Full text of "University of Kansas publications, Museum of Natural History"

See other formats

u/v/

HARVARD UNIVERSITY

LIBRARY

OF THE

Museum of Comparative Zoology

«.i V ^il

The Library •^
Uuseuffl of Cofflparatlve Zoology
Harvard University,

University of Kansas Publications
museum of natural history

VOLUME 20 • 1970-1971

EDITORS

Frank B. Cross, Chairman

William E. Dxiellmax

Philip S. Humphrey

Museum of Natural History

UNIVERSITY OF KANSAS

LAWRENCE

1972

S - N^ - L L^tor^^e^j

Museum of Natural History

unr'ersity of kansas

lawrence

PRINTED BY

UXrV'ERSlTY OF KANSAS

PRINTING SERVICE

LAWRENCE

CONTENTS OF VOLUME 20

1. A taxonomic revision of the leptodaetylid frog genus Syr-
rhophiis Cope. By John D. Lynch. Pp. 1-45, 22 figs. February
20, 1970.

2. A revision of colubrid snakes of the subfamily Homalopsinae.
By Ko Ko Gyi. Pp. 47-223, 38 figs. December 23, 1970.

3. A distributional study of Missouri fishes. By William L. Pflie-
ger. Pp. 225-570, 15 figs., 193 maps. February 16, 1971.

Index, pp. 571-578.

TT ir ^'''^^- ^OMP. ZOOU

University of K a n s a Sli^p^a oy

Museum of Natural History

^v^AR /» I97Q

Vol. 20, No. 1, pp. 1-45, 22 figs. HARVARD
February 20, 1970 UHJVfiWlbiTV

A Taxonomic Revision

of the Leptodactylid Frog Genus

Syrrhophus Cope

BY
JOHN D. LYNCH

University of Kansas

Lawrence

1970

University of Kansas Publications, Museum of Nati^al History

Editors of this number:
Frank B. Cross, Philip S. Humphrey, William E. Duellman

Volume 20, No. 1, pp. 1-45, 22 figs.
Published February 20, 1970

University of Kansas
Lawrence, Kansas

printed by

the university of KANSAS PRINTING SERVICE

LAVi^RENCE, KANSAS

1970

A Taxonomic Revision of the Leptodactylid Frog
Genus Syrrhophus Cope

BY
JOHN D. LYNCH

Introduction

Cope (1878) proposed the genus Syrrliophus for a medium-
sized leptodactylid frog from central Texas; in the ensuing 75 years
the genus was expanded to include a heterogeneous group of frogs
ranging from Texas to Peru. Taylor (1952) and Firschein (1954)
limited the genus to several species of frogs occurring in Guatemala,
Mexico, and Texas. Lynch (1968) provided a definition of the pre-
viously loosely-defined genus.

With the exception of Taylor (1952), who treated the Costa
Rican species, none of these authors dealt with the present status of
the nineteen species erroneously assigned to Sijrrhophus. These
species are listed in Tables 1 and 2 with the name currently applied.
Some of them are new combinations and their justifications will be
published elsewhere. Gorham (1966) is the most recent author to
include South American species in the genus Syrrliophus.

Smith and Taylor (1948) recognized two species groups of the
genus in Mexico, an eastern and a western group (here termed com-
plexes for purposes of discussion), separated on the basis of the
number of palmar (metacarpal) tubercles (three palmar tubercles
in the members of the eastern complex and two in those of the
western complex ) . Duellman ( 1958 ) reviewed the species of the
genus occurring in western Mexico and concluded that there were
five species (two polytypic). Dixon and Webb (1966) described
an additional species from Jalisco, Mexico. The distributions of
some species have been extended, but otherwse the western complex
of species remains unchanged since Duellman's review.

Smith and Taylor (1948) recognized seven species of the genus
in eastern Mexico. Firschein revised the eastern complex (as then
understood), and in so doing added one new species and treated
Syrrhophus verruculatus as a nomen duhium. Dixon (1957) rede-
fined the related genus Tomodactyhis and transferred T. macro-
tympanum Taylor to the genus Syrrhophus. Neill (1965) described
a new subspecies of S. leprus from British Honduras. Two species
(S. gaigeae and S. marnockii) were recognized in Texas until Mil-
stead, Mecham, and McClintock (1950) synonymized S. gaigeae

(3)

4 University of Kansas Publs., Mus. Nat. Hist.

Table 1. — Species Described as Members of the Genus Syrrhophus but Now

Placed in Other Genera.

