high to middle latitudes (Plates 1-7; Table 1; Appendix). The North Pacific
diatom zonations of Akiba (1986), Koizumi (1992), Barron and Gladenkov (1995), and
Yanagisawa and Akiba (1998) are most applicable to this study. A version of the Cenozoic diatom
zonation proposed by Barron and Gladenkov (1995) for the North Pacific (Figure 3) is used here.
In Figure 4 the stratigraphic ranges of selected diatoms are used to correlate micropaleontological
assemblages from the Sandy Ridge section with diatom zones and subzones.

Overall, diatoms are rare to abundant and well preserved to poorly preserved throughout the

TABLE |. Stratigraphic occurrences and relative abundances of selected diatoms from the Sandy Ridge
stratigraphic section

Notes: Stratigraphic interval is in meters above the base of the Milky River Formation, which is marked
by an angular unconformity at Sandy Ridge. The location of sample CAS 60269-s coincides approximately
with that of sample CAS 608382, which was collected from the horizon containing the stratigraphically low-
est occurrence of the bivalve mollusk Astarte. Sample CAS 608372 is from the uppermost part of the Bear
Lake Formation at Sandy Ridge.

Preservation (G = good, M = moderate, P = poor) and relative abundance (A = abundant, C = common,
F = few, R = rare, VR = very rare). Plus sign (+) indicate diatom specimens recorded after the count.

A total number of valves of all diatom taxa counted in each sample are shown. However, because of space
limitations, only occurrences for stratigraphically important marine diatoms and taxa having a most impor-
tance for paleoenvironments are included in Table 1. Other taxa (primary nonmarine and benthic) were also
tabulated, however they are not treated here. Identified taxa of silicoflagellates are also not included.
Interested persons may contact the author for a complete chart of taxonomic occurrences in the Sandy Ridge
stratigraphic section.

Partial boxes within the table indicate the first or/and last occurrences of the biostratigraphically most
important species in the Milky River Formation at Sandy Ridge stratigraphic section.

3} PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 3

; b h fi palde! SIECTCOTTIAVIATHILeSePeIgQelpapal miIainireyss NS =

J é j e atOoper ian or fJoIjatuic irs S ie ; “At i! ]
Elevation above the unconformity (m) J@i<s|<{o ja BIApeir Ie (alias i[ofe|sislSlale
apa~atinin ees ee ee ee ee ee ee ee ee ee ee = ee oe

Seis

20

Actinocyclus curvatulus

A. ochotensis

A. octonarius
Actinocyclus spp.
Actinoptychus senarius

A. splendens

Adoneis pacifica
Bacteriastrum varians
Bacterosira fragilis
Cladogramma dubium
Coscinodiscus marginatus
Cosmiodiscus insignis
Cymatosira debyi
Delphineis angustata group
D. sachalinensis

D. simonsenii

D. surirella

Detonula confervacea
Dicladia capreolus
Fragilariopsis cylindrus

F. oceanica

Hyalodiscus obsoletus
Hyalopyxis concava + Trochosira spinosa
Lithodesmium minusculum
Neodenticula kamtschatica
Nitzschia rolandii
Odoniella aurita

Paralia sulcata

Porosira punctata
Pseudopyxilla americana
Pyxidicula zabelinae
Rhaphoneis angularis
Rhizosolenia hebetata group

+ 6
to
So

a
wn
ew Bmw oo +

mem MIN 1

+

ete HL Rey is)

Stephanogonia hanzawae

Thalassionema nitzschioides
Thalassiosira antiqua
Thalassiosira sp. cf. T. convexa
T. dolmatovae

. eccentrica

. gravida

. hyalina

. jacksonti

. Jouseae

. kryophila

. latimarginata

. leptopus

. manifesta

. marujamica

. nativa + T. sp. cf. nativa
. oestrupti

. orientalis

. praeoestrupti

. sheshukovae

. temperei

. tertiaria

SSSR SSRRRASRAAARAATN

. undulosa + Porosira glacialis

Thalassiothrix longissima

T. robusta 1
Resting spores of Chaetoceros 1 AS 33

: 25 Hooks |
2 17

GLADENKOV: DIATOM BIOSTRATIGRAHY IN SOUTHWESTERN ALASKA 3)3)

S| SS sail lise NPT SE | eg te S1 FG Pata) RN Pst Seal Phat | Sst] | eal pee ows tS ST a oan (gh hiss a
ca | = Joo Slolaltt[alalrtloleldlalr]Jalalta Kalam STPeIrtI TITAS
fou onw Me.2) COMI Me PFO PO RMI MN WH iWMN unit intiwtinian LINFTHofo io NHIninlojot}L,
°° bs 3 ia) ecilriletmjalta

oe) CG m |

28
2

i oe)
min
22) oO
Alia

mir MIA VMISeIATTe A mIiszeIcicliLY Ia te) La QAlroO yo im Toisas io
AIA Tats AAA eIA; eA Ia el eine izalsa wm 1 00 Sinmloicl]s cic], > =~
SISISISISISISISISISISISISISISISISISISISISISIC ISIS Sleleigislgicie
— ae i i a ee SileislslelelsislselslaleS
SIPEIEISIEIOS TET eT eI eI eIerlet~wmTeme~wTe eT erere SpeppepoeoyvepeyyefjCeyelyeijetss

N PPP IPS Ie aN Po Me G R= ge MyM
FG AR: CHE

1 M M P
MECIe ICRC R RRC R:F F R.VR R VRE -F F CEE Ce VO RCV IRI

ies le~ 2 - el + + + + + 2 4+ Ser ere eared Git Awe Galt
+ + 1 1 + + 1 + + + + + + + + + + + + + 4
Desiget 21 + + + Ce ae eee eh eatec cme Bll lpg ete sear al ee Gian Samer
] Tae Sie go, le Pape rare Dh” Afi teil ee hemes 19)
4 3 20. IMSS) sSi= sie ch) Sy = a 3} Te ee Sy al Ne As Se Oaeteee Shale eile il
+ + + 1 all< + + + + + + + +
+ + + + + +
+ + + +
Pap? 9), 920" 2) 4-7. 16 Ney 27/8) ae I Gy PA ass go IP AA0) Sel 2S 22 Snes ade 2.0)
+ + |] + + - + Nae Uaiead Linnie ad Ui 2 + + 1 + + + +
Soe lO) ey 32D): 7 86) 19 PRY abe allt} 48) ts AAO) Sh De TBS HO) PA oA Bes NT tS edhe Ss)
G@ BR Tel TUL GRA OSE AL (cy) 7h ff PI 47 Te Pan el Oy LS Me Ma ete OOP OMe OO meses
[oer 2 5 3 il AN Saas Ae ays RS es eae al besiiatope wee oa el 2
PEON 3. .3 + $21] Ae: PESTA TM Bye Wye Mle NV A Sho alll eben eh hoy ASS 2,
ee ~ + 1 + lL o+ + 4 + Lesiedgcteniartn=tsraribe lack gt ast +
+ 2 1 4 +
eos 4 82S ee 6-15 ya ie Ge shy a) ee GL SSE Ghat 4
eee eS 2 623 Dre a3) If Pile ail ae ye ee + 1 + i 3
- + + + + + + +
tie ae 2 + 1 o+ + + + Poe ae + 1 3 +
+ + + + + + + + + +
+ + + + + + + + + + + + + + +
LOM ADS lS 438712032 1 375 19) 4 6 49)-43) 25.011 MOR ON 4 BQ 48a 284i 31
i + + + + + + + + +
Pee ieaaes A 2 1 + |] ao 2) 2 + | i
+ ~ + + +
See epee SS 1 1 + OD. Mea, Week Reh ee ast mee) PR DR OR SN oi
OSs SONS h) 34 165647 41 19" 41 56° 3) 27 93 33) 323 1622 DAO O TAD AS Gly 295 37/32 252 eal OPS
~ + + + + + + +
+ + + > + + + + + | ener te ee + + +
(a eee OOe Fi) Sp U2 25) 27 255 28). sy fb Sh SyaualS) Sh US a My eS A Ge Oe Ae:
ar ele 2 eee 3) 5S) 2 Deis eel 1 A By OE Wee SBS PE ee gye idle @ 6
pial 2. + + + + + + + + +
- + + + + + + + + +
ASE 2O0 SO e218) 31) 1s 21) wilé: <4 il Qi 2 2AS 6: 2a Om2T a VAM Silane DOR 28 isha:
Sao te 4d p) 3 2 | + 5 + + + + +
+ + + + See
Pe Omeige Oe 3). 3) 14 2.98 aL al 3 By AL ey Sh al ANOS ey OP IG Ogee 16
3. Bi Gy al a ella ah [Ser Seat aD AD oa UO SY ret epereran tect aS) 3
2 + | 2 + Deets tll Bench weit hewalte weal dil a gaat Niece PR 2 I
i 2 + + + + + + + +
+ + ] + + + + 3
(PA Tp ay AI 8} 8 65 Cae Sito te eas Oa ean Uist eS ae RR.
- + 2 + +

wrwet + t+

SF Se SO eae eg SL tet a Li MS as 4) er
aie Ml Dae 4h, se. 12) 33 7B) ssp G5) 5 24 31

2)
bod somjisapon

() Shs} NSS 27) Ds)

Bn 0 BT BT

i)
_
ios)
BE
N
oo
oo
—
&
—
i)

34 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 3

Milky River Formation at Sandy Ridge. The first occurrences of Thalassiosira oestrupii, T. anti-
qua, T. jouseae, T. marujamica, T. sheshukovae, Neodenticula kamtschatica, Bacterosira fragilis,
Detonula confervacea and Rhaphoneis angularis are in sample CAS 617358 from the basal part of
the Milky River Formation (0.9 m above the base). Thalassiosira oestrupii ranges to the top of the
section in sample CAS 608412 (263.0 m). The last occurrences of Neodenticula kamtschatica and
Cosmiodiscus insignis are below the stratigraphic top, in samples CAS 608410 (246.0 m) and CAS
608408 (222.2 m), respectively. The last consistent occurrence of Cosmiodiscus insignis is in sam-
ple CAS 608405 (172.8 m). Lithodesmium minusculum is present in the interval from sample CAS
608376 (3.4 m) to sample CAS 608401 (139.9 m). Delphines simonsenii occurs from sample CAS
608379 (8.8 m) to sample CAS 608392 (85.3 m), and Nitzschia rolandii from sample CAS 608382
(28.0 m) to sample 608425 (93.0 m). Thalassiosira jacksonii and Thalassiosira sp. cf. T. convexa
range from sample CAS 608376 (3.4 m) to sample CAS 608425 (93.0 m), including sample CAS
608393 (88.4 m). Thalassiosira temperei is present in samples CAS 608378 (5.5 m) and CAS
608382 (28.0 m), while sample CAS 60269-s (about 28.0 m) contains Thalassiosira praeoestrupii.
Samples CAS 608426 (95.1 m) and CAS 608408 (222.2 m) yield Thalassiosira latimarginata,
while Thalassiosira tertiaria was found in samples CAS 608426 and CAS 608400 (139.9 m).

AGE OF THE DIATOM FLORA.— As noted above, determining a precise age for the poorly pre-
served assemblage in sample CAS 608372, from the uppermost part of the Bear Lake at Sandy
Ridge, has proven difficult owing to the absence of co-occurring, biostratigraphically significant
taxa characteristic of any North Pacific diatom zone. The co-occurrence of Thalassiosira oestrupii,
Cosmiodiscus insignis and Neodenticula kamtschatica in the Milky River Formation allows recog-
nition of the latest Miocene to early Pliocene subzone b of the North Pacific Neodenticula
kamtschatica Zone of Barron and Gladenkov (1995) in the Sandy Ridge stratigraphic section. The
Neodenticula kamtschatica Zone is the interval from the first occurrence (FO) of N. kamtschatica
to the FO of N. koizumii (Figure 3). This zone is divided into subzones a, b, and c by the FO of
Thalassiosira oestrupii and the last occurrence (LO) of Cosmiodiscus insignis, following Barron
(1980). Thus, based on the find of Cosmiodiscus insignis in sample CAS 608408 (222.2 m), most
of the Milky River Formation at Sandy Ridge section can be correlated with subzone b of the N.
kamtschatica Zone. The boundaries of the N. kamtschatica Zone and its subzones, based on diatom
datum levels, are calibrated with the magnetostratigraphic record in the North Pacific (Barron and
Gladenkov 1995). According to the geochronologic and chronostratigraphic scale of Berggren et
al. (1995), Subzone b has an age range of 5.5 to 4.8 Ma. The Miocene/Pliocene boundary is cur-
rently placed at 5.32 Ma by Berggren et al. (1995). Diatoms from the uppermost part of the Milky
River section, above sample CAS 608405 (172.8 m), however, are rare and generally exhibit poor
preservation. So, it is uncertain whether Cosmiodiscus insignis has its true last occurrence in sam-
ple CAS 608408 (222.2 m), which is its final occurrence, or in sample CAS 608405 (172.8 m),
which is its last consistent occurrence but where this species is represented by a single, possibly
reworked specimen. On the other hand, the absence of Cosmiodiscus insignis above 222.2 m due
to paleoecological changes cannot be excluded. Therefore, a tentative upper boundary of subzone
b, marked by the LO of Cosmiodiscus insignis at level 172.8 m, is proposed for the Sandy Ridge
section.

The occurrence of additional biostratigraphically important diatoms allows refining the age for
the different parts of the stratigraphic section. Not all horizons in this section are characterized by
the consistent occurrence of diatoms, including age-diagnostic marine taxa, but analysis of the
species still is valuable. The most important species for biochronology in the lower part of Milky
River Formation is 7halassiosira temperei (Figure 4). The last occurrence of this species is a use-
ful Neogene biohorizon marker in the North Pacific, including onshore sequences in Japan and

GLADENKOV: DIATOM BIOSTRATIGRAHY IN SOUTHWESTERN ALASKA 35

iat
Sea
@/e|n9z/z DIATOM
é| 2 lls o <
2 os =
v 5 >| a | © ZUNE
< Sena CO); &
= eee Gi Ee LO P curvirostris

| ann oa Neodenticula seminae
; Qo fa Tc as 7m
=| A, oculatus
Actinocyclus oculatus

|
|
? 3 | fee aoe a
= {0 Foal Neodenticula koizumii LON. koizumii
Pre Baz Neodenticula koizumii -
= | C2A Sees Veodenticula kamtschatica KOU, Roiainn
4 las ee LCO N. Kamtschatica
5 = 4 (0S mum = Veodenticula b LO C. insigsnis
6 ae mae Kaimtschatica FO T oestrupii
= a ie
| {
Ti i e__ ;
; ee | GB FO N. kamitschatica ae
fad | : : LCO D. hustedtii
3 — im C4 Thalassionema schraderi
5) = sil CA Denticulopsis katayamae LO D. dimorpha
ea [sees Denticulopsis dimorpha
OR PSs aa ;
7. spe : FO D. dimorpha
‘om (3 Thalassiosira yabei
. = a tect LCO D. praedimorpha
1] Se) = (ae Denticulopsis praedimorpha
a Z. = | GA FO D. praedimorpha
2 sy |e CSA Crucidenticula nicobarica
oO = S| CAB FCO D. hustedtii
< CC : E : . hustedtii
14 — ea U = ——— Denticulopsis hyalina FO D. hyalina
eae ‘
NS aoe = | P ;
Z. = CSB Denticulopsis lauta
FO D. lauta
167 > al Denticulopsis praelauta FO D. praelauta
| (5C mmm Crucidenticula kana
17 | |
oer ag. mes F FO C. kanayae
csp | Crucidenticula sawamurae
18 — ieee
| CSE
19> on ae net FO C. sawamurae
7 | Thalassiosira fraga
i zo
efi aul
t | as
aA i | C= FO T fraga
5 r a eee
22 (6AA ee = Thalassiosira praefraga
| ——
BI t=
ae
I = See ae =
vA : TR eee FO T. praefraga
Sc | Rocella gelida
2 aaa a

FIGURE 3. The North Pacific Neogene diatom zonation after Barron and Gladenkov (1995) correlated to the geomag-
netic polarity time scale and chronostratigraphic subdivisions of Berggren et al. (1995) (see text for additional explanation).

Abbreviations used: FO, first occurrence; LO, last occurrence; FCO, first common occurrence; LCO, last common
occurrence; a-c, subzones; Q, Quaternary; P, Paleogene; O, Oligocene; L, late.

PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 3

ime recorded
in the MRF

T

INUNZ0Y DJHIYUapoany

SnyD]nIO snjr{IOUNIP

DIIUAO JVI DIXHOY

SIJDJHBUIS DAISOISSDIDY J,

sinua] bagay]

absent in SRS

DIVIY4A] DAISOISSDIDY J,
at

DIDULBADUND] DALSOISSDP_Y<§_ —A-A A A A @_

SRS

SILSVAL DHSOLAIDE,

ain

avasuol DAISOISSDIDY J,

uuasuouns siau1ydjag

DawAdafuod DjNUOJag

udnajsa0avad DAISOISSDIDY J,

ficant tax

signi

(CW ) pxaauod DAISOISSD]DY J

MUOSYIDE DAISOISSD]D Y J,

DINUDIMADUL DAISOISSD]DY J

DINDYISHUDY VDINIUapoany

DUDIYUD VAISOISSVDIDY |

Pe

—

—

SS

=)
a

=

lam!

= ee
i} ~ s . co =
~ SLIDJNSuv siauoydvyy ae
of ; a
= avuljagVz D]NIIpIXdd =

= =

= MIN]NISNUNY UNNUSAPOYNT =

—

wD é

S SIUSISUI SHISIPOMUISOD Ta
— 2 = Asa)
jaa) MPUDdIOL DIY IST a

Jadaduia] DAISOISSD]D Y J a

ZumMil -

kamtschatica

Neodenticula
Thalassionema schraderi

Neodenticula seminae

Actinocyclus oculatus
Neodenticula kamtschatica

Neodenticula koizumii

Proboscia curvirostris

Neodenticula koi

HOOdaAAS aL eae ATUVA

HOOda ANAQOIN
(I) ADV _ L Leiter Bt BI i do wee

“
S

GLADENKOV: DIATOM BIOSTRATIGRAHY IN SOUTHWESTERN ALASKA 3)//

Kamchatka. It occurs within the interval corresponding to the lowermost part of subzone b of the
Neodenticula kamtschatica diatom Zone (Barron 1980, 1992a; Akiba et al. 1982; Oreshkina 1985;
Akiba 1986; Akiba and Yanagisawa 1986; Yanagisawa 1990, 1998). The most recent data, based
on extrapolated sedimentation rates and occurrences in deep-sea cores, indicates that Thalassiosira
temperei had its last occurrence at 5.4 Ma in the middle- to high-latitude North Pacific (Yanagisawa
and Akiba 1998) (Figure 4). Therefore, the co-occurrence of Thalassiosira temperei, T: oestrupii,
and Cosmiodiscus insignis indicates a narrow age interval of 5.5 to 5.4 Ma for the lower part of the
Milky River Formation at Sandy Ridge.

The presence of secondary marker species, including Delphines simonsenii, Bacterosira frag-
ilis, Detonula confervacea, and Thalassiosira jouseae, are useful for evaluating the lower age limit
of the Milky River Formation at Sandy Ridge. According to Akiba (1986), the first occurrences of
all of these species at DSDP Site 584 off northeastern Japan, are within the interval corresponding
to subzone b of the Neodenticula kamtschatica Zone of Barron and Gladenkov (1995). More pre-
cisely, these species first occur within a narrow interval just above the FO of Thalassiosira
oestrupii. The total occurrence of Delphines simonsenii is restricted to subzone b at DSDP Site 584
(Akiba 1986). These datum levels have not been directly correlated to magnetostratigraphy, but the
presence of these species in the Sandy Ridge section supports a lower limit for the Milky River
Formation of not more than 5.5 Ma.

As shown in Table 1, Rhaphoneis angularis is typical of diatom assemblages from the Milky
River Formation. In the Harris Grade section, California, this species ranges from between the first
occurrences of Thalassiosira oestrupii and Thalassiosira praeoestrupii, i.e. from about 5.3-5.4 Ma
(Barron and Baldauf 1986; Dumont and Barron 1995). On the other hand, in sequences from the
Oregon continental shelf Rhaphoneis angularis ranges from an older interval corresponding to the
uppermost part of the Thalassionema schraderi Zone (Whiting and Schrader 1985). So, the datum
from California may not apply in more northern regions such as Alaska.

In summary, diatoms imply an age range for lower 28 m of the Milky River Formation, below
sample CAS 608382 (which has the last co-occurrence of Thalassiosira temperei, T. oestrupii,
Neodenticula kamtschatica, and Cosmiodiscus insignis) of 5.5—5.4 Ma.

The most biochronologically important species in the upper part of Milky River Formation are
Thalassiosira latimarginata and Thalassiosira tertiaria. As noted above, T: latimarginata 1s pres-
ent in samples CAS 608426 (95.1 m) and CAS 608408 (222.2 m), and T: tertiaria in samples CAS
608426 and CAS 608400 (139.9 m). The FO of 7: latimarginata has been calibrated with the mag-
netostratigraphic record of deep-sea cores in the subarctic North Pacific at about 5.1 Ma (Barron
and Gladenkov 1995). The FO of T: tertiaria has been correlated with magnetostratigraphy in ODP

FiGureE 4. Published biostratigraphic ranges of selected diatom taxa present in the Sandy Ridge stratigraphic section
(SRS), indicating a latest Miocene to early Pliocene age for the Milky River Formation (MRF).

Black triangle in parentheses indicates that Thalassiosira sp. cf. T: convexa is present in the Milky River Formation (see
Appendix).

Key to published ranges: |. Barron and Gladenkov (1995*), ODP sites in the subarctic North Pacific. 2. Yanagisawa
and Akiba (1998), DSDP Hole 438A, off northeastern Japan. 3. Barron (1980), DSDP sites, off northeastern Japan. 4.
Barron (1992a*), DSDP and ODP sites in the North Pacific. 5. Barron (1992b), California. 6. Oreshkina (1985; personal
communication, 1998), DSDP Site 192, off northeastern Kamchatka. 7. Barron and Baldauf (1986), California. 8. Dumont
and Barron (1995), California. 9. Whiting and Schrader (1985), the Oregon continental shelf. 10. Schrader (1973), DSDP
sites in the Gulf of Alaska. 11. Akiba (1986), DSDP Site 584, off northeastern Japan. 12. Koizumi (1992*), ODP sites in
the Japan Sea. 13. Koizumi and Tanimura (1985*), DSDP sites in the northwestern Pacific. Asterisks (*) indicate a paleo-
magnetic calibration.

The North Pacific diatom zonation of Barron and Gladenkov (1995) for the last 8 myr correlated to the geomagnetic
polarity time scale and chronostratigraphic subdivisions of Berggren et al. (1995).

Abbreviations used: Q, Quaternary; a-c, subzones; DSDP, Deep Sea Drilling Project; ODP, Ocean Drilling Program.

38 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 3

Site 884 in the western part of the subarctic North Pacific, where it has an age of 5.0 Ma (Barron
and Gladenkov 1995). Thus, the presence of these species in the Milky River Formation indicates
an age not older than 5.1-5.0 Ma for the horizons overlying the 95.1-m level.

Lithodesmium minusculum (occurring from 3.4 m to 139.9 m) could have served as an addi-
tional useful age marker for the upper portion of the section studied. The age range of the last
occurrence of this species in California is 4.6-4.8 Ma (Barron 1992b; Dumont and Barron 1995).
However, the presence of Lithodesmium minusculum is documented in younger horizons of the
Neodenticula koizumii-Neodenticula kamtschatica Zone at DSDP Site 192, off northeastern
Kamchatka (T. Oreshkina 1985, personal communication 1998). These data suggest that the datum
from California may not apply in more northerly regions, such as Alaska.

COMPARISON WITH THE DIATOM FLORA OF THE PRIBILOF ISLANDS.— As noted in the
Introduction, a relatively well-preserved and diverse fossil marine diatom flora from an onshore
sequence in Alaska has previously been documented and studied only from the Pribilof Islands,
Bering Sea (Hanna 1919, 1970). Overall, this diatom assemblage is comprised of relatively long-
ranging forms with little or no precise age significance. In contrast to the flora from the Sandy
Ridge section, this diatom assemblage lacks most of the biostratigraphically-important taxa typical
of the Milky River Formation, such as Neodenticula kamtschatica, Thalassiosira oestrupii, T. lati-
marginata, Bacterosira fragilis, Detonula confervacea, Thalassiosira jouseae, and Delphineis
simonsenii. However, both floras have a few species in common, including Cosmiodiscus insignis
and Thalassiosira temperei (referred to as Cymatotheca weissflogii by Hanna 1970). These latter
species suggest that the flora from the Pribilof Islands is older than the one at Sandy Ridge and has
an age older than 5.5 Ma. This conclusion is supported by data from L. Burckle and N. Opdyke,
who studied diatoms and magnetostratigraphy from sequences in the Pribilof Islands (Burckle and
Opdyke, unpublished data). Diatom-bearing sediments exposed in outcrops on St. Paul Island
exhibit the normal polarity of Chron C3A (Burckle, personal communication 2002), which ranges
in age from 6.57 Ma to 5.89 Ma (Berggren et al. 1995). This age range implies that the diatom flora
from of the Pribilof Islands is possibly assignable to the older part of the North Pacific
Neodenticula kamtschatica diatom Zone, which is Subzone a that has an age range of 7.4—-5.5 Ma.
The absence of the zonal species Neodenticula kamtschatica in the Pribilof Islands flora is possi-
bly due to paleoecological exclusion. However, the absence of Neodenticula kamtschatica due to
the possible difference of preparation methods, including the procedures of processing of samples,
between Hanna (1970) and this study also cannot be excluded.

DIATOM PALEOENVIRONMENTS.— In general, the diatom flora from the Milky River
Formation is composed by boreal and subarctic species typical of high to middle latitudes. Overall,
the assemblages are dominated by marine neritic-planktic and sublittoral taxa (including Paralia
sulcata, Chaetoceros spores and Delphineis spp.) that are accompanied by benthic and nonmarine
forms, which implies deposition in productive shallow water (shallower than 100 m) of the Alaskan
continental shelf (Sancetta 1981la-b; Sancetta and Silvestri 1986). The persistence and occasional-
ly common occurrence of Thalassionema nitzschioides accompanied by oceanic planktic elements
suggests incursions of transitional water from the outer shelf zone, where shelf and oceanic waters
are incompletely mixed. The presence of Fragilariopsis oceanica and F. cylindrus, which com-
monly live near sea ice, as well as some marine arctoboreal species (Bacterosira fragilis, Detonula
confervacea, Porosira glacialis, Thalassiosira kryophila, T. hyalina, T. gravida) imply periods of
bloom associated with melting ice near the continental margin of southwestern Alaska. Sparse
occurrences of subtropical or transitional planktonic taxa such as Thalassiosira leptopus, T. linea-
ta, T. temperei, T. oestrupii, Bacteriastrum varians, Auliscus and Azpeitia spp. suggest possible
episodes of relatively warmer conditions.

GLADENKOV: DIATOM BIOSTRATIGRAHY IN SOUTHWESTERN ALASKA 39

SUMMARY

The 263-m-thick stratigraphic section of the Milky River Formation at Sandy Ridge contains
Neogene marine diatom assemblages of different preservation and abundance. A total of about 100
species and varieties of diatoms were identified in the samples examined. The presence of the
marker species Neodenticula kamtschatica, Cosmiodiscus insignis, and Thalassiosira oestrupii
throughout the section allows recognition of the latest Miocene to early Pliocene subzone b of the
Neodenticula kamtschatica Zone of Barron and Gladenkov (1995), which has an age range of 5.5
to 4.8 Ma. Additional age-diagnostic diatoms, including Thalassiosira temperei and T: latimargina-
ta, allow refining the age for the basal part of the Milky River Formation to 5.5 Ma to 5.4 Ma, and
to 5.1 Ma to 4.8 Ma for the upper part of the formation at Sandy Ridge. This information, from
stratigraphically well-controlled samples, is the basis for dating mollusks that occur directly with
the diatoms, including the biogeographically important bivalve Astarte. Diatoms in the stratigraph-
ically lowest Astarte-bearing horizon at Sandy Ridge have an age range of 5.5—5.4 Ma, which is
the basis for dating the earliest opening of Bering Strait (Marincovich and Gladenkov 1999, 2001;
Marincovich et al. 2002; Gladenkov et al. 2002).

Both the dominant and common elements of the Milky River Formation flora imply deposi-
tion on a productive continental shelf, with some incursions of transitional waters from the outer
shelf zone. The presence of species living with sea ice, along with marine arctoboreal species, sug-
gests periods of bloom associated with the melting of ice near the continental margin of southwest-
ern Alaska. Sparse occurrences of subtropical or transitional planktonic taxa suggest transient
warmer conditions.

ACKNOWLEDGMENTS

This study was supported by U.S. National Science Foundation grant OPP 9806461 to Louie
Marincovich, Jr. (California Academy of Sciences). I am grateful to John Barron (U.S. Geological
Survey) and Louie Marincovich, Jr., for stimulating and helpful discussions in the course of this
research. I am also indebted to L. Marincovich, Jr., for sharing his knowledge of geological work
on the Sandy Ridge stratigraphic section and for reviewing an early draft of this manuscript. I thank
Lloyd Burckle (Lamont-Doherty Earth Observatory of Columbia University) for sharing his data
on diatom-bearing sequences of the Pribilof Islands. The consultation, technical assistance and
kind help of Elisabeth Fourtanier, J. Patrick Kociolek, Jean DeMouthe, Elisabeth Ruck, and other
colleagues from the Diatom Laboratory, California Academy of Sciences, are greatly appreciated.
I also thank two anonymous reviewers for their critical reviews of the manuscript and helpful com-
ments.

LITERATURE CITED

AKIBA, F. 1986. Middle Miocene to Quaternary diatom biostratigraphy in the Nankai Trough and Japan
Trench, and modified lower Miocene through Quaternary diatom zones for the middle-to-high latitudes of
the North Pacific. Pages 393-481 in H. Kagami, D.E. Karig, W.T. Coulbourn, et al., eds., Initial Reports,
Deep Sea Drilling Project 87. U.S. Govt. Printing Office, Washington, D.C.

AKIBA, F. AND Y. YANAGISAWA. 1986. Taxonomy, morphology and phylogeny of the Neogene diatom zonal
marker species in the middle-to-high latitudes of the North Pacific. Pages 483-554 in H. Kagami, D.E.
Karig, W.T. Coulbourn, et al., eds., Initial Reports, Deep Sea Drilling Project 87. U.S. Govt. Printing
Office, Washington, D.C.

AKIBA, F., Y. YANAGISAWA Y., AND T. Isuit. 1982. Neogene diatom biostratigraphy of the Matsushima area and
its environs, Miyagi Prefecture, Northeast Japan. Bulletin of the Geological Survey of Japan 33
(5):215—239. (in Japanese with English abstract)

ANDREWS, G.W. 1977. Morphology and stratigraphic significance of Delphineis, a new marine diatom genus.

40 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 3

Beihefte zur Nova Hedwigia 54: 243-260.

ANDREWS, G.W. 1980. Revision of the diatom genus Delphineis and morphology of Delphineis surirella
(Ehrenberg) G.W. Andrews, n. comb. Pages 81—92 in R. Ross, ed., Proceedings of the 6th International
Diatom Symposium. Koeltz Scientific Books, Koenigstein.

ANpDREWS, G.W. 1988. Evolutionary trends in the marine diatom genus Delphineis GW. Andrews. Pages
197-206 in FE. Round, ed., Proceedings of the 9th International Diatom Symposium. Biopress, Bristol &
Koeltz Scientific Books, Koenigstein.

ANDREWS, G.W. AND P. RIvgERA. 1987. Morphology and evolutionary significance of Adoneis pacifica gen. et
sp. nov. (Fragilariaceae, Bacillariophyta), a marine araphid diatom from Chile. Diatom Research 2:\—14.

BARRON, J.A. 1975. Late Miocene-early Pliocene marine diatoms from southern California.
Palaeontographica 151(B):97-170, pls. 1-15.

BARRON, J.A. 1980. Lower Miocene to Quaternary diatom Bostannapny of Leg 57, off Northeastern Japan,
Deep Sea Drilling Project. Pages 641-685 in E. Honza et al., eds., Initial Reports, Deep Sea Drilling
Project 56, 57. U.S. Govt. Printing Office, Washington, D.C.

BARRON, J.A. 1985. Miocene to Holocene planktic diatoms. Pages 641-691 in H.M. Bolli, J.B. Saunders, and
K. Perch-Nielsen, eds., Plankton Stratigraphy. Cambridge University Press, Cambridge, UK.

BARRON, J.A. 1992a. Neogene diatom datum levels in the Equatorial and North Pacific. Pages 413-425 in K.
Ishizaki and T. Saito, eds., Centenary of Japanese Micropaleontology. Terra Scientific Publishing
Company, Tokyo.

Barron, J.A. 1992b. Paleoceanographic and tectonic controls on the Pliocene diatom record of California.
Pages 25—41 in R. Tsuchi and J.C. Ingle, Jr., eds., Pacific Neogene — Environment, Evolution, and Events.
University of Tokyo Press, Tokyo.

BARRON, J.A. AND J.G. BALDAuF. 1986. Diatom stratigraphy of the lower Pliocene part of the Sisquoc
Formation, Harris Grade section, California. Micropaleontology 32:357-371.

BARRON, J.A. AND A.Y. GLADENKOV. 1995. Early Miocene to Pleistocene diatom stratigraphy of Leg 145.
Pages 3-19 in D.K. Rea, I. A. Basov, D. W Scholl, and J. F. Allan, eds., Proceedings of the Ocean Drilling
Program, Scientific Results 145. College Station, TX.

BARRON, J.A. AND A.D. MAHOoop. 1993. Exceptionally well-preserved early Oligocene diatoms from glacial
sediments of Prydz Bay, East Antarctica. Micropaleontology 39:29-4S5.

BERGGREN, W.A., D.V. KENT, C.C. SWISHER II, AND M.-P. AuBRy. 1995. A revised Cenozoic geochronology
and chronostratigraphy. SEPM Special Publication 54:129-212.

BopbeNn, P. 1993. Taxonomy and stratigraphic occurrence of Thalassiosira tetraoestrupii sp. nov. and related
species in upper Miocene and lower Pliocene sediments from the Norwegian Sea, North Atlantic and North
West Pacific. Terra Nova 5:61-—75.

DETTERMAN, R.L., J.E. CASE, J.W. MILLER, F.H. WILSON, AND M.E. YOUNT. 1996. Stratigraphic framework of
the Alaska Peninsula. U.S. Geological Survey Bulletin 1969-A. 74 pp.

DONAHUE, J.G. 1970. Pleistocene diatoms as climatic indicators in North Pacific sediments. Geological
Society of America Memoir 126:121-138.

Dumont, M.P., J.G. BALDAUF, AND J.A. BARRON. 1986. Thalassiosira praeoestrupii — a new diatom species
for recognizing the Miocene/Pliocene Epoch boundary in coastal California. Micropaleontology
32:372-377.

Dumont, M.P., AND J.A. BARRON. 1995. Diatom biochronology of the Sisquoc Formation in the Santa Maria
Basin, California, and its paleoceanographic and tectonic implication. U.S. Geological Survey Bulletin
1995: KI-K17.

DZINORIDZE, R.N. AND I.V. MAKAROVA. 1988. Genera Porosira J@rg., Bacterosira Gran, Detonula Schiitt,
Lauderia Cl. Pages 86-89, pls. 59-60 in S.I. Gleser, I.V. Makarova, A.I. Moisseeva, and V.A. Nikolaev,
eds., The Diatoms of the USSR (Fossil and Recent) I1(1). Nauka Publishers, Leningrad. (in Russian)

FRYXELL, GA. AND G.R. HASLE. 1979. The genus Thalassiosira: T. trifulta sp. nova and other species with tri-
columnar supports on strutted processes. Beihefte zur Nova Hedwigia 64:13—32, 7 pls.

GALLOwAay, W.E. 1974. Deposition and diagenetic alternation of sandstone in northeast Pacific arc-related
basins — Implication for graywacke genesis. Geological Society of America Bulletin 83:379-390.

GLADENKOoV, A.Y. AND L. MARINCOVICH, JR. 1998. New data on the first opening of Bering Strait, based on

GLADENKOV: DIATOM BIOSTRATIGRAHY IN SOUTHWESTERN ALASKA 4]

finds of Neogene marine diatoms in the Bear Lake Formation, SW Alaska. Abstracts 6th International
Conference on Paleoceanography, Lisbon, Portugal, p. 115.

GLADENKOV, A.Y. AND L. MARINCOVICH, JR. 1999. The earliest age of the Bering Strait’s first opening in the
Neogene: paleontological evidence. European Union of Geosciences, March 28th-April Ist, 1999,
Strasbourg, France. Journal of Conference Abstracts 4(1):179-180.

GLADENKOV, A.Y., A-E. OLEINIK, L. MARINCOVICH, JR., AND K.B. BARINOV. 2002. A refined age for the earli-
est opening of Bering Strait. Palaeogeography, Palaeoclimatology, Palaeoecology 183:321—328.

GLADENKOV, Y.B., K.B. BARINOV, A.E. BASILIAN, S.I. BORDUNOV, G.M. BRATSEVA, E.V. ZYRYANOV, N.P.
KURALENKO, D.I. VITUKHIN, T.V. ORESHKINA, S.S. GANZEI, S.I. KTYASHKO, AND V.M. TRUBIKHIN. 1992.
Detailed Division of the Neogene of Kamchatka. Moscow, Nauka Publishers. 208 pp. (in Russian)

Hanna, G D. 1919. Geological notes on the Pribilof Islands, Alaska, with an account of the fossil diatoms.
American Journal of Science 48:216—224.

Hanna, G D. 1970. Fossil diatoms from the Pribilof Islands, Bering Sea, Alaska. Proceedings of the
California Academy of Sciences, ser. 4, 37(5):167—234.

Harwoop, D.M. AND T. MARUYAMA. 1992. Middle Eocene to Pleistocene diatom biostratigraphy of Southern
Ocean sediments from the Kerguelen Plateau, Leg 120. Pages 683-733 in S. W. Wise, Jr., R. Schlich, et
al., eds., Proceedings of the Ocean Drilling Program, Scientific Results 120. College Station, TX.

HASLE, GR. 1993. Nomenclatural notes on marine planktonic diatoms. The family Bacillariaceae. Beihefte zur
Nova Hedwigia 106:315-321.

HASLe, G.R., P.A. SIMS, AND E.E. SYVERTSEN. 1988. Two recent Stellarima species: S$. microtrias and S. stel-
laris (Bacillariophyceae). Botanica Marina 31:195—206.

JORDAN, R.W., R. LiGOwskKI, E.-M. NOTHIG, AND J. PRIDDLE. 1991. The diatom genus Proboscia in Antarctic
waters. Diatom Research 6:63—78.

JORDAN, R.W. AND J. PRIDDLE. 1991. Fossil members of the diatom genus Proboscia. Diatom Research 6:
55-61.

JouseE, A.P. 1961. Miocene and Pliocene marine diatoms from the Far East. Botanicheskie Matererialy Otdela
Sporovykh Rastenii, Botanicheskii Institute, Akad. Nauk SSSR 14:59-70, pls. 1-3. Akad. Nauk SSSR
Publishers, Moscow-Leningrad. (in Russian)

JousE, A.P. 1968. New species of diatoms from bottom sediments of the Pacific and the Okhotsk Sea.
Novitates Systematicae Plantarum Non Vascularum:\12—21. Nauka Publishers, Leningrad. (in Russian)
JOUSE, A.P., V.V. MUKHINA, AND O.G. KozLova. 1969. Diatoms and silicoflagellates in the surface layer of sed-
iments from the Pacific Ocean. Pages 7-47 in P. L. Bezrukov, ed., The Pacific Ocean. Vol. 7: Microflora
and microfauna in the modern sediments of the Pacific Ocean. Nauka Publishers, Moscow. (in Russian)

KANnayA, T. 1959. Miocene diatom assemblages from the Onnagawa Formation and their distribution in the
correlative formations in Northeast Japan. Science Reports of the Tohoku University, Second Ser. (Geol.)
30:1—130, 11 pls.

KANAYA, T. AND I. Koizumi. 1966. Interpretation of diatom thanatocoenoses from the North Pacific applied to
a study of core V20—130 (Studies of a deep-sea core V20—130. Part IV). Science Reports of the Tohoku
University, Second Ser. (Geol.) 37 (2): 89-130.

Koizumt, I. 1973. The late Cenozoic diatoms of Sites 183-193, Leg 19, Deep Sea Drilling Project. Pages
805-855 in J.S. Creager, D.W. Scholl, et al., eds., Initial Reports, Deep Sea Drilling Project 19. U.S. Govt.
Printing Office, Washington, D.C.

Koizumi, I. 1975. Neogene diatoms from the western margin of the Pacific Ocean, Leg 31, Deep Sea Drilling
Project. Pages 779-819 in D-E. Karig, J.C. Ingle, Jr., et al., eds., Initial Reports, Deep Sea Drilling Project
31. U.S. Govt. Printing Office, , Washington, D.C.

Koizumi, I. 1980. Neogene diatoms from the Emperor Seamount Chain, Leg 55, Deep Sea Drilling Project.
Pages 387-407 in E.D. Jackson, I. Koizumi, et al., eds., Initial Reports, Deep Sea Drilling Project 55. U.S.
Govt. Printing Office, Washington, D.C.

KoizuMI, I. 1992. Diatom biostratigraphy of the Japan Sea: Leg 127. Pages 249-289 in K. Pisciotto, J.C. Ingle,
Jr, M. von Breymann, and J.A. Barron, eds., Proceedings of the Ocean Drilling Program, Scientific
Results 127/128 (Pt. 1). College Station, TX.

Koizumi, I. AND Y. TANIMURA. 1985. Neogene diatom biostratigraphy of the middle latitude western North

42 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 3

Pacific, Deep Sea Drilling Project Leg 86. Pages 269-300 in GR. Heath, L.H. Burckle, et al., eds.; Initial
Reports, Deep Sea Drilling Project 86. U.S. Govt. Printing Office, Washington, D.C.

LOHMAN, K.E. 1938. Pliocene diatoms from the Kettleman Hills, California. U.S. Geological Survey
Professional Paper (189-C):81—102, pls. 20-23.

MAKAROVA, I.V. 1988a. New combinations of taxa from the late Miocene-early Pliocene diatom floras of
Sakhalin. Botanicheskii Zhurnal 73(8):1183—1186, pls. 1-2. (in Russian)

Makarova, L.V. 1988b. Diatoms of the USSR seas: genus Thalassiosira Cl. Nauka Publishers, Leningrad. 117
pp. (in Russian)

MakKaArova, I.V. 1988c. Genus Thalassiosira Cl. Pages 58-82, pls. 37-57 in S. I. Gleser, I.V. Makarova, A.I.
Moisseeva, and V.A. Nikolaev, eds., The Diatoms of the USSR (Fossil and Recent) I1(1). Nauka Publishers,
Leningrad. (in Russian) é

MAKAROVA, I.V. 1989. New genus of centric diatoms. Novitates Systematicae Plantarum Non Vascularum
26:34-35, pls. 1-2. Nauka Publishers, Leningrad. (in Russian)

MakArova, I.V. AND A.I. MOISSEEVA. 1986. New species of genus Pyxidicula (Bacillariophyta). Botanicheskii
Zhurnal 71(2):244—245, pls. 1-2. (in Russian)

Marincovich, L., Jr. 2000. Central American paleoceanography controlled Pliocene Arctic molluscan migra-
tion. Geology 28:551—554.

MARINCOVICH, L., JR., K.B. BARINOV, AND A.E. OLEINIK. 2002. The Astarte (Bivalvia: Astartidae) that docu-
ment the earliest opening of Bering Strait. Journal of Paleontology 76:239-245.

