JJ

“ VrV^4ii^7 - ^

m ^ 'ii xtfiiusgx m 3; m '<2L£i^' m

CO \ £ CO — co £ ^

yen libraries Smithsonian institution NouniiiSNi nvinoshiiws saiuvuan libraries smithsoni*

7 ,n 2 ,V . C/> Z CO 2

** ^s\<t 5 ^ .v4&'< s .< S&Z2&:

v\W> -. /Kfcftfc&c ^ ;/ ^

o ^

2 t \S3MAK§/ 2

> X'OmsJX' 2 '®c. > 5 > i|§>“ 2 "*: >

to * 2 j/) k- 2 to 2 a) ^2

JTION NQ!iniIiSNI_NVINOSHllWS S 3 I U V B 3 n_ L I B R AR 1 ES SMITHSONIAN _ INSTITUTION NQlIfUlISNI NVINOSHlll!

< ’>/

cc

^fc-jy “ _ ^,ll}^ Q

id S H ' L I B RAR 1 ES^SMITHSONIAN^iNSTITUTION NOlinillSNI "nVINOSHIIINS^SS I HVH 8 ll^ll B RAR I ES^ SMITHSONI;
~ 2 r- z r~ , 2 r- 2

m rn x:«>vy m * v?- -y m

CO £ CO £ CO £ _

JTION NOlinillSNI NVIN0SH1HNS S3IUVUan LIBRARIES SMITHSONIAN INSTITUTION NOliniliSNI NVINOSHlll

co 2 co . 2 , w 2 co 2

“ —

P

x

co

O

r

k s »% § '%% x 4;

i J CO CO v co S

Mj § e Xw 5 \r

an LIBRARIES SMITHSONIAN INSTITUTION NOIinillSNI_NVINOSHllWS S3IHVB8n LIBRARIES SMITHSONI/

-7 co _ co 2 \ 00 X CO

CO

O XXosv^X ~ O ^VQS>'_>^ — ^ Q

Z _) 2 _J 2 ' -1 2 _

ITION NOlinillSNI NVIN0SH1MS S3IBVH9n LIBRARIES SMITHSONIAN INSTITUTION NOlinillSNI NVINOSHlll
z r- 2 ~ ^ ~

CD
50
>

33

m NT x^uvx rn 3 -^5^ m "<12^ ^ m

co \ £ co — CO £ — CO

liail LIBRARIES SMITHSONIAN INSTITUTION NOlinillSNI NVINOSHIIWS SSlBVHail LIBRARIES SMITHSONI;

2 . CO 2 ^ co Z CO 2 -y, CO

< XX . 2 < .,XX> x. 2 .< S , <

| ^

> s - I I 5 'W i ' X i

CO * 2 00 * 2 CO 2 CO * 2

JTION NOlinillSNI NVIN0SH1IINS S3iaVd8n LIBRARIES SMITHSONIAN INSTITUTION NOlinillSNI NVINOSHlll

^ ~ , C/) ^ 2 - — ^ — -y

w .V^X ^ CO X^Jgx ^ . < co !±! /£^5Di\ co

^(5) < qjo) <

<ouus5X Q NS “ O XX pc^ _ o

_ 2 *' 1 Z _J Z -1 2 *n

nail LIBRARIES SMITHSONIAN INSTITUTION NOlinillSNI NVIN0SH1IWS S3iaVHail LIBRARIES SMITHSONI,
1- 2 r- 2 r* , 2 r~ 2

rn xnqs"J2 ^ m - (ti

JTI0N°° NOlinillSNI - NVINOSHillMS^ S 3 I aVH0ll — LIBRARIES SMITHSONIAn'iNSTITUTION^ NOlinillSNI NVINOSHlll

co 2 co 2 , co 2 » co 2

< 2 ^ S xS^ixvx < a 2 <

x

» o

^ I ■"

2 CO z CO ' Z 01 ••' Z CO

H8I1 LIBRARIES SMITHSONIAN INSTITUTION NOlinillSNI NVIN0SH1IIAIS S3IHVaail LIBRARIES SMITHSONI,

— CO — CO -F > co — CO

m X5^p/ « N^gg^ S X 5 m

05 ~ OS = — tfl X £ co

'UTION NOliniliSNI NVINOSHil WS S3IUVHfll1 LIBRARIES SMITHSONIAN INSTITUTION NOliniliSNI NVINOSHil
in 2 — — . ^ ^ 2 v m 2

y, s .< 1 , ^v<i S x^^rx < s < /<&£&;

\ i */J/B g l -trfjS/v *

\ r/5 :-v«:3X co isffc .^S^l to jo ■ to iof§f 3o! co W| N co to

o

s

> 'W4' S > W 2 ^ > NS^^X 2 > .

/y a nS LS B RAR I ES^SMITHSONIAN ' INSTITUTION NOliniliSNI NVINOSHIIWS^SH I U Wd 8 II2 L8 B RAR I ES^SMITHSONI
to s - to __ r= to ~ to

^■msax q ^ ^ Njtgniv^ o

UTION NOliniliSNI^NVINOSHltNS^a I d VH 8 11 “Yl B RAR I ES^SMITHSONIAN^INSTITUTION NOliniliSNI “NVINOSHil

2 f“ * 2 (r“ 2 P" 2 f“ ,

m ''v$§p x^as^x pi i" m x^-os^x ^ x^amx m

to = to — 2 to £ to

fa an libraries Smithsonian institution NoiiniiiSNi nvinoshiwis ssiavaan libraries smithson

•te, S w I /^v < a I ajA* < xf5^3x i S w 5 ^@22^

§ %% s B %% a 48% g JP%\ a £|<k § %% a « ,

I Veil ° llu.jy 2 /W § J^Sp £ it,.. Ji § ^Jr1 I v®« - '-

UTION^^OiinillSN! NVINOSHHIMS^SgiiJVaan^LIBRAR lES^SMITHSONIAN^INSTITUTION WWOIinillSNI NVINOSHil

(?/5> _ 'xr if) f* /?/«! — » If A — *•©

% _

M*WZ

/‘S /' h
o

2 =J ^

aan libraries Smithsonian jnstitution NoiiniiiSNi nvinoshhiais saiavaan libraries smithsoni

r* 2 P" 2 t“ _ 2 r- 2

' — ^ ~ “ 'v/ - - — - — s O

rn x^a) |jp m ^ x^oiEsgx pi "nr « x^uus^x m

UTION ^ N0linillSNI*_NVIN0SHilWS S3 I H WH 8 1 1 L I B R AR I E S^SMITHSON IAN ~ INSTITUTION °° NOliniliSNI NVINOSHil

CO 2 ^ w z .,, w 2 to 2

a ./.y^g a g a § « i ,*^g ^

\

w IsF 55 i»fr4®r w Wfe" tq (aK asi CO I'M' CO w

I i I t ^ ° lT_ ' 1 %*&y ° £" x 1

^ 5 xo^dc>x > > X^oins^y 5 '"X > 2

fd 8 11 2 L I B RAR I ES^SMITHSONIAN^ INSTITUTION NOliniliSNI NVINOSHil WS^S 3 I 8 V d 0 11 2 L I B R A R I ES^SMITHSONI

5 CO 5 co — co — CO

£ Xosoa^x uj £ UJ

UTION NOliniliSNI NVINOSHillNS S3iavaOI1 LIBRARIES SMITHSONIAN INSTITUTION NOliniliSNI NVINOSHil

Z r- , Z r- Z r-

O xcSTTiTx ^ to O

CO £ CO

faan libraries Smithsonian institution NouniiiSN

Z . CO z . CO

m x^jjos^x Xr x^jAsijx m

co ± co

NVINOSHillNS S3iaV8Bll LIBRARIES SMITHSONI

CO Z . CO

00 ^ S Vv 5 4% ft/'/ 00 IpE. ^5] CO /ft/ 4/5

1 w i '%# i 5# g IJ g ',w 5

5 ■ > X^VA^X 2 ^ > - g X^oct^/ > 5 • > x'A!,y/

UTION ^ NOliniliSNI _ NVINOSHil INS0’ S3 I HV8 8 ll^LI BRAR I ES^SMITHSONIAN INSTITUTION W NOliniliSNI _ NVINOSHil

CO -r \ CO — co — CO

CO

O

z

(PKlfllTI IOAC

o an ip*t’#tpb nvi a

Contents

CONTRIBUTIONS IN SCIENCE
1986

Number 372

Rodents, bats, and insectivores from the Plio-Pleistocene sediments to the east of Lake Turkana,
Kenya. 15 pages, 10 figures, 7 tables. 21 February 1986. ($3.50)

Craig C. Black and Leonard Krishtalka

Number 373

Three new luminescent ostracodes of the genus Vargula (Myodocopida, Cypridinidae) from the
San Bias region of Panama. 23 pages, 8 figures. 21 February 1986. ($4.50)

Anne C. Cohen and James G. Morin

Number 374

Late Miocene and Holocene mammals, exclusive of the Notoungulata, of the Rio Acre region,
western Amazonia. 46 pages, 32 figures, 7 tables. 15 May 1986. ($5.50)

Carl D. Frailey

Number 375

The subfamilies of Eurytomidae and systematics of the subfamily Heimbrinae (Hymenoptera:
Chalcidoidea). 17 pages, 25 figures, 2 maps, 1 table. 15 May 1986. ($4.50)

Gerald I. Stage and Roy R. Snelling

Number 376

Contributions toward a revision of the New World nomadine bees. A partitioning of the genus
Nomada (Hymenoptera: Anthophoridae). 32 pages, 66 figures, 2 tables. 15 May 1986.
($5.00)

Roy R. Snelling

Number 377

Recent discoveries in the Blepharicera tenuipes group, including descriptions of two new species
from Appalachia (Diptera: Blephariceridae). 20 pages, 26 figures.

Charles L. Hogue and Ted Georgian

Description of a new species of the shore fly genus Diedrops (Diptera: Ephydridae) from
Colombia. 6 pages, 10 figures.

Wayne N. Mathis and Charles L. Hogue

3 September 1986. ($5.00)

Number 378

The taxonomy and nomenclature of some Australian paragiine wasps (Hymenoptera: Masaridae).
19 pages, 31 figures. 3 September 1986. ($4.00)

Roy R. Snelling

Number 379

Propodial elaboration in southern African and Indian Ocean Fissurellidae (Mollusca: Proso-
branchia) with descriptions of two new genera and one new species. 12 pages, 26 figures.

3 September 1986. ($3.00)

James H. McLean and R. N. Kilburn

NATURAL HISTORY MUSEUM OF LOS ANGELES COUNTY
900 Exposition Boulevard, Los Angeles, California 90007

Author Index

Black, Craig C.
Cohen, Anne C.
Frailey, Carl D.
Georgian, Ted
Hogue, Charles L.
Kilburn, R. N.
Krishtalka, Leonard
Mathis, Wayne N.
McLean, James H.
Morin, James G.
Snelling, Roy R.
Stage, Gerald I.

Number 372
Number 373
Number 374
Number 377
Number 377
Number 379
Number 372
Number 377
Number 379
Number 373
Numbers 375,376,378
Number 375

The scientific publications of the Natural History Museum o-f L-os Angeles County have been
issued 'at. irregular intervals in three major series: the articles in each series are numbered
individually, and numbers, run consecutively, regardless of the subject matter.

© Contributions- id Science! a •miscellaneous series of technical papers describing orig-
inal research in the life and earth sciences. .

» Science, Bulletin, a raisjsellaoeous series of njiancigiraphs describing original research
in the life and earth sciences. This series was discontinued in 1978 with the issue of
Numbers 29 and 30; monographs are now* published by Ihe Museum in Contributions
in Science.

# Science Series, long articles on natural history topics, generally written for the layman.

Copies .of the publications in these series are sold through the Museum Book Shop. A catalog

is. available o;n request. : i . ; j

JSCIEN I1FIC Pti B Lit: AT IONS COM MITTEE

Craig C. Black, Museum Director
'Richajrd C. Brusca

Daniel M. Cohen, Committee Chairman
;Joh.r< M. Harris,

Charles L. Hogue

George L. Kennedy

Robin A. Simpson, Managing Editor
Errol Stevens 1

ress, Inc., Lawrence. Kansas

RODENTS, BATS, AND INSECTIVORES FROM THE
PLIO-PLEISTOCENE SEDIMENTS TO THE
EAST OF LAKE TURKANA, KENYA

Craig C. Black1 and Leonard Krishtalka2

ABSTRACT. Fossil small mammals collected from five Plio-Pleis-
tocene localities in the Koobi Fora Formation along the east side of
Lake Turkana, Kenya, are described. The assemblage bears close
resemblance to those from Olduvai Bed I in Tanzania and Omo
Members F and G in Ethiopia. The most common elements of the
assemblage are the murids Thallomys quadrilobatus and Praomys
( Mastomys ) minor. The micromammals from one locality, which
dates at approximately 1.6 million years, suggest an arid environ-
ment with Acacia scrub and some riverine forest along intermittent
stream channels.

INTRODUCTION

Paleontological investigations have been carried out along
the eastern side of Lake Turkana, under the auspices of the
Koobi Fora Research Project, National Museums of Kenya,
since 1968. Results of these field studies have appeared in
many journals and are summarized by Coppens et al. (1976)
and Leakey and Leakey (1978). While various studies of the
geology, paleoanthropology, and paleontology have been
pursued by a number of investigators, there have been no
previous studies of microvertebrates from these sediments.
Such studies have been undertaken for a sequence of deposits
of similar age along the Omo River in Ethiopia (Jaeger and
Wesselman, 1976; Wesselman, 1984), for at least Bed I at
Olduvai Gorge in Tanzania (Butler, 1 969; Butler and Green-
wood, 1973, 1976, 1979; Jaeger, 1976), and for a portion of
the Hadar Formation in Ethiopia (Sabatier, 1982).

In order to broaden paleontological knowledge of the East
Turkana faunas to include the smaller mammals, we began
a series of washing and screening operations in the summer
of 1974. A primary goal of the project was to identify the
ecological conditions prevailing during the early evolution
of hominids between 3 and 1 million years ago, a period
informally referred to as the Plio-Pleistocene (Harris, 1983:
3), in the Lake Turkana (formerly Lake Rudolf) basin of
northern Kenya. An understanding of past environmental

Contributions in Science, Number 372, pp. 1-15
Natural History Museum of Los Angeles County, 1986

conditions at Turkana would assist in correlating the East
African Rift Valley Plio-Pleistocene localities.

Small vertebrates such as rodents, insectivores, snakes, and
amphibians are often quite sensitive to local environmental
fluctuations. Their occurrence can be used to interpret local
climatic conditions in both past and present ecosystems. A
history of climatic and environmental change can be recon-
structed by studying the species composition of microver-
tebrate assemblages at each of many fossil localities and com-
paring these groups of species with Recent assemblages of
related species and their environmental requirements. The
degree of reliability of such analogies is dependent upon re-
covery of large samples from temporally restricted horizons
and upon the degree of relationship between the fossil and
Recent taxa. The more diverse the fossil assemblage, and the
closer the fossil taxa are to the Recent species, the greater is
their usefulness in making environmental interpretations.

The initial field party consisted of the authors, together
with John Sutton, Dan Womachel, and several Kenyan assis-
tants. After becoming familiar with the general geological
setting, we sampled eleven localities from the central and
northern part of the area, all of which were above the KBS
Tuff (Harris, 1983). The amount of sediment processed var-
ied with locality, the smallest sample being 1 1 5 kg and the
largest, 3700 kg (Table 1). From this initial sampling three
potentially productive localities were identified. During the
summer of 1975, then again in May and June of 1978, these
localities were intensively sampled, but only one proved to
be sufficiently rich to warrant a large-scale washing operation.
This is our locality 130-A (Figure 1) located below the Okote
Tuff in Area 130 (Isaac and Harris, 1978).

1 . Natural History Museum of Los Angeles County, 900 Expo-
sition Blvd., Los Angeles, California 90007.

2. Carnegie Museum of Natural History, 4400 Forbes Avenue,
Pittsburgh, Pennsylvania 15213.

ISSN 0459-8113

Table 1. Quantities of sediment processed.

Sediment processed (kilograms)

Area

Locality

1974

1975

1978

Species

101

101 -A

180

—

—

None

102

102- A

275

-

—

None

103

103-A

135

-

-

Thryonomys sp.

103-B

180

-

—

104

104- A

115

230

-

Aethomys sp.

130

130-A

700

18,000

2,400

Crocidura cf. C. nana
Crocidura cf. C. dolichura
Scotophilus sp.
Pteropodidae
Nycteris sp.

Tatera sp.

Thyronomys sp.

Hystricid sp.

Arvicanthis sp.

Aethomys sp.

Praomys cf. P. minor
Thallomys quadrilobatus
Mus sp.

131

FxJj20

3,700

—

—

131-A

Chiroptera indet.
Jaculus orientalis
Thryonomys sp.
Hystricid sp.
Aethomys sp.

131-B

450

450

700

Tatera sp.

8

8-A

450

Arvicanthis sp.

Aethomys sp.

Thallomys quadrilobatus

8-B

135

-

-

8-C

115

—

-

Total

6,435

18,680

3,100

The fossiliferous sediment at 130-A is a brown, sandy silt
which changes laterally to a coarse channel sandstone. Within
the stratum, there are lenses of brown, silty clay and many
calcareous root casts. Lithologically the fossiliferous facies is
quite similar to the Type II fossil-bearing facies of Badgley
and Behrensmeyer (1980:143). We interpret the sediments
as overbank or channel margin deposits. Approximately
28,000 kg of matrix was taken from this locality, and most
of the material described here was recovered from this sam-
ple. The sediment processed from FxJj20 (Isaac and Harris,
1978), locality 131-A, was taken from the backdirt of the
archaeological excavation, while that from 131 -B was taken
from a small knoll just to the east of FxJj20. This archaeo-
logical excavation is situated within the Okote Tuff (Leakey
and Leakey, 1978:66) and thus lies stratigraphically above
locality 130-A. A few isolated teeth were recovered from
samples taken from locality 8 in the Ileret Region. These lie

stratigraphically between the samples from 130-A and 131-
A. One Jaculus molar was recovered in 1984 from 10 meters
below the Chari Tuff in Area 3 by Craig Feibel.

All matrix from the various localities was hauled in burlap
sacks to the Koobi Fora camp, where it was placed in Lake
Turkana to soak. The bags were then hand agitated until only
bones, teeth, small calcareous concretions, and a minimum
of silt particles remained in the bags. These were dried and
the resulting bone concentrate sorted at the Koobi Fora base
camp.

Three seasons of fieldwork, approximately 720 man-days
of exploration, washing, and sorting, have resulted in the
recovery of some eighty specimens of rodents, bats, and in-
sectivores, together with a few fragmentary small bird, am-
phibian, and reptile bones. Finding fossiliferous horizons that
represented suitable depositional environments for the ac-
cumulation of small bones and teeth was extremely difficult.

2 Contributions in Science, Number 372

Black and Krishtalka: Turkana Micromammals

The few concentrations of small vertebrate fossils appear to
occur in overbank depressions. Periodic flushing of slow-
moving or intermittent small stream channels during inter-
vals of heavy rains and flooding resulted in water spreading
over the stream banks to form small pools in which fine-
grained sediment and small bones settled out, leaving ac-
cumulations of transported small vertebrates. The apparent
absence of major permanent rivers east of Lake Turkana in
the interval between 1.4 and 1.8 ma was not conducive to
the preservation of the microfauna then living in the region.
Moreover, the prevailing environment, ephemeral rivers
flanked by narrow gallery forest which gave place laterally
to open Acacia scrub or steppe, offered a limited range of
available habitats for small mammals. This might explain
the discrepancy in both sample number and diversity be-
tween the Koobi Fora microfaunal localities and the much
richer but slightly older samples recovered from Olduvai Bed
I or Omo Members F and G.

AGE AND CORRELATION

The principal small mammal producing localities, 1 30-A,
1 31 -A, and 8-A, lie in the lower part of the Upper Member
of the Koobi Fora Formation in the Metridiochoerus an-
drewsi and at the base of the Metridiochoerus compactus
zones of Harris (1983:17). Sediments processed come from
the interval between the KBS and Okote tuffs or from within
the Okote Tuff itself (131 -A). This suggests an age of ap-
proximately 1.6 million years for these small mammal as-
semblages. Harris (1983:18) believes that this portion of the
Turkana section was deposited somewhat later than the sed-
iments of Members F to G of the Shungura Formation in
the Omo valley based upon the correlation of the Shungura
Tuff H2 with the KBS Tuff (Cerling and Brown, 1982). The
F and G assemblages would then date at between 2.4 and
2.2 million years. However, the Turkana rodent assemblages
from 130-A and 131 -A are extremely similar to those from
Omo F and G as well as to the rodents from Olduvai Bed I.

While the Turkana material is insufficient to determine
which species of Tatera, Arvicanthis, Aethomys, and M us are
present, in all cases the material is quite close in size and
morphology to that found at the Omo and/or at Olduvai.
The Turkana rodent assemblage is not as diverse as that from
Members F and G of the Omo, Bed I at Olduvai or the Sidi
Hakoma/Denen-Dora section at Hadar. Both Olduvai (Lav-
ocat, 1965) and Laetoli in Tanzania (Denys, in press) have
a diverse representation of dendromurine rodents, a group
which is not represented in any of the northern faunas. Sci-
urids are present in the Omo and at both Laetoli and Olduvai
but have not been found at Turkana or Hadar.

Based solely on the Muridae, the Turkana assemblages
from localities 1 30-A, 1 3 1 -A, and 8-A postdate but are most
closely comparable to those from the Omo, Members F and
G, and from Oludvai Bed I (Table 2).

ENVIRONMENTAL CONSIDERATIONS

The micromammals from 130-A suggest an arid environ-
ment with some Acacia scrub and perhaps some riverine

Figure 1. The Koobi Fora region showing location of the paleon-
tological collecting areas. From Harris, 1983.

forest in small patches along intermittent stream channels.
The presence of Thallomys strongly suggests Acacia scrub
while Tatera, Arvicanthis, Hystrix, and perhaps Aethomys
are indicative of dry savanna or river floodplain. Praomys,
Mus, and Crocidura all have rather wide ecological tolerance,
but require rather thick vegetative cover whether in mesic
or arid scrub conditions. This assemblage suggests a limited
stream bank forest community rapidly giving place to semi-
arid Acacia savanna and probably considerable true desert
farther removed from the lake shore as indicated by the
Jaculus specimens from 131 -A. A similar environment has
been suggested by Wesselman (1984) for the deposition of
Members F and G at the Omo. As the Turkana assemblages
are younger than those of the Omo perhaps by as much as
500,000-600,000 years, it appears that the arid conditions
noted in the Omo F and G members persisted for a consid-
erable period of time.

METHODS

Measurements were made with an ocular micrometer to the
nearest 0.0 1 mm. Measurements were taken at the maximum
tooth dimension. Tooth terminology follows that of Butler

Contributions in Science, Number 372

Black and Krishtalka: Turkana Micromammals 3

Table 2. Distribution of Koobi Fora rodent taxa at other East African Plio-Pleistocene localities.

Olduvai
Bed I

Omo
F and G

Laetoli

Hadar

Jaculus orientalis

X

X

Tatera sp.

X

T. cf. inclusa

Thryonomys spp.

X

X

X

X

Hystrix sp.

X

X

X

Arvicanthis sp.

A. primaevus

X

Aethomys sp.

A. lavocati

A. deheinzelini

Praomys (Mastomys) minor

X

X

M. cinereus

M. sp.

Thallomys quadrilobatus

X

X

n. sp.

Mus sp.

M. petteri

M. minutoides

X

(1978) and Missone (1969) except that cingulum is used for
both upper and lower teeth following general usage in de-
scribing rodent dentition. All specimens are deposited in the
National Museum of Kenya.

Abbreviations used are:

KNM-ER — Kenya National Museum, East Rudolf

CMNH — Carnegie Museum of Natural History

W— width

L— length

Tri— trigonid

Tal — talonid

mm — millimeter

SYSTEMATIC PALEONTOLOGY
Order Insectivora
Family Soricidae
Genus Crocidura Wagler, 1832

Crocidura sp., cf. C. nana Dobson, 1890

Figure 2

REFERRED MATERIAL. RM, ,-KNM-F.R 5950, 5953;
RM, -KNM-ER 5951.

LOCALITY. 130-A.

DESCRIPTION. M, (L = 1.30; W = 0.80) is relatively
unworn and has a high trigonid that is as long as, but narrower
than, the talonid. The protoconid, the tallest and largest of
the trigonid cusps, is pyramidal, with a long, sloping, antero-
labial face, and occupies approximately two-thirds of the
trigonid. The metaconid is directly lingual to and about half
as high as the protoconid. The well-developed paraconid is
lower than the metaconid and juts anterolingually. The pos-
terior walls of the trigonid and talonid are parallel, and per-
pendicular to the long axis of the tooth. The hypoconid is
an angular cusp at the posterolabial comer of the talonid,
whereas the entoconid is conical and somewhat laterally
compressed. The cristid obliqua meets the trigonid below
and labial to the apex of the protoconid, resulting in a shallow
hypoflexid notch. The posthypocristid runs directly lingually

to the posterior part of the base of the entoconid where it
forms a narrow postentoconid ledge and descends vertically
along the posterolingual comer of the talonid. The anterior
cingulum is broad; a weak lingual cingulum is discernible
along the base of the crown.

M2 (L = 1.20; W = 0.90) repeats the morphology of M„
except that it is slightly smaller, and has a slightly longer
trigonid than talonid. M3 (L = 0.80-0.90; W = 0.65) has a
well-developed trigonid but a markedly reduced talonid; it
is much narrower than the trigonid, unicuspid, and unba-
sined. The single cusp (?hypoconid) bears a round weak facet.
As on M, and M2, a weak cingulum runs along the lingual
part of the base of the crown, and a stronger buccal cingulum
begins anteriorly below the apex of the paraconid.

DISCUSSION. These teeth most closely resemble com-
parable parts of the dentition of C. nana. They are smaller
than those of C. bicolor, C. cyanea, C. flavescens, C. batesi,
and C. fumosa, and slightly smaller than those of C. doli-
chura. M3, without a talonid basin, differs from that of C.
fumosa; M,_2, with a posthypocristid that runs directly lin-
gually rather than obliquely, are unlike those in C. dolichura.

Referral of this material to C. nana is not warranted, how-
ever, considering the lack of preserved parts of the upper and
anterior dentitions, which are crucial to the identification of
crocidurines. P.M. Butler (personal communication) has sug-
gested that this material may belong to a species of Crocidura
recovered from the Kaffir Beer Cave, Transvaal.

Crocidura sp. cf. C. dolichura Peters, 1876

Figure 3

REFERRED MATERIAL. RM, -KNM-ER 5952; RM1-
KNM-ER 5954.

LOCALITY: 130-A.

DESCRIPTION. KNM-ER 5952 is a right mandibular
fragment with M, (L = 1.40; W = 1.00), and broken bases
of the crowns of P4 and M2. The upper half of the protoconid
and the posterior portion of the metaconid are broken away
from M,, but it is apparent that these cusps were higher than
the paraconid in the undamaged state. On the talonid, which
is wider than the trigonid, the entoconid is high, conical, and

4 Contributions in Science, Number 372

Black and Krishtalka: Turkana Micromammals

2

Figures 2 through 5. Cmcidura sp., Scotophilus sp., and Jaculus orientalis. 2. Crocidura sp. cf. C. nana. KNM-ER 5953, RM2_3, x20. 3.
Crocidura sp. cf. C. dolichura. (A) KNM-ER 5952, RM,; (B) KNM-ER 5954, RM', x20. 4. Scotophilus sp. KNM-ER 5955, LM,, x20. (A)
occlusal view, (B) labial view. 5. Jaculus orientalis. (A) KNM-ER 5910, LM2, x 15; (B) KNM-ER 5911, RM1, x20.

Contributions in Science, Number 372

Black and Krishtalka: Turkana Micromammals 5

somewhat laterally compressed. The hypoconid is V-shaped;
the posthypocristid runs lingually and somewhat obliquely
to behind the entoconid where it forms a postentoconid ledge.
The buccal cingulum is strong, whereas the lingual cingu-
lum — along the entire base of the crown — is weaker. A mental
foramen occurs on the mandible below P4.

In occlusal view, the crown of M' (labial L = 1.30; lingual
L = 1.40; posterior W = 2.10) is transverse and trapezoidal:
the straight labial and lingual margins diverge posterobuc-
cally and posterolingually from the anterior margin; the pos-
terior border is deeply excavated between the hypoconal shelf
and metacone. The latter is approximately twice as large and
somewhat higher than the paracone; both cusps are crescentic
and form a W-shaped ectoloph with a mesostyle. The hy-
pocone, a cuspule at the anterior end of the hypoconal shelf,
is almost directly lingual to the apex of the metacone and is
not connected to the postprotocrista. The shelf extends far-
ther posteriorly than does the parastylar salient.

DISCUSSION. This M, is slightly larger than that referred
to Crocidura sp. cf. C. nana ; both M, and M1 are close in
size and morphology to comparative material of C. doli-
chura. Additionally, M1 is more transverse than in C. bicolor
and C. formosa, and has a less expanded hypoconal shelf
than in the former.

Order Chiroptera
Family Vespertilionidae
Genus Scotophilus Leach, 1 82 1

Scotophilus sp.

Figure 4

REFERRED MATERIAL. LM , — KNM-ER 5955; LM1
or LM2— KNM-ER 5956.

LOCALITY. 130-A.

DESCRIPTION. The M, (L = 1.80; L Tri = 1.10; L Tal
= 0.65; W Tal = 1.50) is semi-zalambdodont, with a large
trigonid and much lower, shorter, and slightly narrower tal-
onid. The trigonid leans lingually so that, in anterior view,
the labial slope of the protoconid is long and inclined ap-
proximately 60 degrees to the horizontal. In dorsal view, the
occlusal area of the three trigonid cusps occupies only the
lingual one-quarter of the basal width of the trigonid. The
latter is triangular, with the metaconid and paraconid sub-
equal and situated posterolingual and anterolingual, respec-
tively, to the much higher protoconid. The protoconid is
pyramidal, whereas the paraconid and metaconid are some-
what compressed anteroposteriorly. The buccal slope of the
hypoconid, like that of the protoconid, is acutely inclined
and occupies half of the talonid crown. The hypoconid is V-
shaped, with an extremely short cristid obliqua that meets
the trigonid below and labial to the apex of the protoconid.
The posthypocristid extends posterolingually, parallel to the
posterior wall of the trigonid, to the apex of a small, worn
entoconid, which, in turn, is linked by a short cristid to the

base of the metaconid. A vertical ridge of enamel on the
posterior wall of the talonid below the apex of entoconid may
represent an isolated, reduced hypoconulid. A strong basal
cingulum is continuous along the anterior, buccal, and pos-
terior surfaces of the crown.

The upper molar is a lingual fragment that preserves the
protocone, part of the metacone, and the basolingual wall of
the paracone. Like the lower molar, the protoconal portion
of this tooth is semi-zalambdodont: the protocone is ex-
tremely high, and compressed anteroposteriorly, with a long,
oblique lingual slope; its occlusal surface is subcrescentic and
small, with very short protocristae. The postprotocrista ends
at the lingual wall of the metacone, which is subcrescentic,
anteroposteriorly compressed, and approximately one-third
higher than the protocone. The V-shaped occlusal surface of
the metacone and the preserved portion of the paracone im-
ply that the two cusps were dilambdodont. There are no
conules. A strong, basal postcingulum is slightly expanded
at the posterolingual part of the base of the protocone, where-
as the anterior cingulum is much weaker.

DISCUSSION. These two isolated molars cannot be as-
signed with confidence to a particular species of Scotophilus.
The slope of the protocone on the upper molar is higher than
in S. viridus and also more compressed laterally than in S.
nux and S. nigritellus. The structure of the protocone most
closely resembles that in S. leucogaster and S. dinganii, species
that in addition to 5. viridus also show the development of
a hypoconulid on M1. However, this cusp on KNM-ER 5955
is more strongly developed.

These two teeth mark the first fossil record of Scotophilus
(cf. Butler, 1978).

Family Pteropodidae
Genus and species indet.

REFERRED MATERIAL. KNM-ER 5958, RM,

LOCALITY. 130-A.

DESCRIPTION. This isolated tooth (L = 2.50; W = 1.40)
corresponds to M, of pteropodids in being rectangular in
occlusal outline and in having a high, sectorial labial wall, a
parallel and much lower lingual wall, and a low, median,
longitudinal valley. In external view, the labial wall resembles
a triangle in which the anterior slope rises steeply to an apex
situated approximately one-third of the distance from the
anterior border of the crown. The posterior slope of the labial
wall descends more gradually to the posterior edge of the
tooth. In medial view, the lingual wall mirrors the shape of
the labial one, but its apex, directly opposite that on the labial
wall, is only a third as high as the latter. A weak crest joins
the two apices and forms the highest point of the median
valley where it crosses the crown. The anterior and posterior
borders of the crown are gently rounded and the median
valley is weakly rugose.

DISCUSSION. This tooth closely resembles M, in a num-
ber of pteropodid genera, especially Epomops, Epomopho-

6 Contributions in Science, Number 372

Black and Krishtalka: Turkana Micromammals

rus, and Myonycteris. A more precise identification is not
possible.

?Family Nycteridae

Genus Nycteris Geoffroy and Cuvier, 1795
? Nycteris sp.

REFERRED MATERIAL. KNM-ER 5957, RC,.

LOCALITY. 130-A.

DESCRIPTION. This isolated canine (L = 1 . 1 0; W = 1 .00)
leans posterolingually and is three sided in occlusal cross
section: an arc-shaped anterolabial face and flat posterior and
lingual faces. A broad, continuous cingulum, which is highest
anteriorly, descends sharply ventrolabially and more gently
ventrolingually from that point to rim the base of the crown.
A cuspule is developed on the posterolingual comer of the
cingulum.

DISCUSSION. Referral of this canine to the Nycteridae
and Nycteris is tentative, although the size of the canine and
the structure of the cingulum closely resembles that in the
latter.

Chiroptera indet.

REFERRED MATERIAL. KNM-ER 5959, LM,.

LOCALITY. 131-A.

DESCRIPTION. This isolated lower molar (L = 2.70;
W = 1.90) is worn and much of the enamel on the external
faces of the talonid has been eroded. The tooth is rectangular
in occlusal outline, with the trigonid approximately as long
and wide as the talonid, but twice as high. With the roots
oriented vertically, the labial slope of the protoconid is se-
verely canted (approximately 40 degrees to the horizontal),
so that the occlusal area bounded by the three trigonid cusps
is restricted to the lingual one-third of the basal width of the
trigonid. The paraconid, larger and slightly lower than the
metaconid, is a pyramidal cusp that forms the anterolingual
corner of the crown. The metaconid, more nearly conical
than the paraconid, occurs lingual and slightly posterior to
the protoconid and directly posterior to the paraconid. As a
result, the posterior wall of the trigonid is oriented obliquely
posterolingually; the trigonid basin is large and open lin-
gually, whereas it is closed anteriorly and posteriorly by V-
shaped cristids that join the protoconid to the paraconid and
metaconid.

In posterior view, the external slope of the hypoconid is
more nearly vertical than that of the protoconid (approxi-
mately 60 degrees to the horizontal), so that the occlusal
surface of the talonid occupies approximately the lingual two
thirds of the basal width of the crown. The hypoconid, at
the posterolingual comer of the talonid, is large, broad at the
base and was subcrescentic (if not crescentic) in the unworn,
undamaged condition. The entoconid, directly lingual to the
hypoconid, was much smaller and conical, judging from its
wear facet. A tiny (and mostly eroded) hypoconulid is just
posterior to and twinned with the entoconid. The cristid
obliqua from the hypoconid meets the trigonid medially,

below and slightly labial to the apex of the protoconid. The
posthypocristid runs directly lingually toward the entoconid,
but deflects near its base to join the hypoconulid. A strong
basal cingulum appears to have been present and continuous
along the anterior, labial and posterior aspects of the crown.

DISCUSSION. This tooth most closely resembles M, of
some Rhinolophidae and Hipposideridae in the open tri-
gonid, the posterolingual orientation of the posterior wall of
the trigonid, the trigonid-talonid proportions, and the twin-
ning of the entoconid and hypoconulid. However, first and
second lower molars in these groups are more nearly di-
lambdodont in that the cristid obliqua meets the posterior
wall of the trigonid lingually, near the metaconid, rather than
medially as in KNM-ER 5959. Some specimens of Hipposi-
deros commersoni and Rhinolophus eloquens in the CMNH
collections approach KNM-ER 5959 in size and in the ori-
entation of the cristid obliqua, but the similarities are not
sufficient to warrant positive referral to these taxa.

Order Rodentia
Family Dipodidae
Genus Jaculus Erxleben, 1777

Jaculus orientalis Erxleben, 1977

Figure 5

REFERRED MATERIAL. KNM-ER 5911, RM1 (L =
2.45, W = 2.35); K84-2093a, LM,; KNM-ER 5910, LM2
(L = 2.25, W = 2.10).

LOCALITY. 131-A.

DESCRIPTION. The upper molar is moderately worn but
shows all occlusal features clearly. The protocone is some-
what larger than the other principal cusps. The protocone
and hypocone are set slightly behind the paracone and meta-
cone to which they are connected by strong crests. A short
anterior loph passes from the protocone to the anterointemal
margin of the paracone, forming a shallow notch between
the loph and the protoloph. The valley between the paracone
and metacone and that between the protocone and hypocone
is deep. A short posterior loph passes posteromedially from
the hypocone defining a short valley between it and the meta-
cone.

The first lower molar is fragmentary. The second lower
molar is somewhat worn and moderately high crowned. The
protoconid and hypoconid are set behind the metaconid and
entoconid so that the metalophid and hypolophid slant an-
terolingually. There is a strong lophid from the metaconid
that passes anterior to the protoconid with a deep, narrow
valley between the lophid and the protoconid. The lingual
valley is broad and deep and passes anterobuccally in front
of the protoconid-entoconid lophid. The central buccal valley
passes posterolingually to the base of the entoconid.

DISCUSSION. The M2 is similar to that of Jaculus ori-
entalis. Jaculus has been reported from the Omo Members
F and G, again on very few teeth (6), but is unknown at
Olduvai, Laetoli, and Hadar.

The M1 is similar to that figured by Wesselman (1984:

Contributions in Science, Number 372

Black and Krishtalka: Turkana Micromammals 7

159). In an earlier publication (Black, 1984:1 13) this tooth
was mistakenly identified as that of a ctenodactylid; however,
it clearly represents a jerboa. The three teeth from Turkana
compare well with those from the Omo.

Family Cricetidae
Genus Tatera Lataste, 1882
Tatera sp.

MATERIAL. KNM-ER 2399, LM1; 2394, RM3.

LOCALITIES. 131-A and 131-B.

DESCRIPTION. The upper first molar (L = 2.34; W =
1.60) is extremely well worn but still shows the pattern of
three cross lophs or crests. The anterior loph is the narrowest;
the central loph, the widest. There is the faintest suggestion
of an anteroposterior connection between the three lophs in
the midline of the tooth. The third upper molar (L = 1.22;
W = 1.33) is not as heavily worn and is about as high crowned
as a similar tooth in Tatera afra and is of approximately the
same size. The first molar is not as broad as that of T. afra
but is of the same length.

DISCUSSION. With only two isolated teeth of this taxon
from the entire Turkana sample, it is impossible to identify
accurately the species present. Both teeth come from locality
131, one from the backdirt of the archaeological site FxJj20,
and the other from a sample taken 50 meters to the north.
Species of Tatera are also reported from Olduvai Bed I, Omo
Shungara Members B and F, and Hadar.

Family Thryonomyidae

A number of specimens of Thryonomys, the cane rat, have
been found over the years by others prospecting for hominids
and other larger vertebrate fossils. Our field parties recovered
several isolated teeth while prospecting in area 8-A, near
Ileret, but did not recover any other specimens during our
washing operations. Thryonomys is known to occur at lo-
calities 131, 130, 8, 103, 1-A, and 1. This material has not
been studied in detail but only a single species of Thryonomys
appears to be present.

Family Hystricidae

Hystricids are known from only a few rather fragmentary
specimens from Areas 106, 130, and 131. Whether both

Xenohystrix and Hystrix are present, as they are at the Omo,
has not been determined.

Family Muridae
Genus Arvicanthis Lesson, 1842

Arvicanthis sp.

Figure 6

REFERRED MATERIAL. KNM-ER 2396, partial LM';
5949, LM1; 5944, 5945, LM,-M2; 5946, 5947, M2; 5948,
RM3.

LOCALITIES. 8-A and 1 30-A.

DESCRIPTION. The M1 has the principal cusps arranged
in transverse crests, separated by shallow valleys. T1 and T3
are set only slightly behind T2; they are both smaller than
T4 and T6. There is no T7 while T9 is broadly connected to
T8.

The lower first molars show some variation in the devel-
opment of the labial cingulum and cingular cusps with Cv5
prominent on 5944, but absent on 5945. On each tooth there
is a small cingular ridge between the first and second cross
crests. There is no anterior Sm cusp and the posterior cin-
gulum is reduced. The SV and SI cusps form an oblique crest
separated from the Epd and Eod crest which shows a weak
central ridge directed anteriorly. On M2, Sv is small and there
is no Cv5 on 5945, whereas it is quite small on 5944. The
posterior cingulum cusp, Z, is small. M3 has two transverse
crests, the anterior composed of Epd and Eod while the pos-
terior appears to be formed by a single transversely elongate
cusp.

DISCUSSION. The oblique posterolabial orientation of
the anterior crest on M, relates this sample to Arvicanthis as
does the arrangement of the cusps of M1 in essentially trans-
verse laminae. These specimens are similar in most respects
to those assigned to Arvicanthis primaevus (Jaeger, 1976:71)

Table 3. Measurements in mm of molars of Arvicanthis sp.

KNM-ER

Specimen

L

W

L

W

L

W

5949 M1

2.10

1.47

5944 M,-M2

2.25

1.55

1.55

1.70

5945 M,-M2

2.21

1.47

1.47

1.50

5947 M2

1.81

1.68

5948 M,

1.50

1.42

Figures 6 and 7. Arvicanthis sp. and Aethomys sp. 6. Arvicanthis sp. (A) KNM-ER 5944, LM,_2; (B) KNM-ER 5945, LM,_2; (C) KNM-ER
5949, LM1, x 20. 7. Aethomys sp. (A) KNM-ER 5941, LM,; (B) KNM-ER 2397, M2; (C) KNM-ER 5942, M2; (D) KNM-ER 5939, RM1; (E)
KNM-ER 5940, RM3, x20.

8 Contributions in Science, Number 372

Black and Krishtalka: Turkana Micromammals

Contributions in Science, Number 372

Black and Krishtalka: Turkana Micromammals 9

except that they are somewhat smaller and differ in having
a more variably developed labial cingulum and Cv5 on M,.

The Turkana species is closely related to A. primaevus from
Olduvai but may prove to be specifically distinct when more
adequate material is discovered.

Measurements of molars of Arvicanthis sp. are given in
Table 3.

Genus Aethomys Thomas, 1915

Aethomys sp.

Figure 7

REFERRED MATERIAL. KNM-ER 5939, RM1; 5940,
RM3; 5941, partial LM,; 2395, 2397, 2400, 5942, M2; 5943,
LM3.

LOCALITIES. 8-A, 104, 131-A.

DESCRIPTION. In the upper molars all of the cusps are
inclined toward the rear with the transverse lophs separated
by relatively wide valleys. On M1, Tl, T3, T4, and T6 are
of similar size and smaller than T2, T5, and T8. T9 is smaller
than the other cusps and is slightly anterior to T8. T4 to T9
are strongly connected, while T3 is only connected to T6 at
its base. There is a small cingulum around the base of T2
bearing a small internal cusp. The third upper molar is quite
large. Tl is large and T3 is absent. T4, T5, and T6 are fused
into a strong central crest which is deeply separated from a
smaller but still quite strong T8 and T9 crest.

There is only a partial M,, preserving the posterior two-
thirds of the tooth, that may be assigned to Aethomys. The
four principal cusps are inclined anteriorly with Eod and Epd
not joined, whereas Td and End are joined near their summits
to form a transverse crest. The lingual cusps are set somewhat
ahead of the buccal ones. The buccal cingulum is discontin-
uous. Cv5 is large, while Z is small. M, is a robust tooth with
prominent Sv, Cv3, and Z cusps. The transverse crests are
strong, with broad fusion of Eod-Epd and Td-End. M3 is also
stout and displays two transverse crests with little sign of
individual cusps.

DISCUSSION. These specimens most closely resemble
Aethomys deheinzelini from Omo Members F and G (Wes-
selman, 1984: 133) but they are somewhat larger. As he noted
both the Omo and Turkana species lack any indication of
stephanodonty on M1 such as is seen in Aethomys lavocati
from Olduvai. The posterior cingulum of Mi is reduced as
in Aethomys deheinzelini. However, without additional and
more complete specimens of M1 and M, in the Turkana
sample it is difficult to determine its exact relationships.

Measurements of molars of Aethomys sp. are given in
Table 4.

Genus Praomys Thomas, 1915
Subgenus Mastomys Thomas, 1915

Praomys ( Mastomys ) cf. P. (M.) minor
Jaeger, 1976
Figure 8

REFERRED MATERIAL. KNM-ER 2392, LM,-M2;
2393, RM,-M3; 5912, RM,-M3; 5913, LM,-M3; 5914, LMr
M3; 5915, RM,; 5916-5920, M2; 5921-5922, M2; 5923, RM1-
M-; 5924-5930, M1; 5931-5932, M2; 5933, M3.

LOCALITY. 130-A.

DESCRIPTION. As in the material from Bed I at Olduvai,
the valleys on M1 between T2 and T3, and between T5 and
T6, are deeply notched, much more so than between Tl and
T2 and between T4 and T5. Tl and T4 are set well behind
T2 and T5. T9 is large. There are no cingular cusps. On M2,
T3 is quite small, as is T9.

On the mandible the masseteric crest is a strong ridge
which rises from the ventral border of the mandible below
M, to terminate in a prominent shelf just below the anterior
root of M,. There is a single, large mental foramen just an-
terior to this shelf and just below the symphysis.

The lower first molar has no Sm and only a trace of the
buccal cingulum between Sv and Eod. C5 is present, but small
on M, and M2. Z is not present as a distinct cusp, but is
merely a thickened posterior shelf on both M, and M,. The
four anterior cusps of M , form an X pattern with S 1 displaced
anteromesially. Sv is prominent on M2.

DISCUSSION. The upper molars from 1 30-A appear to
be somewhat smaller than those of Praomys ( Mastomys )
minor from Olduvai Bed I (Jaeger, 1976:89). The lower mo-
lars from the two localities, however, agree quite well in size

Table 4. Measurements in mm of molars of Aethomys sp.

KNM-ER

Specimen

L

W

5939

M1

2.81

2.36

5940

M3

2.14

1.88

5941

M,

—

1.00

5942

m2

2.06

2.00

2395

m2

2.13

2.15

5943

m3

1.55

—

Figure 8. Praomys ( Mastomys ) cf. P. ( M .) minor (A) KNM-ER 5913, LM,_3; (B) KNM-ER 5914, LM,_3; (C) KNM-ER 2393, LM,_3; (D)
KNM-ER 5923, RM,_2; (E) KNM-ER 5928, RM1; (F) KNM-ER 5930, RM1; (G) KNM-ER 5927, RM1; (H) KNM-ER 5929, LM1; (I) KNM-
ER 5932, LM2, x 20.

10 Contributions in Science, Number 372

Black and Krishtalka: Turkana Micromammals

Contributions in Science, Number 372

H

Black and Krishtalka: Turkana Micromammals 11

and the difference in the upper molars may be due simply
to a difference in technique of measurement. Praomys sp.
from Hadar (Sabatier, 1982) is quite close in size to the
Turkana sample but differs significantly in morphology. The
Omo Member B and Member F Mastomys is also referred
to M. minor (Wesselman, 1984). There is certainly nothing
in the occlusal morphology on which to separate the Turkana
and Olduvai samples.

Measurements of molars of Praomys (Mastomys) cf. P.
(M.) minor are given in Table 5.

Genus Thallomys Thomas, 1920

Thallomys quadrilobatus Jaeger, 1976

Figure 9

REFERRED MATERIAL. 5900, 5901, M1; 2398, 5902,
5903; 5908, M2; 5909, M3; 5904, 5905, M,; 5906, M2-M3;
5907, M2.

LOCALITIES. 1 30-A and 8-A.

DESCRIPTION. All teeth are quite similar in occlusal

Table 5. Measurements in mm of molars of Praomys ( Mastomys )
cf. P. (M.) minor.

KNM-ER

Specimen

M1

M2

M3

L

W

L

W

L

W

5923

2.18

1.45

1.41

1.40

—

-

5924

2.10

1.47

-

-

-

-

5925

2.13

1.45

-

-

-

-

5927

2.10

1.43

-

-

-

-

5928

2.13

1.43

-

-

-

-

5929

2.30

1.36

-

-

-

-

5930

2.14

1.33

-

-

-

-

5931

—

—

1.60

1.34

-

-

5932

-

—

1.69

1.31

-

-

5933

-

-

-

-

1.09

1.15

M,

M2

M,

L

W

L

W

L

W

2392

2.13

1.22

1.35

1.24

—

—

2393

—

1.07

1.22

1.10

0.81

0.85

5912

2.00

1.23

1.40

1.22

0.85

0.96

5913

2.11

1.20

1.50

1.25

1.10

0.95

5914

2.10

1.25

1.40

1.25

1.07

0.97

5915

2.13

1.15

-

-

-

-

5916

-

—

1.40

1.24

-

-

5917

—

—

1.45

1.14

-

-

5918

—

-

1.48

1.25

-

-

5919

—

-

1.45

1.35

-

-

5920

—

—

1.40

1.15

-

-

5921

—

—

-

-

1.13

1.00

5922

-

-

—

—

1.10

1.00

morphology to those of Recent Thallomys paedulcus from
Zimbabwe to those of Thallomys quadrilobatus from Bed I
of Olduvai figured by Jaeger (1976). The cheek teeth are
slightly broader and perhaps larger overall than those of T.
paedulcus.

On M', T1 and T4 are set distal to T3 and T6 with a short
crest passing from the base of T3 towards T6. The stephan-
odont crests from T4 to T8 and from T6 to T9 are distinct
but low. The cusp T9 is perhaps larger in the Turkana species
than in the living species and there is a faint indication of
posterior cingular Z, a structure not present in the living
forms. The M2 is strongly stephanodont with well-developed
crests from T4 to T8 and T6 to T9. The T1 is a strong
columnar cusp not flattened transversely as in the living
species. Also, T9 is a much stronger cusp than that on M2
of T. paedulcus and there is a small but distinct posterior
cingular Z present on the Turkana M2s. M3 has a strong T1
and a distinct but transversely flattened T3. On the central
crest T4, T5, and T6 are distinct with T5 displaced towards
the anteroextemal comer of the crown. As the one M3 in this
collection is relatively unworn, both T8 and T9 are discern-
ible in the posterior loph, which is distinctly separated from
the T4-T5-T6 loph by narrow valleys.

The first lower molar is essentially indistinguishable from
M, of T. paedulcus, the only difference being the presence of
a very low, faint Sm ridge between Sv and SI and a slightly
broader posterior shelf beteen the End and Td cusps in the
fossil. The buccal cingulum is continuous from the posterior
border of Sv to the rear of Td. There appears to be a Cv5
cusp on the cingulum. There are anteroposterior crests which
unite SI, Sv, Epd, and Eod along the midline of the tooth
and also a short, anteriorly directed midline crest for End
and Td. The four principal cusps on M2 unite with short
anteriorly directed crests in the midline. Both Sv and C3 cusps
are present as is a strong Z cusp. On M3 a small Sv is present.
On the posterior loph the End cusp is large and transversely
expanded, with only a very small Td component to this loph.

DISCUSSION. The specimens here assigned to Thallo-
mys quadrilobatus are essentially identical to those described
by Jaeger (1976) from Bed I at Olduvai. Jaeger considered
T. quadrilobatus to be directly ancestral to the modem species
of Thallomys and we agree. A reduction in size of T9 and
decrease in size or loss of the posterior cingulum (Z) on M1
and M2, together with reduction of Sm on M, are all that
separate the fossil from the modem species.

Measurements of molars of Thallomys quadrilobatus are
given in Table 6.

Genus Mus Linnaeus, 1758

Mus sp.

Figure 10

REFERRED MATERIAL. 5934-5937, M,; 5938, RM2-
M3.

LOCALITY. 1 30-A.

DESCRIPTION. Only lower molars of this small species
of Mus have been recovered. M, has a short crest directed

12 Contributions in Science, Number 372

Black and Krishtalka: Turkana Micromammals

Figures 9 and 10. Thallomys quadrilobatus and Mus sp. 9. Thallomys quadrilobatus. (A) KNM-ER 5904, RM,; (B) KNM-ER 5906, RM,_5;
(C) KNM-ER 5900, RM1; (D) KNM-ER 5903, RM2; (E) KNM-ER 5902, LM2; (F) KNM-ER 2398, LM2, x2G. 10. Mus sp. (A) KNM-ER
5935, LM,; (B) KNM-ER 5936, RM,; (C) KNM-ER 5938, RM,.,, x20.

Contributions in Science, Number 372

Black and Krishtalka: Turkana Micromammals 13

Table 6. Measurements in mm of molars of Thallomys quadrilo-
batus.

KNM-ER

Specmien

L

W

5900

M1

2.45

1.68

5901

M1

2.44

1.66

2398

M2

1.61

1.57

5902

M2

1.69

1.59

5903

M2

1.60

1.60

5908

M2

1.64

1.72

5909

M3

1.64

1.46

5904

M,

2.21

1.32

5905

M,

2.27

1.45

5907

m2

1.75

1.49

5906

m2

1.76

1.49

m3

1.49

1.41

internally from the anterointemal comer of SI. There is no
anterior Sm cusp, although on KNM-ER 5935, the anterior
crest is somewhat swollen. SI is anterolingual on all the first
molars. The buccal cingulum is continuous from Sv to the
posterior border of M, and Cv5 is present on KNM-5936,
while on the other M,s it is absent. A small Z ridge or ledge
is present. Sv and Sv5 are small on M, and on M3 only a
narrow Sv is present. M3 is not greatly reduced.

DISCUSSION. The specimens here referred to as Mus
are the smallest of the Turkana murids and are just slightly
larger than the teeth of Mus petteri from Olduvai Bed I (Jae-
ger, 1 976). However, in the absence of any upper first molars,
reference to any specific species of Mus is not possible. The
Turkana species differs from Mus petteri in possessing the
anterior crest from Sv and a somewhat more prominent buc-
cal cingulum on M,. A distinct species of Mus may have been
present at Turkana but more material is needed to charac-
terize it.

Measurements of molars of Mus sp. are given in Table 7.

CONCLUSIONS

1. The assemblage from locality 130-A at Turkana, while
somewhat younger than those from the Omo Members F and
G Olduvai Bed I, most closely resembles those local faunas.

Table 7. Measurements in mm of molars of Mus sp.

KNM-ER

Specimen

L

W

5934

M,

1.68

0.96

5935

M,

1.66

1.00

5936

M,

1.69

1.09

5937

M,

1.64

1.00

5938

m2

1.08

1.00

5938

m3

0.71

0.75

2. The Turkana rodents strongly suggest an arid environ-
ment with intermittent stream drainages bordered by sparse
riverine forest and Acacia scrub.

3. Some 7 20 man-days of fieldwork produced only 80 spec-
imens of rodents, bats, and insectivores. Suitable deposi-
tional environments for preservation of small mammals thus,
appear to be quite rare in the Koobi Fora Formation.

ACKNOWLEDGMENTS

Fieldwork was supported by the National Science Founda-
tion (BMS75-22791), the National Geographic Society (Grant
No. 1453), and the M. Graham Netting Research Fund of
Carnegie Museum of Natural History. We thank Richard
Leakey and many members of the Koobi Fora Research
Team for their assistance both in the field and in Nairobi.
We also thank John Harris and Lou Jacobs for their helpful
criticism of this manuscript, and Nancy J. Perkins for the
illustrations.

LITERATURE CITED

Badgley, C., and A.K. Behrensmeyer. 1980. Paleoecology
of Middle Siwalik sediments and faunas. Northern Pa-
kistan. Palaeogeography, Palaeoclimatology, Palaeo-
ecology’ 30:133-155.

Black, C.C. 1984. Systematics and paleoecology of small
vertebrates from the Plio-Pleistocene deposits east of
Lake Turkana, Kenya. National Geographic Society Re-
search Report 16: 1 1 1-1 1 3.

Butler, P.M. 1 969. Insectivores and bats from the Miocene
of East Africa: new material, pp. 1-37, in Fossil Verte-
brates of Africa (L.S.B. Leakey, ed.), Volume 1, Aca-
demic Press, New York.

. 1978. Insectivores and Chiroptera. pp. 56-68, in

Evolution of African Mammals (V.J. Maglio and H.B.S.
Cooke, eds.), Harvard University Press, Cambridge.
Butler, P.M., and M. Greenwood. 1973. The early Pleis-
tocene hedgehog from Olduvai, Tanzania, pp. 7-42, in
Fossil Vertebrates of Africa (L.S.B. Leakey, R.J.G. Sav-
age, and S.C. Coryndon, eds.), Volume 3, Academic Press,
New York.

. 1976. Elephant shrews (Macroscelididae) from Ol-
duvai and Makapansgat. pp. 1-56, in Fossil Vertebrates
of Africa (R.J.G. Savage, and S.C. Coryndon, eds.), Vol-
ume 4, Academic Press, New York.

. 1979. Soricidae (Mammalia) from the early Pleis-
tocene of Olduvai Gorge, Tanzania. Zoological Journal
Linnean Society London 67:329-379.

Cerling, T.E., and F.H. Brown. 1982. Tuffaceous marker
horizons in the Koobi Fora region and in the lower Omo
Valley. Nature 299:216-221.

Coppens, Y., F.C. Howell, G.L. Isaac, and R.E.F. Leakey,
eds. 1 976. Earliest man and environments in the Lake
Rudolf Basin. University of Chicago Press.

Denys, C. (in press) Fossil rodents (other than Pedetidae)
from Laetoli. In: Leakey, M.D. and J.M. Harris (eds.),
Laetoli: a Pliocene site in northern Tanzania. Clarendon
Press, Oxford.

14 Contributions in Science, Number 372

Black and Krishtalka: Turkana Micromammals

Harris, J.M. 1983. Background to the study of the Koobi
Fora fossil faunas. In Koobi Fora Research Project Vol.
2:1-2. J.M. Harris, ed. Clarendon Press, Oxford.

Isaac, G.L., and J.W.K. Harris. 1978. Archaeology. In M.L.

Leakey and R.E. Leakey, eds. Koobi Fora Research Proj-
ect Vol. 1:64-85.

Jaeger, J.-J. 1976. Les Rongeurs (Mammalia, Rodentia) du
Pleistocene Inferieur d’Olduvai Bed I (Tanzanie) Pre Par-
tie: Les Murides. In Fossil Vertebrates of Africa (ed. by
R.S.G. Savage and S.C. Coryndon). Vol. 4, pp. 57-120,
3 pis. Academic Press.

Jaeger, J.-J., and H.B. Wesselman. 1976. Fossil remains of
micromammals from the Omo Group deposits. In Ear-
liest man and environments in the Lake Rudolf Basin
(Y. Coppens, F.C. Howell, G.L1. Isaac, and R.E.F. Leak-
ey, eds.), pp. 351-360. University of Chicago Press.

Lavocat, R. 1965. In O/duvai Gorge 1951-1961. Leakey,

L.S.B., ed., Vol. 1 : Fauna and Background: 1 7-1 8. Cam-
bridge University Press, London.

Leakey, M.G., and R.E. Leakey, eds. 1978. Koobi Fora
Research Project. Volume 1. The fossil hominids and
an introduction to their context, 1968-1974. Clarendon
Press, Oxford.

Missone, X. 1969. African and Indo-Australian Muridae
Evolutionary Trends. Musee royal de l’Afrique cen-
trale— Tervuren Belgique. Annales Science Zoologie 1 72,
219 pp., 27 pis.

Sabatier, M. 1982. Les Rongeurs du site Pliocene a Homi-
nides le Hadar (Ethiopie). Palaeovertebrata 12(1): 1-56,
4 pis.

Wesselman, H.B. 1984. The Omo micromammals. Con-
trib. to Vertebrate Evolution 1-219.

Accepted 9 October 1985.

Contributions in Science, Number 372

Black and Krishtalka: Turkana Micromammals 15

aiiWWii ij,i:|-':i r l!;i I:;-!

rCTIQNS FOR .

feilililillfel:'-'1' JlljK tTl&jii Tjj| S ’K:

The Natural
in the life and

ZnSrfnhs to^ater ^ ^ **** <* *

monographs. ; . -j ■ . ,: ■ ■, ^ y ;y,.-. ■, , . j!||| y. ;s, iV;.; j, ; iy; .bi;; •■■-.! - :, ■ ; pi! 4 - : ■

Manuscripts submitted for publication will undergo anonymous;.!©'

''original resea rc

given to manuscripts written by members dt -ike'' Mus'eu-m' staff. Manuscripts Should:
prepared in accordance with the require me his outlined below and submitted to' * ’

the appropriate Section of the Museum.

Authors must adhere to the articles of the International Codes of Bo
Nomenclature and are urged to comply withal! recommendations of t

Authors proposing new taxa must, indicate chat primary types have been deposited in
accordance with the recommendations of the appropriate 'code, citing the collection by name
and providing the relevant identifying' details. The depository of other study material should
also be indicated.

An abstract is required for all papers. In manuscripts dealing with. Latin American subjects,
the abstract must be presented in Spanish or Portuguese., as appropriate, as well, as in English.
Summaries in other languages are hot required hut are strongly recommended where ap-
propriate. iy||;[ ,gj 111 \ ! ij ; yi; :L|:|!j|!:iy!|l!|l W^m i i| 1 1 : ,,;il | ■;$ ' i : ■ 4 ;;; •’ i : ! : ; . I : . , ■ ■

HBi

PREPARATION OF; MAMtSjCRIPT

text, aejknowiedgnicnis; ref-
pm of 3 prattled pages (as a .
■A •'■■"■ov;.. Indicate italicized

Type manuscript double-spaced, including title page, abstract
erences, tables, and figure, captions. A paper must total a ns '
guideline, 3.5 pages of draft equals about 1 page of J'inal ;
words with underscores; do not use an italic typing dement

Each manuscript component should begin on a new page, in the 'following sequence; title
page; abstracts); text: acknowledgments; literature cited: tables, each table —complete with
title and footnotes— on a separate page; and figure captions. All pages should be' numbered
consecutively.

Submit original illustrations. Line drawings are best prepared for 50 percent reduction
and should not exceed 14 by 17.5 inches (35 by 43 era) in size; photographs are usually
reproduced at full size and should not exceed 7 by 8.7.5 inches (17 by 22 cm. full page width)
or 3.25 by 4.5 inches (8 by 1 1 cm, single column).

SUBMISSION OF MANUSCRIPT

Transmit three copies of the -manuscript in a secure enclosure. Manuscript should be.ac-J
companied by a cover letter signed by the author who will be responsible for correspondence
regarding the manuscript. The covering letter should contain a statement that the manuscript
is based on original research and has:, not been published elsewhere except in abstract or
abbreviated form. The letter should also give information on prior publication of any part
of the' research and whether or not; funds are available to meet 'the cost, of unusual format
or color illustrations. Include- copies of an-| permissions needed to reproduce published
material.

Manuscripts will be reviewed for possible publication with -the. understanding -that -they ;
have not been published, simultaneously submitted, -or already accepted for publication
elsewhere. This does not preclude consideration of a complete report that follows publication
of preliminary findings elsewhere. Copies of any possibly duplicative material should be
submitted with the manuscript that is being sent for consideration.

PAGE CHARGES

J,| ■•!

if

After acceptance of a paper by the Museum, the authors) will receive a statement cf page
charges from the Managing Editor, calculated at a current rate. Partial or complete payment
of this statement will be solicited front; those authors who have funds available for. this

purpose. Authors without access to
should so indicate on the .statement and.
statement only qualifies the paper, for i

publication are in no way dependent on. payntent of the statement.

!r; partial or complete payment of ?:
Tiro it id. the Managing, Editor
n: acceptance

charges

Hi

SERIAL PUBLICATIONS OF THE
'.NATURAL HISTORY MUSEUM OF LOS ANGELES COUNTY

issued ai irregular intervals in three major series; the articles in each series are numbered
individually, .and. numbers run .consecutively, regardless of the subject matter.

# Contributions in Science, a miscellaneous series of technical papers describing orig-
inal research in the life and earth sciences.

® Science Bulletin, a miscellaneous series of monographs describing original research
in the life and earth sciences. This series was discontinued in 1978 with the issue of
Numbers 29 and 30; monographs are now published by the Museum in Contributions

.in Science.

it Science Series, long articles on

Copies of the publications in these series are sold through the Museum Book Shop. A catalog

5 is available on request.

m i

OP

• ; :i j , j;|C 1 \:V. fjjj

Craig c: Black, Museum Director
Richard C. Brusca
Daniel M. Cohen, Committee Chairman
John M. Harris
Charles L. Hogue
George L. Kennedy
Robin A. Simpson,

Errol Stevens

ig Editor

ill

THREE NEW LUMINESCENT OSTRACODES OF THE GENUS
VARGULA (MYODOCOPIDA, CYPRIDINIDAE)

FROM THE SAN BLAS REGION OF PANAMA

Anne C. Cohen1 and Janies G. Morin2

ABSTRACT. Three new species of luminescent myodocopid ostra-
codes, Vargula graminicola, V. shulmanae, and V. contragula from
the San Bias Islands, Panama are described. V. graminicola and V.
shulmanae are sibling species differing in few morphological char-
acters but distinctive in diet, habitat, and bioluminescent patterns.
They are compared to Vargula parasitica (Wilson, 1913), a mor-
phologically similar Jamaican ostracode. V. contragula is morpho-
logically a very distinct species. All three new species produce dis-
tinctive bioluminescent patterns in the water column at night. V.
graminicola occurs within and above shallow marine seagrass beds,
V. shulmanae occurs primarily within and over steep slopes and
walls of deeper coral reefs, and V. contragula is found mainly on
gorgonian dominated shallow low profile coral reefs and slopes.

RESUMEN. Se describen tres nuevas especies de ostracodos mio-
docopidos luminiscentes, Vargula graminicola, V. shulmanae y V.
contragula de las islas de San Bias, Panama. V. graminicola y V.
shulmanae son especies gemelas que difieren en pocos caracteres
morfologicos pero que se diferencian claramente en su dieta, habitat
y patrones de bioluminiscencia. Estas especies son comparadas con
Vargula parasitica (Wilson, 1913), un ostracodo de Jamaica de si-
milar morfologia. V. contragula es una especie morfologicamente
muy diferente. Las tres nuevas especies producen patrones de bio-
luminiscencia distintos en la columna de agua durante la noche. V.
graminicola se encuentra en el interior y por encima de lechos de
praderas de yerbas marinas de poca profundidad, V. shulmanae esta
principalmente en el interior y sobre pendientes pronunciadas y
murallas de arrecifes de coral mas profundos, en tanto que V. con-
tragula habita principalmente en arrecifes y pendientes de poca pro-
fundidad dominados por corales gorgonicos.

INTRODUCTION

Many species within the ostracode genus Vargula Poulsen,
1 962, are known to be luminescent. Luminescent nocturnal
displays, presumably for purposes of sexual communication,
have been shown to occur within the genus in the U.S. Virgin
Islands (Morin and Bermingham, 1980). During nocturnal
surveys of the reefs and surrounding habitats of the San Bias
Islands, Panama, the junior author discovered a wide variety
of distinctive luminescent display patterns produced by os-

Contributions in Science, Number 373, pp. 1-23
Natural History Museum of Los Angeles County, 1986

tracodes that were closely associated with particular habitats.
From our collections, it has become clear that each lumi-
nescent pattern is produced by a distinctly different species
and that each species is associated with a particular habitat;
thus there is a great deal of resource partitioning among these
ostracodes. Furthermore, it appears that all of these species
are undescribed. This paper is the first of several in which
we describe these species and aspects of their biology. Papers
dealing with their population biology, activity patterns, and
bioluminescent displays are in preparation. Three obvious
and abundant luminescent Vargula species in the shallow
Caribbean waters of Panama are described here.

MATERIALS AND METHODS

Specimens were collected using a variety of methods but all
were caught at night when they are most active in the water
column and along the substrate surfaces. Specimens were
caught in the water column 1 ) indiscriminately using a diver-
pushed double plankton net ('/i m diameter, 0.5-mm mesh)
or, more often, by sweeps through the luminescent displays.
Diver-collected sweeps were done either 2) repeatedly using
a 20-cm diameter, 0.5-mm mesh net through the displays or
3) individually using a ‘discrete trap.’ A discrete trap con-
sisted of a double-walled triangular net (24 cm on a side and

0.5-mm mesh) sealed together on two sides and with Velcro
across the top so that it could be sealed or opened into a
cone. After passing the cone through a discrete luminescent
display the contents could be trapped inside by pressing the
Velcro together. In both sweep collection methods the lu-
minescent display type observed consistently correlated with
the expected species actually caught (see general biology sec-

1. Research Associate, Invertebrate Zoology, Natural History
Museum of Los Angeles County, 900 Exposition Blvd., Los Angeles,
California 90007.

2. Department of Biology, University of California, Los Angeles,
California 90024.

ISSN 0459-8113

c

/

G

/

2 Contributions in Science, Number 373

Cohen and Morin: Panamanian Myodocopid Ostracodes ( Vargula )

tions). Epibenthic collections were made by attracting ostra-
codes to 4) fish carcasses enclosed in plastic window screen
(to keep out most of the isopods and other larger organisms)
[=fish trap] or 5) ‘cone traps.’ These cone traps were baited
with dead fish and made from weighted plastic freezer con-
tainers (8x8x6 cm) with four 2-cm holes cut through on
the sides. Each hole was covered by window screen (1.7-mm
mesh) on the outside (to let in ostracodes but keep out iso-
pods) and a cone of 0.5-mm mesh plankton netting with a
2-mm inner orifice on the inside.

Details of the luminescent displays were obtained by direct
observations, timed analysis of underwater tape recordings
of divers comments, and laboratory observations. Details of
all these methods will be presented elsewhere (Morin, in
preparation). Ostracodes were usually preserved by 1) placing
them for 10-15 min in a mixture of 50% seawater and 50%
0.36 M MgCL (isotonic to seawater) [this mixture acted as
an anesthetic to relax but not kill the ostracodes], 2) trans-
ferring them to buffered 4% formaldehyde for 10-20 min for
fixation, and then 3) placing them in 70% ethanol (usually
they were transferred to a second 70% ethanol solution later).
Using a dissecting microscope, overall body measurements
were made to the nearest 0.02 mm on both living and pre-
served ostracodes (both were equal). All other measurements
were made using a compound microscope.

Holotypes and some paratypes are deposited in the United
States National Museum of Natural History (USNM), Smith-
sonian Institution, Washington, D.C. The remaining para-
types are deposited in the USNM and in the Natural History
Museum of Los Angeles County (LACM).

SYSTEMATICS

Vargula graminicola, new species

Figures 1, 2, 3A, 3B, 4A

HOLOTYPE

USNM 193214 adult male, length 1.71, height 1.06 mm; in
alcohol. Sample 19. 6 Sept. 1983 of about 1930 hr, '/t-2 m
above bottom at 5-m depth over grass bed just west of T-Bar

Reef, San Bias, Panama; taken by sweep net; Diana Pilson,
col.

ALLOTYPE

USNM 193215 gravid adult female, length 2.01, height 1.33
mm; in alcohol. Bottle 10 (data below).

PARATYPES

Sample 32. 8 Sept. 1983 at 2010-2020 hr, 1-2 m above
turtle grass at 5-m water depth ca. 10-15 m west of T-Bar
Reef, San Bias, Panama; sweep nets of lower part of 20 lu-
minescent displays; J.G. Morin, col.; 1 male (JM3-A, on slide
and in alcohol), 3 males (JM3-K, with choniostomatid co-
pepods), 91 males (JM3-C-E, J, L), 4 A — 2 males (JM3-F,
N) (LACM). Sample 19. 6 Sept. 1983 at about 1930 hr, ‘/i—
2 m above bottom at 5-m depth over grass bed just west of
T-Bar Reef, San Bias, Panama; taken by sweep net— part of
haul; Diana Pilson, col.; 1 12 males, 1 juv. (USNM 193218,
193219, 193220, 193221). Bottle 10. 31 Aug. 1983 at 1900-
2000 hr, ca. 4 m east of T-Bar Reef, San Bias, Panama, about
5-m depth; “fish trap” in sand blowout; J.G. Morin, col.; 1
female (JM10-10-T-X, with choniostomatid copepod), 1 fe-
male, 1 male (JM10-10-T-W), 10 females (JM10-10-T), 1
female, 1 ?A — 2 ovigerous female (JM10-10-T-Y) (LACM).
Bottle 9. 31 Aug. 1983 at 1 900-2000 hr, on east side of T-Bar
Reef, San Bias, Panama, about 3-m depth, “fish trap” among
corals on reef; J.G. Morin, col.; 1 male, 1 female (USNM
193216, 193217). Bottle 18. 6 Sept. 1983 at 2000 hr on east
side of T-Bar Reef, San Bias, Panama, “fish trap” on grass
bed; J.G. Morin, col.; 1 male, on SEM stub and in alcohol,
1 male on SEM stub, and 1 female on SEM stub (LACM).
JM Box 1#24. 12-13 Dec. 1983 between 2400 and 0200 hr
(combined from 1 collection each night), sweep nets lA-2 m
above turtle grass at 5-m depth ca. 100-200 m west of Vieja
Reef, San Bias, Panama; J.G. Morin, col.; 1 female (B, on
slide and in alcohol), 1 male (A) (LACM). JM Box 2#14-A.
12-13 Dec. 1983, data same as JM BOX 1 #24; 1 male (on
slide and in alcohol) (LACM). Sample 1210.1a. 12 Oct. 1984
at 1900-1915 hr, on grass beds northwest side of Reef 26
(Macaroon), San Bias, Panama, about 4-m water depth; bait-

Figure 1. Vargula graminicola. A. JM3-A, male, valve length 1.75 mm: right lateral view of complete specimen showing valve, lateral eye,
and tips of right mandible and furca. B, C. USNM 193216, male, length 1.71 mm, lateral views of complete specimen showing valve and
lateral eye: B. Left valve; C. Right valve. D. USNM 193219, male, length 1.71 mm, lateral view of complete specimen showing valve and
lateral eye. E. Holotype, USNM 193214, length 1.71 mm, left lateral view of complete specimen showing valve and lateral eye. F. USNM
193219, male, length 1.71 mm, lateral view of complete specimen showing valve and lateral eye. G. 2310.3-A, male, length 1.67 mm, medial
view of left 1st antenna (d-, e-bristles and distal parts of sensory, c-, f-, g-bristles not shown). H, I. JM3-A, male, length 1.75 mm, 2nd antenna:
H. Medial view of endopodite and first 2 joints of exopodite; I. Bristle of 2nd exopodite joint. J. USNM 193216, male, left lamella of furca.
K. JM Box 1-24B, female, length 2.01 mm, left lateral view of complete specimen, showing valve, lateral and medial eyes, furca, and caudal
list (dotted line). L, M. USNM 193217, female, length 2.01 mm, lateral views of complete specimen: L. Left valve and lateral eye; M. Right
valve, lateral eye, and eggs (dotted lines). N-S. JM Box 1-24B, female: N. Inside view of caudal process of left valve; O. Inside view of caudal
process of right valve; P. Tip of left mandible, medial view, showing 3rd and distal 2nd joints of endopodite; Q. Tip of right maxilla, lateral
view, showing 2nd and distal 1st joints of endopodite; R. Exopodite of right 5th limb, posterior view; S. Anterior of body and upper lip.

Contributions in Science, Number 373

Cohen and Morin: Panamanian Myodocopid Ostracodes ( Vargula) 3

Figure 2. Vargula graminicola. A-C. JM Box 1-24B, female, length 2.01 mm: A. Inside view of anterior end of right valve; B. Left mandible,
medial view; C. Tip of 7th limb. D, E. JM3-A, male, length 1.75 mm: D. 6th limb; E. 7th limb.

ed multiple cone trap; J.G. Morin, col.; 3 females (A-C)
(LACM). Sample 2310.3. 23 Oct. 1984 at 1830-1910 hr, on
grass beds ca. 40 m west of Reef 26, San Bias, Panama, about
5-m water depth; baited multiple cone trap; J.G. Morin, col.;
1 male (A), 5 males (B) (LACM).

DIAGNOSIS. Caudal process low (below incisur), and
shaped as triangular curved point. Female carapace length
1.80-2.25 mm; male 1.52-1.85 mm. Furca: 9 pairs of claws;
claws 2 and 4 fused to lamella; claw 4 broader than, but not
longer than claw 3. Second antenna: Bristle on 2nd endo-
podial joint with 6-8 (usually 7-8) stout ventral spines. Man-
dible: No terminal bristles with bulbous base. Maxilla: 2
alpha-, 3 beta-bristles, 4 a-, 3 b-, and 3 c-bristles. Fifth limb:
4th and 5th exopodial joints separated by suture; 4 bristles

on 4th joint, 2 bristles on 5th joint. Infold: Rostrum with
row of 1 9-3 3 bristles parallel to outer margin plus 0-2 bristles
anterior to and 4-12 bristles posterior to row; anteroventral
infold with row of 38-53 bristles plus 20-69 bristles posterior
to row. Seventh limb: Longest of 7 long teeth in comb lateral,
not central; with peg opposite comb; without dorsal jaw. Lip:
Tusks unbranched, with terminal and distal short hairs (spar-
ser in females).

ETYMOLOGY. Vargula graminicola is derived from the
Latin graminis which means “of or pertaining to grass” and
the Latin -icola which means “an inhabitant.” This species
inhabits seagrass bed habitats.

DESCRIPTION OF ADULT MALE. Carapace (Figs. 1 A-
F, 3A, 4A). Similar to that of adult female, only about 90%
as long but more elongate, with short but larger, more pointed

4 Contributions in Science, Number 373

Cohen and Morin: Panamanian Myodocopid Ostracodes ( Vargula)

Figure 3. Vargula graminicola, V. shulmanae, and V. contragula. A, B. Vargula graminicola. A. Bottle 18; 6 Sept. 1983, SEM-1A-179
specimen #3, male, right lateral view showing valve, furcae, and copulatory organ. B. Bottle 18; 6 Sept. 1983, SEM-1B-168 specimen #2,
female, left lateral view showing valve, proximal furcae, and distal setae of first antennae. C. Vargula shulmanae, 21 1 1.6-SEM-1C-237
specimen #6, male, right lateral view showing valve, furcae, and distal 1st antennae and mandibles. D. Vargula contragula, 2310. la-SEM-
3a- 197 specimen #2, male, left lateral view showing valve, furcae, distal 1st and 2nd antennae, and mandibles. Scale bar = 200 >im. r =
rostrum; i = incisur; c = caudal process; 1 = 1st antenna; 2 = 2nd antenna; m = mandible; co = copulatory organ; f = furca.

(shaped as triangular curved point), caudal process, and
straighter dorsal margin.

Infold. Infold of rostrum with about 22-33 double bristles
forming row parallel to rostral margin and continuing pos-
teriorly along incisur margin (about 3-8 along incisur mar-
gin), 0-2 double bristles anterior to row, about 4-12 double
bristles posterior to row, 1-2 double bristles near dorsal edge
of incisur and 1-3 tiny double bristles dorsal to inner edge
of incisur. Anteroventral infold with row of 4-5 double bris-
tles near inner edge of incisur, 1-2 short double bristles near
posterior edge of incisur infold, about 38-53 double bristles
paralleling margin and continuing onto ventral infold (in-
cluding 2-5 bristles on anteroventral incisur margin), about
20-69 double bristles posterior to row, row followed poste-
riorly on left valve by 3-8 widely spaced single bristles; list
becoming broader in vicinity of caudal process with posterior

Contributions in Science, Number 373

fringe (fringe usually with minute pointed processes), list with
10-38 tiny bristles (unclear, about 38 in left valve), possibly
with 1 tiny bristle posteroventral to list, about 10-12 tiny
bristles or projections (unclear in left valve) forming row
paralleling posterior edge of caudal process. Left valve with
caudal list ending dorsally in knob. Right valve similar to
that of female.

Selvage. Selvage with lamellar prolongation with smooth
edge and striations present along ventral and anterior mar-
gins of valves; lamellar prolongation along ventral margin of
incisur broader and with striations more visible than else-
where.

Size. USNM 193214 length 1.71, height 1.06 mm (holo-
tye). Range (n = 113) for all specimens measured: length
1.52-1.85 mm, height 0.91-1.10 mm.

First antenna (Fig. 1G). First joint bare; 2nd joint with

Cohen and Morin: Panamanian Myodocopid Ostracodes ( Vargula ) 5

Figure 4. Vargula graminicola and V. shulmanae. A. Vargula graminicola. Bottle 18; 6 Sept. 1983, SEM-2B-252 specimen #1, male, inside
view; left valve, 1st antenna and 5th limb removed. B. Vargula shulmanae, 21 1 1 .6-SEM-2C-233 specimen #1, male, inside view; left valve,
2nd antenna, mandible, maxilla, 5th and 6th limbs removed. Scale bars = 200 c = caudal process; 1 = 1st antenna; 2 = 2nd antenna;
m = mandible; 4 = 4th limb (maxilla); 7 = 7th limb; co = copulatory organ; f = furca; e = lateral eye; tul = tusk of upper lip; am = severed
ends of central adductor muscles.

spines forming medial rows and on dorsal margin. Third joint
short, ventral margin half or less than half length of dorsal
margin, with medial spines forming row, 1 dorsal spinous
bristle just proximal to middle and 1 long spinous terminal
medial ventral bristle; 4th joint with 1 spinous terminal or
subterminal dorsal bristle and 1 spinous distal or terminal
ventral medial bristle. Sensory bristle of 5th joint with 10
long proximal filaments, 2 more slender distal filaments and
bifurcate tip, bifurcate tip and last filament with narrowed
tips. Sixth joint with short spinous medial bristle near distal
margin. Seventh joint: a-bristle spinous, longer than bristle
of 6th joint; b-bristle with stout proximal filament with bul-
bous base followed by large sucker, short filament with nar-
rowed tip, and small distal process ending in narrowed point-
ed tip, 2 distal slender filaments each bearing 5-6 small suckers
(distal filament not extending beyond tip of bristle); c-bristle
similar to b-bristle but about twice as long, and with larger
basal filament and sucker; bristle with 2 distal filaments with
4-5 small suckers, about 7 long slender bare distal filaments,
and bifurcate tip. Eighth joints: d- and e-bristles bare, fila-
mentous, about same length as b-bristle, tips blunt or with
minute process; f-bristle with 8-9 slender filaments increas-
ing in length distally, some with 1-3 spines; g-bristle with 10
slender filaments increasing in length distally and with 1 spine
each (except last filament), bifurcate tip; c-, f-, and g-bristles
long (g-bristle longest), longer than sensory bristle of 5th
joint.

Second antenna (Figs. 1H, I, 4A). Protopodite with short
spinous distal medial bristle. Endopodite 3-jointed: 1st joint
with 4 proximal bristles (1 short, 3 very short) and 1 short
distal spinous bristle; 2nd joint elongate with 1 short terminal
bristle (missing on 1 limb examined), 3rd joint about V2-V1
length of 2nd joint with long terminal filament (Fig. 1H).
Exopodite: 1st joint with spines on dorsal margin; 2nd joint

6 Contributions in Science, Number 373

with rows of medial and lateral spines crossing ventral mar-
gin, bristle not reaching or almost reaching end of 9th joint,
with 6-7 stout ventral spines (Fig. 1H, I); joints 3-8 with
basal spines becoming longer distally and with bristles with
natatory hairs; lateral spine of 9th joint slightly larger than
spine of 8th joint; 9th joint with 4 bristles (2 longest with
natatory hairs, 1 not as long with natatory hairs and short
spines, 1 shorter with short spines).

Mandible (Fig. 4A). Coxale endite spinous, with bristle
near base. Basale: Ventral margin with 1 short and 1 long
spinous a-bristle with bases on medial side, 1 short b-bristle
with base on lateral side (close to a-bristles), 1 long spinous
and 1 short c-bristle near middle of margin, and 2 distal
d-bristles, both some distance from c-bristles (1 shorter with
short spines and 1 very long with long proximal and short
distal spines); dorsal margin with 1 long spinous distal bristle
with short spines and 2 long unequal subterminal bristles
with short spines. Exopodite with medial hirsute dorsal pro-
cess extending to pointed tip and 2 long ventral bristles with
short spines (distal shorter). First endopodial joint with 4
ventral bristles (1 long with long proximal and short distal
spines, 1 long with short spines, 2 short with short spines).
Second endopodial joint: Medial and lateral rows of spines
or hairs, dorsal margin with 16-18 bristles (5-6 long spinous
bristles, 5 shorter bristles with short spines or hairs, 4-6 short
hirsute medial bristles, 2 short stout medial bristles with stout
spines or possibly hairs); ventral margin with 2 single distal
bristles (bristles with bases together on 1 limb of 1 specimen)
and 1 subterminal bristle equal in length to and with base
on lateral side of base of stout finger-like subterminal un-
ringed process (probably modified bristle) [tips either bifur-
cate or hollow and middle bristle extending beyond base of
subterminal bristle]; process not touching 3rd joint and with
slender rounded point usually extending beyond or to end of

Cohen and Morin: Panamanian Myodocopid Ostracodes ( Vargula)

3rd joint. End joint with 3 long claws with proximal spines
and 4 bristles, none with bulbous base and none longer than
claws (1 dorsolateral bare and slightly ringed claw-like bristle
extending almost to end of dorsal claw, 1 prominent short
bare ventral bristle, 1 long [34 claw length] spinous ventral
medial bristle with bifurcate tip, 1 long stout, but not bul-
bous, spinous lateral bristle extending to tip of claws).

Maxilla. Endite I with 1 2 stout spinous claw-like bristles;
endite II with 4 spinous bristles (3 long stout), endite III with
4-5 spinous bristles (4 stout, long). Coxale with stout dorsal
bristle proximally hirsute and with short distal spines. Basale
with 1 long spinous lateral bristle, 1 long spinous medial
bristle on or near ventral margin and base of exopodite and

1 spinous terminal medial bristle (may be on 1st endopodite
joint). Exopodite with 1 long proximal bristle with long prox-
imal hairs and short distal spines and 2 long terminal bristles
(1 with long proximal hairs and short distal spines, 1 with
short spines only). First endopodial joint with rows of medial
hairs, 2 alpha-bristles (outer bristle longer with long hairs,
inner bristle with short spines), 3 beta-bristles (outer longest
and pectinate, inner 2 with short spines); cutting tooth large
and prominent, bifid (proximal projection with rounded point,
distal projection more than twice as large with rounded
squarish shape); additional short rounded tooth on terminal
lateral margin near ventral margin. Second endopodial joint
with 4 long distally ringed a-bristles (2 longest with proximal
short spines); 3 long, stout, very pectinate claw-like b-bristles
(1 unringed, 2 ringed); 3 distally ringed c-bristles (2 very
pectinate, 1 short); 3 stout, very pectinate d-bristles ( 1 ringed).

Fifth limb. Epipod with 42-48 bristles. Protopodite with
large distal undulate anterior tooth and 1 long anterior bristle
with long proximal and short distal spines. Endite I with 7
stout unequal bristles (6 with long spines, 1 short with shorter
spines); endite II with 5-6 bristles (5 stout with long spines
including 2 distally pectinate); endite III with 7 unequal bris-
tles (most stout with long spines, 5 distally pectinate). First
exopodial joint with 6 pectinate teeth and 1 proximal peg
(smooth except for tiny spines on tip), 1 large spinous bristle
near peg, row of 4 anterior bristles with long proximal spines
or hairs ( 1 of 2 short proximal bristles may be on protopodite,

2 distal bristles longer, stouter); 2nd exopodial joint with 10-
1 1 long stout pectinate claw-like end bristles arranged some-
what into 3 rows (4 bristles unringed, claw-like), and 1 pos-
terior bristle with long proximal hairs and long distal spines.
Inner lobe of 3rd exopodial joint hirsute, with 3 bristles (1
proximal posterior with long proximal hairs and short distal
spines and 2 terminal with short spines); outer lobe hirsute
with 2 terminal bristles with short spines ( 1 with long prox-
imal hairs or spines). Fourth and 5th joints hirsute, separated
by faint but distinct suture, 4th joint with 4 subterminal
bristles with short spines, 5th joint with 2 terminal bristles
with short spines and terminal group of spines (5th joint with
2 lobes, each with 1 bristle in JM2-C, but only 1 lobe in
JM3-A).

Sixth limb (Fig. 2D). Hirsute laterally, 5 bare bristles in
place of epipod. Endite I with 3 bristles (2 short with long
hairs, 1 long with longer hairs); endite II with 5 bristles (3
short with long hairs, 2 long with long proximal hairs and

Contributions in Science, Number 373

short distal spines); endite III with 4 bristles with long prox-
imal hairs and short distal spines (2-3 long, 1-2 short); endite
IV with 3 bristles with long proximal hairs (2 long with short
distal spines, 1 short). End joint spinous, with 6-7 spinous
bristles decreasing in length posteriorly, each with long stout
distal spine-like hairs ( 1 more medial with shorter proximal
spines), followed by space and 3 hirsute posterior bristles
increasing in length posteriorly (anterior of these with stout
distal spines). Limb with suture separating end joint from
proximal part of limb, partial suture at base of endites III
and IV (sometimes II).

Seventh limb (Figs. 2E, 4A). Comb side with 1-2 distal
and 3-4 terminal bristles, each with 2-4 bells; peg side with
3 distal and 3 terminal bristles, each with 2-3 bells. Comb
consisting of 7 long spinous teeth with widened tips (2 lateral
and 1 middle teeth shortest) and 4 short blunt teeth (2 on
each side) with long basal spines and possibly with bumpy
tips. Single procumbent short peg (slightly shorter than short
comb teeth) with hook-like inner basal protuberance and
terminal wreath of about 8 tiny teeth; terminal surface be-
tween comb and peg possibly ridged at base of comb.

Furca (Figs. 1J, 3A, 4A). Each lamella with 9 claws; claws
2 and 4 fused to lamella, remaining claws separated from
lamella by suture; claw 3 more slender than but about same
length as claw 4; claws 1 and 2 with long row of posterolateral
(a few proximal teeth longer on claw 2) and distal row of
medial teeth; all remaining claws with long row of postero-
lateral teeth and at least claws 3-5 with medial teeth.

Bellonci organ. Short, cylindrical, rounded.

Eyes. Medial eye small, pigmented (sometimes restricted
to band). Lateral eye about twice size of medial eye, pig-
mented, with about 1 2 ommatidia. Lateral eye length ranges
in size from 0.23 to 0.30 mm and pigmented area from 0.20
to 0.25 mm (n = 113). Eye pigment brown in reflected light;
maroon or maroon-brown in transmitted light.

Upper lip. Anterior undivided part with numerous glan-
dular processes with unpigmented lobular tips and proximal
bands of maroon pigment; middle part with 2 long tusks, 1
on each side; each tusk with short distal and terminal hairs
and with glandular processes ( 1 terminal and 3-6 posterior;
processes sometimes appearing to have pointed tips); part
posterior to tusks, rounded, hirsute (proximal hairs bunched).

Anterior of body. Rounded projection dorsal to 2-3 shorter
usually pointed projections.

Posterior of body. Smooth.

Copulatory organ (Fig. 3A). Anterior lobe rather conical,
longer than short posterior lobe, which has at least 1 group
of about 3 bristles.

DESCRIPTION OF ADULT FEMALE. Carapace (Figs.
1K-M, 3B). Oval with deep incisur and protruding caudal
process; anteroventral, posteroventral, posterodorsal, and
anterodorsal margins broadly rounded. Greatest height near
middle; not markedly higher in posterior half. Caudal process
small (slightly larger in right valve than in left), dorsal margin
rather straight, forming obtuse angle with rather truncate
posterior margin of valve, dorsal edge of process slightly
shorter than dorsal edge of incisur; both incisur and dorsal

Cohen and Morin: Panamanian Myodocopid Ostracodes ( Vargula ) 7

edge of process at about midheight of valve. Dorsal edge of
incisur slightly overlapping ventral edge at inner end; faint
line on outer surface of valve curving from inner dorsal edge
of incisur to anterior margin of valve ventral to incisur. Tip
of rostrum with few nodes. Outer surface smooth but with
faint pattern resembling overlapping scales, visible at 100x;
scale-like pattern producing minute points on anteroventral
and posteroventral margin. Surface with rather regularly dis-
tributed small pores, some with minute bristles.

Infold (Figs. IN, O, 2A). Similar to that of adult male
except rostrum (Fig. 2 A) with about 7-10 double bristles
posterior to main row of 19-30 double bristles, 0-1 bristle
anterior to row; anteroventral infold with row of about 38-
5 1 double bristles parallel to margin (including about 2-4 on
anteroventral margin of incisur) about 34-64 double bristles
posterior to row; 0-3 bristles posteroventral to list near be-
ginning of caudal process, list of caudal process in left valve
(Fig. 1 N) with numerous tiny bristles and projections forming
irregular row and posterior margin with tiny ripples, caudal
list straight except for slight bend in dorsal half, caudal list
of left valve terminating dorsally with round knob and pos-
sibly with a rounded dorsally directed process with posterior
groove, dorsal and posterior to knob; caudal list of right valve
(Fig. lO) with more than 50 tiny bristles, sometimes with
minute pointed processes or fringe on ventral posterior edge,
caudal list rather straight, ending in raised dorsally directed
bar; small pocket formed in infold posterior to dorsal end of
caudal list, pocket may be half socket (open ventrally) re-
ceiving knob from left valve.

Selvage. Similar to male.

Adductor muscle scars. About 13 irregular scars, some
divided.

Size. USNM 193215 length 2.01, height 1.33 mm (allo-
type). Range (n = 24) for all specimens measured: length
1.80-2.25 mm, height 1.18-1.43 mm.

First antenna. Similar to male except b-bristle of 7th joint
about V1-V2 longer than a-bristle, with 2 short proximal fil-
aments; c-bristle with about 8 filaments (with teeth or spines)
becoming longer distally, about 6 times length of b-bristle;
8th joint with distal filament of f- and g-bristles much longer
than others, small round knob at medial dorsal base of f-bris-
tle.

Second antenna. Similar to male except exopodite with
bristle of 2nd joint not reaching 9th joint, with 7-8 spines
and with narrowed ringed tip.

Mandible (Figs. IP, 2B). Similar to male except bristle
apparent on coxale endite of 1 limb but not other limb of
specimen examined; 2nd endopodite joint with 15-16 dorsal
bristles, only 1 short medial bristle bearing stout spines; ven-
tral margin of 2nd joint (Fig. IP) with 2 single distal bristles
and 1 subterminal bristle together with subterminal unringed
finger-like sclerotized process sometimes extending beyond
3rd joint; 3rd endopodite joint with 1 of 4 bristles (stout
ventral lateral) extending just beyond claws (Fig. IP).

Maxilla (Fig. IQ). Similar to adult male.

Fifth limb (Fig. 1 R). Similar to adult male except proximal
protopodite (coxale?) with pair of small sclerotized teeth
bearing minute teeth or spines; both exopodite joints 1 and

8 Contributions in Science, Number 373

2 appearing to arise independently from protopodite of spec-
imen examined; joint 1 has a more medial base and could
be considered an endopodite; joint 3 arises from joint 2 and
joints 2-5 could be considered joints 1-4 of an exopodite.

Sixth limb. Similar to adult male.

Seventh limb (Fig. 2C). Similar to adult male except comb
side with 4-5 distal and 5 terminal bristles, peg side with 4—
5 distal and 3 terminal bristles. Comb consisting of 7 long
teeth (similar to those in male) and 6 short blunt teeth (3 on
each side). Peg similar to that of male but about same length
as short comb teeth.

Furca. Similar to that of adult male.

Bellonci organ. Similar to that of adult male but with blunt
end.

Eyes. Similar to those of adult male but lateral eye about
1 V2 times as large as medial eye. Overall size and ommatidia
size of lateral eye smaller than in male. Length ranged from
0.21 to 0.26 mm and pigmented area from 0. 1 6 to 0.2 1 mm
(n = 24).

Upper lip. Lip similar to that of adult male except tusk
with very faint short hairs (Fig. IS).

Anterior of body. Anterior of body with rounded projec-
tion dorsal to 3-5 shorter usually pointed projections (Fig.
IS).

Posterior of body. With a few short hairs.

Genitalia. Sclerotized ring with attached spermatophores.

Eggs. JM10-10T-Z with 20 eggs in ovary; the largest egg
was 0.22 mm. 1210.1a, 3 females with 15-18 eggs in ovary.
Other females had smaller eggs.

Ovigerous juvenile. One small ovigerous female (JM10-
1 0-T-1) bears about 18 eggs/ovary (eggs = 0.023-0.027 mm).
It has the diagnostic characters shared by V. graminicola and
V. shulmanae (valve shape; furcal claws; distal bristles of
mandible, maxilla, and 5th limb; 7th limb comb). It has the
specific diagnostic characters of V. graminicola: 6 (lowest
adult number) stout spines on the 2nd exopodite bristle of
the 2nd antenna and rostral infold with 4-6 bristles posterior
to row of 2 1-22 bristles. We identify this female as a juvenile,
probably an A— 1 instar, because it has these juvenile char-
acters: length only 1.22 mm, height 0.80 mm; 7th limb with
only 2 bristles, both terminal, strongly tapered and with only
1 bell; furca with only 7 claws; anteroventral infold with only
14-16 bristles (this high number also shows affinity with V.
graminicola) posterior to row of 27-30 bristles.

GENERAL BIOLOGY. Geographical distribution. Carib-
bean; known from the vicinity of the western San Bias Is-
lands, Panama (9°33'14"N, 78°55'23"W).

Habitat. Vargula graminicola is abundant in shallow (3-
10 m) sea grass ( Thalassia testudinum and Syringodium fil-
iforme ) beds with good water circulation and away from coral
reefs and sand ‘blowouts.’

Ecology, behavior, and bioluminescence. Vargula gramini-
cola is a benthic species that in Panama lives in current-
swept sea grass beds within the turf, rubble, and top few mm
of the sand. Individuals are infaunal by day and epibenthic
and demersal by night. Sex ratios appear to be about 1:1
below the level of the top of the sea grass (ca. 1 5 cm) and in
the substrate, but above the sea grass highly skewed at night

Cohen and Morin: Panamanian Myodocopid Ostracodes ( Vargula )

toward males. Males become demersal zooplankters above
the sea grass at night if a bright moon is not present; females
are only occasionally found up in the water column. The
males appear in the water column in large numbers about
45 min after sunset and remain there most of the night as
long as there is no moonlight. Much of the time they are
found at the level of the sea grass, but, periodically, they rise
obliquely or vertically upward (ca. 50-90°) in linear clusters,
each time producing a species specific group luminescent
display. They swim at a rate of about 11.1 ± 1.4(sd) cm s '
(n = 3 1 laboratory trials of individuals). Each group may
contain from 2 to 40 males. The light occurs as a train of
short duration extracellular secretions left behind by each
rapidly swimming male. Each light pulse, which appears as
a discrete point source, has a duration of 0.28 ± 0.08 s (n =
109). Each cluster of pulses is spaced about 1 5 to 25 cm from
the next; the spacing is even for a given train. There are
usually 8 to 1 2 pulse clusters per train, each train takes about
1 8 s to produce and is about 1 .5 to 2 m in total length. Trains
are repeated at about 60-s intervals. There is loose synchrony
of different trains within a large area of the sea grass (>50
m2) and clusters of displaying males are horizontally sepa-
rated by 1 5 cm to about 1 m or more. These luminescent
displays apparently act as calling signals by the upwardly
swimming males to the sexually receptive epibenthic females
who then swim up into the water column. Presumably the
females copulate with at least one of the males. Details of
mating are unknown. The females are capable of luminescing,
but apparently do not do so during these nocturnal displays.
The displays continue for most of the night until about an
hour before dawn. The frequency of the displays, however,
is most intense during the first hour or two.

It is not known what V. graminicola feeds on naturally,
but they are the only Vargula species in the San Bias strongly
attracted to fish or crustacean carrion. Along with about equal
numbers of Skogsbergia lerneri as many as 10,000 V. gra-
minicola (males, females, and juveniles) have been captured
from a single fish carcass in less than 1 hr at night. They have
been observed to be preyed upon by demersal fishes (in-
cluding Holocentrus rufus) and a cerianthid ( lArachnanthus
nocturnus) and an actiniarian (P.Aiptasia tagetes) anthozoan;
all cases of predation initiated a long-lasting ( > 20 s) lumi-
nescent glow from the ostracode. Papers detailing the lu-
minescent display patterns (Morin, in press) and the popu-
lation biology of V. graminicola are in preparation.

Color of live ostracodes. Males are similar in color to fe-
males and are mostly transparent except for the eyes (see
above), the region of the luminescent organs (=light organs)
and the abdomen. The light organ is a yellow-brown rect-
angular region in the upper lip below the eye. The abdomen
is an orangish-brown to tan color. This color may be affected
by diet since they become paler in captivity when fed solely
on white muscle from fish.

Parasites. Four ostracodes contained copepod parasites
belonging to the Choniostomatidae (identification verified
by T. Bowman). One female (JM10-10-T-X) with 20 eggs in
its ovary had 1 female choniostomatid loosely attached to
the exterior posterodorsal part of the body within the mar-

Contributions in Science, Number 373

supium area of the valves. Three males (JM3-K) each bore
a male choniostomatid loose within the anterior part of the
valve near the base of the 1st and 2nd antennae. One or 2
of these also had a female attached to the dorsal body just
posterior to the heart. Choniostomatid parasites have been
reported previously in cypridinid ostracodes including a
species of Vargula (Komicker, 1975; Bradford, 1975).

Vargula shulmanae, new species

Figures 3C, 4B, 5, 6

HOLOTYPE

USNM 193222 adult male, length 1.94, height 1.25 mm; in
alcohol. Sample 1212.12. 12 Dec. 1984 at 1840-1940 hr,
about 12-14-m depth on reef wall. Sail Rock, San Bias, Pan-
ama; discrete traps of displays; J.G. Morin, col.

ALLOTYPE

USNM 193223 adult female, length 2.35, height 1.65 mm;
in alcohol. Sample 2111.5 (data below).

PARATYPES

Sample 2510.3. 25 Oct. 1984 at 1845-1905 hr, about 4-m
depth on reef wall. Sail Rock, San Bias, Panama; sweep nets
of displays; J.G. Morin, col.; 1 male (A, on slide and in
alcohol), 46 males (C-T) (LACM). Sample 2510.1a. Data as
in Sample 25 10.3; J.G. Morin, col.; 1 female (C, on slide and
in alcohol) (LACM). Sample 2011.2. 20 Nov. 1984 at 1840-
1910 hr, about 6-9-m depth on reef slope, Korbiski Reef,
San Bias, Panama; discrete traps of displays; J.G. Morin,
col.; 2 males (A, B) (LACM). Sample 2111.2. 21 Nov. 1984
at 1838-1902 hr, about 11-15-m depth on reef wall. Sail
Rock, San Bias, Panama; discrete traps of displays; J.G. Mo-
rin, col.; 52 males (A-F) (USNM 193224, 193225, 193226,
193227, 193228, 193229), 3 juv. (G)(USNM 193233). Sam-
ple 2111.3. Data as in Sample 2111.2; J.G. Morin, col.; 1
female (LACM). Sample 2111.5. Data as in Sample 21 1 1.2;
J.G. Morin, col.; 2 juv. (B), 1 ?A— 1 ovigerous female (C)
(LACM). Sample 2111.6. Data as in Sample 2111.2; J.G.
Morin, col.; 1 male, on SEM stub and in alcohol, and 1 male
on SEM stub (LACM). Sample 2311.5. 23 Nov. 1984 at

1842- 1920 hr, about 8-10 m depth on reef slope, Korbiski
Reef, San Bias, Panama; discrete traps of displays; J.G. Mo-
rin, col.; 2 males (A, B), 1 juv. (LACM). Sample 231 1.8. Data
as in Sample 231 1.5; J.G. Morin, col.; 2 males (A, B, with
choniostomatid copepods), 2 males (C, D) (LACM). Sample
1212.9. 12 Dec. 1984 at 1840-1940 hr, about 12-14mdepth
on reef wall. Sail Rock, San Bias, Panama; discrete traps of
displays; J.G. Morin, col.; 2 males (LACM). Sample 1212.12.
Data as in Sample 1212.9; J.G. Morin, col.; 1 male (C), 1
female (A, lost?) (LACM). Sample 1312.1. 13 Dec. 1984 at

1843- 1910 hr, about 4-m depth on low reef. Reef 26 (Ma-
caroon), San Bias, Panama; discrete trap of display; J.G.
Morin, col.; 1 male (B) (LACM).

DIAGNOSIS. Caudal process low (below incisur) and
shaped as triangular curved point. Female carapace length

Cohen and Morin: Panamanian Myodocopid Ostracodes ( Vargula) 9

Figure 5. Vargula shulmanae. A, B. Holotype, USNM 193222, male, length 1.94 mm, right and left lateral views of whole specimen showing
valves, lateral eyes, and caudal infold (dashed line). C. 1312.1 -B, male, length 1.94 mm, left 1st antenna, medial view (distal parts of sensory,
c-, f-, g-bnstles not shown). D. USNM 193223, female (allotype), length 2.35 mm, left view of whole specimen showing valve, lateral eye.

10 Contributions in Science, Number 373

Cohen and Morin: Panamanian Myodocopid Ostracodes ( Vargula)

2.33-2.55 mm; male 1.79-2.01 mm. Furca: 8-9 pairs of
claws; claws 2 and 4 fused to lamella; claw 4 broader than,
but not longer than claw 3. Second antenna: Bristle on 2nd
exopodite joint with only 4-6 proximal and 1 subterminal
stout ventral spines. Mandible: No terminal bristles with
bulbous base. Maxilla: 2 alpha-, 3 beta-, 4 a-, 3 b-, and 3
c-bristles. Fifth limb: 4th and 5th joints separated by suture;
4 bristles on 4th joint, 2 bristles on 5th joint. Infold: Rostrum
with row of 20-33 bristles parallel to margin plus 0-2 bristles
anterior to and 1-6 bristles posterior to row; antero ventral
infold with row of 32-43 bristles plus 0-1 bristle anterior
and 0-1 7 bristles posterior to row. Seventh limb: Males with
only 1-3 proximal bristles on comb side, total of 10-16 bris-
tles on limb; females with 3-4 proximal bristles on comb
side, total of 13-18 bristles; longest of 7 long teeth in comb
lateral, not central; with peg opposite comb; without dorsal
jaw. Lip: Tusks unbranched; with short distal and terminal
hairs.

ETYMOLOGY. Vargula shulmanae is named after Dr.
Myra J. Shulman who, along with one of us (J.G.M.), first
collected the species.

DESCRIPTION OF ADULT MALE. Carapace (Figs. 3C,
4B, 5A, B). Oval with deep incisur and protruding caudal
process; anteroventral, posteroventral, posterodorsal, and
anterodorsal margins broadly rounded. Greatest height near
middle; not markedly higher in posterior half. Caudal process
short, below midheight of valve, but projecting to rounded
point, process slightly larger in right valve than in left, dorsal
margin rather straight, forming obtuse angle with rather trun-
cate posterior margin of valve, dorsal edge of process slightly
shorter than dorsal edge of incisur; both incisur and dorsal
edge of process at about midheight of valve. Dorsal edge of
incisur slightly overlapping ventral edge at inner end; faint
line on outer surface of valve curving from inner dorsal edge
of incisur to anterior margin of valve ventral to incisur. Tip
of rostrum with few nodes. Outer surface smooth but with
faint pattern resembling overlapping scales, visible at 100 x ;
scale-like pattern producing minute points on anteroventral
and posteroventral margin. Surface with rather regularly dis-
tributed small pores, some with minute bristles.

Infold. Infold of rostrum with about 20-31 double bristles
forming row parallel to rostral margin and continuing pos-
teriorly along incisur margin (about 5 bristles along incisur
margin), 0-1 bristle anterior to row, 1-4 double bristles pos-
terior to row, 2 double bristles near inner edge of incisur.
Anteroventral infold with 3 short bristles near inner edge of
incisur; about 20-3 1 double bristles forming row parallel to
margin (about 3-4 of these anteroventral to incisur), 0-1
double bristle anterior to and 1-4 double bristles posterior
to row; about 7 widely spaced bristles on ventral margin.
List becoming broader in vicinity of caudal process with
posterior fringe of minute processes, bristles or pores (right

valve with posterior fringe of irregular minute triangular spine-
like processes); ventral % of left list broader, with 0-1 bristle
anteroventral to list, list ending in dorsal knob; right list
ending in dorsal bar.

Selvage. Typical for genus.

Size. USNM 193222 length 1.94, height 1.25 mm (holo-
type). Range (n = 131) for all specimens measured: length
1.79-2.01 mm, height 1.14-1.28 mm.

First antenna (Figs. 4B, 5C). First joint bare; 2nd joint
with medial spines forming rows and spines on ventral and
dorsal margins. Third joint short, ventral margin half or more
than half length of dorsal margin, with medial spines forming
row and 1 dorsal spinous bristle just proximal to middle and

1 long spinous ventral bristle near middle of joint; 4th joint
with 1 spinous terminal dorsal bristle and 1 spinous distal
ventral medial bristle. Sensory bristle of 5th joint with 10
long proximal filaments (some with minute filaments or
spines), 2 more slender distal filaments and bifurcate tip,
bifurcate tip and last filament with narrowed tips. Sixth joint
with short spinous medial bristle near dorsal distal margin.
Seventh joint: a-bristle spinous, slightly longer than bristle
of 6th joint; b-brist!e with stout proximal filament with bul-
bous base followed by large sucker and a small distal process
ending in a narrowed pointed tip, 2 distal slender filaments
each bearing 5 small suckers (distal filament not extending
beyond tip of bristle); c-bristle similar to b-bristle but about

2 times as long, basal filament and sucker larger, 2 distal
filaments with 4 small suckers, with about 5-6 long slender
bare distal filaments, bifurcate tip. Eighth joint: d- and e-bris-
tles bare, filamentous, almost as long as b-bristle, tips blunt
or with narrowed process; f-bristle with about 7 slender fil-
aments increasing in length distally, some perhaps with 1-3
spines; c-, f-, and g-bristles long (g-bristle longest), longer
than sensory bristle of 5th joint; g-bristle with 9 slender fil-
aments increasing in length distally and with 1 spine each
(except last filament) and process on bifurcate tip.

Second antenna. Protopodite with short spinous distal me-
dial bristle. Endopodite 3-jointed: 1st joint with 4 proximal
bristles (1 short, 3 very short) and 1 short distal spinous
bristle; 2nd joint elongate with 1 short spinous subterminal
bristle, 3rd joint about Vi-lh length of 2nd joint with long
terminal filament. Exopodite: 1st joint with spines on dorsal
margin; 2nd joint with rows of medial and lateral spines
crossing ventral margin, bristle reaching 6th or 7th joint, with
4 proximal and 1 subterminal stout ventral spine and nar-
rowed tip; joints 3-8 with basal spines becoming longer dis-
tally and with bristles with natatory hairs; lateral spine of
9th joint about same length as spine of 8th joint and about
twice length of 9th joint; 9th joint with 4 bristles (3 long with
natatory hairs, 1 shorter apparently bare).

Mandible (Figs. 3C, 4B). Coxale endite spinous, with bris-
tle near base. Basale: Ventral margin with 1 short and 1 long

and muscle scars. E-H. 2510.1a-C, female, length 2.39 mm: E. Right view of whole specimen showing valve, lateral eye, eggs (dashed lines),
tip of 1st antenna, mandible, and furca; F. Inside view of anterior end of right valve; G. Inside view of rostrum of right valve; H. 2nd antenna,
medial view of endopodite, exopodite, and distal protopodite (only bases of bristles on exopodial joints 3-9 shown).

Contributions in Science, Number 373

Cohen and Morin: Panamanian Myodocopid Ostracodes ( Vargula ) 1 1

spinous a-bristle with bases on medial side, 1 short b-bristle
with base on lateral side (close to a-bristles), 1 long spinous
and 1 short c-bristle near middle of margin, and 2 distal
d-bristles, both some distance from c-bristles (1 shorter with
short spines and 1 very long with groups of long proximal
spines and short distal spines); dorsal margin with 1 long
spinous distal bristle with short spines and 2 long unequal
subterminal bristles with short spines. Exopodite with medial
hirsute dorsal process extending to pointed tip and 2 long
ventral bristles with short spines (distal shorter). First en-
dopodial joint with 4 ventral bristles ( 1 long with long prox-
imal and short distal spines, 1 long with short spines, 1 short
with short spines and 1 minute bare). Second endopodial
joint: Medial and lateral rows of spines or hairs, dorsal mar-
gin with 1 7 bristles (5 long spinous bristles, 5 shorter hirsute
bristles, 6 short hirsute medial bristles, 1 short stout medial
bristle with stout spines or possibly hairs); ventral margin
with 2 medium length single distal bristles (bristles with bases
together on 1 specimen) [tips either bifurcate or hollow and
middle bristle usually extending beyond base of subterminal
bristle and finger-like process] and 1 subterminal bristle equal
in length to and with base on lateral side of base of stout
finger-like subterminal unringed process (probably modified
bristle); process not touching 3rd joint and with slender
rounded point extending almost to end of 3rd joint. End joint
with 3 long claws (1 with proximal spines) and 4 bristles,
none with a bulbous base but stoutest reaching just beyond
claws ( 1 dorsal lateral bare and slightly ringed claw-like bris-
tle extending 2A-3/4 length of dorsal claw, 1 prominent short
bare ventral bristle, 1 long [% claw length] spinous ventral
medial bristle, 1 long stout, but not bulbous, spinous lateral
bristle extending just beyond tip of claws).

Maxilla. Endite I with 1 1 stout bristles with long spines;
endite II with 5 bristles with long spines (4 long stout), endite
III with 5 spinous bristles (4 stout, long with long spines, 1
proximal short slender with short spines). Coxale with stout
dorsal bristle proximally hirsute and distally with short spines.
Basale with 1 long spinous lateral bristle, 1 long spinous
medial bristle on or near ventral margin and base of exo-
podite and 1 spinous terminal medial bristle. Exopodite with
1 long proximal bristle with long proximal hairs and short
distal spines and 2 long terminal bristles (1 with long prox-
imal hairs and short distal spines, 1 with short spines only).
First endopodial joint with rows of medial hairs, 2 alpha-
bristles (outer bristle longer with long hairs, inner bristle with
short spines), 3 beta-bristles (outer longest and pectinate, 1
of inner 2 with short spines); cutting tooth rather large and
prominent, bifid (short slender proximal projection with
rounded point, distal square molar-like projection more than
twice as large with 2 points); possibly with 2 additional short
rounded teeth on terminal lateral margin (1 near ventral
margin, 1 near base of beta-bristles). Second endopodial joint
with 4 long distally ringed a-bristles ( 1 with proximal short
spines); 3 long, stout, very pectinate claw-like b-bristles (2
ringed, 1 slightly ringed); 3 distally ringed c-bristles (2 very
pectinate, 1 short); 3 stout, very pectinate d-bristles (2 ringed).

Fifth limb. Epipod with at least 36 bristles. Protopodite
with large distal undulate anterior tooth and 1 long anterior

12 Contributions in Science, Number 373

bristle with long proximal and short distal spines. Endite I
with 6 stout unequal bristles with long spines; endite II with
5 stout bristles with long spines (1 distally pectinate); endite
III with 6 unequal bristles (most stout with long spines, 4
distally pectinate). First exopodial joint with 6 pectinate teeth
and 1 proximal peg (smooth except for tiny spines on tip),

1 large pectinate bristle (with proximal spines) near peg, row
of 3 anterior bristles with long proximal spines or hairs (2
long stout distally pectinate, 1 short); 2nd exopodial joint
with 1 1 long stout pectinate claw-like end bristles arranged
somewhat into 3 rows of 4 (4 bristles unringed, claw-like); 1
posterior bristle with long hairs. Inner lobe of 3rd exopodial
joint hirsute, with 3 bristles (1 proximal posterior with long
proximal hairs and short distal spines and 2 terminal with
short spines); outer lobe hirsute with 2 terminal bristles with
short spines (1 with long proximal hairs or spines). Fourth
and 5th joints hirsute, separated by faint but distinct suture,
4th joint with 4 subterminal and terminal bristles with short
spines, 5th joint with 2 terminal bristles with short spines
and terminal group of spines.

Sixth limb (Fig. 6G). Hirsute laterally, 5 bare bristles in
place of epipod. Endite I with 3 bristles (2 short with long
spines or hairs, 1 long with longer spines); endite II with 5
bristles (3 short with long hairs, 2 long with long proximal
and short distal spines); endite III with 4 bristles with long
proximal spines or hairs and short distal spines (3 long, 1
short); endite IV with 3 bristles with long proximal hairs or
spines (2 long with short distal spines, 1 short). End joint
spinous, with 7-8 spinous bristles decreasing in length pos-
teriorly, each with long stout proximal spines and medium-
long distal spines (1 more medial and slender with shorter
proximal spines), followed by space and 3 hirsute posterior
bristles increasing in length posteriorly (anteriormost of these
with stout distal spines). Limb with suture separating end
joint from proximal part of limb, partial suture at base of
endite III.

Seventh limb. Comb side with 1-2 distal and 4-5 terminal
bristles, each with 2-4 bells; peg side with 2-3 distal and 3-
4 terminal bristles, each with 2-4 bells. Comb consisting of
7 long spinous teeth with widened tips (2 lateral and 1 middle
teeth shortest) and 4 short blunt teeth (2 on each side) with
long basal spines and possibly with bumpy tips. Single pro-
cumbent long peg (slightly longer than short comb teeth) with
distal and terminal tiny teeth.

Furca (Figs. 3C, 4B). Each lamella with 8-9 claws; claws

2 and 4 fused to lamella, remaining claws separated from
lamella by suture; claw 3 more slender than but slightly longer
than claw 4; claws 1-5 or 6 with long row of stout and
approximately equal posterolateral teeth and distal row of
anterior spines.

Bellonci organ. Short, cylindrical, with conical tip.

Eyes (Fig. 4B). Medial eye small, pigmented (sometimes
restricted to band). Lateral eye about twice size of medial
eye, pigmented and about 14 ommatidia. Lateral eye length
ranged from 0.29 to 0.34 mm and pigmented area from 0.20
to 0.28 mm (n = 131). Eye pigment brown in reflected light;
maroon or maroon-brown in transmitted light.

Upper lip (Fig. 4B). Anterior undivided part with numer-

Cohen and Morin: Panamanian Myodocopid Ostracodes ( Vargula)

ous glandular processes with unpigmented lobular tips and
proximal bands of maroon pigment; middle with 2 long tusks,

1 on each side; each tusk with short distal and terminal hairs
and with glandular processes; part posterior to tusks, round-
ed, hirsute.

Anterior of body. Rounded projection dorsal to 2-3 small-
er usually pointed projections.

Posterior of body. Mostly smooth; small group of short
hairs sometimes visible midposteriorly.

Copulatory organ (Fig. 4B). Anterior lobe rather conical,
longer than short posterior lobe which has at least 1 group
of about 3 bristles.

DESCRIPTION OF ADULT FEMALE. Carapace (Fig.
5D, E). Oval with deep incisur and protruding caudal process;
anteroventral, posteroventral, posterodorsal, and anterodor-
sal margins broadly rounded. Greatest height near middle
(at or just posterior to middle, not markedly higher in pos-
terior half). Caudal process small (slightly larger in right
valve than in left), dorsal margin rather straight, forming
obtuse angle with rather truncate posterior margin of valve,
dorsal edge of process about same length as dorsal edge of
incisur; both incisur and dorsal edge of process at about
midheight of valve. Dorsal edge of incisur slightly overlap-
ping ventral edge at inner end; faint line on outer surface of
valve curving from inner dorsal edge of incisur to anterior
margin of valve ventral to incisur. Tip of rostrum with few
nodes. Outer surface smooth but with faint pattern resem-
bling overlapping scales, visible at 100x; scale-like pattern
producing minute points on anteroventral and posteroventral
margin. Surface with rather regularly distributed small pores.

Infold (Fig. 5F, G). Similar to that of adult male except
rostrum (Fig. 5G) with about 3-5 double bristles posterior
to main row of 24-33 double bristles, 0-2 bristles anterior
to row; anteroventral infold (Fig. 5F) with row of about 35-
43 double bristles parallel to margin (about 4 of these ventral
to incisur margin), about 0-17 double bristles posterior to
row; list of caudal process in left valve with numerous tiny
bristles and projections forming irregular row and posterior
margin with tiny ripples, caudal list straight except for slight
bend in dorsal half, caudal list of left valve terminating dor-
sally with round knob, with 0-2 bristles anteroventral to list;
caudal list of right valve with minute pointed processes or
fringe on ventral posterior edge, caudal list rather straight,
ending in raised dorsally directed bar.

Selvage. Similar to male.

Size. 25 10. la-c length 2.39, height 1.63 mm; 22 1 1 .3 length
2.33, height 1.68 mm; USNM 193223 length 2.35, height
1.65 mm (allotype); 1212. 12-A length 2.55, height 1.70 mm.
Only 4 females (and 1 A— 1 ovigerous female) have been
collected.

First antenna. Similar to male except b-bristle of 7th joint
about V3-V2 longer than a-bristle, with 2 short proximal fil-
aments and process on tip; c-bristle with about 8 filaments
(with teeth or spines) becoming longer distally, about 5 times
length of b-bristle, with process on tips of filaments and
bristle; 8th joint with distal filament off- and g-bristles much
longer than others, small round knob at medial dorsal base
of f-bristle.

Contributions in Science, Number 373

Second antenna (Fig. 5H). Similar to male except exopo-
dite with shortest bristle of 9th joint, bearing few spines;
bristle of 2nd joint reaching 7th-9th joint, with 5-6 spines.

Mandible (Fig. 6A). Similar to male except coxale endite
with stout partly annulate spine on tip; 2nd endopodite joint
with 1 5 dorsal bristles (only 4 short hirsute medial bristles);
ventral margin of 2nd joint with 2 single distal bristles and

1 subterminal bristle together with subterminal unringed fin-
ger-like sclerotized process. Third endopodite joint with 1 of
4 bristles (stout ventral lateral) extending just beyond claws.

Maxilla (Fig. 6B). Similar to adult male except 2nd en-
dopodial joint with 2 longer (of 4) a-bristles bearing a few
stout proximal spines, 2 of 3 b-bristles distally ringed.

Fifth limb (Fig. 6C). Similar to adult male except epipodial
appendage with about 46 bristles; proximal part of protopo-
dite (coxale?) with pair of small sclerotized teeth bearing
minute stout spines; exopodite joint 1 with 1-2 bare bristles
near base of peg.

Sixth limb. Similar to adult male except end joint with 6-
8 instead of 7-8 anterior bristles.

Seventh limb (Fig. 6D). Similar to adult male except comb
side with 3-4 distal and 3-5 terminal bristles, peg side with
4-5 distal and 3-4 terminal bristles. Comb consisting of 7
long teeth (similar to those in male) and 4-5 short blunt teeth
(2-3 on each side).

Furca (Fig. 6E). Similar to that of adult male except 3rd
claw both shorter and more slender than 4th claw.

Bellonci organ (Fig. 6F). Similar to that of adult male but
with blunt tip.

Eyes (Fig. 6F). Similar to those of adult male but lateral
eye less than 1 V2 times as large as medial eye. Overall size
and ommatidia size of lateral eye smaller than in male. Length
of lateral eye ranged from 0.23 to 0.29 mm and pigmented
area from 0.18 to 0.23 mm.

Upper lip (Fig. 6F). Lip similar to that of adult male except
tusk with short distal as well as terminal short hair.

Anterior of body (Fig. 6F). With rounded projections dor-
sal to 2-5 smaller, usually pointed projections.

Posterior of body. With a few short hairs.

Genitalia. Sclerotized ring (possibly with small ventral
projections) with attached spermatophores.

Eggs. 25 10. la-c with 26 eggs in ovary; the largest egg is
0.22 mm; 1212.1 2A with 32 eggs in ovary, 0.23-0.25 mm;

21 1 1.5-A with 24 eggs in ovary, 0.058-0.077 mm.

Ovigerous juvenile. One small ovigerous female (21 1 1.5C)
bears about 13 eggs/ovary (eggs = 0.058 mm). It has the
diagnostic characters shared by V. graminicola and V. shul-
manae. It has the specific diagnostic characters of V. shid-
manae: 5 (lowest adult number) stout spines on the 2nd
exopodite bristle of the 2nd antenna; rostral infold with 1-

2 bristles posterior to row of 19-23 bristles; anteroventral
infold with 0-1 bristle posterior to row of 31-33 bristles
(lowest adult number). We identify this female as an A — 1
instar because it has these juvenile characters: length only
1.88 mm, height 1.28 mm; 7th limb with 14-15 bristles
(lowest adult number), all strongly tapered and with only 1-
2 bells; furca with only 8 claws (lowest adult number).

GENERAL BIOLOGY. Geographical distribution. Known

Cohen and Morin: Panamanian Myodocopid Ostracodes ( Vargula ) 13

Figure 6. Vargula shulmanae. A-D. 2510.1a-C, female, length 2.34 mm; A. Right mandible, medial view; B. Tip of right maxilla, lateral
view, showing 2nd and distal part of 1st endopodial joints; C. Tip of right fifth limb, posterior view, showing exopodite; D. Tip of 7th limb.
E, F. USNM 193223, female (allotype), length 2.35 mm: E. Left lamella of furca, lateral view; F. Left view of anterior of body showing lateral
and medial eye, bellonci organ, upper lip. G. 2510.3-A, male, length 2.01 mm, 6th limb.

14 Contributions in Science, Number 373

Cohen and Morin: Panamanian Myodocopid Ostracodes ( Vargula)

only from the vicinity of the type-locality of the western San
Bias Islands, Panama (9°33T4"N, 78°55'23"W).

Habitat. This species is most abundant on slopes and walls
of coral reefs with good water circulation. Most occur at
depths of 4 to >20 m. Individuals generally occur in areas
with dense coral cover and high coral diversity. Almost al-
ways there is a well-developed canopy of large gorgonians
above an understory of massive and foliaceous scleractinian
corals, all on steep slopes.

Ecology, behavior, and bioluminescence. Vargiila shul-
manae is a benthic species of steep coral slopes. On dark
nights males become demersal zooplankters and range up to
5 m above the reef (but above and among the gorgonians).
They appear in the water column about 55 min afer sunset
for about one to two hours. On only 4 occasions have females
been caught in plankton nets. During the period that they
are planktonic, males repeatedly emit specific downward trains
of luminescent signals. For each train, the light is a vertical
sequence of widely spaced extracellular secretions emitted
by one male swimming rapidly downward at about 8.1 ±
1.3 cm s"1 (n = 22 laboratory trials). Each signaling male is
accompanied by as many as 1 2 nonsignaling males or sub-
adult males. Each luminescent pulse has a duration of about
7.2 s (from 10 estimates) and a diameter of about 1-2 mm.
Each train contains about 5-8 pulses and the total train length
is about 1.5-2 m and lasts about 30-40 s. The interpulse
distance between the first two pulses is about 30-50 cm;
subsequent interpulse distances become much closer (the last
pair is about 5 cm apart). Each train is followed by a 30-
40 s nonluminescent dark period before the sequence is re-
peated nearby. There is a loose synchrony of displays over
the reef face. Displays are usually spaced at least 1 m apart.
Presumably these displays act as mating signals by the swim-
ming males to the sexually receptive benthic females. Cop-
ulation probably occurs in the water column when the female
swims to the signaling male. Females are capable of lumi-
nescing but have not been observed to produce any mating
signals. We do not know what they feed on; they are not
attracted to fish or crustacean carrion. A paper discussing
their luminescent displays is in preparation (Morin, in press).

Color of live ostracode. Males are similar in color to fe-
males and are mostly transparent except for the eyes (see
above), the light organ, and the abdomen. Both the light organ
and the abdomen are distinctly brownish in color.

Parasites. Female no. 21 1 1.5-A with eggs in ovary has 2
choniostomatid copepod parasites: 1 female attached to dor-
sum of ostracode and 1 male on anterior of ostracode, dorsal
to 1 st joint of left 1 st antenna and medioanterior to left lateral
eye.

DISCUSSION. Vargula graminicola and V. shulmanae
are so similar morphologically that they may be regarded as
sibling species. The two species have almost identical valves
with a low, short, rather pointed triangular caudal process
(Fig. 3A, C). V. shulmanae is larger, slightly rounder (less
elongate), and has a slightly more triangular caudal process

Contributions in Science, Number 373

than V. graminicola. The two species apparently do not over-
lap in size, but do overlap in ratio of valve length to height
(see below). They differ very little except in the number of
bristles on the infold and the number of stout spines on the
bristle of the 2nd joint of the exopodite of the 2nd antenna.
But the two species live in distinctly different habitats, pro-
duce different luminescent displays and have different diets
( V. shulmanae will not eat carrion while V. graminicola will).
Both V. graminicola and V. shulmanae have limbs which
are almost identical to those of V. parasitica (Wilson, 1913),
described from Jamaica. V. parasitica has a much smaller
(very inconspicuous) and more rounded caudal process than
V. graminicola and V. shulmanae. These three species are
the only Caribbean species of Vargula with a small caudal
process except for two undescribed species from Belize (Co-
hen, manuscript) and V. harveyi, another Jamaican species,
of which only the female is known (Komicker and King,
1965). All six of these species share some morphological limb
characters (discussed below), but V. harveyi is distinguished
by a distal jaw on the 7th limb.

Wilson’s original description of V. parasitica is brief and
does not describe all of the limb characters. Harding (1966)
expanded the description based upon specimens collected by
Wilson from the same host as the holotype. Komicker (1984)
further expanded the description based upon examination of
the holotype (determining that it is an adult female) and other
Jamaican specimens. The senior author of this paper has also
examined the holotype of V. parasitica and four paratypes
including specimens upon which Harding (1966) based his
redescription (males A, C, female B, USNM 43586, 1 1 2672),
and two females (USNM 78656) used by Komicker (1984)
in his supplementary description.

The following is a comparison of Vargula parasitica, V.
graminicola, and V. shulmanae:

Caudal process: Larger, lower, more protruding, more
pointed in V. graminicola and V. shulmanae (particularly
females) than in V. parasitica. In V. graminicola and V.
shulmanae the dorsal margin of the process diverges from
the posterior margin of the valve at a distinct angle, pro-
ducing a triangular shaped caudal process below midvalve
height in both sexes. In V. parasitica the caudal process is
more broadly rounded, shorter and not triangular, and shows
more sexual dimorphism. In females (Wilson, 1 9 1 3:pl. 33,
fig. 30; Harding, 1966:fig. 2; Komicker, 1984:fig. 12a) the
process is high, above midvalve height and very short, form-
ing almost a continuous curve with the posterior valve mar-
gin. In males (Harding, 1966:fig. 1) the process is low (con-
fined to the lower half of the valve) and slightly longer than
in the female).

Valve size: In Wilson’s figure (pi. 53, fig. 303) the greatest
height of the valve of V. parasitica is distinctly posterior to
the middle. In Harding’s and Komicker’s specimens of V.
parasitica and also in V. graminicola and V. shulmanae it
is near the middle. V. parasitica is intermediate in size be-
tween the smaller V. graminicola and larger V. shulmanae,

Cohen and Morin: Panamanian Myodoeopid Ostracodes (Vargula) 15

its length overlapping slightly with males and females of V.
graminicola and overlapping more with males of V. shul-
manae. Wilson reported that the (female) holotype of V.
parasitica is 1.8 mm long and 1.15 mm high. The senior
author remeasured the holotype and found it to be 2.05 mm
long and 1 .47 mm high. The range of valve length for females
of V. parasitica is 2.01-2.17 mm (Komicker, 1984). Females
are 2.43-2.55 mm in V. shulmanae and 1.80-2.01 mm in
V. graminicola. In V. parasitica. Harding’s and Komicker’s
males range from 1.75 to 1.9 mm long. Males of V. grami-
nicola are 1.52-1.75 mm long, while males of V. shulmanae
are 1.82-2.01 mm long. The ratio of male length to height
is 1.65 ± 0.04 (mean ± s.d.) [range = 1.58-1.76] (n = 55)
in V. graminicola and 1.59 ± 0.03 (mean ± s.d.) [range =
1.52-1.65] (n = 60) in V. shulmanae.

Infold: The six specimens of V. parasitica (including four
paratypes) examined by the senior author and the specimens
of V. shulmanae examined all have fewer bristles on the
infold than V. graminicola. However all the specimens of V.
parasitica were collected before 1 924 and their infolds appear
to have deteriorated more (with some bristles broken off)
than the infolds of specimens of V. graminicola and V. shul-
manae which were collected in 1983-1984. The rostrums of
V. parasitica specimens bear 4-9 (male) and 6-13 (female)
bristles in a row parallel to the rostrum compared to V.
shulmanae with 20-31 (10 males) and 24-33 (four females)
and V. graminicola with 22-33 (10 males) and 19-30 (four
females). The specimens of V. parasitica have 0-2 anterior
and 0-3 posterior to the row (this area unclear in some)
compared to 0-2 anterior and 1-6 posterior in 15 V. shul-
manae and 0-2 anterior and 4-12 posterior bristles in 10 V.
graminicola.

The anteroventral infold of the specimens of V. parasitica
is unclear in some specimens and has a row of 30-31 (male),
25-34 (female, possibly more on 1 specimen) bristles com-
pared to 32-43 in 19 V. shulmanae and 38-53 in 14 V.
graminicola. In V. parasitica there are 0-5 bristles posterior
to the anteroventral row compared to 0-4 ( 1 9 males) and 0-
17 (4 females) in V. shulmanae and 20-69 (10 males) and
34-64 (4 females) in V. graminicola.

The caudal infold of these three species of Vargula is sim-
ilar and bears very minute papillae or bristles, difficult to
count.

In order to make more accurate comparisons between the
infolds of V. parasitica and the two new Panamanian species,
fresher specimens of V. parasitica with well-preserved in-
folds are needed.

Second antenna: The number of stout spines on the bristle
of the 2nd exopodite joint is 6-7 (male), 7-9 (female) in V.
parasitica; 7-8 (rarely 6) in V. graminicola; 5-6 (rarely 7) in
V. shulmanae.

Sixth limb: The holotype of V. parasitica has 2 posterior
end bristles; Harding’s specimens of V. parasitica and the
two Panamanian species have 3 posterior end bristles.

Seventh limb: All three species have a terminal comb in
which the central tooth of the 7 long teeth is shorter than the
adjacent long teeth, apparently a unique character among the
Cypridinidae. Wilson’s description and figure (pi. 53, fig. 3 1 ! )

16 Contributions in Science, Number 373

of V. parasitica show about 23 proximal and 2 terminal
bristles, but the 7th limb on the holotype slide has about 7
proximal and more than 2 terminal bristles. Harding reports
14 bristles on the 7th limb of V. parasitica and Komicker
reports that females have 3 proximal and 5 terminal bristles
on the comb side and 4 proximal and 3 terminal bristles on
the peg side; males have 2 proximal and 4 terminal comb
side bristles and 4 proximal and 3 terminal peg side bristles.
In V. graminicola females have 4-5 proximal and 5 terminal
comb side bristles and 4-5 proximal and 3 terminal peg side
bristles; males have 1-2 proximal and 3-4 terminal comb
side bristles and 3 proximal and 3 terminal peg side bristles.
In V. shulmanae females have 3-4 proximal and 3-5 ter-
minal comb side bristles and 4-5 proximal and 3-4 terminal
peg side bristles. Komicker’s specimens of V. parasitica have
a comb with 7-8 long teeth and 4 short teeth on each side.
The two Panamanian species have 7 long teeth and 3 short
teeth on each side.

Furca: In all three species both the 2nd and 4th claws are
united to the lamella. Wilson reported 14 claws; the senior
author found 9 pairs on the holotype. Harding’s and Kor-
nicker's specimens of V. parasitica and also the two Pana-
manian species have 8-9 pairs.

Eyes; More pigmented in specimens of the two Panama-
nian species than in Komicker’s specimens of V. parasitica.
This could be an artifact of preservation differences.

Ovigerous juveniles: Ovigerous but apparently A— 1 ju-
venile females have been found in both V. graminicola and
V. shulmanae. While these bear small ovarian eggs, they do
not fall within the adult size range and may be discriminated
from adults by their 7th limbs, which bear few and strongly
tapered bristles with only 1-2 bells.

Vargula parasitica, V. graminicola, and V. shulmanae are
distinct species but are united by many shared characters. In
all three species (plus two undescribed species from Belize)
the 4th and 5th joints of the 5th limb are separated by a
suture (not fused) and bear 4 bristles on the 4th joint (right
limb of 2510.1a-C female V. shulmanae has only 3 bristles
[Fig. 4C]), 2 on the 5th joint. Both fused and unfused 4th
and 5th joints are present in other species of Vargula and
other members of the tribe Cypridinini. The 4th and 5th
joints are separated in the tribe Gigantocypridinini and in
Codonocera and Pterocypridina (assigned to separate groups
within the Cypridinini by Komicker, 1983). In all three species
(plus V. harveyi and the two species from Belize) the maxilla
bears 2 alpha-, 3 beta-, and 4 a-bristles. Some other species
of Cypridinini share these characters. The presence of 4
a-bristles is also shared with Codonocera, Pterocypridina,
and Monopia (also assigned to a group within the Cypridinini
by Komicker, 1983). None of the three species bears a ter-
minal bristle with a bulbous base on the mandible; this bulbed
bristle is apparently unique, probably derived in some species
of Vargula. All three species share an apparently unique
probably derived character, a 7th limb with a comb in which
the central tooth is shorter than those adjacent to it. The
phylogenetic polarity of the many characters shared by these
three species is still either undetermined or not fully deter-
mined, but it is likely that the three species are closely related.

Cohen and Morin: Panamanian Myodocopid Ostracodes ( Vargula)

A comparison of the figures illustrating the limbs shows that
not only do the species have the same number of bristles on
each joint of each limb, but that the bristles are similar in
length, width and ornamentation, with the very few excep-
tions noted above.

Vargula contragula, new species

Figures 3D, 7, 8

HOLOTYPE

USNM 193230 adult male, length 1.88, height 1.03 mm; in
alcohol. Sample 2310.1a. 23 Oct. 1984 at 1835-1850 hr, at
4-m depth and about 1-2 m above low coral reef on Reef
26 (Macaroon), San Bias, Panama; sweep nets of displays;
J.G. Morin, col.

PARATYPES

Sample 39. 12 Sept. 1983 at ca. 0415 hr, about 8-m depth
and 1 m above sand-coral slope on east side of Ukkup-Tupo,
San Bias, Panama; 3 1 net sweeps of luminescent displays;
J.G. Morin, col.; 1 male (JM2-B, on slide and in alcohol), 2
males (JM2-L, lost) (LACM). Sample 38. 9 Sept. 1983 at
1920-1930 hr, Vi-2 m above coral-sand slope on east side
of Ukkup-T upo, San Bias, Panama; 50 net sweeps of displays,
partial sample; J.G. Morin, col.; 1 male (JM5-B) (LACM).
Sample 1110.2a. 1 1 Oct. 1984 at 1845-1924 hr, 5-m depth,
above coral-sand slope on east side of Ukkup-Tupo, San Bias,
Panama; sweep net of displays, partial sample; J.G. Morin,
col.; 3 males (LACM). Sample 1810.1. 18 Oct. 1984 at 1915-
1930 hr, at 5-m depth and about 1-2 m above coral-sand
slope on southeast side of Taiantupo, San Bias, Panama;
sweep net of displays; J.G. Morin, col.; 2 males (LACM).
Sample 2310.1a. 23 Oct. 1984 at 1835-1850 hr, at 4-m depth
and about 1 -2 m above low coral reef on Reef 26 ( Macaroon),
San Bias, Panama; sweep nets of displays, partial sample;
J.G. Morin, col.; 1 male (A, on slide and in alcohol) (C-F)
22 males, and 1 male on SEM stub (LACM). Sample 2011.3.
20 Nov. 1 984 at 1 840- 1 9 1 0 hr, about 5-8 m depth and about
1-2 m above coral (with some sand) slope on southeast side
of Korbiski Reef, San Bias, Panama; discrete traps of dis-
plays; J.G. Morin, col.; 8 males (USNM 193231). Sample
2311.2. 23 Nov. 1984 at 1840-1920 hr, at 7-10 m depth and
about Vi-2 m above coral-sand slope on southeast side of
Korbiski Reef, San Bias, Panama; discrete traps of displays;
J.G. Morin, col.; 5 males (USNM 193232). Sample 1312.1.
13 Dec. 1984 at 1845-1915 hr, at 4-m depth and about 1-
2 m above low coral reef on Reef 26, San Bias, Panama;
discrete traps of displays; J.G. Morin, col.; 1 male (A, on
slide and in alcohol), 12 males (C-E) (LACM).

DIAGNOSIS. Anteroventral comer of valve not evenly
curved but with rounded projecting bulge; caudal process
large, about half height of valve. Females unknown; male
carapace length 1.71-2.00 mm. Furca: 7-9 pairs of claws;
claw 2 fused to lamella. Mandible: 2nd joint of endopodite
with only 2 ventral bristles ( 1 proximal, 1 distal and lateral
to sclerotized finger-like projection); 3rd joint with 1 terminal
bristle with bulbous base. Maxilla: 1st joint of endopodite
with 1 alpha- and 1 beta-bristle and large bifid cutting tooth;

Contributions in Science, Number 373

2nd joint with only 3 a-, 2 b-, and 3 c-bristles, 3 distal bristles
( 1 beta-, 2 d-bristles) long, stout with conspicuously large
rounded teeth; 2 distal bristles ( 1 beta- and the 3rd d-bristle)
attenuated distally. Fifth limb: 1st exopodial joint with only
3 pectinate teeth and 1 peg; 4th and 5th joints fused with
only 2 bristles. Sixth limb: End joint with 4-5 anterior bristles
and 3-4 hirsute posterior bristles. Seventh limb: Comb con-
sisting of 7 long and 4-6 short teeth (2-4 short teeth/side)
opposite peg; no dorsal jaw. Lip: Tusks unbranched, hirsute
distally.

ETYMOLOGY. Vargula contragula is derived from the
Latin condensatio which means “condensation” and the Lat-
in tragula which means “trail.” High-flying jet aircraft often
leave behind them a visible ‘condensation trail’ of moisture;
the contracted name for this commonly observed pattern is
‘contrail.’ Similarly contragula is the contraction of the same
Latin words and refers to the closely spaced, lateral trail of
luminescent pulses left behind noctumally swimming males
of this species.

DESCRIPTION OF ADULT MALE. Carapace (Figs. 3D,
7A, B). Oval with deep incisur, large protruding caudal pro-
cess, and small protruding process at anteroventral corner;
posteroventral, posterodorsal, anterodorsal, ventral, and
dorsal margins broadly rounded. Caudal process slightly larg-
er in right valve than in left, margin broadly curved forming
obtuse angle with curved posterior margin of valve; dorsal
edge of process shorter than dorsal edge of incisur and above
midheight of valve; inner end of incisur above midheight of
valve. Dorsal edge of incisur slightly overlapping ventral edge
at inner end; faint line on outer surface of valve curving from
inner dorsal edge of incisur to anterior margin of valve ven-
tral to incisur. Valve surface smooth but with faint pattern
resembling overlapping scales, visible at 100x; scale-like
pattern producing minute points on tip of rostrum and on
anteroventral margin of valve. Numerous minute pores scat-
tered rather regularly over valve surface, some with minute
bristles.

Infold (Fig. 7C). Infold posterior to rostrum with row of
about 9-1 1 long double bristles (members of each pair un-
equal in length and very unequal in 2-3 bristles dorsal to
incisur), 1 bristle posterior to these, 1 long and 0-1 shorter
bristle on dorsal inner edge of incisur, 1 tiny bristle posterior
and dorsal to inner edge of incisur. Anteroventral infold with
1-3 short bristles near inner comer of incisur, 1 tiny bristle
posterior to inner comer of incisur, row of about 7-13 long
double bristles on anterior margin (including 3 ventral to
incisur), 0-1 bristles posterior to row; 0-6 (fewer in right
valve) shorter bristles on ventral margin; list becoming broader
in vicinity of caudal process with numerous minute pro-
cesses, crenulations or ripples (including at least 1 5 tiny bris-
tles in left valve); 0-5 bristles ventral to left list; list ending
dorsally in knob (left valve) or bar (right valve); about 10
minute bristles forming widely spaced row on posterior edge
of caudal process of both valves.

Selvage. Selvage with lamellar prolongation with smooth
edge and faint striations present along ventral and anterior
margins; lamellar prolongation along ventral margin of in-
cisur broader and with more visible striations than elsewhere.

Cohen and Morin: Panamanian Myodocopid Ostracodes ( Vargula ) 17

A

E

18 Contributions in Science, Number 373

Cohen and Morin: Panamanian Myodocopid Ostracodes ( Vargula)

Size. USNM 193230 length 1.88, height 1.03 mm (holo-
type). Range (n = 129) for all specimens measured: length
1.71-2.00 mm, height 0.98-1.13 mm.

First antenna (Figs. 3D, 7D). First joint bare. Third joint
short with medial spines, 1 medium long spinous dorsal bris-
tle just proximal to middle and 1 medium long spinous ven-
tral bristle with base near middle or terminal. Fourth joint
with 2 medium long spinous bristles (1 terminal dorsal, 1
distal ventral). Sensory bristle of 5th joint with 10 long prox-
imal filaments, 2 more slender distal filaments, and bifurcate
tip. Sixth joint with medium long spinous medial bristle near
dorsal margin. Seventh joint: a-bristle spinous, about same
length as bristle of 6th joint; b-bristle with stout proximal
filament with bulbous base followed by large distal sucker
and distal process, and 2 slender distal filaments each bearing
2-5 (usually 3-4) smaller suckers, filaments not extending
beyond bristle tip, b-bristle about 2-3 times length of a-bris-
tle; c-bristle with stout proximal filament with bulbous base
(slightly larger than that of b-bristle) followed by large distal
sucker and distal process, 2 proximal slender filaments with
small suckers (each with 3-4), about 4-6 longer, slender bare
distal filaments with spine-like tips, and bifurcate tip. Eighth
joint: d- and e-bristles bare and filamentous, d-bristle longer
than e-bristle, e-bristle longer than b-bristle; f- and g-bristles
with 9-10 slender filaments increasing in length distally and
bearing a few spines, and bifurcate tip; c-, f-, and g-bristles
stouter and much longer than remaining bristles.

Second antenna (Figs. 3D, 7E). Protopodite with very short
distal medial bristle. Endopodite 3-jointed: 1st joint with 4
proximal bristles (1 short, 3 very short) and 1 short distal
bristle with short spines; 2nd joint elongate with 1 very short
distal bristle with short spines; 3rd joint short (about xh length
of 2nd joint) with long terminal filament. Exopodite: 1st joint
with minute terminal process; bristle of 2nd joint reaching
to end of 6th, 7th, 8th or 9th joint, with 6-12 stout more
proximal, mostly ventral spines, and about 0-5 minute spines
on narrower distal portion; joints 3-8 with natatory hairs
only; joints 2-8 with increasingly stouter ventral spines; lat-
eral spine of 9th joint slightly shorter than ventral spine of
8th joint; 9th joint with 4 unequal bristles (3 longest with
natatory hairs).

Mandible (Figs. 3D, 7F). Coxale endite spinous, with bris-
tle near base and stout terminal spine with some annulations.
Basale: Ventral margin with 2 a-bristles (1 long spinous, 1
short) with bases on medial side, 1 short lateral b-bristle
(close to a-bristles), 1 long spinous (short spines) and 1 short

c-bristle near middle of margin, and 2 long distal d-bristles,
separated by space from c-bristles ( 1 with short spines, 1 very
long and stout with long proximal hairs or spines and short
distal spines); dorsal margin with 1 long distal bristle and 2
long subterminal bristles, all with short spines. Exopodite
with medial hirsute dorsal process extending to pointed tip
and 2 spinous bristles. First endopodial joint with 4 ventral
bristles ( 1 longest with few longer and more short spines, 1
long with short spines, 1 not as long with short spines, 1
minute; minute bristle missing and longest bristle distally
bifurcate on illustrated limb of 1312.1 -A). Second endopo-
dial joint: Dorsal margin with 12-17 bristles (5 long bristles
with short spines, 6-10 short hirsute medial bristles, 1 short
stout medial bristle with stout spines, 1 short to medium
length medial bristle with short spines); ventral margin with
1 short distal bristle and 1 shorter subterminal bristle with
base on lateral side of stout finger-like subterminal unringed
process (process almost as long and broad as claws on 3rd
joint). End joint with 3 long claws (2 ventral claws stouter
than dorsal claw and with very stout bases and stout proximal
teeth) and 4 bristles (1 dorsal bristle about half claw length,
1 short slender ventral bristle, 1 slender ventral bristle almost
as long as claws, 1 ventral bristle with large bulbous base
and extending beyond claws).

Maxilla (Fig. 7G). Endite I with about 12 long stout dis-
tally ringed bristles with many long slender spines; endite II
with about 6 mostly stout long distally ringed bristles with
long spines; endite III with about 5 mostly longer bristles
with long proximal and short distal spines and rings. Coxale
with stout hirsute dorsal bristle. Basale with 1-2 long medial
bristles and 1 lateral hirsute bristle at base of exopodite.
Exopodite hirsute with 1 long hirsute proximal bristle and 2
long terminal bristles ( 1 stouter hirsute, both with distal short
spines). First endopodial joint with long slender alpha-bristle
with short hairs and long, stout, claw-like, pectinate beta-
bristle with 8-9 very large stout rounded proximal teeth (dis-
tal V4-V5 of bristle narrow, ringed flexible with about 22 mi-
nute teeth); cutting tooth prominent, usually bifid (usually
composed of 2 rather pointed triangular parts); [1 limb of
13 12.1 -A with trifid tooth, no beta-bristle, 3 long pectinate
c-bristles, 1 long unringed pectinate d-bristle with 8 large
teeth and 2 long pectinate d-bristles with slender ringed tips].
Second endopodial joint with 3 medium-short ringed pec-
tinate a-bristles (shorter than all but 2 end bristles), 2 pec-
tinate b-bristles (1 long, 1 short, both ringed), 1 short slender
ringed and 2 long stout ringed pectinate c-bristles, and 3 long
stout pectinate d-bristles (2 very stout bare, claw-like with

Figure 7. Vargula contragula, males. A. JM2-B, length 1.76 mm, right lateral view of whole specimen showing valve, lateral eye, and caudal
infold (dashed line). B, C. 1 312. 1 -C, length 1 .86 mm: B. Inside of left valve showing valve, infold, and muscle scars; C. Inside view of anterior
part of left valve. D. USNM 193232-A, length 1.86 mm, left 1st antenna, medial view (distal parts of sensory, c-, f-, g-bristles not shown).
E. 2310.1a-A, length 1.98 mm, left second antenna, medial view, showing endopodite, exopodite, and distal part of protopodite (only bases
of bristles on exopodial joints 3-9 shown). F. 1312.1 -A, length 1.94 mm, right mandible, lateral view. G, H. 23 10. la- A: G. Tip of right
maxilla, medial view, showing 2nd and distal part of 1st endopodial joints; H. Tip of right 5th limb, anterior view, showing exopodite.

Contributions in Science, Number 373

Cohen and Morin: Panamanian Myodocopid Ostracodes ( Vargula ) 19

Figure 8. Vargula contragiila, males. A. JM2-B, length 1.76 mm, 6th limb. B. 2310. la-A, length 1 .98 mm, tip of 7th limb. C. JM5-B, length
1.71 mm, right lamella of turca, lateral view. D. USNM 193231-A, length 1.75 mm, left view of anterior of body showing lateral and medial
eyes, bellonci organ, and upper lip.

about 8-10 very large rounded teeth, and 1 longer with about
1 5 small teeth and slender ringed flexible tip with minute
teeth [similar to beta-bristle]).

Fifth limb (Fig. 7H). Epipod with about 45 bristles. Pro-
topodite with 1 hirsute bristle (may be in row on 1st joint of
exopodite). Three endites with about 1 7 bristles, all with long
spines or stout hairs and some also with short spines. Endite

1 with 1-2 long and 5 short bristles with long spines; endite
II with 4-6 rather short bristles; endite III with 3 short and
4 medium-length bristles (2 unringed, 2 pectinate). First ex-
opodial joint with only 3 pectinate teeth and 1 pointed prox-
imal peg (slightly longer than shortest tooth); long pectinate
bristle with proximal long hairs or spines present near peg;
anterior side with 3(— 4 ) bristles forming row near inner mar-
gin (all with long proximal hairs, 2 very long with distal short
spines). Second exopodial joint with 13-14 very long slender
flexible pectinate bristles (all ringed), 1 proximal hirsute an-
terior bristle and 1 proximal posterior bristle (proximally
hirsute with distal short spines). Inner lobe of 3rd exopodial
joint with 2 bristles with long proximal hairs and short distal
spines; outer lobe hirsute with 2 terminal bristles (with short
spines). Fourth and 5th exopodial joints fused, hirsute, with

2 bristles (with short spines) [minute partially sclerotized
process present terminally between bristles of 1312.1 -A],

20 Contributions in Science, Number 373

Sixth limb (Fig. 8A). Three to four bare bristles in place
of epipod. Endite I and II each with 3 bristles (1 long with
long distal spines, 2 short with long hairs); endite III with 4
bristles (2 long with long mid and short distal spines, 2 short
proximally hirsute); endite IV with 2 long bristles (longest
bristle with long mid and short distal spines, other bristle
proximally hirsute). End joint hirsute with 4-5 anterior bris-
tles (decreasing in size to short posteriorly) with long prox-
imal hairs or spines and short distal spines, followed by space
and 3-4 increasingly longer hirsute bristles (shortest with
distal short spines).

Seventh limb (Fig. 8B). Comb side with 0 distal and 4-5
terminal bristles, each with 1-4 bells; peg side with 3 distal
and 3 terminal bristles, each with 2-5 bells. Comb consisting
of 7 long spinous teeth (alate at tip and longest tooth in
middle) and 4-6 short blunt teeth (2-4 on each side). Single
stout short erect curved peg with comb of terminal and lateral
teeth, present opposite long comb.

Furca (Figs. 3D, 8C). Each lamella with 7-9 claws; claw 2
fused to lamella, remaining claws separated from lamella by
suture; claw 1 much longer than claw 2; claw 3 distinctly
longer than claw 4; teeth present on all claws, claws 1-3 with
rather regular teeth on anterior margin and on lateral and
medial posterior margin.

Cohen and Morin: Panamanian Myodocopid Ostracodes ( Vargula)

Bellonci organ (Fig. 8D). Short, cylindrical, distal half con-
ical, narrowing to narrow, blunt tip.

Eyes (Fig. 8D). Medial eye unpigmented. Lateral eye larger
than medial eye, with dark brown or maroon pigment and
about 16 ommatidia. Lateral eye length ranged in size from
0.25 to 0.33 mm and pigmented area from 0. 18 to 0.28 mm
(n = 129).

Upper lip (Fig. 8D). Anterior undivided part unpigmented,
with numerous small pointed or rounded glandular process-
es; middle with 2 long tusks, 1 on each side; distal third of
each tusk with long hairs and few minute processes; part
posterior to tusks, rounded, hirsute.

Anterior of body (Fig. 8D). With rounded projection dorsal
to smaller rounded projection.

Posterior of body. Bare.

Copulatory organ. Posterior lobe helmet-shaped with bris-
tles near tip, and possibly base, lobe curving over distal part
of narrower anterior lobe; rounded posterior part of organ
with many small lobes or teeth.

GENERAL BIOLOGY. Geographical distribution. Known
only from the vicinity of the type-locality of the western San
Bias Islands, Panama (9°33'14"N, 78°55'23''W).

Habitat. V. contragiila is mostly found at depths less than
9 m in association with gorgonian covered reefs, either low
profile patch reefs or shallow sloping reefs, that have sub-
stantial open and rubble areas between the corals and gor-
gonians.

Ecology, behavior, and bioluminescence. Vargiila contra-
gula is a benthic species that lives in low profile patch reefs
and reef slopes with extensive gorgonian populations. Males
become planktonic about 1-3 m above the reefs, but usually
among the gorgonians, for about 1 hour at night. They appear
quickly about 50 min after sunset. No females have been
caught. While they are planktonic V. contragiila males pro-
duce distinctive luminescent displays. These displays are a
series of extracellular light pulses left behind a male as it
swims rapidly (about 9 cm s ') and obliquely upward at an
angle of 30° to 50°. It may be accompanied by up to six
nonluminescing males. Each pulse has a duration of about
4 s. The first two pulses are spaced about 1 5 cm apart, the
second and third about 8 cm, the third and fourth about 5
cm, and subsequent interpulse intervals are about 2 cm apart.
Many dozens of these closely spaced pulses may be produced
during a single display and cover a distance of several meters.
While they are being produced by the male at one end of the
train, they are fading away slowly at the other end (a distance
of about 30 cm and containing about 1 5 visible pulses at any
one time). The effect is like a miniature glowing contrail from
a jet aircraft, moving through the water at a low angle to the
bottom. The paths of the trains are either linear or slightly
curved. Often several males will commence their displays
near one another and then radiate outward. Thus there is
loose synchrony among males. The displays cease rather
abruptly, within about 10 min, approximately an hour after
they begin. As with the two species described above, it is
assumed that these displays are mating signals from plank-
tonic males to sexually receptive benthic females. It is not
known what they feed on; they are not attracted to fish or

Contributions in Science, Number 373

crustacean carrion. No predation upon them has been ob-
served.

Color of live ostracode. The body is mostly transparent
except for the eyes, light organ, and abdomen. The light organ
is an orange-brown color while the abdomen is a distinctly
organgish color.

DISCUSSION. Vargiila contragiila shows the diagnostic
charcters of the genus Vargula (Komicker, 1975). Polarity
of characters within the genus is uncertain. V. contragiila
possesses some apparently unique characters (autapomor-
phies) and shares others with some congeners. V. contragula
is the only species of Vargiila with a 5th limb with a 1st
exopodite joint bearing only 3 teeth plus a peg. Only one
other species of Cypridinidae, Pterocypridina sex Komicker,
1983 ( Pterocypridina group), has 3-4 teeth and a peg (Kor-
nicker, personal communication). Almost all species of Cy-
pridinidae have 6 teeth plus a peg. Reductions have also
occurred in V. dent at a Komicker, 1975 (5 teeth plus a peg)
and other Cypridina group species: Codonocera suensoni
Poulsen, 1962 (4 teeth plus a peg), C. goniacantha Muller,
1906, C. polygonia Muller, 1906, C. weberi Muller, 1906,
and Rugosidoloria serrata Komicker, 1975 (all with 5 teeth
plus a peg).

Only one other species of Vargula, V. spmulosa, has a
valve with a bulge on the anteroventral comer similar to that
of V. contragiila. V. spinulosa also has a furca with the 2nd
claw united to the lamella, but has a 7th limb with a dorsal
jaw and differs from V. contragula in many other respects
including those noted below.

Vargula contragula has some apparently uniquely shaped
distal bristles on the maxilla: 1 beta- and 2 d-bristles are
stout and claw-like with prominently large rounded teeth,
but the beta- and one of these d-bristles also have unusual
very slender ringed minutely toothed distal portions.