Trivial name and author

Current combination

areolatus Boulenger, 1898
calcaratus Andersson, 1945
canjuphijllaceus Barbour, 1928
cocruleus Andersson, 1945
ineptus Barbour, 1928
juninensis Shreve, 1938
luto.sus Barbour and Dunn, 1921
molinoi Barbour, 1928
inontium Shreve, 1938
mijstaceus Barbour, 1922
ohcsus Barbour, 1928
omiltemanus Guntlrer, 1900
pardalis Barbour, 1928

Eleittherodactylus areolatus
Eleutherodactylus anderssoni
Eleiitherodactylus caryophyllaceus
Eleutherodactylus coeruleus
Eleutherodactylus diastema
Eupsoplius junineiisis
Eleutherodactylus lutosus
Eleutherodactylus molinoi
Niceforonia montia
Eleutherodactylus rhodopis
Eleutherodactylus punctariolus
Eleutlierodactylus omiltemanus^
Eleutherodactylus pardalis

' New combination.

Table 2 — Species Incorrectly Regarded as Members of tlie Genus Syrrhophus
but Described as Members of Other Genera.

Trivial name, original generic
assignment, and author

Current combination

chalceus (Phyllobates) Peters, 1873
fesfae (Paludicola) Peracca, 1904
hylaeformis (Phyllobates) Cope, 1875
palmatus (Phyllobates) Werner, 1899
ridens (Phyllobates) Cope, 1866
simonsii (Paludicola) Boulenger, 1900

Eleutherodactylus chalceus
Niceforonia festae
Eleutherodactylus htjlaeformis
Colostcthus palmatus
Eleutherodactylus ridens
Niceforonia simonsii

Table 3 — Nominal Species of Syrrhophus (sensu strictu) and the Name Used

Herein.

Original combination

Current combination

campi, Syrrhophus
cholorum, Syrrhophus lepriis
cystigathoides, Ph yllobates
dennisi, Syrrhophus
gaigeae, Syrrho))lnis
gutiilatus, Malacliylodes
interorbitalis, Syrrhophus
latodactylus, Syrrhophus
leprus, Syrrhophus
longipes, Batrachyla
macrotympanum , Tomodactylus
marnockii, Syrrhophus
modcstus, Syrrhophus
nehulosus, Syrrhoph us
uiiocolimae, Syrrhophus
))alUdus, Syrrhophus modestus
pctrophilus, Syrrhophus
pipilans, Syrrhophus
rubrintacidatus, Syrrhophus
S77iithi, Syrrhophus
teretistes, Syrrhophus
verrticipes, Syrrhophus
verruculatus, Phyllobates

cystignathoides campi

leprus

cystignathoides cystignathoides

dennisi new species

guttilatus

interorbitalis

longipes

leprus

longipes

verrucipes

marnockii

modestus

pipilans nehulosus

nivocolimae

pallidus

guttilatus

pipilans pipilans

rubrimaculatus

guttilatus

teretistes

verrucives

Nomen duhium

Leptodactylid Frog Genus Syrrhophus 5

with S. Duinwckii. Thus, at present, nine species (one polytypic) are
recognized on the eastern slopes and lowlands from central Texas
to British Honduras. These are currently placed on one species
group equi\'alent to the western complex reviewed by Duellman
(1958).

In the course of preparing an account of the species of Eleu-
tJierodoctyhis occurring in Mexico and northern Central America,
it became necessary to reexamine the status of the genus Syrrhophus
and its nominal species. It soon became evident that there were
more names than species, that some previously regarded species
were geographic \ ariants, and that the eastern and western groups
(complexes here) were artificial divisions of the genus. I conclude
that there are seven species (one poltypic) of Syrrhophus in eastern
Mexico, Texas, and El Peten of Guatemala, and seven species (one
polytypic) in western Mexico. The current status of each of the
23 names correctly assigned to the genus is presented in Table 3.