MARINCOVICH, L., JR. AND A.Y. GLADENKOV. 1997. New paleontological information about the first opening
of Bering Strait. Program and Abstracts, Beringian Paleoenvironments Workshop, Florissant, Colorado,
pp. 103-104.

MARINCOVICH, L., JR. AND A. Y. GLADENKOV. 1999. Evidence for an early opening of the Bering Strait. Nature
397:149-151.

MARINCOVICH, L., JR. AND A.Y. GLADENKOV. 2001. New evidence for the age of Bering Strait. Quaternary
Science Reviews 20:329-335.

MukuiNa, V.V. 1965. New species of diatoms from bottom sediments of the Equatorial Pacific. Novitates
Systematicae Plantarum Non Vascularum: 22—25. Nauka Publishers, Moscow-Leningrad. (in Russian)
ORESHKINA, T.V. 1985. Diatom assemblages and stratigraphy of the upper Cenozoic off Kamchatka region of

the Pacific Ocean. Izvestiya Akademii Nauk SSSR, Geology Series 5:60—73. (in Russian)

PERCH-NIELSEN, K. 1985. Silicoflagellates. Pages 811-846 in H.M. Bolli, J.B. Saunders, and K. Perch-
Nielsen, eds., Plankton Stratigraphy. Cambridge University Press, Cambridge, UK.

PLAFKER, G. AND W.O. AppicoTT. 1976. Glaciomarine deposits of Miocene throught Holocene age in the
Yakataga Formation along the Gulf of Alaska margin, Alaska. Pages Q1—Q22 in T.P. Miller, ed., Recent
and Ancient Sedimentary Environments in Alaska. Alaska Geological Society Symposium Proceedings.

SANCETTA, C. 198la. Diatoms as hydrographic tracers: example from Bering Sea sediments. Science
211:279-281.

SANCETTA, C. 1981b. Oceanographic and ecologic significance of diatoms in surface sediments of the Bering
and Okhotsk seas. Deep Sea Research 28A(8):789-817.

SANCETTA, C. 1982. Distribution of diatom species in surface sediments of the Bering and Okhotsk seas.
Micropaleontology 28:221—257.

SANCETTA C. AND S. SILVESTRI. 1986. Pliocene-Pleistocene evolution of the North Pacific Ocean-atmosphere
system, interpreted from fossil diatoms. Paleoceanography 1:163-180.

SCHRADER, H.-J. 1973. Cenozoic diatoms from the Northeast Pacific, Leg 18. Pages 673-797 in L.D. Kulm,
R. von Huene, et al., eds., Initial Reports, Deep Sea Drilling Project 18. U.S. Govt. Printing Office,
Washington, D.C.

SCHRADER, H.-J. AND J. FENNER. 1976. Norwegian Sea Cenozoic diatom biostratigraphy and taxonomy. Pages
921-1099 in M. Talwani, G. Udintsev, et al., eds.; Initial Reports, Deep Sea Drilling Project 38. U.S. Govt.
Printing Office, Washington, D.C.

SHESHUKOVA-PORETSKAYA, V.S. 1964. New and rare marine diatoms from the Neogene of Sakhalin and
Kamchatka. Novitates Systematicae Plantarum Non Vascularum: 69-77. Nauka Publishers, Moscow-
Leningrad. (in Russian)

GLADENKOV: DIATOM BIOSTRATIGRAHY IN SOUTHWESTERN ALASKA 43

SHESHUKOVA-PORETSKAYA, V.S. 1967. Neogene Marine Diatoms of Sakhalin and Kamchatka. Leningrad State
University Press, Leningrad. 432 pp. (in Russian)

VoLosBuEva, V.I., B.V. BELAYA, L.M. DoLMaATova, A.V. GREVTSEV, P.S. MINYUK, V.E. NARKHINOVA, T.P.
POLOVOVA, AND O.A. SHIRAYA. 1992. Key Section of Marine Neogene of Northest Asia in Karaginsky
Island. Part If (Systematic part). SVKNII, Far East Branch of Russian Academy of Sciences, Magadan.
167 pp. (in Russian)

WauitInG, M.C. AND H. SCHRADER. 1985. Late Miocene to Early Pliocene marine diatom and silicoflagellate
floras from the Oregon coast and continental shelf. Micropaleontology 31:249-270.

WORNARDT, W.W., JR. 1967. Miocene and Pliocene marine diatoms from California. California Academy of
Sciences Occasional Papers (63):1—108.

YANAGISAWA, Y. 1990. Diatom biostratigraphy of the Neogene Sendai Group, northeast Honshu, Japan.
Bulletin of the Geological Survey of Japan 41(1):1—25. (in Japanese with English abstract)

YANAGISAWA, Y. 1998. Diatom biostratigraphy of the Neogene Tatsunokuchi Formation in the western
Kitakami City, Iwate Prefecture, Japan. Research Report Iwate Prefectural Museum 14:29-36. (in
Japanese with English abstract)

YANAGISAWA, Y. AND F. AKIBA. 1990. Taxonomy and phylogeny of the three marine diatom genera,
Crucidenticula, Denticulopsis and Neodenticula. Bulletin of the Geological Survey of Japan 41(5):
197-301.

YANAGISAWA, Y. AND F. AKIBA. 1998. Refined Neogene diatom biostratigraphy for the northwest Pacific
around Japan with an introduction of code numbers for selected diatom biohorizons. Journal of the
Geological Society of Japan 104(6):395-414.

YANAGISAWA, Y., K. NAKAMURA, Y. SUZUKI, K. SAWAMURA, F. YOSHIDA, Y. TANAKA, Y. HONDA, AND M.
TANAHASHI. 1989. Tertiary biostratigraphy and subsurface geology of the Futaba district, Joban Coalfield,
northeast Japan. Bulletin of the Geological Survey of Japan 40(8):405—467. (in Japanese with English
abstract)

APPENDIX

Taxonomic Notes and References to Good Figures

Taxonomic references to marine diatom and silicoflagellate taxa from the Milky River
Formation, including those illustrated on Plates 1—7, are listed below. In general, only references
to good figures are listed following a semicolon. Thus, with few exceptions, the reader should look
to these references for the original citation of the taxon. Important synonyms and brief remarks are
included.

In most cases, only generic distinctions were made for nonmarine diatoms encountered in the
Milky River Formation samples. This group includes Achnanthes Bory, Amphora Ehrenberg ex
Kiitzing, Asterionella Hassal (tabulated together with Tabellaria Ehrenberg ex Kiitzing),
Aulacosira Thwaites, Cyclotella Kiitzing ex Brebisson, Cymbella Agardh, Dimerogramma Ralts,
Diatoma De Candolle, Eunotia Ehrenberg, Fragilaria Lyngbye, Gomphonema Agardh, Gyrosigma
Hassal, Melosira Agardh, Meridon Agardh, Navicula Bory, Nitzschia Hassal, Pinnularia
Ehrenberg, Surirella Turpin, Synedra Ehrenberg, Tetracyclus Ralfs, and Tryblionella Smith.

Marine Diatoms

Actinocyclus curvatulus Janisch; Koizumi, 1973, pl. 1, figs. 1-6; Sancetta, 1982, pl. 1, figs. 1-3. Synonyms:
Actinocyclus divisus (Grunow) Hustedt; Coscinodiscus curvatulus Grunow; Coscinodiscus divisus
Grunow. (Plate 5, Figs. 11-12)

Actinocyclus ochotensis Jousé, 1968: pp. 17-18, pl. 2, figs. 2-5; Donahue, 1970, pl. 1, figs. a—b, d; Koizumi,
1973, pl. 2, figs. 3, 6-7; Sancetta, 1982, pl. 1, figs. 4-6. (Plate 5, Fig. 9)

Actinocyclus octonarius Ehrenberg; Wornardt, 1967, fig. 49. Synonym: Actinocyclus ehrenbergii Ralts.
(Plate 5, Fig. 5)

44 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 3

Actinocyclus Ehrenberg spp. Remarks: Some species of Actinocyclus have been not identified precisely. They
have not been described or tabulated specifically here.

Actinoptychus senarius (Ehrenberg) Ehrenberg; Kanaya, 1959, pl. 7, fig. 17; Hanna, 1970, figs. 38-39;
Sancetta, 1982, pl. 1, fig. 7; Akiba, 1986, pl. 29, fig. 2. Synonym: Actinoptychus undulatus (Bailey) Ralfs.
(Plate 5, Fig. 8)

Actinoptychus splendens (Shadbolt) Ralfs; Sheshukova-Poretskaya, 1967, pl. 29, fig. 1; Hanna, 1970, figs.
25-26, 40, 44. (Plate 5, Fig. 1)

Actinoptychus vulgaris Schumann; Sheshukova-Poretskaya, 1967, pl. 28, figs. 2a—d; Sancetta, 1982, pl. 1, fig.
8; Akiba, 1986, pl. 29, fig. 1.

Adoneis pacifica Andrews et Rivera, 1987: pp. 2, 9, figs. 1-26. (Plate 1, Fig. 2)

Arachnoidiscus Bailey ex Ehrenberg spp. Remarks: Only scarce fragments of valves with the typical struc-
ture of Arachnoidiscus were observed. No attempt has been made to subdivide this group.

Aulacodiscus Ehrenberg spp. Remarks: No specific generic distinctions were made for rare specimens of
Aulacodiscus and their fragments found in the material studied.

Auliscus sp. cf. A. grunowii Schmidt. Remarks: Only rare fragments of valves have been found. However,
their structure resembles that of specimens illustrated by Wornardt (1967, figs. 92-94).

Azpeitia Peragallo spp. Remarks: The rare specimens observed here are similar in part to Azpeitia nodulifera
(Schmidt) Fryxell et Sims and Azpeitia tabularis (Grunow) Fryxell et Sims, but they differ to some extent
from these species by a character of valves structure.

Bacteriasrtum varians Lauder; Barron, 1975, pl. 4, fig. 8. (Plate 2, Figs. 8-9)

Bacterosira fragilis (Gran) Gran; Sheshukova-Poretskaya, 1967, pl. 33, figs. 2a—b; Schrader, 1973, pl. 16, fig.
7; Sancetta, 1982, pl. 2, figs. 1-4; Akiba, 1986, pl. 4, figs. 14; Dzinoridze and Makarova, 1988, pl. 60,
figs. 7-12. (Plate 3, Fig. 19; Plate 4, Fig. 18)

Chaetoceros cinctus Gran (spore); Sheshukova-Poretskaya, 1967, pl. 33, fig. 9. Remarks: Because of the sim-
ilarity of Chaetoceros cinctus and Chaetoceros incurvus spores, they have been combined and counted
together.

Chaetoceros furcellatus Bailey (spore); Sheshukova-Poretskaya, 1967, pl. 33, fig. 8; Sancetta, 1982, pl. 2,
Sst.

Chaetoceros incurvus Bailey (spore); Sheshukova-Poretskaya, 1967, pl. 33, fig. 10. Remarks: See Remarks
for Chaetoceros cinctus.

Chaetoceros septentrionales Oestrup (spore); Sancetta, 1982, pl. 2, fig. 8.

Chaetoceros subsecundus (Grunow) Hustedt (spore); Sancetta, 1982, pl. 2, figs. 5-6. Synonym: Chaetoceros
diadema (Ehrenberg) Gran.

Cladogramma dubium Lohman; Sheshukova-Poretskaya, 1967, pl. 24, figs. 6 a—b, pl. 29 figs. 4 a—c; Schrader,
1973, pl. 13, figs. 17-18, 21; Barron, 1975, pl. 5, fig. 10. (Plate 7, Figs. 3, 6)

Cocconeis antiqua Tempére et Brun; Hanna, 1970, fig. 48; Barron, 1975, pl. 5, fig. 11

Cocconeis californica Grunow; Sheshukova-Poretskaya, 1967, pl. 43, fig. 11; Yanagisawa et al., 1989, pl. 6,
iifes NS)

Cocconeis costata Gregory; Sheshukova-Poretskaya, 1967, pl. 44, figs. 4 a—c; Barron, 1975, pl. 5, fig. 12;
Sancetta, 1982, pl. 6, figs. 6-7.

Cocconeis placentulla Ehrenberg; Barron, 1975, pl. 5, fig. 17.

Cocconeis pribiloeformis Hanna, 1970: p. 184, fig. 34.

Cocconeis scutellum Ehrenberg; Sheshukova-Poretskaya, 1967, pl. 44, fig. 7; Barron, 1975, pl. 5, fig. 19.

Cocconeis vitrea Brun; Sheshukova-Poretskaya, 1967, pl. 45, figs. 3a—c; Wornardt, 1967, figs. 183-184;
Barron, 1975, pl. 5, fig. 23.

Coscinodiscus asteromphalus Ehrenberg; Sheshukova-Poretskaya, 1967, pl. 21, fig. 2; Wornardt, 1967, figs.
14-18; Schrader and Fenner, 1976, pl. 34, figs. 2-3.

Coscinodiscus marginatus Ehrenberg; Kanaya, 1959, pl. 4, figs. 4-6; Wornardt, 1967, figs. 27-28; Hanna,
1970, figs. 1-2; Barron, 1975, pl. 7, fig. 1; Akiba, 1986, pl. 1, figs. 14. (Plate 2, Figs. 4-5)

Coscinodiscus oculus-iridis Ehrenberg; Kanaya, 1959, pl. 4, fig. 7; Sheshukova-Poretskaya, 1967, pl. 21, fig.
1; Wornardt, 1967, figs. 34-35; Hanna, 1970, fig. 18; Barron, 1975, pl. 7, fig. 9; Sancetta, 1982, pl. 2, fig. 11.

Coscinodiscus Ehrenberg spp. Remarks: Rare fragments of different Coscinodiscus having indistinct charac-

GLADENKOV: DIATOM BIOSTRATIGRAHY IN SOUTHWESTERN ALASKA 45

ters for identification were assigned to this category.

Cosmiodiscis insignis Jousé, 1961: pp. 67—68, pl. 2, fig. 8; Sheshukova-Poretskaya, 1967, pl. 25, figs. 2 a—c;
Hanna, 1970, figs. 9-11, 31, 32; Koizumi, 1973, pl. 4, figs. 7-11; Barron, 1980, pl. 4, fig. 1; Barron, 1985,
fig. 9.9; Akiba, 1986, pl. 17, fig. 1. Synonym: Thalassiosira insigna (Jousé) Harwood et Maruyama sensu
Barron and Gladenkov (1995). Remarks: Harwood and Maruyama (1992) transferred this taxon from
Cosmiodiscis to Thalassiosira and proposed the new combination Thalassiosira insigna (Jousé) Hawood
et Maruyama, based on observations of Antarctic forms bearing a central structure (dimple or bubble).
Following Harwood and Maruyama (1992), in some cases this new combination has also been used for
North Pacific forms (Barron and Gladenkov 1995; Marincovich and Gladenkov 1999, 2001; and others).
However, recently, specimens of Cosmiodiscus insignis collected by G D. Hanna from the Pribilof Islands
were studied with a SEM by E. Fourtanier (Fourtanier, personal communication, 2001), who observed that
the non-areolate center of the valve does not possess a central process, and that a ring of labiate process-
es is located on the boundary between the mantle and valve face. I also observed the same features when
studying specimens of this taxon from the Milky River Formation with a SEM at the California Academy
of Sciences in 2001. These observations indicate that this North Pacific taxon cannot be included in
Thalassiosira, and that perhaps forms from the North Pacific and Antarctic regions belong to different
taxa. Further study of the Antarctic forms under a SEM should clarify their taxonomic position. (Plate 4,
Fig. 16; Plate 6, Figs. 2-3, 5)

Cosmiodiscis intersectus (Brun) Jousé, 1961: p. 68, pl. 2, figs. 9-10; Sheshukova-Poretskaya, 1967, pl. 25,
figs. 1 a—b; Koizumi, 1973, pl. 4, figs. 12-13.

Cymatosira debyi Tempére et Brun; Sheshukova-Poretskaya, 1967, pl. 40, fig.7, pl. 41, fig. 6; Akiba, 1986,
pl. 19, figs. 19-20; Yanagisawa et al, 1989, pl. 5, fig. 31. (Plate 3, Figs. 3-4; Plate 4, Fig. 8)

Delphineis angustata (Pantocsek) Andrews group. Remarks: Specimens of Delphineis range from elliptical
to elongate with rounded apices, and resemble Delphineis angustata (Pantocsek) Andrews, 1977,
Delphineis ischaboensis (Grunow) Andrews, 1988, and Delphineis karstenii (Boden) Fryxell et Miller
have been placed in this group in part following the example of Delphineis cf. angustata of Akiba (1986).
(Plate 4, Figs. 3-6, 10-11)

Delphineis sachalinensis (Sheshukova) Barron et Baldauf, 1986: p. 363. Synonym: Rhaphoneis sachalinen-
sis Sheshukova: Sheshukova-Poretskaya, 1967, p. 242, pl. 42, fig. 2. (Plate 4, Fig. 13)

Delphineis simonsenii (Mertz) Akiba, 1986: p. 439, pl. 20, figs. 12-13. (Plate 5, Figs. 6-7)

Delphineis surirella (Ehrenberg) Andrews, 1980: pp. 83-85; pl. 1, figs. 1-5, pl. 2, figs. 6-7; Akiba, 1986, pl.
20, figs. 2-3.

Detonula confervacea (Cleve) Gran; Dzinoridze and Makarova, 1988, p. 88, pl. 60, figs. 14-15, text figs.
6.1-6.3. Synonym: Melosira albicans Sheshukova, 1964: Sheshukova-Poretskaya, 1964, p. 69, 71, text
figs. 1-2, pl. 1, fig. 3; 1967, pl. 10, figs. 2a—b, pl. 11, figs. la—b; Akiba, 1986, pl. 4, figs. 11-12;
Yanagisawa et al., 1989, pl. 3, fig. 7. (Plate 1, Fig. 9; Plate 5, Fig. 10)

Dicladia capreolus Ehrenberg; Kanaya, 1959, pl. 11, figs. 1-2; Sheshukova-Poretskaya, 1967, pl. 34, figs.
la—c. Synonyms: Dicladia capreola Ehrenberg; Hanna, 1970, fig. 63; Dicladia pylea Hanna et Grant;
Chaetoceros dicladia Castracane. (Plate 2, Figs. 2-3)

Diploneis smithii (Brebisson) Cleve; Barron, 1975, pl. 8, fig. 17; Sancetta, 1982, pl. 6, fig. 21.

Fragilariopsis cylindrus (Grunow) Krieger; Koizumi, 1973, pl. 7, figs. 1-2; Hasle, 1993, p. 316. Synonyms:
Nitzschia cylindrus (Grunow) Hasle; Koizumi, 1975, pl. 1, fig. 49; Nitzschia cylindra (Grunow) Hasle;
Sancetta, 1982, pl. 3, figs. 6-7. (Plate 3, Figs. 13-15)

Fragilariopsis oceanica (Cleve) Hasle; Hasle, 1993, p. 317; Synonym: Nitzschia grunowiti Hasle; Sancetta,
1982, pl. 3, figs. 8-10; Akiba, 1986, pl. 24, figs. 19-21. (Plate 3, Fig. 16-17)

Grammatophora Ehrenberg spp. Remarks: Observed specimens resemble G. angulosa Ehrenber and G. arcu-
ata Ehrenberg and occur very sporadically.

Hercotheca mamillaris Ehrenberg; Barron, 1975, pl. 9, fig. 13.

Hyalodiscus obsoletus Sheshukova, 1964: Sheshukova-Poretskaya, 1964, p. 71, pl. 1, fig. 1; 1967, pl. 12, fig.
2; Akiba, 1986, pl. 29, fig. 15; Yanagisawa et al., 1989, pl. 3, fig. 12. (Plate 1, Fig. 11)

Hyalopyxis concava (Sheshukova) Makarova, 1989: p. 35, pl. 1, figs. 1-7, pl. 2, figs. 1-6. Synonym:
Trochosira concava Sheshukova: Sheshukova-Poretskaya, 1967, p. 138, pl. 11, figs. 7a—b, pl. 13, fig. 3.

46 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 3

Remarks: Specimens of Hyalopyxis concava were tabulated together with specimens of Trochosira spin-
osa Kitton. (Plate 3, Fig. 21-22)

Liradiscus ellipticus Greville; Barron, 1975, pl. 9, fig. 19.

Lithodesmium minusculum Grunow; Wornardt, 1967, fig. 132; Schrader, 1973, pl. 12, figs. 15, 17; Barron,
1975, pl. 10, fig. 4; Oreshkina, 1985, pl. 3, fig. 15. (Plate 7, Figs. 7, 11)

Navicula glacialis Cleve; Sheshukova-Poretskaya, 1967, pl. 46, fig. 8.

Neodenticula kamtschatica (Zabelina) Akiba et Yanagisawa, 1986: pp. 490-491, pl. 21, figs. 7-21, pl. 22, figs.
1-12; Akiba, 1986, pl. 25, figs. 7-27; Yanagisawa and Akiba, 1990, pl. 7, figs. 27-37. Synonoms:
Denticula kamtschatica Zabelina; Sheshukova-Poretskaya, 1967, pl. 47, figs. 9a—b, pl. 48, figs. 4a—d;
Koizumi, 1973, pl. 5, figs. 14-17; Schrader, 1973, pl. 2, figs. 1-13; Koizumi, 1975, pl. 1, figs. 13-20;
Barron, 1980, pl. 1, figs. 5-8; Denticulopsis kamtschatica (Zabelina) Simonsen; Barron, 1985, fig. 13.16:
Oreshkina, 1985, pl. 1, figs. 16-20; Whiting and Schrader, 1985, pl. 6, figs. 15—16, 18-20. (Plate 5, Figs.
2-4; Plate 6, Fig. 4)

Nitzschia extincta Kozyrenko et Sheshukova group. Remarks: Specimens having a similarity in shape and
structure to Nitzschia extincta Kozyrenko et Sheshukova (in Sheshukova-Poretskaya, 1967, pp. 303-304,
pl. 47, fig. 12) have been placed in this group.

Nitzschia rolandii Schrader; Akiba and Yanagisawa, 1986, pl. 21, figs. 1-6; Akiba, 1986, pl. 25, figs. 1-6;
Yanagisawa and Akiba, 1990, pl. 7, figs. 17-26. (Plate 3, Fig. 20)

Odontella aurita (Lyngbye) Agardh; Sancetta, 1982, pl. 3, figs. 11-12; Akiba, 1986, pl. 17, figs. 2-3. (Plate
2, Figs. 7, 10)

Opephora schwartzii (Grunow) Petit; Wornardt, 1967, figs. 161-165; Barron, 1975, pl. 11, fig. 10.

Paralia sulcata (Ehrenberg) Cleve; Sancetta, 1982, pl. 3, figs. 13-15; Akiba, 1986, pl. 29, figs. 4-5. (Plate 7,
Figs. 12-13)

Pleurosigma Smith spp. Remarks: No specific generic distinctions were made for rare specimens of
Pleurosigma and their fragments found in the material studied.

Porosira glacialis (Grunow) Jorgensen; Koizumi, 1973, pl. 4, figs. 15-18; Barron, 1980, pl. 6, fig. 13;
Dzinoridze and Makarova, 1988, pl. 60, figs. 1-6. Remarks: Specimens of Porosira glacialis were count-
ed together with specimens of Thalassiosira undulosa (Mann) Sheshukova. (Plate 7, Fig. 5)

Porosira punctata (Jousé) Makarova, 1988: Makarova, 1988a, p. 1184, pl. 1, figs. 1-16; Dzinoridze and
Makarova, 1988, pl. 59, figs. 1-10. Synonom: Thalassiosira punctata Jousé, 1961: p. 64, pl. 1, figs. 7-8,
pl. 3, fig. 3; Sheshukova-Poretskaya, 1967, pl. 14, fig. 10, pl. 17, figs. la—b; Hanna, 1970, figs. 5-6;
Koizumi, 1973, pl. 8, figs. 7-9; Schrader and Fenner, 1976, pl. 19, fig. 10; Barron, 1980, pl. 6, fig. 3;
Akiba, 1986, pl. 9, figs. 5—6. (Plate 2, Fig. 1; Plate 6, Fig. 12)

Proboscia alata (Brightwell) Sundstr6m; Jordan et al., 1991, figs. 1-9; Synonom: Rhizosolenia alata
Brightwell; Schrader, 1973, pl. 10, fig. 12; Koizumi, 1975, pl. 1, fig. 38; Akiba, 1986, pl. 18, fig. 6.

Proboscia barboi (Brun) Jordan et Priddle, 1991: p. 56, figs. 1-2; Synonoms: Rhizosolenia barboi (Brun)
Tempere et Peragallo; Schrader, 1973, pl. 24, figs. 4, 7; Barron, 1980, pl. 2, fig. 17; Barron, 1985, fig. 4.5;
Akiba and Yanagisawa, 1986, pl. 42, figs. 3-5, 7, 10-11, pl. 44, figs. 1-8; Rhizosolenia curvirostris vat.
inermis Jousé; Donahue, 1970, pl. 1, figs. b—c; Koizumi, 1973, pl. 5, figs. 32-33.

Pseudopyxilla americana (Ehrenberg) Forti; Sheshukova-Poretskaya, 1967, pl. 39, figs. 2a—b; Schrader, 1973,
pl. 10, fig. 22; Barron, 1975, pl. 11, fig. 12. (Plate 7, Fig. 4)

Pseudopyxilla Forti sp. Remarks: Only scarce fragments of Pseudopyxilla having some similarity to
Pseudopyxilla rossica (Pantoschek) Forti have been observed.

Pyxidicula zabelinae (Jousé) Makarova et Moisseeva: Makarova and Moisseeva, 1986, p. 244-245, pl. 1, figs.
1-15, pl. 2, figs. 1-15. Synonyms: Thalassiosira zabelinae- Jousé, 1961: p. 66-67, pl. 2, figs. 1-7;
Sheshukova-Poretskaya, 1967, pl. 16, figs. 2 a-d; Thalassiosira usatschevii Jousé, 1961: pp. 64, 66, pl. 1,
fig. 10; Sheshukova-Poretskaya, 1967, pl. 15, figs. 3 a-d. (Plate 1, Figs. 13, 15-16)

Rhabdonema japonicum Tempére et Brun; Hanna, 1970, figs. 92-95; Schrader, 1973, pl. 12, fig. 10.

Rhaphoneis amphiceros Ehrenberg; Sheshukova-Poretskaya, 1967, pl. 41, fig. 9; Schrader, 1973, pl. 25, figs.
2—3; Whiting and Schrader, 1985, pl. 6, figs. 4-5; Akiba, 1986, pl. 20, fig. 19.

Rhaphoneis angularis Lohman, 1938: pp. 92-93, pl. 22, figs. 6-8; Schrader, 1973, pl. 26, figs. 9-10; Whiting
and Schrader, 1985, pl. 6, figs. 22-24; Yanagisawa et al., 1989, pl. 5, fig. 38. Synonom: Rhaphoneis amph-

GLADENKOV: DIATOM BIOSTRATIGRAHY IN SOUTHWESTERN ALASKA 47

iceros var. angularis (Lohman) Wornardt, 1967: p. 78, figs. 169-170. (Plate 4, Figs. 1-2, 7; Plate 6, Figs.
6, 10)

Rhizosolenia hebetata (Bailey) Gran group. Remarks: All scarce specimens having a similarity with
Rhizosolenia hebetata f. hiemalis Gran (Akiba, 1986, pl. 17, figs. 10-11, pl. 18, figs. 9-10) and
Rhizosolenia hebetata f. semispina (Hensen) Gran (Akiba, 1986, pl. 18, fig. 8) were placed in this group.
(Plate 4, Fig. 12)

Rhizosolenia setigera Brightwell; Akiba, 1986, pl. 18, fig. 5.

Rhizosolenia styliformis Brightwell; Sancetta, 1982, pl. 4, figs. 7-8; Akiba, 1986, 18, fig. 4.

Stellarima microtrias (Ehrenberg) Hasle et Sims; Hasle et al., 1988, figs. 1-25. Synonym: Coscinodiscus sym-
bolophorus Grunow; Sheshukova-Poretskaya, 1967, pl. 22, figs. 3 a-e; Akiba, 1986, pl. 2, fig. 1.

Stephanogonia hanzawae Kanaya, 1959: pp. 118-119, pl. 11, figs. 3—7; Schrader and Fenner, 1976, pl. 12,
figs. 10, 12, pl. 13, figs. 5, 7-8; Akiba et al., 1982, pl. 2, fig. 36. Synonym: Prerotheca kittoniana var.
kamtschatica Gaponov; Sheshukova-Poretskaya, 1967, pl. 39, figs. 3a—f. (Plate 1, Fig. 14)

Stephanopyxis turris (Greville et Arnott) Ralfs; Kanaya, 1959, pl. 2, figs. 5-7; Koizumi, 1973, pl. 6, figs.
13-16; Schrader, 1973, pl. 15, figs. 1-7; Sancetta, 1982, pl. 4, figs. 9-10.

Stephanopyxis (Ehrenberg) Ehrenberg spp. Remarks: All scarce specimens not referable to Stephanopyxis
turris were assigned to this category.

Thalassionema nitzschioides (Grunow) H. et M. Peragallo; Sancetta, 1982, pl. 4, figs. 11-13; Akiba, 1986, pl.
21, fig. 11. Remarks: No attempt was made to separate varieties of this species. (Plate 3, Figs. 9-10)
Thalassiosira antiqua (Grunow) Cleve-Euler; Sheshukova-Poretskaya, 1967, pl. 14, figs. 3a—b; Schrader,
1973, pl. 11, fig. 25, pl. 25, fig. 25; Barron, 1980, pl. 5, fig. 5; Barron, 1985, fig. 11.2; Akiba, 1986, pl. 12,

figs. 1, 3-4. (Plate 3, Figs. 1-2)

Thalassiosira sp. cf. T: convexa Mukhina, 1965: pp. 22-24, pl. 2, figs. 1-2. Remarks: Specimens resemble T.
convexa but they differ to some extent from typical form by more flat valve, arrangement of areolae and
character of the marginal zone. (Plate 6, Figs. 11, 13-17; Plate 4, Fig. 15)

Thalassiosira decipiens (Grunow) Jorgensen; Sancetta, 1982, pl. 5, figs. 1-3.

Thalassiosira delicata (Barron) Akiba, 1986: p. 440. Synonym: Thalassiosira nidulus var. delicata Barron,
1980: p. 671, pl. 6, figs. 1, 4.

Thalassiosira dolmatovae Oreshkina in Gladenkov et al., 1992: p. 129, pl. 40, figs. 7-8. Synonym:
Thalassiosira sp. 9 of Dolmatova in Volobueva et al., 1992: p. 86, pl. 30, figs. 1-5. (Plate 1, Fig. 1; Plate
3, Fig. 24)

Thalassiosira eccentrica (Ehrenberg) Cleve; Makarova, 1988b, pl. 20, figs. 1-9; pl. 21, figs. 1-12; Makarova,
1988c, pl. 40, figs. 1-9. (Plate 4, Fig. 14; Plate 6, Fig. 9)

Thalassiosira gravida Cleve; Schrader and Fenner, 1976, pl. 16, figs. 5—6, pl. 17, fig. 2; Barron, 1980, pl. 6,
figs. 11, 14; Akiba, 1986, pl. 10, figs. 1-4. Remarks: No separation of the species from Thalassiosira
gravida f. fossilis Jousé (1961: p. 63, pl. 1, fig. 9) was done in the material studied. (Plate 2, Fig. 6)

Thalassiosira hyalina (Grunow) Gran; Sancetta, 1982, pl. 5, figs. 4-5; Akiba, 1986, pl. 5, fig. 9; Makarova,
1988b, pl. 45, figs. 1-10. (Plate 3, Figs. 7-8, 18)

Thalassiosira jacksonii Koizumi et Barron in Koizumi, 1980: p. 396, pl. 1, figs. 11-14; Barron, 1980, pl. 6,
figs. 2, 6; Akiba, 1986, pl. 11, fig. 2. Synonym: Thalassiosira sp. b of Schrader and Fenner, 1976, pl. 17,
figs. 5, 10. (Plate 6, Figs. 7-8)

Thalassiosira sp. cf. T. jacksonii Koizumi et Barron. Remarks: Specimens resemble 7: jacksonii but differ
from the latter by lacking a distinct central hyaline area with a process.

Thalassiosira jouseae Akiba, 1986: p. 440, pl. 6, figs. 8-10. (Plate 1, Figs. 3, 7, 10)

Thalassiosira kryophila (Grunow) J@rgensen; Sheshukova-Poretskaya, 1967, pl. 14, fig. 6; Koizumi, 1973, pl.
8, fig. 3; Makarova, 1988b, pl. 35, fig. 3. (Plate 2, Fig. 11)

Thalassiosira latimarginata Makarova; Makarova, 1988b, pl. 30, figs. 1-12; Makarova, 1988c, pl. 40, figs.
11-17. Synonym: Thalassiosira trifulta Fryxell in Fryxell and Hasle, 1979: pp. 16-19, pls. 1-5, figs. 1-24;
Sancetta, 1982, pl. 5, figs. 10-12, pl. 6, figs. 1-2; Akiba, 1986, pl. 10, figs. 5—7. (Plate 1, Fig. 4; Plate 6,
Fig. 1)

Thalassiosira leptopus (Grunow) Hasle et Fryxell; Akiba, 1986, pl. 9, figs. 3-4; Makarova, 1988b, pp. 49-50;
Makarova, 1988c, pl. 42, fig. 12. (Plate 4, Fig. 17)

48 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 3

Thalassiosira lineata Jousé, 1968: p. 13, pl. 1, figs. 1-2; Akiba, 1986, pl. 14, figs. 7, 9.

Thalassiosira manifesta Sheshukova, 1964: Sheshukova-Poretskaya, 1964, p. 72, pl. 1, figs. 6-7; 1967, pl. 14,
figs. 9a—b; Akiba, 1986, pl. 9, figs. 1-3. (Plate 7, Figs. 8, 10)

Thalassiosira marujamica Sheshukova emend. Makarova, 1988: Makarova, 1988b, pp. 51-52, pl. 24, figs.
1-13; 1988c, pl. 41, figs. 13-21. Synonyms: Thalassiosira decipiens (Grunow) Jé@rgensen sensu
Sheshukova-Poretskaya, 1964, pl. 1, fig. 2; 1967, pl. 14, fig. 2; Thalassiosira borealis Koizumi, 1980: p.
395, pl. 1, figs. 7-10. (Plate 3, Figs. 25, 27)

Thalassiosira nativa Sheshukova, 1964: Sheshukova-Poretskaya, 1964, p. 75, pl. 1, figs. 4-5; 1967, pl. 14,
figs. 7 a—c. (Plate 1, Figs. 5-6)

Thalassiosira sp. cf. T: nativa Sheshukova. Remarks: Small specimens resemble 7: nativa having some indis-
tinct characters for the precise identification were assigned to this category and tabulated together with
specimens of 7. nativa.

Thalassiosira nidilus (Tempere et Brun) Jousé; Akiba, 1986, pl. 6, figs. 5—7.

Thalassiosira oestrupitit (Ostenfeld) Proshkina-Lavrenko; Barron, 1980, pl. 5, fig. 4; Barron, 1985, figs.
11.5—11.6; Akiba, 1986, pl. 14, figs. 1-6; Makarova, 1988b, pl. 25, figs. 1-9; Makarova, 1988c, pl. 42, figs.
13-17. (Plate 3, Figs. 11-12)

Thalalassiosira orientalis Sheshukova emend. Makarova, 1988: Makarova, 1988b, p. 77, pl. 50, figs. 1-9;
1988c, pl. 52, figs. 13-17. Synonyms: Thalalassiosira margaritae (Frenguelli et Orlando) Kozlova sensu
Sheshukova-Poretskaya, 1964, pl. 2, figs. 1-3; Thalalassiosira aff. margaritae (Frenguelli et Orlando)
Kozlova sensu Sheshukova-Poretskaya, 1967, pl. 14, figs. Sa—c. (Plate 7, Fig. 9)

Thalassiosira praeoestrupii Dumont et al., 1986 emend. Bodén, 1993: pp. 67— 68, pl. 1, figs. H—J, pl. 2, figs.
C-G, pl. 3, figs. H, J. (Plate 7, Figs. 1-2)

Thalassiosira sheshukovae Makarova, 1988: Makarova, 1988a, p. 1185, pl. 2, figs. 1-11. Synonym:
Pseudopodosira elegans Sheshukova: Sheshukova-Poretskaya, 1964, p. 75—76, text fig. 3, pl. 2, figs. 4-5;
1967, pl. 24, fig. 3; Pl. 25, fig. 4; Sancetta, 1982, pl.4, figs. 1-2; Akiba, 1986, pl. 4, figs. 5-7. (Plate 1,
Fig. 8)

Thalassiosira temperei (Brun) Akiba et Yanagisawa, 1986: p. 493, pl. 31, figs. 1-7; Yanagisawa, 1990, pl. 1,
figs. 16, 25. Synonyms: Coscinodiscus temperei Brun; Kanaya, 1959, pl. 4, fig. 8; Barron, 1980, pl. 4, fig.
5; Akiba et al, 1982, pl. 1, fig. 5; Cymatotheca weissflogii (Grunow) Hendey sensu Sheshukova-
Poretskaya, 1967, pl. 26, fig. 1; sensu Hanna, 1970, figs. 15—16. (Plate 3, Fig. 23)

Thalassiosira tertiaria Sheshukova, 1967: Sheshukova-Poretskaya, 1967, p. 148, pl. 15, fig. 2.

Thalassiosira undulosa (Mann) Sheshukova, 1967: Sheshukova-Poretskaya, 1967, pp. 148-149, pl. 16, figs.
1 a—c; Koizumi, 1973, pl. 8, figs. 5-6; Akiba, 1986, pl. 9, fig. 4. Synonym: Coscinodiscus undulosus
Mann; Hanna, 1970, fig. 3. (Plate 4, Fig. 19)

Thalassiosira Cleve spp. Remarks: The precise identification of Thalassiosira species having sublinear, tan-
gential or fasciculated areolae pattern, with an indistinct structure of the marginal and central zones, was
not attempted. These specimens were not tabulated separately in the occurrence table.

Thalassiothrix longissima Cleve et Grunow; Schrader, 1973, pl. 23, figs. 7, 17-18; Akiba, 1986, pl. 21, fig.
18.

Thalassiothrix robusta (Schrader) Akiba, 1986: p. 441, pl. 21, fig. 4. (Plate 3, Figs. 5-6; Plate 4, Fig. 9)

Trochosira spinosa Kitton; Sheshukova-Poretskaya, 1967, pl. 11, figs. 6a—b, pl. 13, figs. 4a—b. Remarks: See
Remarks for Hyalopyxis concava (Sheshukova) Makarova.

Xanthiopyxis globosa Ehrenberg; Barron, 1975, pl. 15, figs. 8-9.

Xanthiopyxis ovalis Lohman, 1938: p. 91, pl. 20. fig. 6, pl. 22. fig. 12; Hanna, 1970, fig. 70; Barron, 1975, pl.
[Seis 13)

Xanthiopyxis (Ehrenberg) Ehrenberg spp. Remarks: All scarce specimens not referable to X. globosa or X.
ovalis were assigned to this category.

Silicoflagellates

Distephanus crux s. ampl. (Ehrenberg) Haeckel; Perch-Nielsen, 1985, figs. 18.7—18.10.
Distephanus speculum pentagonus Limmermann; Perch-Nielsen, 1985, fig. 20.1.
Distephanus speculum speculum (Ehrenberg) Glezer; Perch-Nielsen, 1985, figs. 20.8—20.9.

GLADENKOV: DIATOM BIOSTRATIGRAHY IN SOUTHWESTERN ALASKA

PLATES 1-7

49

50 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 3

Plate 1

Fig. 1: Thalassiosira dolmatovae (a—c: valve in different focus). Fig. 2: Adoneis pacifica. Figs.
3, 7, 10: Thalassiosira jouseae. Fig. 4: Thalassiosira latimarginata. Figs. 5—6: Thalassiosira nati-
va. Fig. 8: Thalassiosira sheshukovae. Fig. 9: Detonula confervacea. Fig. 11: Hyalodiscus obsole-
tus. Fig. 12: Actinicyclus sp. Figs. 13, 15-16: Pyxidicula zabelinae. Fig. 14: Stephanogonia hanza-
wae.

Light microscope (LM). Magnification x1000 except of Fig. 4 (x1250).

Figs. 1-3, 7-9, 11-14: CAS 608382. Fig. 4: CAS 608426. Figs. 5, 10: CAS 617358. Fig. 6:
CAS 608378. Fig. 15: CAS 608376. Fig. 16: CAS 608426.

Sil

GLADENKOY: DIATOM BIOSTRATIGRAHY IN SOUTHWESTERN ALASKA

PLATE 1

Sy PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 3

Plate 2

Fig. 1: Porosira punctata. Figs. 2—3: Dicladia capreolus. Figs. 4-5: Coscinodiscus margina-
tus. Fig. 6: Thalassiosira gravida. Figs. 7, 10: Odontella aurita. Figs. 8-9: Bacteriastrum varians.
Fig. 11: Thalassiosira kryophila (a—b: valve in different focus).

LM. Magnification x1000.

Fig. 1: CAS 608378. Figs. 2, 8: CAS 608400. Figs. 3, 5—7, 9-11: CAS 608382. Fig. 4: CAS

608397.

GLADENKOV: DIATOM BIOSTRATIGRAHY IN SOUTHWESTERN ALASKA

PLATE 2

5

PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES

54
Volume 54, No. 3

Plate 3

Figs. 1-2: Thalassiosira antiqua. Figs. 3-4: Cymatosira debyi. Figs. 5-6: Thalassiothrix
robusta. Figs. 7-8, 18: Thalassiosira hyalina. Figs. 9-10: Thalassionema nitzschioides. Figs.
11-12: Thalassiosira oestrupii. Fig. 13-15: Fragilariopsis cylindrus. Figs. 16-17: Fragilariopsis
oceanica. Fig. 19: Bacterosira fragilis (a—b: valve in different focus). Fig. 20: Nitzschia rolandit.
Figs. 21-22: Hyalopyxis concava. Fig. 23: Thalassiosira temperei. Fig. 24: Thalassiosira dolma-
tovae. Figs. 25, 27: Thalassiosira marujamica. Figs. 26, 28: Chaetoceros resting spores.

LM. Magnification x1000 except of Figs. 11—12 (600).

Figs. 1, 6, 8, 10-11, 20, 23-24, 28: CAS 608382. Figs. 2, 18: CAS 60269-s. Figs. 3-4, 13-14,
19: CAS 608376. Figs. 5, 9: CAS 608424. Figs. 7, 12, 26: CAS 617358. Figs. 15—16: CAS 608393.
Fig. 17: CAS 608376. Figs. 21-22, 25, 27: CAS 608371.

55

DIATOM BIOSTRATIGRAHY IN SOUTHWESTERN ALASKA

GLADENKOV

PLATE 3

* |

4
ea

56 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 3

Plate 4

Figs. 1-2, 7: Rhaphoneis angularis. Figs. 3-6, 10-11: Delphineis angustata group. Fig. 8:
Cymatosira debyi. Fig. 9: Thalassiothrix robusta. Fig. 12: Rhizosolenia hebetata group. Fig. 13:
Delphineis sachalinensis. Fig. 14: Thalassiosira eccentrica. Fig. 15: Thalassiosira sp. ct. T. con-
vexa. Fig. 16: Cosmiodiscus insignis. Fig. 17: Thalassiosira leptopus. Fig. 18: Bacterosira fragilis.
Fig. 19: Thalassiosira undulosa.

LM. Magnification x1000.