Vargula contragula and V. hilgendorfii are the only known
species of Vargula with a mandible with the ventral margin
of the 2nd endopodite joint with only 1 single distal bristle
proximal to the usual subterminal bristle paired with a scler-
otized finger-like projection. But a few specimens of some
other species of Vargula occasionally have an aberrant limb
with a reduced number of bristles on this joint.

Vargula contragula shares some characters almost exclu-
sively with a few Caribbean species of Vargula. Of the 22
described species of Vargula (including the three described
herein) six are from the Caribbean Sea. There are 8-10 ad-
ditional Caribbean species which we will describe in future
papers (4-5 from San Bias, Panama and 3-4 from Carrie
Bow Cay, Belize). The 14-16 Caribbean species fall mainly
into three groups of species sharing concordant morpholog-
ical characters: 1) Vargula contragula, V. bullae Poulsen,
1962, 3-4 undescribed species from San Bias and 1-2 un-
described species from Belize; 2) V. harveyi Komicker and
King, 1965, and 1 undescribed species from San Bias; and
3) Vargula parasitica, V. graminicola, V. shulmanae, and
two undescribed species from Belize.

Group 1. The species in the first group (containing V. con-
tragula) share the following characters, most of which occur
in few other species of Vargula: high, usually prominent

Cohen and Morin: Panamanian Myodocopid Ostracodes (Vargula) 21

caudal process, mandible with a bulb-based end bristle; 5th
limb with fused 4th and 5th joints bearing a total of only 2
bristles; maxilla with 1 alpha-, 1 beta-, 3 a-, and a total of 5
b- and c-bristles (except V. bullae has 2 beta-bristles); furca
with only the 2nd claw united to the lamella (except V. bullae
and 1-2 undescribed species from San Bias with both the
2nd and 4th claws united to the lamella); 7th limb with comb
of 7 long teeth (longest in center) and adjacent short teeth.

Apparently the Caribbean species of Vargula are the only
species of the Cypridinidae possessing a mandible with a
bulb-based end bristle except for V. tsujii Komicker and
Baker, 1977, from California, NE Pacific Ocean, V. hilgen-
dorfii (Muller, 1890) from the west Pacific, and V. plicata
Poulsen, 1962, from the Celebes Sea.

The reduction in number of bristles on the maxilla and on
the fused 4th and 5th joints of the 5th limb occurs in some
other members of Cypridinidae (although 1 alpha-bristle oc-
curs only in some members of the Cypridina genera group),
but in Vargula occurs in only 3 of the 16 presently known
non-Caribbean species: V. tsujii (shares all of the listed char-
acters of the first Caribbean group except the high caudal
process and the 7 long teeth in the comb of the 7th limb),
V. hilgendorfii (shares with the first Caribbean group only
the bulb bristle, 2nd claw only of furca fused, high caudal
process, and 1 alpha-bristle), and V. magna Komicker 1984,
from North Carolina and west Florida (shares with the first
Caribbean group only the reduction of maxilla bristles).

Only four non-Caribbean species of Vargula have only the
2nd claw fused to the lamella: V. tsujii, V. hilgendorfii, V.
spinulosa Poulsen, 1962, from Korea (shares with the first
Caribbean group only the high caudal process and fused 2nd
claw of the furca), and V. spinosa Poulsen, 1962, from Japan
(shares only the fused 2nd claw of the furca).

Group 2. The species in the second group share the fol-
lowing characters: low and short caudal process; mandible
with end bristles without or with slightly bulbous base; 5th
limb with unfused or partially fused 4th and 5th joints with
4-5 bristles; maxilla bristles and furca as described in group
3; 7th limb with comb as described in group 1.

Group 3. The species of the third group share the following
characters: low and usually short caudal process; mandible
without bulb-based end bristle; 5th limb with unfused 4th
and 5th joints bearing a total of 6 bristles; maxilla with 2
alpha-, 3 beta-, 4 a-, 3 b-, and 3 c-bristles (number of b- and
c-bristles unclear and possibly different in V. parasitica ); fur-
ca with 2nd and 4th claws fused to lamella; 7th limb with
comb having 7 long teeth (with the longest teeth adjacent to
the center tooth) and short lateral teeth.

Only four other non-Caribbean species of Vargula are also
known to have a 7th limb with a comb with 7 long teeth and
in all of them the central tooth is longest: V. norvegica (Baird,
1860) from the eastern North Atlantic Ocean; and, in the
Pacific Ocean, V. plicata, V. lusca Komicker, 1975, from
south of Tasmania, and V. ascensus Komicker, 1979, from
New Zealand.

The geographically concordant suites of characters suggest
that some Caribbean species of Vargida are closely related
to each other and that the taxon may be a promising subject

22 Contributions in Science, Number 373

for cladistic analysis. However, the high number of new species
which we have found proportional to those presently de-
scribed suggests that many more species may still be unde-
scribed throughout the world. An analysis also will require
better determination of the polarity of the character state
transformations based upon further study of the relationship
of Vargula to other genera in the Cypridinini and their char-
acter states. Study of the possibly undescribed species of
choniostomatid copepods parasitic on some of these species
also might be useful in determining phylogenetic relation-
ships.

ACKNOWLEDGMENTS

We thank the Smithsonian Tropical Research Institute for
the use of their facilities in the San Bias Islands, Panama;
the Kuna Indians for graciously allowing us to work among
their islands; and Myra Shulman, Eric Schultze, and Diana
Pilson who helped collect and observe these ostracodes in
the field. We thank the Crustacean Laboratory, Allan Han-
cock Foundation, University of Southern California for use
of their facilities. We also thank Louis Komicker for advice
and for reviewing the manuscript. Margaret Kowalczyk inked
the figures. Thomas E. Bowman, III verified the identification
of the parasitic copepods. This work was supported in part
by a National Geographic Society Grant, an American Phil-
osophical Society Grant, and a University research Grant
from U.C.L.A. to J.G.M.

LITERATURE CITED

Baird, W. 1860. Note upon the genus Cypridina Milne-
Edwards, with a description of some new species. Pro-
ceedings of the Zoological Society of London 28:199-
202, pi. 71 [on pg. 189],

Bradford, J.M. 1975. New parasitic Choniostomatidae(Co-
pepoda) mainly from Antarctic and Subantarctic Ostra-
coda. New Zealand Oceanograhic Institute Memoir (67):
1-36.

Harding, J.P. 1966. Myodocopan ostracods from the gills
and nostrils of fishes. In: H. Barnes, ed.. Some Contem-
porary Studies in Marine Science: 369-374, George Al-
len and Unwin Ltd., London.

Komicker, L.S. 1975. Antarctic Ostracoda (Myodocopina).
Smithsonian Contributions to Zoology (163): 1-720.

. 1979. The marine fauna of New Zealand: Benthic

Ostracoda (Suborder Myodocopina). New Zealand
Oceanographic Institute Memoir 82:1-58.

. 1983. The Ostracode Family Cypridinidae and the

Genus Pterocypridina. Smithsonian Contributions to
Zoology (379): 1-29.

. 1984. Cypridinidae of the continental shelves of

southeastern North America, the northern Gulf of Mex-
ico, and the West Indies (Ostracoda: Myodocopina).
Smithsonian Contributins to Zoology (401): 1-37.
Komicker, L.S., and J.H. Baker. 1977. Vargula tsujii, a
new species of luminescent Ostracoda from lower and
southern California (Myodocopa: Cypridininae). Pro-
ceedings of the Biological Society of Washington 90(2):
218-231.

Cohen and Morin: Panamanian Myodocopid Ostracodes ( Vargula)

Komicker, L.S., and C.E. King. 1965. A new species of
luminescent Ostracoda from Jamaica, West Indies. Mi-
cropaleontology 1 1(1): 105-1 10.

Morin, J. G. “Firefleas” of the sea: Luminescent signaling
in marine ostracode crustaceans. Florida Entomologist,
in press.

Morin, J.G., and E.L. Bermingham. 1980. Bioluminescent
patterns in a tropical ostracod. American Zoologist 20(4):
851. (Abstract)

Muller, G.W. 1890. Neue Cypridiniden. Zoologische Jahr-
biicher 5:21 1-252.

. 1906. Die Ostracoden der Siboga-Expedition. In

Uitkomsten op Zoologisch, Botanisch, Oceanogra-

phischen on Geologische Gebeid versameld in Neder-
landsch Oost-Indie, 1899-1900, 30: 40 pages, 9 plates.
Leiden: E.J. Brill.

Poulsen, E. 1962. Ostracoda— Myodocopa, Part 1 Cypri-
diniformes— Cypridinidae. Dana-Report (57): 1-414.
Copenhagen: Carlsberg Foundation.

Wilson, C.B. 1913. Crustacean parasites of West Indian
fishes and land crabs, with descriptions of new genera
and species. Proceedings of the United States National
Museum 44:189-277, pi. 18-53.

Accepted 11 October 1985.

Contributions in Science, Number 373

Cohen and Morin: Panamanian Myodocopid Ostracodes ( Vargula) 23

SERIAL PUBUCATI^NS^ OF THE
NAttmAL HtSirORY MUSEUM OF I.OS ANGELES COUNTY

The- scientific publications of the Natural. History Museum of lids Angeles County have been
issued at irregular intervals in three, major series; the articles in each series are numbered
individually, apd numbers run consecutively,' regardless of the subject matter.

®i C ontiibutions i*n Science., a miscellaneous series of technical papers describing orig-
inal research in the life and earth science's.

# Science Bulleliri, a miseelilaneous series of monographs describing original research
in the life and earth sciences. This series was discontinued in 1978 with the issue of
,Nurabprs"39 and 30; monographs are now published by the; Museum in Contributions

i n. Sc ience,

ft Science Series; long articles on natural history topics, generally written for the layman.

; Copies of life publications in these series are sold through the Museum Book Shop. A catalog

is available on request. . ; . i ;j a i f ; TA C'i-

LATE MIOCENE AND HOLOCENE MAMMALS, EXCLUSIVE
OF THE NOTOUNGULATA, OF THE RIO ACRE
REGION, WESTERN AMAZONIA

Carl D. Frailey

Contributions in Science, Number 374
Natural History Museum of Los Angeles County
15 May 1986

ISSN 0459-8113

Natural History Museum of Los Angeles County
900 Exposition Boulevard
Los Angeles, California 90007

CONTENTS

ABSTRACT 1

RESUMEN 1

INTRODUCTION 1

MEASUREMENTS, ABBREVIATIONS, AND LOCALITIES 3

GEOLOGY 4

Tertiary Red Beds 4

Inapari Formation 5

Acre Conglomerate Member 5

Member A 6

Member B 6

Member C 6

Geological Correlation and Previous Work 6

SYSTEMATICS 8

Rio Acre Local Fauna (Late Miocene) 8

Marsupialia 8

Stenodon camp belli n. gen. and sp 9

Megalonychidae 12

Pampatheriinae 12

Erethizontidae 13

Dasyproctidae 13

Cardiomyinae 14

Genus A 14

Genus B 1 4

Kiyutherium orientalis 15

Potamarchus murinus 22

Tetrastylus sp 26

Telicomys amazonensis n. sp 26

Neoepiblemidae 31

Echimyidae 31

Proterotheriidae 31

Macraucheniidae 31

ISynastrapotherium sp 33

IRibodon sp 34

Inapari Member A Local Fauna (Holocene) 34

Nothropus priscus 34

DISCUSSION AND CONCLUSIONS 42

ACKNOWLEDGMENTS 43

LITERATURE CITED 43

LATE MIOCENE AND HOLOCENE MAMMALS, EXCLUSIVE
OF THE NOTCH NGl LATA, OF THE RIO ACRE
REGION, WESTERN AMAZONIA

Carl D. Frailey'

ABSTRACT. Although the Cenozoic fossil record of South America
is relatively good, major fossil-bearing strata are found primarily in
temperate South America and the South American land mammal
ages are established on the faunal sequence found in those deposits.
Studies of vertebrate fossils in other parts of South America, in-
cluding those in this paper, tend to confirm the universality of these
faunal ages.

The discovery of a diverse Late Miocene (Huayquerian) local fau-
na along the Rio Acre on the border between Peru and Brazil pro-
vides the first opportunity to examine a wide representation of the
Late Tertiary fauna of the Amazon. Several taxa are recorded as
fossils in the Amazon Basin for the first time: Dasypodidae (Pam-
patheriinae), Proterotheriidae, Macraucheniidae, Erethizontidae,
Dasyproctidae, Dinomyidae (Potamarchus murinus, Telicomys
amazonensis n. sp., Tetrastylus sp.), Neoepiblemidae, Hydrochoe-
ridae ( Kiyutherium orientalis ), Caviidae (Cardiomyinae), Echimyi-
dae (Heteropsomyinae), and Sirenia (Trichechidae). The affinities of
this local fauna with the Late Miocene savanna fauna of Argentina
suggest that a widespread savanna community extended into the
Amazon Basin. However, a new genus of Mylodontidae and what
may be a new family of Marsupialia underscore major differences
between tropical and temperate faunas of South America during the
Cenozoic.

One Holocene taxon is discussed on the basis of new material.
The discovery of the ground sloth Nothropus priscus in deposits along
the Rio Acre conclusively supports the placement of Nothropus in
the Nothrotheriinae as a close relative of Nothrotheriops, a genus
that preferred a dry habitat. This relationship, and the discovery of
the specimen at the base of a thick formation that has a maximum
radioisotopic date of 1 1,000 years, indicates that major climatic and
environmental changes occurred in the western margin of the Am-
azon Basin during the Holocene.

RESUMEN. Aunque el registro de fosiles del Cenozoico en Sud-
america es bastante amplio, los estratos de fosiles mas importantes
se encuentran primariamente en la zona templada de Sudamerica,
y las Edades Mamiferos de Sudamerica estan establecidas por la
sequencia de fauna en esos depositos. Estudios de fosiles de verte-
brados en otras partes de Sudamerica, incluyendo lo expuesto en
este estudio, tienden a confirmar la universalidad de estas edades
faunesticas.

Contributions in Science, Number 374, pp. 1-46
Natural History Museum of Los Angeles County, 1986

El descubrimiento de una diversa fauna local en la ultima fase del
Mioceno (Huayqueriense) en el Rio Acre cerca de la frontera entre
el Peru y el Brasil, provee la primera oportunidad de examinar una
muestra amplia de la fauna Amazonica de la ultima fase del Ter-
ciario. Varios taxa han sido registrados de fosiles en la Cuenca del
Amazonas por la primera vez: Dasypodidae (Pampatheriinae), Pro-
terotheriidae, Macraucheniidae, Erethizontidae, Dasyproctidae, Di-
nomyidae ( Potamarchus murinus, Telicomys amazonensis n. esp.,
Tetrastylus esp.), Neoepiblemidae, Hydrochoeridae ( Kiyutherium
orientalis), Caviidae (Cardiomyinae), Echimyidae (Heteropsomyi-
nae) y Sirenia (Trichechidae). Las afinidades de esta fauna local con
la fauna de la sabana de la Argentina durante la ultima fase del
Miocene, sugiere que una comunidad de sabana ampliamente dis-
tribuida se extendia hasta la Cuenca Amazonica. Sin embargo, un
nuevo genero de Mylodontidae y lo que pudiera resultar una nueva
familia de Marsupialia hacen notar mayores diferencias entre las
faunas tropicales y templadas de Sudamerica durante el Cenozoico.

Un taxon Holoceno se comenta en base del nuevo material. El
descubrimiento del perezoso terrestre Nothropus priscus en los de-
positos a lo largo del Rio Acre definitivamente sostiene la ubicacion
de Nothropus en los Nothrotheriinae como una relacion cercana de
Nothrotheriops, un genero que mostraba una preferencia por una
habitacion seca. Esta relacion, y el descubrimiento del especimen a
la base de una formacion gruesa que tiene una maxima fecha ra-
dioisotopica de 1 1 ,000 anos, indica que los mayores cambios cli-
maticos y ambientales ocurrieron en el margen del oeste de la Cuenca
Amazonica durante el Holoceno.

INTRODUCTION

The present knowledge of South American faunal history
rests on a fossil record that is largely preserved in Argentina
(Fig. 1). The concentration of major Cenozoic vertebrate
localities in Argentina and their relative paucity elsewhere
in South America has created a situation in which the fossil
record of Argentina has been used to typify conditions

1. Midland College, Midland, Texas 79705, and Research As-
sociate, Section of Vertebrate Paleontology, Natural History Mu-
seum of Los Angeles County.

ISSN 0459-8113

Q Riochican
^ Casamayoran
- Mustersan
D Divisaderan
6 Deseadan
A Colhuehuapian
O Santacruzian
■ Friasian
“O' Chasicoan
^ Huayquerian
^ Montehermosan

Figure 1. Maps of South America with distributions of major vertebrate localities and the location of the Rio Acre. A, Tertiary localities;
B, Pleistocene localities (modified from Patterson and Pascual, 1972).

throughout South America. The Cenozoic faunas of Argen-
tina comprise the types for all but one (the Friasian) of the
South American land mammal ages and smaller faunas found
elsewhere in South America are naturally compared to the
established land mammal ages. However, the extent of zoo-
geographic differences between faunas of South America is
largely unknown and cannot even now be treated in discus-
sions of the development of the South American fauna.

The lack of information regarding Cenozoic mammals is
nowhere more acutely felt than when discussing the Amazon
Basin. The small number of localities and known taxa are
shown in Figure 1 and Table 1. Additionally, the fossils that
were found in previous years were often highly fragmented
and abraded beyond recognition. Stratigraphic placement or
correlation could seldom be determined. Nonetheless, these
fossils have served as holotypes for numerous new genera
and species which are known only from the Amazon Basin
and frequently only from the holotype specimens. The taxa
listed in Table 1 are assuredly a heterochronus grouping, but,
for the most part, the genera listed in Table 1 are found in

Pleistocene deposits in other areas of South America. Gri-
phodon peruvianus is the oldest fossil mammal from the Am-
azon and is dated as probably Mustersan in age (Middle
Eocene) by Patterson (1942). An astrapothere, Synastra-
potherium, is dated morphologically as Oligocene (Paula
Couto, 1976). In this paper, on stratigraphic grounds, I ques-
tion that age assignment and suggest a Late Miocene age.
The three endemic toxodontida and the rodents are with
more certainty of Late Tertiary age although again they do
not necessarily form a single fauna.

Other than incidental recovery of isolated specimens such
as those described by Roxo (1921) and Anthony ( 1 924), only
three more extensive vertebrate fossil collections have been
made in this region. Two of these collections, described by
Spillman (1949) and Willard (1966), were the direct result
of exploration for petroleum in eastern Peru. The extensive
Bassler Collection of Peruvian fossils (collected between 1921
and 1931 but only described much later by Willard, 1966)
is of interest in that it was assembled from several of the
major rivers in eastern Peru (Willard, 1 966, fig. 8). Vertebrate

2 Contributions in Science, Number 374

Frailey: Fossil Mammals of the Rio Acre

Table 1. Faunal list of Cenozoic mammalian taxa previously re-
ported from the Amazon Basin.

Marsupialia

lAbderites, in Willard, 1966
Edentata
Cingulata
Glytodontidae

Panochthus, in Paula Couto, 1956
cf. Hoplophorus, in Willard, 1966
Glyptodontidae, indet., in Spillman, 1949
Dasypodidae, indet., in Spillman, 1949
Pilosa

Eremotherium, in Paula Couto, 1956
Lestodon armatus, in Paula Couto, 1956
Scelidotherium, in Paula Couto, 1956
Megalonychidae, indet., in Paula Couto, 1956
Rodentia
Dinomyidae

*Phoberomys bordasai Patterson, 1942
Echimyidae

**Palaeoechimys peruvianus Spillman, 1949
Hydrochoeridae

*Protohydrochoerus schirasakae Spillman, 1941 (in Spillman,
1949)

*Protohydrochoerus tahuamanii Spillman, 1949
Notoungulata
Toxodontidae

**Trigodonops lopesi (Roxo) 1921
**Abothrodon pricei Paula Couto, 1 944
**Neotrigodon utoquineae Spillman, 1949
Toxodon cf. platensis, in Paula Couto, 1956
Toxodon, in Willard, 1966
Astrapotheria

**Synastrapotherium amazonense Paula Couto, 1976
indet., in Willard, 1966
Pyrotheria

**Griphodon peruvianus Anthony, 1924
Proboscidea
Mastodontidae

Haplomastodon waringi, in Paula Couto, 1956
Cordillerion, in Willard, 1966
Sirenia

Trichechus, in Paula Couto, 1956
Cetacea

**Plicodontinia mourai Miranda-Ribeiro, 1938 {in Paula Couto,
1956)

Perissodactyla

Tapiridae

Tapirus, in Willard, 1966
Artiodactyla
Tayassuidae

*Dicotyles traunmiilleri Spillman, 1949
Dicotyles, in Willard, 1966

fossils, if only as fragments, appear to be widely distributed
in the strata along the eastern margin of the Andes.

Recent years have seen two paleontological expeditions to
the Rio Junta of Brazil (1956 and 1962, in Paula Couto,
1978) and preliminary papers on the fossils collected (Paula
Couto, 1976, 1978). However, as in earlier collections, well-
preserved fossils and a single major assemblage, a true local
fauna, eluded discovery.

In September, 1977, and July, 1979, Dr. Kenneth E.
Campbell of the Los Angeles County Museum and I collected
vertebrate fossils at several localities along the Rio Acre west
of Inapari, Peru (Fig. 2) on the northern border of Madre de
Dios Province. Subsequent short trips to the Rio Acre below
Inapari were conducted by Dr. Campbell alone. The trips
were initially planned following discoveries by local inhab-
itants of several large fossil turtles in strata that are exposed
along the river. Reports of these fossils reached Dr. Campbell
in 1974 while he was making an ornithological collection in
the nearby town of Iberia. Two of the turtles ( Podocnemis
sp.) were eventually recovered; one is now at the University
of Kansas Museum of Natural History (KUVP 49492) and
a second is at the Museo Nacional, “Javier Prado,” in Lima.
In addition, and of greater significance, was the identification
of two formations, one Late Miocene and one Holocene, each
with mammalian fossils (Table 2). This paper describes the
mammalian fossils, other than the notoungulates, that were
discovered along the Rio Acre. The remainder of the fossils
are still under study. These new fossils provide greater insight
into the Cenozoic faunal history of tropical South America
than was ever before possible. At the same time, the nu-
merous single or fragmentary specimens that cannot be iden-
tified below high taxonomic levels yet seem unlike known
genera and specimens indicate that much of the Cenozoic
diversity of the Amazon Basin remains undiscovered.

MEASUREMENTS, ABBREVIATONS,

AND LOCALITIES

All measurements used in this paper are in millimeters unless
otherwise indicated. Teeth measurements, length x width,
were taken at the occlusal surface unless stated otherwise.
Parentheses around a measurement indicate an approximate
measurement. Rodent dental nomenclature follows Patter-
son and Wood (1982) unless otherwise noted. Skeletal mea-
surements are standardized following DeBlase and Martin
(1974) and Driesch (1976). In the illustrations, arrows in-
dicate the anterior and labial directions from a tooth. Ab-
breviatons used are KUVP, University of Kansas, Museum
of Natural History; GB, Servicio Geologico de Bolivia (GEO-
BOL), La Paz, Bolivia; LACM, Natural History Museum of
Los Angeles County; b.p., before present.

Miocene and Holocene localities along the Rio Acre are
on record with the Section of Vertebrate Paleontology, Nat-
ural History Museum of Los Angeles County. Several local-

* Species known only from the Amazon Basin.

** Genus and species known only from the Amazon Basin.

Contributions in Science, Number 374

Frailey: Fossil Mammals of the Rio Acre 3

Generalized Geologic Section of Rio Acre, Peru

Clayey sand, variegated, weathers reddish.
Forms high cliffs along river.

sharp, linear unconformity

Sandy clay, blocky, light brown, weathers reddish,
occasional isolated channel deposit.

Channel deposits of cross-bedded sandstone
with lenses of uniform sand, clay and
lignite. Basal unit is clay-pebble conglomerate.
Replaced laterally with reddish clay.

Clay, variegated, with calcitic stringers that
feather-out towards top of unit.
Eastward dip of 15°. Occasional
grey, cross-bedded sand units.

Figure 2. Generalized geologic section along the Rio Acre (reprinted with permission from Campbell and Frailey, 1984). The Acre Con-
glomerate Member was named later (Campbell et al., 1985). Here it is the lowest conglomeratic lense of Member A.

ities, those that produced the fossils discussed in this paper,
are marked on Figure 3.

GEOLOGY

The surface geology of the northern part of Madre de Dios
Province has recently been mapped by the Oficina Nacional
de Evaluacion de Recursos Naturales (ONERN) of the Re-
public of Peru (ONERN, 1977). Only one formation, the
Inapari Formation of presumed Pliocene-Pleistocene age, is
shown to outcrop along the Rio Acre. Later work indicates
that at least two formations are present (Fig. 2, from Camp-
bell and Frailey, 1984). The upper formation is composed
of four units and represents most of the exposed section along
the Rio Acre. It is this formation that best matches the de-
scription of the Inapari Formation (ONERN, 1977) and our
usage of this name was restricted to it. The age of the Inapari
Formation, however, is Holocene rather than either Pliocene
or Pleistocene (Campbell and Frailey, 1984). The underlying
formation is hence unnamed and was referred to as the Ter-
tiary red beds by Campbell and Frailey (1984) based on a
presumed correlation to some part of the undifferentiated
Tertiary Red Beds of Singewald (1927,1928) and Oppenheim
(1937).

TERTIARY RED BEDS

This formation is principally composed of variegated clays
(green, red, and tan) with calcitic stringers that follow ihe
bedding planes. At Belgica (Fig. 3), gray, cross-bedded well-
sorted sand units replace the variegated clay. In these units
are found fossil pelomedusid turtles in what are apparently
channel deposits. The site at LACM 44 1 8 may also represent
a lensing channel sequence, but at all other places along the
river the lithology is uniform. The red beds have a north-
south strike and an approximate 1 5° dip to the east (see Fig.
3). The top of the formation is blocky clay, weathering red-
dish, in which the irregularly shaped calcitic stringers and
the visible bedding planes feather-out near the undulating
contact with the overlying Holocene Inapari Formation. This
evidently represents a weathering surface, perhaps the B level
of soil formation. The top of the weathered zone forms a
bench in many exposures. The contact between this and the
overlying unit is also the site of a perched water table that
causes numerous small springs to form at this level. The
perched water table indicates that a significant lithologic dif-
ference exists between the two formations which is not ob-
vious in field examination. The red beds are not visible at
every bank exposure along the Rio Acre.

4 Contributions in Science, Number 374

Frailey: Fossil Mammals of the Rio Acre

Table 2. Faunal list of the Rio Acre region.

Table 2. Continued.

Rio Acre Local Fauna
Tertiary

Chondrichthyes'

Batoidea

Potamotrygonidae

Osteichthyes1

Osteoglossidae

Doradidae

Pimelodidae

Characidae

Colossoma

Callichthyidae

Reptilia'

Chelonia

Pelomedusidae

Testudinidae

Crocodilia

Gavialidae

IGavialosuchus

Alligatoridae

ICaiman

IBrachygnathosuchus

Aves!

Mammalia

Marsupialia

Edentata

Mylodontidae

Stenodon camp belli n. gen. and sp.
Megalonychidae
Dasypodidae
Pampatheriinae
Rodentia
Erethizontidae
Dinomyidae

Potamarchus murinus
Telicomys amazonensis n. sp.
Tetrasiylus
Hydrochoeridae

Kiyutherium orientalis
Caviidae
Cardiomyinae
Genus A
Genus B
Neoepiblemidae
Dasyproctidae
Echimyidae
Heteropsomyinae
Litoptema
Proterotheriidae
Macraucheniidae, new genus
Notoungulata1

Astrapotheria

ISynastrapotherium

Sirenia

Trichechiidae

IRibodon

Inapari Member A Local Fauna
Holocene

Reptilia1

Chelonia

Pelomedusidae

cf. Peltocephalus sp.2
Mammalia
Edentata
Megalonychidae

Nothropus priscus

' Not discussed in this paper.

2 Identified by Dr. Eugene Gaffney, pers. comm., 1982.

Fossils are rare in the red beds. They were found in only
two localities, LACM 4418 and 4606. The site stratigraphy
at LACM 4418 is complex and may represent stream de-
position at the top of the red beds. An explanation for the
general rarity of fossils in the red beds is that the fossils are
in fact not derived from the red beds but rather come from
channels which are incised into the red beds and hence are
younger. The stratigraphy at LACM 4418 could be such a
channel sequence in this interpretation. Support for this sug-
gestion comes from the observation near LACM 4418 of a
thick, incised channel sandstone deposited at the top of the
red beds and itself truncated by the unconformity between
the red beds and the Inapari Formation. This channel deposit
may be younger than any part of the red beds. If channels
such as this are the source of the Huayquerian fossils rather
than the red beds, then the red beds are older and may not
even be of Huayquerian age.

For the present, the red beds are considered to be the source
of the Huayquerian fossils in that LACM 4606 produced
fossils ( Potamarchus ) which were also found in Unit A of
the Inapari Formation in LACM 4611. There are no visible
channel-fill deposits in LACM 4606 and the locality appears
to be in the red beds. The fossils found in the red beds and
those re-worked from this formation and re-deposited in the
Inapari Formation are designated the Rio Acre Local Fauna.

INAPARI FORMATION

The Holocene Inapari Formation is composed of four mem-
bers that were designated the Acre Conglomerate Member,
Member A, Member B, and Member C by Campbell and
Frailey (1984) and Campbell et al. (1985). Each will be dis-
cussed separately.

ACRE CONGLOMERATE MEMBER. A clay-pebble
conglomerate is frequently present and readily identifiable at

Contributions in Science, Number 374

Frailey: Fossil Mammals of the Rio Acre 5

BRAZIL

LACM 44 1 8

(Acre 1 ) LACM 4606

Figure 3. The Rio Acre and the location of five sites discussed in the text. LACM locality numbers and equivalent field numbers are listed.
Strike-dip symbols refer only to the Tertiary red beds.

the base of the Inapari Formation. Average pebble size is 1
cm with an occasional cobble up to 20 cm in diameter. The
matrix is gray or brown, well-sorted, unconsolidated sand.
Calcitic inclusions appear to have been derived from the
underlying red beds. Fossils of small animals and broken
pieces of larger animals and fossil wood are common in this
unit although the majority of fossils, all the mammals thus
far known, are evidently re-worked from the red beds. Many
of the fossils are rounded by abrasion but delicate fossils,
such as small fish spines and minute rodent teeth, are also
preserved intact. The Miocene fossils found at LACM 46 1 1
were in this conglomerate.

MEMBER A. Above the Acre Conglomerate Member is
a complex series of channel deposits composed of cross-
bedded brown to buff sandstones with lenses of pure buff
sandstones, variegated clays, and lignites. Fragments of fossil
vertebrates are found throughout this lensing sequence. Some
are probably re-worked from the red beds although their
fragmentary nature prevents positive assertion of this. The
ground sloth, Nothropus priscus, was discovered in a channel-
sand deposit near the base of this unit at LACM 4609. Fossil
vertebrates found in this member are referred to as the Ina-
pari Member A Local Fauna. Large pieces of fossil wood and
well-preserved leaves are abundant in the lignite lenses. These
channel deposits (see Fig. 2) may be replaced laterally by a
variegated, blocky clay that is red in outcrop. This facies is

less extensive than the channel lenses and produced no fos-
sils.

The thickness of Member A varies between 1 and 5 meters
with the least thickness observed where the channel deposits
are absent and replaced by the uniform clay.

MEMBER B. Member B is separated from Member A by
an undulating contact. Member B is blocky, light brown,
sandy clay that weathers reddish. Occasional isolated channel
cuts are present in this unit. Member B is usually horizontally
bedded with rare instances of structural slump. Its thickness
is approximately 5 meters. No fossils were found in Mem-
ber B.

MEMBER C. The uppermost member of the visible sec-
tion forms high, inaccessible cliffs along the river. Member
C is separated from Member B by a sharp, conspicuous, linear
contact. Member C is composed of a clayey sand, variegated
in color (tan, red, and green) that weathers reddish. Thickness
is approximately 30 meters. No fossils were recovered from
Member C.

GEOLOGICAL CORRELATION AND
PREVIOUS WORK

The geology of the western margin of the Amazon Basin is
not well described in the literature and correlation based on

6 Contributions in Science, Number 374

Frailey: Fossil Mammals of the Rio Acre

published descriptions is far from certain. The first descrip-
tion of the geology along the Rio Acre (ONERN, 1977) is
brief as are descriptions of the geology to the south of the
Rio Acre in Madre de Dios Province, Peru (Oppenheim,
1946, 1975; ONERN, 1972) or to the north of the Rio Acre
in the state of Acre, Brazil (Oppenheim, 1937; Paula Couto,
1978). The descriptions of geology are all similar and may
represent the same sequence.

Oppenheim ( 1 946) named the Madre de Dios Formation
for a thick succession of clays and sandstones that are exposed
in high river banks along the Madre de Dios River from the
Bolivian border to the Inambari River and along the lower
part of the Inambari River basin. This is evidently the for-
mation that is called the “Montana Formation” in the geo-
logic map included in the same paper. ONERN (1972) calls
most of this surface stratum simply “Quaternary alluvials”
but does use the term Madre de Dios Formation for expo-
sures along the upper Madre de Dios and Inambari rivers.
Following Oppenheim (1946), these are given a later T ertiary
age designation without elucidation. Either part of the Madre
de Dios Formation of Oppenheim { 1 946) is now called Qua-
ternary alluvium by ONERN (1972) or Oppenheim (1946)
did not include or overlooked the Quaternary alluvium when
he described the geologic section and ONERN included it as
the uppermost, most surficial deposit on their geologic map.
Oppenheim (1975) continued to use the Madre de Dios For-
mation for exposures in the lower part of the Madre de Dios
River after his original usage. From the descriptions of both
Oppenheim (1946) and ONERN (1972), the Madre de Dios
Formation could include both the Ihapari Formation (in its
restricted usage) and the Tertiary red beds. If this proves to
be the case, the name Madre de Dios has priority over the
name Ihapari Formation and should be used over all of Madre
de Dios Province, Peru, and perhaps into Acre, Brazil. In
this paper, as in Campbell and Frailey (1984), the identifi-
cation of separate Madre de Dios and Ihapari formations by
ONERN (1972, 1977) is followed and the uppermost strata
along the Rio Acre are referred to the Ihapari Formation.

Little has been written of the Tertiary geology north of the
Rio Acre. Oppenheim (1937) described riverbank outcrops
of light-to-dark red clays and sandstones with light, greenish
bands of limestones or calcareous clays along the Rio Jurua
in Acre, Brazil, as the Cruzeiro Red Beds. The description
is similar to the appearance of the Ihapari Formation along
the Rio Acre and could be the same formation. On the basis
of the fossils found in a localized overlying deposit, Oppen-
heim (1937) placed the Cruzeiro Red Beds as pre-Pliocene
in age and correlative in part to the Red Beds of Singewald
(1927) and the Formacion de Puca of Steinmann (1929).
Singewaid’s Red Beds were described from along the Ucayali
drainage and produced a new genus of pyro there, Griphodon
peruvianus (Anthony), 1924, which Patterson (1942) placed
as equal to or slightly later than Mustersan (Middle Eocene).
This is much older than the fossils found in the Tertiary red
beds of the Rio Acre and if correct means that the Ihapari
Formation is much younger than the Red Beds of Singewald
(1927, 1928), the Puca Formation of Steinmann (1929), or
the Cruzeiro Red Beds of Oppenheim (1937).

Paula Couto (1978) includes a field sketch of the upper
Jurua River that was made by G.G. Simpson in 1956. This
section is very similar to the Rfo Acre section (Campbell and
Frailey, 1984). A conspicuous element in Simpson’s section
is a “heavy conglomerate” at the position of the Acre Con-
glomerate Member in the Rio Acre section. Above the “heavy
conglomerate,” Paula Couto places two units, yellow-green-
ish clays and silts that are overlain by fine, buff sands. Below
the conglomerate is a “Puca type” stratum that easily cor-
responds to the red beds of the Rio Acre. The use of the
phrase “Puca type” is not explained by Paula Couto but
evidently refers to the thick Puca Formation of Steinmann
(1929) which Steinmann thought was the same as Singewaid’s
Red Beds and which Oppenheim (1937) furthermore thought
was correlative with the Cruzeiro Red Beds. If this is a true
correlation, these beds are of Mustersan age (at least in part)
on the basis of the included pyrothere.

Either similar deposits were formed over an extensive time
period of Middle Eocene to Late Miocene or the “Puca type”
stratum of Paula Couto (1978) is not correlative with any of
the Puca Formation (and Singewaid’s Red Beds) but is in-
stead much younger than either. On the other hand, perhaps
the Griphodon was actually found in much older deposits
that were mistaken for red beds, and the Red Beds of the
Ucayali River, the Puca (or part of it), and “Puca type” beds
are in fact correlative after all. A third possibility, that Gri-
phodon is not Mustersan and the correlation of red beds is
accurate, depends on the unlikely possibility that a primitive
pyrothere survived in the Amazon Basin long after the Oli-
gocene extinction of this group elsewhere in South America.

The Red Beds and the Cruzeiro Red Beds are not the
uppermost strata listed in the sections of Singewald (1928)
and Oppenheim (1937). Above the Red Beds of Singewald
are the unfossiliferous Brown Beds and above the Cruzeiro
Red Beds of Oppenheim is the localized Aquidabam deposit
that he thought was probably Pliocene in age. Patterson ( 1 942)
gave a probable correlation of these with each other and with
the Parana deposits of Argentina. The correlation with the
Parana is based on Tertiary invertebrates in the Aquidabam
deposit, structural folding in the Brown Beds (hence depos-
ited prior to and folded by the Andean uplift), and fossil
vertebrates in the Parana. This correlation was initially un-
certain and made even more so with recent opinions on
mixed taxa in the Parana fauna (Pascual and Bondesio, 1 968).

All of these formations would presumably fall into the
thick and rather broadly defined Late Tertiary Grupo Chi-
riaco of Williams (1949). ONERN (1977) places this group
in the vicinity of the Rio Acre although not actually exposed
along the river.

The angular unconformity that occurs near the low-water
line of the Rio Acre marks the boundary between the Tertiary
red beds and the sediments of the Holocene Inapari For-
mation. This boundary has been noted in numerous publi-
cations on Amazonian geology in the vicinity of the Rio Acre
(Kummel, 1948; Riiegg and Rosenweig, 1949; Mousinho de
Meis, 1971; Paula Couto, 1978; RADAMBRASIL, 1976,
1977) although the identities and ages of the formations in-
volved could not be confidently established. Recent publi-

Contributions in Science, Number 374

Frailey: Fossil Mammals of the Rio Acre 7

Figure 4. Marsupialia, LACM 1 17501, right lower tooth. A, ste-
reoscopic occlusal view; B, labial view; C, lingual view.

cations (Campbell and Frailey, 1984; Campbell et al., 1985)
indicate that the formational relationships are consistent over
a sizeable portion of the western margin of the Amazon Basin
and regional correlation, at least, is not hopeless despite the
present stratigraphic confusion that is outlined here.

SYSTEMATICS

Rio Acre Local Fauna
Late Miocene

Order Marsupialia Illiger, 1811

Family Indeterminate

Figure 4

MATERIAL. LACM 1 17501, lower tooth.

LOCALITY. LACM 4611.

DESCRIPTION. This lower tooth is flat and wide with a
small trigonid and a large, broadly basined talonid. All the
cusps are low and rounded. The cusps of the trigonid are
united by a narrow U-shaped crest that encompasses a shal-
low trigonid basin that is partially filled by a crest that runs
lingually from the protoconid. The metaconid is slightly higher
than the protoconid (the paraconid is missing due to break-
age). The entoconid and hypoconid are of equal height. The
hypoconulid is a widened portion of the posterior crest of
the talonid basin and is situated immediately posteromedial-
ly from the entoconid.

DISCUSSION. The presence of the Marsupialia is indi-
cated by a single tooth that is not referable to known mar-
supials. This tooth has characteristic marsupial features in
that the trigonid is compressed anteroposteriorly and the
hypoconulid lies near the entoconid. However, unlike all

Figure 5. Stenodon campbelli n. gen. and sp., LACM 1 17502, skull,
holotype, ventral view (photograph).

previously described marsupial teeth, this specimen has low,
rounded cusps and a broadly basined talonid which are al-
together very procyonid-like in appearance. This tooth may
represent the existence of a marsupial family that was eco-
logically equivalent to the Procyonidae in South America.
Their fate upon the entry of true procyonids in South Amer-
ica in the Huayquerian is as unknown as their history prior
to their inclusion in the Rio Acre Local Fauna. However, in
some aspects such as crown height and degree of trituberculy,
this tooth is similar to molars of the Prepidolopidae from
the Argentinian Eocene (Pascual, 1981). The little that is
known of the Prepidolopidae or this marsupial group from
the Miocene of the Amazon Basin unfortunately precludes
more than tentative considerations of relatedness.

8 Contributions in Science, Number 374

Frailey: Fossil Mammals of the Rio Acre

Figure 6. Stenodon campbelli n. gen. and sp., LACM 1 1 7502, skull,
holotype, ventral view (reconstruction).

Figure 7. Stenodon campbelli n. gen. and sp., LACM 1 17502, skull,
holotype, dorsal view (reconstruction).

Order Edentata Cuvier, 1798
Family Mylodontidae Gill, 1872
Subfamily Mylodontinae Gill, 1872

Stenodon campbelli, new genus and species

Figures 5, 6, 7, 8, 9A-C

HOLOTYPE. LACM 117502, skull.

TYPE LOCALITY. LACM 4418.

REFERRED MATERIAL. LACM 117503, ulna; LACM
117504, femur; LACM 117505, astragalus.

GENERIC DIAGNOSIS. Dorsal margin of skull is slight-
ly convex and higher than the nasals. The palate extends
anteriorly to M! and is bulbous. Tooth rows diverge slightly
toward the anterior. Teeth are ellipsoid in cross section with
no posterior lobation of M5.

ETYMOLOGY. steno, Gr., narrow; odon, Gr., m. tooth.

SPECIES DIAGNOSIS. Same as for the genus.

ETYMOLOGY. Patronymic, in honor of Dr. Kenneth E.
Campbell, Jr.

DESCRIPTION. Skull. The specimen is fragmented,
crushed and distorted to the left. The malar is missing. The
extreme fragmentation of the skull and its calcareous in-
crustation permits discussion of only the gross features of the
skull.

Dorsally, the skull appears tubular in shape, similar to
Paramylodon but shorter, with postorbital and preorbital
constrictions. Supraorbital bosses are prominent. A low and
wide (10 mm) sagittal crest is indicated between the large
temporal fossae. Prominent lambdoidal and occipital crests
are present and confluent. Thin, paired crests are present
laterally on the occipital and reach the lambdoidal crest. The
occipital is hemispherical and slopes forward. The occipital
condyles are large, bulbous and continue on the ventral line

Contributions in Science, Number 374

Frailey: Fossil Mammals of the Rio Acre 9

Figure 8. Stenodon campbelh n. gen. and sp., LACM 1 17502, skull,
holotype, lateral view (reconstruction).

of the occipital to form the most posterior part of the skull
and extend ventrad as far as the pterygoids.

In lateral view, the dorsal margin is flat or somewhat con-
vex and the nasals are slightly lower than the dorsal margin.
The zygomatic process of the squamosal is wider toward the
anterior, free end and narrows toward the squamosal. The
lacrimal protuberance and lacrimal foramen are large and
situated above M2. The hard palate is approximately 10 mm
ventrad to the alveolar borders or M2-5.

Ventrally, the skull widens anteriorly from its narrowest
point immediately posterior to M5 (67 mm) to its widest
point across the predentary bosses (128). The distance be-
tween the left and right M5s is 32.5; between M's it is 57.3.
The predentary portion of the palate is inflated and bulbous
with a shallow fossa at the midline. The postpalatine fossa
is large, broadly U-shaped and bounded laterally by widely
flaring pterygoids. The basioccipital is trapezoidal in shape.

On the posterior part of the basioccipital, the area of origin
of the M. rectus capitus ventralis is deeply sculpted. Occipital
condyles are fully separate from other basicranial structures.
The paroccipital process and mastoid process are equal in
size and do not extend ventrad to the basioccipital. The
groove for the carotid artery is open. The posterior lacerate
foramen is small, barely larger than the carotid canal. The
condyloid foramen is compressed into a slit rather than the
circular foramen more normal in mylodontids, but this may
be due only to the distortion of the fossil.

Dentition. All teeth are missing on the specimen and the
following discussion is based on the alveoli.

Five upper teeth were present, here designated M1'5. All
the teeth are remarkable in their simple, ellipsoid occlusal
shape with lobation indicated in only one tooth, M1. In M',
the lateral half of the alveolus is enlarged. The second and
third upper alveoli are flexed with a concave border toward
the anterior. The labial margins of the alveoli of M 1-5 lie on
a curving line which converges on the midline of the palate.
Orientation of the teeth also changes progressively toward
the posterior. The M1 is directed posterolabially and M5 is
almost transverse to the midline of the palate. The teeth are
approximately equally spaced with a slight increase in sep-
aration toward the anterior. Measurements of the alveoli
(length x width): M1, 12.8 x 32.5; M2, 12.8 x (24); M\

12.4 x 23.9; M4, 12.6 x 23.5; M5, 12.8 x 20.0. Measure-
ments between alveoli: M1-2, 12.2; M2-3, 8.2; M3-4, 7.6;
M4-5, 7.2. Length of tooth row, 95.

Postcranial skeleton. The astragalus (Fig. 9A-C) has a low
odontoid process which meets the external trochanter at an
approximate 90° angle. The external trochanter is biconvex
when viewed laterally and forms a straight articular surface
when viewed tibially. The fibular facet is planar and extends
between the middle part of the external trochanter and the
distal part of the lateral calcaneal facet. The lateral calcaneal
facet is oval in outline and concave. The proximal end comes
within 10 mm of the proximal end of the external trochanter.
The medial calcaneal facet is a ventral continuation of the
cuboidal articular surface. It is planar and more rectangular
in outline than the lateral calcaneal facet. The calcaneal facets
are well separated (10 mm). The articular surfaces for the
navicular and cuboid are about equal in size. The navicular
surface is concave externally and convex medially. The cu-
boidal facet is triangular in shape and convex. Measurements:
odontoid process (internal trochanter), height, 3 1 , length, 42. 1 ;
external trochanter (length x width), 99.7 x 42.0; fibular
facet, 37.0 x 33.3; navicular facet, 55.9 x 33.1; cuboid facet,

40.4 x 40.0; external calcaneal facet, 66.5 x 38.4; internal
calcaneal facet, 42.8 x 32.8.

DISCUSSION. A large sloth is represented in the fauna
by a skull and several incomplete limb elements. This sloth
is placed in the Mylodontidae on the basis of 1) its large size
which is equal to that seen in Glossotherium; 2) the long,
cylindrical skull; 3) the presence of five molariform teeth
which are subequal in size and subcylindrical in shape; and
4) evenly spaced teeth (i.e., there is no anterior tusk).

The features of the astragalus further substantiate the fa-
milial identification. Astragali with the internal trochanter

10 Contributions in Science, Number 374

Frailey: Fossil Mammals of the Rio Acre

Figure 9. Stenodon campbelli and Megalonychidae. A-C. Stenodon campbelli n. gen. and sp., LACM 1 17505, left astragalus. A, Fibular
view; B, trochlear view; C, calcaneal view, anterior to left. D, Megalonychidae, LACM 117534, distal end of left humerus, anterior view.

modified into an odontoid process are characteristic of My-
lodontidae and Megatheriidae. This astragalus, unlike those
of the Megatheriidae, has a straight (rather than semicircular)
external trochanter and a gently concavo-convex navicular
facet rather than a deep navicular pit. Completely separate
calcaneal facets as in Stenodon is a primitive sloth feature
found in the Scelidotheriinae and in most Mylodontinae (fused
in Glossotherium Owen, 1840; Mylodon Owen, 1840; and
Paramylodon Brown, 1 903). The astragalus of Stenodon is
differentiated from those of the Scelidotheriinae and resem-
bles the Mylodontinae in that the cuboid and navicular facets
are broadly continuous rather than sharply differentiated as
in the Scelidotheriinae, and that the cuboidal facet is convex
rather than concave.

This sloth can further be assigned to the Subfamily My-
lodontinae, rather than the Scelidotheriinae, on the basis of

1 ) a forwardly sloping occiput; 2) a hard palate that is ventral
to the level of the alveoli; 3) an anteriorly widening palate
in which the tooth rows diverge as in Glossotherium and
Glossotheridium Kraglievich, 1934; 4) a low and rounded
paroccipital process; and 5) a groove, rather than a foramen,
for the internal carotid artery.

The specimens are not, however, referable to any of the
known genera of the Mylodontinae and therefore represent
something new. In this genus, all the teeth are subcylindrical
in shape with only a slight tendency to bilobation. The M5
lacks a posterior elongation. The characters indicate that this
genus was more primitive than either Glossotherium or Glos-
sotheridium. The rostrum extends anteriorly to M' and is
bulbous. The teeth appear to have been more compressed in
shape (anteroposteriorly) than in the other mylodonts. This
character is more like that of the Scelidotheriinae and is either

Contributions in Science, Number 374

Frailey: Fossil Mammals of the Rio Acre 11

Figure 10. Pampatheriinae, gen. and sp. indet., LACM 117506,
lower tooth, photograph of lateral side and drawing of occlusal sur-
face.

convergent on the Scelidotheriinae (as I believe) or represents
a shared ancestral (plesiomorphic) character. The dorsal mar-
gin of the skull is essentially flat but slopes forward at the
rostrum as is also seen, although more dramatically, in Glos-
sotherium and Glossotheridium. Stenodon probably is related
to Glossotheridium and G/ossotherium in the Mylodontinae.
It is, however, more primitive than either of these respec-
tively Pliocene and Pleistocene genera in many features al-
though it had developed its own characteristic, derived fea-
tures. As such, Stenodon cannot be placed in a directly
ancestral position to the later genera.

Family Megalonychidae Zittel, 1893

Genus and Species Indeterminate

Figure 9D

MATERIAL. LACM 1 17534, distal end of humerus.

LOCALITY. Sandbar in Rio Acre, no stratigraphic data.

DESCRIPTION. An unusually small ground sloth is iden-
tified by a humerus fragment. The external trochlea is round,
bulbous and has a slight sagittal ridge medially. The internal
trochlea is as broad (anteroposteriorly) as the external troch-
lea. The angle formed between the trochlea when the humerus
is viewed anteriorly is shallow and much less than that seen
in the Megatheriidae. The olecranon fossa is shallow. There
is no evidence of an entepicondylar foramen on the portion
of the humerus that is preserved.

DISCUSSION. The features of this specimen are similar
to humeri of Hapalops although the Rio Acre specimen is
about 50 percent larger than any known species of that genus.
A more striking resemblance, including correct size corre-
spondence, is seen between this fragment and humeri of the
small Pleistocene sloths of the West Indies, Mesocnus Mat-
thew, 1919 and Parocnus Miller, 1929 (but not Acratocnus
Anthony, 1916). On the material available, these humeri are
inseparable. The origin of these small Caribbean ground sloths
is conjectural but their antecedents may have dispersed to
the islands from continental South America during the Mio-
cene or perhaps early Pliocene (Paula Couto, 1967a). This
suggestion is based primarily on the similarity of the Carib-

c

Figure 11. Erethizontidae, gen. and sp. indet., LACM 11 7507, right
M1 or M2. A, lingual view; B, anterior view; C, occlusal view.

bean Megalonychidae to the Early Miocene Megalonychidae.
The Rio Acre megalonychid suggests an alternative possi-
bility in that small megalonychids could have existed in the
Amazon Basin throughout the late Tertiary and migrated to
the Caribbean islands at any time that conditions permitted,
even later than the Early Pliocene. Unfortunately, the Rio
Acre specimen was not found associated with other fossils.
It may in fact not derive from the Tertiary red beds although
its preservation is the same as fossils that were found in place
and is here considered to be part of the same local fauna.

Measurements are: width across trochlea, 58; anteropos-
terior length of external trochlea (capitulum), 33; maximum
length of internal trochlea, 30.

Infraorder Cingulata Illiger, 1811
Family Dasypodidae Bonaparte, 1838

Subfamily Pampatheriinae
Paula Couto, 1954

Genus and Species Indeterminate

Figure 10

MATERIAL. LACM 1 17506, lower tooth.

LOCALITY. LACM 4611.

DESCRIPTION. This tooth is identified as a pampathere
on the basis of its elongate cross section, dumbbell-shaped
lobation and lack of branching of the central osteo-dentine
plate. The degree of molarization places this as one of the
posterior molariform teeth in the lower tooth series, M5-M8.
Measurements, length x width, are 10.2 x 5.0.

DISCUSSION. This tooth is small for a molariform tooth
of Kraglievichia Castellanos, 1927, and more comparable in

12 Contributions in Science, Number 374

Frailey: Fossil Mammals of the Rio Acre

Figure 12. Dasyproctidae, gen. and sp. indet. A, LACM 1 17574, left lower tooth, occlusal and labial views; B, LACM 1 17576, right lower
tooth, occlusal and labial views; C, LACM 1 17577, left M1 or M2, lingual and occlusal views.

size to Vassalia Castellanos, 1927, the earlier (Huayquerian)
pampathere (Castellanos, 1927).

Order Rodentia Bowdich, 1821
Family Erethizontidae Thomas, 1897
Genus and Species Indeterminate

Figure 1 1

MATERIAL. LACM 1 17507, M1 or M2, right.

LOCALITY. LACM 4611.

DISCUSSION. A very small porcupine is represented in
the Tertiary local fauna by a single upper tooth. The tooth
is relatively high-crowned for early porcupines and the crests
are thin and well defined. The metaloph and posteroloph
have united to form a deep oval enamel lake but the pro-
toloph and anteroloph remain free. The sharpness of the
crests and simplicity of the crown pattern is more similar to
Erethizon Cuvier, 1822, or Coendu Lacepede, 1799, than to
Steiromys Ameghino, 1887, and earlier Erethizontidae.

Family Dasyproctidae Smith, 1 842

Genus and Species Indeterminate

Figure 12

MATERIAL. LACM 117574, left P4; LACM 117575, left
P4; LACM 1 17576, right ?DP4; LACM 1 17577, left P4.

LOCALITY. LACM 4611.

DESCRIPTION. Lower teeth. Two lower cheek teeth are
represented in two stages of wear. The teeth are rounded
anteriorly and one (LACM 1 17575) has a wear facet on the
posterior surface which indicates that these teeth are P4s. In
the less worn of the two (LACM 1 17574), the posterolophid
is separated from the anterior lophids by a hypoflexid that

is deep on the labial margin but does not extend to the base
of the crown on the lingual margin. Two additional flexids
are present on the lingual margin that probably correspond
to the mesoflexid and anteroflexid (assuming that the meta-
flexid was incorporated in the hypoflexid when the latter
expanded lingually). On this species, the mesoflexid is per-
sistent during all wear stages of the tooth. The anteroflexid,
on the other hand, would soon disappear beyond the stage
exhibited by LACM 1 17575. An ectolophid, metalophid, and
anterolophid are recognizable. The ectolophid parallels the
posterolophid and is equal to the posterolophid in size. With
wear, as in LACM 1 17575, the ectolophid and posterolophid
become joined lingually (around the shallow end of the hy-
poflexid) and centrally. The metalophid is club-shaped (en-
larged toward its labial end) in these wear stages and joins
the anterolophid, lingually, to create an anterofossettid from
the anteroflexid. From the present height of these teeth and
their advanced stage of wear, the teeth apparently were hyp-
sodont and may never have formed roots.

An isolated lower tooth (LACM 1 17576) is tentatively
referred to this taxon as a DP4. It is only slightly worn, shows
no wear facets from adjacent teeth, and the lophid pattern
is complex and only partially discemable. However, this pat-
tern and the placement and relative heights of the flexids are
comparable to those of the previously described lower teeth.

Upper teeth. The more quadrangular occlusal outline and
the curvature (a concave buccal surface) of LACM 1 17577
indicate that it is an upper tooth. A wear facet is present on
the posterior surface only and suggests that this is a P4. The
anteroloph is separated from the posterior portion of the
tooth by a long hypoflexus that appears to replace the para-
flexus labially. The hypoflexus extends to the base of the
crown on the lingual side. The labial extent of the hypoflexus
is less than half the height of the unworn crown (approxi-
mately 1 mm below this wear stage). Two additional labial

Contributions in Science, Number 374

Frailey: Fossil Mammals of the Rio Acre 13

flexi are present. The posterior flexus (metaflexus) is repre-
sented labially by only a notch on the enamel wall. The more
anterior of the two (mesofiexus) is also indicated only by a
notch in the enamel wall but extends medially as a deep
flexus. The first loph that is posterior to the hypoflexus, the
protoloph, is short and irregular in outline and much larger
toward the labial surface. The protoloph unites with the
metaloph centrally in the tooth. The posteroloph is long and
unites labially with the metaloph and lingually with the com-
bined protoloph and metaloph. Therefore, a large fossette is
enclosed. The irregular shape of this fossette is created in
part by what appears to be an anterior extension of the pos-
teroloph into it. This apparently new loph corresponds to the
neoloph of some authors (see following discussion) and its
appearance here as a new structure may be phylogenetically
significant.

DISCUSSION. These teeth are referred to the Dasyproc-
tidae on the basis of the presence of four distinct lophs and
lophids in which the anteroloph and posterolophid are fully
separate until advanced wear. Furthermore, the posterior
lophs unite and converge on the hypocone and the anterior
lophids unite and converge on the protoconid. These features
are used by Patterson and Wood (1982) in their emended
diagnosis of the Oligocene genus lncamys Hoffstetter and
Lavocat, 1970, from Salla, Bolivia, but these features are still
evident in this much later genus. The distinctness of these
features is remarkable in that the dental similarities are per-
haps closer between lncamys and this unknown genus than
between other known fossil dasyproctids which are closer to
lncamys in age. Furthermore, the similarities are apparent
despite the much larger size and hypsodonty of the teeth
from the identified genus. The two genera differ in their dental
patterns in the placement of the metaloph, which in the Rio
Acre genus unites with the protoloph rather than with the
posteroloph. This feature correlates with the much greater
size of the metaflexus on the Rio Acre specimens. Secondly,
in the upper teeth, lncamys does not have a neoloph but the
Rio Acre form does have a small loph in this position. The
presence of this new loph in a later form but not in one of
the earliest occurring caviomorph genera argues for the prim-
itive presence of only four lophs in Caviomorpha (as in In-
camys) and the later modification of the dental pattern by
the addition of a neoloph as an extension of the posteroloph
(the interpretation of Wood in several papers; see Wood,
1974; and Patterson and Wood, 1982:493-503, for reviews
of this argument).

The alternative interpretation, as expressed by Lavocat
(1974, 1976), is that the primitive caviomorph dental pattern
is five-crested and that the metaloph is a small projection
from the posteroloph. The mesoloph of Lavocat (the meta-
loph of Wood) occupies a central position in the primitive
caviomorph tooth between his protoloph and metaloph. From
this viewpoint, the dental evolution of the Rio Acre dasy-
proctid would not have been from a four-crested Incamys-
type tooth but would have been from a five-crested pattern
in which the metaloph (of Lavocat, the neoloph of Wood)
was reduced. A simpler, four-loph pattern was thus achieved
with a remnant of the metaloph yet remaining (or perhaps

forming again) in this late occurring genus. This four-crested
pattern is thought to be the advanced caviomorph type ac-
cording to Lavocat.

The nearest contemporary dasyproctid is Neoreomys
Ameghino, 1887, of Santacrucian and Friasian ages (Savage
and Russell, 1983). The teeth of Neoreomys are similar to
the Rio Acre specimens in hypsodonty and size and to both
these and lncamys in dental pattern. However, the teeth of
Neoreomys differ markedly in the rapid union of the anter-
oloph and posterolophid with the rest of the enamel pattern
and in the early formation of fossettes and fossettids unlike
the Rio Acre specimens in which these features are only seen
in late wear stages.

Family Caviidae Waterhouse, 1839
Subfamily Cardiomyinae Kraglievich, 1930a
Genus and Species Indeterminate, A

Figure 13

MATERIAL. LACM 1 17508, M1 or M2.

LOCALITY. LACM 4611.

DESCRIPTION. This single tooth is composed of two
cordiform (heart-shaped) prisms of equal size and separated
by a deep internal flexure. Only a faint external flexure, sit-
uated slightly posterior to the termination of the internal
flexure, is present. The enamel is discontinuous on the labial
side of each prism. The tooth is hypsodont and probably
unrooted, although the base is broken in the specimen and
this cannot be clearly determined. In anterior view, the tooth
is strongly convex lingually. Measurement: crown (length x
width), 3.3 x 3.0.

DISCUSSION. The simple biprismatic cordiform pattern
with interrupted enamel identifies this tooth as a specimen
of the Cardiomyinae, the most primitive of the Caviidae
(Pascual et al., 1966). This tooth is smaller than all known
Cardiomyinae and about 25 percent smaller than Procardi-
omys Pascual, 1961.

Subfamily Cardiomyinae Kraglievich, 1930a

Genus and Species Indeterminate, B

Figure 13

MATERIAL. LACM 1 17578, lower tooth.

LOCALITY. LACM 4611.

DESCRIPTION AND DISCUSSION. A second, uniden-
tifiable caviid rodent is represented in this local fauna by a
single, damaged lower tooth. A portion of one prism and one
enamel wall of another together with the cement between is
preserved. The tooth is hypsodont with narrow, cordiform
prisms that are widely separated by a deep hypoflexid. The
hypoflexid maintains a constant width over the length that
is preserved. This genus is not the same as that previously
described (Cardiomyinae, A) because this tooth is taller and
the prisms and hypoflexid wider. Nonetheless, this is a related

14 Contributions in Science, Number 374

Frailey: Fossil Mammals of the Rio Acre

Figure 13. Cardiomyinae. A, Genus A, LACM 117508, right M1 or M2, anterior and occlusal views. B, Genus B, LACM 1 17578, lower

tooth, anterior and occlusal views.

taxon and indicates a diverse cardiomyine fauna was present
at this time.

Family Hydrochoeridae Gill, 1872

Subfamily Cardiatheriinae
Kraglievich, 1930b

Kiyutherium Francis and Mones, 1965

Kiyutherium orientalis
Francis and Mones, 1965
Figures 15, 16

MATERIAL. LACM 117509, M, or M2; 117510, M3;
117511, ?P4; 1 17512, 117513, 117514, 1 17578, M1 or M2;
1 17515, 1 17516, 1 1 75 1 7, M3; 1 17579, tooth fragment.

LOCALITIES. All from LACM 46 1 1 except LACM
1 17578 and 1 17579 which are from LACM 5158.

DESCRIPTION. Lower teeth. A single broken and worn
specimen (LACM 1 17509) is either M, or M2. For this dis-
cussion of characters, an exact identification is not necessary
as characters are essentially the same in these two teeth in
capybaras. Prisms I, Ila, and lib are approximately of equal
width and h.p.i. and h.f.e. (see Fig. 14 for terminology) are
approximately of equal depth. The h.s.i. is shallow, less than
1 mm, but present for the length of the tooth that is preserved.
The h.t.i. is deep, reaching two-thirds the length of Pr. Ila,
and cuts off a small Pr. lie which is triangular in occlusal

outline. The h.p.i. is very deep and reaches the most external
part of the tooth where it is bordered by a thin enamel loop.
The h.f.e. is equally as deep as the h.p.i. The h.s.e. is a shallow,
broad, U-shaped flexid about 1 mm in depth. Measurements
are 8.5 x 5.5 (length x width).

One M3 is preserved in the Rio Acre sample. The M3 is
highly diagnostic in Kiyutherium and in capybaras in general.
The relative proportions of prisms and flexids as described
previously for M, or M, are also seen in M3. The h.s.i. and
h.s.e. are reduced to slight incisions with the h.s.e. slightly
deeper and broader. The h.t.i. extends approximately two-
thirds the length of Pr. Ila and the h.p.i. nearly bisects the
occlusal surface leaving only a narrow enamel loop on its
labial border. The h.f.e. is as deep as the h.p.i. and sharply
bent in occlusal view. The outer surface of the enamel of this
tooth is rugose with small bumps arranged in rows that par-
allel the occlusal surface. Measurements are 11.3 x 7.8
(length x width).

Upper teeth. The upper teeth, P4-M2 and the anterior prism
of M3, are very similar in their structure and will be discussed
as a unit. Each tooth has a deep h.p.e., approximately 2 mm,
that divides the anterior prism into two unequal parts and a
shallow (1 mm deep) h.s.e. Another very shallow flexus (un-
named) is evident in one specimen (LACM 1 17514) on the
external portion of Pr. I posterior to the h.p.e. This flexus is
visible as a shallow groove for the height of the preserved
portion of this tooth. The h.f.i. is deep and reaches the ex-
ternal margin of the tooth where the h.f.i. is bordered by a

Contributions in Science, Number 374

Frailey: Fossil Mammals of the Rio Acre 15

L_

Q.

c

Figure 14. Dental nomenclature of Hydrochoeridae. A, P4-M3, right; B, P4-M,, left; C, M3, left. In order to conform to previous publications
on the Hydrochoeridae, terminology and abbreviations are given in Spanish and follow Pascual and Bondesio (1961) who modified the
terminology of Kraglievich (1940a) and Rusconi (1939).

Lower teeth — h.f.e., hendedura fundamental externa (primary external flexid); h.s.e., hendedura secundaria externa (secondary external
flexid); h.t.e., hendedura terciaria externa (tertiary external flexid). h.p.i., hendedura primaria interna (primary internal flexid); h.s.i. hendedura
secundaria interna (secondary internal flexid); h.t.i. hendedura terciaria interna (tertiary internal flexid); h.c.i. hendedura cuarta interna (fourth
internal flexid). h.s.i. a., hendedura secundaria interna anterior (secondary, anterior internal flexid); h.s.i.p., hendedura secundaria interna
posterior (secondary, posterior internal flexid). h.t.i.pr.s.a., hendedura terciaria interna de la prisma secundaria anterior (this was called the
h.t.i. by previous authors but it is not homologous to the h.t.i. of M,_3). In occlusal view, lower teeth are composed of two moieties called
prisms. The prisms are indicated by Roman numerals. On P4, Prism I separates around a deep h.s.e. in all Hydrochoeridae and creates a third
prism, Pr. Ill or prisma secundaria anterior (pr.s.a.). C indicates a column, a term that is falling out of usage because of its near synonymy
with pnsm but which I have retained in order to designate the characteristic C.3 in P4. Prism II is divided by h.f.i. into an anterior part.
Prism Ila, and a posterior portion. Prism lib. A deep h.t.i. creates another prism, Pr. lie (my usage) from the anterior portion of Pr. Ha in
M,_3 in some genera (ex. Kiyutherium, Cardiatherium, and modified in later genera).

16 Contributions in Science, Number 374

Frailey: Fossil Mammals of the Rio Acre

Figure 15. Kiyutherium orientalis. A, LACM 117510, right M3,
occlusal drawing; B, LACM 1 17509, right M, or M2, occlusal draw-
ing; C-D, LACM 117510, right M,; C, labial view; D, lingual view.

thin loop of enamel. The teeth are hypsodont, with a strong,
labially concave curvature. The posterior prism is more slen-
der than the anterior. All the upper teeth are irregularly pitted
on the labial enamel. The prisms of one specimen (LACM
117511) are less wide than those of the other teeth and this
tooth may be a P4 rather than an M1 or M2. Measurements
(length x width) are: LACM 117511, 7-1 x 5.0; 117512,
7.0 x 5.8; 117513, 8.1 x 7.0; 117514, 8.0 x 5.4.

Three M3s are preserved in this sample although one is
water-tumbled and badly worn (LACM 117517) and another
is broken and preserves only the last six prisms (LACM
117516). The complete M3s (LACM 117515, 117517) have
eight prisms (Pr. I and seven laminae, the last two joined
labially). Prism I has the appearance of those of P4-M2. Prisms
II-VII are narrow bands that are connected labially by a thin
border of enamel. The labial end of Prisms I-VII is marked
by a flexus that is large in Prism I (the h.p.e.) but very shallow
(less than 1 mm) in Prisms II-VII. The prisms gradually
increase in width to Prism VI and then decrease rapidly.
Prism VIII varies in width. In LACM 117515, it is about
one-half the width of Prism VII and about four-fifths the
width of Prism VII in LACM 117516 and 117517. Mea-
surements (length x width) are: LACM 117515, 17.5 x 7.6;
117516, - x 8.6; 117517, 17.8 x (7.3).

DISCUSSION. These specimens are referable to Kiyu-
therium on the basis of laminate lower teeth each with a
minute h.s.i., moderately deep h.t.i., and deep h.f.e. The
lower teeth are very similar to those of the holotype of K.

Figure 16. Kiyutherium orientalis. A, LACM 117514, right M1 or
M2, occlusal drawing; B, LACM 117511, right ?P4, occlusal drawing;
C-E, LACM 117515, left M3; C, occlusal drawing; D, lingual view;
E, labial view.

orientalis (within limits of variation established by Mones,
1975) and reference to this species seems reasonable. How-
ever, in the relative sizes and orientations of some dental
features, the Rio Acre specimens agree more with a specimen
of Kiyutherium from Argentina, a partial cranium and artic-
ulated upper and lower dentitions (Pascual and Bondesio,
1968, 1982). Uncertain of the variation in this species, Pas-
cual and Bondesio (1982) cautiously identified the specimen
as Kiyutherium aff. orientalis.

No P4 is available for comparison but M,_3 of Kiyutherium
in the Rio Acre Local Fauna share several features with Ki-
yutherium aff. orientalis that differ from the holotype. In M,
or M2, for example, the h.s.e. is similarly open, and the depth

Upper teeth — h.p.e., hendedura primaria externa (primary external flexure); h.s.e., hendedura secundaria externa (secondary external flexure),
h.f.i., hendedura fundamental interna (primary internal flexure). Upper teeth are divided by h.f.i. into anterior and posterior prisms, Pr. I and
Pr. II. Additional prisms added to M3 are numbered consecutively beyond II.

Contributions in Science, Number 374

Frailey: Fossil Mammals of the Rio Acre 17

Table 3. Comparative dental features within the Hydroehoeridae.

Upper

teeth

Paleocavia

(Caviidae)

Xenocardia

Cardia-

Anchimysops Kiyutherium therium

Hydrochoe-

ropsis

Protohydro-

choerus

Neochoerus

Hydro-

choerus

P4 h.p.e.

slight

m. deep

?

deep

deep

deep

deep

deep

h.s.e.

m.1 deep

m. deep

m. deep

?

m. deep

deep

deep

deep

deep

h.f.i.

deep

deep

deep

deep

deep

open

open

open

open

M1 h.p.e.

m. deep

m. deep

deep

deep

deep

deep

deep

deep

h.s.e.

m. deep

m. deep

shallow

m. deep shallow

deep

deep

deep

deep

h.f.i.

deep

deep

deep

deep

deep

open

open

open

open

M2 h.p.i.

m. deep

deep

deep

deep

deep

deep

deep

deep

h.s.i.

m. deep

m. deep

shallow

m. deep shallow

deep

deep

deep

deep

h.f.i.

deep

deep

deep

deep

deep

open

deep

open

open

M3 h.p.e.

m. deep

shallow

m. deep m. deep

deep

deep

deep

deep

h.s.e.

m. deep

shallow

shallow shallow

shallow

shallow to

absent

absent

m. deep

h.f.i.

deep

deep

deep

deep

deep

open3

open4

open

open

Number of

prisms

2

6

6

8

10

13

17

16

13

Shape,

excluding

cordiform

cordiform.

lanceolate

lanceolate lanceolate

laminate

laminate

laminate

laminate

Pr. I

slender

Lower

Paleocavia

teeth

(Caviidae)

Phugatherium

Kiyutherium

Anchimys

Procardiatherium

P4 h.f.e.

present

deep

deep

deep

deep

h.s.e.

deep

deep

deep

deep

h.p.i.

present, pos-

deep

deep

deep

reduced

terior to h.f.e.

h.s.i.p.

single, shallow

present, smaller

single, shallow

deepens (as deep

then h.s.i. a.

as h.s.i.

a.)

h.s.i. a.

single, shallow

present

single, shallow

present

h.t.i.

poorly defined

absent

faint

C.3

absent

present

M, h.f.e.

deep

deep

deep

deep

deep

h.s.e.

deep

m.1 deep

slight

m. deep

h.p.i.

present

deep

deep

deep

deep

h.s.i.

shallow

shallow

shallow

m. deep

h.t.i.

deep

shallow

reduced

M, h.f.e.

deep

deep

deep

deep

deep

h.s.e.

deep

small

slight

small

h.p.i.

present, post-

deep

deep

deep

deep

terior to h.f.e.

h.s.i.

small

m. deep

deeper than

small

Phugatherium

h.t.i.

deep

slight

reduced

M3 h.f.e.

deep

deep

deep

deep

deep

h.s.e.

small

small

m. deep

small

h.t.e.

small

footnote 5

h.p.i.

present

deep, narrow

deep

deep

deep

h.s.i.

small

v.2 small

small

m. deep

h.t.i.

deeper

deep

m. deep

Pr. Ila

transverse

transverse

oblique

oblique

oblique

Number of

prisms

2

3, Pr. I thin

3, Pr. I cordiform 3, Pr. II, oblique

3

18 Contributions in Science, Number 374

Frailey: Fossil Mammals of the Rio Acre

Table 3. Continued.

Lower

teeth

Cardiatherium

Protohydro-

choerus

Chapalma-

therium

Hydrochoe-

ropsis

Neochoerus

Hydrochoerus

P4 h.f.e.

deep

v. deep

deep

deep

deep

deep

h.s.e.

deep, anterolingual

deep

deep

deep

deep,

anterolingual

deep,

anterolingual

h.p.i.

deep

deep

deep

deep

deep

deep

h.s.i.p.

deep, as in
Procardiatherium

v. deep

m. deep

deep

deep

deep

h.s.i. a.

present

reduced

deep

m. deep

deep

m. deep

h.t.i.

faint

deep

deep

deep

deep

deep

C.3

present

present

present

ill-defined

present

present

M, h.f.e.

deep

open

open

open

open

h.s.e.

m. deep

m. deep

m. deep

m. deep

m. deep

h.p.i.

v. deep

open

open

v. deep

open

h.s.i.

deep

deep

deep

deep

deep

h.t.i.

deep

deep

deep

open

open

M, h.f.e.

deep

open

open

open

open

h.s.e.

m. deep

m. deep

m. deep

m. deep

m. deep

h.p.i.

v. deep

open

open

open

open

h.s.i.

deeper than
Procardiatherium

deep

m. deep

m. deep

m. deep

h.t.i.

m. deep

v. deep

v. deep

open

open

M3 h.f.e.

deep

open

open

open

open

h.s.e.

small

absent

absent

v. deep

open

h.t.e.

small

open

open

open

open

h.p.i.

v. deep

open

open

open

open

h.s.i.

m. deep

v. deep

deep

v. deep

v. deep

h.t.i.

v. deep

open

open

open

open

Pr. Ha
Number of

oblique

transverse

transverse

transverse

transverse

prisms

3

3

3

3

3

' m. = moderately.

2 v. = very.

3 Last two h.f.i. in series closed.

4 All h.f.i. open except last one in series.

5 Present in “ Eucardiodon .”

of the h.f.e. extensively overlaps the depth of the h.s.i. (par-
ticularly in M,). Also in the M2, Pr. lib is more laminate in
the referred specimens and the anterior lobe is less conspic-
uous than in the holotype. In M3, both share a narrower h.f.e.
(as a result of the deep h.t.i.). These are probably all advanced
features with the exception of the open h.s.e.

The holotype of Kiyutherium orientalis and the Rio Acre
specimens, however, share certain features that differ from
those of Kiyutherium aff. orientalis. For example, the wider
laminae and the lesser obliquity of the h.p.i. on M3 (both
primitive) contrast with the Argentinian specimen. Other
characters that appear to be advanced are found only on the
Rio Acre specimens: the h.p.i. of M,_2 is wider and the h.t.i.
is deeper (i.e., Pr. II is more deeply divided); and the h.s.e.
of M3 is narrower.

In the upper teeth of the Rio Acre specimens, and in par-

ticular M3, the prisms are noticeably more inflated and more
transversely situated (both plesiomorphic characters) than
corresponding features in Kiyutherium aff. orientalis. No up-
per dentition is present with the holotype.

Of the three specimens, those from the Rio Acre Local
Fauna are the smallest but the range does not seem significant
(length of M3 is 11.3 for Kiyutherium from the Rio Acre;
12.0 for Kiyutherium aff. orientalis; and 13.75 for the holo-
type).

The expanded hypodigm of Kiyutherium demonstrates
some variation in dental features. At present, the variation
appears to be within the limits of a species and there is no
consistency in character polarity that would suggest other
than one species. The known specimens of Kiyutherium that
can be identified to the species level are therefore assigned
to K. orientalis.

Contributions in Science, Number 374

Frailey: Fossil Mammals of the Rio Acre 19

N-

Figure 17. Occlusal patterns of genera of Hydrochoeridae discussed in cladograms (Figs. 18, 19). Left, lower right dentition; right, upper
right dentition. Dentitions are not to scale. A, Paleocavia (Caviidae); B, Xenocardia ; C, Anchimysops; D, Phugatherium; E, Anchimys; F,
Procardiatherium ; G, Kiyutherium; H, Cardiatherium; I, Hydrochoeropsis; J, Chapalmatherium; K, Protohydrochoerus; L, Neochoerus; M,
Nothydrochoerus: N, Hydrochoerus. (Sources: Pascual and Bondesio, 1961, 1963, 1968; Pascual et al., 1966.)

20 Contributions in Science, Number 374

Frailey: Fossil Mammals of the Rio Acre

Figure 18. Cladistic relationships among genera within the Hydrochoeridae using features of the upper dentition. A sketch of M3 is included
for each genus. Some genera are known only by the upper dentition.

Kiyutherium was described by Francis and Mones (1965)
from a single ramus from the Kiyu Formation of Uruguay.
This formation was dated faunally by Francis and Mones as
“Mesopotamian.” In the view of Pascual and Bondesio (1968)
the “Mesopotamian” is not a unified faunule but rather is
comprised of several faunules, and correlative in part to the
Huayquerian (Late Miocene) and in part to the Monteher-
mosan (Pliocene). The Argentinian record of Kiyutherium is
from Huayquerian sediments (Pascual and Bondesio, 1968,
1982). A third specimen, a partial M3 with the number of
laminae hypothesized for Kiyutherium (Mones, 1975) and
equal to that seen in M3s from Rio Acre, is briefly mentioned
by Mones (1976) from Huayquerian sediments in Venezuela.
Kiyutherium is then seen to have a wide geographic range,
the broadest known of any of the Cardiatheriinae, that in-
cluded Huayquerian and probable Huayquerian sediments
in La Pampa Province, Argentina (Pascual and Bondesio,
1968); San Jose Department, Uruguay (Francis and Mones,
1965); Estado Falcon, Venezuela (Mones, 1976, a questioned
referral of a single specimen); and now the western margin
of the Amazon Basin.

These new sediments from the Rio Acre permit further
discussion of the position of Kiyutherium within the Car-
diatheriinae. Kiyutherium is acknowledged to be a highly
derived member of the most primitive subfamily of the Hy-

drochoeridae, although probably not in the direct line which
led to the more advanced Subfamilies Protohydrochoerinae
and Hydrochoerinae (Francis and Mones, 1965; Pascual and
Bondesio, 1968). Genera of the Cardiatheriinae are broadly
contemporaneous in the Late Miocene and Early Pliocene
with the morphologically more advanced genera ( Procar -
diatherium Ameghino, 1885, Kiyutherium and Cardiathe-
rium Ameghino, 1883) being the earlier known and longer
ranging (Mones, 1976). Authors have therefore tended to
avoid conventional phylogenetic trees and instead speak of
more primitive and more derived genera (Francis and Mones,
1965; Pascual and Bondesio, 1968; Mones, 1976). The po-
sition of Kiyutherium when viewed cladistically as a pro-
gression of characters within the Hydrochoeridae (Figs. 17,
1 8, and 1 9 and Tables 3, 4, and 5) is therefore roughly equiv-
alent to previous authors’ usages although presentation in
this manner clarifies the relative positions of genera without
introducing implications of ancestry.

The evolutionary development of dental characters (all
that is known of Cardiatheriinae with the exception of the
undescribed partial skull of Kiyutherium) can be seen as an
addition of flexures and flexids that indented the cordiform
prisms inherited from a caviid ancestor, and the anteropos-
terior compression of the prisms that created first lanceolate
and then laminate teeth. The number of laminae of M3 stead-

Contributions in Science, Number 374

Frailey: Fossil Mammals of the Rio Acre 21

Table 4. Characteristics of the upper dentition used in construct-
ing the cladogram of Hydrochoeridae (Fig. 18).

1 . Teeth with 2 prisms, no h.p.e., h.s.e. present, situated posteriorly
to h.f.i., h.f.i. deep, M3 with 2-5 prisms.

2. Caviidae. No h.p.e., M3 with fewer than 6 prisms, prisms cor-
diform.

3. h.p.e. present (strongest on M1-3); M3 has 6 or more prisms;
prisms of all teeth become more slender (lancolate rather than
cordiform).

4. Xenocardia. M3 with 6 prisms, h.p.e. equals h.s.e. in depth on
M'-2.

5. Prisms become more slender, h.p.e. separates Pr. I into anterior
part (Pr. Ia) and small, triangular in cross section, Pr. Ib.

6. Anchimysops. h.f.i. directed posterolabially, deep h.p.e. and
shallow h.s.e.

7. Number of prisms in M3 increases to 8. h.p.e. and h.s.e. deepen.

8. Kiyutherium. Fewer than 1 0 prisms in M3. All flexures shallower
than in sister-group.

9. M3 prisms number 10 or more, h.t.i. present on M1-2, prisms
narrower.

10. Cardiatherium. h.t.i. well developed. Largest of the Cardia-
theriinae.

11. Enamel no longer continuous on labial side of M\ more than
12 prisms on M\ anterior and posterior prisms separated on
Mu2, h.p.i. and h.s.e. equally deep.

1 2. Hydrochoeropsis. h.f.e. not perforate on PJ, 3 posterior laminae
(at least) of M3 connected.

13. h.f.e. perforate on P4, never more than 2 laminae of M3 con-
nected.

14. Chapalmatherium (Protohydrochoerinae). Prisms of P4-M2
widely separated, thick, and directed lingually rather than an-
terolingually. h.p.e. and h.s.i. more medially placed in prism.

15. No labial grooves on M3 (Hydrochoerinae).1

16. Neochoerus. Prisms very thin. M3 with 14-17 prisms. Large size
(twice that of Hydrochoerus).

17. Hydrochoerus. M3 with 12-13 prisms. M2 composed of three
prisms.

1 Ahem and Lance ( 1 980) describe a new species of Neochoerus that

has labial grooves. I suspect this is the first North American record

of Hydrochoeropsis.

ily increases at the generic level although it varies within
limits in individuals (Mones, 1975).

In reference to Figures 18 and 19, Neoanchimys Pascual
and Bondesio, 1961, is synonymized under Phugatherium
Ameghino, 1887, following Mones (1972), and Eucardiodon
Ameghino, 1 89 1 , is synonymized with Anchimys Ameghino,
1886, following Mones (1975). Not hydrochoerus Rusconi,
1935, is poorly known. The revised diagnosis of this genus
(Pascual et al., 1966) relies on features which are highly vari-
able (Mones, 1975). Not hydrochoerus could be referred to

Table 5. Characteristics of the lower dentition used in constructing

the cladogram of Hydrochoeridae (Fig. 19).

1 . h.p.i. present and posterior to h.f.e., no h.s.i., P4 with two prisms
(i.e., no h.s.e.), M3 undifferentiated.

2. Caviidae. Prisms cordiform, h.s.i. absent, h.p.i. shallow, no Pr.
Ill on P4, h.s.e. absent.

3. Prisms narrow, small h.s.i., h.p.i. deep, Pr. Ill present on P4
(deep h.s.e.), slight h.s.e. present.

4. Phugatherium. h.s.e. very deep on M,_2, h.p.i. deep, narrow, Pr.
I thin.

5. Pr. II sits oblique to ramus (rather than transverse), h.s.i. deep-
ens on M|.j, h.t.i. present on M,_3.

6. Anchimys. h.t.e. present.

7. h.t.i. deepens on M,_3, h.s.i. a. and h.s.i.p. present on P4, new
structure, C.3, develops on P4, h.s.e. on M3 reduced.

8. Procardiatherium. h.p.i. on P4 reduced.

9. h.f.e and h.t.i. on M,_3 deepen, h.s.e. and h.f.e. deepen on P4.

10. Kiyutherium. h.s.i., h.s.e. small on M,_3. h.p.i. wide on M,_3.

1 1. h.t.i. deepens on P4, h.p.i. on M,_2 deepens and narrows, h.s.i.
deepens on P4-M3, h.s.e. deepens on M^, laminae thin.

12. Cardiatherium. h.t.i. and h.s.i. on M,_2 only half as deep as in
sister-group. All prisms connected on each tooth.

13. h.t.i. deep on P4, Pr. I and Pr. II of M,^3 separate (h.f.e. and
h.p.i. open, only on M2 and P4 of Neochoerus), h.s.i. and h.s.e.
deep.

14. Hydrochoeropsis.

1 5. h.t.i. perforate on M3, h.s.i. on P4-M3 very deep, h.f.e. perforate
on M,_3.

16. Chapalmatherium (Protohydrochoerinae). h.p.i. open on M,_2,
h.s.e. absent, C.6 on P4 directed posterolingually. (Protohydro-
choerus has h.s.i.p. on P4 deep, h.s.i. a. absent, h.s.i. and h.t.i.
very deep on M,_2, h.s.i. very deep on M3, h.i.e. deep and me-
dially directed.)

17. Deep h.s.e. on M3, h.t.i. open on M2.

18. Neochoerus.

19. Nothydrochoerus, Hydrochoerus. h.s.e. on M3 perforate, h.p.i.
open on M2_3. Nothydrochoerus is much larger than Hydrochoe-
rus but may not be a valid genus.

Neochoerus Hay, 1926 (which is the same size), Hydrochoe-
rus, or simply declared a nomen dubium.

Family Dinomyidae Alston, 1876

Subfamily Potamarchinae Simpson, 1945

Potamarchus Burmeister, 1885

Potamarchus murinus Burmeister, 1885

Figures 20, 2 1

MATERIAL. LACM 117518, right P4; GB 677, left P4;
GB 676, M1 or M2; LACM 117580, M3; LACM 117519,

22 Contributions in Science, Number 374

Frailey: Fossil Mammals of the Rio Acre

Figure 19. Cladistic relationships among genera within the Hydrochoeridae using features of the lower dentition. A sketch of M3, if known,
is included for each genus. Arrows denote taxonomic features discussed in the text. Some genera are known only by the lower dentition.

partial ramus with M,_3; LACM 117520, M3; LACM 1 17521,
117522, 1 17581, teeth fragments.

LOCALITIES. LACM 1 17520-2, LACM 4606; LACM
117518-9, LACM 4611; GB 676-677, Rio Acre, west of
Cobija, Depto. del Pando, Bolivia; LACM 1 17580-1 17581,
LACM 5158.

DESCRIPTION. Lower teeth. The lower molars consist
of parallel lophids that are slightly concave anteriorly and
are oriented obliquely to the line of the ramus. In M2_3 and
possibly M, at an earlier wear stage, the metalophid and
hypolophid are not smoothly concave but rather are S-shaped.
Four lophids are present in M,, apparently a reduction due
to wear, and five each in M, and M3. The anterior border of
each lophid is crenulated. Hypoflexids are persistent and
obvious on each tooth (small on M,) but internal flexids are
present only on M3 at this wear stage. The identification of
these internal flexids is difficult but I suggest that the internal
flexids present on M3 are the mesoflexid and anteroflexid
with a new and unnamed flexid situated anteriorly to the
anteroflexid. The occlusal outline becomes increasingly elon-
gate from M, to M3.

An isolated M3 recovered from LACM 4606, LACM
1 17520, is only slightly worn and shows features of Pota-
marchus that were heretofore unknown. Five lophids are

present and all are separate. The anteriormost lophid is di-
vided into two unequal parts that lie in the same transverse
plane. Three ectoflexids and four entoflexids are present. The
hypoflexid is the most persistent flexid (and the most readily
identifiable) and extends to within 2 mm of the base of the
crown. The anterior two ectoflexids are short (2 mm to less
than 1 mm for the more anterior). All the entoflexids are
short and would disappear after one-third of the crown was
worn away. The posteriormost entoflexid appears to be the
remnant of the metaflexid. The heights of the entoflexids,
from the anterior, are 3.0, 4.6, 4.4, and 2.0. Crown height is
14.8.

Measurements of lower teeth are as follows: M|, 6.3 x 5.6;
M2, 7.5 x 6.4; M3, 8.5 x 6.7 (LACM 117519). M, (GB 677),
6.8 x 5.7; M3 (LACM 1 17520), 8.9 x 6.0 (measured in the
middle of the tooth, a position that corresponds to the wear
stage of the other specimens).

Upper teeth. The least worn P4, LACM 117518 (Fig. 21 A,
B), consists of eight oblique lophs that are gently concave
towards the posterior. The anterior five lophs extend the
width of the tooth; the sixth reaches only three-quarters the
width, the seventh only about half the width of the crown,
and the eighth is minute. The lophs are approximately the
same thickness with the exception of the posterior two lophs

Contributions in Science, Number 374

Frailey: Fossil Mammals of the Rio Acre 23

Figure 20. Potamarchus murinus. A, LACM 1 17580, left M3; B, GB 676, left M1 or M2; C, GB 677, left P4. Lingual and occlusal views of
each.

that are thin and would have been the first lophs to be worn
away. The anterior enamel rim of each loph is thicker than
that of the posterior. The posterior enamel rim is highly
crenulated and variable in thickness. The posterior five lophs
are confluent both labially and lingually. With wear the an-
terior lophs would progressively become united with the pos-
terior block. The occlusal outline is rectangular. A single
broad, shallow groove is present on the vertical midline of
the labial surface. On the lingual surface, a fissure is present
after each of the first three lophs. At this wear stage, the
anteriormost fissure extends to the base of the crown; the
second approximately half this length; and the third approx-
imately one-fifth the distance. Cement is present between the
lophs. Four roots have formed; the labial roots are situated
more toward the crown than are the lingual roots (lateral
hypsodonty). Measurements are 8.7 x 7.6, length x width.

A single isolated tooth (GB 676) is identified as a left M2.
This tooth consists of three complete and one broken lamina
with the postero-extemal portion of the crown missing. The
anterior two laminae are convex to the anterior and the pos-
terior two have a gentle S-curve. The posterior border of each
loph is thin and crenulated. A large, antero-intemal flexus is
present for nearly the entire length of the crown. Two minor
flexi persist, one internal and one external, but each less than
1 mm deep at this wear stage with about half the tooth gone.
Measurements are 9.6 x 6.9, length x width.

The M3 in this sample, LACM 117580, has seven lophs

although the posterior two are not fully formed in this early
wear stage. The thin posterior enamel wall of each loph is
crenulated but not to the extent of other teeth in the sample.
Morphological variation rather than variation due to wear
is indicated in that the thin posterior enamel walls of the
equally little-worn P4 (LACM 117518) are highly crenulated.
The lophs are closely appressed with only a faint indication
in the posterior lophs of what may with wear become S-shaped
flexures. The anterior four lophs are united labially by the
enamel wall. The posterior two are united by enamel on all
margins of the occlusal surface. The labial flexi are short and
with only an additional 1 mm of wear all the lophs would
be united on their labial surface. On the lingual surface of
M\ the flexi extend to the base of the crown. The anterior
portion of the crown is much higher than the posterior and
the lingual enamel surface is longer than the labial surface.
Five roots are present. Two large anterior roots are connected
medially and support the anterior portion of the crown. Be-
hind this root crest are two smaller, medial roots and a large
root that is situated under the posterior apex of the crown.
On the lingual margin of the tooth, lophs 1 and 2 are sup-
ported by the anterior root; 3 and 4 by the medial root, and
5, 6, and 7 by the posterior root. On the labial margin, lophs
1, 2, and 3 are supported by the anterior root; 4, 5, and 6 by
the medial root; and 7 is supported by the terminal root.
Measurements of M3 (length x width) are 8.9 x 6.9.

24 Contributions in Science, Number 374

Frailey: Fossil Mammals of the Rio Acre

Figure 21. Potamarckus murinus. A-B, LACM 1 17518, right P4; A, lateral view; B, occlusal drawing; C, LACM 1 17520, right M3, occlusal
drawing; D-E, LACM 117519, ramus with M,_3; D, occlusal drawing; E, labial view. A and E are to the same scale.

DISCUSSION. These specimens are referred to the Po-
tamarchinae because the teeth are hypsodont and comprised
of closely appressed, parallel lophs and lophids that in the
lower teeth are convex posteriorly, and in which the hypo-
flexid and hypoflexus are the most persistent re-entrants.

In the Potamarchinae, these teeth resemble those of Pot-
amarchus in having the lophids of the lower teeth fully sep-
arated laterally and medially when unworn but united lat-
erally (except for the hypoflexid) with early wear. In the upper
teeth, the degree of obliqueness of the lophs is less than in
the lowers. These teeth differ from those of Simplimus
Ameghino, 1904, the only other genus of Potamarchinae, in
having lophs that are more oblique and that are closely ap-

pressed with no cement evident between them (see Krag-
lievich, 1930c).

The specimens from the Rio Acre have numerous crenu-
lations in the thin anterior enamel blade of each lophid that
are characteristic of Potamarckus murinus Burmeister, 1885,
and not P. sigmodon Ameghino, 1891. Furthermore, the lo-
phids appear not to be as S-shaped as those of P. sigmodon
although a definite S-shape is discemable. Considerable vari-
ation in enamel crenulation and lophid curvature was ap-
parent in the small sample recovered from sites along the
Rio Acre and as a result species referral is somewhat dubious.
With additional specimens, the features that are used to sep-
arate these two species may prove invalid.

Contributions in Science, Number 374

Frailey: Fossil Mammals of the Rio Acre 25

1 c m

I. . -J

Figure 22. Tetrastylus sp., LACM 1 17535, right P4. A, lingual view;
B, labial view; C, occlusal drawing.

Potamarchus murinus is found in the Arroyo Chasico For-
mation (of Chasicoan age) and possibly in the Epecuen For-
mation (Huayquerian age) (Pascual et al., 1966). Potamar-
chus sigmodon is probably contemporaneous but is found in
the “Parana fauna” (Ameghino, 1891). The Parana fauna is
difficult to date precisely due to an apparent mix of Miocene,
Pliocene, and even some Pleistocene taxa (Pascual and Bon-
desio, 1961). The age indicated by the majority of the taxa
is Huayquerian and possibly Montehermosan (Pascual and
Odreman Rivas, 1971).

Subfamily Eumegamyinae Kraglievich, 1932
Tetrastylus Ameghino, 1886

Species Indeterminate

Figure 22

MATERIAL. LACM 1 17535, right P4.

LOCALITY. LACM 4611

DESCRIPTION. This P4 is lophate, hypsodont, and con-
cave on the anterior face. The occlusal surface consists of
four parallel lophids that increase in width toward the pos-
terior (measurements, 2.6, 4.0, 5.0, 5.3) and also in thickness
(0.5, 0.9, 1.1, 1.5). The lophids are closely appressed with a
thin sheet of cement between them. The anterior two lophids
are connected labially. All the other lophs are fully separate
on the preserved part of the crown in this specimen. Mea-
surements, length x width, are 9.0 x 5.5.

As defined by Pascual et al. (1966), four arcuate lophids
that are united on the labial surface only and that decrease
in size anteriorly are characteristic of lower teeth of the Eu-
megamyinae. The size of this tooth dictates its assignment
to the smallest genus of the subfamily, Tetrastylus. Tetra-
stylus is a Chasicoan and Huayquerian genus (Pascual et al.,
1 966) and serves as a good time indicator for the local fauna.

Telicomys Kraglievich, 1926a

Telicomys amazonensis new species

Figures 23, 24, 25

HOLOTYPE. LACM 1 17523, partial skull, missing zy-
gomatic arches, nasal and premaxillary bones, and incisors.

Table 6. Comparative measurements between Telicomys amazo-
nensis and Telicomys giganteus (measured from Rovereto, 1914).

T. T.

giganteus amazonensis

Condylobasal length

297

(263)

Occiput; height (condyle to inion)

74

62

width, maximum

124

84

Maxillary tooth row: length

71

(65)

width of palate between anterior

margins of P4s

7.6

9.6

width of palate between poste-

rior margins of M3s

45

47

Width of braincase

80

(90)

Least interorbital width

92

(75)

Palatal length, from premaxillary-
maxillary suture to anteriormost
point on posterior edge of pal-

ate

105

(96)

Postpalatal length

99

77

Tympanic bulla; length

34

29

width

18

18

P4 (length x width)

16.8 x 12.4

15.3 x 10.8

M1 (length x width)

16.1 x 11.1

10.6 x 10.7

M2 (length x width)

14.8 x 11.4

1 1.2 x 10.0

M3 (length x width)

16.5 x 11.4

15.5 x 9.6

TYPE LOCALITY. LACM 4418.

DIAGNOSIS. Smaller than Telicomys giganteus (Ameghi-
no) 1 904, with narrower occiput and shorter postpalatal length.
P4 approximately equal in size to M3 and both are larger than
M1 and M2 which are of equal size.

DESCRIPTION. Skull. In dorsal view, the skull is flat
and wide. The temporal fossae are deep but narrow and
appear in this view as right triangles that do not meet cen-
trally, i.e., there is no sagittal crest.

In lateral view the dorsal margin of the skull is straight,
the depth of the skull increases to a maximum over the tooth
row. The occiput is flat and nearly vertical.

Ventrally, the tooth rows are straight and well separated.
The palate is triangular and vaulted; the postpalatine notch
is V-shaped and terminates posterior to M3. The basisphe-
noid and basioccipital are short, flat, and broad. The auditory
bullae are large but do not contact the paroccipital processes.

In posterior view the occiput is low and wide. The par-
occipital processes are large but short. Lateral to the occipital
condyles are accessory condyles. These are flat, oval articular
processes that are situated between the paroccipital processes
and occipital condyles and are confluent with the latter.

Upper teeth. The upper cheek teeth each have four major
lophs in which the posterior two are connected lingually and
which are separated by thin sheets of cement. The M2 has a
small fifth loph that would be incorporated into the fourth
loph with slightly more wear. The M3 has six lophs, although

26 Contributions in Science, Number 374

Frailey: Fossil Mammals of the Rio Acre

Figure 23. Telicomys amazonensis n. sp., LACM 1 17523, skull, holotype. A, right upper dentition, occlusal drawing; B, palatal view.

the sixth is nearly worn away even at this early wear stage.
The fifth loph of M3 is comprised of two enamel loops that
are not yet confluent into a single loph but probably would
be with more wear. The fourth and fifth lophs are continuous
on the lingual side. The upper premolar and M3 are approx-
imately equal in size and larger than M'-2 which are also
approximately equal in size. On this skull, the teeth are sep-
arated by 2-3 mm, an apparent distortion due to preserva-
tion.

Comparative measurements of the teeth and skull are given
in Table 6.

DISCUSSION. The Subfamily Eumegamyinae was named
by Kraglievich (1932) but not fully defined until much later
(Pascual et al., 1966; Francis and Mones, 1968). In those
features that can be compared to the description of Pascual
et ah, Telicomys amazonensis agrees in having a flat, broad
skull, restricted temporal fossae and short sagittal crest, and
similar shape of the basicranium and occiput. The lateral,
accessory occipital condyles (paracondyles in the terminology
of Pascual et ah, 1966) are unique in this subfamily of Mam-
malia.

The large specimens of two species of Tetrastylus (T. gi-
ganteus Ameghino, 1904, a skull, and T. gigantissimus
Ameghino, 1909, a mandible, both known only from the
holotypes) were put in a new genus by Kraglievich (1926a).
These two species of Telicomys were synonymized without
comment by Pascual et ah (1966) as Telicomys giganteus.
Kraglievich ( 1 926a) stated that the principal differences, oth-

er than size, between Telicomys and Tetrastylus are the rel-
atively wider and lower occiput, position of the postpalatine
notch posterior to M3, and the disproportionately large size
of the incisors in Telicomys. The incisors are missing on the
Rio Acre specimens but the size, occiput, and postpalatine
notch are like Telicomys giganteus.

The usefulness of a wide and low occiput as a taxonomic
character in large rodents is somewhat doubtful in that the
same relative relationship of this feature in Tetrastylus and
Telicomys can also be seen between Castor and Castoroides.
The latter two genera are related and both are Castoridae,
but they are probably less closely related than are Telicomys
and Tetrastylus. Castoroides, the more derived genus in this
feature, does not follow Castor in age (the lineages are in fact
separate from the Middle Miocene to the Pleistocene; Larry
D. Martin, pers. comm.). It may be that the shape of the
occiput is more a function of size and weight of the head
than an indicator of phylogeny. Curiously enough, the mor-
phological similarity continues in that the occipital condyles
of Castoroides are very flat and laterally expanded to provide
articular surfaces similar to the unique accessory occipital
condyles of the Eumegamyinae. This further suggests that
many of the diagnostic features of these large rodents result
from their very large size.

Those features of the posterior half of the skull of Teli-
comys amazonensis that differentiate this species from T.
giganteus are primitive features. The great lateral expansion
of the occiput and postpalatal length of the skull, and the

Contributions in Science, Number 374

Frailey: Fossil Mammals of the Rio Acre 27

4 cm

li — — - ■ --raj

Figure 24. Telicomys amazonensis n. sp., LACM 1 17523, skull, holotype. A, left lateral view; B, dorsal view; C, occipital view. Note that
C is to a different scale.

28 Contributions in Science, Number 374

Frailey: Fossil Mammals of the Rio Acre

Figure 25. Telicomys amazonensis n. sp., LACM 1 17523, skull, holotype (reconstruction). A, palatal view; B, dorsal view; C, right lateral
view; D, occipital view.

Contributions in Science, Number 374

Frailey: Fossil Mammals of the Rio Acre 29

Figure 26. Neoepiblemidae, Heteropsomyinae, and ISynastrapotherium sp. A-B, Neoepiblemidae, gen. and sp. indet., lower cheek teeth,
occlusal drawings; A, LACM 1 17524; B, LACM 1 17525. C, Heteropsomyinae, gen. and sp. indet., LACM 1 17526, right M1 or M2, occlusal
drawing. D, ISynastrapotherium sp., LACM 1 17531, upper canine, lateral and medial views.

greater importance of the dorsal neck muscles that was prob-
ably concomitant with the enlargement of the incisors, is not
evident in Telicomys amazonensis.

The holotype skull of Telicomys giganteus, the type species,
was found in the Montehermoso Formation and the man-
dible, the holotype of^Tetrastylus gigantissimus," was found

30 Contributions in Science, Number 374

Frailey: Fossil Mammals of the Rio Acre

in the Chapadmalal Formation of Argentina (Kraglievich,
1926a). Telicomys therefore has a known temporal distri-
bution of Montehermosan to Chapadmalalan (Early to Late
Pliocene) in temperate South America. This new species from
the Rio Acre Local Fauna extends both the temporal and
geographic ranges of the genus into the Huayquerian (Late
Miocene) of western Amazonia as it adds a second, more
primitive, species to this genus.

Superfamily Chinchilloidea
Kraglievich, 1940

Family Neoepiblemidae Kraglievich, 1926a

Genus and Species Indeterminate

Figures 26A, B; 28B

MATERIAL. LACM 1 17524, 1 17525, isolated lower teeth.

LOCALITY. LACM 4611.

DISCUSSION. Kraglievich (1926a) erected the Family
Neoepiblemidae to include large rodents with ever-growing
teeth in which the teeth are composed of subparallel laminae
that are fully separated by cement. Two genera were included
by Kraglievich, Neoepiblema Ameghino, 1889, and Dabbe-
nea Kraglievich, 1926a, both known from deposits along the
Rio Parana in the Province of Entre Rios of Argentina. These
deposits are Huayquerian and possibly Montehermosan in
age (Pascual and Odreman Rivas, 1971). Bondesio et al.
(1975) extended the family to include the Santacrucian ro-
dent Scotaeumys Ameghino, 1887, and Perimys Ameghino,

1 887, and thereby expanded the familial diagnosis to include
small forms with rooted teeth in which the enamel laminae
are connected labially (on the upper teeth) or lingually (on
the lower teeth).

The specimens from the Rio Acre Local Fauna are placed
in this family on the basis of the large, ever-growing cheek
teeth in which the laminae are straight, parallel, and uncon-
nected. The laminae are furthermore widely separated by
cement, more so in fact than in either Dabbenea Kraglievich,
1926a, or Neoepiblema Ameghino, 1889, the larger genera
to which the Rio Acre specimens most conform. In addition,
these teeth are intermediate in size between those of Dab-
benea and Neoepiblema and cannot be clearly referred to
either genus. The discovery of these teeth along the Rio Acre
extends the range of the Neoepiblemidae into the Amazon
Basin, identifies the probable existence of another genus of
the larger neoepiblemids, and establishes a temporal con-
nection to the deposits along the Rio Parana in Argentina.

Family Echimyidae Miller and Gidley, 1918
Subfamily Heteropsomyinae Anthony, 1917

Genus and Species Indeterminate

Figure 26C

MATERIAL. LACM 1 17526, right M1 or ML

LOCALITY. LACM 4611.

DESCRIPTION. This tooth is brachydont with four lophs.
The posteroloph and metaloph have joined and closed off

the posterior labial flexure (metaflexus) as a fossette. The
anteroloph and protoloph have united and created a fossette
from the anterior flexure (paraflexus). The protoloph and
metaloph are separate at this wear stage. The hypoflexus is
broad and directed anteriorly. The occlusal shape of the tooth
is quadrate. The tooth has three roots but the medial root is
large and appears to be two fused roots. Measurements
(length x width) are 1.65 x 1.62.

DISCUSSION. Heteropsomyine features seen in this tooth
are the quadrate outline, four separate lophs until moderate
wear is reached on the upper molars, and a broad, anteriorly
extended hypoflexus (Pascual et al., 1 966; Patterson and Pas-
cual, 1968). The Heteropsomyinae range from the Early Oli-
gocene ( Sallamys Hoffstetter and Lavocat, 1970) to the Re-
cent in South America.

Order Litoptema Ameghino, 1889
Family Proterotheriidae Ameghino, 1887

Genus and Species Indeterminate

Figure 27C, D

MATERIAL. LACM 1 17527, right metatarsal III.

LOCALITY. LACM 4611.

DESCRIPTION. The proximal end is gently convex plan-
todorsally and concave transversely for articulation with the
ectocuneiform. Only a small facet for articulation with the
cuboid is present laterally. The lateral margin of the proximal
end is deeply sculpted for metatarsal IV but the medial mar-
gin is only slightly so for metatarsal II. The shaft is flattened
for appression of the secondary metatarsals along the prox-
imal two-thirds of its length and only slightly more rounded
beyond. The shaft is therefore trapezoidal in cross section.
The narrowest part of the shaft is at the proximal one-fifth
of its length. The carina extends with uniform prominence
around the distal articular facet. Two large (7 mm) facets are
placed one each in the lateral and medial fossae of the distal
end. These look like articular facets but are more likely un-
usually smooth attachment areas for the collateral ligaments.

DISCUSSION. This single metatarsal is not sufficiently
diagnostic for generic identification. It has the size and pro-
portions of those of Epecuenia Cabrera, 1939, and Eoau-
chenia Ameghino, 1887, both Huayquerian genera. Mea-
surements: length, 95.5; proximal end, length x width (15)
x 21.2; mid-shaft, length x width, 10.9 x 1 5.2; distal end,
length x width, 15.8 x 22.1.

Family Macraucheniidae Gill, 1872

Unnamed New Genus

Figures 27A-B, E-F and 28E-F

MATERIAL. LACM 1 17528, right mandibular ramus with
DP4-M,; LACM 1 17529, left premolar; LACM 1 17530, right
metatarsal III.

LOCALITIES. LACM 117530, LACM 4418; LACM
1 17528 and 1 17529, LACM 4611.

DESCRIPTION. The ramus maintains a uniform depth

Contributions in Science, Number 374

Frailey: Fossil Mammals of the Rio Acre 31

Figure 27. Macraucheniidae, Proterotheriidae, and Macraucheniidae. A-B, Macraucheniidae, unnamed new genus, LACM 1 17528, right
mandibular ramus with DP4-M,; A, occlusal view; B, labial view. C-D, Proterotheriidae, gen. and sp. index., LACM 117527, right metatarsal
III; C, anterior view; D, posterior view. E-F, Macraucheniidae, unnamed new genus, LACM 1 17530, right metatarsal III; E, anterior view;
F, lateral view.

32 Contributions in Science, Number 374

Frailey: Fossil Mammals of the Rio Acre

(18 mm) beneath the alveoli of DP3. Two mental foramina
are present, one beneath the anterior root of DP4 and another
posterior and dorsal to the first and beneath the center of
DP4.

The DP4 and M, are brachydont, with metastylids separate
from the metaconids, and with prominent anterior, posterior,
and labial intercolumnar cingulae. The prominence of the
labial cingulae, and the openness of the ectoflexids, separate
these teeth from those of the proterotheres. Small accessory
cusps are present posterior to the paraconid (large on DP4)
and anterior to the termination of the hypolophid. With wear
these produce bifurcations in the paralophid and hypolophid.
On DP4, a small accessory cusp is placed between the para-
conid and metaconid on a small crest that extends to the
apex of the protoconid. This situation is analogous to the
placement of the entoconid in the posterior half of the per-
manent teeth of most macraucheniid genera. The appearance
of this additional cusp on a deciduous premolar is evidently
the retention of a primitive character (the true paraconid?)
that is not duplicated in the adult teeth. The entoconid is
either lost or has been reduced and has migrated to form the
small accessory cusp at the termination of the hypolophid.
On DP4 the anterior half of the tooth is elongated and equals
almost twice the length of the posterior half. Measurements,
length x width: DP4, 17.3 x 18.5; M„ 14.0 x 8.9.

The single isolated lower premolar, probably P2, possesses
a very small paraconid that is connected to the protoconid
by a labial crest and also to the metaconid by a lingual crest.
The metaconid is only slightly lower than the protoconid to
which it is connected by a crest. A single crest descends from
the metaconid to the most posterior part of the tooth and
divides the talonid into two oblique surfaces, the lingual
being the smaller. The tooth is two-rooted.

Metatarsal III is slender. It is the same length as that of
Theosodon Ameghino, 1887, but equally as slender as those
of Cullinia Cabrera and Kraglievich, 1931. The articular facet
for the ectocuneiform is narrow and slightly convex plan-
todorsally and slightly concave transversely. The plantar hook
is large and has a large oblique articular facet for metatarsal
IV. The lateral side of the proximal end is sculpted and the
shaft is flat for the proximal two-thirds of its length to receive
metatarsal IV. A small (3 mm) facet is present on the dor-
somedial side of the proximal end for the articulation with
metatarsal II and the shaft is flattened on its medial side for
only one-half of its length. The proximal shaft is therefore
subquadrate and the distal part, approximate one-half, is oval
in cross section. The distal end flares widely toward the ar-
ticular surface. The carina is large and located plantarly and
distally with a slight ridge reaching the dorsal surface. Mea-
surements are: overall length, 120; proximal end, length x
width, 22.0 x 1 3.9; mid-shaft, length x width, 11.7 x 13.1;
distal end, length x width, (18) x 21.3.

DISCUSSION. The Santacrucian-Chasicoan genus Theo-
sodon and the rare Chasicoan genus Cullinia are the geolog-
ically youngest macraucheniids that are comparable in size
to the Rio Acre genus. This Rio Acre macraucheniid is more
derived than either of the known genera in having the meta-
conid and metastylid well separated and in lacking an ento-

conid. The slenderness of the metatarsal is unlike that of
Theosodon but similar to metatarsals of Cullinia. These spec-
imens differ from known macraucheniids to the extent that
a new genus is probably represented. However, additional
and better material is needed to be certain of this.

The discovery of this macraucheniid adds to the diversity
of the family during the Late Miocene. During the Chasicoan,
Theosodon and Cullinia, both small forms, were present while
in the succeeding Huayquerian Land Mammal Age a larger
genus, Promacrauchenia Ameghino, 1904, was the represen-
tative of the family (Savage and Russell, 1983). This small
genus from the Amazon is dentally more advanced than
Promacrauchenia in having lost the entoconid (previously
known to have been lost only in Macrauchenia Owen, 1 840).

Order Astrapotheria Lydekker, 1894

1 Synastrapotherium Paula Couto, 1976

Figure 26

MATERIAL. LACM 1 17531, isolated canine.

LOCALITY. Sandbar in Rio Acre, no lithologic associa-
tion.

DISCUSSION. The Order Astrapotheria is represented in
the Rio Acre Local Fauna by a single tusk that was found as
float material. The latest occurrence of an astrapothere is that
of Xenastrapotherium Kraglievich, 1928, of the Friasian of
Venezuela and Colombia (Cabrera, 1929). This tusk is of the
size expected for Xenastrapotherium although none is de-
scribed for the genus.

The Order Astrapotheria is an indicator of Middle Mio-
cene (Friasian) or older time and is therefore at variance with
the Late Miocene (Huayquerian) age indicated by other fau-
nal members. Either the Astrapotheria survived until more
recent time in the Amazon Basin or the Rio Acre Local Fauna
is not a unified local fauna but rather is mixed. It is my belief
that this is a single local fauna and that this record constitutes
a late occurrence of the Order Astrapotheria. This is based
on the absence of older strata (or any other formation) along
the Rio Acre that could have been the source of the astra-
pothere specimen and in the same preservation of the astra-
pothere tusk and the other fossils.

Synastrapotherium amazonense was described by Paula
Couto (1976) from several cheek teeth found on the Rio Jurua
of Brazil. The Rio Jurua and the Rio Acre are part of adjacent
river systems. Synastrapotherium was dated as Oligocene on
the basis of its relative stage of evolution, but there was no
associated fauna to support this age assignment. However,
the type locality for Synastrapotherium as described by Paula
Couto (1976) and the geologic section of the Rio Jurua as
figured by Paula Couto (1978) appears to be the same as that
cropping out along the Rio Acre. Although there are no com-
parable specimens of the astrapotheres from the Rio Jurua
and the Rio Acre, the tusk from the Rio Acre is questionably
referred to Synastrapotherium on the basis of the probable
origin from the same stratum. A Huayquerian age assignment
for Synastrapotherium amazonense of the Rio Jurua based
on stratigraphy and faunal association of an astrapothere

Contributions in Science, Number 374

Frailey: Fossil Mammals of the Rio Acre 33

from the nearby Rio Acre seems as reasonable as an Oli-
gocene date based on morphological comparisons alone.

Order Sirenia Illiger, 1811
Family Trichechidae Gill, 1872

IRibodon Ameghino, 1883

Figure 28

MATERIAL. LACM 1 17532, left lower molar.

LOCALITY. LACM 4611.

DESCRIPTION. This tooth has a simple crown pattern
that consists of two transverse crests, one anterior cusp, and
one posterior cusp with no accessory cusps. Following the
terminology of Pascual (1953), the anterior cusp is the para-
conid (not present on any of Pascual’s specimens), the first
transverse lophid is comprised of the metaconid and pro-
toconid, the second lophid is composed of the hypoconid
and entoconid, and the single posterior cusp is the hypocon-
ulid.

The paraconid and hypoconulid are similar in that each
points away from the crown (anteriorly and posteriorly, re-
spectively) at an approximate 60° angle. Each cusp is also
attached higher on the labial side but low on the lingual side
(identifiable as a separate crest to the level of the lingual
cingulum). The hypoconulid is the larger cusp and has a slight
terminal bifurcation.

The protoconid is slightly taller than the metaconid and
sits anterior to the metaconid. Both cusps are distinguishable
at this slight wear stage but their distinctness would soon
disappear into a single transverse lophid (actually slightly
oblique because of the position of the protoconid). The pos-
terior transverse crest is the same height as the protoconid,
and the hypoconid and entoconid are only faintly discemable
as distinct cusps. A low crest unites the protoconid with the
center of the posterior lophid. This tooth is much more elon-
gate than those of Ribodon limbatus and has a figure eight
rather than a rectangular occlusal outline. The inferior margin
of the enamel is about 2 mm higher on the lingual side of
the crown. The tooth is double-rooted. The posterior root
bends slightly to the anterior and is approximately twice the
height of the crown. Measurements (length x width): 16.9 x
8.7.

DISCUSSION. The simple bilophate pattern, large hy-
poconulid, and long roots are typical of Ribodon (Pascual,
1953).The bilophate pattern is reminiscent of primitive pe-
rissodactyls and cause for its original placement with the
Tapiroidea (Ameghino, 1 883). The Rio Acre specimen differs
greatly from lower teeth of Ribodon limbatus in being slightly
higher crowned although this may merely reflect less wear,
in having a large paraconid (wholly absent in Ribodon lim-
batus), and in having a smaller hypoconulid. These may,
however, be primitive conditions in the Trichechidae. The
Rio Acre specimen can be only questionably referred to the
genus Ribodon.

Only a very few specimens of Tertiary manatees are known
in South America. The oldest specimen of Trichechidae, Sire-

notherium pirabensis Paula Couto, 1967b, consists of an up-
per tooth and a few skeletal fragments from early Miocene
deposits on the northern coast of Brazil that cannot be di-
rectly compared with the lower molar from the Rio Acre.
However, the upper molar is very similar in its compactness
to those of Ribodon, in which the lower molars are also
compact and not so elongate as the Rio Acre tooth. Sirenothe-
rium is in fact sufficiently similar to Ribodon to perhaps merit
reference to that genus.

The generic holotype of Ribodon and the best specimens
are from exposures along the Rio Parana in Argentina
(Ameghino, 1883; Pascual, 1953). The fossils from these de-
posits are now referred to the Huayquerian or possibly Mon-
tehermosan land mammal ages (Pascual and Odreman Rivas,
1971).

Reinhart (1951) described a single lower molar from the
La Venta Fauna (Middle Friasian) of Colombia as the ho-
lotype of Potamosiren magdalenensis, but Pascual (1953)
thought this species was more likely referable to Ribodon.
The molar of “ Potamosiren” magdalenensis is more similar
to lower teeth in Ribodon than is the Rio Acre molar and
indicates that the features of Ribodon were established at
least by the Middle Miocene if not the Early Miocene (in
Sirenotherium). This would support a generic separation of
the Rio Acre specimen from the Colombian and Argentinian
specimens in that the lineages of these two groups would
necessarily have had to diverge prior to the Middle Miocene.

A second possible specimen of Ribodon is a fused parietal-
supraoccipital from the Rio Jurua, Acre, Brazil. The Rio Acre
forms part of the southern border of Acre, and the Rio Jurua
probably exposes the same geologic section as the Rio Acre
(see discussion of Rio Acre geology). This skullcap could have
come from the same level as did the isolated tooth. Unfor-
tunately, the stratigraphic occurrence of this skullcap is not
known and Paula Couto’s (1956) reference to Trichechus
manatus (a Pleistocene and Recent species) is beyond further
comment at this time.

Inapari Member A Local Fauna
Holocene

Order Edentata Cuvier, 1798
Family Megatheriidae Owen, 1843

Subfamily Nothrotheriinae
Kraglievich, 1923

Nothropus Burmeister, 1882

Nothropus priscus Burmeister, 1882

Figures 29, 30, 31, 32

MATERIAL. LACM 1 17533, skeleton, missing left hind-
limb and both hind feet.

LOCALITY. LACM 4609.

DESCRIPTION. Skull. When discovered, the left side of
the skull was exposed. As a result, the left malar, squamosal,
and most of the left frontal and parietal are missing. The
skull is highly fragmented but intact.

34 Contributions in Science, Number 374

Frailey: Fossil Mammals of the Rio Acre

Figure 28. IRibodon sp., Neoepiblemidae, and Macraucheniidae. A, IRibodon sp., LACM 1 17532, left lower molar, stereoscopic occlusal
view; B, Neoepiblemidae, gen. and sp. indet., LACM 1 17525, lower tooth, side view. C-D, IRibodon sp., LACM 1 17532, left lower molar;
C, lingual view; D, labial view. E-F, Macraucheniidae, unnamed new genus, left premolar, LACM 117529; E, lingual view; F, labial view.

In dorsal view, the skull has a bulbous cranium and a
narrow rostrum. A strong declivity is present between the
lambdoidal crest and the zygomatic process. The rostrum is
constricted anteriorly to the malar and then flares slightly

toward the external nares. The temporal fossa is marked by
a low ridge that extends to within 25 mm of the lambdoidal
ridge and to within 20 mm of the sagittal suture. Postorbital
processes are present but small and are placed over M3. A

Contributions in Science, Number 374

Frailey: Fossil Mammals of the Rio Acre 35

5 cm

Figure 29. Nothropus priscus, LACM 1 17533, skull and right ramus. A, lateral view; B, palatal view with occlusal drawing of teeth.

36 Contributions in Science, Number 374

Frailey: Fossil Mammals of the Rio Acre

5 cm

* — — — ^

Figure 30. Nothropus priscus, LACM 1 17533, skull and mandible, lateral view.

supraorbital foramen is present anterior to the postorbital
process. The malar and zygomatic portions of the squamosal
do not flare widely but instead are nearly parallel to the
cranium.

In lateral view, the skull is domed with the highest point

above the anterior tip of the zygomatic process. The occiput
is vertical with the occipital condyles situated immediately
ventral to the inion. The paraoccipital and mastoid processes
are equal in size, low and rounded. The dorsal margin of the
external auditory meatus, ventral margin of the orbit, and

Contributions in Science, Number 374

Fraiiey: Fossil Mammals of the Rio Acre 37

Figure 31. Nothropus priscus, LACM 1 17533, left ramus. A, lateral view, photograph and drawing; B, medial view.

38 Contributions in Science, Number 374

Frailey: Fossil Mammals of the Rio Acre

alveolar border of the maxilla are on the same plane. The
zygomatic portion of the squamosal narrows anteriorly and
does not contact the malar (10 mm gap). The ascending
portion of the malar is thin and directed dorsocaudally with
a shallow notch for the zygomatic process. The descending
portion of the malar is triangular and directed ventrocau-
dally. The lacrimal foramen is small (6.5 mm). A deep buc-
cinator fossa is present above C‘-M2. The pterygoids are
semicircular in lateral view and extend from a point im-
mediately anterior to the external auditory meatus to within
1 5 mm of M4. They are apparently only slightly expanded
at their posterior base. The rostrum is short and narrows to
its anterior termination.

The palate is constricted anteriorly to M1 (the buccinator
fossa) and elevated between M1-3. The postpalatine fossa is
U-shaped and begins at M4. The facets for the premaxillae
are large (14 mm) and ventrolaterally placed. The premax-
illae are missing in the specimen.

Upper teeth. Five upper teeth are present and are desig-
nated C1, M1'4 to conform to other discussions of Nothro-
theriinae in which only four cheek teeth, the molariforms,
are present.

The upper caniniform tooth is small and oval in cross
section. It is posteriorly recurved with an oblique shear facet
facing posteriorly. Measurements (length x width) are: 6. 1 x
4.9.

The M1 is trapezoidal in occlusal outline with the broad
base medially. The central basin of the tooth is widely open
anteromedially. The anterior border of the tooth extends
farther ventrally than the posterior border. Measurements
(length x width) are: 10.0 x 10.7.

The M2-3 have the same modified trapezoidal occlusal out-
line in which the medial border parallels but is longer than
the external border and the posterior border is curved and
concave posteriorly. The central basin is open postero-
medially. The anterior edge of the tooth is higher than the
posterior edge. Although M‘~4 have a slight external depres-
sion, only M2-3 have a distinct but shallow external groove.
Measurements (length x width) are: M2, 10.6 x 13.2; M3,
10.4 x 13.5.

The M4 is flat and anteriorly recurved. A single, anteriorly
facing oblique shear facet is present on the occlusal surface.
Measurements (length x width) are: 5.0 x 12.2.

Lower teeth. Four lower teeth are present and designated
in this paper as C,, M,_3 to correspond with the upper teeth.

The first tooth, C,, is caniniform, oval in cross section and
with an anterior-facing, oblique shear surface. The lower and
upper caniniform teeth will not occlude when the molariform
teeth are in occlusion. The lower caniniform and M, are
separated by a short diastema (10.5 mm).

The first and second lower molariform teeth are rectan-
gular in cross section with the central basin opening exter-
nally. The anterior and posterior crests are equal in height
in M,. Both internal and external grooves are present in
M,_2. In M2 (and M3) the anterior crest is lower than the
posterior. Measurements of M, and M2 (length x width) are:
9.5 x 12.7; M2, 9.7 x 13.3.

The M3 is rounded medially and the anterior and posterior

crests slightly diverge laterally. The central basin opens an-
terolaterally. Measurements (length x width) are: 10.8 x
12.5.

Mandible. The postdental portion of the ramus is trifid
with strong and elongate coronoid and angular processes. The
condylar process is equally placed between the coronoid and
angular processes. The condyle is flat, narrow, and directed
ventrolaterally. The angular process is sculpted along its me-
dial, ventral margin and the posterior tip is curved medially.
The dentary portion of the ramus is thick and deep and
reaches a maximum curvature beneath M,_3. The posterior
external mandibular foramen is laterally placed at the base
of the ascending ramus immediately posterior to M3. The
posterior internal mandibular foramen is placed on the level
of the alveoli approximately 30 mm posterior to M3. The
predentary portion of the ramus is elongated into a U-shaped
spout. Two mental foramina are present. The larger of the
two is situated near the ventral symphyseal line and the
smaller foramen is posterior and dorsal to the first.

DISCUSSION. The members of the Subfamily Noth-
rotheriinae exhibit numerous primitive sloth features that
are characteristic of Santacrucian genera ( Hapalops Ame-
ghino, 1887, for example, is a frequently mentioned near-
relative although I can find no derived features that are shared
by this genus and the nothrotheres). The most notable of
features in common are the elongate and subcylindrical skull,
simple dentition, trifid postdental ramus, and spout-like pre-
dental ramus. The nothrotheriine genera are larger than Early
Miocene genera, have inflated frontals and pterygoids, ver-
tical grooves on the teeth, and postdental rami that incline
laterally rather than parallel the axis of the tooth row. Noth-
ropus is more primitive than Nothrotherium and Nothrotheri-
ops in that the caniniform teeth are not lost and the angular
process of the ramus is not reduced. However, the shape of
the skull is relatively advanced, and more like the Mega-
theriinae Gill, 1872, in the presence of a domed forehead
and a small, tubular nasal area. The mandible also has Mega-
theriinae features in the long symphyseal spout and the bul-
bous ventral margin. Comparative measurements of noth-
rothere genera are given in Table 7.

Nothropus was previously known from only three speci-
mens, each one a holotype. Nothropus priscus Burmeister,
1882, and Nothropus tarijensis (Burmeister), 1887, are rep-
resented by partial rami in which the distinguishing feature
is the presence of C,. Nothropus nordenskioldi Kraglievich,
1926b, was named for a humerus that is of the proper size
and not referable to any better known ground sloth. Its ref-
erence to this genus is questionable and therefore the validity
of this species is doubtful. Nothropus priscus and N. tarijensis
have typically quadrangular nothrotheriine dentitions with
lateral grooves in the teeth but, unlike Nothrotherium and
Nothrotheriops, an anterior caniniform tooth is present. In
N. tarijensis, this C, is minute. The mandible of LACM
1 17533 has a large C, that is comparable to N. priscus. The
holotype of TV. priscus was probably a juvenile animal
(Ameghino, 1907) and slight variations in tooth shape and
jaw proportions between the holotype and the new specimen
are attributable to age (the length of the tooth series and the

Contributions in Science, Number 374

Frailey: Fossil Mammals of the Rio Acre 39

/ "n
/ \

I \

Figure 32. Nothropus priscus, LACM 1 17533, left ramus, occlusal view, photograph and drawing.

diastema are longer in LACM 1 17533). In fact, other than
the substantial difference in size of the caniniform teeth (that
may be due to sexual dimorphism), teeth of N. priscus and
N. tarijensis differ only slightly in occlusal outline (highly
variable in other ground sloths) and in size (N. tarijensis is
slightly larger).

The lack of good material has retarded systematic consid-
eration of Nothropus. In reference to the size of C, , Ameghino
(1907) suggested that the lineage progressed from Nothropus
priscus to N. tarijensis and ended with the complete loss of
Ci in Nothrotherium (including at that time Nothrotheriops).
This transformation series would require that N. priscus con-

40 Contributions in Science, Number 374

Frailey: Fossil Mammals of the Rio Acre

siderably predated N. tarijensis and other nothrotheriine gen-
era. However, a radiocarbon date from a wood sample taken
from a lignite lense beneath the channel where N. priscus was
found dated this specimen as younger than 7145 ± 150 years
b.p. Other dates from Member A and the underlying Acre
Conglomerate Member range from between 5575 (± 105) and
10,075 (±150) years b.p. (Campbell and Frailey, 1984). This
age spread may represent varying degrees of contamination
or may indicate a cycle of channel cut and fill in this area.
In either event, the maximum age of Nothropus priscus leaves
little time for the sequence envisaged by Ameghino (1907).

Without supporting his proposal, Paula Couto (1971) sug-
gested that Nothropus and Nothrotheriops (which Paula Cou-
to separated from Nothrotherium in the same paper) formed
a collateral group to Nothrotherium. With the additional ma-
terial of Nothropus now available, it appears that Paula Cou-
to’s premonitory placement is accurate. Nothropus and Noth-
rotheriops share (in contrast to Nothrotherium) several derived
features that include a highly domed skull with convex fron-
tals, the presence of a supraorbital foramen, a narrow dorsal
prong of the malar, a posteroventrally directed descending
malar, and possibly (the polarity is uncertain) an elongate
angular process on the ramus. Nothropus resembles Noth-
rotherium in features that I believe have less taxonomic weight
such as the same overall size, a zygomatic process that is
fully separate from the malar and a subsequently shallower
zygomatic notch on the malar, placement of the postorbital
process above M3 rather than M2, and the alignment of the
inferior border of the orbit with the maxillary alveolar bor-
der.

In addition, the generic identification of Nothropus can
now be expanded to include such features as the position of
the postpalatine notch (at M4 and more posterior in Noth-
rotheriops and Nothrotherium) and straight ventral margin
of the predental spout on the mandible as well as a number
of certainly primitive features such as the trifid postdental
ramus, relatively uninflated pterygoids, and the large cani-
niform teeth.

Nothrotheriops is a North American genus that inhabited
an arid or semiarid habitat (Hansen, 1978). If Nothropus
shared the habitat preference of its near relative, this would
strongly suggest an alternative vegetation zone to the tropical
deciduous forest that surrounds the Rio Acre today. The
spread of arid grasslands into the Amazon Basin is postulated
to have occurred several times in the Pleistocene and each
time the forest elements were restricted into patches, the
“tropical forest refugia” (Haffer, 1969, 1974, 1979; Vanzolini
and Williams, 1970; see Prance, 1982, for a review). The
tropical forest refugia model largely depends on the patch-
work distributions of some extant Amazonian species that
are inexplicable by modem physiographc barriers. Evidence
based on fossils that bears on this question is avidly sought.
This discovery of a genus that has an otherwise Andean (N.
nordenskioldi, Pemvian altiplano), Sub-Andean (N. tarijen-
sis, Tarija Basin, Bolivia), and pampean (N. priscus, near
Rosario, Argentina) distribution and which is related to the
desert-dwelling genus ( Nothrotheriops ) argues in favor of drier
conditions in the western part of the Amazon basin during

Table 7. Measurements of skull and ramus of Nothropus priscus
compared with Nothrotherium maquinense and Nothrotheriops shas-
tense.

TV.

priscus

N. maqui-
nense4

TV.

shastense'

Skull

Length from anterior end
of nasals to posterior
end of supraoccipital

272

216, (250)

336.3

Width across anterior
ends of zygomatic pro-
cesses of squamosals

(105)

93,(115)

115.6

Distance between lacri-
mal foramen and the
occipital

177

162, (180)

Distance between lower
border of foramen
magnum and posterior
border of palatine

(120)

106, (120)

Height of skull above
pterygoids

98

84, 94

Length of palate, from
anterior end of maxil-
laries to postpalatine
notch

106

-, 96

135.8

Greatest width across oc-
cipital condyles

72

— , —

77.6

Ramus

Length from anterior end
of symphysis to poste-
rior end of condyle

(120)

(151), 185

272.8

Distance from anterior
end of symphysis to
posterior end of angu-
lar process

(230)

(135), (140)

(268.1)

Greatest height from an-
gle to coronoid process

(90)

77,(80)

110.8

Height of mandible un-
der last tooth (M3)

41

35, 35

45.5

Height of mandible un-
der first molariform

43

35, 29

56.8

C,-M, diastema

10.6

—

—

1 From Paula Couto, 1971. Two specimens of N. maquinense and
an average measurement of 2-5 specimens of N. shastense.

some part of the Holocene. The radiocarbon dates for Mem-
ber A and the underlying Acre Conglomerate Member of the
Inapari Formation indicate a Holocene age for this forma-
tion. Forest fragmentation could have occurred during cool,
dry periods several times during the late Pleistocene and
Holocene. Datable cool, dry periods occurred 20,000-1 3,000

Contributions in Science, Number 374

Frailey: Fossil Mammals of the Rio Acre 41

years b.p., 1 1 ,000-9500 years b.p., and 3500-2800 years b.p.
(Haffer, 1979). Member A of the Inapari Formation, and
Nothropus priscus, could date from the the second of these
periods.

DISCUSSION AND CONCLUSIONS

The strata exposed along the Rio Acre west of Inapari, Peru,
represent two formations and at least two time periods. Late
Miocene and Holocene. These strata appear to be part of a
widespread depositional pattern in the western Amazon Ba-
sin for they correlate with the published description of the
geologic section of the Rio Jurua, Brazil, 500 km northwest
of the explored part of the Rio Acre and with strata exposed
along the Rio Beni, Bolivia, 300 km to the southeast (Camp-
bell and Frailey, 1984; Campbell et al., 1985). The lower
formation, unnamed in this paper and identihed only as the
Tertiary red beds, is composed of variegated clays with cal-
citic stringers and has an eastward dip of 10-15° presumably
as a result of the Andean orogeny.

The fossils recovered from the Tertiary red beds provide
the greatest taxonomic diversity yet known from a single
assemblage in the Amazon Basin and significantly add to the
total number of fossil taxa that are known from this area.
Fourteen new taxa are added to the fossil fauna. Most of this
material is derived from the Tertiary red beds and is a true
death assemblage of fossils rather than isolated elements as
were most previous discoveries. As such, these fossils yield
a more definite age assessment and a better basis for zoo-
geographic and paleoclimatic comparisons than was previ-
ously possible. The Rio Acre Local Fauna, including re-
worked material found in the overlying Inapari Formation,
is dated as Huayquerian (Late Miocene) in age. This age
assignment is made on the presence in the local fauna of such
characteristic Huayquerian taxa as Kiyutherium orientalis
(Francis and Mones, 1965), Tetrastylus (Pascual et al., 1966),
and possibly Potamarchus murinus (Ameghino, 1891). Pot-
amarchus murinus has a Chasicoan and possible Huayque-
rian distribution (Pascual et al., 1966).

The recognition of new taxa is not unexpected in the de-
scription of a local fauna from an area as paleontologically
unknown as the Amazon Basin. The description of only the
second species of the dinomyid genus Telicomys extends the
geographic range of this genus northward and into tropical
South America for the first time. Telicomys amazonensis is
more primitive in features of the basicranium and occiput
than T. giganteus (of Montehermosan-Chapadmalalan age
in Argentina) and the placement of Telicomys amazonensis
with a Huayquerian local fauna is therefore reasonable. In
addition, a new genus and species of mylodontine ground
sloth, Stenodon campbelli, is present that shows affinities
with Glossotherium and Glossotheridium but is unique in the
shape of its palate and upper teeth. No ancestral or descen-
ded forms for this new genus can be identified among known
ground sloths. A new genus of Macraucheniidae is recognized
but not named for lack of adequate material.

A single lower tooth of a marsupial that cannot be referred
to any known marsupial family may represent a wholly new

group that was equivalent ecologically to the Procyonidae.
A compressed trigonid and the position of the hypoconulid
near the entoconid indicate a marsupial, but this tooth also
has a singularly wide and basined talonid that is procyonid-
like, and very unlike marsupials, in its appearance.

A single tooth of a manatee, tentatively referred to Ribo-
don, marks the first appearance of this order in the Tertiary
fossil record of the Amazon Basin. Manatees are rare among
South American fossils with previous Tertiary discoveries in
the Friasian of Colombia (Reinhart, 1951) and Huayquerian
or Montehermosan of Argentina (Pascual, 1953).

One element, although relatively undiagnostic on the basis
of a single canine tooth, is nevertheless interesting at the
ordinal level. The Astrapotheria are thought to have become
extinct at the end of the Friasian (Pascual et al., 1966). The
presence of an astrapothere in the Rio Acre Local Fauna may
indicate that the Astrapotheria survived beyond Friasian time
in the Amazon Basin.

A number of fossils cannot be identified to lower taxo-
nomic levels. Nonetheless, several of these specimens rep-
resent first records of taxonomic categories in the Amazon
Basin owing to the virtual absence of mammalian fossils from
this part of South America. Taxa thus represented include
Pampatheriinae, Proterotheriidae, Macraucheniidae, Ere-
thizontidae, Dasyproctidae, Neoepiblemidae, Caviidae (Car-
diomyinae), and Echimyidae (Heteropsomyinae). These
groups, now known only from isolated teeth and fragments,
indicate that a substantial Amazonian Tertiary fauna existed
and can yet be amplified with additional work.

Overlying the Tertiary red beds is a thick formation to
which the Inapari Formation of ONERN (1977) was restrict-
ed by Campbell and Frailey (1984). This formation com-
prises four readily apparent units. The lowermost identifiable
unit is a characteristic marker bed, the clay-pebble conglom-
erate that is termed the Acre Conglomerate Member by
Campbell et al. (1985). Above the Acre Conglomerate Mem-
ber (and overlying the Tertiary red beds where the Acre
Conglomerate is missing) is a unit composed primarily of
channel-sand deposits with lenses of clay, sand, and lignite,
that can be replaced laterally by buff clay. Wood fragments
from these two members yielded radiocarbon dates between
5575 (±105) and 10,075 (±150) years b.p. (Campbell and
Frailey, 1984) with the older date from the Acre Conglom-
erate Member and the youngest dates from lignite lenses in
Member A.

Member A produced a complete skeleton of the rare noth-
rotheriine ground sloth, Nothropus priscus, that was previ-
ously known only by a single mandibular ramus from Ar-
gentina. In fact, the genus was previously represented by only
three specimens, two rami and a humerus, each named as a
distinct species. Nothropus is thought to have been closely
related (Paula Couto, 1971) to Nothrotheriops that is known
to have inhabited dry regions of southwestern North America
during the Pleistocene (Hansen, 1978). If Nothropus shared
the habitat preference of Nothrotheriops, its presence in the
western margin of the Amazon Basin would indicate a much
drier Holocene environment that is presently found there.
To the extent that Nothropus shared the climatic preference

42 Contributions in Science, Number 374

Frailey: Fossil Mammals of the Rio Acre

of its North American relative, the Rio Acre discovery sup-
ports the forest refugia model of Haffer (1969, 1974), and
Vanzolini and Williams (1970). Studies in a variety of fields,
for example soil science, palynology, and zoogeography, have
delimited several areas in tropical South America where the
tropical lowland forest may have been restricted during pe-
riods of dry conditions in which grassland or savanna pre-
vailed over much of the Amazon Basin. The areas chosen
as probable sites of these refugia differ in location and size
from author to author but one refugium is usually placed in
the area where the skeleton of Nothropus priscus was found.
If this individual was buried near its habitat, Nothropus either
did not share the dry land preference of Nothrotheriops and
was an inhabitant of this forest refugium during the early
part of the Holocene or this refugium was not extant during
this time. A third alternative, that Nothropus occupied both
forest and open habitats and its fossilization in an area that
may have been continually forested during the late Pleisto-
cene and early Holocene, would mean that this specimen
does not contribute to an understanding of vegetational pat-
terns.

The Rio Acre collection has doubled the known Cenozoic
diversity of the Amazon Basin. Although still far from com-
parable to Cenozoic collections elsewhere in South America,
particularly in Argentina, direct faunal comparison between
the Cenozoic faunas of tropical and temperate South America
is feasible and more fruitful than was previously possible. It
can be determined, for instance, that many genera and at
least two species were living both in the Amazon Basin and
in the temperate regions during Late Miocene. Temperate
South America was a savanna during the late Cenozoic (Webb,
1978) and faunal members shared in common suggest that
this savanna extended into the Amazon Basin. Several genera
and species, however, are included that have only been found
in the tropical lowlands and were perhaps endemic. These
are the pyrothere Griphodon (Eocene), an astrapothere, Syn-
astrapotherium (probably Huayquerian), late Tertiary genera
of three toxodontids, Trigodonops, Abothrodon, and Neotri-
godon, and from this paper the ground sloth Stenodon camp-
belli and an unnamed macraucheniid. Higher taxonomic fau-
nal distinctions are not present with the single possible
exception of a new family of Marsupialia. The differences
between temperate and tropical South American faunas dur-
ing the Cenozoic appear to be real but are mostly at the
genus and species levels. This may be a result of greater
aridity in the Amazon Basin for much of its Cenozoic history
that permitted more widespread grassland and grassland sa-
vanna communities than would be possible under the present
climatic regime of South America.

ACNOWLEDGMENTS

I thank Dr. Hernando Macedo, Dr. Guillermo Morales, and
Dr. Gerardo Lamas of the Museo Nacional “Javier Prado”
for courtesies and assistance during work in Peru. I greatly
benefited from discussions with Dr. Rosendo Pascual, Dr.
Pedro Bondesio, and colleagues of the Museo de La Plata,
Argentina, and Prof. Alceu Raney, Universidade Federal de

Acre, Rio Branco, Brazil. I am forever indebted to them for
their advice on several matters. Ing. Raul Carrasco and Ing.
Jorge Arellano of the Servicio Geologico de Bolivia permitted
the use of the GEOBOL collection and were colleagues in
the field. I extend my appreciation to Dr. Larry Martin, Dr.
Robert Wilson, and Dr. Robert Hoffmann of the University
of Kansas for their assistance and critical evaluation during
the preparation of this paper. Dr. Albert E. Wood (professor
emeritus, Amherst College), Dr. George F. Engelmann (Uni-
versity of Nebraska, Omaha), and Dr. Greg MacDonald (Ida-
ho Museum of Natural History) read and improved early
drafts of portions of this paper. Mr. Kenneth Whetstone
(Union Texas Petroleum) and Dr. John Chom (University
of Kansas) provided friendly advice and technical assistance
on several points. Numerous staff members of Midland Col-
lege lent their efforts to the completion of this paper and I
acknowledge the help of Dr. Weldon Horton, Carlotta Kel-
logg, Mohsen Kheir, Robert King, Rebecca Johnson, Julia
Olgin, and JoAnn Stella. A special note of gratitude is ex-
tended to my wife, Julia, for her unfailing support of this
project. Many pleasant and enlightening weeks of field com-
panionship with Dr. Kenneth Campbell (Natural History
Museum of Los Angeles County), Dr. Ronald Wolff (Uni-
versity of Florida), and Dr. Bruce MacFadden (Florida State
Museum) will always be remembered. Fieldwork was sup-
ported by grants from the National Geographic Society (nos.
1776, 2002), the National Science Foundation (DEB 78-
03122), and the Saul Fund and Claude Hibbard Memorial
Fund of the University of Kansas.

LITERATURE CITED

Ahem, M.E., and J.F. Lance. 1980. A new species of Neo-
choerus (Rodentia: Hydrochoeridae) from the Blancan
(Late Pliocene) of North America. Proceedings of the
Biological Society of Washington, Vol. 93, no. 2:435-
442.

Ameghino, F. 1883. Sobre una coleccion de mamiferos
fosiles del Piso Mesopotamia de la Formacion Pata-
gonica. Boletin de la Academia Nacional de Ciencias de
Cordoba 5:101-116.

. 1891. Caracteres diagnosticos de cincuenta especies

nuevos de mamiferos fosiles Argentinos. Revista Ar-
gentina de Historia Natural, Buenos Aires 1:129-167.

. 1907. Notas sobre una pequena coleccion dehuesos

de mamiferos procedentes de las grutas calcareas de
Iporanga en al Estado de Sao Paulo, Brazil. Ameghino
Obras, Vol. 17:103-153.

Anthony, H.E. 1924. A new fossil perissodactyl from Peru.
Novitates, no. 111,3 pp. American Museum of Natural
History.

Bondesio, P., R. Pascual, and M.G. Vucetich. 1975. Los
Neoepiblemidae (Rodentia, Caviomorpha): Su carac-
terizacion y sus relaciones filogenticas con los Chin-
chillidae. In Actas del Primer Congreso Argentino de
Paleontologia y Bioestratigrajia (Tucuman 12-16 de
Agosto de 1974), 2, 431-447. Universidad Nacional de
Tucuman y Asociacion Paleontologica Argentina (Tu-
cuman, Argentina, 1974).

Contributions in Science, Number 374

Frailey: Fossil Mammals of the Rio Acre 43

Burmeister, H. 1882. Nothropus priscus, ein bisher unbe-
kanntes fossiles Faulthier. Sitzungberichte der (Konig-
lich-preussischen) Akademie der Wissenschaften, Berlin
1882(2):6 13-620.

. 1885. Examen critico de los Mamiferos y reptiles

fosiles denominados por D. Augusto Bravard y men-
cionados en su obra precedente. Anales del Museo Na-
cional de Historia Natural, Buenos Aires 3:95-174.

. 1887. Neue beobachtungen an Coelodon. Sitzungs-

berichte der (Koniglich-preussischen) Akademie der
Wissenshaften, Berlin 1 887(2):857-862.

Cabrera, A. 1929. Un astrapotherido de Colombia. Physis
9:436-439.

Campbell, K.E., Jr., and D. Frailey. 1984. Holocene flood-
ing and species diversity in southwestern Amazonia.
Quaternary Research 21:369-375.

Campbell, K.E., Jr., C.D. Frailey, and J. Arellano L. 1985.
The geology of the Rio Beni: Further evidence for flood-
ing in Amazonia. Contributions in Science, Natural His-
tory Museum of Los Angeles County, No. 364:1-18.

Castellanos, A. 1 927. Brevas notas sobre los clamidoterios.
Publication del “Centro Estudiantes de Ingenieria de
Rosario”: 1-8. Rosario, Argentina.

De Blase, A. F., and R.E. Martin. 1974. A Manual of Mam-
malogy with Keys to Families of the World. Dubuque,
Iowa: Wm. C. Brown Company Publishers, 329 pp.

Driesch, A. von den. 1976. A Guide to the Measurement
of Animal Bones from Archaeological Sites. Peabody
Museum Bulletin 1, 137 pp. Peabody Museum of Ar-
chaeology and Ethnology, Yale University, New Haven,
Connecticut.

Francis, J.C., and A. Mones. 1965. Sobre el hallazgo de
Cardiatherium talicei n. sp. (Rodentia, Hydrochoeridae)
en Playa Kiyu, Dept, de San Jose, Republica Oriental
de Uruguay. Kraglieviana 1(1): 3—44.

. 1968. Los roedores fosiles del Uruguay. Boletin del

Laboratorio de Paleontologia de Vertebrados, Vol. 1(2):
35-55. Universidad de la Republica Oriental del Uru-
guay, Montevideo.

Haffer, J. 1969. Speciation in Amazonian forest birds. Sci-
ence 165:131-137.

. 1974. Avian speciation in tropical South America.

Nuttall Ornithological Club, No. 14:390 pp.

. 1 979. Quaternary biogeography of tropical lowland

South America. In The South American Herpetofauna:
Its Origin, Evolution, and Dispersal, ed. W.E. Duellman,
University of Kansas Museum of Natural History
Monograph No. 7, 107-140, Lawrence, Kansas.

Hansen, R.M. 1978. Shasta ground sloth food habits, Ram-
part Cave, Arizona. Paleobiology 4(3):302-319.

Hoffstetter, R., and R. Lavocat. 1970. Decouverte dans le
Deseadien de Bolivie de genres pentalophodontes ap-
puyant les affinites africaines des Rongeurs Cavio-
morphes. Comptes rendus des seances de l’Academie
des Sciences, Ser. D, 271:172-175.

Kraglievich, L. 1926a. Los grandes roedores terciarios de
la Argentina y sus relaciones con ciertos generos Pleis-

tocenos de las Antillas. Anales del Museo Nacional de
Historia Natural, Buenos Aires 34:121-135.

. 1926b. Presencia del genero Nothrotherium Ly-

dekker ( =Coelodon Lund) en la fauna pampeana. Noth-
rotherium torresi, n. sp. Revista del Museo de Historia
Natural en la Plata 29:1-18.

. 1928. Sobre el supuesto Astrapotherium Christi

Stehlin, descubierto en Venezuela ( Xenastrapotherium
magnum y Uruguaytherium Beaulieui). Buenos Aires,
La Editorial Franco: 1-1 6.

. 1930a. Diagnosis osteologico-dentario de los ge-
neros vivientes de la Subfamilia “Caviinae”. Anales del
Museo Nacional de Historia Natural, Buenos Aires 36:
59-96.

. 1930b. Los mas grandes carpinchos actuales y fo-
siles de la Subfamilia “Hydrochoerinae”. Anales de la
Sociedad Cientifica Buenos Aires 1 10:233-250, 340-358.

. 1930c. La Formation Friaseana del Rio Frias, Rio

Fenix, Laguna Blanca, etc. y su fauna de mamiferos.
Physis 10(35): 1-35.

. 1932. Nuevos apuntes para la geologia y paleon-
tologia Uruguayas. Anales del Museo Nacional de Mon-
tevideo, Ser. 2, Vol. 3:257-321.

. 1940a. Description detallada de diversos roedores

Argentinos Terciarios, clasificados por el autor. Lucas
Kraglievich, Obras de Geologia y Paleontologia, La Pla-
ta, Vol. 2:297-330.

. 1940b. Los roedores extinguidos del grupo Neo-

epiblemidae. Lucas Kraglievich, Obras de Geologia y
Paleontologia, La Plata, Vol. 3, no. 88:739-764.

Kummel, G. 1 948. Geological Reconnaissance of the Con-
tamana Region, Peru. Geological Society of America,
Bulletin 59:1217-1266.

Lavocat, R. 1974. What is an hystricognath? In The Biology
of Hystricomorph Rodents, ed. I. W. Rowlands and B.
J. Weir, 7-19 (discussion, 55-60), Symposia of the Zoo-
logical Society of London, no. 34, New York, Academic
Press.

. 1976. Rongeurs caviomorphs de l’Oligocene de

Bolivie, II. Rongeurs du Bassin Deseadien de Salla-Lu-
ribay. Palaeovertebrata 7(3): 15-90. Laboratoire de Pa-
leontologie, Montpellier, France.

Mones, A. 1972. Estudios sobre la Familia Hydrochoeridae
(Rodentia). IV. Elucidation del “status” de Phugatheri-
um cataclisticum Ameghino, 1887. Ameghiniana 9(4):
390-391.

. 1975. Estudios sobre la Familia Hydrochoeridae

(Rodentia). V. Revaluation de sus caracteres morfo-
logicos dentarios con algunas consideraciones sobre la
filogenia del grupo. In Adas del Primer Congreso Ar-
gentino de Paleontologia y Bioestratigrafia (Tucuman
12-16 de Agosto de 1974), 2, 463-476. Universidad
Nacional de Tucuman y Asociacion Paleontologica Ar-
gentina (Tucuman, Argentina, 1974).

. 1976. Filogenia de la Familia Hydrochoeridae

(Mammalia: Rodentia). Preprint, II Seminario sobre
chigiiires y babas (1-4 XII, 1976, Maracay, Venezuela).
Montevideo (circulated by author).

44 Contributions in Science, Number 374

Frailey: Fossil Mammals of the Rio Acre

Mousinho de Meis, M.R. 1971. Upper Quaternary process
changes of the middle Amazon area. Geological Society
of America, Bulletin 82:1073-1078.

ONERN. 1972. Inventario, evaluacion e integracion de los
recursos naturales de la zona de los rios Inambari y
Madre de Dios. Oficina Nacional de Evaluacion de Re-
cursos Naturales (ONERN), Republica del Peru, Lima,
Peru.

. 1977. Inventario, evaluacion e integracion de los

recursos naturales de la Zona Iberia-Inapari. Oficina Na-
cional de Evaluacion de Recursos Naturales, Republica
del Peru, Lima, Peru.

Oppenheim, V. 1937. Geological exploration between up-
per Junta River, Brazil, and middle Ucayali River, Peru.
American Association of Petroleum Geologists, Bulletin
21(1 ):97— 1 10. Tulsa, Oklahoma.

. 1946. Geological reconnaissance in southeastern

Peru. American Association of Petroleum Geologists,
Bulletin 30(2):254-265. Tulsa, Oklahoma.

. 1975. The first (1944) geological exploration of the

upper Amazon Valley of Peru (Valleys of the upper Rio
Madre de Dios, upper Rio Urubamba, Fitzcarrald Pass,
Rio Mishagua and upper Rio Ucayali). Boletin de la
Sociedad Geologica del Peru 45:83-94.

Pascual, R. 1953. Sobre nuevos restos de sirenidos del
Mesopotamiense. Revista de la Asociacion Geologica
Argentina 8(3): 163-1 8 1 .

. 1981. Prepidolopidae, nueva familia de Marsupi-

alia Didelphoidea del Eoceno Sudamericano. Ameghi-
niana, Setiembre de 1980, 1 7(3):2 1 6—242 (published 3
February 1981).

Pascual, R., and P. Bondesio. 1961. Un nuevo Cardiatherii-
nae (Rodentia, Hydrochoeridae) de la Formacion Monte
Hermoso (Plioceno Superior) de la Provincia de Buenos
Aires. Algunas consideraciones sobre la evolucion mor-
fologica de los molariformes de los Cardiatheriinae.
Ameghiniana 2(6):93— 1 1 1.

. 1963. Nuevo tipo de morfologia dentaria en un

Cardiatheriinae (Rodentia, Hydrochoeridae) del Pli-
oceno inferior de Huachipampa (San Juan). Ameghini-
ana 3(2):43-49.

. 1968. Los Cardiatheriinae (Rodentia, Caviomor-

pha) de la Formacion Arroyo Chasico (Plioceno Inferior)
de la Provincia de Buenos Aires. Ameghiniana 5(7):237-
251.

. 1982. Unroedor Cardiatheriinae (Hydrochoeridae)

de la edad Huayqueriense (Mioceno Tardio) de la pam-
pa, sumario de los ambientes terrestres en la Argentina
durante el Mioceno. Ameghiniana, Vol. 19, nos. 1, 2:
19-35.

Pascual, R., N.V. Cattoi, J.C. Francis, O. Gondar, E. Ortega
Hinojosa, E. Tonni, J.A. Pisano, A.G. de Ringuelet, and
J. Zetti. 1966. Vertebrata. In Pa/eontografia Bonae-
rense, ed. A.V. Borello, Fasc. 4, 202 pp. Provincia de
Buenos Aires, Comision de Investigacion Cientifica, La
Plata.

Pascual, R., and O.E. Odreman Rivas. 1971. Evolucion de
las comunidades de los vertebrados del Terciario Ar-

gentine, los aspectos paleozoogeograficas y paleocli-
maticos relacionados. Ameghiniana, Vol. 8, nos. 3, 4:
372-412.

Patterson, B. 1942. Two Tertiary mammals from northern
South America. Novitates, no. 1173, 7 pp. American
Museum of Natural History.

Patterson, B., and R. Pascual. 1968. New echimyid rodents
from the Oligocene of Patagonia, and a synopsis of the
family. Breviora, no. 301, 14 pp. Museum of Compar-
ative Zoology, Harvard University, Cambridge, Mas-
sachusetts.

. 1972. The fossil mammal fauna of South America.

In Evolution, Mammals and Southern Continents, eds.
A. Keast, F.C. Erk, and G. Glass, 247-310, Albany, State
University of New York Press.

Patterson, B., and A.E. Wood. 1982. Rodents from the
Deseadan Oligocene of Bolivia and the relationships of
the Caviomorpha. Bulletin of the Museum of Compar-
ative Zoology, Vol. 149(7):37 1—543. Harvard Univer-
sity, Cambridge, Massachusetts.

Paula Couto, C. de. 1944. Noticla preliminar sobre um
novo toxodonte do Cenozoico do Territorio do Acre.
Boletim do Museu Nacional, N.S., no. 3:1-4.

. 1956. Mamiferos fosseis do Cenozoico da Ama-
zonia. Boletim Conselho Nacional de Pesquisas, Vol. 3,
121 pp. Instituto Brasileiro de Bibliografia e Documen-
ta?ao, Rio de Janeiro, Brasil.

. 1967a. Pleistocene edentates of the West Indies.

Novitates, no. 2304, 55 pp. American Museum of Nat-
ural History.

. 1967b. Contribui9ao a paleontologia do Estado do

Para. Um sirenio na Formaqio Pirabas. Atas do Sim-
posio sobre a Biota Amazonica, Vol. 1 (Geosciencias):
345-357.

. 1971. On two small Pleistocene ground-sloths.

Academia Brasileira de Ciencias, Anais. Rio de Janeiro.
43 (Suplemento):499-513.

. 1976. Fossil mammals from the Cenozoic of Acre,

Brazil. 1. Astrapotheria. 237-249. Anais do XXVIII
Congresso Brasileiro de Geologia, 27 Oct.-2 Nov. 1974,
Porto Alegre, Rio Grande do Sul. Publ. by Sociedade
Brasileira de Geologia, Sao Paulo.

. 1978. Fossil mammals from the Cenozoic of Acre,

Brazil. 2. Rodentia Caviomorpha Dinomyidae. Iherin-
gia, Ser. Geol., Porto Alegre, Vol. 5:3-17.

Prance, G.T. (ed.). 1982. Biological Diversification in the
Tropics. New York, Columbia University Press, 714 pp.

RADAMBRASIL. 1976. “Levantamento de Recursos Na-
turais (Geologia, Geomorfologia, Pedologia, Vegeta^ao,
Uso Potencial da Terra). Folha SC. 19 Rio Branco.”
Vol. 12. Departamento Nacional da Produ^ao Mineral,
Rio de Janeiro.

. 1977. “Levantamento de Recursos Naturais (Geo-
logia, Geomorfologia, Pedologia, Vegetagao, Uso Po-
tencial da Terra). Folha SB/SC. 18 Javari/Contamana.”
Vol. 13. Departamento Nacional da Produ^ao Mineral,
Rio de Janeiro.

Reinhart, R.H. 1951. A new genus of sea cow from the

Contributions in Science, Number 374

Frailey: Fossil Mammals of the Rio Acre 45

Miocene of Colombia. Publications in Geological Sci-
ences, Vol. 28(9):203-214. University of California,
Berkeley, Los Angeles.

Rovereto, C. 1914. Los estratos araucanos y sus fosiles.
Anales del Museo Nacional de Historia Natural, Vol.
25:1-247. Buenos Aires.

Roxo, M.G. de O. 1921. Note on a new species of Toxodon
Owen, T. lopesi Roxo. Rio de Janeiro (Empreza Brasil
Editora):6 pp.

Riiegg, W., and A. Rosenweig. 1949. Contribucion a la
geologia de las formaciones modemas de Iquitos y de la
Amazonia superior. Sociedad Geologica del Peru Vo-
lumen Jubilar, Parte II, 3:1-24.

Rusconi, C. 1939. El premolar inferior de los grandes car-
pinchos extinguidos. Anales de la Sociedad Cientifica
Argentina 1 28(4):233— 2 39.