The fourteen species recognized by me are placed in five species
groups. Two of these groups are presently placed in the western
complex (modeshis and pipilans groups) and three in the eastern
complex (leprus, longipes and mornockii groups). The two com-
plexes do not correspond exactly with the eastern and western
groups of Smith and Taylor (1948), Firschein (1954), and Duell-
man (1958) since S. ruhrimacuJatus is now associated with the
eastern leprus group.

The definitions and contents of the five species groups are as
follows:

leprus group: digital pads not or only slightly expanded, rounded in outline;
first finger longer or shorter than second; snout acuminate or subacuminate,
not rounded; outer metatarsal tubercle conical; digits lacking distinct lateral
fringes,
content: cystipnathoides, leprus and ruhrimactilatus.

longipes group: digital pads widely expanded, triangular in outline; first
finger shorter than second; snout acuminate; outer metatarsal tubercle not
conical; digits bearing lateral fringes,
content: dennisi and longipes.

martiockii group: digital pads expanded, rounded to truncate in outline; first
finger equal in length to second or slightly shorter; snout roimded; outer
metatarsal tubercle not conical; digits lacking lateral fringes; generally stout-
bodied frogs,
content: guttilatus, maruockii, and verrucipes.

modestus group: digital pads expanded, trimcate in outline; first and second
fingers subequal in length, first usually slightly shorter than second; snout
subacuminate; inner metatarsal tubercle twice as large (or larger) as outer
metatarsal tuliercle; digits bearing poorly-defined lateral fringes,
content: interorhitalis, modestus, nivocolimae, pallidus, and teretistes.

6 University of Kansas Publs., Mus. Nat. Hist.

pipilans group: digital pads not or only slightly expanded, truncate in outline;
first finger equal in length to second; snout subacuniinate; metatarsal
tubercles subequal in size; digits lacking lateral fringes,
content: pipilans.

AckuoicJedgments. — For loan of specimens, I am indebted to Richard T.
Baldauf, Texas A & M Universit>' (TCWC); W. Frank Blair, University of
Texas (TNHC); Charles M. Bogert and Richard G. Zweifel, American Museum
of Natural History (AMNH); James E. Bohlke and Edmond V. Malnate,
Academy of Natural Sciences of Philadelphia (ANSP); Robert F. Inger and
Hymen Marx, Field Museum of Natural History (FMNH); Ernest A. Liner
(EAL); Michael Ovchynnyk, Michigan State University collection (MSU);
James A. Peters, United States National Museum (USNM); Douglas A. Ross-
man, Louisiana State LT^niversity Musevnn of Zoology (LSUMZ); Hobart M.
Smith, University of Illinois Museum of Natinal History ( UIMNH ) ; Charles F.
Walker, Universit>' of Michigan Museum of Zoolog>' (UMMZ); and John W.
Wright, Los Angeles Count>' Museum ( LACM ) . Specimens in the collection at
the University of Kansas Aluseum of Natural History are identified as KU.
The abbre\iations EHT-HMS refer to the Edward H. Taylor-Hobart M. Smith
collection and FAS to the Frederick A. Shannon collection. The tspe-specimens
from these collections are now in the Field Museum of Natural History and the
Uni\'ersity of Illinois Museum of Natural History.

I have profited from discussions concerning this problem with several per-
sons, most notably William E. Duellman, Hobart M. Smith, Edward H. Taylor
and Charles F. Walker. Nevertheless, the ideas and conclusions presented here
should not be construed as necessarily reflecting their opinions.

Da\id M. Dennis executed all of the figures, and my wife, Marsha, typed
the manuscript.

Materials and Methods. — In the course of this study, 1003 speci-
mens of the genus were examined. The holotypes of 21 of the 23
nominal species are extant; I have examined 19 of these. Nine
measurements were taken, and five ratios computed for each of 338
specimens. Females are available for all species but one; thus,
measurements were taken on individuals of both sexes.

Analysis of Characters

Size and proportions. — Frogs of this genus range in size from 16
to 40 mm. in snout- vent length. Five species are relatively small:
S. ctjstignathoides, modestus, nivocolimae, palUdus and rubrimacula-
ttis; one, S. longipes, is relatively large, and the remaining eight
species are intermediate in size (22-30 mm.).