Figs. 1-2: CAS 617358. Figs. 3-6: CAS 608379. Figs. 7, 9-13, 16: CAS 608382. Figs. 14, 8,
19: CAS 608376. Fig. 15: CAS 608378. Fig. 17: CAS 608370. Fig. 18: CAS 608371.

GLADENKOV: DIATOM BIOSTRATIGRAHY IN SOUTHWESTERN ALASKA

PLATE 4

>i

58 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 3

Plate 5

Fig. 1: Actinoptychus splendens. Figs. 2-4: Neodenticula kamtschatica. Fig. 5: Actinocyclus
octonarius. Figs. 6-7: Delphineis simonsenii. Fig. 8: Actinoptychus senarius (a—b: valve in differ-
ent focus). Fig. 9: Actinocyclus ochotensis. Fig. 10: Detonula confervacea. Fig. 11-12:
Actinocyclus curvatulus.

LM. Magnification x1000.

Figs. 1, 3-4, 6-7, 9, 12: CAS 608382. Fig. 2: CAS 608376. Fig. 5: CAS 608400. Fig. 8: CAS
608378. Fig. 10: CAS 617358. Fig. 11: CAS 608392.

GLADENKOV: DIATOM BIOSTRATIGRAHY IN SOUTHWESTERN ALASKA 59

PLATE 5

¥
y
bi
aa
-
~
*

60 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 3

Plate 6

Fig. 1: Thalassiosira latimarginata. Figs. 2—3, 5: Cosmiodiscus insignis. Fig. 4: Neodenticula
kamtschatica. Figs. 6, 10: Rhaphoneis angularis. Figs. 7-8: Thalassiosira jacksonii (a—b: valve in
different focus). Fig. 9: Thalassiosira eccentrica. Figs. 11, 13-17: Thalassiosira sp. ct. T; convexa
(a—b: valve in different focus). Fig. 12: Porosira punctata.

LM. Magnification x1000.

Figs. 1: CAS 608408. Figs. 2, 6: CAS 617358. Figs. 3, 5, 9, 11: CAS 608382. Figs. 4, 7-8,
14-17: CAS 608376. Fig. 10: CAS 608379. Fig. 12: CAS 608378. Fig. 13: CAS 60269-s.

61

DIATOM BIOSTRATIGRAHY IN SOUTHWESTERN ALASKA

GLADENKOV

PLATE 6

os

os

*
‘ *
tee cegteee
Ly aa
6

62 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 3

Plate 7

Figs. 1-2: Thalassiosira praeoestrupii. Figs. 3, 6: Cladogramma dubium. Fig. 4:
Pseudopyxilla americana. Fig. 5: Porosira glacialis (a—b: valve in different focus). Figs. 7, 11:
Lithodesmium minusculum. Figs. 8, 10: Thalassiosira manifesta. Fig. 9: Thalassiosira orientalis.
Figs. 12-13: Paralia sulcata.

LM. Magnification x1000.

Figs. 1-2: CAS 60269-s. Figs. 3-4, 8-9, 11-12: CAS 608382. Figs. 5, 7: CAS 608376. Fig. 6:
CAS 608408 . Fig. 10: CAS 617358. Fig. 13: CAS 608392.

63

DIATOM BIOSTRATIGRAHY IN SOUTHWESTERN ALASKA

.
.

GLADENKOV

PLATE 7

64

PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 3

Copyright © 2003 by the California Academy of Sciences
San Francisco, California, U.S.A.

PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES

Volume 54, No. 4, pp. 65-79, 7 figs. March 14, 2003

Caryophyllidia-bearing Dorid Nudibranchs
(Mollusca, Nudibranchia, Doridacea) from Costa Rica

Yolanda E. Camacho-Garcia!2 and Angel Valdés?
!Department of Invertebrate Zoology and Geology, California Academy of Sciences, Golden Gate Park,
San Francisco, California 94118, USA; ?Instituto Nacional de Biodiversidad (INBio), Sub-Programa

de Malacologia, Santo Domingo de Heredia, Costa Rica; Natural History Museum of Los Angeles
County, 900 Exposition Boulevard, Los Angeles, California 90007, USA

The tropical eastern Pacific dorid nudibranch Taringa aivica Ev. Marcus and Er.
Marcus, 1967 is redescribed based on the examination of specimens collected from
the Pacific coast of Costa Rica. The differences between Taringa aivica timia Ev.
Marcus and Er. Marcus, 1967 and Taringa aivica aivica Ev. Marcus and Er. Marcus,
1967 are not consistent and do not justify the existence of two subspecies of Taringa
aivica. Examination of the type material of Peltodoris nayarita Ortea and Llera 1981,
another eastern Pacific species, confirmed that it is a synonym of Peltodoris greeleyi
MacFarland, 1909, originally described from Brazil. The study of Discodoris aurila
Marcus, 1976 from Panama and Costa Rica, shows that this species is characterized
by the absence of jaw elements, the presence of hamate and smooth radular lateral
teeth, and the innermost teeth hamate, elongated, and lacking denticles.

Peltodoris greeleyi and Discodoris aurila are transferred to the genus Diaulula
based on the presence of caryophyllidia, low rhinophoral and branchial sheaths, a
flattened prostate divided into two portions, penis and vagina unarmed, hamate
radular teeth and smooth labial cuticle. The geographic range of Diaulula greeleyi is
extended to Punta Uvita, Costa Rica and the geographic range of Diaulula aurila is
extended from Mexico to Panama.

RESUMEN

La especie de dorido nudibranquio del Pacifico Este tropical Taringa aivica Ev.
Marcus y Er. Marcus, 1967 es redescrita en base al estudio de especimenes recolec-
tados en la costa Pacifica de Costa Rica. Las diferencias entre Taringa aivica timia
Ey. Marcus y Er. Marcus, 1967 y Taringa aivica aivica Ev. Marcus y Er. Marcus, 1967
no son consistentes y no justifican la existencia de dos subespecies diferentes de
Taringa aivica. El estudio del material tipo de Peltodoris nayarita Ortea y Llera, 1981,
otra especie del Pacifico Este, ha confirmado que ésta es un sinénimo de Peltodoris
greeleyi MacFarland, 1909, que fue originalmente descrita de Brasil. El estudio de
Discodoris aurila Marcus, 1976 de Panama y Costa Rica, muestra que esta especie se
caracterizada por la ausencia de uncinos, y la presencia de dientes laterales de la
radula ganchudos y lisos, dientes centrales ganchudos, alargados y sin denticulos.
Peltodoris greeleyi y Discodoris aurila son transferidas al género Diaulula en base a
la presencia de cariofilideos, vainas rinoforicas y branquiales no elevadas, una pros-
tata aplanada dividida en dos regiones, pene y vagina lisos, dientes de la radula en
forma de gancho y cuticula labial lisa. La distribucién geografica de Diaulula gree-
leyi es extendida hasta Punta Uvita, Costa Rica y la distribucion geografica de
Diaulula aurila es extendida desde México hasta Panama.

65

66 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 4

Intensive field work along the Pacific coast of Costa Rica has revealed the presence of three
species of caryophyllidia-bearing dorids. The species, Zaringa aivica Ev. Marcus and Er. Marcus,
1967, Peltodoris nayarita Ortea and Llera, 1981 and Discodoris aurila Marcus, 1976 have already
been reported from other areas in the eastern Pacific, but their anatomy was poorly described and
the presence of caryophyllidia had been overlooked.

Valdés and Gosliner (2001) recently studied the phylogenetic relationships of the caryophyl-
lidia-bearing dorids, which according to these authors are a monophyletic group. They also syn-
onymized several genera previously considered as valid or regarded as uncertain. Therefore, the
caryophyllidia-bearing dorid species from Costa Rica need to be re-examined in light of the new
evidence.

The objective of the present paper is to re-examine the three described species of the
caryophyllidia-bearing dorids present in Costa Rica.

The material examined is deposited at the Department of Invertebrate Zoology and Geology,
California Academy of Sciences, San Francisco (CASIZ), Instituto Nacional de Biodiversidad,
Costa Rica (INBio), the Natural History Museum of Los Angeles County (LACM), Muséum
National d’Histoire Naturelle, Paris (MNHN), and the National Museum of Natural History
(USNM).

SPECIES DESCRIPTION

Genus Taringa Er. Marcus, 1955
Type species: Taringa telopia Er. Marcus, 1955

Taringa aivica Ey. Marcus and Er. Marcus, 1967
(Figs 1A-B, 2A-D, 3A-E)

Taringa aivica Marcus and Marcus, 1967:89—92, figs. 115-119.
Taringa aivica timia Marcus and Marcus, 1967:189-191, figs. 47-51; Behrens and Henderson, 1982:197-199,
figs. 1-4.
MATERIAL EXAMINED

San Miguel, Cabo Blanco, Costa Rica, May 16, 1998, 1 specimen, 12 mm preserved length, 2
m depth, leg. A. Berrocal (INB0001496644); San Miguel Station, Cabo Blanco, Costa Rica,
January 28, 1999, 2 specimens, 30-31 mm preserved length, intertidal, leg. F. Alvarado
(INB0001496686); San Miguel Station, Cabo Blanco, Costa Rica, January 28, 1999, 11 specimens,
10-20 mm preserved length, intertidal, leg. F. Alvarado (INB0001496491); San Miguel Station,
Cabo Blanco, Costa Rica, January 22, 1999, 18 specimens, 9-23 mm preserved length, intertidal,
leg. F. Alvarado (INB0001496522); San Miguel Station, Cabo Blanco, Costa Rica, May 17, 1999,
| specimen, 4 mm preserved length, 2 m depth, leg. S. Avila (INB0001496524); Playa Coralito,
Penon del Coral, Puntarenas, Costa Rica, January 29, 1999, 1 specimen, 12 mm preserved length,
intertidal, leg. F. Alvarado (INB0001496675); Punta Uvita, Puntarenas, Costa Rica, January 15,
2000, 9 specimens, 4-12 mm preserved length, intertidal, leg. M. Calderén (INB0001496174);San
Pedrillo, Osa Peninsula, Costa Rica, January 19, 2000, | specimen, 6 mm preserved length, inter-
tidal, leg. A. Berrocal (INB0001496548); San Pedrillo, Osa Peninsula, Costa Rica, January 15,
2000, 1 specimen, 16 mm preserved length, intertidal, leg. M. Calderén (INB0001495952):; San
Pedrillo, Osa Peninsula, Costa Rica, February 27, 1998, 3 specimens, 9-10 mm preserved length,

FiGuRE 1. Living animals. A-B. Taringa aivica (INB0001496491): C. Diaulula greeleyi (INBO001496508); D. Diaulula
aurila (INB0001495896).

CAMACHO-GARCIA AND VALDES: DORID NUDIBRANCHS FROM COSTA RICA 67

68 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 4

intertidal, leg. A. Berrocal (INB000146490); San Pedrillo, Osa Peninsula, Costa Rica, February
27,1998, | specimen, 9 mm preserved length, intertidal, leg. A. Berrocal (INB0001496489); Playa
Gallardo, Golfo Dulce, Costa Rica, February 12,1997, 1 specimen, 16 mm preserved length, inter-
tidal, leg. S. Avila (INB0001496523); Punta Larga, Golfo Dulce, Costa Rica, November 29,1997,
6 specimens, 5-16 mm preserved length, intertidal, leg. M. Madrigal (INB0001496484); Punta
Larga, Golfo Dulce, Costa Rica, November 27, 1997, 2 specimens, 14-18 mm preserved length,
intertidal, leg. A. Berrocal (INBO001496525).

GEOGRAPHIC RANGE

From Palos Verdes, California (Behrens and Henderson, 1982) to Sonora, Mexico and the
Canal Zone, Panama (Marcus and Marcus, 1967).

EXTERNAL MORPHOLOGY

The body is oval to elongate (Fig. 1A—B). The dorsum is covered with caryophyllidia about
0.15 mm long (Fig. 2D).There are some larger, conical tubercles arranged in two rows on both sides
of the visceral hump. The dorsal color is variable ranging from pale yellow to dark brown. There
is no correlation between the size of the animal and the dorsal color. Normally specimens have
irregularly distributed darker patches on the dorsum; patches may be more densely arranged on the -
center of the dorsum. The larger tubercles are white or cream white in most specimens. There is a
row of white patches on both sides of the dorsal hump, around the larger tubercles, that may be
absent in some specimens and may be larger in others, almost occupying the entire mantle margin.
The rhinophores are dark gray or black with numerous white or cream white spots on the club. The
rhinophores have 17 lamellae. Occasionally there are white spots surrounding the rhinophoral
sheath. The rhinophoral apex is white or cream-white. The gill is composed of six tripinnate
branchial leaves. The leaves are yellowish or cream-white with minute dark spots. Ventrally, the
anterior border of the foot is grooved and notched (Fig. 3E).

ANATOMY

The labial cuticle is smooth. The radular formula is 32 x (39.0.39) in a 26 mm preserved length
specimen (INB0001496522). Rachidian teeth are absent (Fig. 2A). The innermost teeth are hamate
with a wide base and denticles. The lateral teeth are hamate, having a narrow base and a very con-
spicuous prolongation on the upper side (Fig 2B). There are four to nine denticles on each lateral
tooth. These denticles decrease in number towards the central part of the radula. The lateral teeth
increase in size gradually towards the medial portion of the half-row. There are three or four pecti-
nate outermost teeth (Fig. 2C). The stomach is oval and connects distally to the long intestine that
forms a loop and runs to the anal opening (Fig 3A).

The ampulla is very long and convoluted in the middle portion (Fig. 3C). It enters the female
glands near their nidamental opening. The deferent duct is long and convoluted, and enters the flat-
tened prostate. The proximal end of the deferent duct opens into a common atrium with the vagi-
na. The vagina is long and tubular in shape and almost as thick as the deferent duct. At its distal
end, the vagina connects to the large and oval bursa copulatrix. Both, the duct that leads from the
seminal receptacle to the bursa copulatrix and the vaginal duct, are joined. The pear-shaped semi-
nal receptacle is smaller than the bursa copulatrix (Fig. 3D). It is connected to the bursa copulatrix
by a long and coiled duct. The penial cuticle is conical or bell-shaped (Fig. 3B).

CAMACHO-GARCIA AND VALDES: DORID NUDIBRANCHS FROM COSTA RICA

B. Lateral teeth, scale bar= 10pm. C.

5

‘a (INB0001496522), SEM photographs. A. Inner lateral teeth, scale bar= 10um

INZA ALVIC

FIGURE 2. Tar
Outer lateral teeth, scale bar

69

20 um. D. Caryophyllidia, scale bar= 20um.

70 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 4

FIGURE 3. Anatomy of Taringa aivica (INB0001496522 ) A. Dorsal view of the internal organs, scale bar= 1mm; B.
Penis, scale bar= 0.25 mm; C. Reproductive system, scale bar= 1mm; D. Detail of the reproductive system, scale bar=l1mm;
E. Ventral view of the mouth area, scale bar= 1mm. Abbreviations: am=ampulla; b=blood gland; bc=bursa copulatrix;
dd=deferent duct; dg=digestive gland; fg=female gland; h=heart; i=intestine; ot=oral tube; pr=prostate; sh=syrinx; sr=sem-
inal receptacle; st=stomach; t=oral tentacle; v=vagina.

CAMACHO-GARCIA AND VALDES: DORID NUDIBRANCHS FROM COSTA RICA Th

REMARKS

The study of several specimens and the review of the original description confirms that the
external and internal features of this species fit with those of the genus Jaringa. The presence of a
flattened prostate with two portions, an unarmed vagina, a penis armed with a cuticular structure,
inner and mid-lateral hamate radular teeth and pectinated outermost teeth are characteristics of this
genus (Valdés and Gosliner 2001).

According to Marcus and Marcus (1967) the differences between Taringa aivica aivica and
Taringa aivica timia are the presence of an outermost pectinated tooth with a broad spine in
Taringa aivica timia that is not present in 7: aivica aivica and the strong denticles of the lateral teeth
in T: aivica aivica. Also, they found uniformly distributed caryophyllidia and sometimes bicuspid
papillae in Taringa aivica timia, and very small caryophyllidia and conical papillae in Taringa aivi-
ca.

Behrens and Henderson (1982) reported Taringa aivica timia from Palos Verdes, California,
and provided additional anatomical details. In this study Behrens and Henderson found a bell-
shaped papilla in the penial cuticle and 5 to 10 pointed denticles on the outer side of their cusp, but
apparently the broad spine mentioned by Marcus and Marcus (1967) is missing.

In the present study, we have found great variability in dorsal coloration as well as in the
arrangement of the caryophyllidia and papillae (Figs. 1A—B). A spine on the outermost pectinated
teeth was not found in the material study and is also absent in the specimens assigned to Taringa
aivica timia by Behrens and Henderson (1982). The presence or absence of this spine as well as
the shape of the penial cuticle are most likely due to intraspecific variation. We consider that the
features mentioned by Marcus and Marcus (1967) are not enough to justify the existence of two
different subspecies and Taringa aivica timia and Taringa aivica aivica should be synonymized.

Genus Diaulula Bergh, 1878
Type species: Doris sandiegensis Cooper, 1863

Diaulula greeleyi (MacFarland, 1909)
(Figs. 1C, 4A-D, 5A-D)

Peltodoris greeleyi MacFarland, 1909:84-88, pl.15, figs. 77-82; Marcus, 1955; Marcus and Marcus, 1967;
Eyster, 1980:588.
Peltodoris nayarita Ortea and Llera, 1981:47—51, figs. 1-4 (also cited as Anisodoris).

MATERIAL EXAMINED

Ho oryPe of Peltodoris greeleyi: Alagoas, Riacho Doce, Brazil, July 28, 1899, 1 specimen, 9
mm preserved length, leg. A.W. Greeley. (CASIZ 21021).

Ho.otyee of Peltodoris nayarita: Isabel Island, Nayarit, Mexico (21°52’ N, 105°54’ W), 1
specimen, 22 mm preserved length, contracted and dissected (MNHN).

ADDITIONAL MATERIAL EXAMINED

San Miguel Station, Cabo Blanco, Costa Rica, May 17, 1998, 1 specimen, 10 mm preserved
length, 2 m depth, leg S. Avila (INB0001500647); San Miguel Station, Cabo Blanco, Costa Rica,
January 26, 1999, 22 specimens, 3-13 mm preserved length, intertidal, leg. F. Alvarado
(INB0001496509); San Miguel Station, Cabo Blanco, Costa Rica, January 22, 1999, 39 specimens,
2-16 mm preserved length, intertidal, leg. F. Alvarado (INB0001496508); Ballena Island, Uvita,

‘woz =1eq g[eos ‘eipypAydodreD *q ‘wmMo] =1eqg ayRos ‘yVA9} [RsaI] 191NQ “OD
‘wing =1eq gyROs ‘YQ99) [RAOWeT] “g ‘uM QT =Ieq gfeOs ‘yJ99) [RJOIe] JOUUT “YY “sydessojoyd WAS “(SOS96FIOOOAND !4a/aa13 vynynviq “> ANNOY

q

14 ' ea

Volume 54, No. 4

PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES

WZ

CAMACHO-GARCIA AND VALDES: DORID NUDIBRANCHS FROM COSTA RICA V2

Puntarenas, Costa Rica, January 16, 2000, 3 specimens, 2-6 mm preserved length, 6 m depth, leg.
M. Calderon (INB0001496530); Punta Uvita, Puntarenas, Costa Rica, January 15, 2000, 7 speci-
mens, 5—9 mm preserved length, intertidal, leg. M. Calderon (INBO001496515); Playa Ventanas,
Puntarenas, Costa Rica, January 17, 2000, 5 specimens, 5—6 mm preserved length, intertidal, leg.
M. Calderén (INB0001496529); San Pedrillo, Osa Peninsula, January 27, 1998, 1 specimen, 11
mm preserved length, intertidal, leg. A. Berrocal (INBO001496507); San Pedrillo, Osa Peninsula,
Costa Rica, January 20, 2000, 2 specimens, 2—10 mm preserved length, intertidal, leg. M. Calderén
(INB0001496521):; San Pedrillo, Osa Peninsula, Costa Rica, January 19, 2000, 2 specimens, 10-11
mm preserved length, intertidal, leg. A. Berrocal (INB00014965 16).

GEOGRAPHIC RANGE

This species is found in Florida (Marcus and Marcus 1967), Brazil (Marcus 1955), South
Carolina (Eyster 1980), Nayarit, Mexico (Ortea and Llera 1981), Punta Eugenia, Baja California,
México (Bertsch et al. 2000) and the Pacific coast of Costa Rica.

EXTERNAL MORPHOLOGY

The body is oval to elongate (Fig. 1C). The dorsum is covered with long caryophyllidia, about
100 mm long (Fig. 4D). The body is pale yellow to orange. The dorsum is covered with a number
of brown patches that may be darker in some specimens. These patches are more densely arranged
near the mantle edge. On the mantle edge there are some large, opaque white patches. The
rhinophoral sheaths are very inflated and pale cream white or white in color. The gill sheath is also
pale or white and in some specimens it is edged by a thin brown line. The rhinophores are pale yel-
low, with the club brown and the apex opaque white or pale yellow. There are 13 lamellae present
in the rhinophores. The gill is composed of 12 unipinnate branchial leaves. They are yellow to dark
brown. In the living animal the branchial leaves are oriented inwards. Ventrally the anterior border
of the foot is grooved and notched. The oral tentacles are short and conical (Fig. 5D).

ANATOMY

The labial cuticle is smooth. The radular formula is 37 x (55.0.55) in a 10 mm preserved length
specimen (INB0001496508). Rachidian teeth are absent (Fig. 4A). The lateral teeth are hamate,
having a single cusp and lacking denticles (Fig 4B). The teeth increase in size gradually towards
the medial portion of the half-row. The outermost teeth are very small and elongate, also lacking
denticles (Fig. 4C). The esophagus is long and connects directly to the stomach (Fig. 5A).

The ampulla is very long (Fig. 5B). It enters the female glands near their nidamental opening.
The prostate is flattened and granular. It is divided into two different portions that are clearly dis-
tinguishable by their different texture and coloration. The deferent duct is long, and expands into
the wide ejaculatory portion. The deferent duct opens into a common atrium with the vagina. There
are no penial hooks. The vagina is long and wide. At its distal end, the vagina connects to the large
and rounded bursa copulatrix. Another duct, which connects to the seminal receptacle and the uter-
ine duct, leads from the bursa copulatrix. The bursa copulatrix is about ten times larger than the
seminal receptacle (Fig. 5B—C).

REMARKS

MacFarland (1909) described Peltodoris greeleyi from Brazil. Later, Marcus (1955) and
Marcus (1967) redescribed this species from the same locality. According to these authors, the main

74 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 4

FiGure 5. Anatomy of Diaulula greeleyi (INBO001496508). A. Dorsal view of the internal organs, scale bar= 1mm; B.
Reproductive system, scale bar = 1mm; C. Detail of the reproductive system, scale bar= 1 mm; D. Ventral view of the mouth
area, scale bar= 1mm. Abbreviations: am=ampulla; b=blood gland; bc=bursa copulatrix; dd=deferent duct; dg=digestive
gland; fg=female gland; h=heart; i=intestine; ot=oral tube; pr=prostate; sh=syrinx; sr=seminal receptacle; st=stomach;
t=oral tentacle; v=vagina.

distinctive features of this species are the yellowish to orange color of the living animals with some
small brown spots on the center and sides, the prominent branchial and rhinophoral sheaths, the
smooth labial cuticle, the absence of rachidian teeth, the unipinnate gills, the outermost tooth small-
er than the remaining teeth, the absence of denticles and the outermost, and midlateral teeth with
cups.

Ortea and Llera (1981) described the species Peltodoris nayarita from Nayarit, Mexico.
Peltodoris nayarita is characterized by having a yellow mantle with small brown spots all over the
dorsum and sometimes concentrated in the middle portion, high branchial and rhinophoral sheaths,
smooth labial cuticle, unipinnate branchial leaves, outermost teeth smaller and outermost midlat-
eral teeth with a cusp. Ortea and Llera (1981) placed this species in the genera Peltodoris and
Anisodoris at the same time with no clear explanation. They also considered that Peltodoris and
Anisodoris are synonyms.

By studying the type material of Peltodoris greeleyi and Peltodoris nayarita, we found that
both species share the presence of a smaller outermost tooth with a single cusp and no denticles,
hamate lateral teeth with no denticles and a single cusp, and outermost teeth smaller than the

CAMACHO-GARCIA AND VALDES: DORID NUDIBRANCHS FROM COSTA RICA 75

remaining ones. There are several other features shared by the type specimens of these two taxa.
The coloration of the living animals is orange to light brown with small brown spots covering the
mantle, the gills are unipinnate, the labial cuticle is smooth and the rhinophores and branchial
sheaths high (MacFarland 1909; Marcus 1955; Marcus and Marcus 1967; Ortea and Llera 1981).

In the reproductive system drawn by MacFarland (1909), Marcus and Marcus (1967) and our
material from Costa Rica, the deferent duct has a loop and is longer than the slender vagina, the
bursa copulatrix is larger than the seminal receptacle and both are rounded in shape. However, all
of these features are difficult to observe in the incomplete drawing by Ortea and Llera (1981, fig
4).

When comparing all these features with the type material of P. greeleyi, P. nayarita, and the
material studied from Costa Rica, we concluded that these two species are synonyms.

This species is transferred to the genus Diaulula due to a unique combination of features: the
presence of caryophyllidia, the low rhinophoral and branchial sheaths, the penis and vagina
unarmed, the labial cuticle smooth, and the radular teeth hamate and smooth (see Valdés and
Gosliner 2001). Diaulula greeleyi is also different from D. aurila in radular morphology, external
coloration, the presence of high rhinophoral and branchial sheaths and unipinnate branchial leaves.

Diaulula aurila (Marcus, 1976)
(Figs 1D, 6A-D, 7A-D)

Discodoris aurila Marcus, 1976: 85-87; figs 108-111

MATERIAL EXAMINED

SYNTYPES of Discodoris aurila: Deale Beach (Ft. Kobbe Beach), Canal Zone, Panama,
December 1962, 2 specimens 15-20 mm preserved length, leg. F. Bayer and R. Bayer (USNM
576268). Two microslides: F-814 (USNM 576268) and F—-185 (USNM 576268).

ADDITIONAL MATERIAL EXAMINED

San Miguel Station, Cabo Blanco, Costa Rica, January 28, 1999, 3 specimens, 6—14 mm pre-
served length, intertidal, leg. F Alvarado (INB0001482472); San Miguel Station, Cabo Blanco,
Costa Rica, January 26, 1999, 1 specimen, 12 mm preserved length, intertidal, leg. F. Alvarado
(INB0001496506). San Miguel Station, Cabo Blanco, Costa Rica, January 26, 1999, 1 specimen,
9 mm preserved length, intertidal, leg. F Alvarado (CASIZ 159788 ); San Miguel Station, Cabo
Blanco, Costa Rica, January 26, 1999, 1 specimen, 7 mm preserved length, intertidal, leg. F.
Alvarado (INB0001496505); Playa Ventanas, Puntarenas, Costa Rica, January 17, 2000, 1 speci-
men, 12 mm preserved length, intertidal, leg. M. Calder6n (LACM 2922); Ballena Island, Uvita,
Puntarenas, Costa Rica, January 16, 2000, 13 specimens, 6-10 mm preserved length, 6 m depth,
leg. M. Calder6n (INB0001495896); Punta Uvita, Puntarenas, Costa Rica, January 15, 2000, 1
specimen, 8 mm preserved length, intertidal, leg. M. Calderén (INB0001496574); Playa Coralito,
Penon del Coral, Costa Rica, January 29, 1999, 1 specimen, 14 mm preserved length, intertidal,
leg. F. Alvarado (INB0001496280); Punta Voladera, Golfo Dulce, Costa Rica, February 12, 1997,
2 specimens, 8 mm preserved length, 6 m depth, leg. M. Calderén (INB0001496510); Punta
Voladera, Golfo Dulce, Costa Rica, February 11, 1997, 2 specimens, 7-9 mm preserved length,
intertidal, leg. Manuel Lobo (INB0001487296); San Pedrillo, Osa Peninsula, Costa Rica, January
21, 2000, 1 specimen, 10 mm preserved length, intertidal, leg. M. Calderon (INB0001495947).

umgz=ieq gyeos ‘eipryAydodred “gq UM] =1eq aes ‘YI99) [e.a}L] 19INE
‘) ‘uMo] =1eq a[vos ‘yII0} [RIOT] “g ‘UMQ] =I1eq o[eOS ‘YI99) [eIO}e] JoUUT “YW ‘sydersojoyd FAS “(TLPTRPIOOOAND V/LY vinjnvig *9 ANANDA

4,

Volume 54, No. 4

N
ea
O
Z
=
O
YN
ea
'e)
4
=
ea
()
<
UO
<
Z
©
=
O
ea
an
=
jam
e)
N
Oo
Z
Q
ea
ea
Oo
©
(a
a

CAMACHO-GARCIA AND VALDES: DORID NUDIBRANCHS FROM COSTA RICA 77

FicureE 7. Anatomy of Diaulula aurila (INBO001482472). A. Dorsal view of the internal organs, scale bar= 1mm; B.
Reproductive system, scale bar= 1mm; C. Detail of the reproductive system, scale bar= 1mm; D. Ventral view of the mouth
area, scale bar: Imm. Abbreviations: am=ampulla; b=blood gland; bc=bursa copulatrix; dd=deferent duct; dg=digestive
gland: fg=female gland; h=heart; i=intestine; m=retractor muscle; ot=oral tube; pr=prostate; sh=syrinx; sr=seminal recep-
tacle; st=stomach; t=oral tentacle; v=vagina.

GEOGRAPHIC RANGE

This species in known from the Pacific coast of Costa Rica (present study) and Panama
(Marcus, 1976).

EXTERNAL MORPHOLOGY

The body is oval (Fig. 1D). The dorsum is covered with long caryophyllidia, about 190 mm
long (Fig. 6D). The body is pale gray, with a difuse, thick, black line in the middle of the dorsum,
running from the rhinophores to the gill. This line is composed of a number of small black dots.
The viscera are visible as a pale brown patch in the center of the dorsum. There are some black dots
and numerous white dots on the mantle margin. In some specimens the black dots may line up to
form longitudinal, diffuse lines. The rhinophores have 17 lamellae. They are translucent white with
the club dark yellow and the apex and rachis opaque white. The gill is composed of 6 tripinnate
branchial leaves. They are gray to dirty yellow with some orange, apparently glandular spots.
Ventrally the anterior border of the foot is grooved and notched. The oral tentacles are short and
slender (Fig. 7D).

78 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 4

ANATOMY

The labial cuticle is smooth. The radular formula is 20 x (39.0.30) in a 15 mm preserved length
specimen (INB000148247). The radula has no rachidian teeth (Fig. 6A). The lateral teeth are
hamate and smooth (Fig. 6B). The innermost teeth are also hamate and very elongated, lacking
denticles. The outermost teeth are small, elongated and smooth. However, sometimes the tooth has
up to 5 denticles (Fig. 6C). The esophagus is long and connects directly to the stomach (Fig. 7A).

The ampulla is very long and convoluted in the middle portion and divides into the short
oviduct and prostate (Fig. 7B). It enters the female glands near the middle of the mass. The prostate
is large, flattened and granular. It narrows into the short and coiled deferent duct, which expands
again into the muscular and wide ejaculatory portion. The deferent duct opens into a common atri-
um with the vagina. The vagina is long and tubular. The penis is unarmed. The uterine duct is short
and thin, and opens near the center of the female gland mass. The seminal receptacle is slightly
oval, almost as large as the pyriform bursa copulatrix (Fig. 7C).

REMARKS

The anatomical examinations indicate that this species should be placed in the genus Diaulula,
which is characterized by the presence of a dorsum covered with elongated caryophyllidia, low
rhinophoral and branchial sheaths, a flatenned prostate divided into two portions, penis and vagi-
na unarmed, radular teeth hamate and smooth, and a labial cuticle smooth (see Valdés and Gosliner
2001). Marcus (1967) described Discodoris aurila as having jaw elements; however in the speci-
mens studied here from Costa Rica and the microslide of the syntype (F—184) mounted by Marcus
herself, the labial cuticle is smooth.

Other species of Diaulula from the eastern Pacific are Diaulula punctuolata (D’ Orbigny 1837,
Diaulula sandiegensis (Cooper 1863) and Diaulula greeleyi (MacFarland 1909). Diaulula aurila
is different from D. punctuolata and D. sandiegensis in the external coloration. The general color
of the body of Diaulula aurila is pale gray, with a line in the middle of the dorsum composed of a
number of small black dots as well as black or white dots on the mantle margin. In Diaulula punc-
tuolata the dorsum is white to yellowish and has two rows of small dots between the rhinophores
and gill. These small dots can be light or dark in color (Schrédl 1996). In Diaulula sandiegensis,
the general color of the living animals varies from white to pale brown. The dorsum has several
dark brown or dark rings of various sizes often aligned in two longitudinal rows, one on either side
of the body (Valdés and Gosliner 2001). According to Behrens and Valdés (2001) Diaulula
sandiegensis sometimes has denticles on the two outermost lateral teeth, a feature that is not always
present. As in D. sandiegensis, we have found some denticles on the outermost teeth in some spec-
imens of D. aurila. Diaulula aurila is also different from D. greeleyi in radular morphology, exter-
nal coloration, presence of low rhinophoral and branchial sheaths and tripinnate branchial leaves.

ACKNOWLEDGMENTS

The field work in Costa Rica was founded by the Instituto Nacional de Biodiversidad (INBio)
through the Netherlands Government and the project “Development of Biodiversity Knowledge
and Sustainable Use in Costa Rica.” This research was also possible thanks to the Cooperation
Agreement between the Ministry of Environment and Energy (MINAE) and INBio. We are very
grateful to Socorro Avila, Francisco Alvarado, Alcides Berrocal and Mario Calderén, who assist-
ed during the field work in Costa Rica. Terrence M Gosliner took the photographs of the living ani-
mals. The SEM work was carried out in the facilities of the California Academy of Sciences. The
Muséum National d’ Histoire Naturalle of Paris and the National Museum of Natural History made

CAMACHO-GARCIA AND VALDES: DORID NUDIBRANCHS FROM COSTA RICA TD

available the holotype of Peltodoris nayarita and the syntypes of Discodoris aurila respectively.
Benoit Dayrat made constructive comments to this manuscript.

This paper has been supported in part by the National Science Foundation through the PEET
grant DEB—9978155, “Phylogenetic systematics of dorid nudibranchs,” to Terrence M. Gosliner
and Angel Valdés, and the Lakeside Foundation through a scholarship to the senior author at the
California Academy of Sciences.

LITERATURE CITED

BEHRENS, D.W. AND D.R. HENDERSON. 1982. Taringa aivica timia Marcus and Marcus, 1967 (Nudibranchia:
Doridacea) in California. Veliger 24: 197-199.

BEHRENS, D.W. AND A. VALDEZ. 2001. The identity of Doris (s.l.) species MacFarland, 1966 (Mollusca,
Nudibranchia, Discodorididae): A persistent mystery from California solved. Proceedings of the California
Academy of Sciences 52:183-193.

BerTSCH, H., O. ANGULO CAMPILLO, AND J.L. ARREOLA. 2000. New distributional records of Opisthobranchs
from the Punta Eugenia region of the Baja California Peninsula: A report based on 1997-1998 CONABIO-
Sponsored expeditions. The Festivus 32(7):99-104.

EystTer, L.S. 1980. Distribution and reproduction of shell-less opisthobranchs from South Carolina. Bulletin
of Marine Science 30:580-599.

MacFAarLAND, F M. 1909. The opisthobranchiate Mollusca of the Branner-Agassiz expedition to Brazil.
Leland Stanford Junior University Publications, University Series (2):1—104, pls. 1-19.

Marcus, E.R. 1955. Opisthobranchia from Brazil. Boletim da Faculdade de Filosofia, Ciencias e Letras,
Universidade de Sao Paulo, Zoology 20:89-261.

Marcus, E.V. AND E.R. Marcus. 1967. American opisthobranchs mollusks. Studies in Tropical
Oceanography. Miami 6:1-256.

OrTEA, J. AND E.M. LLerRA. 1981. Un nuevo doérido (Mollusca: Nudibranchiata) de la Isla Isabel, Nayarit,
México. [berus 1:47-51.

SCHRODL, M. 1996. Nudibranchia y Sacoglossa de Chile: Morfologia externa y distribuci6n. Gayana
Zoologica 60:17-62.

VALDES, A. AND T.M. GOSLINER. 2001. Systematics and phylogeny of the caryophyllidia-bearing dorids
(Mollusca, Nudibranchia), with descriptions of a new genus and four new species from Indo-Pacific deep
waters. Zoological Journal of the Linnean Society 133:103-198.

Copyright © 2003 by the California Academy of Sciences
San Francisco, California, U.S.A.

80

PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. X

PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES

Volume 54, No. 5, pp. 81—132, 6 figs., 3 tables March 14, 2003

Systematics and Osteology of Leptoglaninae
a New Subfamily of the African Catfish Family
Amphiliidae, with Descriptions of Three New
Genera and Six New Species

Tyson R. Roberts

Research Associate, Smithsonian Tropical Research Institute

Leptoglaninae (type species Leptoglanis xenognathus Boulenger, 1902) is proposed as
a new subfamily of the African catfish family Amphiliidae. It is characterized by an
encapsulated swim bladder with a transverse bony intercapsular bridge between the
two lateral capsules as well as by distinctive coloration and behavior. In contrast to
other amphiliids, all or most of which are rheophilic or even torrenticolous, typical-
ly living in rocky or stony habitats, leptoglanins are arenicolous (sand-dwelling).
Several taxa have been observed buried in sand with just the eyes exposed or diving
into sand. Four of the five genera and 14 of the 16 known species are endemic to the
Congo basin. The Congolese genus Leptoglanis comprises a single species, L. xenog-
nathus. \t differs from all other amphiliids in its highly specialized mouth and jaws
and soft digitiform pharyngeal processes and from all other leptoglanins except
Zaireichthys rotundiceps in having the relatively primitive principal caudal fin ray
count of 7+8.

The Congolese species formerly known as Leptoglanis bouilloni, although it super-
ficially looks like a short-bodied relative of L. xenognathus, does not have an encap-
sulated swim bladder and is not a leptoglanin or even an amphiliid, but rather a
bagrid or claroteid. The species previously known as Leptoglanis camerunensis, L.
mandevillei, and L. rotundiceps are placed in the formerly monotypic leptoglanin
genus Zaireichthys (type species Z. zonatus Roberts, 1967). Zaireichthys differ from
all other leptoglanins and amphiliids in having dorsal and pectoral fins with stout
spines and locking mechanisms, thus forming the classical ‘‘defensive tripod” of
primitive and generalized catfishes. Zaireichthys is the only amphiliid with stout ser-
rae on the pectoral fin spine. The leptoglanin Tetracamphilius pectinatus also has pec-
toral fin serrae, but they are much smaller and the pectoral spine does not have a
locking mechanism. Leptoglanis brevis is placed as a junior synonym of Z. rotundi-
ceps. The Zaireichthys rotundiceps species complex is characterized by having the
pectoral girdle with a large humeral process more or less extensively covered with
sharp conical denticulations. These arise indistinguishably from the bony surface of
the humeral process and are not true teeth. The only other leptoglanin in which such
denticulations have been observed is Z. zonatus, which has a few small ones on the
humeral process of its pectoral girdle.

Tetracamphilius new genus, comprising the species formerly known as Amphilius
angustifrons and A. notatus and two new species T: pectinatus and T. clandestinus, is
distinguished from all other amphiliid genera by tiny fan-shaped jaw teeth with up
to four cusps rather than jaw teeth conical or absent. The species of Tetracamphilius
all have 6+7 principal caudal fin rays. The exceptionally elongate and otherwise dis-
tinctive leptoglanin formerly known as Leptoglanis brieni is designated type species

81

82 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 5

of a new genus, Dolichamphilius. A new species that might be its closest relative is
placed tentatively in Dolichamphilius, D. longiceps. These two species also have 6+7
principal caudal fin rays. The new genus Psammphiletria, comprising two new
species, P. nasuta and P. delicata, differs from all other leptoglanins in having the dor-
sal fin remote from the cranium and in the extreme reduction of its caudal fin rays,
with fewer procurrent rays than in any other leptoglanin and only 5+6 principal cau-
dal fin rays, the lowest count known in Amphiliidae.

Leptoglanis xenognathus Boulenger, 1902, type species of Leptoglanis Boulenger, 1902, is a
highly distinctive catfish found only in the Congo basin with specializations not found in other
known taxa. Leptoglanis has had a long history as a “catch-all” genus. This began with the assign-
ment of Gephryoglanis rotundiceps Hilgendorf, 1905 to Leptoglanis by Boulenger (1911). Several
more highly distinctive species subsequently were described in Leptoglanis by Pellegrin (1926);
and by Poll (1959, 1967). Most of these are amphiliids, but “Leptoglanis” bouilloni Poll (1959)
belongs in Bagridae or Claroteidae.

Previous failure to recognize the heterogeneous nature of Leptoglanis (sensu lato) can be
attributed to lack of osteological information. Osteological study was impeded until very recently
by scarcity of material, and before that by inadequate methods of preparing small fish specimens
for osteological study. This is now somewhat less of a problem, and it has been possible to study
cleared and stained osteological preparations of almost all of the leptoglanin taxa. This study treats
all taxa that are closely related to Leptoglanis. These are recognized herein as forming the subfam-
ily Leptoglaninae of Amphiliidae. Included are Zaireichthys zonatus, the species formerly known
as Amphilius angustifrons, and Amphilius notatus, and several previously undescribed species.

The leptoglanin genera Leptoglanis and Zaireichthys were described originally as Bagridae. A
significant advance in their classification came when it was recognized that they are Amphiliidae
(Bailey and Stewart 1984:9). These authors based their conclusion on examination of the type
species of Leptoglanis and Zaireichthys. A similar conclusion, utilizing more osteology, was
reached by Mo (1991:68—73) but was based on species that are not congeneric with Leptoglanis
and Zaireichthys. For his concept of the genus Leptoglanis Mo examined “Leptoglanis” (now
Zaireichthys) rotundiceps; for Zaireichthys he examined “Zaireichthys rhodesiensis,’ an unpub-
lished taxon that is either a junior synonym or a very close relative of Z. rotundiceps. Not surpris-
ingly, his observations indicated that the two taxa are sister species. Nevertheless, Mo looked at
leptoglanin osteology in greater detail than previous authors and reported several distinctive char-
acters, notably the posteriorly triradiate palatine bone characteristic of all Amphiltidae.

Many of African Bagridae have numerous large, fleshy fingerlike (“digitiform’”) processes in
the pharynx and especially on the branchial arches; Asian Bagridae generally lack such structures
(pers. obser.). I shared this information with Mo, who published it without acknowledgment as part
of his evidence for non-monophyly of the Bagridae and for a sister-group relationship of the sub-
families Claroteinae and Auchenoglanidinae of the African family Claroteidae (Mo 1991:62). What
I also pointed out to him, but what he failed to report, is that apparently identical structures are
found in the amphiliid genus Leproglanis. If these structures (which are generally distributed on the
pharynx and not just on the gill arches as indicated by Mo) are homologous and occur only in the
taxa in which they have been found thus far, they may be evidence (a synapomorphy) indicating
that Amphiliidae are related to (presumably derived from) the African Bagridae or Claroteidae.
This would lessen, although it would not eliminate, the possibility that Amphiliidae are most close-
ly related to other Old World (i.e., Asian) catfish groups such as Sisoridae (see He and Meunier
1998; He et al., 1999), Akysidae, and Parakysidae, which share with Amphiliidae the character

ROBERTS: LEPTOGLANINAE 83

complex of a bilaterally compartmented swim bladder enclosed in auditory bulla, or capsules, mod-
ified from parapophyses of vertebrae 4 and 5. Whether such a character complex is diagnostic of
an ancient monophyletic group of catfishes found in Africa, Asia, and South America or has
evolved independently at least three times is one of the major unresolved issues in siluroid higher
classification. In Africa it occurs only in the family Amphiltidae (Chardon 1968; Mo 1991: He et
al. 1999). Recognition of this character complex in Leptoglanis and Zaireichthys led to the trans-
fer of these genera from Bagridae to Amphiliidae by Bailey and Stewart (1984:9).