Savage, D.E., and D.E. Russell. 1983. Mammalian Paleo-
faunas of the World. Reading, Massachusetts, Addison-
Wesley Publ. Co., 432 pp.

Singewald, J.T., Jr. 1927. Pongo de Manseriche. Geological
Society of America, Bulletin 38:479-492. Boulder, Col-
orado.

. 1928. Geology of the Pichis and Pachitea rivers,

Peru. Geological Society of America, Bulletin 39:447-
464. Boulder, Colorado.

Spillman, F. 1 949. Contribucion a la paleontologia del Peru,

una mamifauna fosil de la region del Rio Ucayali.
Publicaciones del Museo de Historia Natural “Javier
Prado”, No. 1:1-40.

Steinmann, G. 1929. Geologie von Peru. Heidelberg, Carl
Winters Universitatsbuchhandlung, 448 pp.

Vanzolini, P.E., and E.E. Williams. 1970. South American
anoles: The geographic differentiation and evolution of
the Anolis chrysolepis species group (Sauria: Iguanidae).
Arquivos de Zoologia 19(1-4): 1-298.

Webb, S.D. 1978. A history of savanna vertebrates in the
New World. Part II: South America and the Great In-
terchange. Annual Review of Ecology and Systematics,
1978, no. 4147:393-426.

Willard, B. 1966. The Harvey Bassler collection of Peru-
vian fossils. Bethlehem, Pennsylvania, Lehigh Univer-
sity Press, 105 pp.

Williams, M.D. 1949. Depositos Terciarios continentales
del Valle del alto Amazonas. Sociedad Geologica del
Peru, Volumen Jubilar, Parte II, 5:1-13.

Wood, A.E. 1974. The evolution of the Old World and
New World histricomorphs. In The Biology of Histri-
comorph Rodents, ed. I.W. Rowlands and B.J. Weir, 2 1-
54 (discussion, 55-60), Symposia of the Zoological So-
ciety of London, no. 34, New York, Academic Press.

Accepted 28 October 1985.

46 Contributions in Science, Number 374

Frailey: Fossil Mammals of the Rio Acre

INSTfHJCT

The Natural H i story M u seu m of Los Angeles Courtly 'publishes 'the results oi ongnal res^Ilf;
in the life and earth sciences in its Contributions in Science' series* Individ pal Contributions
are issued a l: irregular intervals and range i:r: sire from papers of S printed pages ferilengthy

monographs,. ltlljp|l!i|| ill pi|||l;| || :: A i |;1|; ' ;|

Manuscripts submitted for publication will. undergo anonymous peer fevfirri;' Iriiority' iy
given to manuscripts written by members of. tbeyMusewn: staff; "Mini ujseripts sliKnud 'be
prepared in accordance with the requirements outlined below, and submitted lo-die Head, of
the appro|i“iatei:Seetie!h pi t|e ; flip 'A ■

Authors must adhere to

Norrienelature abd.ririf jprged :*ff||fl|^ qlrihese ebciesi i;.: -.

Authors proposing new taxa must indicate that -primary types have been deposited bp
accordance with the recommendations of .the appropriate code, eating the codec bon by lutme
and providing the relevant identifying details. The depository of other study myteoai shoo Id
also be 'eetev

An abstract required for all papers. In Enanriscripts tlealirig with Larin American sul|bc:i:s.;
the ab^trapf bepresenteddn Spanish or Portuguese, as appropriate . as well dy in; f mgl rrir.
Summaries in Other languages are not required but arc sirdngl v recommended where ap~

*■■£2

ilPKWBfcWA5!,! I , ■■■■

PR EPAR Allots o F M A M LSCRI PT

Type manuscript double-spaced, inducing title page., abstract, text, a ch. n o w I edg me:: > ,
erences, tables, and lilpue captions, A paper must total a minimum of & printed page
guideline, 3.5 pages of draft equals about 1 page: of final 'printed' copy); Indicate italieibed
words with Underscores’ do riot use ahl. italic typing element.: .

Each manuscript component should begin on a new- page,, in. the following, sequence: title ;
page; abstraci(s); text; acknowledgments; literature cited; tables, each table— ecmapietetvritlt
title and footnotes A on a separate page; and figure captions. .All pages should be nuritiifod
I||i™

Submit original illustrations. Line drawings are best. prepared for '50 .percent reduction
and should not exceed 14 by 17.5 inches (3.5 by 43 cm) :i.rs size; photographs ate usually
reproduced at full size and should mbt exceed flbytt. 75 Inches (17 by 12 cm, fit 11 page width:)
or 3.25 by 4.5 inches ;(8 by 1 1 cm, single column).

SUBMISSION OF MANUSCRIPT

Transmit three copies of the.. manuscript in a secure should be ac-

companied by a cover letter stgrred by the author who will be' respdnsitiKriibr correspondence
regarding the manuscript. The covering letter should contain a si|t;erneni, that the manuscript
is based on original research, and has not been published elsewhere except .in- abstract dri-
abbreviated form... The letter should also give ii'Kbtmaripn ion prior ^publication of any part
of the research, and whether or not. funds are available to.meet the cost of- unusual forms l ;
or color illustrations. Include copies of |||| pOfmiSions needed tp -r<^pduri^ pdblish^;
material. !i5iiS|p||||;:;;’;. .. 7 ■■ riyriy. ^

Manuscripts will be reviewed -for possible publication with the understand ingriMt thby
have not been pu bli sh sd, sim ul ta n ecus! y submitted... or already accepted jp| publication
elsewhere. This: does ndSHSiide c(rtii||ic||rntion of a complete report that folldisys; p.u blication :
of prelim in aim findings Ipiilpil Copies of any possibly duplsq|||i:ve material ||fcuM be
submitted;. with the' iri : tivriirig seta t for consideration. ri ylvif- fvriyC |§

'PAGE CHARGES

acceptance of a paper; by the Museum., the authof(s) will receive a statement of page
changes from the Managing- Editor, calculated at a current rate. Partial or complete. payment ,
of this statement will be solicited from those authors who have funds available for this v
purpose. Authors without access to funds; for- partial orcOmpkie payment Of page changes;
should so indicate on the statement and return it to the Managing Editor. The rirtu- ved ,
statement only qualifies the paper for publication: acceptance of a paper.' ;dpd: priririt y of '
publication are. in no way dependent on payment of the statement.

i , : : .

f! j M ip

PWiPfiP

•' i'i'i jj'ii1 i^! ] 'i'j1 !i! wi ii ; !

:l>V'K9h!i .'l!i !i !;i: :-!!'■ ll 'ii , ■■,

Hi!

MR

m I

sHl

mm

ii

II. ,■ i

■BJi -li'i

, I'w . 4 '.

attniMmpi8Ku» - ii

, :

Number 37i

U, , ,,

15 May 1986

—ifetSit ’ s Sp i >
fiiffllT": "ml::

rlUSBTOlMi iVisRii ,i®

«tp P r'|' "j!h

1 .

Ill ll h

iiii; '!!!'!:!;

fflisSsffiw MSS

mm

i

:« Jit A-, -m

«|l»p f l

ililBifl

iMIilnWafii

THE SUBFAMILIES OF
!i|||f : SYSTEMATICS OF THE
I • (HYMENOPTERA: ( HAIA 11)011)1 VI

r‘:‘f R-4i':!E4: "

;d

lillSiiil IjilllElSili1;

1 !

■

1 fey :i

' ;; -fid..
tliUlilid: » »,ji

SWiti ■ >i i ' i u\ (

IjflH

iiJBIMiBi

(If.iJiT't '' > Sl*l ;i 'l: i ;i 11 9 'i lij JW! ' •

Ifl

HBiliiil

SKf? 5 Wjffilii li; imJiliSi! if "it; iji
. ■

'* H'V ®pfpvWw-i‘

r;-'»P;.iUjWfeB8 a*!* t"

: mmt

lit

Tii

tifiHl

MMMi'

■

iilly®!.

411* 'X

ipiifp

‘■Mir

lliiCl Htiliil

T ii 3-\ : j jljfSsIlLoi ^ ^

llai®

Sfc Vi j

VslfH

mm

i

:i

■

Bllllflli!' wift
ini

jljjjj | .-..j js.; i. 'Jr

»geles 'County' • 900 Exposition 'Bottlriinaii » Lets 'Angeles, California 90007

i fi'tnj ! iji I'

fclH

3? ft « 1

ini

Ii!

WliBIl'Iwi

mmm

: ‘Jl i ikfi; 1 ,

I! m

i-:

The KCiiesnl.lIc!; of the Natural Hilitory M^iKfelofLos Angeles County have been

i n divi^Siiijffilliie^aiHtd; h uni hers run hop .smhtvfeiyj, re|^i|i||| dl jffife su bject matter.

C:uair..button$ in Science, a miscellaneous series of technical papers describing brig-.:',
inn!: research in the life and earth sciences.

m Stiicnce Bulletin.' a rruiSefellriineotiis series oi monographs describing original research
in the: life and earth sciences. This series was discontinued in. H>78 with the issue of
Numbers 29 and 30; monographs are now published by the Museum in Contributions

SB .Sue nee.

Science Scries, long articles on natural, history topics, generally written ;1br the layman.

topics oi' the publ ications in these series arc sold; through t ie Museum Book Shop. A catalog

!;: j;; ’ M j$pl

SOEM'I HC P1UB1 .nations committee

Craig’ C. Black, Museum. Director
Richard C. Brusca

Daniel M. Cob eh. Committee Chairman.
John M. Harris' ;

Charles L. Hogue

_:L. Kennedy Cj
Robin A. Simpson, Mki'agiing Editor
Errol. Itewhs . : ■

i IMS MW
pB KllwRT

iiiiiii

UulM w J'ifcillv

Ifill

i&fe

Mini

li i 'O'.:.!; i:. !1 ,! ; IS.

! i : : : :

aiiil HHHi

THE SUBFAMILIES OF EURYTOMIDAE AND
SYSTEMATICS OF THE SUBFAMILY HEIMBRINAE
(HYMENOPTERA: CHALCIDOIDEA1

Gerald I. Stage2 3 4 and Roy R. Snelling3 4

ABSTRACT. The family Eurytomidae consists of three subfamilies:
Rileyinae, Eurytominae (=Harmolitinae = Aximinae = Eudecatom-
inae = Isosominae = Decatominae, all NEW SYNONYMY), and
Heimbrinae. The subfamily Heimbrinae is compared with the other
two and is described in detail for the first time. There are two genera
of Heimbrinae, both limited to the Western Hemisphere: Heimbra
in both North and South America and Symbra (n. gen., type-species:
S. cordobensis n. sp.), known only from Argentina; the genus Heini-
brella is transferred to the subfamily Eurytominae.

The two genera and seven species are separated in a key. Heimbra
includes six known species: H. acuticollis Cameron (the type-species),
H. opaca (Ashmead), H. bicolor Subba Rao, H. nigra Subba Rao,
and two new species, H. parallela and H. pallida. One species, H.
opaca, is confined to western North America, H. bicolor is known
from Mexico and Brazil; the remaining species are known from
semiarid regions of southern South America.

Pertinent illustrations of morphological features supplement the
key and descriptions.

Known distribution data are cited and shown on maps.

INTRODUCTION

The present paper was begun many years ago by the senior
author, while still at the Division of Biological Control, Uni-
versity of California, Berkeley, California. The essentially
finished manuscript languished for some years. The junior
author assumed the responsibility for updating the manu-
script and arranging its publication; however, the bulk of the
systematics and of taxonomic decisions must mostly be cred-
ited to the senior author.

SPECIMENS EXAMINED

During the course of this study we have been able to study
material from the following collections: American Museum
of Natural History (AMNH), British Museum (Natural His-
tory) (BMNH), California Academy of Sciences (CAS), Ca-
nadian National Collections (CNC), Natural History Mu-
seum of Los Angeles County (LACM), United States National
Museum of Natural History (USNM), University of Arizona
(UNAR), University of California at Berkeley (UCB), Davis

Contributions in Science, Number 375, pp. 1-17
Natural History Museum of Los Angeles County, 1986

(UCD), and Riverside (UCR), University of Kansas (UKAN),
and the personal collection of J.A. Halstead (HALS).

HISTORICAL RESUME

The genus Heimbra was originally described by Cameron
(1909) for a single species, H. acuticollis, from Mendoza,
Argentina. He placed the new genus among the typical eury-
tomids in the tribe Eurytomini. Another species, H. opaca
(Ashmead, 1894) originally described in Euperilampus, was
transferred to Heimbra by Burks (1958). At that time Burks
assigned the genus to a position between Eurytoma Illiger
and Eudecatoma Ashmead. Peck ( 1 963) placed Heimbra near
Ipideurytoma Boucek and Novicky in the Eurytominae.

The groupings of the genera within the family Eurytomidae
have been more or less stable since Ashmead treated them
in his 1904 revision of the Chalcidoidea. At that time he
recognized five major eurytomid groups to which he accorded
tribal rank: Aximini, Isosomini, Eurytomini, Rileyini, and
Decatomini. These tribes have all been raised to subfamily
rank by various more recent authors (e.g., Burks, 1971) and
it has been necessary to change some of the names for no-
menclatorial reasons. Nevertheless the group concepts have
remained essentially the same through the years. Thus in
Peck’s catalog of the Nearctic Chalcidoidea (1963), five eur-
ytomid subfamilies were recognized: Harmolitinae (=Iso-
somini), Aximinae, Rileyinae, Eurytominae, and Eudeca-
tominae (=Decatomini). Unfortunately no authors have been
able to offer clear characters to distinguish these groups as a
whole, except in the case of the Rileyinae. In Ashmead’s key
the rileyines are distinguished primarily by their thirteen-

1 . This research was supported, in part, by National Science Foun-
dation Grant 6B-1833.

2. Bowles Road, R.F.D. #1, Stafford Springs, Connecticut 06076.

3. Entomology Section, Natural History Museum of Los Angeles
County, 900 Exposition Blvd., Los Angeles, California 90007.

4. Address reprint requests to R.R. Snelling.

ISSN 0459-8113

segmented antenna with two or three ring segments. In the
same key the other groups are distinguished by characters
far more subtle and variable, such as the shape of the thorax,
whether or not the head is comute, and the nature of the
wing venation. Ferriere (1950) in his key to the subfamilies
occurring in Europe used essentially the same characters as
did Ashmead although his wording in the couplets was more
explicit. Nevertheless, from his key it was still apparent that
Rileyinae was the only group truly distinct from the remain-
ing groups within the Eurytomidae. Claridge (1961) added
his endorsement to Ferriere’s groupings but was similarly
unable to clearly characterize any of the groups except the
Rileyinae. The obvious solution to these problems was sup-
plied by Peck, Boucek, and Hoffer (1964) in their work on
the Chalcidoidea of Czechoslovakia. To them the only group
sufficiently distinct from the rest of the eurytomids to warrant
subfamily rank was the Rileyinae while the other three groups
treated by them, Harmolitinae, Eudecatominae, and Eury-
tominae, were all placed together under the latter name.

Burks (197 1) reviewed the higher classification of the Eury-
tomidae within which he proposed to recognize eight subfam-
ilies; Heimbra was assigned to the new subfamily Heimbri-
nae as its sole genus. One new genus, Heimbrella, was added
to the Heimbrinae by Subba Rao (1980), as well as two new
species of Heimbra from Argentina and Brazil. Burks, in
Krombein et al. (1979), in the most recent catalog of Nearctic
Hymenoptera adhered to his 1971 arrangement.

As Burks (1971) had already noted, it was impossible to
characterize the eight subfamilies that he recognized since
none possessed unique sets of features. “It is an unfortunate
fact that every character I have used here for separating gen-
era will intergrade somewhere in the family. There seem to
be no absolute characters in the Eurytomidae .... The world
genera . . . fall into apparently natural groups . . . [that] . . .
cannot always be segregated by non-integrating key charac-
ters. It has not been possible to take out all the genera of a
group at one place in a key.” Why it should have seemed
necessary to establish or continue the use of these undefined
groups was neither explained nor justified and is, we believe,
unsupportable.

Subba Rao ( 1 980) expressed doubt that a family with about
55 genera worldwide should be divided into eight subfami-
lies. Similarly, Boucek, in Boucek, Watsham, and Wiebes
(1981) noted that such subfamilies as . . Eurytominae, Eu-
decatominae and Harmolitinae . . . are still maintained by
some authors . . . without good reason.” In spite of those
sentiments, no formal steps have been taken to rectify this
situation. We, therefore, propose to formally place the
subfamilies Aximinae, Decatominae, Philoleminae, Har-
molitinae, Prodecatominae, and Eudecatominae in synony-
my with the subfamily Eurytominae (NEW SYNONYMIES),
since none can be differentially defined from that subfamily.

The synonymy of the above six subfamilies within the
Eurytominae includes, of course, the transfer of all their com-
ponent tribes and genera to the Eurytominae. Whether the
Eurytominae, as here recognized, is monophyletic remains
to be determined. We suspect that it is not and that one or

more of the old subfamilies may have to be resurrected,
though differently defined. This problem is beyond the scope
of the present paper.

It is clear that the old subfamilies, as they have been here-
tofore characterized, are not separable. For this reason, we
chose to synonymize them. Future systematists who wish to
revive these subfamilies will be forced to advance ample
justification for doing so.

If we recognize three subfamilies in the Eurytomidae, i.e.,
Eurytominae (s. lat.), Rileyinae, and Heimbrinae, it becomes
relatively easy to characterize and separate them. Table 1
shows the distribution among the three subfamilies of what
we consider to be the more important group characters in
the family. It can be seen that the Heimbrinae possess a
combination of characters as distinctive as do the other two
subfamilies and are perhaps even less similar to the Eury-
tominae than are the Rileyinae.

Subfamily Heimbrinae

DIAGNOSIS

Members of this subfamily may be readily distinguished from
all other eurytomids by the following two features in com-
bination: the peculiarly produced scutellum (Figs. 1 l-14)and
the dorsally flattened, heavily sclerotized, partially fused gas-
ter (Figs. 1,2, 12, 14).

DESCRIPTION

FEMALE. Form robust with heavy sclerotization and large
umbilicate punctation on all body sections. Head tightly ad-
pressed to thorax with width subequal to or greater than
pronotal collar; occiput roundly, deeply incised in middle;
compound eyes large, parocular area elevated above inner
eye margin; antennal scrobes laterally and ventrally carinate,
deeply incised, capable of hiding scapes; well-developed ver-
tical lamella present between antennal sockets; malar area
large, convex; malar groove absent; genal carina well devel-
oped laterally, becoming weakly lamellate near mandibles.
Antenna with 1 1 to 1 3 segments; first flagellar segment often
shorter than others but not forming a ring segment; flagellar
segments and club more or less uniformly subcylindrical.
Mandible resting against shield on procoxa. Pronotal collar
with anterior lateral margin produced forward and carinate,
forming socket to receive head; prepectus subcircular and
very small, diameter less than one-half width of tegula; tegula
large, opaque, heavily sclerotized, and heavily punctured;
mesepistemum anteriorly and ventrally carinate and ven-
trally produced forward along middle and forming flat plate
under procoxa; scutellum strongly, acutely produced poste-
riorly at least to base of gaster; metanotum inverted medially
with anterior margin above and behind corresponding part
of posterior margin. Procoxa strongly carinate and with an-
terior and lateral surfaces concave.

Forewing with costal cell broad; prestigma wider than sub-
marginal vein; marginal vein short, scarcely one-half length
of submarginal; stigmal vein not conspicuously enlarged api-

2 Contributions in Science, Number 375

Stage and Snelling: Systematics of Heimbrinae

Table 1. Distribution of characters in Eurytomidae.

Character

Riieyinae1

Eurytominae2

Heimbrinae

Antennal segmen-
tation

13

9-11

1 1 with undivided club or 13
with tripartite club

Ring segments

2 or 3

1

0

Antennal sexual di-
morphism

Absent

Present (except Eudecatomini)

Absent

Malar groove

Present

Present (except Aximini)

Absent

Prepectus

Small, subcircular to subtriangular

Large, subtriangular

Small, subcircular

Tegula

Weakly sclerotized, smooth or hyaline

Weakly sclerotized, smooth or hya-
line

Heavily sclerotized, punctured
and opaque

Shape of scutellum

Not produced

Not produced (except Acantheuryto-
ma)

Greatly produced posteriorly

Shape of gaster

Subcylindrical

Laterally compressed or subcylindri-
cal

Dorsally flattened

Modifications of

T 2-4 (9) and T 2 and 3 (<5) reduced

Petiole often very long and occasion-

T 2 + 3 fused and large, cover-

gaster

laterally with T 5 covering most of
gaster

al minor fusion of terga along me-
son

ing most of gaster

1 Genera examined: Riley, Neorileya, Calorileya, Archirileya.

2 Genera examined: Eurytoma, Systole, Chryseida, Tenuipetiolatus, Bruchophagus, Prodecatoma, Tetramesa, Phylloxeroxenus, Eudecatoma,
Axima, Aximogastra, Bephrata, Bephratoides, Sycophila, Eurytomocharis, Ipideurytoma, Isosomorpha, Isosomodes, Harmolita, Ailomorpha,
Gahaniola, Aximopsis, Euroxysoma, Conoaxima, Phylloxeroxenoides, Eudoxinna.

cally and without spur; postmarginal vein at least one-half
length of marginal.

Gaster dorsally flattened or weakly concave and heavily
sclerotized; petiole small, rectangular in lateral view, with
large pit near upper edge; terga 2 + 3 indistinguishably fused,
covering most of gaster; terga 4 and 5 short; tergum 6 short
and fused to base of tergum 7 which is large, strongly convex
so that posterior margin lies under anterior margin; tergum
8 primarily ventrally situated and often entirely anterior to
most caudad part of tergum 6 + 7; sterna 3 and 4 fused; sterna
5, 6, and 7 greatly reduced with at most only part of sternum
7 visible externally.

MALE. Head, thorax, wings, and legs identical to those
of female. Gaster with same general aspect as that of female
but differing as follows: terga 2 and 3 fused and large, terga
4, 5, and 6 separate and short; sterna 2 through 8 all distinct,
not fused; sternum 2 much longer than other sterna.

DISCUSSION

It is conceivable that new material referable to this subfamily
will necessitate a broader description of Heimbrinae with
many of the currently included characters perhaps being rel-
egated to generic or specific rank. The unique character com-
bination listed in the diagnosis, however, may be considered
basic in that these features are usually conservative, varying
little within other equivalent groups of eurytomids or other
chalcidoids.

The unique feature of the Heimbrinae is the peculiar fusion
of the gastral segments in both sexes. Specimens of H. opaca
have been dissected and the fused segments identified. In H.
opaca the male (Fig. 2) is least complex. All the sterna are
distinct and visible in an undissected specimen. The terga
are more complex since the second and third segments are
indistinguishably fused. Terga 4-8 are all distinct and easily
identified; laterally they approximate the corresponding ster-
na. The spiracle and cercus are easy means for identifying
terga 7 and 8. Note that tergum 2 + 3 is laterally approximate
to both sterna 2 and 3.

The female of H. opaca (Fig. 1) is more complex than the
male. As is true of the male, all the sterna are distinguishable,
although segments 3-6, and most of 7, are normally hidden
in undissected specimens. Again, as in the male, terga 2 and
3 fused. Terga 4 and 5 are narrow but distinct. Tergum 6 is
fused to 7 as a narrow basal band. Tergum 8 is large and
fully visible.

The only other eurytomid genus in which the scutellum is
prolonged into a spine-like process is the Bornean Acan-
theurytoma Cameron, 1911. This genus is presently believed
to belong to the Eurytominae in the broad sense of this paper;
Burks (197 1 ) placed it in the Prodecatominae. In Acantheury-
toma there is no fusion of the gastral segments, the antenna
is not clavate, and the flagellar segments are elongate.

Although Subba Rao (1980) allied his genus Heimbrella
to Heimbra, and thus within the Heimbrinae, we do not agree
that this genus belongs here. The placement of the antennal

Contributions in Science, Number 375

Stage and Snelling: Systematics of Heimbrinae 3

Figures 1-6. Heimbra species. 1, H. opaca, “exploded” lateral view of female gaster, segments numbered; 2, H. opaca, lateral view of male
gaster, segments numbered; 3, portion of posterior margin of mesopleuron, H. opaca; 4, same, H. bicolor; 5, same H. parallela; 6, same, H.

nigra.

4 Contributions in Science, Number 375

Stage and Snelling: Systematics of Heimbrinae

sockets well above the level of the lower eye margin, the
barely produced scutellum, and the unmodified gastral struc-
ture (tergum 1 large, following segments not fused, and gaster
not dorsally flattened) are all at variance with the Heimbrinae
and we suggest that Heimbrella should be transferred to the
subfamily Eurytominae.

The following key has been prepared to facilitate the iden-
tification of the genera and species in the subfamily Heim-
brinae. Since specimens of both sexes are not available for
all species we do not know if the characters utilized will hold
for both sexes in all species. However, with the correlated
characters between the sexes of H. opaca as a guide we have
endeavored to select for use in the couplets those features
that in our opinion have a high probability of being reliably
constant between the sexes in the other species.

KEY TO GENERA AND SPECIES OF
HEIMBRINAE

la. Antenna 13-segmented with 3-segmented club; flagellar
segments broader than long (Fig. 12); scutellum with
lateral, subapical margins concave in dorsal view (Fig.

1 1) ( Heimbra ) 2

b. Antenna 11 -segmented with undivided club; flagellar
segments longer than broad (Fig. 14); scutellum with
lateral, subapical margins convex in dorsal view (Fig.

13) Symbra cordobensis

2a. Mesopleuron contiguously punctured, appearing dull
along posterior margin (Figs. 3, 4); tergum 6 + 7 (7,6)

with deep, transverse groove (Figs. 10, 12) 3

b. Mesopleuron smooth and shiny along posterior margin
(Figs. 5, 6); tergum 6 + 7 (7,3) without transverse groove

(Figs. 7-9) 4

3a. Umbilicate punctures on head and thorax with inner
surface smooth (Fig. 4); integument generally black,

pronotum orange-red bicolor

b. Umbilicate punctures on head and thorax with inner
surface appearing finely granulose (Fig. 3); integument
unicolorous black or very dark reddish-brown .... opaca
4a. Marginal vein at least 1.5 times as long as stigmal vein;
tergum 6 + 7 in lateral view, with dorsal anterior margin
far behind ventral posterior margin (Figs. 7, 8) .... 5
b. Marginal vein only slightly longer than stigmal vein;
tergum 6 + 7, in lateral view, with dorsal anterior margin
not far behind ventral posterior margin (Fig. 9) .... 6
5a. Integument primarily black; scutellum width about 0.65

times length nigra

b. Integument primarily orange-brown; scutellum width

slightly greater than 0.7 times length pallida

6a. Integument largely ferruginous, marked with blackish
on face, thoracic dorsum, and side of gastral tergum 2 + 3,
mesepimeron and propodeum largely blackish; scutellar
process about 0.7 times longer than wide and moderately
to strongly depressed in profile (Fig. 22) .... acuticollis
b. Integument entirely blackish, except dusky ferruginous
on some appendages; scutellar process about 0.6 times

longer than wide and not at all depressed in profile (Fig.
25) parallela

Heimbra Cameron

Heimbra Cameron, 1909:433-434. Type-species: Heimbra
acuticollis Cameron, 1909; monobasic.

DIAGNOSIS

Heimbra can be distinguished from Symbra, the only other
genus known in the subfamily, by its thick, thirteen-seg-
mented antenna, its peculiar, laterally concave scutellum (Fig.

1 1) and in the female by its relatively short ovipositor (Fig.

12) .

DESCRIPTION

FEMALE. Head subtriangular in front view, malar area
only weakly convex; in side view, round dorsally and acute
ventrally with face relatively flat; length of head (measured
along middle from apex of clypeus to top of vertex) about
0.70 times width; antennal scrobe short and broad. Antenna
short and thick, thirteen-segmented; scape short, not reaching
median ocellus; pedicel and all flagellar segments broader
than long; first flagellar segment shorter than subsequent seg-
ments and asymmetrical, dorsal length greater than ventral
length; diameter of second segment slightly less than first and
length not more than 1.50 times first; length of subsequent
flagellar segments about subequal but diameters gradually
increasing; club three-segmented, not enlarged, greatest di-
ameter subequal to preceding segment.

Scutellum width usually much less than 0.80 times length;
lateral margin in dorsal view sinuate with basal portion con-
vex and subapical portion concave, sides nearly parallel sub-
apically.

Anterior lateral margin of tergum 2 + 3 of gaster broadly
rounded where it overlaps sternum 2 + 3; tergum 6 + 7 broad-
ly convex in lateral profile; visible portion of tergum 8 not
more than about one-half length of gaster, usually much less
and situated entirely anterior to most caudad portion of ter-
gum 6 + 7; ovipositor sheaths short.

MALE. Similar to female in most respects except gaster
is modified as indicated in description of subfamily.

DISCUSSION

It is evident that Heimbra is widely distributed in the West-
ern Hemisphere with specimens having been taken from such
widely separated localities as Mendoza, Argentina and Den-
ver, Colorado. The available distribution data suggest Heim-
bra may be more or less restricted to desert and semiarid
regions. As can be seen from the maps all the North American
records are from or adjacent to the arid parts of the western
United States and the arid parts of northern and central
Mexico, while nearly all of the South American records are
from the arid parts of Argentina and adjacent countries. The

Contributions in Science, Number 375

Stage and Snelling: Systematics of Heimbrinae 5

6 Contributions in Science, Number 375

Stage and Snelling: Systematics of Heimbrinae

resulting disjunct distribution is of interest since it adds
another animal genus to the several botanical genera known
to follow the same pattern. Among the latter, Larrea, Atamis-
quea, Cercidium, Koeberlina, Ephedra, Acacia, Caesalpinia,
Condalia, Baccharis, Lycium, Prosopis, Mendora, and Hoff-
manseggia all occur on both continents while being more or
less limited to xeric regions (Johnston, 1940). Unfortunately,
nothing is known of the biology or hosts of the species of
Heimbra but it would be most interesting if this genus were
associated with any of these plants.

The genus Heimbra as presently understood contains six
species, two from North America and four from South Amer-
ica. These six species fall into two distinct groups. The ex-
clusively South American group includes those species in
which the mesopleuron, along its posterior margin is smooth
and shiny (Figs. 5 and 6), the propodeum is divided into
large areolae, with smooth or weakly sculptured inner sur-
faces, and tergum 6 + 7 (9) or 7 (<5) lacks a distinct transverse
groove (Figs. 7-9). Included in this group are H. acuticollis,
H. nigra, H. pallida, and H. parallela.

The second group, consisting of two species, H. bicolor
and H. opaca, ranges from the United States to Brazil. In
these species, the mesopleuron is punctate to the posterior
margin (Figs. 3 and 4), the propodeum is dull and coarsely
and irregularly sculptured and densely punctate, and tergum
6 + 7 (9) or 7 (<3) has a distinct transverse groove (Figs. 1 0 and
12).

Those features which characterize the first, or acuticollis
group, are shared with the one species of Symbra. Symbra
is, however, readily distinguished from all species of Heimbra
by the very long antennal scape that extends above the level
of the anterior ocellus and the long, slender, nine-segmented
flagellum with segments 2-9 longer than broad and segments
7-9 not forming an apical club. Females of Symbra possess
a long ovipositor that extends beyond gastral tergum 6 + 7
when viewed in profile (Fig. 14). Although males of Symbra
are unknown, they presumably will be similar to the females
in antennal structure and in having the scutellar process short
and, in dorsal view, with the lateral, subapical margins con-
vex, rather than concave as in Heimbra.

Heimbra opaca (Ashmead)

Figures 1, 2, 3, 11, 12, 19

Euperilampus opacas Ashmead, 1894:318; 9. Dalla Torre,
1898:358; Viereck, 1906:227; Snow, 1907:129; Schmie-
deknecht, 1909:82; Essig, 1929:849; Cresson, 1928:28;
Peck, in Muesebeck et a!., 1951:516.

Eurytoma opaca: Crawford, 1914:69.

Heimbra opacus: Burks, in Krombein, 1958:72, 82.

Heimbra opaca: Peck, 1963:832-833. Burks, in Krombein

et al., 1979:846.

DIAGNOSIS

This species may be readily distingished from all other known
species of Heimbra by its unique, dull punctation.

DESCRIPTION

FEMALE. Length 4.4 to 4.6 mm. Primarily black al-
though ventral portions and legs may become dark reddish-
brown. Tarsi orange-brown.

Head with large, shallow, closely spaced, umbilicate punc-
tures over entire surface except on clypeus and in antennal
scrobes; umbilicate punctures appearing weakly and finely
granulose within, each bearing a short silvery-white hair whose
length is generally subequal to diameter of puncture; antennal
scrobes and interspaces between punctures similarly granu-
lose but shiny; clypeus glabrous; antennal scrobes short, length
about 1.25 times greater than maximum distance between
lateral carinae near antennal sockets and about two-thirds
length of eyes; scrobes with lateral carinae curved along lower
half; interantennal lamella small; inner eye margins nearly
parallel. Antennal scape length about three times maximum
width near base; relative dorsal lengths of pedicel and first
three flagellar segments: 6:5:4:4. Mandible black apically,
reddish-brown basally.

Thoracic integument dorsally with punctation and vesti-
ture like that on head except punctures on scutellum some-
what larger; pronotum in side view with dorsal outline con-
vex anteriorly, somewhat flat posteriorly; lateral carina of
pronotal collar very weakly oblique, nearly parallel to pos-
terior margin; scutellum broad, width about 0.75 times length;
scutellum with dorsal prominence conspicuous and acute in
lateral view; mesopleuron finely, contiguously punctured along
posterior margin (Fig. 3); propodeum coarsely sculptured,
with large irregular ridges and finely, contiguously punctured
interspaces.

Wing veins dark brown; marginal vein long (measured
from angle on wing margin to base of stigmal vein), slightly
over 1.5 times length of stigmal vein; postmarginal vein in-
distinct apically but clearly short, not more than 0.90 times
length of marginal vein.

Visible parts of all gastral terga weakly, umbilicately punc-
tate, those on dorsum of tergum 2 + 3 merging into large
reticulations; umbilicate punctures appearing finely, weakly
granulose within; dorsum of tergum 2 + 3 medially flat; pos-
terior lateral margin of tergum 2 + 3 weakly concave in lateral
view; tergum 6 + 7 with convexity asymmetrical in lateral
view, posterior margin at meson far in front of corresponding

Figures 7-14. Heimbrinae. 7, lateral view of female gaster, H. pallida: 8, same, H. nigra male; 9, same, H. parallela female; 10, same, H.
bicolor female; 11-12 dorsal and lateral views of female, H. opaca ; 13-14, dorsal and lateral views of female, Symbra cordobensis.

Contributions in Science, Number 375

Stage and Snelling: Systematics of Heimbrinae 7

19 20

Figures 15-20. Heimbrinae, front view of head: 15, Heimbra bicolor, female; 16, H. nigra, male; 17, H. pallida, female; 18, H. opaca, female;
19, H. parallela, female; 20, Symbra cordobensis, female.

part of anterior margin and convexity scarcely projecting
caudad of posterior margin of tergum 2 + 3, with deep trans-
verse groove and lacking conspicuous keel along meson dor-
sally and posteriorly; exposed portion of tergum 8 short,
slightly over one-half length of hind femur, lateral ridge pro-
duced posteriorly into ventrally projecting broad spine.

MALE. Length 4.0 to 4.5 mm. Similar to female in all
pertinent characters described above except as follows: an-
tennal scape bears one subapical, ocelloid spot on lower sur-
face; scutellum narrower, width about 0.65 times length; ter-
gum 8 without lateral ridges; punctation of visible parts of
sterna 2-8 similar to that on exposed parts of terga 5 and 6;

8 Contributions in Science, Number 375

Stage and Snelling: Systematics of Heimbrinae

sternum 8 with lateral, posteriorly directed, blunt spines sim-
ilar to, but smaller than, those on tergum 8 in female.

TYPE DATA

Heimbra opaca was described from a unique female collected
in June at Denver, Colorado and subsequently deposited in
the collection of the American Entomological Society and
now in the USNM.

The only additional published distribution record (Vie-
reck, 1906; Snow, 1907) is based on a single specimen col-
lected in July at Oak Creek Canyon (6,000 ft. alt.), twenty
miles southwest of Flagstaff, Coconino County, Arizona.

SPECIMENS EXAMINED (Map 1)

UNITED STATES. ARIZONA, Cochise Co.: Id, Texas
Canyon, 5,000-6,000 ft. elev., Sept. 8, 1927 (J.A. Kusche;
CAS). Coconino Co.: 12, Ashfork, June 17 (Barber and
Schwarz; USNM). Gila Co.: 12, Cedar Creek, 15 mi. W Fort
Apache, June 21, 1957 (G. Butler and F. Werner; UNAR).
Maricopa Co.: 12, 5 mi. SE Wickenburg, Sept. 2, 1961 (P.D.
Hurd; UCB), on Euphorbia pediculifera. Pima Co.: 16, Pep-
persauce Canyon, Santa Catalina Mts., Aug. 18, 1940 (J.J.
duBois; LACM); 12, Mt. Lemmon, 9,000 ft. elev., Santa
Catalina Mts., Sept. 5, 1939 (R.H. Crandall; UNAR); 12, 16,
Box Canyon, Santa Rita Mts., Sept. 14, 1964 (L. and C.W.
O’Brien; LACM); 16, Santa Rita Mts., 5,000-8,000 ft. elev.,
July (F.H. Snow; UKAN); 16, 10 mi. E Continental, July 18,
1961 (F. Werner and W. Nutting; UNAR). Pinal Co.: 16,
Oracle, Aug. 25, 1934 (I. Moore; USNM). Santa Cruz Co.:
16, Canelo, July 19, 1958 (M.S. Adachi; UNAR); 12, same
locality, Aug. 3, 1956 (G.D. Butler; UNAR); 12, Nogales,
Aug. 24, 1939 (R.H. Crandall; UNAR); 16, 8 mi. NW No-
gales, Sept. 8, 1957 (T.R. Haig; UCD); 16, 13 mi. NNW
Nogales, same date and collector (USNM); 366, Patagonia,
Aug. 10, 1958 (F.G. Werner, M. Adachi; UNAR); 16, W
side, Patagonia Mts., Aug. 9, 1956 (F.G. Werner and G.D.
Butler; UNAR). CALIFORNIA, Inyo Co.: 16, Antelope
Springs, 8 mi. SW Deep Springs, June 15, 1961 (C. A. Toschi;
UCB), on Chrysothamnus ; 12, 16, same locality, July 1, 1961
(J.A. Powell; UCB), 2 on Eriogonum heermannii. Riverside
Co.: 12, Palm Desert, June 5, 1960 (R.L. Westcott; LACM);
16, Deep Canyon, July 1, 1964 (E.I. Schlinger; UCR). San
Diego Co.: 12, 10 mi. NE Ramona, July 21, 1981 (J.A. Hal-
stead; HALS); 266, 0.9 mi. W Oak Grove, July 25, 1979 (J.
LaSalle; UCR). Santa Barbara Co.: 16, Santa Ynez Mts.,
June 24, 1959 (F.D. Parker; UCD). Santa Clara Co.: 12, no
further data (Harkins Collection; LACM). Tulare Co.: 12,
Kaweah Powerhouse Station 3, Ash Mountain, July 3, 1983
(J.A. Halstead; HALS). COLORADO, Rio Blanco Co.: 12,
Meeker, ca. 6,200 ft. elev., July 20-2 1,1919 (AMNH). KAN-
SAS, Baldwin Co.: 12, Baldwin, July (J.C. Bridwell; USNM).
Scott Co.: 16, no further locality, June 20, 1925 (H.O. Deny;
UKAN). MONTANA, Petroleum Co.: 12, 1.5 mi W Winnett,
July 28, 1970 (N.E. Rees; USNM). NEVADA, Humboldt
Co.: 12, Orovada, July 14, 1962 (M.E. Irwin; UCD). NEW
MEXICO, McKinley Co.: 12, Pinedale, Navajo Reservation,
July 22, 1948 (L.C. Wymann; USNM). Otero Co.: 16, Moun-

Map. 1. Distribution of Heimbra opaca (•) and H. bicolor (O) in
North America.

tain Park, June 27, 1940 (D.E. Hardy; UKAN). Valencia
Co.: 16, Sandia Mts., July 17, 1952 (R.H. and L.D. Beamer,
W.E. LaBerge, C. Liang; UKAN), on Croton; 16, 222, Carrizo
Arroyo, 20 mi. W Los Lunas, Aug. 1-23, 1977 (S. and J.
Peck; CNC). UTAH, Uintah Co.: 16, no further locality, July
13, 1911 (USNM).

MEXICO. HIDALGO: 16, 18 mi. NW Pachuca, 6,400 ft.
elev., June 25, 1971 (Ward and Brothers; USNM), on Pro-
sopis laevigata. MEXICO: 16, Ixtapan la Sol, 5,500 ft. elev.,
Aug. 9, 1954 (J.G. Chilcott; CNC). MORELOS: 16, Cuer-
navaca, Aug. 1959 (N.L.H. Krauss; USNM); 16, same lo-
cality, Aug. 1, 1938 (L.J. Lipovsky; UKAN). NUEVO LEON:
12, Chipinque Mesa, 5,400 ft. elev., near Monterrey, July 8,
1963 (H. and A. Howden; CNC). OAXACA: 266, 7 mi. E
Oaxaca, road to Guelatao, 5,400 ft. elev., July 1 4, 1 963 (L.E.
Caltagirone; LACM). SONORA: 16, Alamos, Aug. 12, 1960
(P.H. Arnaud, E.S. Ross, and D.C. Rentz; CAS). ZACA-
TECAS: 12, 10 mi. N Fresnillo, May 10, 1962 (F.D. Parker
and L.A. Stange; UCD); 222, 15 km E Zacatecas, July 30,
1951 (P.D. Hurd; UCB).

DISCUSSION

The available records indicate that H. opaca is widely dis-
tributed through the arid and semiarid regions of western
North America. The few associated plant records may be a
clue to the actual host or hosts of this species, but in all
probability they merely represent adult feeding behavior since

Contributions in Science, Number 375

Stage and Snelling: Systematics of Heimbrinae 9

such records by these collectors only pertain to flower visits
by the insect.

Heimbra bicolor Subba Rao

Figures 4, 10, 15

Heimbra bicolor Subba Rao, 1980:308; 9 <3.

DIAGNOSIS

This species can be distinguished from all others known in
the genus by the combined characters of the smooth, shiny
umbilicate punctures, the presence of a transverse groove on
tergum 6 + 7 (7,6) and the conspicuous coloring.

FEMALE. Length 4.0 mm. Primarily black except for
bright orange-red dorsal portion of pronotum and some mi-
nor exceptions as noted below.

Head integument with large, deep, closely spaced umbil-
icate punctures over entire surface except clypeus and in
antennal scrobes; umbilicate punctures smooth and shiny
within, each bearing a short, silvery-white hair whose length
is subequal to diameter of puncture; interspaces between
punctures and surface of antennal scrobes appearing weakly,
finely granulose but shiny; clypeus glabrous; antennal scrobe
short, length only slightly longer than maximum distance
between lateral carinae near antennal sockets and less than
0.75 times length of eye; scrobe with lateral carina curved
along lower half; interantennal lamella large; inner eye mar-
gins weakly converging below, nearly subparallel. Antennal
scape length less than three times maximum width near base;
relative dorsal lengths of pedicel and first three flagellar seg-
ments: 6:4:4. 5:5. Mandible nearly black apically, dark red-
dish-brown basally.

Punctation and vestiture of thoracic dorsum similar to that
of head except punctures slightly larger; punctures on pro-
notal collar as large as those on mesonotum or scutellum;
pronotum in side view with dorsal outline flat anteriorly,
convex posteriorly; lateral carina of pronotal collar weakly
oblique to posterior margin; scutellum narrow, width about
0.65 times length; scutellum with dorsal prominence con-
spicuous and acute in lateral view; mesopleuron finely, con-
tiguously punctured along posterior margin (Fig. 4); propo-
deum coarsely sculptured with large irregular ridges and finely,
contiguously punctured interspaces.

Wing veins brown; marginal vein long (measured from
angle on wing margin to base of stigmal vein), about 1.5
times length of stigmal vein; postmarginal vein indistinct
apically but more or less subequal in length to marginal vein.
Legs primarily dark reddish-brown, nearly black but tarsi,
femora, and tibiae distally light orange-brown.

Gastral tergum 2 + 3 laterally and ventrally and tergum
6 + 7 ventrally, umbilicately punctate; punctures on tergum
2 + 3 largest and deepest on anterior lateral portion, becoming
small and indistinct on posterior lateral portion and merging
into large reticulations dorsally; umbilicate punctures within
and interspaces between dorsal reticulations smooth and
shiny; visible portion of terga 4, 5, 6 + 7 anteriorly, and 8
appearing finely, weakly granulose; dorsum of tergum 2 + 3
medially flat; posterior lateral margin of tergum 2 + 3 weakly

concave in lateral view; tergum 6 + 7 with convexity asym-
metrical in lateral view, posterior margin far in front of cor-
responding part of anterior margin and convexity extending
slightly caudad of posterior margin of tergum 2 + 3; tergum
6 + 7 with deep transverse groove but lacking conspicuous
keel along meson dorsally and posteriorly; exposed portion
of tergum 8 short, slightly over one-half length of hind femur;
tergum 8 with lateral ridge produced posteriorly into ven-
trally projecting broad spines.

MALE. Length 3.5 mm. Similar to female in all pertinent
characters described above except as follows: antennal scape
with one subapical ocelloid spot on lower surface; scutellum
width about 0.7 times length; tergum 8 without lateral ridges;
visible parts of sterna 2-8 with punctation similar to that of
exposed parts of terga 5-6; spines of sternum 8 smaller than
those of female.

TYPE DATA

The type series is from Nova Teutonia, Brazil, and is de-
posited in BMNH; the holotype has been examined.

DISTRIBUTION

In addition to the type we have seen material of this species
from Brazil, Paraguay, and Mexico (Maps 1 and 2).

SPECIMENS EXAMINED

BRAZIL. MINAS GERAIS: 399, Pedra Azul, Nov. 1972
(Seabra and Oliveira; CNC).

PARAGUAY. 1 <3, San Bernardino (K. Fiebrig; BMNH).
MEXICO. DURANGO: 19, El Palmito, 6,500 ft. elev.,
July 17, 1964 (J.F. McAlpine; CNC). NAYARIT: 19, Ahua-
catlan, July 18-22, 1951 (P.D. Hurd; UCB), on Donnell-
smithia hintonii. VERA CRUZ: 19, 2 mi. N Cerro Azul on
road to Naranjo, 1,500 ft. elev., June 9, 1963 (L.E. Calta-
girone; LACM).

Heimbra parallela, new species

Figures 5, 9, 19, 25, 26

DIAGNOSIS

Heimbra parallela may be separated from all the other known
species of Heimbra by the evenly convex seventh tergum,
the short marginal vein, and the parallel carinae on the mar-
gins of the antennal scrobes.

DESCRIPTION

FEMALE. Length 5.5 mm. Primarily black but with some
minor exceptions as noted below.

Head with large, deep, closely spaced, umbilicate punc-
tures over entire surface except clypeus and in antennal
scrobes; punctures smooth and shiny within, each bearing a
silvery-white hair that is more than 1.5 times diameter of
puncture; interspaces between punctures appearing weakly,
finely, granulose but still relatively shiny; antennal scrobes
with surface distinctly transversely striate and shiny; clypeus
glabrous; antennal scrobe about twice as long as maximum

10 Contributions in Science, Number 375

Stage and Snelling: Systematics of Heimbrinae

Map 2. Distribution of Heimbrinae in South America: Heimbra acuticollis (♦); H. bicolor (•); H. nigra (•); H. pallida (□); H. parallela (O);
Symbra cordobensis (▲). Inset shows coverage of larger map.

distance between lateral carinae near antennal sockets and
nearly equal to eye length; marginal carinae of scrobes nearly
straight and parallel along lower half; interantennal lamella
large; inner eye margins weakly diverging below. Antennal

scape length nearly 3.5 times maximum width at base; rel-
ative dorsal lengths of pedicel and first three flagellar seg-
ments: 7:7:8:8. Mandible black apically, dark reddish brown
basally.

Contributions in Science, Number 375

Stage and Snelling: Systematics of Heimbrinae 1 1

26

23

Figures 21-26. Heimbra species. 21-24, H. acuticollis: 21, dorsal view of scutellum; 22, lateral view of scutellum; 23, front view of head;
24, lateral view of gaster. 25-26, H. parallela: 25, lateral view of scutellum; 26, forewing.

Punctation and vestiture of thoracic dorsum similar to that
of head except as follows: punctures tend to be somewhat
larger and hairs more than twice longer than puncture di-
ameter; puncture size variable, small punctures on pronotum
anteriorly near meson and on mesoscutum anteriorly, larger
punctures on apical portion of scutellum; pronotum in side
view with dorsal outline slightly more strongly convex an-
teriorly than posteriorly; lateral carina of pronotal collar

strongly oblique to posterior margin; scutellum narrow, width
slightly over 0.60 times length its dorsal prominence con-
spicuous and acute in a lateral view (Fig. 25); mesopleuron
smooth and shiny along posterior margin (Fig. 5); propo-
deum posteriorly with two large, flat, smooth, shiny, sub-
rectangular areolae lateral to large, shallowly depressed, ir-
regularly sculptured diamond-shaped areola in middle.

Forewing veins dark brown; marginal vein short (measured

12 Contributions in Science, Number 375

Stage and Snelling: Systematics of Heimbrinae

from angle on wing margin to base of stigmal vein), only
slightly longer than stigmal vein; postmarginal vein indistinct
apically, length 1. 5-2.0 times length of marginal vein (Fig.
26). Legs primarily black but tarsi and ends of femora and
tibiae light orange-brown.

Gastral terga laterally and ventrally umbilicately punctate,
punctures merging into large reticulations dorsally on tergum
2 + 3; punctures on terga 2 + 3, 4, and 5 very shallow and
comparatively small; punctures on terga 6 + 7 and 8 large and
as deep as on pronotal collar; all umbilicate punctures smooth
and shiny within; area between reticulate ridges on dorsum
irregular but very shiny; dorsum of tergum 2 + 3 medially,
weakly concave, appearing medially flat in lateral view; pos-
terior lateral margin of tergum 2 + 3 weakly sinuate, nearly
straight in lateral view; convexity of tergum 6 + 7 nearly sym-
metrical in side view, posterior margin at meson not far in
front of, but considerably ventrad of, corresponding part of
anterior margin and convexity extending considerably cau-
dad of posterior margin of tergum 2 + 3; tergum 6 + 7 with
conspicuous keel along meson dorsally and posteriorly but
lacking deep tranverse groove; exposed portion of tergum 8
long, subequal to length of hind femur; tergum 8 with lateral
ridge entire, not produced posteriorly into ventrally project-
ing broad spine.

MALE. Length 3. 9-4. 7 mm. Antennal scape without ocel-
loid spot. Similar to female as described above except: scu-
tellum width 0.72-0.75 times length; tergum 8 without lateral
ridges; visible parts of sterna 2-8 with punctation similar to
that of visible parts of terga 5-6; sternum 8 with short, barely
evident lateral ridges instead of spines.

TYPE DATA

Holotype female: Salta, ARGENTINA, Feb. 14, 1951 (Ross
and Michelbacher), in the California Academy of Sciences.
Paratypes: 1<3, Pronunciamiento, Entre Rios, ARGENTINA,
Feb. 1965 (collector unknown; CNC); 1<3, Santa Anna, Mi-
siones, ARGENTINA, no date (Dr. Cemosvitov; BMNFI).

ETYMOLOGY

The specific name, Latin for parallel, refers to the parallel
scrobal carinae.

DISTRIBUTION

Presently known only from Argentina and Brazil. Map 2.
DISCUSSION

In addition to the type series, we have seen one female from
Sao Paulo, S.P., BRAZIL, Jan. 2, 1964 (V.N. Alin; AMNH)
and one male from Curitiba- Villa Velho, Parana, BRAZIL,
Nov. 7, 1970 (J.W. Boyes; CNC). The female is very similar
to the type. The male is likewise similar to the two paratypes,
but is considerably larger, about 4.7 mm long compared to
3.9 and 4.2 mm for the two paratypes.

This species is most similar to H. nigra, and the male is
especially similar since both lack an ocelloid spot on the
antennal scape. Both sexes are readily separable from H.

nigra by the characteristic shape of tergum 6 + 7 and by the
presence of a distinct tubercle on the scutellum.

Heimbra pallida, new species

Figures 7,17

DIAGNOSIS

Heimbra pallida can be distinguished from all others in the
South American species group by the ferruginous color, the
long marginal vein, and the far anterior placement of the
posterior margin of tergum 6 + 7.

DESCRIPTION

FEMALE. Length 3.3 mm. Primarily orange-brown with
appendages palest and antennal scrobes, axillae, mesepister-
num ventrally, tergum 2 + 3 dorsally, and tergum 6 + 7 cau-
dally darkest.

Head with large, closely spaced, umbilicate punctures over
entire surface except clypeus and in antennal scrobes; um-
bilicate punctures smooth and shiny within, each bearing a
large, silvery-white hair whose length is nearly 1.5 times
diameter of puncture; interspaces between punctures smooth
and shiny except on vertex posteriorly where they appear
weakly, finely granulose but still shiny; antennal scrobes with
surface shiny but faintly, transversely striate; clypeus gla-
brous; antennal scrobes short, length slightly less than 1.5
times maximum distance between lateral carinae near an-
tennal sockets and slightly less than 0.9 times length of com-
pound eye; scrobes with marginal carinae weakly diverging
along lower half; interantennal lamella small; inner eye mar-
gins nearly parallel. Antennal scape length about 3.5 times
maximum width near base; relative dorsal lengths of pedicel
and first three flagellar segments: 5:3:3:4.

Punctation and vestiture of thoracic dorsum similar to that
on vertex except puncture size varies as follows: smallest
punctures on pronotum anteriorly and mesoscutum ante-
riorly, largest punctures on apical portion of scutellum;
pronotum in side view with dorsal outline nearly evenly
convex; lateral carina of pronotal collar weakly diverging
ventrally from posterior margin; scutellum narrow, width
slightly over 0.70 times length; dorsal prominence of scu-
tellum weakly developed but with three distinct acuities vis-
ible in lateral view; mesopleuron smooth and shiny along
posterior margin; propodeum with two large, flat, smooth,
shiny subrectangular areas lateral to large, shallowly de-
pressed, irregularly sculptured diamond-shaped area along
anterior meson.

Wing veins dark brown; marginal vein long (measured
from angle on wing margin to base of stigmal vein), nearly
1 .6 times length of stigmal vein; postmarginal vein indistinct
apically and 1.0-1. 5 times length of marginal vein.

Gastral terga laterally and ventrally umbilicately punctate,
punctures merging into large reticulations dorsally on tergum
2 + 3; punctures on terga 2 + 3, 4, and 5 very shallow and
relatively small, interspaces appearing granulose; punctures
on tergum 6 + 7 and tergum 8 laterally as large and deep as

Contributions in Science, Number 375

Stage and Snelling: Systematics of Heimbrinae 13

those on head; all umbilicate punctures smooth and shiny
within; area between reticulate ridges on dorsum irregular
but shiny; dorsum of tergum 2 + 3 broadly concave, appearing
flat in lateral view; posterior lateral margin of tergum 2 + 3
strongly, broadly concave in lateral view; convexity of tergum
6 + 7 asymmetrical in side view, posterior margin at meson
far in front of corresponding part of anterior margin and
scarcely projecting caudad of posterior margin of tergum
2 + 3; tergum 6 + 7 lacking conspicuous keel along meson and
deep transverse groove; exposed portion of tergum 8 less
than 0.66 times length of hind femur; tergum 8 with lateral
ridges pronounced and entire but not produced posteriorly
into ventrally projecting broad spine.

MALE. Unknown but probably similar to female in most
respects.

TYPE DATA

Holotype female: Trancas, Tucuman, ARGENTINA, Dec.
14, on foliage (USNM). The type is in the collection of the
United States National Museum of Natural History, Wash-
ington, D.C.

DISTRIBUTION

Known only from Argentina. Map 2.

DISCUSSION

The long marginal vein of the forewing and the anterior
position of the apical margin of tergum 6 + 7 will readily
separate H. pallida from all species except H. nigra. From
that species, H. pallida may be separated by the uniformly
ferruginous color and the relatively broad scutellum. The
color of H. pallida is similar to that of //. acuticollis, a species
with the marginal vein short and with the apical margin of
tergum 6 + 7 only slightly anterior to the anterior margin.

Heimbra nigra Subba Rao

Figures 6, 8, 16

Heimbra nigra Subba Rao, 1980 (1978):308; 2.

DIAGNOSIS

Specimens of H. nigra can be separated from those of the
other species in the South American species group by the
combined characteristics of uniformly dark color, the long
marginal vein of the forewing and by the apical margin of
tergum 6 + 7 being situated much anterior to the anterior,
dorsal margin (Fig. 8). The latter two features are shared with
H. pallida, an entirely ferruginous species with broader scu-
tellum.

The previously undescribed male is described below.

DESCRIPTION

MALE. Length 3.5-4.15 mm. Integument black except
following orange-brown parts: mandible, legs except coxae,
and tergum 2 + 3 laterally.

Head with large, deep, closely spaced umbilicate punctures
over entire surface except on clypeus and in antennal scrobes;
umbilicate punctures smooth and shiny within, each bearing
a long silvery-white hair whose length is less than 1.5 times
diameter of puncture; interspaces between punctures ap-
pearing finely granulose but shiny; antennal scrobes with sur-
faces shiny but faintly, transversely striate; clypeus glabrous;
antennal scrobes short, length less than 1.25 times maximum
distance between lateral carinae near antennal sockets and
0.80 times length of eyes; scrobes with carinae nearly parallel
along lower half; interantennal lamella small; inner eye mar-
gins weakly diverging below. Antennal scape length about
3.0 times maximum width near base; lacking ocelloid spot;
relative dorsal lengths of pedicel and first three flagellar seg-
ments: 6:3:4:5.

Punctation and vestiture of thoracic dorsum similar to that
on head except both punctures and hairs average slightly
larger with smallest anteriorly on pronotum and largest pos-
teriorly on scutellum; pronotum in side view with dorsal
outline evenly convex; lateral carina of pronotal collar strong-
ly oblique to posterior margin; scutellum narrow, width 0.65
times length; scutellum in side view with dorsal prominence
weak and lacking acute projections; mesepimeron smooth
and shiny along posterior margin (Fig. 6); propodeum with
two large, shiny, finely pitted, elevated areas adjacent to deep
groove along meson.

Wing veins dark brown; marginal vein long (measured
from angle on wing margin to base of stigmal vein), about
1.5 times length of stigmal vein; postmarginal vein indistinct
apically, length 1.0-1. 5 times length of marginal vein.

Gastral terga laterally and ventrally umbilicately punctate,
punctures merging into large reticulations dorsally on tergum
2 + 3; punctures on terga 2 + 3, 4, and 5 shallow and small
compared to those on tergum 6 + 7, punctures on tergum 6 + 7
equivalent to those on pronotal collar; all umbilicate punc-
tures smooth and shiny within; area between reticulate ridges
on dorsum irregular but shiny; dorsum of tergum 2 + 3 broad-
ly concave, appearing flat in lateral view; tergum 6 + 7 with
convexity asymmetrical in side view, posterior margin at
meson far in front of corresponding part of anterior margin
and slightly projecting caudad of posterior margin of tergum
2 + 3; tergum 6 + 7 lacking conspicuous keel along meson and
deep transverse groove; visible parts of sterna with puncta-
tion similar to that on lateral parts of tergum 2 + 3; sternum
8 with two lateral, posteriorly directed, blunt spines.

TYPE DATA

The type of H. nigra, a female from Burzaco, Buenos Aires,
Argentina, is in the British Museum (Natural History).

DISTRIBUTION

This species is known from Argentina and Uruguay. In ad-
dition to the type, we have seen the following specimens.
Map 2.

14 Contributions in Science, Number 375

Stage and Snelling: Systematics of Heimbrinae

ARGENTINA. Buenos Aires: 1<3, San Isidro, Mar. 1957
(J. Daguerre; USNM); 12, 8<3<3, Zelaya, Feb. 1957, Nov. 1958,
Dec. 1962 (J. Daguerre; USNM).

URUGUAY. 13, Paysandu, no date (Silveira; USNM).

DISCUSSION

Subba Rao (1980) gives the length of the holotype female as
2.3 mm. The one female, other than the holotype, that we
have seen is appreciably larger, 3.8 mm. Otherwise it is very
similar to the holotype. In the original description Subba
Rao stressed the absence of a dorsal tubercle or elevation on
the scutellum and this character is consistent in the specimens
we have studied.

The male is immediately separable from those of H. opaca
and H. bicolor by the lack of an ocelloid spot on the lower
surface of the antennal scape. In this, it is similar to the male
of H. parallela, from which it is separable by the lack of a
dorsal prominence on the scutellum and the shape of tergum
6 + 7.

Heimbra acuticollis Cameron

Figures 21-24

Heimbra acuticollis Cameron, 1909:434; 9.

This species is known with certainty only from two type
specimens in the BMNH.

The color is largely ferruginous, with irregular areas of
blackish on the face, the thoracic dorsum, and the side of
the gaster blackish. Color aside, H. acuticollis most closely
resembles the type female of H. parallela; i.e., the marginal
vein is short and the anterior, dorsal margin of gastral tergum
6 + 7, in profile view, is only a little distad of the posterior
margin (Fig. 24). However, the scrobal carinae are not par-
allel, the scutellar process is more slender, with the posterior
one-half more depressed, as seen in profile (Fig. 22). The type
is redescribed below.

DESCRIPTION

FEMALE. Length about 4.6 mm. Color primarily ferru-
ginous, but head mostly blackish, with ferruginous mandi-
bles, side of upper frons, center of vertex, and gena; lateral
and anteromedial spots on mesoscutum, irregular axillar
blotch, median blotch on scutellar process, mesopleuron (ex-
cept ferruginous spot below tegula), metapleuron, and pro-
podeum blackish; large irregular blotch on side of tergum
2 + 3 blackish.

Head with large deep closely spaced umbilicate punctures
over all surfaces except clypeus and antennal scrobes; punc-
tures smooth and shiny within, each bearing a silvery-white
hair that is 1.2-1. 5 times longer than a puncture diameter;
interspaces between punctures finely granulose and slightly
shiny; surface of antennal scrobe slightly shiny and appar-
ently finely transversely striate (not readily visible because
of position of antennal scapes); interantennal lamella large;

inner eye margins weakly diverging below; antennal scape
length about 4.3 times basal width; relative dorsal lengths of
pedicel and first three flagellar segments: 12:8:8:8.

Punctation and vestiture of thorax similar to that of head
except as follows: punctures tend to be a little larger and hairs
about twice longer than a puncture diameter; puncture size
variable, punctures smallest anteromesially on pronotum and
mesoscutum, largest on apical portion of scutellum; prono-
tum, in lateral view, convex, highest at scutal margin; lateral
carina of pronotal collar weakly oblique to posterior margin;
scutellum broad, width about 0.65 times median length; dor-
sal prominence conspicuous and acute in profile (Fig. 22),
apex narrowly truncate in dorsal view (Fig. 21); mesopleuron
smooth and shiny along posterior margin (about as in Fig.
5); posterior face of propodeum not visible. Metacoxa black,
remainder of hind leg ferruginous.

Wing vein dark brown; marginal vein (measured from an-
gle on wing margin to base of stigmal vein) subequal to length
of stigmal vein; postmarginal vein distinct for distance about
equal to length of marginal vein.

Gastral terga laterally and ventrally umbilicately punctate,
punctures merging into large reticulations dorsally on tergum
2 + 3; punctures on terga 2 + 3, 4, and 5 very shallow and
smaller than on pronotal collar; punctures on terga 6 + 7 and
8 a little larger and deeper, but obscured by roughened in-
tegument between punctures; all umbilicate punctures smooth
and shiny within; area between reticulate ridges on dorsum
irregular but very shiny; dorsum of tergum 2 + 3 medially,
weakly concave, appearing nearly flat in lateral view; pos-
terior lateral margin of tergum 2 + 3 weakly sinuate, nearly
straight in lateral view; convexity of tergum 6 + 7 nearly sym-
metrical in side view, posterior margin (in profile) not far in
front of, but considerably ventrad of, corresponding part of
anterior margin and convexity extending considerably cau-
dad of posterior margin of tergum 2 + 3; tergum 6 + 7 with
conspicuous keel along meson dorsally and posteriorly but
lacking deep transverse groove (Fig. 24); exposed portion of
tergum 8 long, subequal to length of hind femur; tergum 8
with lateral ridge entire; not produced posteriorly into ven-
trally projecting broad spine.

DISCUSSION

Two syntypes of H. acuticollis are in the BMNH collection.
Of these, the one in best condition is here designated and
appropriately labeled as the lectotype; the second specimen
is the lectoparatype.

The only additional female, other than the types, that we
have seen is larger than the types, length 5.1 mm, and is
almost wholly ferruginous, without the conspicuous dark-
ened areas on the head and thorax. The gaster is distinctly
brownish-ferruginous over most of tergum 2 + 3 and on the
exposed portions of the remaining terga. This specimen dif-
fers most conspicuously from the types in the profile of the
scutellar process; in profile the posterior one-half is sharply
depressed below the level of the basal one-half and the erect
tubercle is absent. Although this specimen may represent still

Contributions in Science, Number 375

Stage and Snelling: Systematics of Heimbrinae 15

another species, it would be unwise to describe this specimen
as such on such meager evidence.

DISTRIBUTION

Known only from Argentina. Map 2.

SPECIMENS EXAMINED

ARGENTINA. 222, Mendozoa (lectotype and lectopara-
type no. 5.351b, BMNH); 12, Estancia La Noria, Rio San
Javier, Santa Fe, Jan. 2, 1912 (G.E. Bryant; BMNH).

Symbra, new genus

Type-species: Symbra cordobensis, new species.

DIAGNOSIS

This genus can be easily recognized from all the other mem-
bers of the subfamily by its slender, eleven-segmented an-
tenna, the broad, laterally convex scutellum (Fig. 13) and in
the female by its relatively long ovipositor (Fig. 14) and all
features associated with it.

DESCRIPTION

FEMALE. Head subcircular in front view, malar area con-
spicuously convex, in side view subrectangular, face distinct-
ly angled near antennal insertions; length of head (measured
along meson from apex of clypeus to top of vertex) nearly
0.80 times width; antennal scrobe long and narrow. Antenna
long and slender, eleven-segmented; scape long, reaching me-
dian ocellus; pedicel and all flagellar segments except first,
longer than broad; first flagellar segment broader than long
and asymmetrical, dorsal length greater than ventral length;
diameter of second segment subequal to first but length more
than 4.0 times first; subsequent flagellar segments gradually
decreasing in length and diameter; club undivided, subequal
to ninth segment. Scutellum width nearly 0.85 times length;
lateral margin in dorsal view more or less convex, subapical
margins strongly converging.

Anterior lateral margin of gastral tergum 2 + 3 narrowly
rounded where it overlaps sterna 2 and 3+4; tergum 6 + 7
acutely convex in lateral profile and without deep transverse
groove; tergum 8 with visible portion nearly equal to length
of remainder of gaster and apex situated posterior to most
caudad portion of tergum 6 + 7; ovipositor sheaths long.

MALE. Unknown but presumably very similar to female
as in Heimbra males.

DISCUSSION

This genus at present contains only a single species, Symbra
cordobensis, which has been taken near Dean Funes in Ar-
gentina. Symbra is undoubtedly very closely related to Heim-
bra as can be seen by the many shared characters listed in
the subfamily and species descriptions. This affinity is par-
ticularly conspicuous between it and the species of Heimbra

from South America as indicated above in the discussion
under Heimbra.

Symbra cordobensis new species

Figures 13, 14, 21

DIAGNOSIS
Same as for the genus.

DESCRIPTION

FEMALE. Length 4.8 to 4.9 mm. Primarily black but
tegula, ventral portion of gaster, and all appendages except
antenna reddish-brown.

Head with large, closely spaced, umbilicate punctures over
entire surface except clypeus and antennal scrobes; umbili-
cate punctures smooth and shiny within, each bearing a long,
silvery-white hair whose length is about 1.5 times diameter
of puncture; interspaces between punctures appearing weak-
ly, finely granulose but still shiny; antennal scrobes with sur-
face conspicuously transversely striate and shiny; clypeus
glabrous; antennal scrobes very long, length over 2.5 times
maximum distance between lateral carinae near antennal
sockets and somewhat longer than eyes; scrobes with lateral
carinae strongly diverging along lower half; interantennal la-
mella large; inner eye margins weakly convex and diverging
below. Antennal scape long, extending to median ocellus,
length nearly 7.0 times maximum width near base; relative
dorsal lengths of pedicel and first three flagellar segments:
7:4.5:17:16.

Punctation and vestiture of thoracic dorsum similar to that
of head except punctures tend to be somewhat larger and
possess slightly longer hairs; punctures and hairs smallest on
pronotum anteriorly and largest on scutellum posteriorly;
pronotum in lateral view with dorsal outline evenly convex;
lateral carinae of pronotal collar weakly developed but strongly
oblique to posterior margin; scutellum wide, width 0.85 times
length; dorsal prominence of scutellum conspicuous and acute
in lateral view; mesopleuron smooth and shiny along pos-
terior margin; propodeum dorsally with large, strongly car-
inate square, diagonally on meson, otherwise scrobiculate
with general surface shiny. Legs with coxae and tibiae darkest,
tarsi lightest.

Wing veins light brown; marginal vein long (measured
from angle in wing margin to base of stigmal vein), nearly
twice length of stigmal vein; postmarginal vein apically in-
distinct, length 1.25-1.50 times length of marginal vein.

Gastral tergum 2 + 3 laterally umbilicately punctate, punc-
tures most distinct anteriorly and least distinct posteriorly
while merging into large reticulations on dorsal surface; um-
bilicate punctures and dorsal reticulations with inner surfaces
smooth and shiny; posterior margin of terga 2 + 3, 4, and 5
appearing weakly, irregularly granulose; tergum 6 + 7 and most
of 8 somewhat more coarsely punctured than preceding two;
ventral margin of tergum 8 finely reticulate; dorsum of ter-
gum 2 + 3 weakly concave medially, appearing flat in lateral
view; posterior lateral margin of tergum 2 + 3 weakly convex
in lateral view; tergum 6 + 7 acutely convex in lateral view,
posterior margin at meson well behind anterior margin and

16 Contributions in Science, Number 375

Stage and Snelling: Systematics of Heimbrinae

apex of convexity projecting far caudad of posterior margin
of tergum 2 + 3; tergum 6 + 7 without conspicuous keel on
meson or conspicuous transverse groove; exposed portion of
tergum 8 very long, more than twice length of hind femur;
tergum 8 with lateral ridge entire, not produced posteriorly
into ventrally projecting broad spine.

MALE. Unknown but probably like female in most per-
tinent characters described above.

TYPE DATA

Holotype female and one paratype: 5 mi. N Dean Funes,
Cordoba, ARGENTINA, Feb. 8, 1951 (Ross and Michel-
bacher; CAS); the holotype is in the collection of the Cali-
fornia Academy of Sciences, San Francisco, and the paratype
is in the Natural History Museum of Los Angeles County.
Map 2.

ACKNOWLEDGMENTS

We gratefully acknowledge the following individuals who
have made available specimens from collections in their care:
Z. Boucek and J. Noyes (BMNH); B.D. Burks and E.E. Gris-
sell (USNM); G.D. Butler (UNAR); R.L. Doutt (UCB); G.
Gordh (UCR); J.A. Halstead (HALS); C.D. MacNeill and
W.J. Pulawski (CAS); the late A.T. McClay (UCD); C.D.
Michener (UKAN); O. Peck and C.M. Yoshimoto (CNC);
J.G. Rozen, Jr. (AMNH).

The original manuscript by Stage was reviewed by B.D.
Burks, R.L. Doutt, and E.F. Riek; the present version was
reviewed and materially improved by the criticisms of
Z. Boucek and E.E. Grissell.

The illustrations mostly were prepared by Mrs. R.L. Gar-
rett.

To all these people we extend our sincere thanks.

LITERATURE CITED

Ashmead, W.H. 1984. Descriptions of new parasitic Hy-
menoptera. American Entomological Society, Transac-
tions, 21:318-344.

. 1904. Classification of the chalcid flies or the su-
perfamily Chalcidoidea, with descriptions of new species
in the Carnegie Museum collected in South America by
Herbert H. Smith. Carnegie Museum, Memoirs, 1(4):
225-551.

Boucek, Z., A. Watsham, and J.T. Wiebes. 1981. The fig
wasp fauna of the receptacles of Ficus thonningii (Hy-
menoptera, Chalcidoidea). Tijdschrift voor Entomolo-
gie, 124:149-233.

Burks, B.D. 1958. Chalcidoidea, p. 62-84. In K.V. Krom-
bein, Hymenoptera of America North of Mexico, Syn-
optic Catalogue. United States Department of Agricul-
ture, Agriculture Monograph, No. 2, Suppl. 1, 305 p.

. 1971. A synopsis of the genera of the family Eury-

tomidae (Hymenoptera: Chalcidoidea). American En-
tomological Society, Transactions , 97:1-89.

Cameron, P. 1909. A contribution to the knowledge of the
parasitic Hymenoptera of Argentina. American Ento-
mological Society, Transactions, 35:419-450.

. 1911. Description of new genera and species of

Chalcididae. Societas Entomologica, Frankfurt, 26:3-4,
7-8, 1 1-12, 14, 19, 22-23, 28.

Claridge, M.F. 1961. An advance towards a natural clas-
sification of eurytomid genera (Hymenoptera, Chalci-
doidea), with particular reference to British forms. So-
ciety of British Entomology, Transactions, 14(7): 167-
185.

Crawford, J.C. 1914. The species of Perilampidae of Amer-
ica north of Mexico. Entomological Society of Wash-
ington, Proceedings, 16:69-76.

Cresson, E.T. 1928. The types of Hymenoptera in the Acad-
emy of Natural Sciences of Philadelphia other than those
of Ezra T. Cresson. American Entomological Society,
Memoirs, 5:1-90.

Dalla Torre, C.G. de. 1898. Catalogues Hymenopterorum.
V. Chalcididae et Proctotrupidae. Leipzig, 598 p.

Essig, E.O. 1929. Insects of western North America. The
Macmillan Company, New York. 1,035 p.

Ferriere, C. 1950. Notes sur les Eurytoma (Hymenoptera,
Chalcidoidea). I. Les types de Thomson et de Mayr.
Mitteilungen der Schweizerischen Entomologischen Ge-
sellschaft, 23(4):377-410.

Johnston, I.M. 1940. The floristic significance of shrubs
common to North and South American deserts. Arnold
Arboretum Journal, 21:356-363.

Krombein, K.V. 1958. Hymenoptera of America north of
Mexico. Synoptic Catalog. First Supplement. United
States Department of Agriculture, Monograph 2, Wash-
ington, D.C., 305 p.

Krombein, K.V., P.D. Hurd, Jr., D.R. Smith, and B.D. Burks.
1979. Catalog of Hymenoptera in America north of
Mexico, Volume 1. Smithsonian Institution, Washing-
ton, D.C., 1,198 p.

Muesebeck, C.F.W., K.V. Krombein, and H.K. Townes.
1951. Hymenoptera of America north of Mexico. Syn-
optic Catalog. United States Department of Agriculture
Monograph 2, Washington, D.C., 1,420 p.

Peck, O. 1963. A catalogue of the Nearctic Chalcidoidea
(Insecta: Hymenoptera). Canadian Entomologist, Sup-
plement 30, 1,092 p.

Peck, O., Z. Boucek, and O. Hoffer. 1964. Keys to the
Chalcidoidea of Czechoslovakia (Insecta: Hymenop-
tera). Entomological Society of Canada, Memoirs, No.
34, 121 p.

Schmiedeknecht, O. 1909. Family Chalcididae. Fascicle 97,
550 p. In P. Wytsman, Genera Insectorum, Brussels.

Snow, F.H. 1907. List of species of Hymenoptera collected
in Arizona by the University of Kansas Entomological
Expeditions of 1 902, 1903, 1904, 1905 and 1906. Kan-
sas Academy of Sciences, Transactions, 20(2): 127-1 39.

Subba Rao, B.R. 1980 (1978). New genera and species of
Eurytomidae (Hymenoptera: Eurytomidae). Indian
Academy of Science, Proceedings, 87B:293-319.

Viereck, H.L. 1906. Notes and descriptions of Hymenop-
tera from the Western United States in the collection of
the University of Kansas. American Entomological So-
ciety, Transactions, 32:173-247.

Accepted 10 January 1986.

Contributions in Science, Number 375

Stage and Snelling: Systematics of Heimbrinae 17

-V

1

■■ }

mill n
nriltl'1

niu

SUM

SERIAL; PUBLICATIONS OF THE

NATURAL

The scientihe pH.biicaiidr)5.'of the Natural History Museum of Los Angeles County have been
issued at irregular intervals in three major series: the articles hi each series are numbered
individually,' and- numbers run consecutively, regardless of the subject matter.

.^Contributions in Science, a miscellaneous series of technical papers describing orig-
inal research in the life and earth. sciences.

# Science Bulletin, a: miscellaneous series of monographs describing original research
in the life and- earth, sciences: This series was discontinued in 1978 with the issue of
N umbers 29 and 3(.h monographs are now published by the Museum in Contributions
: . in Science. :

® Science Series, long articles on natural history topics, generally written for the layman.

s are sold through the Museum Book Shop. A catalog

is available on request.

CONTRIBUTIONS TOWARD A REVISION OF THE NEW WORLD
NOMADINE BEES. A PARTITIONING OF THE GENUS
NOMADA (HYMENOPTERA: ANTHOPHOR1DAE)

Roy R. Snelling1

ABSTRACT. The higher classification of the New World represen-
tatives of the cleptoparasitic bee genus Nomada is reexamined. As
a result, I propose to divide Nomada into three genera: Nomada
Scopoli, Centrias Robertson, and Hypochrotaenia Holmberg.

The New World genus Hypochrotaenia consists of three subgen-
era: Hypochrotaenia s. str. (=Nomadosoma = Polybiapis), Micro-
nomada (= Cep hen), and Aphelonornada (n. subg.; type-species, No-
mada cruralis Moure).

No subgenera are recognized within Centrias ( =Nomadula , N.
SYN.), an exclusively Nearctic genus.

Within the Holarctic genus Nomada six subgenera are recognized:
Nomada s. str. ( =Heminomada = Xanthidium = Phor = Gnathias,
all N. SYN.), Holonomada, Laminomada, Pachynomada, Noma-
dita {— Callinomada , N. SYN.), and Phelonomada (n. subg.; type-
species, N. belfragei (Cresson).

A key is given for the separation of the genera of Nomadini present
in North America.

IN MEMORIAM
P.H. TIMBER LA Ki: 1883-1981

This Contribution is dedicated to the memory of the late
P.H. Timberlake in recognition of his many years of invalu-
able contributions to the systematics of the bees of North
America. The present study was begun as a cooperative effort
with Tim and it is appropriate that this first part honor an
esteemed colleague.

INTRODUCTION

Nomada has hitherto been recognized as a large genus of
usually brightly marked, wasp-like anthophorid bees. All
species are presumed parasitoids in nests of other bees; fe-
males lack all modifications for collecting and transporting
pollen. Although most abundant, in numbers of species, in
the Holarctic Region, the genus extends as far south as Ar-
gentina in the New' World. In the Old World there are species
in northern Australia, the Solomon Islands and the Philip-
pine Islands, and Cape Province, South Africa.

In general, the sanctity of Nomada as a single, very large

Contributions in Science, Number 376, pp. 1-32
Natural History Museum of Los Angeles County, 1986

genus has never been seriously questioned. A few species that
clearly did not belong here were removed to such genera as
Melanomada (Cockerell, 1 903b) and Hesperonomada (Lins-
ley, 1939) in the New World and Acanthonomada (Schwarz,
1966b) in the Old World. Within the very' extensive Old
World fauna no subgenera have been recognized. However,
Nomadita Mocsary, usually treated as a synonym of No-
mada, is equivalent to the New World subgenus Callino-
mada. In the New World, Robertson (1903) proposed to
divide Nomada into seven genera, but has not been followed
by later systematists. Most recent work, such as that of Mitch-
ell (1962), has adhered to the concept of a single genus with
1 1 subgenera, most of them hazily defined.

GENERAL SYSTEMATICS

The vast majority of cleptoparasitic bees belong to the
subfamily Nomadinae, a very large and complex assemblage
of approximately 45 genera arrayed within 1 3 tribes. Despite
the extreme diversity of the Nomadinae, the group appears
to be a monophyietic one (Stephen, Bohart, and Torchio,
1969; Rozen, Eickwort, and Eickwort, 1978; Michener, 1979).
The Nomadinae attain their greatest diversity in the New
World; six tribes and 24 genera are exclusive to the New
World; six tribes and three genera are shared with the Old
World. In contrast, only one tribe (Ammobatoidini) is lim-
ited to the Old World, as are 1 3 genera, mostly within the
Ammobatini.

Based on studies of the mature larvae, Rozen, Eickwort,
and Eickwort (1978), have examined group relationships
within the New World tribes. Their results, derived through
cladistic analyses, corroborate the distinctness of the various
groups, as well as the apparent monophyly, of the subfamily
Nomadinae.

Wamcke (1977, 1982) proposed to drastically revise the

1. Entomology Section, Natural History Museum of Los Angeles
County, 900 Exposition Blvd., Los Angeles, California 90007.

ISSN 0459-8113

Table 1. Host species of Hypochrotaenia and Centrias.

Parasitoid

Host

Authority

Hypochrotaenia,
subg. Micronomada :

cubensis (Cresson)

Exomalopsis globosa (Fabricius)

Raw, 1977

cubensis (Cresson)

E. similis (Cresson)

Raw, 1977

formula (Viereck)

Agapostemon texana Cresson

Eickwort & Abrams, 1980

gutierreziae (Cockerell)

E. crenulata Timberlake

Parker, 1984

suavis (Cresson)

Nomia melanderi Cockerell

Bohart, 1970

species?

E. solani Cockerell

Linsley, 1954

Hypochrotaenia,
subg. Hypochrotaenia:

pilipes (Cresson)

E. globosa (Fabricius)

Raw, 1977

pilipes (Cresson)

E. similis (Cresson)

Raw, 1977

tomentifera (Ducke)?

E. aureopilosa Spinola

Zucchi, 1973

Centrias:

articulata (F. Smith)

A. radiatus (Say)

Eickwort & Abrams, 1980

articulata (F. Smith)

A. virescens (Fabricius)

Eickwort & Abrams, 1980

crotchii (Cresson)

Andrena prunorum Cockerell

NEW RECORD (pers. obs. & pers.
comm. S.L. Buchmann & J. Cane)

higher classification of the bees. He (1977) recognized only
two families: Andrenidae (with subfamilies Colletinae, Ha-
lictinae, and Andreninae) and Apidae (with subfamilies Me-
littinae, Nomadinae, Anthophorinae, Apinae, Ceratinae, and
Megachilinae). Some modification was made in 1982: the
Apidae acquired an additional subfamily, the Dasypodinae,
and the Anthophorinae apparently were merged with the
Melittinae. The Nomadinae, in Wamcke’s concept, include
all cleptoparasitic anthophorids and are asserted to be closely
allied with the Dasypodinae. Wamcke’s proposals are un-
convincing and are based on character states of a few selected
structural apomorphies that clearly evolved independently
many times. The proposed classification has the effect of
uniting very disparate groups on the basis of superficial sim-
ilarities. His conclusions are rejected here.

The principal tribe, in terms of both distribution and di-
versity, is the Nomadini, with 12 genera. Ten of these genera
are exclusive to the New World. The genus Nomada is Hol-
arctic and one genus {Acanthonomada Schwarz, 1966b) is
limited to the eastern Mediterranean region. Of the strictly
New World genera, six are found only in the Nearctic and
three are wholly Neotropic; only Hypochrotaenia is present
in both regions.

Hypochrotaenia, with three subgenera, ranges from Ar-
gentina to southern Canada and its distribution includes the
Antilles. The nominate subgenus is largely South American,
but does range into the extreme southern United States (the
lower Rio Grande Valley of Texas) and throughout the An-
tilles to Cuba. One subgenus ( Aphelonomada ) is confined to
Cuba. The subgenus Micronomada is primarily North Amer-

ican but is present in the northern Antilles and in South
America to Argentina.

The principal hosts of Hypochrotaenia appear to be the
Exomalopsini, the pollen-collecting group apparently most
closely related to the nomadines. At least some species of
Micronomada have retained the primitive cleptoparasitism
in exomalopsine nests, but others switched to halictid hosts
(see Table 1).

The related genus Centrias is confined to North America.
Some species are inquilines in nests of halictids and others
in nests of Andrena (Table 1 ). Like Nomada, Centrias appears
to be a derivative of Hypochrotaenia through Micronomada.

Nomada is probably derived from Hypochrotaenia ( Mi-
cronomada), possibly by way of such groups as Phelono-
mada, Nomadita, and Pachynomada. Among the New World
Nomada, known hosts are mostly species of Andrena, but
apparently do include species of Ha/ictus (Table 2). Phelo-
nomada may be cleptoparasitic on halictids ( Agapostemon
or Nomia). Some Old World species of Nomada parasitize
other genera of bees ( Halictus , Eucera, Panurgus: Perkins,
1919; Melitta, Colletes: Brauns, 1930), but Andrena seems
to be the host for most species (Perkins, 1919).

Once cleptoparasitism of Andrena “. . . arose in the Nearc-
tic region, Nomada had open to it an enormous trophic and
geographic opportunity. The Holarctic distribution of var-
ious groups of Nomada indicates that the intercontinental
dispersal occurred under more mild climates than the pres-
ent, probably when the Bering Strait area was temperate or
perhaps across the North Atlantic in the Paleogene” (Mich-
ener, 1979).

2 Contributions in Science, Number 376

Snelling: New World Nomadine Bees

Table 2. Host species of Nearctic Nomada.

Parasitoid

Host

Authority

subg. Nomada:

annulata F. Smith

Andrena macro Mitchell

Hurd, 1979

crudelis Cresson

A. obscuripennis F. Smith?

Hurd, 1979

imbricata F. Smith

Halictus rubicundus (Christ)?

Hurd, 1979

imbricata F. Smith

A. vicina F. Smith

Packard, 1868

morrisoni Cresson

A. irana Cockerell

Hurd, 1979

obliquella Fowler

A. suavis Timberlake

Linsley & MacSwam, 1959

obliterata Cresson

A. vicina F. Smith

Packard, 1868

obscurella Fowler

A. complexa Viereck

Linsley & MacSwain, 1955a

obscurella Fowler

A. caerulea F. Smith

Linsley & MacSwain, 1959

opaceila Timberlake

A. caerulea F. Smith

Linsley & MacSwain, 1959

opacella Timberlake

A. suavis Timberlake

Linsley & MacSwain, 1959

rhodalis Cockerell

A. cleodora Viereck

NEW RECORD

rubrica Provancher

A. prunorum Cockerell

NEW RECORD

vallesina Cockerell

A. irana Cockerell ?

Hicks, 1934

“sp. A”

Halictus ligatus Say

Rozen, 1966

“sp. C”

A. bisalicis Viereck

Rozen, 1966

“species”

A. morrisonella Viereck

Rozen, 1977

subg. Holonomada:

edwardsii Cresson

A. perimelas Cockerell

Linsley & MacSwain, 1955b

subg. Pachynomada:

undescribed species

A. helianthi Robertson

Parker & Bohart, 1983

undescribed species

A. haynesi Viereck & Cockerell

Parker & Griswold, 1983

subgenus unknown:

“sp. B”

A. flexa Malloch

Rozen, 1966

In preparation for a revisionary study of the New World
Nomada, I have redefined the various groups recognized by
previous authors. It appears now that Nomada consists of
three major phyletic lines treated here as discrete genera.
While maintaining an overall similarity to one another, they
are easily separable in several details.

The Neotropical Hypochrotaenia, originally described as
a genus, was found to be possibly identical to the Nearctic
Micronornada, and has priority over it. This is a moderately
large group, with numerous species in the Neotropics and
seems to be exclusively New World in distribution.

The second genus includes the two erstwhile subgenera
Centrias and Nomadula. Centrias is a small group and seems
to be limited to the Nearctic Region.

All remaining groups belong to the redefined genus No-
mada, with the previously described subgenera Gnathias,
Phor, Heminomada, Holonomada, Laminomada, Pachy-
nomada, and Nomadita. This genus accommodates the ma-
jority of the Nearctic species and is abundantly represented
in the Old World.

In redefining the subgenera I recognize only those based
on more than a single characteristic. If subgenera were based
on a single, unique feature it would be possible to recognize

many more than is done here, but to no practical purpose.
Although they would represent natural species groups, most
would consist of only one or a few closely related species.

A given apomorphy, such as a spinose procoxa, may occur
sporadically; those species possessing this particular char-
acteristic are not necessarily related. Unless a number of
features, and usually not those utilized in species level tax-
onomy, can be correlated, 1 am not inclined to accord these
groups any formal status. By application of this principle it
is ironic that Gnathias, one of the few North American sub-
genera with heretofore unquestioned validity, must be placed
in synonymy.

I recognize the following:

genus Hypochrotaenia Holmberg

subg. Micronornada Cockerell & Atkins
= Cephen Robertson
subg. Hypochrotaenia Holmberg
= Nomadosoma Rohwer
= Polybiapis Cockerell NEW SYNONYMY
subg. Aphelonomada new subgenus
genus Centrias Robertson

= Nomadula Cockerell NEW SYNONYMY

Contributions in Science, Number 376

Snelling: New World Nomadine Bees 3

genus Nomada Scopoli
subg. Nomada Scopoli

= Gnathias Robertson NEW SYNONYMY
= Phor Robertson NEW SYNONYMY
= Heminomada Cockerell & Atkins NEW SYNONY-
MY

= Xanthidium Robertson (not Ehrenberg) NEW SYN-
ONYMY

subg. Holonomada Robertson
subg. Laminomada Rodeck
subg. Pachynomada Rodeck
subg. Nomadita Mocsary

= Callinomada Rodeck NEW SYNONYMY
subg. Phelonomada new subgenus.

The numerous species here assigned to Hypochrotaenia
share an overall similarity of habitus and this genus is prob-
ably monophyletic. Superficially, the species resemble one
another in that they are usually blackish, sometimes partially
or wholly replaced by ferruginous, and with conspicuous
whitish to yellow maculations. The integument is smooth to
polished between deep, distinctly separated punctures; many
species, however, are very finely and sparsely punctate, es-
pecially on the thorax.

Females of Hypochrotaenia always lack a paraocular ridge
or carina, always present in Centrias and Nomada (except
subgenus Phelonomada). In both sexes, the first flagellar seg-
ment is at least as long as the second, and the male flagellar
segments are always devoid of spines or tubercles. In both
sexes, the procoxa has an inner, apical spine or tooth, though
this has been lost in a few species. Most Centrias and a few
Nomada also possess a procoxal spine.

Hypochrotaenia is also characterized by the subangulate
to angulate anterior margin of the side of the pronotum (Fig.
5), a feature shared with species of Centrias. In Nomada. the
margin of the pronotum is evenly convex (Fig. 64). The
characteristic male genital structures, lack of a patterned
subgenital brush, digitiform inner dorsal lobe of the gono-
coxite, etc. are discussed in the section on terminology, see
also Figs. 1 1-14.

The species comprising the genus Centrias are united by
one unusual feature: the protuberant upper portion of the
metapleuron; the protuberant area is crossed by a sharp ridge
or carina (Fig. 65). In some species the ridge is lamelliform
and deflected downward. In a few small species the protu-
berance and ridge are greatly reduced. Although the meta-
pleuron is uniformly simple in Nomada. it is slightly pro-
duced and submarginate in a few Neotropical Hypochrotaenia.

Females of Centrias are easily separated from those of
Hypochrotaenia by the presence of a distinct paraocular ca-
rina. The clypeus of Centrias species is moderately protu-
berant and the paraocular area, between the clypeus and the
paraocular carina is concave. In species of Hypochrotaenia,
if the clypeus is somewhat convex, the paraocular area usu-
ally is also convex.

In addition to the presence of a metapleural prominence,
Centrias females are separable from those of Nomada by the
presence of procoxal spines. Procoxal spines are present in

a few species of Nomada, including those in the aberrant
subgenus Phelonomada; both sexes in this subgenus are un-
usually robust, with distinct punctures separated by shiny
interspaces, the females lack a paraocular carina, both sexes
have a simple metapleuron, and males lack flagellar spines
and modified subgenital brush. A few rare species of Nomada
s. str. also have short procoxal denticles but possess none of
the other features of Centrias in either sex.

Males of Centrias rarely have the middle flagellar segments
simple; usually, at least segment 3 is spinose, and subsequent
segments may be spinose or bispinose. Less often, segments
3-7 may bear tubercles or narrow ridges. In males of most
species of Centrias, the hairs that comprise the subgenital
brush are arranged in characteristic patterns; in both Hy-
pochrotaenia and Nomada the hairs are suberect to erect and
randomly dispersed.

The genus Nomada, as here defined, includes those species
in which the metapleuron is not protuberant above, the fe-
male possesses a distinct paraocular ridge (except in the sub-
genus Phelonomada ), the anterior margin of the pronotum
is evenly convex, the procoxa has no outer carina and only
very rarely bears an inner, apical spine or tooth, the male
subgenital brush consists of suberect to erect hairs of about
the same length that are randomly distributed over the disc
of sternum 7, and the inner dorsal lobe of the gonocoxite is
short and broad.

The acquisition of a procoxal spine has apparently oc-
curred several times in Nomada, being present in all three
species of Phelonomada, and two or three species of Nomada
s. str. that do not appear to be closely related to one another.
This apomorphy is an acquisition that is independent of the
species of Hypochrotaenia and Centrias.

It is my view that the above grouping represents three
major lineages and that each is probably monophyletic.
Whether or not the latter is true remains to be determined
by a major reexamination of the characters of the entire
subfamily Nomadinae and of the Exomalopsinae, the pre-
sumed ancestral group. Pending such a review, the classifi-
cation proposed here will provide a previously nonexistent
framework.

Following the discussion of each genus and subgenus is a
listing of the New World species-group names presently as-
signed to that taxon. Those pertaining to exclusively Neo-
tropical (including the Antilles) forms are marked with an
asterisk.

IMMATURE STAGES

The larvae of parasitic anthophorids, as a whole, have been
the subject of several papers by Rozen (1966, 1969, 1977),
Rozen, Eickwort, and Eickwort ( 1 978), Rozen and McGinley
(1974), and Ehrenfeld and Rozen (1977). Michener (1953)
described larvae of N. fowled; the same specimens were re-
described by Rozen (1966), as well as those of several ad-
ditional species, including that of Hypochrotaenia ( Micron -
omada) suavis. The larva of H. suavis, as described by Rozen,
is very similar to that of N. fowled, and the significance of
the few differences noted is uncertain. Rozen (1966) provided

4 Contributions in Science, Number 376

Snelling: New World Nomadine Bees

a key to the known postdefecating larvae of Nomadinae, and
McGinley (1981) included the Nomadinae in a key to mature
apoid larvae.

TERMINOLOGY

For the most part, the morphological terminology utilized
in the keys and descriptions is conventional, based on that
of Michener (1944). Explanation is necessary for a few char-
acters.

The paraocular ridge is present in females of all nomadine
genera except Hypochrotaenia. In all these genera, the ridge,
and its lower end, extends across the front of the malar area
between the lower end of the eye and the apicolateral angle
of the clypeus (Figs. 19, 25). The paraocular area, immedi-
ately adjacent to the ridge, is more or less definitely concave.
In Hypochrotaenia (and subg. Phelonomada of Nomada) the
paraocular ridge is low and broad along the inner eye margin
and disappears altogether at its lower end; the junction of
the malar area and the paraocular area is uniformly convex
or, rarely, broadly subangulate (Figs. 1 , 3).

The relative lengths of the first several segments of the
flagellum are sometimes useful in characterizing the various
taxa. The first flagellar segment is conspicuously shorter on
one side than the other; the minimum length of this segment
is measured on this shorter side. The maximum length of
the second segment is measured on the same side as mini-
mum length of the first segment.

Several characteristics of the pronotal collar are used in
defining the various groups. In some groups, such as Mi-
cronomada, it may be swollen and with subhemispherical
lateral lobes. More commonly it is compressed and subcar-
inate or ridged across the middle and with distinctly angulate
lateral lobes. This character is, however, sufficiently variable
that it is unreliable.

On the side of the pronotum there is a more or less sharply
defined vertical ridge immediately behind the anterior mar-
gin, that extends dorsad onto the neck of the pronotum.
Except in Hypochrotaenia this lateral ridge extends mesad
on the pronotal neck, becoming evanescent before attaining
the midline; a distinct groove usually separates the lateral
ridge from the base of the pronotal collar. In Hypochrotaenia
the upper end of the lateral ridge commonly turns toward
the base of the pronotal collar near the lateral lobe; this is
especially clear in Hypochrotaenia s. str. in which the lateral
ridge may become cariniform.

In the genera Centrias and Hypochrotaenia the anterior
rim of the pronotum, at a point approximately directly an-
terior to the lower margin of the posterior lobe of the prono-
tum, is more or less abruptly angulate (Fig. 5). In Nomada
(except Nomadita) the anterior rim is evenly rounded, very
rarely weakly subangulate (Fig. 64).

The upper end of the metapleuron is more or less sharply
protuberant in Centrias (Fig. 65), but not in Hypochrotaenia
and Nomada (Fig. 66). This metapleural prominence is crossed
by a sharp cariniform ridge which, posteriorly, joins the lat-
eral arm of the metanotum. In some species this ridge may
be broad, lamelliform, and sharply deflected ventrad.

Only in the subgenera Holonomada and Laminomada of

Nomada does the propodeum exhibit any group character-
istic of interest. The side of the propodeum, mesad of the
propodeal spiracle, is flat in all except these two subgenera
when viewed in oblique profile. In a few species of Holo-
nomada this area is sharply elevated, but in others (and in
Laminomada) it is merely evenly convex.

The last visible stemite of the males of all groups bears
numerous hairs that differ, in various ways, from those of
the foregoing segments. These hairs comprise the subgenital
brush. In most Nomada the hairs of the subgenital brush are
dense, short, stiff and, when viewed in profile, more or less
definitely inclined cephalad. In a few species they may be
rather long and/or fully erect and/or slightly to strongly curled
at the tips. The disc of the segment, in profile, is flat, more
or less abruptly rounded distad.

In all but a few species of Centrias and Hypochrotaenia,
the hairs of the subgenital brush are mostly long, fully erect,
and distinctly curled; those at the sides are often curled to-
ward the midline and those of the posterior middle curled
cephalad so that a “basket” is formed. The profile of the
segment may be flat or concave, but the apical margin is thin.
In some species of Centrias the subgenital brush may be
complex, composed of distinct clusters of quite long hairs
which may be prostrate for much of their length.

The genitalic capsule of the male offers many characters
of value in separating the various groups. In dorsal view, the
gonocoxite is rather short and broad and the two segments
are not contiguous basad. The area between the two gono-
coxites is termed here the median sinus and the ratio of length
to greatest width is a useful diagnostic feature.

The gonocoxite in Nomada (Fig. 30) and Hypochrotaenia
(Fig. 13), has a more or less distinct inner, dorsal lobe; the
lobe may be short and apically truncate or oblique (Nomada)
or it may be elongate, digitiform and rounded distally (Hy-
pochrotaenia). Between the dorsal lobe and the outer lobe is
the dorsal invagination; the depth and width are useful char-
acteristics. In Centrias (Fig. 24) the dorsal lobe and dorsal
invagination are usually obsolescent.

In most Nomada the gonostylus is usually short and rather
stout; in Centrias and Hypochrotaenia it is most commonly
slender and elongate. A setose basoventral lobe is present on
the gonostylus of most Nomada (Figs. 31,37,51), but absent
in Hypochrotaenia (Fig. 12) and absent or poorly developed
in Centrias (Fig. 24). In many Nomada there is, in addition,
a cluster of short, flattened setae on the inner, basal margin
of the gonostylus; no such basal seta-cluster is present in
either Centrias or Hypochrotaenia.

In most groups of Nomada the gonostylus is thickly cov-
ered, on its distal portion, with long, often plumose, setae.
These setae commonly are attenuated toward the apices and
usually the attenuated portions are curled and/or sinuate (Fig.
30). Although setae may be abundant in Centrias and Hy-
pochrotaenia they never attain the characteristic appearance
of those of Nomada. More commonly the setae are short and
straight and more sparse in these two genera.

The aedeagus is generally similar in all groups, but there
are some differences, especially between species groups, in
the dorsobasal lobes. The subgenus Pachynomada of No-

Contributions in Science, Number 376

Snelling: New World Nomadine Bees 5

mada is unique in that the aedeagus has a distinct ventral
hook (Fig. 49).

SYSTEMATICS

Following is a key to the North American genera of No-
madini. Although Michener ( 1954) indicated that such tribes
as Osirini and Epeolini should be merged with Nomadini,
for purposes of this key, Nomadini is as defined by Michener
(1944). While the tribal characters given by Michener are
not universally applicable, they do work for the North Amer-
ican fauna.

Michener (1944) defined the Nomadini as follows: “The
scopa is absent, the basitibial plates are absent, and the jugal
lobes of the posterior wings are very small. The pterostigma
is rather large, almost always considerably broader than the
distance from the inner margin of the prestigma to the costal
margin. The marginal cell is conspicuously longer than the
distance from its apex to the wing tip and pointed or narrowly
rounded [broadly rounded in Coelioxoides]. The pygidial plate
is present in both sexes. The seventh sternum [i.e., the sixth
gastric sternum] of the female is subtruncate to bilobed at
the apex, which is provided with a few to many coarse, spine-
like setae.”

Coelioxoides is a small Neotropical genus that ranges from
central Mexico to Brazil. It is provisionally included within
the Nomadini but probably belongs in a separate tribe. How-
ever, a new tribe in this complex would be premature and a
complete reevaluation of all groups within the Nomadinae
is needed.

KEY TO NORTH AMERICAN
GENERA OF NOMADINI

la. Thorax normal, maximum width about equal to height;
median flagellar segments at least as broad as long, rarely
broader; integument of head and thorax distinctly punc-
tured, rarely vertex and dorsum of thorax almost im-

punctate; head and body maculate or not 2

b. Thorax flattened, maximum width nearly twice height;
flagellum short, middle segments broader than long; in-
tegument smooth, polished, virtually impunctate; head

and body immaculate Paranomada

2a. Apex of marginal cell pointed and on wing margin or
narrowly rounded and slightly removed from wing mar-
gin; third submarginal cell, when present, no larger than
first; male with metatibial spurs; mesometapleural su-
ture distinct 3

b. Apex of marginal cell broadly rounded and well removed
from wing margin; third submarginal cell larger than
first; male without metatibial spurs; mesometapleural

suture absent Coelioxoides

3a. Anterior and posterior basal angles of mandible almost
equidistant from eye; maxillary palpus not more than
half as long as galea, three- to six-segmented; male gono-
stylus rather long and slender; sixth gastric sternum of
female produced posteriorly as a broad, acute or round-
ed, hairy process; seventh gastric sternum of female with

a pair of small, slender apical lobes with a few coarse

setae along margin 4

b. Anterior basal angle of mandible more distant from eye
than posterior basal angle; maxillary palpus six-seg-
mented, more than half as long as galea; male gonostylus
short and robust (rarely slender); sixth gastric sternum
of female truncate or bilobed at apex; seventh gastric
sternum of female subtruncate or with a pair of broad,
well-separated lobes provided with coarse, spine-like se-
tae 6

4a. Apex of marginal cell acute, on wing margin; maxillary

palp five- or six-segmented 5

b. Apex of marginal cell narrowly rounded and slightly bent
from wing margin; maxillary palpus three- or four-seg-
mented Triopasites

5a. Forewing with three submarginal cells; maxillary palp
six-segmented; middle coxa as long as distance from

summit to base of hind wing Melanomada

b. Forewing with two submarginal cells; maxillary palp five-
segmented; middle coxa shorter than distance from sum-
mit to base of hind wing Hesperonomada

6a. Gastric segments usually more or less conspicuously yel-
low to white maculate, without transverse pubescent fas-
ciae; marginal cell acutely pointed on wing margin; sixth
gastric sternum of female concealed, truncate or weakly
bilobed at apex, with a few spine-like setae apicad; first
gastric tergum projecting as transparent ventral flaps in

dried specimens 7

b. Gastric segments immaculate, with transverse pubescent
fasciae; apex of marginal cell slightly separated from
wing margin, narrowly rounded; sixth gastric sternum
of female exposed, bilobed, lobes margined with a row
of spine-like setae; lateral lobes of first gastric tergum

closely appressed to first sternum Hexepeolus

7a. Upper end of metapleuron not protuberant and without
transverse carina or lamella (Fig. 66); procoxa simple
or, if spinose, then first flagellar segment equals or ex-
ceeds second or gastric terga are weakly and superficially

punctate 8

b. Upper end of metapleuron protuberant, sharply margin-
ate or lamellate along edge (Fig. 65); procoxa spinose or
dentate; first flagellar segment conspicuously shorter than
second; gastric terga sharply punctate (rarely punctures

weak) Centrias

8a. Procoxa seldom spinose. Female: Paraocular ridge pres-
ent (except Phelonomada)\ anterior rim of pronotum
narrow and evenly rounded at side, rarely subangulate
or angulate (Fig. 64). Male: Gonostylus usually with ba-
soventral setose lobe (Figs. 3 1 , 44, 63); inner dorsal lobe
of gonocoxite short and broad (Fig. 30), if somewhat
elongate, outer lobe is short and broad and median sinus

is broader than long Nomada

b. Procoxa usually spinose. Female: Paraocular ridge ab-
sent; anterior rim of pronotum broad, angulate or sub-
angulate at side (Fig. 5) (rarely not, in which case upper
end of lateral ridge turns dorsad onto lobe of collar).
Male: Gonostylus without basoventral setose lobe (Fig.
12); inner dorsal lobe of gonocoxite digitiform (Figs. 1 3,

6 Contributions in Science, Number 376

Snelling: New World Nomadine Bees

14), but if not, outer lobe very slender and median sinus
is longer than broad (Fig. 11) Hypochrotaenia

Genus Hypochrotaenia Holmberg

DIAGNOSIS. Female without paraocular ridge; anterior
rim of pronotum angulate or subangulate at side; procoxa
usually spinose; minimum length of first flagellar segment
equal to, or exceeding, maximum length of second; inner
dorsal lobe of male gonocoxite usually digitiform, never short
and broad, invagination deep; gonostylus of male without
setose basoventral lobe.

DESCRIPTION. (A) Mandible without preapical tooth.
(B) Paraocular ridge of female absent. (C) Minimum length
of first flagellar segment equal to, or exceeding, maximum
length of second. (D) Gena less than half as wide as eye in
profile, posterior margin rounded. (E) Procoxa usually spi-
nose or dentate at apex; outer carina present or absent. (F)
Anterior rim of pronotum subangulate to angulate at side;
pronotal collar somewhat swollen or narrow and with dorsal
crest; lateral ridge at its upper end often turned toward pro-
notal collar. (G) Metapleuron simple or (rarely) protuberant
and margined. (H) Metacoxa rounded above or compressed
and sublamellate to lamellate. (I) Gastric terga almost always
sharply and closely punctate; impunctate apical margins nar-
row to broad. Male. (J) Hairs of subgenital brush usually
fully erect or with apices curled. (K) Inner dorsal lobe of
gonocoxite usually digitiform, rarely another shape, but nev-
er short and broad; outer lobe elongate; invagination deep;
median sinus longer than broad. (L) Gonostylus without se-
tose basoventral lobe or inner, basal seta-cluster.

DISCUSSION. Hypochrotaenia was described by Holm-
berg (1886) for a single South American species. Cockerell
(1916b) recognized that Nomadosoma, based on a Cuban
species, could not be separated from Hypochrotaenia. That
some Neotropical Micronomada should be placed in Hy-
pochrotaenia was correctly noted by Rodeck (in Cockerell,
1949) and Michener (1954) treated all Panamanian species
as members of Hypochrotaenia. Michener also enumerated
the essential differences between Hypochrotaenia and Mi-
cronomada, both treated as subgenera of Nomada.

Polybiapis was described by Cockerell (1916b), as a genus
closely allied to Nomada, for two Brazilian species: mimus
Cockerell and polybioides (Ducke). A third species, abnormis
(Ducke), also belongs here. The principal distinction for Po-
lybiapis is that there are but two submarginal cells, the trans-
verse cubital vein being absent. This is likely not significant;
a few species of Nomada also lack this vein but are otherwise
normal for their groups. Some species of subgenus Hypo-
chrotaenia may, in one or both sexes, lack the second trans-
verse cubital vein. In features other than wing venation. Po-
lybiapis is an unremarkable Hypochrotaenia.

Some South American species of Hypochrotaenia have a
metapleural structure very similar to that of Centrias and,
in some, the protuberance is more or less sharply marginate.
Both sexes may be readily separated from Centrias by the
long first flagellar segment. The females lack a paraocular

ridge and the genitalic structures of the males are very dif-
ferent from those of Centrias.

KEY TO SUBGENERA OF HYPOCHROTAENIA

la. Labrum broader than long (Fig. 4); procoxa usually spi-
nose; outer carina of procoxa, when present, complete

2

b. Labrum distinctly longer than broad (Fig. 2); procoxa

without distal spine or tooth; outer carina of procoxa
present on distal half only (Fig. 7) .... Aphelonomada
2a. Outer carina of procoxa present (Fig. 8); metacoxa of
female sublamellate to lamellate above; metatibia with
long, suberect, coarse spiniform setae, at least in female

(Fig. 10) Hypochrotaenia

b. Outer carina of procoxa absent (Fig. 6); metacoxa of

female rounded above; metatibia with coarse tubercles,
but without spiniform setae, except at apex (Fig. 9)
Micronomada

Subgenus Micronomada Cockerell and Atkins

Figures 3-5, 6, 9, 11-12, 15

Nomada subg. Micronomada Cockerell and Atkins, 1902:
44. Type-species: Nomada modesta Cresson, 1863; origi-
nal designation and monobasic.

Cephen Robertson, 1903:174, 176. Type-species: Nomada
texana Cresson, 1872; original designation and monoba-
sic.

DIAGNOSIS. Procoxa without outer carina; metacoxa
rounded above; lobes of pronotal collar often swollen; meta-
tibia without spiniform setae except at apex; labrum broader
than long.

DESCRIPTION. (1) Mandibular apex subacute to blunt.
(2) Labrum about one-half wider than long, not extending
beyond lower margin of closed mandibles. (3) Clypeus con-
vex to (rarely) flattened. (4) Face, in profile, protuberant be-
low, rarely flattened. (5) Frons and thorax shiny to dull be-
tween dense to contiguous punctures over most areas. (6)
Pronotal collar narrow to broad, usually rounded and some-
what swollen; lateral ridge directed mesad or, rarely, toward
base of collar. (7) Mesoscutum convex to slightly flattened;
scutellum flat or convex, with distinct sloping posterior face.
(8) Procoxa with apical spine or tubercle, rarely aspinose;
outer carina absent. (9) Metacoxa rounded dorsally; meta-
tibia more or less distinctly tuberculate along length, but
without distinctly spiniform setae except at apex. (10) De-
pressed apical margin of second and third gastric terga punc-
tate nearly to margin.

DISCUSSION.The separation between the subgenera Hy-
pochrotaenia and Micronomada may not prove valid when
more species are studied, especially those of the Neotropics.
A few with the rounded metacoxae typical of Micronomada
have a sharp lateral pronotal ridge and the lobes of the pro-
notal collar are angulate dorsally as in Hypochrotaenia. In
other features they are more like Micronomada and are placed
here for the present.

Micronomada is primarily North American. Species fly

Contributions in Science, Number 376

Smelling: New World Nomadine Bees 7

Figures 1-5. Hypochrotaenia spp. 1 and 2, H. ( Aphelonomada ) cruralis, female, frontal and ventral views of head; 3 and 4, H. (Micronomada)
texana, female, same. 5, H. (M.) texana, lateral view of pronotum.

8 Contributions in Science, Number 376

Snelling: New World Nomadine Bees

from spring to autumn. Hosts for species of Micronomada
are, so far as now known, limited to the families Halictidae
and Anthophoridae (Table 1).

All names listed below are new combinations in the genus
Hypochrotaenia.

INCLUDED NAMES

amorphae (Swenk, 1913)
arenicola (Swenk, 1913)
convolvuli (Swenk, 1913)
crassula (Cockerell, 1903d)
crucis (Cockerell, 1 903a)
cubensis (C resson, 1865)*
fervida (F. Smith, 1854)
flavipes (Provancher, 1888)
formula (Viereck, 1903)
garciana (Cockerell, 1 907c)
gutierreziae (Cockerell, 1896)
heiligbrodtii (C resson, 1878)
jamaicensis (Cockerell, 1912)*
lamarensis (Cockerell, 1905c)
lippiae (Cockerell, 1 903a)
melanoptera (Cockerell, 1921)
mitchelli (Cockerell, 1911a)
modesta (Cresson, 1878)
nitescens (Cockerell, 1905c)
neomexicana (Cockerell, 1903a)
pallidelutea (Swenk, 1915)
pampicola (Holmberg, 1886)*
panamensis (Michener, 1954)*
peninsularis (Cockerell, 1925)
putnami (Cresson, 1876)
ridingsii (Cresson, 1878)
rivertonensis (Cockerell, 1 903d)
semisuavis (Cockerell, 1910c)
simplicicoxa (Swenk, 1915)
suavis (Cresson, 1878)
sublippiae (Cockerell, 1907c)
texana (Cresson, 1872)
tiftonensis (Cockerell, 1903d)
trapidoi (Michener, 1954)*
uhleri (Cockerell, 1905c)
vegana (Cockerell, 1 903d)
vierecki (Cockerell, 1903a)
wisconsinensis (Graenicher, 1911)

Subgenus Hypochrotaenia Holmberg

Figures 8, 10, 13, 17-18

Hypochrotaenia Holmberg, 1886:273. Type-species: Hy-
pochrotaenia parvula Holmberg, 1886; original designa-
tion and monobasic.

Nomadosoma Rohwer, 1911:24. Type-species: Pasites pi-
lipes Cresson, 1865; original designation and monobasic.
Polybiapis Cockerell, 1916b:208. Type-species: Polybiapis

mimus Cockerell, 1 9 1 6b; original designation. NEW SYN-
ONYMY.

DIAGNOSIS. Procoxa with complete outer carina; meta-
coxa of female sublamellate to lamellate; metatibia of female
with coarse, spiniform setae along most of length of outer
face.

DESCRIPTION. (1) Mandibular apex blunt, rarely sub-
acute; external surface flattened except on apical one-third.
(2) Labrum about one-half broader than long, in repose not
extending beyond lower margin of closed mandibles. (3)
Clypeus flattened; less commonly weakly convex. (4) Face,
in profile, usually distinctly flat. (5) Frons and thorax shiny
between punctures ranging from dense to nearly absent (latter
condition especially common on mesoscutum). (6) Pronotal
collar narrow, subangulate to crested above; upper end of
pronotal ridge turned toward lateral lobes of pronotal collar.

(7) Mesoscutum, in profile, weakly convex to flat; scutellum
weakly convex or flat, usually without discrete posterior face.

(8) Procoxa spinose or dentate; outer carina present, com-
plete. (9) Metacoxa sharply compressed and sublamellate to
lamellate above (except some males); metatibia with scat-
tered spiniform setae along its length (except some males).

( 1 0) Depressed apical margin of second and third gastric terga
punctate nearly to apical margin.

DISCUSSION. This is an essentially Neotropical group
with many species, most of which seem to be undescribed.
At least one species, presently undescribed, enters the United
States in southern Texas. Some species of Micronomada es-
pecially those allied to H. gutierreziae, closely resemble Hy-
pochrotaenia s. str., particularly in the flattened face and
mesoscutum. However, the structure of the pro- and meta-
coxae and of the metatibiae are as in typical Micronomada.

INCLUDED NAMES

abnormis (Ducke, 1912)*
costalis (Brethes, 1 909)*
costaricensis (Schrottky, 1920)*
hondurasica (Cockerell, 1 949)*

? krugii (Cresson, 1878)*
limata (Cresson, 1878)*
mexicana (Cresson, 1878)*
mimus (Cockerell, 1 9 1 6b)*
nigrescens (Friese, 1921)*
parvula Holmberg, 1886*
pilipes (Cresson, 1872)*
polvbioides ( Ducke, 1908)*
xanthaspis (Cockerell, 1 949)*
xanthopus ( Friese, 1921)*
zamoranica (Cockerell, 1949)*

Aphelonomada new subgenus

Figures 1-2, 7, 14, 16

DIAGNOSIS. Labrum longer than broad, its apex sub-
truncate and extending over base of mouthparts (Fig. 2).

DESCRIPTION. (1) Mandibular apex blunt. (2) Labrum

Contributions in Science, Number 376

Snelling: New World Nomadine Bees 9

Figures 6-10. Hypochrotaenia spp. 6-8, external view of procoxa of H. ( Micronomada ) texana; 7, H. (Aphelonomada) cruralis; 8, H. (H. )
mexicana. 9-10, outer face of female metatibia of: 9, H. (M.) texana; 10, H. (//.) mexicana (pilosity reduced).

longer than broad, in repose covering half or more of pro-
boscidial fossa. (3) Clypeus weakly convex. (4) Face, in pro-
file, flattened. (5) Head and thorax smooth and shiny between
sparse to scattered punctures. (6) Pronotal collar narrow, but

rounded, dorsally. (7) Mesoscutum somewhat flattened; scu-
tellum flat dorsally and with distinct posterior face. (8) Pro-
coxa without apical spine or tooth; outer carina present on
distal half only. (9) Metacoxa, from above, rounded basad.

10 Contributions in Science, Number 376

Snelling: New World Nomadine Bees

Figures 11 — 14. Hypochrotaenia spp., male genitalia. 11-12, H. ( Micronomada ) texana, dorsal and ventral views; 13, H. ( H . ) mexicana,
dorsal view; 14, H. (Aphelonomada) cruralis, dorsal view. Scale line = 0.50 mm.

Contributions in Science, Number 376

Snelling: New World Nomadine Bees 11

Figures 15-18. Hypochrotaenia spp., male sternum seven. 15, H. (Micronomada) texana; 16, H. (Aphelonomada) cruralis ; 17-18, H. (H.)
mexicana, ventral and lateral views. Scale line = 0.25 mm.

compressed apicad; metatibia without tubercles or spiniform
setae along its length. ( 1 0) Depressed apical margin of second
and third gastric terga impunctate.

TYPE-SPECIES. Nomada cruralis Moure, 1960.

ETYMOLOGY. Gr., apheles (smooth or simple) + No-
mada, because of the smooth, shiny integument.

DISCUSSION. Moure (1960) correctly noted that Cres-
son’s specific epithet for this species was preoccupied in No-

mada and he proposed the new name N. cruralis; according
to the International Code of Zoological Nomenclature, a name
that is a primary junior homonym, as is the case here, is
permanently rejected (Art. 59a), presumably even when
transferred to another genus.

Although the only known species of this subgenus closely
resembles those of Hypochrotaenia s. str., Aphelonomada is
distinct. Most notable of the differences is the unusual length

12 Contributions in Science, Number 376

Snelling: New World Nomadine Bees

of the labrum, unique among the New World “ Nomada .”
Unique also within the genus Hypochrotaenia are the broad
impunctate apical margins of the gastric terga which include
the entire depressed apical portion of those segments. The
metatibia is covered, as is much of the body, with long, rather
silky, white hairs; the integument of the outer face is devoid
of tubercles and the coarse, spine-like setae typical of Hy-
pochrotaenia are absent.

The only known species occurs in Cuba. Nothing is known
of the biology of this bee.

INCLUDED NAMES

cruralis Moure, 1 960*
tibialis Cresson, 1865*

UNASSIGNED SPECIES OF HYPOCHROTAENIA

The following species-group names for Neotropical species
evidently all apply to species of Hypochrotaenia. But, until
the types can be studied, their subgeneric placement is un-
certain. Most, if not all, probably belong to Hypochrotaenia,
s. str. All are new combinations in Hypochrotaenia.

bicellularis (Ducke, 1 908)*
carcarensis (Cockerell, 1917)*
ecuadoria (Friese, 1916)*
flavescens (Friese, 1916)*
flaviceps (Cresson, 1865)*
fuliginosa (Friese, 1916)*
infrequens ( F. Smith, 1879)*
nitidiceps (Cockerell, 1931)*
rugicollis (Friese, 1916)*
sangninolenta (Friese, 1916)*
subscopifera (Ducke, 1 908)*
tenuicornis (Cockerell, 1949)*
tomentifera (Ducke, 1 907)*

Genus Centrias Robertson

Figures 19-24, 65

Centrias Robertson, 1903:174, 176. Type-species: Nomada
erigeronis Robertson, 1897; original designation and
monobasic.

Nomada subg. Nomadula Cockerell, 1 903d:6 1 1 . Type-species:
Nomada articulata F. Smith, 1854; original designation.
NEW SYNONYMY.

DIAGNOSIS. Metapleuron protuberant and marginate
above; paraocular ridge present in female; procoxa usually
spinose; anterior rim of pronotum subangulate to angulate
at side; male subgenital brush often complex; inner dorsal
lobe of male gonocoxite virtually absent; segment 3 of male
flagellum often spinose (Fig. 20).

DESCRIPTION. (A) Mandible without preapical tooth.
(B) Paraocular ridge present in female. (C) Minimum length
of first flagellar segment distinctly less than maximum length
of second; male flagellar segments 3-10 often conspicuously
spinose, tuberculate, or sharply ridged beneath. (D) Gena
about half as wide as eye in profile, genal margin sharply

angulate to cariniform. (E) Procoxa usually with distal spine
or tooth; outer carina absent. (F) Anterior rim of pronotum
subangulate to angulate at side; pronotal collar compressed,
more or less crested above. (G) Metapleuron protuberant and
marginate or lamellate above. (H) Metacoxa rounded above.
(I) Gastric terga usually sharply and closely punctate and
with narrow impunctate margins, but may be shallowly and
obscurely punctate. Male. (J) Hairs of subgenital brush erect,
long, curled and often in a complex pattern. (K) Inner dorsal
lobe of gonocoxite virtually absent, outer lobe short, stout;
median sinus broader than long. (L) Gonostylus without se-
tose basoventral lobe and without inner, basal seta-cluster.

DISCUSSION. The generic name Centrias was published
by Robertson (1903) in June; the name Nomadula was pub-
lished, as a subgenus of Nomada, in August of that same year
by Cockerell ( 1 903d).

Nomadula was described as a subgenus allied to Centrias,
the principal justification being that in Nomadula the two
sexes were dissimilar in appearance, rather than very similar
in color and markings as in Centrias. Some western species
are known in which the two sexes have similar color patterns
but are morphologically similar to other species in which this
is not true. There is no basis for continued separation of these
two names.

This appears to be a strictly New World genus limited to
the Nearctic Region. Centrias is probably more nearly related
to the genus Hypochrotaenia than to Nomada. The two gen-
era share a tendency toward spinose procoxae (rare and spo-
radic in Nomada), laterally angulate pronotal rim, sharply
punctate gaster and features of male terminalia. The peculiar
modification of the metapleuron is unique to Centrias, al-
though some South American species of Hypochrotaenia have
the metapleuron somewhat protuberant above, but the prom-
inence is usually not marginate. Presumably, this feature is
independently derived in the two groups. In addition to the
metapleural structure, Centrias consistently differs from Hy-
pochrotaenia in both sexes by the short first flagellar segment.
In addition, male Centrias commonly have a short, acute
spine on at least the third flagellar segment. Females consis-
tently possess a short paraocular ridge.

Some species of Centrias fly in spring months; at least one
of these is known to be cleptoparasitic in nests of Andrena.
Other species are active in summer months and are associ-
ated with halictids as hosts (Table 1). It is possible that there
are two subgenera involved, but the matter will require fur-
ther study; the names Centrias and Nomadula, however,
both apply to the summer group of species.

The names listed below are mostly new combinations in
Centrias.

INCLUDED NAMES

articulata (F. Smith, 1854)
aureopilosa (Swenk, 1913)
australis (Mitchell, 1962)
crotchii (Cresson, 1878)
dacotana (Cockerell, 1903d)
erigeronis (Robertson, 1897)

Contributions in Science, Number 376

Snelling: New World Nomadine Bees 13

23

24

Figures 19-24. Centrias articulata. 19, frontal view of female head (scale line = 1.00 mm); 20, basal segments of male antenna; 21, apex of
female metatibia; 22, male sternum seven (scale line = 0.25 mm); 24-25, male genitalia, dorsal and ventral views (scale line = 0.50 mm).

14 Contributions in Science, Number 376

Snelling: New World Nomadine Bees

erythrochroa (Cockerell, 1903a)
erythrospila (Cockerell, 1916c)
frankei (Cockerell, 1 929b)
frieseana (Cockerell, 1 904)
incerta (C resson, 1863)
martinella (Cockerell, 1 903d)
nigrior (Cockerell, 1903c)
pascoensis (Cockerell, 1903c)
rohweri (Cockerell, 1906a)
rubicunda (Olivier, 1811)
scita (Cresson, 1878)
scitiformis (Cockerell, 1 903d)
semiscita (Cockerell, 1 904)
seneciophila (Mitchell, 1962)
sophiarum (Cockerell, 1903a)
torrida (F. Smith, 1854)

Genus Nomada Scopoli

DIAGNOSIS. Female with paraocular ridge (except sub-
genus Phelonomada)', minimum length of first flagellar seg-
ment usually less than maximum length of second; procoxa
usually without distal spine or tooth; anterior rim of prono-
tum evenly curved laterally; metapleuron neither protuber-
ant nor marginate above. Male. Hairs of subgenital brush
usually short, inclined basad; gonostylus usually with setose
basoventral process.

DESCRIPTION. (A) Mandible with or without preapical
tooth. (B) Paraocular ridge present in female (except subg.
Phelonomada). (C) Minimum length of first flagellar segment
usually less than maximum length of second; male flagellar
segments 3-10 often ridged or tuberculate beneath. (D) Gena
usually at least half as wide as eye in profile, margin usually
angulate to cariniform. (E) Procoxa usually simple at apex;
outer carina absent. (F) Anterior rim of pronotum not an-
gulate at side; pronotal collar compressed, especially in mid-
dle, often crested above. (G) Metapleuron simple above. (H)
Metacoxa rounded above. (I) Gastric terga usually weakly
punctate and with broad impunctate margins, but may be
closely punctate with narrow margins. Male. (J) Hairs of
subgenital brush usually short, straight, and inclined toward
base. (K) Inner dorsal lobe of gonocoxite usually short and
broad, outer lobe usually broad; median sinus usually broad-
er than long. (L) Gonostylus usually with setose basoventral
lobe and often with a cluster of short, flattened setae on inner
basal margin.

DISCUSSION. With the removal of Nomadula and Hy-
pochrotaenia as separate genera, Nomada assumes a uni-
formity previously lacking. It is still a very large and complex
genus and is primarily northern in distribution. In the New
World, no true Nomada are known from South America.
The Old World forms have not been subjected to a large scale
study of the groups and there are no currently named sub-
genera recognized there, except the Holarctic Nomadita.

However, many Old World species can be assigned to ex-
isting subgenera described from the Nearctic fauna.

KEY TO SUBGENERA OF NOMADA

la. Procoxa usually without spine; if spine present, gastric
terga weakly punctate and with broad apical impunctate
bands or gena is sharply marginate; inner dorsal lobe of
male gonocoxite short and broad, at most as long as

broad; paraocular ridge of female present 2

b. Procoxa with short distal spine; gastric terga sharply and
closely punctate, with narrow impunctate margins; gena
narrow, posterior margin rounded; inner dorsal lobe of
male gonocoxite longer than broad, acute; paraocular

ridge of female absent Phelonomada

2a. Minimum length of first flagellar segment equal to, or
longer than, maximum length of second; gastric terga
closely punctate and with narrow impunctate margins;
apex of pygidium broad, without median notch; man-
dible without preapical tooth 3

b. Minimum length of first flagellar segment less than max-
imum length of second; if equal, tergal punctures weak,
impunctate apical margins broad; apex of pygidium sub-
acute to narrowly rounded, margins straight and strongly
convergent distad, with distinct notch, at least in males
(rarely entire); preapical mandibular tooth sometimes

present Nomada s. str.

3a. Propodeum not swollen or elevated mesad of spiracle;
scape of male stout, often subglobose; metatibia of fe-
male with distal setae in a distinct row; erect hairs short

and sparse on head and thorax 4

b. Propodeum swollen or elevated mesad of spiracle; scape
of male stout-cylindrical; metatibia of female with distal
setae in a dense cluster (except one species); erect hairs
of head and thorax usually numerous, long, somewhat

shaggy 5

4a. Posterior margin of gena rounded or weakly angulate;
aedeagus of male with ventral hook; metabasitarsus usu-
ally somewhat swollen, broadest near middle

Pachynomada

b. Posterior margin of gena subcarinate; aedeagus of male
without ventral hook; metabasitarsus slender, flattened

on external face, broadest near base Nomadita

5a. Labrum with extensive smooth and impunctate areas,
at least on basal half; pro- and mesofemur of male greatly
flattened and expanded dorso-ventrally; apex of female
metatibia with a single, irregular row of 8-1 1 ferruginous

setae Larninomada

b. Labrum contiguously punctate; pro- and mesofemur of
male neither flattened nor dorso-ventrally expanded; apex
of female metatibia with a dense cluster of 1 5 or more
setae Holonomada

Subgenus Nomada Scopoli

Figures 25-31, 64, 66

Nomada Scopoli, 1770:44. Type-species: Apis ruficornis
Linne, 1758; designated by Curtis, 1832.

Nomada subg. Heminomada Cockerell and Atkins, 1902:42.

Contributions in Science, Number 376

Snelling: New World Nomadine Bees 15

Figures 25-31. Nomada (TV.) ruficornis. 25, frontal view of female head (scale line = 1.00 mm); 26, basal segments of male antenna; 27,
apex of female metatibia; 28, male sternum seven (scale line = 0.50 mm); 29, apex of female pygidium; 30-31, male genitalia, dorsal and
ventral views (scale line = 1.00 mm).

16 Contributions in Science, Number 376

Snelling: New World Nomadine Bees

Type-species: Nomada obliterata Cresson, 1863; original
designation and monobasic. NEW SYNONYMY.
Gnathias Robertson, 1903: 173. Type-species: Nomada bel/a
Cresson, 1863; original designation. NEW SYNONYMY.
Phor Robertson, 1903: 173. Type-species: Phor Integra Rob-
ertson, 1903; original designation and monobasic. NEW
SYNONYMY.

Xanthidium Robertson, 1903:174 (not Ehrenberg, 1833).
Type-species: Nomada luteola Olivier, 1811; original des-
ignation. NEW SYNONYMY.

DIAGNOSIS. Minimum length of first flagellar segment
less than, rarely equal to, maximum length of second flagellar
segment; preapical mandibular tooth sometimes present;
procoxa usually not spinose; gastric punctures usually more
or less obscure and well separated, apical margin of tergum
2 usually broadly impunctate; apex of pygidium subacute to
narrowly rounded, usually with deep median notch in male.

DESCRIPTION. (1) Mandible with or without preapical
tooth. (2) First flagellar segment no longer, and usually much
shorter, than second; male often with conspicuous ridges or
tubercles on segments 3-10. (3) Posterior margin of gena
angulate to subcarinate. (4) Procoxa rarely with apical spine
or tooth. (5) Propodeum not swollen mesad of spiracle. (6)
Tergum 2 more or less sharply punctate on disc, less distinctly
so on apical depression, impunctate apical margin usually
broad. Male. (7) Apex of pygidium usually distinctly notched.
(8) Process of sternum 8 usually slender, more or less parallel-
sided and with deflected apex. (9) Inner dorsal lobe of gono-
coxite almost always short and broad, with invagination be-
tween inner and outer lobes broad and deep; median sinus
broader than long. (10) Gonostylus with setose basoventral
process and often with inner, basal cluster of very short,
flattened setae. (11) Aedeagus without ventral hook. Female.
(12) Paraocular ridge always present. (13) Metatibia usually
with a row of 2-8 dark, flattened setae at apex. (14) Apex of
pygidium acute to subacute, usually with narrow median
notch.

DISCUSSION. As pointed out above, I do not believe
that a single characteristic should justify a subgenus. Such is
the case with Gnathias , here united with the nominate sub-
genus. There are no other features known which would sup-
port recognition of this heterogenous assemblage.

The subgenus Heminomada was established for a species
with but two submarginal cells in the forewing. Some spec-
imens of TV. obliterata, the type-species, may have three sub-
marginal cells on one wing, two on the other or only partial
loss of the transverse cubital vein in one or both wings. The
equivalence of Heminomada with Xanthidium, a junior
homonym, has long been recognized.

Heminomada has been utilized for those species in which
the head and thorax are more or less richly yellow-marked
and one or more gastric segments have a complete transverse
yellow band. The weakness of this division was recognized
by Mitchell (1962), who observed that “. . . if a species is
known in but one sex it may be difficult to place it with
confidence in one or the other of these groups.” Among the
species in western North America, some species would have

the two sexes in different subgenera. I am certain that Hemi-
nomada may be safely put to rest.

The subgenus Phor was established for a single species with
a long first flagellar segment and a short procoxal spine. The
few other species tentatively placed in Phor seem to have
little relationship to that species or to each other. Both char-
acter states occur sporadically throughout Nomada s. str. It
is hardly surprising that these would combine in a few species,
but a formal name hardly seems necessary.

Nomada is a very large subgenus, including the vast ma-
jority of the North American species. Future study may result
in further splitting, but not along previous lines.

This subgenus is Holarctic and is largely northern and
montane in the Western Hemisphere. Those few species whose
hosts are known are mostly cleptoparasites in nests of An-
drena (Table 2). In addition to the hosts listed in Table 2,
Barrows (1978) has implicated N. luteola, N. maculata, and
N. illinoensis as possible cleptoparasites of Andrena ( Ptilan -
drena) erigeniae Robertson.

INCLUDED NAMES

accepta Cresson, 1878
agynia Cockerell, 1905c
a/bofasciata F. Smith, 1879
aldrichi Cockerell, 1910a
alpha Cockerell, 1905c
amoena Cresson, 1863
angelarum Cockerell, 1903c
angulata Swenk, 1913
annulata F. Smith, 1854
aprilina Swenk, 1913
arizonica Cockerell, 1911b
armatella Cockerell, 1903d
ashmeadi Cockerell, 1903c
astori Cockerell, 1903a
atrofrontata Cockerell, 1903c
augustiana Mitchell, 1962
autumnalis Mitchell, 1962
avalonica Cockerell, 1938
azaleae Mitchell, 1962
banksi Cockerell, 1907b
bella Cresson, 1863
bethunei Cockerell, 1903d
beulahensis Cockerell, 1903a
bicincta Howard, 1902!
bicrista Swenk, 1913
bifurcata Cockerell, 1903c
bilobata Swenk, 1913
binotatus Robertson, 1903
bisetosa Swenk, 1913
bishoppi Cockerell, 1911a
bisignata Say, 1824
californiae Cockerell, 1903c
calloxantha Cockerell, 1921
callura Cockerell, 1911a
capillata Mitchell, 1962
capitalis Mitchell, 1962

Contributions in Science, Number 376

Snelling: New World Nomadine Bees 17

carinicauda Cockerell, 1921
carolinae Cockerell, 1 903d
ceanothi Cockerell, 1907b
citrina Cresson, 1878
civilis Cresson, 1878
clarescens Cockerell, 1 92 1
clarkii Cockerell, 1903a
collinsiana Cockerell, 1905c
coloradella Cockerell, 1905c
coloradensis Cockerell, 1903d
colorata Mitchell, 1962
composita Mitchell, 1962
concinnulaC ockerell, 1921
contractula Cockerell, 1 903d
coquilletti Cockerell, 1903c
cordleyi Cockerell, 1903b
cornelliana Cockerell, 1908c
corvallisensis Cockerell, 1 903a
crawfordi Cockerell, 1905c
cressonii Robertson, 1893
crudelis Cresson, 1878
cuneatus Robertson, 1903
custeriana Cockerell, 1911b
cymbalariae Cockerell, 1906a
davidsoni Cockerell, 1 903c
debilis Timberlake, 1954
decemnotatus Robertson, 1903
decempunctata Cockerell, 1903c
decepta Mitchell, 1962
dentariae Robertson, 1903
denticulata Robertson, 1902a
depressa Cresson, 1863
depressicauda Cockerell, 1 908a
detrita Mitchell, 1962
dialpha Cockerell, 1921
dilucida Cresson, 1878
dreisbachi Mitchell, 1962
eastonensis Cockerell, 1903d
ednae Cockerell, 1907a
electa Cresson, 1863
electella Cockerell, 1 903d
e/egantula Cockerell, 1903c
elrodi Cockerell, 1903d
erythraea Dalla Torre, 1896!
festiva Cresson, 1863
flagellaris Cockerell, 1 903d
jlammigera Cockerell, 1 906c
flavomarginata Swenk, 1913
florilega Lovell & Cockerell, 1905
fontis Cockerell, 1910b
fowleri Cockerell, 1903a
fragariae Mitchell, 1962
fragilis Cresson, 1878
fuscicincta Swenk, 1915
gibbosa Viereck, 1905
gdletti Cockerell, 1905c
gracilis Cresson, 1863
graenicheri Cockerell, 1905a

grayi Cockerell, 1 903a
heterosticta Cockerell, 1921
honorata Cockerell, 1922a
hoodiana Cockerell, 1903d
hydrophylli Swenk, 1915
idahoensis Swenk, 1913
illinoiensis Robertson, 1 900
illinoensis Cockerell, 1911a!
illinoensis Cockerell, 1921!
imbricata F. Smith, 1854
indusata Mitchell, 1962
inepta Mitchell, 1962
infantula Cockerell, 1907b
infrarubens Cockerell, 1905c
integerrima Dalla Torre, 1896!

“ integra Robertson, 1893”

“ intermpta Fowler, 1899”

interruptella Fowler, 1902!

itamera Cockerell, 1910c

jocularis C resson, 1879

kincaidiana Cockerell, 1903d

kingstonensis Mitchell, 1962

klamathensis C. Fox, 1926

krombeini M. Schwarz, 1 966a!

lachrymosa Cockerell, 1921

laratniensis Swenk, 1913

latifrons Cockerell, 1 903c

lehighensis Cockerell, 1 903d

lepida Cresson, 1863

leucozona Rodeck, 1931

lewisi Cockerell, 1903a

libata Cresson, 1863

limbata Dalla Torre, 1896!

louisianae Cockerell, 1903d

luteola Olivier, 1811

luteoloides Robertson, 1895

luteopicta Cockerell, 1905c

maculata Cresson, 1863

maculiventer Swenk, 1915

malonella Cockerell, 1910c

malonina Cockerell, 1910c

marginel/a Cockerell, 1903c

mckenziei Timberlake & Cockerell, 1937

media Mitchell, 1962

mediana Swenk, 1913

melanosoma Cockerell, 1916c

melliventris Cresson, 1878

mendica Mitchell, 1962

mera Cockerell, 1908c

mesosticta Cockerell, 1939

mimula Cockerell, 1908b

miniata F. Smith, 1854

minima Mitchell, 1962

“ minuta Swenk, 1913”

modocorum Cockerell, 1903b

morrisoni Cresson, 1878

munda Cresson, 1878

nigrociliata Swenk, 1913

18 Contributions in Science, Number 376

Snelling: New World Nomadine Bees

nigrofasciata Swenk, 1913

“ obliqua Fowler, 1899”

obliquella Fowler, 1902!

obliterata Cresson, 1863

“ obscura Fowler, 1899”

obscurella Fowler, 1902!

obstusata Swenk, 1915

obtusata Swenk, in Muesebeck et al., 1951!

ochlerata Mitchell, 1962

ochrohirta Swenk, 1913

octomaculata Robertson, 1903

octonotatus Robertson, 1903

odontocera Cockerell, 1916c

opacella Timberlake, 1954

opposita Cresson, 1878

orba Mitchell, 1962

orcusella Cockerell, 1910c

oregonica Cockerell, 1903a

ornithica Cockerell, 1 906d

orophila Cockerell, 1921

osborni Cockerell, 1911b

ovata Robertson, 1903

pacata Cresson, 1878

packardie/la Cockerell, 1906b

pallidella Cockerell, 1905c

pailidipicta Swenk, 1913

parallela Swenk, 1913

paralpha Cockerell, 1921

par at a Cresson, 1878

parva Robertson, 1900

perbella Viereck, 1905

perivincta Cockerell, 1905c

perplexa Cresson, 1863

perplexans Cockerell, 1910c

physura Cockerell, 1903d

placitensis Cockerell, 1903b

plenus Robertson, 1903

propinqua Swenk, 1913

proxima Cresson, 1863

pseudops Cockerell, 1905a

pulchella F. Smith, 1854

pulsatillae Cockerell, 1 906c

pygmaea Cresson, 1863

pyrrha Cockerell, 1916c

quadrimaculatus Robertson, 1903

quadrisignatus Robertson, 1903

rhodalis Cockerell, 1 903d

rhodomelas Cockerell, 1903d

rhodosoma Cockerell, 1903c

rhodosomella Cockerell, 1903c

rhodoxantha Cockerell, 1905c

rivalis Cresson, 1878

robertsonella Cockerell, 1 903d

rubi Swenk, 1915

rubrella Cockerell, 1905c

“ rubra Provancher, 1888”

rubrica Provancher, 1896!

rufula Cockerell, 1903d

ruidosensis Cockerell, 1903d
salicola Swenk, 1913
salicis Robertson, 1 900
sanctaecrucis Cockerell, 1903c
sayi Robertson, 1893
schwarzi Cockerell, 1903d
sedae Cockerell, 1919a
sedi Cockerell, 1 920!
semirufula Cockerell, 1905c
semirugosa Cockerell, 1929a
septentrionalis Swenk, 1913
sexmaculatus Robertson, 1903
sexnotatus Robertson, 1903
siccorum Cockerell, 1919a
simplex Robertson, 1902b
siouxensis Swenk, 1913
skinneri Cockerell, 1 908a
sobrina Mitchell, 1962
sphaerogaster Cockerell, 1903d
spokanensis Cockerell, 1910c
stevensi Swenk, 1913
subaccepta Cockerell, 1907c
subangusta Cockerell, 1903c
subgracilis Cockerell, 1903c
subnigrocincta Swenk, 1915
subpacata Swenk, 1913
subrubi Swenk, 1915
subrutila Lovell & Cockerell, 1905
subsimilis Cockerell, 1903c
subvicinalis Cockerell, 1903c
suda Cresson, 1878
sulphurata F. Smith, 1854
swenki M. Schwarz, 1966a!
taraxace/la Cockerell, 1903d
tintinnabulum Cockerell, 1903c
townesi Mitchell, 1962
trevoriana Cockerell, 1905b
tricurta Swenk, 1915
truttorum Cockerell, 1 909b
tyrrellensis Mitchell, 1962
ulsterensis Mitchell, 1962
ultima Cockerell, 1903a
ultimella Cockerell, 1903c
undulaticornis Cockerell, 1 906c
utensis Swenk, 1913
unicolor Robertson, 1 903
valida F. Smith, 1854
vallesina Cockerell, 1906b
velutina Swenk, 1913
vernonensis Cockerell, 1916a
vexator Cockerell, 1 909b
viburni Robertson, 1897
vicina Cresson, 1883
vicinalis Cresson, 1878
volatilis F. Smith, 1879
vulpis Cockerell, 1921
washingtoni Cockerell, 1903d
wootonel/a Cockerell, 1 909b

Contributions in Science, Number 376

Snelling: New World Nomadine Bees 19

wyomingensis Swenk, 1913
xantholepis Cockerell, 1911b
xanthoparius Robertson, 1928
xanthophilus Cockerell, 1 900
xanthura Cockerell, 1908c
ziziae Swenk, 1915

NOTE. “ ” denotes preoccupied name. ! denotes variant
spelling, including emendations and new names.

Subgenus Holonomada Robertson

Figures 32-37

Holonomada Robertson, 1903:174, 177. Type-species: No-

mada superba Cresson, 1863; original designation.

DIAGNOSIS. Minimum length of first flagellar segment
equal to, or exceeding, maximum length of second; propo-
deum convex and prominent mesad of spiracle; distal setae
of female metatibia slender, ferruginous, forming a dense
cluster; apex of pygidium broadly rounded; medium sized to
moderately large species.

DESCRIPTION. (1) Mandible without preapical tooth.
(2) Minimum length of first flagellar segment as long as, or
longer than, maximum length of second. (3) Genal margin
abrupt, but not cariniform. (4) Procoxa without distal spine.
(5) Propodeum weakly to strongly swollen or protuberant
mesad of spiracle. (6) Tergum 2 densely and finely punctate,
with very narrow impunctate margin; punctures often ob-
scured by transverse lineation of interspaces. Male. (7) Apex
of pygidium without median notch. (8) Process of sternum
8 elongate, narrow, broadened distad. (9) Inner dorsal lobe
of gonocoxite about as long as broad, invagination between
inner and outer lobes moderate to deep; median sinus as long
as, or a little longer than, broad. (10) Gonostylus with setose
basoventral lobe and with seta-cluster or inner basal margin.
(11) Aedeagus without ventral hook. Female. ( 1 2) Paraocular
ridge present. (13) Distal setae of metatibia long, slender,
ferruginous, forming dense cluster. (14) Apex of pygidium
broadly rounded, without median notch.

DISCUSSION. Holonomada is a small group of primarily
western, relatively large and robust species. Previous authors
have stressed the propodeal structure as diagnostic of Ho-
lonomada. At its greatest development, in such species as N.
superba, it is certainly obvious. But, in some western species,
such as N. edwardsii and N. hemphilli, the swelling above
the spiracle is very subtle. Some species of Nomada s. str.
approach this condition, but may be separated by the shorter
first flagellar segment, the notched pygidium of the males,
and the lack of a definite seta-cluster on the metatibia of the
females.

Holonomada is a weak subgenus and it may be necessary
to merge it with the nominate subgenus when the fauna of
western North America becomes better known. Some west-
ern species, now assigned to subgenus Nomada, are as large
as Holonomada, equally hairy and the notch of the male
pygidium is poorly defined. The apparent females, however,
do not have the metatibial seta-cluster characteristic of Ho-
lonomada.

The only known host for any species of Holonomada is a
species of Andrena (Table 2). However, it seems possible
that the anthophorid, Synhalonia amsinckiae Timberlake,
may be a host for N. hemphilli. This is a presumption based
on the coincidence of flight activity and the fact that in the
area where I observed both species for more than ten years,
no Andrena of suitably large size was encountered at the
appropriate time of year. It is possibly relevant that a Jap-
anese species of Holonomada, N. daimio Matsumura, is
known to be a parasitoid in nests of Tetralonia, a eucerine
genus closely related to Synhalonia (H. Nagase, pers. comm.).

Holonomada was revised by Evans (1972). Eleven species
were recognized; three of these were each divided into two
subspecies. There are additional species yet to be described.

INCLUDED NAMES

affabilis Cresson, 1878
australior Cockerell, 1903c
dallasensis Cockerell, 1911a
edwardsii Cresson, 1878
excellens Cockerell, 1903c
excurrens Cockerell, 1903c
grandis Cresson, 1875
hemphilli Cockerell, 1903c
henningeri Evans, 1972
hurdi Evans, 1972
intercepta F. Smith, 1879
linsleyi Evans, 1972
magnifica Cockerell, 1 903d
ma/vastri Swenk, 1913
nebrascensis Swenk, 1913
parkeri Evans, 1972
pecosensis Cockerell, 1903d
rhodotricha Cockerell, 1903c
superba Cresson, 1863
vinnula Cresson, 1879

Subgenus Laminomada Rodeck

Figures 38-44

Nomada subg. Laminomada Rodeck, 1947:266. Type-
species: Nomada hesperia Cockerell, 1903c; original des-
ignation.

DIAGNOSIS. Minimum length of first flagellar segment
greater than maximum length of second; procoxa with con-
ical distal process; pro- and mesofemur of male expanded;
labrum shiny on basal half between distinctly separated punc-
tures; female metatibia with a single irregular row of 8-1 1
coarse setae at apex.

DESCRIPTION. (1) Mandible without preapical tooth.
(2) Minimum length of first flagellar segment greater than
maximum length of second; segments 3-10 of male without
conspicuous ridges or tubercles. (3) Posterior margin of gena
subangulate but not cariniform. (4) Procoxa with short, con-
ical process at apex. (5) Propodeum weakly swollen mesad
of spiracle. (6) Tergum 2 densely and finely punctate, with
very narrow impunctate margin. Male. (7) Apex of pygidium

20 Contributions in Science, Number 376

Snelling: New World Nomadine Bees

o

Figures 32-37. Nomada ( Holonomada ) superba. 32, frontal view of female head (scale line = 1 .00 mm); 33, basal segments of male antenna;
34, apex of female metatibia; 35, male sternum seven (scale line = 0.50 mm); 36-37, male genitalia, dorsal and ventral (scale line = 0.50
mm).

Contributions in Science, Number 376

Snelling: New World Nomadine Bees 21

Figures 38—14. Nomada (Laminomada) hesperia. 38, frontal view of female head (scale line = 1 .00 mm); 39, basal segments of male antenna;
40, apex of female metatibia; 41, apex of female pygidium; 42, male sternum seven (scale line = 0.50 mm); 43—44, male genitalia, dorsal and
ventral (scale line = 0.50 mm).

22 Contributions in Science, Number 376

Snelling: New World Nomadine Bees

without median notch. (8) Process of sternum 8 short, slen-
der, broadened distad, apex deflected. (9) Inner dorsal lobe
of gonocoxite broad, invagination between inner and outer
lobes deep; median sinus longer than broad. (10) Gonostylus
with setose basoventral lobe and with seta-cluster on inner
basal margin. (11) Aedeagus without ventral hook. Female.
(12) Paraocular ridge present. (13) Metatibia with a single
irregular row of 8-11 long, slender ferruginous setae. (14)
Apex of pygidium broadly rounded, without median notch.

DISCUSSION. This subgenus is very close to Holono-
mada and perhaps should be merged with it although there
are conspicuous differences, especially in the structure of the
pro- and mesofemur and the metabasitarsus of the male. The
female is much more like that of such Holonomada species
as N. edwardsii and the differences are those which normally
differentiate species.

The single species of Laminomada ranges from south-
eastern Washington to southern California. The flight period
is from late March to early June. The host is unknown.

INCLUDED NAMES

falconis Rodeck, 1947
flavopicta Swenk, 1913
hesperia Cockerell, 1903c

Subgenus Pachynomada Rodeck

Figures 45-5 1

Nomada subg. Pachynomada Rodeck, 1945:180. Type-

species: Nomada vincta Say, 1837; original designation.

DIAGNOSIS. Minimum length of first flagellar segment
equal to, or exceeding, maximum length of second; genal
margin rounded to weakly angulate; female metabasitarsus
swollen, broadest near middle; female pygidium broadly
U-shaped; male aedeagus with ventral hook.

DESCRIPTION. (1) Mandible without preapical tooth.
(2) Minimum length of first flagellar segment at least equal
to maximum length of second; male without conspicuous
ridges or tubercles on segments 3-10. (3) Posterior margin
of gena rounded to weakly angulate. (4) Procoxa without
distal spine. (5) Propodeum not swollen mesad of spiracle.
(6) Tergum 2 densely and finely punctate, with narrow im-
punctate apical margin (except N. victrix). Male. (7) Apex of
pygidium broad, rounded, sometimes with weak median
notch. (8) Process of sternum 8 slender, apex weakly deflect-
ed. (9) Inner dorsal lobe of gonocoxite very broad and weak;
invagination between inner and outer lobes broad, weak;
median sinus longer than broad, margins straight or nearly
so. (10) Gonostylus without setose basoventral process or
inner basal seta-cluster. (11) Aedeagus with ventral hook.
Female. (12) Paraocular ridge present. (13) Metatibia with
short to very short, well separated apical spines. ( 1 4) Apex
of pygidium broadly U-shaped, without median notch.

DISCUSSION. To this subgenus are assigned a few un-
common summer and fall species. The species are mostly
western. Males are further distinguished from those of other
subgenera by the distinctly swollen antennal scape. The

hooked aedeagus is apparently unique. Females are less easily
characterized. The metabasitarsus is somewhat swollen and
widest near the middle, the pygidium is usually broadly
U-shaped and the posterior margins of the gena are rounded
or, less commonly, weakly angulate.

This, and the two remaining subgenera, Nomadita and
Phelonomada, share an unusual feature within the genus. In
all three the face, adjacent to the inner eye margins, is more
or less distinctly convex. In the subgenera Nomada, Ptolo-
nomada, and Laminomada, the face is concave between the
inner eye margins.

There are other parallels between these three subgenera:
they are summer or fall flying, are largely western in distri-
bution, have reduced erect pubescence on head and thorax,
and all possess a relatively long first flagellar segment.

Thus, Pachynomada, Nomadita, and Phelonomada rather
closely resemble many species of Hypochrotaenia ( Micro -
nomada). These similarities prompted Moalif (1979) to
opine that Pachynomada and Nomadita (as Callinomada )
were probably more closely allied to Micronomada than to
other groups of Nomada. However, the pronotal and genitalic
characteristics of these three subgenera are unlike those of
Micronomada. Micronomada females lack the paraocular
ridge (true, also, of Phelonomada). These parallels are inter-
esting but do seem to be examples of character convergence
rather than of phyletic propinquity.

Pachynomada has been revised by Moalif (1979). Eleven
species were recognized, four of which are yet undescribed.
Two species (N. morrisoni and N. adducta), formerly placed
in Pachynomada, were excluded; N. morrisoni is here as-
signed to subgenus Nomada and N. adducta is tentatively
placed in Nomadita.

An undescribed species has been reported to be clepto-
parasitic in nests of Andrena ( Callandrena ) helianthi by Par-
ker and Bohart (1983). The same species of Pachynomada
has also been seen leaving a nest of A. (C.) haynesi (Parker
and Griswold, 1983).

INCLUDED NAMES

asteris Swenk, 1913
aztecorum Cockerell, 1903a
besseyi Swenk, 1913
heterochroa Cockerell, 1921
pratensis Cockerell, 1919b
suffosa Cockerell, 1922b
victrix Cockerell, 1911a
vincta Say, 1837
vitticollis Cresson, 1878
zebrata Cresson, 1878

Subgenus Nomadita Mocsary

Figures 52-57

Nomadita Mocsary, 1894:37. Type-species: Nomadita mon-

tana Mocsary, 1894; monobasic.

Nomada subg. Callinomada Rodeck, 1 945: 181. Type-species:

Contributions in Science, Number 376

Snelling: New World Nomadine Bees 23

Nomada antonita Cockerell, 1909a; original designation.

NEW SYNONYMY.

DIAGNOSIS. Minimum length of first flagellar segment
equal to, or exceeding, maximum length of second; propo-
deum not swollen mesad of spiracle; genal margin subcari-
nate; male sternum 8 rather broad, margins tapering distally;
female metabasitarsus broadest near base; propodeum bare
or nearly so, with little or no erect hair.

DESCRIPTION. (1) Mandible without preapical tooth.
(2) Minimum length of first flagellar segment equal to, or
greater than, maximum length of second; male without con-
spicuous ridges or tubercles on segments 3-10. (3) Posterior
margin of gena usually subcarinate. (4) Procoxa simple or
with distal process. (5) Propodeum not swollen mesad of
spiracle. (6) Tergum 2 finely and densely punctate, apical
impunctate margin very narrow. Male. (7) Apex of pygidium
rounded, entire or with broad, shallow emargination. (8) Pro-
cess of sternum 8 broad at base, margins convergent distad.
(9) Inner dorsal lobe of gonocoxite poorly defined, subacute;
invagination between inner and outer lobes broad, shallow;
median sinus longer than broad. (10) Gonostylus without
setose basoventral lobe or inner, basal cluster of short, flat-
tened setae. (11) Aedeagus without ventral hook. Female.
(12) Paraocular ridge present. (13) Metatibia with 4-6 short,
acute, flattened spines at apex. ( 1 4) Apex of pygidium broadly
rounded to subacute.

DISCUSSION. The North American components of No-
madita have in the past been placed in Callinomada. Com-
parison of these with such European species as N. montana
(Mocsary), N. roberjeotiana Panzer, and N. rufipes Fabricius,
convinces me that Callinomada is not essentially different
from Nomadita. The pygidium of the females of the North
American species is more broadly rounded than in the Eu-
ropean species, but the two groups are otherwise sufficiently
similar that their separation seems pointless.

Although externally very similar to species of Pachyno-
mada, those of Nomadita differ sharply in characters of the
male genitalia and associated structures. Both sexes differ
from Pachynomada in smaller size, slender metabasitarsi,
and the sharply angulate to subcarinate posterior genal mar-
gin. The broad, distally tapering sternum 8 of the male is
apparently unique in Nomada, but similar to that of Hy-

pochrotaenia. All other features place Nomadita with No-
mada rather than Hypochrotaenia.

The few species of Nomadita appear to be uncommon.
They are summer and autumnal fliers and mostly western.
The hosts are unknown but may be an autumnal group of
Andrena, possibly of the subgenus Cnemidandrena.

The North American species were revised (as Callino-
mada) by Rodeck ( 1 949). One species was added by Mitchell
(1962).

INCLUDED NAMES

? adducta Cresson, 1878
antonita Cockerell, 1909a
aquilarum Cockerell, 1903a
cockerelli Graenicher, 1911
dacotensis Swenk, 1913
mutans Cockerell, 1910c
omahaensis Swenk, 1915
placida Cresson, 1863
rodecki Mitchell, 1962
snowii Cresson, 1878
verecunda Cresson, 1879

Phelonomada new subgenus

Figures 58-63

DIAGNOSIS. Minimum length of first flagellar segment
equal to maximum length of second; procoxa with short distal
spine; gastric terga sharply, closely punctate and with narrow
impunctate apical margin; female without paraocular ridge;
inner dorsal lobe of male gonocoxite elongate, triangular.

DESCRIPTION. (1) Mandible without preapical tooth.
(2) Minimum length of first flagellar segment equal to max-
imum length of second; segments 3-10 of male without con-
spicuous ridges or tubercles. (3) Gena narrow, posterior mar-
gin rounded. (4) Procoxa with short distal spine. (5)
Propodeum not swollen mesad of spiracle. (6) Tergum 2
sharply, closely punctate on disc and apical depression, im-
punctate apical margin narrow. Male. (7) Apex of pygidium
entire or with weak median notch. (8) Process of sternum 8
slender, more or less parallel-sided and with deflected apex.
(9) Inner dorsal lobe of gonocoxite elongate, apex acute; in-
vagination between inner and outer lobes broad, deep; me-

Figures 45-51. Nomada ( Pachynomada ) vincta. 45, frontal view of female head (scale line = 1.00 mm); 46, basal segments of male antenna;
47, apex of female metatibia; 48, male sternum seven (scale line = 0.25 mm); 49-51, male genitalia, lateral, dorsal, and ventral (scale line =
0.50 mm).

Figures 52-57. Nomada (Nomadita) montana. 57, frontal view of female head (scale line = 1 .00 mm); 53, basal segments of male antenna;
54, apex of female metatibia; 55, male sternum seven (scale line = 0.25 mm); 56-57, male genitalia, dorsal and ventral (scale line = 0.50
mm).

Figures 58-63. Nomada (Phelonomada) belfragei. 58, frontal view of female head (scale line = 1 .00 mm); 59, basal segments of male antenna;
60, apex of female metatibia; 61, male sternum seven (scale line = 0.25 mm); 62-63, male genitalia, dorsal and ventral (scale line = 0.50
mm).