Males are generally smaller than females and have proportionate-
ly longer heads and usually larger tympani. No significant diflFer-
ences were found among proportions, except that S. longipes has
a larger tympanum/eye ratio than any other species. Frogs in the
Syrrhophus marnockii group tend to have shorter shanks and feet,
thereby giving those species a more stocky appearance. However,
the differences are not significant.

A summary of the data on size and proportions for the frogs of
the genus Syrrhophus is given in Tables 4, 5, and 6.

Leptodactylid Frog Genus Syrriiophus 7

Hands and Feet. — Taylor and Smith (1945), Smith and Taylor
(1948), Firschein (1954) and Duellman (1958) discussed the value
of the palmar tubercles in identifying frogs of this genus. The
eastern complex in general has a well-developed outer palmar
tubercle (Fig. 1) in distinction to the western complex in which
the outer palmar tubercle is reduced or absent (Fig. 2). Dixon and
Webb ( 1966 ) imply that the outer palmar tubercle is rarely absent
but is usually smaller than the first supernumerary tubercle of the
fourth finger. My study of the western species demonstrates that
the outer palmar tubercle is indeed usually present and smaller than
the first supernumerary tubercle.

Differences in interpretation of the terms "unexpanded" and
"narrow," as well as differences in techniques of preser\'ation, have
led to confusion of the reported digital shapes in various species.
Constant specific differences are evident in the hands (Fig. 1).
Except in the cases of excessive uptake of fluids, all species have a
terminal transverse groove at the tip of each digit. Taylor ( 1940b )
stated that S. sfiiitlii lacked grooves, but examination of the holotype
reveals faint grooves at the tops of the digits. Syrrhophus giitfdatus,
leprus, pipdans, and verrucipes lack lateral fringes on the fingers.
Lateral fringes are well developed in the longipes and modestus
groups but poorly defined or absent in the other members of the
genus. The digital pads of the frogs of the longipes group are much
broader than those of the other species and are narrowest in the
frogs of the leprus group. Supernumerary tubercles are present on
the palmar surfaces of all species of the genus.

Table 4 — Size and Proportions in the Frogs of the Syrrhophus leprus Group.

Snout-

vent

Tibia

Head

Tym-

Evelid/

length

length/

width/

panum/

Inter-

Species

Sex N

(SVL)

SVL

Eye

orliital

cijstignathoides

16.3-23.5

41.3-49.6

34.0-40.1

43.7-66.5

43.2-89.6

campi

(4,5.8)

(37.0)

(,56.2)

(61.5)

16.0-25.8

41. ,5-5 1.0

.33.0-,38.0

42.8-60.0

48.2-69.2

(4,5.8)

(,35.0)

(51.2)

(60.1)

c. ctfstignathoides

16.8-22.1

45.1-,50.4

,33 2-40.7

44.3-68.7

44.6-65.4

(47.3)

(,37.8)

(,54.8)

(60.0)

19.6-24.2

46.4-,50.0

,34.1 -,38.1

43.,3-.56.5

,53 2-65.4

(47.6)

(,36.2)

(46.9)

(,59 2)

leprus

20.6-26.4

42.,3-.52.3

35.0-40.3

47.,5-62.5

,58.2-72.5

(46.8)

(37.4)

(56.5)

(67.3)

22.1-29.2

43.4-,53.3

,32.6-38.9

38.6-57.9

,50.2-86.9

(47.1)

(,35.8)

(47.1)

(68.1)

ruhrimaculatus

18.2-23.5

40.4-46.2

31.8-35.5

35.,5-46.5

65.1-78.5

(43.4)

(33.8)

(41.7)

(71.7)

University of Kansas Publs., Mus. Nat. Hist.