A ventral bony bridge (here named “‘intercapsular bridge’) linking the ossified swim bladder
capsules was found in Leptoglanis xenognathus, Zaireichthys zonatus, Z. rotundiceps, and
Amphilius notatus by Bailey and Stewart (1984:9) (Fig. 1). They did not find a similar bridge in
other amphiliids, and predicted that the taxa with the bridge represented a monophyletic assem-
blage. This monophyletic grouping, as they envisioned it, corresponds to the Leptoglaninae as rec-
ognized herein. An intercapsular bridge is present in all of the leptoglanin species for which oste-
ological study material has been available. However, a well-developed intercapsular bridge also is
present in the doumein amphiliid Andersonia leptura (pers. obser.; He et al. 1999:133, fig. 7), so it
is not a character unique to Leptoglaninae. The presence of the intercapsular bridge in Andersonia
and leptoglanins has led to the hypothesis that Doumeinae is the sister-group of Leptoglanis and
the leptoglanins (here Leptoglaninae) (op cit. 117,142). Their concept of Leptoglanis and lep-
toglanins is based upon study of a single species, Leptoglanis camerunensis (here Zaireichthys
camerunensis).

Most catfish families probably have ribs with a single head and relatively simple articulation.
In Leptoglaninae, however, the proximal end of each rib is “bicipital” or two-headed. The ribs
articulate with a lateral process or an anteroventral process arising from the corresponding verte-
bral centrum. In addition, at least in Leptoglanis xenoganthus, each rib is attached to an anterodor-
sal process of the corresponding vertebra by a ligament arising just below its bicipital head (Fig.
Dy:

Amphiliidae are bottom-dwelling catfishes ranging in size from 20 to 150 mm. Eleven genera
and about 64 species (including those described herein) are known. The better-known species live
in the parts of swift-flowing streams with rocky bottom including mountain streams and rapids of
large lowland rivers. These belong to the subfamilies Amphiliinae (comprising the genera
Amphilius and Paramphilius, with some 25 species) and Doumeinae (Andersonia, Belonoglanis,
Doumea, Phractura, and Trachyglanis, with some 23 species). These two subfamilies (especially
Amphiliinae) are very widely distributed is tropical Africa. The third subfamily, Leptoglaninae,
with 5 genera and at least 16 species, occurs mainly in the Congo basin. Its species live mainly in
rivers with extensive stretches of sandy bottom. Amphiliinae and Doumeinae frequently occur
together, but are seldom (almost never) found together with Leptoglaninae. Synonymies and infor-
mation on distribution of Amphiliidae are provided by Skelton and Teugels (1986). For an illustrat-
ed key to the genera see Poll and Gosse (1994:186—190).

Due to their small size, near restriction to the Congo basin, and tendency to inhabit large
stretches of open sandy habitat with low fish species diversity, Leptoglaninae are among the least
known fishes of the African ichthyofauna. When this study began, only two genera (Leptoglanis
and Zaireichthys) and about 10 valid species had been described. The present account adds three
new genera and six new species, and more taxa surely remain to be discovered. All of the lep-
toglanin genera are present in the Congo basin, as are 15 of the total of 16 species recognized here
(Fig. 3). Fourteen of the species are known only from the Congo basin and presumably are endem-
ic to it. The only known leptoglanin species not present in the Congo basin is Zaireichthys
camerunensis, known only from the Niger basin..

84 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 5

Knowledge of the biology of Leptoglaninae is virtually non-existent. Their feeding and repro-
ductive behavior have not been observed. Zaireichthys rotundiceps (as Leptoglanis rotundiceps)
and Zaireichthys cf. dorae (as Leptoglanis cf. dorae) were reported as occurring over fine sand,
“lying buried with only the eyes protruding” by Skelton (1993:218, 220). The actual sand-diving
behavior has not been previously recorded. At Banda (on the Ubangui River upstream from
Bangui) I placed several leptoglanins about one inch long in a small bowl with sand on the bottom.
When undisturbed the fish rested quietly on top of the sand. If mildly disturbed, they darted about
on the sand with incredible rapidity. When badly disturbed, they instantly disappeared below the
sand and stayed there for several minutes. While buried in the sand their eyes did not project.
Although the species thus observed were not identified in the field at Banda, they were at least two
and possibly three species, probably Zaireichthys mandevillei, Tetracamphilius angustifrons or T.
clandestinus, and perhaps Psammphiletria nasuta. On this and other occasions I noted the extraor-
dinarily copious mucus secretion exuding from the pectoral gland. When handling live specimens
at Banda, two of the fish slipped out of my fingers and were suspended by a long transparent strand
of mucus 10-12 inches long. The mucus strand was nearly equal in thickness to the body diameter
of the fish, and its origin from just above the pectoral fin (i.e., from the axillary pore of the pec-
toral gland) was observed quite clearly. Zaireichthys heterurus was observed diving into the sand,
apparently as an escape reaction, in the riviere Lulindi. I have observed in several species that the
eyes of live species have pupils with slit-like horizontal openings, at least in fish caught in bright
daylight. This is perhaps an adaptation related to their sand-diving behavior. Upon preservation the
slit-like shape of the pupils fairly rapidly disappears and they appear to be round. From superficial
observation of gut contents of cleared and stained specimens, leptoglanins appear to be mainly car-
nivores, preying on very small arthropods (insects and crustaceans).

This paper takes into consideration literature on Leptoglanis and Zaireichthys recorded in
Eschmeyer’s “On Line Fish Species Catalogue” up to 9 September 2002.

MATERIALS AND METHODS

MATERIAL EXAMINED.— Specimens used in this study are deposited in the following institu-
tions: AMNH, American Museum of Natural History, New York; BMNH, The Natural History
Museum, London; CAS, California Academy of Sciences, San Francisco; MNHN, Muséum
National d’ Histoire Naturelle, Paris; MRAC, Musée Royale pour I’ Afrique Centrale, Tervuren; and
UMMZ, University of Michigan Museum of Zoology.

Specimen lengths given throughout this paper are standard length, SL (from tip of snout to end
of hypural plate). Specimens for osteological study were macerated with KOH and stained with
alizarin, or macerated with trypsin and stained with alcian blue and alizarin. Radiographs were pre-
pared in dorsal as well as lateral view.

LocaLitIEs.— Locality data presented in this paper represents nearly all known samples of
every species of Leptoglaninae except for the species or species complex Zaireichthys rotundiceps.
This is the only species with an extensive range outside as well as inside the Congo basin, and it is
by far the most frequently collected leptoglanin. It is likely that Z. rotundiceps represents several
closely related species (pers. obser.; Skelton 1993; Seegers 1996). In presenting the localities, the
type locality or localities of the species is given first, generally followed by other localities approx-
imately grouped from N to S and from W to E. The locality dated are presented in the language and
wording of the original collector(s). My own original locality records for specimens collected in
the Congo basin are a mixture of English and French. Localities of leptoglanin specimens exam-
ined from the Congo basin are indicated in Fig. 3 (in some cases two or more localities very close

ROBERTS: LEPTOGLANINAE 85

together for the same species are indicated by a single symbol).

SYNONYMY.— Synonymies include primary synonymy (original descriptions and name pro-
posals) and secondary synonymy (name changes). Additional references may be included in a syn-
onymy if they have direct bearing on systematics and identification has been confirmed by exam-
ination of voucher specimens or can be determined from the reference itself. In secondary syn-
onyms and misidentifications the name is followed by a comma.

Counts.— Counts of gill rakers and vertebrae, previously unused by catfish workers, have
proved to be extremely useful in distinguishing species in group after group of catfishes: Akysis
(Roberts 1989a, table 9); Bagarius (Roberts 1983); Chiloglanis (Roberts 1989b); Gagata (Roberts
and Ferraris 1998); Kryptopterus (Roberts 1989a, table 10); Leiocassis (Roberts 1989a, table 5);
Pangasiidae (Roberts and Vidthayanon 1991); Malapterurus (Roberts, 2000) Mystus (Roberts
1989a, tables 7-8; Roberts 1992; Roberts 1994); Nangra (Roberts and Ferraris 1998); Ompok
(Roberts, 1989a, table 11); and Si/urichthys (Roberts 1989a, table 12).

GILL RAKERS.— Counts of gill rakers are extremely useful in distinguishing catfish species. In
most catfishes, as is usual in teleosts, the number of gill rakers becomes fixed at a fairly early stage,
but in some groups (e.g., Clarias, Mystus) there are species in which the number of gill rakers
increases throughout life. In Leptoglaninae, however, gill raker counts do not seem to be very use-
ful. Due to the small size of the specimens and the small size of the gill rakers, it is difficult or
impossible to obtain counts from whole specimens. Counts can be obtained only after the gill arch-
es have been removed from cleared and stained specimens. But the main reason gill raker counts
are not useful in Leptoglaninae is because the rakers are few in number as well as very small.

VERTEBRAE.— As in other catfishes, vertebral counts are useful for distinguishing species and
recognizing evolutionary trends in Leptoglaninae. It is informative to give counts of abdominal and
postabdominal (or “caudal’’) as well as total vertebrae. Counts given in this paper were all obtained
from cleared and stained specimens. They include four anteriormost vertebrae incorporated into the
Weberian apparatus. The hypural fan is counted as one vertebra. In amphiliids as in most catfishes
the first rib-bearing vertebra is almost invariably vertebra 6. Only one specimen of Leptoglaninae
has been noted in which vertebra 5 had ribs (the holotype of Dolichamphilius longiceps), and they
were much smaller than the ribs on vertebra 6.

In practice, whether counting vertebrae from cleared and stained specimens or from radi-
ographs, the count begins from the first vertebra bearing ribs, i.e. vertebra 6. Abdominal and
postabdominal vertebrae are distinguished by the relationship of their hemal spines to the anal fin
pterygiophores. Abdominal vertebrae are all those with hemal spines lying anterior to the anal fin
pterygiophore that extends furthest anteriorly. If a vertebra has its hemal spine with the tip exactly
meeting the anterior tip of the first anal fin pterygiophore, it is included in the count of abdominal
vertebrae. In practice the abdominal and postabdominal vertebrae are readily distinguished in
amphiliids by this method, whether in cleared and stained specimens or in radiographs. Vertebrae
in the caudal peduncle or peduncular vertebrae are those lying posterior to a vertical line through
the posteriormost anal fin pterygiophore. The terms used here and in other papers by me, abdomi-
nal and postabdominal, are anatomically correct and easily understood. The definitions are such
that they apply to all catfish groups and to nearly all groups of teleosts. One major advantage of
these simple definitions is that they can be readily used to obtain vertebral counts from radiographs.
Another is that they can be used to obtain counts from fossils of most teleosts, including catfishes
(Roberts and Jumnongthai 2000). Total vertebral counts of incomplete fossil fish specimens,
although often reported, are almost worthless. When part of a fossil is missing, exact counts often
can be obtained for either the abdominal or postabdominal vertebrae.

Counts of abdominal and postabdominal vertebrae are particularly useful in distinguishing cat-

86 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 5

fish species. In many instances related catfish species have the same total number of vertebrae
composed of quite different numbers of abdominal and postabdominal vertebrae.

FIN RAYS.— All of the rayed fins of amphiliids provide valuable meristic data for systematic
study except the pelvic fins, which usually have 6 rays on both sides in all species of Amphiliidae
(I have observed a few specimens with 7 rays on one side or the other, and one with only 5 rays on
one side). In Amphiliidae, and especially Leptoglaninae, the most useful fin ray counts are those
of the caudal fin. None of the amphiliids exhibit the most primitive count of principal caudal fin
rays known in catfishes, 9+9, reported only in the primitive Neotropical catfish family
Diplomystidae. But the family includes species with nearly all of the principal caudal fin ray counts
otherwise known in catfishes (see Lundberg and Baskin, 1969), from 8+9 to 5+6. Leptoglaninae
have counts ranging from 7+8 to 5+6. In several taxa the upper principal rays are more numerous
than the lower; the new species Zaireichthys heterurus has 7+5. Although caudal fin ray counts dif-
fer considerably within the family Amphiliidae, and also within some leptoglanin genera (cf.
Zaireichthys), they usually are constant within a species and in several instances within a genus.
The leptoglanin genus Tetracamphilius has five species, all with 6+7.

In the dorsal and anal fins, as in most teleosts, sometimes the last ray only and sometimes the
last two rays articulate with the posteriormost pterygiophore. The latter condition is usually (but
incorrectly) described as “last ray divided to base.” Depending upon the author the last ray divid-
ed to base may be included or excluded from the total count fin rays. The last ray, when separate-
ly notated, is referred to as “/2” (also incorrect, because it is in itself a complete ray). Here the con-
vention 1s followed of indicating the additional last ray, when it is present, by “2”.

Principal caudal fin rays may be defined as all of the branched rays articulating with the upper
and lower halves of the hypural fan, plus one upper and one lower unbranched ray. Principal rays
of the upper and lower caudal fin lobes are readily distinguished in Amphiliidae, since they artic-
ulate with either the upper or the lower half of the hypural fan and there is invariably a distinct gap
between them. All of the rays anterior to the uppermost and lowermost principal caudal rays are
the procurrent caudal fin rays. A caudal fin ray count given as 10—12,6/7,11—13 indicates 6 upper
and 7 lower principal rays, and 10—12 upper and 11—13 lower procurrent rays.

Scope.— This work represents a thorough revision of the subfamily Leptoglaninae to the
species level with the exception of the species complex Zaireichthys rotundiceps. The latter group
is found in eastern and southern Africa including Angola and also in the eastern part of the Congo
basin (Seegers, 1989; Skelton, 1993; Seegers, 1996; some new locality records for eastern part of
Congo basin herein). BMNH and MRAC have many samples that I have examined only superfi-
cially, mainly to verify that they have an elongate denticulated humeral process. Several institu-
tions in eastern and southern Africa have samples that I have not examined. The only other lep-
toglanin with an extra-Congolese distribution is Zaireichthys camerunensis in the Niger basin in
West Africa. Its distribution is thus widely disjunct from the rest of the known Leptoglaninae.
Leptoglanins are unknown from the Ogooué and other West African coastal basins.

My original intention was to publish a more extensively illustrated osteological account of the
leptoglanin taxa together with the systematic revision. Due to difficulties in completing observa-
tions on some of the taxa and other commitments the osteological work has been curtailed. Enough
has been done, however, to provide evidence for monotypy of the Leptoglaninae, reassign
“Leptoglanis” bouilloni to Bagridae or Claroteidae, and contribute to diagnoses of leptoglanin gen-
era and species.

ORDER OF TAXONOMIC PRESENTATION.— The taxa are presented in the following order:
Leptoglanis, type genus and earliest described genus of Leptoglaninae is given first, followed by
Zaireichthys, the only other previously described leptoglanin genus. After Zaireichthys the three

ROBERTS: LEPTOGLANINAE Q7

new genera Dolichamphilius, Psammphiletria, and Tetracamphilius are given in alphabetical order.
Within each genus the generic type species is treated first, followed by the other species (includ-
ing new species) in alphabetical order.

SUBFAMILY NAMES.— In this as in my other papers, the subfamily termination is given as “-in”
(rather than “-ine preferred by some authors), i.e. “leptoglanin,” rather than “leptoglanine”. This
might not be classically correct Latin, but since “-id” is now universally used instead of the more
classically correct “-ide” for families, it is consistent to use -in for subfamilies.

ILLUSTRATIONS.— For ease of use, the figures are grouped together in a separate section fol-
lowing the bibliography.

SYSTEMATIC ACCOUNT

LEPTOGLANINAE Roberts, new subfamily

Type species: Leptoglanis xenognathus Boulenger, 1911

DIAGNOsIS.— Amphiliid catfishes with the left and right osseous swim bladder capsules con-
nected ventrally by a transverse intercapsular bony bridge. Cranium narrow anteriorly. Anterior
fontanelle either small or absent; posterior cranial fontanelle absent. Branchiostegal rays 3-8. Gill
rakers poorly developed, a maximum of 10 rakers on leading edge of lower element of first gill
arch. Upper elements of gill arch without gill rakers (present in Amphiliinae). Outermost pectoral
and pelvic fin rays without elaborate lepidotrichia and unculiferous pads (both highly developed in
all Amphiliinae and most Doumeinae). Pectoral fin with only one rod- or plate-like ossified radial
element (most other amphiliids and other catfishes with two). Adipose fin invariably elongate. Ribs
relatively few, from 3 to 6 pairs (ribs also few in Doumeinae, more numerous in Amphiliinae).
Caudal fin, variable in form, from deeply forked to truncate. Principal caudal fin rays 7+8, 7+7,
6+7, 7+5, 6+5, or 5+6 (not 8+9). Principal caudal fin ray counts of Leptoglaninae and other
Amphiliidae are presented in Table 1.

COMMENTS.— Leptoglaninae comprises arenicolous or sand-dwelling amphiliids occurring
mainly in the Congo basin. They are all relatively small species. Leptoglanis xenognathus, the

TABLE |. Principal caudal fin ray counts in Leptoglaninae and other amphilids
(data from Lundberg and Baskin 1969; Skelton 1986, 1989, 1993; and pers. obser.)

n/n+1:

8+9=17 Amphilius baudoni, A. cryptobullatus, A. kivuensis, A. uranoscopus, A. zairensis, Doumea
alua, Paramphilius teugelsi, Phractura sp.

7T+8—15 Amphilius longirostris, A. opisthophthalmus, Belonoglanis sp, Doumea thysi, Leptoglanis
xenognathus, Paramphilius trichomycteroides, Zaireichthys rotundiceps

6+7=13 Amphilius atesuensis, A. brevis, A. lentiginosus, A. maesii, A. pictus, Dolichamphilius
brieni, D. longiceps, Paramphilius firestonei, Tetracamphilius angustifrons, T. notatus,
T. pectinatus, Zaireichthys zonatus

5+6=11 Psammphiletria delicata, P. nasuta

n/n:

74+7=14 Andersonia leptura, Zaireichthys camerunensis

n/n?1:

64+5=11 Zaireichthys mandevillei

n/n?2:

74+5=12 Zaireichthys heterurus

88 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 5

largest, attains only slightly over 60 mm in standard length. The smallest species, Psammphiletria
nasuta, P. delicata, and Tetracamphilius clandestinus, are only 20 mm or a bit longer. The largest
Zaireichthys are only 40 mm. At least some species dive completely into the sand (behavior
observed for several species collected by me in the Ubangui and elsewhere), an activity unknown
in other amphiliids. All or many of them have a large humeral or axial pectoral gland producing
mucus. Amphiliinae and the “naked” (plateless) doumein genus Doumea, on the other hand, are
known for their ability to inhabit torrential streams and to cling to rock substrates by means of their
paired fins. The armored (plate-bearing) genera of Doumeinae cling to submerged grassy vegeta-
tion in swift current. Such clinging 1s facilitated by thickened epidermal pads on the ventral sur-
face of the paired fins. When the pads are observed with scanning electron microscopy they are
found to be composed of thousands of unculi, or unicellular keratinous hooklets (Roberts 1982). In
Amphiliinae and in most Doumeinae (Andersonia an exception), these pads are supported by great-
ly enlarged outer pectoral and pelvic fin rays with numerous expanded lepidotrichia. The pads are
either absent or very feebly developed in Leptoglaninae, in which the lepidotrichia of the outer pec-
toral and pelvic fin rays are simple and relatively few. Unculiferous pads were not observed in
specimens of Tetracamaphilius pectinatus, T. clandestinus and Psammphyletria nasuta examined
with SEM.

With the exception of some of the species of Zaireichthys, nearly all of the Leptoglaninae have
very narrow heads. The crania of all Leptoglaninae (including Zaireichthys) are characterized by
narrow frontal bones (or frontal bones narrowed at least anteriorly) and no anterior or posterior
fontanels. Most Amphiliinae and Doumeinae have anterior and posterior fontanels; Andersonia has
the anterior fontanel only. Leptoglaninae have milky white, opaque or translucent (but not glasslike
or transparent) bodies with delicately banded or spotted color patterns; some species have very lit-
tle coloration. As is typical of small sand-dwelling catfishes in South America and Asia, the body
sometimes has a faint yellowish tinge and the spotting or banding tend to be yellowish, brownish,
or even orangish. Amphiliinae and Doumeinae have darkly opaque bodies with variously mottled
cryptic color patterns typical of catfishes inhabiting rocky streams. As in many other sand-dwelling
catfishes, leptoglanins are capable of modifying the openings of their pupils, reducing them to
small horizontal slits in bright light. Similar pupillary modification has not been observed in
Amphiliinae or Doumeinae.

Structure of the dorsal and pectoral fins and their supporting elements differ so much among
leptoglanin genera that it might easily be supposed they belong in different subfamilies or in dif-
ferent families. Zaireichthys was described as a genus of Bagridae (Roberts 1967a) because it has
a stout dorsal fin spine with a locking mechanism and a strong, serrated pectoral fin spine. But
Zaireichthys is very similar osteologically to other genera of leptoglanins and especially to
Tetracamphilius, one species of which has a weakly serrate pectoral fin spine but no pectoral lock-
ing mechanism (7. pectinatus).

KEY TO GENERA OF THE SUBFAMILY LEPTOGLANINAE

| Dorsal and pectoral fins without locking mechanisms; dorsal fin with single elongate flexible
spine; pectoral fin spine flexible, without serrae or with small serrae in one species...... 2

Dorsal and pectoral fins with locking mechanisms; dorsal fin with two stout spines, anterior
one small: pectoral fin spine stout, with larse serrae = 5. =... --- oe a ee Zaireichthys

2 Nawitecthiconicalias.aei ewe so ok hob e ss does MR oon. eee ee eee 3

ROBERTS: LEPTOGLANINAE 89

3. Upper jaw without fleshy lobes; entire premaxillary with teeth; lower jaw with teeth ...... 4
Upper jaw with fleshy lobes or fimbriae; anterior portion of premaxillary and lower jaw
UE OUGLC CLINE Een Ents ee rt snc taut, (at. hue. cota ye Sie ae Stee tthe Leptoglanis

4 Insertion of dorsal fin near head; upper caudal fin lobe with 6 principal rays; snout without

ELE AE OW Gere eer ets, Sessa A ec ean Sat Nag yrs panto Dolichamphilius
Insertion of dorsal fin nearer mid-body, over vertebra 11-13; upper caudal fin lobe with
5 principal rays; snout prolonged by a discrete rhinal lobe .............. Psammphiletria

Genus Leptoglanis Boulenger 1902
Leptoglanis Boulenger, 1902:42 (type species Leptoglanis xenognathus Boulenger, 1902, by monotypy).

DIAGNosIs.— Leptoglanis, here regarded as a monotypic genus, 1s in several respects the most
highly modified member of the Leptoglaninae. It has the following specialized characters unique
within Amphiliidae: (1) anterior margin of mouth with a single large medial fleshy lobe or fimbria
and several lateral fleshy lobes or fimbriae of variable size (Fig. 4b); (2) premaxillary with tooth-
less anterior portion, posterior portion with slender conical teeth (Figs. 4—6); (3) lower jaw tooth-
less and exceptionally elongate (Figs. 4-6); (4) coronomeckelian bone exceptionally large and well
defined, immovably articulated to Meckel’s cartilage and lower jaw, and with a distinct posterior
process for ligamentous attachment to basicranium (Fig. 7b); (5) hyomandibular and quadrate
joined by a complex joint consisting of a cartilaginous ball partially enclosed by bony sockets
formed in the hyomandibular and quadrate (Fig. 7a); (6) an exceptionally short braincase (Figs.
4—6): (7) pharynx and gill arches with numerous soft fleshy fingerlike or digitiform structures; and
(8) pectoral and pelvic fin rays excessively branched (Fig. 9a). Principal caudal fin rays 7+8 (the
only other Leptoglaninae with 7+8 principal caudal fin rays 1s Zaireichthys rotundiceps). Vertebrae
20—23+17-19=38-41.

COMMENTS.— The oral fimbriae and digitiform pharyngeal structures of Leptoglanis, possi-
bly homologous to each other, may be adaptations to feeding on organisms living in the sand.
Comparable structures have not been found in any other species of Leptoglaninae or Amphiliidae.
Structures comparable to the pharyngeal digitiform structures of Leptoglanis occur in many species
of the African bagrid catfish subfamilies Claroteinae and Auchenoglanidinae. Structures compara-
ble to the oral fimbriae have not been found in any other African catfish. In other amphilids virtu-
ally the entire oral surface of the premaxilla bears teeth. No other amphiliids are known with a
toothless lower jaw. The excessive branching of the pectoral and pelvic fin rays results in terminal
branches that are exceptionally long and fine (Fig. 9a), and quite unlike those present in other lep-
toglanins (Figs. 9b-f). Thus, the paired fins of Leptoglanis are exceptionally flexible. The mem-
branes connecting the fin rays of the paired fins are very thin. There is no sign of the large unculif-
erous pads present on the ventral surface of the pectoral and pelvic fin rays in species of Amphilius.
Leptoglanis xenognathus, Zaireichthys camerunensis, and Z. dorae are the only leptoglanins with
abdominal vertebrae notably more numerous than postabdominal vertebrae. Zaireichthys
camerunensis and Z. dorae usually have fewer total vertebrae than Leptoglanis, only 34-39. In
other leptoglanins except Dolichamphilius brieni the ratio of abdominal to caudal vertebrae is more
nearly 1:1. In D. brieni (formerly placed in Leptoglanis) number of total vertebrae is slightly more
(43-44) than in Leptoglanis, and caudal vertebrae considerably outnumber abdominal vertebrae.
Vertebral counts of Leptoglaninae are presented in Table 2.

90 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 5

TABLE 2. Leptoglaninae vertebral counts.
Abdominal Postabdominal Total

16 17 18 19 20) 21:22, 23 LOW MSP 1L9N2 0212223024525) 33 34 35 36 37 38 39 40 41 42 43 44

Leptoglasnis xenognathus

Ubangui (holotype) 1 i 1
Stanley Pool Aint Teel Opel a) itl
Lualaba, Wagenias 1 1 1
Lualaba, Panga l 1 1

Zaireichthys zonatus
Kinshasa (paratypes) 1 1 1
Zaireichthys camerunensis

Benoué (syntypes) 23} Dt 1 a i

1
to
tv
as

_
nn

=

Guinea, Upper Niger 5
Zaireichthys dorae
Luachimo (holotype) 2 2 1
Zaireichthys flavomaculatus
Kamaiemby (holotype) ? 1 ?
Zaireichthys heterurus
Lulindi es 4 ils <3}
Avokoko (types) 2 2 2
Zaireichthys mandevillei
Stanley Pool (holotype) g 2 l
Ubangui De YP 33}. Il
Zaireichthys rotundiceps
Bubu (syntypes) 8 72 dl 2 it it
Luwoyeye I | l
Dolicamphilius brieni
Stanley Pool (types) 2 lane i
Dolicamphilius longiceps
Lualaba (holotype) | l 1

Psammphyletria nasuta

vs)
2
2

Ubangui (types)
Psammphyletria delicatus

Stanley Pool (paratype) l ! l
Tetracamphilius pectinatus

Lulua (types) DB 3) SvOwe! l 9

Ubangui (types) ieee) if il 2
Tetracamphilius angustifrons

Ubangui (syntypes) 2 l I I I

Ubangui 2 4 5) 1 ES)

Lualaba, Wagenias l l I
Tetracamphilius cladestinus

Ubangui (types) 5 5 5
Tetracamphilius notatus

Faradje (holotype) l I l

Chinko | 1 I

ROBERTS: LEPTOGLANINAE 9]

Leptoglanis xenognathus Boulenger, 1902
(Figs. 1-2, 4-9)

Leptoglanis xenognathus Boulenger, 1902:42, pl. 14, fig. 1 (type locality “Ubangi 4 Banzyville [=Mobaye]”).

DIAGNOsIS.— Same as diagnosis of monotypic genus Leptoglanis.

COMMENTS.— Specimens from Ubangui and from Stanley Pool usually have 4—9 large dermal
lobes or fimbriae (together with several very small fimbriae) on anterior margin of the mouth. The
only two specimens known from the upper Congo (57.4 and 47.7 mm) have only 3 large lobes. In
other features examined, these specimens are similar to L. xenognathus from Ubangui and Stanley
Pool, but their status should be reconsidered when more material becomes available. It should be
noted that these very soft structures are highly susceptible to drying, after which they can be diffi-
cult to detect.

COLORATION.— The freshly collected 35.8-mm specimen from Ubangui at Bawili had the
body milk-white (opaque, not translucent) in life. The most noticeable color features were on the
fins: two broad oblique bands on the dorsal fin, two broad vertical bands on the caudal fin, and a
single broad band in the middle of the pectoral fin, all composed of brownish melanophores lying
superficially on the fin rays (not in the interradial membranes). Anal fin with a few fine
melanophores on some rays, otherwise colorless. Adipose and pelvic fins without notable color fea-
tures, almost devoid of melanophores. Dorsal part of head especially snout, basal portion of bar-
bels, and cheeks with fine, faint melanophores; occiput (over hindbrain) with large, brownish
melanophores; distal portion of barbels and ventral surface of head without melanophores. Body
with fine, faint melanophores concentrated just above and below lateral line canal and in dorsal
portions of myoseptal troughs. Three faint oval spots composed of fine melanophores at beginning,
middle and end of adipose fin base. Two faint mid-dorsal oval spots between dorsal and adipose
fins. Ventrolateral and ventral surfaces of body almost entirely without melanophores. A thin,
sharply demarcated black line in midventral myopsepta from vent about half-way to anal fin ori-
gin.

GILL RAKERS AND DIGITIFORM PHARYNGEAL STRUCTURES.— Gill rakers numerous for a lep-
toglanin, with soft fleshy covering and slender axial core staining faintly blue in alcian-alizarin
preparations. The 53.4-mm cleared and stained specimen from Stanley Pool has the following
counts of gill rakers on leading/trailing edge of gill arches 1—5: (1) 8/0; (2) 10/0; (3) 0/8; (4) 5/3;
and (5) 0. The leading edge of arch 4 lacks rakers with an axial core, but its distal portion bears a
close-set series of 4 fleshy digitiform structures (Fig. 8b).

Gill rakers and especially digitiform structures are much more numerous in the 62.2-mm spec-
imen (not cleared and stained) from Stanley Pool. It has the following counts of gill rakers of the
leading/trailing edge of gill arches 1—5: (1) 12/4; (2) 12/5; (3) 4/8; (4) 2+8/4; and (5) 1. Digitiform
structures, mostly much larger than the gill rakers, are as follows: roof of pharynx near posterior
margin of hyoid arch with 7 irregularly scattered or clustered; gill arch 1 with 3 on trailing edge of
lower part of arch and a row of 6 on upper part of arch and continuing onto roof of mouth; gill arch
2 with 2 on trailing edge of lower limb and 10 irregularly scattered on upper limb and extending
onto roof of mouth; gill arch 3 with 5 on dorsomedian part of lower limb; arch 4 and arch 5 with-
out digitiform structures.

VERTEBRAL COUNTS (Table 2).— Specimens from the lower Congo mainstream (valene Pool
or Stanley Pool) have consistently fewer vertebrae (38-39) than those from the upper Congo at
Panga (40) and from the Ubangui River (41). This might indicate that they represent different
species. The problem should be investigated when more material becomes available.

92 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 5

MATERIAL EXAMINED

REPUBLIQUE CENTRAFRICAINE: CAS 92614, 35.8 mm, R. Ubangui near Bawili, 68-69 km
upriver from Bangui, 5 March 1988, T.R. Roberts; REPUBLIQUE DU CONGO (KINSHASA): MRAC
118463-118488, 22:29.9-64.8 mm, Stanley Pool[=Malebo Pool at KINSHASA], passe devant la
refuge Jipo, 17 Oct. 1957, P. Brien, M. Poll, J. Bouillon; UMMZ 196084, 4:41.4-60.5 mm, Stanley
Pool, passe devant la refuge Jipo, 17 Oct. 1957, P. Brien, M. Poll, J. Bouillon; MRAC
118458-118461, 3:29.3-62.2 mm, Stanley Pool, le long de ile Funa, 7 Oct. 1957, P. Brien, M.
Poll, J. Bouillon; MRAC 118453—118456, 2:33.3—36.3 mm, Stanley Pool, 27 Oct. 1957, P. Brien,
M. Poll, J. Bouillon; MRAC 29644, 57.4 mm, Panga [=Lualaba?], Bock; MRAC 90-29-P-123,
47.7 mm, Chutes Wagenia near Kisangani, 1989, V. Nyangombe.

SKELETAL STUDY MATERIAL.— MRAC 118463-118489, 2:42.9-47.4 mm and UMMZ 196084,
2:41.4-47.4 mm, Stanley Pool, passe devant le refuge Jipo, 17 Oct. 1957, P. Brien, M. Poll, J.
Bouillon (cleared and stained with alizarin); MRAC 118463—118489, 53.4 mm, Stanley Pool, passe
devant le refuge Jipo, 17 Oct. 1957, P. Brien, M. Poll, J. Bouillon (cleared and stained with alcian
and alizarin).

Genus Zaireichthys Roberts, 1967
Zaireichthys Roberts, 1967b:124 (type species Zaireichthys zonatus, by original designation and monotypy).

DIAGNOSIS.— Zaireichthys differs from all other Amphiliidae in having “a defensive tripod”
consisting of stout dorsal and pectoral fin spines with mechanisms for locking them into erect posi-
tion. First dorsal fin spine (part of locking mechanism) short, second large, non-serrate; dorsal fin
branched rays 6. Pectoral fin spine with 4-10 strong serrae on its inner margin; pectoral fin
branched rays 7-8. Humeral process of pectoral girdle variably developed, from very short and
smooth (non-denticulate) to elongate and extensively covered with small denticulations (the
species Z. zonatus exhibits an intermediate condition, with a short humeral process bearing a few
small but well-defined denticulations). Caudal fin shape highly variable from deeply forked (as in
other leptoglanins) to rounded or truncate. Principal caudal fin ray counts highly variable (albeit
nearly invariable within each species): 7+8, 7+7, 7+5, 6+7, or 6+5.

CRANIUM.— Compared to that of Leptoglanis, the cranium of Zaireichthys is generalized. The
drawing of the dorsal surface of the cranium of Z. zonatus (Fig. 11) is not entirely satisfactory,
mainly because the single specimen available for clearing and staining did not stain very well and
was damaged. Thus it was not possible to distinguish clearly between the pterotic and supraclei-
thrum in the dorsal view, and no attempt was made to draw the cranium in ventral view. The dor-
sal view (Fig. 11) indicates the striking difference between the skull of a generalized leptoglanin
(Zaireichthys) and that of one of the most specialized forms, Leptoglanis xenognathus (Fig. 5).

COMMENTS.— A small first dorsal fin spine is lacking in all other members of the family
Amphiliidae, none of which have a stout elongate dorsal fin spine or can lock the dorsal fin into
erect position. The pectoral fin spine is non-serrate in all other amphiliids except Tetracamphilius
pectinatus, which has small serrations on its pectoral fin spine. No other amphiliids can lock the
pectoral fin into erect position.

KEY TO SPECIES OF ZAIREICHTHYS

This key should work for all of the species included in Zaireichthys except for Z. flavomaculatus
and Z. dorae which are too poorly known to be adequately characterized. Specimens of these two
species should key out under Z. rotundiceps. Some meristic and other characters distinguishing
the species of Zaireichthys (except Z. dorae and Z. flavimaculatus) are presented in Table 3.

ROBERTS: LEPTOGLANINAE 93

—

. Head not bulbous; eyes visible from the side; adipose fin ending well before caudal

REL MOOG YASPOULEd LA Misra seta wa seca as Recta eR eee | Se wel) ee aie. Morea, ee ove ep 2
Head bulbous due to greatly enlarged mandibular muscles; eyes visible from above only;
adipose fin connected posteriorly to caudal fin; body banded................. Z. zonatus
2. Humeral process short, without denticulations; principal caudal fin rays fewer than 7+8.... 3
Humeral process long and spinelike, more or less extensively covered with denticulations:
PRM IPalnCauG alohiMEnAay SNES sess es evel Nic 2 ca eure Jo eee ena level oh ayy ace: Z. rotundiceps

(OS)

. Abdominal and postabdominal vertebrae nearly equal in number; 3—5 pairs of ribs; principal

SAUCAL BAT LAY GOIN O Lato) eke oe ee SIR tans RNS ee eae tem nes SER ne tenn anne pat a
Abdominal vertebrae always more numerous than caudal; 6 pairs of ribs; principal caudal

[TWD TRENVOU/S5 [Las eae is cil eich ear erase ara heaters mend eect eB ley een Z. camerunensis

4. Principal caudal fin rays 7+5; head broad, without dermal ridges ............. Z. heterurus

Principal caudal fin rays 6+5; head narrow, usually with dermal ridges....... Z. mandevillei

TABLE 3. Comparison of the species of Zaireichthys (excepting Z. dorae and Z. flavomaculatus)

zonatus camerunensis heterurus mandevillei rotundiceps
head shape bulbous narrow broad narrow rounded or blunt
branchiostegal rays 7 6-7 7 6-7 6-7
pectoral spine serrrae 3-4 3-5 6-9 7-8 8
pectoral fin rays 8 7 6-7 7-8 8
humeral denticulations few - - - many
dorsal fin rays 5-6 4-6 5-6 6 6
anal fin rays 10 9 8-10 9-11 10
caudal fin shape rounded emarginate forked forked truncate
principal caudal fin rays 6/7 7/7 7/5 6/5 7/8
procurrent caudal fin rays 10/11 11-13/10-13 13-17/13-15 13-18/12-14 14/12
ribs 5 6 3-5 3-5 5-6
vertebrae 34 35-38 33-34 34-36 36-38
coloration banded spotted spotted spotted spotted
collar - . + + =
largest specimen (mm SL) 24.5 33:3 34.6 26.2 31:9

Zaireichthys zonatus Roberts, 1967
(Figs. 9b; 10; 11; 12a)

Zaireichthys zonatus Roberts, 1967b:124, figs. 3-4 (type-locality lower rapids of Congo River, just below
Stanley Pool, at Kinsuka village, within Leopoldville city limits).

DIAGNOSIS.— Zaireichthys zonatus is immediately distinguished from all other leptoglanins
by greatly expanded oral musculature, giving the head a characteristically bulbous appearance (Fig.
10a), and greatly enlarged jaw teeth (Fig. 12a). In the holotype, which has the oral musculature
much better developed than in the paratype, the eyes are only visible when the head is viewed from
above (Fig. 10a). In all other Zaireichthys the eyes are also visible when viewed from the side. Eyes
small, eye diameter 11-12 times in head length. Snout broadly rounded. Body with vertical bands
having darkened margins (similar coloration in Tetracamphilius pectinatus but not in other
amphiliids). Dorsal fin spine and pectoral fin spines very stout and short. Pectoral spine with only
4 serrae. Humeral process short, with a few fine denticulations on ventral margin of its tip. Adipose

94 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 5

fin very long, low, with gently rounded (not triangular) margin, confluent posteriorly with procur-
rent rays of upper caudal fin lobe (adipose fin entirely separate from caudal fin in all other lep-
toglanins). Gill rakers on leading/trailing edges of gill arches 1-5: 1) 3/0; 2) 3/0; 3) 0/4; 4) 3/3; and
5) 3/0 (from cleared and stained 18.1-mm paratype). Caudal fin rays 11,6+7,11. Vertebrae 16+18
= 34.

COMMENTS.— Placement of the nominal species of Leptoglanis with stout dorsal fin spine and
serrate pectoral fin spines in Zaireichthys was proposed by Mo (1991:12). Reservations have been
voiced by Seegers (1996:199) who retained the species in Leptoglanis, where they clearly do not
belong. Presence of minute serrations on the humeral spine of Z. zonatus, a character unknown to
Seegers, is an additional reason for following the placement proposed by Mo. The alternative is to
designate a new genus for L. rotundiceps and related forms with denticulated humeral process. This
procedure, however, would leave L. camerounensis, L. heterurus, and L. mandevillei without
generic placement. The solution adopted here, retaining the species in Zaireichthys, seems best at
least until further study has been done.

DISTRIBUTION.— Zaireichthys zonatus is known only from the two type specimens collected
in the mainstream rapids of the River Congo just below Stanley (or Malebo) Pool. Although the
habitat was predominantly rocky, the specimens were collected on the edge of a sandy area within
a meter or so of each other.

MATERIAL EXAMINED

TYPE MATERIAL.— REPUBLIQUE DU CONGO (KINSHASA): CAS(SU) 64126, 24.5 mm, lower
rapids of Congo River at Kinsuka village, just below Stanley Pool, 21 July 1964, T.R. Roberts
(holotype); CAS(SU) 64127, 18.1 mm, collected with the holotype (paratype; cleared and stained
with alizarin).

ADDITIONAL MATERIAL EXAMINED

None; species known only from the two type specimens.

Zaireichthys camerunensis (Daget and Stauch, 1963), new comb.
(Figs. 8d, 12b, 13)

Leptoglanis camerunensis Daget and Stauch, 1963:94—95, fig. 1(type locality R. Benoué a Lakdo, Cameroun).

DIAGNOSIS.— Zaireichthys camerunensis differs from all other Zaireichthys and all other lep-
toglanins in having 7+7 principal caudal fin rays, and from all except Dolicamphilius longiceps in
having 6 pairs of ribs rather than only 3—5. First pair of ribs on vertebra 6 (first pair on vertebra 5
in D. longiceps). Neural and hemal spines tend to be simple and slender rather than complexly lam-
inate (Fig. 13). Branchiostegal rays 6/6 or 7/7. Pectoral fin spine with 3 to 5 serrae. Humeral
process of pectoral girdle short, without denticulations. No broad black collar just behind head. Gill
rakers on leading/trailing edges of gill arches 1-5 moderately large: (1) 7/0; (2) 7/0; (3) 0/6; (4)
3/3; (5) 0/0 (from cleared and stained specimen). Two proximal gill rakers on the leading edge of
the first gill arch lie on the leading edge itself; four medial rakers lie on top of the flattened portion
of the arch; and one distal raker lies on or nearly on the trailing edge of the arch. On the other arch-
es gill rakers are more clearly located on leading and trailing edges (Fig. 8d). Procurrent caudal fin
rays 11—13/10-13. Vertebrae 18—21+16—18 = 34-39. Largest known specimen 33.3 mm.

COMMENTS.— The only other amphiliid observed with 7+7 principal caudal fin rays is the
armored doumein Andersonia leptura (Table 1).

DISTRIBUTION.— Zaireichthys camerunensis is the only species of leptoglanin known from

ROBERTS: LEPTOGLANINAE 95

north of the Congo basin. It has been found only in the Niger basin, in the Upper Benue in
Cameroun, and in the Upper Niger in Guinea (for map of distribution map see Risch, 1992:418).

MATERIAL EXAMINED

NIGER BASIN, CAMEROUN: MNHN 1962-1272, 6:17.0-21.7 mm, R. Benoué a Lakdo, May
1960, Stauch (syntypes); NIGER BASIN, GUINEA: MNHN 1988-1151, 3:20.7—23.8 mm, R. Milo a
Boussolé, 10 Dec. 1986, D. Paugy; MNHN 1988-1150, 15:23.4-31.7 mm, and CAS 92615,
9:22.4-31.3 mm (3:24.1, 27.2, and 30.0 mm cleared and stained), R. Dele, tributary of R. Niandan,
on road from Firawa to Kissidougou, March 1988, B. Hugueny; MNHN 1988-1152, 6:27.4-31.4
mm, R. Bouyé at Bouye (Niandan watershed, Niger basin), 30 April 1987, B. Hugueny; MNHN
1988-1167, 3:29.4-31.4 mm, R. Niandan at Fermessoudou, B. Hugueny; MNHN 1988-1168,
3:28.3-33.3 mm, R. Niandan at Sougounbaya, March 1988, B. Hugueny.