24 Contributions in Science, Number 376

Snelling: New World Nomadine Bees

Contributions in Science, Number 376

Snelling: New World Nomadine Bees 25

26 Contributions in Science, Number 376

Snelling: New World Nomadine Bees

Contributions in Science, Number 376

Smelling: New World Nomadime Bees 27

Figures 64-66. Thoracic structures of: 64, Nomada (N.) sp., pronotum; 65, Centrias articulata, metapleuron; 66, Nomada (N.) sp., meta-
pleuron.

dian sinus broader than long. (10) Gonostylus with weak
setose basoventral lobe and with inner, basal cluster of very
short, flattened setae. (11) Aedeagus without ventral hook.
Female. (12) Paraocular ridge absent. (13) Metatibia with a
row of 4-5 slender, pale, well-separated setae at apex. (14)
Apex of pygidium narrowly rounded, entire.

TYPE SPECIES. Nomada belfragei Cresson, 1878.

ETYMOLOGY. Gr., phelos (false or deceitful) + Nomada.

DISCUSSION. Superficially, the forms placed here very
closely resemble species of Hypochrotaenia, subg. Microno-
mada, with which they have usually been associated. The
male terminalia and the structure of the pronotum, however,
ally Phe/onomada with Nomada. This is the only group of
Nomada in which the females lack the paraocular ridge.

The few species of Phelonomada fly in late spring (April
and May) and range from Kansas to central Texas. Hosts are
unknown, but probably will be found to be halictids, either
Agapostemon or Nomia.

INCLUDED NAMES

belfragei Cresson, 1878
engelmanniae Cockerell, 1911a
wheeleri Cockerell, 1 903d
xanthogaster Cockerell, 1911a

ACKNOWLEDGMENTS

This phase of the Nomada revision is based on the exami-
nation of approximately 14,000 specimens which must rep-
resent the majority of the Nearctic species. Some Neotropical
and Old World species were also studied. Specimens were
loaned by: P.H. Amaud, Jr., California Academy of Sciences;
G.E. Bohart, Utah State University; G.C. Eickwort, Cornell
University; M. Favreau, American Museum of Natural His-
tory; S. Frommer and the late P.H. Timberlake, University
of California, Riverside; the late P.D. Hurd, Jr., National
Museum of Natural History; W.E. LaBerge, Illinois Natural

28 Contributions in Science, Number 376

Snelling: New World Nomadine Bees

History Survey; U.N. Lanham, University of Colorado; C.D.
Michener, University of Kansas; A.R. Moldenke, Oregon
State University; J.A. Powell and E.G. Linsley, University
of California, Berkeley; B. Ratcliff, University of Nebraska;
R.O. Schuster, University of California, Davis; M. Was-
bauer, California Department of Agriculture; F.G. Werner,
University of Arizona. I am also indebted to Dr. M. Schwarz
for sending material of critical Palearctic species.

I am pleased to acknowledge important critical comments
by colleagues on an earlier draft of this paper, especially the
late P.D. Hurd, Jr., E.G. Linsley, C.D. Michener, and J.G.
Rozen, Jr.

LITERATURE CITED

Barrows, E.M. 1978. Male behavior in Andrena erigeronis
(Hymenoptera: Andrenidae) with comparative notes.
Kansas Entomological Society, Journal, 51:798-806.
Bohart, G.E. 1970. The evolution of parasitism among
bees. Utah State University, Logan, 33 pp.

Brauns, H. 1930. Neue Bienen aus Sud-Afrika. Konowia
9:43-54.

Brethes, J. 1909. Hymenoptera Paraguayensis. Museo Na-
tional de Buenos Aires, Anales, 1 2:225-256.

Cockerell, T.D. A. 1896. New species of Nomada and Chy-
photes. Canadian Entomologist, 28:284-285.

. 1898. Tables for the determination of New Mexico

bees. University of New Mexico, Bulletin, 1:41-73.

. 1900. Observations on bees collected at Las Vegas,

New Mexico, and in the adjacent mountains. Annals and
Magazine of Natural History, (ser. 7) 5:401-416.

. 1903a. New American Hymenoptera, mostly of the

genus Nomada. Annals and Magazine of Natural His-
tory, (ser. 7) 12:200-214.

. 1903b. North American bees, and a new homop-

teran. Annals and Magazine of Natural History, (ser. 7)
12:442-455.

. 1903c. Bees of the genus Nomada from California.

Philadelphia Academy of Natural Sciences, Proceedings,
55:559-579.

. 1903d. North American bees of the genus Nomada.

Philadelphia Academy of Natural Sciences, Proceedings,
55:580-614.

. 1904. Descriptions and records of bees 1. Annals

and Magazine of Natural History, (ser. 7) 14:21-30.
. 1 905a. Some bees of the genus Nomada from Wis-
consin. Canadian Entomologist, 37:189-191.

. 1905b. In, H.L. Viereck, Synopsis of bees of Or-
egon. Canadian Entomologist, 37:285-287, 313-314,
320-321.

. 1905c. The bees of the genus Nomada found in

Colorado, with a table to separate all the species of the
Rocky Mountains. Colorado Agricultural Experiment
Station, Bulletin, 94:69-85.

. 1906a. The bees of Florissant, Colorado. American

Museum of Natural History, Bulletin, 22:419-455.

. 1906b. Descriptions and records of bees, 8. Annals

and Magazine of Natural History, (ser. 7) 17:222-230.

— . 1906c. Descriptions and records of bees, 12. Annals
and Magazine of Natural History, (ser. 7) 18:69-75.

— . 1 906d. New Rocky Mountain bees, and other notes.
Canadian Entomologist, 38:160-166.

— . 1906e. Some bees from Washington State. Cana-
dian Entomologist, 38:277-282.

— . 1907a. Descriptions and records of bees, 14. Annals
and Magazine of Natural History, (ser. 7) 19:531-540.
— . 1907b. New American bees, IV. The Entomologist,
40:97-100.

— . 1907c. New American bees, V. The Entomologist,
40:265-269.

— . 1908a. Bees of the genus Nomada, belonging to the
group of N. depressa Cresson. Entomological News, 19:
323-324.

— . 1908b. A deceptive bee. Entomological Society of
Washington, Proceedings, 10:66-67.

— . 1908c. Three new bees of the genus Nomada. En-
tomological Society of Washington, Proceedings, 1 0:83—
85.

— . 1 909a. Two new bees. Canadian Entomologist, 4 1 :

35-37.

— . 1909b. New American bees, VIII. The Entomol-
ogist, 42:92-95.

— . 1910a. Descriptions and records of bees, 27. Annals
and Magazine of Natural History, (ser. 8) 5:361-369.
— . 1 9 1 0b. Some insects from Steamboat Springs, Col-
orado, II. Canadian Entomologist, 42:366-370.

— . 1910c. Some bees of the genus Nomada from

Washington State. Psyche, 17:91-98.

— . 1911a. Bees in the collection of the United States
National Museum, 1. United States National Museum,
Proceedings, 39:635-658.

— . 1911b. Names applied to bees of the genus Nomada
found in North America. United States National Mu-
seum, Proceedings, 41:225-243.

— . 1912. Descriptions and records of bees, 47. Annals
and Magazine of Natural History, (ser. 8) 10:484-494.
— . 1916a. Some bees in the British Museum. Cana-
dian Entomologist, 48:272-274.

— . 1916b. Some Neotropical parasitic bees. Ento-

mological News, 27:208-210.

— . 1916c. New and little known bees from California.
Pomona Journal of Entomology and Zoology, 8:43-64.
— . 1917. Descriptions and records of bees, 76. Annals
and Magazine of Natural History, (ser. 8) 20:235-241.
— . 1919a. The bees of the Rocky Mountain National
Park. Entomological News, 30:286-294.

— . 1919b. Bees in the collection of the United States
National Museum, 3. United States National Museum,
Proceedings, 5:167-221.

— . 1920. Descriptions and records of bees, 89. Annals
and Magazine of Natural History, (ser. 9) 6:201-21 1.

— . 1921. Western bees obtained by the American Mu-
seum Expeditions. American Museum Novitates, no. 24,
pp. 1-15.

— . 1922a. Descriptions and records of bees, 95. Annals
and Magazine of Natural History, (ser. 9) 10:265-269.

Contributions in Science, Number 376

Snelling: New World Nomadine Bees 29

. 1922b. Bees in the collection of the United States

National Museum, 4. United States National Museum,
Proceedings, 60 (no. 24 13): 1-20.

. 1925. Some bees collected by Mr. P.H. Timberlake

at Riverside, California. Pan-Pacific Entomologist, 1:
179-180.

. 1926. Descriptions and records of bees, 1 09. nnals

and Magazine of Natural History, (ser. 9) 1 7:30 1-309.

. 1929a. Descriptions and records of bees, 1 19. An-
nals and Magazine of Natural History, (ser. 10) 4:296-
304.

. 1929b. New bees from the Mesa Verde National

Park, Colorado. New York Entomological Society, Jour-
nal, 37:441-448.

. 1931. Descriptions and records of bees, 130. A nnals

and Magazine of Natural History, (ser. 10) 8:537-553.

. 1937. Bees collected in Arizona and California in

the spring of 1937. American Museum Novitates, no.
948, pp. 1-15.

. 1938. Descriptions and records of bees, 169. Annals

and Magazine of Natural History, (ser. 1 1) 2:146-154.

. 1939. Descriptions and records of bees, 171. Annals

and Magazine of Natural History, (ser. 1 1) 3:177-185.

. 1 949. Bees from Central America, principally Hon-
duras. United States National Museum, Proceedings, 98:
429-490.

Cockerell, T.D.A., and E. Atkins. 1 902. On the bees of the
family Nomadidae of Ashmead. Annals and Magazine
of Natural History, (ser. 7) 10:40-46.

Cresson, E.T. 1863. On the North American species of the
genus Nomada. Entomological Society of Philadelphia,
Proceedings, 2:280-312.

. 1869. Notes on Cuban Hymenoptera, with descrip-
tions of new species. American Entomological Society,
Transactions, 2:293-298.

. 1872. Hymenoptera Texana. American Entomo-
logical Society, Transactions. 4:153-292.

. 1875. Report upon the collections of Hymenoptera

made in portions of Nevada, Utah, Colorado, New Mex-
ico and Arizona during the years 1872, 1873 and 1874.
In Report of the Wheeler Expeditions West of 100th.
Meridian, 5:707-728.

. 1876. List of Hymenoptera collected by J. Duncan

Putnam of Davenport, Iowa, with descriptions of two
new species. Davenport Academy of Sciences, Proceed-
ings, 1:206-211.

. 1878. Descriptions of new North American Hy-
menoptera in the collection of the American Entomo-
logical Society. American Entomological Society, Trans-
actions, 7:61-136.

. 1879. Descriptions of new North American Hy-
menoptera in the collection of the American Entomo-
logical Society. American Entomological Society, Trans-
actions, 7:201-214.

Dalla Torre, C.G. de. 1896. Catalogus Hymenopterorum,
10:1-643.

Ducke, A. 1908. Contribution a la connaissance des Hy-

menopteres des deux Ameriques. Revue dEntomologie,
Caen, 27:28-55.

. 1912. Die natiirlichen Biene-Genera. Zoologische

Jahrbuch, Jena Arbeiter fur Systematische, 34:51-1 16.

Ehrenfeld, J., and J.G. Rozen, Jr. 1977. The cuckoo bee
genus Kelita, its systematics, biology, and larvae (An-
thophoridae, Nomadinae). American Museum Novi-
tates, no. 2631, pp. 1-24.

Evans, D.L. 1972. A revision of the subgenus Holonomada
of the genus Nomada (Hymenoptera: Anthophoridae).
Wasmann Journal of Biology, 30: 1-34.

Fowler, C. 1899. California bees of the genus Nomada.
Entomological News, 10:157-162.

. 1902. The long-tongued bees (Apidae) of Califor-
nia. California Agricultural Experiment Station, Report
1898-1901, pp. 338-347.

Fox, C.L. 1 926. A new Nomada. Pan-Pacific Entomologist,
2:212.

Friese, H. 1916. Zur Bienenfauna von Costa Rica (Hym.).
Stettiner Entomologische Zeitung, 77:287-350.

. 1921. Nachtrag zur Bienenfauna von Costa Rica.

Stettiner Entomologische Zeitung, 82:74-98.

Graenicher, S. 1911. Bees of northwestern Wisconsin. Pub-
lic Museum of Milwaukee, Bulletin, 1 :22 1-249.

Hicks, C.H. 1934. Some reared insect parasites and their
hosts. University of Colorado Studies, Boulder, 21:265-
271.

Holmberg, E.L. 1886. Sobre Apidos Nomadas de la Re-
publica Argentina. Sociedad Cientifica Argentina, An-
ales, 22:231-240, 272-286.

Howard, L.O. 1 902. The Insect Book. Doubleday, Page and
Company, New York, 429 pp.

Linsley, E.G. 1939. Some new genera and species of epeo-
line and nomadine bees (Hymenoptera, Nomadidae).
Pan-Pacific Entomologist, 15:1-11.

Linsley, E.G., and J.W. MacSwain. 1955a. Synonymy of
Nomada fowleri Cockerell with N. obscurella Fowler
(Hymenoptera: Nomadidae). Canadian Entomologist,
87:440.

. 1955b. The habits of Nomada opacella Timberlake

with notes on other species (Hymenoptera: Anthopho-
ridae). Wasmann Journal of Biology’, 13:253-276.

. 1959. Ethology of some Ranunculus insects with

emphasis on competition for pollen. University of Cal-
ifornia Publications, Entomology, 16:1-46.

Lovell, J.H., and T.D.A. Cockerell. 1905. The nomadine
and epeoline bees of southern Maine. Psyche, 12:39-42.

McGinley, R.J. 1981. Systematics of the Colletidae based
on mature larvae with phenetic analysis of apoid larvae
(Hymenoptera: Apoidea). University of California Pub-
lications, Entomology, 91:1-307.

Michener, C.D. 1944. Comparative external morphology,
phylogeny, and a classification of the bees (Hymenop-
tera). American Museum of Natural History, Bulletin,
82:151-326.

. 1953. Comparative morphological and systematic

studies of bee larvae with a key to the families of hy-

30 Contributions in Science, Number 376

Sneiling: New World Nomadine Bees

menopterous larvae. University of Kansas Science Bul-
letin, 35:987-1102.

. 1954. Bees of Panama. American Museum of His-
tory, Bulletin, 104:1-175.

. 1 979. Biogeography of the bees. Missouri Botanical

Gardens, Annals, 66:277-347.

Mitchell, T.B. 1962. Bees of the eastern United States.
North Carolina Agricultural Experiment Station, Tech-
nical Bulletin, 152:1-557.

Moalif, A.S. 1979. A revision of the subgenus Pachyno-
mada of the genus Nomada (Hymenoptera: Anthoph-
oridae). Thesis, Utah State University, Logan, 54 pp.

Mocsary, A. 1894. E fauna apidarum Hungariae. Ter-
meszetrajzi Fuzetek, 17:34-37.

Moure, J.S. 1960. Notes on the types of the Neotropical
bees described by Fabricius (Hymenoptera: Apoidea).
Studia Entomologia, 3:97-160.

Muesebeck, C.F.W., K.V. Krombein, and H.K. Townes.
1951. Hymenoptera of America north of Mexico. Syn-
optic Catalog. United States Department of Agriculture,
Monograph 2, Washington, D.C. 1,420 p.

Olivier, G. A. 1811. Encyclopedic methodique. Histoire na-
turelle. Publiee par une societe de gens de lettres, de
savans et d’artistes. Insectes. Vol. 8. Paris and Liege.

Parker, F.D. 1984. Biological notes on the bee Exomalopsis
crenulata Timberlake (Hymenoptera: Anthophoridae).
Pan-Pacific Entomologist, 60: 1 88-192.

Parker, F.D., and G.E. Bohart. 1983. Notes on the biology
of Andrena ( Callandrena ) helianthi Robertson. Pan-Pa-
cific Entomologist, 58:1 1 1-1 16.

Parker, F.D., and T. Griswold. 1983. Biological notes on
Andrena ( Callandrena ) haynesi Viereck and Cockerell
(Hymenoptera: Andrenidae). Pan-Pacific Entomologist,
58:284-287.

Perkins, R.C.L. 1919. The British species of Andrena and
Nomada. Entomological Society of London, Transac-
tions, 1919:218-317.

Provancher, L. 1888. Additions et corrections au volume
2 de la faune entomologique du Canada . . . Hyme-
nopteres, pp. 273-440. Quebec.

. 1896. Les demieres descriptions de l’Abbe Pro-
vancher. Naturaliste Canadien, 23:8-10, 27, 28.

Raw, A. 1977. The biology of two Exomalopsis species
(Hymenoptera: Anthophoridae) with remarks on soci-
ality in bees. Revista Biologia Tropical, 25:1-1 1.

Robertson, C. 1893. Notes on bees, with descriptions of
new species. American Entomological Society, Trans-
actions, 20:245-248, 273-276.

. 1895. Notes on bees, with descriptions of new

species. Third paper. American Entomological Society,
Transactions, 22:1 15-128.

. 1897. North American bees— descriptions and syn-
onyms. Academy of Science of St. Louis, Transactions,
7:315-356.

. 1900. Nomada sayi and two related new species.

Canadian Entomologist, 32:293-295.

. 1902a. Some new or little known bees. Canadian

Entomologist, 34:48-49.

. 1902b. Some new or little known bees. Entomo-
logical News, 8:79-81.

. 1903. Synopsis of Nomadinae. Canadian Ento-
mologist, 35:172-179.

Rodeck, H.G. 1931. Rocky Mountain bees. IV. American
Museum Novitates, no. 496, pp. 1-1 1.

. 1945. Two new subgenera of Nomada Scopoli

(Hym.: Apoidea). Entomological News, 56:179-181.

. 1947. Laminomada, new subgenus of Nomada

(Hym.: Apoidea). Entomological Society of America,
Annals, 40:266-270.

. 1949. North American bees of the genus Nomada,

subgenus Cal/inomada. Entomological Society of Amer-
ica, Annals, 42:174-186.

Rohwer, S.A. 1911. A new genus of nomadine bees. En-
tomological News, 22:24-27.

Rozen, J.G., Jr. 1966. The larvae of the Anthophoridae
(Hymenoptera; Apoidea). Part 2. The Nomadinae.
American Museum Novitates, no. 2244, pp. 1-38.

. 1969. The larvae of the Anthophoridae (Hyme-
noptera, Apoidea). Part 3. The Melectini, Ericrocini, and
Rhathymini. American Museum Novitates, no. 2382, pp.
1-24.

. 1977. Immature stages of and ethological obser-
vations on the cleptoparasitic bee tribe Nomadini (Apoi-
dea, Anthophoridae). American Museum Novitates, no.
2638, pp. 1-16.

Rozen, J.G., Jr., K.R. Eickwort, and G.C. Eickwort. 1978.
The bionomics and immature stages of the cleptopar-
asitic bee genus Protepeolus (Anthophoridae, Nomadi-
nae). American Museum Novitates, no. 2640, pp. 1-24.

Rozen, J.G., Jr., and R.J. McGinley. 1974. Systematics of
ammobatine bees based on their mature larvae and pu-
pae (Hymenoptera, Anthophoridae, Nomadinae).
American Museum Novitates, no. 2551, pp. 1-16.

Say, T. 1824. Natural History. In W.H. Keating, Narrative
of expedition to the source of St. Peter’s River, etc. Vol.
2. H.C. Carey and I. Lea, Philadelphia.

. 1837. Descriptions of new species of North

American Hymenoptera, and observations on some al-
ready described. Boston Journal of Natural History, 1 :
362-416.

Schrottky, C. 1920. Himenopteros nuevos o poco cono-
cidos sudamericanos. Museu Paulista, Revista, 12:177-
227.

Schwarz, M. 1966a. Zur Kenntnis der Gattung Nomada
Scop. (Hymenoptera Apoidea). Nachrichtenblatt der
Bayerischen Entomologen, 1 5: 1 7-24.

. 1966b. Beitrag zur Subfamilie Nomadinae (Hy-
menoptera, Apoidea). Polskie Pis mo Entomologiczne,
36:383-394.

Scopoli, J.A. 1770. Historico naturalis. Leipzig, annus 4.

1 50 pp.

Smith, F. 1854. Catalogue of hymenopterous insects in the

Contributions in Science, Number 376

Snelling: New World Nomadine Bees 31

collection of the British Museum. London, vol. 2, pp.
199-465.

. 1879. Descriptions of new species of Hymenoptera

in the collection of the British Museum. London, 240

pp.

Stephen, W.P., G.E. Bohart, and P.F. Torchio. 1969. The
biology and external morphology of bees, with a synopsis
of the genera of northwestern America. Agricultural Ex-
periment Station, Oregon State University, Con’allis, 140

pp.

Swenk, M.H. 1913. Studies of North American bees. I.
Family Nomadidae. University of Nebraska Studies, 12:
1-113.

. 1915. Studies of North American bees. III. Families

Nomadidae and Stelididae. University of Nebraska
Studies, 15:155-193.

Timberlake, P.H. 1954. Two new species of Nomada, sub-
genus Gnathias, from California (Hymenoptera: Apoi-
dea). Pan-Pacific Entomologist, 30:133-136.

Viereck, H.L. 1905. Synopsis of bees of Oregon, Washing-
ton, British Columbia and Vancouver. IV. Canadian
Entomologist, 37:277-287.

Wamcke, K. 1977. Ideen zum natiirlichen System der Bie-
nen. Mitteilungen Mimchen Entomologische Gesell-
schaft, 67:39-63.

. 1982. Zur Systematik der Bienen— die Unterfam-

ilie Nomadinae (Hymenoptera, Apidae). Entomofauna,
Zeitschrift fur Entomologie, 3:97-126.

Zucchi, R. 1973. Aspectos bionomicos de Exomalopsis au-
reopilosa e Bombus atratus incluindo consideracoes sobre
a evolucao do comportamento social (Hymenoptera—
Apoidea). Thesis, Faculdade de Filosofia, Ciencias e
Letras de Ribeirao Preto, Ribeirao Preto (S.P.), Brazil,
172 pp.

Accepted 8 January 1986.

32 Contributions in Science, Number 376

Snelling: New World Nomadine Bees

IlllillS

relevant identifying .details; The depository of other study' material sliouid

: ^ . iiaM

An abstract is required for all- papera, In.oianuforipts dealing with I.. aim American subjects'

must be presented in Spanish or Portuguese, as 4|j>|5ffi|||ia1e ,j a s wel \' as; ,» Epjjgl ish .

other languages are no

INSTRUCTIONS FOR AUTHORS

> Museum of Los' Angeles County publishes the results of original research
s in its Contributions in Science series., Individual ;Cbnlrifc:ut'idns:
.ular intervals and range in size from, papers of S jnihted pages to lengthy

bmitt-H ■

for' publication will undergo anoriytnqus: peer -review. Priority ss
uscrtp'S. wrritfeh by. members of the Musaifo staff lyjfanuscripts should be
sccordance with the requirements outlined below and submitted to the Head of
Section of the .Museum. ,

lere to the articles- of the J nteriiationa! ; Codes, of Botanical or Zoological
.re urged to comply with -all recommendations of those codes,
new taxa must Indicate that pri.ma.ty types have bberr deposited in

e reco mmendaiions of the appropriate code., citing the collection bv name

1 ' km’ ir K * ^ : w

but are strongly, recprrrmended where ap-

iifti;

SB!

■cript double-spaced, including. title page, abstract, text,, acknowledgments, re li-
es, and ligure captions. A paper must total a minimum of 8 primed pages (as a.

draft equals about 1 page of .final printed copy). -Indicate ffolitfeed

s; do not use an italic typing element.

component should begin on a new- page, in the following sequence: title

acknowledgments; literature cited; tables, each lablc— complete .with
:i. a separate page;, and ligure captions. All pages should be numbered

s. Line drawings are best prepared for 50 percent' reduction.
17.5 inches: (35 by 43 alii iiriiake; ptbprbgraphs are usually
should not exceed f:;i» cm, ftfll page width)

<USCR.IPT

secure enclosure. Marmse'rip': pi§ukl be ,ac-
r who wd! be i|kp a risible for correspondence
r should contain a stateme.r;i that Lite m ant;! scrips
been published blaM.jfefo :dicc»( In i ■bHPitot j or :
in forma tion. on. pips; syoti-bb cation of any part
available to meet the cost of unusual format

be reviewed tor possible publication: with itbepnde
, ashed. simultaneously ip;|foaMed; or . already ■ accepted : for publication
does no s precl odd ednsi dera t i or. of a. complex report that follow s publication :
findings elsewhere. . :Copi es of any possibly d up 1 ica t i.x 'em a I e r:i If s h o uk- be;
the manuseripl that is being sent for c^a|i|dkltibh; ;||||j|||| iljjlgiyjy (If

: ^ ! :' |L

.wild receive a statement of page

■ , *« »PiMlS««33

.fa." • . i-wa I '? :'b{; !7;;| ;,;J r;|| pii . J f . b f i: ri: i I : :i IF! : :j f ! ' ; j • I :■ j :[ -1 -rt ;jp;j!!!j |!!;:ji ji':|; >:

xep.tance of a paper by the ; •Musnmif d,H M
t from the Managing Editor calculated at a current rate. Partial or coiopl
, Mi „ c.- - "i1 * -|j|" whp have funds a

* :fW partial or complete: pay"'’''*'*
im it to the Managing
publication; acceptance of a

iyrnent

Sii

-.fo |i|i||»niyi:^

yment of the

ill

l[ i !f ;

If' Ml

IS

■ 'ill® r

w3Wila

11 rtf 'll of! nli#,, i ,„!■< S i jlj't i

■ISlf fll if 'I I ■;!!

IBiPi ill I

■ ■ :l:-l,1i ■Hr"

'j]i!j!!i|il;Hpiii!i!i

llll 1 1

HHBwifl

I If |u:.;:- i«i i

i :j! { .. i •: 1 ;;; ; ],■ r .: ! • : it. .. I ,

■ i 'i t 1 iVm1, -u w

: ii "i

t HliMliilll
u tSigiii 19 o MB Mil

■ ; ii

a L -ii ' i !• . •: Mvii.i.

m

m-r^BBKJ jufflfe * tJB

IP.Jllil

iiiiiiliiiii

ifliiin

IfHrffr

SERIAL PUBLICATIONS, OF THE
I NATURAL ii ISTORY MUSEUM O# LOS ANGELES COUNTY

The sd*

lave been

issued at irregular intervals1 in three major series; the articles’ in each series are numbered-
individually, and numbers run consecutively, regardless of the subject, matter. ,’|*^

^Contributions- in .Science, a. miscellaneous series of technical papers, describing1 orig-
inal'research in the life and earth sciences.

Itejence Bunetin , a- miscellaneous series of monographs describing original .research
iirt the life and- earth sciences; This series was discontinued in 1978 with the issue of

Numbers 29 and 30; monographs are now published by the Museum in Contributions

m Science.

<§ Science Series.. long articles on natural history topics, generally written for the layman .

Copies of the publications in these series are sold through, the Museum Book Shop. A catalog
is available on request.

: SCIENTIFIC PUBLICATIONS COMMITTEE

Craig C. Black, Museum Director
Richard C. Brusca

Daniel M. Cohen, Committee Chairman
John M. Harri$

Charles L, Hogue

Gejorge I. . Kennedy ; \ j

Robin' A. Simpson, Managing Editor

Enrol . StevehS;,'

llSpl-i
l! iSinHi!ii:::‘

li.Kl'.l '

!; M Rill

■

i Afp|y§;® BMiiWr

SililllsS A., mf |{tnf itjijmaill |

i! I will HI UBWH, . ....

'M 1 ir at .V.; sen Pr«js, itnof La wer,.ce. ' Kansas j

! i«ijp PiH!Si:PiW!!.IMI

■gl .!'■

' I ''fil'T I'n I

il

11M jilniS

Ay a.:tI!

!ii

Hi

CiU'tilKfc,.

iiife

W8M

hi

1 * pfltattilr-'iAl! L tj <„fts

.. i I i r i)lrplf i 11 "ffl NbPlfll

- ft|h!

liiii

;^D

RECENT DISCOVERIES IN THE BLEPHARICERA TENUIPES GROUP,
INCLUDING DESCRIPTIONS OF TWO NEW SPECIES FROM
APPALACHIA (DIPTERA: BLEPHARICERIDAE)

Charles L. Hogue and Ted Georgian

DESCRIPTION OF A NEW SPECIES OF THE SHORE FLY GENUS
DIEDROPS (DIPTERA: EPHYDRIDAE) FROM COLOMBIA

Wayne N. Mathis and Charles L. Hogue

Contributions in Science, Number 377
Natural History Museum of Los Angeles County
3 September 1986

ISSN 0459-8113

Natural History Museum of Los Angeles County
900 Exposition Boulevard
Los Angeles, California 90007

CONTENTS

RECENT DISCOVERIES IN THE BLEPHARICERA TENU1PES GROUP, IN-
CLUDING DESCRIPTIONS OF TWO NEW SPECIES FROM APPALACHIA (DIP-
TERA: BLEPHARICERIDAE)

Charles L. Hogue and Ted Georgian 1

DESCRIPTION OF A NEW SPECIES OF THE SHORE FLY GENUS DIEDROPS
(DIPTERA: EPHYDRIDAE) FROM COLOMBIA
Wayne N. Mathis and Charles L. Hogue 21

RECENT DISCOVERIES IN THE BLEPHA RICERA TENUIPES GROUP,
INCLUDING DESCRIPTIONS OF TWO NEW SPECIES FROM
APPALACHIA (DIPTERA: BLEPHARICERIDAE)

Charles L. Hogue1 and Ted Georgian2

ABSTRACT. Two new species of Blepharicera (B. appalachiae and
coweeiae ) are described in the Blepharicera tenuipes group. Both
occur in the southern half of the Appalachian Mountains in eastern
North America, the former of wide distribution, the latter restricted
to a small portion of the upper Little Tennessee drainage. A key to
all the known stages in the B. tenuipes group is provided along with
new information on ecology, distribution, and phylogeny (for which
the sister Blepharicera micheneri group is newly recognized).

INTRODUCTION

At the time of his review of the net- winged midges of eastern
North America, Hogue (1978) preferred not to make definite
identifications of the immatures of any species because of
the uncertainty of stage associations until better material
could be acquired. He was also unable to distinguish more
than one pupal type. The recent availability of numerous
specimens, including some valuable individual rearings, col-
lected in the vicinity of the United States Forest Service’s
Coweeta Hydrologic Laboratory (Macon County, North Car-
olina) by Georgian and associates, has now made it possible
to separate pupae of five of the eastern species. Using integ-
umentary characters of this intermediate stage, we can also
identify several of the larvae by inspection of prepupal larvae
and pharate adults.

These integumentary characters include the patterns and
detailed structures of the papillae and cuticular ornamenta-
tion of the abdominal tergites. Streams in the Coweeta lo-
cality produce two very distinctive pupal types, the adults of
which show small, but consistent features that distinguish
them from existing species. They are considered new species
and are described here.

Other discoveries in the Blepharicera tenuipes group also
are presented. Complete synonymies are reserved for a
monographic treatment of the Blephariceridae of North
America soon to be published (Hogue, in press).

MATERIALS AND METHODS
MATERIALS

Most of the new material used for this study was collected
by T. Georgian and D.H. Ross, and placed in the University

Contributions in Science, Number 377, pp. 1 20
Natural History Museum of Los Angeles County, 1986

of Georgia Entomological Museum and the Entomology Sec-
tion of the Natural History Museum of Los Angeles County.
However, some paratypes are deposited in the U.S. National
Museum of Natural History. See the Acknowledgment sec-
tion for explanations of abbreviations used for sources of
other specimens.

A number of larvae and adults of B. appalachiae were
considered by Hogue ( 1 978) as a variant of B. tenuipes (Walk-
er, 1848). These specimens have been reevaluated and some
identifications modified. All those that are now determined
as belonging to the new species are listed here; the others
cited in that paper as “Atypical adults” (p. 23) and “Larva
E” (p. 30) are still of uncertain identity.

TERMINOLOGY

The terminology used in this paper is mostly that established
in Hogue’s (1978, 1981) earlier treatments of blepharicerid
anatomy. One important change is the application of the
name “ lateral tine ” to the prong-like, tubular rods lateral to
and paralleling the aedeagal filaments and which have been
called “parameres” in most previous works on the family.
The tines appear to be secondary developments of the ventral
plate (whose presence is probably plesiomorphic for the fam-
ily) as proposed by Stuckenberg (1958:101). They are rep-
resented either by ( 1 ) no more than mere pigmented straps,
or (2) as columnar projections without a lumen. The first
case occurs in Blepharicera micheneri and its close relatives
in western North America; the second is the condition in
Paltostoma, as seen in Hogue’s (1979) figures of Costa Rican
species. Blepharicerid tines are possibly homologous to sim-
ilar structures given the same name in the terminalia of tab-
anids by Bonhag (1951:161, 198).

The neutral term “ gonites ” is used here for the large, wide
plates subtending and sometimes surrounding the base of the

1. Entomology Section, Natural History Museum of Los Angeles
County, 900 Exposition Blvd., Los Angeles, California 90007.

2. Biology Department, Saint Bonaventure University, Saint
Bonaventure, New York 14778.

ISSN 0459-8113

globose lanciform taeniaform oviform

pyriform

setiform

short

claviform

Figure 1. Various forms taken by the modified, dorsal secondary sensilla of the larvae of the Blepharicera tenuipes group. Names as used
in descriptions.

aedeagus that have been also called “parameres” by authors
(McAlpine, 1 98 1 : 5 1 f). According to G.C.D. Griffiths (pers.
comm.), the latter term should be reserved presently for quite
distinct organs in other insect orders and not applied to the
Diptera until controversies in homologies can be settled.

Leg segment proportions are expressed “progressively,”
that is, each segment is compared to its proximal neighbor
rather than to one standard segment (customarily the most
basal segment).

The various forms taken by the modified (generally short,
peg-like), dorsal secondary sensilla of the larvae are described
with adjectives whose precise definition can be appreciated
by the shapes labelled in Figure 1 (only those displayed by
species included in this paper are given). See Hogue (1978)
for designations of primary setae.

MEASUREMENTS

All measurements are made from topotypic material, unless
otherwise noted, and cited in millimeters. The first value
represents the mean of the number of specimens indicated
by N; values following in parentheses are range extremes.
Larval body length is taken only on prepupal specimens (i.e.,
those showing some trace of pupal branchiae). Adult pro-
boscis length is approximate, measured from a point level
with the venter of the eyes to the apex of the labella. The

formula for expressing comparative sizes of male vs. female
pupae is (L)(W) male/(L)(W) female. Meristic counts of sen-
silla are cited for one side of bilateral structures as absolute
values “(20)” or ranges “(15-20)”; where it is necessary to
cite different values for each side, a double parenthetic no-
tation is used “(5)(15)”.

Blepharicera tenuipes Group

PHYLOGENY

With the addition of the two new species described in this
paper the Blepharicera tenuipes group contains eight species,
all restricted in geographic occurrence to the eastern United
States (Appalachians, northward to Hudson’s Bay and Lab-
rador, westward to eastern Minnesota). We consider them
very closely related and all descended from a single ancestor,
as did Hogue (1978).

It has not yet been possible to arrive at a clear phylogeny
of the species within the group, because of their homogeneity
and incongruous variation in many characters. These bleph-
aricerids are so alike morphologically that relatively few syn-
apomorphies can be distinguished, except one found in the
male terminalia. Also, a reticulate pattern of the following
synapomorphies emerges when they are applied toward con-
struction of a cladogram:

2 Contributions in Science, Number 377

Hogue and Georgian: New Appalachian Blepharicera

1. Inner margin of the Xth tergite lobe convexly expanded
medially. A simple, straight inner margin is the plesiomor-
phic condition.

2. A slight to well-developed apicomedial, dorsal carina
present on the tegmen. The plesiomorphic tegmen is flat
apically.

3. Apex of lateral tine with a conspicuous recurved hook.
A simple apex is plesiomorphic but other autapomorphies
are common.

4. Apex of tegmen strongly emarginate on either side of
median carina. The plesiomorphies entire to weakly emar-
ginate, may actually be reduction apomorphies correlated
with small size in those species with this condition. The deep
emargination nevertheless is a strongly positive apomorphy.

5. Median aedeagal filament decidedly heavier than lat-
erals.

6. Many integumentary papillae of pupal tergites occurring
in diads. These papillae are normally more or less evenly
spaced on the sclerites.

Most of the species are closely related to and similar to
tenuipes, judging from the common shape of the Xth tergite
lobe (quadrate with distinct triangular apical sublobe). This
configuration may be considered plesiomorphic to the type
found in wi/liamsae Alexander, 1953, capitata Loew, 1863,
and similans Johannsen, 1929, that have a greatly expanded
medial margin, and therefore not a reliable state for indi-
cating relationships. Among these latter three species, the
heavy (synapomorphic) median aedeagal filament would seem
to indicate a close relationship between the first two, but the
very different similans could have closer affinities with the
tenuipes cohort than with the other two species because of
its somewhat more similar Xth tergite lobe. Among the species
clearly similar to tenuipes, cherokea Hogue, 1978, and dimi-
nutiva Hogue, 1978, segregate by their smallness, a dubious
synapomorphy when reduction in size can occur often in this
midge family. Complicating the picture still more is the pres-
ence of a fairly strong apical carina on the tegmen in wil-
liamsae which relates it toward tenuipes and its close rela-
tives, but away from capitata. Among the latter, appa/achiae
and tenuipes seem closely related on the basis of the common
occurrence of irregular diad formations in the patterns of
integumentary papillae on the pupae.

Therefore, at this stage of knowledge of the Blepharicera
tenuipes group, we refrain from attempting a cladogram. More
analysis of characters is needed before a clear selection of
phylogenetically significant states can be ascertained. The
possibility of hybridization between species may also have
affected their evolution. Ultimately, it may be necessary to
resort to chromosomal or biochemical analyses to detect in-
terrelationships.

ZOOGEOGRAPHY

Zwick (1984) proposed a preliminary phylogeny of the genus
Blepharicera. He considered the species of the western United
States (with the exception of B. ostensackeni, here called the
“ Blepharicera micheneri group”), the sister group to the B.
tenuipes group. The ancestral stock of both groups probably

arrived in North America from Asia because most Blepha-
ricera reside there. The single European species belongs to a
distinct group. This hypothesis conflicts with Hogue’s earlier
(1978:1-2) supposition of a fundamental division between
the two stocks. He suggested that the former arrived in North
America from the northwest, while the latter dispersed sep-
arately, and probably earlier, over an eastern connection with
the Palaearctic Region.

Zwick’s arguments are convincing. The problem remains,
however, of connecting the two lines paleogeographically,
i.e., to answer the question, by what route did the B. tenuipes
group reach eastern North America from the northwest? Also,
what disruptive events forced the separation of the two groups?

Only a northern path across the Canadian Shield would
seem a plausible answer to the first question, because a south-
ern highland corridor was present no later than the Permian.
It is unlikely that the Blepharicerini would have evolved
before that time. Invasion of North America and movement
eastward could have been associated with the mid-Cenozoic
spread of the temperate deciduous forests, as supposed by
Ross (1956) for various mountain caddisflies, such as the
“Rhyacophila Siberica Group” (and “Wormaldia Anilla
Group,” and subgenus Doloclanes — p. 181-182), with dis-
tributions identical to or concordant with the Blepharicera
in question. Baumann (1975) also noted similar patterns in
some rheophilic stoneflies (Podmosta). A newly discovered
montane genus of limnephilid caddisfly bears very strong
affinities to western forms and is believed to have dispersed
across the northern deciduous forest and undergone subse-
quent speciation in the Appalachians (Huryn and Wallace,
1984). The absence of Blephariceridae from the Ouachita
and Ozark mountains, remnants of that intervening orogenic
zone, and which surely would preserve some vestige of any
past blepharicerid fauna, also provides evidence, albeit neg-
ative, in favor of the northern alternative.

The absence of a continuous east-west mountain chain
along which these generally monticolous midges could have
dispersed does not preclude the hypothesis of a northern
route. Blepharicera tenuipes extends over low postglacial ter-
rain in southeastern Canada, indicating the ability of these
insects to colonize rapidly well-watered land with little relief.
The distance between the most southern extreme of the Wis-
consin ice sheet across Appalachia and the northernmost
recorded Recent occurrence of the species is approximately
1500 km. The ice sheet began retreating from its terminal
position about 15,000 years ago (Davis, 1983), giving the
flies a dispersal rate of at least 0.1 km per year. To traverse
the distance from the Rocky Mountains to the Appalachian
Mountains (3500 km) would require about 35,000 years, an
easy march, geologically speaking.

Increasingly arid conditions in the late Cenozoic undoubt-
edly forced the division of the eastern and western groups.

In North America, the greatest variety of species is con-
centrated in the southern Appalachian Highlands, particu-
larly in the Blue Ridge area, an acknowledged important
center of speciation and a refugium from glaciation (see the
various papers listed in Holt, 1969). Although Blepharicer-
idae are poorly vagile insects, three species (B. tenuipes, sim-

Contributions in Science, Number 377

Hogue and Georgian: New Appalachian Blepharicera 3

Hans, and capitata) dispersed considerable distances north-
ward in the postglacial era. This seems to show that the adults
can move across drainage boundaries fairly easily. Stream
capture, while a common phenomenon in the geologic history
of the area and significant for the dispersal of strictly aquatic
organisms, is not their only means of colonization of new
watersheds. However, diminutiva may have entered the Sa-
vannah drainage from the Little Tennessee when Stekoa Creek
diverted the southern portion of the latter (south of Rabun
Gap, Georgia) (Ross, 1971:32). The restricted distributions
of cherokea and coweetae to the Little Tennessee (Ohio drain-
age) and northward indicates their probable origin subse-
quent to this stream-capture event.

TAXONOMY

The Blepharicera tenuipes group was defined by Hogue (1978:
6-8) on the basis of the adult flies. A more complete de-
scription and definition will appear in Hogue (in press). Zwick
(1984) demonstrated some new diagnostic features, in par-
ticular some correlated states in the mesosomal complex of
the male terminalia, namely, the short, heavy lateral tines,
small apodeme of the sperm pump, and anteriorly expanded
gonite from which arises a large muscle narrowly inserting
at the base of the aedeagal tine and lateral to which the outer
aedeagal filament passes. In the B. micheneri group this mus-
cle passes lateral to all the aedeagal filaments, which are much
reduced in size. The former arrangement is an autapomorphy
for the group.

Identification of larvae in the group is somewhat problem-
atical. The following keys only apply to typical specimens in
the mature, fourth instar. Atypical larvae may show a con-
siderable variety of conditions of the shape and distribution
of the dorsal, secondary sensilla which are the primary iden-
tifying characters. There are many intermediates and non-
conformants (species appearing to be out of their range or
associated with stages of distinct species) and these cannot
be assigned with certainty to any species. Therefore, many
of the larval records cited by Hogue (1978) are suspect. Only
detailed biological work with these Diptera can hope to shed
light on the causes and significance of larval variability. In-
trogression among the closely related members of the group
is suspected.

KEY TO ADULT MALES

(Based on the terminalia; slide mounts normally
required to see structures)

la. Medial aedeagal filament distinctly longer than laterals
and with asymmetrical, hooked apex. Apex of lateral
tine incurved, asymmetrically bifurcate. Outer comer of
Xth tergite lobe slightly obtusely angulate, inner margin

shallowly convex similans

b. Medial aedeagal filament about same length as laterals
and with simple apex. Apex of lateral tine not bifurcate.
Shape of Xth tergite lobe varied 2

2a. Medial aedeagal filament distinctly thicker than laterals.
Inner margin of Xth tergite lobe convexly expanded

3

b. Medial aedeagal filament similar to laterals. Inner mar-
gin of Xth tergite lobe straight 4

3a. Inner wall of lateral tine incurved, apex acute and ex-
tending well beyond aperture. Inner margin of Xth ter-
gite lobe expanded angularly to or beyond midline. Larg-
er species, wing length 4. 9-6. 6 mm williamsae

b. Inner wall of lateral tine very slightly produced, truncate,
with a minute, pointed, dorsal projection, aperture ter-
minal. Inner margin of Xth tergite lobe expanded evenly
and moderately, far short of midline. Smaller species,

wing length 4. 8-5. 3 mm capitata

4a. Apex of tegmen deeply incised on either side of strong,
medial dorsal carina. Tip of lateral tine complex, with

dorsally recurved, stout hook (Figs. 5, 9) 5

b. Apex of tegmen weakly incised or entire on either side
of weak, medial dorsal carina. Tip of lateral tine sim-
ple 7

5a. Interlobular space (between Xth tergite lobes) U-shaped

(Fig. 8) 6

b. Interlobular space V-shaped (Fig. 8) coweetae

6a. Posteromedial triangular sublobe of Xth tergite lobe sit-
uated midway along posterior margin tenuipes

b. Posteromedial triangular sublobe of Xth tergite lobe dis-
placed nearly to inner corner (Fig. 4) .... appalachiae
7a. Posteromedial triangular sublobe of Xth tergite lobe con-
spicuous, well produced; outer comer of lobe obtusely

rounded cherokea

b. Posteromedial triangular sublobe of Xth tergite lobe mi-
nute and barely produced; outer comer of lobe acutely
rounded diminutiva

KEY TO ADULT FEMALES

(Slide mounts required to see details of terminalia;
wing lengths given for confirmation)

la. Two spermathecae, medial reduced to a narrow cylin-

drical rod. Wing length 5. 4-7. 3 mm similans

b. Three spermathecae 2

2a. Ducts of lateral spermathecae sclerotized for a short dis-
tance before genital aperture. Wing length 5. 2-6. 4 mm

capitata

b. Ducts of spermathecae membranous throughout ... 3
3a. Scutum with a contrasting, light brown, quadrate area
anterior to scutellum. VUIth stemite lobe of terminalia
devoid of setae. Spermathecae spherical. Large species;
wing length usually 7.2 mm or greater .... williamsae
b. Scutum generally unicolorous. VUIth stemite lobe of
terminalia usually with several setae, at least one. Small-
er species; wing length 7. 1 mm or less 4

4a. VUIth stemite lobes of terminalia contiguous, medial
depression V-shaped. Very small species; wing length

about 5.1 mm diminutiva

b. VUIth stemite lobes of terminalia disjunct, medial
depression U-shaped. Larger species; wing length 5.5
mm or greater 5

4 Contributions in Science, Number 377

Hogue and Georgian: New Appalachian Blepharicera

Note: Remaining species very difficult to distinguish. Fol-
lowing continuation of key is for typical specimens only.
Wing lengths overlap.

5a. Accessory gland narrow throughout, apparently (one
specimen available) sclerotized and pigmented ante-
riorly cherokea

b. Accessory gland widened at some point, unpigmented

throughout 6

6a. Accessory gland very wide posteriorly, twice narrow'ed
anteriorly. Lobe of hypogynial plate apically truncate .

tenuipes

b. Accessory gland moderately wide posteriorly, narrowing
once anteriorly. Lobe of hypogynial plate variously

shaped 7

7a. Anterior portion of accessory gland straight, parallel-
sided. Lobe of hypogynial plate broadly rounded apically

appalachiae

b. Anterior portion of accessory gland dilated. Lobe of hy-
pogynial plate narrowly rounded coweetae

KEY TO PUPAE

(Pupae of capiiata, diminutiva, and cherokea unknown)

la. Integument of abdominal tergites with a fine, reticulate
pattern (Figs. 20, 25); surface dull. Larger (length usually

5.5 mm or greater) williamsae

b. Integument of abdominal tergites homogeneous or oth-
erwise sculptured, no reticulate pattern; surface shiny.

Smaller (length usually less than 5.5 mm) 2

2a. Integument of abdominal tergites showing minute wavy
furrows between and connecting the papillae. Papillae

minutely spinulate (Figs. 18, 22) coweetae

b. Integument of abdominal tergites between papillae un-
interrupted by cuticular furrow's or lines. Papillae smooth

3

3a. Papillae, especially in region lateral to muscle scars of
abdominal tergites clustered, unevenly distributed, often
arranged in pairs or diads (Figs. 12, 17, 21, 24) .... 4
b. Papillae all more or less evenly spaced (Figs. 19, 23). .

similans

4a. Frequent papillar diads in medial zone of abdominal
tergites (between muscle scars) as well as lateral (Figs.

17, 21) appalachiae

b. Papillae of median zone of abdominal tergites all more
or less evenly spaced (Figs. 12, 24) tenuipes

KEY TO LARVAE

(Larvae of capitata, cherokea, and diminutiva unknown.
Larva “F” (Hogue, 1978) apparently an aberration
of other species; see below')

la. Dorsal secondary sensilla mostly setiform (Fig. 1 3) (tae-
niaform to elongo-fustiform3 in many specimens) ....
appalachiae

3. Latin: fustis = cudgel + forma = shape.

b. Dorsal secondary sensilla mostly modified (capitate to

claviform) present dorsally 2

2a. Modified sensilla large, globose, apically spinulate, clus-
tered in vicinity of st and tp primary sensillae and an-

teromedially on segment (Fig. 15) similans

b. Modified sensilla small to medium, mostly elongate
forms, apically smooth, clustered or diffusely arranged

3

3a. Modified sensilla short claviform to pyriform, arranged
generally in transverse anterior and posterior clusters,
the latter including primary sensillum st (Figs. 10, 11)

tenuipes

b. Modified sensilla more or less generally distributed over
segment (may be in broad transverse patterns but never

clustered) 4

4a. Modified sensilla lanciform to fusiform, arranged in 3
broad transverse fields medially (Fig. 1 6). Dorsum lon-
gitudinally dark pigmented medially, light laterally . . .

williamsae

b. Modified sensilla pyriform, glandiform or oviform, dif-
fusely distributed (Fig. 14). Dorsal pigmentation uni-
form coweetae

DESCRIPTIONS OF NEW SPECIES

Blepharicera appalachiae , new species

Figures 2-5, 13, 17, 21, 26

Blepharicera tenuipes “Atypical (southern type)” of Hogue,

1978:21, fig. 24.

Larva “E” (partim) Hogue, 1978:29-30.

Larva “C” Hogue, 1978:28-29, fig. 35.

DIAGNOSIS. In the male terminalia, while all structures
are otherwise virtually identical to those of B. tenuipes and
other members of the B. tenuipes group, the posterior margin
of the Xth tergite lobe is uniquely shaped: the outer comer
is acutely rounded, followed medially by a deep concavity
and central sublobe strongly displaced toward the inner cor-
ner. The secondary sensilla on the larval dorsum are typically
setiform rather than peg-like or otherwise modified, as in
tenuipes and other members of the group. (A few modified
sensilla may intrude in some specimens.) The arrangement
of the dorsal papillae distinguishes the pupa. The papillae
are very irregularly arranged, neighboring pairs often co-
alescing into diads over the entire surface of the abdominal
tergites, but more so in the outer areas lateral to the main
muscle scars.

DESCRIPTION. Adult a small, sturdily built, well-scler-
otized blepharicerid. Very similar to B. tenuipes and co-
weetae.

Male (Figs. 2, 4-5). Coloration. Generally dull gray-brown,
pruinose. Mesoscutum concolorous, dull gray. (With illu-
mination at some angles, lateral portions become translucent,
reddish-brown, medial area longitudinally brown to black-
ish.) Comers of posterior pronotum and scutellum contrast-
ing light brown, pleurites similar to scutum. Wing membrane
completely hyaline. Size. A medium-sized Blepharicera.

Contributions in Science, Number 377

Hogue and Georgian: New Appalachian Blepharicera 5

Figures 2-5. Blephahcera appalachiae, new species. 2. Male head. 3. Female head. 4. Male terminals external structures (dorsal view). 5.
Male terminalia, phallic structures (dorsal view).

6 Contributions in Science, Number 377

Hogue and Georgian: New Appalachian Blepharicera

Measurements (from non-topotypical material) (N = 10):
Wing length 5.3 (5. 0-6.0). Leg segment lengths:

femur
tibia
tarsus 1
2

3

4

5

foreleg

3.2 (3. 1-3.7)
3.0 (2.9-3. 3)
1.6 (1.5-1. 8)
0.80 (0.76-0.89)
0.55 (0.51-0.64)
0.29 (0.27-0.36)
0.29 (0.27-0.32)

midleg

3.4 (3.2-3. 9)
2.9 (2.8-3. 3)

1.5 (1.4-1. 7)
0.74 (0.66-0.81)
0.52 (0.48-0.57)
0.29 (0.27-0.34)
0.29 (0.27-0.34)

hindleg

4.6 (4.3-5. 1)
4.3 (4.0-4.8)

1.7 (1. 5-1.9)
0.53 (0.46-0.62)
0.35 (0.29-0.42)
0.24 (0.22-0.29)
0.28 (0.25-0.33)

Head (Fig. 2). Structure: Normal type, subholoptic. Su-
prafrontal carina a rounded convexity. Clypeus elongate,
L/W = 2.0. Eyes approximate dorsally, interocular distance
equal to combined diameters of 5 upper ommatidia; eye
divided, upper division well differentiated from lower (callis
oculi narrow), much smaller (0.4 x) than lower in area, 10-
1 1 rows of ommatidia along mid-meridian; upper ommatidia
slightly larger ( 1 .4 x ) than lower in diameter. Proboscis short,
free portion about 0.4 x head width; mandibles completely
absent; palpus 5-segmented, distal 4 palpal segment propor-
tions 1.0-1. 0-1. 3-3. 4. Antenna 1 5-segmented, flagellar seg-
ments elongate throughout, ultimate longer ( 1 .4 x ) than pen-
ultimate, apical 3 segment proportions 1.0-1. 0-1. 4. Sensilla:
Setiform groups on head capsule as follows: clypeals several
(10), medium, only a few, small setae medially, more nu-
merous and longer along distolateral margins. A single, me-
dium strong seta over antennal socket; facial groups other-
wise absent. Postocellars few (2-3), small, lateral.
Medioccipitals absent. Supracervicals very numerous (30),
tiny. Occipitals separated from and larger than postgenals,
numerous (30), long. Postgenals numerous ( 1 7-26), smaller
toward center of group.

Thorax and appendages. Structure: Wing venation typical
for Blepharicera. Tibial spurs 0-0- 1 . Progressive leg-segment
proportions: foreleg 0.9-0. 6-0. 5-0. 7-0. 6-0. 9; midleg 0.8-0. 5-
0.5-0.7-0. 6-1.0; hindleg 0.9-0.4-0.3-0.7-0.7-1 . 1 . Sensilla:
Macrotrichia on wing veins as follows: complete ventrally
on R4 and dorsally on R5; apicodorsally only on M„ M2, and
CuAl. Setiform groups on thoracic sclerites as follows: an-
terior pronotals apparently absent. Humeral callus with 4-5
small setae. Acrostical series short. Dorsocentral series ap-
parently absent. Supraalars few (6-7), restricted to posterior
and medial portions of sclerite. Prescutellars several (6). Scu-
tellars medium-sized, forming a dense group on outer comer,
numerous, similar, more widely spaced setae dispersed to-
ward the midline. Metapleurals absent. Suprametapleurals
few (3-4), small.

Terminalia (Figs. 4-5). Structure: Abdominal segment VIII
greatly reduced, mostly membranous; tergite consisting of a
short, medial, ligulate sclerite, stemite and pleurites undif-
ferentiated. Epandrium simple, emarginate posteromedially.
Xth tergite lobes well developed, prominent, parallel; inter-
lobular depression deep, U-shaped; individual lobe shape
quadrate, apex trilobate: outer sublobe acutely rounded; mid-
dle sublobe small, displaced toward inner comer, also slightly
angled inwardly; inner sublobe right-angled; inner margin

straight, paralleling outer and extending directly to base of
lobe; distal margin of lobe between outer and middle lobes
slightly concave; inner arm poorly sclerotized, elongate, apex
lobate, disjunct from same member opposite. Fused gono-
coxites and hypandrium well sclerotized, forming capsule
about as wide as long; posterolateral comers of gonocoxite
strongly produced. Outer gonostylus large (length about 0.7 x
mid-line length of hypandrium), an entire, subrectangular,
lobe with concave inner margin. Inner gonostylus a narrow,
porrect, smooth, simple, digitiform projection. Phallic com-
plex straight, supinate, not recurved. Aedeagal filaments equal,
with slightly flared apices, outer filaments strongly bowed
outward near bases; lateral tine longer than rods, broad
throughout, tapering slightly to complex apex; latter with
dorsal hook on inner wall; canal wide toward the base and
containing numerous long spiculae, arising from inner wall
and directed toward the apex; aperture subapical. Sperm pump
and piston poorly developed, former without internal spines,
latter with conspicuous, tubular ejaculatory atrium attached
to apodeme; apodeme a small, ventral, vertical flange. Sub-
anal pouch wide, bowl-shaped; tegmen broad, apex trilobate:
deeply incised on either side of strong, vertical, medial carina.
Gonite large, broad, ovate; gonocoxal lobe poorly developed,
posterior lobe elongate, anterior lobe expansive, ventral bridge
complete beneath basiphallus. Sensilla: Epandrium with nu-
merous, short to long setiforms generally, these shorter to-
ward the anterior. Xth tergite lobe with medium setiforms
generally and evenly spaced over central portion dorsally
these longer toward the apex; inner arm with few to several
(5-8), small setiforms near apex. Fused gonocoxites and hy-
pandrium ventrally with numerous, medium setiforms, these
spaced generally only over posterior half. Outer gonostylus
with very numerous, medium setiforms generally and evenly
spaced over outer surface and marginally, absent from mid-
dle of inner face. Epiproct with few (4-6) alveoliforms in
dorsolateral group. Hypoproct with few (3-4) long setiforms
apically.

Female (Fig. 3). Coloration and measurements from non-
topotypic material. Coloration. As in male, paired medial
longitudinal lines present under some illumination. Size. A
medium-sized Blepharicera. Measurements (N = 6): Wing
length 6.7 (6. 2-7. 6). Leg segment lengths:

foreleg

midleg

hindleg

femur

3.8 (3. 6-4.2)

3.8 (3.4-4. 2)

5.3 (4.9-5. 8)

tibia

3.2 (2. 9-3. 5)

3.2 (2.9-3. 5)

4.8 (4.4-5. 3)

tarsus 1

1.6 (1.5-1. 7)

1.5 (1.4-1. 7)

2.1 (1. 9-2.3)

2

0.78 (0.72-0.82)

0.77 (0.71-0.81)

0.72 (0.65-0.78)

3

0.49 (0.44-0.54)

0.51 (0.46-0.55)

0.45 (0.38-0.52)

4

0.34 (0.29-0.38)

0.32 (0.28-0.37)

0.32 (0.28-0.38)

5

0.40 (0.37-0.42)

0.40 (0.36-0.44)

0.40 (0.36-0.44)

Head (Fig. 3). Structure: Normal type, subholoptic. Su-
prafrons narrow, with a long, strongly convex carina. Clypeus
elongate L/W = 2.5. Eyes approximate dorsally, interocular
distance equal to combined diameters of 1-2 upper omma-
tidia; eye divided, upper division well differentiated from
lower (callis oculi very broad, anterior portion strongly striate),
approximately equal to lower in area, somewhat flattened

Contributions in Science, Number 377

Hogue and Georgian: New Appalachian Blepharicera 7

dorsally, 15-16 rows of ommatidia along mid-meridian; up-
per ommatidia much larger (2.3 x) than lower in diameter.
Parietal sclerite broad, trapezoidal in outline shape. Probos-
cis short, free portion about 0.6 x head width; mandibles
present and complete; palpus 5-segmented, distal 4 palpal
segment proportions 1.0-1. 0-1. 0-2. 4. Antenna 15-segment-
ed, flagellar segments narrowly elongate throughout, ultimate
longer (2.0 x ) than penultimate, apical 3 segment proportions
1.0-0. 9-1. 7. Sensilla: Setiform groups on head capsule as
follows: clypeals numerous (23-24), medium, only a few small
setae medially, these more numerous and longer along dis-
tolateral margins. A single, moderately strong seta over an-
tennal socket. Parietal setae several (5-7 usually, rarely more,
14-16). Postocellars few (2-3), small, lateral. Medioccipitals
absent. Supracervicals numerous (23-25), tiny. Occipitals
separated from and larger than postgenals, numerous (18),
long. Postgenals numerous (14-15), smaller toward center of
group.

Thorax and appendages. Structure: Wing venation as in
male. Tibial spurs 0-0-2 (inner twice length of outer). Hind
basitarsus short. Progressive leg segment proportions: foreleg
0.8-0. 5-0. 5-0. 6-0. 7-1 .2; midleg 0.8-0.5-0.5-0.7-0.6-1 .3;
hindleg 0.9-0. 4-0. 4-0. 6-0. 7-1 .3. Sensilla: Macrotrichia of
wing veins and setiform groups on thoracic sclerites as in
male.

Terminalia. Structure: Posterior margin of Vlllth stemite
lobe broadly bilobate, medial depression shallow; scleroti-
zation in base of latter rectangular (much wider than long).
Hypogynial plate subquadrate, base slightly wider than apex;
with weak transverse creases across base; apex broadly
rounded (outer angle not distant from level of tip). Accessory
gland elongate, posterior sides subparallel (not dilated an-
teriorly). Spermathecae 3 in number, equal in size and shape,
ovoid; necks very short; ducts completely unsclerotized. Sen-
silla: Medial group of Vlllth sternite with several to numer-
ous (7-20, usually more than 1 5), small setae; posterior lobe
with several (7-10), medium setae dispersed generally. Short,
stout apicodorsal setae of hypogynial lobe several to nu-
merous (9-20). Epiproct with 2 apical setae. Alveoliforms of
hypoproct few (4-6).

Pupa (Figs. 17, 21). Integument. Dorsum well sclerotized.
Pleural margins not sclerotized ventrally. Frontal, scutal,
branchial, and alar sclerites smooth, completely without pa-
pillae. Metascutal (except lateral third), scutellar, and ab-
dominal tergites densely papillose. Individual papillae
smooth, rounded, oval convexities. Pattern disperse, papillae
unevenly spaced, often arranged in pairs or diads (some con-
fluent), both in central area of abdominal tergites and areas
lateral to muscle scars. Cuticle between papillae uniform in
structure. Coloration. Surface reflection shiny; pigmentation
more or less even throughout. Size. Medium. Measurements,
male (N = 48): body length 4.1 (3. 5-4. 7), width 2.5 (1.9-
2.7); female (N = 26): body length 4.7 (3. 9-5. 2), width 2.9
(2. 3-3. 2); male about 0.7 x size of female. Structure. Outline
shape almost ovoid, L/W male = 1.7, female = 1.6; cross
section convex, sides declivous all around. Dorsal sclerites:
abdominal tergite margins symmetrically convex, of II and

III slightly wider than thorax, equally projecting all around.
Branchial sclerite smoothly curving. Ventral sclerites: anten-
nal case extending well beyond base of wing case in both
sexes (about 0.3 x length), apex straight. Apices of leg cases
in male coterminate; in female tip of hindleg most extended,
foreleg and midleg much less, these almost coterminate.
Mandibular case small in male, much longer and fuller in
female. Branchiae: medium, erect, approximate, parallel,
projecting forward just short of plane of anterior margin.
Plates of each branchia lobate, rigid, parallel, inner two slight-
ly smaller and thinner than outers, slightly spreading; indi-
vidual plates angular seculate in outline.

Larva (Fig. 13). Integument. Dorsally with distinct, but
fine, linear and zigzagging corrugations, these absent ven-
trally. Coloration. Trunk evenly pigmented, medium-brown,
sclerotized portions dark brown to black. Size. Medium for
the group. Measurements (N = 20): body length 5.9 (4.7-
7.1), head capsule width 1.24 (1.14-1.41), antennal segment
lengths, basal 0. 12 (0. 1 1-0. 1 5), apical 0. 18 (0. 1 7-0.21). Gen-
eral shape. Cylindrical. Head. Antenna short, 2-segmented,
intersegmental membrane wide; segment proportions 1.0-
1.45. Trunk. Structure: Anterior division spheroid. Lateral
margins of abdominal segments inclinous, truncate (pseu-
dopods extending well beyond); dorsopseudopodal lobes small
and poorly developed. Anal division trilobate: lateroterminal
lobe acutely rounded; pleuron straight, medioterminal lobe
truncate posteriorly; terminal incision shallow, widely
V-shaped. Sensilla: Primary trunk sensilla: tP submedial, tae-
niaform, tM-T submedial, small, taeniaform, tl-VII sub-
medial, taeniaform, stP lateromedial, taeniaform, stM-T
lateromedial, elongo-elliptoid, stl lateromedial, elongo-fus-
tiform, stll-VII lateromedial, large, setiform. Inner tpP far-
lateromedial, disjunct from stP, setiform, inner tpI-VII far-
mediolateral, disjunct from and anterior to stl-VII, large,
setiform. Outer tpI-VII near and slightly lateroposterior to
inner tp, smaller, setiform. pdpodl-VII indiscernible. Inner
and outer dpodM-T proximate, those of M sublateral, of T
mediolateral, setiform, dpodl-VII on minute, anterolateral
tubercle (= dorsopseudopodal lobe), setiform, inner and out-
er dpodVIII contiguous, at apex of lateroterminal lobe, se-
tiform. ssP slightly lateral to ssM-T, decidedly larger than
latter. Dorsal secondary trunk sensilla: numerous, more or
less generally distributed, usually all setiform with a few tae-
niaform to elongo-fustiform types intermingled. Terminal
setae 3-3, marginal.

VARIATION. Larval specimens are common in which a
few or most of the central dorsal sensilla are modified (elon-
go-fustiform, elongo-pyriform to subclaviform, larval type
“C,” of Hogue, 1978:fig. 35). These are probably referable
to appalachiae and some are confirmable as that species by
the presence of pharate pupal structure observable beneath
the cuticle. It might be theorized that these represent indi-
viduals in which introgression is occurring from other re-
gional species which normally have modified setae ( william -
sae, coweetae ). Further analysis of large samples and reared
material are needed to explain these chaetotactic variations.

SPECIMENS EXAMINED. Types. HOLOTYPE male

8 Contributions in Science, Number 377

Hogue and Georgian: New Appalachian Blepharicera

(extracted from pupal skin), in alcohol with skin (terminalia
and head dissected and mounted on slides nos. CLH 84-35
and 84-94, respectively): NORTH CAROLINA, Macon
County. Coweeta Hydrologic Laboratory, Lower Shope Fork,
9 May 1981, T. Georgian (LACM).

ALLOTYPE female (extracted from pupal skin), in alcohol
with skin (terminalia dissected and mounted on slide no.
CLH 84-33): same locality and collector as holotype, 28 May
1982 (LACM).

Eight PARATYPE males and females (extracted from pu-
pal skins) on slides and in alcohol, with skins: same locality
and collector as holotype, 9 May 1981,28 May 1982 (LACM,
UGAM, USNM).

Additional specimens. GEORGIA, Rabun County. Betty’s
Creek: 23 March 1 984, T. Georgian (37 larvae); 7 April 1 984,
T. Georgian (22 larvae, 6 pupae).

MARYLAND, Frederick County. Little Catoctin Creek, 1
mi. N Harmony, 30 May 1958, P.H. FreytagU pupa: OSU).
0.5 mi. E Yellow Springs, 31 May 1958, P.H. Freytag (1
larva, 8 pupae: OSU). Cunningham Falls State Park, Little
Hunting Creek, 17 May 1984, R.W. Baumann and C.R. Nel-
son (2 larvae, 10 pupae: BYU).

NORTH CAROLINA, Macon County. Clear Creek, 3200
ft., Highlands, 2 July 1958, J.G. Franclemont (2 females:
CU). Highlands, 3000-5000 ft.. May 1936, R.C. Shannon (3
males: USNM). Coweeta Hydrologic Laboratory, Lower
Shope Fork, 2300 ft., T. Georgian: 9 May 1981 (23 larvae,
5 pupae); 24 April 1982 (34 larvae, 29 pupae); 28 May 1982
(10 pupae); April 1984 (1 1 larvae). Dryman’s Fork, 2100 ft.,
T. Georgian: 1 April 1980 (9 larvae); 21 April 1980 (25
larvae, 5 pupae); 21 April 1980 (3 larvae, 2 pupae). Swain
County. Deep Creek, Great Smoky Mountains National Park,
Bryson City, Deep Creek Campground, 21 May 1970, G.B.
Wiggins and T. Yamamoto (1 pupa, 2 males: ROM).

TENNESSEE, Sevier County. Great Smoky Mountains
National Park, Greenbrier Cove, 2000 ft., 22 May 1938,
Williams (1 male: USNM). Great Smoky Mountains Na-
tional Park, 18 May 1957, J.R. Vockeroth (1 male: CNC).

VIRGINIA, Bath County. Blowing Springs Camp, 8 mi.
W Warm Springs, 18-20 May 1963, C.M. and O.S. Flint (2
males, 1 female: USNM). Wilson Creek, Route 629, 12 May
1979, C.R. Parker (3 males, 1 female: BCK). Giles County.
Stoney Creek, 1800-2000 ft., 26 May 1962, J.R. Vockeroth
(1 larva, 33 pupae: CNC). Sinking Creek, 29 May 1941, A.
Stone (1 pupa: USNM). Madison County. Shenandoah Na-
tional Park, White Oak Creek, 24 June 1951, B.D. Burks (4
larvae, 5 pupae: USNM). Nelson County. Crabtree Creek,
Crabtree Falls, Highway 56, 28 May 1984, R.W. Baumann
and C.R. Nelson (2 males, 1 female: BYU). Page County.
Luray, 21-24 June 1933, A. Melander (2 males, 3 females:
USNM). Patrick County. Confluence of Little Rock Castle
Creek and Rock Castle Creek, Route 605, 31 May 1980, B.
Kondratieff ( 1 male, 2 females: BCK). Rockbridge County.
Pedlar River, below Panther Falls, 28 May 1984, R.W. Bau-
mann and C.R. Nelson (9 males, 4 pupae: BYU). Shenan-
doah County. Passage Creek, Camp Roosevelt, 21 May 1973,
S. Fiance (3 larvae: USNM). Smyth County. N Fork Holston

River, Route 42, junction Route 633, 9 May 1981, B. Kon-
dratieff(l male: BCK). Washington County. S Fork Holston
River, Highway 58, near Damascus, 27 May 1984, R. Bau-
mann and C.R. Nelson (2 pupae: BYU).

ETYMOLOGY. The name is considered a noun in the
genitive case and refers to the southern Appalachian Moun-
tains where the species is of wide occurrence.

DISTRIBUTION. Blepharicera appalachiae is wide-
spread in most major drainages throughout the central and
southern Appalachian Mountains (Fig. 26). It is replaced
largely by tenuipes to the north, except for a zone of overlap
in Maryland and Virginia (to Giles County). Its distribution
southward ends in northern Georgia (Rabun County), but it
has not been found in the Savannah River drainage, although
it is present less than 1 7 km away in the southernmost portion
of the Little Tennessee system.

ECOLOGICAL NOTES. The larval ecology of B. appa-
lachiae is similar to that of other members of the family (cf.
Hogue, 1981). The stream at the type locality is a fourth-
order mountain stream with average width 5.5 m, average
velocity 50 cm/sec, and a bed predominantly of schist and
quartzite boulders averaging 21 cm in greatest dimension.
At least three other blepharicerid species (B. coweetae, sim-
ilans, and williamsae) are also present. Some habitat segre-
gation based on rock size and water depth has been observed
(Georgian, unpubl. data), but individuals of more than one
species have been collected from single rocks. The four larval
instars can be identified by measuring the width of the dorsal
head sclerite, permitting analysis of life cycles (Georgian and
Wallace, 1983). Of the four species present at the Coweeta
Hydrologic Laboratory, B. williamsae is the most abundant
and the earliest to appear. Larvae are first seen in early winter
(first instars December-March, mature larvae February-April
or early May); they pupate in early spring (April-May). B.
appalachiae and coweetae complete their larval development
from early April through the end of May, and are much less
abundant than williamsae. The last species to become evi-
dent in the annual cycle is B. similans, noticeable larvae
appearing in early May, with pupae present throughout the
summer until mid-September. The rate of growth and timing
of life cycles changes with elevation (and is presumably con-
trolled by correlated temperatures). At a site near the type
locality only 490 m higher (Coweeta Hydrologic Laboratory,
Watershed #27, 1158 m), mature larvae and pupae of B.
williamsae were found and other species were absent as late
as the first of June.

Blepharicera coweetae , new species

Figures 6-9, 14, 18, 22, 26

DIAGNOSIS. B. coweetae differs from other members of
the B. tenuipes group in the unique shape of the Xth tergite
lobe: outer comer acutely rounded, followed medially by a
shallow concavity and central sublobe displaced toward the
midline; the inner margin is straight but angled toward the
midline so that the interlobular depression is V-shaped. The
larva is most similar to that of B. tenuipes but is smaller,

Contributions in Science, Number 377

Hogue and Georgian: New Appalachian Blepharicera 9

Figures 6-9. Blepharicera coweetae, new species. 6. Male head. 7. Female head. 8. Male terminalia, external structures (dorsal view). 9.
Male terminalia, phallic structures (dorsal view).

10 Contributions in Science, Number 377

Hogue and Georgian: New Appalachian Blepharicera

evenly pigmented, and has disperse, longer, predominantly
fustiform dorsal sensilla whereas this stage of that species is
much larger, usually with longitudinal light bands in the
otherwise dark pigmentation and has short claviform dorsal
sensilla in series. The pupa displays a unique pattern of light
lines in the generally darkly pigmented integument, com-
posed of numerous, very fine, closely set, longitudinal, nar-
rowly wavy, transparent furrows and short, transverse “hy-
phens,” the former appearing macroscopically as strong striae.

DESCRIPTION. Adult a small, sturdily built, well-scler-
otized blepharicerid. Very similar to B. temiipes and appa-
lachiae.

Male (Figs. 6, 8-9). Only a single, freshly emerged, alcohol-
preserved specimen (holotype) available for coloration and
measurements; character values of latter may be insufficient
due to incomplete development. Other structural characters
from pharate adults dissected from pupae. Coloration. Gen-
erally dull gray-brown, pruinose. Mesoscutum unicolorous,
dull gray, comers of posterior pronotum and scutellum con-
trasting light brown, latter becoming darker toward the sides;
pleurites similar to scutum but paler. Wing membrane com-
pletely hyaline. Size. A medium-sized B/epharicera. Mea-
surements (N = 1): Wing length 6.1. Leg segment lengths:

foreleg

midleg

hindleg

femur

3.6

3.9

5.2

tibia

3.2

3.0

4.7

tarsus 1

1.8

1.6

1.8

2

0.91

0.83

0.66

3

0.64

0.60

0.44

4

0.33

0.32

0.28

5

0.32

0.33

0.32

Head (Fig. 6). Structure: Normal type, subholoptic. Su-
prafrons narrow, with long, moderately convex carina. Clyp-
eus elongate, L/W = 2.0. Eyes approximate dorsally, inter-
ocular distance equal to combined diameters of 2.8 upper
ommatidia; eye divided, upper division well differentiated
from lower (callis oculi narrow), approximately equal to low-
er in area, 19-20 rows of ommatidia along mid-meridian;
upper ommatidia slightly larger (1.5 x) than lower in di-
ameter. Proboscis short, free portion about 0.4 x head width;
mandibles completely absent; palpus 5-segmented, distal 4
palpal segment proportions 1.0-1. 0-1. 0-3.0. Antenna 15-
segmented, flagellar segments elongate throughout, ultimate
longer ( 1 .4 x ) than penultimate, apical 3 segment proportions
1.0-0.9-1.25. Sensilla: Setiform groups on head capsule as
follows: clypeals several (9-10), medium, only a few, small
setae medially, more numerous and longer along distolateral
margins. No single, medium strong seta over antennal socket;
parietal setae few (1-2). Postocellars few (2-5), small, lateral.
Medioccipitals absent. Supracervicals very numerous (30),
tiny. Occipitals separated from and larger than postgenals,
numerous (26-30) and long; upper series of longer and heavi-
er setae than lower. Postgenals numerous (12-26), small to-
ward center of group.

Thorax and appendages. Structure: Typical wing venation
for B/epharicera. Tibial spurs 0-0- 1 . Progressive leg-segment
proportions: foreleg 0.9-0. 6-0. 5-0. 7-0. 5-1.0; midleg 0.8-0. 5-

0.5-0. 7-0. 5-1 .0; hindleg 0. 9-0.4-0. 4-0. 7-0.6- 1 . 1 . Sensilla:
Macrotrichia on wing veins as follows: complete ventrally
on R4 and dorsally on R5; apicodorsally only on M„ M2, and
CuAl. Setiform groups on thoracic sclerites as follows: an-
terior pronotals absent. Humeral callus with 3 very small
setae. Acrostical series short, divergent anteriorly. Dorso-
central series complete. Supraalars few (5), restricted to pos-
terior and medial portions of sclerite. Prescutellars absent.
Scutellars medium-sized, forming a dense group on outer
comer; numerous similar, widely spaced setae dispersed to-
ward the midline. Metapleurals present, few (2-6). Supra-
metapleurals 2-3, minute.

Terminalia (Figs. 8-9). Structure: Abdominal segment VIII
greatly reduced, mostly membranous; tergite consisting only
of a short, medial, ligulate sclerite; pleurites and stemite
undifferentiated. Epandrium simple, emarginate postero-
medially. Xth lobes well developed, prominent, slightly di-
vergent; interlobular depression deep, V-shaped; individual
lobe quadrate, apex trilobate: outer sublobe acutely rounded,
middle subsublobe small, central, angled inwardly, inner sub-
lobe obtusely angled; inner margin straight, extending me-
dially and directly to base of opposite margin; distal margin
of lobe between outer and middle lobes slightly concave;
inner arm poorly sclerotized, elongate, apex lobate, disjunct
from same member opposite. Fused gonocoxites and hypan-
drium well sclerotized, forming capsule about as wide as long;
posterolateral corners of gonocoxite strongly produced. Outer
gonostylus moderately large (length about 0.7 x midline length
of hypandrium), an entire, subrectangular, lobe with concave
inner margin. Inner gonostylus a narrow, porrect, smooth,
simple, digitiform projection. Phallic complex straight, su-
pinate, not recurved. Aedeagal filaments equal, with slightly
flared apices, outer filaments strongly bowed near bases; lat-
eral tine longer than rods, broad throughout, tapering slightly
to complex apex; latter with dorsal hook on inner wall; canal
wide toward the base and containing numerous long spiculae
arising from inner wall and directed toward the apex; aperture
subapical. Sperm pump and piston poorly developed, former
without internal spines, latter with conspicuous, tubular, dor-
sal ejaculatory atrium attached to apodeme; apodeme a small,
ventral, vertical flange. Subanal pouch wide, bowl-shaped;
tegmen broad, apex trilobate, i.e., deeply incised on either
side of strong, vertical, medial carina. Gonite large, broad,
ovate; gonocoxal lobe poorly developed, posterior lobe elon-
gate, anterior lobe expansive, ventral bridge complete be-
neath basiphallus. Sensilla: Epandrium with very numerous,
short (toward anterior) to long (toward posterior) setiforms
generally. Xth tergite lobe with medium setiforms generally
and evenly spaced over central portion dorsally, longer to-
ward apex; inner arm with few to several (5-8), small seti-
forms near the apex. Fused gonocoxites and hypandrium
ventrally with numerous, medium setiforms generally spaced
only over posterior half. Outer gonostylus with very numer-
ous, medium setiforms generally and evenly spaced over
outer surface and marginally, absent from middle of inner
face. Epiproct with few (4-8) alveoliforms in dorsolateral
group. Hypoproct with few (2) long setiforms apically.

Contributions in Science, Number 377

Hogue and Georgian: New Appalachian Blepharicera 1 1

Female (Fig. 7). Only a single, freshly emerged, alcohol-
preserved specimen (allotype) available for coloration and
measurements. Structural characters from pharate adults dis-
sected from pupae. Coloration. Apparently as in male. Size.
A medium-sized Blepharicera. Measurements (N = 1): Wing
length 6.3. Leg segment lengths:

foreleg

midleg

hindleg

femur

3.7

3.6

5.2

tibia

3.1

2.9

4.3

tarsus 1

1.4

1.4

1.9

2

0.67

0.66

0.64

3

0.42

0.41

0.38

4

0.30

0.29

0.32

5

0.44

0.43

0.38

Head (Fig. 7). Structure: Normal type, subholoptic. Su-
prafrons narrow, with a long, strongly convex carina. Clypeus
elongate L/W = 2.3. Eyes approximate dorsally, interocular
distance equal to combined diameters of 1-2 upper omma-
tidia; upper division well differentiated from lower (callis
oculi very broad, anterior portion strongly striate), approx-
imately equal to lower in area, somewhat flattened dorsally,

1 7 rows of ommatidia along mid-meridian; upper ommatidia
larger (2.5 x) than lower in diameter. Parietal sclerite broad,
trapezoidal in outline. Proboscis short, free portion about
0.4 x head width; mandibles present and complete; palpus
5-segmented, distal 4 palpal segment proportions 1 .0-1.0-
1.0-3. 4. Antenna 1 5-segmented, flagellar segments narrowly
elongate throughout, ultimate longer (1.8 x) than penulti-
mate, apical 3 segment proportions 1.0-0. 9-1. 7. Sensilla:
Setiform groups on head capsule as follows: clypeals nu-
merous to very numerous (26-45), general basally, more nu-
merous and longer along distolateral margins. Seta over an-
tennal socket. Parietal setae numerous to very numerous ( 1 8-
38). Postocellars few (3-5), small, lateral. Medioccipitals ab-
sent. Supracervicals numerous (20), tiny. Occipitals separat-
ed from and larger than postgenals, numerous (32), long,
uppers longer and heavier than lowers. Postgenals numerous
(28-30), smaller toward center of group.

Thorax and appendages. Structure: Wing venation as in
male. Tibial spurs 0-0-2 (inner twice length of outer). Hind
basitarsus short. Progressive leg segment proportions: foreleg
0.8-0. 5-0. 5-0. 6-0. 7-1 .5; midleg 0.8-0. 5-0. 5-0. 6-0.7-1 .5;
hindleg 0.8-0. 4-0. 3-0. 6-0. 8-1 .2. Sensilla: Macrotrichia of
wing veins and setiform groups on thoracic sclerites as in
male.

Terminalia. Structure: Posterior margin of VUIth stemite
lobe broadly bilobate, medial depression shallow; scleroti-
zation in base of latter rectangular (much wider than long).
Hypogynial plate subhexagonal, base slightly wider than apex;
with weak transverse creases across base; apex acutely round-
ed (outer angle far from level of tip). Accessory gland elon-
gate, decidedly dilated anteriorly. Spermathecae 3 in number,
equal in size and shape, ovoid; necks very short to absent;
ducts completely unsclerotized. Sensilla: Medial group of
VUIth stermte with few (3) or numerous (12-18) small setae;
posterior lobe with few to numerous (2-12), medium-sized

setae restricted to outer one-third. Short, stout apicodorsal
setae of hypogynial lobe usually numerous ( 1 2-19). Epiproct
with 1-2 apical setae. Alveoliforms of hypoproct few (3-6).

Pupa (Figs. 18, 22). Integument. Dorsum well sclerotized.
Pleural margins not sclerotized ventrally. Frontal, scutal,
branchial, and alar sclerites smooth, completely without pa-
pillae. Metascutal (except lateral one-third), scutellar, and
abdominal tergites densely papillose. Individual papillae finely
spiculate, rounded, oval convexities. Pattern on abdominal
tergites disperse, papillae more or less evenly spaced, slightly
more dense toward the middle of sclerites, absent laterally.
Cuticular structure between papillae irregular, thin and
grooved areas (corresponding to pigmentary pattern as de-
scribed below). Coloration. Surface reflection shiny; dark pig-
mentation broken by numerous, very fine, close-set, longi-
tudinal, narrowly wavy, transparent furrows or lines and short,
transverse “hyphens” between, the former appearing mac-
roscopically as strong striae. Size. Medium. Measurements,
male (N = 27): body length 4.5 (3. 8-5.0), width 2.5 (1.9-
2.8); female (N = 27): body length 5.2 (4. 7-5. 7), width 2.8
(2.5-3. 1); male about 0.8 x size of female. Structure. Outline
shape elongate, L/W male = 1.8, female = 1.9; cross section
convex, sides declivous all around. Dorsal sclerites: lateral
margins of abdominal segments convex, of II and III slightly
wider than thorax, of IV often projecting slightly. Branchial
sclerite smoothly curving. Ventral sclerites: antennal case
extending well beyond base of wing case in both sexes (about
0.3 x length), apex straight. Apices of leg cases in male co-
terminate; in female tip of hindleg most extended, foreleg
and midleg much less, these almost coterminate. Mandibular
case small in male, much longer and fuller in female. Bran-
chiae: medium, erect, approximate, parallel, projecting for-
ward just short of plane of anterior margin. Plates of each
branchia lobate, rigid, parallel, inner two plates slightly smaller
and thinner than outers, slightly spreading; individual plates
angularly seculate in outline.

Larva (Fig. 14). Similar to tenuipes. Integument. Dorsally
with distinct, but fine, mostly linear, corrugations, these ab-
sent ventrally. Coloration. Trunk evenly pigmented, medium
brown, sclerotized portions dark brown to black. Size. Me-
dium for the group. Measurements (N = 4): Body length (N =
3), 5.4 (5. 3-5. 6), antennal segment lengths, basal 0.16 (0.13-
0.17), apical 0.19 (0.18-0.20). General shape. Cylindrical.
Head. Antenna short, 2-segmented, intersegmental mem-
brane wide; segment proportions 1.0-1. 2. Trunk. Structure:
Anterior division spheroid. Lateral margins of abdominal
segments inclinous, truncate (pseudopods extending well be-
yond); dorsopseudopodal lobes small and poorly developed.
Anal division trilobate: lateroterminal lobe acutely rounded;
pleuron slightly convex, medioterminal lobe convex poste-
riorly, evenly rounded; terminal incision deep, V-shaped.
Sensilla: Primary trunk sensilla: tP submedial, elongo-ellip-
toid, tM-T submedial, small, taeniaform, tl-VII submedial,
lanciform, obscured by multiplication. stP lateromedial,
taeniaform to setiform, stM-T lateromedial, elongo-ellip-
toid, stl lateromedial, acutiform, stll-VII lateromedial, se-
tiform. Inner tpP far-lateromedial, disjunct from stP, seti-

12 Contributions in Science, Number 377

Hogue and Georgian: New Appalachian Blepharicera

Figures 10-12. B. tenuipes. 10. Larva (dorsum left, venter right). 11. Detail of dorsum of second abdominal segment of larva. 12. Pupa
(dorsum left, venter right).

Contributions in Science, Number 377

Hogue and Georgian: New Appalachian Biepharicera 13

2 mm

1

form, inner tpI-VII far-mediolateral, disjunct from and
anterior to stl-VII, setiform. Outer tpI-VII near and slightly
posterior to inner tp, lanciform to elongo-elliptoid. pdpodl-
VII indiscernible. Inner and outer dpodM-T proximate, those
of M sublateral, of T mediolateral, setiform, dpodl-VII on
minute, anterolateral tubercle (= dorsopseudopodal lobe),
setiform, inner and outer dpodVIII contiguous, at apex of
lateroterminal lobe, setiform. ssP slightly lateral to ssM-T,
decidedly larger than latter. Dorsal secondary trunk sensilla:
numerous, more or less disperse but tending to be in two
broad, transverse series medially, these merging laterally.
Almost all elongo-elliptoid to elongo-claviform or lanciform,

these short medially, becoming fairly long laterally; a few
long setiforms intermingled laterally. Terminal setae 3-3,
marginal.

SPECIMENS EXAMINED. Types. HOLOTYPE male
(emerged from pupa), in alcohol with pupal skin (terminalia
not dissected): NORTH CAROLINA, Macon County. Co-
weeta Hydrologic Laboratory, Lower Shope Fork, T. Geor-
gian. Individually reared: collected 7 April 1984 as mature
larva; pupated 13 April; emerged 5 May 1984 (LACM).

ALLOTYPE female (emerged from pupa), in alcohol with
pupal skin (terminalia dissected and mounted on slide CLH
84-71). GEORGIA, Rabun County. Betty’s Creek, T. Geor-

14 Contributions in Science, Number 377

Hogue and Georgian: New Appalachian Blepharicera

E

£

CM

E

E

CM

Figures 17-20.

Pupae, dorsal/ventral aspects. 17. B. appalachiae. 18. B. coweetae. 19. B. similans. 20. B. williamsae.

Contributions in Science, Number 377

Hogue and Georgian: New Appalachian Blepharicera 15

gian. Individually reared: collected 7 April 1984 as mature
larva; pupated 16 April 1984; emerged 12 May 1984(LACM).

Ten PARATYPE males, 5 PARATYPE females (extracted
from pupal skins), variously dissected and mounted on slides
no. CLH (males) 81-8, 84-1, 84-2, 84-39, 84-43, 84-44, 84-
45, 84-46, 84-83a-d, 84-84a-d, (females) 84-47, 84-48, 84-
49, 84-50, and 84-5 1 . Same locality as holotype, 9 May 1981,
T. Georgian (LACM, UGAM, USNM).

Additional specimens. NORTH CAROLINA, Macon
County. Coweeta Hydrologic Laboratory, Lower Shope Fork,
2300 ft., 9 May 1981 (66 pupae); 24 April 1982 (9 larvae,
15 pupae). Dryman’s Fork, 2100 ft., 21 April 1980 (26 pu-
pae).

ETYMOLOGY. This species is named for Coweeta Creek
in whose tributaries it was first found. “Coweeta” is treated
as a noun in the genitive case. It is a Cherokee word associated
with a major tribal settlement in the area, near Franklin,
North Carolina. It has more recently been used by the U.S.
Forest Service for its Coweeta Hydrologic Laboratory, also
the type locality of the insect.

DISTRIBUTION. B. coweetaeis known to date only from
a roughly circular area about 1 7 km in diameter in the south-
ern Appalachians, including the type locality and two other
streams (Dryman’s Fork, Macon County, North Carolina and
Betty’s Creek, Rabun County, Georgia), all within the Little
Tennessee River drainage (Fig. 26).

ECOLOGICAL NOTES. The species is uncommon at the
known sites of occurrence and little is known concerning its
biology (See Ecological Notes under B. appalachiae).

LARVAE AND PUPAE OF OTHER SPECIES

Detailed descriptions of the larvae of the other species are
being published by Hogue (in press). The following will verify
the provisional identifications made by Hogue (1978), to
which reference can be made for brief descriptions and il-
lustrations. Minimal diagnostic features only are repeated
here for clarification (see also the keys above). Diagnostic
features consist primarily of the shape and distribution of
the modified dorsal sensilla. The pupae are described in full
and figured.

The immatures of B. cherokea, diminutiva, and capitata
are still unknown.

Blepharicera similans Johannsen, 1929

Figures 15, 19, 23

Larva “B,” Hogue, 1978:27, fig. 33.

The dorsal modified sensilla are large, of various sizes, den-
tate and basically short capitate to almost spherical in shape.
They are arranged generally in two transverse rows, a longer
series between the large subtergal setiform sensilla and a

shorter anteromedial row; a few also are associated with the
large, inner, tergopleural setiform sensilla.

Pupa (Figs. 19, 23). Integument. Dorsum well sclerotized.
Pleural margins not sclerotized ventrally. Frontal, scutal,
branchial, and alar sclerites smooth, without papillae. Meta-
scutal (medially only), scutellar, and abdominal tergites mod-
erately densely papillate. Individual papillae smooth, round-
ed, oval convexities. Pattern on abdominal tergites disperse,
papillae more or less evenly spaced, slightly more dense me-
dially, absent far laterally. Cuticle between papillae uniform
in structure. Coloration. Surface reflection shiny; pigmen-
tation even. Size. Medium. Measurements (northern mate-
rial, specimens from North Carolina, Shope Creek distinctly
smaller, 0.7), male (N = 10): body length 4. 1 (3.8-4. 5), width
2.4 (2. 2-2. 8); female (N = 10): body length 5.0 (4. 4-5. 3),
width 2.9 (2. 5-3. 3); male about 0.7 x size of female. Struc-
ture. Outline shape almost ovoid, L/W male = 1.4, female =
1 .4. Cross section convex, sides declivous all around. Dorsal
sclerites: lateral margins of abdominal segments symmetri-
cally convex, those of II and III slightly wider than thorax,
equally projecting all around. Branchial sclerite smoothly
curving. Ventral sclerites: antennal case extending well be-
yond base of wing case in both sexes (about 0.3 x length),
apex straight. Apices of leg cases in male coterminate; in
female tip of hindleg most extended, midleg less, hindleg less
than either; mandibular case small in male, much longer and
fuller in female. Branchiae: medium, erect, approximate, par-
allel, projecting forward just short of plane of anterior margin.
Plates of each branchia lobate, rigid, parallel, inner two slight-
ly smaller and thinner than outers, slightly spreading; indi-
vidual plates angular seculate in outline.

Blepharicera tenuipes (Walker, 1 848)

Figures 10-12, 24

Larva “A” Hogue, 1978:27, fig. 30.

The dorsal modified sensilla are short, elongo-claviform. They
are arranged in two irregular transverse rows, a longer pos-
teromedial series including the subtergal sensilla, and a short-
er anteromedial row. Laterally the sensilla are mostly seti-
form (Figs. 10-11).

Pupa (Figs. 1 2, 24). Integument. Dorsum well sclerotized.
Pleural margins not sclerotized ventrally. Frontal, scutal,
branchial, and alar sclerites smooth, completely without pa-
pillae. Metascutal (except lateral one-third), scutellar, and
abdominal tergites moderately densely papillate. Individual
papillae smooth, rounded, oval convexities. Pattern disperse,
unevenly spaced, papillae often arranged in pairs but not in
fused diads, mainly in areas of abdominal tergites lateral to
muscle scars, more or less evenly spaced in medial area be-
tween scars; slightly more dense medially, absent laterally.

Figures 21-25. Pupae, Blepharicera tenuipes group. Photomicrographs of tergites of abdominal segment II, showing arrangements of integ-
umentary papillae. Dorsomedial third of sclerite at right, detail of area lateral to major muscle scar at left. 21. B. appalachiae. 22. B. coweetae.
23. B. similans. 24. B. tenuipes. 25. B. williamsae.

16 Contributions in Science, Number 377

Hogue and Georgian: New Appalachian Blepharicera

21

22

23

24

- • © ' .

Q

\©

©

:«l

©

*' <i

© «

1

I'my I

■ $ i ,: ■ “ #

• $

0

^ &

©

0

©

©

i .g •

' % \ *• *

■*?'*-<• *

-■ ,*JV ‘

!• ¥ \ * f

*•• *

shhhbsImBHSb

V)

_ 8

P • «» 9

-*

9-

9

• «

«, •

&

1

*

<*>

^ fT> .>

<0 3 1

9 • #«

•# • f >
f# %r>i

<* 9

m t

»•- »

©

©©

©

©

£

©

9

. *

a

<s>

A *>

0

©

« •«>

©Q 5

>m

% .;• ©

' O -:■

l •

•3

0

«

^ 0

0 .. €

i? s~ ° °°

I & O cTsi^ (XO

s * lb ^ 4#>° ' ° cO Oe

°°- o ®0^ «*T

~ Oof

*o ® %

0 * .. <?>

0* «S* **

00

•.« . \ a ■■ .

B; V.*. •- . •

. • / v : * . :.\ :

*■> • • /♦.. v ..•,«.*•

» ♦ « . . , - i_

O - ast 3 ~© »

% ou a Q 9 ®>- . ~

-, 3. 3 ■» J O 0

a J * *,

. a

A--* g>

®0 03 *

a j ° G J3

- o

00 o «j

V'

^ "a

■ ?ff,/

J*

(9 %'./ v ...

”/sv

« 0 3 (9 ® ..,

i ■# - 0 ,a jj> ©

© ® j J ® _ i

O © ® 0 G #

O Oo ^ 0| ? >

i5 ©

t d)k

,j

* s-

t)

■v&f* °K »5 Jj : “ %n . . v

iS-a

Contributions in Science, Number 377

Hogue and Georgian: New Appalachian Blepharicera 17

Figure 26. Drainage systems of eastern United States and Canada showing distribution of the Blepharicera appalachiae (solid circles) and
coweetae (open circle). Shaded area represents range of Blepharicera tenuipes. Location of Coweeta Hydrologic Laboratory area indicated by
open circle.

18 Contributions in Science, Number 377

Hogue and Georgian: New Appalachian Blepharicera

Cuticle between papillae uniform in structure. Coloration.
Surface reflection shiny; pigmentation even. Size. Medium.
Measurements, male (N = 12): body length 4.6 (4. 1-4.9),
width 2.5 (2. 3-2. 8); female (N = 14): body length 5.5 (5.3-
5.8), width 3.2 (3.0-3. 5); male about 0.6 x size of female.
Structure. Outline shape almost ovoid, L/W male = 1.8, fe-
male = 1.8. Cross section convex, sides declivous all around.
Dorsal sclerites: lateral margins of abdominal segments sym-
metrically convex, of II and III slightly wider than thorax,
equally projecting all around. Branchial sclerite smoothly
curving. Ventral sclerites: antennal case extending well be-
yond base of wing case in both sexes (about 0.3 x length),
apex straight. Apices of leg cases in male coterminate; in
female tip of hindleg most extended, foreleg and midleg much
less, these almost coterminate. Mandibular case small in male,
much longer and fuller in female. Branchiae: medium, erect,
approximate, parallel, projecting forward just short of plane
of anterior margin. Plates of each branchia lobate, rigid, par-
allel, inner two slightly smaller and thinner than outers, slightly
spreading; individual plates angular seculate in outline.

Blepharicera willia msae Alexander, 1953

Figures 20, 25

Larva “D” Hogue, 1978:29, fig. 31.

The larva of this species is extraordinarily large for the B.
tenuipes group (mean body length 8.6 mm, range extremes
7.2-10.5 mm). The integument is also pigmented unevenly,
i.e., dark dorsomedially, light cream laterally, presenting a
longitudinally striped appearance. The dorsal modified sen-
silla are short, oviform, elliptoid or pyriform and more or
less densely distributed evenly over the entire dorsal surface
(slightly more dense laterally) (Fig. 16).

Pupa (Figs. 20, 25). Integument. Dorsum well sclerotized.
Pleural margins not sclerotized ventrally. Frontal, scutal,
branchial, and alar sclerites smooth, completely without pa-
pillae. Metascutal (except lateral half), scutellar, and abdom-
inal tergites densely papillose. Individual papillae finely spic-
ulate, small, rounded, oval convexities; pattern on abdominal
tergites general, papillae more or less evenly but closely spaced,
slightly more dense medially, absent marginally. Cuticle be-
tween papillae finely reticulate. Coloration. Surface reflection
dull; pigmentation irrorate, in concordance with reticulate
structure. Size. Medium. Measurements, male (N = 10): body
length 5.5 (5. 1-5.9), width 3.1 (2. 8-3. 4); female (N = 10):
body length 6.7 (6. 1-7.5), width 3.7 (3.4-4. 1); male about
0.8 x size of female. Structure. Outline shape almost ovoid,
L/W male = 1.8, female = 1.8. Cross section convex, sides
declivous all around. Dorsal sclerites: lateral margins of ab-
dominal segments asymmetrically convex, of II and III slightly
wider than thorax, of IV projecting slightly. Branchial sclerite
smoothly curving. Ventral sclerites: antennal case extending
well beyond base of wing case in both sexes (about 0.3 x
length), apex straight. Apices of leg cases in male coterminate;
in female tip of hindleg most extended, foreleg and midleg
much less, these almost coterminate. Mandibular case small
in male, much longer and fuller in female. Branchiae: me-
dium, erect, approximate, parallel, projecting forward just

short of plane of anterior margin. Plates of each branchia
lobate, rigid, parallel, inner two slightly smaller and thinner
than outers, slightly spreading; individual plates angular sec-
ulate in outline.

Larva F Hogue, 1978

Larva “F” Hogue, 1978:30, figs. 34, 36.

The existence of very distinctive larvae with gross, dorsal,
conical protuberances in the center of the abdominal seg-
ments with which no equally distinct adults can be associated
is a puzzling phenomenon. Such larvae appear among normal
larvae of other species in the tenuipes group ( tenuipes , ap-
palachiae), suggesting that they are variants expressing a de-
velopmental anomaly. This is the conclusion of Zwick (pers.
comm.), who also finds the same condition among larvae of
European Blepharicera and species of Liponeura. There is a
tendency in some larval individuals and even populations
for a centripetal crowding and multiplication of secondary
sensilla on the disc of abdominal segments, indicating hy-
peractive epidermal cell growth in the region which might
be expressed maximally by hypertrophy of the entire integ-
ument. Larva F, therefore, should not be considered a distinct
species until more knowledge can be acquired regarding the
morphological significance of this type of larva.

ACKNOWLEDGMENTS

Dr. Wayne Swank kindly provided access to streams at the
U.S. Forest Service’s Coweeta Hydrologic Laboratory.
Equipment for rearings was made available by Dr. J.B. Wal-
lace of the Department of Entomology, University of Geor-
gia. A. Huryn and J. O’Hop gave invaluable assistance in
the field.

For the use of material we wish to acknowledge the fol-
lowing repositories and institutions, including our own (listed
alphabetically by the abbreviation cited under Specimens
Examined for each of the species), and thank their respective
curators who kindly arranged loans and provided informa-
tion:

BCK personal collection, Boris Kondratieff, New Ellen-
ton, South Carolina.

BYU Bean Life Science Museum, Brigham Young Uni-
versity, Richard W. Baumann.

CNC Canadian National Collection, B.V. Peterson.

CU Cornell University, L.L. Pechuman.

LACM Natural History Museum of Los Angeles County.
UGAM University of Georgia Entomological Museum.
OSU Ohio State University, Paul H. Freytag.

ROM Royal Ontario Museum, Glenn B. Wiggins.

USNM U.S. National Museum of Natural History, Alan
Stone.

LITERATURE CITED

Baumann, R.W. 1975. Revision of the stonefly family Ne-
mouridae (Plecoptera): a study of the world fauna at the
generic level. Smithsonian Contributions to Zoology 211:
i-iii, 1-74.

Contributions in Science, Number 377

Hogue and Georgian: New Appalachian Blepharicera 19

Bonhag, P.F. 1951. The skeletal-muscular mechanism of
the head and abdomen of the adult horsefly (Diptera:
Tabanidae). Transactions of the American Entomolog-
ical Society 77:131-202.

Davis, M.B. 1983. Holocene vegetational history of the
eastern United States, p. 166-181 In H.E. Wright, Jr.,
editor. Late-Quaternary environments of the United
States, vol. 2, The Holocene (University of Minnesota
Press, Minneapolis) xvii, 277 p.

Georgian, T., and J.B. Wallace. 1983. Seasonal production
dynamics in a guild of periphyton-grazing insects in a
southern Appalachian stream. Ecology’ 64:1236-1248.

Hogue, C.L. 1978. The net-winged midges of eastern North
America, with notes on new taxonomic characters in the
family Blephariceridae (Diptera). Contributions in Sci-
ence, Natural History Museum of Los Angeles County
291:1-41.

. 1979. The family Blephariceridae in Costa Rica.

Contributions in Science, Natural History Museum of
Los Angeles County 3 1 1 : 1-22.

. 1981. Blephariceridae, p. 191-197 In J.F. Mc-

Alpine, B.V. Peterson, G.E. Shewed, H.J. Teskey, J.R.
Vockeroth, and D.M. Wood, editors. Manual ofNearctic
Diptera, vol. 1. Agriculture Canada Monographs 27:i-
vi, 1-674.

. In press. Blephariceridae, Flies of the Nearctic Re-
gion, G.C.D. Griffiths, editor (E. Schweizerbart’sche Ver-
lagsbuchhandlung, Stuttgart).

Holt, P.C., editor. 1969. The distributional history of the
biota of the southern Appalachians. Part I. Inverte-

brates. Research Division Monographs, Virginia Poly-
technic Institute, State University, Blacksburg. No. 1.

Huryn, A. D., and J.B. Wallace. 1984. New eastern Nearctic
limnephilid (Trichoptera) with unusual zoogeographical
affinities. Annals of the Entomological Society of Amer-
ica 77:284-292.

McAlpine, J.F. 1981. Morphology and terminology— adults,
p. 9-63 In J.F. McAlpine, B.V. Peterson, G.E. Shewell,
H.J. Teskey, J.R. Vockeroth, and D.M. Wood, editors.
Manual ofNearctic Diptera, vol. 1. Agriculture Canada
Monographs 27:i-vi, 1-674.

Ross, H.H. 1956. Evolution and classification of the moun-
tain caddisflies (University of Illinois Press, Urbana) vii,
213 p.

Ross, R.D. 1971. The drainage history of the Tennessee
River, p. 1 1-42 In P.C. Holt, editor. The distributional
history of the biota of the southern Appalachians. Part
III. Vertebrates, Research Division Monographs, Virgin-
ia Polytechnic Institute, State University, Blacksburg.
No. 4.

Stuckenberg, B.R. 1958. Taxonomic and morphological
studies on the genus Pau/ianina Alexander (Diptera:
Blephariceridae). Memoires de l' Inst it ut Scientifique de
Madagascar, Serie E 10:97-198, pi. I— III.

Zwick, P. 1 984. Phylogeny and biogeography of net winged
midges of the genus Blepharicera (Diptera: Blepharicer-
idae). Paper presented at the XVII International Con-
gress of Entomology, Hamburg.

Submitted 9 October 1985; accepted 30 April 1986.

20 Contributions in Science, Number 377

Hogue and Georgian: New Appalachian Blepharicera

DESCRIPTION OF A NEW SPECIES OF THE SHORE FLY GENUS DIEDROPS
(DIPTERA: EPHYDRIDAE) FROM COLOMBIA

Wayne N. Mathis1 and Charles L. Hogue2

ABSTRACT. Adults and immatures of Diedrops roldanorum, new
species, were recently collected in Colombia (Tolima: 3 km west of
Boqueron). This species, including the puparium (the first for the
tribe Dagini), is described, and a revised key to the species and a
revised diagnosis of the genus is provided.

RESUMEN. Adultos y inmaduros de Diedrops roldanorum, especie
nueva, fueron colectados recientemente en Colombia (Tolima: 3 km
oeste de Boqueron). Se describe esta especie, incluyendo la puparia
(la primera para la tribu Dagini), y se provee una clave revisada para
las especies y una diagnosis revisada del genero.

INTRODUCTION

For the past decade the genus Diedrops has received consid-
erable attention, beginning with its description (Mathis and
Wirth, 1976). Since then the genus was reviewed, as part of
a generic review of the tribe Dagini (Mathis, 1 982), and later
it received further study in the form of a new species de-
scription with additional notes on the genus (Mathis, 1984).
Each of these contributions resulted directly from fieldwork,
especially the collection of new specimens, which also pro-
vides the impetus for the present study.

Over 20 years ago, while collecting rheophilic Diptera in
Costa Rica, C.L. Hogue collected immatures of a Diedrops
species for the first time. Although the specimens were then
determined to be a shore fly (family Ephydridae), neither the
species nor the genus was recognizable from the material.
Hogue sent the specimens to W.N. Mathis, who likewise did
not recognize the genus or species. Further study of these
insects was then held in abeyance, pending collection of ad-
ditional material, especially adults.

Subsequent to his original collection, Hogue found im-
matures, principally puparia, at several other sites in Mexico,
Costa Rica, Peru, and Colombia. One site (near Boqueron)
in the last country yielded extremely large numbers of im-
matures, but adults eluded capture despite several attempts
to find them. Finally in June of 1984, Hogue returned to this
locality in Colombia and succeeded in finding additional

Contributions in Science, Number 377, pp. 21-26
Natural History Museum of Los Angeles County, 1986

immatures and, more importantly, in rearing 60 adults from
some 800 puparia he collected. This paper presents our study
of this material in the form of a new species description,
including that of the puparium, a revised key, and a slightly
revised characterization of the genus. The genus Diedrops
now includes four species, although differences in puparia
from the several disjunct populations indicate that additional
species exist. These populations need further sampling and
rearing of adults from mature puparia to establish the full
range of species diversity in the genus. Puparia usually occur
in very large numbers and, if carefully removed in quantity
to damp absorbent paper in an enclosed container, they will
yield adults suitable for taxonomic study.

Perspective for this paper is provided in the papers referred
to previously, and further details concerning generic place-
ment, etc., can be found in them. For convenience and con-
tinuity, the descriptive format adopted here essentially fol-
lows that of Mathis (1982, 1984).

Four head ratios and two venational ratios used in the
species’ descriptions are defined here for the convenience of
the reader. Frontal ratio: frontal height (from the anterior
margin of the frons to a line between the posterior pair of
ocelli)/frontal width (at the level of the anterior ocellus);
facial-head ratio: facial width between the eyes (narrowest
measurement)/overall head width (greatest measurement);
eye-to-cheek ratio: genal height (immediately below the eye)/
eye height; eye-to-face ratio: face length (in profile from an-
terior margin of eye to anterior margin of face)/eye width
(greatest length along plane of eye); costal vein ratio: the
straight line distance between R2+3 and R4+5/distance be-
tween R, and R2+3; M vein ratio: the straight line distance
along M basad of crossvein dm-cu/distance apicad of cross-
vein dm-cu.

1. Department of Entomology, NHB 169, Smithsonian Institu-
tion, Washington, D.C. 20560.

2. Entomology Section, Natural History Museum of Los Angeles
County, Los Angeles, California 90007.

ISSN 0459-8113

Figures la-c. Type locality of Diedrops roldanorum new species, a. General area. Puparia most abundant below and to right of lower strike
zone of waterfall, b. Close-up of puparia on dry portions of rock faces, c. Slanted sedimentary beds where puparia concentrated.

SYSTEMATICS

Genus Diedrops Mathis and Wirth

Diedrops Mathis and Wirth, 1976:1 26 [type species: Diedrops
aenigma Mathis and Wirth, by original designation].—
Mathis, 1977:555 [generic key]; 1982:6-10 [review].—
Mathis, 1984:349-353 [discussion, key, new species de-
scription].

DIAGNOSIS. Head. Ocellar bristles lacking; lateroclinate
fronto-orbital bristles 2, sometimes weakly developed; me-
sotrons in depression; arista moderately long, although not
twice length of 1st flagellomere, minute hairs on at least basal
%, sometimes to apex, but generally becoming bare apically;
1st flagellomere nearly twice length of 2nd antennal segment;
face shield-like, shallowly and evenly protrudent over entire

height; facial setae uniformly sparse and subequal in size
except those along oral margin, the latter setae longer, es-
pecially laterally; lacking facial series of setae extended from
midfacial height to posteroventral angles of face; genal bristle
present and conspicuous.

Thorax. Prescutellar acrostichal bristles 1 pair; scutellar
bristles variable as to comparative length; postpronotum bare
of setulae; anterior notopleural bristle only slightly smaller
than posterior one; level of insertion of posterior notopleural
bristle, especially as compared to anterior bristle, variable;
proepistemum with scattered setulae; katepistemal bristle
conspicuously weaker than anepisternal bristle; apex of vein
R.2+3 approximate to vein R4+5, distance between these at
apex less than V2 that between veins R4+5 and M; coloration
of halter, especially knob, variable; armature of forefemur
variable.

22 Contributions in Science, Number 377

Mathis and Hogue: A New Diedrops

Abdomen. Male abdomen and terminalia as follows: 5th
sternum divided, each stemite with setulae more densely
clustered toward posteromedian angle. Epandrium shield-
like, cerci and cereal cavity occupying dorsal V2 to V3, ventral
margin emarginate, setulose; gonite at least 3 times higher
than wide, with posterodorsal gonal arch, latter with ventro-
median process; aedeagal apodeme comparatively large,
J-shaped, ventral portion wider; aedeagus 2-3 times longer
than wide, in lateral view, variously shaped depending on
species.

DISCUSSION. In a recent review of the tribe Dagini,
Mathis (1982) hypothesized that the lineage giving rise to
the genus Diedrops belonged to an unresolved trichotomy;
the other two lineages are those from which Psilephvdra and
Dagus + Physemops arose. With the addition of a third and
fourth species to Diedrops, one character used previously by
Mathis was determined to be invalid (Mathis, 1984). Within
the tribe Dagini, Mathis stated (1982:5) that only in speci-
mens of the Dagits + Physemops lineage was the posterior
notopleural bristle inserted more dorsad compared to that
of the anterior bristle. In the new species of Diedrops, how-
ever, the posterior bristle is also distinctly elevated. In the
other two species of Diedrops, D. aenigma and D. hitchcocki,
the posterior bristle is inserted at a very slight elevation from
the level of the anterior bristle. But in specimens of D. steineri
the posterior bristle is distinctly inserted at an elevated level,
similar to specimens of Dagus or Physemops. We are still of
the opinion that an elevated insertion is an apomorphic char-
acter state, and consequently, that Diedrops is closely related
to the Dagus + Physemops lineage. Repositioning Diedrops
resolves the trichotomy with Psilephvdra and Dagus + Phy-
semops ( Psilephydra is now the sister group to the remaining
lineages of the tribe) but the relationships between Diedrops,
Dagus, and Physemops remain unresolved, although each
genus in this trichotomy is well characterized.

BIONOMICS. All the immatures of this genus have been
taken from small to medium-sized mountain streams, spe-
cifically from smooth rocks in the impact and splash areas
at the bases of small waterfalls. In Costa Rica puparia were
being pounded by water falling from some 2-3 m height and
were thus under intense hydraulic pressure. In other places
Hogue has found puparia at the base of smaller falls, and at
the type locality the main concentration of the population is
centered directly below and immediately beside the strike
area of a narrow fall of approximately 10 m, on sloping, fine
sandstone beds (Fig. la-c). The collections there were made
during regressive, dry season stream states, when water vol-
ume was low. Nevertheless, these stages are definitely capable
of surviving very swift current conditions and should be
added to the guild of torrenticolous dipterous groups such
as the Blephariceridae, Deuterophlebiidae, Mamina (Psy-
chodidae), etc. They exhibit some of the same morphological
adaptations, although not all to the extreme degrees of these
examples: compact and streamlined (flattened in the case of
the puparium) shape; erect external respiratory organs; suc-
tion disc venter (larva) and adhesive perimeter (puparium);
slightly lobulate segmentation; papillose (puparium) and spi-
nulate (larva) integument; and thickened, tough body wall.

Only Diedrops and one other species [Scatella ( Apulvillus )
cheesmanae; Craig, pers. comm.] exhibit these conditions
among the Ephydridae, a family with early stages more nor-
mally developing in quieter, sometimes saline water.

DISTRIBUTION. The addition of the new locality data,
noted previously, does not extend the known distributional
limits (southern Mexico to Peru) for the genus, but does
provide several localities in between, especially in Costa Rica
and Colombia.

KEY TO SPECIES OF DIEDROPS

la. Face distinctly bicolored, with a vertical, wide, brown,
median stripe about the width of the distance between
the eyes, otherwise face silvery gray to whitish; fore- and
midfemora of male with row of prominent, robust setae

along posteroventral surface (Panama)

D. steineri Mathis

b. Face unicolorous, silvery gray; leg setation of sexes sim-
ilar, weak, lacking row of robust setae 2

2a. Anepimeron with 1 to several setulae near anterior mar-
gin; wing apex bluntly rounded; apex of vein R2+3 slightly
sinuate; length of basitarsus larger than combined length

of remaining tarsomeres for each leg (Peru)

D. hitchcocki Mathis and Wirth

b. Anepimeron bare of setulae; wing apex more narrowly
rounded; vein R2+3 nearly parallel to vein R4+5; length
of basitarsus equal to or shorter than combined length

of remaining tarsomeres for each leg 3

3a. Face wider, facial-head ratio averaging 0.50; brownish
coloration of mesonotum extended to posterodorsal cor-
ner of anepistemum; anterior scutellar bristles subequal
to length of apical pair; length 4 to 4.5 mm (Mexico)

I), aenigma Mathis and Wirth

b. Face narrower, facial-head ratio averaging 0.40; anepi-
sternum entirely silvery gray, lacking any brownish col-
oration; anterior scutellar bristles about Vi length of api-
cal pair; length 2.3 to 3.3 mm (Colombia)

D. roldanorum, new species

Diedrops roldanorum new species

Figures 1-10

DESCRIPTION. Adult male and female. Moderately small
to medium-sized shore flies, length 2.3 to 3.3 mm.

Head. Frons width-to-length ratio 0.30; vestiture of frons
uniformly microtomentose, appearing dull, vestiture of me-
sofrons not distinguished from that of parafrons; face, in
lateral view, conspicuously inclined anteroventrally, arched,
just below facial prominence, thereafter very shallowly arched,
nearly flat; facial setae comparatively longer and more con-
spicuous immediately below antennae and along oral margin;
face unicolorous, grayish silver, lacking a median, vertical,
brown stripe; eye width-to-face ratio 0.40; anteroventral
margin of eye bluntly rounded; eye-to-cheek ratio 0.60.

Thorax. Setae of dorsocentral and acrostichal series gen-
erally weakly developed; only a larger pair of prescutellar
acrostichal setae and the posterolateral dorsocentral bristle
well developed; anteroventral scutellar bristle more weakly

Contributions in Science, Number 377

Mathis and Hogue: A New Diedrops 23

Figures 2-6. Diedrops roldanorum: 2, epandrium and cerci, posteroventral view; 3, same, lateral view; 4, internal male genitalia, posterior
view; 5, same, lateral view; 6, sternum 5 of male.

developed, about xh length of apical scutellar bristle; posterior
notopleural bristle inserted at about same level as anterior
bristle; anepimeron bare of setulae; anepistemum entirely
silvery gray, lacking any brownish coloration. Leg setation
of sexes similar, weak, lacking row of robust setae; length of
basitarsus slightly longer or subequal to combined length of
remaining tarsomeres for each leg; knob of halter mostly pale.

yellowish. Wing with apex more narrowly rounded; vein R2+3
evenly and very shallowly arched on basal 3/4, thereafter very
shallowly dipping toward vein R4+5 distally; costal vein ratio
15.50; vein M ratio 1.20.

Abdomen (Figs. 2-6). Dorsum slightly lighter in color than
mesonotum; tergum 1 and anterior lh of tergum 2 grayish,
other terga blackish brown; length of 3rd tergum of male only

24 Contributions in Science, Number 377

Mathis and Hogue: A New Diedrops

I 1 . Omm 1

Figures 7-10. Diedrops roldanorum: 7, puparium, dorsal view; 8, puparium, ventral view; 9, same, lateral view; 10, cephalopharyngeal
skeleton of 3rd instar larva, lateral view.

slightly shorter than combined length of 4th and 5th terga;
5th tergum of male (Fig. 6) with posterior margin deeply
emarginate; male terminalia as in Figures 2-5.

Puparium. Shape. Generally oval in dorsal and ventral
views (Figs. 7, 8) with 9 ventrolateral, rounded welts forming
a crenulate lateral margin, each welt fringed with short setulae
(welts probably used for locomotion); retreated margins be-
tween welts extended dorsally as shallow, gradually indented
furrows that become weaker dorsally to a small black spot,
thereafter nonexistent, furrows apparently delimiting seg-
ments; in lateral view dome-shaped (Fig. 9), dorsum gently
and evenly rounded, venter mostly flat; 2 dorsal, more or
less prominent, digitiform projections near anterior (anterior
spiracles) and posterior ends (respiratory tubes), anterior spi-
racles with small funnel-like structures around lateral margin;
respiratory tubes larger than anterior spiracle but unadorned,
apical Vs to Vi completely dark brown.

Coloration. Dorsum generally dark brown; venter paler,
yellowish to whitish laterally, darker medially. Surface pat-
tern; dorsum appearing shagreened, with granulations ellip-
tical; medial longitudinal Vs with granulation pattern oriented
from side to side, orientation of lateral pattern from front to
back; a regular pattern of small black spots, as in Figures 7
and 9, in addition to granulations.

Dimensions. Length 3-3.5 mm; width 1.9-2. 1 mm; height
0. 8-1.1 mm.

Cephalopharyngeal skeleton of third instar larva. Man-
dibles paired, not connected dorsally, length of anteroventral
projection variable (Fig. 10 shows it at its longest); dental
sclerite and other detached ventral sclerite as in Figure 10;
hypopharynx with slender, delicate, dorsal bridge, sclerite
broadly fused posteriorly with dorsal and ventral cornua;
both dorsal and ventral cornua pigmented dark brown, pos-
terior portions of each cornu with irregularly shaped win-
dows, these paler; parastomal bar bifurcate, with 2 slender
connections to dorsal cornu; ventral cornu more robust than
dorsal cornu, becoming irregularly thicker posteriorly and
posterior margin with a median, shallowly pointed projection
(not evident on some specimens). Length 1.9 mm.

TYPE MATERIAL. The holotype male is labeled “CO-
LOMBIA, Tolima [Departamento]: 3 km W Boqueron 22
June 1984/see field notes Chas. L. Hogue No: CLH 343.1.”
The holotype is pinned directly, is in good condition (slightly
teneral), and is in the Natural History Museum of Los An-
geles County. The allotype female and 44 paratypes (11<5,
339; LACM, USNM) bear the same locality data as the ho-
lotype. All adult specimens of the type series were reared
from mature puparia.

ADDITIONAL MATERIAL. In addition to the type se-
ries of adults from the type locality, there are several hundred
larvae, puparia, and puparial skins preserved in alcohol
(LACM, USNM).

Contributions in Science, Number 377

Mathis and Hogue: A New Diedrops 25

DISTRIBUTION. Presently known only from the type
locality in Colombia.

BIONOMICS. The type locality is a small stream feeding
into the Rio Sumapaz through a deep, steep gorge (Fig. la).
The area is heavily vegetated near the stream and located
generally in Dry Tropical Forest (Holdridge System). A stream
of water drops over the face of a large (30-40 m) undercut
cliff and continues onto the wide face of strongly tilted (20°),
stratified, fine grain, compact, black and gray sandstone (Fig.
1 c). Large boulders broken from this bed and the surrounding
cliffs clutter the course of the stream below, before it runs
under the nearby highway in a concrete culvert. The largest
numbers of puparia are located on these beds, especially
along the drier, eroded edges, wet only with spray and mist
(Fig. lb). The larvae were found within the direct strike zone
of the fall, amidst a fairly thick growth of algae. They ap-
parently migrate to the drier periphery of the inundated area
for pupariation. Many puparia, especially those farthest from
the fall, were completely dry, and often found to be empty
skins from which the adults had already emerged. Hogue
netted many hundreds of adult ephydrids of several genera
from the immediate vicinity of these microhabitats in June
of 1983, but failed to turn up any Diedrops; adults were only
obtained by rearing puparia. The latter continued issuing
from their puparia for a period of 1 1 days before emergence
ceased.

Several hymenopterous parasitoids emerged from larvae
and puparia that were being reared. They are as follows:
Family Diapriidae: Trichopria new species (determination
provided by Dr. L. Masner); Family Eucoilidae: new genus
(determination provided by Dr. G. Nordlander).

ETYMOLOGY. It is a pleasure to name this species after
Dr. and Mrs. Gabriel Roldan P., both of whom figured prom-
inently in the success of fieldwork conducted in Colombia
through their gracious hospitality, detailed knowledge of the
terrain, and significant contributions to the study of aquatic
insects in Latin America.

REMARKS. The narrower face, which is unicolorous,
grayish silver; the unicolorous, grayish anepisternum; the
bare anepimeron; the short anterior scutellar bristle; the more
narrowly rounded wing apex; and characters of the male
terminalia distinguish this species from congeners.

The type locality (Fig. 1) is situated immediately upstream
from a road culvert passing under the highway connecting
Boqueron and Melgar, 3.3 km west of the bridge crossing
the Rio Sumapaz at Boqueron (2.4 km west of Nariz del
Diablo, a prominent rock jutting over the pavement) (4° 1 6'N,
74°34'W), 500 m elevation. Extremely large numbers of im-
matures have been observed here by Hogue over a period of

several years, in June-July, the driest months of the first of
the region’s biphasic dry seasons.

ACKNOWLEDGMENTS

We thank Drs. Norman E. Woodley and Douglas A. Craig
for reviewing the manuscript. The illustrations were pro-
duced by George Venable and are gratefully acknowledged.
For assistance in providing determinations of the parasitoids
we thank Dr. Arnold S. Menke (SEL, USDA), Dr. L. Masner
(CNC), and Dr. Goran Nordlander (Swedish University of
Agricultural Sciences, Uppsala, Sweden). We are indebted
particularly to the Servicio de la Eradicacion de la Malaria,
Bogota (Drs. Michael J. Nelson and Marco F. Suarez) for
providing logistical assistance in obtaining specimens of the
new species, and to INDERENA (Instituto National de Re-
cursos Naturales del Ambiente), University of Antioquia
(Laboratorio Alexander Humboldt— Dr. Gabriel Roldan P.),
and Museo de Historia Natural, Universidad Nacional de
Colombia (Dr. Reuben Restrepo), for cooperation in col-
lecting collateral material at various times in Colombia. Hogue
wishes to acknowledge the general financial help of the Los
Angeles County Museum of Natural History Foundation and,
in other countries, specifically the following individuals and
agencies for making it possible to obtain Diedrops: Organi-
zation for Tropical Studies, Eric Fischer and D. Baird (Costa
Rica); Dr. Gerardo Lamas M., Museo de Historia Natural,
Universidad Nacional Mayor de San Marcos (Peru); Mr.
George V. Caldwell, Los Angeles and Direccion General de
la Fauna Silvestre, Secretaria de Agricultura y Recursos Hid-
raulicos (Mexico).

LITERATURE CITED

Mathis, W.N. 1977. Key to the neotropical genera of Pa-
rydrinae with a revision of the genus Eleleides Cresson
(Diptera: Ephydridae). Proceedings of the Biological So-
ciety of Washington 90(3):553— 565.

. 1982. Studies of Ephydrinae (Diptera: Ephydridae),

VI: Review of the tribe Dagini. Smithsonian Contri-
butions to Zoology No. 345:1-30.

. 1984. Notes on the shore fly genus Diedrops (Dip-
tera: Ephydridae). Proceedings of the Entomological So-
ciety of Washington 86(2):349-353.

Mathis, W.N. , and W.W. Wirth. 1976. A new neotropical
shore fly genus with two new species (Diptera: Ephyd-
ridae). The Pan-Pacific Entomologist 52(2): 126-1 32.

Submitted 8 October 1985; accepted 13 January 1986.

26 Contributions in Science, Number 377

Mathis and Hogue: A New Diedrops

PIT*

fNSTiWCriONS FOR AUTHORS

' !'! ! I ;

The Natural History Mu^isrn: of Lbf An.geles: County publishes the ibfoii of original rest arch
in the life and earth sciences :in its Contributions in Science series. Individual (kntrib locus
are issued at irregular intervals and range in size from papers of 8 printed pages to lengthy

mon ographs. .jffM iSlf illlllillllf vliHil Ly. j :}p S:j ; }" | fft; rg ■:

Manuscripts -submit ted for publication ’wili undergo anonymous peer review; Priority ..is
given to manuscripts vftpltten.iby members I§§|;fi|§ Museum staff Manuscripts should be
prepared in accordance with ' fie below and sufeniitted: to tile Head of
the appropriate Section of the Museum. gP;; f.f; it yiPrb: Pi; -fi :1 idu;;l::;;iir!lt:P'i::Pt 1 ;-..}:A;j, ;

Authors must adhere to the a nicies qf tlikifnieipatidnal Cbdefidf Botanical di; Zoological
^|m4hCiatU!to and|&e ur||d to coinsip^ • a^| re||2i®&4arioUs or these cedes.

Authors proposing new tax a must indicate mat primary' types have been deposited in
accordance with the recommendations of the appropriate code, citing the collect ion, by -nUme
and. providing the relevant . identify^::dfet|Ssi. The depository of other- study material should
also be indicated. dij ; i l§||j$i:|| iSll !

An d subjects.,

the abstract must be presented in • Spam sh of Portuguese, as appropriate, as well, as in E.iig|;i.$h.
Summaries in other languages are not- required- but: are strongly recommended where ap-
propriate.

PREP A RATION OF MAN U SCR l PT

Type manuscript double- spaced*. ihSluding tide page, abstract, text, acknowledgments, ref-
erences, tables, and figure: cap lions. - A pi-tpef must lota! a minimum of 8 printed, pages fas a
guideline, 3.5 pages of draft equals, about l page of final printed copy}.. Indicate italicized
words; ^ With underscores;. typing dement. :

Each manuscript component should begin on a new j||ge, m the ix||pwJng|Smk title
page; abstract(s); text; :actedwledgrnents.; literature, .cited; tables, each table— complete; with
title and footnotes— on a separate page; and figure. captions, All pages! should be numbered

Submit original illustrations. Line drawings -are best prepared Tor 50 percent -'reduction.;,
and should not exceed 14 by 17.5 inches (3:5 .by 43 cm) in size: photographs are usually
reproduced at full size and should not exceed 7 by:;fi.75 inches (1 7 by 22cm, full page width)
or 3.25 by 4,5 inches (8 by 11 cm, sii||ld CQ’litiiifi

SUBMISSION i|» AT AN' UiSCBXPT

Bjj.A'fl --A

Transmit three copies of . tire maniispfipt in a secure enclosure. Manuscript should be ac-
companied byiatoyeit-letjier- idpned by the author who will -be responsible for correspondence
regarding the manuscript. The covering letter should cof4®la stulilHi. that the. manuscript :
is based, on original research and. has not been, published bl'^w-Bi'ere except. in abstract or
abbreviated form. The letter should: also give information on prior publication of any part
of the research and whether or not funds are available to meet the cost of unusual format
or color illustrations. Include copies, of any . permissions •: needed to reproduce publi she di

material, i ■ll||q;:||S; : ilfiiB

Manuscripts will be reviewed for possible publication -with the understanding that they}
have not been published, simultaneously submitted, or already accepted for .-publication
elsewheffe: This doesnot preglpde consideration of a complete report that follows publication
of preliminary findings' elsewhere. Copies of any possibly duplicative material should bif:i
submitted with the mmrasbri.pl fhat.i&.tkibg, "II- f! i:;i; j j :t:dr : av! 'll1:: ■ Bri

PAGE CHARGES

ter: acceptance of a paper by ihe-Museum, the aul,hor(s} will receive a slatement.of page
charges from the |||K Partial or complete payment i

of this statement will be solicited from those authors who have funds available; for- ihis
purpose. Authors without : access to fuhdslfof partial or complete payment of -page charges"
should so indicate on the statement and return it to the Managing Editor. The rctu ved
statement only qualifies the: paper for publication; acceptance :pf a paper and priorit, of

publication are in no way

- «.$j . L

•«I,‘

HmPBKi

on payment: re

.. , j ; ■ -

the stare metre

1 ; ...

........ if

:=: c • r i: <! .!■ •

Iiiss

i!:iri!|!m!|i IfflH
11 8 ilplHii

ills

1jb«I4ih

it

lj|

iiSBii

iPpI

1

iHH

jil

jiiikCySi'tf

MM

i]|! !;ri |s|fi

' ,H1iSii i, !it

iiiiiiiiWll

iSPii&ipiW

'll Iff I ill
litff

• iffii [ ijlpi ji!1!' fc1 h

SSillMaM 11111

"“aSISsMfflpw i

ip; Sliilli ' ‘ '

\mmm

E RIAL PUBLICATIONS OF THE

NATURAL HISTORY MUSEUM OF LOS: ANGELES COUNTY

; The- suit:- iiii ik - publications -of the Natural' Hi story Museum of Los Angeles County have been
tssjuekl at imigular iriten*als in three major series; the articles in each series are numbered
'iduiftl y, and numbers run consecutively. regardless of the subject matter.

«; Cioni;ribi;.ttioin,s in Science, a miscellaneous scries of technical papers describing orig-
inal research in. the Hie arid earth sciences.

t’ Spieace Bulletin, a miscellaneous series of monographs describing original research
. in. the life and earth sciences. Thus- series was discontinued in 1978 with the issue of
Numbers 39 and 30; monographs are- now published by the Museum in Contributions

in Science.

#;ScitjaCe Series, long articles on nahmaL hi. story topics, generally written for the layman.

Copies i;:if the publications in these series are sold through the Museum Book Shop. A catalog

.is.av&Mhte:.o«: request: ' . .''C.': it :"u :C hi ki ii !-:jih:'l| C irb I ' L

'SCIENTIFIC PUBLICATIONS COMMITTEE

Craig C. Black, Museum Director

— m

THE TAXONOMY AND NOMENCLATURE OF SOME AUSTRALIAN
PARAGIINE WASPS (HYMENOPTERA: MASARIDAE)

Roy R. Snelling1

ABSTRACT. New taxa are described among the Australian masarid
wasps. The genera known to occur in Australia are separated by a
key. A new genus, Ammoparagia (type-species: A. hua, new species),
is described. Two new species of Rolandia, R. borreriae and R.
houstoni, are described and Riekia angulata Richards is transferred
to Rolandia; the four known species of Rolandia are separated in a
key.

The genus Metaparagia and the subgenera Cygnaea and Paragiella
of Paragia are synonymized under Paragia. Three new species of
Paragia are described: P. oligomera, P. conjluens, and P. monocesta.
Comments on distribution and synonymy are made on P. magdalena
Turner, P. nasuta F. Smith, P. sobrina F. Smith, and P. walkeri
Meade- Waldo. Important taxonomic features of the new taxa are
illustrated.

INTRODUCTION

The masarid wasps of the world were reviewed by Richards
(1962); three subfamilies were recognized: Gayellinae, Eu-
paragiinae, and Masarinae. The first two are small groups
limited to the Western Hemisphere. The worldwide subfam-
ily Masarinae was divided into two tribes, the Paragiini and
the Masarini. Only the tribe Paragiini was known to be pres-
ent in Australia, where there were four genera to accom-
modate about two dozen species.

Most recently, Carpenter (1 982) has demonstrated that the
Masaridae, as conceived by Richards, are an unnatural group,
since the Euparagiinae are a sister group to the Masarinae +
Gayellinae of Richards. In Carpenter’s view, the several for-
mer families of Vespoidea are united into the single family
Vespidae, with the Masarinae + Gayellinae as a subfamily,
these groups assuming tribal rank; the erstwhile Paragiini of
Richards is merged with the Masarini.

While, on the whole, I am in agreement with Carpenter’s
proposals, for purposes of discussion of the various taxa here,
I am following the scheme of Richards (1962). This is, at
present, the only recent monograph of the group. Therefore,
it seems expedient for discussions to relate to the hierarchy
used there. The higher classification of the masarids is pres-
ently being studied by Carpenter.

Contributions in Science, Number 378, pp. 1-19
Natural History Museum of Los Angeles County, 1986

The following new taxa are described so that the names
might be available to Terry Houston for his studies on their
biologies.

SPECIMENS EXAMINED

Most of the material recorded below is from the collections
of the Western Australian Museum, Perth (WAM); other
specimens are from the collections of the British Museum
(Natural History), London (BMNH); Australian National In-
sect Collections, Canberra (ANIC); National Museum of Vic-
toria, Abbotsford (NMV); University of Queensland, St. Lu-
cia (UQLD).

TERMINOLOGY

All measurements were made by means of a micrometer disc
within one eyepiece of a binocular microscope.

Ammochaetae. Ammochaetae are present in a few genera
of masarids. These are long bristles or setae, somewhat flat-
tened, that form a definite fringe along the genal margin of
the head and the lower margin of the mandibles. Since their
tips curl inward, the ammochaetae form a “basket.” Al-
though the function of the ammochaetae is presently un-
known, presumably they form a psammophore used to trans-
port sand particles excavated from a nest. Note that most
masarids have hairs on the underside of the head, but these
are irregularly distributed, uneven in length, and do not form
a definite, close-set row along the genal margin.

Clypeal length is measured along the midline, from base
to apical margin.

Clypeal width is the maximum width across the clypeus at
the level of the lateral angles. The distance between the clyp-
eus and the inner eye margin is measured from the lateral
angle of the clypeus to the nearest point on the eye margin.

1 . Entomology Section, Natural History Museum of Los Angeles
County, 900 Exposition Blvd., Los Angeles, California 90007.

ISSN 0459-8113

Head length is measured along the midline of the face and
is the maximum measurable distance between the apical
margin of the clypeus and the dorsal (preoccipital) margin
of the head, when both are in focus in frontal view.

Head width is the maximum measurable width, in frontal
view, across the eyes.

Occipital carinae( Figs. 25-27) have been used by Richards
as both generic and specific diagnostic characters; the two
carinae were designated the “dorsal occipital keel” and the
“ventral occipital keel.” The “dorsal occipital keel” is the
preoccipital carina (Fig. 27, prc) and begins on the preoccip-
ital margin a little mesad of the summit of the compound
eyes and follows the genal margin ventrad, and usually be-
comes obsolete before attaining the hypostomal carina. Rich-
ard’s “ventral occipital keel” extends dorsad from the pos-
terior mandibular articulation along the posterior eye margin.
I prefer the term postocular carina (Fig. 27, poc) for the latter
since it better describes the position of the carina; use of the
term “keel” suggests a higher and thinner structure than is
the case. When the gena is greatly narrowed, as in most
Rolandia, the preoccipital and postocular carinae become
confluent.

Pretegidar groove and carina are used to describe features
on the dorsal, posterior portion of the pronotum anterior to
the tegula. These are best developed in species of Paragia
and consist of a broad, shallow groove followed by a low,
but usually sharp, carina immediately anterior to the pos-
terior margin of the segment. The description used by Rich-
ard (e.g. “Spiracular lobe of pronotum well marked off by a
furrow”) seems less efficient than “pretegular groove pres-
ent”; the terms used here are in accord with usage elsewhere
among the vespoid wasps (e.g. Bohart, 1984; Giordani Soika,
1978).

Punctures are described according to size, as follows: fine
(0.02-0.035 mm diameter), moderate ( 0.036-0.055 mm di-
ameter), coarse (0.056-0.070 mm diameter), or very coarse
(over 0.070 mm diameter). Distances between punctures are:
contiguous (punctures so crowded as to often be deformed
and interspaces are sharp-edged), subcontiguous (interspaces
up to 0.30 times a puncture diameter), dense (interspaces
ranging between 0.30 and 0.70 times a puncture diameter),
close (interspaces 0.70-1 .50 times a puncture diameter), sparse
(interspaces from 1 .50-3.00 times a puncture diameter), scat-
tered (interspaces irregular and ranging 3.00-6.00, or more,
times a puncture diameter). Variations in puncture size and/
or density may be described by combining terms (punctures
moderate to coarse; punctures subcontiguous to dense, etc.).

Total length is the sum of head length + thorax length +
gaster length. Because gastric segments may be contracted or
extended, this measurement is inexact and is made only to
the nearest tenth of a millimeter. Note that head length is
included, not from “front of head” or some such equally
imprecise method; consequently, total lengths given here will
seem larger than those of other authors. For example, Rich-
ards (1962) gives length for Paragia excellens (as P. sobrina)
as 13.0-14.0 mm; my method for this species yields 1 6.6—
17.2 mm.

KEY TO AUSTRALIAN GENERA OF PARAGIINI

la. Parategula absent (Fig. 15); disc of last visible gastral
tergum not extended over apical margin (i.e. apical mar-
gin is posteriormost portion of segment) (Fig. 12) . . 2
b. Parategula present (Fig. 8); disc of last visible gastral
tergum extended so that true apical margin is ventral
and anterior to posteriormost portion of segment (Fig.

3) Ammoparagia, new genus

2a. Glossa always exposed, not retractile into prementum;
gastral tergum 2 constricted at base (Fig. 14); ammo-
chaetae present (Figs. 25, 26) or absent (Fig. 27) along
posterior genal margin; propleuron with or without lat-
eral groove 3

b. Glossa retractile into prementum, thus often concealed;
gastral tergum 2 not constricted at base (Fig. 1 3); pos-
terior margin of gena without ammochaetae; propleuron

with lateral groove Riekia

3a. Anterior margin of pronotum convex (Fig. 28); gena

without ammochaetae (Fig. 27) Paragia

b. Anterior margin of pronotum angulate (Fig. 29); gena
with ammochaetae along posterior margin (Figs. 25, 26)
Rolandia

SYSTEMATICS

Ammoparagia, new genus

Figures 1-8

DIAGNOSIS

A member of the tribe Paragiini as defined by Richards ( 1 962),
recognizable by the following combination of characteristics:
glossa retractile into prementum; parategula present; scutel-
lum produced over metanotum and base of propodeum; gas-
tral terga 2-4 (9) or 2-5 (6) with transverse basal groove;
ocular sinus moderately deep; head hollowed out beneath
and distinct genal ammochaetae present.

DESCRIPTION

Mandible with three acute apical teeth, decreasing in size
from lower to upper; blade broad, narrowest at base. Max-
illary palpus (Fig. 6) four-segmented, first segment short and
stout, second segment longer and narrower; next two pro-
gressively a little shorter, each with fine bristles, most nu-
merous on second. Prementum triangular in profile, little
longer than high, uniformly sclerotized ventrally; labial pal-
pus (Fig. 5) four-segmented, first segment longest, second
and third subequal, last segment much narrower than others,
third segment (9) with two stout recurved bristles that extend
much beyond tip of last segment; paraglossa and glossa with
terminal sclerotized pads; paraglossa little longer than men-
tum; glossa about three times as long as prementum, bifur-
cated portion a little longer than basal portion, without dorsal
comb-like processes proximally, entire dorsal surface finely
imbricate, less strongly so distad. Hypostomal apodeme tri-
angular; hypostomal bridge moderately broad, not depressed.
Labrum flat, apical margin transverse (5) or rounded (9).

2 Contributions in Science, Number 378

Snelling: Australian Paragiine Wasps

8

Figures 1-8. Ammoparagia hua. 1, 2, male genital capsule, ventral and lateral views, respectively; scale lines = 0.50 mm. 3, female gastric
apex, lateral view. 4, male gastric apex, ventral view. 5, 6, labial and maxillary palpi, respectively. 7, male protrochanter and -femur base. 8,
female thorax, dorsal view; scale line = 1.00 mm.

Contributions in Science, Number 378

Snelling: Australian Paragiine Wasps 3

Clypeus about 1.4 times as broad as long, apical margin
transverse and wider than distance from its lateral angle to
lateral angle of clypeus, disc raised above paraocular area.
Malar space absent. Antennal club stout and moderately de-
fined ($) or flagellum weakly broadened distad (3).

Ocular sinus moderately deep, obtuse. Interantennal area
flat, without median tubercle; antennal sockets separated from
inner eye margin by less than a socket diameter; interantennal
distance about five times antennal socket diameter. Preoc-
cipital carina low, postocular carina weak to absent; suboc-
cipital furrow deep. Ammochaetae present along genal mar-
gin and lower margin of mandible.

Side of pronotum without oblique impressed line; humeral
margin obtuse and ill-defined; pretegular carina absent; pre-
tegular groove shallow (more so in 3). Outer margin of tegula
subcircular. Notauli of mesoscutum well marked anteriorly,
but not extending much distad of level of anterior margin of
tegula; parapsidal line long, narrow; parategula present (Fig.
8). Anterior margin of scutellum confluent with posterior
margin of mesoscutum; disc of scutellum convex along mid-
line and obtusely produced over metanotum and base of
propodeum (Fig. 8); disc evenly rounded onto axilla. Median
portion of metanotum hidden under scutellum, vertical. An-
gles of propodeum obtuse.

Pterostigma of forewing about three times longer than
broad, lower margin rounded, prestigma about one-third as
long as stigma; marginal cell about 2.5 times longer than
wide, apex abruptly curved away from wing margin; junction
of M and Rs moderately swollen; first submarginal cell, on
Rs, about one-third longer than second, second receiving
both recurrent veins; cu-v considerably distad of separation
of M and Cu; junction of Cula and 2 m-cu rounded. Anal
lobe of hindwing less than one-third as long as cell Cu; veins
1A and cu-a rounded together; 13 hamuli present. Tegula
subcircular.

Protrochanter of male (Fig. 7) with distal process elongate
and spine-like in ventral view, its lower margin cariniform;
first three segments of protarsus of both sexes strongly asym-
metrical, posterior lobe, much the longer, especially on sec-
ond and third segments; mesotibial spurs short and stout;
femora without basal ring; outer metatibial spur much short-
er than inner, inner spur trifid at apex; first and second meta-
tarsal segments combined longer than metatibia; tarsal claws
with small erect tooth at about midlength.

Gastral tergum 1, in dorsal view, more than twice wider
than long; terga 2-5 (3) or 2-4 (2), depressed at basal and
apical margins, especially in male; distal margin of last ter-
gum (both sexes) hidden under protuberant “hood” of disc
(Fig. 3); basal groove of sternum 2 very shallow and poorly
defined; sterna flat (both sexes) and male without discal pro-
tuberances on any segment; sternum 6 of female broadly
rounded at apex; apical margin of last visible sternum of
male broadly rounded.

MALE GENITALIA (Figs. 1, 2). Parameral spine dor-
soventrally flattened, apex simple, not recurved; digitus of
volsella long and with tubercles along outer margin; cuspis
of volsella prominent and fused to paramere.

TYPE-SPECIES. Ammoparagia hua, new species.

ETYMOLOGY

The generic name is combined from the Greek ammos (sand)
plus the generic name Paragia.

DISCUSSION

The presence of parategulae and the four-segmented maxil-
lary palpus will immediately separate Ammoparagia from
all other known Australian paragiine genera. The genus is
monotypic at the present time.

In the key to Australian genera of Paragiini by Richards
(1962) Ammoparagia fails at the first couplet. It agrees with
Riekia Richards, 1962, in that the glossa is retractile into the
prementum, a feature previously believed unique to Riekia
in the Australian fauna. Ammoparagia differs from Riekia
in that the head is moderately concave beneath and there is
a strong fringe of ammochaetae along the genal and lower
mandibular margins, there is no lateral pronotal furrow, the
scutellar disc extends over the metanotum and base of the
propodeum, there is a definite propodeal angle, the gastral
terga are depressed at the base, and the apical margin of the
last exposed tergum is hidden under the disc of the segment
in both sexes.

From Rolandia Richards, 1962, Ammoparagia is further
separable by the lack of a lateral pronotal furrow, the lack
of an impressed line on the propleuron, the shape of the
scutellum, the shape of the last visible tergum, the lack of a
definite basal furrow on sternum 2, and the lack of comb-
like processes on the glossa.

Ammoparagia differs, also, from Metaparagia Meade-
Waldo, 191 1, in the presence of ammochaetae, the presence
of notaulices, the subcircular tegula, asymmetrical protarsus,
the presence of an erect tooth on each tarsal claw, the shape
of the scutellum, the anteriorly constricted gastric terga, and
the shape of the last gastric tergum.

Finally, from Paragia Shuckard, 1837, Ammoparagia may
be further separated by the moderately developed ocular si-
nus, the presence of ammochaetae, the lack of a groove setting
off the spiracular lobe of the pronotum, the lack of an im-
pressed line on the side of the pronotum, the poorly devel-
oped furrow at the base of gastric sternum 2, the shape of
the last visible tergum, the much longer glossa, the non-
recurved parameral spines of the male, and the broadly
rounded last exposed male sternum.

Outside of the Australian fauna there are two paragiine
genera: Ceramius Latreille, 1810, in Eurasia and Africa, and
Ceramiopsis Zavattari, 1 9 1 0, in South America. Both of these
genera differ from Ammoparagia in the absence of parateg-
ulae, the shape of the scutellum, the lateral (rather than dor-
sal) placement of the propodeal spiracle, the shape of the last
gastral tergum as well as in many details of the mouthparts,
wing venation, and male genital structure.

So far as I have been able to determine, no other masarid
wasp possesses parategular processes on the mesoscutum,
except in the Euparagiinae which have a minute process.
However, in Euparagia Cresson, 1879, the mesoscutum has
a sharp, raised margin adjacent to the tegula and the para-
tegular process appears to be an extension of this margin. In

4 Contributions in Science, Number 378

Snelling: Australian Paragiine Wasps

Ammoparagia the mesoscutum is not marginate and the
parategular process is digitiform and bends down over the
anterior portion of the scutellar axilla. This is a common
feature in the related family Eumenidae. The shape of the
last gastric tergum appears to be unique within the Masari-
dae.

Aside from these two unusual features, Ammoparagia seems
to be nearest to Riekia and Rolandia. The retractile glossa
is shared with Riekia, as well as the truncate clypeus and the
poorly developed occipital ridges. The more prominent dif-
ferences have already been noted above. There is a resem-
blance, too, to Rolandia, especially in the presence of genal
ammochaetae and the reduced number of segments in the
maxillary palpus, though Rolandia has lost only one, rather
than two segments.

Ammoparagia hua, new species

Figures 1-8

DIAGNOSIS
Same as for the genus.

DESCRIPTION

MALE. Measurements. Holotype head width 2.03; head
length 1 .95, wing length 6.09; total length 9.4 mm. Paratypes:
HW 2.06-2.13; HL 1.77-1.87; WL 5.85, TL 8.8-9. 1 mm.
Head 1.14-1.16 times as broad as long; mandible short and
broad, tridentate, teeth acute and on strongly oblique margin,
inner tooth much the shortest, dorsal margin straight, broadly
rounded onto apical margin. Clypeal disc moderately shiny
between subcontiguous to contiguous coarse, elongate punc-
tures; vertex similar but punctures less elongate, moderate
in size. Interocellar distance about twice diameter of anterior
ocellus; ocellocular distance about 1.7 times diameter of an-
terior ocellus; ocelloccipital distance about 2.5 times diam-
eter of anterior ocellus. Antennal scape shorter than inter-
antennal distance; first flagellar segment a little more than
twice as long as wide and about two-thirds as long as scape;
following flagellar segments a little shorter than first; flagel-
lum gradually broadened distad, apical segments about 1.5
times as broad as apical breadth of first segment.

Anterior margin of side of pronotum straight, not angled
below midlevel; pronotal humerus obtuse at side; pronotum
moderately shiny, interspaces slightly sculptured between
moderate, subcontiguous punctures. Mesoscutum moderate-
ly shiny between moderate punctures that tend to be arranged
in short rows, punctures more elongate cephalad. Scutellum
dull, punctures coarse and elongate, mostly subcontiguous,
but with linear impunctate area on center of disc, integument
microstriate, appearing somewhat silky. Metanotum visible
only at sides, visible portion concave, shiny, coarsely areo-
late. Mesopleuron shiny between dense to subcontiguous,
moderate to coarse punctures, and with sparse, minute se-
tigerous punctures. Propodeal angle obtuse; disc moderately
shiny between coarse, dense punctures and with short, irreg-
ular rugulae; side shiny, with irregular impunctate areas in

middle, punctures coarse and dense above, finer and more
separated below.

Ventral process of protrochanter (Fig. 7) longer than seg-
ment, apex narrowly rounded in profile, ventral margin com-
pressed and subcarinate.

Gastral terga moderately shiny between coarse, dense
punctures, last tergum subcontiguously to contiguously punc-
tate; sternal punctures coarse, dense to subcontiguous inter-
spaces with minute piliferous punctures, surface appearing
dull due to numerous fine hairs.

Erect hairs yellowish and abundant on all body surfaces,
especially long and dense on frons, vertex, and dorsum of
thorax; hairs conspicuously shorter, but no less abundant,
on gastral terga; gastral sterna with hairs shorter and less
abundant and with numerous short appressed hairs.

Head black, the following yellowish with a weak orange
tint: mandible; clypeus; ocular sinus, narrowly extended along
inner orbit nearly to base of mandible; large, irregular frontal
band, narrowly separated from clypeus and from ocular sinus
mark; linear band along upper outer orbit.

Thorax black, the following yellowish with a weak orange
tint: most of pronotum except anterior margin of propleuron,
mark along margin of collar adjacent to anterior margin of
mesoscutum, and narrow mark in front of tegula; parategula
of mesoscutum; tegula, except transparent central spot; most
of upper plate of mesopleuron; linear bar in middle of scu-
tellum, not reaching anterior margin; metanotum with mi-
nute median spot; dorsolateral spot on propodeum, including
angle. Legs principally yellowish, but with irregular black
marks ventrally on all coxae. Wings dusky, veins dusky fer-
ruginous basad to brownish distad.

Gaster largely orange-yellow; base and obscure submedian
spot on dorsal surface of tergum 1 (absent in one paratype)
blackish; tergum 2 with small, irregular dark blotch on either
side of middle; terga 3-6 with large, submedian blackish spots
that extend nearly to apical depression, their respective inner
margins obliquely convergent basad; tergum 7 more ferru-
ginous, with a pair small, submedian dark blotches at base;
ventral segments yellowish ferruginous, a little dusky basad.

FEMALE. Measurements. Head width 2.16-2.19; head
length 1.94-1.97; wing length 5.97; total length 9. 2-9. 5 mm.

Head similar to that of male with the following differences:
about 1.11-1.12 times as broad as long; frons and vertex
duller, contiguously rugosopunctate; preocciput similar,
punctures a little coarser. First flagellar segment about twice
as long as wide and about one-half as long as scape; remainder
of flagellum forming a stout club, antepenultimate segment
about twice broader than long. Interocellar and ocellocular
distances about 2.5 times diameter of anterior ocellus; ocel-
loccipital distance about 2.7 times diameter of anterior ocel-
lus.

Thoracic structure similar to that of male, but dorsum
duller, interspaces minutely lineolate; angles of propodeum
distinctly short-dentiform.

Legs as described for male, but without protrochanteral
process.

Gaster as described for male, with usual sexual differences.

Pilosity much shorter than in male, that of front of head

Contributions in Science, Number 378

Snelling: Australian Paragiine Wasps 5

about as long as an ocellar diameter; that of pronotal dorsum
a little shorter, the hairs of the middle portion more or less
spatulate at tips, a few longer hairs along posterior portions
of humeral angle; hairs of mesoscutum distinctly shorter;
hairs of side of thorax longer than on pronotal dorsum; gastral
hairs very short and sparse (longer and more abundant cau-
dad), but with abundant fine, appressed hairs.

Head black, the following yellow-orange: mandible, except
reddish margins and apical teeth; clypeus; broad band across
lower frons, ending about midway between clypeus and an-
terior ocellus, and with a narrow stripe along inner orbit
nearly to base of mandible; broad upper outer orbital mark.

Thorax similar to that of male, but marks yellow-orange
and with the following additional markings: narrow, rect-
angular posteromedian mark on mesoscutum; most of dor-
sum of scutellum (but not reaching anterior margin); large
spot on lower plate of mesopleuron; propodeal mark ex-
tended onto side. Legs yellow-orange. Gastral terga yellow-
orange, with irregular, obscure basal areas dusky, last seg-
ment mainly dusky; ventral segments ferruginous.

TYPE MATERIAL (All Western Australia)

Holotype male, eight male and four female paratypes: 43 km
ENE Landor Homestead (25°08'S, 1 16°54'E), 23 Aug. 1984
(T.F. Houston and B.P. Hanich, no. 596-1), on flowers of
Goodenia berardiana (Goodeniaceae). Additional paratypes:
1 <3, 2 9$, 16 km WSW Lyons River Homestead (24°38'S,
1 1 5°20'E), 30 Aug.-l Sept. 1980 (C.A. Howard and T.F.
Houston, no. 344-9), on flowers of G. berardiana; 5 68, 9 km
SW Gifford Creek Homestead (20°03'S, 116°13'E), 2 Sept.
1980 (C.A. Howard and T.F. Houston, no. 350-1), on flowers
of G. berardiana. Holotype and most paratypes in WAM;
additional paratypes in BMNH and LACM.

ETYMOLOGY

The specific name is an arbitrary combination.
DISCUSSION

All the males are very similar to one another, except for
minor variations in color. One of the Gifford Creek males
has a minute preapical yellowish spot on the right antennal
scape; presumably, males with more conspicuously maculate
scape will eventually be discovered. Two males have the
lower plate of the mesopleuron wholly black and two (in-
cluding the holotype) have a small irregular yellowish spot
a short distance in front of the posterior margin.

The six available females are very similar to one another.
The distally clavate hairs on the pronotal dorsum are an
unusual feature; they are not present in the males, nor in
females of other Australian paragiines that I have examined.

Riekia Richards

Riekia Richards, 1962:54-55. Type-species: Riekia nocatun-
ga Richards, 1962; monotypic and original designation.

This genus was established by Richards (1962) for a species

known only from females from New South Wales. Additional
females of R. nocatunga were recorded from New South
Wales by Richards ( 1 968). In the latter paper a second species,
R. angulata, was described from females from Queensland
and New South Wales; in my opinion this is not a Riekia,
but rather a species of Rolandia (see below).

In his key to the Australian paragiine genera, Richards
stated that the first recurrent vein (1 m-cu) is received by the
first submarginal cell. In the description of Riekia, however,
it is stated that the vein is received in the second submarginal
cell. In the two specimens of R. nocatunga (including a para-
type) that I have seen, the latter statement is correct, and
presumably that in the key is a lapsus.

A further source of possible confusion lies in the charac-
terization of the occipital carinae or “keels.” In the key Rich-
ards states: “Occipital keels not developed.” The detailed
description of Riekia does not agree: “Dorsal occipital keel
not developed, ventral occipital keel long and strong . . .”.
In the specimen I have studied the preoccipital carina
(= “dorsal keel” of Richards) is present and well developed
along the posterior genal margin from the occiput to the base
of the mandible. There is no postocular carina (see TER-
MINOLOGY). The table by Richards (1962) on p. 12 in-
dicates that both are absent in Riekia.

In addition to the characters cited by Richards that dis-
tinguish between Riekia and Rolandia, it should be noted
that in the latter genus the anterior margin of the pronotum
is abruptly angulate opposite the base of the procoxa (Fig.
29). In Riekia the margin is evenly and very broadly curved
(much as in Fig. 28). The lateral furrow of the pronotum is
sharply defined, crenate, and extends to the posterior margin
of the segment, above the lower comer, in Riekia. There is
no lateral furrow in Rolandia. Instead there is a blunt ridge
that extends dorsomesad from the angulation of the anterior
margin, behind which the disc is very broadly and shallowly
depressed. The depressed portion is crossed by a few widely
spaced, weak rugulae.

Riekia sp.

Figures 10, 19-21

A male specimen from 43 km ENE Landor Homestead
(25°08'S, 1 16°54'E), Western Australia, 23 Aug. 1984 (T.F.
Houston and B.P. Hanich, TFH #596-1; WAM), on flowers
of Goodenia berardiana, is obviously a species of Riekia.
There is, however, no certainty that it is the opposite sex of
R. nocatunga, and I suspect that it is not. More material
must be available before the specific status of this male can
be determined.

The male will key to Riekia and agrees generally with the
original description of Riekia, except for the expected sexual
differences. Agreement with females of R. nocatunga, sexual
differences aside, is good, but the shape of the scutellum is
different; the posterior margin is broad and abruptly decliv-
itous, rather than gradually descendant to the metanotum.
The pretegular carina is distinct. There are differences, too,
in surface sculpturing but these might be sexual differences.

The last three flagellar segments are slightly concave be-

6 Contributions in Science, Number 378

Snelling: Australian Paragiine Wasps

15

Figures 9-15. Riekia and Rolandia spp. 9-11, apex of male antenna of Rolandia houstoni, Riekia sp., and Rolandia borreriae, respectively.
12, female gastric apex, Rolandia houstoni. 13, 14, base of gaster of Riekia nocat unga and Rolandia houstoni. respectively. 15, female thorax,
dorsal view, Rolandia houstoni.

Contributions in Science, Number 378

Sneliing: Australian Paragiine Wasps 7

neath and the apical segment is long and distinctly curved
(Fig. 10). The protrochanter lacks a ventral process and the
profemur is narrow at the base. Genitalic and associated
structures are shown in Figures 1 9-2 1 ; the parameral spine
is broad and thick, not at all hook-like.

Rolandia Richards

Rolandia Richards, 1962:57. Type-species: Paragia macu-
lata Meade- Waldo, 1910; monotypic and original desig-
nation.