Table 5 — Size and Proportions in the Frogs of the Synliophtis longipes and

S. marnockii Groups.

Snout-

vent

Tibia

Head

Tym-

Eyelid/

length

length/

M'idth/

panum/

Inter-

Species

Sex

(SVL)

SVL

Eye

orbital

dennisi

22.8-28.4

43.9-49.7

35..3-41.2

53.9-64.2

55.3-74.0

(47.4)

(38.8)

(58.9)

(6.5.1)

25.9-32.0

46.3-50.8

35.6-40.3

50.6-.58.7

58.1-70.9

(48.2)

(37.7)

(.54.9)

(63.6)

longipes

22.1-33.2

45.8-51.7

38.7-44.4

61.1-87.2

61.5-83.0

(48.4)

(41.8)

(72.0)

26.8-39.6

44.3-51.0

36.3-40.8

49..5-72.1

.55..3-85.9

(47.2)

(39.1)

( .59.5 )

(67.9)

gtittilatus

20.6-29.0

41.2-48.1

36.9-44.9

55.1-75.7

.53..3-79.5

(44.5)

(40.6)

(64.1)

(66.0)

2.5.7-31.0

41.4-46.8

35.9-42.3

47.6-61.7

62..3-79.8

(43.6)

(.38.5)

(54.0)

(72.9)

marnockii

18.4-28.9

42.3-47.2

36.1-43.0

47.2-68.3

51.6-74.4

(44.1)

(39.6)

(61.2)

(663)

20.4-35.4

38.7-46.4

.35.9-41.3

45.8-73.3

52.1-70.5

(42.7)

(.38.2)

(60.3)

(60.7)

verrucipes

17.5-29.2

42.7-49.5

.36.2-42.4

.56.1-82.2

,56.8-82.8

(46.3)

(39.1)

(67.8)

(70.4)

26.5-31.7

42.4-47.7

36.0-.38.1

458-57.8

61.0-77.9

(44.6)

(37.0)

(53.9)

(69.0)

Table 6 — Size and Proportions in the Frogs of the Stjrrhophus pipilans and

S. modcstus Groups.

Snout-

vent

Tiliia

Head

Tym-

Eyelid/

length

length/

width/

panum/

Inter-

Species

Sex

(SVL)

SVL

Eye

orbital

pipilans

22.9-28.5

38.1-42.0

34.4-37.2

36.6-47.8

56.1-82.4

nchtdosus

(40.0)

(.3.5.4)

(43.6)

(682)

21.1-22.7

42.1-44.5

33.2-35.8

.36 6-47.6

64.3-65.4

pipilans pipilans

22.6-27.8

37.9-44.0

322-36.5

.38,0-54.0

56.1-79.5

(41.4)

(33.0)

(46.2)

(67.3)

29.4

.38.4

.32.5

44 6

55.0

modestus

15.8-20.1

38.5-42.6

.32.1-38.1

26 8-39 3

.57 0-86.9

(40.6)

(.34.2)

(31.5)

(69.1)

18.5

44.2

.36.0

24.0

52.1

pallidus

17.9-19.3

41.0-44.9

326-36.2

27.0-35.6

.59.4-67.7

(43.4)

(35.2)

(30.9)

(65.2)

tcretistes

19.2-23.2

41.5-45.3

32.5-36.4

28 6-438

51.2-75.0

(43.7)

(.34.0)

(.33.7)

(62.2)

24.8

41.8

.30.8

.37.9

60.5

nivocolimae

18.9-21.1

42.2-48.6

30 9-37.1

.30 0-.39.3

42.6-69.1

(45.0)

(.33.7)

(34.7)

(55.0)

24.1

40.9

.33.5

27.6

.56.5

intcrorhifalis

25.6

43.0

39.4

57.6

20.2-26.7

.39.9-47.1

32.6-39.3

29.1-41.2

58.2-76.9

(43.2)

(35.8)

(36.4)

(69.2)

Leptodactylid Frog Genus Syrrhophus

\ -

Fig. 1: Palmar views of hands of six species of the eastern complex of Syrr-
hophus. (A) verrucipes (UIMXH 15995), (B) ruhrimacidatiis (KU 58911),
(C) dennhi sp. nov. (holotype, UMMZ 101121), (D) guttilatiis (UIMNH
55520), (E) marnockii (TCWC 4782), and (F) longipes (TCWC 12179).

All x6.5.

University of Kansas Publs., Mus. Nat. Hist.

> ">

-\'

Fig. 2: Palmar views of hands of two species of the western complex of

SijrrhopJnis. pipilans (left, KU 58908, x6) and teretistes (center, KU 75269,

and right, KU 75263, respectively, X9).