Zaireichthys dorae (Poll, 1967), new comb.

Leptoglanis dorae Poll, 1967:211, fig. 95 (type locality “rapides de la Luachimo, dans les flaques d’eau
residuelles” [Congo basin], Angola).

COMMENTS.— The 27-mm holotype and only known specimen of Zaireichthys dorae proba-
bly is immature. Pectoral fin spine with 4 or 5 large serrae. Humeral process moderately long, den-
ticulations poorly developed. Neural and hemal processes relatively slender (not laminate).
Vertebrae 20?+17?=37. The species should be redescribed from more material.

Note that “Leptoglanis cf. dorae (non Poll 1967)” briefly described and figured reported from
the “Okavango, Kwando, Chobe, and Zambesi systems) by Skelton (1993:220) 1s not conspecific
with L. dorae Poll (1967).

DISTRIBUTION.— Known only from the rio Luachimo, Congo basin, Angola.

MATERIAL EXAMINED

TYPE MATERIAL.— MRAC 161646, 27 mm, rapides du rio Luachimo, aval du barrage, [Congo
basin], Angola, 10 Feb. 1957, D. Machado (holotype L. dorae).

ADDITIONAL MATERIAL EXAMINED

None; species known only from holotype.

Zaireichthys flavomaculatus (Pellegrin, 1926), new comb.

Leptoglanis flavomaculatus Pellegrin, 1926:204 (type locality “Kamaiembi” [ =Sankuru, Kamaiembi, riviere
Lulua, Congo basin]; Pellegrin, 1928:29, fig. 17.

COMMENTS.— The holotype and only known specimen is in poor condition. Humeral process
long and denticulate (as in Z. rotundiceps). Radiographs show caudal fin with 15,7+8,15 rays; ver-
tebrae 18?+18=36? The species should be redescribed from more material.

A single specimen has been reported as L. flavomaculatus from the rio Lufume, a tributary of
the riviere Luele (in Angola) by Poll (1967:211, fig. 94). It differs markedly from the holotype of
Z. flavomaculatus in having a finely spotted coloration overall and a truncate caudal fin, and evi-
dently represents an undescribed species.

DISTRIBUTION.— Zaireichthys flavomarginatus is known only from the riviére Lulua in south-
ern Congo (Kasai).

96 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 5

MATERIAL EXAMINED

TYPE MATERIAL.— MRAC 19721, 39.3 mm, Kamaiembi, H. Schouteden (holotype L. flavo-
maculatus).

ADDITIONAL MATERIAL EXAMINED

None; species known only from holotype.

Zaireichthys heterurus Roberts, new species
(Fig. 14)

HOLOTYPE.— REPUBLIC DU CONGO (KINSHASA): MRAC 87-42—P-1140, 31.4 mm, R.
Avokoko, affluent rive droit fleuve Zaire, km 21 route Kisangani-Wanie Rukula, 29 March 1987,
L. De Vos and A. Kimbembi.

DIAGNOSIS.— Zaireichthys heterurus differs from all other amphiliids in having 7+5 principal
caudal fin rays. Humeral process moderately long and stout but not denticulated. Barbels attenu-
ate, maxillary barbel extending posteriorly almost to end of pectoral fin spine. Head broadly round-
ed. Eyes small. Color pattern spotted, with a broad dark collar (as in Z. mandevillei) immediately
posterior to head. Caudal peduncle moderately slender; caudal fin deeply forked, upper lobe much
shorter and smaller, and falcate, lower lobe rounded; fin rays in lower lobe noticeably thicker than
those in upper lobe. Largest known specimen 34.6 mm. With growth, individual fish become
increasingly stouter, more perhaps than any other leptoglanin species (compare Fig. 14b of the
stout 31.4-mm holotype with Fig. 14c of a cleared and stained and much more slender 23.0-mm
specimen). Vertebrae 16—17+17=33-—34.

COMMENTS.— The two non-type specimens from the Lufira both have 7+5 principal caudal
fin rays, but the caudal peduncle seems more slender than in other Z. heterurus.

DISTRIBUTION.— Zaireichthys heterurus is widely distributed in the eastern sector of the
Congo basin. It occurs in clear, swift flowing tributaries lying to the east of the Congo River main-
stream.

ETYMOLOGY.— The name /eferurus is from the Greek heteros, different, and oura, tail.

MATERIAL EXAMINED

HOLOTYPE.— REPUBLIC DU CONGO (KINSHASA): MRAC 87-42-P-1140, 31.4 mm, R.
Avokoko, affluent rive droit fleuve Zaire, km 21 route Kisangani-Wanie Rukula, 29 March 1987,
L. De Vos and A. Kimbembi.

PARATYPES.— REPUBLIC DU CONGO (KINSHASA): MRAC 87-42-P-1140-43, 3:23.2—26.6 mm,
MRAC 90-47—P-538-549, 12:23.4-30.1 mm, R. Avokoko, 27 Dec. 1987, L. De Vos; MRAC
88—24—P—1-8, 8:22.4—32.5 mm, R. Avokoko, same collection data as holotype (2:26.0-30.4 mm
cleared and stained with alcian and alizarin); CAS 92617, 84:17.6-31.5 mm, R. Lulindi, 21 km by
road N of Lusangi (route Kongolo-Kasongo), 28 Aug. 1986, T.R. Roberts (4:22.4—23.2 mm cleared
and stained with alcian and alizarin); CAS 92618, 24.7 mm, R. Lukuga just downstream from
Niemba, 20-21 Aug. 1986, T.R. Roberts; MRAC 90-47—P-—524—-537, 6:24.9-29.9 mm, R.
Avokoko, km 22 route Kisangani-Wanie Rukula, 17 Dec. 1989, L. De Vos, Kambasu; MRAC
90-47—P-564—580, 8:20.7-27.3 mm, R. Avokoko, 19 Jan. 1990, L. De Vos, Kambasu; MRAC
90—30—P—1198—256, 2:24.4-32.3 mm, Chutes Wagenia, Kisangani, 7 Feb. 1990, L. De Vos.

ROBERTS: LEPTOGLANINAE 97

ADDITIONAL MATERIAL EXAMINED

REPUBLIC DU CONGO (KINSHASA): MRAC 87-61—P-1-2, 2:24.3-26.3 mm, R. Lufira, Oct.
1947, G FE. de Witte.

Zaireichthys mandevillei (Poll, 1957), new comb.
(Figs. 8e, 12c; 15a)

Leptoglanis mandevillei Poll, 1959:98, pl. 25 fig. 2 (type locality Stanley Pool [Congo R. near Kinshasa]).

DIAGNOsIS.— Zaireichthys mandevillei is the most slender-bodied and perhaps the smallest
species of Zaireichthys. Largest known specimen 26.3 mm. Snout more pointed than in other
Zaireichthys. Eyes small. Barbels attenuate, maxillary barbel extending posteriorly to middle of
pectoral fin spine. Most specimens have entire dorsal surface of head with longitudinally oriented,
interrupted ridges (presumably keratinous) not observed in other leptoglanins (small, scattered
ridges, presumably keratinous, present in some specimens of Z. heterurus and Z. rotundiceps).
Pectoral fin spine elongate, with 7—9 strong serrae. Humeral process very short, without denticula-
tions. Color spotted, with a prominent dark collar just behind head (as in Z. heterurus). Caudal
peduncle very slender; caudal fin forked, lower lobe larger and longer than upper lobe; principal
caudal fin rays 6+5.

COMMENTS.—Ridges similar those of Z. mandevillei occur on the dorsal surface of the head
of an African mochokid catfish, Chiloglanis reticulatus (Roberts 1989:159, 169, fig. 5).
Zaireichthys mandevillei has perhaps the most reduced gill rakers of any leptoglanin. Gill rakers
(very small) on leading/trailing edge of gill arches 1—5: (1) 1/0; (2) 1/0; (3) 1/2; (4) 2/1; and (5) 0
(in 22.3-mm cleared and stained specimen) (Fig. 8e).

DISTRIBUTION.— Previously known only from Stanley Pool in the Congo River (type locali-
ty), Z. mandevillei is now reported from the Ubangui and Lualaba. So far as known, it is confined
to sandy reaches of the mainstream of the Congo River and its largest tributaries.

MATERIAL EXAMINED

TYPE MATERIAL.— REPUBLIC DU CONGO (KINSHASA): MRAC 118533—536, 4:19.9-22.8 mm,
Stanley Pool, entrée de la passe de Limbili, 19 Sept. 1957, P. Brien, M. Poll, J. Bouillon
(paratypes).

ADDITIONAL MATERIAL EXAMINED

REPUBLIC DU CONGO (KINSHASA): MRAC 1313, 19.8 mm, R. Ubangui at Banzyville, 1901,
Royaux (poor condition; originally identified as Amphilius angustifrons); MRAC 90-30—P-1192,
26.2 mm, Chutes Wagenia, Kisangani, 23 Jan. 1990, L. De Vos; MRAC 90-30—P—1193—1197,
2:22.6-24.8 mm, Chutes Wagenia, Kisangani, 25 Jan. 1990, L. De Vos; MRAC
90-30—P-1198—256, 23:21.2-26.3 mm, Chutes Wagenia, Kisangani, 7 Feb. 1990, L. De Vos;
MRAC 90-47-P-58 1-632, 21:21.9-24.9 mm, Chutes Wagenia near Kisangani, 3—8 June 1990, L.
De Vos; MRAC 90-47—P-633-651, 8:22.1-23.3 mm, Chutes Wagenia, Kisangani, 24 June 1990,
L. De Vos.

REPUBLIQUE CENTRAFRICAINE: CAS 92619, 22:16.3—-24.2 mm, R. Ubangui at Isle Molenge 82
km upriver from Bangui, 27 Feb. and 7 March 1988, T.R. Roberts; CAS 92620, 19.4 mm, R.
Ubangui near Isle Baskiki, 75 km upstream from Bangui, 6 March 1988, T.R. Roberts; CAS 92621,
49:13.6-23.9 mm, Ubangui R. near Banda, 72 km upriver from Bangui, 5 March 1988, T.R.
Roberts (7:17.0—23.5 mm cleared and stained with alcian and alizarin); CAS 92622, 11:15.3-21.3
mm, R. Ubangui near Bawili, 68-69 km upstream from Bangui, 5 March 1988, T.R. Roberts.

98 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 5

Zaireichthys rotundiceps (Hilgendorf, 1905), new comb.
(Fig. 15b)

Gephyroglanis rotundiceps Hilgendorf, 1905:412 (type locality “Im Bubu bei Irangi” [Rufiji basin,
Tanganyika])

Leptoglanis rotundiceps, Boulenger, 1911:352, fig. 273.

Leptoglanis brevis Boulenger, 1915:169 (type locality riv. Lumumbashi a Elisabethville).

?Leptoglanis wamiensis Seegers, 1989:285, fig. 1 (type locality Kisangata-Bach bei Mvumi, 32 kilometer SW
Kidete, am Wege nach Kimamba/Kilosa im Wami-Einzug NW Morogoro, Tanzania).

?Leptoglanis sp Seegers, 1989:284 (see Seegers 1996:196)

DIAGNOSIS.— Zaireichthys rotundiceps of this account, which probably represents several
species, has humeral process of shoulder girdle well developed, elongate, with its ventral margin
and sometimes entire surface covered with fine denticulations. Pectoral fin spine stout with strong
serrae (7 or more in larger specimens). Dorsal fin spine stout, with a well-developed locking mech-
anism. Caudal fin shape variable, from slightly forked through, emarginate, truncate, or even
slightly rounded, but not deeply forked. Principal caudal fin rays 7/8 (only checked in a few spec-
imens). Head broad, snout rounded or blunt, body relatively stout. Adipose fin elongate, its margin
rounded. Coloration highly variable, from abundant dark spots in several rows to light pale spot-
ting pattern (never banded). Marks often present on head and fins as well as body.

Gill rakers moderately well developed, leading/trailing edges of gill arches 1—5 with 1) 6/0; 2)
6/0; 3) 0/5; 4) 5/5; and 5) 5 rakers; rakers on leading edge of gill arch 5 (lower pharyngeal) not
stained with alizarin or alcian, but clearly discernible (24.1-mm cleared and stained specimen from
Luwoyeye). Coloration of the Zaireichthys rotundiceps species complex is highly variable, involv-
ing more or less numerous spots of variable size and distribution over most of the body. There is
no humeral collar or band of dark pigmentation just behind the hind.

COMMENTS.—A large denticulate humeral process was observed in all of the type and
non-type specimens listed as material examined of Z. rotundiceps. Detailed study of all of this
material has not been attempted, but some preliminary comments are in order. I have examined the
humeral denticulations in a 24.1-mm specimen of Z. rotundiceps from rivi¢re Luwoyeye, CAS
92623. Superficially resembling sharp conical teeth, the denticulations are concentrated along the
lower margin of the humeral process. Apparently they are not true teeth. They arise directly from
the bony humeral process; there are no tooth sockets; and, although the denticulations are numer-
ous and differ in size, there is no sign of any stages of tooth formation or of tooth replacement.
Direct comparison of the BMNH syntypes of L. rotundiceps and L. brevis reveals that they are
closely similar, possibly conspecific. For further discussion of L. rotundiceps and L. brevis see
Seegers (1996).

A large denticulated humeral process also occurs in Z. dorae and Z. flavomaculatus, two poor-
ly characterized species known only from the holotypes, that clearly are not conspecific with Z.
rotundiceps. A smaller humeral process bearing relatively few denticulations is present in Z. zona-
tus. Similar denticulations have not been reported in any other members of the family Amphiliidae.

Leptoglanis wamiensis Seegers, 1996 undoubtedly is a Zaireichthys. It may be tentatively
regarded as a junior synonym of Z. rotundiceps. Seegers emphasized its small size and unique color
pattern as species characters. Size and color are highly variable in the samples of Zaireichthys iden-
tified here as Z. rotundiceps. Problems with such characters as body size and number and size of
spots in the color pattern (and their intensity) are 1) they are continuously variable and difficult to
describe and record, let alone compare; 2) they are probably influenced by non-genetic factors such
as water clarity, food availability, and physiological condition; and 3) they differ in virtually every

ROBERTS: LEPTOGLANINAE 99

population sample. Characters that might prove more useful in distinguishing the different species
include shape and width of premaxillary tooth patch, length of lateral line canal; and numbers of
branched dorsal and caudal fin rays (D. H. Eccles, in lit., 23 Oct. 1979).

Two species of Zaireichthys are reported (as Leptoglanis rotundiceps and L. cf dorae) from the
Zambesi and other southern African localities by Skelton (1993:218—220). The behavior of lying
buried in fine sand with only the eyes protruding is recorded for both species by Skelton.
Zaireichthys cf rotundiceps in Lake Malawi lives in shells of the snail Lanistes (D. J. Stewart, in
litt., 18 June 1979).

MATERIAL EXAMINED

TYPE MATERIAL.— EASTERN AFRICA: BMNH 1905.7.25. 43-46, 4: 20.7—30.1 mm, and MRAC
76—50—P—1, 26.1 mm, Bubu River near [rangi, Rufiji basin, July 1897, O. Neumann (syntypes G.
rotundiceps); REPUBLIC DU CONGO (KINSHASA) BMNH 1920.5.26.93, 25.1 mm, R. Lubumbashi,
Elizabethville, Congo basin, 9 June 1911, L. Stappers (syntype L. brevis).

ADDITIONAL MATERIAL EXAMINED

REPUBLIC DU CONGO (KINSHASA): CAS 92623, 3:24.1—-33.3 mm, riv. Luwoyeye, a tributary of
riviere Lukuga, about 10 km by road S of Nyunzu, 22 Aug. 1986, T.R. Roberts (24.1-mm speci-
men cleared and stained); MRAC 89-43—P-—2242, riv. Kawe II, affl. riv. Lubutu, km 260 route
Kisangani-Bukavu, & July 1989, L. De Vos; MRAC 182734, 25.4 mm, vallee riv. Lupweshi, afflu-
ent de la riv. Lufira, 8 June 1958, N. Magis; MRAC 183297-—340, 44:18.8—34.8 mm, au pont route
de la riv. Lubumbashi, camp forestier de la Kipopo, | and 8 May 1960, T. van den Audenaerde;
MRAC 18341-—342, 2:22.4—25.2 mm, étang de la station de la Kipopo, June 1960, D.F. Thys van
den Audenaerde; MRAC 183343-44, 2:28.7—31.1 mm, Kiubo, dans les rapides au-dessus des
chutes des riv. Lufira et Luwilombo, 8 Aug. 1960, T. van den Audenaerde; MRAC 79—1—P-4382,
30.2 mm, riv. Bowa, affl.droit de la Kalule Nord et 1/ affl. droit du Lualaba, pres de Kiamalwa, 2—3
March 1949, GF. de Witte; MRAC 83416, 35.3 mm, Elizabethville, Parc Heenen, 7 Jan. 1947,
Mission Pisc. Katanga; MRAC 90262-90263, 2:34.4—36.3 mm, riv. Koki, entre Mulenge et
Kamulenge, 28 Sept.-1 Oct. 1947, M. Poll; MRAC 90264, 37.4 mm, riv. Kitwe, affluent de la
Lukuga a Kokompwa; MRAC 93050, 28.9 mm, riv. Luvubu, 9 Sept. 1952, G. Marlier; MRAC
152631-33, 3:24.0-26.4 mm, riv. Kilobelobe, Katanga, 21 June 1963, M. Lips.

OUTSIDE CONGO BASIN.— BMNH_ 1987.7.13.77, 25.7 mm, Cubango (Okavango) R. at
Nkurunkuru, M. J. Penrith; BMNH 1979.12.6.6—10, 5:22.3-25.9 mm, Kunene R. one mile E of
Epupa Falls, Oct. 1971, M. J. Penrith; BMNH 1979.12.6.2—5, 4:20.3— 25.4 mm, Kwando R. at
Choyi near Kongolo, Caprivi, 3 June 1975; BMNH 1979.12.6.30, 30.0 mm, Impalilay Stream near
confluence of Zambesi and Chobe, E Caprivi, 7 Aug. 1975, B. van der Waal; BMNH 1979.12.6.1,
23.2 mm, Kafue R., Zambesi system, 16 Dec. 1963; BMNH 1976.12.6.11—13, 3:18.0-19.5 mm,
Mazinzi Bay, Lake Malawi (trawled in shells at 11-13 m), Oct. 1974, D. H. Eccles; BMNH
1979.12.6.17—21, 5:20.3-31.7 mm, western affluent streams of Lake Malawi, Oct. 1978, D.
Tweddle and N. J. Willoughby; BMNH 1979.12.6.22—25, 4:27.0-33.1 mm, Sabi R., eastern
Rhodesia, 4 Dec. 1960; MRAC 90-46-P-1-6, 3:29.9-37.9 mm, Rwimi R., road between Portal
and Kasese, Uganda, 13 June 1990, R. Wildekamp.

100 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 5

Dolichamphilius Roberts, new genus
Type species: Leptoglanis brieni Poll, 1959

DraGcnosis.— Dolichamphilius was placed in Leptoglanis, but differs in numerous respects
from the generic type-species L. xenognathus: (1) body and especially caudal peduncle extremely
elongate; (2) pectoral fin with 9-11 instead of only 8 branched rays; (3) pectoral- and pelvic-fin
branched rays with long simple branches medially and extremely foreshortened compound branch-
es distally (this character is shared with Psammphyletria); (4) outermost branched ray of pectoral
and pelvic fins more elongate, so these fins have a falcate shape; (5) principal caudal fin rays 6+7
instead of 7+8; (6) upper and lower jaws with teeth (vs. lower jaw toothless); and (7) premaxillary
bone of generalized usual shape for amphiliids, with entire ventral surface dentigerous (vs. premax-
illary with an extensive non-dentigerous area anteriorly).

ETYMOLOGY.—From the Greek dolichos, long or elongate; ammos, sand; and philos, beloved,
dear. Gender masculine.

Dolichamphilius brieni (Poll, 1959), new comb.
(Figs. 9e, 16a)

Leptoglanis brieni Poll, 1959:96, pl. 24 fig. 2 (type locality “Stanley Pool, passe devant le refuge Jipo”).

DiAGnosis.— Dolichamphilius brieni is readily distinguished from all other leptoglanins by
its extremely slender body and elongate caudal peduncle; pectoral fin with a flexible spine and 11
rays; first branched ray of pectoral and pelvic fins distinctly longer than other rays; both jaws with
conical teeth; lower jaw with only a single short row of 8-10 fine sharp conical teeth; caudal ver-
tebrae more numerous than abdominal, vertebrae 19+24—25=43—44; peduncular vertebrae 17.

DISTRIBUTION.— Dolichamphilius brieni is known only from the type locality, Stanley Pool
[Congo River at Kinshasa].

MATERIAL EXAMINED

TYPE MATERIAL.— REPUBLIC DU CONGO (KINSHASA): MRAC 118504—-118506, 26.9 mm,
Stanley Pool, extérieur des tiles de V'archipel N°’ Dyili vers Vile de Cristal, 7 Oct. 1957, P. Brien, M.
Poll, J. Bouillon (paratype); MRAC 118504, 31.2 mm, Stanley Pool, passe devant le refuge Jipo,
7 Oct. 1957, P. Brien, M. Poll, J. Bouillon (paratype; cleared and stained with alcian and alizarin).

ADDITIONAL MATERIAL EXAMINED

None; species known only from type specimens.

*Dolichamphilius” longiceps Roberts, new species
(Fig. 16c)

HOLOTYPE.—REPUBLIC DU CONGO (KINSHASA): MRAC 90-47—P-704-708, 42.2 mm, Chutes
Wagenia, Kisangani, 24 June 1991, L. De Vos.

DIAGNOsIs.— Dorsal fin rays 1542; anal fin rays i117; pectoral fin rays 19. Caudal fin rays
10,6+7,12. Vertebrae 20+20=40; 6 pairs of ribs. First pair of ribs, on vertebra 5, smaller than ribs
on vertebra 6 (in all other leptoglanins, first pair of ribs are largest and occur on vertebra 6). Caudal
peduncle elongate, but much less so than in D. brieni, with only 12 vertebrae. Body with four or
five elongate oval spots centered on midline; spot nearest middle of body considerably enlarged on

ROBERTS: LEPTOGLANINAE 101

both sides of body in holotype would be a diagnostic feature if characteristic of the species.

COMMENTS.— This distinctive species is unfortunately known only from the holotype, which
has been radiographed but not cleared and stained for osteology. It resembles in some ways the
species of Zetracamphilius but differs from them in having simple conical jaw teeth and more
numerous vertebrae. It is provisionally assigned to Dolichamphilius, even though it differs in sev-
eral respects from the generic type species. Because of doubt that this species belongs to
Dolichamphilius, its characters have not been taken into account in the diagnosis of
Dolichamphilius.

DISTRIBUTION.— Dolichamphilius longiceps is known only from the type locality, Chutes
Wagenia on the R. Lualaba near Kisangani.

ETYMOLOGY.— The name longiceps is from the Latin, Jongus, long; and -ceps, derived from
caput, head.

Psammphiletria Roberts, new genus
Type species: Psammphiletria nasuta Roberts, new species.

DIAGNOsIS.—Small, sand-dwelling amphilid catfishes (largest known specimen 23.7 mm)
with a prominent fleshy rhinal lobe; basal half of maxillary barbel included in maxillary mem-
brane; branchiostegal rays 4—5; tubiferous portion of lateral line incomplete, ending above anal fin;
dorsal and pectoral fins without locking mechanisms, their fin rays all slender and flexible; dorsal
fin origin very far behind head and pectoral fin, over vertebra 11—13; adipose fin triangular, mod-
erately elongate, its origin over base of third anal fin ray; 3-4 pairs of ribs; pectoral- and pelvic-fin
branched rays with long simple branches medially and extremely foreshortened compound branch-
es distally (this character is shared with Dolichamphilius); pelvic fin origin below vertebra 10-11;
principal caudal fin rays 5/6; lower lobe of caudal fin distinctly prolonged; vertebrae 17—18+
16—18=33-—36; neural and hemal spines expanded distally (lamellar).

COLORATION.— Head and body overall white, translucent but not transparent. There is an
almost continuous median row of brown spots or marks on the dorsal surface of the body. Head,
fins, and side of body without markings except for a diamond-shaped peduncular spot and some
smaller indistinct markings just anterior to it.

COMMENTS.— Psammphiletria looks superficially like the young of Amphilius, but there are
numerous differences. Amphilius differs from Psammphiletria in having no rhinal lobe; anterior
and posterior cranial fontanelles present; branchiostegal rays 6—8; gill rakers present on upper por-
tion of gill arches (absent in Psammphiletria); 10 or more pairs of ribs; pelvic fin origin very much
farther posterior; pectoral girdle with two ossified radials instead of only one; principal caudal fin
rays 6+7 or more; lower caudal fin lobe not prolonged.

Zaireichthys differs in having very stout dorsal and pectoral fin spines with well-developed
locking mechanisms; serrate pectoral fin spine; principal caudal fin rays always more than 5+6; and
procurrent caudal fin rays much more numerous. Dolichamphilius differs in having a more elon-
gate body, more numerous vertebrae, dorsal fin origin immediately behind cranium; pectoral fin
with 11—12 rays; first branched ray of pectoral and pelvic fins prolonged.

ETYMOLOGY.—From the Greek psammo “sand” and philetria “lover of’. Gender feminine.

Psammphiletria nasuta Roberts, new species
(Figs. 8f, 9d, 17a—d)

HOLOTYPE.— REPUBLIQUE CENTRAFRICAINE: CAS 92624, 23.5 mm, Central African Republic,

102 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 5

sandy riffle in R. Ubangui near Banda, 72 km upriver from Bangui, 5 March 1988, T.R. Roberts.

DIAGNOSIS.—A species of Psammphyletria with 4-6 unbranched anal fin rays. Dorsal fin with
only 5 rays; dorsal fin pterygiophores 4. Rhinal lobe very large. Gill rakers on first gill arch 34.
Vertebrae 18+17—18=34—36.

DISTRIBUTION.— Psammphyletria nasuta is known only from the Ubangui River upstream
from Bangui.

ETYMOLOGY.— The name nasuta is from the Latin nasutus, long-nosed.

MATERIAL EXAMINED

HOLOTYPE.— REPUBLIQUE CENTRAFRICAINE: CAS 92624, 23.5 mm, Central African Republic,
sandy riffle in R. Ubangui near Banda, 72 km upriver from Bangui, 5 March 1988, T.R. Roberts.

PARATYPES.— REPUBLIQUE CENTRAFRICAINE: CAS 92625, 11:14.8-22.7 mm, same data as
holotype (3:16.1—23.7 mm cleared and stained with alcian and alizarin); CAS 92626, 16.5 mm, R.
Ubangui at Isle Basiki, 75 km upriver from Bangui, 6 March 1988, T. R. Roberts.

Psammphiletria delicata Roberts, new species
(Fig. 17e)

HOLOTYPE.—MRAC 118455, 20.5 mm, Stanley Pool, 27 Sept. 1957, P. Brien, M. Poll, J.
Bouillon.

DIAGNOSIS.—Psammphiletria delicata is distinguished from P. nasuta, its only congener, by
slight differences in counts of dorsal and anal fin rays, and in having dorsal and anal fin pterygio-
phores and laminar portions of neural and hemal spines more expanded. Dorsal fin rays 7 (5 in P.
nasuta). Dorsal fin pterygiophores 7 (4 in P. nasuta). All three type specimens have anal fin with
three simple rays, three branched rays, and a simple ray (last simple ray only ray on last anal fin
pterygiophore). Rhinal lobe large, but not so large as in P. nasutus; this may be partly due to dif-
ferences in preservation and condition of the specimens, those of P. delicata being in less good con-
dition. Vertebral count of the single cleared and stained paratype 17+16=33.

DISTRIBUTION.— Psammphyletria delicata is known only from Stanley Pool (Malebo Pool).

ETYMOLOGY.— The trivial name delicata is Latin for dainty or delicate.

MATERIAL EXAMINED

HOLOTYPE.—MRAC 118455, 20.5 mm, Stanley Pool, 27 Sept. 1957, P. Brien, M. Poll, J.
Bouillon.

PARATYPES.—MRAC 118456, 19.5 mm, same data as holotype (cleared and stained with
alcian and alizarin); MRAC 118489, 18.6 mm, Stanley Pool, passe devant la refuge Jipo, 17 Oct.
1957, P. Brien, M. Poll, J. Bouillon.

Tetracamphilius Roberts, new genus
Type species: Tetracamphilius pectinatus Roberts, new species.

DIAGNOsIs.— Tetracamphilius differs from all other leptoglanins and from all other amphili-
ids in having pedicellate jaw teeth with a fan-shaped distal end typically armed with two to six tiny
flattened triangular cusps (Fig. 19). All other amphiltids have simple conical jaw teeth or jaw teeth
absent. Adipose fin triangulate, its origin well in advance of a vertical line through anal fin origin.
Caudal peduncle slender, peduncular vertebrae 9-12. Caudal fin deeply forked, lower lobe slight-
ly larger than upper, principal caudal fin rays 6+7.

ROBERTS: LEPTOGLANINAE 103

COMMENTS.— The cusps on the jaw teeth of Tetracamphilius are possibly unique not only in
catfishes but perhaps also in Ostariophysi. I do not know of any other catfishes with comparable
cusps. Multicuspid characoid jaw teeth are superficially similar but are much larger, and usually
formed by combination of individual conical elements (Roberts 1967b), which does not appear to
be so in Jetracamphilius. Scanning electron microscopic observations of the jaw teeth in a speci-
men of 7: pectinatus reveals that the cusps are flattened, not really conical, and that they are fre-
quently broken off, so that it is difficult to find a tooth with more than one or two intact cusps;
many of the teeth have all of the cusps broken off, so that they are squared off distally or peg-like
in shape. The number of cusps increases with growth. The largest cleared and stained specimen
examined, a 37.2 mm 7: angustifrons, had jaw teeth with as many as six cusps. When the cusps of
such a tooth are broken off, the crown has a discoid shape. The multicuspid teeth of
Tetracamphilius are present on the upper as well as lower jaw, but not in the pharynx (upper and
lower pharyngeal teeth in all leptoglanins including Tetracamphilius are simple conical teeth).

EtTyMoLoGy.—From the Greek tetra, four, and akis, point, in reference to the multicuspid
teeth, and Amphilius, a generical name for this group of catfishes.

KEY TO SPECIES OF TETRACAMPHILIUS

Some meristic and other characters distinguishing the species of Tetracamphilius,
presented in Table 4, should be used in conjunction with this key.

PPeeccioraluiim spine without semae:-body, Spotted eis... vse a5 des to aes eel eae 2
Pectoral fin spine with fine serrae on inner margin; body with dark-margined, pale-centered
BYERS rapes ass eb as hocks ene eaei ga aes Sy yousde Bihgs Fo ane Ae dU MeN AN she Sh ON al ams T. pectinatus
2. Lamellar portion of olfactory organ large, its length nearly equal to eye diameter; posterior
Hake smmMUCKoAarcerthan ANCELOL MATES jcyeeveretas. acer, 3 Sia pees Hole enere ola Seals co eagiensn se one ere 3
Lamellar portion of olfactory organ not enlarged, its length less than half eye diameter;
AME HM Olga yNHOSLCH ORM ATES (CCU Al MMISIZE mr Neuen aie acl oy ones eee our ie T. notatus
3. Dorsal fin rays 8—9; anal fin rays 9-10; pectoral fin rays 8-9 .............. T. angustifrons
Dorsal fin rays 6—7; anal fin rays 8; pectoral fin rays 7-8 ................. T. clandestinus

TABLE 4. Comparison of the species of Tetracamphilius

pectinatus angustifrons clandestinus notatus
pectoral spine serrae 6-7 - = =
pectoral fin rays 8-9 ) 7-8 8
dorsal fin rays 8-9 6-7 6-7 8
anal fin rays 8-10 9-10 8 10
procurrent caudal finrays = 13-14/13-14 9-11/10-12 9-12/9-10 13/11
branchiostegal rays 7-8 6-7 6-7 6-7
ribs 4 4-5 3-4 5
total vertebrae 34-38 36-38 35 35-37

largest specimen (mm SL) a3e 39.4 OES; B25

104 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 5

Tetracamphilius pectinatus, Roberts, new species
(Figs. 18)

HOLOTYPE.— REPUBLIC DU CONGO (KINSHASA): CAS 92627, 33.7 mm gravid female, Congo
basin, tributary of R. Luala 26 km N of Kibunzi (8 km S of turnoff to Kibunzi, Luozi dist.), 45°6.5’
S, 13°48’ E, 24 July and 17 Aug. 1973, T.R. Roberts and D.J. Stewart.

DriAGnosis.— Tetracamphilius pectinatus differs from the other species of the genus and from
all other amphiltids except Zaireichthys in having a serrated pectoral fin spine. Unlike Zaireichthys,
which has very large pectoral spine serrae, the serrae are very small, and the pectoral fin spine does
not possess a locking mechanism. It differs from all other leptoglanins (and all other amphiliids)
except Z. zonatus in having color pattern with a series of bands with paler inner part and darkened
margins.

COLORATION IN LIFE.— Tetracamphilius pectinatus is the most colorful known leptoglanin. In
life, the dark narrow bands are a dark chocolate brown, the dorsum of the head and areas between
the narrow bands tan or orangish tan, and the abdomen and other pale areas cream-colored.

HABITAT NOTES.— The tributary of the riviere Luala in Bas-Congo where adult T. pectinatus
were collected in breeding condition was clear, moderately swift flowing, with sand, gravel, and
rock rubble bottom, 6—8 m wide and | m deep, in rolling or hilly savannah. On 24 July 1973 col-
lecting was done with a square frame net, by pushing it into the sandy or other bottom, or by hold-
ing it in place and kicking rocky rubble and gravel 1—3 m upstream from it. Some 8 specimens of
T. pectinatus were collected together with ornately banded mochokid catfish, Chiloglanis reticula-
tus Roberts, 1989, along the interface of sand and gravel. A second mochokid species, the mottled
C. batesii Boulenger, 1904, was collected only in rocky rubble and riffles along with a species of
the rheophilic cyprinid genus Garra. No other catfishes were collected at this locality. On 17
August 1973 a larger collection was made at the same spot using toxaphene.

DISTRIBUTION.— TJetracamphilius pectinatus is known from the mainstream of the Ubangui
upstream from Bangui, from a small tributary of the riviere Luala in the Lower Congo basin
(Bas-Congo), and from the Luala in southern Congo (Kasai).

EtTyMoLoGy.— The trivial name pectinatus is Latin for comb-like, in reference to the small
serrations on the pectoral fin.

MATERIAL EXAMINED

HOLOTYPE.— REPUBLIC DU CONGO (KINSHASA): CAS 92627, 33.7 mm gravid female, Congo
basin, tributary of R. Luala 26 km N of Kibunzi (8 km S of turnoff to Kibunzi, Luozi dist.), 45°6.5’
S, 13°48’ E, 24 July and 17 Aug. 1973, T. R. Roberts and D. J. Stewart.

PARATYPES.— REPUBLIQUE CENTRAFRICAINE: CAS 92628, 31:26.3—33.7 mm, collected with
the holotype (3:27.4-29.2 mm cleared and stained with alcian and alizarin; 7:26.3-29.6 mm,
cleared and stained with alizarin); CAS 92629, 2:23.4—24.5 mm, R. Ubangui, riffles in sand bar at
Isle Basiki, 75 km upstream from Bangui, 6 March 1988, T. R. Roberts; CAS 92630, 5:21.2—26.3
mm, R. Ubangui, riffles in shallow sandy area near Banda, 72 km upstream from Bangui, 5 March
1988, T. R. Roberts (2:21.9—24.0 mm cleared and stained with alcian and alizarin); CAS 92631,
23.3 mm, R. Ubangui upstream from Bangui, 1 March 1988. REPUBLIC DU CONGO (KINSHASA): T.
R. Roberts; CAS 92632, 28.6 mm, R. Lulua just below Chutes de Mbombo, near Luluaburg
[=Kananga], 7 Sept. 1986, T. R. Roberts.

ROBERTS: LEPTOGLANINAE 105

Tetracamphilius angustifrons (Boulenger, 1902), new comb.
(Figs. 19, 20a)

Anoplopterus angustifrons Boulenger, 1902:42, pl. 10 fig. 4 (type locality “Banzyville” [=Mobaye.on the riv-
i¢re Ubangu1]).
Amphilius angustifrons Boulenger, 1911:362, fig. 282.

DraGnosis.—A Tetracamphilius differing from all other species except 7: clandestinus in hay-
ing a relatively elongate snout with an enlarged olfactory organ; length of lamellar portion of olfac-
tory organ nearly equal to eye diameter; and anterior and posterior nostrils widely separated, pos-
terior much larger than anterior. From 7: clandestinus it differs in having more dorsal fin rays (8-9
vs. 6-7; more anal fin rays (9-10 vs 7-8); and more pectoral fin rays (9 vs 7-8).

NOTES ON SYNTYPES.—BMNH and MRAC syntypes are all in very poor condition. Probably
they were accidentally dried when the species was being described, because the drawing of the
whole specimen evidently was based on a specimen in very good condition, while the drawing of
the dorsal view of the head evidently was based on a specimen that had dried. This apparently led
to the erroneous illustration of the posterior nostrils and the statement “posterior nostril midway
between eye and end of snout” (Boulenger 1911:362). On the MRAC syntype I did not find poste-
rior nasal openings in the position indicated by Boulenger. I found the large posterior nostril only
after careful searching with fine forceps. The openings were tightly pressed to the dorsal surface of
the lateral ethmoids, only slightly in advance of the eyes, and were very difficult to see until gen-
tly lifted.

Amphilius angustifrons was known only from the original series of 5 syntypes, from the
Ubangui River near Banzyville (near Bangui), all of them now in very poor condition (Skelton,
1986: 266: present observations). Judging from the original figure, at least one of the specimens
was in good condition when studied by Boulenger (1902; 1911).

One of the two BMNH syntypes was cleared and stained for this study. The preparation was
not satisfactory, however. Bone and cartilage stained well, but soft tissues were stained deep black-
ish blue. The specimen broke into pieces, and some bits were lost. The caudal fin skeleton and most
of the fins rays disintegrated. The head has been dissected and sketches and observations made on
the jaws, vomer, gill arches, branchiostegal rays, and auditory capsules. The multicuspid jaw teeth
(most fallen out) were observed and drawn. In this specimen, the multicupsid jaw teeth included
individual teeth with up to 6 cusps; teeth with the cusps all broken off are left with broadly round-
ed distal ends. Vertebral counts were obtained from radiographs of three syntypes.

REMARKS ON IDENTIFICATION OF SPECIMENS.— Identification of freshly collected material
from Ubangui as T. angustifrons is based mainly on direct comparison with one of the syntypes, a
larger specimen in very poor condition. Identification of the specimens from the Chutes Wagenia
is problematic. The head is narrow anteriorly, as in 7: angustifrons, but the color pattern is some-
what more like that of specimens identified as 7: notatus. The eyes are smaller than in specimens
identified as T. angustifrons or T. notatus. If correctly identified, these are the only specimens of 7:
angustifrons known from a locality other than the Ubangui.

DISTRIBUTION.—Tetracamphilius angustifrons is known only from the Ubangui mainstream
above Bangui, with the exception of one lot of specimens tentatively identified as T. angustifrons
from the Chutes Wagenia, Lualaba.

MATERIAL EXAMINED

TYPE MATERIAL.— REPUBLIC DU CONGO (KINSHASA): MRAC 1313, 39.4 mm, Banzyville,
1901, Royaux (syntype; the other specimen from this lot, 19.8 mm, has been reidentified as

106 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 5

Zaireichthys mandevillei); BMNH 1901.12.27.40, 37.2 mm, Banzyville, 1901, Royaux (syntype;
cleared and stained with alcian and alizarin, but specimen did not clear and disintegrated, see
below).

ADDITIONAL MATERIAL EXAMINED

REPUBLIQUE CENTRAFRICAINE: CAS 92633, 9:17.4—31.7 mm, R. Ubangui at Isle Molinge, 82
km upstream from Bangui, 27 Feb. and 7 March 1988, T.R. Roberts; CAS 92634, 6:17.4-27.3 mm,
R. Ubangui at Isle Basiki, 75 km upstream from Bangui, 6 March 1988, T.R. Roberts; CAS 92652,
6:18.0-24.8 mm, sandy riffle in R. Ubangui near Banda, 72 km upriver from Bangui, 5 March
1988, T R. Roberts (cleared and stained with alcian and alizarin); CAS 92635, 15:15.2—26.6 mm,
R. Ubangui in shallow sandy area at Bawili, 68-69 km upstream from Bangui, 5 March 1988, T.
R. Roberts. REPUBLIC DU CONGO (KINSHASA): MRAC 90-47—P-704-708, 4:26.0— 29.3 mm, Congo
River in Chutes Wagenia near Kisangani, 24 June 1990, L. De Vos.

Tetracamphilius clandestinus Roberts, new species
(Fig. 20b)

HOLOTYPE.— REPUBLIQUE CENTRAFRICAINE: CAS 92653, 17.8 mm, riffles in shallow sandy
area of Ubangui near Banda, 72 km upstream from Bangui, 5 March 1988, T. R. Roberts.

DIAGNOsIS.— Tetracamphilius clandestinus, with largest known specimen 19.5 mm, is per-
haps the smallest leptoglanin species. It is very similar to 7: angustifrons but has slightly fewer ver-
tebrae, ribs, dorsal and pectoral fin rays, and differently shaped auditory capsules. Dorsal branched
fin rays 5 or 542 (6% or 72 in T. angustifrons). Pectoral fin branched fin rays 6 or 7 (usually 8, very
rarely 7 or 9 in T: angustifrons). Usually 3 pairs of ribs, sometimes 4 (usually 5 pairs of ribs, some-
times 4 in 7: angustifrons). In T. clandestinus and T: angustifrons of the same size, the fleshy ros-
trum usually is more pronounced in 7: clandestinus. Total vertebrae 17+18=35 in five cleared and
Stained paratypes (7: angustifrons and other species of Tetracamphilius usually with 36 or more
vertebrae; Table |). Consult diagnosis of 7: angustifrons.

DISTRIBUTION.— Tetracamphilius clandestinus is known only from the Ubangui mainstream
above Bangui.

ETYMOLOGY.— The trivial name clandestinus is Latin, meaning secret or hidden.

MATERIAL EXAMINED

HOLOTYPE.— REPUBLIQUE CENTRAFRICAINE: CAS 92653, 17.8 mm, riffles in shallow sandy
area of Ubangui near Banda, 72 km upstream from Bangui, 5 March 1988, T.R. Roberts.

PARATYPES.— REPUBLIQUE CENTRAFRICAINE: CAS 92654, 22:14.4—-19.5 mm, collected with
the holotype (5:17.3-19.2 mm cleared and stained with alcian and alizarin); CAS 92655,
2:15.2-15.7 mm, R. Ubangui near Bawili, 68-69 km upstream from Bangui, 5 March 1988, T.R.
Roberts.

Tetracamphilius notatus (Nichols & Griscom, 1917), new comb.
(Fig. 20c)

Amphilius notatus Nichols & Griscom, 1917:715, fig. 24 (type locality Faradje [=R. Uele, a large tributary of
the Ubangui]).

DIAGNOsIs.— A Tetracamphilius species with non-serrate pectoral fin spines, spots instead of
bands on the body, and olfactory organ not greatly enlarged.