In describing Rolandia, Richards stressed the reduced num-
ber of segments in the maxillary palpus (five, or rarely six,
segments) and the obtusely truncate clypeus, as seen in pro-
file. The following additional species agree generally with the
description of Rolandia, but have the clypeal profile normal
for a paragiine, i.e. low and somewhat convex for most of
its length and flattened or a little concave toward the apex.

Other departures from the original characterization of Ro-
landia include the presence of a minute tooth on the tarsal
claws; the propodeal angle may be obtuse and not at all spine-
like; the hind wing may possess a very small anal lobe (this
actually is a correction, for while Richards stated that the
anal lobe is absent, it is present though quite small in R.
maculata).

The four species of Rolandia may be separated as follows.

KEY TO SPECIES OF ROLANDIA

la. Apical portion of clypeus, in profile, flattened and thin
(Figs. 26, 27); longest hairs in center of second gastral
tergum about one-half as long as transverse diameter of
anterior ocellus; scutellum finely, contiguously punctate
and with longest hairs on disc shorter than diameter of

anterior ocellus 2

b. Anterior portion of clypeus, in profile, obtuse (Fig. 25);
longest hairs in center of second gastral tergum about as
long as transverse diameter of anterior ocellus; disc of
scutellum coarsely rugosopunctate and with erect hairs
very long, some more than twice diameter of anterior

ocellus maculata (Meade-Waldo)

2a. Punctures on discs of second and third gastral terga clear-
ly delimited by slightly shiny interspaces; hairs of female
clypeus whitish and distinctly shorter than transverse
diameter of anterior ocellus; male profemur broad at
base (Fig. 18) and flagellar segments beyond first no

longer than broad 3

b. Discs of second and third gastral terga contiguously
punctate and without definite interspaces; hairs of female
clypeus yellow to reddish, mostly distinctly longer than
transverse diameter of anterior ocellus; male profemur
narrow at base (Fig. 1 7) and flagellar segments beyond

first distinctly longer than broad

borreriae, new species

3a. Frons weakly shiny, contiguously finely punctate and
without shiny, linear, raised interspaces, midline raised
and shiny, ending short of anterior ocellus by about its
own length; scutellum, in profile, evenly sloping to meta-

notum, without definite posterior face; maxillary palpus

six-segmented angulata (Richards)

b. Frons with numerous irregular, elongate, shiny inter-
spaces between otherwise contiguous punctures; midline
present as a short interanntenal tubercle separated from
anterior ocellus by several times its own length; scutel-
lum, in profile, with a short, abruptly descending pos-
terior face; maxillary palpus five-segmented

houstoni, new species

Rolandia angulata (Richards),
new combination

Riekia angulata Richards, 1968:101-102. 2.

I have been able to examine a paratype of this species and
cannot agree with its placement in Riekia; it is possible that
its description in Riekia is a lapsus. The paratype differs from
a paratype of Riekia nocatunga Richards, 1962 (the type-
species of Riekia), in the following characteristics, all shared
with Rolandia: the glossa is not retractile; ammochaetae are
present along the genal margin; the anterior margin of the
propleuron is angulate; the propodeal angle is obtuse and
ends in an obtusely tooth-like projection; gastral tergum 2 is
constricted at the base.

This species was described from females only from
Queensland and New South Wales. I have examined one
paratype female; it differs from R. houstoni in the features
discussed under that species. Although males of R. angulata
are unknown, they will presumably differ from those of R.
houstoni by the same features that separate the female in the
key.

In addition to a paratype of R. angidata, I have examined
eight females from Cunnamulla, Queensland, 7 Oct. 1949
(N. Geary; NMV).

Rolandia houstoni, new species

Figures 9, 12, 14, 15, 18, 22, 23, 26, 29

DIAGNOSIS

Mandibular teeth of female subacute and upper margin not
conspicuously concave basad of inner tooth; clypeus not ob-
tuse in profile and apical margin broadly truncate; scutellum
with short, abruptly descendant posterior face; midline of
face reduced to short interantennal tubercle.

DESCRIPTION

MALE. Measurements. Holotype head width 2.61; head
length 2.58; wing length 6.77; total length 11.2 mm. Para-
types: head width 2.48-2.63; head length 2.42-2.58; wing
length 6.56-6.82; total length 10.7-11.4 mm.

Head 1.01-1.03 times as broad as long; mandible short
and broad, tridentate, teeth subacute, inner tooth smallest,
upper margin basad of inner tooth weakly convex. Clypeus
about 1.2 times as broad as long, gently convex in profile,
merging into weakly concave apical one-third; apical margin,
in frontal view, truncate, truncate portion subequal to dis-

8 Contributions in Science, Number 378

Snelling: Australian Paragiine Wasps

Figures 16-24. Rolandia and Riekia sp. 16-18, protrochanter and femur base of Rolandia maculata, R. borreriae, and R houstoni. 19-21,
male gastric apex (ventral) and genital capsule (ventral and lateral) of Riekia sp. 22-24, male genital capsules of Rolandia houstoni (ventral
and lateral) and R. borreriae; scale line for 19-24 = 0.50 mm.

Contributions in Science, Number 378

Snelling: Australian Paragiine Wasps 9

tance from end to truncation to lateral angle of clypeus; disc
moderately shiny between dense fine to moderate punctures.
Interantennal tubercle low, obtuse, its lower facet moderately
shiny and impunctate. Frons dull, finely rugosopunctate, with
occasional flat interspaces that are dull and finely lineolate;
vertex and preoccipital areas finely, contiguously to subcon-
tiguously punctate, interspaces shinier than on frons. Inter-
ocellar distance about two times diameter of anterior ocellus;
ocellocular distance subequal to interocellar distance; ocel-
loccipital distance about three times diameter of anterior
ocellus. Antennal scape shorter than interantennal distance;
first flagellar segment a little less than four times as long as
thick, about four-fifths as long as scape, and about twice as
long as following segment; segments gradually broader dis-
tad, antepenultimate segment about twice as thick as first
and about as long as broad; apical segment short and stout
(Fig. 9).

Dorsum of pronotum dull, moderately rugosopunctate;
pretegular groove well defined; side of pronotum with broad,
shallow impressed area behind anterior margin that does not
reach anterior angle, depressed area with a few very short
longitudinal ridges that originate at anterior margin (Fig. 29).
Mesoscutum slightly shiny, finely to moderately rugoso-
punctate. Scutellum dull and roughened between dense to
subcontiguous, obscure, fine punctures; axilla with a few, fine,
longitudinal rugulae. Mesopleuron slightly shiny and rough-
ened between close to dense fine punctures. Side of propodeal
angulation, in dorsal view, straight, its terminus bluntly
rounded; dorsal area moderately shiny and roughened and
moderately shiny between sparse obscure punctures that are
a little coarser than those of dorsal area; side shiny and weakly
roughened between scattered fine punctures.

Process of protrochanter large, broadly digitiform in pos-
terior view (Fig. 18) and slightly curled forward; profemur
compressed over basal one-half, ventral margin acute, an-
gulate near base.

Gastral tergum 1 with dorsal face moderately shiny be-
tween dense, fine punctures; punctures of remaining terga
subcontiguous, progressively finer on succeeding segments;
sterna moderately shiny, with mixed minute and fine punc-
tures that become very obscure toward margins; middle of
juncture of sterna 7 and 8 deeply and narrowly impressed;
genitalia as in Figures 22 and 23.

Pilosity abundant, hairs fine and mostly shorter than di-
ameter of anterior ocellus, but a few longer hairs on side of
frons, on mesopleuron, and on gastral sterna; mesoscutum
with some very short, erect, brownish hairs.

Head black, the following yellowish white: mandible; clyp-
eus; side of face from above ocular sinus nearly to base of
mandible; large frontal mark, very narrowly separated from
lateral face mark. Underside of scape and large spot on upper
outer orbit yellow-orange.

Thorax black, with yellow-orange marks: pronotum (dusky
ferruginous near tegula and on propleuron); tegula; obscure
posterolateral spot on mesoscutum and adjacent axilla; most
of disc of scutellum, not attaining anterior margin; middle
one-third of metanotum; large spot on mesopleuron below
wing base; lateral angle of propodeum. Legs mostly dusky

ferruginous, but coxae largely blackish and profemur yellow-
ish apicad and along lower margin. Wings dusky, veins red-
dish basad, becoming brownish apicad.

Gastral segments dusky ferruginous, becoming darker cau-
dad; terga 1-5 with lateral yellow blotches, largest on second,
progressively smaller on segments 3-5; terga 1-6 with more
or less quadrate yellow blotch or spot on apical middle; sterna
without defined yellowish marks, all segments more or less
ferruginous to dusky.

FEMALE. Measurements. Head width 2.48-2.65; head
length 2.45-2.58; wing length 6.56—7. 1 3; total length 1 0.9—
12.5 mm.

Head 1.00-1.04 times as broad as long. Otherwise similar
to male, with following differences: frons and vertex duller,
a little more coarsely rugulose; ocellocular distance distinctly
greater than interocellar distance; first flagellar segment about
twice as long as thick; flagellar club weakly defined.

Thorax about as described for male, but mesoscutum more
sharply rugosopunctate; protrochanter and profemur simple.

Gaster about as described for male, except usual sexual
differences.

Pilosity about as in male, but everywhere shorter, espe-
cially on head and thorax; profemur with ventral fringe of
long hairs, some longer than apical thickness of femur.

Markings about as in male, and uniformly orange-ferru-
ginous, but: frontal and lateral face marks confluent; entire
pronotum colored; mesoscutum with irregular lateral marks
along posterior three-fourths; scutellum largely colored; dor-
sal area and disc of propodeum colored; yellow marks of
gastral dorsum weakly defined.

TYPE MATERIAL (All Western Australia)

Holotype male, four male and four female paratypes: 10 km
ESE Meedo Homestead (25°40'S, 114°37'W), 23-26 Aug.
1 980 (C.A. Howard and T.F. Houston, no. 336-8), on flowers
of Goodenia berardiana (Goodeniaceae). Additional para-
types: 1 <5, 1 2, 7 km N Boologooro Homestead (24°29'S,
1 13°42'W), 27-29 Aug. 1980 (C.A. Howard and T.F. Hous-
ton, no. 338-13), on G. berardiana; 1 <3, same data except no.
338- 1 1 5, on Pileanthus peduncu/aris (Myrtaceae); 1 2, 1 6 km
WSW Lyons River Homestead (24°38'S, 1 1 5°20'W), 30 Aug.-
1 Sept. 1980 (C.A. Howard and T.F. Houston, no. 344-9)
on G. berardiana; 1 2, same data except no. 344-24, ex burrow
in sand. Holotype and most paratypes in WAM; three para-
types in LACM.

ETYMOLOGY

This species is dedicated to Terry F. Houston who collected
many of the type series and who has made available much
of the material on which this paper is based.

DISCUSSION

Both sexes of R. houstoni and R. angulata may be separated
from R. maculata by the shape of the clypeal profile; in R.
maculata the clypeus is obtusely truncated at the apex (Fig.
25) while that of R. angulata and R. houstoni (Fig. 26) is not

10 Contributions in Science, Number 378

Snelling: Australian Paragiine Wasps

much different from that of other paragiines. The sculpture,
especially of the dorsum of the thorax, is much less coarse
in R. angulata and R. houstoni.

In R. maculata the mandibular teeth are broad, with
rounded apices and the upper margin, basad of the inner
tooth is concave for a short distance. The mandibular teeth
are narrower and more acute, and the upper margin does not
have the characteristic profile in R. angulata and R. houstoni.

I have seen males of R. maculata. and the genital structures
suggest further differences between the two species. The dig-
itus (Fig. 23) is shorter and less digitiform in R. houstoni.
The ventral process of the paramere is sharply narrowed to
an acute apex in R. maculata; in R. houstoni the process is
approximately parallel-sided for most of its length and the
apex is narrowly rounded (Fig. 23).

This species is very similar to R. angulata. Although males
off?, angulata are not presently known, they will presumably
differ from those of R. houstoni in the same characters as do
the females.

In addition to the characters cited in the key, females of
C. houstoni differ from the paratype of R. angulata in the
following: clypeal interspaces shiny (dull in R. angulata ); in
dorsal view, anterior margin of pronotal collar slightly con-
vex and humeral margin broadly rounded (in R. angulata,
straight, humeral margin subangulate); long hairs on pronotal
dorsum about 0.1 mm long (about 0.2 mm long in R. an-
gulata).

Rolandia borreriae, new species

Figures 11, 17, 24

DIAGNOSIS

Clypeus of both sexes low-convex; male flagellar segments
all longer than broad, last segment slightly curved and acute
at apex (Fig. 11); male profemur narrow at base (Fig. 17);
hairs of female clypeus stout and bristle-like, distinctly longer
than transverse diameter of anterior ocellus.

DESCRIPTION

MALE. Measurements. Holotype head width 2.65; head
length 2.48; wing length 6.23; total length 11.1 mm. Para-
types: HW 2.45-2.68; HL 2.33-2.50; WL 5.91-6.33; TL 10.2-
10.9.

Head 1.05-1.07 times as broad as long; mandible short
and broad, tridentate, teeth acute on strongly oblique margin,
inner tooth much the shortest, dorsal margin weakly convex.
Clypeal disc moderately shiny and microscopically lineolate
between contiguous to subcontiguous, round to oblong punc-
tures. Interocellar distance 2.00-2.30 times anterior ocellus
diameter; ocellocular distance 3. 1 2-3.40 times anterior ocel-
lus diameter. Antennal scape a little shorter than interanten-
nal distance; first flagellar segment about 0.8 times scape
length and about 1.7- 1.8 times length of second segment; all
flagellar segments longer than broad, last segment about 2.4
times as long as broad, slightly curved in profile, apex acute
(Fig. 1 1).

Pronotal humerus broadly rounded, transverse carina ex-
tended nearly to posterior margin; propleural ridge short,
weak, curved; pretegular groove distinct and narrow, preteg-
ular carina weak; disc slightly shiny, punctures moderate to
coarse, mostly contiguous. Mesoscutum slightly shiny, finely
to coarsely vermiculate-punctate. Disc of scutellum dull and
roughened between subcontiguous to close moderate punc-
tures; lateral portions slightly shiny between coarse, longi-
tudinal rugae; disc, in profile, evenly and strongly curved
from anterior declivity to posterior margin. Metanotum dull
and densely tessellate across middle, laterally a little shinier
between coarse, longitudinal rugae. Disc of mesopleuron dull
and densely tessellate between subcontiguous to close, fine
punctures, becoming moderate and contiguous above. Pro-
podeal angle obtuse; disc and laterobasal surfaces slightly
shiny between fine punctures that are mostly contiguous on
disc, becoming close to sparse on laterobasal area; side shiny
between scattered minute punctures.

Process of protrochanter much longer than segment (Fig.
17); profemur narrow at base.

Dorsal face of gastral tergum 1 slightly shiny between con-
tiguous to subcontiguous minute punctures; following terga
dull, punctures in middle of tergum 2 so dense as to be
indistinguishable, becoming slightly more separated laterad
and on following segments; all segments with scattered, bare-
ly discemable, fine, setigerous punctures. Gastral sterna shin-
ier, but otherwise similar, fine setigerous punctures more
distinct. Genital capsule shown in Figure 24.

Very short, fine hairs abundant on thorax and gaster, most-
ly appressed, but erect on mesoscutum; brownish on meso-
scutum, lighter elsewhere and white on mesopleuron. Front
of head with numerous long, slender hairs, longest on clypeus
equal to about diameter of anterior ocellus and longest on
vertex equal to about twice diameter of anterior ocellus; hairs
on pronotal dorsum a little longer than on clypeus; longest
hairs on gastral dorsum equal to about one-half diameter of
anterior ocellus.

Head black, the following whitish: mandible, except red-
dish teeth; clypeus; large supraclypeal mark; narrow stripe
along inner orbit almost to top of eye; large postocular spot;
underside of scape. Upperside of scape and flagellum black-
ish, flagellar segments irregularly dull reddish beneath.

Thorax black, the following yellowish white: broad band
on pronotal collar, short stripe anterior to tegula, joined to
stripe across posterior lobe; tegula, except small ferruginous
spot; large subtegular mark; small mesoscutal spot adjacent
to tegula; disc of scutellum; large mark along lateral ridge of
propodeum. Legs mostly yellowish, but upper side of femora
yellowish ferruginous. Wings slightly brownish, veins brown,
becoming reddish toward base.

Gaster blackish brown, the following yellowish white: api-
cal bands on terga 1-5, narrowed sublaterally and broadened
at extreme side; similar on tergum 6, but only weakly, or not
at all, narrowed sublaterally; tergum 7 brownish ferruginous;
sterna dusky ferruginous, with irregular blackish bands basad
and transverse, preapical whitish mark on middle of seg-
ments 2-5.

FEMALE. Measurements. Head width 2.60-2.83; head

Contributions in Science, Number 378

Snelling: Australian Paragiine Wasps 11

length 2.50-2.65; wing length 6.70-6.93; total length 1 0.6—
1 1.7 mm.

Head 1.04-1.07 times as broad as long. Otherwise similar
to male, with following differences: interocellar distance 2.08-
2.27 times anterior ocellus diameter; ocellocular distance
2.75-3.00 times anterior ocellus diameter; interantennal dis-
tance 1.31-1.35 times scape length; first flagellar segment
0.51-0.56 times scape length and 2.09-2.30 times length of
second flagellar segment; flagellar segments, except first,
broader than long.

Thorax about as in male but mesoscutum more rugoso-
punctate; protrochanter and profemur simple.

Gaster about as described for male except usual sexual
differences.

Pilosity about as in male, but everywhere shorter; clypeus
with abundant suberect, bristle-like, slightly brownish hairs
with attenuate apices. Profemur abundantly hairy on ventral
and posterior faces, but without definite fringe.

Markings about as in male, but some black partially re-
placed by ferruginous, especially on pronotum and gaster;
lateral mark on mesoscutum over half of segment length;
axilla yellowish; tergum 6 ferruginous. Legs yellowish fer-
ruginous, becoming yellowish externally on tibiae and basi-
tarsi.

TYPE MATERIAL

Holotype, 3 male and 4 female paratypes: Cooper Creek
(12°06'S, 133°04'E), 19 km E by S Mt. Borradaile, Northern
Territory, 5-6 June 1973 (J.C. Cardale), on flowers of Bor-
reria exserta (Rubiaceae). Holotype and most paratypes in
ANIC; one paratype pair in LACM.

ETYMOLOGY

Of, or pertaining to, the plant genus Borreria on which the
type series was collected.

DISCUSSION

Males are easily separable from those of other Rolandia species
by the elongate flagellum, with the apical segment curved
and pointed, and by the narrow profemur base and com-
paratively large protrochanteral process. Similarly, the fe-
male, although superficially similar to R. angulata and R.
houstoni, is characterized by the peculiar, bristle-like hairs
of the clypeus. In both of the other species the hairs are
slender and flexuous and not notably different from those of
the frons except by being shorter. They are whitish in color,
rather than brownish yellow, as in R. borreriae.

Paragia Shuckard

Paragia Shuckard, 1837:81. Type-species: Paragia decipiens
Shuckard, 1837; monotypic.

Metaparagia Meade-Waldo, 191 1:748. Type-species: Para-
gia pictifrons F. Smith, 1857; original designation. NEW
SYNONYMY.

Paragia subg. Cygnaea Richards, 1962:53, 60. Type-species:

Paragia vespiformis F. Smith, 1865; monotypic and orig-
inal designation. NEW SYNONYMY.

Paragia subg. Paragiella Richards, 1962:53, 67. Type-species:

Paragia odyneroides F. Smith, 1850; original designation.

NEW SYNONYMY.

Metaparagia was established for two species, both apparently
rare and known from unique female type specimens. These
types are in the BMNH and 1 have examined both. While I
agree with Richards that they are distinct species, I do not
agree that there is any justification for the separation of Meta-
paragia from Paragia.

Several of the features used by Richards to distinguish
between Metaparagia and Paragia are based on the relative
expression of a given character. Thus, the development of
the ocular sinus, amount of ventral “excavation” of the head,
the shape of the pterostigma, the development of the furrow
at the base of gastral sternum 2, development of the occipital
carinae, the relative size of the hindwing anal lobe, and the
development of the tooth of the tarsal claw, all are characters
of this type. Their expression is matched by one or more of
the more typical species of Paragia, and, in any case, none
of these characteristics can be readily quantified, and I attach
no significance to them.

The structure of the pronotum is not quite like that of
other Paragia: the pretegular groove and carina are absent,
as in the groove of the propleuron. All other Paragia do
possess a pretegular groove, but it is weak in some species
that also lack the pretegular carina. Similarly, the groove
across the disc of the propleuron seems always to be present,
though it is weak in some species. Again, I believe the char-
acter states seen in the two Metaparagia are the end point
of a continuum and attach to them no significance beyond
species level value. Much the same may be said of the la-
melliform humeral carina, a conspicuous feature not stressed
by Richards.

The only truly distinctive characteristic is the pyriform
tegula of Metaparagia, contrasted with the subcircular tegula
of Paragia. In the two species assigned to Metaparagia, the
inner, posterior angle of the tegula is distinctly produced.
The surface of the tegula bears a few irregular rugulae, but
is otherwise smooth and shiny. Since variations in tegular
shape are not unusual in other vespoid genera, I do not
believe this character state to be of value above species-group
level.

Metaparagia is therefore to be considered a junior syn-
onym of Paragia and its two species, M. doddi Meade-Waldo,
1911, and M. pictifrons (F. Smith), transferred to that genus
as NEW COMBINATIONS.

Richards (1962), in his study, recognized a total of 19
species of Paragia in Australia, and added two more in 1 968.
These he assigned to three subgenera: Paragia Shuckard,
1837, and two new subgenera, Cygnaea and Paragiella.

The most distinctive of these is the subgenus Cygnaea,
largely because of unusual modifications of the abdominal
segments. However, I am philosophically opposed to mono-
typic subgenera, especially within genera as small as Paragia.
For this reason I treat Cygnaea as a synonym of Paragia.

12 Contributions in Science, Number 378

Snelling: Australian Paragiine Wasps

The only included species in Cygnaea is P. vespiformis F.
Smith, 1865.

The distinctions that Richards attempted to enumerate
between the subgenera Paragia and Paragiella are weak and
unconvincing. Even Richards noted these difficulties when
he cited species in each of his subgenera that were exceptional
to the characterization in one or more features. The several
new species described below further weaken the differences
between these two groups. No distinctive features were cited
by which these two subgenera were to be separated. Most
characteristics were cited as “usually” or as “more or less.”
Others, described in less ambiguous terms, were noted to
have exceptions. The two subgenera are thus perceived to be
coextensive through continuous series of character state vari-
ation.

Richards’ key to the species of Paragiella is not satisfactory
since many of the features are cited in subjective terms and
a user without access to a good representative collection will
experience difficulty and uncertainty in interpreting the key
statements. Since I have not been able to examine all the
species, I cannot offer a better key at this time. Similarly, the
treatment of previously described species is unsatisfactory
since Richards usually commented largely on the color pat-
terns of these species, rather than important morphological
characteristics. Color patterns in these masarids are very un-
reliable since there is a considerable amount of variation in
most species. There are also several complexes of mimetic
color patterns.

Paragia oligomera, new species

DIAGNOSIS

Separable from all other Paragia by the reduced palpal seg-
mentation: three labial and five maxillary palpal segments;
in addition: propodeal process obtuse; distance between clyp-
eus and eye equal to diameter of anterior ocellus; mesopleu-
ron rugosopunctate and propodeal disc irregularly rugose be-
tween close, coarse punctures; metanotum protuberant and
with horizontal basal face.

DESCRIPTION

FEMALE. Measurements. Holotype head width 5.03; head
length 4.36; wing length 12.21; total length 22.8 mm. Para-
type: head width 4.51; head length 3.85; wing length 11.13;
total length 18.4 mm.

Plead 1.15-1.17 times as broad as long. Mandibular teeth
broad and blunt. Clypeus separated from inner eye margin
by about diameter of anterior ocellus; about 1 . 1 times as
wide as long; apical truncation broadly rounded, truncation
narrower than distance from its end to lateral angle of clyp-
eus; disc dull, anterior portion subcontiguously, finely punc-
tate, grading to coarsely rugosopunctate over most of disc.
A short carina-like ridge present supramesad of each antennal
socket. Frons and vertex moderately to coarsely rugoso-
punctate; preocciput similar but punctures coarser and sub-
contiguous in some areas, especially laterad. Preoccipital and

postocular carinae both present and well developed. Inter-
ocellar distance two times diameter of anterior ocellus; ocel-
locular distance about three times diameter of anterior ocel-
lus; ocelloccipital distance a little greater than ocellocular
distance. Antennocular distance about twice diameter of an-
tennal socket; interantennal distance more than three times
diameter of antennal socket; scape a little longer than inter-
antennal distance and about 1.7 times longer than first fla-
gellar segment; flagellum not clavate, antepenultimate seg-
ment a little broader than long.

Humeral angles of pronotum present, obtuse; dorsal face
and propleuron above lateral furrow dull to slightly shiny,
coarsely rugosopunctate; area anterior to lateral furrow mod-
erately shiny between dense, moderate punctures. Mesoscu-
tum moderately rugosopunctate, with numerous linear in-
terspaces. Scutellum dull, coarsely rugosopunctate to anterior
margin; lateral scutellar furrow distinct. Metanotum with
both horizontal and vertical faces, their juncture subcarinate,
fully visible in dorsal view. Mesopleuron moderately shiny
between moderate to coarse, subcontiguous to contiguous
punctures. Propodeal angles obtuse; dorsal face slightly shiny
between subcontiguous moderate punctures; disc moderately
shiny between short irregular rugae and close moderate punc-
tures; side moderately shiny between dense, moderate punc-
tures.

Probasitarsus a little more than twice longer than wide,
margins nearly parallel; tibiae each with apical row of short,
flattened ferruginous setae, those of metatibia difficult to see
under long, dense, reclinate hairs that cover most of outer
face; metabasitarsus without paired row of stout ferruginous
setae along anterior margin; tarsal claws each with a long
tooth.

Anal lobe of hindwing small but distinct; r-m perpendic-
ular to Rs and M.

Gastral terga slightly shiny between moderate, dense punc-
tures that become fine on tergum 5; sterna moderately shiny
between close to dense moderate punctures, interspaces
densely minutely punctate.

Pilosity abundant on frons and vertex, longest hairs on
frons shorter than diameter of anterior ocellus and longest
on occiput distinctly longer than diameter of anterior ocellus;
long hairs of genal fringe more than twice diameter of anterior
ocellus. Pronotal dorsum with a few long hairs across middle
one-third, longest about equal to ocellar diameter, hairs oth-
erwise very short and sparse; mesoscutum with a few scat-
tered moderately long hairs, about as long as on frons, but
mostly with sparse, very short stiff hairs; sides of thorax with
numerous erect, slender hairs, some equal to ocellar diam-
eter, or a little greater. First gastral tergum with sparse, mod-
erately long, erect, slender hairs at sides and across base,
hairs short, stiff, scattered on dorsal face; second tergum with
scattered very short, stiff hairs; following terga similar, but
hairs progressively longer on succeeding segments. Femora
and tibiae with dense, stout, bristle-like hairs on all except
dorsal faces of femora, some of which are more or less dis-
tinctly curled at tips; tarsi with dense, shorter, straight hairs.

Contributions in Science, Number 378

Snelling: Australian Paragiine Wasps 13

Black, the following yellowish: clypeus, except margins and
large, bifurcate mark on median lobe; stripe mesad to an-
tennal socket; large irregular blotch on inner orbit at ocular
sinus; large outer orbital spot; broad band across pronotal
collar, extended posteriorly at each end, and on posterior
comer in front of tegula; most of tegula; small posterolateral
spot on mesoscutum; band across posterior scutellar disc,
sharply narrowed in middle; large mesopleural spot; pro-
podeal angles; broad posterior bands on gastric terga 1-4,
slightly broadened on each side; large median spot on tergum
6; broad marginal bands on sterna 2-4, that on sternum 3
broadest. Legs black, apices of femora and tibial stripe yel-
low. Wings brownish, with dark brown veins; most of mar-
ginal cell and parts of first submarginal, radial, and first dis-
coidal cells clouded.

The paratype is similar in color to the holotype, but the
mandibles are ferruginous, the clypeus does not have a black
preapical spot, a dull reddish interantennal spot is present;
the pronotal dorsum is almost wholly yellow with an obscure,
reddish area anterior to tegula, mesoscutum has a large pos-
teromedian yellow spot, disc of scutellum largely yellow, and
legs are largely yellow.

TYPE MATERIAL (All Western Australia)

Holotype female: 10 km NE Wanneroo, 4 Jan. 1982 (T.F.
Houston, no. 418-3), on flowers of Regelia ciliata (Brome-
liaceae). Paratype: 9, Mogumber, no date (A. Douglas, no.
53-3310). Both in WAM.

ETYMOLOGY

Combines the Greek oligos (few) with meros (part), in ref-
erence to the reduced palpal segmentation.

DISCUSSION

This species differs from all other Paragia in the reduced
number of palpal segments: all others have four labial and
six maxillary palpal segments. This is a member of the sub-
genus Paragiella as defined by Richards (1962) and in his
key to the species will fail at the third couplet because it does
not accord with either alternative. The three species in that
section of the key ( P . venusta F. Smith, 1865, P. generosa
Richards, 1962, and.P. australis Saussure, 1853) are all small-
er, less than 1 5 mm long, and are more finely and less closely
punctate; in none are the head and thorax rugosopunctate.

Paragia magda/ena Turner

Paragia magda/ena Turner, 1908:89. 9. Richards, 1962:67,
68-69.

This species is still known only from the type in the BMNH,
from Mackay, Queensland. It is separable from all species
of Paragiella, except P. confluens, by the lack of a propodeal
process; for differences between these two species, see the
description of P. confluens. Richards noted that the whole
body is covered with pubescence; in particular, the discs of
the gastric terga bear abundant suberect hairs, the longest

hairs on the third segment longer than the diameter of the
anterior ocellus. Conspicuously longer hairs are present on
the first tergum and on the propodeum. Surface sculpture
tends to be coarsely and contiguously punctate to rugoso-
punctate, but the clypeus is shiny between contiguous to
subcontiguous punctures and the frons is roughened between
moderate to coarse, mostly subcontiguous punctures. The
second tergum is likewise dull, the punctures fine to moderate
and subcontiguous to dense. Total length is 16.7 mm. The
wings are too badly frayed to be accurately measured, but
Richards gives wing length as 10.5 mm. The color pattern
has been described by Richards.

Paragia confluens, new species

Figure 28

DIAGNOSIS

Belongs to Paragiella as defined by Richards (1962); differs
from other species (except P. magdalena) in lacking lateral
processes on propodeum; differs from P. magdalena in pos-
sessing anal lobe on hindwing and entire posterior face of
propodeum without obvious punctures. This species is fur-
ther characterized by the short, broad clypeus, asymmetrical
protarsal segments 2-4, possessing only a single ferruginous
seta at apex of meso- and metatibiae, and there are few or
no erect, apically hooked, hairs on the ventral femoral sur-
faces.

DESCRIPTION

FEMALE. Measurements. Holotype head width 2.65; head
length 2.35; wing length 7.28; total length 10.8 mm. Paratype:
head width 2.42; head width 2.19; wing length 6.77; total
length 9.9 mm.

Head 1.10-1.12 times as broad as long; mandibular teeth
broad and blunt. Clypeus nearly touching inner eye margin;
about 1.6 times as broad as long; apical truncation straight,
wider than distance from its end to lateral angle of clypeus;
disc slightly shiny, finely lineolate between dense, fine, elon-
gate punctures, becoming subrugose in apical area. Frons
slightly shiny, interspaces distinctly roughened between sub-
contiguous, fine punctures; vertex and preocciput similar, but
with some larger punctures. Preoccipital and postocular ca-
rinae congruent. Interocellar distance about two times di-
ameter of anterior ocellus; ocellocular distance about three
times diameter of anterior ocellus; ocelloccipital distance
subequal to ocellocular distance. Antennocular distance less
than diameter of antennal socket; interantennal distance more
than six times diameter of antennal socket; scape about two-
thirds as long as interantennal distance and a little more than
twice as long as first flagellar segment; flagellum stout and
clavate, antepenultimate segment about twice as wide as long.

Pronotum without humeral angles, broadly rounded onto
side; pretegular groove absent, but with a distinct pretegular
carina; propleural furrow distinct; pronotal dorsum and pro-
pleuron above furrow dull, finely, contiguously punctate; pro-
pleuron below furrow moderately shiny, roughened between
dense to close, fine punctures. Mesoscutum slightly shiny,

14 Contributions in Science, Number 378

Snelling: Australian Paragiine Wasps

Figures 25-31. Rolandia and Paragia spp. 25-27, lateral view of head of female, R. maculata, R. houstoni, and P. monocesta. 28, 29, lateral
view of pronotum, P. confluens, and R. houstoni. 30, 31, P. monocesta. male gastric apex and genitalia, ventral view; scale line = 0.50.
Abbreviations: poc = postocular carina, prc = preoccipital carina.

Contributions in Science, Number 378

Snelling: Australian Paragiine Wasps 15

finely rugosopunctate anteriorly, becoming finely, contig-
uously punctate; a pair of raised, impunctate submedian welts
at level of anterior margins of tegulae. Scutellum dull, disc
contiguously minutely punctate; no furrow between disc and
side of scutellum. Metanotum dull, middle portion vertical,
roughened but without distinct punctures. Upper plate of
mesopleuron dull, moderately rugosopunctate; lower plate
slightly shiny and sharply roughened between close, fine
punctures. Metapleuron slightly shiny, contiguously micro-
punctate. Propodeal angles absent; posterior face slightly
shiny, subcontiguously micropunctate and with a few fine
oblique rugules in center of disc; side similar, but punctures
tending to be aligned so surface appears to be finely obliquely
striate near juncture of lateral and posterior surfaces.

Protibia with two or three stout, ferruginous spine-like
setae at apex; probasitarsus slightly more than twice wider
than long; protarsal segments 2-4 with anterior distal corner
produced so that segments are asymmetrical; meso- and
metatibiae each with a single stout, ferruginous, spine-like
seta on outer side of apex; metabasitarsus about four-fifths
as long as metatibia; tarsal claws each with minute erect
tooth.

Hindwing with distinct anal lobe.

Gastral terga slightly shiny and roughened between sub-
contiguous to contiguous minute punctures; sterna similar
but punctures a little coarser.

Pilosity abundant on head and thorax, especially on frons,
mostly about as long as, or slightly longer than, diameter of
anterior ocellus; hairs mostly slender and slightly curled at
tips; side posterior face of propodeum and side of first gastric
tergum with numerous longer hairs of variable length; re-
mainder of gaster with hairs mostly short, stiff, and reclinate
and with abundant fine, appressed pubescence, especially
distad and on sterna. Femora and tibiae with abundant fine,
appressed pubescence and scattered short erect hairs, spe-
cifically lacking dense cover of fully erect, fine, hairs with
curled apices.

Black, the following orange-yellow: mandible, except mar-
gins and teeth; clypeus, except margins; supraclypeal area +
frons, extending nearly to top of eye on each side, but in
middle only a little over midpoint between clypeus and an-
terior ocellus, and not extended ventrad along inner eye mar-
gin; large blotch on upper gena; dorsum and side, down to
furrow, of pronotum; tegula; axillar spot; disc of scutellum;
median portion of metanotum; upper plate of mesopleuron;
large spot on each side of propodeal disc; dorsal face of first
gastral tergum; second tergum, except narrow basal band and
small, obscure, sublateral, preapical spots; small transverse,
preapical median blotch and lateral band on third tergum;
fourth and fifth terga except very narrow basal band; legs,
except coxae. Second to fourth gastral sterna wholly yellow-
ish-orange. Wings transparent, light brown and with dark
brown veins and stigma.

Markings of the paratype are similar to those of the ho-
lotype but the apical one-third of the clypeus is black, the
propleuron is wholly black, the median spot on tergum 3 is
larger and triangular in shape and sterna 2-4 are yellowish-
orange in the middle one-half only.

TYPE MATERIAL (All Western Australia)

Holotype female: 36 km ESE Minnie Creek Homestead,
24°38'S, 1 1 5°42'E, 2 Sept. 1980 (C.A. Howard and T.F.
Houston, no. 349-2), at pool in creek. Paratype: 1 2, 16 km
WSW Lyons River Homestead, 24°38'S, 1 1 5°20'E, 30 Aug.-
1 Sept. 1980 (C.A. Howard and T.F. Houston, no. 344-9),
on Goodenia berardiana (Goodeniaceae). Both in the collec-
tion of the Western Australian Museum.

ETYMOLOGY

The specific name is Latin and refers to the confluence of
the preoccipital and postocular carinae.

DISCUSSION

This is an unusually distinctive species. Especially notewor-
thy is the confluence of the postocular and preoccipital ca-
rinae. In other species of Paragia, when they do possess
occipital carinae, the postocular carina remains close to the
outer eye margin, while the preoccipital carina passes behind
it along the margin of the gena. In P. confluens the postocular
carina follows the genal margin also, so that the two carinae
merge. Another interpretation may be that the postocular
carina is, in fact, wholly absent and that its position along
the lower genal margin is taken over by the greatly extended
preoccipital carina. The effect is the same.

Females of other species of Paragia have densely hairy
legs. Typically, many or most of the hairs are fully erect,
somewhat stiff and bristle-like, and are distinctly curved at
their tips. Hairs of this type are present on the meso- and
metalegs as well, especially on the femora and tibiae. Such
hairs are not present on the legs of P. confluens.

Other unusual features of P. confluens include the short,
broad clypeus, the presence of only a single spine-like seta
at the apex of the outer face of the meso- and metatibiae,
and the lack of propodeal angles. The latter characteristic is
shared with P. magadalena Turner, in which the anal lobe
is lacking, the posterior face of the propodeum is punctured
throughout, the entire body is densely hairy, the legs possess
the usual curled hairs, and the color pattern is very different;
this species is apparently known only from the type from
Queensland (Mackay).

Paragia monocesta, new species

Figures 27, 30, 31

DIAGNOSIS

Runs to P. nasuta in key by Richards (1962) but differs in
color pattern, especially much more extensive orange marks,
particularly of face and abdomen, the latter with tergum 2
mostly black, with a basal orange band.

DESCRIPTION

FEMALE. Measurements. Holotype head width 3.23; head
length 2.87; wing length 8.77; total length 12.8 mm. Para-

16 Contributions in Science, Number 378

Snelling: Australian Paragiine Wasps

types: head width 3.03-3.49; head length 2.62-3.03; wing
length 8.10-9.18; total length 12.0-13.8 mm.

Head 1.12-1.20 times as broad as long. Mandibular teeth
broad and blunt, inner tooth subacute. Clypeus separated
from eye margin by more than one-half diameter of anterior
ocellus; about 1.2 times as wide as long; broadly convex
truncation wider than distance from end of truncation to
lateral angle of clypeus; disc slightly shiny between subcon-
tiguous fine to moderate punctures, becoming subrugose to-
ward apical margin. Supraclypeal area broadly elevated be-
tween antennal sockets, shiny and with punctures separated
by up to a puncture diameter in middle; a raised carina-like
ridge adjacent to each antennal socket and extending well
above socket; frons slightly shiny, moderately to coarsely
rugosopunctate, becoming contiguously, moderately punc-
tate on vertex and preocciput. Preoccipital and postocular
carinae present. Interocellar distance about two times di-
ameter of anterior ocellus; ocellocular distance about three
times diameter of anterior ocellus; ocelloccipital distance
subequal to ocellocular distance. Antennocular distance dis-
tinctly greater than antennal socket diameter; interantennal
distance about 4.5 times antennal socket diameter; scape
about as long as interantennal distance and about twice as
long as first flagellar segment; flagellum gradually thickened
distad, antepenultimate segment about 1.3 times as broad as
long.

Humeral angles of pronotum weak, rounded; pretegular
groove distinct; dorsal face dull, contiguously moderately
punctate; propleuron above furrow dull between subcontig-
uous moderate punctures, surface below furrow slightly shiny,
punctures fine to moderate, irregularly spaced up to two di-
ameters apart; furrow distinct, but anterior edge blunt. Meso-
scutum slightly shiny, subrugose to contiguously, moderately
punctate, with a pair of submedian, mostly impunctate, welts
anteriorly; notauli distinct to posterior margin. Disc of scu-
tellum slightly shiny between very coarse, subcontiguous
punctures, narrow anterior margin of disc nearly impunctate,
but distinctly sculptured; side separated from disc sharply
defined crenate furrow and without longitudinal ridges. Mid-
dle portion of metanotum slightly convex and visible in dor-
sal view. Upper plate of mesopleuron slightly shiny between
moderate to coarse, subcontiguous, elongate punctures; lower
plate with punctures dense to subcontiguous, mostly mod-
erate, some interspaces with minute to fine punctures. Meta-
pleuron a little shinier, punctures very irregularly spaced,
many minute punctures in interspaces between fine punc-
tures. Propodeal angles prominent but broadly obtuse; pos-
terior face slightly shiny and distinctly roughened between
dense, fine punctures; lateral face similar, but spacing vari-
able, sparse along anterior margin, becoming close to dense
distad. Probasitarsus about three times longer than wide,
segments 2-4 symmetrical; metabasitarsus about 0.8 times
length of metatibia; tarsal claws each with a large, oblique
tooth.

Anal lobe of hindwig small but distinct; r-s perpendicular
to Rs and M.

Second gastral tergum distinctly constricted at base; dor-
sum of tergum 1 dull, appearing velvety, and roughened be-

tween dense, moderate punctures; remaining terga similar
but punctures a little smaller, but still moderate, spacing
variably close to dense; sterna distinctly shinier, punctures
similar, a little smaller, but still moderate, spacing variably
close to dense; sterna distinctly shinier, punctures similar but
variably sparse to close.

Erect hairs short and sparse on clypeus, hairs longer and
denser on frons, longest hairs of vertex distinctly longer than
diameter of anterior ocellus; hairs on thoracic dorsum sparse,
longest on mesoscutum about as long as diameter of anterior
ocellus; hairs more abundant on sides of posterior face of
propodeum; first tergum with numerous erect hairs, longest
slightly longer than diameter of anterior ocellus; second ter-
gum with a few short, erect hairs basad at side, otherwise
second and third with scattered very short, stiff, reclinate
hairs; following segments with sparse, short erect hairs, longer
distad.

Prolegs with normal complement of short, stiff, erect, api-
cally curled hairs; mesofemur with similar hairs (except on
upper face), mesotibia with similar hairs on inner face and
some along anterior margin of outer face.

Black, the following orange: clypeus, except dark margins;
frons from clypeus to over one-half of distance to anterior
ocellus, but extending nearly to top of eye at side; narrow
stripe along inner eye margin to lateral angle of clypeus; most
of gena, including narrow stripe almost to base of mandible;
antennal scape; pronotum; short lateral stripe and postero-
median spot on mesoscutum; scutellar disc and side, except
along lateral furrow; most of upper mesospleural plate; pro-
podeum except anterolaterad and large quadrate blotch on
posteromedian portion of disc; legs; entire first, third, fourth
(except small basomedian spot), narrow apical band on fifth,
and all of sixth terga; second tergum with broad basal band,
narrowed in middle; second (except large triangular patch on
each side), third, and sixth sterna.

Forewings transparent, yellowish brown at base, becoming
brown distad; apical portion of costal cell, most of first sub-
marginal cell, marginal cell, all dark brown; veins yellow-
orange to about level of prestigma, brown beyond.

MALE. Measurements. Head width 2.72-2.97; head length
2.44-2.56; wing length 7.54-8.41; total length 10.1-1 1.4mm.

Head 1.12-1.18 times as broad as long. Mandibular teeth
acute. Clypeus separated from eye margin by about 0.5 times
diameter of anterior ocellus; about 1 . 1 times as wide as long,
apical truncation gently convex and slightly longer than dis-
tance from end of truncation to lateral angle of clypeus; disc
shiny between dense to subcontiguous fine to moderate punc-
tures. Interantennal tubercle obtuse, shiny, nearly impunc-
tate on lower facet. Frons, vertex, and preocciput slightly
shiny between contiguous to subcontiguous, fine to moderate
punctures. Interocellar distance about two times diameter of
anterior ocellus; ocellocular distance about 2.5 times diam-
eter of anterior ocellus; ocelloccipital distance about two times
diameter of anterior ocellus. Antennocular distance less than
antennal socket diameter; interantennal distance about 2.0
times antennal socket diameter; scape about 1 .6 times longer
than either interantennal distance or first flagellar segment,
latter about three times as long as wide at apex; flagellum

Contributions in Science, Number 378

Snelling: Australian Paragiine Wasps 17

moderately thickened distad, antepenultimate segment slightly
broader than long.

Thorax about as in female, but posterior face of propo-
deum with distinct short rugulae extending laterad from me-
dian groove.

Process of protrochanter somewhat quadrate in anterior
view, not extending over base of femur; in profile, thin and
with anterior face slightly concave, lower edge subcarinate.

Gaster as described for female; sterna without tubercles.

Apical sterna and genitalia as in Figures 30 and 31.

Pilosity similar to that of female but hairs longer, longest
on vertex and dorsum of pronotum about twice as long as
diameter of anterior ocellus, those of mesoscutum not quite
so long; propodeum, pleura, and first tergum also with nu-
merous long hairs. Legs with numerous, but not dense, short,
erect hairs, none curled at tips.

Black, the following yellowish: mandibular spot; clypeus;
vertical bar on frons extending more than halfway between
base of clypeus and anterior ocellus; narrow stripe along inner
orbit from lateral angle of clypeus to level of upper end of
frontal bar; antennal scape; large postocular spot (somewhat
orange). The following orange: pronotum, except spiracular
lobe; minute posterolateral spot on mesoscutum; posterior
two-thirds of scutellar disc; large spot below wing base; angles
of propodeum; legs, except coxae in large part; gastral terga
as in female, but segments 5-7 black. Second sternum wholly
yellowish, and following segments dark.

TYPE MATERIAL (All Western Australia)

Holotype female, 10 female and four male paratypes; 10 km
ESEMeedo Homestead, 25°40'S, 1 14°37'E, 23-26 Aug. 1980
(C.A. Howard and T.F. Houston, no. 336-19), on Calythrix
oldfieldii (Myrtaceae). Additional paratypes: 1 2, 7 km N
Boologooro Homestead, 24°39'S, 1 1 3°42'E, 27-29 Aug. 1 980
(C.A. Howard and T.F. Houston, no. 338-5), on Verticordia
forrestii (Myrtaceae); 1 2, Gascoyne Junction, 2 Sept. 1968
(F.H.U. Baker) (this locality is about 150 km E Carnarvon
at 25°03'S, 1 1 5° 1 3'E; T. Houston, pers. comm.). Holotype
and most paratypes in WAM; 2 22, 1 <5 in LACM.

ETYMOLOGY

The specific name is Greek (one band or girdle) and refers
to the single black band on metasomal tergum 2.

DISCUSSION

Although this will run to P. nasuta in the key by Richards
(1962) it does not resemble that species; P. nasuta is a pri-
marily black wasp with limited orange marks, while P. mono-
cesta is largely orange. In addition to being larger (total length
13.4-16.5 mm), females of P. nasuta have a largely rugo-
sopunctate clypeus, the humeral carina reaches the posterior
margin of the pronotum, and the fine to moderate punctures
on the disc of the second gastral tergum are mostly separated
by three or more times a puncture diameter.

Males of the two species, except in the clypeal sculpturing,
differ from one another in the same characteristics as do their

females. Additionally, males of P. nasuta have the median
lobe of the eighth gastric sternite a little longer than broad
and narrower than the emargination between it and the lateral
lobes. In males of P. monocesta the median lobe is much
broader than long and much broader than the emargination
between it and the lateral lobes.

Paragia nasuta F. Smith

Paragia nasuta F. Smith, 1868:252. 3, 2. Richards, 1962:68,
76-77. 3, 2.

Richards has described the color pattern of both sexes and
recorded specimens from Queensland and Western Australia.

New records of this species in Western Australia are as
follows: 5 22, 4 33, 70-75 km ENE Norseman, 10-16 Nov.
1978 (T.F. Houston et al., no. 220-18), on flowers of Me-
laleuca fulgens (Myrtaceae); 1 3, 3. 5-5. 5 km S Yellowdine,
31°18'S, 1 19°39'E, 27 Oct. 1978 (T.F. Houston, no. 216-15),
on flowers of Grevillea paradoxa (Proteaceae) (all in WAM).

Paragia sobrina F. Smith

Paragia excellens F. Smith, 1869:309. 3, 2.

Paragia sobrina F. Smith, 1869:309. 2. Richards, 1 962:75—
76. 3, 2.

Paragia excellens and P. sobrina were described on the same
page, with P. excellens appearing first and the more com-
pletely described of the two. The description of P. sobrina is
based on those characteristics by which it differs from P.
excellens. Richards (1962) correctly recognized that the two
forms were conspecific, but for unknown reasons chose to
use P. sobrina as the name for this species.

Smith (1869) cited the type locality for both names as
“Champion Bay,” in Western Australia. According to Be-
quaert (1928) the female types of both are labelled “Swan
River.” Richards (1962), however, cited Champion Bay as
the type locality for P. sobrina. I have examined both types
and confirm Bequaert’s statement: the types of both P. ex-
cellens and P. sobrina are clearly labelled “Swan River.”
Two additional specimens of P. sobrina from Western Aus-
tralia were examined; 2 22, near Emu Rock, 53 km E Hyden,
32°27'S, 1 19°25'E, 8-10 Nov. 1979 (T.F. Houston, no. 280-
1), on flowers of Beaufortia bracteosa (Myrtaceae) (WAM).

These have been compared with the types of both P. ex-
cellens and P. sobrina. In both types, the propodeum is most-
ly orange; the propodeum is black in the Emu Rock females,
with the lateral processes orange and, in one specimen, there
is a narrow orange mark in the middle of the posterior face.
The pronotum is entirely orange in the types, but in the Emu
Rock specimens it is largely black below the humeral carina
and with an extension of black in front of the posterior lobe.
The second tergum of the P. sobrina type is wholly black.
Broad orange bands are present across the second tergum in
the type of P. excellens and the two Emu Rock specimens,
though narrower in the latter two. These two specimens also
bear small orange marks on the third tergum, which is black
in the two types.

18 Contributions in Science, Number 378

Snelling: Australian Paragiine Wasps

The most characteristic morphological feature of P. so-
brina is the shape of the propodeal process. Specialized,
hooked hairs are abundant on the prolegs and mesofemur
and the protarsal segments are robust and are weakly asym-
metrical; the last protarsal segment is especially stout, about
two-thirds as broad as long.

Paragia walkeri Meade-Waldo

Paragia walkeri Meade-Waldo, 1910:32. 6. Richards, 1962:

68, 72-73. 5, 2.

A single female, although not associated with males, probably
belongs to this species. It shares with the described males the
peculiarly colored wings (“yellow with dark tips”), the vel-
vety appearance of the black areas of the body, the bluntly
spine-like propodeal process, the sparsely punctate gastral
terga, and the large size. The specimen is 25.6 mm long; the
forewing is 18.0 mm long.

This is a very attractively marked specimen. The head
(except around the ocelli), pronotum, scutellum, metanotum,
most of the propodeum, prepectus, all of gastral tergum 1
and sternum 1, a narrow basal band on tergum 2, and two
small basolateral spots on sternum 2, are dull yellowish. The
mandible, antenna legs, and fourth to sixth abdominal seg-
ments are reddish yellow, except that tergum 4 is blackish
across the base.

The specimen was collected 9 km N by E Mudginberry
Homestead, 12°31'S, 132°54'E, Northern Territory, 10-11
June 1973 (J.C. Cardale; ANIC).

ACKNOWLEDGMENTS

The specimens from the Western Museum were made avail-
able for study by Terry F. Houston. Important types and
other specimens from the collections of the British Museum
(Natural History) were loaned by Colin R. Vardy. Other
specimens were loaned by I. Naumann (ANIC) and K. Walk-
er (NMV).

My thanks also to J.M. Carpenter for a spirited and thought-
provoking interchange on the systematics of this group, even
though we are not in complete accord.

LITERATURE CITED

Bequaert, J.C. 1928. A study of certain types of diplopter-
ous wasps in the collection of the British Museum. An-
nals and Magazine of Natural History, (10)2: 1 38-176.
Bohart, R.M. 1984. Gastrodynerus, a new genus of Eu-
menidae from western North America (Hymenoptera:
Eumenidae). Pan-Pacific Entomologist, 60:12-15.
Carpenter, J.M. 1982. The phylogenetic relationships and
natural classification of the Vespoidea (Hymenoptera).
Systematic Entomology, 7: 1 1-38.

Cresson, E.T. 1879. (no title). American Entomological So-
ciety, Proceedings, 7:vi-vii.

Giordani Soika, A. 1978. Revisione degli Eumenidi neo-
tropicali appartenenti ai generi Eumenes Latr., Omicron
(Sauss.), Pararaphidoglossa Schulth. ed affine. Museo
Civico di Storia Naturale di Venezia, Bolletino, 29:1-
420.

Latreille, P.A. 1810. Considerations generates sur l’ordre
naturel des crustaces, des arachnides et des insectes. Paris.

Meade-Waldo, G. 1910. New species of Diploptera in the
collection of the British Museum. Annals and Magazine
of Natural History, (8)5:30-51.

. 1911. Notes on the Masaridae (Hymenoptera), with

descriptions of a new genus and three new species. An-
nals and Magazine of Natural History, (8)8:747-750.

Richards, O.W. 1962. A revisional study of the masarid
wasps (Hymenoptera, Vespoidea). British Museum,
London, 294 pp.

. 1968. New records and new species of Australian

Masaridae (Hymenoptera: Vespoidea). Australian En-
tomological Society, Journal, 7:101-104.

Saussure, H., de. 1853. (No title). Societe Entomologique
de France, Annales, (3)6, Bulletin:xxvii-xxvi.

Shuckard, W.E. 1837. Descriptions of new exotic aculeate
Hymenoptera. Entomological Society of London,
Transactions, 2:68-82.

Smith, F. 1850. Descriptions of two new species of exotic
Hymenoptera. Entomological Society of London,
Transactions, (2)1:41-42.

. 1857. Catalogue of the hymenopterous insects in

the collection of the British Museum. 5, pp. British Mu-
seum, London.

. 1865. Descriptions of some new species of hy-
menopterous insects belonging to the families Thynni-
dae, Masaridae and Apidae. Entomological Society of
London, Transactions, 2:389-399.

. 1868. Descriptions of aculeate Hymenoptera from

Australia. Entomological Society of London, Transac-
tions, 1868:231-258.

. 1869. Descriptions of new genera and species of

exotic Hymenoptera. Entomological Society of London,
Transactions, 1869:301-31 1.

Turner, R.E. 1908. Two new diplopterous wasps from
Queensland. Entomological Society of London, Trans-
actions, 1908:89-91.

Zavattari, E. 1910. Un nuovo Masaride dell'America meri-
dionale. Museo Civico di Storia Naturale Genova, An-
nali, 44:531-535.

Submitted 31 October 1985; accepted 24 March 1986.

Contributions in Science, Number 378

Snelling: Australian Paragiine Wasps 19

i 1 1 if ili

siRili Publications of toe

■ N ATORAl/MS'fOKV MUSEUM OF LOS ANGELES COUNTY

MBai

liKl

liill Pllillli

■MHi

1: :he scjeiti1:ic pobHcaisonis of the Natural History Museum of Los Angeles' County have been
issued at irregular intervals. in three major series; the articles in each series are numbered
' individually. and numbers run consecutively, regardless- of the subject matter.

# Contributions in Science, 'a miscellaneous series of technical papers describing orig-
inal research in the life and earth sciences.

# Science Bulletin, a miscellaneous ; series' bf monographs describing original research
ra the life and earth sciences. This series was discontinued in 1978 with the issue of
Numbers 3$ and 30; monographs are now published by the Museum in Contributions

: in Science. , "Tj

# Spien^;|ffi|l|oha-^:iclcs on j^aiural history topics, generally written for the layman.

■ Copies of the.puMications in these series are sold through the Museum Book Shop. A catalog
■us available oh request.

SCIEN T IF if: FU B I JCAT l CMS COM MUTT EE j

."Hi

Craig C. Black. Museum Director
Richard C. B rosea;

Daniel. M. Cohen, Committee Chairman
John M. Harris
Charles I.. -Hogue

PROPODIAL ELABORATION IN SOUTHERN AFRICAN AND INDIAN
OCEAN FISSURELLIDAE (MOLLUSCA: PROSOBRANCHIA)
WITH DESCRIPTIONS OF TWO NEW GENERA
AND ONE NEW SPECIES

James H. McLean1 and R. N. Kilburn2

ABSTRACT. Species previously assigned to Amblychilepas Pilsbry,
1890, in southwestern, southern, and eastern Africa and the western
Indian Ocean are reviewed. Amblychilepas has an unmodified pro-
podium and is represented in southern Africa by A. platyactis, new
species. Two kinds of elaborate propodial processes form the basis
for new genera. Dendrofissurella, type species Patella scutellum Gme-
lin, 1791, from southern Africa, has a large body and a trunklike
propodium with side branches. The genus is monotypic, although
two subspecies are recognized. Medusafissurella, type species Fis-
sured la salebrosa Reeve, 1 850, has a smaller body and a propodium
of radiating tentacles. Three species are known: M. salebrosa (Reeve),
in the Arabian Sea and east Africa, M. dubia (Reeve, 1849), in
southern and eastern Africa, and M. chemnitzii (Sowerby, 1835) in
southwestern Africa. The function of the propodium remains to be
investigated.

INTRODUCTION

The propodium (anterior end of the foot) in fissurellids has
not heretofore been reported to have unusual features. Here
we describe elaborate propodial tentacles in four fissurellid
species occurring in the Arabian Sea, and along the eastern,
southern, and southwestern coasts of Africa. These species
were previously assigned to Amblychilepas Pilsbry, 1 890, the
type species of which lacks these processes. Two genera are
proposed, each strikingly different in shell and body pro-
portions, as well as propodial elaboration. Three of the species
are poorly known and have rarely been discussed subsequent
to their original descriptions.

The genera treated here (Amblychilepas, Medusafissurella,
new genus, and Dendrofissurella, new genus) are closely re-
lated in radular and shell characters and are assigned to the
subfamily Fissurellinae. We follow Thiele (1929) and McLean
(1984a, 1984b) in recognizing two subfamilies in the Fissu-
rellidae: Emarginulinae and Fissurellinae. The Emarginuli-
nae are the oldest, originating in the Mesozoic. Despite major

Contributions in Science, Number 379, pp. 1-12
Natural History Museum of Los Angeles County, 1986

differences among genera in shell characters, the radula in
the Emarginulinae has unifying features; further subdivisions
are therefore recognized only at the tribal level. The relatively
few genera in the Fissurellinae are relatively young, appearing
in the Cenozoic.

Fissurelline genera differ from emarginuline genera in the
following features: 1) the rachidian plate of the radula has a
broad base and a narrow tip, rather than the broad or mod-
erately broad tip of the emarginuline rachidian; 2) the large
outer lateral tooth is so long that it is aligned with the inner
lateral teeth of the row above, rather than the same row as
in emarginuline genera (Hickman, 1984); 3) the shell muscle
lacks the inwardly directed hook-shaped process of emargi-
nuline genera (Odhner, 1932); 4) the selenizone that is present
at least in the early juvenile of emarginuline genera is lacking
(McLean, 1984a, 1984b).

This paper treats those fissurelline genera in which the
body tends to be larger than the shell. Shells in this group
have been confused with emarginuline genera in the tribe
Fissurellidini; the latter —Lucapinella Pilsbry, 1890; Leuro-
lepas McLean, 1970; Fissurel/idea Orbigny, 1841 ; Pupillaea
Sowerby, 1835; and Buchanania Lesson, 1 830 — were treated
by McLean (1984a).

Material upon which this paper is based is housed in the
following museums or collections: AMS, Australian Mu-
seum, Sydney; BMNH, British Museum (Natural History),
London; CAS, California Academy of Sciences, San Fran-
cisco; JC, J. Christiaens collection, Hasselt, Belgium; LACM,
Los Angeles County Museum of Natural History, Los An-
geles; MNHN, Museum National d’Histoire Naturelle, Paris;

1 . Invertebrate Zoology Section, Natural History Museum of Los
Angeles County, 900 Exposition Blvd., Los Angeles, California 90007.

2. Natal Museum, Pietermaritzburg 3201, Natal, South Africa.

ISSN 0459-8113

Figures 1-4. SEM views of radulae, showing a narrow rachidian with expanded base, four narrow laterals, a large quadricuspid outer lateral,
an uncusped lateromarginal plate, and slender marginals; all x200. 1. Medusafissurella dubia (Reeve, 1849). Salt Rock, Umlahli District,
Natal, South Africa, NM B9928. 2. Dendrofissurella scutellum hiantula (Lamarck, 1822). Algoa Bay, Eastern Cape Province, South Africa,
NM B99 19.3. Amblychilepas platyactis, new species. Paratype, Kwelera, Eastern Cape Province, South Africa, NM B9929/T3058. 4. Fissurella
nimbosa (Linnaeus, 1758). Puerte La Cruz, Venezuela, LACM 76-30.

NM, Natal Museum, Pietermaritzburg; USNM, United States
National Museum, Washington, D.C.

Family Fissurellidae
Subfamily Fissurellinae

Three genera having no reported modification of the pro-
podium are currently recognized in the subfamily Fissurel-
linae: 1) Fissurella Bruguiere, 1798 (with several subgenera),
in which the size of the body does not greatly exceed that of
the shell and the large outer lateral radular plate has four
cusps; 2) Amblychilepas Pilsbry, 1890, with an oval shell,
central foramen, and large body; the large outer lateral has

four cusps as in Fissurella; 3) Macrochisma Sowerby, 1839
(with several subgenera), in which the shell is narrow, the
foramen is elongate and posterior, the body much longer than
the shell and the large outer lateral has three cusps (one very
small); Kilbum has noted (unpublished observation) that
Macrochisma africana Tomlin, 1932, has a deeply bifid pro-
podium. McLean (1970) included the monotypic genus Leu-
rolepas McLean, 1970, in the Fissurellinae, but later (1984a)
transferred it to the Emarginulinae, tribe Fissurellidini.

Radulae within the Fissurellinae show few generic and
specific differences. Radulae of members of four genera
( Fissurella , Amblychilepas, Medusafissurella, and Dendro-
fissurella) are illustrated here (Figs. 1-4). Each has a narrow
rachidian with expanded base, four narrow laterals, a large

2 Contributions in Science, Number 379

McLean and Kilburn: African Fissurellids

Number 356

Number 357

Number 358

Number 359

Number 360

Number 361

Number 362

Number 363

Number 364

Number 365

Number 366

Number 367

Number 368

Contents

CONTRIBUTIONS IN SCIENCE
1985

Canine function in Smilodon (Mammalia; Felidae; Machairodontinae)

William A. Akersten

Systematics and distribution of the skinks allied to Eumeces tetragrammus (Sauria: Scincidae)

Carl S. Lieb

Systematics of the South American freshwater fish genus Adontosternarchus (Gymnotiformes,
Apteronotidae)

Francisco Mago-Leccia, John G. Lundberg, and Jonathan N. Baskin

Biomechanical implications of the variation in Smilodon ectocuneiforms from Rancho La Brea

Christopher A. Shaw and Antonia E. Tejada-Flores

A review of the gobiid fish genus Monishia Smith, 1949, from the western Indian Ocean and
Red Sea, with description of a new species

Menachem Goren

The systematics of the hylaeine bees (Hymenoptera: Colletidae) of the Ethiopian zoogeographical
region: The genera and subgenera with revisions of the smaller groups

Roy R. Snelling

Review of the Late Pleistocene avifauna from Lake Manix, central Mojave Desert, California
George T. Jefferson

Fossil pontoporiid dolphins (Mammalia: Cetacea) from the Pacific Coast of North America
Lawrence G. Barnes

The geology of the Rio Beni: Further evidence for Holocene flooding in Amazonia

Kenneth E. Campbell, Jr., Carl David Frailey, and Jorge Arellano L.

Vascular plants of the Channel Islands of southern California and Guadalupe Island, Baja California,
Mexico

Gary D. Wallace

Rugose corals (Coelenterata, Anthozoa) from the Lower Permian McCloud Limestone at Tomb-
stone Mountain, northern California

Edward C. Wilson

The Late Miocene dolphin Pithanodelphis Abel, 1905 (Cetacea: Kentriodontidae) from California
Lawrence G. Barnes

Evolutionary relationships of the Atherinopsinae (Pisces: Atherinidae)

Brian N. White

NATURAL HISTORY MUSEUM OF LOS ANGELES COUNTY
900 Exposition Boulevard, Los Angeles, California 90007

Number 369

Number 370

Number 371

A review of the genera of cleptoparasitic bees of the tribe Ericrocini (Hymenoptera: Anthophoridae)
Roy R. Snelling and Robert W. Brooks
The status of Smilodon in North and South America
Annalisa Berta

A new record og giant short-faced bear, Arctodus simus from western North America with a re-
evaluation of its paleobiology

Steven D. Emslie and Nicholas J. Czaplewski

Author Index

Akersten, William A.
Arellano L. Jorge
Barnes, Lawrence G.
Baskin, Jonathan N.
Berta, Annalisa
Brooks, Robert W.
Campbell, Kenneth E.
Czaplewski, Nicholas J.
Emslie, Steven D.

Frailey, Carl David
Goren, Menachem
Jefferson, George T.

Lieb, Carl S.

Lundberg, John G.
Mago-Leccia, Francisco
Shaw, Christopher A.
Snelling, Roy R.
Tejada-Flores, Antonia E.
Wallace, Gary D.

White, Brian N.

Wilson, Edward C.

Number 358
Number 370
Number 369
Number 364
Number 371
Number 371
Number 364
Number 360
Number 362
Number 357
Number 358
Number 358
Number 359

Number 359
Number 365
Number 368
Number 366

Number 356
Number 364

Numbers 363, 367

Number 361, 369

quadricuspid outer lateral, an uncusped lateromarginal plate,
and slender marginals. The marked asymmetry of the fis-
surellid radula, which places the laterals of the left side of
the ribbon higher than those of the right, has been discussed
by Hickman (1981).

The five genera (two new) here recognized in the subfamily

Fissurellinae may be keyed as follows:

la. Propodium without tentacles 2

b. Propodium with elaborate tentacles 4

2a. Body not or not greatly exceeding size of shell

Fissurella

b. Body at least twice shell length 3

3a. Foramen oval, central Amblychilepas

b. Foramen elongate, posterior Macrochisma

4a. Propodium with numerous subequal tentacles

Medusafissurel/a

b. Propodial outgrowth with main trunk and side branches
Dendrofissurella

Medusafissurella new genus

Type species: Fissurella salebrosa Reeve, 1 850. Recent, Ara-
bian Sea, Indian Ocean.

DESCRIPTION. Shell markedly narrowed anteriorly; an-
terior end raised; anterior shell edge thinner and sharper than
elsewhere; posterior end only slightly or not at all raised;
anterior slope concave. Foramen oval, interior callus not
truncated posteriorly. Sculpture of strong, scabrous ribs. Pos-
terior portion of foot covered by shell; shell edge only slightly
enveloped by mantle folds; propodium with radiating ten-
tacles, sometimes branched, subequal in length. Large outer
lateral tooth of radula quadricuspid.

REMARKS. On the basis of shell characters alone, the
species grouped here have been variously assigned to other
fissurellid genera. The shell of Medusafissurella differs from
that of most Fissurella species in having a prominently raised
anterior end with thin edge, and from Amblychilepas and
Dendrofissurella in having strong, scabrous primary ribs, and
a less raised posterior end. Body differing in being nearly
covered by the shell and in having numerous propodial ten-
tacles, not the single main, branching structure of Dendro-
fissurella.

The quadricuspid outer lateral tooth of Medusafissurella
is similar to that in Dendrofissurella, Amblychilepas , and Fis-
surella.

Medusafissurella comprises three allopatric species: M. sa-
lebrosa (Reeve, 1 850), M. dubia (Reeve, 1 849), and M. chem-
nitzii (Sowerby, 1835).

ETYMOLOGY. The prefix is suggested by the propodial
tentacles, which recall the serpentine locks of Medusa in
Roman mythology; gender feminine.

Medusafissurella salebrosa (Reeve, 1850)

Figures 5-7, 14, 26

Fissurella salebrosa Reeve, 1850:pl. 11, sp. 78; Bosch and

Bosch, 1982:29, 3 figs. Type locality: Karachi, Pakistan.

Glyphis salebrosa; Pilsbry, 1890:208, pi. 39, fig. 7 [copy Reeve,
1 850:fig. 78],

Diodora salebrosa; Christiaens, 1974:91.

Lucapinella salebrosa; Biggs, 1969:202 [checklist only],

“ Fissurella subrostrata Guilding”; of Sowerby II, 1862:192,
fig. 215, not F. subrostrata “Gray”; Sowerby, 1835b:6, fig.
35.

DESCRIPTION. Shell markedly narrowed anteriorly; an-
terior end raised; anterior slope concave; sculpture of strong,
scabrous ribs; rib interspaces broad, foramen nearly circular,
broader posteriorly. Posterior portion of foot covered by shell,
shell edge only slightly enveloped by mantle folds; exterior
color gray-brown, interior white. Propodium with about 1 5
radiating tentacles, subequal in size, anteriormost the largest.
Maximum shell length 35 mm (JC colln.).

The description of the propodium is based on the only
preserved specimen available (CAS 031984, Figs. 14, 26).

TYPE MATERIAL. Lectotype here designated, BMNH
1975078 (Fig. 5), one of two original syntypes so catalogued.

DISTRIBUTION. Indian Ocean, northernmost Arabian
Sea, Pakistan to Somalia.

MATERIAL EXAMINED. PAKISTAN: Buleiji Point,
Sind Province (LACM 79-34) (Fig. 6); Goth Jafar, west of
Karachi (CAS 031984) (Figs. 14, 26); Karachi (NM H5382).
OMAN: Muscat (JC colln.); Masirah Island (NM J3845).
SOMALIA: Alula (JC colln.); Socotra Island (JC colln.) (Fig.
7).

COMPARISONS. The anterior end of M. salebrosa is
more tapering and the primary ribs are more strongly defined
and broadly separated than those of M. dubia and M. chem-
nitzii.

REMARKS. This enigmatic species is poorly known, part-
ly because of its localized distribution. The first illustration
with a properly documented locality subsequent to that of
Reeve’s original figure is that of Bosch and Bosch (1982),
who indicated it as common, “distributed generally on rocks
or in crevices.”

Sowerby II (1862) incorrectly placed F. salebrosa in the
synonymy of F. subrostrata “Guilding,” Sowerby, 1835, a
yet unrecognized taxon (see synonymy above), said to be
from St. Vincents, West Indies. The original illustration shows
more numerous, less pronounced radial ribs than those of F.
salebrosa. Christiaens (1973:91) tentatively placed F. sale-
brosa in the synonymy of F. subrostrata. He now considers
(pers. comm.) the specimen figured by Perez-Farfante (1943:
20, pi. 6, figs. 9-1 1) as “F. subrostrata ” to be a mislocalized
specimen of Medusafissurella salebrosa.

Medusafissurella dubia (Reeve, 1849)

Figures 1, 8-10, 15

Fissurella dubia Reeve, 1 849:pl. 6, fig. 35. Type locality: Port
Natal [= Durban], South Africa.

Lucapina dubia; Sowerby II, 1862:193, pi. 9, fig. 208.
Glyphis dubia; Pilsbry, 1890:217, pi. 39, fig. 6 [copy Reeve,
1849:fig. 35],

Amblychilepas dubia; Kilbum and Rippey, 1982:35, pi. 6,
fig. 2.

Contributions in Science, Number 379

McLean and Kilburn: African Fissurellids 3

Fissurellidea genevievae Dautzenberg, 1929:546, pi. 1, figs.

3-7. Type locality: Madagascar (several localities cited).

DESCRIPTION. Shell outline oval, slightly narrowed an-
teriorly; anterior end raised, concave at early stage, all slopes
convex at later stage. Sculpture of strong scabrous ribs, rib
interspaces relatively narrow; foramen oval. Posterior por-
tion of foot covered by shell, shell edge slightly enveloped
by mantle folds. Propodial tentacles approximately 12 on
ventral side, some branching up to 4 times so that 2 or 3
layers of tentacles project in anterior view. Maximum shell
length 36 mm (Kilbum and Rippey, 1982).

The description of the animal is based on preserved spec-
imens from Salt Rock, Umhlali District, Natal (NM B9915)
(Fig. 15), collected by R.N. Kilburn, November, 1970.

TYPE MATERIAL. Holotype, BMNH 198495 (Fig. 9).
Type material of Fissurellidea genevievae has not been lo-
cated. It is not in the Dautzenberg collection at the Brussels
Museum.

DISTRIBUTION. East Africa to South Africa: Mogadis-
cio, Somalia, to Kelso, Natal, South Africa; Madagascar.
Living in submerged rock crevices, intertidal fringe to a depth
of several meters.

MATERIAL EXAMINED. SOMALIA: Mogadiscio
(USNM 673793; LACM 25051); Socotra Island (JC colln.).
MOZAMBIQUE: Bazaruto Island, near lighthouse (NM
G4599) (Fig. 8); Cabo de Santa Maria, Bay of Maputo (NM
6238); Mozambique (LACM 25206). SOUTH AFRICA,
NATAL: Kosi Bay, Zululand (NM B2092); Mvoti River
mouth (NM 9350); 3 mi. off Umhlanga Rocks, 12-13 fm.
(NM A269); Salt Rock, Umhlali district (NM B9928) (Fig.
15); Tongaat (NM 7132); Durban (NM 9489, NM B2093,
NM 8989, NM 267, NM 7 1 39, NM 9004) (Fig. 1 0); Isipingo
(NM 5883); Umkomaas (NM 9 1 32); Park Rynie (NM 5870);
Kelso (NM B2095).

COMPARISONS. Shell differing from that of M. sale-
brosa in having a less tapered anterior end, a greater number
of radial ribs, a more elongate foramen; propodium differing
in having the tentacles branched.

REMARKS. Kilburn and Rippey (1982) were the first to
recognize the occurrence of this species subsequent to the
original description of Reeve. They noted that it differed
anatomically from “ Amblychilepas ” scutellum but did not
provide details. Fissurellidea genevievae Dautzenberg has not
been noticed in subsequent literature, but the original figures
are adequate to allocate it to Medusafssurella dubia, with
which it was not originally compared.

MedusafissureUa chemnitzii (Sowerby, 1835)

Figures 11-13, 16

Fissurella chemnitzii Sowerby, 1835a; 126; Sowerby, 1835b:
5, fig. 55; Reeve, 1849:pl. 1, fig. 1 (part). Original locality
unknown [“Benguela, West Africa,” cited by Reeve, 1 849],

Fissurellidea chemnitzii; Sowerby II, 1862:202, pi. 2, fig. 29,

pi. 8, fig. 192 (part).

Megatebennus (section Amblychilepas) chemnitzii; Pilsbry,

1890:185, pi. 39, fig. 90 [copy Reeve, 1849:fig. 1].

DESCRIPTION. Shell outline oval, slightly narrowed an-
teriorly; anterior end markedly raised; posterior end slightly
raised. All slopes straight to slightly concave. Sculpture of
strong irregular, finely scabrous ribs; rib interspaces narrow.
Foramen elongate oval. Posterior portion of foot projecting
‘A shell length beyond shell, shell edge slightly enveloped by
mantle folds. Propodial tentacles numerous, irregular. Max-
imum shell length 52 mm (East London Museum 11001).

The description of the animal is based on four poorly
preserved specimens from Baia de Santa Maria, Benguela
Prov., Angola (MNHN uncat.), collected by S. Gofas, De-
cember, 1982 (Figs. 12, 16).

TYPE MATERIAL. Neotype (here designated), MNHN
uncat., Baia de Santa Maria, Angola (Fig. 12). Dimensions:
24.8 x 15.2 x 7.2 mm. Sowerby’s original material is un-
known; it has not been recognized in the British Museum.

DISTRIBUTION. Southwestern Africa: Pointe Noire,
Congo, to Kunene River Mouth, Angola/Namibia border.
Rocky intertidal to 2 m.

MATERIAL EXAMINED. CONGO: Pointe Noire, Cote
Sauvage (JC colln.) (Fig. 1 1). ANGOLA (all MNHN uncat.,
collected by S. Gofas): Ambrizete, Zaire Prov.; Baia de Santa
Maria, Benguela Prov. (Figs. 12, 16); Sao Nicolau, Mo-
camedes Prov.; Chapeu Armado, Mocamedes Prov.; Luan-
da, Luanda Prov. (JC colln.). NAMIBIA: Kunene River mouth
(NM B6040; East London Museum 11001) (Fig. 13).

COMPARISONS. MedusafissureUa chemnitzii differs from
M. salebrosa and M. dubia in its larger size, more elongate
foramen, less scabrous ribs, and in having red rather than
brown or gray coloration.

REMARKS. This is the most enigmatic of the three Me-
dusafissurella species. A neotype is here designated because
original material cannot be located and none of the illustra-
tions purported to represent this species are completely ac-
curate. The history of the name Fissurella chemnitzii is de-
tailed below.

The descriptions of Sowerby (1835a) and Reeve (1849)
agree with the species as here interpreted. Sowerby (1835a)
remarked that “the only specimen I have ever seen of this
species was in the Tankerville Collection, from which after
several vicissitudes, it has at length found its way to Mr.
Cuming’s.” Although he also noted: “This remarkable shell
is represented by Martini (I, t. xi, fig. 100),” the shell illus-
trated by Sowerby (1835b) is not copied from Martini’s figure
100 (Martini and Chemnitz, 1769-1795) (which we consider
to represent the Mediterranean Diodor a italica (Defrance,
1820)), but could only be the Tankerville shell. The original
figure of Sowerby (1835b) does not show the primary ribs as
sufficiently prominent and the foramen is somewhat too large;
however, the proportions are correct and it conceivably could

4 Contributions in Science, Number 379

McLean and Kilburn: African Fissurellids

Figures 5-13. Shells of Medusafissurella species; exterior and interior views with anterior at top; lateral views of left side. Figs. 5-7. M.
salebrosa( Reeve, 1 850). 5. Lectotype, Karachi, Pakistan, BMNH 1975078,27.0 x 20.7 x 8.9 mm. 6. 7 km WNW Buleiji Point, Sind Province,
Pakistan, LACM 79-34, 20.0 x 12.5 x 5.4 mm. 7. Socotra Island, Somalia, JC colln., 29.2 x 26.7 x 10.2 mm. Figs. 8-10. M. dubia (Reeve,
1849). 8. Bazaruto Island, Mozambique, NM G4599, 22.4 x 15.5 x 4.0 mm. 9. Holotype, Durban, Natal, South Africa, BMNH 1984195,
22.1 x 13.7 x 5.3 mm. 10. Durban, NM 9489, 19.4 x 12.1 x 4.6 mm. Figs. 11-13. M. chemnitzii (Sowerby, 1835). 11. Pointe Noire, Cote
Sauvage, Congo, JC colln., 23.1 x 16.4 x 5.9 mm. 12. Neotype, Baia de Santa Maria, Angola, MNHN uncat., 24.8 x 15.2 x 7.2 mm. 13.
Kunene River Mouth, Namibia, NM B6040, 42.9 x 28.9 x 10.7 mm.

have been based on the species treated here. The specimen
figured by Reeve from “Benguela, West Africa, collected by
Dr. Tams” has not been located in the British Museum; this
illustration is inaccurate because it depicts primary ribs that

are too prominent, too few, and with interspaces too broad,
and a foramen that is broader posteriorly. A later illustration
(Sowerby II, 1862), which has no documentation, is based
on a still different specimen; in fact, it is a better rendition

Contributions in Science, Number 379

McLean and Kilburn: African Fissurellids 5

Figures 14-16. Preserved bodies of Medusafissurella species. 14. M. salebrosa, dorsal and ventral views of preserved body out of shell, Goth
Jafar, 10 mi. west of Karachi, Pakistan, CAS 031984, shell dimensions 26.5 x 20.0 x 9.6 mm. 15. M. dubia. dorsal and ventral views of
preserved body out of shell, Salt Rock, Umhlali Dist., Natal, South Africa, NM B9928, shell dimensions 29.0 x 16.9 x 7.2 mm. 16. M.
chemnitzii, Baia de Santa Maria, Angola, MNHN uncat., lateral view of preserved specimen, shell dimensions 27.7 x 16.5 x 7.5 mm.

of M. dubia than anything else. Despite these discrepancies,
we retain the name of Sowerby (1835), basing it on his de-
scription (1835a) and his figure (1835b), to which we relate
the neotype specimen designated here.

Dendrofissurella new genus

Type species: Patella scat ellum Gmelin, 1791. Recent, south-
ern Africa.

DESCRIPTION. Shell oval, anterior end narrower than
posterior, ends raised: foramen nearly central, elongate-oval.
Sculpture of fine radial ribs. Foot projecting posterior to shell
for distance greater than length of shell; shell edge slightly
enveloped by mantle fold; propodium with single tapering,
trunklike elongation, with approximately 9 irregularly placed
lateral branches. Large outer lateral tooth of radula quadri-
cuspid.

REMARKS. Differing from Atnblychilepas in having an
elaborate propodium, minimal envelopment of the shell by
the mantle, and minor development of the papillae of the

upper lobe of the mantle. Differing from Medusafissurella in
having a trunklike propodium rather than a broad propo-
dium with subequal tentacles. The foot projects posteriorly
to a greater extent than in Medusafissurella.

The quadricuspid outer lateral tooth of Dendrofissurella
scutellum is similar to that of species of Fissurella, Ambly-
chilepas, and Medusafissurella.

The genus contains a single species, for which we recognize
two geographic subspecies.

ETYMOLOGY. The prefix is suggested by the trunklike
propodium, a Greek word for tree; gender feminine.

Dendrofissurella scutellum (Gmelin, 1791)

(a) D. scutellum scutellum

Figures 17, 18

Patella scutellum Gmelin, 1791:3731. Original locality un-
known [Table Bay, designated by Kilbum and Rippey,
1982:211],

Fissurella scutellum; Krauss, 1848:63.

6 Contributions in Science, Number 379

McLean and Kilburn: African Fissurellids

Megatebennus (section Amblychilepas) scutellum; Pilsbry,
1890:184, pi. 39, fig. 89, pi. 44, figs. 99, 100, 1, 2.
Fissurellidaea scutella [sic]; Turton, 1932:207.
Amblychilepas scutella [sic]; Barnard, 1963:286, figs. 21b,
22d-f.

Amblychilepas scutellum; Kensley, 1973:29, fig. 30 [drawing
of animal]; Tietz and Robinson, 1974:48, pi. 26 [photo-
graph of animal],

Amblychilepas scutellum scutellum; Kilburn and Rippey,
1982:35, 212, pi. 6, fig. 1 (part).

Not Fissurella scutella of Reeve, 1849:pl. 6, fig. 33 [= A.
javanicensis (Lamarck)].

Not Fissurellidea scutella of Sowerby II, 1862:203, pi. 9, fig.

207 [= A. javanicensis (Lamarck)].

Fissurellidea sella Sowerby , II, 1862:203, pi. 8, fig. 197; Tur-
ton, 1932:207. Type locality: “South Africa.”
Megatebennus (section Amblychilepas) sella; Pilsbry, 1890:
185, pi. 62, fig. 3 [copy Sowerby II].

(b) D. scutellum hiantula (Lamarck, 1 822)

Figures 2, 19-22

Fissurella hiantula Lamarck; 1822:14; Mermod, 1950:708,
figs. 18.1, 18.2, 18.3 [syntypes, Lamarck Collection], Type
locality: “Mer des Indes” [Algoa Bay, here designated],
Fissurellidea hiantula; Sowerby II, 1862:202, pi. 8, figs. 1 93—
195.

Not Fissurellidea hiantula of Pilsbry, 1 890: 1 79, or other au-
thors treating species from Argentina [= Fissurellidea me-
gatrema Orbigny, 1841],

Amblychilepas scutellum hiantula; Kilburn and Rippey, 1982:
35, 211, ph 6, fig. 1 (part).

Fissurella incarnata Krauss, 1848:65, pi. 4, fig. 7; Janus,
1961:3, pi. 1, figs. 4-6. Type locality: “In sinu tabulari et
falso, in litore natalensi.”

Fissurellidea incarnata; Sowerby II, 1862:203, pi. 8, fig. 109.
Megatebennus (section Amblychilepas) incarnata; Pilsbry,
1890:186, pi. 35, figs. 4, 5 [copy Krauss],

Fissurellidea incarnata maculata Turton, 1932:206, pi. 53,
fig. 1431. Type locality: Port Alfred.

Fissurellidea multilineata Turton, 1932:206, pi. 53, fig. 1432.
Type locality: Port Alfred.

Fissurellidea albanyana Turton, 1932:207, pi. 54, fig. 1433.
Type locality: Port Alfred.

Fissurellidea nigrostrigata Turton, 1932:207, pi. 54, fig. 1435.
Type locality: Port Alfred.

DESCRIPTION. Shell saddle-shaped, anterior and pos-
terior ends raised; sculpture of fine raised ribs of nearly equal
size, but with some ribs twice the width of adjacent ribs.

Following Kilburn and Rippey (1982), we recognize two
geographic subspecies, with distributions overlapping in False
Bay, where they intergrade. To the west (on the cold tem-
perate Atlantic coast) occurs the typical D. scutellum scutel-
lum, which has a large thick shell (attaining 40 to 50 mm in
length) with moderately to extremely raised ends, and a drab
pattern of dark gray to olive-brown. To the east (on the warm
temperate south coast) it is replaced by D. scutellum hiantula,

which is smaller (length 30 mm or less), thinner-shelled, with
ends less raised, and a more vividly and delicately colored
pattern, generally pink or brown with white rays or spots.

TYPE MATERIAL. As noted by Pilsbry ( 1 890) and Kil-
bum and Rippey (1982:211), Gmelin (1791) cited a clearly
recognizable figure of Meuschen (1782:pl. 2, fig. 3) to rep-
resent Patella scutellum. Meuschen’s illustration is a type
figure; a type specimen is not available.

There are three specimens labeled Fissurella hiantula in
the Lamarck collection, as discussed by Mermod (1950).
These specimens clearly relate the name to the eastern sub-
species of D. scutellum. Of three syntypes figured by Mermod,
specimen number 1 , at 3 1 mm in length, is close to Lamarck’s
cited dimensions. Lamarck’s (1822) reference to an illustra-
tion of Born (which we now identify as Pupillaea aperta
(Sowerby, 1825)) is discounted because type material of F.
hiantula is available.

Type material for F. incarnata Krauss, was not located by
Janus (1961), who designated a neotype in the Stuttgart Mu-
seum. Janus selected a specimen that agrees with our concept
of F. scutellum scutellum, but Krauss’s figures agree with F.
scutellum hiantula. We therefore retain the name in the syn-
onymy of F. scutellum hiantula.

The holotype of Fissurellidea sella, from “South Africa”
is in the BMNH.

Turton’s types, all from Port Alfred, are in the Oxford
University Museum.

DISTRIBUTION. South Africa: Natal North Coast to Sal-
danha Bay, Cape Province.

Shells of both subspecies are well represented in museum
collections from numerous localities. The following pre-
served specimens of the typical subspecies have been ex-
amined: Saldanha Bay (NM B99 1 8); Table Bay (NM B99 1 7);
Cape Town (LACM 25203). These specimens tend to have
the propodium contracted, some to the point of nearly con-
cealing the feature. Preserved specimens of D. scutellum
hiantula are scarce in museum collections. Our evidence that
the eastern subspecies has a propodium similar to that of D.
scutellum is based on: 1) the specimen illustrated by Tietz
and Robinson ( 1 974:pl. 46), presumably from the Tsitsikama
coast, which shows a pink, rather than drab shell, as expected
in the eastern subspecies; 2) a single specimen from Still Bay
(NM A2536), which has an intact, dried animal with well-
developed propodium; and 3) three preserved specimens from
Algoa Bay (NM B9919).

Some specimens provisionally identified as D. scutellum
hiantula have fewer ribs overall and have ribbing of alter-
nating strength (Figs. 20, 21). Color variation is similar to
that of D. scutellum hiantula. Until preserved specimens be-
come available, the identity of this form is not certain. It
could be a variant of this species, or it could prove to be yet
another species of Amblychilepas.

MATERIAL EXAMINED, (a) D. scutellum scutellum:
SOUTH AFRICA: Saldanha Bay (NM 5584); Shell Bay (NM
8725); various localities in Table Bay (NM 638 1 , NM 1271,
NM 6641, NM A2975); Kommetjie, west coast Cape Pen-
insula (NM 5586, NM A4189). Simonstown, shallow dredg-
ings (NM 8976, NM 6769) (Figs. 17, 18); Strandfontein, False

Contributions in Science, Number 379

McLean and Kilburn: African Fissurellids 7

Figures 17-21. Shells of Dendrofissurella species; exterior and interior views with anterior at top; lateral views of left side. Figs. 17, 18.
Dendrofssurella scutellum scutellum (Gmelin, 1791). 17. Simonstown, False Bay, Western Cape Province, South Africa, NM 6769, 33.7 x
18.8 x 8.4 mm. 18. Same lot, 32.5 x 19.8 x 9.8 mm. Figs. 19-21. D. scutellum hiantula (Lamarck, 1822). 19. East London, Eastern Cape
Province, South Africa, NM A4186, 30.0 x 17.6 x 8.0 mm. 20. Cape Agulhas, Western Cape Province, South Africa, NM A2535, 20.0 x
1 1.6 x 4.8 mm. 21. Port Elizabeth, Eastern Cape Province, South Africa, LACM 3765, 15.0 x 8.8 x 3.9 mm.

Bay (NM A3961); off Macassar Beach, False Bay, 10 fm.
(NM A3 123).

(b) D. scutellum hiantula: SOUTH AFRICA, CAPE
PROVINCE: Cape Agulhas (NM A2535); Still Bay (NM
A2536, NM A2531); Mossel Bay (NM A5307); Jeffreys Bay
(NM 5886, NM B6 1 43); Algoa Bay (NM B2868, NM B 1 795,
B439); Port Alfred (NM B440-2, NM B6142, NM B6355);
East London (NM A4 1 86, NM 8363) (Fig. 1 9). TRANSKEI:
Kei River mouth (NM C3476); Nxaxo River mouth (NM
C3773); Qolora River mouth (NM C3406); Sandy Point (NM
C3665); Dwesa (NM C596 1); Lwandile/Mdumbi (NM C75);
Hluleka (NM Cl 500); Coffee Bay (NM A801, NM B6139);
Nthlonyane (NM B1430); Mkambati (NM C5608); Mbotyi
(NM A2534); Mzamba (NM B4550, NM 7142). NATAL:
Port Shepstone (NM 8986, NM 9128); Kelso (NM 5885);
Mtwalume (NM B8605); Umkomaas (NM 1273, NM 7151);
Durban (NM B4884); Tongaat (NM 7141 1); 3 mi. offUm-
hlanga Rocks, 12-13 fm. (NM A272); Umhlali beach (NM
A4575); Mvoti River mouth (NM 7144).

COMPARISONS. On shell characters both subspecies of
D. scutellum differ from the three species of Medusafissurella
in having less prominent, nonscabrous ribs and in having the
posterior end more raised.

REMARKS. Reeve (1849) and Sowerby II (1862) incor-
rectly considered F. scutellum an earlier name for the Aus-
tralian A. javanicensis; their figures show the latter. Krauss
(1848) and Barnard (1963) used A. scutellum for the South
African species, but incorrectly placed A. javanicensis in its
synonymy. Pilsbry (1890) erroneously used the name Fis-
surellidea hiantula for an Argentinean species, which Mc-
Lean (1984a) relegated to the synonymy of Fissurellidea me-
gatrema Orbigny, 1841.

The prominent dendritic propodium of D. scutellum has
previously been figured by Kensley (1973) and by Tietz and
Robinson (1974); the latter figure is reproduced here (Fig.
22). Kensley (1973) illustrated a bifurcation in the propo-
dium, although all specimens that we have examined have
a single central trunk.

Genus Amblychilepas Pilsbry, 1890

Amblychilepas Pilsbry, 1890:184. Type species: Fissurella
trapezina Sowerby, 1835 (= F. javanicensis Lamarck, 1822).
Recent, Australia.

8 Contributions in Science, Number 379

McLean and Kilburn: African Fissurellids

Figures 22, 23. Living animals of Dendrofissurella and Amblychilepas. 22. Dendrofissurella scutum hiantula, dorsal view of living animal,
Tsitsikama Coastal National Park, Eastern Cape Province, South Africa; overall length about 80 mm; after Tietz and Robinson (1974). 23.
Amblychilepas nigrita (Sowerby, 1835), dorsal view of living animal, New South Wales, Australia, overall length about 80 mm; photograph
by John Fields, courtesy AMS.

Sophismalepis Iredale, 1 924:2 1 9. Type species: Fissurella ni-
grita Sowerby, 1835. Recent, Australia.

DESCRIPTION. Shell oval, anterior end narrower than
posterior, ends raised; sculpture of radial striae; foramen nearly
central, elongate oval. Foot projecting posterior to shell for
distance greater than length of shell; shell edge enveloped by
upper fold of mantle, from which long papillae extend toward
foramen; propodium unmodified. Massive outer tooth of
radula quadricuspid.

REMARKS. Except for A. platyactis new species, species
of Amblychilepas are restricted to Australia. Figured here for
comparison (Fig. 23) is a living animal of the Australian A.
nigrita (Sowerby, 1835). As in the new species here described,
the papillae of the upper fold of the mantle of A. nigrita are
well developed and project toward the foramen. The pro-
podium of the Australian type species (A. javanicensis) is
unmodified (based on AMS C. 1 17375).

The following species has been confused with the common
south African species here reassigned to Dendrofissurella scu-
tellum hiantula.

Amblychilepas platyactis new species

Figures 3, 24, 25

Amblychilepas scutellum [non Gmelin, 1791]; Odhner, 1932:

298, fig. 31 [drawing of animal].

DIAGNOSIS. Shell to 30 mm in length, saddle-shaped,
narrowed anteriorly; anterior markedly raised; posterior
slightly raised; radial ribs alternating in strength, separated
by deeply incised grooves. Primary ribs broad, low, and fiat-

topped, up to twice as broad as secondary ribs and 3 times
broader than tertiary ribs; primary ribs broader on sides of
shell than on ends; concentric growth lamellae thin and raised,
especially prominent on broad primary ribs. Color dark red
with white rays.

Propodium unmodified; body extending posteriorly for one
shell length; upper lobe of mantle enveloping edge of shell
and having long, elaborate papillae corresponding to broad
primary ribs.

DESCRIPTION. Shell outline elongate-oval in dorsal view,
slightly narrowed anteriorly; anterior and posterior ends
raised, lateral profile of base evenly curved. Radial sculpture
of flat-topped ribs separated by narrow incised grooves rather
than interspaces. Primary ribs 1 4, twice as broad as secondary
ribs, which in turn are twice as broad as tertiary ribs. Primary
ribs at anterior and posterior ends more elevated than those
of sides, particularly in early growth stages. Concentric sculp-
ture of thin, raised incremental growth lines, arched upon
crossing primary ribs, to a lesser extent on crossing secondary
and tertiary ribs. Foramen elongate-oval, slightly broader
posteriorly, ‘/5 shell length. Color brick red, with white rays
that tend to emerge at later growth stages and correspond to
primary ribs. Margin faintly crenulate to correspond with
exterior sculpture, shell edge angulate at sides, thinner an-
teriorly and posteriorly. Callus surrounding foramen faintly
outlined in pink, more bluntly terminating posteriorly than
anteriorly.

Propodium unmodified, outline of foot elongate-oval, body
extending posteriorly for one shell length. Sides of foot with
pustulose tubercles; epipodial tentacles of neck weakly de-
veloped; mantle lobe with tubercles like those on foot sides
but smaller; lower edge of mantle lobe with projecting pa-

Contributions in Science, Number 379

McLean and Kilburn: African Fissurellids 9

Figures 24, 25. Amblychilepas platyactis, new species, exterior and interior views with anterior at top; lateral views of left side. 24. Holotype,
Port Alfred, Eastern Cape Province, South Africa, NM B6397/T2744, 1 7.2 x 10.3 x 3.9 mm. 25. Paratype, False Bay, Eastern Cape Province,
South Africa, NM 6769/T3009, 20.5 x 1 1.0 x 4.5 mm.

pillae having numerous projecting points; upper edge of man-
tle with fewer papillae that are about three times the size of
the lower edge papillae, about 14 major papillae altogether,
corresponding to broad primary ribs. Major papillae sepa-
rated by less prominent papillae. Tips of cephalic tentacles
reddish in preservative.

DIMENSIONS. Holotype, shell length 17.2, width 10.3,
height 3.9 mm. Maximum length 30.8 mm (NM 6769/T3009).

TYPE MATERIAL. Holotype, NM B6397/T2744, Port
Alfred, Eastern Cape Province, South Africa, collected by R.
Kilbum, 1966 (Fig. 24). Paratype 1, LACM 2108, Port Alfred,
collected by E. Warren, July 1912. Paratype 2, NM B9916/
T3008, Kommetjie, collected by C.M. Connolly. Paratypes
3-5, NM B9929/T3055, Kwelera, E of East London, col-
lected by C.M. Connolly. Paratypes 6-1 1, NM 6769/T3009
(Fig. 25) Simonstown, False Bay, collected by C.M. Connolly.
Holotype and paratypes wet-preserved, except for paratypes
6-1 1 , shells only.

DISTRIBUTION. South Africa: East London to Kom-
metjie, Atlantic coast of Cape Peninsula.

COMPARISONS. On anatomical characters, there is no
difficulty in distinguishing intact specimens of A. platyactis
from both subspecies of Dendrofissurella scutellum, the for-
mer having elaborate papillae on the upper lobe of the mantle
and lacking the propodial elaboration; the latter having small,
simple papillae on the upper lobe and having the trunklike
propodium. Shells, however, closely approach those of D.
scutellum hiantula, of which there are specimens with broad
primary ribs (Figs. 20, 21) up to twice the breadth of the
lesser ribs. However, no specimens of D. scutellum hiantula
are known with the extremely broad ribs of A. platyactis,
which may be three times the breadth of the lesser ribs. The
development of the scaly sculpture on the primary ribs is
much more extreme in the young stages of A. platyactis.
Shells of all specimens are red and white rayed, which in-
dicates that the range of color variation is minimal in A.
platyactis. The posterior end of the shell of A. platyactis is
less raised than that of D. scutellum.

REMARKS. We name this species with reluctance, con-
sidering the large number of synonyms pertaining to D. scu-

tellum hiantula. However, original descriptions of all these
taxa make no reference to broad ribbing and it is clear that
none of them has sculpture to match the type specimens of
A. platyactis. The specimen figured by Odhner ( 1 932) is pre-
sumed to be this species on the basis of lack of propodial
development and the papillae of the upper lobe of the mantle.

ETYMOLOGY. The name is Greek, meaning with wide
or fiat spokes or rays, suggested by the extremely broad ribs
of this species.

DISCUSSION. Dendrofissurella scutellum (both subspe-
cies) and the three species of Medusafissurella have very dif-
ferent expressions of the propodium, here considered diag-
nostic at the generic level. However, the vexing question of
function remains. One might speculate that the propodial
tentacles serve to trap or hold food, such as drifting algae.
Yet, in these species the snout, which is no larger than in
other fissurellid genera, is dorsal to the proportionally longer
propodium (Fig. 26). A role in feeding would be more readily
understood if these tentacles were attached to the snout in-
stead of the propodium, access to the mouth thereby being
closer. In the fissurellids described here, the snout would
presumably have to be extended through the tentacles to
reach the food, but the propodium is not bifid, as would be
expected if this were the case. Yet, the mouth and the snout
are more laterally compressed than in fissurellid genera lack-
ing the propodium.

Stomach contents of a preserved specimen of Medusafis-
surella dubia included branched coralline algae, suggesting
that the diet of this species is comprised of encrusting algae,
not unlike the diet of Fissurella barbadensis Gmelin, ex-
amined by Ward (1966). Thus, there is no indication that
propodial development corresponds to an unusual diet in
these species.

The two fissurelline genera having these propodial modi-
fications occur in southern Africa and the western Indian
Ocean, a region central to the distribution of genera in the
subfamily. All five of the genera keyed above are known from
South Africa. Fissurella, with its unmodified propodium, has
more numerous species in the western hemisphere, on both
sides of the Atlantic and in the eastern Pacific. Amblychilepas

10 Contributions in Science, Number 379

McLean and Kilburn: African Fissurellids

and Macrochisma are better represented to the east, Ambly-
chilepas in Australia and Macrochisma in Japan and Aus-
tralia.

ACKNOWLEDGMENTS

We thank J. Christiaens of Hasselt, Belgium for reading the
manuscript and loaning specimens of several species from
his collection. Barry Roth, California Academy of Sciences,
first brought the preserved specimen of Medusafissurella sa-
lebrosa to our attention. Preserved specimens of M. chem-
nitzii were furnished on loan by Philippe Bouchet of the Paris
Museum. Winston Ponder of the Australian Museum ar-
ranged the loan of comparative material and the photograph
of Amblychilepas nigrit a. We thank B. Hayes and F. Graeve
of Port Elizabeth for preserved Dendrofssurella scutellum
hiantula and Mrs. S. Muller of the East London Museum for
shells of M. chemnitzii. The drawing of M. salebrosa recon-
structed as living was made by LACM volunteer artist Yvonne
Albi. SEM photographs were made at the Center for Electron
Microscopy and Microanalysis of the University of Southern
California with the assistance of Alicia Thompson. LACM
museum photographers Dick Meier and John DeLeon as-
sisted in the preparation of the illustrations. We thank our
anonymous reviewers for helpful suggestions.

LITERATURE CITED

Barnard, K.H. 1963. Contributions to the knowledge of
South African marine Mollusca. Part IV. Gastropoda:
Prosobranchiata: Rhipidoglossa, Docoglossa; Tecti-
branchiata, Polyplacophora, Solenogastres, Scaphopo-
da. Annals of the South African Museum 47(2):20 1-360.
Biggs, H.E.J. 1969. Marine Mollusca of Masirah Island,
South Arabia. Archiv fiir Molluskenkunde 99(3):20 1—207.
Bosch, D., and E. Bosch. 1982. Seashells of Oman. London:
Longman, 206 pp.

Christiaens, J. 1973. Les Fissurelles Europeennes. II. Le
genre Fissurella Bruguiere, 1798. Informations de la So-
ciete Beige de Malacologie, Serie 2, no. 5, pp. 67-98,
pis. 1-4.

. 1974. Le genre Diodora (Gastropoda): especes non-

europeennes. Informations de la Societe Beige de Ma-
lacologie, Serie 3, nos. 6-7, pp. 73-97, pis. 1-3.
Dautzenberg, P. 1929. Mollusques testaces marins de
Madagascar. Faune des Colonies franqaises 3:321-636,
pis. 4-7.

Gmelin, J.F. 1791. Caroli a Linne Systema naturae per
regna tria naturae. Editio decima tertia. Leipzig, vol. 1,
pt. 6, Vermes, pp. 3021-3910.

Hickman, C.S. 1981. Evolution and function of asymmetry
in the archaeogastropod radula. The Veliger 23(3): 1 89—
194.

. 1 984. Implications of radular tooth-row functional

integration for archaeogastropod systematics. Malaco-
logia 25(1): 143-160.

Iredale, T. 1924. Results from Roy Bell’s molluscan col-
lections. Proceedings of the Linnean Society of New South
Wales 49:179-278, pis. 33-36.

Figure 26. Right lateral view of Medusafissurella salebrosa, based
on specimen in Fig. 14, reconstructed to show the mantle folds in
contact with shell, tentacular propodium, snout, and cephalic ten-
tacles with eyes at base, drawn by Yvonne Albi.

Janus, H. 1961. Die Typen und Typoide sudafrikanischen
Meeresmollusken im Staatlichen Museum fur Natur-
kunde. I. Gastropoda. Beitrag fiir Naturkunde 70:1-19,
3 pis.

Kensley, B. 1973. Sea-shells of Southern Africa, Gastro-
pods. Cape Town: Maskey Miller Limited, 236 pp., 910
figs.

Kilburn, R., and E. Rippey. 1982. Sea shells of southern
Africa. Johannesburg: Macmillan South Africa, 249 pp.,
46 pis.

Krauss, F. 1 848. Die Sudafrikanischen Mollusken. Ein Bei-
trag zur Kentniss des Kap- und Natallandes und zur geo-
graphischen Verbreitung derselben, mit Beschreibung und
Abbildung der neuen Arten. Stuttgart: Ebner and Seubert,
140 pp.

Lamarck, J.B.P.A. de M. de. 181 5-1822. Histoire naturelle
des animaux sans yertebres. Paris, v. 1-7.

Martini, F.H.W., and J.H. Chemnitz. 1769-1795. Neues
systematisches Conchylien-Cabinet. Niiremburg, v. 1-
11.

McLean, J.H. 1970. Descriptions of a new genus and eight
new species of eastern Pacific Fissurellidae, with notes
on other species. The Veliger 12(3):362-367.

. 1984a. Shell reduction and loss in fissurellids: a

review of the genera and species in the Fissurellidea
group. American Malacological Bulletin 2:21-34.

. 1984b. Systematics of Fissurella in the Peruvian

and Magellanic faunal provinces (Gastropoda: Proso-
branchia). Contributions in Science, Natural History
Museum of Los Angeles County, No. 354, 70 pp.

Mermod, G. 1950. Les types de la collection Lamarck au
Museum de Geneve. Mollusques vivantes, I. Revue Suisse
de Zoologie 57(34):687-756.

Meuschen, F.C. 1782. Conchyliologische Briefe. Der Na-
turforscher 18:1 1, pi. 2.

Odhner, N.H. 1932. Zur Morphologie und Systematik der
Fissurelliden. Jenaische Zeitschrift fiir Naturwissen-
schaft 67:292-309.

Perez-Farfante, I. 1943. The genus Diodora in the western
Atlantic. Johnsonia 1(11): 1-20.

Contributions in Science, Number 379

McLean and Kilburn: African Fissurellids 1 1

Pilsbry, H.A. 1890. Stomatellidae, Scissurellidae, Pleuro-
tomariidae, Haliotidae, Scutellinidae, Addisoniidae,
Cocculinidae, Fissurellidae. Manual of Conchology,
Philadelphia, v. 12, 321 pp., 65 pis.

Reeve, L. 1849-1850. Monograph of the genus Fissurella.
Conchologia Iconica, London, v. 6, 16 pis., text not pagi-
nated.

Sowerby, G.B. 1835a. Shells collected by Mr. Cuming on
the western coast of South America, and among the
islands of the South Pacific Ocean. Proceedings of the
Zoological Society of London, for 1834:123-128.

. 1835b. A catalogue of the Recent species of Fis-
surella. The Conchological Illustrations. London, pp. 1-
8, 77 figs, on 12 unnumbered plates.

Sowerby, G.B., [II]. 1862. Monograph of the family Fis-

surellidae. Thesaurus Conchyliorum, or Monographs of
the Genera of Shells. London, v. 3:183-206, pis. 1-9.

Thiele, J. 1929. Handbuch der systematischen Weichtier-
kunde. Jena: Gustav Fischer, 1 154 pp.

Tietz, R.M., and G.A. Robinson. 1974. The Tsitsikama
shore, a guide to the marine invertebrate fauna of Tsitsi-
kama National Park. National Parks Board, Republic
of South Africa, 115 pp., 50 pis.

Turton, W.H. 1932. The marine shells of Port Alfred, S.
Africa. London: Oxford University Press, xvi + 331 pp.

Ward, J. 1966. Feeding, digestion, and histology of the
digestive tract in the keyhole limpet Fissurella barba-
densis Gmelin. Bulletin of Marine Science 16:685-695.

Accepted 24 March 1986.

12 Contributions in Science, Number 379

McLean and Kilburn: African Fissurellids

The scientific publications of the Natural History Museum of Los Angeles County have been
issued a: irregular intervals in three major series; the issues in each series are numbere
individually, and numbers run consecutively, regardless of the subject matter.

Contributions in Science, a miscellaneous series of technical papers describing orig-
inal research in the life and earth sciences. . . jjjfr

Science Bulletin, a miscellaneous series of monographs describing original research
in the .life and earth sciences. .This series; was discontinued in 1978 with the issue of
M timbers 29 and 30; monographs -are. nowf published by the Museum in Contributions
.it Science.' I .-'h ; '-d. bi' J

... : • , j h.i | :

Scion, cd Series.: tong articles on natural history topics, generally written for the layman.

Copies of the publications in these series -are sold through the Museum Book Shop. A catalog

is available on request. I :• h , ‘ ; | ; ; ' i i1;1 j . : ; ' ; • : j- : Li: L.;! j;

i : li t

Velati ve to scholarly inquiry and collections but not. reporting the fesiilifs of original research.
Issue is authorized by the museum 's Scientific Publications Committee; however, manuscripts
do not receive anonymous peer review. Individual Technical Reports may be obtained from

the relevant Section of the museum.

trio

EVOLUTION AND CLASSIFICATION OF THE
LATE CRETACEOUS-EARLY TERTIARY
GASTROPOD PERISSITYS

W.P. Popenoe and L.R. Saul

Contributions in Science, Number 380
Natural History Museum of Los Angeles County
12 May 1987

ISSN 0459-8113

Natural History Museum of Los Angeles County
900 Exposition Boulevard
Los Angeles, California 90007

CONTENTS

ABSTRACT 1

INTRODUCTION 1

OCCURRENCE 3

MORPHOLOGIC CHANGE 5

Patterns of Change 5

Rate of Morphologic Change 8

CLASSIFICATION 9

Species 9

Genus 9

Family 9

Superfamily 10

Order 11

SYSTEMATIC PALEONTOLOGY 11

Family Perissityidae new family 11

Genus Perissitys Stewart, 1927 12

Perissitys cretacea (Cooper, 1 896) 12

Perissitys elaphia new species 13

Perissitys brevirostris (Gabb, 1864) 16

Perissitys pacific a new species 17

Perissitys colocara new species 19

Perissitys si antoni { Stewart, 1927) 20

Perissitys stewarti (Zinsmeister, 1983) 21

Genus Pseudocymia new genus 21

Pseudocymia aurora new species 22

Pseudocymia (?) aitha new species 22

Pseudocymia(l) cf. R.(?) aitha new species 24

Pseudocymia^ ) cahalli new species 24

Pseudocymiaif) kilmeri new species 24

Genus Murphitys new genus 25

Murphitys michaeli new species 25

Murphitys corona new species 27

Murphitys madonna new species 27

Genus Christitvs new genus 28

Christitys delta new species 28

Christitvs medica new species 29

Christitys martini new species 30

ACKNOWLEDGMENTS 30

LITERATURE CITED 31

LOCALITIES CITED 34

EVOLUTION AND CLASSIFICATION OF THE
LATE CRETACEOUS-EARLY TERTIARY
GASTROPOD PERISSITYS

W.P. Popenoe and L.R. Saul1

ABSTRACT. A new neogastropod family Perissityidae is proposed
for several lineages of American Pacific Coast gastropods having
species of early Senonian age that are very similar but late Senonian
species that are disparate. Four genera of Perissityidae are discussed:
the type genus Perissitys Stewart, 1927, Pseudocymia new genus,
Murphitys new genus, and Christitys new genus. Seven species are
assigned to Perissitys: P. cretacea (Cooper), P. elaphia new species,
P. brevirostris (Gabb) (type species), P. pacifica new species, P. co-
locara new species, P. stantoni (Stewart), and P. slewarti (Zins-
meister). The genus ranges in age from Coniacian through early
Paleocene; and its evolutionary changes include reduction and dis-
appearance of tubercles on the inner apertural margins, comparative
increase in length of the siphonal canal, and extension of the inner
lip callus to finally encompass nearly the entire shell in the later
Maastrichtian, followed subsequently by reduction of this callus in
the Paleocene.

Pseudocymia new genus, with type species P. aurora new species
includes P.(?) ait ha new species, P.(?) cahalli new species, and P.(?)
kilmeri new species and ranges from Turonian into the Maastrich-
tian. Evolutionary changes in Pseudocymia include reduction of den-
ticles on the inner apertural margins and elongation of the entire
shell.

Murphitys new genus comprises M. michaeli new species (type
species), M. corona new species, and M. madonna new species, and
ranges from Coniacian to Maastrichtian. Its evolutionary changes
include increasing angularity of whorl profile, development of col-
umellar folds, and expansion of the outer lip to form a rimmed
aperture.

Christitys new genus, comprising C. delta new species, C. medica
new species (type species), and C. martini new species ranges from
Coniacian to Campanian. Its evolutionary changes include devel-
opment of a more pyriform shape and reduction of the denticulations
of the outer lip.

Sequence and rate of morphologic change in these molluscan groups
have significance in principles of gastropod taxonomy and utility for
correlation.

INTRODUCTION

Erection of a family, Perissityidae, to encompass the enig-
matic neogastropod lineages of Perissitys, Cophocara, and
“ Hindsia nodulosa (Whiteaves)” was first suggested by W.P.

Contributions in Science, Number 380, pp. 1-37
Natural History Museum of Los Angeles County, 1987

Popenoe at the 1971 Cordilleran section meeting of the Geo-
logical Society of America (Popenoe, 1971, p. 179; Saul,
1971, p. 189). Popenoe intended to propose the family and
describe half of the perissityid genera; Saul was to (and will)
describe the remainder. This paper treats those genera and
species of Perissityidae that Popenoe intended to describe.
Some problems of biostratigraphic correlation, which we be-
lieve are now resolved, delayed the writing of this paper
beyond the time allotted to Popenoe; the paper has, therefore,
been completed by Saul.

The genera Perissitys and Cophocara were described by
Stewart (1927), but neither was assigned to a family, both
being consigned to “doubtful systematic position.” Earlier,
Cossmann (1901) and later, Wenz (1943), Erickson (1974),
and Zinsmeister (1983) placed these genera near Tudicla, an
association that may be more of form than of phylogeny. As
traced in this paper the evolutionary sequence of which Peris-
sitys brevirostris (Gabb, 1 864) and Cophocara stantoni Stew-
art, 1927, form a part, extends from Coniacian into early
Selandian (middle Paleocene). End members of this lineage
would scarcely be included in the same family by most sys-
tematists were intermediate stages unknown.

Three new genera, Pseudocymia, Murphitys, and Christi-
tys, are represented in the Coniacian by species resembling
Perissitys cretacea (Cooper). These genera diverged from Pe-
rissitys and from each other in the Santonian, and by Cam-
panian time all were represented by species of decidedly dif-
ferent aspect.

Six previously described West Coast species are unhesi-
tantly allotted to the family Perissityidae: Perissitys brevi-
rostris (Gabb, 1864), Fusus kingii Gabb, 1864, Cophocara
stantoni Stewart, 1927, C. stewarti Zinsmeister, 1983, Fas-

1 . At the time that the research described in this paper was con-
ducted, both authors were with the Department of Earth & Space
Sciences, University of California, Los Angeles. Popenoe died in
1981. Saul is currently with the Invertebrate Paleontology Section
of the Natural History Museum of Los Angeles County.

ISSN 0459-81 13

Figure 1. Index map to geographic occurrence of perissityids de-
scribed in this paper. 1. Agua Caliente Canyon, San Rafael Mts.,
Santa Barbara Co., California. 2. Antelope Creek, Mill Creek, and
Tuscan Springs, Tehama Co., California. 3. Benson (= Brannan
Creek), Blundon Point, and Elkhom Creek, Nanaimo Basin, Van-
couver Island, British Columbia. 4. Browns River, Comax Basin,
Vancouver Island, British Columbia. 5. Butte Creek, Chico Creek,
and Pentz, Butte Co., California. 6. Cottonwood Creek in Cholame
Valley Quadrangle, Monterey Co., California. 7. Deer Valley and
vicinity of Martinez, Contra Costa Co., California. 8. East side of
the Diablo Range from Stanislaus Co. south to Fresno Co., Califor-
nia. 9. Point Loma, San Diego Co., California. 10. Punta San Jose,
Baja California Norte, Mexico. 11. East of Redding, Shasta Co.,
California. 12. San Antonio del Mar and Johnson’s Ranch, Baja
California, Mexico. 13. Santa Ana Mountains, Orange Co., Califor-
nia. 14. Santa Catarina Landing area, Baja California, Mexico. 15.
Lake Nacimiento, southern Santa Lucia Range and Santa Margarita
Lake, La Panza Range, San Luis Obispo Co., California. 16. Santa
Monica Mountains, Los Angeles Co., California. 17. Simi Hills, Los
Angeles and Ventura cos., California. 18. Sucia Island, San Juan
Co., Washington. 19. Warm Springs Mountain, northwestern San
Gabriel Mts., Los Angeles Co., California. 20. Vicinity of Carlsbad,
San Diego Co., California. 21. Rumsey Hills, Yolo Co., California.
22. Punta Banda, Baja California, Mexico. 23. Chignik Lagoon, Alas-
ka Peninsula, Alaska. 24. East of San Juan Capistrano, Orange Co.,
California. Place names mentioned in text listed alphabetically: 1.
Agua Caliente Canyon; 2. Antelope Creek; 11. Bear Creek; 13. Bee
Canyon; 17. Bell Canyon; 3. Benson (= Brannan) Creek; 3. Blundon
Point; 4. Browns River; 21. Buckeye Creek; 5. Butte Creek; 15.
Cantinas Creek; 20. Carlsbad; 5. Chico Creek; 23. Chignik Lagoon;
8. Coalinga; 11. Clover Creek; 6. Cottonwood Creek in Cholame

ciolaria nodulosa Whiteaves, 1874, = Hindsia nodulosa
Whiteaves, 1879, and Sistrum ( Ricinula ?) cretaceum Cooper,
1896. Probably Nekewis simiensis Zinsmeister, 1983, and
possibly N. io (Gabb, 1864), N. washingtoniana (Weaver,
1912), and N. nehalemensis (Anderson & Martin, 1914) are
also perissityids. T. Kase has provided plaster casts of spec-
imens of some Japanese perissityids, including Surculites ?
cf. 5.? fusoides Nagao, 1939, and Pseudoperissitys bicarinata
Nagao & Otatume, 1938. Additionally, descriptions and fig-
ures suggest that Trachytriton sachalinensis Schmidt, 1873
[= Serrifususl sachalinensis (Schmidt) in Hayami & Kase
( 1977, p. 63) non Serrifusus sachalinensis (Nagao, 1932)], T.
duiensis Schmidt, 1873, Pyropsis sp. indet. of Nagao (1939),
Pyrifusus ( NeptuneUa ) kawakamiensis Nagao, 1939, and Fu-
sus (s.l.) volutodermoides Nagao, 1939, from the northwest-
ern Pacific are Perissityidae, as is Perisso/ax brevirostris Gabb
of Martin (1926) from Alaska. Fusus dusenianus Wilckens,
1907, Stmthiolariopsisl tunuda Wilckens, 1907, and Het-
erotenna praecursor (Wilckens, 1907) from the upper Cre-
taceous of southern Patagonia resemble perissityids, but the
aperture of none has been described. Perissitys? sp. A and B
of Erickson (1974), however, from the Fox Hills Formation
of North Dakota do not show any characteristics of this
family. We have not seen specimens of the undescribed
Cophocara sp. of Sohl (1967, p. 34) from the upper Cam-
panian of Wyoming. In previous classifications species as-
signable to the Perissityidae have been distributed among
Cassididae, Thaididae, Buccinidae, Fasciolariidae, Xancidae
or Vasidae, Tudiclidae, and Turridae.

Much of this diverse family assignment results from the
divergent evolutionary paths followed by the several lineages.
The lineage with changes best documented by specimens is
that of Perissitys. Adequate specimens, sufficiently well dis-
tributed through the stratigraphic sections, indicate style, pat-
tern, pace, and rate of evolution within this group.

Abbreviations used with catalog and locality numbers are:

ANSP = Academy of Natural Sciences of Philadelphia
CAS = California Academy of Sciences, San Francisco
CGS = Canada Geological Survey

CIT = California Institute of Technology (collections now at

LACMIP)

LACMIP = Los Angeles County Museum of Natural History,

Invertebrate Paleontology

Valley Quadrangle 17. Dayton Canyon; 7. Deer Valley; 3. Departure
Bay; 8. Diablo Range; 3. Elkhom Creek; 12. Johnson’s Ranch; 8.
Laguna Seca Hills; 17. Lang Ranch; 7. Martinez; 2. Mill Creek; 11.
Morley; 1 1 . Oak Run; 8. Oil Canyon; 1 1 . Old Cow Creek; 8. Ortigalita
Creek; 7. Pacheco; 8. Pacheco Pass; 8. Panoche Hills; 5. Pentz; 11.
Pine Timber Gulch; 9. Point Loma; 11. Price Hollow; 22. Punta
Banda; 14. Punta Canoas; 10. Punta San Jose; 11. Redding area; 7.
Riggs Canyon; 21. Rumsey Hills; 11. Salt Creek; 12. San Antonio
del Mar; 24. San Juan Capistrano; 21. Sand Creek; 13. Santa Ana
Mts.; 14. Santa Catarina Landing area; 15. Santa Margarita Lake;
16. Santa Monica Mts.; 17. Simi Hills; 18. Sucia Island; 11. Swede
Basin; 2. Tuscan Springs; 19. Warm Springs Mt.; 23. Whaler’s Creek.

2 Contributions in Science, Number 380

Popenoe and Saul: Perissitys

Figure 2. Stratigraphic occurrence of Perissitys, Murphitys, Christitys, and Pseudocymia in four California sections: Redding area (area 1 1 ),
based on Popenoe, 1943, and Matsumoto, 1960; Chico Creek (area 5), after Saul, 1983; Simi Hills (area 17), after Saul, 1983; and east side
Diablo Range (area 8), simplified from Saul, 1983. The columns are not all to the same scale. The Redding area, Simi Hills, and east side
Diablo Range columns are composites and each represents a considerable area; the Chico Creek column is based on outcrops along Chico
Creek. Campanian reversed magnetozone (Chron 33r) is from Ward et al. (1983). The position of some fossil localities is indicated by locality
number.

LSJU = Stanford University (collections now at CAS)
SDSNH = San Diego Society of Natural History
UCB = University of California, Berkeley
UCBMP = University of California, Berkeley, Museum of
Paleontology

UCLA = University of California, Los Angeles (collections
now at LACMIP)

UCR = University California, Riverside
USGS = United States Geological Survey
USNM = United States National Museum of Natural His-
tory

OCCURRENCE

Specimens of Perissitys, Pseudocymia, Murphitys, and
Christitys have been collected from more than 200 localities
from Alaska to Baja California, Mexico. Only seven areas

have provided stratigraphic successions of species: in British
Columbia on Vancouver Island (Figure 1, area 3), and in
California east of Redding (area 1 1 ), along Chico Creek (area
5), on the east side of the Diablo Range (area 8), near Warm
Springs Mountain (area 19), in the Simi Hills (area 17), and
in the Santa Ana Mountains (area 13). Stratigraphic position
of the perissityid occurrences at Redding, Chico Creek, Simi
Hills, and in the Diablo Range is diagrammed in Figure 2.
The probable chronologic ranges of species, derived from
their stratigraphic occurrences are plotted on Figure 3 with
the position of some Pacific West Coast zonal indicators.

Stratigraphic nomenclature of the Cretaceous east of Red-
ding (area 1 1 ) has been revised by Haggart (1986). Popenoe
( 1 943), in a progress report, placed the strata in six Members
I-VI. Type sections were left undesignated, but disposition
of the members has been shown on maps (e.g., Popenoe,
1943; Matsumoto, 1960; Trujillo, 1960). M.V. Kirk (fide

Contributions in Science, Number 380

Popenoe and Saul: Perissitys 3

CRETACEOUS

Turonianl I Santonian j
Coniacian

Campanian

Maastrichtian

VAVJ Pacifica j, y ,y.V/ • Stanton/

V ///// /A cntnrnrn V / /// Z/
I

PALEOCENE

Danian

Selandian

elaphia

cretacea
I i

12 1 aitha
aurora □ z □

I I

c

brevirostris

r~r.

Stewart/

Perissitys

rzTza

cahatti

I michaeti

' I

PZ\mec

'etta\ZA

^7 corona

l EZ 2 10 I L'/J

delta

I I
I I

I I
I I

med/co y / / \

| martini
, c hoizana

kitmeri
1 □

| madonna

I
I

Pseudocymia

Murphitys

c pescaderoensis

Christitys

webbi

chicoensis

I

c redder i

capensis

I chicoensis

'/////. 1

I

inornatus

■ i ' | r lufjniLfjiuuGiiiiiscruo |

J^roniceras Submortoniceras Metop/acenticeras

□ I I EZZZZZZZZI KZvZ/Z/1

CZZZZZZZZZ3'

Hoplitoplacenticeras \

Z

. iomaensis

w paynei

c. orienda

tzzzzzzF2

chaneyi

p quayiei

ZL

I K ‘ t__L *

p. adeiaidana pen
coiumna

n

Subprionocycius

! i

O

|oo [cn

| Chron 33r
-P* ro

IOO

fi—i

T777\

stock/

i pachecoensis

_jf' 'ilZJ

’ insular is infra-

granuioto

Turritella
chicoensis stock

Turritella
chaneyi stock

Bacu/ites

other

ammonites

P4

Time Scale

Figure 3. Geologic range of species of perissityids described in this paper, Turritella chaneyi and T. chicoensis stocks and some ammonites
fitted to the time scale of Palmer ( 1983) and Berggren et al. (1985a). Arrows indicate approximately the range allotted to B. capensis by Haggart
and Ward (1984) and Haggart (1984). Turritella zonation from Saul (1983), and Campanian reversed magnetozone, Chron 33r, from Ward
et al. (1983). In Harland et al. (1982, p. 74) Chron 33r is dated as 78.53-82.93 Ma.

Matsumoto, I 960, p. 4) suggested the name Redding For-
mation for these members. Popenoe (in Jones, et al., 1978)
named Members I— III, in ascending order Bellavista Sand-
stone, Frazier Siltstone, and Melton Sandstone. J.M. Haggart
(1986) delimited the outcrop area of the Redding Formation
and named Members IV-VI. In his Bear Creek Sandstone
he includes Member III and part of Member V of Popenoe
(1943) and the Melton Sandstone of Popenoe (in Jones et al.,
1978); in the Hooten Gulch Mudstone the mudstones of
Members IV and VI of Popenoe ( 1 943), and in the Oak Run
Conglomerate part of Member V of Popenoe (1943). Vir-
tually all of the Redding area specimens used in this study
were collected by Popenoe, and the recorded locality descrip-
tions reflect his view of the stratigraphy. Conversion to Hag-
gart's stratigraphy would require a more detailed map than
that available in Haggart (1986, fig. 4) and would result in a
complete revision of recorded names. As Popenoe is senior
author of this paper and his view of Haggart’s revision un-
known, the Redding area column (Figure 2, area 11) is a
modification of Popenoe (1943) compatible with his last ex-
pressed opinions. Despite the differences regarding strati-
graphic nomenclature between Popenoe and Haggart, age
assignments are mainly in accord, and stratigraphic position
of the specimens within sections exposed in the several creeks
is not disputed. The differences arise in recognizing lithologic
units present from section to section in the several creeks.

Perissityid gastropods are found in coarse- to very fine-
grained sandstone. Perissitys is most abundant in medium-

to fine-grained sandstone, and is associated with very shal-
low-water molluscan assemblages characterized by species
of Yaadia, Meekia, and Cymbophora; slightly deeper water
assemblages characterized by Calva, Pterotrigonia, and To-
ne a; and softer bottom assemblages characterized by Cras-
satella, Clisocolus, Cucullaea, Turritella, and Anchura. Peris-
sitys regularly occurs with naticids and may, like them, have
been confined to soft substrates (Taylor et al., 1980, p. 380).
Its occurrence in several assemblages suggests a fairly wide
water-depth range. Perissityids are usually a minor compo-
nent of these faunas and are inferred to have been predators.
Although morphologic changes in the Perissitys lineage are
apparent, there are neither correlative substrate nor marked
faunal association changes suggestive of adaptation to a new
habitat. The modifications of form in Perissitys may be fur-
ther adaptations to an infaunal, predatory life style, but not
to change in habitat.

The largest specimens of Perissitys are from near the top
of the Chico Formation on Chico Creek (area 5)— diameter
over 45 mm; Pleasants Sandstone Member of the Williams
Formation, Bee Canyon, Santa Ana Mountains (area 13)—
diameter 51.5 mm; Moreno Formation near Ortigalita Creek
(UCB loc. A-6618) (area 8)— diameter about 35 mm; and
the top of the Great Valley Series near Martinez (area 7)—
diameter 36 mm. Large specimens of Perissitys spp. have
thus been found in beds of mid Campanian through mid
Maastrichtian age.

The geologically oldest species are the most geographically

4 Contributions in Science, Number 380

Popenoe and Saul: Perissitys

restricted; whereas those of Campanian age are most widely
distributed. However, Campanian sandstones crop out more
widely than those of Coniacian-Santonian age, and the ab-
sence or abundance of perissityids may be an artifact of the
geological record rather than an indication of their place of
origin and a record of dispersal. The geologically oldest (Tu-
ronian-Coniacian) perissityids from western North America
are all from east of Redding (Figure 1, area 11). Ages of
Japanese perissityids are roughly contemporaneous with those
of the West Coast. Japanese perissityids from the Upper Yezo
Group, Pyropsis sp. indet. of Nagao, 1939, = Tudicla ( Peris -
sitys ) sp. in Hayami & Kase (1977); Pyrifusus ( Neptunella )
kawakamiensis Nagao, 1939, = Rhombopsisl kawakamien-
sis (Nagao) in Hayami & Kase (1977); Surcu/ites fusoides
Nagao, 1939; and Fusus (s.l.) volutodermoides Nagao, 1939,
are considered by Hayami and Kase (1977) to be of Coniacian
or Santonian age. Trachytriton sachalinensis Schmidt, 1873,
= Serrifusus ? sachalinensis (Schmidt) in Hayami & Kase
(1977)and T. duiensis Schmidt, 1873, = SP. duiensis (Schmidt)
in Hayami & Kase (1977) from Sachalin are “Campanian or
thereabouts” (Hayami and Kase, 1977). Pseudoperissitys bi-
carinata Nagao & Otatume, 1938, is Campanian or Maas-
trichtian (Hayami and Kase, 1977).

The earliest and most characteristic perissityids are from
the North Pacific and this family may be of North Pacific
origin as well as of predominantly North Pacific occurrence.
Zinsmeister (1983) indicates confamiliality for Cophocara,
Heteroterma, Nekewis, and Tudiclana; and, as Heteroterma
spp. have been recognized in the New Zealand Paleocene and
Patagonian Cretaceous (Finlay and Marwick, 1937), peris-
sityid distribution apparently extended to South America and
New Zealand in the latest Cretaceous and Early Tertiary.

MORPHOLOGIC CHANGE

The earliest perissityids are similar in shape and in aperture
and differ mainly in sculpture (Figures 5-9, 104). The latest
perissityids are, however, dissimilar, each lineage having
evolved a distinctly different aspect. And in most lineages
the geologically youngest form differs markedly from the
oldest. With the exception of columellar folds (commonly
ranked as a generic or familial feature) the characters that
change in a lineage are those usually considered to be of no
more than specific importance. Sequential ordering of species
in the lineages is based upon stratigraphic superposition of
specimens in perissityid-bearing sections and biostratigraph-
ic correlation of these and other deposits in which perissityids
occur. Although perissityid occurrences in these deposits are
spotty, sufficient specimens at close enough intervals have
been obtained to make patterns of change within the group
apparent. Surprisingly, between the five similar Coniacian
species and their disparate Campanian descendents, there
are no more than three species in any lineage. In each lineage
the species arise by phyletic transitions; lineage splitting, be-
cause of the trends in the development of some characteristics
within each lineage, results in recognizable genera. Forma-
tion of new morphologic species through phyletic transitions
is more than four times as common as is lineage splitting;
divergence between lineages is more prevalent than para-

Figure 4. Plot of time versus changed characteristics in the Peris-
sitys lineage. Bars indicate recognized range of each specific taxon.
Brackets show range of possible genera if the genera were defined
according to characteristics commonly considered to be of generic
importance. Because there are specimens of intergrading form be-
tween the recognized species a sloping line is drawn connecting the
mid-point of each range bar. The more horizontal the line the faster
the species is changing; the more nearly vertical the slope, the more
the species approaches stasis. Two changes are diagrammed within
cretacea during the Coniacian; six accumulate during the earlier part
of the Santonian between cretacea and elaphia ; five during the later
Santonian and earliest Campanian between elaphia and brevirostris;
four through the early and mid Campanian between brevirostris and
pacifica; two during latest Campanian and earliest Maastrichtian
between pacifica and colocara; and three from early Maastrichtian
into late Maastrichtian between colocara and stantoni. Three changes
are diagrammed between stantoni and steward although the fossil
record between these two is incomplete.

lellism, and convergence is rare. To determine whether the
style of evolution is punctuated or gradual, the rate fast or
slow, and its pace steady or varied requires a specimen-rich
lineage such as Perissitys. Plotting morphologic change against
biostratigraphic zonations combined with a time scale, pro-
vides apparent rate and pace of change in the Perissitys lin-
eage (Figure 4).

PATTERNS OF CHANGE

Divergence in the four lineages Perissitys, Pseudocymia,
Murphitys, and Christitys (compare Figures 7, 104, 9, and 8
to 93, 100, 127, 160, and 1 82) is accomplished through mod-

Contributions in Science, Number 380

Popenoe and Saul: Perissitys 5

Figures 5-9. Coniacian perissityid species. 5. Pseudocymia( ?) ait ha
new species, holotype, xl, UCLA 39440 from CIT loc. 1007. 6.
“ Fusus ” aff. kingii Gabb, hypotype, x 1, LACMIP 7245 from

CIT loc. 1007. 7. Perissitys cretacea (Cooper), neotype, x 1, UCLA
59588 from UCLA loc. 4104. 8. Christitys delta new species, ho-
lotype, xl.5, UCLA 59455 from UCLA loc. 4209. 9. Murphitys
michaeli new species, paratype, x 1.5, LACMIP 7271 from CIT loc.
1230.

est modifications to different sets of characteristics in each
lineage. The degree of differentiation achieved by incremen-
tal changes within three perissityid lineages across 1 0 million
years can be seen by comparing Figures 5-9 of Coniacian
species with Figures 10-1 5 of mid Campanian forms. Amount
of alteration within each lineage is suggested by comparing
Figure 7 to 93 and 100, 104 to 127, 9 to 160, and 8 to 182.
Although the morphologic differences between Perissitys cre-
tacea (Figure 7), P. brevirostris (Figure 10), and P. stantoni
(Figure 93) have been considered to warrant recognition of
distinct genera, these forms result from phyletic transition
within a lineage. The differences between species resulting
from a lineage split (Figures 5 and 7) are not greater than
those between adjacent species in a lineage (Figures 7 and
27).

The genus Pseudocymia changes from shortly fusiform in
shape to moderately elongate fusiform. The pseudofolds on
the columella are reduced as is the posterior sinus in the
outer lip. Unifying features include the several nearly equal,
distinct spiral ribs that override short, strong axial ribs, a
siphonal fasciole, and the thickened outer lip with several
subequal, liriform denticulations. The genus Christitys de-
velops a more pyriform shape with a broader, more expanded
periphery upon which the number of strong spiral ribs is
reduced, and weaker outer lip denticulations. Unifying fea-
tures include the posterior sinus of the growth line, a rela-
tively abapical position for the strong outer lip denticula-
tions, and the columellar fold at the base of the previous
whorl. The genus Murphitys develops a roundly expanded
inner lip, a growth line straightened across the ramp with the

Figures 1 0-1 5. Three species of mid Campanian age from the three
most complete perissityid lineages. 10, 11. Perissitys brevirostris
(Gabb), hypotype, x 1.5, UCLA 59661 from CIT loc. 1400. 12, 13.
“ Hindsia nodulosa (Whiteaves),” hypotypes, x 1; 12, UCLA 39464
from CIT loc. 1158; 13, UCLA 39466 from CIT loc. 1402. 14, 15.
“ Fusus ” cf. “F.” kingii Gabb, hypotype, x 1, LACMIP 7246 from
UCLA loc. 7003.

sinus displaced posteriorly and becoming adjacent to the
suture, and more nodulose axial sculpture combined with
finer spiral sculpture. Unifying features include the two col-
umellar folds, the rimmed outer lip, and the plumply buc-
ciniform shape. The genus Perissitys changes from buccini-
form to pyriform, develops an expanded inner lip that
eventually envelops most of the spire and last whorl in callus,
and reduces and then eliminates columellar pseudofolds and
outer lip denticulations.

The sums of these changes are impressive, especially so
because the differences between species are not large. Absence
of specimens, regarded as constituting a species, from the
geological record would produce a punctuational style to the

6 Contributions in Science, Number 380

Popenoe and Saul: Perissitys

evolution of these lineages; absence of two such adjacent
species would prevent recognition of the lineage.

A gap between the early late T uronian Pseudocymia aurora
and the five Coniacian perissityid species (Figures 5-9) leaves
unrepresented a short period of time during which, if these
five species are derived from P. aurora, there is an increase
in species diversity. The Coniacian-Santonian Perissitys rec-
ord following this gap is of relatively rapid evolution (Figure
4). Similar rates of evolution could produce the five Conia-
cian perissityid species— P. cretacea, M. michaeli, /J. (?) ait ha,
C. delta, and “F.” aff. “F.” kingii— from P. aurora. Near the
Coniacian-Santonian boundary, another lineage split and in-
crease in species diversity is suggested by the similarity of
Coniacian Christitys delta (Figures 8, 165-168) and Santoni-
an C. medica (Figures 169-170, 172-177) to the early San-
tonian member of the “ Hindsia nodulosa ” lineage (Saul, in
prep.). These possible increases in species and generic di-
versity involve no greater morphological modifications from
the putative ancestors than are present between species of
Perissitys. Additional such alterations render these lineages
recognizable as distinct genera. The possibility of prolifera-
tion of species at the margins of geographic ranges (Shuto,
1974) is suggested by the presence of perissityids in Japan.
Thus, although the style of morphologic change in West Coast
perissityids is predominantly gradual, apparent punctuation
may be developed in other geographic areas.

Evolution of “specific” characteristics in the Perissitys lin-
eage has sufficed to produce morphologic dissimilarity re-
gardable as of “generic” degree, but the style of change re-
mains gradual within and between species of Perissitys and
the characteristics evolve independently. Individual speci-
mens have some but not all features more advanced than do
other specimens from the same locality. A collection from a
single locality probably represents more than one generation
of mollusks, resulting in variety of form, but the heteroge-
neous expression of the evolving characteristics suggests het-
erochronous spread through the populations of incremental
alterations.

The changes within and between species of Perissitys are
modest. The bucciniform Perissitys cretacea becomes slightly
shorter spired, and the inner lip broadens posteriorly.

P. elaphia has one or two more strong peripheral ribs than
does P. cretacea. Typically in P. elaphia the parietal lip ex-
pands to cover the apertural face of the last whorl, there is
an additional medial denticle within the outer lip, the pos-
terior growth-line sinus is diminished, the anterior canal is
lengthened, and P. elaphia has a more pyriform shape. The
earliest P. elaphia have, however, a bucciniform shape, a
moderate expansion of the parietal lip, only a small addi-
tional medial denticle within the outer lip, and no notable
change in the anterior canal. Although there is usually a third
spiral rib about the periphery, it is weaker than the other
two. The third peripheral rib is rapidly strengthened upsec-
tion and some late early Santonian P. elaphia (Figure 38)
have four strong peripheral spirals. The parietal callus also
expands adapically so that P. elaphia from the Inoceramus
schmidti horizon of Mill Creek, the Baculites capensis Zone
on Chico Creek and later Santonian localities of the Redding

area (e.g., UCLA loc. 4217 on Clover Creek) have callus up
the apertural face of the spire (Figure 40). Additionally, these
late Santonian P. elaphia are of more pyriform shape as the
whorl becomes more strongly contracted immediately an-
terior to the abapical strong peripheral spiral. Finally, the
youngest P. elaphia have weakly developed pseudofolds and
denticulations. The species cannot, therefore, be considered
to be in a state of stasis. Morphologic variation within each
locality collection suggests that the changes are gradually
acquired and not abruptly developed.

Perissitys brevirostris lacks the apertural ornaments of P.
elaphia, has a longer, straighter canal, a shorter spire, no
more than three strong peripheral ribs, and a more expansive
parietal lip. The earliest Campanian P. brevirostris (Figure
43) differs from typical P. brevirostris and resembles late
Santonian P. elaphia in having a relatively high ramp and a
more angulate peripheral profile. Although specimens found
through 600+ m of the Ten Mile Member of the Chico
Formation on Chico Creek (area 5) are identified as P. bre-
virostris, typical P. brevirostris with the longer straighter canal
and usually rounder whorl profile are found through the up-
per 427 m of this thickness (Figures 54, 56). Despite some
variation of whorl profile, spacing of the peripheral ribs, and
height of spire, P. brevirostris is morphologically relatively
stable through this section, in contrast to the greater number
of changes in P. elaphia, which has been found through only
200+ m of the Musty Buck Member (Figure 2).

The transition from P. brevirostris to P. pacifica involves
shortening of the spire, crowding of the peripheral spirals to
form a narrower and more angulate periphery, and increased
coverage of the shell by the expanded callus. Many specimens
from the Chatsworth Formation in Bell Canyon (area 17)
have two strong peripheral spirals, typical of P. pacifica, but
the rounder whorl profile of P. brevirostris (Figure 61). In the
upper part of its stratigraphic range P. pacifica gradually ac-
quires the characteristics of P. colocara applying increased
callus over the shell and developing less prominent peripheral
spirals and nodes.

So many specimens of P. colocara have been identified as
Cophocara stantoni that the differences between P. pacifica
and P. colocara might be expected to be much greater. The
differences consist, however, mainly of the addition of more
callus and the further reduction of peripheral spirals and
nodes, trends which ultimately result in P. stantoni. The more
enveloping callus on P. colocara is typically deposited, at
intervals, thickly near the aperture. When further growth
occurs, this lump of callus distorts the growth spiral and
causes the shell to appear deformed. Some specimens of P.
colocara (Figures 84, 85) either produced less of a callus coat
or were interrupted before they completed their coating and
in their seminudity resemble P. pacifica. Although an en-
veloping callus coat gives an impression of definite specific
difference, the coat developed over a period of time during
which more individuals are more callused and specimens
with less coating are rarer in the collections.

P. colocara and P. stantoni are difficult to distinguish be-
cause both are coated with callus. The crowding of the pe-
ripheral spiral ribs reaches its ultimate in P. stantoni, and

Contributions in Science, Number 380

Popenoe and Saul: Perissitys 7

only one rib is dominant; the callus coat is thicker and applied
more evenly; the spire is consistently short, and the periphery
less angulate.

Specimens from the late Maastrichtian beds at the base of
the San Francisquito Formation on Warm Springs Mountain
(area 19) (Figure 95) and from the Lower Laguna Seca For-
mation, UCB loc. A-3262 (area 8) (Figure 96) have charac-
teristics intermediate between P. stantoni and P. stewarti,
but are, unfortunately, poorly preserved. UCB loc. A-3262
has not yielded turritellas, and Perissitys has not yet been
recovered in association with Turritel/a peninsularis adelai-
dana Merriam. The oldest turritellas with which P. stewarti
is associated are late Turrit ella quay lei and T. peninsularis
and are probably not older than late Danian (Saul, 1983),
but specimens from UCB loc. A-3262 and some other lo-
calities in the Panoche Hills (area 8) may be from older
horizons. P. stewarti differs most evidently from P. stantoni
in having two well-developed peripheral spiral rows of nodes,
giving it a blunt, biangulate periphery. It also has a shorter
spire and a thinner callus coating.

Perissitys is well represented in the fossil record and changes
within as well as between its species are documented by
specimens. Of the four genera, Perissitys supplies most evi-
dence for style and rate of change. Four of the Perissitys
species have fossil records that are stratigraphically complete
enough to give an indication of the pace of change.

RATE OF MORPHOLOGIC CHANGE

The range of Perissitys is Coniacian to early Selandian (Pa-
leocene). The Coniacian age is based on ammonite zonation
and the Selandian on turritellid zonation. Several time scales
have been proposed recently for this interval. Stage lengths
vary on the various time scales, and estimates of chronologic
duration of Perissitys species and the rates of change within
the lineage are dependent on the time scale chosen. The
DNAG 1983 Time Scale (Palmer, 1983) for this interval is
in close agreement with a 1984 version of Berggren et al.
(1985b), Kent and Gradstein (1985), and Berggren et al.
(1985b) except for the placement of the Danian-Selandian
boundary. If this boundary is put at 62.3 Ma (Berggren et
al., 1985a; Berggren et al., 1985b) rather than at 63.6 Ma
(Palmer, 1983), Perissitys stewarti is probably mainly Danian
in occurrence. The range of P. stewarti (Figure 3) is plotted
with respect to its association with Turrite/la peninsularis
and is independent of the position of the Danian-Selandian
boundary. At the earlier end of the Perissitys lineage, P. cre-
tacea ranges through the Pacific West Coast Coniacian Stage
defined by ammonite correlations. The early Campanian ma-
rine magnetic anomaly 33-34, Chron 33r, encompasses ap-
proximately 260 m of the Ten Mile Member in the Chico
Creek section (Ward et al., 1983) (see Figure 2). Early P.
brevirostris is found through this part of the Chico Creek
section and Chron 33r is coincident with the earlier part of
the range of P. brevirostris. Chron 33r is considered to have
begun shortly after the Santonian-Campanian boundary and
to last for 3.5-5 Ma. It is dated at 78.53 to 82.93 Ma by
Harland et al. (1982, p. 74) but is considered to be nearly 2
Ma older on the DNAG 1983 Time Scale (Palmer, 1983).

Based on these correlations and plotted against the DNAG
1983 Geologic Time Scale, Perissitys has a duration of rough-
ly 26 Ma (Figure 3). Seven successive specific taxa are named
in this lineage. Were a constant rate of change assumed each
species would have a time span of 3.7 Ma, or slightly more
than one-third the 10 Ma considered to be the mean species
duration for gastropods (Stanley, 1985, p. 16 ). Perissitys cre-
tacea (Cooper), however, has been recognized only from the
Coniacian which may be no longer than 1 Ma (Obradovich
and Cobban, 1975, p. 46). But, as P. cretacea may not be
descended from the only pre-Senonian perissityid herein rec-
ognized, Pseudocvmia aurora new species of early late Tu-
ranian age, the earliest possible appearance of P. cretacea
remains undefined and its range may have extended into the
Turanian and exceeded one million years. Perissitys elaphia
new species occurs through the Santonian, which is estimated
to represent about 3.5 Ma (Palmer, 1983). Specimens iden-
tified as P. brevirostris (Gabb) are of early to mid Campanian
age, and the taxon ranges through approximately 6 Ma. Peris-
sitys pacifica new species is of late Campanian age and thus
has a duration of roughly 4 Ma. P. pacifica changes less
dramatically and more slowly than does P. elaphia but more
rapidly than P. brevirostris. Early Maastrichtian P. colocara
new species also has a probable duration of 4 Ma. It is suc-
ceeded near the middle of the Maastrichtian by P. stantoni
(Stewart), which has a duration of at least 3.5 Ma. Unfor-
tunately specimens of Perissitys have not yet been found in
deposits certainly recognized as earliest Paleocene and the
biochron of P. stantoni (Stewart) is probably not completely
known, nor is that of the Paleocene species P. stewarti (Zins-
meister).

The number of altered characteristics in the genus Peris-
sitys are plotted against geologic time in Figure 4. Each of
the above-mentioned differences between species is given
equal weight. The slope of the line is an indication of the
rate of change in Perissitys. Change is most rapid through
the Coniacian-Santonian and slowest through the early to
mid Campanian. Although Pseudocymia aurora may or may
not be ancestral to Perissitys cretacea, there are essentially
four morphologic changes between the two. P. cretacea has
fewer denticles on its outer lip, a slightly longer canal, a
deeper posterior sinus, and a periphery marked by two strong
spirals. Had Pseudocymia aurora been included on Figure 4,
the slope of line between it and Perissitys cretacea would
have been similar to that between P. cretacea and P. elaphia.
Forms intermediate between P. aurora and P. cretacea are,
however, unknown.

The Perissitys lineage exhibits its most rapid accrual of
morphologic change through the Coniacian and Santonian,
but the other lineages, although not as well documented, do
not evince equal degrees of evolution. The least change rec-
ognized for this same Coniacian-Santonian interval is in the
Murphitys lineage whose species M. michaeli ranges through
the Coniacian and Santonian stages.

In the Perissitys lineage the varied pace of gradual change
produces morphologic differences of generic degree, but the
recognition of different genera has not occurred at times of
relatively rapid evolution and between species of short tem-
poral duration. Previous workers have identified P. pacifica

8 Contributions in Science, Number 380

Popenoe and Saul: Perissitys

as P. brevirostris and P. colocara as Cophocara stantoni, thus
recognizing a generic distinction between these two new
species, each of which has a duration of approximately 4 Ma.
Occasionally late P. pacifica has been identified as Cophocara
stantoni and early P. colocara as P. brevirostris reflecting
independent and unequal development of characteristics.
Perceived generic differences are not, therefore, dependent
on rapid evolution nor great morphologic change.

If the average duration of a gastropod species is 10 Ma
(Stanley, 1985), the perissityids are evolving more rapidly
than is usual for their class. Their morphologic changes do
not appear to reflect major habitat changes, although the
different lineages probably had somewhat different require-
ments. Judging from their occurrence in the geologic record,
Pseudocymia, Murphitys, and Christitys were less widely
adapted than Perissitys. The continuing direction of changes,
such as the increasing envelopment of the spire by the inner
lip callus and the lengthening of the anterior siphonal canal
in Perissitys, produce the distinct morphologies of this group.
That so much directional change occurred during the Late
Cretaceous, a time of relative stability, suggests biological
rather than physical pressure. Greater arthropod predation
and increasingly infaunal habitat of bivalves (Vermeij, 1977)
that may have been prey of the perissityids would both have
promoted infaunal adaptations in these siphonate gastro-
pods.

CLASSIFICATION

Gastropods that we propose to include in the family Peris-
sityidae form a morphologically compact group in the Conia-
cian, and five species of that age could readily be assigned
to a single genus (Figures 5-9). Lineages based on these five
species, traced through the Late Cretaceous, evolve new forms
by modifying species-defining characteristics to the extent
that by the mid Campanian the three most complete lineages
are represented by species— Perissitys brevirostris (Gabb),
“ Fusus ” cf. “F.” kingii Gabb, and “ Hindsia nodulosa ”
(Whiteaves) (Figures 10-15) — which have not been consid-
ered congeneric, nor assignable to the same family. Clearly
the approach of the systematist will affect the ordinal, fa-
milial, generic and even specific placement of these fossils.
The classification presented here has its basis in stratigraph-
ically ordered specimens collected from Late Cretaceous and
early Paleocene sequences of the Pacific Coast of North
America. The successions of fossils assigned to the perissi-
tyids document a part of the Late Cretaceous siphonate gas-
tropod radiation (Sohl, 1964; Taylor et al., 1980).

SPECIES

A paleontological species is a morphologic species, and if
there are gaps in the record of a lineage, the preserved seg-
ments of the lineage may readily be recognized as species. If
there are no gaps, or the gaps are short relative to the rate
of evolution of the lineage, a grading continuum of mor-
phologies may result. This is the case with Perissitys. Because
of the range of variation present at any stratigraphic level,
and the changes between levels, Perissitys specimens iden-
tified by comparison to those from super- and subjacent strata

are alike enough to be identified as being of the same species.
If artificial gaps are created, however, by comparing speci-
mens of early Campanian age to those of Comacian or mid
Maastrichtian age, three easily distinguishable species, suf-
ficiently different to have been previously allotted to three
genera (Figure 4), are apparent. The West Coast lineage of
Perissitys is an evolutionary continuum, and its species are
arbitrarily named segments, but arbitrary only in the sense
that there are no abrupt morphologic changes in this contin-
uum despite the distinctness of the end members. Each lin-
eage has several characteristic features that vary and evolve
independently, producing a clinal rather than punctuational
series of phenotypes. Species recognition is dependent upon
an aggregate of changing features, and each remains recog-
nizable for various lengths of time (Figures 3, 4).

Should these forms be considered to be species if speciation
events are absent? This paper deals only with specimens from
the West Coast of North America, but there are perissityids
in Japan and possible perissityids in New Zealand and Chile.
Although the present West Coast record is one of phyletic
transitions, the lineages may have split near the geographic
range margins of the species. The North American perissi-
tyids pass through sufficient morphologic changes that seg-
ments of these lineages must be considered to be species.
Additionally, the named species have biostratigraphic integ-
rity and are useful for correlating Late Cretaceous and Early
Tertiary sections.

GENUS

If the genus Perissitys were narrowly restricted to those species
having the apertural characteristics of the type species, P.
brevirostris (Gabb), only it and P. pacifica new species could
be included therein. Of the remaining species, P. cretacea
(Cooper) could be placed in a new genus along with the four
contemporary species of Figures 5-9; P. elaphia might con-
stitute a new monotypic genus; and P. colocara new species,
P. stantoni (Stewart), and P. steward (Zinsmeister) would
make up Cophocara (Figure 4). Such genera are horizontal
in that their distribution tends to lie along restricted time
planes and resemble many Recent genera in which aggre-
gations of similar but distinct species are recognized, but their
phylogenetic relationships through time are not. The peris-
sityid genera recognized are linear (or vertical) genera. Re-
taining a single generic name for each lineage organizes the
species; recognizing successive chronospecies provides gno-
mons along the genetic line.

FAMILY

Perissityidae are, or are derived from, bucciniform gastro-
pods that have two folds or pseudofolds on the columella, a
parietal welt or denticulations near the posterior end of the
aperture, a posterior sinus to the growth line, and strong
median denticulations within the outer lip. The outer lip
flares anteriorly and is thickened by a varix. Impressions of
former outer-lip thickenings and denticulations are found on
steinkerns, even though such varices may not be obvious on
the shell exterior. Compared to turritellid lineages of the same
period and time span, the perissityid lineages evolve rapidly

Contributions in Science, Number 380

Popenoe and Saul: Perissitys 9

and divergently, each lineage following its own course and
arriving at forms suggestive of disparate families. The ap-
ertures of early perissityids resemble those of columbellinids,
and several late Jurassic columbellinid trends— constricted
and then flared apertures, denticulate outer and inner aper-
tural lips, inner lip callus expanded and covering one-half to
the entire exterior of the shell (Taylor et al., 1980)— resemble
those of perissityids. These suggest a columbellinid deri-
vation, possibly within the Early Cretaceous.

Perissit\>s stewarti (Zinsmeister, 1983) from the Paleocene
is pyriform in shape with a moderately long, nearly straight
anterior canal, and lacks apertural pseudofolds and dentic-
ulations; it resembles Tudiclidae. Pseudocymia( ?) kilmeri new
species of early Maastrichtian age is subfusiform in shape
with a short, flexed anterior canal, strong denticulations on
the outer lip, and weak pseudofolds on the columella, and
resembles tropical Buccinidae. Christitys martini new species
of early Campanian age is pyriform in shape, has weak den-
ticulations on the outer lip and a fold on the columella. Its
form, except for that of its outer lip, is similar to some Vas-
ldae. Murphitys madonna new species, of early Maastrichtian
age, is angulate bucciniform in shape and has a thickened
outer lip and two folds on the columella. Except for these
columellar folds it resembles some Cymatiidae. Zinsmeister
( 1 983, p. 1297) includes those forms with short to moderately
elevated spire, i.e., Perissitys, Cophocara, and Heteroterma,
in Tudiclidae, and moves the higher spired Nekewis from the
Turridae to the Tudiclidae because of its similarities to Het-
eroterma. None of these genera have the tudiclid fold at the
base of the columella; neither Heteroterma nor Nekewis has
the characteristic expanded inner lip of Tudicla spirillus (Lin-
naeus), and all have an eye-shaped rather than a well-rounded
aperture. The growth line of Heteroterma and Nekewis is
sinused posteriorly and resembles that of Christitys (Figures
166 and 176) and “ Hindsia nodulosa (Whiteaves)” (Figure
1 3), and these genera are closer to Perissityidae than to T udic-
lidae. Tudiclana simulator Findlay & Marwick, 1937, which
resembles Perissitys brevirostris but lacks the subsutural welt,
is described has having a low fold at the base of its columella,
and it may belong with Tudicla as indicated by Finlay and
Marwick (1937). Abbott (1959, p. 20-471) suggests that Tu-
dicla is usually included in the Vasidae because its overall
shape resembles that of Tudicula spp., although there is no
anatomical evidence to support this placement. Thus Finlay
and Marwick (1937) and Zinsmeister (1983) may well be
correct in divorcing the Tudiclidae from the Vasidae. Al-
though we do not consider Perissitys to be a tudiclid, the
Tudiclidae and Perissityidae may be more closely related
than either is to the Vasidae.

SUPERFAMILY

At the family level the perissityids are grouped together be-
cause of their early characteristics in common, several of
these characteristics also serving to distinguish them from
other families, but at the next hierarchical level we place
them in the superfamily into which these diverging lineages
appear to evolve. Placing them in Muricacea of Ponder (1973)

would be simplest as he includes within the Muricacea the
Buccinacea and Volutacea of Wenz (1941). This large poly-
morphic superfamily grouping tends, however, to obscure a
number of relationships. The families Muricidae, Thaididae,
Magilidae, and Columbariidae, which constitute the Muri-
cacea of Wenz (1941) and Taylor and Sohl (1962), form a
distinct group within Ponder’s Muricacea. With Muricacea
constituted as of Taylor and Sohl ( 1 962), and the Buccinacea
and Volutacea of Wenz (1941) and Taylor and Sohl (1962)
regarded as distinct, choice of superfamily placement for the
Perissityidae is increased. The Perissityidae have little in
common with Late Cretaceous muricaceans and more sim-
ilarity to buccinaceans and volutaceans. Buccinaceans and
volutaceans appear in the late Early Cretaceous and both are
relatively common in the Late Cretaceous (Taylor et al.,
1980). Wenz (1943) and Zinsmeister (1983) place Perissitys
and Cophocara in the Volutacea, apparently because the pyr-
iform shape resembles that of Tudicla, which was included
in the Volutacea by Thiele (1929), Wenz (1943), and others,
and presumably by Taylor and Sohl (1962) who, following
Wenz, placed the Vasidae — in which Tudicla is often in-
cluded—in the Volutacea. Clearly the earliest perissityids
have neither a volutid shape nor columella folds similar to
those of volutes, whereas volutes from the same beds already
display characteristic volute morphology. Perissitys evolves
toward a shape similar to that of some genera included by
Wenz in the Vasidae [e.g., Pyropsis (Conrad, 1860; Sohl,
1964), Tudicla ( Roding, 1798; Abbott, 1959), and Tudiclana
(Finlay and Marwick, 1937)], but Sohl (1964) later moved
the Vasinae, including Tudicla, to the Buccinacea, and it is
within the Buccinacea that the Perissityidae are placed.

As already indicated perissityids resemble the Cymatiidae
and the Columbellinidae. Cossmann (1904), Wenz (1940),
and Sohl (1960) assigned the Columbellinidae to the Strom-
bacea, but, because columbellinids lack the apertural sinuses
characteristic of strombids and aporrhaiids, we concur with
Fischer (1884) and Taylor et al. (1980) and include Col-
umbellinidae in the Tonnacea. It is the earliest of the ton-
nacean families to appear; Taylor et al. ( 1 980, p. 385) derive
it from the Strombacea within the Jurassic. The tonnacean
families Cymatiidae, Bursidae, and Cassididae are reported
from the mid Cretaceous (Taylor et al., 1980, p. 387), but
described forms are not compellingly similar to perissityids,
and the similarity of some perissityids to cymatiids results
from ancestry within the same family. The characteristics of
perissityids are in many respects intermediate between those
of columbellinids and buccinaceans.