In S. cystignathoides and lepnis, the first finger is longer than the
second, and the first two fingers are equal in length in guttiJatus and
marnockii. In the other species the first finger is shorter than the
second.

Supernumerary tubercles are weU de\'eloped on the plantar
surfaces in all species, except S. guttiJatus, in which they are poorly
defined (Fig. 3). The relative sizes of the metatarsal tubercles has
been used in the classification of the species and species groups of
Sijrrhophus. The metatarsal tubercles are similar in all species of
the eastern complex (including ruhrimaculatus) ; the outer tubercle
is always about one-half the size of the ovoid inner metatarsal
tubercle. In the leprus group the outer tubercle is conical and com-
pressed. The metatarsal tubercles of pipilans are about the same
size, or the outer is slightly smaller than the inner. In the modestus
group the outer metatarsal tubercle is about one-third the size of
the inner.

All species, except guttiJatus, have well-defined to poorly defined
lateral fringes on the toes. All species have expanded toe pads. The
fifth toe is usually shorter than the third, but the second is equal in
length to the fifth in some specimens of S. cystignatJioides and S.
marnocJxii. SyrrJwpJuis nivocoJimae is the only species with tubercles
along the outer edge of the tarsus; this is merely a reflection of the
highly tuberculate nature of the skin in this species.

SJcin texture. — The skin of the dorsum is smooth or very weakly
pustular in all species of the genus except nivocoJimae and verru-
cipes. The dorsal surfaces of nivocoJimae are warty; in verrucipes
the skin is pustular. The skin of the venter is areolate in cystignatJi-

Leptodactylid Frog Genus Syrrhophus

oides cystignothoides, dennisi and verrttcipes but is smooth in all
other species of the genus.

Color pattern. — As is evident in the diagnoses, the color patterns
of given populations ha\'e been regarded as useful in separating the

Fig. 3: Plantar views of feet of four species of the eastern complex of St/rr-
hophus. (A) guttilatus (UIMNH 55519, X6), (B) leprus (UIMNH 42726,
X6), (C) verrucipes (UIMNH 15995, x6), and (D) longipes (TCWC 12179,

X4.6).

12 University of Kansas Publs., Mus. Nat. Hist.

species and subspecies. Duellman ( 1958 ) suggested that the colora-
tion, with the exception of modestus, was a dark ground color with
pale markings. It is a moot point whether the frogs have light spots
on a dark background or have a light background with an extensive
reticulate dark pattern. The venters are gray or white, and the
vocal sac is nearly black in some species. Interorbital dark bars or
triangles are absent in only two species of the eastern complex,
cijstignathoides campi and marnockii; the latter lacks a supratym-
panic stripe, which is present in the other members of the eastern
complex. Sijrrhophus interorbitalis and nivocoJimae have light in-
terorbtal bars; these bars occur in only one other population of the
genus (S. c. cystignatJioides). Bars on the thighs are ill defined or
absent in the members of the marnockii and part of the modestus
groups. The color in life is noted in the species accounts.

Voice. — The voices of all Sijrrhophus can be described as a
single short chirp or peep; without audiospectrographic analyses
the significance of the differences between a chirp, peep, or short
whistle cannot be appreciated. Martin (1958) and Wright and
Wright (1949) reported multi-noted calls, and one collector of S.
verrucipes noted the frog "trilled."

Fouquette (1960) presented analyses of two species (marnockii
and pipilans nehulosus). The voices were very similar; both frogs
were reported to "trill" and "chiip."

Systematic Account

The genus Sijrrhophus has been defined (Lynch, 1968) and
limited to the group of species occurring in Guatemala, Mexico and
the United States. The closest relatives of Sijrrhophus are the frogs
of the genus Tomodactyhis (Dixon, 1957; Firschein, 1954). Lynch
(1968) implied there were no osteological bases for the separation
of Eleiitlwrodactijhis, SyrrlwpJiiis, and Tomodactyhis. At that time,
I believed such to be the case and derived SyrrhopJius and Tomo-
dactyhis from the rhodopis complex of EleutJierodactijhis, with
which they share terrestrial habits and relatively short limbs. In the
rhodopis complex there is a tendency for the loss of the outer palmar
tubercle, a not uncommon condition in Syrrhophus and Tomodacty-
his.