ROBERTS: LEPTOGLANINAE 107

REMARKS ON IDENTIFICATION OF SPECIMENS.— This species was known only from the holo-
type. Specimens from the Chinko and the Mbomou were identified as 7: notatus by comparing
them directly with the holotype. Holotype has dorsal fin rays i642, anal ii7/2, pectoral i8/i7, caudal
14,6/7,13. Color pattern evidently has faded somewhat, but melanophores are still visible and the
original color pattern is therefore discernible. All features of color pattern identical with those in
freshly preserved specimens.

DISTRIBUTION.— Tetracamphilius notatus is known from the mainstream of the Ubangui,
from several of its larger and smaller tributaries, and from the R. Lufira in eastern Congo.

MATERIAL EXAMINED

TYPE MATERIAL.— REPUBLIC DU CONGO (KINSHASA): AMNH 6711, 28.5 mm, Faradje,
Jan.1913, Lang and Chapin (holotype).

ADDITIONAL MATERIAL EXAMINED

REPUBLIQUE CENTRAFRICAINE: CAS 92656, 2:24.9-26.7 mm, R. Chinko at Rafai, Republique
Centrafricaine, 30 Jan. 1987, T.R. Roberts (25.9 mm cleared and stained); CAS 92657, 25.2 mm,
rapids in R. Mbomou about 10 km SW of Rafai, Republique Centrafricaine, | Feb. 1987, T.R.
Roberts; CAS 92658, 19.1 mm, R. Ubangui, riffles in sand bar at Isle Basiki, 75 km upstream from
Bangui, 6 March 1988, T.R. Roberts; CAS 92659, 4:15.8—24.2 mm, R. Ubangui, riffles in sandy
shallow area near Banda, 72 km upstream from Bangui, 5 March 1988, T.R. Roberts; CAS 92650,
23.9 mm, R. Ngougofon near where it flows into R. Topia, about 120 km due E of Berberati, 23
March 1988, T.R. Roberts. REPUBLIC DU CONGO (KINSHASA): MRAC 87-61—P-3, 32.5 mm,
Kilwezi, affluent droit, de la riviére Lufira, alt. 800 m, 30 Aug. 1948, GF. de Witte.

OSTEOLOGY

This is a summary account of leptoglanin osteology. Adequate drawings for a reasonably full
osteological account of even a single species have not been completed. This is due to the inordi-
nate amount of time and effort involved. A comprehensive osteological account of just the crani-
um of one species should include four drawings: dorsal, ventral, lateral, and occipital view. A full
set of osteological drawings for one species would involve 15 or more figures. A comprehensive
osteological account of Leptoglaninae, including descriptions, comparisons, and analysis of rela-
tionships just within Leptoglaninae and Amphiliidae, could easily run to 100 pages.

Enough leptoglanin osteology is presented here to introduce the topic. It is intended to permit
discussion of morphological trends within Leptoglaninae. It may also be useful in discussions of
relationships within Amphiliidae. Ichthyologists primarily interested in the catfishes of Asia and
South America recently have investigated the possible relationships of Amphiliidae and
Leptoglaninae to various South American and Asian catfish families, in particular to the Asian fam-
ily Sisoridae (Pinna 1996; He and Meunier 1998; He et al. 1999). This is beyond the scope of this
paper.

The osteological features of Leptoglaninae showing the greatest diversity and specialization
are the cranium, jaws and dentition, pectoral fin girdle, and axial skeleton. The jaws of Leptoglanis
xenognathus are perhaps most specialized, but the palatine arch appears to be generalized (Figs.
4-7). In Leptoglaninae, teeth are found only on the jaws and pharyngeal arch. The multicupsid jaw
teeth of the genus Tetracamphilius are a notable specialization, especially for such small species,
but the palatine arch of Tetracamphilius (Fig. 19) and pharyngeal arches are relatively generalized.
The pharyngeal teeth are large and conical, not multicuspid. The gill arches of Leptoglaninae, char-

108 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 5

acterized by reduced gill rakers, appear to be morphologically generalized and relatively unspecial-
ized. Those of Leptoglanis (Fig. 8b) are morphologically very similar to those of Zaireichthys (Fig.
8c—e). The hyoid arch also appears to be relatively generalized and exhibits little osteological vari-
ation. Catfish families (e.g., Bagridae, Schilbeidae) often exhibit considerable range in the number
of branchiostegal rays. Leptoglaninae all have 4-8 branchiostegal rays. The lowest observed count
of 4 occurs in the tiny species Psammphyletria nasuta; the highest counts of 6-8 occur in the
largest species Leptoglanis xenognathus.

The pectoral fin girdle, as in catfishes generally, comprises three main bony elements: postem-
poral-supracleithrum, scapulocoracoid, and cleithrum (Diogo et al. 2001). There are also a meso-
coracoid bone, a cartilaginous complex radial, and a single bony proximal radial. Some catfishe
families, such as the African Claroteidae and Asian Bagridae, normally have two bony proximal
radials, but Amphiliidae including Leptoglaninae usually have only one (Fig. 9a—b). This nomen-
clature follows Diogo et al. (2001) and authors cited therein. Much of the variation in leptoglanin
pectoral girdle morphology (Fig. 9a—c) evidently is related to the variation in pectoral fin spines
from stout, rigid, and serrate to slender, flexible, and non-serrate, and the corresponding presence
or absence of a mechanism to lock the pectoral fin spine in erect position.

The paired fins, dorsal fin, and caudal fin show great variety of structure and numerous spe-
cializations. This is of course reflected in their girdles and other bony supporting structures.

One outstanding feature is the presence of dorsal and pectoral fin spines with locking mecha-
nisms in Zaireichthys. In this genus the pectoral fin spine is also strongly serrate. All other lep-
toglanins lack locking mechanisms for the dorsal and pectoral spines, and only one other species,
Tetracamphilius pectinatus, has a serrated pectoral fin spine (Fig. 9c). In Psammphyletria the pec-
toral fins rays are morphologically virtually the same as the pelvic fin rays. These fish effectively
have two pairs of pelvic fins, an anterior pair (the pectoral fins) and a posterior pair (the true pelvic
fins).

The axial skeleton exhibits numerous striking morphological differences. The range of verte-
bral counts from 33 to 44 is only a pale reflection of this. In some species of Zaireichthys the neu-
ral and hemal spines are relatively slender and morphologically generalized (Figs. 5c, 9—11b). In
Leptoglanis, Dolicamphilius, and Psammphyletria, on the other hand, these processes are hugely
expanded and morphologically highly specialized (Figs. 1d; 12;13d—e; 16a,c). Tetracamphilius and
some species of Zaireichthys are intermediate in this respect (Figs. lla, 14b). The degree of spe-
cialization of the dorsal and anal fin pterygiophores is correlated closely with that of the neural and
hemal processes (Figs. 1d; 5c; 9; 10c; 11; 12a,b: 13d—e; 14; 16a,c).

The caudal fin ray formulas of leptoglanins exhibit a remarkable range of principal fin ray
counts, from 7/8 in Leptoglanis xenognathus and Zaireichthys rotundiceps down to 5/6 in
Psammphyletria (Table 1). Caudal fin shape also varies markedly, from deeply forked, moderate-
ly forked, truncate or rounded (Figs. lc; 5b; 9; 10b; 11; 12a—b; 13c,e; 14; 16).

Despite these great differences in the caudal fin shape and ray counts, the caudal fin skeleton
shows remarkably little morphological variation (Figs. 1d; 5c; 9; 10c; 11a,b; 12a; 13d,e; 14b; 16c).
It consists mainly of a single upper and single lower hypurals. Sometimes the two elements are
entirely separate, as in L. xenognathus (Fig. 1d). They may be partially fused, as in species of
Tetracamphilius (Figs. 14b, 16a,c), or entirely fused, as in Psammphyletria (Figs. 13d,e). The prim-
itive principal caudal fin count in catfishes is 8/9 (Lundberg and Baskin 1969). This number occurs
in several species of Amphilius and other Amphiliinae (Table 1). It is reasonable to assume that 7/8
is the most primitive caudal fin ray count in leptoglanins. This primitive count, however, does not
appear to be associated with an equally primitive arrangement of the caudal fin skeleton. In lep-
toglanins the caudal fin skeleton consists mainly of a single upper and single lower hypural ele-

ROBERTS: LEPTOGLANINAE 109

ment. This may indicate that the presence of 7/8 principal caudal fin rays in Leptoglanis and in one
species of Zaireichthys is due to independent re-expression of a primitive catfish character trait.
The dorsal and pectoral fin spine locking mechanisms and pectoral fin spine serrae of Zaireichthys
may be further examples of the same phenomenon.

DISCUSSION

AFRICAN CATFISH FAMILIES AND LEPTOGLANINAE.— The freshwater catfishes of Africa cur-
rently are classified in seven families: Bagridae, Claroteidae, Mochokidae, Schilbeidae, Clariidae,
Malapteruridae and Amphiliidae. Their phylogenetic relationships and higher classification are not
well understood. All members of the first four families generally differ from Leptoglaninae and
agree with each other in sharing the primitive catfish characteristics of the defensive tripod, an
unencapsulated swim bladder, and 8/9 principal fin rays. The defensive tripod, an effective anti-
predator device, consists of more or less stout, serrated dorsal and pectoral fin spines that can be
locked in erect position. Bagridae is a large family present only in Africa and Asia. The large
endemic African family Claroteidae was recently removed from Bagridae (Mo 1991). Thus African
Bagridae now includes only the endemic African genus Bagrus. Bagrus differs from all other
bagrids, from Claroteidae, and from all other African catfishes in having a dorsal fin with 10-11
soft rays. There are only six or seven species. They are all large, with flat head and long barbels.
Claroteidae, Mochokidae, and Schilbeidae, with diverse head and body shape, usually have 7 soft
dorsal fin rays. The exclusively African Mochokidae have highly specialized jaws with pedicellate
multicupid teeth for browsing on algae, and strongly branched barbels. No other African catfishes
have branched barbels. Osteological characters of Mochokidae, especially of the jaws, cranium,
and pectoral girdle, indicate that they probably are related to the Auchenoglanidinae, a subfamily
presently assigned to Claroteidae. Close relationship between Mochokidae and Doradidae, a South
American family with branched barbels, seems unlikely. Schilbeidae and Clariidae are shared by
Africa and Asia. Schilbeidae all have a laterally compressed body, a very long anal fin, and long
barbels. Clariidae are distinguished from all other African catfishes by having a more or less elon-
gate, eel-like body form and eel-like locomotion and a highly apomorphic cranium. Clariid pec-
toral fins have a stout serrated spine with a mechanism to lock it into erect condition, but the dor-
sal fin is entirely soft-rayed. The clariid dorsal fin has numerous rays, and extends the entire length
of the body, unless interrupted posteriorly by the adipose fin. Then the dorsal fin with the adipose
fin extends virtually the entire length of the body. The caudal fin invariably is rounded. Given the
distinctive and specialized characteristics of Mochokidae, Schilbeidae, Clariidae, and
Malapteruridae, it is difficult to sustain hypotheses of close relationships between any of them and
Leptoglaninae.

The most viable hypothesis of a close relationship between Leptoglaninae and another family
of African catfishes is with Amphiliidae. Leptoglaninae agrees with Amphiliidae and differs from
other African catfish groups in two major ways: (1) absence of the defensive tripod (except in
Zaireichthys); and (2) bony encapsulation of the swim bladder. The hypothesis of Amphiliidae-
Leptoglaninae relationship is strengthened by presence of a “transverse interscapsular bony bridge”
in Andersonia leptura (He 1999; pers. obser.), a member of the rheophilic amphiliid subfamily
Doumeinae. Apart from Andersonia, this character is known only in Leptoglaninae. Leptoglaninae
differs from the two other amphiliid subfamilies, Amphiliinae and Doumeinae, in having an entire-
ly different life style and in lacking unculiferous pads on the ventral surface of their paired fins.
Andersonia is a highly specialized plated doumein, present in the Nile and other Sudanic drainage
systems. It is not closely related to Leptoglaninae. The possibility that Amphiliidae or

110 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 5

Leptoglaninae might be closely related to South American or Asian catfishes 1s beyond the scope
of the present work.

PLEOMERISM IN LEPTOGLANINAE.— Pleomerism, the tendency among related fish species for
vertebral number to be correlated with maximum body length (Lindsey 1975), has been document-
ed in several catfish families: Lindsey (1975; Ictaluridae and Mochokidae); Roberts (1983; sisorid
genus Bagarius); Roberts and Vidthayanon (1991; Pangasiidae). The phenomenon often can be
used by catfish systematists to distinguish closely related species differing in size. It may some-
times prove useful in predicting adult size of species known only from very small or immature
specimens, and in detecting valid species formerly placed in synonymy (Lindsey 1975). Maximum
standard length of leptoglanins ranges from a little under 20 mm to 65 mm (consult material exam-
ined) and total number of vertebrae from 34 to 44 (Table 2).

To simplify the present discussion, leptoglanins may be divided into four size classes: 1) less
than 20 mm; 2) 20-30 mm; 3) 30-50 mm; and 4) over 50 mm. The ranges of total vertebral counts
recorded for these classes are, respectively, 35 (only a single species, 7: clandestinus); 34-37 (six
species); 34-44 (six species); and 38-41 (only a single species, Leptoglanis xenognathus).

It may be noted that 34 is close to the lowest known vertebral count recorded in catfishes.

The very lowest vertebral counts in P. delicata (33), Z. zonatus (34), and T: clandestinus (35)
indicate that these probably are indeed very small species. The highest counts, 43-44 in
Dolichamphilius brieni, may indicate that this very rare species gets considerably larger than the
26.9-31.2 mm standard length of the only two specimens known. This count is also almost certain-
ly related to the exceptionally elongate or slender body of this species (see discussion of relation-
ship between pleomerism and body elongation in Lindsey, 1975). While the number of abdominal
vertebrae (19) is not exceptional, the numbers of postabdominal vertebrae (24—25) and peduncular
vertebrae (17) are the highest found in Leptoglaninae. The only known species that might be con-
generic with D. brieni, D. longiceps, has a less elongate body with only 40 vertebrae.

The lowest total vertebral counts recorded in Leptoglaninae occur in the smallest species, P.
delicatus (33), and in the stoutest species, Z. heterurus (33-34).

ZOOGEOGRAPHY OF LEPTOGLANINAE.— The geographical distribution of Leptoglaninae con-
trasts strikingly with that of the two other amphiliid subfamilies. Basic information on distribution
of Amphiltidae is provided by Skelton and Teugels (1986). All three subfamilies are well represent-
ed in the Congo basin. Amphiliinae occur throughout virtually all of tropical Africa, including the
Upper and Lower Guinean coastal areas, the Ogooué basin, Angolan coastal basins, and most of
eastern and southern Africa. Doumeinae also occur in Upper and Lower Guinea, the Ogooué, and
Angolan coastal basins, but are absent in southern and eastern Africa.

The only leptoglanin known from north of the Congo basin, Zaireichthys camerunensis, has
been reported only from the Niger basin (Risch 1992). The Niger basin is part of the Nilo-Sudanic
ichthyofaunal province recognized by Roberts (1975). It embraces the Nile, Chad, Niger, Senegal
and Volta basins. Most Nilo-Sudanic fish genera occur in all of these basins, and further collecting
may result in discovery of Z. camerunensis in other Nilo-Sudanic drainages. It is highly unusual
for an essentially Congolese fish group to have close phyletic relationship to fishes otherwise found
only in the Niger basin. Another instance of disjunct distribution involving Nilo-Sudanic and
Congolese fishes is provided by the rheophilic cichlid genus Steatocranus, with six endemic
species in the Congo basin and one in the Volta basin (Roberts and Stewart 1976).

The striking difference in the distribution patterns of Amphiliinae and Doumeinae versus
Leptoglaninae may well lay in their different habitat preferences. Amphiliinae and Doumeinae are
current-loving or rheophilic fishes typically living in high gradient streams with rocky substrate.
They often occur at high elevations in mountain streams, and some species may be classed among

ROBERTS: LEPTOGLANINAE 111

the oribatic or mountain-loving African fishes (Roberts 1975). Such fishes often occur in separate
drainages on opposite sides of drainage divides, possibly because of numerous stream captures as
mountain tops are eroded by their drainages. Leptoglaninae, on the other hand, are nearly all
restricted to very large lowland streams, inhabiting extensive low-gradient areas where the sub-
strate is predominantly or entirely fine sand. Such species are unlikely to cross over mountain
divides. On the other hand, they are likely to have very extensive ranges within any particular river
basin, as indeed seems to be so for several Congolese leptoglanins. Leptoglanis xenognathus and
Zaireichthys mandevillei are known from the Lower Congo (Stanley Pool), Ubangui, and Lualaba.
Tetracamphilius pectinatus also has an extensive range within the Congo basin.

A SPECIALIZED LIFESTYLE.— This account of Leptoglaninae may be concluded by a consider-
ation of the unusual “lifestyle” of the more specialized taxa. We are concerned here with all of the
species in the genera Leptoglanis, Dolichamphilius, and Psammphyletria, two species of
Tetracamphilius, and one of Zaireichthys. These are all small or very small species. The largest
known specimen of Leptoglanis is 65 mm and the largest known Dolichamphilius only 32 mm.
Psammphiletria nasuta reaches only 23.7 mm. Psammphyletria delicatus is perhaps the smallest
catfish in Africa. The largest specimen is only 20.5 mm. Although only two specimens are known,
this probably is close to its maximum size. Tetracamphilius angustifrons is known up to 39.4 mm.
Tetracamphilius clandestinus, at only 19.2 mm, is possibly the smallest of all known African cat-
fishes. Finally, we have Zaireichthys mandevillei, of which the largest known specimen is only 26.7
mm. Despite the fact that these species are all small, and that several of them are among the small-
est of all of the African freshwater fish species, they are known only from the mainstreams of two
of the largest rivers in Africa, the Congo and its largest tributary, the Ubangui.

These species live only in what may be described as great aquatic deserts, the outstanding fea-
ture of which is sand. These sandy deserts or plains often appear to be featureless. Seemingly uni-
form sandy stretches inhabited by Leptoglaninae are often very extensive, but individual fish are
not uniformly scattered over wide areas. Just as in the great African deserts, the sandy terrain has
some features frequented by more organisms than others. In the Ubangui River, where the greatest
diversity of Leptoglaninae has been found, they are most abundant in (1) sand riffles, where water
flows in streams from vast shallow areas into deeper water; (2) edges of sand bars, where vast fea-
tureless sand flats suddenly slope into deeper water; and (3) edges of hollows in the sand caused
by large rocks. An effective way to collect leptoglanins is to search for any kind of unusual feature
in the sand, then push a fine-meshed push-net deeply into it. The little fish often come out on top
of the sand when the net is removed from the water, or it may be necessary to let the sand filter out
through the mesh of the push-net. This is particularly effective for collecting small species and
small specimens of the larger species, but apparently is not so effective for catching the larger
species such as Leptoglanis xenognathus, presumably because they are too mobile. These may be
caught occasionally by seining at night. Fishing with a push-net in large uniform sandy areas (sand
flats) produces almost no specimens of Leptoglaninae, even if there is a good current. Leptoglanins
only occur in habitats with flowing water.

The aquatic insect larvae, crustaceans, and other small animals fed upon by leptoglanins prob-
ably are most abundant in the special niches in the aquatic sand desert occupied by the leptoglanins.
Leptoglanins are perhaps the only fishes in the Congo basin that can successfully exploit this spe-
cial food resource. Total or near total exemption from predation by fish may be another benefit con-
veyed by their small size and habitat. Their sand-diving behavior, mentioned above, may be relat-
ed to their feeding as well as to predator avoidance. Also possibly involved in detecting potential
food organisms and predators is the sense of olfaction. Leptoglanins have large olfactory organs.

While living in large rivers and having small body size have some obvious advantages, there

ie PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 5

probably are some disadvantages. How do they adjust to the marked habitat changes during the
annual cycle of low and high water periods? How do such tiny fish locate each other when it is time
to mate? During high water periods characterized by swift current and massive changes in the sand
banks, the leptoglanins presumably are widely dispersed. How do they manage to come together
again? During periods of low water they are found mainly in special places in the sand deserts, such
as riffles, sand-bar ridges or fall-offs, and troughs or holes created by isolated logs or rocks.
Individuals finding such a place are likely to encounter conspecifics if they remain there. As men-
tioned above, at least some leptoglanins have fairly large olfactory organs. These might function in
intraspecific communication as well as in finding food and avoiding predators. Probably all lep-
toglanins produce copious amounts of mucus. This mucus, secreted by the pectoral axillary gland,
might include pheromones or might itself act as a pheromone.

Sand-dwelling catfishes with a lifestyle and sand-diving behavior comparable to that of
Leptoglaninae apparently do not exist in the rivers of tropical Asia. I have searched for them with-
out success, especially in the mainstream of the Mekong and in its larger tributaries, but also in
other large Asian rivers. The only ecological equivalents among catfishes appear to be some of the
very small tropical South American Trichomycteridae and Pimelodontidae.

ACKNOWLEDGMENTS

Specimens were cleared and stained for skeletal study by Dave Catania (CAS), James
Chambers (BMNH), and Monique Beaunier (MNHN). Radiographs were prepared by Dave
Catania (CAS), Jeanine Abel (MNHN), and Miguel Parent (MRAC). Study material was loaned by
Dirk FE. Thys van den Audenaerde, MRAC; M.L. Bauchot, MNHN; Norma Feinberg, AMNH;
and Gordon Howes, BMNH. David Eccles kindly provided a copy of his unpublished studies of
eastern and southern African Zaireichthys. Donald J. Stewart also provided his unpublished notes
on Leptoglaninae. I am especially grateful to Luc De Vos (MRAC) for permitting me to study
Leptoglaninae collected by him in the Chutes Wagenia, riviére Avokoko, and other localities near
Kisangani in eastern Congo.

Many of the specimens were collected during my fieldwork in the Congo basin sponsored by
the Committee for Research and Exploration of the National Geographic Society 1n 1973 (grant no.
1107-72) and 1987-88 (grant no. 329-86). Additional financial support was provided by the
Institute for Biological Exploration, Smithsonian Tropical Research Institute, and Muséum
National d’Histoire Naturelle, Paris. The paper was written mainly in the Department of
Ichthyology, California Academy of Sciences. I am especially indebted to Dr. Tomio Iwamoto
(CAS) who, in my absence, kindly reviewed proof and dealt with the Proceedings editor on all mat-
ters relating to the final publication of this paper. However, I alone take responsibility for errors of
commission and omission that may be found herein.

LITERATURE CITED

BAILEY, R.M., AND D.J. STEWART. 1984. Bagrid catfishes from Lake Tanganyika, with a key and descriptions
of new taxa. Miscellaneous Publications, Museum of Zoology, University of Michigan, Ann Arbor (168):i-
iv, 141.

BOULENGER, G.A. 1902. Additions 4 la faune ichthyologique du Congo. Matériaux pour la faune du Congo.
Annales du Musée du Congo, Zoologie-Série 1, 2(2):19-57, pls. 7-16.

BOULENGER, G.A. 1911. Catalogue of the Fresh-water Fishes of Africa in the British Museum (Natural
History), vol. 2. Trustees of the British Museum (Natural History), London. 527 pp.

ROBERTS: LEPTOGLANINAE 113

BOULENGER, GA. 1915. Diagnoses de poissons nouveaux. II. Mormyrides, Kneriides, Characinides,
Cyprinides, Silurides. (Mission Stappers an Tanganika-Moero.) Revue zoologique africaine 4(2):162-171.

DacET, J., AND A. STAUCH. 1963. Poissons de la partie Camerounaise du bassin de la Bénoué. Mem. [FAN
68:84—107.

Dioco, R., C. OLIVEIRA, AND M. CHARDON. 2001. On the osteology and myology of catfish pectoral girdle,
with a reflection on catfish (Teleostei: Siluriformes) plesiomorphies. Journal of Morphology 249(2):
100-125.

He, S-P., AND J.P. MEUNTER. 1998. The biogeography of Amphiliidae and Sisoridae (Teleostei: Siluriformes).
Acta Zootaxonomica Sinica 23, suppl., 1-8.

He, S-P., M. GAYET, AND F.J. MEUNTER. 1999. Phylogeny of the Amphiliidae (Teleostei: Siluriformes). Ann.
Sci. Nat. 20(4):117-146.

HitGenporr, F. 1905. Fische von Deutsch und Englisch Ost-Afrika. Gesammelt von Oscar Neumann
1893-1895. Zoologischen Jahrbiichern 22(4):404—420.

LINDsEy, C. 1975. Pleomerism, the widespread tendency among related fish species for vertebral number to
be correlated with maximum body length. Journal of the Fisheries Research Board of Canada 32:
2454-2469.

LUNDBERG, J.G, AND J.N. BASKIN. 1969. The caudal skeleton of catfishes, order Siluriformes. American
Museum Novitates 2398, 49 pp.

Mo, T. 1991. Anatomy and systematics of Bagridae (Teleostei), and siluroid phylogeny. Koeltz Scientific
Books, Koenigstein. vili+216 pp., figs. 6-63 on unnumbered pages.

PELLEGRIN, J. 1926. Description de Siluridés, d’un Cyprinodontidé, et d’un Tetrodontidé. Récoltés au Congo
Belge par le Dr. Schouteden. Revue zoologique africaine 14:201—208.

PINNA, M.C.C. DE. 1996. A phylogenetic analysis of the Asian catfish families Sisoridae, Akysidae, and
Amblycipitidae, with a hypothesis on the relationships of the Neotropical Aspredinidae (Teleostei,
Ostariophysi). Fieldiana: Zoology, new ser., 84:1v+83 pp.

PINNA, M.C.C. DE. 1928. Poissons du Chiloango et du Congo receuillis par l’expédition de Dr. Schouteden
(1920-1922). Annales du Musée royal du Congo Belge (1)3(1):49 pp.

Poti, M. 1959. Recherches sur la faune ichthyologique de la région du Stanley Pool. Annales du Musée Royal
du Congo Belge, Teruvren (Belgique), série 8°, Sciences zoologiques 71:74—174, pls. 12-26.

PoLL, M. 1967. Contribution a la faune ichthyologique de 1’ Angola. Companhia de Diamantes de Angola
(Museu de Dundo), Publs. cult. 75, 381 pp., 162 figs., 20 pls.

Poti, M, AND J.-P. Gosse. 1995. Genera des poissons d’eau douce de l'Afrique. Mémoire de la Classe des
Sciences, Collection in-8°, 3°, Tome IX. Académie royale de Belgique, Bembloux. 324 pp.

Riscu, L. 1992. Bagridae. Pages 394-431 in C. Leveque, D. Paugy, and GG Teugels, eds., Faune des pois-
sons d’eaux douces et saumdatres de l'Afrique de l'Ouest. ORSTOM/MRAC (Organisation pour la
Recherche scientifique et technologique d’ outre-Mer/Musée Royale pour I’ Afrique Centrale).

Roserts, T.R. 1967a. Rheoglanis dendrophorus and Zaireichthys zonatus, bagrid catfishes from the lower
rapids of the Congo River. Ichthyologica 1967(July-Dec.):119-131.

RoBerTs, T.R. 1967b. Tooth formation and replacement in characoid fishes. Stanford Ichthyological Bulletin
8(4):23 1-247.

Roserts, T.R. 1975. Geographical distribution of African freshwater fishes. Journal of the Linnean Society,
London 54(4):249-3 19.

Roserts, T.R. 1982. Unculi (horny projections arising from single cells), an adaptive feature of the skin of
ostariophysan fishes. Zool. Scripta 11:54—-76.

Roperts, T.R. 1983. Revision of the Asian sisorid fish genus Bagarius, with description of a new species from
the Mekong. Copeia 1983(2):434-445.

Rosperts, T.R. 1989a. The freshwater fishes of western Borneo (Kalimantan Barat, Indonesia). California
Academy of Sciences, Memoir 14. 210 pp.

Roperts, T.R. 1989b. Systematic revision and description of new species of suckermouth catfishes
(Chiloglanis, Mochokidae) from Cameroun. Proceedings of the California Academy of Sciences 46:
151-178.

Roperts, T.R. 1992. Revision of the striped catfishes of Thailand misidentified as Mystus vittatus (Bloch

114 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 5

1794), with descriptions of two new species with numerous gill rakers (Pisces, Bagridae). Ichthyological
Exploration of Freshwaters 3(1):77-88.

ROBERTS, T.R. 1994. Systematic revision of the tropical Asian bagrid catfish genus Mystus sensu stricto, with
description of a new species from Thailand and Cambodia. Ichthyol. Expl. Freshw. 5(3):241—256.

ROBERTS, T.R., AND C.J. FERRARIS. 1998. Review of the south Asian sisorid catfish genera Gagata and Nangra,
with descriptions of a new genus and five new species. Proceedings of the California Academy of Sciences
50(14):314-345.

ROBERTS, T.R., AND J. JUMNONGTHAI. 2000. Miocene fishes from Lake Phetchabun in north-central Thailand,
with descriptions of new taxa of Cyprinidae, Pangasiidae, and Chandidae. Natural History Bulletin, Siam
Society 47:153-189. \

Roserts, T.R., AND D.J. STEWART. 1976. An ecological and systematic survey of fishes in the rapids of the
lower Zaire or Congo River. Bulletin of the Museum of Comparative Zoology, Harvard University 147(6):
239-317.

RoBerts, T.R., AND C. VIDTHAYANON. 1991. Revision of the tropical Asian catfish family Pangasiidae with
biological observations and descriptions of three new species. Proceedings of the Academy of Natural
Sciences of Philadelphia 143:97-144.

SEEGERS, L. 1989. Leptoglanis wamiensis (Bagridae, Siluriformes) aus dem Wami-Einzug in Tanzania. Die
Aquarien- und Terrarien- Zeitschrift (Datz) 42:284—-287.

SEEGERS, L. 1996. The Fishes of the Lake Rukwa Drainage. Annales du Musée royale pour |’ Afrique Centrale,
Tervuren, 278. 407 pp.

SKELTON, P. 1984. A systematic revision of species of the catfish genus Amphilius (Siluroidei, Amphiltidae)
from east and southern Africa. Annals.of the Cape Province Museum of Natural Histtory 16(3):41-71.
SKELTON, P. 1986. Two new Amphilius (Pisces, Siluroidei, Amphiliidae) from the Zaire River system. Africa.

Revue zoologique africaine 99:263—291.

SKELTON, P. 1989. Descriptions of Two New Species of West African Amphiliid Catfishes (Siluroidei:
Amphiliidae). J.L.B. Smith Institute of Ichthyology, Spec. Publ. 48. 13 pp.

SKELTON, P. 1993. A Complete Guide to the Freshwater Fishes of Southern Africa. Southern Book Pub.,
Harare. xiv + 388 pp.

SKELTON, P., AND G. G. TEUGELS. 1986. Amphiliidae. Pages 54-65 in J. Daget, J.-P. Gosse, and D.F.E. Thys
van den Audenaerde, eds., Check-List of the Freshwater Fishes of Africa, vol. I. ISNB/MRAC/ORAS-
TOM (Institut des Sciences naturalles de Belgique/Musée Royale pour |’ Afrique Centrale/Organisation
pour la Recherche scientifique et technologique d’outre-Mer). xiv + 520 pp.

ROBERTS: LEPTOGLANINAE

FIGURES 1-20

115

116

PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Voiume 54, No. 5

ROBERTS: LEPTOGLANINAE

bony swimbladder capsule

first rib

centrum of fifth vertebra

centrum of fourth vertebra

intercapsular bony bridge

FicureE |. Leptoglanis xenoganthus, 41.4 mm, Stanley Pool. Dorsal and ventral views of
the bony swim bladder capsules. The intercapsular bridge is a shared specialization or synapo-
morphy apparently common to all leptoglanins and to the doumein amphiliid genus
Andersonia.

dorsal fin rays

dorsal fin spine

neural spine

anterodorsal process

lateral process
ligament hemal spine

anteroventral processes

ribs 1-4

FiGureE 2. Leptoglanis xenognathus, 54.5 mm, Stanley Pool. Anterior portion of vertebral col-
umn and associated structures. Note specialized bicipital and ligamental attachment of ribs.

117

118

© L. xenognathus
@) Z. zonatus

@ Z. dorae

® z. flavomaculatus
® Z. heterurus

@® Z. mandevillei
©) Z. ef rotundiceps
(6) D. brieni

® D. longiceps

@) P. nasuta

© P. delicata

@®) T. pectinatus
@ T. angusticeps
© T. clandestinus
® T. notatus

(0)
SE
@Q,_fO®
i ica

PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 5

5°

ge

10°

20° De 30°

FiGuRE 3. Distribution of Leptoglaninae in the Congo basin. Note clustering of species records on the Congo
mainstream near Kinshasa (formerly Leopoldville), on the upper Congo mainstream or Lualaba at Kisangani (for-
merly Stanleyville), and especially in the Ubangui River upstream from Bangui. This reflects collecting activity as
well as presence of favorable habitat.

ROBERTS: LEPTOGLANINAE 119

Ficure 4. Leptoglanis xenognathus: a and c, holotype, total length 55 mm, Ubangui, dorsal view of head and full lat-
eral view; b, 62.2 mm, Stanley Pool, ventral view of head; d, 53.4 mm, Stanley Pool, axial skeleton (vertebrae 21+18=39)
(a and c from Boulenger 1911, Fig. 272).

120 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Voiume 54, No. 5

pterotic supracleithrum
frontal

lateral ethmoid
nasal
mesethmoid

sphenotic

metapterygoid
posttemporal

infraorbitals {

premaxilla supraoccipital

neural process

swimbladder

maxilla eonetis
lacrimal
dentary yy
palatine < ;
F hyomandibular
coronomeckelian SS yo

angulo-articular

anaes quadrate Z opercle
——— interopercle preopercle

FiGure 5. Leptoglanis xenognathus, 53.4 mm, Stanley Pool. Cranium, suspensorium and jaws (dorsal view).

interopercle

quadrate preopercle
metapterygoid opercie
ectopterygoid hyomandibular

supracleithrum
basioccipital

mandible

vertebrae 2+3+4
vertebra 5

vertebra 6

first rib

[ERP Aa
gamed aes ets: ne
“s ‘
S R PEE |S Fe ;
Z a
een
———S

premaxilla S | ie : eae SEs.

/ ay / / ; : e ot: rate ; :

vomer / Wiss 5 teenie intercapsular
/ arasphenoid % PELE E

mesethmoid =| | E \\ AES bridge

lateral ethmoid is ee prootic Sete Ro)” swimbladder
orbitosphenoid frontal sphenotic / N capsule
l 3mm pterotic exoccipital

FIGURE 6. Leptoglanis xenognathus, 53.4 mm, Stanley Pool. Cranium, suspensorium and jaws (ventral view).

ROBERTS: LEPTOGLANINAE 121

ectopterygoid metapterygoid hyomandibular

quadrate

| 2mm | See opercle
a interopercle
preopercle

coronomeckelian

dentary

. Meckel's cartilage
b angulo-articular aan

FiGureE 7. Leptoglanis xenognathus, 53.4 mm, Stanley Pool. a, suspensorium and jaws (medial view); b, lower jaw

(lateral view).

1D PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 5

ceratohyal

basibranchials 1-3
branchiostegal rays 1-8

hypobranchials 1-3

ceratobranchials 1-4

epibranchials 1-4

upper pharyngeal toothplate lower pharyngeal

1mm pharyngobranchials 3-4
[Se

Ficure 8. Leptoglaninae, hyoid and gill arches. a—b, Leptoglanis xenognathus, 54.5
mm, Stanley Pool; c. Zaireichthys zonatus, 18.1 mm, Kinsuka.

ROBERTS: LEPTOGLANINAE 123

FIGURE 8 (cont.). d, Zaireichthys camerunensis, 30.0 mm, riviére Dele; e, Zaireichthys mandev-
illei, 22.3 mm, Banda; f. Psammphyletria nasuta.

124 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 5

scapulocoracoid

bony proximal radial

cleithrum

cartilaginous complex radial

Ficure 9. Leptoglaninae, paired fins and their girdles. a, Leptoglanis xenognathus; pectoral girdle and
fin, 54.9 mm, Stanley Pool; b, Zaireichthys zonatus, pectoral girdle and fin (pectoral fin spine locked in
erect position), 18.1 mm, Kinsuka.

ROBERTS: LEPTOGLANINAE

FIGURE 9 (cont.). c, Tetracamphilius pectinatus, pectoral girdle and fin, Luala watershed; d,
Psammphyletria nasuta, pectoral girdle and fin, 23.2 mm, Banda: e, Dolicamphilius brevis, pelvic girdle

and fin, 31.2 mm, Stanley Pool.

125

126 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 5

———— =

FIGURE 10. Zaireichthys zonatus, rapids below Stanley Pool. a, dorsal view of head (24.5 mm holotype); b, lateral view
of body (24.5 mm holotype); c, lateral view of axial skeleton (18.1 mm paratype) (vertebrae 16+18=34).

pterotic

posttemporal
mesethmoid

———

eee (\ ‘ Ss

| | - / /
Va Sat
/ 1 A>

X\

; dorsal fin
lacrimal - \ /

spines 1-2

palatine ;
swimbladder capsule

infraorbitals

fe dae SU Seely supraoccipital

FiGuRE 11. Zaireichthys zonatus, 18.1 mm, rapids beow Stanley Pool. Dorsal view of cranium (with upper jaw).

ROBERTS: LEPTOGLANINAE 17

hyomandibular

metapterygoid

ectopterygoid

opercle

coronomeckelian

reopercle :
Ce angulo-articular

\
interopercie quadrate (ieee RCTS a

FicureE 12. Zaireichthys. Suspensorium and lower jaw. a, Z. zonatus, 18.1 mm
paratype, rapids below Stanley Pool; b, Z. camerunensis, 30.0 mm, riviére Dele; c, Z. man-
devillei, 22.3 mm, Banda.

SESERSGEEEG

= — SS AROS

FiGureE 13. Zaireichthys camerunensis, 24.1 mm, riviére Dele. Axial skeleton (vertebrae 20+18=38).

128 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 5

FiGureE 14. Zaireichthys heterurus. a, dorsal view of head, 31.4 mm holotype; b, lateral view, 31.4 mm
holotype; c) lateral view of axial skeleton, 23.0 mm paratype (Lulindi) (vertebrae 16+17=33).

FiGuRE 15. Zaireichthys, axial skeleton. a, Z. mandevillei, 21.5 mm, fleuve Ubangui near Banda (verte-
brae 17+17=34); b, Z. rotundiceps, 24.1 mm, riviere Luwoyeye (vertebrae 18+17=35).

ROBERTS: LEPTOGLANINAE 129)

Ficure 16. Dolicamphilius. a, D. brieni, 31.2 mm paratype, Stanley Pool, axial skeleton (vertebrae 18+24—43): b
and c, D. longiceps, 42.2 mm holotype, Chutes Wagenia near Kisangani, lateral view; dorsal view of head and paired
fins.

130

PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES

Volume 54, No. 5

C SS Non ‘ \ - . | [ wel
NY

AN YAS
== Uae)

aa:
SSS

eral view; d, P. nasuta,

2 ¥%9)

FiGure 17. Psammphyletria. a-c, P. nasuta, 23.5 mm holotype, Ubangui, dorsal and ventral view of head, full lat-
paratype, Stanley Pool, axial skeleton (vertebrae 17+16=33).

mm paratype, Ubangui, axial skeleton (vertebrae 18+18=36); e, P. delicata, 19.5 mm

ROBERTS: LEPTOGLANINAE 131

Ficure 18. Tetracamphilius pectinatus. a, 33.7 mm gravid female holotype, Luala watershed, full lateral view; b,
28.3 mm paratype, Luala watershed, axial skeleton (vertebrae 18+18=36).

hyomandibular

quadrate : 1mm
ectopterygoid [eee Bh Ziel Rca
LEE

palatine

metapterygoid maxilla

7

opercle

dentary

preopercle angulo-articular

: : Meckel's cartilage
interopercle coronomeckelian

FicureE 19. Tetracamphilius angustifrons, 22.3 mm, Ubangui, medial view of suspensorium and lower jaw.

132 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES

Volume 54, No. 5

a
DTA TEE ;

s \ SS —
SX
SS

ee

FiGureE 20. Tetracamphilius. a, T. angustifrons, 19.1 mm, Ubangui, axial skeleton (vertebrae 18+17=35); b, T. clandes-

tinus, 17.8 mm holotype, Ubangui, full lateral view; c, T. notatus, 25.9 mm, riviére Chinko, axial skeleton (vertebrae
19+18=37).

Copyright © 2003 by the California Academy of Sciences
San Francisco, California, U.S.A.

PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES

Volume 54, No. 6, pp. 133-140, 23 figs. March 14, 2003

Four Species of the Spider Genus Steatoda (Araneae:
Theridiidae) from the Gaoligong Mountains, Yunnan, China

Chang-Min Yin!, Charles E. Griswold?, You-Hui Bao! and Xiang Xu!
! College of Life Science, Hunan Normal University, Changsha, Hunan Province 410081 P. R. China;
? California Academy of Sciences, San Francisco, California 94118, U.S.A.

Four species of the genus Steatoda are described from the Gaoligong Mountains
region of Yunnan Province, China. Three new species, i.e., Steatoda mainlingoides,
Steatoda pardalia, and Steatoda tortoisea, are described as is the previously unknown
female of Steatoda terostiosa Zhu, 1998.

There are 16 species of the genus Steatoda known from China (Zhu 1998; Platnick 2002). In
this paper we describe three new species and the hitherto unknown female of Steatoda terostiosa
Zhu, 1998, which were collected in the Gaoligong Mountains by the first and second Sino-America
expeditions, 1998 and 2000 respectively. The type specimens are deposited in the College of Life
Science, Hunan Normal University, and some paratypes are in California Academy of Sciences.

MATERIALS AND METHODS

Specimens were fixed in 75% ethanol for 24 hours, then transferred to 85% ethanol for preser-
vation. Vulvae were cleared in lactic acid. Examination was via an Olympus Tokyo BH-2 stereomi-
croscope. The following leg segment measurements were taken (femur, patella + tibia, metatarsus,
and tarsus) and these measurements summed to give total length. All measurements are in mm. All
length and width measurements are of the largest dimension of the structure in question. All scale
bars equal 1.0 mm except figs. 22 and 23 in which they equal 0.1 mm.

ABBREVIATIONS USED.— AER=anterior eye row; ALE=anterior lateral eye, AME=anterior
median eye, MOQ=median ocular quadrangle, MOQA=MOQ anterior, MOQP=MOQ posterior,
PER=posterior eye row, PLE=posterior lateral eye, PME=posterior median eye, PME-PME=inter-
val between PMEs, PME-PLE=interval between PME and PLE.

Steatoda mainlingoides, new species
Figures 1-8

TypEs.— Holotype 2 and paratype ¢ from native forest at pass over Gaoligongshan at
2100m, Nankang, 36 air km SE TengChong, 24°50’ N, 98°47’ E, Baoshan Prefecture, Yunnan,
China, collected 4-7 October 1998 by C. Griswold, D. Kavanaugh and C.-L. Long, deposited in
HNU (No. 98-TC-7).

ETYMOLOGY.— The specific name refers to its similarity to Steatoda mainlingensis (Hu and
Li 1987).

DIAGNOsIS.— This new species is similar to Steatoda mainlingensis, but can be distinguished
by: (1) the narrow median markings on the abdominal dorsum (Fig. 1), whereas that of S. mainlin-

133

134 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 6

gensis has broad median and posterior triangles (Zhu 1998: fig. 229A); (2) the subtriangular epig-
ynum (Fig. 5), whereas that of S. mainlingensis is ellipsoid (Zhu 1998: fig. 229B); (3) the vulval
tube is longer and curved (Fig. 8), whereas that of S. mainlingensis is very short and straight; and
(4) the S-shaped conductor and short embolus of male palpal organ (Figs. 2, 3) are very different
from S. mainlingensis (Zhu 1998: fig. 229D, E).