Ponder (1973) argues that the neogastropods are derived
from the Subulitacea rather than from the advanced ton-
nacean mesogastropods considered ancestral by Wenz (1938,
p. 65; 1941, p. 1082) and others. Subulitaceans have, in the
Paleozoic, already lost the median sinus of the outer lip (Pon-
der, 1973, p. 302), whereas early perissityids and volutes
have a shallow notch which may be a posteriorly displaced
remnant of the median sinus. Taylor et al. (1980, p. 385-
386) suggest that instead of Subulitidae, the Purpurinidae are
antecedent to neogastropods. The perrissityids, however, ap-
pear to be derived from a columbellinid ancestor, probably

10 Contributions in Science, Number 380

Popenoe and Saul: Perissitys

in advance of the cymatiids, and perhaps in consort with
Tudiclidae and Colubrariidae. These latter two families have
been variously classified but are probable Buccinacea (Sold,
1964; Ponder, 1973). The perrissityids, thus, suggest deri-
vation of buccinaceans from early tonnaceans.

ORDER

Cox (1960) considered the line between mesogastropods and
neogastropods to be arbitrary and included both orders in
Caenogastropoda, but Ponder (1973) separates mesogastro-
pods and neogastropods. Much of the evidence for separation
lies in the soft parts and includes differences in chromosome
numbers (Patterson, 1969; Ponder, 1973, p. 296), whereas
similarities of neogastropods and some mesogastropods (ton-
naceans) are evident in the shell. Both the soft and hard part
comparisons involve interpretations as to derivation of struc-
tures, possible parallel evolution, etc. Unfortunately, for most
soft part evolution there is no geologic record, and parallel
and diverging evolution of shell form make the geologic rec-
ord of the hard parts difficult to interpret. Neither Ponder’s
anatomical nor conchological criteria can be used for sepa-
ration of fossil mesogastropod cymatiids and bursids from
neogastropod buccinaceans. The fossil record suggests that
perrissityids evolved during the Late Cretaceous from forms
that resembled some cymatiid mesogastropods into forms
that resemble some buccinacean and turrid neogastropods.
The line between meso- and neogastropods is, of course,
arbitrary, as it is drawn across evolving lineages irrespective
of whether neogastropods derive from “primitive” or “ad-
vanced” mesogastropods. In shell form the perissityids ap-
pear transitional between Mesogastropoda and Neogastrop-
oda, but in a hierarchical classification there is no place for
transitional forms, and we include them in the Neogastropo-
da. The evolution of the Perissityidae suggests that most
(perhaps all) neogastropods are derived from within the Col-
umbellinidae, and that the superfamilies and families of the
Neogastropoda arose both from different columbellinids and
sequentially from columbellinid stocks.

SYSTEMATIC PALEONTOLOGY
Phylum Mollusca Linnaeus, 1758
Class Gastropoda Cuvier, 1797
Order Neogastropoda Wenz, 1938
Superfamily Buccinaceae Rafinesque, 1815
Family Perissityidae new family

DIAGNOSIS. Bucciniform gastropods with two folds or
pseudofolds on the columella, a parietal welt or denticula-
tions near the posterior end of the aperture, a posterior sinus
to the growth line, and strong median denticulations within
the outer lip. The outer lip flares anteriorly and is thickened
by a varix. Impressions of former outer-lip thickenings and
denticulations are found on steinkerns, even though such
varices may not be obvious on the shell exterior.

The new family Perissityidae is proposed for several genera
of gastropods that range from Late Cretaceous through Early
Tertiary. The early Senonian species of these genera are of
moderate size and bucciniform shape, with fine to coarse
spiral ribs and short but strong axial ribs about the whorl
periphery. All forms thus far studied have fairly large pau-
cispiral protoconchs. The growth line has a shallow antispiral
sinus adapical to the mid whorl. The outer lip is slightly
thickened, rimmed, and flared, especially from mid whorl to
the anterior siphonal constriction. The anterior sinus is near-
ly two-thirds as long as the eye-shaped aperture. The aper-
tural armature, which is very characteristic in early Senonian
forms, is ontogenetically intermittently developed and best
displayed in the adult stage. Impressions of the typical outer
lip denticles are found on natural casts, spaced as though
indicating varices. Most characteristic is the strong denticle
at mid whorl with commonly a lesser one anterior to it. This
set of denticles is just anterior to the anterior end of the
antispiral sinus and at the posterior end of the outer lip flare.
The denticles oppose a pseudofold or fold on the columella.
A moderately strong tubercle (or set of tubercles) in con-
junction with similar tubercular structures on the inner lip
constricts the posterior end of the aperture. Additionally there
are usually small tubercles on the outer lip adjacent to the
anterior siphonal canal. The inner lip is clearly demarked,
fairly thick, and in some species forms a pseudoumbilicus at
the anterior end of the anterior siphon with the siphonal
fasciole. Medially on the columella there are two or three
subequal pseudofolds or one or two folds.

Folds and pseudofolds do not differ in apertural view; both
appear to be spirally elongate plicae on the columella. Folds
spiral uninterrupted on the columella of the teleoconch (Fig-
ures 149, 152, 163), but pseudofolds are short, extending less
than a quarter turn (Figure 20) into the shell anterior and
are absent within earlier whorls (Figures 31, 117). Pseudo-
folds, like varices, are developed at growth halts, and shells
that have not developed a thickened and denticulate outer
lip do not show pseudofolds (Figures 17, 35), but folds are
present even though the outer lip has not been thickened
(Figure 170). The strongest folds occur, however, within ap-
ertures that have a well-developed varix, and folds are ap-
parently enhanced by additional callus at growth halts.

Turonian and early Senonian perissityids resemble Col-
umbellinidae in overall shape, in having a well-armed ap-
erture, and in the position of the posterior sinus on the outer
lip just adapical to the whorl periphery. The posterior sinus
of perissityids is not as narrow and elongate as that of the
columbellinids; later Senonian Perissitys spp. and Christitys
spp. are of more pyriform shape than are columbellinids; and
the anterior segment of the outer lip of perissityids flares.

Although many perissityids have varices, none has the
regular varices of the Cymatacea. The range of shape in peris-
sityids is similar to that in cymatiids, but cymatiids lack folds
on the columella and do not cover the shell with callus.

Perissityidae, especially Turonian and early Senonian Pe-
rissityidae, resemble some members of the large and poly-
morphic family Buccinidae in shell shape, sculpture, and
shape of aperture. Buccinids do not cover the shell with

Contributions in Science, Number 380

Popenoe and Saul: Perissitys 1 1

callus, lack columellar folds, and do not have strong medial
outer lip denticulations and an anteriorly flared outer lip.

Like Fasciolariidae, some perissityids have columellar folds,
and some later Senonian perissityids develop a fusiform shape
and/or a long anterior siphonal canal. Unlike fasciolariids
some perissityids become strongly pyriform in shape and
deposit callus over the shell.

Perissityids have been included in the Vasidae which some
of them resemble in being pyriform and having folds on the
columella. Perissityids do not have spinose sculpture, have
an anteriorly expanded outer lip, and a typical pattern of
outer lip denticulations which includes a strong medial den-
ticulation.

Tudiclidae differ from pyriform Perissityidae in having a
round aperture and a columellar fold at the posterior end of
the anterior canal.

The posterior growth-line sinus of some perissityids re-
sembles that of some Turridae, but turrids lack apertural
armaments, external callus deposits, and columellar folds.

GENERA INCLUDED. Perissitys Stewart, 1927 (includ-
ing Cophocara Stewart, 1927), Pseudocymia new genus,
Murphitys new genus, Christ itys new genus, and Pseudoperis-
sitys Nagao & Otatume, 1938. Probably Heteroterma Gabb,

1 869, will prove to be a perissityid; possibly Nekewis Stewart,

1 927, belongs here.

DISTRIBUTION. North Pacific (Japan, Sakhalin, Alaska
Peninsula, British Columbia, Washington, Oregon, Califor-
nia, Baja California) and eastern South Pacific (Chile, New
Zealand).

GEOLOGIC AGE. Late Cretaceous (Turonian) to Early
Tertiary (mid Selandian and possibly into Oligocene).

Genus Perissitys Stewart, 1927

TYPE SPECIES. Perissitys brevirostris (Gabb, 1864), by
original designation (Stewart, 1927, p. 426).

DIAGNOSIS. Bucciniform to pyriform gastropods of me-
dium to moderately large size with a subangulate whorl pe-
riphery which is ornamented by fewer than five spiral ribs.
The crossing of these spiral ribs by short axial ribs produces
the characteristic noded periphery. The edge of the outer lip
anterior to the periphery is fimbriated. With the exception
of the geologically earliest species, P. cretacea (Cooper), all
species have an expanded inner lip that is developed at growth
halts and is the inner lip equivalent of a varix.

REMARKS. Stewart (1927) did not assign this genus to a
family, and he mentioned only one other species, Pyropsis
hombroniana (d’Orbigny) from the Quiriquina Formation of
Chile (Wilckens, 1904, p. 213, pi. 18, figs. 8, 9), as possibly
belonging to Perissitys. Plaster casts of specimens of this
species sent to us by Dr. E. Perez d’A. of Chile have a rounded
aperture indicating that P. hombroniana is not a Perissitys ,
nor can it be included in the Perissityidae. Pyropsis sp. indet.
of Nagao (1939, p. 228, pi. 21, fig. 5-5b) from “Upper Am-
monites bed” = Upper Yezo Group, Abeshinai-gawa, Teshio
Pref., Japan, is probably a Perissitys.

Morphologic changes which develop in West Coast Peris-
sitys of Coniacian through Danian age include shortening of

the spire and lengthening of the anterior canal so that the
shell shape changes from bucciniform to pyriform, expansion
of the inner lip over the apertural face and envelopment of
the spire and most of the body whorl in a callus coat, and
reduction and disappearance of pseudofolds from the colu-
mella and tubercles from the outer lip. The change from
bucciniform to pyriform shape includes displacement of the
noded periphery from mid whorl toward the base of the
whorl. The rows of nodes on the periphery become reduced
in prominence, although the number of rows of nodes ini-
tially increases and then decreases.

Coniacian Perissitys, P. cretacea (Cooper, 1896) resembles
other Coniacian perissityids, but typical Perissitys, P. brevi-
rostris (Gabb, 1864), differs from all other perissityid genera
in having the inner lip expanded over the apertural face and
spire of the shell. In pyriform shape Perissitys resembles
Tudicula H. & A. Adams, 1863, Tudiclana Finlay & Mar-
wick, 1937, Tudicla Roding, 1798, and Pyropsis Conrad,
1860. But none of these has the inner lip expanded onto the
spire as in Perissitys. Except for Tudiclana all have an ap-
erture posteriorly broadened resulting in a rounder aperture,
and an inner lip which is folded onto the whorl. Tudiclana
has a low fold at the base of the columella, and lacks the
callus coat and the raised subsutural welt of Perissitys.

RANGE. Along the west coast of North America Perissitys
is found in sandstones of Coniacian (Late Cretaceous) to early
Selandian (middle Paleocene) age. The species are discussed
in chronologic order from earliest to latest.

Perissitys cretacea (Cooper, 1896)

Figures 7, 16-26

Sistrum ( Ricinulal ) cretaceum Cooper, 1896, p. 330, pi. 47,

figs. 1, 2; Coan, 1981, p. 161.

DIAGNOSIS. Bucciniform Perissitys with two strong spi-
ral ribs about the periphery, the inner lip not markedly ex-
panded, two pseudofolds on the columella, and denticles on
the outer lip.

DESCRIPTION. Shell of medium size, bucciniform and
thick walled; spire about one-third of total shell height, con-
sisting of five or six whorls including a smooth, mammillate
protoconch of nearly three whorls succeeded abruptly by a
whorl strongly sculptured by raised spiral threads and axial
ribs; next and succeeding whorls gently concave below the
appressed suture, with strong double angulation about the
mid whorl accentuated by two strong spirals crossing short
strong axial ribs; last whorl concave below the median an-
gulation, narrowing to form a sturdy anterior siphonal canal
of moderate length, which is twisted to the left (apertural
view) and backward at its tip; suture at the abapical periph-
eral angulation.

Sculpture above the peripheral angulation of eight to ten
subequal spiral threads which on the last whorl are narrower
than the interspaces; peripheral angulation accentuated by
two strong spiral ribs which form nodes across 12-14 short
but strong axial ribs; spiral ribs abapical to periphery slightly
stronger than on ramp, strongest on whorl base, becoming
finer toward anterior siphon.

12 Contributions in Science, Number 380

Popenoe and Saul: Perissitys

Outer lip with shallow antispiral sinus adapical to the pe-
riphery and slight spiral antisinus abapical to the periphery;
flaring moderately, flare thickened with callus deposit which
continues around the posterior end of the aperture and rounds
into the sharply demarked inner lip; inner lip of nearly equal
width from posterior to anterior end of aperture, forming a
pseudoumbilicus along the anterior siphonal canal at the
siphonal fasciole. Outer lip with one strong, medial denticle
interior to the periphery, a moderate denticle or denticle pair
adapical to the strong denticle, and a group of small denticles
next to the anterior canal. Inner lip and columella with two
nearly equal pseudofolds, the more adapical one opposite the
strong denticle of the outer lip and a moderately strong den-
ticle just abapical to the posterior end of the aperture.

NEOTYPE. UCLA 59588, here designated. Cooper (1896,
p. 330) stated that he had four specimens, but Coan (1981,
p. 161) was unable to find any of them.

HYPOTYPES. UCLA 59589-59594 from UCLA loc.
4104; UCLA 59601 from UCLA loc. 5990.

DIMENSIONS. Of neotype: UCLA 59588-height 30.5
mm (incomplete), diameter 19.9 mm, height of spire 11.2
mm; of hypotypes: UCLA 59590— height 19.4 mm (incom-
plete), diameter 13.8 mm, height of spire 6.7 mm; UCLA
59592 — height 27.8 mm, diameter 17 mm, height of spire
9.9 mm; UCLA 5960 1 —height 2 1 .4 mm, diameter 14.1 mm
(incomplete), height of spire 8.7 mm.

TYPE LOCALITY. Cooper (1896, p. 330) indicated only
Morley, Shasta County, California, and the precise locality
for his specimens is indeterminable. Morley School is shown
in sec. 25, T33N, R2W, Millville Quadrangle (U.S.G.S., 1 954),
on the Oak Run Road more than 4 mi. northeast of UCLA
loc. 4 1 04 in sec. 1 6, T32N, R2W, Millville Quadrangle, Shas-
ta Co., California. No outcrops yielding gastropods similar
to P. cretacea have been found in the immediate vicinity of
“Morley School,” and Cooper’s specimens may have come
from the vicinity of UCLA loc. 4104.

DISTRIBUTION. Sandy beds of Member IV (Popenoe,
1943), CIT Iocs. 1007, 1289, and 1596, and UCLA Iocs.
4104 (= CIT loc. 1034) and 5990, Redding area, Shasta Co.,
California.

GEOLOGIC AGE. Coniacian.

REMARKS. None of Cooper’s (1896, p. 330) four speci-
mens is available (Coan, 1981), and it seems unlikely that
Cooper’s description and figures referred to any other species,
but a neotype is designated to obviate confusion with other
species of Perissityidae. His specimens probably came from
outcrops of Member IV along Oak Run. More than 40 spec-
imens are at hand, the largest of these, UCLA 59595 from
UCLA loc. 4 1 04, has a diameter of 2 1 .2 mm. All specimens
upon which the above description is based are from Redding
area outcrops in Swede Basin, Oak Run, and Clover Creek,
Millville Quadrangle, and along Old Cow Creek, Pine Timber
Gulch, and Bear Creek, Whitmore Quadrangle (area 1 1 ). The
species has not yet been recognized elsewhere; it is the earliest
known of the inferred Perissitys lineage. It may possibly be
derived from a Turanian form of the Redding area, Pseu-
docymia aurora new genus and new species, from the Frazier
Siltstone below Collignoniceras sp. (Jones, et al., 1978, p.
XXII. 8).

Progressive changes within P. cretacea include shortening
and broadening of the spire and adapical movement of the
suture from anterior to the abapical strong peripheral rib to
just covering this rib. The inner lip becomes slightly broader,
especially on the body whorl, its outer edge becomes straight-
er, loses the angular bend, and rounds convexly to join the
outer lip. The adapical edge of the inner lip moves from just
touching the abapical strong peripheral rib to touching or
covering the adapical strong peripheral rib. These are small
changes, but the time period is apparently short.

Perissitys cretacea differs from Pseudocymia aurora in hav-
ing its suture upon, rather than abapical to, the peripheral
angulation, finer spiral ribs, a more contracted and concave
base, a longer more twisted siphonal canal, and fewer den-
ticulations on the outer lip. P. cretacea differs from all later
species of the genus in having an inner lip callus which is of
nearly equal width from posterior to anterior end of the
aperture. The shape of P. cretacea is similar to that of Can-
tharus occidentalis (Gabb, 1864) of Cenomanian age (Mur-
phy and Rodda, 1960, p. 845), but C. occidentalis lacks the
pseudofolds and strong denticulations on the outer lip of P.
cretacea. C. occidentalis resembles P. cretacea in having a
flaring outer lip rather than the unflared lip of modern Can-
tharus.

Perissitys elaphia new species

Figures 27-41

DIAGNOSIS. Pyriform Perissitys with three to four strong
spiral ribs about the periphery, an inner lip expanded to cover
the apertural face of the last whorl, two columellar pseudo-
folds, and denticles along the outer lip.

DESCRIPTION. Shell of medium size, pyriform; spire
about one-fourth of the height of the shell, comprised of 1.5
to 2 smooth, globose nuclear whorls, succeeded by two gently
concave whorls, about twice as wide as high; last whorl three-
fourths or more of the shell height, having a narrow tumid
band just below the suture, a concave ramp and rounded
periphery accentuated by three or four strong spiral ribs, and
contracting abruptly abapically to form anterior siphonal ca-
nal; anterior siphonal canal about as long as the eye-shaped
aperture, curving gently to the left (apertural view); suture at
or just below the adapical strong spiral rib.

Sculpture on subsutural welt of fine spiral threads, on ramp
of very fine distant spiral threads; strong ribs of periphery
made nodular by 11-14 strong short axial ribs, interspaces
of strong ribs having fine threads; base of whorl with three
to five narrow riblets grading into threads and very fine threads
on the anterior siphonal neck.

Outer lip with slight antispiral sinus adapical to the pe-
riphery and slight spiral antisinus at the periphery; flaring
slightly especially at, and abapical to, the periphery, lip flare
thickened with callus deposit that also fills the antispiral sinus
and continues up onto the spire at the posterior end of the
aperture, inner lip expanded posteriorly, overlapping the su-
ture, then spreading roundly to cover the apertural face of
the last whorl, curving at the most abapical strong rib back
toward the anterior end of the siphonal canal and crossing
at the siphonal fasciole to form a pseudoumbilical chink;

Contributions in Science, Number 380

Popenoe and Saul; Perissitys 13

14 Contributions in Science, Number 380

Popenoe and Saul: Perissitys

edge of expanded inner lip well demarked. Outer lip with a
strong median denticle interior to the adapical strong pe-
ripheral rib flanked by a moderately strong denticle on either
side, and next to the anterior canal a group of small denticles.
Inner lip and columella with two nearly equal pseudofolds,
the more adapical one opposite the strong denticles of the
outer lip and a moderately strong denticle just abapical to
the posterior end of the aperture.

HOLOTYPE. UCLA 59606 from CIT loc. 1 232.

PARATYPES. UCLA 59607-59608 from CIT loc. 1232,
59609-59610 from UCLA loc. 3298, 59612-59613 from CIT
loc. 1227, 59614-59615 from CIT 1246; 59618 from UCLA
loc. 4107; 59620 from UCLA loc. 3623; 59621-59622 from
CIT loc. 1016, 59623-59624 from UCLA loc. 3624; 59633-
59634 from CIT loc. 1017.

HYPOTYPES. UCLA 59638 from UCLA loc. 3633; 59639
from UCLA loc. 42 1 7; CAS 3 1 325.0 1 from CAS loc. 31325.

DIMENSIONS. Of holotype — height 31.4 mm, diameter
1 9.7 mm, height of spire 8.4 mm; of paratypes— UCLA 59609,
height 31.2 mm (incomplete), diameter 20.7 mm, height of
spire 7.8 mm; 59633, height 26.5 mm (incomplete), diameter
17.6 mm, height of spire 7.9 mm.

TYPE LOCALITY. CIT loc. I 232: Price Hollow, 2250 ft.
N, 2000 ft. E of SW cor. sec. 2, T32N, R2W, Millville Quad-
rangle, Shasta Co., California.

DISTRIBUTION. Member V and lower Member VI in
Oak Run, Price Hollow, and Clover Creek of the Redding
area (area 1 1 ); Chico Formation on Antelope Creek (area 2);
Kingsley Cave Member of Chico Formation on Mill Creek
(area 2); Musty Buck Member on Chico Creek (area 5) from
approximately 360 m to 600 m above the base of the Chico
Formation; lower Forbes Formation on Buckeye Creek (area
21); lower Bostrychoceras elongatum Zone of the Haslam
Formation on Elkhom Creek (area 3) and Browns River (area
4).

GEOLOGIC AGE. Santonian.

REMARKS. More than 100 specimens are assigned to this

species. The largest of these, UCLA 59609 from UCLA loc.
3298, has a diameter of 20.0 mm. Geologically older spec-
imens have rounder whorls (Figures 32, 36, 38), later ones
have a longer, more concave ramp, narrower, more angulate
peripheral swelling, and the parietal callus expanding farther
up onto and around the spire (Figures 40, 41). A specimen
(CAS 31325.01) from CAS loc. 31325 on Buckeye Creek in
the Rumsey Hills (area 21) resembles those from UCLA loc.
3633 on Chico Creek (area 5), but it is more abruptly con-
stricted abapical to the periphery, and has only the more
abapical of the two pseudofolds on the columella. Specimens
from UCLA loc. 3633 are not large. None has denticles on
its outer lip, but two low pseudofolds are present on the
columella. The absence of Bostrychoceras elongation and In-
oceramus schmidti from the Chico Creek section above this
locality but below the earliest occurrence of Baculites chi-
coensis has led to the inference that the late Santonian is
missing from the Chico Creek section (Ward et al., 1983;
Haggart and Ward, 1 984; Haggart, 1 984). The specimen from
Buckhorn Creek is from the lower Forbes Formation, and,
if B. elongation and I. schmidti Zone equivalents are missing
from the Chico Creek section (Figure 2), it may be younger
than those from UCLA loc. 3633 at which Baculites capensis
is present.

The relative stratigraphic positions of P. elaphia and P.
brevirostris are the same in the Chico Formation (area 5) and
in the Nanaimo Basin (area 3). In the Nanaimo Basin P.
elaphia occurs in the lower B. elongation Zone and P. brevi-
rostris occurs in the overlying I. schmidti Zone with Cana-
doceras yokoyamai; on Chico Creek P. elaphia occurs in the
Baculites capensis Zone and P. brevirostris in the overlying
B. chicoensis Zone with Canadoceras yokoyamai. On Mill
Creek (area 2), P. elaphia occurs in the I. schmidti Zone of
the Kingsley Cave Member. These P. elaphia from the I.
schmidti Zone on Mill Creek and the B. elongation Zone of
the Nanaimo Basin resemble the P. elaphia of the upper B.
capensis Zone on Chico Creek. Inoceramus schmidti and C.

Figures 16-61. Three species of Perissitys: P. cretacea (Cooper), P. elaphia new species, and P. brevirostris (Gabb). All figures x 1 unless
otherwise noted. Photos 19-23, 26-32, 35-36, 38, 40, 41, 48-50, 60, 61 by T. Susuki. As nearly as possible the figures are arranged from
geologically oldest to youngest. Figures 16-26, P. cretacea (Cooper, 1896), Coniacian, 16-25, hypotypes; 16, 18, specimen with protoconch
preserved, UCLA 59601 from UCLA loc. 5990, xl.5; 17, young individual with unadorned aperture, UCLA 59494 from UCLA loc. 4104,
x 1.5; 19, UCLA 59589 from UCLA loc. 4104; 20, specimen having aperture cut away to show pseudofolds evanescing apically, UCLA 59593
from UCLA loc. 4104, xl.5; 21, 23, UCLA 59590 from UCLA loc. 4104, x2; 22, 26, UCLA 59592 from UCLA loc. 4104; 24, varix on
ablabral side of shell, UCLA 59591 from UCLA loc. 4104; 25, neotype, UCLA 59588 from UCLA loc. 4104. Figures 27^11, Perissitys elaphia
new species, Santonian; 27-29, holotype, UCLA 59606 from CIT loc. 1232; 30, paratype, UCLA 59612 from CIT loc. 1227, xl.5; 31,
specimen cut to show pseudofolds at aperture and smooth columella of earlier whorls, paratype, UCLA 59607 from CIT loc. 1232; 32,
paratype, UCLA 59614 from CIT loc. 1246, xl.5; 33, ablabral view showing inner lip margin— compare to Figures 24 and 43, paratype,
UCLA 59633 from CIT loc. 1017; 34, paratype, UCLA 59621 from CIT loc. 1016; 35, juvenile, paratype, UCLA 59614 from CIT loc. 1246,
x3; 36, 38, paratype, UCLA 59609 from UCLA loc. 3298; 37, paratype, UCLA 59623 from UCLA loc. 3624; 39, paratype, UCLA 59621
from CIT loc. 1016; 40, 41, hypotype, UCLA 59639 from UCLA loc. 4217. Figures 42-61, Perissitys brevirostris (Gabb, 1864), hypotypes;
42, 44, early Campanian, juvenile with complete anterior canal, UCLA 59649 from UCLA loc. 3637, x 1.5; 43, typical earliest Campanian
profile resembling that of P. elaphia, UCLA 59648 from UCLA loc. 3637; 45-47, specimen figured by Stanton, 1896, USNM 21254 from
near Pentz; 48-50, UCBMP 1 1069 from near Pentz; 51-53, LACMIP 7247 from UCLA loc. 3641; 54, 56, LACMIP 7248 from UCLA loc.
3643; 55, 59, 60, UCLA 59661 from CIT loc. 1400, 60, x2; 57, 58, topotype, UCLA 59666, from UCLA loc. 4082; 61, mid Campanian form
intermediate to P. pacifica, UCLA 28715 from CIT loc. 1 158.

Contributions in Science, Number 380

Popenoe and Saul: Perissitys 1 5

yokoyamai from the Dobbins Shale on Sand Creek (area 2 1 )
are below Chron 33r and of Santonian age (Ward et al., 1983);
but in the Nanaimo Basin they occur with P. brevirostris and
are undoubtedly of Campanian age. Early P. brevirostris oc-
curs with Baculites chicoensis and C. yokoyamai within the
early Campanian Chron 33r on Chico Creek (area 5); on Mill
and Antelope creeks (area 2) P. elaphia similar to those of
the Baculites capensis Zone on Chico Creek occurs in the I.
schmidti Zone. Although /. schmidti Michael, 1899, has not
yet been recognized in the Chico Creek section, other species
that occur with it of both late Santonian and early Campanian
age are present there, and the absence of I. schmidti is more
probably a result of deposition in shallow water rather than
indicative of a hiatus in the section.

If the late Santonian zones of Bostrvchoceras elongatum
and Inoceramus schmidti are missing from the Chico Creek
section the evolution of P. elaphia must be more rapid in
the early Santonian and slower in the late Santonian. I.
schmidti, however, ranges from late Santonian through early
Campanian (Haggart, 1984); its absence in the Chico Creek
section may be related to ecologic factors rather than com-
plete absence of strata of appropriate age, and the morpho-
logic changes in P. elaphia were developed through the entire
Santonian.

This species differs from P. cretacea in its more pyriform
shape, expanded inner lip callus which covers the apertural
face of the last whorl, and in typically having three denticles
at the periphery on the outer lip. Although the growth line
on the whorl has an antispiral sinus similar to that of P.
cretacea, the edge of the outer lip is straightened by callus
adapical to the periphery and the outer lip of P. elaphia has
a straighter profile than that of P. cretacea. P. elaphia differs
from P. brevirostris in having a weaker sutural welt which is
not noded, denticulations in the aperture, and a more bent
siphonal canal.

Pyropsis sp. indet. (Nagao, 1939, p. 228, pi. 2, figs. 5, 5a-
b) is similar in shape to Perissitys elaphia, and the growth
line description is similar, but no columellar pseudofolds or
outer lip denticulations are mentioned. The specimen is small
and “imperfect” (Nagao, 1939, p. 228) and may not yet have
formed the characteristic structures. Pyropsis sp. indet. is
probably a Perissitys new species; it is said to have about ten
peripheral nodes whereas Perissitys elaphia has 11 to 14.

ETYMOLOGY. Elaphos, Greek, deer or stag, for its oc-
currence in the Musty Buck Member of the Chico Formation.

Perissitys brevirostris (Gabb, 1864)

Figures 10, 11, 42-61

Perissolax brevirostris Gabb, 1 864, p. 9 1 , pi. 1 8, fig. 43; Tryon,
1881, p. 104, pi. 30, fig. 67; Stanton, 1896, p. 1047, pi.
67, fig. 4; Whiteaves, 1903, p. 356, pi. 43, fig. 3; Arnold,
1909, p. 104, pi. 1, fig. 2; Grabau and Shimer, 1909, p.
769, fig. 1119; Arnold and Anderson, 1910, p. 274, pi. 23,
fig. 2.

Tudicla ( Perissolax ) brevirostris (Gabb): Tryon, 1 883, p. 141,
pi. 51, fig. 59.

Tudicula ( Perissolax ) brevirostris (Gabb): Cossmann, 1901,
p. 71, text-fig. 21.

Perissitys brevirostris (Gabb): Stewart, 1927, p. 426, pi. 20,
fig. 4; Schenck and Keen, 1940, pi. 17, fig. 5; Popenoe,
1954, p. 17, fig. 4(5); Popenoe, 1973, p. 20, pi. 2, fig. 16;
Saul and Alderson, 1981, p. 36, pi. 3, fig. 5.

Tudicla (Perissitys) brevirostris ( Gabb): Wenz, 1943, p. 1304,
abb. 3718.

Not Perissitys brevirostris (Gabb); Sundberg and Riney, 1 984,
p. 105, fig. 3.3 = Murphitys madonna new species.

DIAGNOSIS. Pyriform Perissitys with three strong spiral
ribs about the periphery, a noded subsutural welt, inner lip
expanded to cover all of the apertural face of the shell, and
no apertural ornaments.

DESCRIPTION. Shell of medium to moderately large size,
pyriform; spire about one-fifth of the height of the shell,
comprised of 1.5 to 2 smooth, globose nuclear whorls, suc-
ceeded by 2-4 gently concave whorls, nearly three times as
wide as high; last whorl about half of the shell height, having
a noded, tumid band just below the suture, a concave ramp,
a rounded periphery accentuated by three strong spiral ribs,
and contracting abruptly to form anterior siphonal canal;
anterior siphonal canal at least as long as the eye-shaped
aperture, curving slightly to the left (apertural view); suture
at the adapical strong rib.

Sculpture on noded subsutural welt of fine spiral threads,
on margin of ramp of very fine distant spiral threads, mid
ramp nearly smooth; strong ribs of periphery subequal, two
adapical ribs equal and stronger, third (abapical) rib closer
to second rib than second is to first; ribs made nodular by
about 1 3 strong short axial ribs, interspaces of strong spiral
ribs having fine threads; base of whorl with about four riblets;
siphonal neck with faint distant spiral threads.

Outer lip nearly straight, with slight antispiral sinus and
spiral antisinus, barely thickened; inner lip callus expanding
up onto spire and covering most of apertural face of shell
and lapping onto abapertural surface of siphonal neck. Ap-
erture without denticles or pseudofolds.

HOLOTYPE. ANSP 4188 (Stewart, 1927, p. 426).
HYPOTYPES. USNM 21254 and UCBMP 11069 from
near Pentz, Butte Co. (area 5); CGS 5792 from Sucia Island,
Washington (area 18); UCLA 28715 from CIT loc. 1158,
Bell Canyon (area 17), 59648-59649 from UCLA loc. 3637,
Chico Creek (area 5), 5966 1 from CIT loc. 1 400, Sucia Island
(area 1 8), and 59666 from UCLA 4082, Tuscan Springs (area
2); LACMIP 7247 from UCLA loc. 3641, and 7248 from
UCLA loc. 3643, both Chico Creek (area 5).

DIMENSIONS. Of hypotypes- UCBMP 11609, height
47 mm, diameter 30.3 mm, height of spire 10 mm; LACMIP
7247, height 33.7 mm, diameter 20.4 mm, height of spire 6
mm; LACMIP 7248, height 40 mm, diameter 27.2 mm,
height of spire 8.6 mm; UCLA 59661, height 22 mm (in-
complete), diameter 1 7.3 mm, height of spire 6 mm; UCLA
28715, height 41.5 mm, diameter 30.8 mm, height of spire
9.8 mm.

TYPE LOCALITY. Tuscan Springs, Tehama Co., Cali-
fornia (area 2).

16 Contributions in Science, Number 380

Popenoe and Saul: Perissitys

DISTRIBUTION. Chico Formation at Tuscan Springs
(area 2); associated with Submortoniceras chicoense (Trask)
in the Ten Mile Member, Chico Formation on Chico Creek
and Butte Creek (area 5), and the Chico Formation at Pentz
Ranch (area 5); upper Holz Shale Member of the Ladd For-
mation, Santa Ana Mountains (area 13); found below and
with Hop/itoplacenticeras va ncou veren.se (Meek) in the Cedar
District Formation on Sucia Island (area 1 8); and in the upper
Inoceramus schmidt i Zone of the Haslam Formation at Blun-
den Point, on Brannan Creek, and the north shore of De-
parture Bay (area 3). Specimens from the lower Chatsworth
Formation in Bell Canyon (area 1 7) are intermediate between
P. brevirostris and P. pacifica.

GEOLOGIC AGE. Early and mid Campanian.

REMARKS. This is the most abundantly represented and
most widely distributed species of Perissitys. The largest of
the more than 100 specimens at hand is UCLA 59658 from
UCLA loc. 3647, which despite lacking most of the shell on
the last whorl, has a diameter of 44.6 mm. Pyropsis sp. indet.
of Nagao (1939, p. 228, pi. 21, fig. 5-5b), which is of Conia-
cian or Santonian age (Hayami and Kase, 1977, p. 65), ap-
pears similar in shape to P. brevirostris, but, except that the
inner lip is said to be expanded, the aperture is not described.
The Alaskan specimens from the Chignik Formation listed
as Perissolax brevirostris Gabb by Martin (1926, p. 304) are
neither brevirostris nor Perissitys. Although they have a pyr-
iform shape similar to P. brevirostris, they have a fold on the
columella and other apertural denticulations, and they lack
the widely expanded inner lip of P. brevirostris. They are
described as Christitys martini new species.

The cowl of callus in P. brevirostris is usually deposited at
intervals and its development probably records the onset of
resting or non-growth periods, and, as it can be seen on the
spire (Figures 55, 59), it is an inner lip equivalent of a varix.
There is no apparent regularity to these growth halts.

Perissitys brevirostris remains a recognized taxon through
a longer time (6± Ma) than any other species of Perissitys,
and its evolutionary changes are more subtle than those of
its predecessor, P. elaphia. P. brevirostris differs from late P.
e/aphia in reaching a larger size, having the siphonal neck
smoother and the subsutural welt more swollen and nodular,
and in having a straighter anterior canal. The parietal lip is
more expanded, but its margin is less distinct. No specimen
assigned to P. brevirostris has denticulations or pseudofolds
arming the aperture. In reducing apertural armaments Cam-
panian Perissitys moves contrary to the apertural strength-
ening trend noted by Vermeij ( 1 977), which he suggests helps
to reduce shell breakage by arthropods. The enlargement of
the callus to cover the apertural face may strengthen the shell,
but it leaves the outer lip and at least half of the last whorl
undefended. The callus becomes thicker and more envel-
oping in P. pacifica. P. pacifica and P. brevirostris are alike
in providing the largest available specimens of Perissitys. In
addition to the thicker callus, P. pacifica differs from P. brevi-
rostris in having the three strong peripheral ribs closer to-
gether, the middle rib strongest and the abapical rib weakest,
and in having a better developed posterior siphonal notch.

Specimens identified as P. cf. P. brevirostris (Figure 43)

from the lower 100 m of the Ten Mile Member on Chico
Creek (area 5) have a higher spire and more concave ramp
than typical P. brevirostris. They are similar to the latest P.
elaphia except for the apertural armaments. A specimen from
UCLA loc. 3635, which is at the base of this interval, has a
low pseudofold on the columella like that of the specimen
(CAS 3 1 325.0 1 ) from CAS loc. 3 1 325, Buckeye Creek, Rum-
sey Hills (area 21). Specimens of P. cf. P. brevirostris from
above the base of this 100-m interval have no pseudofolds.

Ammonites have not been found associated with P. brevi-
rostris at its type locality at Tuscan Springs (area 2). The
specimens of P. brevirostris from there resemble those from
Pentz (area 5), where Submortoniceras chicoense is common
and from the upper 500 m of the Chico Formation on Chico
Creek (area 5). Outcrops of the Chatsworth Formation in
Bell Canyon, Simi Hills (area 17), are believed (Saul and
Alderson, 1981, Saul, 1983) to be older than the zone of
Metaplacenticeras pacificum. Perissitys from Bell Canyon are
intermediate between P. brevirostris and P. pacifica (Figure
61).

Perissitys pacifica new species

Figures 62-69

DIAGNOSIS. Pyriform Perissitys which have two strong
and one weak spiral ribs about the periphery, a waved sub-
sutural welt, the inner lip expanded to cover about half of
the shell, and no apertural ornaments.

DESCRIPTION. Shell of medium to moderately large size,
pyriform; spire about one-fifth of the height of the shell,
comprised of 1.5 to 2 smooth, globose nuclear whorls, suc-
ceeded by 2-4 gently concave whorls more than three times
as wide as high; last whorl about half of shell height, having
a waved, tumid band adjacent to the suture, a concave ramp,
and subangulate periphery accentuated by two strong and
one weaker spiral ribs, contracting abruptly to form anterior
siphonal canal; anterior siphonal canal slightly longer than
the eye-shaped aperture, curving slightly to the left (apertural
view) and abaperturally; suture at the adapical strong rib.

Sculpture on waved sutural welt of about five close-set
spiral riblets, on ramp of distant spiral threads; peripheral
spiral ribs made nodular by 12-13 strong, very short axial
ribs, interspaces of peripheral spiral ribs having riblets; base
of whorl with about four spiral riblets; siphonal neck with
distant spiral threads.

Outer lip nearly straight with a slight posterior siphonal
sinus at the sutural welt, a faint antispiral sinus and spiral
antisinus at the periphery, edge thickened by callus which
coats a narrow strip of the outside of the whorl, expands up
over the spire, and covers the apertural face of the whorl.
Aperture without denticles or pseudofolds.

HOLOTYPE. UCLA 59691.

PARATYPES. UCLA 59692 from CIT loc. 1 1 59; 59696-
59698 from CIT loc. 974; 59712 from UCLA loc. 4207;
59715 from UCLA loc. 2415; 59719 from UCLA loc. 7110.

DIMENSIONS. Of holotype— height 30.5 mm (incom-
plete), diameter 24.4 mm, height of spire 6.5 mm; of para-
type— UCLA 59719, height 33.8 mm, diameter 19 mm (in-

Contributions in Science, Number 380

Popenoe and Saul: Perissitys 1 7

18 Contributions in Science, Number 380

Popenoe and Saul: Perissitys

complete), height of spire 5.5 mm; UCLA 59696 — height
36.5 mm (incomplete), diameter 29.5 mm, height of spire 9
mm.

TYPE LOCALITY. CIT loc. 1 159, Dayton Canyon, Simi
Hills, Los Angeles Co., California (area 17).

DISTRIBUTION. Associated with Metaplacenticeras cf.
M. pacificum (Smith) in the Pleasants Sandstone Member of
the Williams Formation, Santa Ana Mountains (area 1 3); the
Chatsworth Formation in Dayton Canyon, Simi Hills (area
17); and the Tuna Canyon Formation in the Santa Monica
Mountains (area 16). Also collected from near the top of the
Debris Dam Sandstone, Agua Caliente Canyon (area 1), and
an unnamed formation on the Salsipuedes arm of Santa Mar-
garita Lake (area 1 5).

GEOLOGIC AGE. Late Campanian, Metaplacenticeras
pacificum Zone.

REMARKS. A greater length of anterior siphonal canal is
present on the specimen (Figure 63) from Garapito Creek,
Santa Monica Mountains (area 16) than on any of the other
50+ available specimens.

The largest specimen, UCLA 597 1 5 from UCLA loc. 24 1 5,
has a diameter of 51.5 mm. P. pacifica is very similar to P.
brevirostris in shape and sculpture. It has a shorter, more
callus-covered spire; the peripheral ribs are closer together;
the subsutural welt is waved rather than noded; and the
posterior siphonal notch is better developed. Steinkerns of
the two species can be distinguished even if most of the shell
is missing; those of P. brevirostris have a roundly convex
profile, whereas those of P. pacifica are flattened to slightly
concave adapical to the more angulate periphery. Some spec-
imens of P. pacifica have less prominent peripheral nodes
and are closer to P. colocara than to P. brevirostris.

ETYMOLOGY. Pacifica, for its occurrence in the Meta-
placenticeras pacificum Zone.

Perissitys colocara new species

Figures 70-86

Cophocara n.sp. Popenoe, 1973, p. 24, fig. 39; Saul and Al-

derson, 1981, p. 36, pi. 3, figs. 6, 7; Sundberg and Riney,

1984, p. 105, fig. 3.1.

DIAGNOSIS. Pyriform Perissitys which have two to three
closely spaced spiral ribs about the periphery, a strong wrin-
kled subsutural welt, inner lip expanded to cover at least two-
thirds of the shell, and no apertural ornaments.

DESCRIPTION. Shell of medium to moderately large size,
pyriform; spire variable but usually less than one-fourth of
height to shell, comprised of 2-3 globose, smooth nuclear
whorls, succeeded by two convexly rounded whorls com-
monly four times as wide as high, followed by 1-2 barely
concave whorls of similar height to width, all callus coated;
last whorl constituting half of shell height, having a tumid
band just below the suture, a slight concavity to the ramp
profile adjacent to the tumid band, and broadly expanded,
subangulate periphery accentuated by 2-3 spiral ribs, con-
tracting abruptly to form anterior siphonal canal; anterior
siphonal canal longer than the triangulate aperture; suture
adapical to the peripheral ribs, just abapical to the slight
concavity of the ramp.

Sculpture on sutural welt of 2-5 spiral riblets, on ramp of
distant spiral threads; peripheral spiral ribs made nodular by
10-14, extremely short, axial ribs; base of whorl commonly
with four fine spiral riblets; siphonal neck with distant fine
spiral threads.

Outer lip with posterior siphonal notch at the sutural welt,
a slight antispiral sinus just adapical to the periphery and
spiral antisinus at the periphery, edge slightly thickened by
callus which covers at least two-thirds of the shell. Aperture
without denticles or pseudofolds.

HOLOTYPE. LACMIP 7249.

PARATYPES. UCLA 59082 from UCLA loc. 6534, 592 1 7
from UCLA loc. 3814; LACMIP 7253-7254 from LACMIP
loc. 7792, 7256 from UCLA loc. 3268, 7250-7252 from
UCLA loc. 6534, 7255 from UCLA loc. 7149, and 7257
from LACMIP loc. 2852; UCBMP 37992 from UCB loc.
A-66 18, 37993 from UCB loc. B-532 l;and CAS 463.01 from
CAS loc. 463.

DIMENSIONS. Of holotype- LACMIP 7249, height 36.7
mm, diameter 19 mm, height of spire 8 mm; of paratypes—
LACMIP 7250, height 38.4 mm, diameter 24.8 mm, height
of spire 12.3 mm; UCLA 59082, height 23.8 mm, diameter
14.3 mm, height of spire 5.6 mm; LACMIP 7256, height

Figures 62-103. Four species of Perissitys: P. pacifica new species, P. colocara new species, P. stantoni (Stewart), and P. stewarti (Zinsmeister).
All figures x 1 unless otherwise noted. Photos 72, 74-76, 84, 85, 88, 93, 94 by T. Susuki. As nearly as possible, the figures are arranged from
geologically oldest to youngest. Figures 62-69, P. pacifica new species, late Campanian; 62, 64, holotype, UCLA 5969 1 from CIT loc. 1 1 59;
63, paratype, UCLA 59719 from UCLA loc. 71 10; 65, 66, paratype, UCLA 59712 from UCLA loc. 4207, x 1.5; 67, 69, paratype, UCLA
59696 from CIT loc. 974; 68, paratype, UCLA 59715 from UCLA loc. 2415. Figures 70-86, P. colocara new species, early Maastrichtian;
70, paratype, LACMIP 7252 from UCLA loc. 6534; 71, 74, holotype, LACMIP 7249 from LACMIP loc. 7962; 72, 73, paratype, UCLA 59082
from UCLA loc. 6534; 75, 76, paratype, LACMIP 7250 from UCLA loc. 6534; 77, juvenile with protoconch, paratype, LACMIP 7254 from
LACMIP loc. 7792, x4; 78, 79, 81, juvenile with protoconch, paratype, LACMIP 7253 from LACMIP loc. 7792, x2.5; 80, 83, paratype,
LACMIP 7255 from UCLA loc. 7149, 83, x 1.5; 82, paratype, CAS 463.01 from CAS loc. 463, x 1.5; 84, 85, paratype, LACMIP 7256 from
UCLA loc. 3268; 86, hypotype, UCBMP 37992 from UCB loc. A-66 18. Figures 87-95, P. stantoni (Stewart), hypotypes, late Maastrichtian;
87, 89, LACMIP 7259 from CIT loc. 1602; 88, UCBMP 37994 from UCB loc. 249; 90, 92, USNM 400974 from USGS loc. 7059; 91, UCBMP
37995 from UCB loc. A-3216; 93, 94, LACMIP 7258 from CIT loc. 1602; 95, LACMIP 7260 from UCLA loc. 1594. Figures 96-103, P.
stewarti (Zinsmeister), hypotypes, early Paleocene; 96, UCBMP 37996 from UCB loc. A-3262; 97, 101, CAS 61617.01 from LSJU loc. 1068,
x 1.5; 98, 103, CAS 61619.01 from LSJU loc. 2245, 103, x 1.5; 99, 100, CAS 61616.01 from LSJU loc. 460; 102, CAS 61616.02 from LSJU
loc. 460.

Contributions in Science, Number 380

Popenoe and Saul: Perissitys 19

31.2 mm (incomplete), diameter 25.4 mm, height of spire

9.7 mm; LACMIP 7255, height 20 mm (incomplete), di-
ameter 17 mm, height of spire 5 mm; CAS 463.01, height

20.7 mm, diameter 16.5 mm, height of spire 4.8 mm.

TYPE LOCALITY. LACMIP loc. 7962, Carlsbad Re-
search Park, San Diego Co., California (area 20).

DISTRIBUTION. Rosario Formation in Arroyo Santa
Catarina (area 14), at Punta San Jose (area 10), near San
Antonio del Mar (area 12), at Point Loma (area 9), and
vicinity of Carlsbad (area 20); Chatsworth Formation on
Lang Ranch, Simi FTills (area 17); upper Panoche Formation
north of Coalinga, Fresno Co. (area 8); “Moreno Grande”
Formation of Huey (1948) on Ortigalita Creek, Merced Co.
(area 8).

GEOLOGIC AGE. Early Maastrichtian.

REMARKS. The greatest abundance of this species is from
localities in Arroyo Santa Catarina (area 14), but specimens
are also common at UCLA loc. 5902 north of Coalinga (area
8). Most are small, but a few medium-sized specimens are
available and a large partial specimen from UCB loc. A-66 1 8
has a diameter of 33.7 mm.

The sculpture on the sutural welt and the ramp is subdued,
but the callus coating is applied so as to accentuate the spiral
threads and riblets and P. colocara appears to have stronger
spiral sculpture than does P. pacifica. On many specimens
such a thick welt of parietal callus is applied at the growth
pauses that the spiral curve of the shell is offset. The speci-
mens look lumpy and distorted. P. colocara is unusually
variable for this group in height of spire and peripheral rib-
bing. A few specimens have a spire height closer to one-third
of the height of the shell (Figures 75, 76) rather than the more
common under one-fourth proportion. Some specimens have
peripheral ribs similar to those of P. pacifica, but most have
finer ribs, more closely spaced, similar to those of P. stantoni.
P. colocara is more callus coated than is P. pacifica and a
little less so than P. stantoni. The subsutural welt of P. col-
ocara is stronger than that of P. pacifica, is adapical to the
peripheral nodes and less undulating than that of P. pacifica.
The suture on adult P. colocara is more deeply channeled
than it is on P. pacifica or P. stantoni, and the whorl of P.
colocara is more angulate than is that of P. stantoni.

ETYMOLOGY. Kolos, Greek, docked, curtailed, short-
ened, stunted, and kara, Greek, head, top, referring to the
usually short spire of the species.

Perissitys stantoni (Stewart, 1927)

Figures 87-95

Cophocara stantoni Stewart, 1927, p. 428, pi. 20, figs. 1-3;

Saul, 1986, p. 27, fig. 21.

Tudicla(Cophocara)stantoni(Stewar\):'Wenz, 1943, p. 1305,

fig. 3719.

Not Cophocara stantoni Stewart: Smith, 1975, p. 475, pi. 2,

figs. 17, 18 = P. stewarti (Zinsmeister).

DIAGNOSIS. Pyriform Perissitys with one dominant spi-
ral rib about the periphery, a weakly noded subsutural welt,
the shell well coated with callus except for last quarter of last
whorl, and apertural ornaments lacking.

DESCRIPTION. Shell of medium to moderately large size,
pyriform; spire about one-fifth of height of shell, comprised
of 3-4 post-nuclear whorls about four times as wide as high,
all callus coated; last whorl about half of shell height, having
a tumid band adjacent to the suture, a slight concavity to
the ramp profile adjacent to the tumid band, and an ex-
panded, subangulate periphery accentuated by 1-3 spiral ribs
of which one is dominant; whorl contracting abapically
abruptly to form anterior siphonal canal; anterior siphonal
canal nearly as long as eye-shaped aperture; suture just abapi-
cal to the slight concavity of the ramp.

Sculpture nearly obliterated by callus coating except on
last quarter of last whorl; sutural welt of large specimens
slightly nodulose; perhipheral spiral ribs made nodular by
15-17 extremely short axial ribs; base of whorl with about
three spiral riblets.

Outer lip with posterior siphonal notch at the sutural welt;
growth line strongly bent at the posterior notch, otherwise
straight; inner lip expanded to cover all but last quarter of
last whorl. Aperture without denticles or pseudofolds.

HOLOTYPE. USNM 73399.

PARATYPES. USNM 73400 and 73403 from USGS loc.
1258; whereabouts of the ANSP specimen figured by Stewart
(1927, pi. 20, fig. 2) from near Martinez, Contra Costa Co.
(area 7), is unknown (Elana Banamy, in litt., 17 Oct. 1985).

HYPOTYPES. USNM 400974 from USGS loc. 7059 (area
8); UCBMP 37994 from UCB loc. 249 (area 7) and 37995
from UCB loc. A-3216 (area 8); LACMIP 7258-7259 from
CIT loc. 1 602 (area 7), and 7260 from UCLA loc. 1 594 (area
19).

DIMENSIONS. Of holotype— height 42.2 mm (incom-
plete), diameter 26 mm, height of spire 17.5 mm; of hypo-
types— USNM 400974, height 34.6 mm (incomplete), di-
ameter 27.4 mm, height of spire 7.5 mm; UCBMP (37994),
height 30.5 mm (incomplete), diameter 24 mm, height of
spire 8.7 mm; LACMIP 7258, height 47.3 mm, diameter 27
mm, height of spire 9 mm.

TYPE LOCALITY. USGS loc. 1 258, near Blum and Com-
pany’s Pacheco warehouse, 1 V2 mi. north of Pacheco, Contra
Costa Co., California (area 7).

DISTRIBUTION. Near or at the top of the Great Valley
Series in the vicinity of Martinez, Deer Valley, and Riggs
Canyon (area 7); Garzas and Volta Sands near Pacheco Pass,
Merced and Stanislaus cos. (area 8); Asuncion Formation on
Cantinas Creek, southern Santa Lucia Range, San Luis Obis-
po Co. (area 1 5); basal San Francisquito Formation on Warm
Springs Mountain (area 19).

GEOLOGIC AGE. Late Maastrichtian.

REMARKS. The type species of Cophocara Stewart, 1 927,
is C. stantoni. As the most notable difference between it and
Perissitys pacifica is the thicker, more enveloping callus layer
on C. stantoni, we include stantoni in Perissitys. P. stantoni
has fewer and weaker spiral ribs, weaker nodes and thicker
callus covering than any other described species of Perissitys.
It is very similar to P. colocara, but has a less concave whorl
profile abapical to the sutural welt and more rounded whorl
profile abapical to the periphery, more and smaller nodes on
the periphery, a more nodulose subsutural band, and better

20 Contributions in Science, Number 380

Popenoe and Saul: Perissitys

developed posterior siphonal notch. Although a few speci-
mens of P. colocara are large, the average size of the 40 +
specimens studied of P. stantoni is about twice that for spec-
imens of P. colocara , and the largest specimen recorded was
one of Gabb’s, which was a diameter of 36 mm (Stewart,
1927, p. 428).

The specimens from UCLA loc. 1 594 on Warm Springs
Mountain are poorly preserved. They resemble P. stantoni,
but may have had a more tumid subsutural welt and a spiral
row of nodes on the ramp (Figure 95).

Perissitys stewarti (Zinsmeister, 1983)

Figures 96-103

Cophocara stantoni Stewart: Smith, 1975, p. 465, pi. 2, figs.

17, 18. Not Cophocara stantoni Stewart, 1927.

Cophocara stewarti Zinsmeister, 1983, p. 1298, fig. 3 J; not

fig. 7>K = “ Heteroterma trochoidea Gabb” of Nelson,

1925, non Gabb, 1869.

DIAGNOSIS. Pyriform Perissitys with two spiral ribs about
the periphery, a nodulose subsutural welt, the inner lip ex-
panded to cover nearly two-thirds of the shell, and no aper-
tural ornaments.

HOLOTYPE. UCR 6670/9.

HYPOTYPES. CAS 61619.01 from LSJU loc. 2245, CAS

61616.01 and 61616.02 from LSJU loc. 460, and CAS

61617.01 from LSJU loc. 1068; UCBMP 37996 from UCB
loc. A-3262.

DIMENSIONS. Of holotype— height 28 mm (incom-
plete), diameter 29 mm, height of spire 5 mm; of hypotypes—
LSJU 10240, height 25.9 mm, diameter 19.3 mm, height of
spire 4.3 mm; LSJU 10241, height 23 mm (incomplete),
diameter 24.5 mm, height of spire 5.9 mm; CAS 61617.01,
height 16.9 mm (incomplete), diameter 4 mm, height of spire
3 mm.

TYPE LOCALITY. UCR loc. 6670, Simi Hills, Ventura
Co., California (area 17).

DISTRIBUTION. Lower Santa Susana Formation, Simi
Hills (area 17) (Zinsmeister, 1983, p. 1298); near the base of
the Laguna Seca Formation, Panoche Hills, Fresno Co. (area
8) (Smith, 1975, p. 468); San Francisquito Formation on
north side of East Fork Fish Canyon (UCLA loc. 1 581) (area
1 9); Sepultura Formation at Punta Canoas (UCLA loc. 6368)
(area 14).

GEOLOGIC AGE. Early Paleocene, late Turritella pen-
insularis quaylei (area 19) and T. peninsularis zones (area
1 7). Specimens from near the base of the Laguna Seca For-
mation (area 8) may be older; they are probably below lo-
calities with T. peninsularis, but the available collections
have no associated turritellas. Misidentified as Cophocara
stantoni, P. stewarti has been used to indicate Cretaceous
age. Some of the localities are recorded as being from the
Moreno Formation, others as being from the Laguna Seca
Formation. The formational name seems in some cases to
have been chosen on a faunal rather than lithologic basis:
Moreno, if the fossils were considered to indicate Cretaceous
age. Laguna Seca (or equivalent), if a Paleocene age seemed

indicated. Plotted on a map, the purported younger locality
may appear to be downsection from the older, and strati-
graphic range of P. stewarti cannot be inferred.

REMARKS. The spire makes up one-seventh to one-eighth
of the height of the shell. Post-nuclear whorls are about six
times as wide as high, and all are callus coated; about half
of shell height is the last whorl. The two strong peripheral
spirals are made nodose by 14-17 extremely short, axial ribs;
and the base of the whorl has 4-5 spiral riblets adjacent to
the siphonal neck. The growth line is strongly bent at the
posterior notch but is otherwise straight. The aperture is
without denticles or pseudofolds. The anterior canal is straight
except for a backward bend near its tip.

P. stewarti is the youngest recognized Perissitys. It has the
shortest spire of known species, and were its ancestry un-
known, its classification in the Tudiclidae (Zinsmeister, 1 983,
p. 1298) would be apt. P. stewarti is very similar to, and
easily confused with, P. stantoni, but has a thinner, less ex-
tensive callus, more strongly marked varices, a deeper pos-
terior siphonal notch, and stronger nodes on the subsutural
band. The peripheral nodes are in two nearly equal rows,
whereas P. stantoni has one dominant row. The anterior
siphonal canal is much shorter and has a stronger backward
flexure near its tip. A virtually shell-less specimen from UCB
loc. A-3262 has a diameter of 26. 2 mm and, with shell, would
be larger than the holotype.

Perissitys stewarti resembles Heteroterma trochoidea Gabb,
1869, in its two spiral rows of nodes and short spire, but H.
trochoidea lacks the callus coating over the shell, the strong
rounded welt at the suture, and has a more sloping ramp. P.
stewarti resembles H. gabbi Stanton, 1896, in whorl profile,
but H. gabbi lacks the callus coating and has longer axial ribs
crossed by several spiral ribs. H. striata Stanton, 1896, lacks
the sutural welt and the callus coating and has at least four
spiral ribs about the periphery.

Genus Pseudocymia new genus

TYPE SPECIES. Pseudocymia aurora new species.

DIAGNOSIS. Small to medium sized, shortly fusiform
gastropods with shouldered whorls having a concave ramp
and base straightly sloping into an anterior siphonal neck of
short to moderate length which is bent to the left and abaper-
turally, and has near its anterior end a well-developed si-
phonal fasciole. The shell appears to be out-of-round, a little
flattened aperture-to-back and expanded laterally. It is or-
namented by narrow spiral ribs which override strong, short
axial ribs. The axial ribs are most pronounced on the pe-
riphery. The spire is less than half of the total shell height.
The aperture is elongate oval with well-demarked inner lip
and flaring outer lip. The inner lip has two pseudofolds near
the base of the whorl and another near the posterior end of
the aperture. The outer lip is thickened and denticulated with
the strongest denticulations opposite the base of the whorl.

RANGE. Turonian to Maastrichtian.

DISCUSSION. This genus is based on P. aurora of Tu-
ronian age. It resembles species of Columbellina but lacks
the marked, oblique posterior sinus in the outer lip. The outer

Contributions in Science, Number 380

Popenoe and Saul: Perissitys 2 1

lip denticulations are, however, separated into posterior and
medial sets by a groove similarly placed to that of Colum-
bellina's posterior sinus. P. aurora has a remarkable resem-
blance to Cymia tecta (Wood, 1828) of the modern Panamic
fauna, but lacks the medial columellar fold of that species
and is slightly compressed. The aperture-to-back flattening
of the shells is suggestive of Cymatiidae and Bursidae. Another
group in the Perissityidae which shares this characteristic is
that of “ Fusus ” kingii Gabb. Pseudocymia is uniangulate
about the periphery rather than distinctly biangulate as is the
“ Fusus ” kingii group. Pseudocymia remains fusiform rather
than becoming pyriform as do Perissitys and Christitys. It
lacks the columellar folds of Christitys and Murphitys.

The genus is named for its resemblance to the type species
of Cymia, C. tecta (Wood, 1828).

Pseudocymia aurora new species

Figures 104-108

DIAGNOSIS. Thick-shelled Pseudocymia with a broad
apical angle and an outer lip that is considerably thickened
and internally strongly lirate.

DESCRIPTION. Shell of medium size, broadly fusiform,
rather massive, somewhat compressed aperture-to-back; spire
of about four whorls, convex-conical in outline; apical angle
approximately 60°; whorl outline subangular, shouldered just
abapical to the middle, ramp gently concave; last whorl with
a fairly broad, steeply sloping, slightly concave ramp adapical
to the shoulder, abapical to which the whorl contracts evenly
to the anterior end; suture linear, impressed, sinuous; si-
phonal fasciole well developed, rounded, enclosing a very
shallow umbilical pit.

Spiral sculpture of the body whorl consisting of about 12
moderately strong revolving round-topped ridges below the
whorl shoulder, and two or three much weaker spirals above
the whorl shoulder, separated by interspaces somewhat wider
than the spirals, and generally with two or three very fine
revolving lirae in the interspaces; axial sculpture of about 1 2
blunt rounded ribs on the shoulder, obsolete or absent adapi-
cally and abapically.

Inner lip with a moderately thick callus coating, concave
parietally, nearly straight anteriorly to the narrow and twisted
canal, ornamented near the posterior suture with a pro-
nounced rounded swelling bearing three or four transverse
denticulations, and near its mid-length by two low rather
distantly placed pseudofolds of which the posterior is slightly
stronger; outer lip much thickened and bearing internally a
series of strong transverse wrinkles which increase in size
progressively from anterior end to whorl shoulder; outer lip
rather deeply notched internally opposite the whorl shoulder,
posterior to which two smaller wrinkles oppose the swelling
on the inner lip and enclose adapically a narrow anal gutter
at the juncture of the lips.

HOLOTYPE. UCLA 59586.

PARATYPE. UCLA 59587 from CIT loc. 1 532, Salt Creek,
Millville Quadrangle (area 1 1).

DIMENSIONS. Of holotype— height 37.5 mm; height of
last whorl 25 mm; diameter 23.5 mm; height of spire 14.4
mm.

TYPE LOCALITY. CIT loc. 1212, Salt Creek, Millville
Quadrangle, Shasta Co., California (area 1 1).

DISTRIBUTION. Represented by two specimens, both
from the upper half of the Frazier Silt on Salt Creek, Shasta
Co., California.

GEOLOGIC AGE. Late Turonian, horizon of Subpriono-
cyc/us neptuni, below occurrences of Collignoniceras sp. (Jones
et al., 1978, p. XXII. 8, figs. 5, 6).

REMARKS. The holotype is a nearly perfect shell, and is
remarkably modern looking for a Turonian gastropod. It is
very similar in shape and sculpture to Cymia tecta, type
species of Cymia. from which it differs chiefly in its circum-
apertural denticulations and lack of the strong fold about
midway of the inner lip. The sculpture of P. aurora resembles
that of Murphitys michaeli. M. michaeli has a longer anterior
canal which lacks the well-developed fasciole of P. aurora.
Although the outer lip of P. aurora is thickened, it is not
rimmed like that of M. michaeli.

ETYMOLOGY. Aurora, Latin, dawn, morning.

Pseudocymia^ ?) aitha new species

Figures 5, 109-1 12

DIAGNOSIS. Small Pseudocymia (?) with three strongest
spiral ribs on the periphery, two of which are clearly evident
on the spire. Within, the outer lip has three strong denticles
near the periphery and small denticles near the anterior canal.

DESCRIPTION. Shell shortly fusiform, small; spire about
half of height of shell, comprised of four post-nuclear whorls,
each about twice as wide as high; apical angle approximately
42°; whorl outline angulate with shallowly concave ramp,
peripheral angulation accentuated by three strong spiral ribs,
base sloping straight to the anterior siphonal fasciole; anterior
canal short, bent to the left and abaperturally; suture coin-
cident with the abapical peripheral spiral cord.

Sculpture of narrow spiral ribs, weak posterior to the pe-
riphery, moderately strong on the periphery and base; inter-
spaces wider than the ribs; peripheral spiral ribs made nodu-
lose by about ten short but strong axial ribs.

Aperture broad posteriorly; outer lip expanded and thick-
ened, with a posterior siphonal notch nearly midway between
the suture and the periphery, bearing a posterior denticle just
anterior to the siphonal notch, two strong denticles at the
periphery, the posterior of which is the stronger, and three
small denticles at anterior end of aperture; inner lip well
marked, bearing two wrinkle-like denticles near anterior end
of aperture, two pseudofolds anterior to the mid-point and
three small denticles at beginning of anterior siphonal canal.

HOLOTYPE. UCLA 39440.

DIMENSIONS. Of holotype— height 28 mm (incom-
plete); diameter 18 mm; height of spire 14 mm.

TYPE LOCALITY. CIT loc. 1007, Oak Run, Millville
Quadrangle, Shasta Co., California (area 1 1).

DISTRIBUTION. One specimen from Member IV of the
Redding Formation in Oak Run, Millville Quadrangle (area
1 1) and a poorly preserved specimen from Member V, Bear
Creek, Whitmore Quadrangle (area 1 1).

GEOLOGIC AGE. Coniacian.

22 Contributions in Science, Number 380

Popenoe and Saul: Perissitys

Figures 104-131. Species of Pseudocymia new genus. All xl unless otherwise noted. Figures 104-108. Pseudocymia aurora new species,
Turanian, holotype, UCLA 59586 from CIT loc. 1212; 104 and 108 photos by T. Susuki. Figures 109-112. Pseudocymia (?) aitha new species,
Coniacian, holotype, UCLA 39440 from CIT loc. 1007; 112, xl.5. Figures 113-118. Pseudocymia (?) cf. P.(l) aitha new species, Santonian,
hypotypes, x2, from UCLA loc. 6496; 113, 115, 116, 118, LACMIP 7261; 114, 117, LACMIP 7262; 117, the back of the last whorl is broken
off and the pseudofolds and median outer lip denticulation are visible. Figures 119-126. Pseudocymia( ?) cahalli new species, Campanian; 119,
121, 124, 125, the anterior canal is broken off and the outer lip is crushed, holotype, LACMIP 7263 from UCLA 2324; 124, x 1.5; 120, 122,
123, paratype, UCLA 39473 from CIT loc. 83; a badly leached shell preserving only remnants of sculpture but more of the anterior canal
including part of the anterior siphonal fasciole, breakage of the posterior end of the outer lip gives a false impression of its posterior outline;
126, paratype, CAS 61618.01 from LSJU loc. 1860 with posterior end of outer lip preserved. Figures 127-131. Pseudocymia (?) kilmeri new
species, early Maastrichtian, holotype, UCBMP 37997 from UCB loc. B-5323, edge of outer lip broken posterior to the periphery.

REMARKS. The holotype is well preserved and nearly
complete, lacking part of the nucleus and the anterior end of
the anterior siphonal canal. Although the growth line has an
antispiral sinus, this sinus is not as sharply bent and deep as
the posterior siphonal notch suggests that it should be. The

notch may, therefore, have been deepened by an injury, and
so definite a notch may not be characteristic of the species.
Despite this deeper siphonal notch and the less fortified ap-
erture, the species is included in Pseudocymia because of
its overall shape and well-developed anterior fasciole.

Contributions in Science, Number 380

Popenoe and Saul: Perissitys 23

Pseudocymiaip) aitha differs from P. aurora in its more
angulate whorl profile, slightly more concave ramp, and fewer
denticles on the outer lip. It is very similar to Perissitys
cretacea, from which it differs in being less expanded about
the periphery, having slightly stronger and fewer spiral ribs
and longer axial ribs, more than one median denticle on the
outer lip, and the most posterior portion of the inner lip
abapical to the peripheral spirals rather than lapping over
them.

ETYMOLOGY. Aitha, Greek, burnt, fiery, reddish-brown,
for its occurrence near Redding, California.

Pseudocymia( ?) cf. /*.(?) aitha new species

Figures 113-118

HYPOTYPES. LACMIP 7261-7262 from UCLA loc.
6496.

DIMENSIONS. Of LACMIP 7261 —height 16.8 mm, di-
ameter 9.6 mm, height of spire 7 mm; of LACMIP 7262-
height 17.4 mm, diameter 10 mm, height of spire 7 mm;
both specimens are incomplete.

DISTRIBUTION. Two specimens from the Panoche For-
mation, Howard Ranch Quadrangle, Merced Co., California
(area 8).

GEOLOGIC AGE. Santonian.

REMARKS. Two small incomplete specimens resemble
P.(?) aitha, but are slightly more slender, have a less angulate
whorl profile, and one or two more axial ribs per whorl. The
posterior section of the outer lip is sulcate, but the notch is
broader than that of P.(?) aitha. As in P.(?) aitha there is a
posterior denticle just anterior to the siphonal notch and two
strong denticles at the periphery, the posterior of which is
stronger. These specimens are superficially similar to M. mi-
chaeli, but they are more slender, have fewer axial ribs per
whorl, and the denticles on the outer lip are placed as in P.(?)
aitha.

Pseudocymia(?) cahalli new species

Figures 119-126

DIAGNOSIS. Elongate Pseudocymiaip. ) with the outer lip
ornamented by two strong denticles at the periphery, a pos-
terior denticle of moderate size and small denticles near the
anterior siphon.

DESCRIPTION. Shell of medium size, fusiform; spire of
about five post-nuclear whorls, each about twice as wide as
high; apical angle approximately 30°; whorl outline suban-
gular with shallowly concave ramp, peripheral angulation
accentuated by short axial ribs; suture just abapical to the
periphery.

Sculpture of narrow subequal spiral ribs; interspaces wider
than the ribs; periphery made nodulose by about ten short
but strong axial ribs.

Aperture widest posteriorly; outer lip expanded and thick-
ened by varix, bearing a pair of strong denticles opposite the
periphery, a posterior denticle, and at least two smaller ones
near the anterior siphon; inner lip of moderate width, thick,
with a strong pseudofold opposite the periphery, a lesser one

near the anterior siphon, one or two denticles at the anterior
siphon, and one or two denticles near the posterior end.

HOLOTYPE. LACMIP 7263.

PARATYPES. UCLA 39473 from CIT loc. 83 and CAS
61618.01 from LSJU loc. 1860.

DIMENSIONS. Of holotype— height 35.7 mm (incom-
plete); maximum diameter 18.3 mm (incomplete); height of
spire 1 7 mm.

TYPE LOCALITY. UCLA loc. 2324, Santa Ana Moun-
tains, Orange Co., California (area 13).

DISTRIBUTION. Cedar District Formation, Sucia Island
(area 1 8); and uppermost Holz Shale Member of Ladd For-
mation, Santa Ana Mountains (area 13).

GEOLOGIC AGE. Mid Campanian.

REMARKS. The above description is based upon three
incomplete specimens; one from Sucia Island and two from
the Santa Ana Mountains. None of the specimens retains the
anterior siphonal canal, but there is an indication of a well-
developed siphonal fasciole. The spiral ribs appear to be
strongest anterior to the periphery. The short axial ribs are
most numerous on the spire (about 14) but are reduced to
about ten stronger ones on the ultimate whorl. The outer lip
is poorly preserved on a paratype (UCLA 39473). Denticles
within the outer lip are described from impressions on
the infilling of the holotype and the second paratype (CAS
61618.01).

P.(l) cahalli is most similar to P.('l) kilmeri, from which
it differs in having a less strongly angulate whorl profile, the
suture closer to the noded periphery, and stronger pseudo-
folds on the columella.

ETYMOLOGY. The species is named for C.A. Hall.

Pseudocymia( ?) kilmeri new species

Figures 127-131

DIAGNOSIS. Elongate Pseudocymiap) expanded periph-
erally and contracted suturally, with six evenly spaced linear
denticles within the outer lip between the periphery and the
anterior canal. There are two weak pseudofolds on the col-
umella.

DESCRIPTION. Shell medium sized, subfusiform with
backward flexed short open anterior canal and small pseu-
doumbilicus; spire almost half of height of shell, comprised
of five post-nuclear whorls, each about twice as wide as high;
apical angle approximately 38°; profile of last three whorls
concave adapical to the periphery, prior whorls convex; base
sloping straight into constriction adapical to the siphonal
fasciole; suture abapical to the periphery; shell broadened by
indistinct varices at about 1 80°.

Overall sculpture of spiral riblets strongest on the periph-
eral angulation, progressively weaker from periphery to both
sutures with weakest riblets on the sutural ramp and three
or four strongest on the angulation, one to three threads
between the riblets; about ten axially elongated nodes per
whorl on the periphery; suture on fourth abapical strong rib.

Aperture widest posteriorly; outer lip thickened by varix,
expanded medially, and bearing a strong denticle plus one
or two smaller ones near the posterior end, two strong den-

24 Contributions in Science, Number 380

Popenoe and Saul: Perissitys

tides at the periphery and four of decreasing strength ante-
riorly; inner lip of moderate width and thickness with small
denticles bordering the posterior notch, two faint pseudofolds
just anterior to the midpoint and a low welt near the anterior
end.

HOLOTYPE. UCBMP 37997.

DIMENSIONS. Height 47.4 mm; diameter 24.9 mm;
height of spire 22.4 mm; height of penultimate whorl 10.3
mm; diameter of penultimate whorl 16 mm.

TYPE LOCALITY. UCB loc. B-5323: Punta San Jose,
Baja California (area 10).

DISTRIBLITION. Rosario Formation, Punta San Jose
(area 10).

GEOLOGIC AGE. Early Maastrichtian.

REMARKS. The above description is based upon the only
known specimen, the nearly complete holotype. The outer
lip is broken along the sutural ramp; the tip of the anterior
canal is chipped; some shell is missing from the back of the
body whorl, and the apex is eroded.

P. (?) kilmeri has much more subdued apertural ornament
than P. aurora, but the denticulations are similarly placed.
P.(?) kilmeri is also much more elongate, has a more fusiform
shape, and slightly longer anterior canal. In sculpture and
shape of base it resembles P.(l) ait ha, but it is also more
fusiform than P.(?) ait ha, has a more abapical suture, and
has stronger spiral ribs adapically to the three prominent
ones on the periphery.

In shape and sculpture P.(?) kilmeri bears a strong resem-
blance to Solenosteira spp., especially S. gatesi Berry, 1963,
of the modern Panamic fauna. It differs from S. gatesi most
notably in the flare of its outer lip and the strength of the
denticles within.

ETYMOLOGY. The species is named for F.H. Kilmer
who collected the holotype at Punta San Jose, Baja California,
Mexico.

Genus Murphitys new genus

TYPE SPECIES. Murphitys michaeli new species.

DIAGNOSIS. Plumply bucciniform gastropods of small
to medium size which have a rounded to doubly subangulate
whorl profile. Whorls are ornamented by spiral ribs, two of
which predominate, and short axial ribs. Outer lip expanded
to form a rim which is thickened at its edge and has posterior,
medial, and anterior denticles within. Parietal lip is roundly
expanded. Columella has two folds. Anterior canal is mod-
erately long, slightly curved, and has a fasciole near its tip.

DISCUSSION. Characteristic of this genus is the rimmed
outer lip with its inner surface rolled outward. The parietal
lip is roundly expanded, although that of the earliest species,
M. michaeli, is only slightly more expanded than the parietal
lip of Perissitys cretacea or Pseudocymia aurora, both of
which it resembles in shape and sculpture. The parietal lip
is more roundly expanded in later species of Murphitys but
is not as expansive as the parietal lip of Perissitys spp. Murph-
itys differs from Perissitys and Pseudocymia in having folds
on the columella. The folds are faint on M. michaeli but

become strong on M. madonna. Additional trends in this
lineage lead to the loss of the posterior sinus on the outer lip
and to a more angulate whorl prohle.

The illustration of Fusus (s.l.) volutodermoides Nagao, 1939
(p. 231, pi. XXII, fig. 5) from the upper Yezo Group, Abe-
shinai, Teshio Prov., Japan, resembles M. michaeli, but that
species is described as having a smooth columella.

RANGE. Coniacian to early Maastrichtian.

ETYMOLOGY. The name is a compound of Murphy (for
M.A. Murphy) and itys, Greek, rim or felly of a wheel.

Murphitys michaeli new species

Figures 9, 1 32-147

DIAGNOSIS. Small Murphitys which have a rounded
whorl profile and peripheral spirals that are scarcely stronger
than the other spirals. The outer lip bears a strong medial
denticulation. The two columellar folds are, although dis-
tinct, not strong.

DESCRIPTION. Shell small, bucciniform; spire about one-
third of total shell height, consisting of about five whorls
including a smooth mammillate protoconch of one whorl
succeeded by whorl sculptured by raised spiral threads and
axial ribs; all whorls roundly convex; base of whorl narrowing
to form a sturdy anterior siphonal canal of moderate length
and twisted to the left (apertural view) and backward at its
tip; suture adapical to the whorl mid-point.

Sculpture of alternating spiral ribs and threads with inter-
spaces usually a little wider than the ribs; two or three pe-
ripheral ribs slightly stronger and the suture on the most
anterior of these; spiral sculpture undulated by arcuate col-
labral ribs which decrease in number to about 15 and in
regularity and strength on the last whorl.

Outer lip thickened and slightly expanded to form a rimmed
aperture with shallow antispiral sinus adapical to the pe-
riphery and slight spiral antisinus anterior to the periphery;
inner lip fairly broad and thin, parietal portion roundly ex-
panded onto the body whorl. Outer lip with two posterior
denticles, a strong medial denticle and a clump of small
denticles just posterior to the anterior canal; inner lip with
a posterior denticle and two folds, the posterior one opposite
the strong median denticle of the outer lip and the anterior
one just posterior to the clump of outer lip denticles; an
additional denticle sometimes present at the posterior end
of the anterior canal.

HOLOTYPE. LACMIP 7264.

PARATYPES. LACMIP 7265-7269 from UCLA loc.
4106, 7270 from CIT loc. 1247, and 7271-7272 from CIT
loc. 1230.

DIMENSIONS. Of holotype — height 19.7 mm, diameter
13.0 mm, height of spire 7.7 mm; of paratypes— LACMIP
7265, height 28.4 mm, diameter 13.2 mm, height of spire
8.6 mm; LACMIP 7267, height 22.0 mm, diameter exclusive
of outer lip flare 12.6 mm (including outer lip flare 14.3 mm),
height of spire 8.0 mm; LACMIP 7270, height 23.0 mm,
diameter 12.9 mm, height of spire 7.8 mm; LACMIP 7271,
height 19.6 mm, diameter 13.7 mm, height of spire 5.4 mm;

Contributions in Science, Number 380

Popenoe and Saul: Perissitys 25

Figures 132-164. Species of Murphitys new genus. All x 1 unless otherwise indicated. As nearly as possible the figures are arranged from
geologically oldest to youngest. Figures 132-147. Murphitys michaeli new species. 132-136, Coniacian; 137-147, Santonian; 132, 135, paratype
LACMIP 7272 from CIT loc. 1230, xl.5; 133, 134, 136, paratype, LACMIP 7271 from CIT loc. 1230; 133, 134, xl.5; 136, x2; 137-139,
143, holotype, LACMIP 7264 from UCLA loc. 4106; 137, 138, 143, xl.5; 139, x2; 140, paratype with more of anterior canal preserved,
LACMIP 7268 from UCLA loc. 4106; 141, 145, paratype with subdued axial sculpture, LACMIP 7270 from CIT loc. 1247; 142, paratype,
LACMIP 7266 from UCLA loc. 4106, x 1.5; 144, paratype with moderate axial sculpture, LACMIP 7267 from UCLA loc. 4106; 146, paratype
with part of last whorl cut away and columeliar folds exposed, LACMIP 7269 from UCLA loc. 4106, x 1.5; 147, paratype with nearly complete
anterior canal, LACMIP 7265 from UCLA loc. 4106. Figures 148-156. Murphitys corona new species, early Campanian, all from CIT loc.
1053. 148-151, 156, holotype, LACMIP 7273; 156, xl.5; 152, paratype with back cut away and columeliar folds exposed, LACMIP 7275;
153, 155, paratype with more fully developed outer lip, LACMIP 7274; 153, xl.5; 154, paratype with nearly complete anterior canal and
anterior siphonal fasciole, LACMIP 7276. Figures 157-164. Murphitys madonna new species, early Maastrichtian. 157, 161, paratype, SDSNH
25942 from SDSNH loc. 3162A; 158, paratype, LACMIP 7277 from LACMIP loc. 4898, xl.5; 159, 163, paratype with last whorl broken
and columeliar folds exposed, SDSNH 27871 from SDSNH loc. 3162A; 160, 162, holotype, SDSNH 25958 from SDSNH loc. 3161A; 164,
paratype, steinkem with impressions of outer lip armaments at varices (arrows), LACMIP 7278 from UCLA loc. 5431.

26 Contributions in Science, Number 380

Popenoe and Saul: Perissitys

LACMIP 7272, height 19.3 mm, diameter 1 1.9 mm; height
of spire 7.4 mm.

TYPE LOCALITY. UCLA loc. 4106, hillside north of
Clover Creek, Millville Quadrangle, Shasta Co., California
(area 1 1 ).

DISTRIBUTION. Member IV along Oak Run (area 1 1)
and Members V and VI along Clover Creek (area 1 1 ); ?Kings-
ley Cave Member of the Chico Formation on Mill Creek
(very poor specimens) (area 2).

GEOLOGIC AGE. Coniacian and Santonian.

REMARKS. Murphitys michaeli resembles Perissitys cre-
tacea but differs from it in having a more convexly rounded
whorl profile with a concave ramp only on the last third of
the last whorl. The first and second post-nuclear whorls of
M. michaeli are virtually indistinguishable from those of
Perissitys elaphia. Juvenile specimens of the two species then
become distinguishable as P. elaphia develops a concave
ramp. M. michaeli does not have as expanded an inner lip
as P. elaphia nor as long an anterior canal. M. michaeli is
similar to Pseudocymia aurora in its apical profile and sculp-
ture but differs in being more constricted anterior to the
periphery, having a rounded whorl profile and a broader,
thinner, roundly expanded inner lip, lacking a pseudoum-
bilicus, and having fewer denticles on the outer lip.

Strength of spiral ribbing is variable in M. michaeli. Pe-
ripheral spirals are in some specimens scarcely stronger than
those anterior and posterior, but in other specimens two
peripheral spirals are strong enough to give the whorl a sub-
angulate profile.

In apertural view the anterior fold is the more prominent
one, its outer end being farther out on the inner lip, and the
posterior fold is much less noticeable, its outer end being
well within the aperture. The posterior is the more prominent
fold away from the aperture within the spire. Both are, how-
ever, not strong except at varices.

ETYMOLOGY. The species is dedicated to Michael A.
Murphy in recognition of his work on the Cretaceous of the
Ono area, Shasta Co., California.

Murphitys corona new species

Figures 146-156

DIAGNOSIS. Plump Murphitys which have a doubly sub-
angulate periphery demarked by the two strongest spiral ribs.
The outer lip bears about five nearly equal denticulations.

DESCRIPTION. Shell of medium size, roundly buccini-
form; spire less than one-third of total shell height, consisting
of about four post-nuclear whorls; protoconch unknown;
whorls roundly convex with slightly concave ramp and dou-
ble angulation at the periphery; last whorl narrowing to form
a sturdy, nearly straight, anterior canal of moderate length;
suture between the peripheral angulations.

Sculpture of alternating spiral ribs and threads with inter-
spaces about twice as wide as the ribs; two slightly stronger
ribs on the two angulations; ribbing anterior to the periphery
a little stronger than that posterior to the periphery; spiral
sculpture undulated by slightly arcuate, broadly rounded col-
labral ribs that number 1 1-14 on the last whorl; angulations
noded at intersections with collabral ribs.

Outer lip thickened, slightly expanded, and somewhat re-
curved to form a rimmed aperture with shallow antispiral
sinus adapical to the periphery and slight spiral antisinus
anterior to the periphery; inner lip broad and thin, roundly
expanded onto the body whorl. Outer lip with about five
nearly equal denticles and 2-3 denticulations adjacent to the
anterior canal; columella with two folds and a parietal den-
ticle, the stronger fold at the base of the previous whorl, the
weaker one near the posterior end of the anterior canal.

HOLOTYPE. LACMIP 7273.

PARATYPES. LACMIP 7274-7276 from CIT loc. 1053.

DIMENSIONS. Of holotype — height 26.8 mm; diameter
18 mm, height of spire 7.5 mm.

TYPE LOCALITY. CIT loc 1053: spur north of Santiago
Creek, El Toro Quadrangle, Santa Ana Mountains, Orange
Co., California (area 13).

DISTRIBUTION. Ladd Formation, upper Holz Shale
Member, Santa Ana Mountains (area 13); and Chatsworth
Formation, Bell and Dayton canyons, Simi Hills (area 17).

GEOLOGIC AGE. Early Campanian, associated with
Canadoceras cf. C. yokoyamai (Jimbo) (Matsumoto, 1960,
p. 102), zone of Turritella chicoensis holzana Saul, 1983,
through late Campanian Metaplacenticeras pacificum Zone.

REMARKS. The double corona of distinct nodes at the
periphery distinguishes this species from all other perissi-
tyids. The species is based on nine specimens from CIT loc.
1053. A poorly preserved specimen from CIT loc. 94, Ladd
Formation, uppermost Holz Shale Member, T. chicoensis
Zone, of Williams Canyon, Santa Ana Mountains, and ju-
venile specimens from Bell (CIT loc. 1 158) and Dayton can-
yons (CIT loc. 1 1 59), Chatsworth Formation, T. chicoensis
and Metaplacenticeras pacificum zones, Simi Hills (area 17)
are also probably this species.

The more posterior of the two columellar folds appears
weaker in apertural view because it begins farther back within
the aperture, whereas the stronger appearing anterior fold not
only begins farther out toward the inner lip but is also at its
strongest at that point and weakens within the whorl.

Juveniles of Perissitys brevirostris and P. pacifica are sim-
ilar to M. corona, but the latter is more distinctly bicoronate.
Adult M. corona lack the sutural welt of adult P. brevirostris
and P. pacifica.

ETYMOLOGY. Corona, Latin, both for its corona of nodes
on the periphery and its occurrence in the old Corona Quad-
rangle.

Murphitys madonna new species

Figures 157-164

Perissitys brevirostris (Gabb) Sundberg & Riney, 1984, p.

105, fig. 3.3. Not Perissolax brevirostris Gabb, 1864.

DIAGNOSIS. Moderately large, biangulate Murphitys
which have four subequal denticulations along the outer lip
and two strong folds on the columella.

DESCRIPTION. Shell large, doubly angulate about the
broad periphery; spire about one-third of the total shell height,
consisting of about six post-nuclear whorls; protoconch ap-
parently smooth and relatively large; whorls concave on the

Contributions in Science, Number 380

Popenoe and Saul: Perissitys 27

ramp with double angulation at the periphery; last whorl
narrowing to form a sturdy, slightly backward deflected an-
terior canal, bent abaperturally at its tip, longer than the
height of the last whorl, and having a siphonal fasciole at its
tip; suture at or just abapical to the adapical angulation, on
the first four post-nuclear whorls, dropping to just above the
abapical angulation on the last whorl.

Sculpture of fine alternating spiral ribs and threads with
interspaces about three times as wide as the ribs; ribbing
anterior to the abapical peripheral angulation a little stronger
than that posterior to the abapical peripheral angulation; on
third post-nuclear whorl about nine short collabral ribs form-
ing strong nodes at their intersections with the two peripheral
angulations, increasing to 16 on sixth whorl.

Outer lip thickened, forming a rimmed leaf-shaped aper-
ture; inner lip broad and of moderate thickness, parietal por-
tion roundly expanded onto the body whorl. Outer lip with
a pair of denticles posterior to the notch at the adapical
angulation, a pair of denticles between the angulations and
three to four smaller denticles between the abapical angu-
lation and the anterior canal; columella with two folds and
a parietal denticle, the stronger fold at the base of the previous
whorl, the weaker one near the posterior end of the anterior
canal.

HOLOTYPE. SDSNH 25958.

PARATYPES. LACMIP 7277 from LACMIP loc. 4898,
7278 from UCLA loc. 5431, 7279 from UCLA loc. 7137,
7280 from LACMIP loc. 1215, and 728 1 from LACMIP loc.
8159; SDSNH 25942 and 27871 from SDNH loc. 3162-A.

DIMENSIONS. Of holotype— height 64.5 mm, diameter
35.1 mm, height of spire 23.6 mm.

TYPE LOCALITY. SDSNH loc. 3162-A, Carlsbad Re-
search Park, San Diego Co., California (area 20).

DISTRIBUTION. Chatsworth Formation on Lang Ranch,
Simi Hills (area 1 7); Pleasants Sandstone Member of Wil-
liams Formation, Bean Creek, east of San Juan Capistrano
(area 24); Rosario Formation near Carlsbad (Point Loma
Member) (area 20), at Punta Banda (area 22), near Punta San
Jose (area 10), near San Antonio del Mar (area 12), and at
Arroyo Santa Catarina (area 14).

GEOLOGIC AGE. ?Latest Campanian and early Maas-
trichtian.

REMARKS. Although more than 20 specimens of this
species are at hand, most are poorly preserved. The best
specimens are from the Carlsbad Research Park. Small spec-
imens are roundly inflated like Murphitys corona, but have
fewer and stronger nodes about the periphery. Large speci-
mens are much more angulate and resemble stout “ Fusus ”
kingii Gabb, 1 864. They have, however, the rimmed aperture
and roundly expanded inner lip of Murphitys. As in many
other perissityids, steinkems of the species have at intervals
impressions of the outer lip denticles (Figure 164) left by the
irregular development of the varices.

A medium-sized specimen (LACMIP 7281) from LAC-
MIP loc. 8 1 59, east of San Juan Capistrano (area 24) occurs
with Baculites cf. B. subanceps pacificus Matsumoto & Obata,
1963. This baculite is considered to be of late Campanian
age (Matsumoto with Miyauchi, 1984, p. 75), but some other
mollusks from LACMIP loc. 8159 are not typical of late

Campanian faunas: Calva peninsularis (Anderson & Hanna,
1935), Paleomoera dyskritos, Dailey & Popenoe, 1966, and
a large undescribed parallelodontid have all been found in
beds of early Maastrichtian age.

ETYMOLOGY. From Madonna Hill on El Camino Real,
east of Carlsbad, San Diego Co., California.

Genus Christitys new genus

TYPE SPECIES. Christitys medica new species.

DIAGNOSIS. Compactly turriculate to pyriform gastro-
pods of small to moderate size with an angulate whorl pe-
riphery made nodose by short axial ribs. Aperture deltoid in
shape with a broad posterior sinus adapical to the whorl
periphery. Outer lip with posterior and medial denticula-
tions. Columella with a fold near the base of the previous
whorl and posterior and anterior denticulations.

DISCUSSION. Christitys resembles Murphitys in having
a columellar fold but is more sharply angulate about the
periphery and has fewer denticulations on the outer lip. The
outer lip denticulations are weaker in the geologically younger
C. martini than they are in C. delta or C. medica. Conversely,
the posterior denticulations of the inner lip are stronger in
C. martini than in C. delta. C. martini has a shorter spire
and more pyriform shape and resembles Heteroterma Gabb,
1869. Heteroterma differs from Christitys in lacking colu-
mellar folds. CominellaP. praecursor Wilckens, 1907, from
the late Senonian of southern Patagonia was placed in Het-
eroterma by Finlay and Marwick (1937), but the aperture is
not described (Wilckens, 1907, p. 1 17). The figures of C.?
praecursor are suggestive of Christitys. The lack of columellar
folds in Heteroterma is not considered by us to preclude a
close relationship to Christitys.

ETYMOLOGY. The generic name is a compound of
Christie (for J.M. Christie), and itys, Greek, rim or felly of
a wheel.

Christitys delta new species

Figures 8, 165-168

DIAGNOSIS. Compactly turriculate Christitys which have
a sharply angulate periphery.

DESCRIPTION. Shell small, compactly turriculate; spire
consisting of about four whorls and a smooth, mammiform
protoconch of one and a half whorls succeeded by a whorl
sculptured by raised spiral threads; first two whorls roundly
convex; others concave on the ramp, angulate at the periph-
ery, and abruptly contracted abapically.

Sculpture of spiral ribs narrower than the interspaces,
strongest on the angulate periphery, weakest on the ramp;
periphery made nodose by 14 very short axial ribs.

Outer lip with a shallow antispiral sinus adapical to the
periphery and moderate spiral antisinus anterior to the pe-
riphery; inner lip well demarked and of moderate and nearly
constant width. Outer lip with two tiny posterior denticles
and a moderately strong coalesced pair just anterior to the
peripheral angulation; inner lip with a moderate parietal den-
ticle, a fold on the columella just at the base of the previous
whorl and an anterior pseudofold.

HOLOTYPE. UCLA 39455.

28 Contributions in Science, Number 380

Popenoe and Saul: Perissitys

Figures 165-182. Species of Christitys new genus. All x 1.5 unless otherwise noted. Figures 165-168. Christitys delta new species, holotype,
Coniacian, UCLA 59455 from UCLA loc. 4209. Figures 169, 170, 172-177. Christitys medica new species, Santonian, all from UCLA loc.
4106; 169, 172, 173, holotype, LACMIP 7282; 170, paratype, LACMIP 7286, x 2, a juvenile; 174, paratype LACMIP 7284, a young specimen
with nearly complete outer lip; 175, 176, paratype LACMIP 7283; 177, paratype LACMIP 7285, x2. 171, 178-182. Christitys martini new
species, Campanian, all from USGS 5795, x 1; 171, paratype USNM 400976; 178-182, holotype, USNM 400975.

DIMENSIONS. Height 20.2 mm; diameter 14.9 mm;
height of spire 10.2 mm.

TYPE LOCALITY. UCLA loc. 4209, north side of Clover
Creek valley, Millville Quadrangle, Shasta Co., California
(area 1 1).

DISTRIBUTION. Known only from the type locality in
Member IV on Clover Creek (area 1 1).

GEOLOGIC AGE. Coniacian.

REMARKS. The above description is based upon the ho-
lotype and only known specimen. The spire is very well
preserved, but unfortunately the anterior canal is broken off.
This specimen is higher spired, more sharply angulate at the
periphery, and more coarsely sculptured than any specimen
assigned to C. medica. It differs from Perissitys cretacea and
Murphitys michae/i in having its suture very close to its an-
gulate whorl periphery, which produces a nearly triangular
spire.

ETYMOLOGY. The species is named delta, fourth letter
of the Greek alphabet, for its triangular spire.

Christitys medica new species

Figures 169, 170, 172-177

DIAGNOSIS. Compactly turriculate Christitys which have
an angulate periphery broad enough to display three to four
spiral ribs.

DESCRIPTION. Shell small, compactly turriculate; spire
consisting of about four whorls (protoconch not observed),
first whorl convexly rounded and spirally sculptured; second
whorl with flat ramp, spiral sculpture and angulate, noded
periphery; third and fourth whorls with slightly concave ramp;
angulate noded periphery abruptly contracted anteriorly,
produced into siphonal neck of moderate length.

Sculpture of spiral ribs narrower than the interspaces, three
to four strongest on the angulate periphery made nodose by
about 1 3 very short axial ribs.

Outer lip with a shallow antispiral sinus between the suture
and the periphery and moderate spiral antisinus anterior to
the periphery, slightly flared anterior to the periphery; inner
lip well demarked and of moderate and nearly constant width.
Outer lip with small denticle posterior to the periphery, a
strong bifid denticle just anterior to the periphery, abapically
a moderate denticle, and tiny denticles adjacent to the an-
terior canal; inner lip with a well-developed parietal denticle,
a fold opposite the periphery, and an anterior pseudofold,
all of apparently equal strength.

HOLOTYPE. LACMIP 7282.

PARATYPES. LACMIP 7283-7286, all from UCLA loc.
4106.

DIMENSIONS. Of holotype— height 21.8 mm (incom-
plete), diameter 15.2 mm; height of spire 9.3 mm (incom-
plete); of paratypes— LACMIP 7283, height 20.2 mm (in-

Contributions in Science, Number 380

Popenoe and Saul: Perissitys 29

complete), diameter 15.2 mm, height of spire 9.7 mm;
LACMIP 7286, height 8.6 mm, diameter 9.9 mm, height of
spire 4.2 mm.

TYPE LOCALITY. UCLA loc. 4106, hillside north of
Clover Creek, Millville Quadrangle, Shasta Co., California
(area 1 1).

DISTRIBLITION. All specimens are from the type locality
in Member VI on Clover Creek, Shasta Co., California.

GEOLOGIC AGE. Santonian.

REMARKS. None of the seven specimens is complete. C.
medica differs from C. delta in its slightly broader periphery
and in having the suture slightly more abapical to the pe-
riphery, giving the spire a slightly more turriculate shape.
Except for the stronger peripheral ribs, the ribbing is overall
of more even strength on C. medica. The denticulations of
the outer lip, especially the large bifid denticle, resemble those
of the “ Fusus ” kingii lineage (Figures 6, 1 4). In external shape
and sculpture the species is most similar to Santonian age
“ Hindsia nodulosa but C. medica has a shorter spire, ap-
pears to have had a shorter anterior canal, and has the large
bifid denticle within the outer lip rather than the two denticles
present in “//. nodulosa" (Figure 12).

ETYMOLOGY. The specific name is from Latin, medica.
clover from Media, for its occurrence on Clover Creek.

Christitys martini new species

Figures 171, 178-182

DIAGNOSIS. Pyriform Christitys which have a bluntly
angulate periphery defined by two spiral ribs.

DESCRIPTION. Shell of moderate size, apparently pyr-
iform; spire consisting of about four whorls (protoconch un-
known); last whorl with slightly concave ramp, bluntly an-
gulate periphery, contracted abruptly anterior to the periphery
to form an anterior canal of unknown length.

Sculpture of spiral riblets narrower than the interspaces
and two strong ribs defining anterior and posterior sides of
the periphery; periphery made nodose by about 1 2 very short
axial ribs.

Outer lip with a shallow antisiphonal sinus between the
suture and the anterior side of the periphery and slight spiral
antisinus anterior to the periphery, not flared; inner lip well
demarked, a little expanded posteriorly. Outer lip thickened
with denticle about midway along the ramp, a moderately
strong bifid? denticle at the anterior side of the periphery,
and abapically a moderate denticle; inner lip with a thick
tripartite parietal denticle, a fold opposite the periphery, and
an anterior pseudofold.

HQLOTYPE. USNM 400975.

PARATYPES. USNM 400976-400977 from USGS loc.
5795 and 400978 from USGS loc. 5796.

DIMENSIONS. Of holotype — height 27 mm (incom-
plete), diameter 25 mm, height of spire 8.5 mm.

TYPE LOCALITY. USGS loc. 5795, Whalers Creek, near
Chignik Lagoon, Alaska Peninsula, Alaska (area 23).

DISTRIBUTION. Lower member of Chignik Formation,
Inoceramus schmidti Zone near Chignik Bay, Alaska Pen-
insula.

GEOLOGIC AGE. Early Campanian. Ward et al. (1983)
place the Inoceramus schmidti Zone in the Santonian because
magnetic anomaly 33-34 occurs above I. schmidti in the
Great Valley Series of California. Jones (1963, p. 432), how-
ever, records Canadoceras newberryanum throughout the
Chignik Formation, and C. newberryanum occurs above
magnetic anomaly 33-34 on Chico Creek (area 5). The Chig-
nik Formation specimens may be of early Campanian age.

REMARKS. The species is based on four specimens from
the Chignik Formation listed by Martin (1926, p. 304) as
Perissolax brevirostris. These specimens are worn and bro-
ken, but the apertural characteristics are those of Christitys.
C. martini differs from C. medica in having a broader pe-
riphery, shorter spire, and more expanded inner lip; from
Perissitys brevirostris in having apertural decorations, and a
much less expanded inner lip; from P. elaphia in having a
fold on the columella, a deeper antispiral sinus to the growth
line, two rather than three or four strong spirals at the pe-
riphery, and a much less expanded inner lip. The strong outer
lip denticles of C. martini are at and abapical to the periphery
as in C. delta and C. medica rather than at and adapical to
the periphery as in P. cretacea and P. elaphia. C. martini
resembles Murphitys corona in shape, but has a more pro-
nounced antispiral sinus to the growth line; the outer lip is
not rimmed, and the inner lip is less expanded.

Christitys martini resembles Pseudoperissitys bicarinata
Nagao & Otatume, 1938 (pi. 4, fig. 3-5a), but P. bicarinata
lacks columellar folds and outer lip thickening. It is said to
be of Campanian or Maastrichtian age, and may be closely
related to Christitys. In shape C. martini also resembles Het-
eroterma trochoidea Gabb, but the latter is without apertural
ornament and has a thin inner lip, is from the Paleocene
“Martinez Stage,” and is considerably younger. The apertural
ornaments, which appear to be reduced in C. martini, could
have been lost by Paleocene time, and Heteroterma may be
related to Christitys. The relationships of these three appar-
ently similar genera need further study.

ETYMOLOGY. The species is named for G.C. Martin
who recorded the specimens from the Chignik Formation.

ACKNOWLEDGMENTS

The study of later Cretaceous formations and faunas of the
Pacific Coast, of which this report is a small part, has been
supported by grants-in-aid from the Penrose Fund of the
Geological Society of America, the National Science Foun-
dation, and the Academic Senate of the University of Cali-
fornia, Los Angeles. T. Susuki prepared part of the Creta-
ceous collections for study and provided some of the
photographic illustrations. Figures were drafted by J.
Guenther. Essential comparative material was loaned by D.R.
Lindberg, J.W. Durham, and the late J.H. Peck, University
of California, Berkeley; P.U. Rodda, California Academy of
Sciences, San Francisco; A.M. Keen, Stanford University;
E.C. Wilson, Los Angeles County Museum of Natural His-
tory; D.L. Jones, E.J. Moore, J.T. Smith, and J.G. Vedder,
U.S. Geological Survey, Menlo Park; H.G. Richards, Acad-
emy of Natural Sciences of Philadelphia; F.J. Collier, U.S.

30 Contributions in Science, Number 380

Popenoe and Saul: Perissitys

National Museum of Natural History; J.A. Jeletzky, Geo-
logical Survey of Canada; T.A. Demere, San Diego Society
of Natural History; and P.D. Ward, University of California,
Davis. Additional specimens were donated by J.M. Alderson
of California State University, Northridge; J.L. Loch of the
University of Southern California; Victor Miller of San Diego
State University; and G. Smith, University of California,
Santa Barbara. The Koll Company is commended for allow-
ing the San Diego Society of Natural History to make a
salvage collection of beautifully preserved fossils from the
Carlsbad Research Park site. Plaster casts of comparative
material from the Quiriquina beds of Chile were generously
provided by E. Perez d’A., Servicio Nacional de Geologia y
Mineria of Chile. T. Kase, National Science Museum, Tokyo,
Japan, has sent plaster casts of probable Japanese perissi-
tyids, including Surculitesl cf. S.l fusoides Nagao, 1939,
Pseudoperissitys bicarinata Nagao & Otatume, and Pseudo-
galeodea tricarinata Nagao, 1 932. Preprints of Berggren, Kent,
and Flynn (1985a) provided by Mark Filowicz of Union Oil
were of considerable assistance in adopting a time scale. This
paper has benefited greatly from the criticism of H. Tappan,
G.L. Kennedy, R.B. Saul, E.C. Wilson, W.J. Zinsmeister,
and P.U. Rodda. Remnant errors are the responsibility of
L.R. Saul.

LITERATURE CITED

Abbott, R.T. 1959. The family Vasidae in the Indo- Pacific.
Indo-Pacific Mollusca, vol. 1, no. 1, p. 20-403-20-473,
10 pi.

Anderson, F.M., and B. Martin. 1914. Neocene records in
the Temblor Basin, California, and Neocene deposits of
the San Juan District, San Luis Obispo County, Cali-
fornia. California Academy of Sciences, Proceedings, se-
ries 4, 4:15-122, pi. 1-10.

Arnold, R. 1909. Paleontology of the Coa/inga District,
Fresno and Kings Counties, California. United States
Geological Survey, Bulletin 396, 173 p., 30 pi.

Arnold, R., and R. Anderson. 1910. Geology > and oil re-
sources of the Coalinga District, California. United States
Geological Survey, Bulletin 398, 354 p., 52 pi., 9 text-
fig-

Berggren, W.A., D.V. Kent, and J.J. Flynn. 1985a. Paleo-
gene geochronology and chronostratigraphy. In Geo-
chronology and the geological record, ed. N.J. Snelling.
Geological Society of London, p. 141-186, Memoir 10.
Berggren, W.A., D.V. Kent, J.J. Flynn, and J.A. Van Cou-
vering. 1985b. Cenozoic geochronology. Geological
Society of America, Bulletin 96:1407-1418.

Berry, S.S. 1 963. Notices of new eastern Pacific Mollusca—
V. Leaflets in Malacology 1:139-146.

Coan, E.V. 1981. James Graham Cooper, pioneer western
naturalist. Moscow, Idaho: University Press of Idaho,
255 p.

Conrad, T.A. 1860. Descriptions of new species of Creta-
ceous and Eocene fossils of Mississippi and Alabama.
Academy of Natural Sciences of Philadelphia. Journal,
2nd ser., 4:275-298, 2 pi.

Cooper, J.G. 1896. On some new Cretaceous (and Eocene?)
Mollusca of California. California Academy of Sciences,
Proceedings, series 2, 6:330-337, pi. 47-48.

Cossmann, M. 1901. Essais de pa/eoconcho/ogie comparee.
Paris. Vol. 4, 293 p., 10 pi.

. 1 904. Essais de paleoconcho/ogie comparee. Paris.

Vol. 6, 151 p„ 9 pi.

Cox, R.L. 1960. Thoughts on the classification of the Gas-
tropoda. Malacological Society of London, Proceedings
33:239-261.

Erickson, J.M. 1974. Revision of the Gastropoda of the Fox
Hills Formation, Upper Cretaceous (Maestrichtian) of
North Dakota. Bulletins of American Paleontology 66:
131-253, pi. 14-20, 5 text-fig.

Finlay, H.J., and J. Marwick. 1937. The Wangaloan and
associated molluscan faunas of Kaitangata-Green Is-
land Subdivision. New Zealand Geological Survey, Pa-
laeontological Bulletin 15, 140 p., 18 pi.

Fischer, P. 1 880-1 887. Manuel de conchy/iologie et de pa-
leontologie conchyliologique. Paris: F. Savy, 1369 p., 23
pi., 1 1 38 text-fig.

Gabb, W.M. 1864. Description of the Cretaceous fossils.
California Geological Survey, Palaeontology 1:57-243
[1864], pi. 9-32 [1865],

. 1869. Cretaceous and Tertiary fossils. California

Geological Survey, Palaeontology v. 2, 299 p., 36 pi.

Grabau, A.W., and H.W. Shimer. 1909. North American
index fossils. New York: A.G. Seiler & Company, v. 1,
viii + 853 p., 1210 fig.

Haggart, J.W. 1984. Upper Cretaceous (Santonian-Cam-
panian) ammonite and inoceramid biostratigraphy of
the Chico Formation, California. Cretaceous Research
5:225-241, 8 fig.

. 1986. Stratigraphy of the Redding Formation of

north-central California and its bearing on Late Creta-
ceous Paleography. In Cretaceous stratigraphy western
North America, ed. P.L. Abbott, p. 161-178, 6 text-fig.
Society of Economic Paleontologists and Mineralogists,
Pacific Section.

Haggart, J.W. , and P.D. Ward. 1984. Late Cretaceous (San-
tonian-Campanian) stratigraphy of the northern Sac-
ramento Valley, California. Geological Society of Amer-
ica, Bulletin 95:618-627 , 11 fig.

Harland, W.B., A.V. Cox, P.G. Llewellyn, C.A.G. Pickton,
A.G. Smith, and R. Walters. 1982. A geologic time
scale. Cambridge University Press, xi + 131 p.

Hayami, I., and T. Kase. 1977. A systematic survey of the
Paleozoic and Mesozoic Gastropoda and Paleozoic Bi-
va/via from Japan. University of Tokyo, University Mu-
seum Bull. 13, 156 p., 1 1 pi., 2 fig.

Huey, A.S. 1948. Geology of the Tesla Quadrangle, Cali-
fornia. California Division of Mines, Bulletin 140, 75
p., 1 1 pi., 3 fig.

Jones, D.L. 1963. Upper Cretaceous (Campanian and
Maestrichtian) ammonites from southern Alaska. U.S.
Geological Survey, Professional Paper 432, iv + 53 p.,
41 pi., 25 text-fig., 2 tab.

Jones, D.L., W.V. Sliter, and W.P. Popenoe. 1978. Mid-

Contributions in Science, Number 380

Popenoe and Saul: Perissitys 3 1

Cretaceous (Albian to Turonian) biostratigraphy of
northern California. Museum d’Histoire Naturelle de
Nice, Annales, v. 4-1976, p. XXII.1-XXII.13, 2 pi., 7
fig., 1 tab.

Kent, D.V., and F.M. Gradstein. 1985. A Cretaceous and
Jurassic geochronology. Geological Society of America,
Bulletin 96:1419-1427.

Martin, G.C. 1926. The Mesozoic stratigraphy of Alaska.
U.S. Geological Survey, Bulletin 776, 493 p.

Matsumoto, T. 1 960. Upper Cretaceous ammonites of Cal-
ifornia. Part III. Kyushu University, Faculty of Science,
Memoirs, Ser. D, Geology, Special vol. 2, 204 p., 2 pi.,
20 text-fig.

Matsumoto, T. with T. Miyauchi in Part II. 1984. Some
ammonites from the Campanian (Upper Cretaceous) of
Northern Hokkaido. Palaeontological Society of Japan,
Special Papers, Number 27, 94 p., 31 pi., 13 text-fig.

Michael, R. 1899. Ueber Kreidefossilien von der lnsel Sa-
chalin. Kdniglich Preussischen geologischen Landesan-
stalt und Berkakademie zu Berlin 19:153-164, pi. 5-6.

Murphy, M.A., and P.U. Rodda. 1960. Mollusca of the
Cretaceous Bald Hills Formation of California. Journal
of Paleontology’ 34:835-858, pi. 101-107, 2 text-fig.

Nagao, T. 1932. Some Cretaceous Mollusca from Japanese
Saghalin and Hokkaido (Lamellibranchiata and Gas-
tropoda). Hokkaido Imperial University, Faculty of Sci-
ence, Journal, Ser. 4, Geology & Mineralogy 2:23-50,
pi. 5-8.

. 1939. Some molluscan fossils from the Cretaceous

deposits of Hokkaido and Japanese Saghalien. Part II.
Gastropoda. Hokkaido Imperial University, Faculty of
Science, Journal, Ser. 4, Geology & Mineralogy 4:215-
240, pi. 21-22.

Nagao, T., and K.-i. Otatume. 1938. Molluscan fossils of
the Hakobuti Sandstone of Hokkaido. Hokkaido Im-
perial University, Faculty of Science, Journal, Ser. 4,
Geology & Mineralogy 4:31-56, pi. 1-4, text-fig. 1.

Nelson, R.N. 1925. A contribution to the paleontology of
the Martinez Eocene of California. University of Cali-
fornia Publications, Department of Geological Sciences,
Bulletin 15:397-466, pi. 49-61.

Obradovich, J.D., and W.A. Cobban. 1975. A time-scale
for the Late Cretaceous of the Western Interior of North
America. Geological Association of Canada, Special Pa-
per 13, p. 31-54.

Palmer, A.R. 1983. The Decade of North American Ge-
ology 1983 geologic time scale. Geology 1 1:503-504.

Patterson, C.M. 1969. Chromosomes of molluscs. Marine
Biological Association of India, Symposium on Mollus-
ca, Proceedings, Part II, p. 639-689.

Ponder, W.F. 1973. The origin and evolution of the Neo-
gastropoda. Malacologia 12:295-338, 6 fig.

Popenoe, W.P. 1943. Cretaceous: East side Sacramento
Valley, Shasta and Butte Counties, California. American
Association of Petroleum Geologists, Bulletin 27:306-
312.

. 1954. Mesozoic formations and faunas, southern

California and northern Baja California. California Di-

vision of Mines, Bulletin 170, Chapter III, Historical
Geology, p. 15-21, fig. 1-4.

. 1957. The Cretaceous gastropod genus Biplica.

University of California Publications in Geological Sci-
ences 30:425-454, pi. 50-51, 1 text-fig., 1 map.

. 1971. Cophocara-Perissitys succession. Pacific Coast

Cretaceous. Geological Society of America, Abstracts
with Programs 3(2): 179-1 80.

. 1973. Southern California Cretaceous formations

and faunas with especial reference to the Simi Hills and
Santa Monica Mountains. In Cretaceous stratigraphy of
the Santa Monica Mountains and Simi Hills, Southern
California, ed. A.E. Fritsche, p. 15-29, 4 pi., 2 text-fig.
Society of Economic Paleontologists and Mineralogists,
Pacific Section, Geological Guidebook 1973 Fall Field
Trip.

. 1983. Cretaceous Aporrhaidae from California:

Aporrhainae and Arrhoginae. Journal of Paleontology
57:742-765, 6 fig.

Popenoe, W.P., L.R. Saul, and T. Susuki. 1987. Gyrodi-
form gastropods from the Pacific Coast Cretaceous and
Paleocene. Journal of Paleontology. 60:70-100, 7 fig.

Rafinesque, C.S. 1815. Analyses de la nature ou tableau de
I’ universe et des corps organises. Palermo, 224 p.

Roding, P.F. 1798. Museum Boltenianum . . . : Pars se-
en nda continens Conchylia .... Hamburg (J.C. Trap-
pii), vii + 109 p.

Saul, L.R. 1971 “Hindsia nodulosa (Whiteaves)”: Another
Upper Cretaceous gastropod lineage? Geological Society
of America, Abstracts with Programs 3(2): 189-1 90.

. 1974. Described or figured West Coast species of

Cymbophora. Journal of Paleontology 48:1068-1095, 3
pi., 7 text-fig.

. 1978. The North Pacific Cretaceous trigoniid genus

Yaadia. University of California Publications, Geolog-
ical Sciences, v. 119, 65 p., 12 pi., 20 text-fig., 12 tab.

. 1983. Turritella zonation across the Cretaceous-

Tertiary boundary, California. University of California
Publications, Geological Sciences, v. 125, 164 p., 7 pi.,
27 text-fig.

. 1986. Mollusks of latest Cretaceous and Paleocene

age. Lake Nacimiento, California. In Geology of Upper
Cretaceous and Lower Tertiary rocks near Lake Naci-
miento, California, ed. K. Grove and S.A. Graham, p.
25-32, 60 fig. Society of Economic Paleontologists and
Mineralogists, Pacific Section, Book 49.

Saul, L.R. , and J.M. Alderson. 1981. Late Cretaceous Mol-
lusca of the Simi Hills: An introduction. In Simi Hills
Cretaceous turbidites, southern California, ed. M.H. Link,
R.L. Squires, and I.P. Colburn, p. 29-41, 3 pi., 1 fig.
Society of Economic Paleontologists and Mineralogists,
Pacific Section, Pall 198 1 Field Trip Volume and Guide-
book.

Schenck, H.G., and A.M. Keen. 1 940. California fossils for
the field geologist. Stanford, California, privately pub-
lished and Stanford Univ. Press, 86 p., 56 pi.

Schmidt, M.F. 1873. Uber die Petrefakten der Kreidefor-
mation von der lnsel Sacha/in. Academie Imperiale des

32 Contributions in Science, Number 380

Popenoe and Saul: Perissitys

Sciences de Saint Petersbourg, Memoires, ser. 7, v. 1 9(3):
1-33, pi. 1-8.

Shuto, T. 1974. Larval ecology of prosobranch gastropods
and its bearing on biogeography and paleontology. Le-
thaia 7:239-256, 8 fig.

Smith, J.T. 1975. Age, correlation, and possible Tethyan
affinities of mollusks from the Lodo Formation of Fresno
County, California. In Conference on future energy ho-
rizons of the Pacific Coast: Paleogene symposium and
other selected technical papers, ed. D.W. Weaver, G.R.
Homaday, and A. Tipton, p. 464-483, 2 pi., Pacific
Sections, American Association of Petroleum Geolo-
gists, Society of Economic Paleontologists and Miner-
alogists, and Society of Exploration Geologists, Long
Beach, California.

Sohl, N.F. 1960. Archaeogastropoda, Mesogastropoda, and
stratigraphy of the Ripley, Owl Creek, and Prairie Blujf
Formations. U.S. Geological Survey, Professional Paper
331-A, 152 p., 18 pi., 11 fig., 1 tab.

. 1964. Neogastropoda, Opisthobranchia, and Ba-

sommatophora from the Ripley, Owl Creek, and Prairie
Bluff Formations. U.S. Geological Survey, Professional
Paper 331-B, p. iv + 153-344, pi. 19-52, fig. 12-18, 2
tab.

. 1967. Upper Cretaceous gastropod assemblages of

the Western Interior of the United States. In A sympo-
sium on paleoenvironments of the Cretaceous seaway in
the Western Interior, eds. E.G. Kauffman and H.C. Kent,
p. 1-37, Golden, Colorado, Colorado School of Mines,
Department of Publications.

Stanley, S.M. 1985. Rates of evolution. Paleobiology > 11:
13-26.

Stanton, T.W. 1896. The faunal relations of the Eocene
and Upper Cretaceous on the Pacific Coast. U.S. Geo-
logical Survey, 1 7th Annual Report, pt. l,p. 1011-1060,
pi. 63-67.

Stewart, R.B. 1927. Gabb's California fossil type gastro-
pods. Academy of Natural Sciences of Philadelphia, Pro-
ceedings 78(1926):287-447, pi. 20-32.

Sundberg, F.A., and B.O. Riney. 1984. Preliminary report
on the Upper Cretaceous macro-invertebrate faunas near
Carlsbad, California. In Upper Cretaceous depositional
systems Southern California-Northern Baja California,
ed. P.L. Abbott, p. 103-107, 3 fig. Society of Economic
Paleontologists and Mineralogists, Pacific Section. Vol-
ume and Guidebook 36, for Meetings in San Diego.

Taylor, D.W. , and N.F. Sohl. 1962. An outline of gastropod
classification. Malacologia 1:7-32.

Taylor, J.D., N.J. Morris, and C.N. Taylor. 1980. Food
specialization and the evolution of predatory proso-
branch gastropods. Palaeontology’ 23:375-409.

Thiele, J. 1929-1931. Handbuch der Systematischen
Weichtierkunde, vol. 1. Jena: Gustav Fischer, 111 p.,

783 text-fig. (1-376, 470 text-fig., 1929; 377-779, 313
text-fig., 1931).

Trujillo, E.F. 1960. Upper Cretaceous Foraminifera from
near Redding, Shasta County, California. Journal of Pa-
leontology 34:290-346, pi. 43-50, 3 text-fig.

Tryon, G.W., Jr. 1881. Manual of concho/ogy.v ol. 3. Phila-
delphia: by the author, 310 p., 87 pi.

. 1883. Structural and systematic conchology, vol.

2. Philadelphia: by the author, 430 p., pi. 23-91.

Vermeij, G.J. 1977. The Mesozoic marine revolution: Evi-
dence from snails, predators and grazers. Paleobiology
3:245-258.

Ward, P.D., K.L. Verosub, and J.W. Haggart. 1983. Marine
magnetic anomaly 33-34 identified in the Upper Cre-
taceous of the Great Valley Sequence of California. Ge-
ology 1 1 :90-93.

Weaver, C. A. 1912. Preliminary report on the Tertiary pa-
leontology’ of western Washington. Washington Geolog-
ical Survey, Bulletin 15, p. 1-80, pi. 1-15.

Webster, M.L. 1983. New species of Xenophora and An-
chura (Mollusca: Gastropoda) from the Cretaceous of
Baja California Norte, Mexico. Journal of Paleontology
57:1090-1097, 3 fig.

Wenz, W. 1938-1944. Gastropoda. In Handbuch der Pa-
laozoologie, ed. O.H. Schindewolf, Band 6, 1639 p., 43 1 1
fig-

Whiteaves, J.F. 1874. Notes on the Cretaceous fossils col-
lected by Mr. James Richardson at Vancouver and the
adjacent islands. Canada Geological Survey, Report of
Progress 1873-1874, p. 260-268, pi. of fossils.

. 1879. On the fossils of the Cretaceous rocks of Van-
couver and adjacent islands in the Strait of Georgia.
Canada Geological Survey, Mesozoic Fossils 1:93-190,
pi. 1 1-20.

. 1903. On some additional fossils from the Vancou-
ver Cretaceous, with a revised list of species therefrom.
Canada Geological Survey, Mesozoic Fossils 1 : 309-4 1 5,
pi. 40-51.

Wilckens, O. 1904. Revision der fauna der Quiriquina-
Schichten. Neues Jahrbuch fur Mineralogie, Geologie,
und Palaeontologie 18 (supplement): 1 8 1-284, tafel 17-
20.

. 1907. Die Lamellibranchiaten, Gastropoden, etc.

der obern Kreide Sudpatagoniens. Naturforschenden Ge-
sellschaft zu Freiburg im Breisgau, Berichte 15:97-166,
tafel 2-9.

Wood, W. 1828. Supplement to the Index Testaceologicus:
Or a catalogue of shells, British and foreign. London,
vi + 59 p., 8 pi.

Zinsmeister, W.J. 1983. New late Paleocene molluscs from
the Simi Hills, Ventura County, California. Journal of
Paleontology 57:1282-1304, 4 fig.

Submitted 24 January 1986; accepted 6 October 1986.

Contributions in Science, Number 380

Popenoe and Saul: Perissitys 33

LOCALITIES CITED

Approximately one-third of the cited localities have been
previously published; these are cited briefly with a reference.
The italicized number to the left refers the locality to an area
of Figure 1.

13

13

7

8

8

13

11

5

5

11

13

8

83 CIT (= UCLA 2949): Fine ss just above shale,
section line fence gate on old road, lA mi. W of
Schultz Ranch, S side of Williams Canyon, ap-
prox. 435 ft. S of NW cor. sec. 20, T5S, R7W,
El Toro Quad., Santa Ana Mts., Orange Co.,
Calif. Coll.: B.N. Moore, 1926. Ladd Forma-
tion, uppermost Holz Shale Member. Mid
Campanian, Turritella chicoensis Zone.

94 CIT: Sandstone just above shale about 300 ft.
E of section line on ridge N of Williams Can-
yon, 500 ft. N, 300 ft. E of SW cor. sec. 17,
T5S, R7W, El Toro Quad., Santa Ana Mts.,
Orange Co., Calif. Coll.: B.N. Moore, 1928.
Ladd Formation, uppermost Holz Shale Mem-
ber. Mid Campanian, Turritella chicoensis
Zone.

249 UCB: % mi. W of Muir Station, 'A mi. N of
Franklin Creek, Concord Quad., Contra Costa
Co., Calif. Coll.: J.C. Merriam. Great Valley
Series. ?Late Maastrichtian.

460 LSJU: 300 ft. S, 650 ft. W of NE cor. sec. 24,
T14S, R1 IE, Chounet Ranch Quad., Panoche
Hills, Fresno Co., Calif. Basal Laguna Seca For-
mation, 300 ft. below base of Domengine. ?Late
Danian.

463 CAS (= CAS 2362): Center E side sec. 4, T20S,
R14E, north side Los Gatos Creek, Coalinga
Quad., Fresno Co., Calif. Coll.: F.M. Ander-
son. Panoche Formation, “Ragged Valley Shale
Member.” Early Maastrichtian.

974 CIT: Aliso-Santiago Creek divide, El Toro
Quad., Orange Co., Calif. Williams Formation,
Pleasants Sandstone Member. Late Campani-
an. (Matsumoto, 1960, p. 99.)

1007 CIT: Oak Run, Millville Quad., Shasta Co.,
Calif. Member IV. Coniacian. (Matsumoto,
1960, p. 100.)

1016 CIT: Chico Creek, Paradise Quad., Butte Co.,
Calif. Chico Formation, Musty Buck Member.
Santonian. (Matsumoto, 1960, p. 101.)

1017 CIT: Chico Creek, Paradise Quad., Butte Co.,
Calif. Chico Formation, Musty Buck Member.
Santonian. (Matsumoto, 1960, p. 101.)

1034 CIT: See UCLA 4104.

1053 CIT (= UCLA 4191): N of Santiago Creek, El
Toro Quad., Orange Co., Calif. Ladd Forma-
tion, upper Holz Shale Member. Early Cam-
panian. (Matsumoto, 1960, p. 102.)

1068 LSJU: S of Chaney Ranch Canyon, WV2, SW'/i
sec. 19, T14S, R12E, Chounet Ranch Quad.,
Panoche Hills, Fresno Co., Calif. Coll.: R.T.

17

17

11

14

11

11

11

11

11

7

11

White. Laguna Seca Formation. ?Late Danian
or ?early Selandian.

1 1 58 CIT (= UCLA 6020): SE slope of Simi Hills,
N bank Bell Canyon, 1.15 mi. due W of Los
Angeles-Ventura Co. line on boundary (ex-
tended) between T1 and 2N, Calabasas Quad.,
Ventura Co., Calif. Coll.: W.P. Popenoe, June
18, 1935. Chatsworth Formation. Late mid
Campanian.

1 1 59 CIT: Prominent fossil bed on crest of spur be-
tween forks of Dayton Canyon, about 400 ft.
E of Los Angeles-Ventura Co. line, approx. 400
ft. N, 2350 ft. W of SE cor. sec. 28, T2N, R 1 7 W,
Calabasas Quad., Los Angeles Co., Calif.
Chatsworth Formation. Late Campanian. (Po-
penoe et al., 1987, p. 98.)

1212 CIT: Little Cow Creek, Millville Quad., Shasta
Co., Calif. Frazier Siltstone (= Member II),
hard sandy concretions, in shale. Late Tura-
nian. (Popenoe, 1983, p. 765.)

1215 LACMIP: Mollusks and echinoid from Arroyo
Santa Catarina, approx. 6 km N, 6 km E of
mouth of Rio Santa Catarina, Baja California,
Mexico. Coll.: H. Dushane, July, 1958. Rosario
Formation. Early Maastrichtian.

1227 CIT (= UCLA 3296): Massive sandstone in-
terbedded with conglomerate on S side of high
E-W trending ridge, S side Oak Run Valley,
3275 ft. S 54°50'W of SE cor. sec. 10, T32N,
R2W, Millville Quad., Shasta Co., Calif. Coll.:
Popenoe & Ahlroth, 1936. Member V. Early
Santonian.

1 230 CIT : SE side Oak Run valley, SE of Dry Creek,
S side of hill on section line between sections
1 1 & 14, 200 ft. S, 1750 ft. E of NW cor. sec.
14, T32N, R2W, Millville Quad., Shasta Co.,
California. Coll.: Popenoe & Ahlroth, 1936.
Member V. Coniacian.

1 232 CIT: Price Hollow, Millville Quad., Shasta Co.,
Calif. Member V, cross-bedded sandstones in-
terbedded with thick conglomerate. Early San-
tonian. (Saul, 1978, p. 56.)

1246 CIT: Clover Creek, loose boulder on hillslope
on E side of 1000 ft. hill, about 0.3 mi. NE of
Reinicke’s Ranch house, SEVA, NEVA sec. 13,
T32N, R2W, Millville Quad., Shasta Co., Cal-
if. Coll.: Popenoe, 1936. Member V. Santoni-
an.

1247 CIT: N side Clover Creek, Millville Quad.,
Shasta Co., Calif. Member V. Santonian. (Mat-
sumoto, 1960, p. 104.)

1258 USGS: Near Blum and Company’s Pacheco
warehouse, about 1 V2 mi. N of Pacheco, Contra
Costa Co., Calif. Coll.: T.W. Stanton, 1894.
Great Valley Series. Late Maastrichtian.

1289 CIT: Bear Creek, at stone culvert on road to
Aldridge’s, SW of Bonnie Crags, about Vi mi.

34 Contributions in Science, Number 380

Popenoe and Saul: Perissitys

18

18

11

19

19

11

7

18

8

13

W of Aldridge’s and 1.3 mi. E of road forks
and 2 mi. E of road crossing of N fork Bear
Creek, 2000 ft. S, 500 ft. E ofNW cor. sec. 18,
T3 1 N, R 1 E, Whitmore Quad., Shasta Co., Cal-
if. Coll.: W.P. Popenoe & EEV. Church, August
17, 1936. Member V. Coniacian.

1 400 CIT : Sucia Island, Orcas Island Quad. , San J uan
Co., Wash. Cedar District Formation. Mid
Campanian, Hoplitoplacenticeras Zone. (Mat-
sumoto, 1960, p. 107.)

1402 CIT: Float from base of cliffs on S side of Fossil
Cove near W end, Sucia Island, sec. 26, T38N,
R2W, Orcas Island Quad., San Juan Co., Wash.
Cedar District Formation. Mid Campanian.

1532 CIT: Salt Creek, Millville Quad., Shasta Co.,
Calif. Frazier Siltstone (= Member II), near
top. Turonian. (Matsumoto, 1960, p. 107.)

1581 UCLA: N side East Fork Fish Canyon, Warm
Springs Mountain Quad., NW San Gabriel Mts.,
Los Angeles Co., Calif. San Francisquito For-
mation. Late Danian. (Saul, 1983, p. 1 12.)

1 594 UCLA: N side Warm Springs Mt., Warm
Springs Mountain Quad., Los Angeles Co.,
Calif. San Francisquito Formation. Late late
Maastrichtian. (Saul, 1983, p. 113.)

1596 CIT: Forks of Snow and North Bear creeks,
2750 ft. S, 1250 ft. E of NW cor. sec. 7, T32N,
R1E, Whitmore Quad., Shasta Co., Calif. Coll.:
W.P. Popenoe & W.M. Tovel, September 10,
1941. Member V. Coniacian.

1602 CIT: NW side of crest of ridge just W of prom-
inent bend in highway between Martinez and
Crockett, 500 ft. NW of CIT 1601, approx. 1.8
mi. N 75° W of shore end of ferry pier at Mar-
tinez, and S of Benicia city wharf, Benicia Quad.,
Contra Costa Co., Calif. Coll.: W.P. Popenoe,
Nov. 14, 1944. Great Valley Series, upper
sandstone just below Martinez. Late Maas-
trichtian.

1860 LSJU: Southend of Sucia Island also northwest
islet of Sucia Group, Puget Sound, Orcas Island
Quad., San Juan Co., Wash. Coll.: F.L. Cole,
1893. Cedar District Formation. ?Mid Cam-
panian.

2245 LSJU: Between Marca Canyon and Chaney
Ranch Canyon, 100 ft. S, 700 ft. W of NE cor.
sec. 24, T14S, R11E, Chounet Ranch Quad.,
Panoche Hills, Fresno Co., Calif. Coll.: R.A.C.
Brown, S. Muller, M. Payne & H.G. Schenck,
1940. Basal Laguna Seca Sand, 300 ft. below
base of Domengine. ?Late Danian or ?early Se-
landian.

2324 UCLA: Just E of Santiago Creek, S of Williams
Canyon and N of Modjeska Canyon, limey ss
at top of Holz exposed in abandoned road cut
in large gully, 4200 ft. N 70° W of mouth of

13 2415

14 2852

20 3162A

8 A-3216
8 A-3262

14 3268

11 3298

5 3623

5 3624

Harding Canyon, NW14, NW'A sec. 29, T5S,
R7W, El Toro Quad., Santa Ana Mts., Orange
Co., Calif. Coll.: W.P. Popenoe, 1946. Ladd
Formation, uppermost Holz Shale Member.
Mid Campanian.

UCLA: Spur NW side Bee Canyon, El Toro
Quad., Orange Co., Calif. Williams Formation,
Pleasants Sandstone Member. (Saul, 1978, p.
56.)

LACMIP: West side Arroyo Santa Catarina,
Estado de Baja California, Mexico. Rosario
Formation. Early Maastrichtian. (Webster,
1983, p. 1096.)

SDSNH: Carlsbad Research Center, gently dip-
ping (3-5° SW), locally faulted sequence of
blue-gray sandy claystone with interbedded,
well-cemented sandstones exposed during de-
velopment of industrial park, fossils collected
through a 70+ ft. section, stratigraphic hori-
zons denoted by letter “A” being lowest, sec-
tion was best exposed along Palmer Way ap-
prox. 2000 ft. W of intersection with El Camino
Real (no longer accessible), 33°8'2"N lat.,
1 17°16'41"Wlong., San Luis Rey Quad., Carls-
bad, San Diego Co., Calif. Coll.: B.O. Riney,
May 1982, with permission of Koll Company.
Rosario Formation, Point Loma Member. Ear-
ly Maastrichtian.

UCB: In banks along Los Banos Creek, SE’A,
SE'/4, NE'/4 sec. 12, T1 IS, R9E, Volta Quad.,
Merced Co., Calif. Moreno Formation, Volta
Sands. Mid Maastrichtian.

UCB: W branch of small draw a short distance
N of large unnamed gully about 3.5 mi. S of
Ortigalita Creek near intersection of S border
sec. 1 and meridian 120°50'W, approx. 700 ft.
E of SW cor. sec. 1, T12S, R10E, Charleston
School Quad., Merced Co., Calif. Coll.: A. Ben-
nison, December, 1939. Laguna Seca Forma-
tion. Danian.

UCLA: Olive silts exposed along banks of small
gully about 3 mi. NE ofSanta Catarina Landing
and ‘A mi. S of the channel of Arroyo Santa
Catarina, 5.9 km N, 5.9 km E ofSanta Catarina
stream at Pacific Ocean, N 18° W of Punta
Canoas, Baja California, Mexico. Coll.: W.P.
Popenoe & party, February, 1949. Rosario
Formation. Early Maastrichtian.

UCLA (= CIT 1006): Between Basin Hollow
and Clover creeks, Millville Quad., Shasta Co.,
Calif. Member V. Santonian. (Popenoe, 1957,
p. 448.)

UCLA: Chico Creek, Paradise Quad., Butte Co.,
Calif. Chico Formation, Musty Buck Member.
Early Santonian. (Matsumoto, 1960, p. 155.)
UCLA: First ravine to S of Mickey’s Place on
W side of Chico Creek about 500 ft. upstream

Contributions in Science, Number 380

Popenoe and Saul: Perissitys 35

5

5

5

5

5

5

17

2

11

11

11

from UCLA 3622 and 100 ft. below fork in
ravine, 800 ft. N, 900 ft. E of SW cor. sec. 1,
T23N, R2E, Paradise Quad., Butte Co., Calif.
Coll.: L.R. & R.B. Saul, 1952. Chico Forma-
tion, Musty Buck Member. Early Santonian.

3633 UCLA: E of Chico Creek county road. Paradise
Quad., Butte Co., Calif. Chico Formation, top
of Musty Buck Member. ?Late Santonian, Bac-
ulites capensis Zone. (Matsumoto, 1960, p. 15,
156.)

3635 UCLA: On E bank of Chico Creek W from

House and approx. 400 ft. S of twin meadows,
1 800 ft. S, 400 ft. E of NW cor. sec. 1 3, T23N,
R2E, Paradise Quad., Butte Co., Calif. Coll.:
L.R. & R.B. Saul, 1952. Chico Formation, bas-
al Ten Mile Member. Early Campanian.

3637 UCLA: Chico Creek, Paradise Quad., Butte Co.,
Calif. Chico Formation, Ten Mile Member.
Early Campanian. (Matsumoto, 1960, p. 15,
156.)

364 1 UCLA: Chico Creek, Paradise Quad., Butte Co.,
Calif. Chico Formation, Ten Mile Member.
Early Campanian. (Saul, 1983, p. 121.)

3643 UCLA: Chico Creek, Paradise Quad., Butte Co.,
Calif. Chico Formation, middle Ten Mile
Member. Early Campanian zones of Submor-
toniceras chicoense and Turrit ell a chicoensis
holzana. (Saul, 1974, p. 1093.)

3647 UCLA: E side Chico Creek up ravine which is
1.7 mi. from Ten Mile House site by Chico
Creek county road, leached fossils in ss approx.
150 ft. up ravine from county road and near
large block of basalt, 1750 ft. S, 2750 ft. E of
NW cor. sec. 35, T23N, R2E, Paradise Quad.,
Butte Co., Calif. Coll.: L.R. & R.B. Saul, 1953.
Chico Formation, Ten Mile Member. Late ear-
ly Campanian zones of Submortoniceras chi-
coense and Turrit el la chicoensis.

3814 UCLA: Dark gray calcareous ss stringers in buff
ss in gully channel 250 ft. N of private road on
Lang Ranch, 2800 ft. N, 175 ft. W of SE cor.
sec. 26, T2N, R19W, Thousand Oaks Quad.,
NW end Simi Hills, Ventura Co., Calif. Coll.:
D.H. Dailey, 1958. Chatsworth Formation.
Early Maastrichtian.

4082 UCLA (= CIT 1038): Tuscan Springs, Tuscan
Springs Quad., Tehama Co., Calif. Chico For-
mation. Early Campanian. (Saul, 1978, p. 57.)

4104 UCLA (= CIT 1034): Hard, cemented ss slabs
weathering out of siltstone in pasture a little W
of N of Hathaway Ranch house, about 1 500
ft. S, 1400 ft. E of NW cor. sec. 16, T32N,
R2W, Millville Quad., Shasta Co., Calif. Coll.:
W.P. Popenoe, 1959. Member IV. Coniacian.

4106 UCLA: Clover Creek, Millville Quad., Shasta
Co., Calif. Member VI. Early Santonian. (Po-
penoe, 1983, p. 760, 765.)

4107 UCLA: Ss in bed of South Cow Creek, about

13 4207

11 4209

11 4217

20 4898

10 B-5323

10 5431

23 5795

23 5796

8 5902

4400 ft. W of Hunt Ranch House and just
downstream from old ruined iron bridge over
creek, 1700 ft. S, 2100 ft. W of NE cor. sec.
17, T31N, R2W, Millville Quad., Shasta Co.,
Calif. Coll.: W.P. Popenoe, 1954. Member V.
Santonian.

UCLA (= CIT 976): S side Williams Canyon,
El Toro Quad., Santa Ana Mts., Orange Co.,
Calif. Williams Formation, lower Pleasants
Sandstone Member. Late Campanian. (Mat-
sumoto, 1960, p. 99.)

UCLA: N side of Clover Creek valley along
SW slope of NE trending hill, 1300 ft. N, 500
ft. W ofSE cor. sec. 31,T32N, R2W, Millville
Quad., Shasta Co., Calif. Coll.: W.P. Popenoe,
1954. Member IV, siltstone with interbedded
fossiliferous concretionary sandstone slabs.
Coniacian.

UCLA: Clover Creek Millville Quad., Shasta
Co., Calif. Member VI. Late Santonian. (Po-
penoe et al., 1987, p. 99.)

LACMIP: Near base of east-facing road cut
opposite Madonna Hill Guest Home, 5392 El
Camino Real, 0.85 mi. N of Palomar Airport
Road, San Luis Rey Quad., San Diego Co.,
Calif. Coll.: G.L., C.B. & L.M. Kennedy & W.
& L. Hindes, 1 973. Rosario Group, Point Loma
Formation. Early Maastrichtian.

UCB: S side Punta San Jose, next promontory
eastward from one on which old fish camp is
located and at end of road on point; promon-
tory at UCB B-5322 blocks travel along beach
except at lowest tides, approx. 2 m strat. below
UCB B-5322 and farther E along beach, Baja
California, Mexico. Coll.: E.C. Allison & F.H.
Kilmer, 1957. Rosario Formation. ?Early
Maastrichtian.

UCLA: Siltstone and sandstone exposed in lit-
toral zone and 10 to 20 ft. above high tide, S
side Punta San Jose, about Vi mi. E of the point
and 30 mi. airline S of Ensenada, Pacific coast
of Baja California, Mexico. Coll.: W.P. Popen-
oe & W.V. Sliter, 1965. Rosario Formation.
Early Maastrichtian.

USGS: Whalers Creek, about 2 mi. from Chig-
nik Lagoon. Series of sandstones about 100 ft.
below the coal. Alaska Peninsula. Coll.: W.W.
Atwood, 1908. Chignik Formation, Lower
Member (Martin, 1926, p. 304). Campanian,
Inoceramus schmidti Zone.

USGS: North side of Chignik Bay 1 mi. N of
base of sand spit, Alaska Peninsula. Coll.: W.W.
Atwood, 1908. Chignik Formation, Lower
Member (Martin, 1926, p. 304). Early Cam-
panian, Inoceramus schmidti Zone.

UCLA: In gullies tributary to Oil Canyon at its
source, just W of center of sec. 7, T 1 9S, R 1 5E,
Joaquin Rocks Quad., Fresno Co., Calif. Coll.:

36 Contributions in Science, Number 380

Popenoe and Saul: Perissitys

11 5990

14 6368

8 6496

14 6534

8 A-6618

17 6670

18 7003

8 7059

W.P. Popenoe, 1949. Panoche Formation,
“Ragged Valley Shale,” about 200 ft. above
contact with Joaquin Rocks Member. Early
Maastrichtian, Pachydiscus ( N codes m oceras )
catarinae Zone and Goudkoff’s E Zone.
UCLA: Sandstone cropping out in bed of small
NW-ward-flowing gully tributary to French
Creek, near S end of Swede Basin, 300 ft. S,
1800 ft. E of NW cor. sec. 9, T33N, R2W,
Millville Quad., Shasta Co., Calif. Coll.: W.P.
Popenoe, 1959. Member IV. Coniacian.
Blocks fallen from bluffs, first large point N of
Punta Canoas, about 3 mi. S of Santa Catarina
Landing, Baja California, Mexico. Coll.: Weir,
Gugnian & Stager, Feb. 14, 1948. Sepultura
Formation. Early Selandian.

UCLA: Approx. I'A mi. SW of Howard Ranch
and V2 mi. N of Romero Creek, 2700 ft. N,
2160 ft. W of SE cor. sec. 21, T9S, R8E, Pa-
checo Pass Quad., Merced Co., Calif. Coll.: F.A.
Schilling, Jr., 1958. Panoche Formation, boul-
der in conglomerate 8200 ft. below top. San-
tonian.

UCLA: S side of short cul de sac having prom-
inent bare vertical outcrop on N side, W side
Arroyo Santa Catarina almost opposite road
into Arroyo, approx. 8.1 km N, 4.2 km E of
mouth of stream NW side Arroyo Santa Ca-
tarina, N 19° W of S tip Punta Canoas, Baja
California, Mexico. Coll.: L.R. Saul, R.L. Saul
& W.P. Popenoe, 1968. Rosario Formation,
buff silts below gritty sands with fossils, cobble
conglomerate above. Early Maastrichtian.
UCB: Moreno Grande concretionary sand-
stone along road in S bank of Ortigalita Creek
in SW'/4, NE'/4, NW‘/4 sec. 28, T11S, R10E,
Ortigalita Peak Quad., Merced Co., Calif. Coll.:
L.I. Briggs, 1948-49. Moreno Formation. Ear-
ly Maastrichtian.

UCR: E side Meier Canyon, 3030 ft. SW 20°
from hill 1314, Calabasas Quad., Simi Hills,
Ventura Co., Calif. Coll.: W.J. Zinsmeister,
1974. Santa Susana Formation. Early “Mar-
tinez” probably = early Selandian.

UCLA: East end of Little Sucia (SW islet of
Sucia Is.) approx. 200 ft. S from N side of islet
and at base of northern E-facing cliff outcrop,
sec. 26, T38N, R2W, Orcas Island Quad., San
Juan Co., Wash. Coll.: R.B. Saul, 1972. Cedar
District Formation, ? just below Hoplitopla-
centiceras. Mid Campanian.

USGS: Foothills between Ortigalita and Little
Panoche creeks, about 4 mi. SE of mouth of
Ortigalita Creek, on hilltop in NW'/i, NE‘/4 sec.
12, T12S, R10E, about 1500 ft. SW of 875 ft.
hill, Ortigalita Peak Quad., Merced Co., Calif.

Coll.: R.W. Pack & R. Anderson, 1910. More-
no Formation, within 100 ft. of base of lower
gray sandstone. Mid Maastrichtian.

16 7110 UCLA: N side of Garapito Creek just above

1300 ft. contour, 900 ft. N, 725 ft. E of SW
cor. sec. 33 (proj.), TIN, R16W, San Vicente
y Santa Monica Grant, Topanga Quad., Santa
Monica Mts., Los Angeles Co., Calif. Coll.: J.M.
Alderson, 1983. Tuna Canyon Formation, up-
per sandstone. Late Campanian, zones of
Metaplacenticeras pacificum and Turritella
chicoensis pescaderoensis.

22 7137 UCLA: Fossiliferous, hard calcareous concre-

tion from ss beds in grabenf?) (Coralliochama
beds in sea cliff on both sides of graben?), about
200 ft. eastward along shore from sandy ravine
with road to beach, N side Punta Banda Pen-
insula, Baja California, Mexico. Coll.: J.M. Al-
derson, 1984. Rosario Formation. Early Maas-
trichtian.

12 7149 UCLA: Fossils at beach level from sea cliff ex-

posure approx. 2 km S of San Antonio del Mar,
Baja California, Mexico. Coll.: Victor Miller,
1984. Rosario Formation. Early Maastrich-
tian.

20 7792 LACMIP: 5 m from base of a temporary cut

bank which has exposed 17 m of section, N
side Faraday Ave., E of intersection with Ruth-
erford Road, approx. 3570 ft. N, 9730 ft. W of
SE cor. San Luis Rey Quad., Carlsbad Research
Park, Agua Hedionda Grant, San Diego Co.,
Calif. Coll.: James Loch, spring 1984. Rosario
Formation, Point Loma Member. Early Maas-
trichtian.

20 7962 LACMIP: Carlsbad Research Park, cut bank

N side of Faraday Ave. W of intersection with
Rutherford Road, W of fault of unknown offset,
approx. 3575 ft. N, 9770 ft. W of SE cor. San
Luis Rey Quad., Agua Hedionda Grant, San
Diego Co., Calif. Coll.: James Loch, spring
1984. Rosario Formation, Point Loma Mem-
ber. Early Maastrichtian.

24 8159 LACMIP (= CIT 1824): Shales in canyon back

of mine office of Gladding McBean Co. North
Mine. This is the foot wall of fault by mine,
N'/2, NE'/4 sec. 1, T8S, R7W, Canada Gober-
nadora Quad., Orange Co., Calif. Coll.: C.R.
Stauffer, 1945. Williams Formation, Pleasants
Sandstone Member. ?Late Campanian-?early
Maastrichtian.

21 31325 CAS: Head of Buckeye Creek, north branch of

creek SE of Lee Bow Well, sec. 22, T12N, R3W,
Rumsey Quad., Rumsey Hills, Yolo Co., Calif.
Basal Forbes?, float from lower conglomerate.
Late Santonian or early Campanian.

Contributions in Science, Number 380

Popenoe and Saul: Perissitys 37

11

L52X

NH

A NEW SPECIES OF BARI SI A (SAURIA, ANGUIDAE)
FROM OAXACA, MEXICO

John P. Karges and John W. Wright1

ABSTRACT. A new species of Barisia from the cloud forest of the
Sierra Juarez, Oaxaca, Mexico, has both paired postmental scales
and complete series of superciliary scales and therefore is assigned
to the gadovi species group. It differs from the other species of this
group ( gadovi and antauges) by having distinctly barred labial re-
gions. It also has two or more subocular scales and smooth dorsal
scales (only one subocular and distinct keeling in B. gadovi) and a
dorsal color pattern with at least traces of dark, posteriorly directed
chevrons (absent in B. antauges). With a maximum snout-vent length
of 77 mm (N = 95). this is the smallest member of the gadovi species
group and previously has been confused in collections with the sym-
patric, similar-sized B. viridiflava of the moreletii group. All mem-
bers of the moreletii group have single postmental scales and B.
viridiflava has acutely keeled dorsal scales. Like congeners, B. juarezi
new species is viviparous.

RESUMEN. Una nueva especie de Barisia del bosque nebuloso de
la Sierra Juarez, Oaxaca, Mexico, tiene escamas postmentales pa-
readas y una serie completa de escamas superciliares; y por lo tanto,
es asignada al grupo gadovi. Esta difiere de las otras especies de este
grupo ( gadovi y antauges) por tener las regiones labiales distinti-
vamente barreadas. Esta tambien tiene dos o mas escamas subo-
culares y escamas dorsales sin quillas (solamente una escama su-
bocular y quillas distintivas en B. gadovi) y el patron del color dorsal
con por lo menos trazas oscuras en forma de cheurones (galones)
oscuros, dirigidas posteriormente (ausentes en B. antauges). Con una
longitud hocico-cloaca maxima de 77 mm (N = 95), este es el miem-
bro mas pequeno del grupo gadovi y en colecciones anteriores, ha
sido confundido con B. viridiflava, una especie del grupo morletti,
que es simpatrica y de tamano similar. Todos los miembros del
grupo morletti tienen solamente una escama postmental y B. viri-
diflava tiene escamas dorsales agudamente aquilladas. Como sus
congeneros, B. juarezi nueva especie, es vivipara.

INTRODUCTION

As generally understood, the anguid lizard genus Barisia con-
sists of seven species arrayed in three species groups (Tihen,
1 949). These species occur from northern Mexico southward
to western Panama. Some species have comparatively wide
distributions, such as Barisia imbricata, which occurs over

Contributions in Science, Number 381, pp. 1-11
Natural History Museum of Los Angeles County, 1987

a broad area including the Sierra Madre Oriental, Sierra Madre
Occidental, Mesa Central, and Sierra Madre del Sur; others
are restricted to single mountains or isolated ranges ( B . viridi-
flava, B. levicollis, and B. antauges). The habitats occupied
by Barisia are primarily seasonally wet montane llanos, and
forest-edge grassland in pine and/or oak woodlands or cloud
forest. These habitats have been relatively poorly explored
by biologists, in large part due to inaccessibility, and may
remain poorly known owing to the rapid deforestation and
habitat alterations by man. The extensive highlands of south-
ern Mexico are not contiguous and consist of isolated moun-
tain ranges and intervening valleys that constitute effective
ecological barriers for montane organisms. These areas have
a high level of species endemism. We describe here a new
species of Barisia endemic to the Sierra Juarez in northern
Oaxaca that was independently discovered by each of us. We
name it in honor of one of Mexico’s greatest heroes, Benito
Juarez, a man who was also indigenous to the region.

Barisia juarezi new species

Figures 1-5

HOLOTYPE. University of Texas at Arlington (UTA)
R-8485, an adult male, from the northern slope of the Sierra
Juarez between 6.1 and 1 1.6 km (3.8 and 7.2 mi.) N crest of
Cerro Pelon, Ixtlan District, Oaxaca, Mexico, 7 April 1979,
by J.A. Campbell, L.S. Ford, J.E. joy, Jr., and J.P. Karges
at approximately 2500-2700 m elev. (original number, JAC-
3708).

PARATYPES. Ninety-four, see Specimens Examined, be-
low.

DIAGNOSIS AND DEFINITION. Barisia juarezi is a
member of the gadovi species group as defined by Tihen

I . Section of Herpetology, Natural History Museum of Los An-
geles County, Los Angeles, California 90007, USA. Present address
of Karges: Fort Worth Nature Center, Route 10, Box 53, Fort Worth,
Texas 76135, USA.

ISSN 0459-81 13

Figure 1. Dorsolateral view of an adult male (LACM 130277, 77 mm SVL), paratype of Barisia juarezi new species photographed in life.

Figure 2. Dorsolateral views of adult female Barisia gadovi (upper, LACM 129612) and paratype of B. juarezi new species (lower, LACM
129615) photographed in life. The differences in facial markings, dorsal color pattern, and size are conspicuously illustrated.

(1949), with both paired postmental scales and complete
series of superciliary scales. It is the smallest member of the
group (maximum observed snout-vent length, 77 mm) and
differs from the other species of the group, B. gadovi and B.
antauges (here including B. modesta), by having two distinct
dark diagonal labial bars with an intervening white bar, ex-
tending from the posterior supralabials to the orbit and along
the lower border of the subocular (Figs. 1 and 2). The much

larger (to 102 mm SVL, Tihen, 1949) B. gadovi has a single,
horizontal light supralabial stripe, and B. antauges has dark
and light mottled supralabials but no stripes or bars. In ad-
dition, B. juarezi differs from B. antauges by the presence of
at least traces of dark, posteriorly directed chevrons (Fig. 3),
a characteristic shared with B. gadovi where the chevrons are
generally more distinct (Fig. 4). Barisia juarezi and B. an-
tauges both differ from B. gadovi in the presence of two or

2 Contributions in Science, Number 381

Karges and Wright: New Species of Barisia

Figure 3. Variation in dorsal color pattern of male Barisia juarezi new species. Specimens (all paratypes) range in snout-vent length from
53 mm to 71 mm and are UTA R-8432, 8425, 5806, 8489, and LACM 130277 from left to right.

more suboculars (only one in B. gadovi) and smooth dorsal
scales (distinctly keeled in B. gadovi).

DESCRIPTION OF HOLOTYPE. An adult male, 65 mm
snout- vent length, with complete tail (89 mm) and total length
(154 mm) to tail ratio of 0.58; head width, 9.75 mm; head
length to anterior auricular margin, 13.0 mm; greatest head
depth, 7.35 mm; horizontal length of right orbit, 2.45 mm;
axilla-to-groin length, 39.5 mm; longest toe (fourth) on right
hind foot, 5.92 mm, on right front foot, 4.57 mm.

Three anterior intemasal scales, right side with large me-
dian and smaller lateral scale, occupying same amount of
space as single left intemasal (Fig. 5); two anterior intemasals
in medial contact posterior to rostral; rostral, first supralabial
(by small lateral intemasal on right side), nasals, posterior
intemasals, and supranasals in contact with anterior inter-

nasals; no postrostral; supranasals longer than wide, not in
medial contact; right posterior intemasal longer than wide
and in medial contact with anterior element of divided left
posterior intemasal; anterior element of left posterior inter-
nasal generally rhomboidal and slightly larger than posterior
element, effectively forming an anterior canthal; two left in-
ternasals equal in size to right; two postnasals on each side,
lower nearly twice size of upper and more than half as large
as nasal; single large loreal on each side, contacting third and
fourth supralabials, both postnasals, posterior intemasals
(posterior element on left), single preocular, first superciliary,
and narrowly contacting first medial supraocular on right
side; frontonasal rhomboidal, about as long as wide, con-
tacting both posterior intemasals and both prefrontals; paired
prefrontals in medial contact posteriorly, about as large as

Contributions in Science, Number 381

Karges and Wright: New Species of Barisia 3

Figure 4. Dorsal views of representative pairs of adult specimens of members of the Barisia gadovi species group. From left to right the
specimens are B. an/a«grs-BMNH 1903.9.30.122, 72 mm SVL and MBS 3685, 82 mm; B. juarezi new species— LACM 130277, 71 mm
and UTA R-8489, both paratypes, 67 mm; and B. gadovi — LACM 121921, 93.5 mm and UTA R-5794, 74 mm.

frontonasal, abutting frontal; frontal large, twice as long as
wide; live medial supraoculars and two small lateral supra-
oculars on each side; six superciliaries over each eye, first
largest; frontoparietals rhomboidal, much smaller than ad-
jacent frontal and parietals; supralabials 9/9, antepenulti-
mate largest and last to reach orbit at posterior margin of
second subocular; two elongate suboculars and three discrete
postoculars on each side; primary temporals 4/4, lowest nar-
rowly contacting posterior subocular on each side; secondary
temporals 5/5; tertiary temporals 4/4, with small azygous
scale on right between uppermost tertiary temporal and right
paraoccipital; rosette of scales around interoccipital, includ-
ing two parietals, interparietal (containing parietal eye in
posterior half), two paraoccipitals and medially paired pos-

toccipitals; infralabials 8/7; sublabials 5/5, extending in series
anteriorly to second infralabial and second pair of chin-
shields; five pairs of chinshields posterior to mental, first
being medially paired postmentals; second pair of chinshields
only other pair in medial contact; single medial gular scale
between second and third pairs of chinshields.

Eight smooth nuchal scales across narrowest part of nape;
granular scales extending from auricular opening posteriorly
along lateral surfaces of neck and along lateral fold; dorsal
and ventral body scales in parallel, regular (not oblique) rows
both transversely and longitudinally; transverse dorsal scale
rows from postoccipitals (inclusive) to level of posterior mar-
gin of thigh, 48; all dorsal scales smooth, in 14 longitudinal
rows and of uniform size across midbody; an additional row

4 Contributions in Science, Number 381

Karges and Wright: New Species of Barisia

Figure 5. Variation in head scutellation of representative Barisia juarezi new species. Specimens (all paratypes) from A to E are: UTA
R-4317, 8456, 8488, 4315, and 5806. See text for descriptions and terminology.

of slightly smaller scales along dorsolateral border of lateral
fold on each side; ventral scales across midbody in 12 lon-
gitudinal rows; ventrals smooth, in 55 transverse rows from
first gular to anterior edge of vent; tail complete with 70
caudal whorls and 23 scales around base; caudal scales arched
transversely near base of tail, giving tail a longitudinally keeled
texture.

Adpressed limbs separated by 1 1 transverse dorsal scale
rows; enlarged, smooth, imbricate scales on anterior and dor-
sal surfaces of forelimbs and anterior and ventral surfaces of
hind limbs, posterior surfaces with granular scales; toes clawed
except for clawless second toe (injured) and freshly severed
third toe on hind foot; 16 subdigital lamellae under fourth
toe of right hind foot.

Dorsum above third and fourth longitudinal dorsal scale
rows with uniform longitudinal pale brown stripe with darker
brown spots on several middorsal scales; sides, between lat-

eral fold and third longitudinal scale row, darker brown than
dorsum, with irregular, non-continuous dark brown bars (Figs.
1-3), consisting of dark brown individual scales, some white
spots on lower posterior margins; individual dark brown scales
scattered along dorsolateral portion of body, from nape onto
tail, forming spotted line along body; dark brown spots scat-
tered along middorsum and onto tail; ground color of dorsal
surfaces of limbs and tail similar to dorsum; venter dark and
mottled, individual scales usually containing much more dark
pigment than pale; ventral surfaces of tail and limbs similar
to venter; lighter areas of ventral scales metallic silver-blue
in life and preservative; top of head and posterior part of
jaws with pale reddish-brown ground color; preorbital region
of face dark brown, including first superciliary scales, upper
part of loreal and preocular scales above white suborbital
stripe; suborbital stripe an anterior extension of diagnostic
white labial bar, covering posterior half of sixth and anterior

Contributions in Science, Number 381

Karges and Wright: New Species of Barisia 5

third of seventh supralabial; dark brown diagonal bar, pos-
terior to white bar (also diagnostic) on posterior half of sev-
enth and anterior half of eighth supralabial; anterior five and
posteriormost supralabials pale brown; anterior half of sixth
supralabial also brown, completing diagnostic pattern of two
dark diagonal bars with intervening pale bar on upper jaw;
lateral surfaces of neck uniform brown.

VARIATION

We attempted to select from the type series a holotype spec-
imen that was modal in as many features as possible. The
following analysis of variation is thus focused on departures
from these modal conditions in the same sequence as con-
tained in the preceding description of the holotype. The anal-
ysis is based on all 95 specimens contained in the hypodigm
(see Specimens Examined, below).

SIZE AND BODY DIMENSIONS

The largest specimen is a female (77 mm in SVL) but there
is no significant difference between the mean sizes (Mann-
Whitney U = 81.82, P < 0.05) of the 15 largest females
(mean ± SE, 70. 1 ± 0.90) and 1 5 largest males (69.6 ± 0.79).
Adult male B. juarezi have wider heads and larger jaws than
females. In 43 individuals with SVL greater than 60 mm (21
33 and 22 99) there is a highly significant difference between
the sexes in the snout-vent length/head-width ratio (Mann-
Whitney U = 208.72, P 0.01, <53 6.48 ± 0.12, 5.77-7.83
and 99 7.52 ± 0.09, 6.64-8.30). There is also a significant
difference in relative tail lengths between the sexes ( | Z | —
transformed Mann-Whitney U = 4.334, P <k 0.01) calculated
for 30 specimens greater than 50 mm SVL, with complete,
unregenerated tails (35 1.38 ± 0.02, 1.23-1.50, N = 17 and
99 1.33 ± 0.03, 1.09-1.50, N = 13). The relatively longer
tail of males results in a mean difference of one percent
between males (57%) and females (58%).