However, the skulls of Syrrhophus and Tomodactyhis show de-
partures from the pattern observed in the Middle American Eleii-
therodactyhis, as well as many of those species in western South
America. Baldauf and Tanzer (1965) reported that the fronto-

Leptodactylid Froc Genus Syrriiophus 13

parietals and prootics were fused in Syrrhophus marnockii and that
the prootics and exoccipitals appeared to be one bone ( otoccipital ) .
The otoccipital is not uncommon in cleutherodactyline frogs, but
the fusion of the frontoparietals with the prootics ( regardless of the
fusion of the latter with the exoccipital ) is uncommon in the family.
I ha\'e found the frontoparietal-prootic fusion only in Syrrlwphus
(all species), Tomodactyhis (all species), and Eleuthewdactylus
(West Indies species). None of the Middle American Eleuthew-
dactylus has the two bones fused. Examination of the character is
difficult in dried skeletal preparations. Cleared and stained or
macerated preparations are satisfactory for checking this character.
Thus, in addition to the presence of numerous plantar super-
numerary tubercles in the frogs of the genera Syrrhophus and Tomo-
dactylus, these two genera can be separated from other Middle
American eleutherodactylines by the fusion of the frontoparietals
and prootics. This character not only further strengthens the argu-
ment that the two genera are closely related but poses a problem
of zoogeographic analysis of the distribution of the character, which
will be discussed fully elsewhere.

Key to the Species of the Frog Genus Syrrhophus

1. Three large, well-developed palmar tubercles 2

Two large palmar tubercles; outer (third) palmar tubercle reduced

in size or absent 9

2. Digital pads more than twice (usually three or more) times \\'idth of

digit 3

Digital pads less than tvvice width of digit 4

3. Males having vocal slits; dorsum vermiculate; diameter of tympanum

in males about one-half diameter of eye S. dennisi

Males lacking vocal slits; dorsum flecked, spotted, or blotched; diam-
eter of t>mpanum in male about three-fourths that of eye S. longipes

4. First finger longer than second 5

First finger shorter than or equal to second '

5. \^enter smooth; dorsum spotted or vermiculate - S. leprtis

Venter areolate, or if smooth, dorsum flecked and interorbital bar
lacking ^

6. \^enter areolate; interorbital bar present; ground color yellowish

S. ajstipnathoides cijstignathoides

Venter smooth; interorbital bar absent; ground color brown

S. cijstignathoides campi

7. First finger shorter than second; digital tips only slightly dilated; green

in life with darker green spots S. verntcipes

First finger equal to second; digital tips slightly to moderately expanded 8

14 University of Kansas Publs., Mus. Nat. Hist.

8. Dorsum vermiculate; interorbital bar present; ground color cream to

brown in life S. giittilatus

Dorsum punctate or flecked; interorbital bar absent; ground color green

in life S. marnockii

9. Dorsum dark with pale (red in life) spots; digital pads not expanded

S. rubrimactdatus

Dorsum pale \\ith dark markings and digital pads slightly to widely
expanded 10

10. Digital tips not widely expanded; tympanum well-defined; outer meta-
tarsal tubercle more than one-half size of inner 1 1

Digital tips widely expanded, truncate in outline; tympanum poorly de-
fined; outer metatarsal tubercle less than one-half size of inner 12

11. Dorsum dark brown with large light spots or blotches; tympanum/eye

ratio usually greater than 43 percent S. pipilans pipilans

Dorsum dark brown with small light spots; t>'mpanum/eye ratio less
than 48 percent S. pipilans neiihlostis

12. Light interorbital bar present 13

Light interorbital bar absent 14

13. Adults small, less than 22 mm. snout-vent length with a broad mid-
dorsal stripe; dark bands on shank narrower than light interspaces

S. nivocolimae

Adults larger, more than 22 mm. snout-vent length; dorsum vermi-
culate; dark bands on shank broader than light interspaces