FEMALE.— Carapace and sternum black. Carapace integument with small punctures along
grooves, margins ornamented with hairs. Cervical and radial grooves distinct, caput somewhat ele-
vated. Fovea transversely concave. AER recurved, PER almost straight. Chelicerae, palpi, endites
and leg femora black-brown, other leg segments red-brown. Chelicerae with fang orange-yellow,
promargin with two teeth and no retromarginal tooth (Fig. 7). Abdominal dorsum deep grayish
black, covered with white-yellow hairs; with white forming a crescent pattern anteriorly, a series
of small markings on the median line, and two pairs of markings laterally (Fig. 1). Center region
of venter grayish black, with yellow-brown striae laterally. Spinnerets brown to yellow-brown, sur-
rounded by white membranous ring. Epigynum subtriangular (Fig. 5), protruding posteromedially
(Fig. 6), spermathecae ovoid, connective tube curved (Fig. 8).

MALE.— Body coloration, patterns and eye arrangement as in female, only leg formula dif-
fers. Palpal bulb with S-shaped conductor (Fig. 3) enlarged at distal end (Fig. 2), embolus conical,
tapered (Figs. 2, 4).

MEASUREMENTS.— 2 total length 7.45—11.20. 3 6.20—-7.50. Holotype ¢ total length 9.90.
Cara-pace length 2.70, width 2.00; abdomen length 6.50, width 5.70. Eye sizes and intervals:
AME=ALE=PME=PLE 0.25; AME-AME =AME-ALE 0.05; PME-PME= PME-PLE 0.15; MOQ
length 0.44, anterior width 0.50, posterior width 0.55; CH 0.55. Leg measurements: ? I: 4.30 +
4.75 + 3.50 + 1.70 = 14.15, II: 3.90 + 3.50 + 2.50 + 1.40 = 11.30, III: 2.60 + 2.90 + 2.00 + 1.20'=
8.70, IV: 3.50 + 4.50 + 3.30 + 1.80 = 13.10; leg formula I, IV, I, Ill. % I: 5.30 + 5.50 + 4.10 + 1.90
= 16.80, I: 5.00 + 4.50 + 3.30 + 1.80 = 14.40, II: 3.50 + 3.50 + 2.50 + 1.50 = 11.00, IV: 4.70 +
4.90 + 3.90 + 1.90 = 12.40; leg formula I, I, IV, II.

MATERIAL EXAMINED

CHINA.— Yunnan: Baoshan Prefecture: pass over Gaoligongshan at 2100m. Nankang, 36 air
km SE TengChong, 24°50’ N, 98°47’ E, native forest, 4-7 October 1998, C. Griswold, D.
Kavanaugh and C.-L. Long (holotype 2, paratype ¢, HNU No. 98-TC-7); native forest at 2300m,
Luoshuidong, 28 air km E TengChong, 24°57’ N, 98°45’ E, flight trap over stream, 26-31 October
1998, C. Griswold, D. Kavanaugh and C.-L. Long (121d CAS, 22 HNU No. 98-TC-6). Nujiang
Prefecture: Nujiang State Nature Reserve, Qiqi He, 9.9 air km W Gongshan, 27°43’ N, 98°34’ E,
2000m, 9-14 July 2000, H.-M. Yan, D. Kavanaugh, C.E. Griswold, H.-B. Liang, D. Ubick and D.-
Z. Dong (1& subadult HNU No. 00-QF-56).

Steatoda pardalia, new species
Figures 9-13

TypEs.— Holotype ¢ and paratype ¢ from Gaoligongshan in Nujiang State Nature Reserve
at 2775m near No. 12 Bridge Camp area, 16.3 air km W of Gongshan, 27°43’ N, 98°30’ E, Nujiang
Prefecture, Yunnan Proy., China, collected 15—19 July 2000 by H.-M. Yan, D. Kavanaugh, C.E.
Griswold, H.-B. Liang, D. Ubick and D.-Z. Dong, deposited in HNU (No. 00-QD-12).

ETYMOLOGY.— The specific name refers to the abdomen dorsal pattern, which is similar to
leopard’s spots (Fig. 11).

DIAGNOsSIS.— This new species is very distinct from other Chinese Steatoda. The pale network

YIN, GRISWOLD, BAO AND XU: NEW CHINESE STEATODA SPIDERS 135

FicureEs 1-8. Steatoda mainlingoides, new species. 1, 5—8. Female. 2-4. Male left pedipalpus. 1. Habitus, dorsal. 2.
Prolateral. 3. Retrolateral. 4. Ventral. 5. Epigynum, ventral. 6. Epigynum, lateral. 7. Chelicera, retrolateral. 8. Vulva, dorsal.

pattern on the abdominal dorsum (Fig. 11) is somewhat similar to Steatoda tortoisea, new species
(Fig. 21), but S. pardalia can be distinguished by (1) lacking the epigynal scape (Fig. 12) charac-
teristic of S. tortoisea (Fig. 22) and (2) the thick, meandering copulatory duct (Fig. 13) that differs
from the short, curved duct of S. tortoisea (Fig. 23).

FEMALE.— Carapace brown, margin black-brown, lateral margins smooth. Cervical groove
indistinct, head region slightly elevated. AER procurved, PER slightly procurved. Fovea trans-
versely concave, two V-shaped markings behind eye region to fovea. Sternum brown, with hairs.
Chelicera brown, promargin having three teeth and no retromarginal tooth. Endite and labium
brown proximally, milk white distally. Palpi and legs yellow-brown but middle section of femora
and distal ends of patellae and tibiae black-brown. Abdomen ovoid, dorsum gray-black, clothed
with pale network pattern, dividing the dorsum into 11 black patches, five in middle and three on
each side (Fig. 11). Abdomen brown ventrally with one pair of white scale markings. Periphery of
spinnerets black-brown, colulus distinct, color same as spinnerets. Posterior margin of epigynum
procurved, with thickened rim, rim of atrium protruding (Fig. 12). Spermathecae spherical, connec-
tive duct V-shaped, inner arm extending forward to atrium (Fig. 13).

MALE.— Habitus as in female except lateral margins of carapace serrate. Palpal bulb from pro-
lateral view (Fig. 9) with anterior part of median apophysis robust, cubical, posteriorly elongate,
curved into a sickle-shaped part, from retrolateral view (Fig. 10), embolus short, conical, accom-
panied by the conductor on its right side.

136 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 6

Ficures 9-13. Steatoda pardalia, new species. 9, 10. Male left pedipalpus. 11-13. Female. 9. Prolateral. 10.
Retrolateral. 11. Habitus, dorsal. 12. Epigynum, ventral. 13. Vulva, dorsal.

MEASUREMENTS.— & total length 2.93. Carapace length 1.70, width 0.96; abdomen length
1.76, width 1.43. 3 total length 2.85, Carapace length 1.11, width 1.00; abdomen length 1.74,
width 1.36. Eye sizes and intervals: 2/3: AME=ALE 0.10, PME=PLE 0.11; AME-AME 0.04,
AME-ALE 0.03, PME-PME 0.09, PME-PLE 0.04; MOQ length= anterior width 0.20< posterior
width 0.23, carapace height 2 0.26, 3 0.23. Leg measurements: 2 I: 1.20 + 1.34 + 0.89 + 0.69 =
4.12, Il: 1.00 + 1.14 + 0.71 + 0.63 = 3.48, III: 0.91 + 0.94 + 0.54 + 0.60 = 2.99, IV: 1.14 + 1.30 +
0.80 + 0.63 = 3.87; leg formula I, IV, II, III. 3 I: (missing) + 1.57 + 1.49 + (missing) = (not calcu-
lated), II: 1.29 + 1.11 + 0.80 + 0.59 = 3.79, III: 1.06 + 0.97 + 0.67 + 0.53 = 3.23, IV: 1.34 + 1.29
+ 0.81 + 0.68 = 4.12; femur I probably longer than femur IV, leg formula probably I, IV, I, II.

DISTRIBUTION.— Yunnan, China.

MATERIAL EXAMINED

Only the types.

YIN, GRISWOLD, BAO AND XU: NEW CHINESE STEATODA SPIDERS By,

Steatoda terastiosa Zhu, 1998
Figures 14-20

Steatoda terastiosa Zhu 1998:347, fig. 232A-C. Platnick 2002.

Note.— The male was described by Zhu (1998:346). We describe the female for the first time.

FEMALE.— Carapace, sternum, endites and labium flaming red (Fig. 20). Carapace clothed
with short black hairs that are denser on the lateral margins than on the median. Cervical and radi-
al grooves distinct, darker, caput elevated, with long black hairs. AER recurved, PER slightly
recurved; bases of ALE and PLE contiguous. Clypeus protruding. Fovea shallow, subtriangular.
Sternum with dense black hairs, triangular, posterior end long and pointed between coxae IV.
Chelicerae red brown, with two small promarginal teeth and no retromarginal tooth. Distal ends of
labium and endite pale white, endite with black margin. Palpal trochanter and coxa and trochanters
of legs flaming red, remaining segments black-brown. Abdomen subspherical, dorsum brown with
three distinct pairs of muscular depressions, the second largest and third smallest; white forming
two transverse bands on anterior half, one longitudinal median band and one pair of round marks
laterally on the posterior half (Fig. 14). Abdominal venter with large white, square mark before
spinnerets, gray patches beside white mark and surrounding spinnerets; base of anal tubercle with
regularly arranged brown hairs. Epigynum trapezoid, an ellipsoid atrium on anterior half, the pos-

Ficures 14-20. Steatoda terastiosa Zhu, 1998. 14-17. Male. 18-20. Female. 14. Habitus, dorsal. 15. Abdomen, ven-
tral. 16. Palpal organ, ventral. 17. Palpal organ, retrolateral. 18. Epigynum, ventral. 19. Vulva, dorsal. 20. Habitus, lateral.

138 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 6

terior half recurved, thickened, with black margin (Fig. 18). Spermathecae spherical, connecting
duct anterior, funnel-shaped (Fig. 19).

MALE.— (Figs. 14-17): Body smaller than female; for description see Zhu (1998:347, fig.
232A-C).

MEASUREMENTS.— 2 total length 11.43. Carapace length 3.43, width 2.97; abdomen length
5.77, width 5.09. Eye sizes and intervals: AME=ALE 0.17; AME-AME 0.11; AME-ALE 0.20.
PME 0.19; PLE 0.21, PME-PME 0.14, PME-PLE 0.23, MOQ length 0.40, anterior width 0.43, pos-
terior width 0.47; carapace height 0.29. Leg measurements: I: 4.06 + 4.69 + 3.57 + 1.83 = 14.15,
Il: 3.43 + 3.60 + 2.71 + 1.54 = 11.28, III: 2.91 + 2.86 + 2.09 + 1.31 = 9.17, IV: 4.37 + 4.63 + 3.31
+ 1.66 = 13.97; leg formula: I, IV, II, III.

DISTRIBUTION.— Yunnan, China.

MATERIAL EXAMINED

CHINA.— Yunnan Province: Kunming Prefecture: Kunming, Heilongtan District, Kunming
Institute of Botany botanical gardens, 25°08’ N, 102°45’ E, 1950m, 21-23 June 2000, D.
Kavanaugh and C.E. Griswold (22 HNU No. 00-KB-10, 12130 CAS).

Steatoda tortoisea, new species
Figures 21—23

Types.— Holotype 2 from Nujiang, 30°06’ N, 97°12’ E, Yunnan Province, China, collected
30 July 1981 Jia-Fu Wang, deposited in HNU (No. 97-102). Paratype, 12 subadult, from Liuku,
Nujiang Prefecture, Lushui Co., Liuku Township, 800m, 25°52’ N, 98°51’ E, Yunnan Prov., China,
collected 25—26 June 2000 by D. Kavanaugh, C.E. Griswold, H.-M. Yan and D. Ubick, deposited
in HNU (No. 00-LK-9).

ETYMOLOGy.— The specific name is derived from the tortoise shell-like pattern of the
abdomen dorsum (Fig. 21).

DIAGNOsIs.— The vulva is most similar to Steatoda cingulata (Thorell 1890), but S. cingula-
ta lacks the scape characteristic of S. fortoisea (Fig. 22), and the abdominal patterns of these two
species are distinct.

FEMALE.— Carapace pear-shaped, red-brown to black-brown. Cervical groove clear, caput
elevated. AER strongly recurved, PER almost straight, ALE and PLE contiguous at base. Fovea
transversely concave, radial grooves indistinct. Thoracic region clothed with red-brown hairs.
Sternum, chelicerae, endites and labium red-brown, chelicerae with 2 promarginal teeth and no
retromarginal tooth. Labium wider than long. Palpi and legs red-brown to black-brown. Abdominal
dorsum ovoid, black-brown, with tortoise shell-like pattern, three pairs of distinct muscle impres-
sions, the second pair largest (Fig. 21). Abdominal venter brown, with a square white marking
behind epigastric groove and a pair of pale subaxial striae, with ring-like mark surrounding the
spinnerets. Epigynum with scape, its lateral margins turned up towards the median (Fig. 22). Vulva
with two large spherical spermathecae (Fig. 23).

MALE.— Unknown.

MEASUREMENTS.— Holotype 2 total length 9.27; carapace length 3.97, width 3.76; abdomen
length 5.51, width 4.55. Eye sizes and intervals: AME=ALE=PME=PLE 0.23; AME-AME =PME-
PLE 0.09; AME-ALE 0.06; PME-PLE 0.13; MOQ length 0.51, anterior width 0.49, posterior width
0.46; carapace height 0.17. Leg measurements: I: 4.50 + 5.20 + 3.50 + 1.40 = 14.60, II: 3.50 + 3.50
+ 2.60 + 1.00 = 10.60, III: 2.60 + 2.50 + 2.05 + 1.05 = 8.20, IV: 3.90 + 4.55 + 3.40 + 1.05 = 12.90;

YIN, GRISWOLD, BAO AND XU: NEW CHINESE STEATODA SPIDERS 139

FIGURES 21—23. Steatoda tortoisea, new species, holotype female. 21. Habitus, dorsal. 22. Epigynum, ventral. 23. Vulva,
dorsal.

leg formula: I, IV, II, I.
DISTRIBUTION.— Yunnan, China.

MATERIAL EXAMINED

Only the types.

ACKNOWLEDGMENTS

Support for this research came from the China Natural History Project, the Foundation of
Natural Science of the Education Department of Hunan Province (China), the California Academy
of Sciences and the US National Science Foundation grant DEB BSI 074632456. We are grateful
to Prof. Heng Li and Prof. Chun-Lin Long for support for the 1998 and 2000 Sino-American-expe-
ditions to the Gaoligong Mountains and to Prof. Zhi-ling Dao for organizing the expeditions. We
also thank all the collectors who did the difficult fieldwork and provided us with specimens. This
is Scientific Contribution no. 23 from the California Academy of Sciences Center for Biodiversity
Research and Information (CBRI) and contribution no. 16 from the China Natural History Project
(CNHP).

140 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 6

REFERENCES

PLATNICK, N.I. 2002. The World Spider Catalog, version 2.5. American Museum of Natural History, online at
http://research.amnh.org/entomology/spiders/catalog8 1-87/index.html
ZHU, M.S. 1998. Fauna Sinica (Arachnida: Araneae: Theridiidae). Science Press, Beijing: 324-347.

Copyright © 2003 by the California Academy of Sciences
San Francisco, California, U.S.A.

PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES

Volume 54, No. 7, pp. 141-153, 7 figs., 1 table. March 14, 2003

A New Species of Bufo (Anura: Bufonidae) from Myanmar
(Burma), and Redescription of the Little-Known Species
Bufo stuarti Smith 1929

Guinevere O. U. Wogan!, Htun Win2, Thin Thin2, Kyi Soe Lwin?,
Awan Khwi Shein2, Sai Wunna Kyi?, Hla Tun2
! Department of Herpetology, California Academy of Sciences, Golden Gate Park, San Francisco CA 94118
2 Nature and Wildlife Conservation Division, Forest Department, Ministry of Forestry, Myanmar

A new species of Bufo is described from the Rakhine State near the Rakhine Yoma
Elephant Range in western Myanmar. This species is morphologically similar to
Bufo melanostictus, but it is distinguished by its smaller body size at maturity, the
bright breeding coloration of the males, and its advertisement call. The new species
brings the total number of bufonid species known to occur in Myanmar to seven; all
are referred to the genus Bufo. Additionally, Bufo stuarti Smith 1929 is redescribed
from the type specimen and more recently acquired specimens from the vicinity of
the type locality. Lastly, a key to the Bufo of Myanmar is provided.

Recent fieldwork in western Myanmar has resulted in the discovery of a new species of
bufonid that appears to be closely related to Bufo melanostictus. Further, the senior author during
a visit to the Zoological Survey of India, Calcutta was afforded the opportunity to examine the type
specimen of Bufo stuarti, one of the few southern Asian bufonid types that was not examined and
redescribed by Dubois and Ohler (1999). Bufo stuarti is a little known species that has been report-
ed only from northern Myanmar and Assam, India (Smith 1929; Frost 2002). Because of the dearth
of information on the species, and because we have obtained new material from the vicinity of the
type locality, “Putao Plain, N.E. Burma,” the holotype is herein redescribed with additional notes
based on our newly acquired materials.

MATERIALS AND METHODS

Animals were collected by hand by the authors. Latitude and longitude were recorded using a
Garmin 12 GPS with the datum set to WGS 84. Animals were euthanized and then preserved in
10% buffered formalin before being transferred to 70% ethanol. Specimens are deposited in the
California Academy of Sciences (CAS). Museum symbolic codes follow Leviton et al. (1985).

Measurements were taken using digital calipers to the nearest 0.1 mm (instrumental error of +
0.2 mm). Measurements follow Dubois and Ohler (1999). The following measurements were used
in the analysis: snout vent length (SVL), head length (HL), head width (HW), jaw to nares length
(MN), jaw to eye (MPE), jaw to front of eye (MAE), distance between front of eyes (IFE), distance
between back of eyes (IBE), internarial space (IN), distance from eye to nares (EN), snout length
(SL), nostril to tip of snout (NS), tympanum diameter (TympD), eye to tympanum distance (ET),
forelimb length (FLL), hand length (HAL), first finger length (FFL), third finger length (TFL),
tibia length (TL), foot length (FL), femur length (F), fourth toe length (FTL), inner metatarsal

141

142 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 7

tubercle length (IMT), outer metatarsal tubercle length (OMT), parotoid length (ParL), and paro-
toid width (ParW).

Ranges and means are provided in Table | for all of the above measurements and ratios for
HL/SVL, HW/SVL, ParW/ParL, FEM/SVL, TL/FEM, OMT/IMT, EN/MN, ParL/SVL, ParW/HW
are included.

Statistical analysis was carried out on measurements recorded for the lectotype of Bufo
melanostictus (ZMB 3462), given in Dubois and Ohler 1999, and from seventeen Bufo melanos-
tictus from throughout Myanmar (N=18), the new species (N=5), and Bufo stuarti (N=3) in the col-
lections of the California Academy of Sciences. To eliminate the need for allometric corrections
(Hayek et al. 2001), only adult specimens were included in analysis. SPSS Base 10.0 for Macintosh
(SPSS Inc. 1999) was used for statistical analysis.

Calls were analyzed using Signal 3.1/RTSD 1.1 (Engineering Design 1999) software and a
Kay Elemetrics DSP SonaGraph model 5500.

SPECIES DESCRIPTION

Bufo crocus Wogan sp. nov.
Figures 1-3

HoLotyPre.— CAS 220192 adult male from Myanmar, Rakhine State, Gwa Township,
Rakhine Yoma Elephant Sanctuary, Ye Bya Stream Camp, 17°41’47.0” N, 94°38’48.7” E., elev. 83
m, collected 25 April 2001 by J.B. Slowinski, GO.U. Wogan, Htun Win, Thin Thin and Kyi Soe
Lwin.

PARATYPES (4 specimens).— CAS 220193-220195 from Myanmar, Rakhine State, Gwa
Township, Rakhine Yoma Elephant Sanctuary, Ye Bya Stream Camp, 17°4147.0” N, 94°38’48.7”
E, collected 25 April 2001 by J. B. Slowinski, GO.U. Wogan, Htun Win, Thin Thin and Kyi Soe
Lwin. CAS 220331 from Myanmar, Rakhine State, Gwa Township, Rakhine Yoma Elephant
Sanctuary, 17°42’14.0” N, 94°38’54.3” E, collected 26 April 2001 by J.B. Slowinski, GO.U.
Wogan, Htun Win, Thin Thin and Kyi Soe Lwin.

DIAGNOsIs.— A moderately sized Bufo, snout-vent length in males 54.6—-59.1 mm females
67.6-67.8 mm, body stout, cornified canthal, pre-orbital and post-orbital crests present, and weak-
ly defined uncornified parietal ridges (terminology follows Dubois and Ohler 1999). Bufo crocus
possesses a protruding snout and a subterminal mouth and its head is wider then long. It has promi-
nent singular subarticular tubercles. Breeding males are bright yellow in coloration.

DESCRIPTION OF HOLOTYPE (Fig. |).— For ease in comparison with other Asian bufonids, the
description follows the format of Dubois and Ohler (1999).

(A) SIZE AND GENERAL ASPECT: (1) specimen of medium size (SVL 54.6 mm).

(B) HEAD: (2) head width (18.7 mm) is greater than head length (16.1 mm), (3) snout protrud-
ing (SL 5.9 mm), (4) canthus rostralis distinct, (5) interorbital space concave, distance between
front of eyes (7.9 mm) slightly greater than half of distance between back of eyes (15.1 mm), (6)
nostrils rounded, closer to tip of snout (NS 2.1 mm) than to eye (3.8 mm), (7) pupil horizontal, (8)
tympanum (TYMPD 2.2 mm) ovular, verticle, distinct and located 1.0 mm from eye (ET), (9)
pineal ocellus absent, (10) vomerine ridge absent, (11) tongue longer than wide, and unnotched,
(12) supratympanic fold absent, (13) parotoid glands present, oval, prominent, and less than twice
as long (ParL 10.7 mm) as wide (ParW 5.7 mm), (14) canthal, preorbital, supraorbital, postorbital,
supratympanic and (faint) parietal ridges present.

(C) FORELIMBs: (16) forelimb (12.7 mm) longer than hand (11.0 mm), enlarged, (17) fingers
long, (18) relative length of fingers, shortest to longest: Il<I<IV<III, (19) tips of fingers rounded,

WOGAN ET AL.: ANEW BUFO FROM MYANMAR 143

nena,

220193 female.

Ficures 1-2. (1) Bufo crocus, holotype, CAS 220192 male; (2) Bufo crocus, paratype, CA

Photos by Hla Tun.

not enlarged, without grooves, (20) fingers without dermal fringe, webbing absent, (21) subarticu-
lar tubercles prominent, rounded, single, (22) prepollex oval; one palmar tubercle large prominent
and reniform; numerous small rounded and pointed tubercles covering palm and fingers.

(D) HINDLIMBs: (23) femur (20.3 mm) and tibia (20.4 mm) of approximately equal length,
(24) toes long, toe IV half the length of the femur, (25) relative length of toes shortest to longest
I<Il<V<IH<IV, (26) tips of toes rounded, not enlarged, without grooves, (27) webbing (follows
Myers and Duellman, 1982) [2-3111-2 1/211111/2-3 1/21V3 2/3-1 1/2V, (28) dermal fringe along toe
V absent, (29) subarticular tubercles more prominent than the superneumerary tubercles, (30) inner
metatarsal tubercle prominent and round (1.5 mm), (31) tarsal fold absent, (32) outer metatarsal
tubercle prominent (absent from left foot), numerous tubercles on the plantar surface of the foot
and toes.

(E) SKIN: (33) anterior of snout and region between eyes and anterior portion of dorsum with
dense flattened keratinized warts, posterior portion of dorsum with less dense keratinized flattened
warts, flank with scattered glandular warts, (34) dorso-lateral folds absent, (35) forelimb, tarsus,
and leg with flattened keratinized warts, tips of fingers and toes keratinized, (36) row of keratinized
warts along lower lip, rest of ventral side with glandular flattened warts, not spiculate, (37) paro-
toid glands present.

(F) COLORATION IN PRESERVATIVE: (38) dorsal and lateral parts of head and body are light tan,
keratinized warts are black as are ridges, (39) dorsal parts of limbs are light tan, keratinized warts
are black, (40) ventral parts of head body and limbs are light with indistinct dark mottling, throat
region blue.

COLORATION IN LIFE: (38B) dorsal and lateral parts of head and body are bright yellow with
black keratinized warts, (39B) dorsal parts of limbs are bright yellow with black keratinized warts,
tips of toes and fingers are black, (40B) ventral parts of head body and limbs are white with pink,
throat yellow.

(G) MALE SEXUAL CHARACTERS: (41) nuptial spines on prepollex and fingers I and II, dorsal
side of finger I covered entirely, (42) inner openings of vocal sac present, round, (43) forearm
enlarged.

VARIATION WITHIN THE PARATYPIC SERIES.— The tubercles on the hands and feet vary in num-
ber and density and prominence. The palmar tubercle is reniform or round. The subarticular tuber-
cles of CAS 220195 are more prominent than those of the holotype.

SKIN: All of the paratypes agree with the description of the holotype, but the degree of black
keratinizations varies. CAS 220193 and 220194 have spiculate warts on the flanks and ventral side.

144 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 7

Goes Sima |

0) 20 40
KILOMETERS

FiGuRE 3. Maps with collection localities of Bufo crocus and Bufo stuarti. Collection localities of holotypes are indicat-
ed by stars.

In both of these specimens, the ventrum is almost entirely covered in minute spiculate asperities.

COLORATION: The other males of the paratypic series agree with the coloration of the holotype
except that they are a slightly darker tan shade, and the degree of mottling on the ventrum varies.
The females of this species differ markedly in coloration. In preservative (38) dorsal and lateral
parts of the head and body are tan to brown, in CAS 220193 there is an indistinct pattern of black
mottling intermixed with an umber background, keratinized warts are black as are ridges, (39) dor-
sal parts of limbs are brown, (40) ventral parts of head, body and limbs are light with dark mot-
tling, the mottling varies in degree

In life (38B) the dorsal and lateral parts of head and body are brown with black and red mar-
bling, (39B) dorsal parts of limbs are light brown, (40B) ventral parts of head, body, and limbs are
white with dark mottling.

NATURAL History.— Description of habitat: Bufo crocus is found in primary evergreen for-
est. Thus far, itis known only from two localities in the Rakhine Division in Myanmar. It is expect-
ed that as additional surveys are carried out the range will be extended, however, surveys conduct-
ed to the north and south of this region have not produced any individuals of this species.

REPRODUCTIVE BEHAVIOR.— Males in breeding condition are bright yellow. This degree of
sexual dichromatism has been observed in several other species of Bufo: B. macrotis (Taylor 1962),

WOGAN ET AL.: ANEW BUFO FROM MYANMAR 145

B. kisoloensis (Drewes and Vindum 1994), B. periglenes (Savage 1966), B. leutkeni (Villa 1972),
B. peripatetes (Savage and Donnelly 1992), and Bufo stuarti (Smith 1940). Drewes and Vindum
(1994) suggest that in voiced species (as is B. crocus) it is unlikely that the dichromaticism plays
a roll in mate selection.

Breeding occurs in congruence with the earliest rains of the monsoons, generally in late April
to early May. This species is an explosive breeder, hundreds were observed after heavy rains dur-
ing the night of April 26, 2001. Individuals were gathered around pooled water bodies. Males were
actively calling and amplexus was observed. Neither egg deposition nor larvae were observed.

COMPARISONS AND REMARKS.— Bufo crocus differs from all of the following Asian Bufo
belonging to the Bufo bufo, Bufo viridis, Bufo stomaticus, Bufo stejnegeri, and Bufo orientalis
groups (sensu Inger 1972) in possessing cephalic crests: B. bufo, B. bankorensis, B. gargarizans,
B. wrighti, B. japonicus, B. minshanicus, B. tibetanus, B. viridis, B. calamita, B. raddei, B. surdus,
B. latastei, B. luristanica, B. orientalis, B. dodsoni, B. stomaticus, B. olivaceus, B. dhufarensis, B.
beddomi. It differs from B. asper and B. juxtasper (Inger’s 1972 Bufo asper group) in lacking a
tarsal ridge. It can be distinguished from Bufo biporcatus, B. divergens, B. philippinicus, B. parvus,
B. quadriporcatus, and B. claviger, the members of the Bufo biporcatus group (Inger 1972),
through the presence of melanophores in tissue surrounding the vocal sac (Inger 1972). It is distin-
guished from the B. scaber group Dubois and Ohler 1999) (B. scaber, B. parietalis, B. silent-
valleyensis, B. atukoralei, B. kotagamai) in that it lacks very prominent parietal ridges.

Bufo crocus is tentatively placed in the Bufo melanostictus group based on Inger’s (1972) suite
of characters, but differs from all other species in this group. Included herein are comparisons with
all currently recognized species from Inger’s 1972 defined group (Bufo melanosticus, B. parietal-
is, B. stuarti, B. himalayanus), and Dubois and Ohler’s (1999) redefined Bufo melanostictus group
(B. melanostictus, B. himalayanus, B. cyphosus, B. microtympanum, B. noellerti, B. stuarti). From
Bufo stuarti, B. crocus can be distinguished by its very distinct cephalic ridges, which are lacking
in B. stuarti. From B. microtympanum, it differs in having a tympanum that is much larger in rela-
tion to the eye and in the absence of double distal tubercles on fingers three and four. It is readily
distinguishable from B. noellerti both on the basis of geography, B. noellerti is endemic to Sri
Lanka, and in the absence of a dark patch extending from the tympanum onto the flanks (see
Manamendra et al. 1998): from B. cyphosus by the absence of symmetrical stripes on the female’s
ventrum (Dubois and Ohler 1999); and from B. himalayanus in that it possesses a distinct
supratympanic cephalic crest. Bufo crocus differs from B. melanostictus in its smaller body size, in
possessing weak parietal ridges and singular subarticular tubercles, and in the absence of paired
warts between the parotoid glands.

Of the above species, B. crocus appears most similar to Bufo melanosticus. Besides differing
in the above mentioned characters, it also differs in its call, reproductive behavior, and habitat pref-
erence. The advertisement calls of the two species differ markedly in structure, dominant frequen-
cy, duration and pattern (Figs. 4-5). The B. melanostictus call recorded from Myanmar is consis-
tent with findings reported by Grosjean and Dubois (2001) in that it is comprised of a series notes
made up of a variable number of pulses. The call of Bufo crocus consists of a series of notes with-
out the distinct pulses.

Smith (1917), van Kampen (1923), and Church (1960) reported Bufo melanostictus as breed-
ing throughout the year. Boulenger (1912) however, reported the breeding season as March and
April on the Malay Peninsula. Bufo crocus as reported above is an explosive breeder in which
breeding ocurrs with the earliest monsoon rains. Unlike B. melanostictus, it is also not commonly
found outside of this breeding period. Additional searches carried out at the type locality in June
of 2001, and late January 2002 yielded only one individual. Lastly, the two species are found in

146 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 7

=
3
=
=
€
a=
Tina fare}
5.0
aS
6.6
PIE
=~
= a0 ; fF 1
ti iy
3 i f « ‘ 3 je} 2 tei ti
z j
= as ; ; } i
| 7.
3.0 : : ; cj
i ty wet FL 6. fe 4
2.5 A } i& AA gil’ 4. a | it,
ay qs Be 6g A } 4
oe St ; 3 mi a. ; & : + =.
2.91 ——3 — ap
1.0 11 12 1.3 1A 13 1.6 1.7 13 1.3 2.0
Time (220)
OF: 28Hs OT: 1h eme T-inc: Gams FFT 26 Wind:HANN
HEE OFF Le-t9 db Hi:-1db
yes
S 3
2 2 1
<

“We
aft =

2s 139 1.92 1.34 135 7.48 1.40 1.42 744 1,48
Tene (eec}

FiGurRE 4. Call of Bufo crocus (CAS 220331) recorded on April 26, 2001 at 22:35 hrs.
Temperature 28.05° C, 90% humidity after heavy rain. Myanmar. (A) spectrogram; (B) oscillogram;
(C) detailed oscillogram of notes 3 and 4.

very different habitat types. Bufo melanostictus is primarily a commensalistic species (Inger 1966;
Berry 1975; Inger et al. 1984; Dutta and Manamendra-Arachchi 1996; Dutta 1997; Inger 1997;
Iskandar 1998) and Bufo crocus is found in primary evergreen forest.

ETYMOLOGyY.— The specific epithet, crocus, is Latin for yellow or saffron. This is in reference
to the bright yellow breeding coloration of the males.

REDESCRIPTION OF BUFO STUARTI SMITH, 1929
Figures 3, 6-7

In 1929, Malcolm Smith described Bufo stuarti!, with the type being deposited at the
Zoological Survey of India in Calcutta (ZSI 19985). The following redescription of the type spec-
imen (Fig. 6) follows the format of Dubois and Ohler 1999:

(A) SIZE AND GENERAL ASPECT: (1) a medium sized frog (SVL 71.3 mm) (note that difference
in SVL from Smith’s description (73.0 mm) may differ due to time in preservation, or the current
use of a more precise measuring device).

(B) HEAD: (2) head of medium size, wider (26.2 mm) than long (21.9 mm) (3) snout obtuse,
protruding beyond mouth, (4) canthus rostralis distinct, (5) interorbital space slightly concave, (6)

WOGAN ET AL.: ANEW BUFO FROM MYANMAR 147

Araylitid Clcaltey

freepioncy [Kilzt

1.3 14 15 7.6 4.7 1.6 1.5 20
Thin {ree

OF. SSHz OT 10.2msa T-ine: 20ma FFT OS Wid:HANh

Si-Fit: OFF La: -? dh Hi: i mh

Amptiteds (Voit)

1.
Tima (sec)

FiGure 5. Call of Bufo melanostictus (USNM/field series 35608) recorded on March 2, 2002 at
21:00 hrs. Temperature 22.0° C. Myanmar .(A) spectrogram; (B) oscillogram; (C) detailed oscillo-
gram of notes between 1.4 and 1.5 seconds.

nostrils closer to snout than eye, (7) pupil not observed, (8) tympanum 2.4 mm in diameter, oval
and vertical, distinct, (9) pineal ocellus absent, (10) vomerine ridge not observed, (11) tongue not
observed, (12) supratympanic fold absent, (13) parotoid glands present, reniform and long (18.4
mm), (14) cephalic ridges absent.

(C) FORELIMBs: (16) arms robust and enlarged, (17) fingers long and thin, (18) relative length
of fingers 3>4>1>2, (19) tips of fingers not enlarged, rounded, grooves absent, (20) fingers with-
out dermal fringe, (21) subarticular tubercles prominent, (22) prepollex rounded and very promi-
nent, one prominent rounded palmar tubercle.

(D) HINDLIMBs: (23) leg measurements not recorded, (24) toes long, (25) relative length of
toes 1<2<5<3<4, (26) tips of toes rounded, not enlarged without grooves, (27) webbing formulae
11-2 1/2112-21/21112-31V2 1/2-1 1/2 V (28) dermal fringe along toe V absent, (29) subarticular
tubercles indistinct, (30) inner metatarsal tubercle long and prominent, (31) tarsal fold absent, (32)
outer metatarsal tubercle present, round and prominent; numerous tubercles on the plantar surface
of the foot.

(E) SKIN: (33) head and anterior of dorsum granular, posterior of dorsum with large kera-
tinized spiculate warts, (34) dorso-lateral fold absent, (35) dorsal parts of limbs with keratinized
warts, (36) throat, chest, belly and thighs granular, (37) parotoid glands present.

148 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 7

DS

FIGURES 6-7. (6) Bufo stuarti, holotype ZSI 19985. Photo by G.O.U. Wogan.
Photo by Hla Tun.

.

(7) Bufo stuarti, CAS 224954.

(F) COLORATION: (38) dorsal and lateral parts of the head and body are tan olive with brown
on warts, (39) dorsal parts of limbs are tan olive, (40) ventral parts of head, body, and limbs are
whitish tan.

(G) MALE SEXUAL CHARACTERS: (41) nuptial spines present on digits I, II, III (42) vocal sacs
present.

Recent herpetological surveys made in the vicinity of the type locality have yielded three adult
Bufo stuarti (Fig. 5). Herein I include these specimens in order to provide a more thorough descrip-
tion of B. stuarti. Ranges and means are provided in Table 1.

Additional observations on the recent specimens: (5) interorbital space distinctly concave, (7)
pupil horizontal (Fig. 7), (10) vomerine ridge absent, (11) tongue not emarginate, (13) parotoid
glands of CAS 224954 have a slightly triangular shape, CAS 221317 and 221439 are elongate and
approximately twice as long as wide, (30, 32) measurements of the inner and outer metatarsal
tubercles show that the outer metatarsal tubercle is larger, (33) the dorsal and lateral parts of the
head and body of CAS 221317 and CAS 224954 are more coarsely granular and warty, however,
CAS 224954 lacks asperities on the head, (34) dorsal parts of limbs are very granular with large
asperities and warts, (38) CAS 221317 dorsal and lateral parts of the head and body are dark brown
to tan, with a faint mid-dorsal line, parotoids umber, sides are white with tan and dark brown warts.
CAS 224954 dorsal and lateral parts of the head and body are dark brown to umbre with a black w
shaped pattern, black triangular mark between eyes, faint mid-dorsal line, parotoids umber, (39)
CAS 221317 and CAS 224954 dorsal parts of limbs are barred with light tan, (40) CAS 221317
ventral surfaces of head, body, and limbs light with dark brown mottling varying in density, CAS
224954 is heavily mottled with dark brown on a light tan background.

! Original description (Smith 1929): “Adult male collected on the Putao plain, N.E. Burma, near the Tibetan frontier
by Dr. Murray Stuart, after whom it is named. Habit like that of B.melanostictus. Crown without bony ridge; snout as long
as the upper eyelid, prominent, projecting forwards beyond the lower jaw; canthus rostralis distinct; loreal region almost
verticle; interorbital space a little broader than the upper eyelid; tympanum distinct, half the diameter of the eye. First fin-
ger a little longer than second, third finger nearly twice as long as second; two well marked carpal tubercles, the outer larg-
er and flatter than the inner. Toes half webbed, the membrane not reaching the tips of the third and fifth toes, subarticular
tubercles, not very prominent; no tarsal fold, two well marked metatarsal tubercles. The tarso-metatarsal articulation reach-
es to the tip of the snout; the heels meet when the legs are folded at right angles to the body. Skin of the head fairly smooth,
of the back and limbs above with smooth warts of moderate size. Parotoids well developed, elongate, twice as long as
broad, parallel with each other. Skin of the lower parts coursely granular. Nuptial asperities on the inner three fingers. Pale
73 mm. Bufo stuarti is closely related to B. stomaticus
Lutken from which it can be distinguished by the more prominent snout, the strong canthal ridges and by the absence of a
tarsal fold.”

olive above, uniform: brownish-white below. From snout to vent

WOGAN ET AL.: ANEW BUFO FROM MYANMAR 149

TABLE |. Means and ranges for morphometric characters.

Character Bufo crocus Bufo crocus Bufo melanostictus — Bufo melanostictus Bufo stuarti Bufo stuarti
sigle males (N=3) females (N=2) males (N=9) females (N=9) male females (N=2)

Min. Mean Max. Min. Mean Max. Min. Mean Max. Min. Mean Max. Min. Mean Max.
SVL 54.6 56.533 59.1 67.6 67.7 67.8 70.5 77.878 93.4 67.8 91.644 150.4 60.7 61.0 68.6 76.2
HL 16.1 17.233 18.0 19.2 19.35 19.5 21.7 23.867 28.5 20.7 28.611 46.1 17.2 170)" 18395: 20:9
HW 18.7 20.567 21.5 242 242 242 27 30.822 38.9 26.2 36.122 62.2 21.6 23.3 25.05 26.8
MN 13.6 15.033 169 160 164 168 186 20.222 23.1 16.8 24.611 37.3 14.2 1359) 16:0» ase
MPE 5.4 6.0 CSCO GSS) TOSesess 7956 wes 4 O23 Se 8S 3.6 4.0 5.3 6.6
MAE ED WE S67 ES = WO 1225126 152 16244 SS ADO 1871 28 10.4 Oe HLS ill
IFE Dee GT OF 28:98 9105 892 er:80) ee OO 4 Se e2 SS 838 8.2 8.1 8.8 9.5
IBE Sele LG: 0s Gr elie 1 S:450 192 LON 20:7 89 25s Meas 228225) 35.1 15.6 15970) 181
IN A)- S}VRR)) 9 Shy) A) 4.2 AD) 325145225) S60 Bile 52445 oe 847 3.6 35) 3.9 4.3
EN 3.8 4867 68 45 455 46 41 4.744 59 3.7 5.667 8.3 Bl) 4.1 4.7 53
SL SOF 6567-71 84 8.5 8.6 6.9 8.278 10.4 7.0 9.5 15.9 6.5 6.2 7.1 8.0
NS Pail 228M Py Tl PUR = PES EPI PATS ess DO je SEIDEL Se BY0) 6.5 22 4.2 6.2
TympD D2 2433) 9 3:0) 219 3.3 B42) S027) 9s OMe F425 2M aed 2.2 Lee 2:4 5 SD
ET OR O67 2 lee 9 1555 OP O!S m7 27 2) lds 3: OIE ae 6.2 1D) 14 1.65 1.9
ParW Sse O253— 06s 1-9 7.9 US. O33 8.0 95 5.6 7.144 11.0 13.4 10.9 12.35 13.8
ParL 1057 11-133 11-4 11-9 12.4 12:9 14.1 ~16:878 21.1 13:8" 21.211 37.2 5.3 6.4 14.25 22.1
BEE 2Siiag 13:39 13°9P 15:3) 15955 16:6" Won 187252 424 2053892987) 16.1 16.3 18.15 20.0
HAL PO-2 13:2 = 14°8) 15:8" 16:0) 16:2 1610) 18'478" 212-1537") 21-3" 345 15.9 15.4 17.9 204
Ie 20.4 20.9 21.3 24.7 248 249 25.5 29.944 36.0 24 32.564 51.7 25.8 24.9 28.45 32.0
BE 20.3 21.3 22.3 23.7 23.85 24 25.8 28.589 32.4 22.2 31.878 52.8 26.3 24.7 28.55 32.4
FEM 20.3 21.567 22.8 24.6 26.35 28.1 28.4 33.175 41.5 247 34.811 55.8 26.8 25.7 26.15 26.6
IMT tS 2.033025) 225 2255) 2167) ea eS STS 43 elk 23.9227 16:6 3.2 25 3.5 4.5
OMT OG) 12:0) 18

UT 1.8 19 2.1 2916 41 23 2.967 4.0 3.8 Bh HebIR A SH
MPE/MAE 0.48 0.5181 0.55 0.5 0.518 0.53 0.37 0.4885 0.66 0.36 0.6375 1.54 0.35 0.4 0.4539 0.5

HL/SVL 0.29 0.3049 0.32 0.28 0.2858 0.29 0.29 0.3064 0.32 0.3 0.3132 0.33 0.28 0.27 0.2765 0.28
HW/SVL 0.34 «0.3636 0.38 0.36 0.3575 0.36 0.37 0.3945 0.43 0.37 0.3927 0.41 0.36 0.35 0.3668 0.38
ParW/ParL 0.53 0.5593 0.6 0.61 0.6381 0.66 0.42 0.4753 0.52 0.27 0.349 0.43 2193 0.49 1.3247 2.16
FEM/SVL 0.37 0.3817 0.41 0.36 0.3892 0.41 04 0.4203 0.45 0.36 0.3813 0.42 0.44 0.35 0.3852 0.42
TL/FEM 0.93 0.9705 1.0 0.89 0.9451 1.0 0.87 0.9234 0.99 0.89 0.9379 0.97 0.96 0.97 1.0859 1.2

OMT/IMT 0.4 0.5638 0.72 0.68 0.7054 0.73 0.22 0.7531 1.08 0.61 0.7962 1.14 1.19 OFS 72
EN/MN 0.27 0.3186 0.4 0.27 0.2775 0.28 0.21 0.234 0.26 0.21 0.2316 0.27 0.26 0.29 0.2939 0.29
ParL/SVL 0.19 0.197 0.2 0.18 0.1832 0.19 0.2 0.2162 0.23 0.2 0.2294 0.29 0.09 OS 02197510229
ParW/HW_ 0.29 0.3032 0.32 0.33 0.3264 0.33 0.24 0.26 0.29 0.16 0.2016 0.23 0.62 0.41 0.4995 0.59

DISCUSSION

Despite the taxonomic confusion within the Asian Bufonidae and particularly the genus Bufo,
recent work (Dubois & Ohler 1999; Macey et al. 1998; Liu et al. 2000) has started to untangle the
confusion surrounding these taxa. The taxonomic position of Bufo stuarti has remained uncertain
because of its rarity in collections. Inger (1972) originally placed it in the Bufo melanostictus group
thereby indicating the presence of cephalic ridges, however, Smith (1929) mentions a similarity
with Bufo stomaticus in the original description (refer to footnote 1). In our examination of the type

150 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 7

and subsequent collections, no cephalic ridges were found. We suggest that additional studies
should be undertaken before reassigning it to a different species group. Bufo crocus as mentioned
previously is tentatively assigned to the Bufo melanosticus group based on the prescribed suite of
characters from Inger’s 1972 designation. Its taxonomic position, however, is uncertain pending
additional analysis.