SCUTELLATION

In general, we use the scale terminology of Tihen ( 1 949) with
modifications proposed by Waddick and Smith (1974). Oc-
casionally we have found it necessary to redefine some con-
ditions as they pertain to B. juarezi. The standard positions
of several scales (e.g., frontal, nasals, rostral) of gerrhonotines
provided reference points for identifying other scales.

The anterior intemasals are most frequently paired (45 of
95 specimens); occasionally single (22 specimens including
the holotype) or one or both (2 1 specimens) are divided to
form three or four discrete scales (Fig. 5C). In four specimens,
a single medial scale is present between the anterior inter-
nasals and constitutes a postrostral scale. In animals with
four scales (both anterior intemasals divided) in a lateral
series behind the rostral, the medial pair of scales is best
termed the medial intemasals rather than medially paired
postrostrals, since the postrostral is defined as a single medial
azygous scale (Waddick and Smith, 1974), or if two post-
rostrals are present, they form a longitudinal series. We have
not observed a longitudinal series of anterior and posterior

postrostrals in B. juarezi, but one specimen (UTA R-6096)
does have a postrostral scale posterior to a nearly medial
element of a divided anterior intemasal. When divided into
lateral and medial elements, the medial scale is usually larger
than the lateral. The anterior intemasals of each side contact
medially behind the rostral in most specimens (85 of 94,
90.5%); in four specimens the medial contact of the anterior
internasals is precluded by the contact of the rostral and
enlarged supranasals (Fig. 5A); and in two others, medial
anterior intemasal contact is precluded by the rostral-post-
rostral suture.

The supranasals are separated from medial contact by the
anterior intemasals, posterior intemasals, and postrostrals
when present, in 72 specimens (77.4%). Individual supra-
nasal scales are roughly triangular and usually longer than
wide, although an individual specimen may have one su-
pranasal wider than long on one side. Both supranasals are
longer than wide in 53 (57%) specimens, both wider than
long in 22 (23.6%), and one longer than wide and the other
wider than long in 18 (19.4%). The number of animals with
medial contact of the supranasals (21) approximately equals
the number of specimens with both supranasals wider than
long (22). On the dorsal surface of the snout, the paired scales
posterior to the anterior intemasals, postrostral, and supra-
nasals are the posterior intemasals, among the most variable
head scales. Anterior contacts of the posterior intemasals
variously include the anterior intemasals (rarely. Fig. 5D),
postrostral (if present), the supranasals (if expanded. Figs.
5A, B), and anterior medial contact of the two posterior
intemasals (Fig. 5D). The posterior intemasals vary in size
relative to adjacent scales, and are often asymmetrical, with
the scale on one side larger than its counterpart (Fig. 5C).
Occasionally one of the posterior intemasals is divided into
anterior and posterior elements, the anterior element effec-
tively forming a postrostral and the posterior element form-
ing a canthal.

The frontonasal scale is either present (55 specimens), ab-
sent (37), or partially fused with the prefrontals (1). The
frontonasal is bordered by the postrostral (if present and
posteriorly placed), the posterior intemasals, and prefrontals.
The frontonasal is usually rhomboidal and about as long as
wide (Fig. 5D), but sometimes longer than wide (Fig. 5C).
The frontonasal varies from one-half to equal in size to the
adjacent prefrontals. When the frontonasal is absent, the pre-
frontals are large and occupy the area where the frontonasal
would be on other individuals (Figs. 5A, B). In six of the 37
specimens lacking a frontonasal scale, the prefrontals are
partially fused, usually along the anterior half of the medial
contact. The remaining 3 1 specimens lack frontonasals, and
the paired posterior intemasals and paired, fully divided pre-
frontals form a linear pattern of paired scales, medially di-
vided, which in some cases also includes the enlarged su-
pranasals and anterior intemasals (Figs. 5A, B). There are
usually five, occasionally four, medial supraoculars and usu-
ally two or three lateral supraoculars but rarely none, one,
or four.

The scales on the side of the head of B. juarezi are also
variable and their definitions require some clarification. Im-

6 Contributions in Science, Number 381

Karges and Wright: New Species of Barisia

mediately posterior to the nasals are the postnasals. These
are usually divided into upper and lower elements but oc-
casionally are fused to a single scale, or, more rarely, divided
into three small elements. Usually, a single loreal is present
on each side but occasionally there are two on one or both
sides of the head. When divided, the loreal is usually divided
horizontally into upper and lower elements. The upper ele-
ment is usually the smaller scale and owing to its position it
forms what others (Waddick and Smith, 1974; Campbell,
1 982) have called an anterior canthal or cantholoreal. Rarely,
the upper or lower loreal elements may be further divided
into anterior and posterior elements; in these cases (four
specimens) there are three loreal scales. Between the loreals
and the orbit is a single preocular (rarely divided into upper
and lower elements).

The complete series of superciliary scales extends from the
canthus to the posteriormost medial supraocular. Most spec-
imens (91 of 94, 95.8%) have six superciliaries in both series
(68 specimens) or in at least one series (923 specimens).
Twenty-one animals have five in one or both series, four
have seven in one or both series, and one specimen has only
four on one side. The anteriormost superciliary scale extends
anterior to the eye and is somewhat expanded. In this respect,
it could be called the posterior canthal (see Waddick and
Smith, 1 974, fig. 3 for Gerrhonotus liocephalus), but we prefer
to include it in the superciliary series as did Bogert and Porter
(1967, fig. 5) for Abronia mixteca.

In decreasing frequency, there are two, three, or four sub-
oculars, of which the posteriormost is occasionally in contact
with the lowest primary temporal. The small but discrete
postocular scales occur in a vertical series of two, three, or
rarely four scales bordering the posterior margin of the eye.

There are typically four primary temporals, sometimes
three or five, in a linear series extending from the supralabial-
subocular margin upward to and abutting the frontoparietals.
The four secondary temporals, sometimes five, are posterior
to the primary temporals. The two uppermost tertiary tem-
porals are in medial contact across the back of the head,
forming the postoccipitals (and the first row of transverse
dorsal scale rows, see below).

Most specimens have either nine or 10 supralabials on at
least one side, but 38 (40.4%) have 10/10, 20 (21.3%) have
9/9, and 26 (27.7%) have nine on one side and 10 on the
other. Ten specimens have 1 1 supralabials on at least one
side, and two specimens have 1 1 on both sides. The most
frequent number of infralabials is eight on at least one side
(85.1%), and the modal class (41 specimens) has eight on
both sides. Other specimens have nine on at least one side
(33 specimens, 35.1%), nine on both sides (seven specimens),
seven on one or both sides (13 specimens), and a single
specimen (LACM 130277) has five infralabials on one side.

We began the counts of transverse rows with the row im-
mediately behind the occipital scale and ended at the level
of the posterior edge of the hind limbs. The count was made
along the middorsal row. The range for all 95 specimens was
42-55 (x = 47.34 ± 2.20). The number of longitudinal dorsal
scale rows, counted at midbody, was usually 14 enlarged
scales of similar size (84.21%, 80 of 95 specimens). There

was frequently an additional row of slightly smaller scales on
each lateral border. In fact, 62 specimens (65.3%), including
the holotype, have this extra row on both sides, or rarely on
only one side. Nineteen specimens have only 14 discrete
longitudinal scale rows, 14 specimens have 16 rows, and 59
specimens have 14 discrete rows and two additional lateral
rows of smaller scales. Two specimens have 14 rows and an
additional row only on one side, and one specimen has 12
rows and a lateral row of reduced scales on each side.

The ventral scales are arranged in parallel rows in both
longitudinal and transverse series. The transverse series was
counted from the most anterior medial gular scale behind
the postmentals (wedged between enlarged chinshields) to
and including the scale row lining the anterior margin of the
vent (preanals). This count was possible on 93 specimens
and the number ranged from 50 to 59 (x = 54.3 ± 1.92).
The number of ventral rows, counted at midbody between
the lateral folds, was 12 in 91 specimens (95.79%), 13 in one
specimen, and 14 in three. The preanal margin consists of
four, rarely three, large scales with smaller scales lining the
lateral edges of the vent. Fifteen specimens have complete,
unregenerated tails, with 69-77 (7 1 .0 ± 0.59) caudal whorls.
There are 16 to 19 subdigital lamellae under the fourth toe
of the hind foot.

COLOR AND PATTERN VARIATION

In life, the dorsum is uniformly light brown, but varies from
reddish-brown to tan, often with a bronze tint. The dark
brown posteriorly directed dorsal chevrons form a “herring-
bone” pattern, which may be bold or faint (Figs. 1-3). This
pattern is the predominant dorsal pattern of B. juarezi and
occurs in both sexes (23 of 35 males; 26 of 39 females). Less
frequently, the dorsal pattern is obscured to such an extent
that the middorsum is entirely uniform (10 males; 12 fe-
males), contrasting with the darker, barred dorsolateral sur-
face. In five specimens, a distinct narrow, dark middorsal
stripe is present, extending from the occipital or nuchal region
posteriorly onto the tail. Frequently, the middorsal area is
dark, forming an indistinct stripe through the centers of the
chevrons, particularly distinct posteriorly and onto the tail.

The sides of the body are darker than the dorsum, espe-
cially in subadult and juvenile specimens. Dark brown scales
mark the dorsolateral border from the nape posteriorly onto
the tail. In adults, the sides have dark erratic bars from the
dorsum to the lateral fold, consisting of dark brown scales,
some with white or cream-colored flecking along the poste-
rior scale margin. The lateral bars are most distinct in adult
males and much less distinct or occasionally absent in fe-
males. The lateral pattern is obscured in juveniles by the
entirely dark lateral surface. The ground color of the lateral
fold and granular scales on the sides of the neck is gray-
brown, with some scattered white flecks, sometimes covering
entire scales.

The ventral color pattern is sexually dichromatic. In males,
the venter is darkly mottled, with limited patches of pale
pigmentation usually confined to the posterior edges of in-
dividual scales. In life, and to a lesser extent in preservative.

Contributions in Science, Number 381

Karges and Wright: New Species of Barisia 7

Figure 6. The geographical distribution of the three members of the Barisia gadovi species group ( antauges— A , gadovi— M, and juarezi new
species—#). All three species occur principally in pine-oak habitats in portions of the Sierra Madre del Sur in Guerrero, Oaxaca, and Veracruz,
Mexico.

the paler areas have an iridescent, metallic blue tinge. Rarely,
a male will have more pale ventral pigment than dark. Fe-
males, on the other hand, have much paler iridescent blue-
gray or cream venters, with mottling so diffuse that the venter
is nearly uniform. Some dark mottling occurs in the gular
region of some females, and occasionally large dark blotches
occur on individual scales. In juveniles, the venter is dark
brown, like the lateral surfaces, with heavy mottling. The gular
region of juveniles is paler than the remainder of the venter,
with distinct sublabial stripes, contiguous with the diagnostic
supralabial bars.

In all individuals, the dorsal surfaces of the limbs and tail
are the same as the dorsal ground color, and the ventral color
of the appendages is that of the venter. Regenerated portions
of autotomized tails are a much lighter brown or tan than
the original portions of the tail, and the tail is frequently
enlarged into a bulbous swelling at the point of regeneration.

COMPARISONS

The two species most similar to B. juarezi are the two other
members of Tihen’s (1949) gadovi group, B. gadovi Boulen-
ger and B. antauges Cope [here including B. modesta (Cope)
fide H.M. Smith, in lift.]. The gadovi group was defined by
Tihen ( 1 949) as Barisia having paired postmental scales and
complete superciliary series. The other two species groups,
moreletii and imbricata, have either single postmentals or
incomplete superciliary series. The gadovi group occurs west
of the Isthmus of Tehuantepec in the highlands of south-

central Mexico (Fig. 6). Within the group, B. juarezi is most
similar to B. antauges in size, color pattern of the head, and
squamation. The following comparisons are based on the
seven known specimens of B. antauges (see Specimens Ex-
amined, below). The maximum snout-vent length in B. juare-
zi (N = 95) is 77 mm (UTA R-8696) whereas the largest B.
antauges (USNM 30221, holotype) is 85 mm. The number,
placement, and variation in head scales are similar in both
B. juarezi and B. antauges. There is a rosette of apparently
homologous scales surrounding the interoccipital scale in both
species. A postrostral scale is present in all specimens of B.
antauges but is absent as a discrete medial scale in more than
half (55.5%) of the specimens of B. juarezi. The presence of
a single medial frontonasal and an anterior canthal on one
or both sides is variable in both species. Additional shared
similarities in head scales include: eight to 10 supralabials
on each side; five or six superciliaries per side; two or three
lateral supraoculars; five or six medial supraoculars; four or
five primary and secondary temporals; and five to seven
tertiary temporals (see Variation, above). In B. juarezi, the
preocular is usually a large scale extending from the upper
anterior portion of the orbit forward to the distinct loreal(s)
and downward to the supralabials. The preocular in B. an-
tauges is less distinct and usually fused with the lower and
posteriormost loreal. The smooth dorsal scales are arrayed
in 14 to 16 longitudinal rows.

The presence of smooth dorsal scales easily distinguishes
both B. antauges and B. juarezi from the other group mem-
ber, B. gadovi, which has acutely keeled middorsal scales.

8 Contributions in Science, Number 381

Karges and Wright: New Species of Barisia

Additional scale characteristics which differentiate B.juarezi
from B. gadovi are: usually six (range 4-7) superciliaries in
B.juarezi and usually four (range 3-5) in B. gadovi; the loreal
is usually single on each side in B. juarezi and paired (upper
and lower elements, with upper forming an anterior canthal)
in B. gadovi (and B. antauges ); Barisia juarezi has two or
three suboculars, whereas B. gadovi has a single, elongate
scale; and the longitudinal dorsal scale rows of B. juarezi
number 14 (sometimes 16, see Variation, above) whereas
there are usually 18 (occasionally 16) in B. gadovi. The dif-
ferences in color pattern of the side of the head (cheek) be-
tween B. juarezi and B. gadovi are distinct. In B. gadovi.
there is a long, horizontal, white stripe on the supralabials
extending from the angle of the jaw to the nostril. The face
stripes are both light and dark and diagonal in B. juarezi.
The most conspicuous similarity between B. juarezi and B.
gadovi is the otherwise unique dorsal color pattern of pos-
teriorly directed dark chevrons, most distinct in adult males
(Fig. 4). These are not present in B. antauges.

Two other species of Barisia. each allocated to different
species groups (Tihen, 1 949), occur in the highlands of north-
central Oaxaca. Barisia imbricata (of the imbricata species
group) is known from several localities in the region (Guil-
lette and Smith, 1982), including the Sierra Juarez. Barisia
viridiflava, member of the moreletii group (Tihen, 1 949), and
B. juarezi are sympatric (see Habitat, below). Barisia im-
bricata is easily distinguished from B. juarezi by its large size
(maximum SVL approximately 140 mm), low number (2-
4) of superciliaries in incomplete series, rugose head scales
deeply indented at sutures, acutely keeled middorsal scales
in less than 45 transverse rows, and the lack of diagonal face
stripes. More similar to B. juarezi is B. viridiflava, as borne
out by the presence of many specimens of B. juarezi mas-
querading as B. viridiflava in some collections. Overall sizes
of B. juarezi and B. viridiflava are similar (maximum SVL
of viridiflava, 75 mm). The differences in color and pattern
are distinct; B. viridiflava usually has at least one (middorsal)
or three prominent longitudinal dark stripes in the paler brown
dorsal area, whereas B. juarezi has dark chevrons. The light
jaw stripe in B. viridiflava is horizontal and extends poste-
riorly from the loreals across the suboculars to the lower
temporals. The middorsal scales of B. viridiflava are distinct-
ly keeled, unlike B. juarezi. Most important of the scale
differences is the single postmental, a character which is di-
agnostic for separating the moreletii and gadovi groups.

HABITAT AND BIOLOGY

Barisia juarezi occurs in cloud forest at elevations from 2000
to approximately 2800 m on the northern slopes of the Sierra
de Juarez (Fig. 6). The cloud forest merges with pine-oak
forest at elevations from 2800 to 3000 m on the summits
and adjacent ridges. The dominant forest plant species are
oaks ( Quercus sp.), some pine ( Pinus ayachuite and Finns
sp.), madrono {Arbutus xalapensis), sweetgum (Liquidamber
sp.), and a tree fern (Cyathea mexicana ). Epiphytic mosses,
bromeliads, ferns, and orchids are abundant and dense on
trees, and unlike the higher more open pine-oak forest, there

is a dense understory consisting of small trees, shrubs, vines,
and herbs. Second growth is dense in areas of fallen trees
and lumbered clearings. Rain occurs in the habitat in all
months but is heaviest during the period of June through
October or November. Most of the precipitation comes from
moisture-laden winds from the Gulf of Mexico.

Barisia juarezi is almost strictly terrestrial, occurring on
the forest floor and in clearings, seeking refuge under rocks,
logs, and in the leaf litter. Specimens caught on the surface
were sunning or actively foraging in the patches of dappled
sunlight penetrating the canopy.

Like congeners, Barisia juarezi is viviparous. A pregnant
female (UTA R-8488) collected 7 April 1979, had three well-
developed, pigmented embryos in the oviducts. The degree
and pattern of pigmentation and size of these embryos was
similar to those of the smallest field-collected individuals
(25, 27, 28.5 mm SVL), indicating that they were nearly full-
term. Additionally, several B. juarezi have given birth in
captivity in our laboratory.

The female reproductive cycle apparently involves over-
winter gestation, with parturition occurring in late spring,
immediately prior to the summer monsoon season. Females
attain sexual maturity at approximately 60 mm SVL, and
enlarged ovarian follicles were present in a single female
collected 2-3 April and seven collected in June and early
July. Oviductal eggs (embryologic stage not determined but
still largely yolk masses) were found in four females preserved
in mid-July and August. A female collected in winter (early
January) also contained large oviductal eggs. The smallest
field-collected juveniles were obtained in early April, further
corroborating pre-monsoon natality. Young-of-the-year, be-
tween 30 and 40 mm SVL, have been collected throughout
the summer months.

Rainfall appears to modify substantially the seasonal ac-
tivity of B. juarezi. Most specimens were collected during
the wet season. In April 1979, a normal dry season, they
were common in leaf litter and under surface rocks and logs.
The soil was moist, indicating some recent rain. Following
a severe drought across southern Mexico and a delayed mon-
soon in the winter of 1982-83, we found no specimens in
early June 1983, and the soil was dry, even under surface
objects.

ACKNOWLEDGMENTS

We thank the following persons and institutions for loans of
specimens: C.W. Myers (American Museum of Natural His-
tory, AMNH), W.F. Pyburn (University of Texas at Arling-
ton, UTA), D.B. Wake and H.W. Green (University of Cal-
ifornia, Berkeley, Museum of Vertebrate Zoology, MVZ),
J.R. Dixon (Texas Cooperative Wildlife Collection, TCWC),
A.G. Kluge (University of Michigan Museum of Zoology,
UMMZ), W.G. Degenhardt (Museum of Southwestern Bi-
ology, UNM), A.E. Leviton (California Academy of Science,
CAS), W.R. Heyer and R.W. McDiarmid (National Museum
of Natural History, USNM), H.M. Smith (University of Col-
orado Museum, UCM), W.E. Duellman (University of Kan-
sas Museum of Natural History, KU), Andrew Stimson [Brit-

Contributions in Science, Number 381

Karges and Wright: New Species of Barisia 9

ish Museum (Natural History), BMNH], R.G. Webb and C.S.
Lieb (University of Texas at El Paso, UTEP), D.F. Hoff-
meister (University of Illinois Museum of Natural History,
UIMNH), P. Alberch (Harvard University, Museum of
Comparative Zoology, MCZ), R.F. Inger (Field Museum of
Natural History, FMNH), C.J. McCoy (Carnegie Museum,
CM), and the Reptile Department of the Naturhistorisches
Museum Basel (MBS). We also thank C.M. Bogert, J.A.
Campbell, and H.M. Smith for unpublished observations.

The photography was executed by J. DeLeon. R.L. Bezy,
J.A. Campbell, C. Ceron H., L.S. Ford, J.S. Godley, J.E. Joy,
Jr., C.S. Lieb, R.W. McDiarmid, and M.B. Ruggles assisted
in aspects of the field research.

SPECIMENS EXAMINED

Barisia juarezi (paratypes, 94). MEXICO: Oaxaca; Ixtlan
District; 51 km (by rd.) N Ixtlan de Juarez, LACM 15158;
54 km N Guelatao on Mex. Hwy. 175, 2510 m, LACM
122496; 60.3 km (34.7 mi. by rd.) NE Guelatao, Mex. Hwy.
175, 2316.5 m (7600 ft.), LACM 125365; 52.2 km (32.4 mi.
by Hwy. 1 75) N Guelatao, 25 1 5 m, LACM 1296 14-1 5; 55.6
km (34.5 mi. by Hwy. 175) N Guelatao, 12.6 km (7.8 mi.
by rd.) NE jet. of rd. to Comaltepec, LACM 130271; 17.9
km (11.1 mi. by rd.) S Vista Hermosa, LACM 130272-84;
7.2 km (4.5 mi. by rd.) N Cerro Pelon, UTA R-3409-15,
CAS 142597; 3.2 km (2 mi.) N Cerro Pelon, UTA R-3832;
9.4 km (5.8 mi.) N crest Cerro Pelon, UTA R-4309-17; 9.8
km (6.1 mi.) N crest Cerro Pelon, UTA R-4305-07, 4863-
64; 10.5 km (6.5 mi.) N crest Cerro Pelon, UTA R-5803-
06; 1 1.4 km (7.1 mi.) N crest Cerro Pelon, UTA R-5807-
09; 9.7 km (6 mi.) N crest Cerro Pelon, UTA R-60996-100;

1 1 km (6.8 mi. by rd.) N Cerro Pelon, UTA R-7741, 7879,
8695-96; 10.2 km (6.3 mi.) N crest Cerro Pelon, UTA R-
8425, 8430-32; 8. 1-1 1 .6 km (5-7.2 mi.) N Cerro Pelon, UTA
R-8464; 6.1-11 .6 km (3. 8-7. 2 mi.) N crest Cerro Pelon, UTA
R-8484, 8486-95; 11.6 km (7.2 mi.) N crest Cerro Pelon,
UTA R-8503-04, 8512-15; 12.1 km (7.5 mi.) N crest Cerro
Pelon, UTA R-851 1; ca. 16.1 km (10 mi.) SE Llano de las
Flores, 2819.4 m (9250 ft.), AMNH 100711; 9.7 km (6 mi.)
NE Cerro Pelon, 2194.6 m (7200 ft.), AMNH 100719; 61
km NE (by Mex. Hwy. 175) Guelatao, MVZ 1 12393; 52 km
NE (by Mex. hwy. 175) Guelatao, MVZ 1 12394; Cerro Juar-
ez, TCWC 36536-37; 10 km (6.2 mi.) NE Cerro Pelon, 52.4
km (32.5 mi.) SSW Valle Nacional, UMMZ 134015; 11.3
km (7 rd. mi.) NE Cerro Pelon, 2133.6 m (7000 ft.), UNM
25505; 4.8 km (3 mi.) E Cerro Pelon, UNM 30770-71; 1 1.9
km (7.4 mi.) E Cerro Pelon, UNM 30772-73; Vista Hermosa,
Comaltepec, UCM 49305.

Barisia antauges (including B. modesta, 7). MEXICO: Ve-
racruz; Pico de Orizaba, USNM 30221 (holotype), USNM
7084A-C (syntypes of B. modesta ), CAS 98681, BMNH
1903.9.30.122, MBS 3685.

Barisia gadovi levigata (54). MEXICO: Oaxaca; Vic. Tejo-
cotes, AMNH 102658-72, 102674-83, UNM 22756, 2278 1-
82, UTA R-5793-98, LACM 102984-86, 121917-24,
121926; 13.2-15.5 km W (by rd.) San Vicente Lachixio,

LACM 125366-67; 10.4 km WSW (by rd.) San Vicente La-
chixio, LACM 129613, MVZ 164778; Sierra de Cuatro Ve-
nados, LACM 125368-69; 25-25.3 km (15.5-15.7 mi. by
rd.) W Zaachila, LACM 12961 1-12.

Barisia gadovi gadovi (46). MEXICO: Guerrero; Omilteme
(and vicinity), MCZ 42703-15, 96804, UNM 6016, 26444,
TCWC 9900-08; Chilpancingo, KU 23792-94; 7.2 km (4.5
mi.) W Mazatlan, TCWC 9897-99, 1 1383-84, 1 1536; Aso-
leadero, 45 km (airline) WNW Chilpancingo, KU 105837-
39; 6-12 km SW Filo de Caballo, KU 182656-64.

Barisia viridiflava (2 1 7). MEXICO: Oaxaca; Sierra de Juarez;
Cerro Pelon, LACM 109296-99, 109384, 121931-37, CM
41256; 1.1 km (0.7 mi.) NE Cerro Pelon, LACM 121939-
40, UTA R-58 12-13, 7996, 848 1-82; E Cerro Pelon, AMNH
98043; 49 km (by rd.) N Ixtlan de Juarez, LACM 1 5 1 57; 49
km (by Mex. Hwy. 175) NE Guelatao, CAS 139904; 37 km
(by rd.) N Guelatao, LACM 122500; 1 1.3 km (7 mi.) NNW
Ixtlan de Juarez, CAS 87286, 87290; 14.5 km (9 mi.) NNE
Guelatao, UTEP 5167; Cerro Humo Chico, UCM 38782-
85, 44347-49; Llano de Las Flores, UMMZ 118808-09,
1 19631 (6), 124093, 125876 (4), AMNH 89647-51, 89827-
30, 98044-45, UIMNH 60152-61, KU 70805-14; Llano de
English, UIMNH 60137-51; 6.4 km (4 mi.) S Llano de las
Flores, AMNH 89641-43, 90992-93, 90995, 100720; 11.8
km (7.3 mi.) NE Cerro Pelon, AMNH 102716; 8.7 km (5.4
mi. by rd.) N El Carrizal, AMNH 100717; 8.1 km (5 mi.) N
El Carrizal, AMNH 102718-19, 102721-25; Cerro San Fe-
lipe, LACM 122497-99, 125372-77, UCM 4 106 1-62, 48327-
29, USNM 1 13220, UMMZ 126256 (5), 126257, 126258
(3), FMNH 99023-36, 112026, MVZ 112389-92, 140643-
48. 162291-98, AMNH 90982-91, 103703, UNM 15495-
504, UIMNH 60122-36; 12.9 km (8 mi.) W Cumbre del
Estudiante, AMNH 9721 1; 2.4 km (1.5 mi.) N Campo Co-
conal, 27.4 km (17 mi. by rd.) NW Tamazulapan, LACM
131525-26; Mt. Zempoaltepec, LACM 15154-56, 62446,
130790, AMNH 90096, USNM 47184-85, 47599.

Barisia moreletii (5). MEXICO: Chiapas; 1 1 km (6.8 mi.) SE
San Cristobal de las Casas, LACM 58102-04; 12.6 km (7.8
mi.) SE San Cristobal de las Casas, LACM 61209; 1.6 km
(1 mi.) S San Cristobal de las Casas, LACM 74284.

Barisia imbricata (5). MEXICO: Oaxaca; 1 km (0.6 mi. by
rd.) N Machin, 41.9 km (26 mi. by Mex. Hwy. 175) N Gue-
latao, LACM 130125; Veracruz; Pico de Orizaba, LACM
121929, 131443, 131445-46.

LITERATURE CITED

Bogert, C.M., and A.P. Porter. 1967. A new species of
Abronia (Sauria, Anguidae) from the Sierra Madre del
Sur of Oaxaca, Mexico. American Museum Novitates,
no. 2279, 21 pp. American Museum of Natural History.
Campbell, J.A. 1982. A new species of Abronia (Sauria,
Anguidae) from the Sierra Juarez, Oaxaca, Mexico. Her-
petologica 38(3):355— 36 1 .

Guillette, L.J., Jr., and H.M. Smith. 1982. A review of the
Mexican lizard Barisia imbricata, and the description

10 Contributions in Science, Number 381

Karges and Wright: New Species of Barisia

of a new subspecies. Transactions of the Kansas Acad-
emy of Sciences 85( 1 ): 1 3-33.

Tihen, J.A. 1949. A review of the lizard genus Barisia.

University of Kansas Science Bulletin 33(3):2 17-256.
Waddick, J.W., and H.M. Smith. 1974. The significance

of scale characters in evaluation of the lizard genera
Gerrhonotus, Elgaria and Barisia. Great Basin Natu-
ralist 34(4):25 7-266.

Submitted 30 October 1985; accepted 17 November 1986.

Contributions in Science, Number 381

Karges and Wright: New Species of Barisia 11

mk

! tO jofttHP ;

! 3 • W« *#4«S;f; Sl ••"•• lll! '

■■■

! 1 |i > fi III
mi ! till! Mil
'i

ala ,

'll III, VP ill!

'

sposii

)F MA]f

an

pitted, in ; ;:
jiffi name

nli i’I. : £

ih Latin Ameipcan subjects.

il as in English

■llferrl

Z'RIPT

, i' : ! :

■H

HI

is

| j|

;U

l,‘" mm m ":8 m

Mi!

fcl| §>

ip

.a;>5b

•i PB ii W il 1 11 HI |l i liU'li li I1 |

ifllji I ij! i lj! MjT.il :| I ; M>:

simAi

ii ;h

l| !, .1

I;!l:Jil l!i'

Wm

||&iq|ihj|i;

I:!1-;

■Ill

lj {i-;:’- 1 Rl-'.il-ii-; ■’ j.'v • r!i

age, width)

j , u.i .i;i;4 4/u

jlSlHp

I’:|m MSiwfc'

Illlllail

. $ ■ ! , i : S .

il Bllliili

.! j;l' siitMi) ■ .. . 1 (.fc.fl

CfMAM-SICKIPT .

m ;Srp; M V:.: M Mi ' : ;i i;i \

pi in a secure enclosure. Manuscript should be
:'M Author whoiwiijl be responsible for correspon.de
tier should cos!) tain a statement thdtthe manure
• elsewhere ereept in flfife

I

I'PA,,

M l

ill'll

111

ilSiil'

in.,'

■ M

111!

■ I' ''hi I o

,jye a statement of page,
rter fijitia or complete payijn'biif
“re funds available for thfe
lent of page charges
for. The ret rnled
" and prior y p:l

ISSlIil

■ ■iliil i

i'llUWHTO fi ll
111

iiiiiilfi

III;'!

Ill

ill

Hli! ;r I

»1?

1 ' ™™

fAL PUBLICATIONS OF THE

NATURAL HISTORY MUSEUM OF LOS ANGELES COl

: ; ' ■■■■•'■■

The scientific. publications.' of the Natural History Museum of Los Angeles County have been

issued; at irregular intervals in. three major series; the issues in each series are numbered

mm

Iv.

in Science, a miscellaneous series of technical papers describing orig-
inal research in the life and earth Sciences.

# Science Bulletin, a miscellaneous series- of monographs describing original research

sn the life and earth sciences. This series was discontinued in 1978 with the issue of

Numbers 29 and 30; monographs are now published by the Museum in Contributions

in Science.

(_ !

• Science Series, long articles on. natural history topics, generally written for the layman.

I

Copies of the publications ip these series.' are sold through the Museum Book Shop. A catalog
available o.n request.

■ Ki

I f Hi ; m , HPIlRrTOtt T'. .. < ,, ir rr- nimir

The 1V1 iiseitpi. also publishes Tech nical Reports, & miscellaneous series containing information

Issue is authorized by the: museum's Scientific Publications Committee; however, manuscripts
do not receive anonymous peer review. Individual Technical Reports may be obtained from

the relevant ; Section of. the museum.

SCIENTIFIC PUBLICATIONS COMMITTEE

l\

SMb, r
PllllWlji

[giff

!;ShS

■ r

kiwll

Craig C. Black., Museum Director !

R ichard Cl Brusca

Daniel M. Cohen, Committee Chairman

John M. Harris

Charles L. Hogue .

*■' ■ ■ ' : : ■

George L. Kennedy

itf-p

Errol Stevens

ia&Ing Editor

ills

!. V)

if

PISSPPPPc

M "C .'Mew

llh

[ft' 1

HH ip

Spjitlll

li

f

tUlUL'.'tllfil

so f

lilill

1 m

ill !::!!:

i.'.f'lip

T v.

m

i \]yl ; . .}

, flrSSr reS

MWJi

III

jptejs

;L;i "■

ill!

AN EARLY MIOCENE PINNIPED OF THE GENUS
DESMATOPHOCA (MAMMALIA: OTARIIDAE)
FROM WASHINGTON

Lawrence G. Barnes1

ABSTRACT. The extinct otariid pinniped genus Desmatophoca
Condon, 1906, is represented by two rare sea lion-like species that
are relatively large and highly derived, in comparison with most
other contemporaneous fossil otariids. The type and only previously
described species of the genus is Desmatophoca oregonensis Condon,
1906, which was known only by the holotype skull and a question-
ably referred humerus from late Early to early Middle Miocene rocks
referred to the Astoria Formation in Oregon, U.S.A. Another partial
skull from the same formation is referred here to the same species.

A new species of Desmatophoca, D. brachycephala, has been dis-
covered in slightly older rocks of Early Miocene age, also referred
to the Astoria Formation, that are exposed on the north side of the
Columbia River in Washington, U.S.A. Desmatophoca brachyceph-
ala, like D. oregonensis, is known by a skull of a male. It is distin-
guished from the latter species by having a relatively wider skull
with a shorter rostrum, larger diameter canines, more derived cheek
tooth structures, wider interorbital region, and larger mastoid pro-
cesses. These and other differences make D. brachycephala a more
derived species than D. oregonensis, and indicate that the distinctive
otariid subfamily Desmatophocinae existed earlier in time than has
previously been documented and must have had a significant early
evolutionary history. The group is only known by fossils from the
eastern. North. Pacific margin. Contrary to some previously published
statements, there are enough substantial morphological differences
between the Desmatophocinae and another fossil subfamily, the
Allodesminae, to continue to classify them separately.

INTRODUCTION

Desmatophoca oregonensis Condon, 1906, was one of the
first fossil otariid pinnipeds described in the scientific liter-
ature, and was for many years also the geologically most
ancient one known. This sea lion-like animal has been dis-
cussed by many subsequent writers, and has figured in dis-
cussions of the origin of the family Otariidae (sensu lato;
including sea lions, fur seals, walruses, and fossil taxa), yet
it is still relatively poorly understood. It appears to be a rare
species and is known in the previously published literature
only by the holotype cranium (associated with an unde-
scribed mandible fragment and some incomplete postcranial
bones; see Packard and Kellogg, 1934:20) and a subsequently

Contributions in Science, Number 382, pp. 1-20
Natural History Museum of Los Angeles County, 1987

referred isolated humerus (Packard, 1947). These fossils are
all from rocks in the Newport Embayment that have been
referred to the Astoria Formation (late Early to early Middle
Miocene age), and are exposed on the coast of Oregon near
Newport, Lincoln County (Howe, 1926; Packard and Kel-
logg, 1934:5-19, fig. 1; Moore, 1964; Ray, 1976). I consider
the identity of the humerus to be tenuous, and shall not make
further reference to it in this study.

Desmatophoca oregonensis has been the subject of widely
differing taxonomic opinions. Condon (1906) believed that
it had features of the family Otariidae as well as of the Pho-
cidae (true seals). Wortman (1906) made the same claim. He
correctly interpreted the species as the most primitive fossil
pinniped then known, but gave its age as probably Oligocene
and, citing characters he interpreted as similar to those of
Patriofelis Leidy, 1870, reiterated his earlier (1894) theory
that the pinnipeds evolved directly from oxyaenid creodonts.
Kellogg (1922), while incorrectly concluding that D. oregon-
ensis was younger geologically than another early species,
Allodesmus kernensis Kellogg, 1922, stated unequivocally
(1922:62) that it was a true otariid and not related to phocids.
Packard and Kellogg (1934:24) later concluded that D. ore-
gonensis was geochronologically older than A. kernensis, a
currently held view (Barnes, 1972). Mitchell ( 1 966) suggested
that D. oregonensis was in some ways a suitable ancestor of
later true sea lions, or (1968:1888, fig. 16) of all the Otariidae
(sensu lato), but later, after the discovery of enaliarctines, he
considered it to be an early and aberrant offshoot of the
otariid lineage (Mitchell, 1975; and see Barnes, 1972:62).
Hay (1930), indicating a more distant relationship, named a
new family, the Desmatophocidae, to contain it. Simpson
(1945) suggested that Hay’s family might instead best be
regarded as a subfamily of the Otariidae, but he did not use
such a rank in his classification. Mitchell (1966) and Barnes

1. Section of Vertebrate Paleontology, Natural History Museum
of Los Angeles County, 900 Exposition Blvd., Los Angeles, Cali-
fornia 90007.

ISSN 0459-8113

(1972) both used the subfamily rank, but they included with-
in the subfamily Desmatophocinae the even more aberrant
and highly derived species of Allodesmus Kellogg, 1922. Re-
penning and Tedford (1977:74; see also Repenning, 1976;
King, 1983:129-130) recognized a separate family Desma-
tophocidae within the superfamily Otarioidea, but they re-
tained Allodesmus within it. Mitchell (1968, 1975) and Barnes
(1979, and in press) classified Desmatophocinae and Allo-
desminae as equal units, along with several other subfamilies
(both extant and extinct), within a single, broadly defined
pinniped carnivore family, the Otariidae (see Barnes, Dom-
ning, and Ray, 1985:table 1). It is in the latter taxonomic
context that I treat the subfamily Desmatophocinae in the
present study. The Recent sea lions, fur seals, and walruses
were also classified by Hall (1981) within one family, the
Otariidae, but he used the subfamily name Rosmarinae rath-
er than Odobeninae for the walruses, and included phocids
and otariids in the order Pinnipedia instead of the Carnivora.

Most of the various other fossil species, both named and
unnamed, that have been classified in the Desmatophocinae
by Mitchell (1966) and Barnes (1972), have subsequently
been reassigned to other subfamilies (Mitchell, 1975; Re-
penning and Tedford, 1 977; Barnes, 1 979, and in press). The
concept and content of the Desmatophocinae has thus be-
come much reduced to include only one named species, Des-
matophoca oregonensis.

A new fossil species belonging to the genus Desmatophoca
Condon, 1906, has recently been discovered in Washington,
and it is the purpose of this paper to describe and diagnose
it. The new specimen is from the Astoria Formation, which
is part of a sequence of rocks on the north side of the Co-
lumbia River that has been prospected extensively by J.L.
Goedert and G.H. Goedert. The type locality of the new
species of Desmatophoca has produced a diverse assemblage
of vertebrates, including other mammals and fishes. Among
these are at least two additional species of otariid pinnipeds,
but they are not yet known by enough material to be more
precisely identified. The invertebrates from the Lincoln Creek
Formation, which directly underlies the Astoria Formation
near Knappton, have been recently described by Zullo (1982 —
barnacles), Rigby and Jenkins (1983 — sponges), and Moore
( 1984 — mollusks). The latter author reviewed the history of
collecting and research in the Knappton area and described
the geographic and geologic setting (Moore, 1984:figs. 1, 2).

METHODS AND MATERIALS

The holotype of Desmatophoca brachycephala, new species,
was originally discovered in at least three separate sections
of a broken concretion, and some of its parts are still missing.
The different sections were joined with plastic resin, and the
same material was also used to fill some vacuities. At the
rostral extremity, remnants of the canines and incisors were
only tenuously attached by rock matrix to the rest of the
cranium. They now retain their original positions relative to
the mostly missing extremity of the snout only because of
this resin, which was poured in prior to removal of the sur-
rounding rock. For economy and to avoid jeopardizing the

integrity of the specimen, the rock was left within the right
orbit. Because of the crushing of the bullae and the hardness
of the matrix, it was decided not to attempt to open either
tympanic cavity of the holotype at this time.

In the restorations of the skulls (Figs. 3, 4b, 5, 7, 9), only
those parts that are preserved on at least one side of a skull
are shown in solid lines. All other missing parts are repre-
sented by dashed lines. The anatomical terminology used
here is adapted from that used by Howell (1928), Miller,
Christensen, and Evans (1964), Mitchell (1966, 1968),
Mitchell and Tedford ( 1973), Barnes (1972, 1979), and Re-
penning and Tedford (1977). Those skull measurements in
Table 1 which are the same as those that were defined by
Sivertsen (1954:18-20) are identified by the same numbers,
in brackets, that were given them by Sivertsen. Other mea-
surements are as defined by Barnes (1972:fig. 1; 1979:4-5).

A complete synonymy and an emended diagnosis of the
subfamily Desmatophocinae are given here. The latter is
partly based on characters listed by Mitchell (1968:1893-
1 894) and by Repenning and Tedford (1977: 10-1 1 , 74), with
appropriate modifications. Geochronologic ages of fossil pin-
nipeds cited herein are modified from those given by Re-
penning and Tedford (1977) and Barnes (1979) following the
revised radiometric scale of Dalrymple (1979), and the cor-
relations proposed by Addicott (1976), Ray (1976), Armen-
trout (1981), and Moore and Addicott (1987).

The acronym, LACM, is for the Natural History Museum
of Los Angeles County, Los Angeles, California. An author
and date in parentheses following a taxonomic name indi-
cates my use of that name at a different rank than originally
proposed. Millions of years ago (mega-annum) is abbreviated
ma. Anatomical abbreviations used in the illustrations are
explained as follows:

ac— alisphenoid canal

at— auditory tube (= musculotubular canal, including eus-
tachian tube)

Bo— basioccipital
Bs— basisphenoid

cc— carotid canal (posterior aperture)
earn— external acoustic meatus
fh— hypoglossal foramen
fi — incisive foramen (= palatine fissure)
fio— infraorbital foramen

fla— anterior lacerate foramen (joined with foramen rotun-
dum as an orbital fissure)
flp— posterior lacerate foramen
fo— foramen ovale
fop — optic foramen
fpal— palatine foramen

fpp— posterior aperture of palatine foramina
Fr— frontal

fsm— stylomastoid foramen
gf— glenoid fossa

hf— tympanohyal pit (= hyoid fossa)

Ju— jugal

mp— mastoid process
Mx— maxilla

2 Contributions in Science, Number 382

Barnes: Desmatophoca from Washington

Na— nasai
Oc— occipital
occ— occipital condyle
Pa— parietal
Pal — palatine
Pmx — premaxilla
Ps— presphenoid
Pt — pterygoid

pp— paroccipital (= jugular) process
Sq— squamosal
tb— tympanic bulla

SYSTEMATICS

Class Mammalia Linnaeus, 1758

Order Carnivora Bowdich, 1821

Infraorder Arctoidea Flower, 1869

Parvorder Ursida Tedford, 1976

Family Otariidae Gill, 1866

INCLUDED SUBFAMILIES. Enaliarctinae Mitchell and
Tedford, 1973; Otariinae (Gill, 1866); Desmatophocinae
(Hay, 1930); Allodesminae (Kellogg, 1931); Imagotariinae
Mitchell, 1968; Dusignathinae Mitchell, 1968; Odobeninae
(Allen, 1880).

Subfamily Desmatophocinae (Hay, 1930)
Mitchell, 1966

Desmatophocidae Hay, 1930:557, as a family of the suborder
Pinnipedia, order Carnivora, to include Desmatophoca.
Desmatophocinae (part). Mitchell, 1966:4, 39, 40; Barnes,
1972:5, as a subfamily of the family Otariidae, to include
Desmatophoca, Allodesmus, and less precisely identified
species, and, according to Mitchell (1966:39, 40), Dusi-
gnathus Kellogg, 1927, as well.

Desmatophocinae (Hay, 1930). Mitchell, 1968: 1839; Barnes,
1979:38, as a subfamily of Otariidae, exclusive of Allo-
desminae.

Desmatophocidae (part). Repenning and Tedford, 1977:10,
74-7 6, as a family of the superfamily Otarioidea, to include
Desmatophoca, Allodesmus, and less precisely identified
species.

EMENDED DIAGNOSIS OF SUBFAMILY. A subfam-
ily of the family Otariidae differing from Enaliarctinae by
having crania without an embayment in the lateral edge of
the basioccipital for a loop of the median branch of the
internal carotid artery and by lacking camassial teeth; dif-
fering from all other subfamilies except Allodesminae by
having tympanic crest not projecting into tympanic cavity,
and by having nasal bones elongate and tapering posteriorly
and inserted between frontals (character not yet determined
for Imagotariinae and Dusignathinae); differing from Ota-
riinae, Dusignathinae, Imagotariinae, and Odobeninae by
having large, elongate posterolaterally directed paroccipital

Table 1. Measurements of holotype skull, LACM 120199, of Des-
matophoca brachycephala, new species, in mm. Parentheses indicate
estimated measurements. Brackets indicate measurements explained
by Sivertsen (1954:18-20), and the method of taking the other mea-

surements follows Barnes (1972, 1979, in press).

Total length (283)

Anterior border of orbit to tip of snout (80)

External acoustic meatus to anterior border of orbit 1 35.0

Anterior border of orbit to tip of nasals (28)

Post-palatal length (palatal notch to basion) 121.3

Basion to anterior edge of zygomatic root [18] 188.0

Length of tooth row, C to M1 (88)

Length of tooth row, P'-M' 62.5

Width of rostrum across canines [12] (80)

Width of palate across base of I’ alveoli (34)

Width of palate across alveoli of P2 (64)

Width of palate across anterior alveoli of P4 (80)

Width between infraorbital foramina 65.0

Width across greatest interorbital constriction [6] 40.0

Width across supraorbital processes [7] 37.5

Width across greatest intertemporal constriction (30.0)

Width of braincase at anterior edge of glenoid fossa [8] (73)

Zygomatic width [1 7] 184.0

Auditory width [19] 142.0

Mastoid width [20] 168.0

Paroccipital width (126)

Greatest width across occipital condyles 69.8

Greatest width of anterior nares (35)

Greatest height of anterior nares (3 1 )

Width of zygomatic root of maxilla [14] 22.0

Greatest width of foramen magnum 32.1

Greatest height of foramen magnum 22.5

Least depth of jugal 12.0

Transverse diameter of infraorbital foramen 1 5.0

Anteroposterior diameter left C alveolus (26.5)

Anteroposterior diameter right P1 alveolus (12.8)

Anteroposterior diameter right PJ alveoli 12.2

Anteroposterior diameter right M1 alveolus 7.2

process which is separate from mastoid process and not joined
by a crest; differing from most Enaliarctinae, all Otariinae,
Imagotariinae, and Odobeninae by having pterygoid process
of maxilla enlarged so palate expands ventral to orbit (char-
acter not determinable for Dusignathinae); differing from
Allodesminae and Odobeninae by having vertical posterior
and medial crista on canine crowns; differing from Imago-
tariinae and Odobeninae by having hyoid fossa separated
from stylomastoid foramen by only a thin bridge of bone,
not joined by an elongate, narrow sulcus; differing from Ota-
riinae by having posterior lacerate foramen not greatly elon-
gate anteroposteriorly; differing from Allodesminae by hav-
ing more inflated tympanic bulla, posterior lacerate foramen
not expanded transversely, smaller orbit with anterior margin

Contributions in Science, Number 382

Barnes: Desmatophoca from Washington 3

flared anterodorsally instead of being retracted posteriorly,
unreduced incisive foramina, and squamosal-jugal contact
not greatly expanded dorsoventrally; and differing further
from Odobeninae by having canines which are not modified
as tusks.

TYPE AND ONLY INCLUDED GENUS. Desmato-
phoca Condon, 1906.

Desmatophoca Condon, 1906

Desmatophoca Condon, 1906:3.

EMENDED DIAGNOSIS OF GENUS. Identical with
that for the subfamily as given above until additional genera
are described.

TYPE SPECIES. Desmatophoca oregonensis Condon,
1906; type by original monotypy.

INCLUDED SPECIES. Desmatophoca oregonensis Con-
don, 1906, late Early Miocene and/or early Middle Miocene,
Oregon; and Desmatophoca brachycepha/a, new species. Ear-
ly Miocene, Washington.

Desmatophoca oregonensis Condon, 1 906

Figures 1, 9a

Desmatophoca oregonensis Condon, 1906:3, 1 1, figs. 1-3 on

p. 9, unnumbered figures on pp. 3, 10.

HOLOTYPE. University of Oregon (Eugene), Museum of
Natural History (UOMNH) F735, cranium, partial dentary,
and postcranial bones.

TYPE LOCALITY. UOMNH locality 1153, coast near
Astoria, Lincoln County, Oregon. The precise collecting lo-
cality of the holoytpe is not known, except that it apparently
was west of Newport in the sea cliff where the Astoria for-
mation is exposed (Packard and Kellogg, 1934:20, fig. 1).

REFERRED SPECIMEN. LACM 123285, the anterior
part of a cranium with right canine, P2-3, and left I3, canine,
and P1-3; collected by D.J. Martel, 18 April 1983.

LOCALITY OF REFERRED SPECIMEN. LACM 4851,
among boulders on beach midway between the mouths of
Schooner and Moloch creeks, approximately 6.5 km north
of Newport, Lincoln County, Oregon, as shown on Yaquina,
Oregon, United States Geological Survey topographic map,
1 :62,500 scale, 1957 edition. Approximately 44°41'45" north
latitude and 124°03'55" west longitude. This locality is ap-
proximately 0.75 km north of Schooner Point, which was
the collecting locality of a referred specimen of the fossil
mysticete, Cophocetus oregonensis Packard and Kellogg, 1934
(see Packard and Kellogg, 1934:21, fig. 1, item 6).

FORMATION AND AGE. Astoria Formation, late Early
Miocene and/or early Middle Miocene. The rocks that have
been referred to the Astoria Formation which crop out on
the Oregon coast in Lincoln County, near Newport (Packard
and Kellogg, 1934:20, fig. 1; Ray, 1976:fig. 2), produced the
mollusks that were used to characterize the Newportian Stage
(Addicott, 1976:102, 104, fig. 4). The holotype of D. ore-
gonensis was found at an unspecified horizon within the As-
toria Formation, and the referred rostrum (LACM 123285)

was found at a location near the sea cliff outcrop of the “Iron
Mountain bed,” a distinctive horizon within the Astoria For-
mation which produced the dome-skulled chalicothere fossil
that was described by Munthe and Coombs (1979). The land
mammals and other lines of evidence derived from this same
coastal Oregon part of the Astoria Formation provide a cor-
relation with the Hemingfordian or the early part of the
Barstovian North American land mammal ages (Ray, 1976:
fig. 2; Munthe and Coombs, 1979:78-80). The concordance
between these mammal ages and the Newportian Stage rep-
resents an interval of time spanning from approximately 19
ma to 15 ma and from the late part of the Early Miocene
through the early part of the Middle Miocene (Armentrout,
1981). This then is also the best current estimate of the geo-
logic age of Desmatophoca oregonensis.

DISCUSSION. Condon’s (1906) original publication (see
also Condon, 1910) on Desmatophoca oregonensis includes
a brief description, photographs of the dorsal and right lateral
surfaces of the partially prepared holotype cranium, and
drawings of the PJ. These have been greatly supplemented
by additional observations by Wortman (1906), Downs (1956:
124-125), Mitchell (1966:36-37; 1968:1883, 1893-1894, ta-
ble V; 1975:12-14), Barnes (1972:63; 1979:35), and Repen-
ning and Tedford (1977:74). E. Mitchell is supervising further
preparation of the holotype. He intends to publish a detailed
reassessment of the species (Mitchell, 1975:12), and has al-
ready presented illustrations that show the major cranial
characters of the holotype (Mitchell, 1966:pl. 29; 1975:fig.
2). The U.S. National Museum of Natural History has pro-
vided casts of the holotype of D. oregonensis.

The recently collected referred specimen of Desmatophoca
oregonensis (Fig. 1 ) closely matches the size and morphology
of the holotype, and yields additional information on the
structure of the P1, the P3, and of the posterior part of the
palate. Both of the presently available specimens appear to
represent male individuals because of the relative sizes of
their canines and the development of bony rugosities and
tubercles. All of the sutures between the bones of the referred
snout remain unfused, but the individual had attained the
same size as the holotype specimen at the time of its death.
Therefore, the referred specimen must represent a young
adult individual. This new specimen helps to confirm as
diagnostic for the taxon certain characters of the holotype
as: rostrum parallel-sided; cranium high between orbits and
sloping anteriorly toward nasal bones; nasals elongate and
slightly depressed medially, with closely appressed and ta-
pering posterior ends extending posteriorly to a point dorsal
to the middle of the orbit; narial opening sloping gently and
narrower ventrally than at the dorsal part; presence of a slight
fossa on the anterior surface of the rostrum on the maxilla-
premaxilla suture dorsal to the diastem between the I3 and
the canine; zygomatic portion of the jugal short and thick,
mortised by relatively elongate splints of bone dorsolaterally
and ventromedially with corresponding dorsomedial and
ventrolateral splints from the maxilla; palate flat posteriorly
but slightly excavated anteriorly, bearing scattered palatine
foramina, of which the anterior one on each side is the largest;
canine crown slightly recurved, covered by slightly irregular

4 Contributions in Science, Number 382

Barnes: Desmatophoca from Washington

c

Figure 1. Desmatophoca oregonensis Condon, 1906, referred specimen, rostrum, LACM 123285, LACM locality 4851; a, dorsal view; b,
left lateral view; c, ventral view.

enamel surface, and bearing a vertical crista posteriorly and
a less prominent one medially; and cheek teeth with prom-
inent central cusps and lingual cingulae with small cuspules.
The P' has a single, cylindrical root, and on LACM 123285,

its crown has both anterior and posterior cristae. The cin-
gulum is most prominent on the posteromedial side of the
crown and extends slightly around the posterolateral side of
the crown. The root of P2 is clearly bifid on the holotype.

Contributions in Science, Number 382

Barnes: Desmatophoca from Washington 5

but on LACM 123285 it is bilobed with a deep vertical sulcus
on the lateral side. On both specimens, the P3 has two dis-
tinctly separate roots, of which the posterior one is bilobed.
The tooth is present on both sides of LACM 123285, and
its central cusp is high, broad, triangular, and slightly re-
curved posteromedially. The cingulum is prominent, extends
around the medial and posterior sides of the crown, and
bears, variably, eight to nine cuspules, of which the most
prominent is at the posterior side of the tooth in line with
the posterior crista on the crown. On both specimens, the PJ
has two separate roots, of which the posterior one is larger
and transversely bilobed as on the P3. The roots of M1 on
both specimens have patterns similar to those of P3-4, except
that they are smaller and the bilobed posterior root is oriented
obliquely.

The referred specimen shows that the species has an M2
with a single root, round in cross section, and that the pter-
ygoid process of the maxilla, at the posterolateral comer of
the palate, is extensive and thin beneath the orbit. These
features of the referred specimen have been incorporated into
a modified restoration of the ventral view of the skull of the
species (Fig. 9a). The lateral borders of the external narial
opening (comprised of the premaxillae) are more rounded
on the referred specimen than on the holotype. The zygo-
matic arch of the referred specimen is not complete enough
to provide any additional information on the structure of the
squamosal-jugal articulation, which, as shown by a plaster
impression on the holotype, was slightly expanded dorso-
ventrally.

Desmatophoca brachycephala , new species

Figures 2-8, 9b

DIAGNOSIS OF SPECIES. A species of Desmatophoca
differing from D. oregonensis by having cranium with shorter
and wider rostrum, which is more expanded laterally around
larger canines; ventral part of external narial opening wider;
interorbital region wider, especially in anterior part; zygo-
matic arch more slender; optic foramina located more pos-
teroventrally within orbits; palate wider, especially in pos-
terior part and having larger pterygoid process beneath orbit;
M1 with only one small, bilobed root rather than two separate
roots; M2 absent; external acoustic meatus wider and directed
more laterally; mastoid process larger and extended more
posterodorsally; paroccipital process directed more laterally,
instead of posteriorly; posterior lacerate foramen larger and
more circular in outline.

HOLOTYPE. LACM 120199, incomplete cranium with
crowns of left I1-3, parts of both canines, lacking other teeth
and parts of the rostrum, and the right dorsolateral part of
the braincase, collected by J.L. Goedert and G.H. Goedert
in 1979.

TYPE LOCALITY. LACM 4584 (= LACM Invertebrate
Paleontology [LACMIP] locality 5864), east of Knappton,
Pacific County, Washington.

FORMATION AND AGE. The type locality of Desma-
tophoca brachycephala near Knappton is in the lower part
of a marine rock unit which has been referred to the Astoria

Formation and is Early Miocene, but not earliest Miocene,
in age, is correlative with the Pillarian Molluscan Stage, the
Vertipecten fucanus Molluscan Zone, indirectly correlated
with the Saucesian Foraminiferal Stage, and with either the
late part of the Arikareean or the early part of the Heming-
fordian North American land mammal ages, and therefore
is probably between approximately 20 and 23 million years
old. This horizon is stratigraphically above the Eocene to
earliest Miocene age, marine, Lincoln Creek Formation which
crops out downdip and to the west along the Columbia River
shoreline. The bases for these determinations are as follows.

The type section of the Astoria Formation is at Astoria,
Oregon, directly across the Columbia River south of Knapp-
ton, Washington. In her study of the mollusks from the upper
part of the Lincoln Creek Formation near Knappton, Moore
(1984:1) commented on the age of the Astoria Formation,
which directly overlies the Lincoln Creek in this sequence of
strata (see also Wells, 1979). Moore also explained that as
fossils weather out of exposures of these two formations in
the bluffs along the Columbia River, they retain their ap-
proximate stratigraphic positions on the beach. This phe-
nomenon is clearly demonstrated by the fact that, over a
period of several months, and from a fairly small area, J.
and G. Goedert were able to assemble the holotype skull of
D. brachycephala, as well as other fossils, from broken and
extremely weathered parts of concretions.

The type locality of D. brachycephala (= LACMIP locality
5864) is approximately one-half km southeast along the
north shore of the Columbia River from LACMIP locality
5863, which is shown on the map in Moore’s publication
(1984:fig. 2). Owing to the dip of the rocks here, the two
localities are nearly along strike from one another, and if
there is any difference in their stratigraphic positions, the
type locality of D. brachycephala is possibly only slightly
higher in the section than LACMIP locality 5863. These
localities are low in the Astoria Formation, have produced
fossil mollusks and, based primarily on the occurrence of the
bivalve Acila ( Acila ) gettysburgensis, are assigned to the Pil-
larian Molluscan Stage (Moore, 1984:1, figs. 2, 3; written
communication, 13 August 1986). The mollusk, Vertipecten
fucanus, which characterizes the Pillarian Stage (Addicott,
1976), has not yet been found at the type locality of D. bra-
chycephala, however. The boundary between the Pillarian
Stage and the younger Newportian Stage in the Astoria For-
mation near Knappton, although not yet located biostrati-
graphically, is presumably north of these localities, and higher
stratigraphically within the formation. Much of the type sec-
tion of the Astoria Formation on the south side of the Co-
lumbia River is also referable to the Pillarian Stage (Addicott,
1976:101).

There are some minor conflicts regarding the epochal age
designation of these rocks. Following Addicott (1976) and
Moore (1984), the Pillarian Stage and the localities in the
base of the Astoria Formation near Knappton would fall in
the later part of the Early Miocene, but following Armentrout
(1981) and Moore and Addicott (1987) they would represent
the early part of the Early Miocene.

Regardless of this, the entirety of the Pillarian Stage is older

6 Contributions in Science, Number 382

Barnes: Desmatophoca from Washington

than the Newportian Stage by definition (Addicott, 1976:
102, 104, fig. 4). Therefore, the section of the Astoria For-
mation near Knappton that yielded D. brachycephala is older
than the rocks that are referred to the Astoria Formation near
Newport on the Oregon coast, and which produced D. ore-
gonensis and the mollusks that were used to characterize the
Newportian Stage (see also Addicott, 1976:104). The out-
crops of the Astoria Formation near Knappton in Washing-
ton that yielded D. brachycephala and which contain mol-
lusks of the earlier Pillarian Molluscan Stage may, therefore,
represent the early part of the Early Miocene, but are younger
than the underlying earliest Miocene part of the Lincoln Creek
Formation, are indirectly correlated with the Arikareean
North American Land Mammal Age, and may, therefore, be
between 20 and 23 million years old. This age determination
is most in accordance with the correlations that were pro-
posed by Armentrout (1981).

ETYMOLOGY. The species name, brachycephala, is de-
rived from Greek; brachys, short, and kephale, head; and is
in reference to the short snout of this species as compared
with Desmatophoca oregonensis.

DESCRIPTION AND COMPARISONS. The holotype
of Desmatophoca brachycephala consists of a nearly complete
cranium that was assembled from several pieces of broken
fossiliferous rock. These pieces were found within a localized
area. From the effects of predepositional erosion and post-
depositional breakage, the specimen is lacking all of the cheek
teeth, parts of the right and left sides of the palate, parts of
the premaxillae, parts of the incisors and canines, the right
paroccipital process, and the central and right side of the
upper surface of the braincase. It appears to have lain on the
seafloor for some time prior to final burial and fossilization,
during which time organic and/or inorganic factors caused
the erosion of much of the bone surface, principally on the
dorsal surface, and the loss of the medial, bony walls of both
orbits, the distal ends of the nasal bones, and parts of the
nuchal crest. Subsequent tectonic distortion compressed the
braincase, the interorbital region, and the tympanic bullae.
The holotype of D. brachycephala includes structures of the
pterygoid hamuli and the zygomatic arches that are not known
for D. oregonensis. All other parts of the cranium of D.
brachycephala are directly comparable with corresponding
parts on the holotype and the referred specimen (LACM
123285) of D. oregonensis.

The holotype of D. brachycephala represents an individual
in the adult or Group I age class extrapolating from the cranial
suture closure method adopted by Sivertsen (1954: 10-13) to
determine ages of Recent specimens. Of the nine sutures that
he found useful for age evaluation, at least seven are closed
or mostly closed on the fossil. Because most of the roof of
the braincase is missing, only a short section of the inter-
parietal suture is preserved, but it appears to be completely
closed. The original degree of closure of the squamosal-pa-
rietal suture is difficult to determine. Its course is partly vis-
ible on the left side, but distortion has caused the braincase
to break along much of that suture. The interfrontal suture
is closed, although it appears to be open in the photograph
(Fig. 2) because, between the orbits, it is flanked on each side

by an elongate ridge of bone (Fig. 3). The exact suture age
of the holotype following Sivertsen’s method is therefore not
determinable, but it was at least 28. Recent otariids with
suture ages of between 19 and 36 are adults (Group I) ac-
cording to Sivertsen’s method.

In addition to the shared diagnostic characters listed for
the subfamily (and for the genus), D. brachycephala and D
oregonensis are demonstrably congeneric because they share
the following cranial characters: narrow interorbital region
lacking supraorbital processes; interfrontal suture between
the orbits flanked by low, elongate ridges; low-vaulted brain-
case; wide squamosal fossa between the braincase and the
zygomatic arch; laterally flaring mastoid process; anterior
aperture of the infraorbital foramen slightly overhung by the
anterior edge of the orbit; rostrum wider at the canines than
at P3; palate relatively flat and expanded posterolaterally ven-
tral to the orbit; more than one posterior palatine foramen
on each side, rather than only a single one connected to an
elongate and well-developed palatine sulcus; glenoid fossa
with a large anteroventrally projecting postglenoid process
and small, ventrally directed preglenoid process at the lateral
edge only; tympanic bulla relatively flat, with tuberosities
limited to the anteromedial part; broad concavity between
the middle part of the tympanic bulla and the postglenoid
process bearing small postglenoid foramen near the postgle-
noid process; thin but wide shelf of squamosal bone pro-
jecting laterally from the braincase dorsal to the external
acoustic meatus; external acoustic meatus entering the bulla
in a posteromedial direction; mastoid process approximately
cubic in shape, but excavated posteriorly and with a rugose
surface facing ventrolaterally; paroccipital process projecting
posterolaterally but slightly deflected medially at its termi-
nation, excavated ventrally and with a thin anterolateral edge;
basioccipital expanded posteriorly, with raised muscular tu-
bercles adjacent to the tympanic bullae, and with a narrow
median pharyngeal tubercle which is flanked by relatively
deep, hemispherical fossae; hypoglossal foramen small and
close to the posteromedial side of the posterior lacerate fo-
ramen and with its aperture facing antero-ventrolaterally;
occipital condyles projecting prominently from occipital shield
with very convex articular surfaces and separated ventrally
by a deep intercondylar notch; and foramen magnum wide
and low with a thick dorsal margin.

Most of the cranial morphology of D. oregonensis has now
been documented but, because the descriptions are varied
and widely scattered in the literature, a relatively detailed
account of the cranial morphology of D. brachycephala is
warranted here. As in D. oregonensis, the rostrum expands
gradually anteriorly from its narrowest point in the cheek
region around P3, immediately anterior to the infraorbital
foramina. The rostrum of D. brachycephala is relatively
shorter than that of D. oregonensis, considering that the cra-
nia of the two species have approximately the same zygo-
matic widths. The external narial opening is widest ventrally
and narrower at its apex, the reverse of the condition in D.
oregonensis. That part of the premaxilla forming the lower
lateral border of the narial opening is wide and rounded, as
on the referred specimen of D. oregonensis (LACM 123285,

Contributions in Science, Number 382

Barnes: Desmatophoca from Washington 7

8 Contributions in Science, Number 382

Barnes: Desmatophoca from Washington

Figure 2. Desmatophoca brachycephala, new species, holotype, skull, LACM 120199, LACM locality 4584, dorsal view.

u

Contributions in Science, Number 382

Barnes: Desmatophoca from Washington 9

Figure 3. Desmatophoca brachvcephala, new species, restoration of skull based on holotype, LACM 120199, dorsal view. Abbreviations used in this and following
illustrations are explained in the Methods and Materials section.

10 Contributions in Science, Number 382

Barnes: Desmatophoca from Washington

Figure 4. Desmatophoca brachycephala, new species, holotype, skull, LACM 120199, LACM locality 4584, left lateral view; a, photograph of original specimen;
b, restoration.

( \
S- — '

o

T3

u

<

3

P.

a -e

^ t-

-O o
<3 a

*u <u

Q 3

Ifi £;
0, tfl

9 5

as o

E -S

Fig. la), not formed into a narrow crest as on the holotype
of D. oregonensis. This feature may be variable, therefore,
rather than of taxonomic significance. The narial apertures
of the incisive foramina are elongate anteroposteriorly and
closely appressed, but shorter and more divergent posteriorly
than in D. oregonensis.

The nasals of D. brachycepha/a are narrower anteriorly
than those of D. oregonensis. Their combined form is of a
tapered wedge extending posteriorly between the frontals, but
the sutures at that location are indistinct, and the posterior
margins of the nasals cannot be exactly determined. On either
side of the nasal bones, the premaxilla-maxilla sutures are
obliterated by fusion.

At the anterior margin of the orbit, the dorsal margin of
the zygomatic arch flares anteriorly over the anterior opening
of the infraorbital foramen and forms a “cup” for the eyeball,
much as in the enaliarctines and otariines. This structure is
very different from the posteriorly receding dorsal margin of
the zygomatic arch in species of Allodesmus. The dorsal mar-
gin of the zygomatic arch in D. brachycepha/a is narrower
and projects farther anteriorly than in D. oregonensis.

The interorbital region of D. brachycepha/a is wider and
shorter (measuring approximately 20 mm less, anteropos-
teriorly, between the braincase and the anterior border of the
orbit) than in D. oregonensis. The sulci on either side of the
braincase, which mark the locations of the pseudosylvian
sulcus on the brain of D. oregonensis (a character also present
in Enaliarctos mea/si Mitchell and Tedford, 1973), are much
shallower and less distinct in D. brachycepha/a. There are
also small fossae on either side of the midline on the anterior
dorsal part of the braincase in D. oregonensis. These are
probably homologues of the parasagittal fossae (Mitchell and
Tedford, 1973:225) of Enaliarctos mea/si, and their presence
or absence cannot be determined in D. brachycepha/a due to
breakage. The nuchal crest is broken off in the holotype of
D. oregonensis and, therefore, not known for that species.
That part of the crest which is preserved on the holotype of
D. brachycepha/a is thick and projects posteriorly over the
occipital shield nearly as far as the posterior surfaces of the
occipital condyles (Figs. 2, 4a).

The squamosal fossa is the broad recess in the squamosal
between the braincase and the zygomatic arch, dorsal to the
glenoid fossa, and which floors the temporal fossa. It appears
to be characteristically wide in species of Desmatophocinae
compared with most other otariids, and is relatively wider
in D. brachycepha/a than it is in I), oregonensis. Its anterior
part, between the anterolateral corner of the braincase and
the zygomatic arch, is nearly flat and creates a bony shelf
over the anterior part of the glenoid fossa (Fig. 3). The lateral
expansion which affected this part of the cranium of D.
brachycepha/a also involved the mastoid process, which
therefore has a much more extensive, flat dorsal surface than
in D. oregonensis.

The complete zygomatic arch has not been previously
known for the genus, because the structure is missing on the
holotype of D. oregonensis. This has been the source of some
speculation, particularly because the zygomatic arches of the
species in the subfamily Allodesminae are greatly modified.

Contributions in Science, Number 382

Barnes: Desmatophoca from Washington 11

12 Contributions in Science, Number 382

Barnes: Desmatophoca from Washington

Figure 6. Desmatophoca brachycephala, new species, holotype, skull, LACM 120199, LACM locality 4584, ventral view.

f sm

Contributions in Science, Number 382

Barnes: Desmatophoca from Washington 13

Figure 7. Desmatophoca brachycephala, new species, restoration of skull based on holotype, LACM 120199, ventral view.

A matrix impression (see Condon, 1906:3) of the missing
part of the right zygomatic arch of the holotype of D. ore-
gonensis shows the zygomatic process of the squamosal to
be slightly expanded dorsoventrally, but it appears to be
basically of the primitive type (Mitchell, 1966:pl. 29d; 1968:
1893; 1975:fig. 2). The right jugal-squamosal junction is
completely preserved on the holotype of D. brachycephala,
and is close to the generalized, primitive carnivore type. The
zygomatic process of the squamosal tapers anteriorly to a
blunt point which abuts the posterior side of a tapered, dor-
sally projecting postorbital process of the jugal (Fig. 4b). In
the primitive carnivore condition (as in enaliarctines and
otariines), the squamosal does not abut the postorbital pro-
cess of the jugal. Overall, the zygomatic arch of D. brachy-
cephala is relatively thick, although not quite as thick as in
the preserved comparable parts of the holotype of D. ore-
gonensis. The condition in D. brachycephala is more prim-
itive than that in D. oregonensis, and neither species has the
exaggerated vertical expansion of the squamosal and jugal at
their junction which is so characteristic of the species of
Allodesminae.

The palate of D. brachycephala is flatter and broader pos-
teriorly, but narrower and more arched anteriorly than that
of D. oregonensis (Fig. 9). The incisive foramina of D. brachy-
cephala are also more divergent posteriorly and recessed more
deeply into the palate. The largest posterior palatine foramen
on each side opens into an irregularly shaped, elongate pal-
atine sulcus. This pair of sulci differs from those which are
characteristic of the species of Enaliarctos, which owe their
prominence to being sharply defined, parallel, nearly sym-
metrical, and to not being flanked by other, smaller foramina
on the palate. In D. brachycephala. as in D. oregonensis, each
of the large posterior palatine foramina is associated with
two or more smaller ones. The expansive pterygoid process
of the maxilla beneath the orbit of D. brachycephala is even
larger than in D. oregonensis. It forms a thin infraorbital shelf
with a prominent comer. This structure is not as prominent
as in the Early Miocene enaliarctine, Pinnarctidion bishopi
Barnes, 1979, but it is more prominent than in D. oregonensis
and in the Middle Miocene allodesmine, Allodesmus pack-
ardi Bames, 1972.

The crowns of the incisors of D. brachycephala are aligned
transversely at the front of the palate (Figs. 6, 7). Some dis-
tortion may have occurred during fossilization, however, be-
cause much of the crown of I1 is actually farther posterior
than the two lateral ones, but the root of this tooth is not
present to confirm its exact original alignment. Both I1 and
I2 have crowns with their apices worn down to flat surfaces.
The crown of I1 is one-half the diameter of I2. More of I1 is
preserved than of the two previous teeth. The tooth is very
procumbent and its root is oval in cross section. The root is
not entirely preserved, but it measures at least 1 1 mm trans-
versely and 1 5 mm anteroposteriorly in cross section at the
alveolar rim. The crown shows neither cingulae nor cusps as
preserved because it has an extensive posterolabial wear facet
and its apex is broken off.

Neither canine is complete. All that remains of the right
one is the lingual half of the root within the alveolus and.

projecting from the alveolus, is the lingual part of the root
at the base of the enamel. The left one is more complete.
The lingual side of each canine root has a shallow longitudinal
sulcus. The canine of D. brachycephala must have been oth-
erwise essentially circular in cross section, and at least 25
mm in diameter at the alveolar rim. I estimate from the
remnant of the left canine and the curvature of its alveolar
margin that its diameter there was approximately 20 percent
greater than in D. oregonensis.

There are no cheek teeth preserved on the holotype of D.
brachycephala. The alveolus for P1 is located posterolingual
to the canine, and is closely appressed to it, so much so that
the alveoli of these two teeth are partly contiguous and ex-
tremely procumbent (derived conditions). In D. oregonensis,
this alveolus is more separate from the canine and posterior
to it. The lingual side of the alveolus of P1 has a faint crest,
indicating that the root of this tooth had a slight longitudinal,
lingual sulcus. The root was otherwise nearly circular in cross
section and approximately 10 mm in diameter.

The areas that were occupied by the alveoli for P2 and P3
have been broken away on both sides of the skull (Fig. 6),
and there is no way to determine the morphologies of these
teeth. The space that they occupied on the right side, between
the alveoli for P1 and P4, spans only 25 mm anteroposteriorly.
This is nearly 1 5 mm less than the corresponding space on
the holotype of D. oregonensis, so these two teeth in D.
brachycephala must have been more crowded together (Fig.
9). Judging from its alveoli, the root morphology of P4 in D.
brachycephala is nearly the same as in D. oregonensis. A
separate anterior root on this tooth occupied a roughly four-
sided alveolus. The primitively separate two posterior roots
were fused into a single, transversely bilobed structure (Fig.
7), of which the medial lobe is the homologue of the separate
root that is above the protocone in fissipeds and primitive
enaliarctine otariids (see Mitchell and Tedford, 1973; Bames,
1979). In D. brachycephala, the P4 is located more anteriorly
relative to the zygomatic arch than in D. oregonensis. The
anterior and posterior alveoli of this tooth in D. brachy-
cephala are closely appressed, indicating that the P4 roots
were approaching the stage of fusion (derived character). A
diastem of approximately 6 mm, such as is also present in
D. oregonensis, separates the alveoli of P4 and M1.

The single M1 alveolus of D. brachycephala measures 8
mm anteroposteriorly by 6 mm transversely. It is bilobed,
clearly reflecting the root structure of the M1. The posterior
lobe of the root of this tooth was larger than the anterior one.
In contrast, as noted previously, the more primitively con-
structed M1 alveolus of D. oregonensis is double, clearly in-
dicating that the same tooth in that species had two separate
roots (Fig. 9). Of these, the anterior root was small and round,
and the posterior one was larger and expanded (bilobed)
transversely.

An M2 was never present in D. brachycephala, in contrast
to D. oregonensis, in which its presence is indicated by an
essentially circular alveolus, approximately 7 mm in diam-
eter, on the referred snout (LACM 123285). This loss of M2
in D. brachycephala is possibly an individually variable char-
acter, of course, but if a larger population sample in the future

14 Contributions in Science, Number 382

Barnes: Desmalophoca from Washington

should confirm that it is diagnostic, then it is a derived feature
of D. brachycephala. Some individuals of living species of
Otariinae experience pathologic or fortuitous loss of the M3,
but for species in other subfamilies, for example Allodestnus
packardi and Recent walruses, Odobenus rosmarus (Lin-
naeus, 1758), similar tooth loss is an example of convergence.

The width of the pterygoid of D. brachycephala at the level
of the pterygoid hamulae is enhanced by a lateral protuber-
ance, which is also present in most otariids except the otar-
iines and Enaliarctos mealsi. The pterygoid hamulus is rel-
atively thick, short, and bends ventrolaterally. There is a
broad concavity between it and the above-mentioned lateral
protuberance.

Most parts of the medial walls of both orbits of the holotype
of D. brachycephala were broken away and/or decomposed
prior to fossilization, and much anatomical information has
therefore been lost. No lacrimal foramen is detectable on
either the holotype or the referred specimen (LACM 123285)
of D. oregonensis. Enough of the bone forming the anterior
margin of the orbit remains on the holotype skull of D.
brachycephala to demonstrate the absence of the same fo-
ramen in that species as well. In both species there is a small
nutrient foramen in the orbit just ventromedial to the orbital
aperture of the infraorbital foramen, and, located more pos-
teriorly, is the posterior aperture of the canal leading to the
posterior palatine foramina (Fig. 5). Posterodorsal to the lat-
ter, on the referred specimen of D. oregonensis is a round
sphenopalatine foramen, but the holotype of D. brachyceph-
ala is too damaged in this area to ascertain the location of
such a structure.

The area of the orbital aperture of the optic foramen is
likewise damaged in D. brachycephala, but judging by the
curvature of the surrounding bone (Fig. 5), it appears to have
been located more posteroventrally (derived character) than
it is in D. oregonensis. Compared with D. oregonensis, the
anterolateral corner of the braincase of D. brachycephala
protrudes farther laterally, as does the pterygoid ventrolateral
to the alisphenoid canal, resulting in the creation of a broader,
more recessed orbital fissure, which contains the anterior
lacerate foramen, the foramen rotundum, and anterior end
of the alisphenoid canal. In the posteroventrally recessed
position of the orbital fissure as well, D. brachycephala is
again the more derived of the two species.

Desmatophoca oregonensis and D. brachycephala both have
the aperture of the foramen ovale recessed in an elongate
fossa that is confluent with the posterior end of the ali-
sphenoid canal. This fossa is separated from the medial part
of the glenoid fossa by an oblique strut of bone extending to
the pterygoid. In both species, the posterior end of the ali-
sphenoid canal is located at a point adjacent to the middle
of the glenoid fossa. This is a relatively posterior location
(derived) when compared with the anatomy of primitive car-
nivores, and with that of enaliarctine and otariine pinnipeds,
but is not as far posterior as in species of Allodestnus (see
Barnes, 1972), which have the most derived condition among
the otariids.

The glenoid fossa of the squamosal is approximately 45
mm wide and is bordered posteriorly by a prominent, but

relatively thin postglenoid process which projects antero-
ventrally and is widest medially. In the lateral part of the
glenoid fossa, the anterior border is deflected ventrally to
create a small preglenoid process. The medial edge of the
glenoid fossa, as in Allodestnus spp., is marked by a sulcus
that curves posteriorly into a small pit, which is probably
the anterior end of the canal for the chorda tympani nerve.
The lateral ends of the glenoid fossae are canted anterolat-
erally on the holotype skulls of both species of Desmatopho-
ca, but with the present sample, it is not possible to assess
whether this is a diagnostic character for the genus or whether
it is individually variable as in Allodestnus kernensis (see
Barnes, 1 972). The glenoid fossae on the skulls of both species
of Desmatophoca are not as rectangular in shape as in species
of Allodestnus. They enclose nearly 180 degrees of rotational
movement of the mandibular condyle by virtue of the prom-
inent postglenoid process and the smaller preglenoid process
laterally. The position of the glenoid fossa on the cranium
of D. brachycephala is farther lateral relative to the tympanic
bulla than in D. oregonensis (Fig. 9).

The bulla is fused to the postglenoid process over a broad,
concave area, at the anterior edge of which is a very small,
vestigial postglenoid foramen. Both bullae are damaged on
the holotype of D. brachycephala, but the right one is the
most complete. It is prolonged anteromedially in the form
of irregular styliform processes that extend ventral to the
auditory tube (which in life contained, in part, the eustachian
tube), the median lacerate foramen, and the anterior end of
the carotid canal. A transverse crest on the surface of the
bulla extends from this anteromedial part of the bulla to a
point ventral to the external acoustic meatus. Anterior to this
crest, the bulla surface slopes anterodorsally, but posteriorly
it is nearly flat. As in species of Allodestnus, and as in En-
aliarctos mealsi and D. oregonensis, the posteromedial comer
of the bulla is retracted ventral to the posterior end of the
carotid canal, but unlike those species, the bulla is also re-
tracted dorsal to the end of the same canal as well. The effect
of this is an enlarged anteromedial part of the posterior lac-
erate foramen (Fig. 7).

The posterior lacerate foramen is approximately 50 percent
larger than in D. oregonensis (Fig. 9), and in both species the
foramen has an ovoid shape, although it is not as elongate
as in Enaliarctos mealsi or in species of Otariinae. As is
apparently characteristic of the genus Desmatophoca, the pit
(hyoid fossa) for attachment of the tympanohyoid is not very
close to the posterior lacerate foramen (as it is in Allodestnus
spp.), but instead is positioned on the posterolateral side of
the bulla, close to the stylomastoid foramen, and between
the bulla and the mastoid process. The tympanohyal pit is
separated from the small stylomastoid foramen by only a
small crest of bone, but it is separated from the posterior
lacerate foramen by a broad, convex surface.

The external acoustic meatus of D. brachycephala is wider
and directed more laterally than it is in D. oregonensis. This
is a more primitive morphological state in D. brachycephala.
The mastoid process is relatively and absolutely larger (de-
rived character), than in D. oregonensis and is more rounded
in shape (Fig. 9). In contrast to D. oregonensis, the process

Contributions in Science, Number 382

Barnes: Desmatophoca from Washington 15

descends ventrally well below the ventral surface of the tym-
panic bulla, and projects farther dorsolaterally and is con-
tinuous with a prominent but narrow extension of the nuchal
crest (Fig. 4). The rugose ventrolateral face of the mastoid
process is larger than in D. oregonensis.

The paroccipital process differs in both its orientation and
shape from that of D. oregonensis. This process on both
species of Desmatophoca resembles those of Allodesmus spp.
only insofar as this process is large and is separated from the
mastoid process by a wide notch. Otherwise, the paroccipital
process of D. brachycephala (and of D. oregonensis ) differs
from those of species of Allodesmus by projecting more pos-
teriorly, not posteroventrally, by being much more flattened
dorsoventrally, and by bending slightly medially at the pos-
terior extremity. There is no fossa present on the ventral
surface of the skull between the paroccipital process and the
tympanic bulla as in species of Enaliarctinae and Otariinae,
but the bone surface is instead virtually flat and, in D. brachy-
cephala. it is even slightly convex when compared with D.
oregonensis.

The basioccipital and basisphenoid expand posteriorly in
both species of Desmatophoca. and the widest part of the
basioccipital is between the tympanic bullae (Fig. 9). The
holotype skulls of both species are virtually the same width
between the bullae. The basioccipital is proportionally wider
at this point in D. brachycephala than in D. oregonensis,
however, and the appearance of this is enhanced by the bas-
ioccipital being medially constricted posteriorly between the
larger posterior lacerate foramina and being more tapered
anteriorly toward the basisphenoid. Both species have very
prominent fossae for insertion of the rectus capitus longus
muscles anterior to the condyles, but in D. brachycephala
these fossae are wider and shallower and the median pha-
ryngeal tubercle between them is more prominent. Species
of allodesmines, enaliarctines, and otariines have less prom-
inent fossae and tubercles, and their extensive development
in species of Desmatophoca is a derived character.

In both species of Desmatophoca, the occipital condyles
do not tilt strongly laterally (Fig. 8). They project prominently
from the occipital shield and extend ventrally below the ven-
tral surface of the basioccipital. The intercondylar notch is
relatively narrow and deep in both species of Desmatophoca,
especially so in D. brachycephala, and somewhat as in En-
aliarctos mea/si. The articular surfaces of the condyles are not
bilaterally symmetrical on the holotype of D. brachycephala.
The one on the left side is abnormal because it does not
extend ventromedially as near to the midline as does the one
on the right side, nor as do the surfaces on both sides of the
holotype of D. oregonensis. The foramen magnum of D. brachy-
cephala is compressed dorsoventrally (Fig. 8), even more
than it is in D. oregonensis. There are prominent fossae dorsal
to each occipital condyle, and another fossa is centered just
below the apex of the occipital shield. Between these, a prom-
inent tubercle extends obliquely dorsolaterally toward the
nuchal crest from above each side of the foramen magnum.

RELATIONSHIPS

The subfamily Desmatophocinae can be accommodated
within the expanded concept of the family Otariidae that was

used by Mitchell (1968, 1975), Barnes (1972:62; 1979),
Mitchell and Tedford (1973), and Hall (1981). There is noth-
ing in the known morphology of either species of Desma-
tophoca that would seem to negate the previously postulated
origin of the subfamily from or near a taxon in the Enaliarc-
tinae. Prior characterizations of the desmatophocines as being
large, moderately derived, early sea lion-like animals with
some retained primitive characters also seem to be correct
(Mitchell, 1968, 1975; Barnes, 1972; Repenning and Ted-
ford, 1977; King, 1983:129-130, fig. 3.1). Multiple-rooted
posterior cheek teeth, widely separated mastoid and paroc-
cipital processes, relatively primitive and non-mortised (e.g.,
fissiped-like) zygomatic arches, and small supraorbital pro-
cesses of the frontal are all primitive characters that the des-
matophocines share with the enaliarctines. The relatively
elevated position of the optic foramina within the interorbital
region of D. oregonensis, the loss of M- in D. brachycephala,
the presence of the anteriorly flared dorsal margin of the
zygomatic arch at the front of the orbit, the different shape
of the paroccipital process, and other derived characters of
both species are some of the ways in which desmatophocines
differ from any of the known species in the subfamily Al-
lodesminae. This situation, plus the fact that the unique com-
bination of derived characters of the allodesmines (e.g., ex-
ceptionally bulbous-crowned and nearly homodont cheek
teeth, lack of vertical carinae on the canines, dorsoventrally
expanded (mortised) squamosal-jugal contact, large orbit with
its anterior border retracted posteriorly above infraorbital
foramen, flatter tympanic bulla, transversely expanded pos-
terior lacerate foramen, and posteroventrally positioned op-
tic foramina within the interorbital region) is not shared with,
nor uniquely derivable from the characters of the Desma-
tophocinae, are part of the argument against either synony-
mizing the two subfamilies or uniting them in a higher cat-
egory exclusive of the family Otariidae. In fact, there has
been no evidence presented that any members of the other
later otariid subfamilies (Otariinae, Imagotariinae, Dusi-
gnathinae, Odobeninae) evolved from the Desmatophocinae,
and this subfamily appears to have become extinct without
leaving any descendants (see Mitchell, 1975; Repenning and
Tedford, 1977; Barnes, 1979).

Within the genus Desmatophoca, D. oregonensis is, in most
of its cranial characters, the more primitive of the two known
species. Among its most obvious primitive characters are the
elongate skull, a smaller posterior lacerate foramen, smaller
canine, two separate roots on the M1, and the retained M2.
Desmatophoca brachycephala, the geochronologically earlier
taxon, has only a few characters in which it is more primitive
than D. oregonensis: a broader interorbital region, a more
slender jugal, and a less dorsoventrally expanded squamosal-
jugal articulation, for example. It is interesting, though, that
in the majority of its cranial characters, D. brachycephala is
clearly the more derived species of the two because, for ex-
ample, it has lost the M2, has a larger and more cylindrical
canine, a single (albeit multilobed) root on the M1 formed
by fusion of previously separate roots, a P' which is crowded
adjacent to the posteromedial side of the canine, a shorter
rostrum, a more anterodorsally flaring anterior margin of the
orbit, a larger mastoid process, a smaller and more laterally

16 Contributions in Science, Number 382

Barnes: Desmatophoca from Washington

II I

5 cm

Figure 8. Desmatophoca brachycephala, new species, holotype, skull, LACM 120199, LACM locality 4584, posterior view.

directed paroccipital process, the hyoid fossa closer to the
stylomastoid foramen and separated from it by only a thin
ridge of bone, a broader squamosal fossa, a broader brain-
case, the mastoid process projecting farther laterally and more
deeply excavated posteriorly, and a larger posterior lacerate
foramen. The various differences between the two species of
Desmatophoca, both in geochronological occurrence and in
suites of primitive and derived characters, rule out any sort
of ancestral-descendant relationship between them. These
differences do not preclude the possibility of their sharing a
common ancestor, however.

I interpret the above-cited differences as being clearly di-
agnostic at the species level, and not attributable to individ-
ual variation or sexual dimorphism. The holotypes of both
species have the kinds of cranial rugosities, crests, and pro-
cesses that appear as male secondary sex characters in in-
dividuals of Recent otariid species, and both holotypes rep-
resent adult individuals. Both skulls are nearly the same size,
although the holotype of D. brachycephala, which is broader
and has larger canines, is slightly shorter than that of D.
oregonensis.

Not only does the discovery of D. brachycephala expand
the contents and diagnoses of both the genus Desmatophoca
and the subfamily Desmatophocinae, but it also adds another
taxon to the list of examples of cranial convergence between
species in the families Otariidae and Phocidae (true seals).
Various examples of phocid-otariid convergence have been
pointed out by Mitchell (1968:1887-1888, 1892; 1975) and
Barnes (1972:63, 64, 66; 1979:30). Desmatophoca brachy-
cephala has general cranial proportions similar to those of
some phocids, especially the Recent monk seals of the genus
Monachus Fleming, 1822. The skull of D. brachycephala is
also broad and dorsoventrally flattened, somewhat as in the
allodesmine, Allodesmus packardi Barnes, 1972, but differs
notably by having more widely flared zygomatic arches and
a broader snout.

It is tempting to speculate on the possible life habits of

Desmatophoca brachycephala, but additional specimens
would first be necessary. Obvious pinniped characters of the
species (e.g., enlarged nares, orbit, braincase, cranial foram-
ina and posterior part of the palate, relatively uninflated
bulla, reduced or lost lacrimal bone and lacrimal foramen,
near homodonty) indicate that it must have been relatively
well adapted to an aquatic existence, but in the absence of
information on its middle ear, dentition, mandible, and post-
cranial skeleton, comments on its diving ability, prey items,
mode of feeding, and locomotion would be mostly specu-
lation.

CONCLUSIONS

The extinct subfamily Desmatophocinae is a useful taxo-
nomic grouping within the marine carnivore family Otariidae
because it contains morphologically distinctive animals that
had a considerable evolutionary history separate from other
recognized subfamilies in the family. The Desmatophocinae
are apparently not so closely related to the subfamily Allo-
desminae that the two should be synonymized, an idea that
has been supported by some previous authors, including my-
self. The subfamily Desmatophocinae, now represented by
two named species from rocks bordering the eastern North
Pacific Ocean, has a known geochronologic range from the
Early Miocene to the early Middle Miocene. These are rel-
atively large pinnipeds, apparently relatively rare, which, for
their time, and in comparison with contemporaneous and
slightly older and smaller species in the primitive subfamily
Enaliarctinae, are relatively highly derived. The desmato-
phocines apparently represent the earliest lineage within the
Otariidae to have evolved large size and some derived char-
acters that are convergent with some of those present in the
other, later otariid subfamilies. The Desmatophocinae ap-
parently died out without evolving into any later group.

Desmatophoca Condon, 1906, is the type genus of the
subfamily. Its type species is Desmatophoca oregonensis Con-

Contributions in Science, Number 382

Barnes: Desmatophoca from Washington 17

Figure 9. Comparative ventral views of holotype skulls of species of Desmatophoca; a, D. oregonensis Condon, 1906 (modified from Mitchell,
1975:fig. 2, with additional information from referred specimen, LACM 123285); b, D. brachycephala, new species; reduced to the same
cranium length.

18 Contributions in Science, Number 382

Barnes: Desmatophoca from Washington

don, 1 906, known only by specimens from late Early Miocene
and/or early Middle Miocene age rocks (circa 15-19 ma) that
are referred to the Astoria Formation, and which are exposed
near Newport on the Pacific coast of Oregon.

A second species, Desmatophoca brachycephala, new
species, has been found in slightly older. Early Miocene strata
(circa 20-23 ma), that are exposed near Knappton in Wash-
ington. The rocks that yielded the holotype and only known
specimen of this species have also been referred to the Astoria
Formation, but they were deposited in a different sedimen-
tary realm from those on the coast of Oregon, and are cor-
relative with the type section of the formation directly across
the Columbia River from Knappton at Astoria, Oregon. Al-
though D. brachycephala is the earlier of the two species,
judging by most of its characters (broader cranium, shorter
rostrum, and other derived dental and cranial features), it is
more highly derived than D. oregonensis. Neither species is
suitable as a morphological or chronological ancestor of the
other, but they could share a common ancestor. Like some
previously described fossil otariids, some general features of
the skull of D. brachycephala are convergent with those of
true seals in the family Phocidae. It is unfortunate, but prob-
able, that studies of the more numerous specimens of D.
oregonensis, already in museum collections, will, by infer-
ence, provide more information about D. brachycephala be-
fore additional specimens of this rare pinniped become avail-
able.

The discovery of D. brachycephala extends the geochro-
nologic range of the genus and of the subfamily farther back
in the Early Miocene and indicates that there must have been
a considerable, but as yet undocumented, evolutionary his-
tory of desmatophocines prior to that time.

ACKNOWLEDGMENTS

I owe my sincere thanks to the discoverers of the holotype
of D. brachycephala, James L. Goedert (LACM Museum
Associate) and Gail H. Goedert (LACM Field Associate).
Over the course of several months they located the various
fragments of rock containing parts of the specimen during
low tides that exposed the muddy, algae-covered shore of the
Columbia River near Knappton, Washington. We know of
this species of pinniped primarily because of their tireless
dedication. I also thank Donald J. Martel, who collected the
important referred specimen of D. oregonensis on the Oregon
coast. Michael D. Quarles and Rodney E. Raschke (LACM
Museum Technicians) painstakingly prepared, respectively,
the holotype of D. brachycephala and the referred snout of
D. oregonensis. Their work was financed by a National Sci-
ence Foundation Grant (BSR 82-18194). The photographs
were prepared by John De Leon and Donald Meyer, LACM
staff photographers. I am also indebted to Susan E. Barnes
and Lorraine Vasquez for their help in preparing the manu-
script for submittal, and to Ann E. Morey for helping to lay
out some of the illustrations. All artwork is by the author. I
thank Edward Mitchell and James Goedert for helpful dis-
cussions during the course of this work and John M. Harris,
Ellen J. Moore, Charles A. Repenning, and Richard H. Ted-
ford for reviews of the manuscript.

LITERATURE CITED

Addicott, W.O. 1 976. Neogene molluscan stages of Oregon
and Washington, pp. 95-1 15. In A.E. Fritsche, H. Ter
Best, Jr., and W.W. Womardt (eds.). The Neogene Sym-
posium. Society of Economic Paleontologists and Min-
eralogists, 160 pp.

Armentrout, J.M. 1981. Correlation and ages of Cenozoic
chronostratigraphic units in Oregon and Washington.
Geological Society of American Special Paper 184: 1 37-
148.

Barnes, L.G. 1972. Miocene Desmatophocinae (Mamma-
lia: Carnivora) from California. University of California
Publications in Geological Sciences 89:1-68, pis. 1-6.

. 1979. Fossil enaliarctine pinnipeds (Mammalia:

Otariidae) from Pyramid Hill, Kern County, California.
Contributions in Science, Natural History Museum of
Los Angeles County 318:1-41.

. In press. A new enaliarctine pinniped from the

Astoria Formation, Oregon; and a classification of the
Otariidae (Mammalia: Carnivora). Contributions in Sci-
ence, Natural History Museum of Los Angeles County.

Barnes, L.G., D P. Domning, and C.E. Ray. 1985. Status
of studies on fossil marine mammals. Marine Mammal
Science 1 ( 1 ): 1 5-53.

Condon, T. 1906. A new fossil pinniped (Desmatophoca
oregonensis) from the Miocene of the Oregon coast. Uni-
versity of Oregon Bulletin, suppl. 3(3): 1-14.

. 1910. Oregon geology: A revision of “The two

islands.” J.K. Gill Co., Portland, Oregon, 187 + xvi pp.

Dalrymple, G.B. 1979. Critical tables for conversion of
K-Ar ages from old to new constants. Geology 7:558-
560.

Downs, T.D. 1956. A new pinniped from the Miocene of
southern California: With remarks on the Otariidae.
Journal of Paleontology 30: 1 1 5-13 1 .

Hall, E.R. 1981. The mammals of North America, second
edition. Vol. 1, i-xv, 1-600 pp.; vol. 2, i-vi, 601-1 181
pp., John Wiley and Sons, New York, New York.

Hay, O.P. 1930. Second bibliography and catalogue of the
fossil Vertebrata of North America. 2 vols. xiv + 1074
pp.; viii + 916 pp.

Howe, H.V. 1926. Astoria: Mid-Tertic type of Pacific Coast.
Pan-American Geologist 45:295-306, pi. 22.

Howell, A. B. 1928. Contribution to the comparative anat-
omy of the eared and earless seals (genera Zalophus and
Phoca). Proceedings of the United States National Mu-
seum 73:1—142, pi. 1.

Kellogg, R. 1 922. Pinnipeds from Miocene and Pleistocene
deposits of California . . . and a resume of current the-
ories regarding origin of Pinnipedia. University of Cal-
ifornia Publications in Geological Sciences 13:23-132.

King, J. E. 1 983. Seals of the world. Second edition. British
Museum (Natural History), Comstock Publishing As-
sociates, Ithaca, New York, 240 pp.

Miller, M.E., G.C. Christensen, and H.E. Evans. 1964.
Anatomy of the dog. W.B. Saunders Company, Phila-
delphia and London, xii + 941 pp.

Mitchell, E.D. 1966. The Miocene pinniped Allodesmus.

Contributions in Science, Number 382

Barnes: Desmatophoca from Washington 19

University of California Publications in Geological Sci-
ences 6 1 : 1-46, pis. 1-29.

. 1968. The Mio-Pliocene pinniped Imagotaria.

Journal of the Fisheries Research Board of Canada 25:
1843-1900.

. 1975. Parallelism and convergence in the evolution

of Otariidae and Phocidae. Rapports et Proces-verbaux
des Reunions Conseil international pour FExploration
de la Mer 169:12-26.

Mitchell, E.D., and R.H. Tedford. 1973. The Enaliarctinae.
A new group of extinct aquatic Carnivora and a consid-
eration of the origin of the Otariidae. Bulletin of the
American Museum of Natural History 151 (3):20 1—284.

Moore, E.J. 1964 (for 1963). Miocene marine mollusks
from the Astoria Formation in Oregon. United States
Geological Survey Professional Paper 419:i-iv, 1-109,
pis. 1-33.

. 1984. Molluscan paleontology and biostratigraphy

of the Lower Miocene upper part of the Lincoln Creek
Formation in southwestern Washington. Contributions
in Science, Natural History Museum of Los Angeles
County 351:1-42.

Moore, E.J., and W.O. Addicott. 1987. The Miocene Pil-
larian and Newportian (Molluscan) stages of Washing-
ton and Oregon and their usefulness in correlations from
Alaska to California. In E.L. Yochelson (ed.). Shorter
Contributions to Paleontology and Stratigraphy. United
States Geological Survey Bulletin 1 664 (Chapter A) A 1 -
A1 3, pis. 1-4.

Munthe, J., and M.C. Coombs. 1979. Miocene dome-skulled
chalicotheres (Mammalia, Perissodactyla) from the
western United States: A preliminary discussion of a
bizarre structure. Journal of Paleontology 5 3( 1 ):77— 9 1 .

Packard, E.L. 1947. A pinniped humerus from the Astoria
Miocene of Oregon. Oregon State College, Oregon State
Monographs, Studies in Geology 7:23-32.

Packard, E.L. , and R. Kellogg. 1934. A new cetothere from
the Miocene Astoria Formation of Newport, Oregon.

Carnegie Institution of Washington Publications 447: 1-
62, pis. 1-3.

Ray, C.E. 1976. Fossil marine mammals of Oregon. Sys-
tematic Zoology 25(4):420-436.

Repenning, C.A. 1976. Adaptive evolution of sea lions and
walruses. Systematic Zoology 25(4):375-390.

Repenning, C.A., and R.H. Tedford. 1977. Otarioid seals
of the Neogene. United States Geological Survey Profes-
sional Paper 992:i-vi, 1-93, pis. 1-24.

Rigby, J.K., and D.E. Jenkins. 1983. The Tertiary sponges
Aphrocallistes and Eurete from western Washington and
Oregon. Natural History Museum of Los Angeles Coun-
ty, Contributions in Science 344:1-13.

Simpson, G.G. 1945. The principles of classification and
a classification of mammals. Bulletin of the American
Museum of Natural History 85:1-350.

Sivertsen, E. 1954. A survey of the eared seals (family
Otariidae) with remarks on the Antarctic seals collected
by M/K “Norvegia” in 1928-1929. Det Norske Viden-
skaps-Akademii Oslo 36:1-76, pis. 1-10.

Wells, R.E. 1979. Geologic map of the Cape Disappoint-
ment-Naselle River area. Pacific County, Washington.
United States Geological Survey, Open-File Report 79-
389, 1:48,000.

Wortman, J.L. 1894. Osteology of Patriofelis, a Middle
Eocene creodont. Bulletin of the American Museum of
Natural History 6:129-164, pi. 1.

. 1906. A new fossil seal from the marine Miocene

of the Oregon coast region. Science, New Series, 24:89-
92.

Zullo, V.A. 1982. Arcoscapellum Hoek and Solidobalanus
Hoek (Cirripedia, Thoracica) from the Paleogene of Pa-
cific County, Washington, with a description of a new
species of Arcoscapellum. Natural History Museum of
Los Angeles County, Contributions in Science 336:1-9.

Submitted 30 June 1986; accepted 3 February 1987.

20 Contributions in Science, Number 382

Barnes: Desmatophoca from Washington

Mi

fell

Kara -f ra n'lj MUamnni ; ii*:^ i urt ; » .. ■

“fiiSiiilllliS

ji i]iffl|Wl!|!|i1iif

M i 'I JJiBlVflil

■iiiiPlll 13

INSTRUCTIONS FOB AUTHORS

■»4fkilii,M I|”1 1 1 'MM

ili' j

■res®

h -scienibhs iSiitS^onl^ulion* m Scieft^! seribslidisshdif of Cdrilinlkisiofjs-
;p,u3ar intervals, and range m size is^fa:rt :l^

i'OTl wi Hi uj|dje||| drfooynmu-s .review.. Priority is
ten by members of, the Museum staff. Mmmscripts diouJd be
■ fo-seins o'uTrl.i:-n.e:d b*e3o w and submitted |i| t be Head of

Section of the Mdseurn.

9,'thcrp !/t the iit'f fl',,:.

or Zo'i

ohs of the appropriate Code, citing the col lection,, 'by

name

The depository o f other si udy pmteriabshoo id

m

. in Spanish or Portuguese, as appropriate, as well asin English,
are not required, but are strongly recommended where ap-

ii

**41*1

I|n

le-spaced, including title! page;-; ab4rai4; teirt, tmkhovdedgniients, rdf-

* ‘ " : ? ’• -J 1 ! 41 11 f *• ’ ;:ll!l

TIT m

®t component should begin -on- athe>v::g«a|ecUj;- the foD owing; seqtpjiiijuce: title f

>$lract{s); text: acknowledgments; literature cited: tables, each labile- complete! with

i footnotes- on a separate page; and figure j^ptipiis. All; ftages should be xrurrt fered

' Ip ||l . I-’’ j.j |:c; ;] If : ' ' g|i|l jui k;| : d: |; ;| ■:= ;! 'M

l original illustrations. Line drawings are best prepared for 50 percent reduction
14 by 17.5 inches (35 by 43 cm) iti sw;e:, .photographs- are usually!

iiiaiiis

re

full size and should not exceed 7 by 8.75 inches (.1 ? by 22 cm, full page: width)'
W4JS inched: (8 by 1 1 cm, ijjllpll ootiitSbiq).

i iSSlPi ij: iii'rtrii'Tt

STi

SURmSSlOtS OF MANtfSCRJFF

•-'•ittie Aiiuscfipt in a secure enclosure. Manuscript should be ae~
- mm fodeTstgned by the .author who will be responsible for cent spondehde
inuscripL The covering letter should contain a statement that the manuscript
not been published elsewhere except in abstract or

any pari:

P lil

»ij

Mil

■

. ... .... .... ; : . or not funds- are'-tfvailal^^ ^n^siialiifemtat !

or illustrations. Include copies of anv permissions needed to reproduce pobli.shed

:at. : : i : ::!■■ id. hm To: ;.i: ijhj aTjltiri'j.ijIlviJp:!:: ■ ;

uscripts will be reviewed for possible . publication with the •understanding that they.
iot been published., siraukanebirsly submitted, or atteady accepted for pifoiieatiou
vh ere . T h i s does ndt'-pf ehtudh cajikiadhatipri ! of a complete report Ipl follows :||i blieaiion
bminary findings elsewhere. Copies of any possibly duplicative material should be
tted with the -manuscript- that is, being shut for cons ideratton..

flf ;(! U- !flS?ilS?dmw i ' W'llir J y?'i ’ i iTc: ; ; f : p-ii! f #

p a ii » k ru .'!. nu: c 's

• i i > i ; • i > • " M: . .il

■ !•?:

mil

iseuiin., the authc>jfs)-Wiil: receive! a- statement of

pdithr; calculated at a -current, rate. Partial dr complete pay njieni

from those authors who ha ve funds available for this

ss to funds for partial or complete payment, of page charges

,pi ,:e :on' the stp!en:i.ent an<i return it to the Managing 'Editor.. The rem ned .

■ n

ili®
ii

■M

! 1 11 ill

’I i

ii.l!;! lil;

■!

‘ ‘ ilu,.i if '• i ;i. r' i ,,•! i «i M( .Wit{'|

iilHSilll''

{IHflji ! i]|]||ij!i :j

'ISiliS

I iflllmlra

m " ' Hi:!

SERIAL PUBLICATIONS OF THE
NATURAL HIS-TORY MUSEUM- OF LOS ANGELES COUNTY

The scientific publications of the Natural History Museum of Los Anjjebs CoUAty have been,
issued at irregular intervals in three major series; the issues in each series are numbered
individually and, numbers run consecutively, regardless of the sutgiefef matter.

® Contributions in Science, a miscellaneous series of technical papers describing orig-
inal research in the, life and earth sciences. . ■ .

: i® Science, flu Het in ^ a miscellaneous series research

in the Sifeuind . earth scienriesiJ This series was discontinued in 197:8 with the issue of
Num.be re 2 9 and 30; monographs are now published by the M use dm in Contri butions

I

in Science.!

Bin

, , , |;'! i :;j,

. u: :, I ■ I I "II II il I l illill, „ I f !" I1, P

,'IMilBii Livy ^=8l!!i0liwiftBlliiB8ftww^WlroHBBBHMu

"sis iiiliilSlillHl

ft Science Scries, kmc: ariicles on natural history topics, generally written for the layman.

Copies of the publications Ltii tj^ese -series jflm sold, through the Museum Book Shop. A catalog
is pvailisii® p:p request.

The Museum also publishes Technical Reports, a miscellaneous' : series Containing information
rolat fye sc Ii atari y inquiry anjl collections but not reporting the results of original research.
Issue, is. aulbourid by'the muiseym’s Scientific Publications Committee; however, manuscripts
do not reeei ve .anonym cius peer review.. Individual Technical Reports may be obtained from

the relevant See!. top. of the museum.

SCIENTIFIC PUBLICATIONS COMMITTEE

■■ Craig C,

i«; Museum Director

f •••• . V'fE .B.I! tiSiHiii " i ■ 1 -.| '

. ||l Coke** | ' ;Cbm«|iittee Cha kmiii ;

; Jolim M. Harris '

Charles L. Hogue
Geo fee L. Kennedy '

- : I Or, ,| i r , ; , . , - - ' ' J ' - , ' . ", ' ,

Rqti id A,; Si mpson;, -Managing Editor
Errol Steveiis .

l|l

liiifii! ,
■ ii ii
iieili

MIDDLE MIOCENE MARINE BIRDS FROM THE
FOOTHILLS OF THE SANTA ANA MOUNTAINS,
ORANGE COUNTY, CALIFORNIA

Hildegarde Howard1 and Lawrence G. Barnes2

ABSTRACT. A mixed marine and non-marine fossil vertebrate as-
semblage of Middle Miocene age has been recovered from several
localities on both sides of Oso Creek in the foothills of the Santa
Ana Mountains in Orange County, California, U.S.A. The fossils
were found during grading operations for construction of Upper Oso
Dam, and were derived from a coarse clastic marine rock unit. This
rock unit, which crops out near the southeast margin of the Los
Angeles Basin (both a topographic and depositional feature), has
been mapped as the Topanga Formation. It overlies rocks assigned
to the Early Miocene Vaqueros Formation and is in turn overlain
by rocks mapped as the Late Miocene Monterey Formation.

Among the available sample of 54 avian bones in the assemblage
from Upper Oso Dam are fossils representing Diomedea 7 mi lleri
Howard (albatross), Puffinus priscus Miller, Puffinus ?new species,
Puffinus spp. (shearwaters), Sulidae (boobies and gannets, two species),
Anatidae (undescribed large diving duck), and Alcodes, alf. A. ulnulus
Howard (flightless auk). This is the oldest fossil avifauna yet de-
scribed from Cenozoic marine strata in the Orange County area, and
the only marine avifauna in California of Middle Miocene age that
has yet been reported from south of the outcrops of the Sharktooth
Hill Bonebed in Kern County. The birds in the assemblage from
Upper Oso Dam have their closest affinities with the species from
the Sharktooth Hill Bonebed. No species of alcids are known from
that bonebed, however, and the primitive flightless auk genus, Al-
codes, had been previously recorded only in slightly younger. Late
Miocene fossil assemblages from the Monterey Formation in Orange
County.

INTRODUCTION

From 1978 to 1983, field parties organized by the Natural
History Museum of Los Angeles County (LACM) collected
fossils from Middle Miocene marine sediments in the Oso
Creek area (Fig. 1 ) in the foothills of the Santa Ana Mountains
near Mission Viejo, Orange County, California (Raschke,
1984). The fossil-bearing sediments were exposed during the
mechanized grading of hills on both sides of Oso Creek during
construction of an earth-fill dam, named Upper Oso Dam
(33°39'30" north latitude, 117°37'35" west longitude). The
subjects of the present study are the 54 bird bones thus ob-
tained which represent probably 34 individuals.

Contributions in Science, Number 383, pp. 1-9
Natural History Museum of Los Angeles County, 1987

The bones were found in a gray and yellowish-red colored,
pebble-bearing, coarse sand deposit that is several meters
thick and has been identified on geologic maps (e.g., Morton
et al., 1973:pl. 1) as belonging to the Topanga Formation.
These sediments at Upper Oso Dam directly underlie shales
and siltstones that have been mapped as the Monterey For-
mation, which is Late Miocene in age. The Monterey For-
mation has elsewhere yielded the birds that were previously
reported by Howard ( 1966, 1968, 1976, 1978) from localities
in Laguna Hills (Leisure World) and Laguna Niguel, ap-
proximately 10 km to the southwest in Orange County (Fig.
1).

It may no longer be correct to refer to the coarse sands at
Upper Oso Dam as the Topanga Formation because the type
section of that formation is in the Santa Monica Mountains,
approximately 100 km to the north within the Los Angeles
Basin, where the name has been elevated to the status of a
group (Yerkes and Campbell, 1979). Because the Topanga
Formation is no longer used in the Santa Monica Mountains,
and because of lithologic differences and the lack of lateral
continuity between the two areas, the formation name is
probably no longer applicable to rocks anywhere in Orange
County.

Raschke (1984) provided a preliminary notice of the oc-
currence of vertebrate fossils at Upper Oso Dam, and future
reports will be published on the geology in the vicinity and
on the other fossils. The fossil assemblage from the Middle
Miocene sediments at Upper Oso Dam is correlated with the
Barstovian North American Land Mammal Age and the
Sharktooth Hill Local Fauna from the Round Mountain Silt
in Kern County, farther north in California (Raschke, 1984:
64). Some evidence for these age determinations derives from

1. Chief Curator Emeritus, Earth Science Division, Natural His-
tory Museum of Los Angeles County, Los Angeles, California 90007.

2. Curator and Section Head, Vertebrate Paleontology, Natural
History Museum of Los Angeles County, Los Angeles, California
90007.

ISSN 0459-8113

Figure 1. Location of the sites that produced the fossil birds described in this study, at Upper Oso Dam, Oso Creek, Orange County,
California.

unpublished data from other fossil specimens in the LACM
from Upper Oso Dam, including those of sharks, bony fishes,
turtles, and mammals. Among these other taxa, the sharks
and mammals are probably the most useful age indicators,
considering the state of our knowledge of Pacific coast marine
and terrestrial vertebrates. The shark assemblage is typically
Middle Miocene in aspect and includes Isurus hastalis (Agas-
siz, 1843); I. planus (Agassiz, 1856); Isurus spp.; Procar-
charodon megalodon (Agassiz, in Charlesworth, 1837); Car-
charinus sp. and/or Negaprion sp.; Galeocerdo aduncus
Agassiz, 1843; Hemipristis serra Agassiz, 1843; Squalus ser-
riculus Jordan and Hannibal, 1923; Squatina lerichei Jordan
and Beal, 1913; and Myliobatis sp. These have similar mor-
phologies and occur in approximately similar relative abun-
dances as do the same taxa from the Sharktooth Hill Bone-
bed. Terrestrial mammalian taxa from Upper Oso Dam, not
strictly limited to, but characteristic of the Barstovian Land
Mammal Age, include a horse, Merychippus sp.; a stilt-legged
camel, Aepycamelus sp.; and a large paleomerycid (Raschke,
1984:64). The mammals also include at least three species
of otariid pinnipeds: Allodesmus sp., Neotherium sp., and an
undescribed genus and species. The same or similar species
of all three of these pinniped genera occur in the Sharktooth

Hill Bonebed. Among the odontocete cetaceans from Upper
Oso Dam are Kentriodon sp., cf. “ Squalodon ” errabundus
Kellogg, 1931, and a large, unnamed species; again, all similar
to, or the same as, taxa from the Sharktooth Hill Bonebed.
The age of that bonebed, as determined on the basis of several
lines of evidence (see Barnes, 1976; Barnes and Mitchell,
1 984), is late Middle Miocene, correlated with the Barstovian
North American Land Mammal Age (Wood et al., 1941, the
“Temblor” provisional provincial mega-invertebrate stage
of Addicott (1972), and the Relizian or Luisian foraminiferal
stage, and is between approximately 1 3 and 1 5 million years
old. Additional, indirect evidence for the Barstovian age of
the vertebrate-bearing sands at Upper Oso Dam is the fact
that they overlie strata referred to the Vaqueros Formation
(Morton et al., 1973:pl. 1), which have produced land mam-
mals that are characteristic of the Hemingfordian North
American Land Mammal Age (Raschke, 1984:62, fig. 1;D.P.
Whistler, personal communication).

Unfortunately, many of the bird bones from Upper Oso
Dam that we describe here are fragmented, crushed, and
distorted. Some of this is the result of geologic phenomena,
but some resulted from the fossils’ being scraped and run
over by machinery. Specific identifications have, therefore,

2 Contributions in Science, Number 383

Howard and Barnes: Miocene Birds from California

been difficult. Five families are represented, the most abun-
dant of these being the shearwaters (family Procellariidae,
26 bones, 16 individuals), and the booby-gannet family (Su-
lidae, 22 bones, 1 0 individuals). Other fossils belong to mem-
bers of the albatross family (Diomedeidae, two bones of one
individual), waterfowl family (Anatidae, one bone), and auk
family (Alcidae, three bones of three individuals). At least
eight species are represented. (We did not calculate minimum
numbers of individuals, but made our determinations on the
basis of the stratigraphic and geographic separation of the
specimens.)

The families Procellariidae and Sulidae are well repre-
sented in the fossil record, with seven species of the Sulidae
and eight species of the Procellariidae (including two fulmars)
having been previously described from California Miocene
deposits (Howard, 1983). A thorough study of each of these
families is needed and, therefore, identifications herein of the
shearwater and sulid specimens should be considered ten-
tative. The Sulidae are now in the process of being reviewed
by K.I. Warheit of the Department of Paleontology at the
University of California, Berkeley. Fossil avian species have
been described from the Sharktooth Hill Bonebed by Wet-
more ( 1930), Miller (1961, 1962), Howard ( 1 966, 1 984), and
Warter (1976). Howard (1983) summarized these and other
published records of fossil birds from California.

MATERIALS AND METHODS
Acronyms used for institutions are:

CAS: California Academy of Sciences, San Francisco, Cali-
fornia.

LACM: Natural History Museum of Los Angeles County,
Los Angeles, California.

UCLA: University of California at Los Angeles, California.
UCMP: University of California Museum of Paleontology,
Berkeley, California.

UCMVZ: University of California Museum of Vertebrate
Zoology, Berkeley, California.

USNM: National Museum of Natural History, Smithsonian
Institution, Washington, D.C.

In addition to the fossil material in the LACM collections,
referred specimens and casts of holotypes from CAS, UCMP,
UCLA, and USNM have been available for this study. An
accurate cast of the holotype of Puffinus inceptor Wetmore,
1930, was used for comparisons instead of the original spec-
imen (CAS 5223). Modem skeletal material is from the LACM
collections, with the exception of specimens of Aethia pusilla
(Pallas, 1811), which are from UCMVZ. The fossils described
from Upper Oso Dam are from LACM localities 4545, 4546,
and 4547 which, in that order, are progressively older and
stratigraphically lower within the deposit. The locality num-
ber LACM 4464 is assigned to fossils that were found weath-
ered out and loose at LACM localities 4545, 4546, or 4547,
and therefore lacking more precise stratigraphic data. The
classification herein follows that of the Sixth Edition of the

American Ornithologists’ Union Checklist of North Amer-
ican Birds (1983).

SYSTEMATICS
Order Procellariiformes
Family Diomedeidae (albatrosses)
Diomedeci Linnaeus, 1758

Diomedea 1 miller i Howard, 1966

Figure 2C

REFERRED SPECIMENS. Associated proximal half of
right ulna and distal end and shaft fragments of right radius
from one individual, LACM 120439, from LACM locality
4547.

DISCUSSION. Albatross bones have been previously re-
ported from several Miocene age sites in California, but only
two species, both first recognized on the basis of fossils from
the Sharktooth Hill Bonebed, have been named: Diomedea
californica Miller, 1962, and D. milleri Howard, 1966. Di-
omedea californica has also been found in the Late Miocene
Monterey Formation in the Laguna Hills in Orange County
(see Howard, 1978). The holotype of the species is a tarso-
metatarsus, slightly larger than that of the living Short-tailed
Albatross, D. albatrus Pallas, 1769, but not as large as that
of the extant Wandering Albatross, D. exulans Linnaeus,
1758. Diomedea milleri is a considerably smaller species, and
is still known only from the Sharktooth Hill Bonebed. Its
holotype is the proximal end of an ulna having a proximal
breadth of 14.4 mm, as compared to 17.6 mm for the same
bone of the living Black-footed Albatross, D. nignpes Au-
dubon, 1839 (see Howard, 1966:4).

The albatross ulna, LACM 120439, from Upper Oso Dam
(Fig. 2C), is 15.8 mm for the same measurement. The radius
associated with this ulna is also narrower of shaft than is the
same bone of a Recent D. nigripes. The area immediately
distal to the ulnar cotylae, which contains the diagnostic
characters of the species D. milleri, is crushed on LACM
1 20439. However, a clearly defined ridge on the palmar side
of the shaft below this area suggests that there was a distinct
angle between the palmar and internal surfaces of the shaft
immediately below the internal cotyla, as was described for
D. milleri. Although the present specimen, LACM 120439,
appears to have a more prominent acromion and external
cotyla than does the holotype of D. milleri, these areas are
slightly eroded in the latter, possibly altering its morphology.

Family Procellariidae (shearwaters)

Puffinus Brisson, 1760

Puffinus prisons Miller, 1961

Figures 2A, B, F, G

REFERRED SPECIMENS. Left humerus, LACM
1 204 1 5; associated right and left humeri and incomplete right

Contributions in Science, Number 383

Howard and Barnes: Miocene Birds from California 3

Figure 2. A, B, Puffinus priscus Miller, 1961, right humerus, LACM 120416, palmar and anconal views; C, Diomedea 1 miller i Howard, 1966,
right ulna, LACM 120439, internal view; D, E, Puffinus ?new species, right humerus, LACM 120422, palmar and anconal views; F, G, Puffinus
priscus Miller, 1961, right ulna, LACM 120416, palmar and internal views; H, I, external views of left ulnae of Alcodes Howard, 1968; H, A.
ulnulus Howard, 1968, holotype, LACM 18277, from Monterey Formation in Laguna Hills, LACM locality 1945; I, Alcodes, aff. A. ulnulus
Howard, 1968, LACM 120434, from Upper Oso Dam, LACM locality 4546; J-L, Alcodes, aff. A. ulnulus Howard, 1968; J, left humerus,
LACM 1 204 1 4, anconal view; K, L, right coracoid, LACM 121010, posterior and anterior views. All views natural size, except H and I, which
are x 1.5.

4 Contributions in Science, Number 383

Howard and Barnes: Miocene Birds from California

ulna from one individual, LACM 120416; distal end of right
humerus, LACM 120417; proximal fragment of left ulna,
LACM 1 20424; all from locality LACM 4546. Proximal three-
fourths of right humerus, LACM 1 26 1 59, from locality 4464.

TENTATIVELY REFERRED SPECIMENS. Right and
left distal ends of tibiotarsi, LACM 120425 (right) and 120426
(left); right tarsometatarsus lacking proximal end, LACM
120419; all from locality 4546.

DISCUSSION. The referred humeri from Upper Oso Dam
agree in their general conformation, in their size, and in the
flatness of their shafts (Figs. 2A, B) with the holotypic distal
end of this element (UCMP 58 1 85) from the Sharktooth Hill
Bonebed and/or with additional unpublished specimens (three
of which-LACM 18140, LACM 44556, and LACM
126219— are complete) from the same stratum, that are now
referable to Puffinus priscus Miller, 1961. Howard (1968:7-
8) questionably referred a partial humerus from the Late
Miocene age Monterey Formation in Orange County to the
same species. The ranges of measurements of the humeri
from Upper Oso Dam are: length — 79.3 to 81.2 mm, breadth
of distal condyles— 7.4 to 7.8 mm, distance from distal sur-
face of condyles to proximal surface of ectepicondylar prom-
inence—9.5 to 9.9 mm. The same measurements of humeri
from the Sharktooth Hill Bonebed which are assigned to P.
priscus are: length — 80.3 to 82.5 mm (n = 3), breadth of
distal condyles— 7.1 to 7.7 mm (n = 6, holotype— 7.5 mm),
and distance from distal condyles to proximal surface of
ectepicondylar prominence— 8.9 to 10.0 mm (n = 5, holo-
type—9.5 mm).

Each of the two ulnae, one of which (Figs. 2F, G) is as-
sociated with humeri (LACM 120416), has a high, pointed
olecranon (possibly exaggerated by crushing) and a long,
pointed, and upturned external cotyla. These characters are
suggestive of those of the extant Black-vented Shearwater,
P. opisthomelas Coues, 1864, although to an exaggerated
degree.

The tarsometatarsus, which measures 6. 1 mm in distal
breadth, is tentatively assigned to P. priscus on the basis of
favorable comparisons between it and tarsometatarsi of the
species from the Sharktooth Hill Bonebed. Of the three tar-
sometatarsi of Puffinus now available from the latter locality,
one specimen (LACM 104058) is incomplete, comprising
only the distal half. It agrees in size with LACM 1 204 1 9 from
Orange County, is slightly smaller than the other two, and,
we believe, is assignable to P. priscus.

The tibiotarsi resemble the only distal end of this element
assignable to Puffinus that has been found in the Sharktooth
Hill Bonebed, LACM 122994. That specimen is tentatively
referred to P. priscus on the basis of its size relative to the
tarsometartarsi noted above.

Puffinus ?new species

Figures 2D, E

REFERRED SPECIMENS. Associated, complete, but
poorly preserved right humerus, crushed proximal end of
ulna, and proximal and distal fragments of radius from one

individual, LACM 120422, from locality LACM 4546; col-
lected 23 April 1980 by L.G. Barnes.

DESCRIPTION. Although in general characters the hu-
merus (Figs. 2D, E) resembles those of species in the subgenus
Puffinus. the shaft is more rounded and less flattened through-
out. It is, therefore, clearly distinguishable from the holotype
of P. priscus Miller, 1961 (UCMP 58185), but slightly less
so from the holotype of P. mitchelli Miller, 1961 (UCMP
58184) (casts of both holotypes are at hand). In length (90.4
mm), LACM 120422 is intermediate between three Shark-
tooth Hill specimens of P. priscus (80.2 to 82.5 mm) and the
one measurable specimen (LACM 17500) of P. mitchelli (98. 1
mm). Although the ulna from Upper Oso Dam is severely
crushed, the breadth across the cotylae could not have ex-
ceeded 7.5 mm.

Poor preservation of this humerus from Upper Oso Dam
makes difficult any detailed comparison with the holotype
(CAS 5223) of P. inceptor Wetmore, 1930, which is the distal
end of a humerus. The specimen from Orange County (LACM
120422) is broken through the internal condyle, but appears
to lack the notable internal thrust of the distal end that is
characteristic of P. inceptor, and is reflected in the measure-
ments given by Wetmore (1930:87): breadth of shaft— 5.9
mm, greatest distal breadth— 12.2 mm. In the humerus from
Upper Oso Dam, the shaft breadth is 6.3 mm, the greatest
distal breadth 1 1.2 mm.

DISCUSSION. Many other Miocene species of the genus
Puffinus have been previously recorded from California (see
Howard, 1983). Among these, P. diatomicus Miller, 1925,
from the Late Miocene age Sisquoc Formation in Santa Bar-
bara County has the humerus represented in both the ho-
lotype and referred skeletal impressions. Miller (1925) mea-
sured the lengths of these humeri as 80 to 84 mm (considerably
less than for LACM 120422). The humerus is also the ho-
lotype for each of the two previously named Late Miocene
species from Orange County; P. calhouni Howard, 1968, and
P. barnesi Howard, 1978. Each of these is smaller and has a
more compressed shaft than does LACM 1 20422 from Upper
Oso Dam. Compared with the holotypic distal end of a hu-
merus of the Early Miocene species, P. micraulax Brodkorb,
1963, from Florida, LACM 1 20422 is broader. The holotypic
distal end of the humerus of P. conradi Marsh, 1970, from
the Middle Miocene Calvert Formation of Maryland, is
broader and has a thicker shaft.

Although the humerus of LACM 120422 appears to be
distinct from all of the above, the poor condition of the
available material precludes naming a new species. Should
more specimens assignable to this shearwater be discovered,
comparisons should also be made with the French Middle
Miocene species, P. aquitanicus (Milne-Edwards, 1874) and
P. antiquus (Milne-Edwards, 1874). In fact, Olson ( 1 985:2 1 1)
has noted the need for a worldwide revision of fossils assigned
to the genus Puffinus.

Puffinus, spp. undetermined

REFERRED SPECIMENS. Complete right ulna, LACM
120420; distal three-fourths of left ulna, LACM 124360;

Contributions in Science, Number 383

Howard and Barnes: Miocene Birds from California 5

proximal and distal ends of right ulna, LACM 120421; frag-
ment of proximal end of left ulna, LACM 126155; associated
fragments of distal end of ulna and proximal end of radius
and carpometacarpus of one individual, LACM 126156;
complete radius, LACM 120423; right carpometacarpus,
LACM 1 204 1 9; associated fragments of distal end of left ulna
and tarsometatarsal shaft of one individual, LACM 126157;
all from locality LACM 4546. Distal fragment of radius,
LACM 120412; incomplete right humerus, LACM 126160;
both from locality LACM 4464.

DISCUSSION. The size range alone among these speci-
mens suggests that at least two species are represented. Most
are of small size and in this respect are suggestive of P. priscus.
One poorly preserved humerus (LACM 126160) and one
ulna (LACM 120420) are, however, notably larger. This hu-
merus has been partly restored, but appears to have a more
flattened shaft toward the distal half of the bone than do
humeri of P. mitchelli from the Sharktooth Hill Bonebed.
The length of this poorly preserved humerus, from the prox-
imal edge of the brachial impression to the head is 83.2 mm.
The same measurement of a Sharktooth Hill humerus, LACM
17500, referred to P. mitchelli, is 86.6 mm. The ulna from
Upper Oso Dam, LACM 120420, has a length of 95.7 mm.
In contrast, the largest ulna known from the Sharktooth Hill
Bonebed (LACM 1 24359) measures only 88.2 mm long. Both
are markedly larger than the incomplete ulna (LACM 1 204 16)
from LTpper Oso Dam that is associated with the humerus
that we assigned above to P. prisons.

Order Pelecaniformes
Family Sulidae (boobies and gannets)
Sulidae, spp. undetermined

REFERRED SPECIMENS. Associated right and left hu-
meri, right and left ulnae, and a badly crushed radius of one
individual, LACM 120433; associated left coracoid and
proximal fragment of shaft of left humerus of one individual,
LACM 120428; right tarsometatarsus lacking part of prox-
imal end, LACM 120432; proximal end of radius, LACM
120429; proximal end of humerus, LACM 120430; distal
end of humerus, LACM 120431; skull fragment, LACM
1 20427; all from locality 4546. Associated incomplete radius
and ulna of one individual, LACM 1 20438; humerus, LACM
123910; seven associated incomplete and poorly preserved
bones of one individual lying together in matrix with the
following elements represented: incomplete mandible, prox-
imal end of right humerus, three fragments of left humerus,
fragment of left ulna, distal end of right radius, proximal end
of left radius overlying left carpometacarpus, LACM 126161;
all from locality LACM 4547.

DISCUSSION. These specimens will be included in War-
heit’s study of the family Sulidae. A brief examination of the

material from Upper Oso Dam reveals two humeri (LACM
1 20433, 1 2043 1 ) whose measurements exceed those of spec-
imens of Morus vagabundus Wetmore, 1930, the only species
of sulid that has been described from the Sharktooth Hill
Bonebed. The distal breadth of the holotypic humerus of M.
vagabundus is 18.3 mm, whereas the same measurement of
LACM 1 2043 1 is 2 1 .0 mm. The length of a referred humerus
of M. vagabundus is 170 mm (Howard, 1966:6), and the
length of LACM 120433 from Upper Oso Dam is 183 mm.

A second species of sulid from the deposits at Upper Oso
Dam is indicated by specimens LACM 1 20428, 1 20432, and
120438. Although difficult to measure exactly, they suggest
a smaller species. That more than one species of sulid is also
represented in the Sharktooth Hill Bonebed was noted by
Miller (1961:401), who reported a proximal fragment of an
ulna that is larger in size than the one referred to M. vaga-
bundus by Wetmore (1930). Howard (1966:7) also reported
from the same bonebed two tarsometatarsi that differ from
each other in qualitative characters as well as in size.

Order Anseriformes

Family Anatidae (ducks, geese, and swans)
Subfamily Anatinae (ducks)

Tribe Mergini (divers)

Mergini, genus and species undetermined

REFERRED SPECIMEN. Distal end of left femur, LACM
125994, from locality LACM 4546.

DESCRIPTION. The fibular condyle of this femur is thrust
laterally, and the popliteal area is deeply depressed and an-
gular, as is the case in diving ducks such as those of the
genera Somateria Leach, 1819; Mergus Linnaeus, 1758; and
Melanitta Boie, 1822. The fossil specimen is significantly
larger than the femora of extant species of these genera, how-
ever, and the breadth of its distal end is 20.7 mm.

DISCUSSION. This femur represents a large diving an-
atid that was the size of a goose. In its general conformation
it is suggestive of femora from the Sharktooth Hill Bonebed,
which represent a large anatid species. Compared with them,
however, the femur from Upper Oso Dam is more slender
and the muscle scar on the internal side of the posterior face
is more proximally located. Anteriorly, the intercotylar area
appears to be more deeply excavated, but allowance must be
made for the poor preservation of the Orange County spec-
imen.

The femora and other bones of the anatid from the Shark-
tooth Hill Bonebed, which are now under study by Howard,
appear to represent a new species, and probably a new genus.
The single femur from Upper Oso Dam indicates the pres-
ence of a large diving anatid in the Orange County area

6 Contributions in Science, Number 383

Howard and Barnes: Miocene Birds from California

contemporaneous with a probably congeneric, but specifi-
cally distinct bird in the Kern County area.

Order Charadriiformes
Family Alcidae (auks et al.)

A/codes Howard, 1968

Alcodes, aff. A. ulnulus Howard, 1968

Figures 2I-L

REFERRED SPECIMENS. Left ulna, LACM 120434
from locality LACM 4546; left humerus, LACM 120414,
from locality LACM 4545; right coracoid, LACM 121010,
from locality LACM 4547.

DESCRIPTION. The holotype of Alcodes ulnulus How-
ard, 1968, is a complete left ulna (LACM 18277, Fig. 2H).
The distal end of a right ulna (LACM 18279) and the prox-
imal end of a right carpometacarpus (LACM 18278) were
also referred to the same species by Howard (1968). An in-
complete proximal portion of a right ulna (LACM 18280),
10 percent larger than the holotype, was tentatively referred
to the species in the same publication.

The ulna (LACM 120434, Fig. 21) from Upper Oso Dam
is not as well preserved as the holotype, but resembles it by
having a relatively thick shaft for its length. It is, however,
slightly longer, and relatively narrower than the holotype of
A. ulnulus. In contrast to LACM 120434 from Upper Oso
Dam, the ulna of a Recent specimen of the Least Auklet,
Aethia pusilla (UCMVZ 60629), is longer and narrower, and
that of the Pliocene, flightless, auk-like swimmer, Mancalla
milieu Howard, 1970, is shorter and broader (see Table 1).
Although Alcodes ulnulus clearly had evolved in the direction
of flightlessness, it was not as derived as any species in the
genus Mancalla Lucas, 1901.

The humerus, LACM 120414, although incomplete (Fig.
2J), resembles the paratypic specimen (LACM 2813) of the
Pliocene species, Mancalla milleri, in the compression of its
distal end and in the height of its ectepicondylar prominence.
The deltoid crest, however, is not as extensive as that of M.
milleri, and the shaft is narrower throughout. Other distinc-
tive characters, peculiar to the Upper Oso Dam specimen,
are: proximally the pectoral attachment and external tuber-
osity are distinctly external to the capital shaft ridge and
separated from the ridge by a depression, and the ridge is
more angular; the depression below the head, internal to the
capital shaft ridge extends distally to the line of the distal
edge of the incomplete bicipital crest. Unfortunately, the
bone is too poorly preserved to make meaningful measure-
ments.

The coracoid from Upper Oso Dam, LACM 121010, re-
sembles that of M. milleri in its straight shaft and narrow
sternal facet (Figs. 2K, L). It is shorter, however, than the

Table 1. Measurements of alcid ulnae (in mm).

Alcodes

ulnulus

(holotype)

Alcodes,
aff. A. ul-
nulus
(LACM
120434)

Aethia

pusilla

(UCMVZ

60629)

Mancalla

milleri

(minimum)

Internal length

21 A

29.4

32.5

23.6

Breadth of shaft

2.9

2.9

1.6

3.2

Depth of shaft

4.9

3.9

2.7

5.3

smallest known coracoid of M. milleri, from which it further
differs as follows: triosseal canal angular at the anterior edge,
with the tip of the furcular facet projecting farther anteriorly;
anterior surface of the shaft evenly rounded throughout its
length, lacking the flattening and internal projection distally
that is characteristic of M. milleri. Its greatest length is 34.3
mm; whereas the minimum length of any specimen of M.
milleri is 38.5 mm (Howard, 1970:8).

DISCUSSION. These three bones are the only alcid spec-
imens collected at Upper Oso Dam. Indeed, they are the first
reported from any Middle Miocene deposit in California,
including the Sharktooth Hill Bonebed. Alcids were much
more abundant and diverse by Late Miocene time in southern
California, however, as evidenced by the presence of at least
five genera among the 27 alcid bones that have been recorded
(Howard, 1968) from the lower part of the Late Miocene age
Monterey Formation at a Laguna Hills site in Orange County
(LACM locality 1945, the type locality of Alcodes ulnulus).

The genus Alcodes had not been previously recorded else-
where, until the present study. When originally described,
because of its apparent flightlessness, it was placed, somewhat
questionably, by Howard in the subfamily Mancallinae. This
subfamily had been used traditionally for flightless species
in the genera Mancalla and Praemancalla Howard, 1976.
Recently, his studies of the flightless alcids have led Olson
(1985), however, to question the need to recognize the
subfamily.

SUMMARY AND CONCLUSIONS

Fossil birds comprise part of a vertebrate assemblage dis-
covered in marine Middle Miocene rocks at Upper Oso Dam,
near Mission Viejo, Orange County, California. The fossils
were collected from coarse clastic rocks that underlie shales
and siltstones mapped as the Monterey Formation. The
Monterey Formation has been the source of most of the
previously described Late Miocene birds from localities else-
where in Orange County. The fossil bird bones from Upper
Oso Dam described in this report are from a sandstone unit
that has been mapped as the Topanga Formation. Recent
changes in the nomenclature of the rocks in the type area of

Contributions in Science, Number 383

Howard and Barnes: Miocene Birds from California 7

the Topanga Formation, to the north in Los Angeles County,
resulted in discontinuance of its use as a formation name.
Because of this, and for other reasons, its use is probably no
longer correct for rocks in Orange County, and at the present
time, there is not an appropriate formational name for the
source rock of these fossils.

The age of the sandstone deposit bearing these fossils at
Upper Oso Dam was determined on the basis of relationships
with overlying and underlying rock units, and identifications
of marine and terrestrial vertebrates. The deposit is Middle
Miocene in age, correlative with the Barstovian North Amer-
ican Land Mammal Age, and approximately correlative with
the Sharktooth Hill Bonebed, which is in the upper part of
the Round Mountain Silt in Kern County, California. Special
attention was given, therefore, to comparisons between the
avian material from these two faunal assemblages.

The poor preservation of the avian material from Upper
Oso Dam has made difficult the specific assignment of the
54 avian bones represented. They are, however, identifiable
as follows (numbers of individuals are given in parentheses):

Family Diomedeidae (albatrosses)

Diomedea ‘l miller i Howard, 1966 (1)

Family Procellariidae (shearwaters)

Puffinus priscus Miller, 1961 (5)

Puffinus ?new species ( 1 )

Pujfmus, spp. undetermined (10)

Family Sulidae (boobies and gannets)

Sulidae, two spp. undetermined (10)

Family Anatidae (waterfowl)

Tribe Mergini, genus and species undetermined (1)
Family Alcidae (auks et al.)

Alcodes, aff. A. ulnulus Howard, 1968 (3)

The fossil bird assemblage from Upper Oso Dam is more
like that from the Sharktooth Hill Bonebed than any other
yet documented from the eastern North Pacific margin. How-
ever, only four of the nine avian families now represented
by species in the Sharktooth Hill Local Fauna (Howard, 1984)
are represented in the collection from the Upper Oso Dam
site. The albatross from Upper Oso Dam is questionably
assigned to Diomedea milleri, a species that is known only
from Sharktooth Hill. A shearwater, Puffinus priscus, defi-
nitely occurs at both localities. Possibly one or both of the
other species of Puffinus that have been described from
Sharktooth Hill are also represented by the other material
that we assign to this genus from Upper Oso Dam. The sulid
bones (although not yet identified to the species level) suggest
the presence of two species at the Orange County locality, as
is also the case for the Sharktooth Hill Bonebed.

A single anseriform femur documents the occurrence of a
large diving anatid in the deposits at Upper Oso Dam. The
only previously named marine anatid from California is Pres-
bychen abavus Wetmore, 1930, first described from the
Sharktooth Hill Bonebed, then later found in the Late Mio-
cene Monterey Formation in Orange County (Howard, 1 968).
The species from Upper Oso Dam is not P. abavus, but is
instead probably congeneric with another, undescribed species

that has been recently discovered in the Sharktooth Hill
Bonebed (Howard, manuscript in preparation).

Of particular interest are the three bones from Upper Oso
Dam belonging to a species of the family Alcidae. We have
assigned these to the primitive flightless auk genus, Alcodes,
and they are similar to, but not demonstrably conspecific
with specimens of A. ulnulus. Significantly, no alcid speci-
mens have ever been found in the Sharktooth Hill Bonebed.
This contrasts with relatively abundant and diverse Late
Miocene alcids, including A. ulnulus, in the geologically
younger Monterey Formation in Laguna Hills, not far from
Upper Oso Dam in Orange County.

The fossil bird assemblage from Upper Oso Dam is the
oldest one yet known in Orange County. It is the southern-
most documented occurrence of Middle Miocene marine birds
from the Pacific border of the United States, and indeed, is
the only Middle Miocene avifauna yet described from Cal-
ifornia aside from that of the Sharktooth Hill Bonebed, which
is located approximately 322 km to the north.

We have noted taxonomic differences as well as similarities
between the bird assemblages from these two sites. Unfor-
tunately, poor preservation of some of the bones from Upper
Oso Dam has caused us to qualify several identifications.
The apparently real absence of alcids in the Sharktooth Hill
Local Fauna, however, is a striking difference between that
assemblage and those of Middle and Late Miocene age from
Orange County. The site of deposition of the Upper Oso
Dam avifauna in Orange County is on the opposite (west)
side of the San Andreas Fault from the Sharktooth Hill Bone-
bed and, accounting for estimates of prior lateral (northward)
displacement (see Hill and Dibblee, 1953; Crowell, 1975),
might have originally been as much as 300 km farther south
in the Middle Miocene than at the present time. This means
that the two sites of deposition may have been over 600 km
apart in Middle Miocene time, and this latitudinal difference
must be one of the factors accounting for the faunal differ-
ences. The flightless alcids may not have had such wide dis-
tributions as the flying pelagic species. They may have been
restricted to a southerly distribution, and the significant lat-
itudinal differences between the two areas may explain why
members of the group have never been found in the Kern
County fossil record but are relatively common in Orange
County sediments.

ACKNOWLEDGMENTS

Our thanks go to present and former members of the Ver-
tebrate Paleontology staff of LACM for help in collecting this
material and in documenting the geology of the area. These
people include R.M. McKenzie, R.E. Raschke, D.H. Spitzer,
and B.J. Welton. The following volunteers also assisted in
collecting the material: J.C. Brown, C. Heron, E. Kukla, M.
Morgan, R.L. Price, D.M. Weir, and M.E. Zimmerman. S.L.
Olson and K.E. Campbell read a previous draft of the manu-
script and offered helpful suggestions. Representatives of the
Mission Viejo Company and the Santa Margarita Water Dis-
trict allowed us access to land under their control. S.E. Barnes

8 Contributions in Science, Number 383

Howard and Barnes: Miocene Birds from California

and L. Yasquez helped in preparing and editing the manu-
script for publication. Photographs are by J. De Leon, LACM

staff photographer. Figure 1 was drafted by A.E. Morey, and

Figure 2 was produced by M. Butler, both of LACM.

LITERATURE CITED

Addicott, W.O. 1972. Provincial middle and late Tertiary
moiluscan stages, Temblor Range, California. In Pro-
ceedings of the Pacific Coast Miocene Biostratigraphic
Symposium, ed. E.H. Stinemeyer, 1-26. Society of Eco-
nomic Paleontologists and Mineralogists, Bakersfield,
California.

American Ornithologists’ Union. 1983. Check-list of North
American Birds, sixth edition. Allen Press, Lawrence,
Kansas, xxix + 877 pp.

Barnes, L.G. 1976. Outline of eastern North Pacific fossil
cetacean assemblages. Systematic Zoology 25(4):32 1—
343.

Barnes, L.G., and E.D. Mitchell. 1984. Kentriodon obscurus
(Kellogg, 1931), a fossil dolphin (Mammalia: Kentrio-
dontidae) from the Miocene Sharktooth Hill Bonebed
in California. Contributions in Science, Natural History
Museum of Los Angeles County 353:1-23.

Brodkorb, P. 1963. Miocene birds from the Hawthorne
Formation. Quarterly Journal of the Florida Academy
of Sciences 26(2): 160-1 67.

Crowell, J.C. 1975. The San Andreas Fault in southern
California. In San Andreas Fault in southern California,
a guide to the San Andreas fault from Mexico to Carrizo
Plain, ed. J.C. Crowell, 7-27. California Division of
Mines and Geology, Special Report, 1 18.

Hill, M.L., and T.W. Dibblee, Jr. 1953. San Andreas, Gar-
lock, and Big Pine faults. A study of the character, his-
tory, and significance of their displacement. Geological
Society of America Bulletin 64(4):443-458.

Howard, H. 1966. Additional avian records from the Mio-
cene of Sharktooth Hill, California. Contributions in
Science, Natural History Museum of Los Angeles Coun-
ty 114:1-11.

. 1968. Tertiary birds from Laguna Hills, Orange

County, California. Contributions in Science, Natural
History Museum of Los Angeles County 142:1-22.

. 1970. A review of the extinct avian genus, Man-

calla. Contributions in Science, Natural History Mu-
seum of Los Angeles County 203:1-12.

. 1976. A new species of flightless auk from the

Miocene of California (Alcidae, Mancallinae). Smith-
sonian Contributions to Paleobiology 27:141-146.

. 1978. Late Miocene marine birds from Orange

County, California. Contributions in Science, Natural
History Museum of Los Angeles County 290:1-28.

. 1983. A list of the extinct fossil birds of California.

Bulletin of the Southern California Academy of Sciences
82(1): 1—1 1 .

. 1984. Additional avian records from the Miocene

of Kern County, California with the description of a new
species of fulmar. Bulletin of the Southern California
Academy of Sciences 83(2):84-89.

Marsh, O.C. 1870. Notice of some fossil birds from the
Cretaceous and Tertiary formations of the United States.
American Journal of Science, Ser. 2, 49( 1 46):205— 2 1 7.

Miller, L. 1925. Avian remains from the Miocene of Lom-
poc, California. Carnegie Institute of Washington Pub-
lication 349:107-1 17.

. 1961. Birds from the Miocene of Sharktooth Hill,

California. Condor 63:399-402.

. 1962. A new albatross from the Miocene of Cali-
fornia. Condor 64:421-472.

Milne-Edwards, A. 1874. Observations sur les oiseaux fos-
siles des faluns de Saucats et de la mollasse de Leognan.
Bibliotheque Ecole pratique des hautes Etudes, Paris,
Section des Sciences Naturelles XI (3): 1—12, pis. 1 1-12
(for 1874, publication date sometimes given as 1875).

Morton, P.K., R.V. Miller, and D.L. Fife. 1973. Prelimi-
nary geo-environmental maps of Orange County, Cali-
fornia. California Division of Mines and Geology Pre-
liminary Report 1 5:pls. 1-4.

Olson, S.L. 1985. The fossil record of birds. In Avian Bi-
ology, Vol. VIII, ed. D.S. Farner, J.R. King, and K.C.
Parkes, 79-238, Academic Press, Inc., Orlando, Florida.

Raschke, R.E. 1 984. Early and Middle Miocene vertebrates
from the Santa Ana Mountains, California. In The Nat-
ural Sciences of Orange County, ed. B. Butler, J. Gant,
and C. Stadum. Memoirs of the Natural History Foun-
dation of Orange County 1:61-67. Huntington Beach,
California.

Warter, S.L. 1976. A new osprey from the Miocene of
California (Falconiformes: Pandionidae). In Collected
Papers in Avian Paleontology Honoring the 90th Birth-
day of Alexander Wetmore, ed. S.L. Olson. Smithsonian
Contributions to Paleobiology 27:133-139.

Wetmore, A. 1930. Fossil bird remains from the Temblor
Formation near Bakersfield, California. Proceedings of
the California Academy of Sciences, Ser. 4, 19:85-93.

Yerkes, R.F., and R.H. Campbell. 1979. Stratigraphic no-
menclature of the central Santa Monica Mountains, Los
Angeles County, California. With a section on age of the
Conejo Volcanics by D.L. Turner and R.H. Campbell.
United States Geological Survey Bulletin 1457-E:i-iv,
E1-E31, pis. 1-3.

Wood, H.E., 2nd, R.W. Chaney, J. Clark, E.H. Colbert, G.L.
Jepsen, J.B. Reeside, Jr., and C. Stock. 1941. Nomen-
clature and correlation of the North American conti-
nental Tertiary. Bulletin of the Geological Society of
America 52:1-48, pi. 1.

Received 20 June 1986; accepted 6 January 1 987.

Contributions in Science, Number 383

Howard and Barnes: Miocene Birds from California 9

Hi

ii jijj ijj

mus

ture

Ills

j j« 1

Mi

ij SM'SPffi ■■fi'iii i : i..u: .■ ! . v-.i i. ■

■Ii wliMnftiiirf JIB I’.i fal ;

ipiWMiliiifid:! I

jiff 11

flHI ,

.^s !> ;; :;

fe:; kl -lib

llifj

i;ijUn 'lijil'ilijli!!

pH!

llffllilBll

■Bn

Wmm

lirlijllji;! i!:j!lii;i|'!i!I!i;i:jl,,;j

:;!iriti,,iI;ll;;;.!.ii-."1! j

i. I.i :;!j:iiil:i:

f :*i i ; »«• rttW ■; 4j *• if UipvJ. W^tUlllg vv »• UJ IJ V- \ :Jl, ■3'IUU'J I..» UJ

jffc-iiiir .Portuguese, as appropriate,, ap well as so H:ag1ii:li
ot required but are strongly recommended where ail

"" •;!' ; 1 !• = . - : « i ; • ■ • • .if: . ch ■! - : ; ■ = • • 1 ' ;I ! i"j : • 'i

v '|K|
: i1 n iHi'lif :i :--i ;■

' ' ' ION OF if ANUSCMPT j .! eJ . ;j ; :;f ;j if; fi f

WWW

Jljjjj

flip life • .i| ; ' .! j ■ ■ : j i ", ; > . - . 'j , ,i " ' | 1 '( j | !'|i ■ [ .2

"A including title page, . abstract, t«?tl t: , ;

3'Wl<

11

18 Mb

' captions/ Indicate :i tain:! zed words wim ’ ■! |;d

should begin on a piew page., in the : |:U ;

d| rnems; literature1' cited; tables, each

i"

i; and figure captions. All pages shot

iiiiaiiiilli!illl!lll .IIMii 'ii#^ ' " ' " '

. ill m '

liii'jtSi

!!|i;i !:ii:ii;:1

. :

It iiii ll I . , .

* drawings are best prepared for SO pe-rejinrl ;;

pcli|s;ii:5'-by '43 cm) in size; photographs are ;ji$:uaji|yi SA
8.75 inches (i 7 by 22 cm r lull piigejybdth);:; i ;

I

MljssjtpN Of- iySA.NUSCRi.pr :•;;;! ;; i ■ ;i!:f|||gj

lli HlSiill If ' ! iy Itf v I 111 II ! ; i |;i vHMbI. U;I|iS|||i

<«Rsqr||| in; a >j|[jl I

iiiia

ill!

ney

iHniiE

Mu:.-,

| ii |

HH

Bii/Iiti

mi

si

r. C:

:: »s

: irt'iau-ria: should: be

■fill

ii

111®

toMifeSi (i

- ;

i!l il

m

wij re

■ii

im

■emebt c-f' page : -

BBfll

il I! I l i l

mmm j

Il 'OIK Wl,!

'ii;;;,,,

CONTRIBUTIONS TOWARD A REVISION OF THE NEW WORLD
NOMADINE BEES. 2. THE GENUS MELANOMADA
(HYMENOPTERA: ANTHOPHORIDAE)

Roy R. Snelling and Jerome G. Rozen, Jr.

Contributions in Science, Number 384
Natural History Museum of Los Angeles County
18 June 1987

ISSN 0459-8113

Natural History Museum of Los Angeles County
900 Exposition Boulevard
Los Angeles, California 90007

CONTENTS

ABSTRACT 1

INTRODUCTION 1

SPECIMENS EXAMINED 1

SYSTEMATICS 1

Key to Adults of the Nearctic Genera, Melanomada Complex 2

Melanomada Cockerell 2

Key to Species of Melanomada 3

Melanomada annectens. New Species 3

Melanomada chica. New Species 6

Melanomada grindeliae (Cockerell) 6

Melanomada melanantha (Linsley) 8

Melanomada nimia, New Species 8

Melanomada sidaefloris (Cockerell) 9

Melanomada Species A 9

Melanomada Species B 9

Immature Stages 9

HOST ASSOCIATIONS AND BIONOMICS 12

ACKNOWLEDGMENTS 12

LITERATURE CITED 12

CONTRIBUTIONS TOWARD A REVISION OF THE NEW WORLD
NOMADINE BEES. 2. THE GENUS MELANOMADA
(HYMENOPTERA: ANTHOPHORIDAE)

Roy R. Snelling1 and Jerome G. Rozen, Jr.2

ABSTRACT. The cleptoparasitic bee genus Hesperonomada is syn-
onymized with Melanomada. Melanomada, together with Triopa-
sites and Paranomada in North America and Brachynomada and
perhaps Kelita in South America, is a closely knit group (the Me-
lanomada complex) that is morphologically distinctive, both as adults
and larvae, within the tribe Nomadini as it presently is constituted.
A key to the North American genera of this complex is given, as is
a key to the six known species of Melanomada. Three new species
of Melanomada are described and illustrated: M. annectens (Cali-
fornia, U.S. A.), M. chica (Baja California Sur, Mexico), and M. nimia
(Kansas, U.S. A.).

The mature larvae of M. annectens is described and illustrated
and the relationships of the genus to other members of the complex,
and to the Nomadini as a whole, are discussed in terms of larval
characteristics.

As far as known, all North American members of the Melanomada
complex are cleptoparasites of the anthophorid genus Exomalopsis.
Known host relationships are discussed, suspected relations are men-
tioned, and the relationships of both the South American Brachy-
nomada and Kelita to the Panurginae are outlined.

INTRODUCTION

Recently, we undertook a study of the nesting biology of
Exomalopsis nitens Cockerell, near Corona, Riverside Coun-
ty, California, in hope of recovering information about, and
immature stages of, its reported cleptoparasite Hesperono-
mada melanantha (Rozen, 1984; Rozen and Snelling, 1986).
Upon examining the cleptoparasite from the nesting area, we
determined that the species was not H. melanantha. and,
although the two species were obviously similar, this new
species shared a number of features with the genus Mela-
nomada. These characteristics immediately placed into ques-
tion the distinctiveness of the genera Hesperonomada and
Melanomada. This present contribution, in which Hespe-
ronomada is synonymized with Melanomada. results from
our attempt to resolve the relationships of these two genera,
and makes known a number of distinctive new species be-
longing to Melanomada. We not only describe the new species
but also present information about the host relationships and
immature stages of one of these.

Contributions in Science, Number 384, pp. 1-12
Natural History Museum of Los Angeles County, 1987

Snelling has been primarily responsible for preparing the
Systematics section of this manuscript, whereas Rozen has
drafted the sections on Immature Stages and Host Associa-
tions and Bionomics.

SPECIMENS EXAMINED

Adult specimens examined during the course of this study
were from the following collections: Academy of Natural
Sciences of Philadelphia (ANSP), American Museum of Nat-
ural History (AMNH), California Academy of Sciences (CAS),
Natural History Museum of Los Angeles County (LACM),
University of Kansas (UKAN), United States National Mu-
seum of Natural History (USNM), and Utah State University
(UTSU). Larval specimens are in the American Museum of
Natural History.

SYSTEMATICS

The Nearctic genera Melanomada Cockerell, 1903b (includ-
ing Hesperonomada Linsley, 1939, synonymized herein),
Triopasites Linsley, 1939, and Paranomada Linsley and
Michener, 1937, together with the South American Brachy-
nomada Holmberg, 1886, form a close-knit and distinctive
group within the Nomadini. An additional South American
genus, Kelita Sandhouse, 1943, may be a part of this same
complex, but possesses some unique features that are greatly
at variance with this group and it is, therefore, not included
in the following discussion.

The following apomorphies (many cited by Ehrenfeld and
Rozen, 1977, and Rozen, 1977) define this group within the
tribe Nomadini; plesiomorphic states are indicated in brack-
ets. ( 1 ) Anterior and posterior basal angles of mandible about
equally distant from eye [anterior angle more distant than

1 . Entomology Section, Natural History Museum of Los Angeles
County, 900 Exposition Blvd., Los Angeles, California 90007.

2. American Museum of Natural History, Central Park West at
79th Street, New York, New York 10024.

ISSN 0459-8113

basal angle in other Nomadini]. (2) Maxillary palpus with
three to six segments and less than one-half as long as galea
(except in Melanomada chica) [other Nomadini with six-
segmented maxillary palpus more than one-half as long as
galea]. (3) Metacoxa with inner, dorsal carina or lamella [ab-
sent in other Nomadini (small but present in Hexepeolus)].
(4) Metasomal sternum 6 of female produced posteriorly as
a short, acute or rounded, hairy process (Fig. 4) [truncate or
bilobed in other Nomadini]. (5) Metasomal sternum 7 of
female with a pair of slender, apical lobes with a few coarse
setae arising from their inner apical margins (Fig. 5) [sternum
7 subtruncate or with a pair of broad, well-separated lobes
with coarse, spine-like setae at apex or along outer margins],
(6) Metasomal sternum 7 of male without basal apodeme
(Figs. 8, 10, 12, 14) [basal apodeme present in other No-
madini]. (7) Gonostylus of male genitalia long and narrow
(Figs. 16-18) [short and broad in most other Nomadini, but
with some exceptions in Hypochrotaenia and Nomada ]. (8)
Integument smooth and shiny; punctures, when present, not
confluent or contiguous [integument dull to shiny; punctures
of head and thorax commonly confluent, at least in some
areas]. (9) Mesepisternum of female, and often male as well,
with dense, prostrate highly plumose pubescence on upper
one-half or more, lower one-half or less with pubescence
sparse and simple, or nearly so [in other Nomadini, mes-
episternal pubescence usually uniform, and usually weakly
plumose, but sometimes densely pubescent above and be-
coming less dense ventrad]. (10) Prepygidial brush of female
metasomal tergum 5 suberect, the hairs acute or subacute
[prepygidial brush of female suberect to prostrate, hairs, ex-
cept anteriorly, spatulate or apically broadened and reflec-
tive]. (11) Mature larva3 with well-developed, paired dorsal
tubercles (Fig. 19) [no dorsal tubercles in other Nomadini].
( 1 2) Mature larva with well-developed hypostomal ridge [ridge
weak in other Nomadini], (13) Vertex of mature larva with
protrusions [vertex recessed in other Nomadini]. (14) Man-
dibles of larva long (Figs. 22-24) [short in other Nomadini].

Although the Melanomada complex appears highly dis-
tinctive, we do not believe it should be accorded tribal status
because of the limited number of synapomorphies defining
it, because of the unclear relationship of Kelita to it, and
because of our limited knowledge of Brachynomada. None-
theless, these genera are very different from what we know
of the other members of the Nomadini.

Within the complex, Paranomada is the most distinctive,
easily recognized by the greatly flattened body and the smooth
and shiny, virtually impunctate integument. The remaining
genera are more similar one to another, although Triopasites,
with its characteristic marginal cell and the patch of plumose
hairs on the second metasomal sternum, is difficult to confuse
with the others. Brachynomada , in South America, seems to
be very close to Melanomada, but is too poorly known for
an adequate assessment of its characters.