As the Myanmar Herpetological Survey moves into new areas, assuredly new and interesting
bufonids will be found in Myanmar. To date, no representatives of the genera Ansonia,
Leptophryne, and Pedostibes have been found within Myanmar, although each has been reported
in neighboring countries. Frost (2002), for instance, suggests that at least three bufonid species, two
of Ansonia (A. penangensis, A. malayana) and one Leptophryne (L. borbonica) should occur in
Myanmar. In addition, as work is carried out on widespread composite species, such as Bufo
melanostictus, it is likely that additional species masquerading under this name will be discovered
within Myanmar.

KEY TO THE BUFO OF MYANMAR

Pe€ephalie nidgesipresent si... oe ae. Ste wets «ere ees ee ee ae 2
Cephalic ndges not present... 7. ta. 3. ee. Bae Ss Moo. Ba eee 3
paCephalic ridges mn parallelitoparotoids. S. -wiegyac tae soe eee. oe eee B. parvus
Cépahlic ridges not parallel... 36 ass ses. be ont cage CaS eee Oa ee eee 5
Se MyM PANU GISHIMCE <0... 282.5 fis se cae eusucheh save, ols keues “elo ouss ekeneke: ore wc oo Sere AY aise. oy See 4
iv MpPanuMU NOC WISIOLE. |. ous Gy. - axsstuarsiacielter ca nick een one Oe B. burmanus?
4. Tympanum of equal size to eye, parotoids small and round.................... B.macrotis
Tympanum much smaller thenveye, parotoids clongate....4. ...- 25-141. oe B. stuarti
)., larsalinidge present, supratympanic ridges prominent .....- i... -\.2cts- ee B. asper
Marsalinid @evabSents...0..055, aye cithelevesonsts clceseate seas ee > onuay= seek ence ose Gal rec 6
6. Adult size, 62-150 mm; paired dorsal warts between parotoid glands ....... B. melanostictus
Adult size, 54-67 mm; paired warts between parotoid glands absent.............. B. crocus

MATERIAL EXAMINED

Bufo andrewsi: China, CAS 214911—214915, Yunnan Province, Nu Jiang Prefecture, village S
of Gongshan, 27°42’13.7” N, 98°42’10.2” E, ca 1451 m.

Bufo asper: Malaysia, CAS-SU 14833, Pinang; Thailand, CAS 73689, Sritamarat, Nakon;
Myanmar, CAS 222196, Mon State, Kyaik Hto Township, Kyaik-Hti-Yo Wildlife Sanctuary, trib-
utary of Moe Baw Chaung, 17°29’48.5” N, 97°04’49.6” E.

Bufo bufo: China, CAS 16484-16487, Shanghai.

Bufo burmanus: Myanmar, MCZ 23440-23444, Chin State

Bufo crocus: Myanmar, CAS 220192-—220195, Rakhine State, Gwa Township, Rakhine Yoma
Elephant Sanctuary, Ye Bya stream camp, 17°41’47.0” N, 94°38’48.7” E; CAS 220331, Rakhine
State, Gwa Township, Rakhine Yoma Elephant Sanctuary, 17°42’14.0” N, 94°38’54.3” E; CAS

? Bufo burmanus is maintained in this key in accordance with Frost 2002, but it should be noted that Dubois and Ohler
(1999) synonymized this species with B. pageoti. B. burmanus is known only from Myanmar and has not been obtained in
recent collections.

WOGAN ET AL.: ANEW BUFO FROM MYANMAR 151

222923, Rakhine State, Gwa Township, Rakhine Yoma Elephant Wildlife Sanctuary, 17°35’03.9”
N, 94°40'52.6” E; CAS 222970, Rakhine State, Gwa Township, Daung stream, 17°35’03.9” N,
94°40'52.6” E.

Bufo dhufarensis: Saudi Arabia, CAS 134168—134169, vicinity of Matri and Khasawiyah
[16°58’ N., 42°42’ E — 16°56’ N, 42°37’ E], 20-25 m.

Bufo gargarizans: China, CAS 194198—194201, Anhui Province, Luan Prefecture, 23 km SW
and 10 km south of Yanzhihe, also 20 km WSW of Manshuihe (31°08’ N, 116°01’ E), Beimazhei.

Bufo himalayanus: China, CAS 177560—177568, Xizang (Tibet) Autonomous Region, Xigaze
Prefecture, between Chinese check point at Zhangmu (Khasa) [28°07’ N, 85°59’ E] and the Nepal
border on the Lhasa-Kathmandu Rd., 2100-2300 m.

Bufo macrotis: Myanmar, CAS 213551—213558, Yangon Division, Hlaw Ga Park,
Mingalardon Township, 17°02’08.2” N, 96°06’00.6” E.

Bufo melanostictus: Myanmar, CAS 205186—205187, Rakhine State, Gwa Township, ca 300
m E of Pleasant Beach Resort, 17°43’28.8” N, 94°32’36.1” E; CAS 222981, Rakhine State, Taung
Gok Township, Taung Gok Town, 18°50’31.8” N, 94°12’31.8” E, 11 m; CAS 213763, Magwe
Division, Shwe Set Taw Wildlife Sanctuary, Pwint Byu Township, Oak Pho Camp, 20°19’38.5” N,
94°34’27.1” E; CAS 213621, Magwe Division, Shwe Set Taw Wildlife Sanctuary, Mimbu (Sagu)
Township, 20°03’34.3” N, 94°35’37.7” E; CAS 220562, Magwe Division, Saw Township, Saw
Town, Forest Department office, 21°08’52.9” N, 94°09’36.6” E; CAS 219956-219957, Chin State,
Min-Dat District, Kanpetlet Township, Nat Ma Taung National Park, Saw stream, 21°10°16.0” N,
94°04’26.1” E; CAS 219775, Ayeyarwady Division, Pya Bon District, Bogalay Township, Mein
Ma Hla Kyun Wildlife Sanctuary, West Htaw Pai Camp, 15°51’06.3” N, 95°18’40.6” E; CAS
213880, Magwe Division, Shwe Set Taw Wildlife Sanctuary, Lap Pan Taw Camp, Minbu (Sagu)
Township, 20°06’56.1” N, 94°36'38.4” E.

Bufo microtympanum: India, CAS 94446, 12.9 km NE of Munnar; CAS 104227-104228,
India, Madras State, Palni Hills, Kodaikanal.

Bufo orientalis: Saudi Arabia, CAS 139542-139543, near Al Hada at 20°22’ N, 40°16’ E, ca
2000 m.

Bufo parvus: Malaysia, CAS-SU 3242-3243, Perak; CAS-SU 3237, Perak, Gunong Kledang.

Bufo pentoni: Saudi Arabia, CAS 119214, Al Lith, 20°09’ N, 40°17’ E, ca 20 m.

Bufo raddei: China, CAS 166809-166819, Ningxia Hui Autonomous Region, Yinnan
Prefecture, along the south shore of the Yellow River (Huang He), Shenjiatan (37°28’ N, 105°18’
E).

Bufo scaber: Sri Lanka, CAS 85271, India, CAS 104139, Madras State, Kanyakumari District
(Cape Comorin District) Mylaudi Village, 4 mi. W. Jamestown.

Bufo stomaticus: Afghanistan, CAS 96172, Khost; Nepal, CAS-SU 15315, above Tambur
River.

Bufo stuarti: Myanmar, CAS 221436, Kachin State, Putao District, Naung Mon Township,
Aureinga Camp, 27°17'36.4” N, 97°51’50.0” E; CAS 221432, Kachin State, Putao District, Naung
Mon Township, Aureinga Camp, 27°17'32.4” N, 97°51'56.4” E; CAS 221317, Kachin State, Putao
District, Machanbaw Township, Ahtonga Village, 27°15’41.8” N, 97°48’06.9” E; CAS 2214835,
Kachin State, Putao District, Machanbaw Township, rd between Ahtonga and Babaw, 27°15’27.2”
N, 97°50’32.4” E; CAS 224954, Kachin State, Putao District, Ba Bawt Village, 27°21’33.4” N,
97°54’56.1” E, 640 m; CAS 224955, Kachin State, Putao District, Ba Bawt Village, 27°22’39.3” N,
97°53'46.7” E, 556 m; ZSI 19958, “Putao Plain, N. Myanmar” (Kachin State).

Bufo surdus: Yran, CAS 158247, Seistan and Baluchestan Province, W slope of Kuh-e-Taftan,
2250 m.

Sy PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 7

Bufo viridis: Turkey, CAS 217837-217845, Icel Province, Anamur, 36°04’17.7” N, 32°52’
03.47 E, 6 m.

ACKNOWLEDGEMENTS

Specimens for this report were obtained by the Myanmar Herpetofaunal Survey, a joint pro-
gram of the Myanmar Nature and Wildlife Conservation Division, the California Academy of
Sciences, and the Smithsonian Institution’s National Museum of Natural History, with support
from National Science Foundation grant DEB-9971861. The authors thank U Shwe Kyaw, Director
General, Forest Department, Ministry of Forestry, and U Khin Maung Zaw Director of the
Myanmar Nature and Wildlife Conservation Division, Ministry of Forestry for continued assis-
tance and support. We thank George Zug for the use of his Bufo melanostictus recording, Michelle
Koo for the preparation of the map, and Jens Vindum, Ricka Stoelting, and Jeff Wilkinson for read-
ing and commenting on the manuscript. Financial support for Wogan, Htun Win, Thin Thin, Hla
Tun, Kyi Soe Lwin, Awan Khwi Shein, and Sai Wunna Kyi was provided by NSF grant DEB-
9971861. Additional support for Wogan was obtained from the California Academy of Sciences
Lindsay Fund, DAPTF/Conservation International Seed Grant and San Francisco State University
—COSE Student Project Funds. Additional funding was made possible through the generous dona-
tions to the Department of Herpetology by Malcolm and Jeanne Miller, David L. Jameson, Emily
Date and the late Richard Dumke. Finally, we would like to thank our dear friend and advisor, the
late Dr. Joseph Bruno Slowinski, for his tireless patience, his guidance, and most of all for instill-
ing in us much of his enthusiasm for the study of the Myanmar fauna.

LITERATURE CITED

BOULENGER, G.A. 1912. A Fauna of the Malay Peninsula from the Isthmus of Kra to Singapore Including the
Adjacent Islands. Reptilia and Batrachia.Yaylor and Francis, London, UK. xiii + 294 pp.

Berry, P.Y. 1975. The Amphibian Fauna of Peninsular Malaysia. Tropical Press, Kuala Lumpur, Malaysia. x
+ 130 pp.

Cuurcu, G. 1960. Annual and lunar periodicity in the sexual cycle of the Javanese toad, Bufo melanostictus
Schneider. Zoology 44:81-188.

DREWES, R.C. AND J.V. VINDUM. 1994. Amphibians of the Impenetrable Forest, Southwest Uganda. Journal
of African Zoology 108:55—70.

Dusots, A. AND A. OHLER. 1999. Asian and Oriental toads of the Bufo melanostictus, Bufo scaber and Bufo
stejnegeri groups (Amphibia, Anura): a list of available and valid names and redescription of some name
bearing types. Journal of South Asian Natural History 4:133-180.

Dutta, S.K. 1997. Amphibians of India and Sri Lanka (Checklist and Bibliography). Odyssey Publishing
House, Orissa, India. 348 pp.

Dutta, S.K. AND K. MANAMENDRA-ARACHCHI. 1996. The Amphibian Fauna of Sri Lanka. Wildlife Heritage
Trust of Sri Lanka, Colombo, Sri Lanka. 230 pp.

ENGINEERING DESIGN. 1999. SIGNAL 3.1/RTSD 1.1. Engineering Design Belmont, MA.

Frost, D.H. 2002. Amphibian Species of the World. An online reference V2.21 (July 15,2002) Electronic data-
base available at http://research.amnh.org/herpetology/amphibian/index.html

GROSIEAN, S. AND A. Dusols. 2001. Description of advertisement calls of five Bufo species (Bufonidae) from
South and South-east Asia. Hamadryad 26:235-—246.

HAYEK, L.-A.C., W.R. HEYER, AND C. Gascon. 2001. Frog morphometrics, a cautionary tale. Alytes 18:
153-177.

INGER, R.F. 1966. The systematics and zoogeography of the Amphibia of Borneo. Fieldiana: Zoology 52:1—
401.

INGER, R.F. 1972. Bufo of Eurasia.Pages 102-118 in W-F. Blair, ed., Evolution of the genus Bufo. University
of Texas Press, Austin, Texas.

WOGAN ET AL.: ANEW BUFO FROM MYANMAR 153

INGER, R.F., H.B. SHAFFER, M. KOSHY, AND R. BAKDE. 1984. A report on a collection of amphibians and rep-
tiles from the Ponmudi, Kerala, South India. Journal of the Bombay Natural History Society 81:406-427.

INGER, R.F. AND R.B. STUEBING. 1997. Field Guide to the Frogs of Borneo. Natural History Publications. Kota
Kinabalu, Borneo. 1x + 205 pp.

ISKANDAR, D.T. 1998. The Amphibians of Java and Bali. Research and Development Centre for Biology-LIPI,
Indonesia. xix + 117 pp.

Leviton, A.E., R.H. Gipss, JR., E-HEAL, AND C.E. DAwson. 1985. Standards in herpetology and ichthyology:
part 1. Standard symbolic codes for institutional resource collections in herpetology and ichthyology.
Copeia 1985:802-832.

Liu, W., A. LATHRop, J. FU, D.YANG, AND R.W. Murpuy. 2000. Phylogeny of East Asian bufonids inferred
from mitochondrial DNA sequences (Anura: Amphibia). Molecular Phylogenetics and Evolution 14:
423-435.

Macey, J.R., J.A. SCHULTE, A. LARSON, Z. FANG, Y. WANG, B.S. TUNIYEV AND T.J. PAPENFuUsS. 1998.
Phylogenetic relationships of toads in the Bufo bufo species group from the eastern escarpment of the
Tibetan plateau: a case of vicariance and dispersal. Molecular Phylogenetics and Evolution 9:80-87.

MANAMENDRA-ARACHCHI, K., AND R. PETHIYAGODA. 1998. A synopsis of the Sri Lankan Bufonidae
(Amphibia: Anura), with description of two new species. Journal of South Asian Natural History 3(1):
213-246.

Myers, C.W. AND W.E. DUELLMAN. 1982. A new species of Hyla from Cerro Colorado, and other tree frog
records and geographical notes from Western Panama. American Museum Novitates (2752):1—25.

SavaGE, J.M. 1966. An extraordinary new toad (Bufo) from Costa Rica. Revista de Biologia Tropical 14:
153-167.

SAVAGE, J.M. AND M. DONNELLY. 1992. The second collection of, and variation in the rare Neotropical toad
Bufo peripates. Journal of Herpetology 26:72-74.

SmitH, M.A. 1917. On tadpoles from Siam. Journal of the Natural History Society of Siam 2:261—275.

SmitH, M.A. 1929. On a collection of amphibians and reptiles from the upper reaches of the Brahmaputra.
Records of the Indian Museum 31:77-80.

SmitH, M.A. 1940. The amphibians and reptiles obtained by Mr. Ronald Kaulback in upper Burma. Records
of the Indian Museum 42:465-486.

SPSS Inc. 1999. SPSS Base 10.0. Chicago IL.

TAYLor, E.H.1962. The amphibian fauna of Thailand. University of Kansas Science Bulletin 43:265-599.

VAN KAMPEN, P.N. 1923. The Amphibia of the Indo-Australian Archipelago. E.J. Brill, Ltd., Leiden, Germany
xii + 304 pp.

VILLA, J. 1972. Anfibios de Nicaragua. Instituto Geografico Nacional & Banco Central de Nicaragua.
Nicaragua. 216 pp.

Copyright © 2003 by the California Academy of Sciences
San Francisco, California, U.S.A.

154 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 7

PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES

Volume 54, No. 8, pp. 155-160, 4 figs., 1 table March 14, 2003

Cottoclinus canops, a New Genus and Species of Blenny
(Perciformes: Labrisomidae) from the Galapagos Islands

John E. McCosker', John S. Stephens’, and Richard H. Rosenblatt*
1 California Academy of Sciences, San Francisco, California 94118; 2 Research Associate,
California Academy of Sciences; 3 Scripps Institution of Oceanography, La Jolla, California 92093

We describe Cottoclinus new genus, type species C. canops novum, from two speci-
mens from Espanola Island, Galapagos Islands. The new genus is assigned to the
labrisomid tribe Mnierpini Hubbs. The tribe is redefined and considered sister to the
Labrisomini. It is united by nine synapomorphies: 1) body elongate to moderately
elongate (depth 13%-22% of standard length); 2) spinous dorsal fin with no incision
in outline, anterior rays soft and flexible; 3) anal fin rays all thickened, and fleshy at
tips; 4) head blunt with upper jaw projecting; 5) lower lip interrupted by a pair of
grooves outlining a frenum on chin; 6) lips thick and inflated; 7) scales expanded
posteriorly forming pockets; 8) membrane of anal fin rays incised almost to base of
rays; 9) belly and pectoral base naked. The two currently recognized monotypic gen-
era in the tribe, Dialommus Gilbert 1891 and Mnierpes Jordan and Evermann 1896,
are considered synonyms. The name Mnierpini is retained for the tribe. Cottoclinus
differs from Dialommus in that there is no fleshy pigmented vertical bar across the
eye separating two flat windows, and the posterior mandibular teeth are not abrupt-
ly smaller and set lower than the anterior teeth. C. canops is considered the basal
member of the clade leading to D. fuscus and D. macrocephalus.

In 1964 Boyd W. Walker and Edmund S. Hobson of the University of California, Los Angeles,
participated in a University of California expedition to the Galapagos Islands aboard the M/V
Golden Bear, operated by the California Maritime Academy. They were able to make large collec-
tions of fishes at a number of the islands. Included in the material that they returned with were two
specimens of a labrisomid blenny from a tidepool on Espanola (Hood Island), the southernmost
Galapagos Island. Until recently, these two undistinguished-looking specimens had languished on
a Shelf with other unidentified material. Eventually they were brought to the attention of one of us
(JSS). In concert with the other two authors of this paper, it was determined that they represented
an undescribed genus and species. It is the purpose of this paper to describe the genus and species
and to present a hypothesis of relationships.

MATERIALS AND METHODS

Type specimens of the new species are deposited in the Department of Ichthyology of the
California Academy of Sciences (CAS), San Francisco. Measurements are straight-line (point to
point) and made with dial calipers and recorded to the nearest 0.1 mm. Length of specimens is
given as standard length (SL), the distance from the front of the upper lip to the base of the caudal
fin. Body depth is measured vertically from the origin of the anal fin; body width is taken just pos-
terior to the gill opening. Head length (HL) is measured from the front of the upper lip to the pos-

* The order of authorship was randomly determined.

IS)5)

156 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 8

terior end of the opercular membrane, and snout length is from the upper lip to the fleshy edge of
the orbit. Orbit diameter is the greatest fleshy diameter, and interorbital width is the least fleshy
width.

Cottoclinus McCosker, Stephens, and Rosenblatt, new genus
Type species Cottoclinus canops, new species.

DIAGNOsIS.— That of the single included species.
ETYMOLOGY.— From the generic names Coftus and Clinus, in reference to the cottid-like
appearance of this blenny. Gender masculine.

Cottoclinus canops McCosker, Stephens, and Rosenblatt, new species
Figs. 1-4, Table |

MATERIAL EXAMINED.— Holotype: CAS 217107 (formerly W64—35), an immature male, 45.2
mm SL, from a low tidepool on the south side of Punta Suarez, Isla Espanola (Hood), Galapagos
Islands, collected on 18 Feb. 1964 by M. Castro. Paratype: CAS 217108, an immature female, 54.4
mm SL, collected with the holotype.

DIAGNOosIs.— A moderate-sized mnierpin labrisomid blenny with the following characteris-
tics: body moderately elongate (depth 4.4—5.1 in SL); head bulbous, slightly depressed; lips fleshy;
posterodorsal corner of maxillary slips into a cheek pouch; nasal, orbital, and nuchal cirri small and
fimbriate; breast and belly scaleless; scales membranously expanded posteriorly to form pockets;
spinous dorsal fin rays nearly level, lower and less numerous than dorsal fin soft rays; one anal fin
spine; interradial membrane of anal fin incised almost to base of each ray; all fin-rays unbranched;
cornea convex, eyeball lacking a vertical bar and heavy pigmentation; teeth conical, stout, slightly
recurved; medial teeth of dentary notably larger than lateral teeth; vomerine teeth present, palatine
teeth absent.

DESCRIPTION.— Meristic values listed are those of the holotype, those for the paratype are in
parentheses. (We have selected the larger of the two specimens to be the paratype in that it is bent
and in poorer condition.) Counts: D XXI, 10 (XX, 10); AT, 19; Pi xiii; P2 I, 3; C xiii; branchioste-
gal rays 6, pored lateral line scales 47; vertebrae 10 + 25. Body depth about 4.5—5.1 in length. Head
roughly quadrangular when viewed from front; its depth about 90% of width. Eyes set high, enter-
ing dorsal profile, cornea not divided. Lips fleshy, lower lip with a free ventral flap attached to a
medial frenum. Mouth short; maxilla, which slips into cheek pouch, ends at level of posterior mar-
gin of pupil.

Premaxilla with a small, conical symphyseal tooth, followed by an outer row of 10 conical
teeth that decrease in size posteriorly. A few vomerine teeth, palatine teeth absent. An interior patch
of setiform teeth at head of premaxillary. Dentary of holotype with a pair of enlarged incisors at

FiGuRE 1. Lateral and dorsal views of holotype of Cottoclinus canops, CAS 217107, 45.2 mm SL.

McCOSKER ET AL.: NEW GENUS AND SPECIES OF GALAPAGOS BLENNY js)

symphysis, followed by 5 conical TABLE 1. Measurements in millimeters of the holotype and

recurved teeth, increasing in size posteri- paratype of Cortoclinus canops, new species.
orly; sixth tooth about 1/3 length of fifth, Holotype Parahipe
followed by a row of small, increasingly — grandard length 45.2 554
flattened teeth. A small medial patch Of esa jenieth 15.0 16.8
villiform teeth near symphysis. Dentition efanal cicttn ne 1AA
of paratype somewhat different: symphy-
seal teeth not so conspicuously enlarged; ca ae te oe
second through fifth teeth about 3/4 as Gro oe ae
long as symphyseals, then teeth 6 and 7 Preanal distance Zehi) oe:
are conspicuously enlarged, about twice Longest pectoral ray 13.2 14.3
as long as symphyseal teeth, with teeth 8 Body depth at anal origin 8.8 12.2
through 11 similar in size to teeth 2 to5. — Caudal peduncle depth 4.3 4.9
Posterior nostril with a raised rim, Longest dorsal fin spine 5.1 —
anterior nostril with a short tube sur- — [ongest caudal ray 98 11.7

mounted by a palmate tentacle with about
7 fimbriae. A small fimbriated supraorbital cirrus dorsally on the eyeball, and a pair of fimbriated
nuchal cirri arising from a common base on the nape on either side of the first dorsal fin spine.

Dorsal fin spines 1—20 subequal, but spines 18—20 decrease in length, resulting in a distinct
notch between the spinous dorsal fin and the soft dorsal fin. Margin of soft dorsal fin rounded, 2nd
through 5th rays longest. Last dorsal fin ray free from caudal peduncle. Anal fin spine 1/3 (1/2) as
long as first soft-ray. Anal fin soft-rays thickened, their membranes incised almost to the base,
except for the last 6 rays. Pectoral fan-shaped, rays 7—9 longest; rays 9-13 thickened. Pelvic spine
splint-like, soft-rays one and two stout, soft ray 3 slender, but almost as long as second.

Body scaled except for breast, belly, an area anterior to pectoral fin, and an area at nape
extending back along first 4 or 5 dorsal fin spines. Scales with posterior membranous caudal expan-
sions, forming pockets, particularly on lower sides. Lateral-line scales pored, the pore openings
being located along the mid-dorsal scale margins; posteriorly, pores become obsolete.

Cephalic sensory pores (Fig. 2) mostly minute, numerous, many appearing as skin pricks. The
following is based on the condition of the holotype. Mandibular (M) series with six pores on each
side, the first along the symphysis, the following four evenly spaced, the last more separated. Six
preopercular (PO) pores closely following the posteri-
ormost mandibular pore. A series of four pores above
maxilla, followed by an upward-arching row of approx-
imately eight small pores. Two pores above upper lip in
advance of the anterior nostril, followed by three pores
(the middle pore lacking on left side of holotype above
the anterior frontal region). A single median interorbital
or commissural (C) pore is followed by a pore pair, then
a single median supratemporal (MST) pore that is
closely followed by another pore pair. The orbits are
surrounded by numerous minute infraorbital (IO), pos-
torbital (PO), and supraorbital (SO) pores. Additional
rows of minute pores are located between the POP and
PO regions. There are few preorbital pores, although

FIGURE 2. Lateral (left), dorsal (right) and ven-
tral (below) views of cephalic sensory pores of : :
holotype of Cottoclinus canops, CAS 217107, two nasal pores (N) lie between, and just medial to, the

45.2 mm SL. anterior and posterior nostrils. A row of minute pores

158 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 8

passes from the mid-PO region to form a “T”-shaped juncture alongside the ST pores. Several
minute pores lie distal to the nuchal cirri, which are followed by a large lateral supratemporal (ST)
pore.

Body coloration in isopropyl alcohol tan, with a suggestion of 3 or 4 diffuse bars. Head of
holotype profusely spotted with brown flecks, these less conspicuous in the paratype. Upper lip
with a brown mark at the corner of mouth and another near midline, the latter one continued on
lower lip. Middle and lower pectoral rays brown. Caudal fin of holotype with the suggestion of 3
vague bars. Eyeball pigmented dorsally, but no suggestion of.a pigmented vertical bar across eye.

ETYMOLOGY.— From the Greek kanon, rule or standard, and ops, eye, in reference to the nor-
mal structure of the cornea.

RELATIONSHIPS.— Hubbs (1952) established the tribe Mnierpida (sic) within his subfamily
Labrisominae (now Labrisomidae) for the reception of the genera Mnierpes and Dialommus. The
tribe was defined by: 1) “a fleshy bar runs vertically across pupil of eye; 2) body elongate (depth
13%-17% of standard length); 3) spinous dorsal fin with anterior rays soft and flexible, with no
incision in outline; 4) anal fin rays all enlarged and fleshy at tips; 5) head blunt with upper jaw pro-
jecting.” To these we can add: 6) posterior dentary teeth abruptly smaller and set lower than the
anterior teeth; 7) lower lip interrupted by a pair of grooves outlining a frenum on chin; 8) lips thick
and inflated; 9) scales expanded posteriorly forming pockets; 10) membrane of anal fin rays incised
almost to base of rays; and 11) belly and pectoral base naked. Although no cladistic analysis of rela-
tionships of all labrisomid genera has been attempted, the classification of Stepien et al. (1993,
1997) based on molecular data indicates that the tribe Labrisomini (genera Malacoctenus and
Labrisomus) 1s sister to the tribe Mnierpini (sensu Hubbs). The above 11 character states are
synapomorphies of the Mnierpini with respect to the Labrisomini.

Cottoclinus shares all of these synapomorphies except | and 6. (Although C. canops is not as
elongate as D.fuscus and D. macrocephalus — body depth 19.5% and 22% in holotype and
paratype respectively —it is still relatively elongate in comparison to other labrisomids.)
Relationships are best indicated by placing Cottoclinus as the basal member of the clade leading to
Mnierpes and Dialommus (Fig 3). Those genera were separated by Hubbs on the basis of number
of anal fin spines (1 versus 2), presence or absence of a supraorbital, and meristic and proportion-

Di niommus onaGsocanueTan Uotad eae tracts al characters more suitable for the distinction of

wa species than genera. Dialommus then would be

3 Pe characterized by the presence of | anal fin spine

BRB PE SEEEER: « cia and the lack of an orbital cirrus (an autapomorphy),
Pie ‘ phen a and Mnierpes by the presence of a supraorbital cir-
rus and 2 anal fin spines. One anal fin spine and no

K we a orbital cirri are each apomorphic features within

the Labrisomidae. However, Cottoclinus shares

FIGURE 3. Cladogram of hypothesized relationships One apomorphic and one plesiomorphic character
within the Mnierpini. Characters are: 1) orbital cirri state with each.
present; 2) 1 anal fin spine; 3) 2 anal fin spines; 4) body Inasmuch as Cottoclinus has 1 anal fin spine,
elongate (depth 19-22% of SL); 5) anterior spinous dor- a . ; . :
sal fin rays flexible, without incision; 6) anal fin rays and if the phylogeny diagrammed in Figure 3 is
enlarged, tips fleshy; 7) head blunt, upper jaw project- Valid, two hypotheses can be erected. Either the
ing; 8) posterior dentary teeth smaller and lower than common ancestor of Dialommus and Mnierpes had
anterior; 9) chin frenum present; 10) lips thick and one anal fin spine and an anal fin spine was gained
inflated; 11) posterior scale pockets; 12) membrane of :
anal rays incised; 13) belly and pectoral base naked; 14) in Mnierpes as a reversal, or the common ences
body very elongate (depth 13-17% of SL); 15) a verti- had two anal fin spines and an anal fin spine was
cal fleshy bar across pupil: 16) orbital cirri absent. lost independently in Cottoclinus and Dialommus.

McCOSKER ET AL.: NEW GENUS AND SPECIES OF GALAPAGOS BLENNY 159

In either case, the synapomorphies of the com-
plex corneal modifications (Graham and
Rosenblatt 1970) and the condition of the
mandibular dentition would have been present
in the common ancestor of Dialommus and
Mnierpes. The anal fin spine-pterygiophore
relationship in Mnierpes is essentially identi-
cal to the plesiomorophic condition in
Labrisomus and Malacoctenus, and both
Cottoclinus and Dialommus have lost the first
spine but retained the first pterygiophore (Fig.
4). (The phylogeny as diagramed in figure 3
reflects the independent loss of an anal fin
spine.) Therefore, we cannot reject either
hypothesis on the basis of available evidence.

The common possession of the dentition
and corneal specializations in the monotypic
genera Dialommus and Mnierps indicates
common ancestry and the species D. fuscus
and M. macrocephalus should therefore be
regarded as congeneric. Dialommus Gilbert Ficure 4. Lateral view of radiographs of anterior anal fin
1891 (type species D. fuscus Gilbert 1891) is and supporting elements. (Images have been slightly enlarged
senior to Mnierpes Jordan and Evermann 1896 __ or reduced to facilitate comparison.) Upper left, Cottoclinus
(type species Clinus macrocephalus Giinther 425, holotype, CAS 217107, 45.2 mm SL. Upper right,

ASG deNing ie ARIUS SVOAVGIIOE Dialommus fuscus, CAS 23754, 68 mm SL. Lower left,
), Lenpes: aout y ye ©) Labrisomus dendriticus, CAS 46577, 88.5 mm SL. Lower

Dialommus. right, Dialommus macrocephalus, CAS 213855, 63 mm SL.
REMARKS.— The membranous posterior

prolongation of the scales to form pockets and the separation and thickening of the anal fin rays
have been hypothesized as adaptations to amphibious life; the former would hold water, and the
latter would aid in clinging to sloping surfaces and in terrestrial locomotion (Clark 1932; Graham
1970). These modifications are already present in Coftoclinus, indicating that the ancestor of the
fuscus-macrocephalus clade must have attained considerable terrestriality even before the modifi-
cations of the eye had developed.

In January 2002, JMc and JSS returned to Espanola and collected in tidepools contiguous with
the type locality of C. canops. They were unable to access the pool sampled in 1962 but were able
to collect from two pools within a km of that site. In general, the pools were devoid of large fish.
The following species were collected: one each of Muraena lentiginosa, Echidna nocturna and
Uropterygius macrocephalus; several Scorpaena mystes, Acanthemblemaria castroi, Starksia gala-
pagensis and Ogilbia deroyi; numerous Dialommus fuscus (in the 1962 collection, 24 D. fuscus
were collected from the same pool as the two C. canops) and Malacoctenus es eae and hun-
dreds of small Tomicodon chilensis.

Recent El Nifio events (1982-1983 and 1997-1998) have had strong effects on tidepool and
shallow subtidal species at the Galapagos. For example, the abundant shallow-water chaenopsid,
Acanthemblemaria castroi, had largely disappeared following the most recent El Nino (JMc and C.
C. Baldwin, personal observations), but it is once again abundant, and the conspicuous midwater
pomacentrid, Azurina eupalama, has not been seen at Galapagos since 1977. We consider the pos-
sibility that Cottoclinus canops might be a naturally-occurring casualty of a recent El Nino event.

160 PROCEEDINGS OF THE CALIFORNIA ACADEMY OF SCIENCES
Volume 54, No. 8

The known distribution of Dialommus macrocephalus is from Cape San Lucas to the Bay of
Panama and D. fuscus is widely distributed within the Galapagos archipelago and Cocos Island
(Costa Rica). Cottoclinus canops is known only from the Galapagos. Based on the basal position
of C. canops, it could be argued that the tribe evolved on the Galapagos and that the progenitor of
D. macrocephalus invaded the mainland from there or from Cocos Island.

The tribal name Mnierpini, based on a generic junior synonym, 1s retained in accordance with
Article 40.1 of the Code of Zoological Nomenclature: ““When the name of a type genus of a nom-
inal family-group taxon is considered to be a junior synonym of the name of another nominal
genus, the family-group name is not to be replaced on that account alone” (International
Commission on Zoological Nomenclature 1999).

ACKNOWLEDGMENTS

We thank: Miguel Castro, then of the Charles Darwin Foundation, who captured the speci-
mens; Dan Pondella, who brought the specimens to our attention; Philip Hastings and Tomio
Iwamoto, for reading this manuscript; Dong Lin and Tinya Hoang for assistance with photography
and figure preparation; Eliecer Cruz and Mario Piu of the Galapagos National Park for permission
to perform research in the GNP; Robert Bensted-Smith and the CDRS for general assistance in the
Galapagos and for the use of R/V Beagle; and the Fundacion El Viejo and the Griffith Family
Foundation for financial assistance.

LITERATURE CITED

Clark, H.G.W. 1932. The Templeton Crocker Expedition of theCalifornia Academy of Sciences, 1932. No. 29.
New and noteworthy fishes. Proceedings of the California Academy of Sciences 21(29):383-396.

Gilbert, C.H. 1891. Scientific results of explorations by the U. S. Fish Commission steamer “Albatross”. No.
19. A supplementary list of fishes collected at the Galapagos Islands and Panama, with descriptions of a
new genus and three new species. Proceedings of the United States National Museum 13:449-455.

Graham, J.B. 1970. Preliminary studies on the biology of the amphibious clinid Mnierpes macrocephalus.
Marine Biology 5(2):136—-140.

Graham, J.B., and R.H. Rosenblatt. 1970. Aerial vision: Unique adaptation in an intertidal fish. Science
169:586-S88.

Giinther, A.C.L.G. 1861. Catalogue of the Acanthopterygian Fishes in the Collection of the British Museum,
vol. 3. Trustees of the British Museum, London, UK. xxvi + 586 pp.

Hubbs, C.L. 1952. A contribution to the classification of theblennioid fish family Clinidae, with a partial revi-
sion of the eastern Pacific forms. Stanford Ichthyological Bulletin 4(2):41-165.

International Commission on Zoological Nomenclature. 1999. International Trust for Zoological Nomencla-
ture, London. 306 pp.

Jordan, D.S., and B.W. Evermann. 1896. A check-list of the fishes and fish-like vertebrates of North and
Middle America. Report of the U.S. Commission on Fisheries 21[{1895], Appendix 5:207—584.

Stepien, C.A, M.T. Dixon and D.M. Hillis. 1993. Evolutionaryrelationships of the blennioid fish families
Labrisomidae, Clinidae, and Chaenopsidae: Congruence among DNA sequence and allozyme data.
Symposium on Evolution of Percomorph Fishes, American Society of Ichthyologists and Herpetologists.
Bulletin of Marine Sciences 52:496-515.

Stepien, C.A, A.K. Dillon, M.L. Brooks, K.L. Chase and A.N. Hubers. 1997. The evolution of blennioid fish-
es based on an analysis of mitochondrial 12S rDNA sequence data. Pages 245-270 in T.D. Kocher and
C.A. Stepien, eds., Molecular Systematics of Fishes. Academic Press, San Diego.

Copyright © 2003 by the California Academy of Sciences
San Francisco, California, U.S.A.

INSTRUCTIONS TO AUTHORS

Authors expecting to submit papers for consideration for publication in the Academy’s Proceedings,
Occasional Papers, or Memoir series should follow the directions given below in preparing their submissions.
Under some circumstances, authors may not be able to comply with all the computer-based requirements for sub-
mission. Should this be the case, the author is advised to contact the Editor or Associate Editor for guidance on how
best to present their materials.

The Scientific Publications Office of the Academy prepares all materials for publication using state-of-the-art,
computer-assisted, page-description-language software. Final copy is sent to the printer for printing. The printer
does not modify the files sent for printing. Therefore, it falls to the authors to check carefully page proof when it is
returned for approval. Ordinarily, all communication with authors is done via email and galley and page proofs of
manuscripts, including figures, are transmitted as attachments to email communications. Again, exceptions to this
will be made in the event that an author is unable to communicate in this way.

Authors will be expected to provide digital copies of both manuscript text files and images, as well as a paper
printout of their manuscript. Please note the following:

TEXT: Text can be in Microsoft Word, as a Word document file, WordPerfect, also as a WP document file, or,
best of all, as a “rtf” (rich text format) file, which can be produced by most wordprocessors. Authors who use non-
standard fonts must include file copies of those fonts so that their symbols can be reproduced accurately.

IMAGEs: Images, both color and grayscale, should be in either JPG (JPEG), or TIF (TIFF) format. Resolution
for grayscale images should be at least 600 ppi (1200 ppi if possible, especially for photomicrographs), and 300 ppi
for color. All images should be sized so that none exceeds a maximum print size of 5.5” x 7.875” (140 mm x
200 mm).

TABLES: Our processing software allows for direct importation of tables. This reduces the chances for errors
being introduced during the preparation of manuscripts for publication. However, in order to use this feature, tables
must be prepared in Microsoft Excel or in Microsoft Word using Word’s table feature; do not prepare tables using
tabs or space bars. Complex tables not prepared as described above will be returned to the author for revision.

DiciTAL FiLes: IBM or MAC formatted disks will be accepted subject to the following conditions: (a) floppy
disks must not exceed 1.4 mb and (b) zip disks, preferably IBM format, must not exceed 100mb. Authors are
encouraged to submit their digital files on CD-ROM (CD-R formatted disks NOT CD-RW) inasmuch as these can
be read by nearly all CD-ROM drives.

BIBLIOGRAPHY FORMAT: Three bibliographic styles are accommodated in the Academy’s scientific publications,
one commonly used in scientific journals publishing papers in systematic and evolutionary biology, a second used
mainly in the geological literature, and lastly, the format most commonly used in the humanities by historians of
science. On request, the author will be sent a style sheet that includes samples of the three formats. Authors are also
encouraged to examine a copy of the lastest published Proceedings. In all instances, however, authors should not
abbreviate journal names but spell them out completely. For books, the reference must include the publisher and city
of publication. It is recommended that the total number of pages in the book also be given.

SUBSCRIPTIONS

The Proceedings series of the California Academy of Sciences is available by exchange or subscription. For
information on exchanges, please contact the Academy’s Librarian via regular mail addressed to the Librarian,
California Academy of Sciences, Golden Gate Park, San Francisco, CA 94118, U.S.A. or via email addressed to
amalley@mail.calacademy.org. Subscription requests, including information on rates, should be addressed to
Scientific Publications, California Academy of Sciences, Golden Gate Park, San Francisco, CA 94118, U.S.A. or
via email to the Editor at scipubs@mail.calacademy.org.

COMMENTS

Send editorial correspondence to the Editor, Scientific Publications Office, California Academy of Sciences,
Golden Gate Park, San Francisco, CA 94118, U.S.A. or via email to the Editor at scipubs@mail.calacademy.org.

LIB

Wi
39088 013027404

TABLE OF CONTENTS

VINDUM, JENS V., HTUN WIN, THIN THIN, Ky1 SOE LWInN, AWAN KHwWI SHEIN, HLA TUN:
A new Calotes (Squamata: Agamidae) from the Indo-Burman Range of western Myanmar ....

WILKINSON, JEFFERY A., HTUN WIN, THIN THIN, Kyl SOE Lwin, AWAN KHWI SHEIN, HLA TUN:
A new species of Chirixalus (Anura: Rhacophoridae) from western Myanmar (Burma)

GLADENKOV, ANDREY YU.: Diatom biostratigraphy of the Neogene Milky River Formation,
Alaska Peninsula, southwestern Alaska

CAMACHO-GARCIiA, YOLANDA E. AND ANGEL VALDES: Caryophyllidia-bearing dorid
nudibranchs (Mollusca: Nudibranchia: Doridina) from Costa Rica

TYSON R. ROBERTS: Systematics and osteology of Leptoglaninae, a new subfamily of the
African catfish family Amphiliidae, with descriptions of three new genera and six new
species

YIN CHANG-MIN, CHARLES E. GRISWOLD, BAO YOu-HUI, AND XU XIANG: Four species of the
spider genus Steatoda (Araneae: Theridiidae) from Gaoligong Mountains, Yunnan, China... 133

GUINEVERE O. U. WOGAN, HTUN WIN, THIN THIN, Ky1 SOE LWIN, AWAN KHwWI SHEIN, SAI
WUNNA Ky1, HLA TUN: A new species of Bufo (Anura: Bufonidae) from Myanmar, and
redescription of the little-known species Bufo stuarti Smith 1929....................0- 141

JOHN E. MCCOSKER, JOHN S. STEPHENS, JR., RICHARD H. ROSENBLATT: Cottoclinus canops, a
new genus and species of blenny (Perciformes: Labrisomidae) from the Galapagos Islands . . 155

CALIFORNIA
ACADEMY
SCIENCESS

ESTABLISHED 1853