As mentioned above, Kelita shares some characters with
the members of this complex. Of the features cited above,

3. Larval features are known only for representatives of Mela-
nomada, Triopasites, Nomada, and Hypochrotaenia.

Kelita differs from the Paranomada-Melanomada-Triopa-
sites-Brachynomada complex in characters 2 (maxillary pal-
pus more than one-half as long as galea), 3 (metacoxa without
inner, dorsal carina), 7 (gonostylus short and broad), 8 (in-
tegument of head and mesosoma pebbled), 9 (entire mes-
epistemum covered with prostrate, plumose pubescence), and
10 (prepygidial hairs short and broad). Sternum 5 of the
female has a transverse ridge across the base and the ridge
is produced distally along the midline as a narrow, variably
acute process. Sternum 6 of the female is similar to that of
the genera of the Melanomada complex, but bears at its apex
a few very short, broad, blunt setae. Additional unusual fea-
tures of Kelita are cited by Ehrenfeld and Rozen (1977).

As a group, the members of the Melanomada complex
may be recognized by the following abbreviated suite of char-
acters: anterior and posterior basal angles of mandibles al-
most equidistant from eye; maxillary palpus not more than
one-half as long as galea (except in M. chica), three- to six-
segmented; male gonostylus long and slender; metasomal
sternum 6 of female produced posteriorly as an acute or
rounded, hairy process and sternum 7 with a pair of slender
apical lobes with a few coarse setae along their margins;
integument smooth and shiny between fine, well-separated
punctures.

KEY TO ADULTS OF THE NEARCTIC GENERA,
MELANOMADA COMPLEX

la. Mesosoma about as deep as wide; integument conspic-
uously punctate (exceptionally, vertex and thoracic dor-
sum very sparsely and minutely punctate), mesepister-
num always sharply punctate; mesocoxa as long as, or
a little longer than, distance from its base to base of hind

wing 2

b. Mesosoma greatly flattened, width distinctly greater than
depth; integument smooth and shiny, all areas virtually
without punctures; mesocoxa much longer than distance

from its base to base of hind wing Paranomada

2a. Forewing with two or three submarginal cells and apex
of marginal cell acute, ending on wing margin; margin
of metasomal sternum 2 not abruptly depressed and

without broad, dense patch of plumose hairs

Melanomada

b. Forewing with three submarginal cells and apex of mar-
ginal cell narrowly rounded and bent away from wing
margin; margin of metasomal sternum 2 abruptly de-
pressed and, across middle one-third, with a conspicu-
ous, dense patch of pale, profusely plumose hairs ....

Triopasites

Melanomada Cockerell

Nomada subg. Melanomada Cockerell, 1903b:587. Type-
species: Nomada grindeliae Cockerell, 1903a. Monotypic
and original designation. Cockerell, 191 la:649. Cockerell,
191 lb:226. Swenk, 1913:8, 15. Linsley and Michener, 1937:
285.

2 Contributions in Science, Number 384

Snelling and Rozen: New World Genus Nomada

Hesperonomada Linsley, 1939:5. Type-species: Hesperono-
inada melanantha Linsley, 1939. Monotypic and original
designation. Linsley and Michener, 1937:291-292. Mich-
ener, 1944:274-275. Snelling, 1986:6. NEW SYNONY-
MY.

Melanomada: Michener, 1944:275. Rodeck, 1945:202-203.
Snelling, 1986:6.

Both Michener (1944) and Snelling (1986) separated Hes-
peronomada from Melanomada by its five-segmented (rather
than six) maxillary palpus and the presence of two (rather
than three) submarginal cells in the forewing. Michener fur-
ther noted that in Melanomada the mesocoxa is about as
long as the distance from its summit to the base of the hind
wing; the mesocoxa was said to be shorter in Hesperono-
mada. Two of the new species described below exhibit in-
termediate states and effectively eliminate the distinctions
that formerly separated these two genera.

Both of these new species described below (M. annectens,
M. chicd) possess two submarginal cells in the forewing, as
in Hesperonomada melanantha, but they have a six-seg-
mented maxillary palpus, as in Melanomada. In both, also,
the mesocoxa is about as long as the distance from its summit
to the base of the hind wing. One of the new species (M.
annectens) (Fig. 6) in particular, is annectant between Mel-
anomada and Hesperonomada. It has a small vein stub pres-
ent on vein M that is evidently the remnant of crossvein 1st
r-m (especially well developed in males), for it is in the correct
position.

Rodeck (1945) united the two genera, though it is not clear
whether he intended to treat Hesperonomada as a synonym
or subgenus of Melanomada. However, he later treated the
two genera as separate (Rodeck, 1951). In our view, Hes-
peronomada can no longer be considered separate and dis-
tinct, either as a genus or as a subgenus of Melanomada.

KEY TO SPECIES OF MELANOMADA

la. Forewing with two submarginal cells 2

b. Forewing with three submarginal cells 4

2a. Maxillary palpus six-segmented; large, triangular me-
diobasal area of clypeus smooth, shiny, and impunctate

3

b. Maxillary palpus five-segmented, with segments 4 and
5 fused (Fig. 2); clypeus uniformly densely punctate,
sometimes with narrow median impunctate line on basal

one-half melanantha (Linsley)

3a. Maxillary palpus about one-half as long as galea, apical
segment distinctly shorter than preceding segment and
uniformly cylindrical (Fig. 1 ); mesoscutal punctures about
0.01 mm diameter or more and separated by much less

than a puncture diameter in center of disc

annectens, new species

b. Maxillary palpus slightly longer than galea, apical seg-
ment distinctly longer than preceding segment and
abruptly wider over apical one-half (Fig. 3); mesoscutal
punctures minute, hardly greater in diameter (about 0.006
mm) than hairs arising from them and separated by two

or more puncture diameters across middle of disc ....

chica, new species

4a. Clypeal disc weakly convex or flattened, but if flattened,
then disc of mesoscutum is uniformly and densely punc-
tate; sides of declivitous face of propodeum partially con-
cealed by dense, prostrate, highly plumose hairs .... 5
b. Clypeal disc weakly concave and disc of mesoscutum
polished, punctures variable in size and variably spaced,
with extensive impunctate areas; side of declivitous face
of propodeum thinly covered by subappressed, short-

plumose hairs that do not conceal integument

grindeliae (Cockerell)

5a. Clypeal disc weakly convex; punctures in center of meso-
scutum variably spaced, with some interspaces exceed-
ing three puncture diameters; surface of mesostemum,
especially near coxae, partially concealed by dense, pros-
trate, highly plumose pubescence; middle of propodeal
triangle granulose and without evident longitudinal ru-

gulae sidaejloris (Cockerell)

b. Clypeal disc flattened; mesoscutal punctures uniform in
size, evenly spaced in center with interspaces 1.5 punc-
ture diameters or less; mesosternum not concealed by
dense pubescence; middle of propodeal triangle with sev-
eral fine, longitudinal rugulae nimia, new species

Melanomada annectens , new species

Figures 1,6, 10, 11, 17, 25

DIAGNOSIS. This species is separable from M. grindeliae
and M. sidaefloris by the presence of two, rather than three,
submarginal cells in the forewing and from M. melanantha
by the six-segmented, rather than five-segmented, maxillary
palpus. It is further distinguished from all three by the pres-
ence of a distinct, median impunctate area on the clypeus.
From M. chica it is separated by its coarser punctation, max-
illary galea being about twice as long as the maxillary palpus,
densely punctate mesoscutum, and larger size.

DESCRIPTION. Female. Measurements. Head width
1.86-2.05; head length 1.35-1.67; wing length 4. 7-4. 8; total
length 6. 0-6. 6 mm.

Head 1.2-1. 4 times broader than long; vertex strongly
arched above tops of eyes; inner eye margins strongly con-
vergent below; lower interocular distance 0.85-0.87 times
upper interocular distance. Posterior ocelli separated from
preoccipital margin by much less than their diameter; ocel-
locular distance 2. 6-3. 8 times anterior ocellus diameter; in-
terocellar distance 1. 8-2.0 times anterior ocellus diameter.
Scape about 2.7 times longer than distance between antennal
sockets and, in full frontal view, its apex on level of anterior
margin of anterior ocellus; first flagellar segment slightly lon-
ger than broad; second flagellar segment 1.15-1.31 times
longer than first; middle flagellar segments about one-third
longer than first and about one-third longer than broad.

Glossa normal, i.e., much longer than combined lengths
of first two labial palpal segments; labial palpus flattened,
first segment about 1 1.5 times longer than wide and about
2.3 times longer than second; maxillary palpus six-segment-
ed, less than one-half as long as galea, second segment about

Contributions in Science, Number 384

Snelling and Rozen: New World Genus Nomada 3

-I

4 Contributions in Science, Number 384

Snelling and Rozen: New World Genus Nomada

1.6 times longer than sixth, sixth segment normal in shape,
i.e., approximately parallel-sided and narrowed at apex.

Integument smooth and shiny; clypeus with broad, im-
punctate median line, disc otherwise conspicuously punctate,
with most punctures about 0.02-0.03 mm diameter, sepa-
rated by 1-2 puncture diameters, and with smaller punctures
(about 0.0 1 mm diameter) irregularly scattered between; frons,
vertex, and preocciput similarly punctate, but much of face
obscured by densely plumose, appressed pubescence. Meso-
scutum and scutellum subcontiguously punctate, but with
irregular interspaces up to a puncture diameter; punctures
0.02-0.03 mm diameter. Upper, densely pubescent portion
of mesepisternum with punctures contiguous to subcontig-
uous, mostly about 0.03 mm diameter; lower, sparsely pu-
bescent area with punctures similar but subcontiguous to
dense. Metasomal tergal discs slightly roughened between
punctures; those of second tergum distinct, slightly trans-
verse, about 0.0 15 mm diameter; punctures becoming slight-
ly coarser and closer on succeeding segments.

Forewing (Fig. 7) with two submarginal cells, second with
1st r-m stub arising from M.

Flead and mesosoma blackish, metasoma ferruginous;
mandible, labrum, apical margin of clypeus, underside of
first flagellar segment, pronotal lobe, and most of legs (except
irregular darker areas on femora and protibia), dusky fer-
ruginous. Tegula ferruginous. Wings transparent brownish,
marginal cell and broad apical area darker; veins and stigma
dark brown.

Male. Measurements. Head width 1.84-2.13; head length
1.48-1.74; wing length 4. 7-5. 3; total length 5. 5-6. 3 mm.

Head about 1.2 times broader than long; vertex strongly
convex above tops of eyes; lower interocular distance 0.85-
0.88 times upper interocular distance. Posterior ocelli sep-
arated from preoccipital margin by less than their diameter;
ocellocular distance 2. 4-2. 5 times anterior ocellus diameter;
interocellar distance 2. 1-2.3 times anterior ocellus diameter.
Scape as described for female; first flagellar segment slightly
longer than broad; second flagellar segment 1.0- 1.1 times
longer than first.

Otherwise about as described for female, but impunctate
area of clypeus inverted-deltoid in shape, apex almost reach-
ing apical margin. Abdomen blackish, with segment margins
reddened, and legs dark reddish brown.

TYPE MATERIAL. Holotype female, allotype, and 3 fe-
male paratypes: Indian Truck Trail, 12 mi. S Corona, Riv-
erside Co., California, 24 May 1985 (J.G. Rozen and R.R.
Snelling), at nest site of Exomalopsis nitens Cockerell. Para-
types (all from same locality): 4 59, 19 June 1984 (R.R. Snell-
ing); 2 99, 4 33, 20 June 1984 (R.R. Snelling and E. Williams);
1 9, 2 May 1985 (R.R. Snelling); 1 9, 4 <33, 19 May 1985 (R.R.
Snelling); 2 99, 1 3, 22 May 1985 (J.G. Rozen); 1 9, 1 3, 25

May 1985 (J.G. Rozen and R.R. Snelling); 3 99, 3 <33, 26 May
1985 (J.G. Rozen and R.R. Snelling). Holotype and allotype
in LACM; paratypes in AMNH, LACM, UKAN, USNM.

ADDITIONAL MATERIAL (not paratypes). Los Angeles
Co.: 1 3, Hollywood, 12 Aug. 1956 (D.L. Dahlsten; LACM);
1 3, Westwood Hills, 27 July 1935 (paratype of H. mela-
nantha; UCB). Orange Co.: 1 9, Irvine Park, 3 Sept. 1962
(M.E. Irwin; UTSU). San Diego Co.: 1 9, Laguna Beach, no
date (Baker; paratype of H. melanantha; UCB?); 1 3, La Jolla,
14 Aug. 1911 (J.C. Bridwell; USNM); 1 9, near La Mesa, 28
June 1954 (F.X. Williams; USNM).

ETYMOLOGY. The specific name is a Latin adjective,
meaning joining or linking and refers to the intermediate
position of this species in revealing the relationship of Me-
lanomada and Hesperonomada.

DISCUSSION. Both sexes of M. annectens closely resem-
ble those of M. melanantha. The most conspicuous difference
lies in the number of maxillary palpal segments, five in M.
melanantha and six in M. annectens. In M. melanantha the
fifth segment may be seen to result from the fusion of seg-
ments five and six; this combined segment is longer than the
second segment (Fig. 2).

The clypeus is uniformly densely punctate in both females
and males of M. melanantha. although there may be a very
narrow median line that is free of punctures. Females of M.
annectens have approximately the middle one-fourth to one-
third of the clypeal disc impunctate or nearly so. In the males,
the impunctate area is an inverted delta, with the base ex-
tending between the subantennal sutures and the apex reach-
ing almost to the apical margin of the clypeus.

Other differences between the two species include the
slightly longer first flagellar segment of M. annectens that is
little, if any, shorter than the second; in M. melanantha the
first segment is distinctly shorter than the second. Males of
both species have the outer, dorsal margins of the pro- and
mesocoxae more or less carinate, but the carinae are sharper
and higher in M. annectens than in M. melanantha. Simi-
larly, the inner dorsal carina of the metacoxa is better de-
veloped in M. annectens.

The second submarginal cell of the forewing is consistently
different between the two species. In M. annectens there is
a small remnant of crossvein 1st r-m (Fig. 6). This remnant
is consistently lacking in the forewing of M. melanantha.

Genitalic structures of the males of the two species are very
similar. Sternum 7 of M. annectens (Fig. 10) is about twice
as long as broad, with its apex more broadly rounded; in M.
melanantha (Fig. 8) sternum 7 is about as broad as long, its
apex more acute. Similarly, the apical process of sternum 8
is broader and more rounded in M. annectens (Fig. 1 1) than
in M. melanantha (Fig. 9). Melanomada annectens has a
slight distal broadening of the gonostylus (Fig. 13), whereas

Figures 1-7. Melanomada species. 1, maxillary palpus and galea of M. annectens ; 2, maxillary palpus of M. melanantha: 3, maxillary palpus
and galea of M. chica; 4, 5, metasomal sterna 6 and 7, M. grindeliae, female; 6, 7, forewing of M. annectens and M. grindeliae. respectively.
Scale line (Figs. 1-3) = 0.25 mm.

Contributions in Science, Number 384

Snelling and Rozen: New World Genus Nomada 5

in M. melanantha it is uniformly slender (Fig. 16). The apex
of the penis valve is narrower in M. melanantha (Fig. 16)
than in M. annectens.

Melanomada chica, new species

Figures 3, 25

DIAGNOSIS. This species may be separated from M.
grindehae and M. sidaefloris by its small size, exceptionally
fine and sparse punctures, two (rather than three) submar-
ginal cells in the forewing, and six-segmented (rather than
five-segmented) maxillary palpus. From M. annectens and
M. melanantha, M. chica may also be distinguished by its
small size and minute, sparse punctures. It is further distin-
guished from M. melanantha by the six-segmented maxillary
palpus. From all species it is identifiable by the maxillary
palpi being as long as the galea.

DESCRIPTION. Female. Measurements. Head width
1.16; head length 0.93; forewing length 2.9; total length 3.5
mm.

Head about 1.3 times wider than long; vertex strongly
convex above summits of eyes; inner eye margins strongly
convergent below; lower interocular distance 0.78 times up-
per interocular distance. Posterior ocelli separated from
preoccipital margin by less than their diameters; interocellar
distance 1.7 times anterior ocellus diameter; ocellocular dis-
tance 2.4 times anterior ocellus diameter. Scape about 1.7
times longer than distance between antennal sockets and, in
frontal view, short of anterior ocellus by about its apical
width; first flagellar segment broader than long; second fla-
gellar segment about 1 .4 times longer than first; middle fla-
gellar segments about as long as broad.

Glossa shorter than combined lengths of first two segments
of labial palpus; labial palpus flattened, first segment about
five times longer than wide and about three times longer than
second; maxillary palpus six-segmented, slightly longer than
galea (Fig. 3), second and sixth segments longest and sub-
equal, sixth segment narrow and parallel-sided for about one-
half its length and abruptly broader in apical one-half.

Punctures everywhere minute; clypeal punctures separated
by a puncture diameter or more, a few relatively coarser
punctures (up to about 0.02 mm diameter), but most much
finer (about 0.006 mm diameter), punctures densest near
apical margin. Lower part of face smooth and shiny, with
only sparse minute punctures, most hardly greater in di-
ameter than hairs arising from them; upper frons, vertex, and
preocciput largely bare and with scattered barely perceptible
punctures. Mesoscutal and scutellar punctures variably spaced
across discs, but most interspaces more than two puncture
diameters, most punctures about 0.006 mm diameter, a few
up to 0.01 mm. Upper mesepisternum smooth and shiny
between close punctures up to about 0.02 mm diameter,
lower portion smooth, shiny, and sparsely pubescent, punc-
tures sparse, most less than 0.01 mm diameter. Metasomal
terga shiny, but not polished, between sparse, obscure, trans-
verse, piligerous punctures that are best defined, densest, and
coarsest (up to 0.02 mm diameter) on fourth and fifth terga.

Forewing with two submarginal cells.

Upper half of head and entire thorax blackish; lower half
of head, antennae, and legs dusky ferruginous, tibiae and
following segments lighter; metasoma ferruginous. Tegula
clear yellowish red. Wings transparent brownish, darker along
apical margins; veins and stigma dark brown.

Male. Unknown.

TYPE MATERIAL. Holotype female: vicinity of Estacion
Microondas “Ligui,” 425 m elev., 48 km S Loreto, Baja
California Sur, MEXICO, 7 Sept. 1977 (R.R. Snelling), in
LACM.

ETYMOLOGY. The specific name is a Spanish adjective
meaning small, and is appropriate for this diminutive species.

DISCUSSION. The diminutive size and shortened
mouthparts, except the maxillary palpus, will immediately
separate M. chica from all other species of Melanomada. The
short mouthparts are a common characteristic of bees as-
sociated with prostrate species of the plant genus Euphorbia.

Presumably the bee host of M. chica will prove to be one
of the smaller species of Exomalopsis, such as E. rufiventris,
which occurs in the vicinity of the type locality.

Melanomada grindeliae (Cockerell)

Figures 4, 5, 7, 12, 13, 18, 25

Nomada grindeliae Cockerell, 1903a:210; <3.

Nomada ( Melanomada ) grindeliae: Cockerell, 1903b:587.
Nomada ( Melanomada ) heleniella Cockerell, 1911 a:648; 2

<3. NEW SYNONYMY.

Melanomada grindeliae: Michener, 1944:275.

Melanomada heleniella: Rodeck, 1951:1207.

DIAGNOSIS. Melanomada grindeliae is separable from
M. annectens, M. chica. and M. melanantha by the presence
of three, rather than two, submarginal cells in the forewing;
it is further separable from M. chica by the short maxillary
palpus, only about one-half as long as the galea, and notably
larger size (head width more than 1.25 mm); it differs from
M. sidaefloris and M. nimia by the sparsely pubescent pro-
podeal declivity and other characters noted in the key.

The types of N. grindeliae (ANSP) and N. heleniella
(USNM) have been examined and compared directly with
one another. Aside from the slightly smaller size of the N.
heleniella types, there are no notable differences between the
two.

In addition to the types, we have seen several specimens
of both sexes from Lincoln, Nebraska, the type locality of N.
grindeliae. Other material that we examined includes the
following.

COLORADO, Pueblo Co.: 1 <3, Pueblo, 5000 ft. elev., 19
Aug. 1960 (R.H. Dreisbach; UTSU). KANSAS, Chase Co.:

1 2, 2 mi. W Elmdale, 30 Sept. 1975 (Michener and Chabot;
UKAN). Douglas Co.: 1 2, Lawrence, 5 Sept. 1954 (C.W.
Rettenmeyer; UKAN), on Solidago sp.; 1 2, same locality, 7
Sept. 1954 (E.A. Cross; UKAN), on Bidens sp. OKLA-
HOMA, Alfalfa Co.: 2 <3<3, 6 mi. S Driftwood, 27 Sept. 1969
(K.L. Johnson; UKAN), on Aster subulatus.

Two females collected at Lincoln, Nebraska, by J.C. Craw-
ford (USNM) bear the notation “at Nomia place.” We as-
sume that this is a casual association. The true host of M.

6 Contributions in Science, Number 384

Snelling and Rozen: New World Genus Nomada

8

10

12

14

Figures 8-18. Males, Melanomada species. Metasomal sterna 7 and 8, respectively, of: 8, 9, M. melanantha; 10, 11, M. annectens ; 12, 13,
M. grindeliae; 14, 15, M. nimia. Genitalia, ventral view: 16, M. melanantha ; 17, M. annectens; 18, M. gnndeliae. Scale line = 0.50 mm.

Contributions in Science, Number 384

Snelling and Rozen: New World Genus Nomada 7

grindeliae will probably be a species of Exomalopsis, such
as E. pygmaea (Cresson), which occurs throughout the known
range of M. grindeliae.

Melanomada melanantha (Linsley),
new combination

Figures 2, 8, 9, 1 6, 25

Hesperonomada melanantha Linsley, 1939:6-7; 2 <5.

DIAGNOSIS. This species is distinguished from all other
Melanomada. except M. annectens and M. chica, by the
presence of two, rather than three, submarginal cells in the
forewing; from M. annectens and M. chica it may be distin-
guished by the five-segmented, rather than six-segmented,
maxillary palpus.

The type locality of M. melanantha is in Monterey County,
California. Paratypes were from Contra Costa, Los Angeles,
and San Diego counties. The specimens from the latter two
counties are not M. melanantha, but rather M. annectens.
In addition to the types, we have seen the following speci-
mens of M. melanantha, all from California.

Alameda Co.: 1 2, Pleasanton, 20 Sept. 1 957 (O.A. Stevens;
USNM), on Centaurea sp. Contra Costa Co.: 1 2, Antioch,
25 Aug. 1955 (C.D. MacNeill; UCB); 1 2, 1 <3, same locality,
22 Sept. 1954 (J.G. Rozen; AMNH), on Grindelia sp.; 1 <3,
same locality, 5 Sept. 1936 (M.A. Embury; UCB); 2 22, same
locality, 22 Sept. 1954 (P.D. Hurd, C.D. MacNeill; UCB,
USNM), on Grindelia sp.; 4 22, 1 3, same locality, 14 Oct.
1954 (P.D. Hurd, M.S. Wasbauer; UCB, USNM), on Grin-
delia sp.; 3 22, 2 mi. W Pittsburg, 19 Sept. 1957 (J.A. Powell;
UCB), 2 on Grindelia camporum, 1 on Baccharis g/utinosa.
San Joaquin Co.: 1 3, Stockton, 12 June 1916 (R. Stinchfield;
UCB), on Grindelia (camporum?). Santa Clara Co.: 1 3, no
locality or date (Harkins; UCB).

The differences between this species and the superficially
similar M. annectens are discussed under the latter. The host
of M. melanantha is unknown.

Melanomada nimia, new species

Figures 14, 15, 25

DIAGNOSIS. This species can be readily distinguished
from others with three submarginal cells in the forewing ( M .
grindeliae and M. sidaefloris) by the uniformly densely punc-
tate mesoscutum (sparsely punctate in other species).

DESCRIPTION. Female. Measurements. Head width
1.45-1.65; head length 1.22-1.36; wing length 3. 9-4. 3; total
length 4. 5-4. 9 mm.

Head 1.2-1. 3 times broader than long; vertex moderately
arched above tops of eyes; inner eye margins strongly con-
vergent below; lower interocular distance 0.82-0.84 times
upper interocular distance. Posterior ocelli separated from
preoccipital margins by much less than their diameter; ocel-
locular distance 2. 4-2. 7 times anterior ocellus diameter; in-
terocellar distance 2.0 times diameter of anterior ocellus.
Scape about 1.5- 1.6 times longer than distance between an-
tennal sockets and, in frontal view, its apex distinctly short

of level of anterior ocellus; first flagellar segment broader
than long; second segment 1.5-1. 6 times longer than first.

Glossa much longer than combined lengths of first two
labial palpal segments; labial palpus flattened, first segment
more than 10 times longer than wide and more than twice
longer than second; maxillary palpus six-segmented and less
than one-half as long as galea.

Integument smooth and shiny; clypeus with sparse, fine
punctures and irregularly distributed minute punctures; su-
praclypeal area mostly impunctate, with scattered fine and
minute punctures laterad; frons and vertex with sparse, mi-
nute punctures; face, including clypeus, below midlength
largely obscured by dense, appressed, copiously plumose pu-
bescence. Mesoscutum and scutellum shiny between fine
punctures (about 0.02 mm diameter) that are mostly sepa-
rated by less than their diameters. Mesepistemal punctures
fine and subcontiguous, integument largely obscured by co-
piously plumose, appressed hairs. Sides of propodeal disc
largely obscured by copiously plumose appressed hairs; me-
dian, apubescent area of basal triangle dull, with a few fine
longitudinal rugulae. Basal abdominal terga shiny between
scattered, obscure, minute punctures; succeeding terga with
surface increasingly roughened, hence duller.

Forewing with three submarginal cells.

Head and thorax blackish, metasoma and legs beyond coxae
ferruginous; mandible (except piceous distal two-thirds), la-
brum, apical margin of clypeus, pronotal lobe, and tegula,
also ferruginous. LInderside of flagellum dark reddish brown,
basal segment paler. Wings transparent, slightly brownish,
marginal cell darker; veins and stigma dark brown.

Male. Measurements. Head width 1.44-1.57; head length
1.17-1.27; wing length 3. 7-4.0; total length 4. 4-4. 8 mm.

Head 1 .2 times broader than long, vertex moderately arched
above tops of eyes; inner eye margins strongly convergent
below, lower interocular distance 0.81-0.83 times upper in-
terocular distance. Ocelli as described for female. Scape as
described for female; first flagellar segment slightly longer
than broad and about 1.3 times length of second; median
flagellar segments about as broad as long.

Otherwise about as described for female, but legs and meta-
somal segments 3-6 dark brownish to blackish.

TYPE MATERIAL. Holotype female, allotype, and two
female paratypes from 2 mi. W Elmdale, Chase Co., KAN-
SAS, 30 Sept. 1975 (Michener and Chabot); 2 22, 2 33, Yates
Center, Woodson Co., KANSAS, 7 Sept. 1949 (Michener
and Beamer). Holotype, allotype, and most paratypes in
UKAN; one female paratype in LACM.

ETYMOLOGY. The specific name is a Latin adjective,
meaning too much or excessive and is in reference to the
densely punctate mesoscutum.

DISCUSSION. This species is most similar to the sym-
patric M. grindeliae, but in the latter species the punctures
of the mesoscutum are finer and, at least in the middle of
the disc, separated by three times a puncture diameter or
more. Other differences include the more convex clypeus,
much denser mesepistemal and propodeal pilosity of M. ni-
mia, as well as the red-marked metasoma of the M. nimia
male.

8 Contributions in Science, Number 384

Snelling and Rozen: New World Genus Nomada

From M. sidaefloris both sexes differ in the much sparser
mesosternal pubescence. Although some plumose hairs are
present anterior to the mesocoxae, they are not appressed to
form a dense, surface-concealing patch, as in M. sidaefloris.
The mesoscutal punctation of M. nimia is much denser than
that of M. sidaefloris, as noted in the key.

The finely, longitudinally rugulose median area of the pro-
podeal triangle separates M. nimia from both M. grindeliae
and M. sidaefloris. From the West Coast species, M. mela-
nantha, M. annectens, and M. chica, M. nimia is immediately
separable by the presence of three submarginal cells in the
forewing.

The genitalic capsule of M. nimia (not illustrated) is very
similar to that of M. grindeliae, but with much sparser setae
along the shaft of the gonostylus. The hidden sterna, however,
are different. Sternum 7 (Fig. 14) of M. nimia is proportion-
ately broader, though similar in shape. Elongate, slender, and
apically acute, sternum 8 of M. nimia (Fig. 1 5) is very dis-
tinctly different from the short, broad, and truncate to sub-
truncate segments of most other species of Melanomada.

Melanomada sidaefloris (Cockerell)

Figure 25

Nomada pennigera var. sidaefloris Cockerell, 1898:59; <3.
Melanomada sidaefloris: Rodeck, 1945:203. Rozen, 1977:7-

9; larva.

DIAGNOSIS. This species is separable from M. annec-
tens, M. chica, and M. melanantha by the presence of three
submarginal cells in the forewing. Although very similar to
M. grindeliae, both sexes are readily separable by the pres-
ence of dense patches of profusely plumose, mainly prostrate
pubescence on the side and venter of the thorax and on the
side of the propodeal disc. The mesoscutum of M. grindeliae
is uniformly very sparsely punctate, while in M. sidaefloris
the punctures are moderately dense in most areas, becoming
irregularly spaced and sparse posteromedially. This species
is also similar to M. nimia, but in that species the mesoscu-
tum is very uniformly punctate, the punctures separated by
about one-half a puncture diameter, and the propodeal tri-
angle bears several short longitudinal rugulae.

In addition to the type, from Mesilla, New Mexico, we
have seen a few specimens all from New Mexico (Hidalgo
County) and Arizona (Cochise County), but presumably the
range of this species will include western Texas and the north-
ern Mexican States of Chihuahua and Sonora. Rozen (1977)
found M. sidaefloris to be a cleptoparasite in the nests of
Exomalopsis sidae Cockerell.

Melanomada species A

Figure 25

Two specimens from Kingsville, Kleberg County, Texas,
a female collected 30 May 1977 and a male collected 6 May
1976, both by J.E. Gillaspy (UTSU), do not agree with any
of the above species and probably represent an additional
species in this genus. We are reluctant to name this form
from such inadequate material. In our key above, these will

run to M. sidaefloris, which they somewhat resemble. The
mesosternum is unlike that of M. sidaefloris in that it is thinly
pubescent, the smooth shiny integument fully visible and
bearing minute, irregularly scattered punctures. In M. sidae-
floris the highly plumose hairs are dense enough to conceal
the surface, which is densely and finely punctate. The hy-
postomal area is also minutely and sparsely punctate, unlike
that of M. sidaefloris in which punctures are dense and small.

Melanomada species B

Figure 25

This is known from a single male collected by P.D. Hurd
at Villa Guadalupe, Jalisco, Mexico, 26 July 1951 (UCB). It
closely resembles the male of M. sidaefloris, but the meso-
scutal punctures are more uniformly spaced. Its most out-
standing feature is the presence of extensive, median, pol-
ished and apubescent areas on metasomal sterna 3-5. This
characteristic will distinguish this from all other Melano-
mada we have seen. As with the preceding, we refrain from
naming the species at this time because of the paucity of
available material.

IMMATURE STAGES

The mature larva of only Melanomada sidaefloris was pre-
viously described and compared with other nomadine larvae
(Rozen, 1977; Rozen et al., 1 978). The larva of M. annectens,
treated below, confirms the similarity of this species to M.
sidaefloris and to members of the Melanomada complex.
The larvae of this complex are quite dissimilar to those of
Nomada and Hypochrotaenia. Although pupae of the Mela-
nomada complex are poorly known, those of M. sidaefloris
(Rozen, 1977) and Paranomada velutina are quite similar to
one another and different from those of Nomada.

Mature larva of Melanomada annectens
Snelling and Rozen
Figures 19-24

DIAGNOSIS. The mature larva of this species is very
similar to that of M. sidaefloris and can be distinguished
from it only with some difficulty. The slightly darker head
capsule, somewhat less produced vertex, and somewhat more
pronounced dorsal body tubercles of M. annectens seem to
be reliable differences.

DESCRIPTION. Head (Figs. 20, 2 1 ). As described for M.
sidaefloris (Rozen, 1977) except for following: Integument
more darkly pigmented than in M. sidaefloris so that scler-
otized part of head capsule appears darker than rest of body.
Vertex as seen in lateral view only slightly produced on each
side above antenna, less so than in M. sidaefloris; produced
area somewhat wrinkled. Antennal papilla with four sensilla;
antennal protuberances less developed than in M. sidaefloris.
Mandible (Figs. 22-24) with dorsal apical edge finely and
evenly serrate; ventral apical edge perhaps somewhat more
coarsely serrate, but teeth shorter and more rounded than
those on dorsal edge.

Contributions in Science, Number 384

Snelling and Rozen: New World Genus Nomada 9

Figure 25. Distribution of species of Melanomada: M. annectens (★); M. chica (0); M. grindeliae (•); M. melanantha (O); M. nimia (-&);
M. sidaefloris (©); M. species A ( + ); M. species B (©).

Body. Integument spiculate in some areas. Paired dorsal
tubercles (Fig. 19) slightly more pronounced than those of
M. sidaefloris; first four pairs of tubercles somewhat trans-
verse rather than conical (this was particularly evident on

one specimen that had not yet entered diapause). Other fea-
tures of body as described for M. sidaefloris.

Material studied. Four postdefecating larvae, 4 predefe-
cating larvae, 12 mi. S of Corona, Riverside Co., California,

Figures 19-24. Melanomada annectens, postdefecating larva. 19, body, side view; 20, head, side view; 21, same, frontal view; 22-24, right
mandible, dorsal, adoral, and ventral views. Scale line (1 mm) refers to Figure 19 only.

Contributions in Science, Number 384

Snelling and Rozen: New World Genus Nomad a 11

25 May 1985, J.G. Rozen, Jr. and R.R. Snelling. From nests
of Exomalopsis nitens Cockerell.

HOST ASSOCIATIONS AND BIONOMICS

All known North American cleptoparasites of the Melano-
mada complex are nest associates of members of the genus
Exomalopsis (Rozen, 1984). In Melanomada, M. sidaefloris
has been associated with Exomalopsis consobrina Timber-
lake (Rozen, 1977, 1984) by collections made 26 mi. south
of Animas, New Mexico in 1976, and to Exomalopsis sidae
Cockerell at 5 mi. north of Willcox, Arizona in 1983 (Rozen,
1 984). Although M. annectens (as Hesperonomada mela-
nantha) was associated with Exomalopsis nitens (Rozen, 1 984,
table 2), the cleptoparasite had been misidentified and is in
reality M. annectens (Rozen and Snelling, 1986). Other pos-
sible cleptoparasite-host associations are mentioned in the
section on Systematics.

The bionomics of M. sidaefloris were presented by Rozen
(1977, 1984) and of M. annectens by Rozen and Snelling
(1986).

The South American genera Kelita and Brachynomada are
nest associates of Panurginae (Andrenidae), a further dis-
tinction from the Melanomada complex. Brachynomada near
argentina Holmberg was associated with Psaenythia annu-
lata (Gerstaecker) (Rozen, 1977). Kelita chilensis (Friese) is
a known parasite of Lipanthus parvulus (Friese) (Rozen, 1970),
and Kelita tuberculata Ehrenfeld and Rozen is possibly as-
sociated with Lipanthus alicahue Ruz and Toro (Ehrenfeld
and Rozen, 1977; Ruz and Toro, 1983).

ACKNOWLEDGMENTS

For the opportunity to study specimens from the collections
in their care, we are indebted to: J.A. Chemsak (UCB), S.I.
Frommer (UCR), S.S. Shanks and R.J. McGinley (USNM),
C.D. Michener (UKAN), and W.J. Pulawski (CAS). We are
particularly indebted to S.S. Shanks and W.J. Pulawski for
making available important type material. The type of M.
grindeliae was made available by D. Azuma (ANSP).

LITERATURE CITED

Cockerell, T.D.A. 1898. Tables for the determination of
New Mexico bees. New Mexico University, Bulletin, 1 :
41-71.

. 1 903a. New American Hymenoptera, mostly of the

genus Nomada. Annals and Magazine of Natural His-
tory, (7)12:200-214.

. 1903b. North American bees of the genus Nomada.

Academy of Natural Sciences of Philadelphia, Proceed-
ings, 55:580-614.

. 1911a. Bees in the collection of the United States

National Museum 1 . United States National Museum,
Proceedings, 39:635-658.

— . 1911b. Names applied to bees of the genus Nomada

found in North America. United States National Mu-
seum, Proceedings, 41:225-243.

Ehrenfeld, J., and J.G. Rozen, Jr. 1977. The cuckoo bee
genus Kelita, its systematics, biology, and larvae. Amer-
ican Museum Novitates, no. 2631, 24 pp.

Holmberg, E.L. 1886. Sobre apidos Nomadas de la Re-
publica Argentina. Sociedad Cientifica Argentina, An-
ales, 22:231-240, 272-286.

Linsley, E.G. 1939. New genera and species of epeoline and
nomadine bees (Hymenoptera, Nomadidae). Pan-Pa-
cific Entomologist, 15:1-11.

Linsley, E.G. , and C.D. Michener. 1937. A revision of the
North American Nomadidae (Hymenoptera). American
Entomological Society, Transactions, 65:265-305.

Michener, C.D. 1944. Comparative external morphology,
phylogeny, and a classification of the bees (Hymenop-
tera). American Museum of Natural History, Bulletin,
82:1-326.

Rodeck, H.G. 1945. Genus Melanomada Cockerell, new
designation (Hym.: Apoidea). Entomological News, 56:
202-203.

. 1951. Tribe Nomadini, in C.F.W. Muesebeck, K.V.

Krombein, and H.K. Townes, Hymenoptera of America
north of Mexico. Synoptic Catalog. United States De-
partment of Agriculture, Monograph No. 2, Washington,
D.C., pp. 1189-1207.

Rozen, J.G. , Jr. 1970. Biological observations on the para-
sitic bee Kelita (Hymenoptera: Apoidea). New York En-
tomological Society, Journal, 78:146-147.

. 1977. Immature stages of and ethological obser-
vations on the cleptoparasitic bee tribe Nomadini (Apoi-
dea, Anthophoridae). American Museum Novitates, no.
2638, 16 pp.

. 1984. Comparative nesting biology of the bee tribe

Exomalopsini (Apoidea: Anthophoridae). American
Museum Novitates, no. 2798, pp. 1-37.

Rozen, J.G., Jr., K.R. Eickwort, and G.C. Eickwort. 1978.
The bionomics and immature stages of the cleptopar-
asitic bee genus Protepeolus (Anthophoridae, Nomadi-
nae). American Museum Novitates, no. 2640, 24 pp.

Rozen, J.G. , Jr., and R.R. Snelling. 1986. Ethology of the
bee Exomalopsis nitens and its cleptoparasite (Hyme-
noptera: Anthophoridae). New York Entomological So-
ciety, Journal, 94:480-488.

Ruz, L., and H. Toro. 1983. Revision of the bee genus
Liphanthus (Hymenoptera: Andrenidae). University of
Kansas Science Bulletin, 52:235-299.

Sandhouse, G.A. 1943. The type species of the genera and
subgenera of bees. United States National Museum, Pro-
ceedings, 92:51 9-6 1 9 .

Snelling, R.R. 1986. Contributions toward a revision of the
New World nomadine bees. A partitioning of the genus
Nomada (Hymenoptera: Anthophoridae). Natural His-
tory Museum of Los Angeles County, Contributions in
Science, no. 376, 32 pp.

Swenk, M.H. 1913. Studies of North American bees. I.
Family Nomadidae. University of Nebraska Studies, 12:
1-113.5.

Submitted 17 June 1986; accepted 22 January 1987.

12 Contributions in Science, Number 384

Snelling and Rozen: New World Genus Nomada

;

SERIAL .PUBLICATIONS OF THE
NATURAL HISTORY IVOJSEpM OF LOS ANGELES COUNTY

The scfenti Ik publications, of the Natofajl History Museum of Los Angeles County have been
issued hit irregular intervals in three major series:1 the issues in each series are numbered
indiyi^ dally:, add numbers run consecutively, regardless of the' subject matter.

• 'p. , '

■ W Contribution's in Science, a miscellaneous series of technical pajpers .describing orig

ittal research ih tiM life and. earth science, s

' Science

I'etin. a rrii see Han ecus series of monographs describing original research

■ in. the life jaiid dairtft sciences.' This series was discontinued in 1978 with the issue of

■ Nu'rr.beri29.an;d 30; monographs are now publ ished by the Museum in Contributions

in Science. d - i

generally written for the layman.

Copifs of the publications in these series are sold through the Museum Book Shop. A catalog

is available on request.

Tin :: hf usclum also publishes Technical Reports, a miscellaneous series containing information
relative, to scholarly inquiry and collections but not reporting the results of original research.
ISssiiuyis authorized by tfe museum’s Scientific Publications Committee; however, manuscripts
do hot receive anonymous peer review. Individual Technical Reports maybe obtained from

the relevant SjeildltMf of the. museum.

KCTENIl 1E.IC PUBLICATIONS COMMITTEE

Craig C. Black. Museum Director
Richard C. Brusca

Daniel M. Cohen. Committee Chairman
John M. Harris
Charles L. Hogue
George L. Kennedy
Robin a\. Simpson, Managing Editor
Errol Stevens

'< flilfiU M-'- f Yjdfjjll

Mmm

NOTES ON THE BYTHITID FISH GENUS SACCOGASTER WITH A
NEW SPECIES FROM THE GULF OF MEXICO

Daniel M. Cohen

NEW BYTHITID FISH, DINEMATICHTHYS MINYOMMA,
FROM THE CARIBBEAN SEA

Allegra N. Sedor and Daniel M. Cohen

Contributions in Science, Number 385
Natural History Museum of Los Angeles County
18 June 1987

ISSN 0459-8113

Natural History Museum of Los Angeles County
900 Exposition Boulevard
Los Angeles, California 90007

CONTENTS

NOTES ON THE BYTHITID FISH GENUS SACCOGASTER WITH A NEW
SPECIES FROM THE GULF MEXICO

Daniel M. Cohen 1

NEW BYTHITID FISH, DINEMATICHTHYS MINYOMMA, FROM THE CA-
RIBBEAN SEA

Allegra N. Sedor and Daniel M. Cohen 5

NOTES ON THE BYTHITID FISH GENUS SACCOGASTER WITH A
NEW SPECIES FROM THE GULF OF MEXICO

Daniel M. Cohen1

ABSTRACT. A new bythitid fish species, Saccogaster rhamphido-
gnatha, is named from the Gulf of Mexico. Diagnostic characters
are a hook-like projection on the maxillary, seven branchiostegal
rays, and 12 pectoral fin rays. A revised diagnosis for Saccogaster
is presented, with a key to the eight species in the genus. The third
known specimen of S. staigeri is recorded from the northwest Gulf
of Mexico, a 1600-km range extension. Data are given for two Aus-
tralian specimens, from New South Wales and Western Australia,
which are tentatively identified as S. tuberculata.

INTRODUCTION

This paper reports on four specimens of the rarely encoun-
tered bythitid genus Saccogaster , which recently have come
to hand. Two of these small ophidiiform fishes, caught in the
Gulf of Mexico nearly two decades ago, were discovered in
the Collection of Fishes of Texas A&M University and sent
to me for study. Each represents a different species. One of
them is the third known specimen of S. staigeri, and the
other is undescribed and represents the eighth known species
of Saccogaster and the second to be discovered since the
genus was reviewed by Cohen and Nielsen ( 1 972). Two other
specimens, both recently caught, one off Sydney and the other
off the northwest shelf of Australia, were sent to me from
the Australian Museum and are tentatively identified as S.
tuberculata, known also from the South China Sea and Ha-
waii. In this paper I rediagnose the genus based on additional
material, present a new key to the eight species, describe one
new species, and present new information about S. staigeri
and S. tuberculata.

Terminology and methods follow Cohen and Nielsen (1972,
1978). Study material is housed in the Australian Museum,
Sydney (AM), the Natural History Museum of Los Angeles
County (LACM), and Texas A&M University (TCWC).

Saccogaster Alcock

DIAGNOSIS. Ophidiiform fishes of the family Bythitidae,
subfamily Bythitinae as defined by Cohen and Nielsen (1978:
42). Scales absent from head, reduced or absent on body.
Tail section of body not attenuate. Anal fin originating on

Contributions in Science, Number 385, pp. 1-4
Natural History Museum of Los Angeles County, 1987

posterior half of body. Pectoral fin supported by elongate
radials (enclosed in a free peduncle in some species, adnate
in others). Maxillary vertically expanded posteriorly. Oper-
cular spine present. Palatine teeth present. Developed rakers
on first arch 0-6. Ventral fins with 1 ray in each. Pectoral fin
rays 12-23. Branchiostegal rays 7, 8, or 9. Precaudal verte-
brae 1 2-20. Males with a stalked intromittent organ.

KEY TO SPECIES OF SACCOGASTER

la. Scales absent 4

b. Scales present on body 2

2a. Spine on opercle with three points; pectoral fin rays 22;

dorsal fin rays 92 S'. Hawaii

b. Spine on opercle with a single point; pectoral fin rays

16-18; dorsal fin rays 75-88 3

3a. Gill filaments on first arch not notably reduced in size;
palatine tooth row several teeth wide; dorsal fin rays 75-

83 S. maculata

b. Gill filaments on first arch notably reduced in size; pal-
atine teeth in a single row; dorsal fin rays 87-88

S. staigeri

4a. An antrorse, hook-like projection at posteroventral angle
of maxillary; pectoral fin rays 12; branchiostegal rays 7

S. rhamphidognatha new species

b. No hook-like projection at posteroventral angle of max-
illary; pectoral fin rays 14-23; branchiostegal rays 8 or

9 5

5a. Pectoral fin borne on a free, fleshy peduncle; gill opening

extending above level of top pectoral ray 6

b. Pectoral fin without a free, fleshy peduncle; gill opening
restricted, not extending above level of top pectoral ray

7

6a. Pectoral fin rays 20-23; anal fin rays 56-61; developed

rakers on first gill arch 3 S. tuberculata

b. Pectoral fin rays 14; anal fin rays 64; developed rakers
on first arch 6 S. parva

i. Division of Life Sciences, Natural History Museum of Los
Angeles County, 900 Exposition Blvd., Los Angeles, California 90007.

ISSN 0459-8113

Figure 1. Saccogaster rhamphidognatha new species. Holotype, LACM 44189-1. Standard length 59.7 mm. Drawn by Sharon Belkin.

7a. Posterior end of maxillary sheathed dorsally; anal fin

rays 53-54; nasal chambers not sooty black

5. normae

b. Posterior end of maxillary not sheathed; anal fin rays
37; nasal chambers sooty black .... A. melanomycter

Saccogaster rhamphidognatha new species

Figure 1

HOLOTYPE. The only known specimen, LACM 44189-1
(formerly TCWC 3915.4), <3, Gulf of Mexico, off Mississippi
delta, 29°1 3.5'N, 88°16.5'W, 210 m, RV ALAMINOS, stat.
69A13-43, Oct. 4-16, 1969.

DIAGNOSIS. An antrorse, hook-like projection at the
posteroventral angle of the maxillary; branchiostegal rays 7;
pectoral fin rays 12.

DESCRIPTION. Counts and measurements are presented
in Table 1 . Body slender, greatest depth about 8 in SL (stan-
dard length). Preanal about 1.8 in SL. Head compressed,
about 4.7 in SL. Skin translucent to transparent. Scales ab-
sent.

Lateral line obsolete on body. Head pores apparently re-
stricted to 1 in supraorbital series over upper lip medial to
anterior nostril, 2 in the infraorbital series on the upper lip,
and 2 in the mandibular series near the tip of the lower jaw.

Snout not notably fleshy or inflated. Posterior nostril large,
about equal in diameter to clear window over eye; anterior
nostril a small, simple pore above upper lip and lateral to
supraorbital pore. Lower jaw slightly included. A short, strong,
compressed non-emergent spine on the opercle. Upper rear
part of maxillary free, not sheathed; a prominent antrorse
hook present at the posteroventral angle.

Gill opening extending dorsally to above the level of the
relatively narrow pectoral fin base, about even with the dorsal
margin of the adnate skin-covered pectoral peduncle.

First gill arch with 2-3 flat spiny plates on the upper arm,
a short spiny protuberance at the angle followed by 1 (right
arch) or 2 (left arch) similar protuberances on the lower arch,
followed by 6 spiny plates. Gill filaments on first arch short.
Pseudobranch of 2 filaments present.

Dentition on premaxillary a narrow, exterior band of gran-
ular teeth and an inner band of larger, needle-like teeth.
Vomer with a broadly U-shaped band of small, pointed teeth,
which continues posteriorly onto the palatines. Dentary with
a narrow exterior row of small, slightly compressed teeth and
a single interior row of larger, pointed teeth.

Dorsal fin originating over centrum 7, anal fin under cen-

trum 24. Pectoral fin rays broken off short. Pectoral radials
elongate but pectoral peduncle adnate. The pectoral fin base
is narrow and does not occupy the entire distal margin of
the peduncle. A single, short ray in each ventral fin, the rays
immediately adjacent.

The type is an apparently mature male. The basal part of
the intromittent organ is a fleshy, posteriorly directed pro-
tuberance, longer than wide, with the vent on the antero-
ventral surface; a prominent, 6-mm-long penis protrudes from
the round, distal aperture.

Coloration pale on the head and body; vertical fins brown-
ish distally.

Vertebrae 20 + 38 = 58; neural spines 5-8 variously de-
pressed and truncate. First parapophyses on centrum 7.

RELATIONSHIPS. Saccogaster rhamphidognatha re-
sembles four other Saccogaster species: tuberculata, parva,
normae, and melanomycter (description of the first three in
Cohen and Nielsen, 1972, of the last in Cohen, 1981) in a
derived character, loss of scales, which is reductive and could
have occurred independently in any combination of the five.
The new species shares with two of these scaleless species,
V. normae and melanomycter, two additional derived char-
acters that are not reductive, an adnate pectoral fin peduncle
and a restricted gill opening. Although all three of these species
are well distinguished by autapomorphies, I interpret them
as being more closely related to each other than to any other
Saccogaster species.

ETYMOLOGY. The species name rhamphidognatha is
coined from the Greek rhamphidos, a hook, and gnathos, the
jaw.

Saccogaster staigeri Cohen and Nielsen, 1972

STUDY MATERIAL. TCWC 6146.1, 9, northwest Gulf
of Mexico, 27°44.9'N, 95°20.1'W, 338-384 m, RV ALA-
MINOS, stat. 68A13-19, 2-m dredge, Nov. 19, 1968.

This specimen is the third known, and the first taken since
the species was described. Counts and measurements are
presented in Table 1 and agree well with the type material.

Body scales are sparse and non-imbricate and thus resem-
ble the pattern found in the female paratype rather than in
the male holotype. The lateral canal has 1 pore, above the
upper angle of the gill opening; supraorbital pore 1, above
the upper lip and medial to the tubular anterior nostril; in-
fraorbital pores 2, along the upper lip posterior to the anterior
nostril. There are 7, larger teeth at the tip of the premaxillary,
another character in which this specimen is more similar to

2 Contributions in Science, Number 385

Cohen: Gulf of Mexico Saccogaster

Table 1. Measurements and counts on Saccogaster rhamphidognatha, S. staigeri , and S. tuberculata.

S. rhamphidognatha

5.

staigeri

S. tuberculata

Holotype
LACM 44189-1

3

TCWC 6146.1

9

AM 1 26002-005

9

AM I 22814-029

9

mm

% SL

mm

% SL

mm

% SL

mm*

% SL*

SL

59.7

65.2

84.5

160

Snout to dorsal fin

14.1

23.6

19.6

30.1

25.6

30.3

48.0

30.0

Snout to anal fin

33.1

55.4

39.4

60.4

45.5

53.8

91.0

56.9

Snout to ventral fin

11.1

18.6

15.6

23.9

16.4

19.4

33.7

21.1

Body depth at vent

6.0

10.1

8.1

12.4

13.7

16.2

25.0

15.6

Head length

12.6

21.1

19.0

29.1

21.8

25.8

40.0

25.0

Snout length

2.8

4.7

4.1

6.3

4.7

5.6

8.0

5.0

Eye diameter

0.9

1.5

1.8

2.8

2.7

3.2

4.4

2.8

Interorbital width

2.4

4.0

3.9

6.0

5.0

5.9

8.8

5.5

Upper jaw length

6.0

10.1

9.3

14.3

10.9

12.9

19.4

12.1

Depth of maxilla

2.4

4.0

3.2

4.9

3.6

4.3

6.4

4.0

Cleithrum to ventral fin

1.3

2.2

2.7

4.1

2.0

2.4

7.0

4.4

Ventral fin length

3.7

6.2

7.5

1 1.5

8.7

10.3

17.6

11.0

Pectoral peduncle length

-

-

4.0

6.1

4.0

4.7

7.3

4.6

Pectoral fin base width

1.5

2.5

3.0

4.6

4.3

5.1

9.5

5.9

Pectoral fin length

-

-

9.4

14.4

1 1.8

14.0

21.1

13.2

Dorsal fin rays

77

88

89

85

Anal fin rays

49

53

56

61

Pectoral fin rays

12

18

22

23

Caudal fin rays

12

13

12

12

Branchiostegal rays

7

8

-

(

)

Vertebrae

20 +

38

= 58

14

+ 38

= 52

12 + 43 = 55

56

* Specimen is bent at midlength; hence, longer measurements and proportions are imprecise.

the female paratype with 9 teeth rather than to the male
holotype with 20. Whether this variation might be related
to sex or size cannot yet be determined, as the holotype is
93.1 mm SL and the paratype 61.2 mm SL, similar to 65.2
mm SL for the present specimen.

DISTRIBUTION. The holotype was caught off the east
coast of Florida, the paratype in the eastern Gulf of Mexico
south of Tortugas Island off the tip of Florida. The present
specimen is an extension of known range of about 1600 km
to the northwestern Gulf of Mexico.

Saccogaster tuberculata (Chan, 1966)

STUDY MATERIAL. AM I 26002-005, 9, New South
Wales, 33°34'S, 152°02'E, 658 m, FRV KAPALA, stat. no.
K 86-01-06, 11 Feb., 1 986. AM 228 14-029, 9, Western Aus-
tralia, 18°29'S, 1 16°36'E, 696-704 m, mud bottom, RV SOE-
LA, 6 April, 1982.

This species has been known previously from four speci-
mens, the holotype from the South China Sea, and three other
examples from a single locality in the Flawaiian Islands (Co-

hen and Nielsen, 1972). The two specimens (measurements
and counts presented in Table 1) here discussed (recorded
from Australia by Paxton et ah, in press) are tentatively
identified as S’, tuberculata. Although they agree with the
species diagnosis and in most respects with the description
given by Cohen and Nielsen (1972), there are some differ-
ences between the two Australian specimens and between
Australian and other specimens. Some of these differences
may be attributed to size as specimen AM I 26002-005 from
New South Wales (NSW) is smaller than any others, and AM
I 22814-029 from Western Australia (WA) is larger.

The skin in NSW is thin with a moderate degree of fine
pigmentation giving the fish a light brownish cast; the skin
over the venter is clear and the viscera are visible as there
is neither peritoneal pigmentation nor musculature around
this area. WA has thick pale skin as described for other
specimens and a pigmented peritoneum that conceals the
viscera.

The lateral line in NSW has only about a dozen papillae;
in WA it is obsolete. There is a pore slightly posterior to the
rear margin of the maxillary in WA; no such pore exists in

Contributions in Science, Number 385

Cohen: Gulf of Mexico Saccogaster 3

either NSW or in previous specimens. WA is apparently
unique in having a prominent pore, about as large as the eye,
on the rear margin of the preopercle.

There are nine branchiostegal rays in WA; an accurate
count is not possible in NSW; there are eight branchiostegals
in other specimens.

ACKNOWLEDGMENTS

I thank the following for the gift or loan of specimens, for
information, or for other assistance: M. McGrouther, J. Pax-
ton, and D. Hoese (AM), M. Retzer (TCWC), J. Seigel
(LACM). The manuscript was read by D. Hoese, R. La-
venberg, J. Nielsen, A.N. Sedor, and C. Swift.

LITERATURE CITED

Chan, W.L. 1966. A new genus and species of deep-sea
brotulid from the South China Sea. Japanese Journal of
Ichthyology 14:4-8.

Cohen, D.M. 1981. Saccogaster melanomycter (Ophidi-
iformes: Bythitidae), a new fish species from the Carib-
bean. Proceedings of the Biological Society of Washing-
ton 94:374-377.

Cohen, D.M., and J.G. Nielsen. 1972. A review of the
viviparous fishes of the genus Saccogaster. Proceedings
of the Biological Society of Washington 85:445-468.

Cohen, D.M., and J.G. Nielsen. 1978. Guide to the iden-
tification of genera of the fish order Ophidiiformes with
a tentative classification of the order. National Oceanic
and Atmospheric Administration Technical Report, Na-
tional Marine Fisheries Service Circular, no. 417, 72 pp.
U.S. Department of Commerce.

Paxton, J.R., D.F. Hoese, G.R. Allen, and J.E. Gates. In
press. Zoological Catalogue of Australia, Vol. 6, Fishes
Part 1. Bureau of Flora and Fauna, Australian Govern-
ment Publishing Service. Canberra.

Submitted 12 November 1986; accepted 13 February 1987.

4 Contributions in Science, Number 385

Cohen: Gulf of Mexico Saccogaster

NEW BYTHITID FISH, DINEMATICHTHYS MINYOMMA,
FROM THE CARIBBEAN SEA

Allegra N. Sedor1-2 and Daniel M. Cohen"

ABSTRACT. Dinematichthys minyomma is described as a new
species, based on a single collection of 51 specimens from a coral
reef in the Bay Islands, Honduras. It is tentatively assigned to the
genus on the basis of a single character, the position of the anterior
nostril high on the snout. Unique diagnostic characters include a
small eye ( 10.2-1 5 in head), few lateral scale rows (79-89), few dorsal
fin rays (73-80), and few anal fin rays (55-62).

INTRODUCTION

The specimens here described were collected some 20 years
ago, and their lack of a name has been recognized for nearly
as long. The chief reason for their anonymity has been un-
certainty about appropriate generic placement. Although this
uncertainty remains, the problem is now better defined, and
we describe this species in order to provide a name for use
in forthcoming studies, and to call attention to the species’s
existence.

METHODS

Counts of vertebrae were taken from radiographs, and ex-
clude the ural centrum. Dorsal and anal fin ray counts were
also taken from radiographs, and include the last unsup-
ported ray. Caudal fin ray counts include principal and pro-
current rays. Measurements are to the nearest 0. 1 millimeter.
Eye measurements were taken on the spectacle. Lateral scale
row counts were taken along a midlateral pale line originating
near the opercular angle. Because the scales are deciduous,
it is not possible to obtain exact counts for all specimens, so
minima and maxima, including counts of scale pockets where
possible, are reported for these individuals. Clearing and
staining methods followed Dingerkus and Uhler (1977). Fig-
ure 2 was drawn with a camera lucida.

The term “sclerified” is used here in its most general sense
(Wake, 1979) to refer to the hardening that occurs in the
tissues of the copulatory apparatus of these fishes (“ossified”
of Cohen and Nielsen, 1978). Staining failed to indicate the
presence of cartilage or bone in any of these tissues except

Contributions in Science, Number 385, 5-10
Natural History Museum of Los Angeles County, 1987

the penis (see below). Because only one specimen was cleared
and stained, the possibility cannot be ruled out that the failure
of these tissues to stain is an artifact of preservation, though
this is not indicated by the quality of the staining of the other
tissues in the specimen. In any case, a more precise char-
acterization of these tissues cannot be made without clearing
and staining of additional specimens or, possibly, histological
studies of fresh specimens.

All other methods follow Cohen and Nielsen (1978).

Abbreviations: USNM: National Museum of Natural His-
tory; LACM: Natural History Museum of Los Angeles Coun-
ty; TABL: U.S. National Marine Fisheries Service Tropical
Atlantic Biological Laboratory, Miami; SL: standard length.

GENERIC PLACEMENT

The tribe Dinematichthyini of the bythitid subfamily Bros-
mophycinae as defined by Cohen and Nielsen ( 1 978) contains
nine nominal genera, including Dinematichthys. The prob-
lematical status of Dinematichthys is discussed in detail by
Cohen and Nielsen ( 1 978), who noted that there are no known
extant specimens of the type species, D. ilucoeteoides Bleeker,
1855, and that none of the numerous literature records for
the species agrees with the original description. Although
several species have at one time or another been assigned to
Dinematichthys , Cohen and Nielsen (1978) restricted the ge-
nus to the type species. They further defined Dinematichthys
on the basis of the following six characters: caudal fin free
from the dorsal and anal fins, male intromittent organ with
two pairs of ossified pseudoclaspers, opercle with a sharp-
pointed spine, seven branchiostegal rays, imbricate body
scales, and an anterior nostril high above the upper lip. Of
these, only the last is diagnostic for the genus; the other

1. Division of Life Sciences, Natural History Museum of Los
Angeles County, 900 Exposition Blvd., Los Angeles, California 90007.

2. Department of Biological Sciences, University of Southern
California, Los Angeles, California 90089-0371.

ISSN 0459-8113

Figure 1. Dinematichthys minyomma, paratype, USNM 280123, 71 mm SL, male, lateral view.

characters may be shared by the nominal dinematichthyine
genera Ogi/bia, Brotulina, and Monothrix. It is on the basis
of this one diagnostic character that the species here described
is provisionally placed in Dinematichthys.

In addition to the type species, knowledge of which is based
solely on the original description, the genus presently con-
tains one other species, D. dasyrhynchus Cohen and Hutch-
ins, 1982. Because a single diagnostic character does not
provide convincing evidence of monophyly among these three
species, we stress that the placement of D. minyomma in
Dinematichthys is provisional.

Dinematichthys minyomma new species

Figures 1-3

DIAGNOSIS. A Dinematichthys that dilfers from D. ilu-
coeteoides and D. dasyrhynchus in having a smaller eye (10.2-

15 vs. 5 and 6. 9-9. 4 in head), fewer lateral scale rows (79-
89 vs. 100 and 140), fewer dorsal fin rays (73-80 vs. 83 and
96-103), and fewer anal fin rays (55-62 vs. 69 and 62-69).
It further differs from D. iluocoeteoides in having an un-
sheathed maxillary, and from D. dasyrhynchus in having
fewer vertebrae (10 + 28-30 vs. 13-14 + 33-34), fewer
cephalic pores (19 vs. 22-26), fewer pectoral fin rays (22-24
vs. 25-28), and fewer caudal fin rays (16 vs. 17-18). These
differences are summarized in Table 1, and data for the new
species are shown in Tables 2 and 3.

STUDY MATERIAL. Holotype. USNM 280122, for-
merly TABL 67-124, 66.1 mm SL, male, collected in the
Caribbean Sea, Honduras, Bay Islands, Guanaja (Bonaca)
Island, at 1 6°26'38"N, 85°53'02"W, from a coral reef at 0-9
m, 21 April 1967, by George Miller.

Paratypes. USNM 280123, 38 specimens, 35.5-74.5 mm
SL: 12 males, 47.1-74.5 mm SL, 26 females 35.5-70.7 mm

Table 1. Comparison of the three species of Dinematichthys.'1

Character

D. minyomma

D. iluocoeteoides

D. dasyrhynchus

Head length/eye diameter

10.2-15.0

A little over 5

6. 9-9. 4

Lateral scale rows

79-89

About 100

About 140

Dorsal fin rays

73-80

83

96-103

Pectoral fin rays

22-24

22-23

25-28

Posterior end of maxillary

Unsheathed

Sheathed

Unsheathed

Vertebrae

10 + 28-30

—

13-14 + 33-34

Anal fin rays

55-62

69

62-69

Caudal fin rays

14 + 2

14

17-18

Lateral pores

1

-

2

Supratemporal pores

0

-

1 or 2

Supraorbital pores

3

-

4

Infraorbital pores

5

—

6, 7 or 8

1 . Data for D. iluocoeteoides and D. dasyrhynchus are taken from the original descriptions; data for D. minyomma are summarized for all
51 type specimens.

2. Measurements in mm.

6 Contributions in Science, Number 385

Sedor and Cohen: Caribbean Sea Dinematichthys

Table 2. Data for selected specimens, and summary data for 18 male and 32 female specimens of Dinematichthys minyomma.'

Males,

summary

USNM

280122

USNM

280123

USNM

280123

USNM

280123

Females,

summary

USNM

280123

USNM

280123

LACM

44201-1

LACM

44201-1

Vert.

10 + 29-30

10 + 29

10 + 30

10 + 29

10 + 29

10 + 28-30

10 + 29

10 + 29

10 + 29

10 + 29

D. rays

74-79

75

79

76

76

73-80

77

73

75

73

A. rays

56-61

58

59

59

57

55-62

58

55

57

56

P. rays

22-23

23

23

22

23

22-24

23

23

23

23

L. scale rows

81-86

84

81-85

84-86

79-89

79-82

87-89

G. rakers total

2-4 + 12-16

3 + 16

3+14

3+13

2+12

2-4 + 10-16

4+14

2+16

2+13

3+14

Dev. G. rakers

0-3

1

2

1

3

0-3

2

0

2

3

SL

47.1-74.5

66.1

61.5

72.3

54.1

35.5-70.7

57.7

68.8

61.0

52.1

Depth

14.4

13.6

15.7

10.8

12.9

15.5

13.5

11.2

Pre-dorsal

21.6

21.2

24.2

16.4

18.7

23.6

19.8

18.0

Pre-anal

32.8

31.3

36.3

26.4

28.0

34.4

28.6

26.8

Pect. length

9

8.7

9.5

7.4

7.9

9.7

8.0

7.5

Vent, length

18.5

15.2

17.1

14.0

14.8

17.3

14.2

Head length

19.3

18.0

21.6

15.0

16.3

20.3

17.4

15.9

Eye diameter

1.5

1.4

1.5

1.4

1.4

1.6

1.7

1.2

Snout length

4.3

4.3

5.2

3.5

3.9

5.0

4.1

3.9

U. jaw length

10.1

10.0

10.7

7.6

8.3

11.0

9.0

7.5

G. max. width

3.6

3.8

3.4

2.7

3.0

3.6

3.2

2.5

1. Measurements in mm.

SL; collected with the holotype. LACM 44201-1, 12 speci-
mens, 52.1-70.8 mm SL: 6 males, 53.3-70.8 mm SL, 6 fe-
males, 52.1-70.5 mm SL; collected with the holotype. One
LACM male specimen, 63.3 mm SL, was cleared and double-
stained.

Description. Body compressed, elongate, deepest at or near
origin of dorsal fin, depth at vent averaging 4.8 in standard
length. Dorsal profile sloping gradually downward to rounded
snout. Snout and lower jaw fleshy, with dermal folds and
cirri. Upper jaw slightly protruding, maxillary expanded and
unsheathed posteriorly. Opercle with stout spine pointing
caudad, penetrating the skin slightly anterior to the opercular
margin, and just ventral to the upper opercular angle. Eye
small, sunken, covered at the surface by a round, translucent
spectacle that goes 10.2-15 in head. Posterior nostril, adja-
cent to eye, with a raised rim longest at the medial and ventral
margins. Anterior nostril tubular, 1.5 to 2 eye diameters
anterior to posterior nostril, same distance above upper lip.

Fleshy snout with a dermal fold inserted at the symphysis
of upper jaw about one eye diameter above upper lip, ex-
tending ventrally on each side of symphysis to overhang the
premaxillary. The dermal flap formed by this fold extends
posteriorly along the premaxillary to a point beneath the
posterior margin of the anterior nostril. A second dermal fold
extends ventrally from insertion at the lateral margin of the
anterior nostril; this fold is somewhat produced anteriorly
into an elongate lobe hanging down well below the upper lip
and slightly below the horizontal extension of the first dermal
fold. The second fold extends posteriorly along the upper
jaw, to a point ventral to the eye. The edges of these skin
folds may appear fringed due to the presence of dermal cirri.

On the lower jaw, a circular fold lies on the inferior surface
of the dentary on either side of the symphysis. Just behind
this fold a second semicircular fold extends around the an-
terior tip of the isthmus. As on the upper jaw, these folds
are edged with cirri that also extend along the anterior half
of the inferior surface of the dentary.

Table 3. Frequency distributions for fin ray and vertebral counts of
Dinematichthys minyomma.

Dorsal fin rays

73

74

75

76

77

78

79

80

X

Males

—

2

6

5

2

1

2

—

76.0

Females

5

5

6

6

2

3

-

1

75.3

Total

5

7

12

1 1

4

4

2

1

75.6

Anal fin

rays

55

56

57

58

59

60

61

62

X

Males

—

1

4

6

6

—

1

—

58.2

Females

1

4

7

5

6

4

-

1

58.0

Total

1

5

1 1

1 1

12

4

1

1

58.1

Pectoral fin

rays

Caudal vertebrae

22

23

24

28

29

30

Males

4

14

—

Males

—

15

1

Females

8

22

1

Females

2

22

4

Total

12

36

1

Total

2

37

5

Contributions in Science, Number 385

Sedor and Cohen: Caribbean Sea Dinematichthys 7

Figure 2. Arrangement of cephalic sensory pores in Dinematich-
thys minyomma, 68.7 mm SL female paratype, USNM 280123.
Pores located beneath dermal folds are rendered in broken outlines.

The lateral canal has one pore dorsal to the gill opening
(Figure 2). No supratemporal pores. Three supraorbital pores:
one large pore (with a diameter about equal to that of the
nostril) hidden beneath the first dermal fold that overhangs
the upper lip, one small pore medial to the anterior nostril,
one posterodorsal to the eye on the frontal region. Five in-
fraorbital pores: one lies lateral to the anterior nostril near
the insertion of the second dermal fold on the snout, a second
pore (with a diameter about two times that of nostril) lies
beneath the flap of skin overhanging the upper lip. The outer
edge of this pore is vertically bisected by a septum. Three
smaller infraorbital pores are evenly spaced above the pos-
terior half of the upper jaw. Ten preoperculomandibular pores:
one large pore in the circular dermal fold on the inferior
surface of the dentary, one large pore opening supralaterally
within the dermal fold at the tip of the isthmus, one small
pore on the anterior outer mandibular margin just beneath
the dentary, three small pores spaced along the posterior half
of the mandible, two large pores beneath a flap of skin at the
angle of the preopercle, one at the preopercular margin just
above the angle, one at the juncture with the lateral canal.

Numerous papillae are present on the head, some stout
and fleshy, some smaller and less conspicuous; these latter
are often obscured by a mucoid covering on some regions of
the head. Beginning at the symphysis of the upper jaw: one
large papilla at the origin of the first dermal fold on the snout;
a second smaller papilla just above the point where this fold
meets the upper lip. Beneath the lateral extension of this fold,
separated by the third supraorbital pore, are two stalked
papillae in a medial position, and three in a lateral position
at the posterior end of the fold. Three small papillae sur-
rounding the anterior nostril at the medial, dorsal, and lateral
margins. Two small papillae bordering the eye, one on the
medial, one on the ventral margin. Two more lie near the
medial margin of the posterior nostril, a third medial to these,
above the anterior nostril. One small papilla in the infraor-
bital region, about halfway between the posterior nostril and
the upper lip. One small papilla posterodorsal to the eye,
three in the interorbital region on each side of the midline.

One on frontal region of the head, just posterior to the first
supraorbital pore. One at the dorsal border of the cheek scale
patch, in line with the first preopercular pore, five traversing
the top of the head from the supraorbital region to the su-
pratemporal region. Three small papillae on opercle: one
ventral to the last (tenth) preoperculomandibular pore at the
border of the preoperculum, one posterior to the ninth pre-
operculomandibular pore, one anteroventral to the opercular
spine. Two conspicuous papillae near the upper angle of the
maxillary at the lower anterior border of the patch of cheek
scales: one at the maxillary angle, one along the upper pos-
terior edge of the maxillary. One smaller papilla near the
edge of the preoperculum directly posterior to the maxillary,
and three small papillae slightly ventral, along the edge of
the preoperculum. Seven small papillae distributed irregu-
larly along the mandible. One stout stalked papilla associated
with the first preoperculomandibular pore on the dentary.

In addition to the consistent pattern of papillae described
above, irregular clusters of numerous smaller papillae may
be present on the snout, and near the lower jaw synphysis
anterior to the third preoperculomandibular pore.

Lateral line inconspicuous, in two parts. One series of neu-
romasts originates at the dorsal margin of the head antero-
dorsal to the lateral pore, and extends caudad to a point above
the anal fin origin. The second, midlateral series originates
above the anal fin and extends to the caudal peduncle.

All but the posterior quarter of the premaxillary bears an
irregular series of elongate conical teeth along the inner mar-
gin. These generally decrease posteriorly in size and density.
The outer margin and posterior quarter of the premaxillary
are covered by granular teeth. The dentary also bears an outer
band of granular teeth and an inner band of sharp conical
teeth in an irregular series. On both the premaxillary and the
dentary, these conical teeth are generally longest at the medial
margin and most densely distributed near the symphysis.
The vomer bears a triangular patch of conical teeth, longest
at the posterior border. Palatines also with conical teeth in
an irregular series, longest at the medial margin.

Each gill arch bears a row of flat plates covered with tiny
granular teeth on both the outer and inner surfaces; 0-3
developed rakers on the outer face of the ceratobranchial.
Three suprabranchial patches of granular teeth, located be-
tween the first and fourth arches on each side. One basi-
branchial patch of granular teeth between the second and
third arch on each side.

Dorsal fin originating over the anterior quarter of the pec-
toral fin, with four cartilaginous predorsals; anal fin origi-
nating about equidistant from snout and caudal fin. In both
dorsal and anal fins the posterior rays are longest, all are
branched, and the last is unsupported by radials. The caudal
fin is slightly rounded. The caudal skeleton has two hypural
plates, one parhypural, one epural; each hypural bears one
short procurrent ray, and one unbranched and six branched
principal rays. The rounded pectoral fin is about one-half the
length of the head. Ventral fins are inserted immediately
adjacent to each other, more than one eye diameter behind
the symphysis of the cleithra; rays nearly reach the vent, or
extend beyond it.

8 Contributions in Science, Number 385

Sedor and Cohen: Caribbean Sea Dinematichthys

Small, overlapping cycloid scales arranged in regular rows
cover the body completely, but are lacking on the head except
for a large patch on the cheek.

Ribs articulating with the first five centra; subsequent ab-
dominal vertebrae have ribs articulating with parapophyses.
Ribs on the third and fourth centra with anterior and pos-
terior flanges.

The male copulatory apparatus (Figure 3) consists of a
penis flanked by two pairs of pseudoclaspers, and by an ac-
cessory sclerified structure (the accessory cornified body of
Turner, 1946). The apparatus is partially covered anteriorly
by a thick fleshy genital flap that originates on the posterior
margin of the anus. Embedded in the tissue on the inner
(dorsal) side of this flap is another sclerified structure, which
apparently provides an anchor for dense tissue connections
extending from the pseudoclaspers to the genital flap tissue.
This structure is clearly discernable only in cleared speci-
mens, although its contours can be traced in the flap tissue
of preserved specimens.

The copulatory apparatus is the distal outgrowth of the
urogenital sheath. This sheath originates in the sperm duct
near the fusion of the two central canals of the testes, and
terminates in the hollow, cylindrical penis. The urinary and
sperm ducts pass through the urogenital sheath and through
the penis, terminating in a pore at the tip of the penis. The
penis is cartilaginous, with a ventral ossified portion near the
tip.

Although there appear to be two pseudoclaspers on either
side of the penis, dissection of the copulatory apparatus re-
vealed that the two are joined at their base (Figure 3b), so
they are here referred to as lobes of a single pseudoclasper.
The branched pseudoclasper is anchored in the tissue of the
lateral wall of the urogenital sheath. The external lobe of the
pseudoclasper appears about twice as large as the internal
lobe (Figure 3a). The internal lobe appears similar in shape,
but has a spur at its anterior margin; however, dissection
reveals that the true shape of the internal pseudoclasper lobe
(Figure 3b) is obscured by the dense connective tissue of the
genital flap.

The posterior edges of the external lobes of the pseudo-
claspers are connected distally by a membrane that passes
posteriorly around the base of the penis. Attached to this
membrane by a thin ligament is the accessory sclerified struc-
ture. This large, rounded structure lies either right or left of
the midline, skewing the orientation of the penis by slightly
displacing its base (Figure 3a).

Color of holotype in alcohol: an overall pale brown, darker
on top of head, lighter on sides of head. Bases of dorsal, anal
and pectoral fins darker brown; posterior margins of body
scales also darker.

REMARKS. Presence of a single pair of unossified branched
pseudoclaspers in this species would seem to invalidate the
pseudoclaspers, as used by Cohen and Nielsen (1978), as a
defining character for the genus. However, the branched
pseudoclaspers possessed by all Dinematichthys, Ogiibia,
Monothrix, and Brotulina species are quite distinct from the
unbranched pseudoclaspers present in other dinematichthy-
ine species (Sedor, 1985). Clearly, further work is needed to

ventral view, 7 1 mm SL male paratype, USNM 280 123. b, dissected
right pseudoclasper, lateral view. Abbreviations: ac, accessory body;
fs, flap structure; gf, distal tip of genital flap; pe, penis; pel, pseu-
doclasper, external lobe; pil, pseudoclasper, internal lobe. Bar in-
dicates 0.5 mm.

characterize properly the pseudoclaspers and assess their
taxonomic and phylogenetic significance.

ETYMOLOGY. The species is named minyomma (from
the Greek minys, small and omma, eye) in reference to its
small eye, and is treated as a noun in apposition.

ACKNOWLEDGMENTS

We thank J.G. Nielsen, R.J. Lavenberg, J.A. Seigel, C.C.
Swift, and M.R. Meisler for their comments on the manu-
script.

LITERATURE CITED

Bleeker, P. 1855. Bijdrage tot de kennis der ichthyologische
fauna van de Batoe Eilnanden. Natuurkundig Tijdschrift
voor Nederlandsch- Indie 8:305-328.

Cohen, D.M., and J.B. Hutchins. 1982. Description of a
new Dinematichthys (Ophidiiformes: Bythitidae) from
Rottnest Island, Western Australia. Records of the West-
ern Australia Museum 9(4):34 1-347.

Contributions in Science, Number 385

Sedor and Cohen; Caribbean Sea Dinematichthys 9

Cohen, D.M., and J.G. Nielsen. 1978. Guide to the iden-
tification of genera of the fish order Ophidiiformes with
a tentative classification of the order. National Oceanic
and Atmospheric Administration Technical Report, Na-
tional Marine Fisheries Service Circular no. 417, 72 pp.

Dingerkus, G., and L.D. Uhler. 1977. Enzyme clearing of
alcian blue stained whole small vertebrates for dem-
onstration of cartilage. Stain Technology’ 52(4):229— 232.

Sedor, A.N. 1985. A phylogenetic hypothesis based on the
male copulatory complex in dinematichthyine fishes (or-

der Ophidiiformes, family Bythitidae). M.S. thesis, Uni-
versity of Southern California, Los Angeles, 58 pp.

Turner, C.L. 1946. Male secondary sexual characters of
Dinematichthys iluocoeteoides. Copeia 1 946(2):92— 96.

Wake, M.H., ed. 1979. Hyman’s Comparative Vertebrate
Anatomy. Third Edition. Chicago: University of Chi-
cago Press, 788 pp.

Submitted 12 November 1986; accepted 13 February 1987.

10 Contributions in Science, Number 385

Sedor and Cohen: Caribbean Sea Dinematichthys

Q I

11

L52X I

NH I

A

■A lV- A«gP*

SERI At PUBLICATIONS OF THE .

NATURAL HISTORY MUSEUM OF LOS ANGELAS COUNTY

The scientific publications of the Natural History Museum of Los Angeles County have been
issued'. at irregular intervals in three- major series; the issues in each series are numbered
individually, 'apd numbers 'run consecutively, regardless of the subject matter.

® Contributions in Science, a miscellaneous series -of technical papers describing orig-
inal research in the life and earth sciences.

# Science.' Bulletin, a miscellaneous series of monographs describing original research
in the life and earth sciences. This series was discontinued in 1978 with the issue of
. Numbers 29 and 30: monographs .are now published by the Museum in Contributions

in Science. Y ! ' ■i-’CC: - ';i' v ' Cda

# Science Series, long articles on natural history topics, generally written for the layman.

-Copies of the publications in 'these scries are sold through the Museum Book Shop. A catalog
is. available on request.

T he Museum also publ ishes Technical Reports, a miscellaneous series containing information
relative to scholarly inquiry and collection’s but not reporting the results of original research,
issue is authorized by .the museum's Scientific Publications Committee; however, manuscripts
do houeieeiVe anonymous peer review, individual Technical Reports may be- obtained from

* , I . ;.ij i '! T ’ • i'T', ’^1 f I' 1 1 *' if

:#ie relevant: Section of "the museum. \

bCI ENTJ FilC FU B UCA TflONS COM M 1TTEE

Dmk-Yvri'Mtov E«s.; Inc::, Isaju. * R^jwtfiriK.rti

Craig C. Black. Museum1 Director

Richard C. B-rusca. : .

Daniel M. Cohen, Committee Chairman

John M. Harris

Charles I... Hogue

George L. Kennedy

Robin' A. Simpson, Managing Editor

Errol Stevens : ■ i !: i :: U 1 \ ' pa i Siijlii !

MOLES OF THE SCAPANUS LATIMANUS GROUP
(TALPIDAE, INSECTIVORA) FROM THE PLIOCENE AND
PLEISTOCENE OF CALIFORNIA

J. Howard Hutchison1

ABSTRACT. All the California Pliocene and Pleistocene moles of
the genus Scapanus belong to the subgenus Scapanus. Two informal
species groups of S. (Scapanus) are proposed: the Orarius group with
S. orarius and 5. townsendii and the Latimanus group with S. ma-
latinus new species and S. latimanus. Only fossil members of the
Latimanus group are known from California. Scapanus malatinus
new species from Irvingtonian age deposits (about 1.8 Ma) of San
Diego County, California, is the closest morphological ancestor of
S. latimanus. It exhibits an early stage of enamel excursion below
the gum line (hypsodonty) in the M,_2. Scapanus latimanus is not
known earlier than later Irvingtonian and appears to mark the most
recent acquisition of advanced hypsodonty in the Talpidae.

INTRODUCTION

Moles of the genus Scapanus are presently confined to the
moister areas of North America from British Columbia to
northern Baja California and extend inland nowhere more
than 600 kilometers (375 miles) from the ocean (Hall and
Kelson, 1959; Palmer, 1937) (Fig. 1). Three living species
are commonly recognized: 5. latimanus (Backman), S’, town-
sendii (Backman), and S. orarius True. The fossil record
expands the range of the genus, principally for members of
the subgenus Xeroscapheus Hutchison, well into the Great
Basin. Xeroscapheus represents a group of moles convergent
on Scalopus Geolfroy St. Hilaire in many respects (Hutchi-
son, 1968:93) and is characterized by relatively shorter jaws,
more hypsodont teeth, and more powerful forelimbs. Al-
though Xeroscapheus became extinct by the end of the Hemp-
hillian (latest Miocene), S. latimanus appears, in part, to have
taken over the ecologic role of Xeroscapheus. particularly in
the more southerly part of its range. Scapanus latimanus does
not presently occupy most of the areas vacated by Xerosca-
pheus. undoubtedly because these areas have now progressed
into deserts and semideserts. Since the Clarendonian (late
Miocene), progenitors of the modern species, subgenus Sca-
panus, appear to have remained more or less within the
present range of the genus.

The only extinct previously named species of Scapanus,

Contributions in Science, Number 386, pp. 1-15
Natural History Museum of Los Angeles County, 1987

S. shultzi Tedford (1961) and S. proceridens Hutchison ( 1 968)
belong to the subgenus Xeroscapheus (Hutchison, 1 968). Lit-
tle was known of the prehistory of the subgenus Scapanus
except for scattered records of the extant species, S. latimanus
in the Rancholabrean (late Pleistocene). Pre-Rancholabrean
records of S. (Scapanus) are rare and fragmentary (Hutchi-
son, 1968). All California records of the subgenus appear to
be on the lineage of or closely related to S. latimanus, a
hardly surprising result considering the present wide distri-
bution of this species. Previously undescribed fossils provide
a temporal framework for the acquisition of some diagnostic
features of this lineage.

METHODS AND MATERIALS

Terminology of the teeth is from Hutchison (1974). Mea-
surement definitions follow Hutchison (1968) for the bones
and Palmer (1937) for the skull. All measurements are in
millimeters (mm). Ages are abbreviated as ka (kilo-annum)
and Ma (Mega-annum). Super and subscripted numbers (x
for indeterminate tooth number) with the following abbre-
viations refer to upper and lower teeth respectively: I — in-
cisor, P— premolar, M — molar. Detailed site descriptions are
available from the cited literature or may be obtained by
qualified investigators from the repository institution.
Institutional acronyms for specimens cited in text are:

JHH: author’s field number.

IVCM: Imperial Valley College Museum, El Centro, Cali-
fornia.

LACM: Natural History Museum of Los Angeles County,
Section of Vertebrate Paleontology, Los Angeles, Califor-
nia.

MVZ: University of California Museum of Vertebrate Zo-
ology, Berkeley, California.

1. Museum of Paleontology, University of California, Berkeley,
California 94720.

ISSN 0459-8113

2 Contributions in Science, Number 386

Hutchison: Scapanus from California

SBCM: San Bernardino County Museum, Redlands, Cali-
fornia.

SDNHM: San Diego Natural History Museum, San Diego,
California.

UCMP: University of California Museum of Paleontology,
Berkeley, California.

UCR: University of California, Department of Geological
Sciences, Riverside, California.

UO: University of Oregon Museum of Natural History, Eu-
gene, Oregon.

USNM: United States National Museum, Washington, D.C.

SYSTEMATICS

Class Mammalia Linnaeus, 1758

Order Insectivora Illiger, 1811

Family Talpidae Vicq d’Azyr, 1792

Subfamily Talpinae
Fischer von Waldheim, 1817

Tribe Scalopini Trouessart, 1879

Genus Scapanus Pomel, 1848

Subgenus Scapamis Pomel, 1848

Scapanus malatinus new species

Figures 2, 3a-c, 4a, 6a

Scapanus, Downs and White, 1968:44, fig. 2 (a faunal list).

DIAGNOSIS. Small size (humerus total length = 12.2-
13.3 mm); lower molars with longitudinally compressed tri-
gonids and anterior cingulid and entocingulid spanning the
base of the paraconid; M, with enamel extending down labial
side of root below gum line, at least below the trigonid; M,_3
with sharp and well-developed ectocingulids spanning hy-
poflexids; M, does not show enamel excursion down the root;
upper molars with metacingulum joining lingual edge of
metacone; trough between base of infraorbital bridge and
ridge for insertion of masseter muscles narrow and shallow;
infraorbital bridge strap-like rather than rod-like; frontal re-
gion convexly arched; shaft length of the radius about 60%
of humerus length.

HOLOTYPE. LACM 3451, left M,_3, M, broken into two
parts and heavily damaged. The teeth were found as isolated
fragments in a small sieved sample from a 2-3 square foot
area and originally were catalogued together. Because some

Figure 2. Scapanus malatinus, IVCM 15/61, partial skull with
right M2"3. Hatched areas represent the major areas where the bone
was spalled otf exposing a natural endocast. Cross-hatched areas
represent matrix or broken surfaces. Scale line equals 1 mm.

of the fragments fit together and all the teeth are in the same
stage of wear and of the same side, there is little doubt that
they are from the same individual. Collected by Harley Gar-
bani, 18 January 1954.

TYPE LOCALITY. LACM locality 1114 (Rodent Hill).
Latitude 32025'21''N; longitude 1 16°12'7"W. Vallecito bad-
lands of the Anza-Borrego Desert, San Diego County, Cali-
fornia. Palm Springs Formation, Huesos “Member,” T.
Downs “zone” 57.6. From about a foot thick whitish-gray-
greenish, semi-consolidated, micaceous, unstratified silt or
mudstone (T. Downs’ notes of 13 December 1954).

AGE. Irvingtonian Land Mammal Age (Downs and White,
1968), Irvingtonian I of Repenning (1980, 1985), Matuyama
Epoch of the paleomagnetic scale (Opdyke et ah, 1977), with-
in the Olduvai event (Downs, pers. comm. 1 983), about 1.7-
1.9 Ma (Repenning, 1985).

ETYMOLOGY. Malatinus is an anagram of latimanus.

HYPODIGM. LACM locality 1114 (=IVCM loc. 15):
Type; IVCM 15/61, associated individual with skull lacking

Figure 1. Distribution map of the Latimanus group species of Scapanus (Scapanus) in California (modified from Palmer, 1937, and Hall
and Kelson, 1959). Fossil sites are: 1, Samwell Cave (UCMP 1008); 2, Galen’s Pit (UCMP V75068); 3, Potter Creek Cave (UCMP 1055); 4,
Hawver Cave (UCMP 1069); 5, Teichert (UCMP V69129); 6, Mercer’s Cave (UCMP V67108); 7, Rodeo area (UCMP V6312, V71001-
V7 1003); 8, Pacheco 2 (UCMP V78027); 9, Maxum (UCMP V6869); 10, Irvington (UCMP V3604) and Prune Avenue (UCMP V5301); 11,
Tranquility (UCMP V4401); 12, Calico Lakes (SBCM 1.76.35); 13, Solid Waste site (SBCM 1.76.33); 14, Zuma Creek (LACM 1754); 15,
Rancho La Brea (LACM 6909 and Pit 91); 16, Emery Borrow Pit (LACM 7053); 17, San Pedro (UCMP 2047); 18, Newport Mesa (LACM
3877); 19, Redec A (USGS M1451); 20, San Diego area (SDNHM 3181 -E, 3131 -B); 21, Anza-Borrego Desert (LACM and IVCM sites). Small
map of the distribution of the extant species of Scapanus north of California from Hall and Kelson, 1959.

Contributions in Science, Number 386

Hutchison: Scapanus from California 3

Figure 3. a-c. Teeth of Scapanus malatinus, LACM 345 1 (Ho-
lotype); a, occlusal views of left M (trigonid) to M,; b, M, trigonid,
anterior, labial, and lingual views; c, M3, labial and lingual views,
d-j, M, trigonids of Scapanus (Scapanus), anterior views; d, 5. la-
timanus cf. caurinus, UCMP 1 3 1 165; e, S. latimanus, UCMP 8 1 729,
Irvington site; f, S. 1. occultus, MVZ 4036; g, S. /. anthonyi, MVZ
6216; h, 5. /. cf. anthonyi, SDNHM 26232, Glen Abbey site; i, S.
townsendii, JHH “G”; j, 5. orarius, JHH “A.” Scale line equals 1
mm.

anterior and left side of rostrum but with right M2-3, right
humerus, anterior part of manubrium, proximal right femur,
proximal right radius, fragment of right dentary, scapular
blade fragment, a thoracic and caudal vertebra, proximal rib
fragment, medial and ungual phalanx of manus; LACM
122848, M' fragment; LACM 122849, Mv talonid; LACM
3450, 9804, 122850-122851, humeri fragments; LACM
122852, ulna fragment; LACM 122853-122855, radii; LACM
122856, lunar; LACM 122857, magnum; LACM 122858-
122859, metacarpals I; LACM 122860, metacarpal II; LACM
122862-122863, metacarpals III; LACM 122863-122864,
metacarpals IV; LACM 122865, metacarpal V; LACM
122866-127867, 122880, manus sesamoids; LACM 1 22868—
122872, manus proximal phalanges; LACM 122873-122879,
manus medial phalanges; LACM 122881-122887, manus
ungual phalanges; LACM 122888, femur; LACM 122889,
femur head; LACM 122890, proximal tibia; LACM 122891,
astragalus; LACM 122892, cuboid; LACM 122893, navic-

Table 1 . Stratigraphic sequence of selected Scapanus malatinus lo-
calities, from the Anza-Borrego Desert, San Diego County. Localities
between the dashed lines indicate the level of the Olduvai event.

Land

Mammal Age

T. Downs Zone

LACM locality

59.2

1 191

Irvingtonian

58.5

1 142

57.7

1461

57.6

1114 (type)

Blancan

55.5

1615

50.4

1428

ular; LACM 122896, metatarsal II; LACM 122894, meta-
tarsal III; LACM 122895, metatarsal IV; LACM 122900-
122905. pes phalanges; LACM 122897-122899, pes ungual
phalanges.

REFERRED MATERIAL. LACM locality 1191: LACM
3596, distal humerus. LACM locality 1245: LACM 4424,
distal humerus. LACM locality 1428: LACM 1229 10, heavi-
ly worn right M,. LACM locality 1442: LACM 9806, distal
humerus. LACM locality 1461: LACM 122906, magnum;
LACM 122907, metacarpal V; LACM 122908, manus me-
dial phalanx; LACM 122909, manus ungual phalanx. LACM
locality 1615: LACM 4450, distal humerus; LACM 12291 1,
carpal; LACM 122912, manus ungual phalanx. LACM lo-
cality 1768: LACM 4451, distal humerus. IVCM locality
124: IVCM 1060, 1061, incomplete humeri. IVCM locality
333: IVCM 1362, humerus fragment. All the referred spec-
imens are from the Vallecito badlands and same formation
and “member” as the type. They are all in the Matuyama
Epoch but vary in stratigraphic position (see Table 1).

DESCRIPTION. The skull (Fig. 2), although chipped, pre-
serves all the cranium and most of the posterior part of the
rostrum. Because the skull is only slightly distorted, some of
the standard skull measurements used on extant species may
be taken (Table 2). The frontal region is noticeably convex
dorsally as in some specimens of 5. latimanus (see Jackson,
1915, pi. 5, fig. 4). Aside from the dentition, size, and relative
rostral proportions, the skulls of extant species of Scapanus
closely resemble one another. The preserved parts of the
fossil skull share those features common to the three extant
species. LTnfortunately the tip of the rostrum is lacking in
IVCM 1 5/6 1 so its relative length or number and size of the
antemolar teeth cannot be determined. The arching of the
frontal region appears to be related to shortening of the snout
in recent species so some shortening of the snout is inferred
in 5. malatinus. The bridge of bone forming the lateral wall
of the infraorbital foramen is relatively narrow and strap-
like as in S. orarius and unlike the rod-like structure in S.
latimanus. Lateral to the base of the infraorbital bridge is a
longitudinal and slightly dorsally deflected ridge of bone for
insertion of the masseter muscles. A shallow and relatively

4 Contributions in Science, Number 386

Hutchison: Scapanus from California

Table 2. Measurements of the skull of Scapamts malatinus, IVC 15/61, compared to selected S. latimanus samples.

Measurement

IVC 15/16

anthonyi

occultus

all latimanus

Mastoidal breadth

15.9 mm

15.3-16.0

15.3-16.3

15.3-18.6

Interorbital breadth

7.3

6. 9-7. 4

6. 9-7. 7

6.9-8. 6

Maxillary breadth

9.1*

8. 9-9. 4

8. 9-9. 8

8.9-10.8

Depth of skull

8.5

8. 3-8. 9

9.0-9. 8

8.3-10.8

Length of orbit

7.8

8.2-8. 7

7.6-8. 8

7.6-10.0

* Only the right side preserved and measured to midline then doubled to give standard maxillary breadth.

narrow trough lies between this ridge and the base of the
infraorbital bridge and resembles that in S. latimanus. In S.
orarius and S. townsendii this trough is wider and the ridge
is better developed dorsally.

The upper molars strongly resemble those of S. townsendii
and S. orarius. although the pre-ectoflexus of the M3 forms
a more open angle. The metacingulum on all the upper mo-
lars joins the lingual-most part of the metacone whereas in
S. latimanus it usually joins the metacone anterior and labial
io this point.

Of the five lower molar specimens (Figs. 3, 4), only the
M , is complete. The trigonid of the left M, of the type lacks
the tip of the paraconid. There is a small but distinct lingual
parastylid at the base of the anterior part of the paraconid
from which a short entocingulid extends posteriorly to the
metaconid. This cingulid is absent in extant species, and,
excepting S. latimanus, the parastylid is smaller or absent in
extant species. The base of the trigonid valley is about the
same level as in the M,. A small hairline-like precingulid,
absent in extant species, spans the base of the paraconid but
falls short of joining the parastylid. The most striking feature
of the tooth is the extent to which the enamel extends down
the labial side of the root (Fig. 3b), a condition approaching
or equalling that of S. latimanus rnthonyi . The arch of the
labial margin of the protoconid is uniform, unlike the asym-
metrical curvature that occurs there in S. orarius.

The M2 (LACM 345 1 ) lacks the parastylid due to breakage
but shows that the trigonid is compressed longitudinally to
an equal or slightly greater extent than in observed specimens
of S. latimanus which in turn has the most compressed tri-
gonids of the extant species (Fig. 4). The floor of the trigonid
basin is at the same level as the crest of the crista obliqua at
its juncture to the metaconid. Although the parastylid along
with the root and basal enamel margin is broken away, there
is a trace of a precingulid extending labially at least two-
thirds the distance across the base of the paracristid. The
trigonid of the M, (LACM 345 1 and 7LACM 1 22849) shows
the same strong development of the entocingulid and enamel
margins as in the M3. The entoconid and crista obliqua are
sharply defined and continuous with the entoconid and hy-
poconid respectively. The entocristid is lower than the crista
obliqua. As in the extant species there is a distinct entostylid
and no ento- or postcingulid. The enamel margins show no
tendency to extend onto the roots with the possible exception
of the margin (not complete) below the protoconid.

The M,, while resembling that in the extant species in the
disposition and size of the major cusps, differs from them in
having 1) a sharp and well-developed ectocingulid spanning
the hypoflexid (variable in S. latimanus but when present
not strongly united to the protocone), 2) the parastylid con-

Figure 4. Comparison of the left M, of four species of Scapanus
( Scapanus ), anterior, labial, posterior, and lingual views; a, S. ma-
latinus, LACM 345 1 (holotype); b. S', latimanus cf. caurinus, UCMP
131 165, dashed line indicates limit of cementum; c, S. townsendii,
JHH “G”; d, S. orarius, JHH "A." Scale lines equal 1 mm.

Contributions in Science, Number 386

Hutchison: Scapanus from California 5

Radius shaft length (mm)

Figure 5. Plots of the radius to humerus proportions and regression
lines in three Recent species of Scapanus. The plots (open symbols)
of S. malatinus are based on the one isolated measurable humerus
with each of the three isolated radii from LACM locality 1114.
Scapanus latimanus plots based upon specimens in the UCMP and
MVZ including all the subspecies mentioned by Palmer (1939) ex-
cepting .S’, insularis , S. sericatus, S. monoensis, and 5. anthonyi.

tinuous with a precingulid which extends along the lingual
two-thirds of the base of the paracristid, 3) entocingulid well
defined along the entire lingual base of the paraconid, and
4) an entocristid which is distinctly lower than the crista
obliqua and is truncated by a notch between it and the ento-
conid. The enamel margins at the base of the M3 crown are
only moderately sinuous as in S. townsendii and S. orarius
and do not extend down the labial sides of the roots as in S’.
latimanus. The trigonid is more compressed longitudinally
than in S. orarius and resembles that of S. latimanus.

The radius resembles S. latimanus and S. townsendii in
proportions and falls within the size range of S. latimanus.
The total length of the only complete humerus (IVCM 15/
16) is plotted in Figure 5 with each of the three radii from
the same site; the resulting proportions all fall within or
adjacent to the plots of small S. latimanus and away from
those of S. orarius. The smallest measurable humerus (IVCM
106 1) is 1 2.2 mm in length, smaller than any living specimen
measured.

One ulna (LACM 1 22853) lacking the proximal and distal
ends resembles that of S', latimanus and S. townsendii in its
relatively short robust shaft and short fusiform brachialis

Figure 6. Right metacarpal II of Scapanus {Scapanus), lateral view;
a, S. malatinus. LACM 122860; b, S. latimanus. UCMP 131 166; c,
S. orarius, JHH "A”; d, S. townsendii, JHH “G.” m = magnum
articular facet, pp = proximal process, pt = palmar tuberosity. Scale
line equals 1 mm.

scar. It differs from the observed extant species in its rela-
tively longer radial facet and more open semilunar notch.

Metacarpal I (LACM 122858) resembles those of the ex-
tant species but is relatively short as in S. orarius and S.
latimanus. Metacarpals II-V resemble those of S. townsendii
and V. latimanus in their relative shortening. Metacarpal II
resembles that of 5. latimanus in the less protuberance of
the palmar tuberosity (Fig. 7) and lack of a deep trough-like
depression which separates the palmar tuberosity from the
trapezoid facet. The proximal prominence also resembles
that of V. latimanus in extending away from the metacarpal
axis at nearly a right angle, not being curved as in S. town-
sendii and 5. orarius (Fig. 6). There is a medial low tuberosity
on the medial part of the palmar side of metacarpal III just
distal to the magnum facet. This tuberosity bears a roughly
circular palmar facing facet as in S. latimanus, unlike the
rounded surface in the other species.

The femur is relatively short but not noticeably different
in morphology from the modern species. The humerus, tibia,
anterior part of the manubrium, palmar sesamoids, and me-
dian and ungual phalanges resemble those of the extant species.

REMARKS. Whereas all of the specimens come from the
Vallecito Creek “Local Fauna” of Downs and White (1968),
two of the localities (LACM 1245, LACM 1768) are of un-
certain stratigraphic provenience. The other sites (Table 1)
are arrayed on either side of the T. Downs “zone” 56, which
probably coincides with the Olduvai magnetic polarity event
(Opdyke et al., 1977; T. Downs in letter, 1 977). The Olduvai
event is dated elsewhere at about 1.72 to 1.88 Ma (N. John-
son, in prep).

Scapanus latimanus (Backman, 1842)

Scalops latimanus Backman, 1842: Boston Journal of Nat-
ural History, 4:34.

See Hall and Kelson. 1959:69-71 for additional synonymies.

The extant species, S’, latimanus, is moderately common
in Rancholabrean microvertebrate sites but several new rec-
ords can be added to those previously published. The most
interesting record of the species, however, is from the type
locality of the Irvingtonian and the oldest record of the species
thus far.

In most cases I do not attempt to describe in detail all the

6 Contributions in Science, Number 386

Hutchison: Scapanus from California

new material referable to this species but list it briefly. The
records are ordered by county and institutional locality num-
ber. The asterisk preceding some localities indicates the first
published record of S. latimanus from these localities. All
the localities are Rancholabrean in age except UCMP locality
V3604.

ALAMEDA COUNTY. *1 CMP locality V3604 (Irving-
ton site 2). Specimens: UCMP 81729, trigonid of right M,;
UCMP 81730, left humerus lacking the head and tips of
minor processes.

Remarks: The lower molar fragment (Fig. 3e) exhibits ex-
tensive enamel on the labial side of the root, loss of the
precingulid, and high paracristid. These features are diag-
nostic for S. latimanus and distinguish it from S. maiatinus.

This site is within the present geographic range of S. I.
latimanus, a subspecies which the fossil closely resembles.
The locality is the type of the Irvingtonian Land Mammal
Age (Stirton, 1939; Savage, 1951; Casteel and Hutchison,

1 973; Kurten and Anderson, 1 980) from the Irvington Grav-
els. This is the earliest record of Scapanus that can be as-
signed with confidence to S. latimanus. Lindsay et al. (1975)
placed the type locality in the upper portion of the Matayama
magnetic interval, which has a time range of about 0.7 to
1.5 Ma. Kurten and Anderson (1980) place the Irvingtonian
locality in the middle Irvingtonian.

*LICMP locality V5301 (Prune Avenue). Specimens: UCMP
43247, 1 19270-1 19278, including a dentary with P,_2, den-
tary fragments (4), M\ lunar, and proximal, medial and un-
gual phalanges of the manus.

Remarks; The site lies within the range of S. I. latimanus.
Fish from this locality were described by Casteel and Hutch-
ison (1973).

CALAVERAS COUNTY. *UCMP locality V67108 (Mer-
cer’s Cave). Specimen: UCMP 1 19249, dentary fragment.

Remarks: The site falls between the ranges of .S'. /. sericatus
Jackson and S. 1. minusculus Bangs. Sinclair (1905) described
a sloth Megalonyx sierrensis (=M. jeffersoni Leidy) from this
site. Other mammals from the locality in the UCMP collec-
tions include the extinct Euceratherium collinum and extant
Neotoma, Myotis, Lepus, Sylvilagus, Marmot a, Spermophi-
lus. and Peromyscus.

CONTRA COSTA COUNTY. UCMP locality V6312
(Garretson). Specimens: UCMP 92478-92481, 94309-9431 1,
including a dentary fragment, Mh M3, radius and ungual
phalanges of manus.

Remarks: These specimens and those of the following three
localities form the basis of the records of S. latimanus listed
by Wolff (1973, 1975) from the Montezuma Formation of
the Rodeo area. All the sites lie within the present range of
S. 1. latimanus.

UCMP locality V71001 (Centra! Channel). Specimens:
UCMP 92483-92487, 92489, 119250-119252, including a
dentary fragment, M1, M3, incomplete humerus, metacarpal
IV, calcaneum, and ungual phalanges.

I CMP locality V71O02 (Lawlers Boundary). Specimens:
UCMP 93951-93952, proximal humerus and phalanx 2 of
manus.

UCMP locality V71003 (Northern Cove). Specimens:

Figure 7. a, Scapanus latimanus cf. anthonyi, SDNHM 26231,
ventral view of rostrum: b-d, labial, lingual, and posterior views of
M,; b, 5. /. cf. anthonyi, SDNHM 26232; c, S. 1. anthonyi, MVZ
6216; d, 5. /. occultus, MVZ 4036. Scale line equals 1 mm.

UCMP 92476, 92477, 93983, M1, dentary fragment with
M,_3, and ulna fragment, respectively.

*UCMP locality V78027 (Pacheco 2). Specimens: UCMP
1 19287, 125567, including clavicle, dentary fragments, up-
per and lower molars, humerus, distal femur, medial phalanx
of manus.

Remarks: This site is a concentrated pocket of mostly small
vertebrates. No extinct vertebrates have yet been identified
(with the possible exception of a bird— D. Steadman, pers.
comm.) from the sample which is not yet completely sorted.
The pocket occurs in a series of cross-bedded alluvial deposits
that contain Megalonyx, Mammut, Mammuthus, and Equus.
The presumption that the microvertebrate site is also Ran-
cholabrean is tempered by the possibility that the deposit is

Contributions in Science, Number 386

Hutchison: Scapanus from California 7

a later cut and fill episode than that represented by the extinct
larger vertebrates, although the preservation is the same as
for the extinct forms. The site is within the present geographic
range of S. I. latimanus.

EL DORADO COUNTY. UCMP locality 1069 (Hawver
Cave). Specimens: UCMP 21143,11 9242-1 1 9248, including
humerus, femur, and pelvis.

Remarks: Locality listed as Wisconsin in Kurten and An-
derson (1980) and Harris ( 1 985) and lies between the present
geographic ranges of S. I. dilatus True and S. I. sericatus.
Stock (1918, 1925) thought that the original specimens
(UCMP 2 1 1 43, a humerus and edentulous dentary fragment,
since lost) might represent a new species. Additional material
and the variation in the living species do not support the
significance of the characters of the humerus that Stock (1918)
used to indicate distinction from S. latimanus.

FRESNO COUNTY. UCMP locality V4401 (Tranquili-
ty). Specimen: UCMP 37698, incomplete humerus.

Remarks: Hewes (1946) listed this occurrence as Scapanus
sp. and gave an accurate and detailed locality description.
Brattstrom (1959) replicated the UCMP files as to location
of the site but these appear to be somewhat misplaced by
modern map standards. It is not known if the Scapanus is
the same age as the Pleistocene fauna listed by Hewes ( 1 946)
because human artifacts and burials also are found in this
general locality and the surface collections represent a mix-
ture of the two assemblages. The human remains were ra-
diocarbon dated at about 2550 years (Berger et al., 1971).
The site lies west of the geographic range of S. I. campi
Grinnell and Storer.

LOS ANGELES COUNTY. *LACM locality 1754 (Zuma
Creek). Specimen: LACM 124286, dentary fragment with
alveoli of P3-M3.

Remarks: The locality lies within the present geographic
range of S'. /. occult us Grinnell and Swarth. Additional fauna
are recorded in Kurten and Anderson (1980).

LACM locality 6909 (Rancho La Brea Pit 91). LACM
R-2 1 363, right humerus, LACM R-52 1 77, heavily worn M3.

Remarks: The humerus was figured and described by Ak-
ersten et al. (1979:fig. lc). Chemical or radiometric dates
from this pit range from 25,100 to greater than 33,000 years
before present (Marcus and Berger, 1984). Considering the
vast quantity of small vertebrates from this and other pits
in the Rancho La Brea deposits, S. latimanus was exceedingly
rare in this area. Woodard and Marcus (1973) placed the
Rancho La Brea deposits in the upper part of Member B of
the Palos Verdes Sand. Rancho La Brea lies within the pres-
ent geographic range of S'. /. occultus.

*LACM locality Pit A (Rancho La Brea). LACM 20420,
distal femur; LACM 20419, synsacrum.

Remarks: Howard (1962) postulated on the basis of the
bird remains that Pit A was “probably active nearer the end
of the Pleistocene than the other eight pits” studied from
Rancho La Brea.

UCMP locality 2047 (San Pedro). Specimens: UCMP (if
initially properly identified) uncatalogued and since lost.

Remarks: The location and stratigraphy of this locality are
given in Miller (1971), although the locality number is er-

roneously listed as V-2047. The locality is in a marine terrace
of the Palos Verdes Sand. The Palos Verdes Sand has been
radiometrically dated (Fanale and Schaeffer, 1965:314) and
ranges in age from 100 to 130 ka. Wehmiller et al. (1977:fig.
18) estimated a more restricted age range of about 122 to
1 32 ka for the Palos Verdes Sand. A general summary of the
fauna from this general area of the Palos Verdes Sand is given
in Langen waiter (1975). The site lies within the present geo-
graphic range of S’. /. occultus.

ORANGE COUNTY. LACM locality 3877 (old number
1067) (Newport Mesa). Specimen: LACM 20839, distal hu-
merus, radius, palmar sesamoid and phalanges of manus.

Remarks: The locality and specimens are described by
Miller (1971), Hudson and Brattstrom (1977), and Kurten
and Anderson (1980). The site lies within the present geo-
graphic range of S. /. occultus and is in the Palos Verdes Sand
and the same age as given for UCMP locality 2047.

*LACM locality 7053 (Emery Borrow Pit I). Specimen:
LACM 27099, humerus.

Remarks: The locality is in the La Habra Formation, see
Kurten and Anderson (1980). Site lies within the present
geographic range of S'. /. occultus.

SACRAMENTO COUNTY. UCMP locality V69129
(Teichert Gravel Pit East 1). Specimens: UCMP 1 1 9253—
1 19256, M2 fragment, radius, ulna fragment, and proximal
phalanx of manus respectively.

Remarks: Hutchison and Williams reported this material
in a faunal list published by Hansen and Begg (1970). The
site is within the Riverbank Formation (San Joaquin loam)
and has been dated by uranium and actinium series method
at 1 03 ± 6 ka or greater. The locality lies between the present
geographic range of .S'. /. latimanus and -S’. /. minusculus.

SAN DIEGO COUNTY. *SDNHM locality 3131-B (Glen
Abbey). Specimens: SDNHM 2623 1-26238, anterior part of
rostrum, M, (2), humeri (2), ulnae (3), femur, and clavicle.

Description: The anterior part of the rostrum (Fig. 7A) is
well preserved and retains eight of the anterior teeth. The
I1-3 are preserved on both sides as well as the C and P2 on
the left side. Both sides preserve complete or partial alveoli
of the P3 to M1. The P2 alveolus is absent on the right side
but a diastema equivalent to the distance between left C-P3
remains. The Pps are absent and there are no spaces marking
their loss. The left P2 is the smallest antemolar and does not
fill the space between P3 and C.

Of the two M/s from the site, one (SDNHM 26232, Fig.
3h, 7b) is little worn and well preserved, the other is heavily
worn and abraded but agrees with the former in general fea-
tures. The M, closely resembles that of S. 1. anthonyi Allen
in degree of enamel excursion down the roots and general
conformation and size. If differs from 5. /. anthonyi and
resembles S. malatinus in presence of a narrow and incom-
plete precingulid and distinct parastylid. A remnant of the
ectocingulid occludes the base of the hypoflexid unlike 5. /.
anthonyi, and possibly resembling that of S. malatinus but
this area is not preserved in known specimens of S. mala-
tinus.

The humerus is small, measuring 12.8 mm in total length.

Remarks: The rostrum resembles S. I. anthonyi in size.

8 Contributions in Science, Number 386

Hutchison: Scapanus from California

loss of the P1 on both sides and the loss of the right P2 as
well. Scapanus l. occultus may resemble S. I. anthonyi in size
and in the loss of one or both P1 (Palmer, 1937:312) but no
cases of a P2 being lost as well have been reported. This
presumably derived pattern of tooth loss would thus nor-
mally justify assignment of the San Diego specimen to S. I.
anthonyi. The area of San Diego is presently occupied by S.
1. occultus with S. 1. anthonyi occurring to the south in the
mountains in Baja California (Palmer, 1937; Hall and Kel-
son, 1959). Presence of S. 1. anthonyi in San Diego would
thus constitute a range extension of the species during the
late Pleistocene. This picture is complicated, however, by the
rather primitive morphology of the M, of the San Diego
fossil, the only lower tooth thus far known of the taxon. The
degree of enamel extension down the roots more closely re-
sembles that of S. malatinus than S. 1. anthonyi (Fig. 3). The
presence of a small parastylid, narrow precingulid, and dis-
tinct transverse ectocingulid cuspid resemble the primitive
conditions as those in S. malatinus and were not observed
in S. 1. anthonyi (Fig. 9). The small ectocingulid cuspid may,
however, occur on the M, of other 5. latimanus.

Chandler (1982) described a passenger pigeon, Ectopistes
migratorius, from this locality but the rest of the fauna is
undescribed. The locality lies in unnamed river terrace de-
posits (Chandler, 1982). The age of the site is not precisely
determined. The only temporally significant species presently
known from the site is an Equus occidentalis tooth identified
by Chandler (1982) but the basis of the identification was
not given. Whereas Equus occidentalis is best known from
the Rancholabrean (especially Rancho La Brea), it may be
present in earlier deposits (Kurten and Anderson, 1980). As
the morphology of the M, of the San Diego mole appears to
be intermediate between S', malatinus and S. I. anthonyi, the
Rancholabrean age of the site should be considered as ten-
tative. The San Diego specimens are here tentatively referred
to S. 1. cf. anthonyi on the basis of dental formula, but further
comparison with Recent material from San Diego (not seen)
is desirable. Moreover, more and better material of S. ma-
latinus and the fossil San Diego form may indicate another
assignment.

SAN BERNARDINO COUNTY. Reynolds and Reynolds
(1984) and Reynolds (1984) published a record of S. lati-
manus in a list with the other vertebrates and invertebrates
from SBCM locality 1.76.34 (Luz Foundation). The tooth
fragment upon which this was based is indeterminate as to
genus and is not here regarded as a valid record.

SBCM locality 1.76.33 (Solid Waste site). Specimen: SBCM
A 1776-173, clavicle.

Remarks: This specimen was reported in a list with the
other vertebrates and invertebrates by Reynolds and Reyn-
olds (1985) who provided a radiocarbon date for the site of
12,210 ± 430 years. The site lies northeast of the nearest
approach of the present geographic range of S. I. occultus.

SBCM locality 1.76.35 (Calico Lakes Phase I). Specimen:
SBCM A 1855-1, clavicle.

Remarks: Published in a list with the other vertebrates and
invertebrates by Reynolds and Reynolds (1985). This ma-
terial is associated with a radiocarbon date of 9050 ±350

years. The site lies northeast of the nearest approach of the
present geographic range of S. 1. occultus.

SHASTA COUNTY. UCMP locality 1008 (Samwell Cave
1). Specimens: UCMP 9623, 35685, 1 19257-1 19261, an-
terior part of skull with right and left M2, dentary, rostral
fragment with M1, incomplete femur, pelvi.

Remarks: Furlong (1906), Stock (19 18), and Hibbard (1958)
recorded these specimens. Graham (1959) and Treganza
(1964) provided detailed maps of the cave and Treganza
documents the lack of positive evidence for human and Pleis-
tocene faunal associations. Harris (1985) presumed the fossil
material to be late Wisconsinan in age. Graham (1967) sug-
gested that the fossil moles from Samwell Cave were distinct
enough to merit a new fossil subspecies, but formal descrip-
tion has not yet appeared. The cave lies within the present
geographic range of 5. /. dialatus.

UCMP locality 1055 (Potter Creek Cave). Specimens:
UCMP 5630, 33164, dentaries with M1-3; UCMP 7648, 7046,
pelvi; UCMP 7532, ulna; UCMP 5343, 6395, 6538, 6539,
6819, 6821, 7464, 7465, 7506, 7576-7578, 7668, 7853, 8249,
33167, humeri.

Remarks: The locality was described and the fauna was
listed by Sinclair ( 1 904) and Miller (1912). Hibbard (1958),
Brattstrom (1954, 1958), and Hutchison (1967) added ad-
ditional taxa to the list. Sinclair (1904) and Harris (1985)
regarded the site as late Wisconsinan or older in age. Sinclair
listed the Potter Creek Cave mole as Scalops cf. townsendii
in 1903 but corrected this to S’. californicus(1) Ayres in 1904
(now a junior synonym of S. latimanus). Stock (1918) pre-
sented a revised list from this cave and listed the mole as S.
latimanus for the first time. The cave lies within the present
geographic range of 5. /. dialatus.

*UCMP locality V75068 (Galen’s Pit). Specimens: UCMP
1 19262-1 19269, including pelvis fragment, ulnae fragments,
scaphoid, trapezium, metacarpals II and III, and calcaneum
fragment.

Remarks: This site has not been previously published. It
is a cave in the Hosselkus Limestone collected by Jens Munthe.
Specimens came from a flowstone breccia including such
extinct taxa as Equus, Mammuthus, Euceratherium, and ex-
tant genera Gymnogvps (cf. G. californicus). Odocoileus,
Aplodontia, Thomomys, Neotoma, and Peromyscus. The cave
lies within the present geographic range of S'. /. dialatus.

Scapanus new species?

Figures 8a-d

MATERIAL. UCMP 126753, left M2 fragment (Fig. 8c);
UCMP 126754, damaged right M, (Fig. 8a); UCMP 126755,
damaged left M, (Fig. 8b); UCMP 126756-126778, scapula
fragment, humerus, lunar, metacarpals I-V, proximal, me-
dial, and ungual phalanges of the manus, astragalus, meta-
tarsals, and pedal phalanx.

UCMP locality V6869 (Maxum), Contra Costa County,
California. Tassajara Formation.

AGE. Pliocene, earliest Blancan, Blancan I of Repenning

Contributions in Science, Number 386

Hutchison: Scapanus from California 9

Figure 8. a-c, Scapanus ( Scapanus ) new species? from the Maxum
locality; a, UCMP 126754, right M,, occlusal, lingual, labial, and
anterior views; b, UCMP 126755, left M,, occlusal, lingual, labial,
and posterior views; c, UCMP 126753, left M2, posterior view; d,
Scapanus ( Scapanus ) new species?, USNM 264301, trigonid of left
M2, occlusal, anterior, lingual, and labial views, e-j, comparison of
posterior views of M:; e, S. (S.) latimanus cf. caurinus, UCMP
131 165, dashed line indicates limit of cementum; f, S. (Scapanus)
townsendii, JHH “C”; g. S'. (Scapanus) orarius, JHH "A”; h, S.
(Xeroscapheus) proceridens. UCMP 124433, worn; i, Scapanus (X.)
shultzi, UCMP 124434; j, Scapanoscapter simplicidens, UO 24290.
Scale line equals 1 mm.

(1980), 4. 3-4. 8 Ma, Gilbert reversed magnetic polarity ep-
och, upper chron 2 of May and Repenning (1982).

REFERRED MATERIAL. USNM 264301, trigonid of
?M, (Fig. 8d) USGS locality Ml 451 (Radec A), Riverside
County, California. Temecula Arkose Formation, ascribed
to the late Blancan, 2. 2-2. 7 Ma by Golz et al. (1977), but
considered to be early Blancan by Repenning (pers. comm.)
and close to the Maxum locality in age.

DESCRIPTION. The M2 fragment (UCMP 126753) re-
sembles that tooth of 5. orarius and S. townsendii and differs
from S. shultzi (Fig. 8i) in greater size of the lingual root
with respect to the labial roots, strong ventrally directed notch
in the enamel margin on the posterior side of the protocone
and lack of well-defined conules. The M,, although damaged,
appears to be rather high crowned compared to S. orarius
and S. townsendii and has a more compressed trigonid. There
is no enamel excursion down the roots as in 5. malatinus,
S. latimanus, and S. proceridens. The region of the parastylid
is broken away and there is no indication of a precingulid
on the remaining part of the tooth. The M, (UCMP 126755)
also exhibits a high-crowned condition with compressed tri-
gonid but retains a well-developed precingulid as in S. shultzi
and unlike the extant species. The ?M: trigonid of the Radec
specimen lacks a defined precingulid.

Metacarpal II (UCMP 126760) resembles that of S. ma-
latinus and S’, latimanus in the separation of the proximal
tuberosity (the proximal prominence is broken away).

REMARKS. Whereas the above material is not referable
to a known species, the material is too fragmentary to form
the basis of a new taxon at present.

Scapanus species indet.

MATERIAL. SDNHM locality 3181 -E (J Street II), San
Diego County. “Upper” member San Diego Formation,
Blancan (Demere, 1983; Domning and Demere, 1984). Spec-
imen: SDNHM 26230, humerus lacking part of proximal
end.

REMARKS. The humerus is in the size range of S. ma-
latinus, S. orarius, and small S. latimanus but is not specif-
ically diagnostic. The age of the site would suggest relation-
ship to the questionable new species Scapanus described
above.

The locality is a marine shell bed containing a variety of
sharks, a ray and such terrestrial vertebrates as rodent, snake,
lagomorph, felid, Equus, and an artiodactyl. The locality lies
very high in the San Diego Formation (“upper” member of
Demere, 1983) and was deposited in a very shallow nearshore
marine environment (Demere, pers. comm., 1985).

DISCUSSION

Whereas a respectable suite of genetic (Yates and Green-
baum, 1982), gross proportional, and soft tissue character
states (Jackson, 1915; Palmer, 1937; Hall and Kelson, 1959)
distinguish the extant species of Scapanus from each other,
these are not preserved in typical fossil remains. Fortunately,
a number of dental and osteological features may also be
used (Table 3). Scapanus latimanus is more distinct from S.
orarius and S. townsendii in skull and dental features than
either of the latter species are from each other. Scapanus
townsendii and S. orarius are in part sympatric (Fig. 1) and
probably partition the resource in their range on the basis of
size (McNab, 1971). Scapanus townsendii is distinctly larger
(195-237 mm body length) than S’, orarius (162-175 mm).
Scapanus latimanus occupies a larger range than either of
the other species and shows a greater range (Hall and Kelson,

10 Contributions in Science, Number 386

Hutchison: Scapanus from California

Table 3. Selected character state matrix for species of Scapanus, using Scapanoscapter as the immediate primitive outgroup. Primitive, derived

and variable states indicated by 0, 1, and V respectively.

Derived character state

Scapano-

scapter

Scapanus

shultzi

proceri-

dens

orarius

town-

sendii

(Maxum)

malatinus

latimanus

1. Sinuous enamel M,_, margin

0

1

i

i

1

1

1

1

2. PU2 single-rooted

0

1

i

i

1

1

1

1

3. P4 with fused roots

0

1

i

0

0

0

0

0

4. P4 single-rooted

0

0

0

1

1

1

1

1

5. M i_2 conules reduced

0

0

0?

1

1

1

1

1

6. Crown hypsodonty

0

1

1

0

0

1

1

1

7. Cingulids lost

0

0

0

1

1

0

0

1?

8. Trigonids open

0

0

0

1

1

0

0

0

9. M, labial enamel excursion

0

0

0

0

0

0

1

1

10. M,_j with labial excursion

0

0

0

0

0

0

0

1

1 1 . Radius elongate

7

0

0

1

0

7

0

0

12. Reduction in premolar no.

0

1

1

0

0

?

?

V

13. Hypertrophy of I2

0

1

1

0

0

?

?

1

14. Cementum development

0

0

1

0

0

0

0

1

15. Infraspinatus fossa of scapula not ex-

tending to acromion process

?

0

0

1

1

1

1

1

16. Deltoid process of tibia spine-like

?

0

0

1

1

7

1

1

17. Small vascular notch on clavicle

?

0

0

1

1

7

7

1

1959) in body size (132-192 mm). In general S. latimanus
shows a decrease in size and tooth formula from north to
south (Palmer, 1 937; Ziegler, 1 97 1) and this is generally coin-
cident with increasing aridity to the south. Whereas it is clear
that hypsodonty in moles is a derived character (Hutchison,
1968), the polarity of the limb proportions is unclear. The
very large size attained in S. townsendii is probably derived
but simple size characters are notoriously unreliable indices
of relationship. The size trend of 5. latimanus to S. town-
sendii can be viewed as an example of Bergmann’s rule
(Palmer, 1937). The radii of both show some negative al-
lometry with regard to the humerus (Fig. 5), that may reflect
a simple size/power relationship. Scapanus malatinus, S. la-
timanus, and S. townsendii appear to lie along approximately
the same trend line. The sample of S. orarius is small but
the humerus/radius plots are clearly transposed to the right
of the other’s curve. Whereas S. latimanus exhibits clear
specializations such as hypsodonty and tooth crowding or
reduction with regard to the other two species, the relation-
ship of the latter to each other is far from clear.

The hypothesized cladistic relationships of the five species
are presented in Figure 9 with the numbered nodes indicating
the derived character states at each level (Table 3). A pos-
tulated phylogeny of the five species based on the cladistic
analysis, stratigraphic sequence, and geography is also pre-
sented in Figure 9.

The species 5. latimanus can be traced back in time to the
later Irvingtonian. Scapanus malatinus from the early Ir-
vingtonian provides a morphological link between S. lati-

manus on one hand and S. townsendii and S. orarius on the
other. The high-crowned lower molars with evidence of
enamel excursion down the roots, at least in M,_2, arching
of the skull and probable shortening of the snout are derived
character states shared with S. latimanus, but not with the
other two. The radius in S', malatinus is also relatively short
(Fig. 5) as in S. latimanus and S. townsendii. Scapanus ma-
latinus has very well developed ectocingulids, remnants of
precingulids, and compressed trigonids — all character states
that can be considered as primitive (using S. shultzi and
Scapanoscapter as outgroups). The loss of these character
states in the case of the cingulids or modification in the case
of the spreading of the trigonid cusps in both S. orarius and
S. townsendii suggest that these species share a common
ancester not shared by S. malatinus and S. latimanus. This
arrangement is also favored by the distribution of the species.
Shortening of the radius is a widespread phenomenon in
diverse lineages of talpids but, based on comparison to Xe-
roscapheus and its distribution in modern Scapanus, it is
here tentatively considered as primitive for Scapanus. On
the basis of the above features and supported by their geo-
graphic array, the species of the S'. ( Scapanus ) are here di-
vided into two informal groups, the Orarius group consisting
of S. orarius and S. townsendii and the Latimanus group
consisting of S. malatinus and S. latimanus plus all other
known fossils of the subgenus in California.

The Blancan specimens from the Maxum locality that rep-
resent an unnamed species are high crowned with compressed
trigonids but lack enamel excursion down the roots. The

Contributions in Science, Number 386

Hutchison: Scapanus from California 1 1

Figure 9. Cladogram of the species of Scapanus using Scapanoscapter as an outgroup and the favored phylogeny for the same species.
Numbers indicate derived characters in Table 3. Characters derived twice are underlined, “v” indicates variable.

cingulids vary from poorly to well defined. In general, the
teeth closely resemble 51. malatinus and the location of the
Maxurn site in central California also favors a relationship
with the Latimanus group.

The subspecies of S', latimanus (Palmer, 1937; Hall and
Kelson, 1 959) exhibit a wide range of tooth morphology with
the smaller southern species generally less hypsodont than
the northern forms. The presence in the Irvingtonian of
northern California of a very derived type of tooth mor-
phology such as that in S. /. latimanus and a more primitive
state in 5. malatinus (and S. I. cf. anthonyi) from the south
may indicate a long-standing north-south cline in hypso-
donty in which S’. /. anthonyi and S'. /. occultus are the most
conservative extant members. The reduction in tooth num-
ber may be an even older chronological phenomenon and
may persist through ancestor descendant morphospecies
erected on tooth morphology. The presence of a highly re-
duced tooth count in late Pleistocene species (S. I. cf. an-
thonyi) from San Diego coupled with a relatively primitive
tooth morphology may presage a reduced tooth count in 5.
malatinus which will become recognizable only when the
species is better known.

The development and degree of hypsodonty in Scapanus
may be partly size dependent. The large northern subspecies
have the most extreme development of enamel excursion
down the lower molar roots and the smaller subspecies, S.
1. anthonyi, the least. The very large S. townsendii is also
higher crowned than the small and partly sympatric S. or-
arius. This general pattern in S', latimanus may have ex-
tended back in time as far as the Irvingtonian. Some Pleis-
tocene S. latimanus occur in a few areas not presently within
the range of the living species, indicating a greater range

during at least the Wisconsin glacial interval. Scapanus I.
anthonyi is a likely relict of this greater Pleistocene range.

The cladistic relationships of populations are not neces-
sarily determinable or meaningful where there is a sufficient
genetic flow between them to allow character states derived
in one area to dilfuse into another. Populations may be di-
vided into several practically recognizable taxa (subspecies,
demes, etc.) but if character state (genetic) interchange con-
tinues, a cladistic analysis may provide a set of pseudotaxa
determined principally by the choice of characters used and
subject to the vagaries of historical and ecological changes.
The incidence of the character state could indicate the timing
of relationships (cladistic division) or only differential ac-
ceptance of the state throughout the varied range of the species
group. A much denser and well-dated fossil record may re-
solve these opposing hypotheses. Whereas S. malatinus is
just outside of the range of variation of extant species, a better
fossil record may show that it is only a segment of a heter-
ochronically evolving set of character states within the bio-
logical concept of 5. latimanus. Meanwhile practicality is
served by treating S. malatinus as a paleospecies and not
blurring the morphological definition of an extant species.

The earliest evidence of “true” molar hypsodonty (enamel
below the gum line, cementum) in S. ( Scapanus ) appears
between 1.2 and 1.7 Ma. This constitutes the latest apparent
acquisition of hypsodonty in the Talpidae. “True” molar
hypsodonty occurs in two other lineages of talpids: Scapanus
(Xeroscapheus) and Scalopus (and also in the proscalopids,
Mesoscalops Reed). Two other lineages (Domninoides Green,
Scaptochirus Milne-Edwards) also show extreme crown hyp-
sodonty but do not develop enamel excursion or cementum
in the lower molars. The development of high-crowned or

12 Contributions in Science, Number 386

Hutchison: Scapanus from California

hypsodont molars is associated in each case with hypertrophy
of the L, reduction in the length of the antemolar region with
crowding and frequently loss of some antemolars (Hutchison,
1968; Ziegler, 1971). All the high-crowned moles are also
highly derived burrowers but not all highly derived burrowers
show a tendency toward hyposodonty (e.g., S. orarius, S.
townsendii, Talpa europea Linnaeus, Parascalops breweri
(Bachman), Proscapanus sansaniensis (Lartet)). If abrasive
characteristics of the diet were held constant then acquisition
of hypsodonty would allow a longer tooth life (i.e., potential
animal longevity). There is no indication among any of the
living members of this group of a longevity advantage over
the non-hypsodont burrowing moles. If all the moles live to
about the same actuarial age, the hypsodonty is of little func-
tional advantage except where the environment is dentally
more abrasive. Such adaptations would permit these moles
to expand into more xeric and mountainous areas where
coarse-grained soils predominate. Hypsodonty in these oth-
erwise marginal edaphic environments would compensate
for increased tooth wear and allow the moles to live normal
life spans. Crown hypsodonty in Domninoides develops in
the Great Plains during the Clarendonian, a time of devel-
oping aridity in the Great Plains of North America (Gregory,
1971; Elias, 1942; Hutchison, 1982). The Great Basin sub-
genus Xeroscapheus apparently originates also during the
Clarendonian (Hutchison, 1968) and also during a time of
increased aridity (Shotwell, 1964; Axelrod, 1980) and con-
tinued volcanism. Scalopus aquaticus (Linnaeus) presently
occupies the largest geographic range of any living North
American mole including many humid and humus-rich soils
(Hall and Kelson, 1959). Scalopus successfully utilizes coarse-
grained edaphic environments such as the deep loose sands
in central Florida and upper beach sands (both of which may
be considered as ecologically xeric) and the humus-poor soils
of the Great Plains. The earliest fossil Scalopus (Hemphil-
lian) are from areas of Kansas (Hutchison, 1968:7) and Texas
(Dalquest, 1983) that show sedimentary indications of dis-
tinct seasonal aridity such as development of caliches or
prevalence of volcanic ash. The extant S. latimanus in gen-
eral occupies the least humid areas of the range of the genus,
although much of the Pacific slope is edaphically abrasive
(high in silica and other hard minerals). Scaptochirus occu-
pies mountainous areas in the northern province of China
(Schwartz, 1948) where soil profiles are probably shallow or
young and therefore coarse.

The hypertrophy of the L and shortening of the rostrum,
which is concurrent with the development of high-crowned
or hypsodont teeth, is not as readily explained by coarsening
soils because these teeth are used more in grasping than
crushing. Development of powerful piercing teeth along with
short snouts is likely to reflect changes in the nature of the
food resource itself. In semiarid and coarse-grained environ-
ments as noted by Palmer (1937:283) the edaphic fauna is
likely to consist of fewer soft-bodied animals (worms, slugs,
and thin-shelled insect larvae) and more hard-shelled forms
(hard-shelled larvae, shelled snails, and burrowing adult in-
sects) adapted to resist soil abrasion and periodic desiccation.
Because the hardness of chitin and shell is much less than

enamel, this does not readily account for development of
hypsodonty but would indicate a need for a more powerful
bite at the tip of the rostrum in order to hold and pierce the
prey. This is accomplished by concentrating the bite to a
single point (clustered I' + L) for prey penetration and by
shortening of the rostrum to increase the mechanical advan-
tage of the incisors at the tip of the dentary. It would be
interesting to know if there is an increase in the effective size
of the prey in these areas. If this were so, bite strength at the
tip of the snout would be useful in subduing larger prey. An
increase in gape angle might also be expected in short-snouted
moles, if only to maintain the vertical span of the gape.

ACKNOWLEDGMENTS

Research support was provided by the Annie M. Alexander
Endowment to the University of California Museum of Pa-
leontology. T. Downs, aside from being instrumental in the
development of the Vallecito collections, also provided
stratigraphic information. T.A. Demere, San Diego Natural
History Museum, and C.A. Repenning, U.S. Geological Sur-
vey, Menlo Park, called the unpublished Blancan material
to my attention and provided age information. The following
people generously loaned material in their care: T. Downs,
D.P. Whistler, G.T. Jefferson, C.A. Shaw, and R.L. Reynolds
(Los Angeles County Museum of Natural History), G. J. Mil-
ler (Imperial Valley College Museum), R.E. Reynolds (San
Bernardino Natural History Museum), J.L. Patton and R.E.
Jones (University of California Museum of Vertebrate Zo-
ology), and M.O. Woodburne (University of California, Riv-
erside). A.D. Barnosky and M.J. Novacek provided construc-
tive criticisms.

LITERATURE CITED

Akersten, W.A., R.L. Reynolds, and A.E. Tejada-Flores.
1979. New mammalian records from the late Pleisto-
cene of Rancho La Brea. Bulletin of the Southern Cali-
fornia Academy of Sciences 78:141-143.

Axelrod, D.I. 1980. Contributions to the Neogene paleo-
botany of central California. University of California
Publications in Geological Science 121:1-212.

Berger, R., R. Protsch, R. Reynolds, C. Rozaire, and J.R.
Sackett. 1971. New radiocarbon dates based on col-
lagen of California Paleo-Indians. Contributions of the
University of California Archaeological Research Facility
12:43-49.

Brattstrom, B.H. 1954. The fossil pit-vipers (Reptilia Cro-
talidae) of North America. Transactions of the San Diego
Society of Natural History 1 2:3 1-46.

. 1958. New records of Cenozoic amphibians and

reptiles from California. Bulletin of the Southern Cali-
fornia Academy of Sciences 57:5-13.

. 1959. A new species of fossil turtle from Western

North America. Bulletin of the Southern California
Academy of Sciences 58:65-71.

Casteel, R.W., and J.H. Hutchison. 1973. Orthodon (Ac-
tinopterygii, Cyprinidae) from the Pliocene and Pleis-
tocene of California. Copeia 1973:358-361.

Contributions in Science, Number 386

Hutchison: Scapanus from California 13

Chandler, R.M. 1982. A second record of Pleistocene pas-
senger pigeon from California. Condor 84:242.

Dalquest, W. W. 1 983. Mammals of the Coffee Ranch local
fauna Hemphillian of Texas. The Pearce-Sellards Series
38: 1-41.

Demere, T.A. 1983. The Neogene San Diego Basin: a re-
view of the marine Pliocene San Diego Formation. In
Cenozoic Marine Sedimentation. Pacific Margin, U S. A.,
Society of Economic Paleontologists and Mineralogists,
eds. D.K. Larue and R.J. Steel, 187-195.

Domning, D.P., and T.A. Demere. 1984. New material of
Hydrodamalis cuestae (Mammalia: Dugongidae) from
the Miocene and Pliocene of San Diego County, Cali-
fornia. Transactions of the San Diego Society of Natural
History 20: 1 69-1 88.

Downs, T., and J.A. White. 1968. A vertebrate faunal
succession in superposed sediments from late Pliocene
to middle Pleistocene in California. In Proceedings of
XXIII International Geological Congress, Section 10,
Tertiary-Quaternary Boundary , 41-47. Academic Press,
Prague.

Elias, M.K. 1942. Tertiary prairie grasses and other herbs
from the High Plains. Geological Society of America,
Special Paper 41:1-176.

Fanale, F.P., and O.A. Schaeffer. 1965. Helium-uranium
ratios for Pleistocene and Tertiary fossil aragonites. Sci-
ence 149:312-316.

Furlong, E.L. 1906. The exploration of Sarnwel Cave.
American Journal of Science 22:235-247.

Golz, D.J., G.T. Jefferson, and M.P. Kennedy. 1977. Late
Pliocene vertebrate fossils from the Elsinore fault zone,
California. Journal of Paleontology 51:864-866.

Graham, R. 1959. Additions to the Pleistocene fauna of
Sarnwel Cave, California. I. Canis lupus and Canis lat-
rans. Cave Studies 10:54-67.

Graham, S.A. 1967. Moles of the Sarnwel Cave fossil de-
posits: additions to the Pleistocene fauna of Sarnwel Cave,
California II. Caves and Karst 9:49.

Gregory, J.T. 1971. Speculations on the significance of fos-
sil vertebrates for the antiquity of the Great Plains of
North America. Hessischen Landesamtes fur Bodenfor-
schung, Abhandlungen 60:64-72.

Hall, E.R., and K.R. Kelson. 1959. The mammals of North
America. New York: Ronald Press, 1084 pp.

Hansen, R.O., and E.L. Begg. 1970. Age of Quaternary
sediments and soils in the Sacramento area, California
by Uranium and Actinium series dating of vertebrate
fossils. Earth and Planetary Science Letters 8:41 1-419.

Harris, A.H. 1985. Late Pleistocene vertebrate paleontology’
of the West. Austin: University of Texas Press, 293 pp.

Hewes, G.W. 1946. Early man in California and the Tran-
quility Site. American Antiquity 1 1:209-216.

Hibbard, C.W. 1958. Summary of North American Pleis-
tocene mammalian local faunas. Papers of the Michigan
Academy of Science, Arts, and Letters 43:3-32.

Howard, H. 1962. A comparison of avian assemblages from
individual pits at Rancho La Brea, California. Contri-

butions in Science, no. 58, 24 pp. Natural History Mu-
seum of Los Angeles County.

Hudson, D.M., and B.H. Brattstrom. 1977. A small her-
petofauna from the late Pleistocene of Newport Beach
Mesa, Orange County, California. Bulletin of the South-
ern California Academy of Sciences 26:16-20.

Hutchison, J.H. 1967. A Pleistocene vampire bat (Des-
modus stocki) from Potter Creek Cave, Shasta County,
California. Paleobios 3:1-6.

. 1968. Fossil Talpidaeflnsectivora, Mammalia) from

the later Tertiary of Oregon. Bulletin of the Oregon Mu-
seum of Natural History 11:1-117.

. 1974. Notes on type specimens of European Mio-
cene Talpidae and a tentative classification of Old World
Tertiary Talpidae (Insectivora: Mammalia). Geobios 7:
21 1-256.

. 1982. Turtle, crocodilian, and champsosaur diver-
sity changes in the Cenozoic of the north-central region
of western United States. Palaeogeography, Palaeocli-
matology, Palaeoecology 37:149-164.

Jackson, H.H.T. 1915. A review of the American moles.
North American Fauna, no. 38, 100 pp.

Kurten, B., and E. Anderson. 1980. Pleistocene mammals
of North America. New York: Columbia University Press,
442 pp.

Langenwalter, P.E., II. 1975. The fossil vertebrates of the
Los Angeles-Long Beach Harbors region. In Paleonto-
logic record of areas adjacent to the Los Angeles and
Long Beach Harbors, Los Angeles County, California,
ed. G.L. Kennedy, 36-54. Marine Studies of San Pedro
Bay, California. Part 9. Allan Hancock Foundation Har-
bors Environmental Projects, Sea Grant Programs, Uni-
versity So. California, Los Angeles, California.

Lindsay, E.H., N.M. Johnson, and N.D. Opdyke. 1975.
Preliminary correlation of North American Land Mam-
mal ages and geomagnetic chronology. University of
Michigan Papers on Paleontology’ 12:1 1 1-1 19.

Marcus, L.F., and R. Berger. 1984. The significance of ra-
diocarbon dates for Rancho La Brea. In Quaternary Ex-
tinctions—a prehistoric revolution, eds. P.S. Martin and
R.G. Klein, 1 59-183. Tucson: University Arizona Press.

May, S.R., and C.A. Repenning. 1982. New evidence for
the age of the Mount Eden fauna, southern California.
Journal of Vertebrate Paleontology 2: 109-1 1 3.

McNab, B.K. 1971. On the ecological significance of Berg-
mann’s Rule. Ecology’ 52:845-854.

Miller, L.H. 1912. Contributions to avian paleontology from
the Pacific Coast of North America. University of Cal-
ifornia Publications Bulletin of the Department of Ge-
ology’ 7:61-1 15.

Miller, W.E. 1971. Pleistocene vertebrates of the Los An-
geles Basin and vicinity (exclusive of Rancho La Brea).
Bulletin of the Los Angeles County Museum of Natural
History 10:1-124.

Opdyke, N.D., E.H. Lindsay, N.M. Johnson, and T. Downs.
1977. The paleomagnetism and magnetic polarity stra-
tigraphy of the mammal-bearing section of Anza Bor-

14 Contributions in Science, Number 386

Hutchison: Scapanus from California

rego State Park, California. Quaternary Research 7:31 6—
329.

Palmer, F.G. 1937. Geographic variation in the mole Sca-
panus latimanus. Journal of Mammalogy’ 18:280-314.

Repenning, C. A. 1980. Faunal exchanges between Siberia
and North America. Canadian Journal of Anthropology
1:37-44.

. 1985. Pleistocene mammalian fauanas: climate and

evolution. Acta Zoologica Fennica 170:173-176.

Reynolds, R.E. 1984. Paleontologic salvage, Solar Trough
site, Wismer and Becker Solar Electric Generating Sta-
tion 1 (Luz), Daggett, San Bernardino County, Califor-
nia. Redlands: San Bernardino County Museum, 33 pp.

Reynolds, R.E., and Reynolds, R.L. 1984. Paleontologic
salvage, Solar Trough site, Wismer and Becker Solar
Electric Generating Station 1 (Luz), Daggett, San Ber-
nardino County, California. Redlands: San Bernardino
County Museum, 127 pp.

. 1985. Late Pleistocene faunas from Daggett and

Yermo, San Bernardino County, California. In Geologic
Investigations Along Interstate 15, Cajon Pass to Manix
Lake, California, Compiler R.E. Reynolds, 175-191.
Redlands: San Bernardino County Museum.

Savage, D.E. 1951. Late Cenozoic vertebrates of the San
Francisco Bay Region. University of California Publi-
cations, Bulletin of the Department of Geological Sci-
ences 28:215-314.

Schwarz, E. 1948. Revision of the Old-World moles of the
genus Talpa Linnaeus. Proceedings of the Zoological So-
ciety of London 1 1 8:36-48.

Shotwell, J.A. 1964. Community succession in mammals
of the Late Tertiary. In Approaches to Paleoecology’, eds.
J. Imbrie and N. Newell, 135-150. New York: John
Wiley and Sons.

Sinclair, W.J. 1903. A preliminary account of the explo-
ration of the Potter Creek Cave, Shasta County, Cali-
fornia. Science (new series) 17:708-712.

. 1904. The exploration of the Potter Creek Cave

University of California Publications American Archae-
ology and Ethology > 24:339-410.

. 1905. New Mammalia from the Quaternary caves

of California. University of California Publications Bul-
letin of the Department of Geology 6:145-161.

Stirton, R.A. 1939. Cenozoic mammal remains from the
San Francisco Bay region. University of California Pub-
lications Bulletin of the Department of Geological Sci-
ences 24:339-4 10.

Stock, C. 1918. The Pleistocene fauna of Hawver Cave.
University of California Publications Bulletin of the De-
partment of Geology 10:461-515.

— . 1925. Cenozoic gravigrade edentates of western

North America, with special reference to the Pleistocene
Megalonychinae and Mylodontidae of Rancho La Brea.
Carnegie Institution of Washington Publication 331:1—
206.

Tedford, R.H. 1961. Clarendonian Insectivora from the
Ricardo Formation, Kern County, California. Bulletin

of the Southern California Academy of Sciences 60:57-
76.

Treganza, A.E. 1964. An ethno-archaeological examination
of Samwel Cave. Cave Studies 12:1-29.

Wehmiller, J.F., K.R. Lajoie, K.A. Kvenvolden, E. Peterson,
D.F. Belknap, G.L. Kennedy, W.O. Addicott, J.G.
Vedder, and R.W. Wright. 1977. Correlation and chro-
nology of Pacific Coast marine terrace deposits of con-
tinental United States by fossil amino acid stereochem-
istry-technique evaluation, relative ages, kinetic model
ages, and geologic implication. United States Geological
Survey Open-File Report 77-680:106 pp.

Wolff, R.G. 1973. Hydrodynamic sorting and ecology of a
Pleistocene mammalian assemblage from California
(U.S.A.). Palaeogeography, Palaeochmatology, Palaeo-
ecology ’ 13:91-101 .

. 1975. Sampling and sample size in ecological anal-
ysis of fossil mammals. Paleobiology ; 1:195-204.

Woodard, G.D., and L.F. Marcus. 1973. Rancho La Brea
fossil deposits: re-evaluation from stratigraphic and geo-
logical evidence. Journal of Paleontology 47:54-69.

Yates, T.L., and I.F. Greenbaum. 1982. Biochemical sys-
tematics of North American moles (Insectivora: Talpi-
dae). Journal of Mammalogy’ 63:368-374.

Ziegler, A.C. 1971. Dental homologies and possible rela-
tionships of Recent Talpidae. Journal of Mammalogy’
52:50-68.

Received 12 December 1985; accepted 18 December 1986.

Contributions in Science, Number 386

Hutchison: Scapanus from California 1

INSTRUCTIONS- FOR AUTHORS

The Natural History Museum 6$Lo$ nty the.resthts ofpriginal research,

in the life and earth sciences in its Contributions; in Science series. Issues of Contributions

are published at irregular intervals sub range sin size from short papers to lengthy monographs.

Manuscripts submitted, for publication will Undergo anonymous peer review. Priority is
given to .manuscripts written by. members of the Museum staff. Manuscripts should be
prepared in accordance with the requirements outlined below and' submitted to the Head of
the appropriate Section of the Museum:..

Authors must adhere to the articles of the International -Codes of Botanical or Zoological
Nomenclature; attd are urged to comply with all recommendations of these codes.

Authors proposing: new faixa mhst fiadictste: tha.f primary types. Have been deposited •»
accbrdahCei.iyi^i . the recommendations of the appropriate code, Citiitg:'th.e|:f:tdle<h:ic5i;i; hymame
and providing tile pelevantridehti lying details. The depository of other study ma terial should
also be indicated. ■; |j| ■ VTA

j An all: papefA- in manuscr ipts dealing with Latin American subjects,

theAbstraet must: §e presented m Spain sh ;<Jr Portpibesc.; as; appropriate;; d.s? well as in English .
Summaries in other languages are not; required bat ile strong! iy re-co mm ended. where ap-
: propriAtevjjJitif i ;f§t|f If d f ;;;: |;i;Vj ;j yy

PREPARAi'ibi OF MAN USCRl PI"

Type itlilp® texiyacl^

erences. tables, and figure captions. Indicate italicized words with imrierscores; do not use

an malic typing element. .

Each manuscript comppien.t ^b'ouM:1x:gin on a new page, in |Sfe to; i owing sequence: title:
page; abstract(s): text; acknowledgments; literature cited; tables, 'each table “-complete, wi#
title and footnotes- on a separate page; and figure captions. All. pages should, be numbered
consecutively.

: , ■ ; Submit or iginal, 'illustrations;,. Line drawings are best prepared j for 50 percent red ucuon ■
and should not exceed 14 by I 7.5 by & ;jlm| ■infsKe; ;,pbdldgfapbs: ate usually

reproduced at full size and should not exceed 7 by 8.75 inches ( 1 7 by 22 cm. full page widtltl
or 3 25 by 4.5 inches (8 by 1 1 cm, single column). 1

St| RMISSION OF ; MANUSCRIPT

Transmit three copies of the 'manuscript; in a |pdrg:AlRibsum ; Manuscript • should l>e at- . I: 'f
companied. by a; cover letter signed by the' an thor who vri I): be resporisibkibr correspomience
regarding the man uscript. The covering letter should contain a vatcnr.rnl that the mac men pt
is based on original research and has not been published elsewhere • except .in abstract- or ••
abbreviated form. The letter should also giyja information or prior publication of any pan:
of the research and whether or not funds are available to meet the cost of unusual format
or color illustrations. Include copies of i*my permissions needed to .reproduce;: published

tTyafATiialL:;. rie-etd'eri.:.' |f

Manuscripts wiill be reviewed ■ .for' possible publication svith the undeAfaoding that they' '■■
have not been published. tfpjlB accepted for publication

elsewhere. This does not preclude consideration of a complete report that i dhows publication
of preliminary .findings elsewhere. Copies of any possmly dh-pbeative jri:4tena| should be
submitted H qlh Ii! ;:i! j!i!| [!l!i lihini :l! iilliilinjll!!! I:; i:i!i li:!:! lili! i!;!ji:!!:li:i!!us:ij

PAGE CHARGES

er acceptance of. a. paper by die Mti scum., . ihe a ut hp:r{s) wiil , rpcea vg; ;:a ii||itlph;t of page-
charges. front i he M anaging Sdiidr ; cbfcufeteb at a current pj&p Partial M comp 'etc pavrn.cn t
of this statement will be solicited bom those authors who hate furu s available for this
purpose. Authors wifhgpt for partial or j epp||jif .payment of page |bhtgp

should so indicate on tire statement and return it to the MpApging- Editor,- : Tfepric ned
statement pp§§;i|it ; g||:ri • Of

publication ' isle, in ho. jffaiy *

iili

mm

bIWbmW

1:1 UjiltM

iMWlli

Kil

' in ' . ‘in, ..b i« i * •' 1 1 ’i i

.. O' :V:

li'IiW'i

„ jpHH

llliiilBiwtibiijii
it"!!" sjiii::

( . I

to *■. tz w — <n ~ to

avaan libraries Smithsonian institution NouniiiSNi nvinqshiiws saiavaan libraries smithsonI

1 ^ 5 1 J I ,, A ^

^4

CO

</}

sJrXAP^y t " t

> Vloiiis^X 5 ^ 2

CO * 2 to * 2 CO 2 CO

TITUTION NOlinillSNI NVIN0SH1IIMS S3iava8l1 LIBRARIES SMITHSONIAN INSTITUTION NOlinillSNI NVINOSHil
co — co ~ to — _co

a

CO

X

CO

o

_J 2 _ _ _ __

avaan libraries SMiTHsoNiAN^iNSTiTUTiON^NouniiiSNi^NViNOSHiiiNS^sa i avaa n libraries smithson
2 ro 2 5 1% 2 Z o

x\ 3 *■ fC^7 ^

(/)

m ^ x^vas^x m

Tl TUTIQN^NOlinillSN I _ NVINOSHil WIS S3IHVaail~UBRARSES SMITHSONIAN'lNSTITUTION^NOliniliSNI ” NVIN0SH1
co 2 w 2 w 2 co 2

X O X M" Wk Q m X ^o\ o x O

CO

CO

co

x vs^ -J±7 o

t— J 2

> S XftosHX > 2

avaan librari es^smithsonian institution ^NouniiiSNi nvinoshiiws^sb i a va a ii^li b rar i es^smithson

5 CO ... — co -* \ CO “ CO

IVAS^ o ' 5 “ XjVASt^ O " y “ X^tVAStC^ 5 2 ■

-J 2 _

TITUTION NOlinillSNI NVIN0SH1I/JS S3 I avaail LIBRARIES SMITHSONIAN INSTITUTION NOlinillSNI NVINOSHli

2 r- z r- 2 r* _ 2

,<cvcsoa^x O ~ „

CO ,.mTS — co /S^V«*WV^ .— y * V 03

33

'7/ 33 — Vs V, , . S3 Si 33 Z 30

' yojas!ix m ^ m ocJX w m ^ pi

avaan u b rar i Est°SMiTHS0NiAN“iNSTiTUTi0N NouniiiSNi nvinoshiiiais S3 1 ava a n~U B RAR I ES^SMITHSON

2 .,. CO 2 Vi co 2 ^ co 2 .... co

< vX\ . 2 < x£S\. 2 .< S . <

< 33 /y'.i rvXifVVV ~ :

X
= 1 CO

^111^ t i _

> N^oIjr^X 5 ^ 2 ^SiLeS^ > 5

CO * 2 to * 2 CO 2 CO *.2 ‘

TITUTION NOlinillSNI NVINQSH1IWS S3iaVaai*l LIBRARIES SMITHSONIAN INSTITUTION NOlinillSNI NVIN0SH1

co — <o = co ‘-T ^ co

^ UJ /<5»5Sv t th w

O dc^2 Z "s^cgcSX q

avaan li b rar i es“smithsonian'jinstitution NoiinniSNi^NViNOSHiiiNS saiavaan^LiBRARi esz smithson

2 r- 2 r; . 2 r~ 2

ss. t ®''»\ S «"i(IWX 1-

.p

^VAs^y> m Xi^osv^X m xnvosv^x ^ n^vask^ m '■ ^ x^jvas^x' rn

CO ? CO _ CO — CO

TITUTION NOlinillSNI NVIN0SH1IWS S3iava9l1 LIBRARIES SMITHSONIAN INSTITUTION NOlinillSNI NVIN0SH1

CO 2 CO 2 „ CO 2 V to 2

> i " | i I -"

2 CO 2 CO z CO ■" 2 CO

avaan_LIBRARIES SMITHSONIAN^INSTITUTION N0linillSNI_NVIN0SHlltMSwS3iaVaan_UBRARIES SMITHSON

h /Sffl J0fKA H \l^vvv^ Z /HinWA hi /Stf ittKv z

X
CO

^ §

— iv m r

o s<i*oi*£y _

2

IICTITI ITiriM MnimillCMI MWIKIDQU I

0 9®tI

:$S?0X m XgTgcJX ^ m Xondc^X £ X^SS^X m X» ^ x^ius^/ m

— to _ to _ to X 5 to

TITUTION NOIlfllllSNI NVINOSHXItAIS S3iava8ll LIBRARIES SMITHSONIAN INSTITUTION NOIlflillSNI NVIN0SH1

to 2 . to z ...•■ to 2 ... to z

2 AWWA -h

to
O

> S '2X2£ — > "‘W*®- 5 '' > x^Qiiisgx 2 '\ >

k 2 CO 2 CO * 2 CO k 2 CO

avaan libraries Smithsonian institution NouniusNi nvinoshiiiais saiavaan libraries smithson

5 „ w 2 ^ ^ to 5 CO - CO

to ^ , „ .« to ^ to ,kmk\ ^ to

^Ocits^X O X^v_dC^/ __ '<07*' O

TITUTION NOlinillSNl'VlVINOSHlIlAIS S3iaVdOI1 LIBRARIES^ SMITHSONIAN”1 INSTITUTION NOIlfllllSNI ^NVINOSHl

I § xV.. I I I i I 5

!>

30

03

33 Xx%i§L

” ^

m 'W 52 x*as*sx m m xovo^ix ^ rn

to _ co ± co ± to

avaan libraries Smithsonian institution noikuiisni nvinoshiiiais saiavaan libraries smithson

W 2 v CO 2 CO 2 CO

X MP~'Wk o ’ :m X /£i

to to v sfcSSg. CO si

.§ Jyp! s ^ gt g %wt

> 5* ^ X >• 2 ' P > xwi>

riTUTION OTN0linilISNI = NVIN0SHi!lNSOTS3 1 dVa 8 ll\| BRAR I ES^SMITHSONIAN INSTITUTION ^NQIi ISNl^NVSNQSMI

CO -2 \ to ___ X CO ” to

to

o

2

avaan libraries Smithsonian institution NouniusNi nvinoshiiiais saiavaan libraries smithson

2 8” 2 F” . 2 r* 2

m Ng.ocx 2 rn g xioia.s'^y tn

TITUTiON NOIiniliSNS°”NVINOSHllWS S3 I a VH 8 8 1 L 1 B RAR I ES^SMITHSONlArONSTlTUTION W NOlimilSN) NVIN0SH1
m 2 Z .*<■.• <2 2 ,v-. co z

> i^r S > w' s ^ > xffiacx s ^ >

aVaen^LJBRAR! ESWSMITHS0NIAN INSTITUTION NQIlfUllSNI NVIN0SHII!ASs“s3 I HV8 8 l*|Z LI B RAR I ES^SMITHSON

c/> 5 - w = ,.„.«/» 5 co

K w . _4 Jo “ >^^X. S iKSx w fk w

O xy oc^x ~ xyujisr^ q ''X ^ O

TITUTION N0SiniliSNI_NVIN0SHillAIS2S3 I UVU 8 II^LI B RAR I ES^SMITHSONIAN^INSTITUTION NOlinillSNI^NVINOSHI
Z C * ? r* 2 r; . 2 “

1 >s*

m ^ frg *2 *" m X^osxj NX ^ rm

CO — co 2 to £ to

avaan libraries Smithsonian institution NoiinniSNi nvinoshiiiais saiavaan libraries smithson

■“ 52 , 2» CO 2 C/> . Z •, w

^ 1 ^ ^ g

■Sr > '<S£SS>' 2 s N^5IE^ !" s

niUTION-’ N0110111SN I NVINOSHIHAIS^ S3 I d VH 0 ll^L! B RAR S ES^SMITHSONIAN^INSTITUTION ^NOimillSNI NVINOSHi
_ to 5 \ 5 _ to - ^ _ <o 5 \

w w -

CO

o

2

avaan libraries Smithsonian jnstitution NouniusNi nvinoshiiiais s^iavaan libraries smithson