S. interorbitalis

14. Dorsum spotted \\'ith discrete black spots; pattern definite S. modestus

Dorsum reticulate or vermiculate, pattern poorly defined 15

15. Adults small, less than 21 mm. snout-vent length; upper arm not

banded S . paUidus

Adults larger, usually greater than 21 mm. snout-\ent length; upper
arm banded S. teretistes

Species Accounts

The following accounts do not include complete descriptions of
each taxon, because a more than adequate number of descriptions
is available in the recent (1940-1966) literature. An abbreviated
synonymy, in which are listed all combinations and emendations of
names and significant contributions to our knowledge of the taxon,
is given for each. For each species and subspecies the following are
given: descripti\'e diagnosis, statement of range, remarks on taxon-
omy, fist of specimens examined, illustration of color pattern, and
distribution map.

Syrrhophus cystignathoides (Cope)

Phyllohates a/stigrmthoides Cope, 1877:89-90 [Syntypes.— Originally USNM
32402-32409, ( 32405 now in MCZ ) from Potrero, near Cordoba, Veracruz,
Mexico, Francis Sumichrast collector.]

Leptodactylid Frog Gf.nus Syrrhophus 15

Diagnosis. — Adults small, males 16.0 to 23.5 mm. in snout- vent length,
females 16.0-25.8 mm. in snout-vent length; vocal slits present in males; finger
tips slightK- expanded; first finger longer than second; outer metatarsal tubercle
one-half size of inner, conical, compressed; skin of dorsum weakly pustular,
that of venter smooth to areolate; tympanum 44 to 69 per cent diameter of eye
(mean 55.5 per cent); ground color >ello\\- to brown in life with brown to
black fleckings on dorsum and flanks; limbs banded; interorbital bar present
or not.

Remarks. — Two geographic races (subspecies) are herein recognized; pre-
viously these were held by various authors to be species ( campi and cijstignath-
oidcs). Intergradation occurs in southern Tamaulipas and eastern San Luis
Potosi, Mexico. The hvo subspecies can be distinguished on the basis of color
pattern and the condition of the skin of the venter.

Distribution. — Low to moderate elevations from the Rio Grande embay-
ment to central Veracruz, Mexico (Fig. 5).

Syrrhophus cystignathoides campi Stejneger, New combination

Syrrhophus campi Stejneger, 1915:131-32. [Holotype.— USNM 52290, from

Brownsville, Cameron Co., Texas; R. D. Camp collector, March 31, 1915].

Smith and Taylor, 1948:52. Martin, 1958:50.

Diagnosis. — Venter smooth; usually no interorbital light and dark bars pres-
ent; ground color brown in life (Fig. 4a).

Remarks. — Martin ( 1958) \\'as the first author to point out that S. campi was
probably a subspecies of the more southern S. cystignathoides. Various ref-
erences in the literature might lead one to belie\e that the two were sympatric
over much of northeastern Mexico; this error was created by the use of a single
character (condition of the skin of the venter) to characterize the two popula-
tions. Specimens from southern Texas have a smooth venter, lack interorbital
bars and have, in general, a brown ground color, whereas specimens from
central Veracruz have an areolate venter, interorbital light and dark bars and
a yellow ground color. In southern Tamaulipas and eastern San Luis Potosi,
these characters \ary discordantly, thereby strongly suggesting that the two
populations intergrade. Both populations agree in other morphological char-
acters; therefore, they are here treated as geographic variants.

Etymology. — Named for the collector of the type specimens, Mr. R. D.
Camp of Browns\ille, Texas.

Distribution. — Lower Rio Grande embayment in Texas to central Nuevo
Leon and Tamaulipas, Mexico. Intergrades are known from southern Tamauli-
pas and adjacent San Luis Potosi, Mexico (Fig. 5).

Specimens examined.— (113) TEXAS, Cameron Co.: MCZ 10277-85, 10286
(10); Brownsville, AMXH 3215, 3218-20, 3221 (3), 5376, 62117, FMNH
105336, KU 8135-39, MCZ 3738-42, 3743 (10), TCWC 5908, 7139, TNHC
92-94, 20909, UMMZ 51760, 54031 (5), USXM 52290 (holot>'pe); 22 mi. SE
Brownsville, TNMC 14223; 8 mi. SW Brownsville, UMMZ 101127
(3); Harlingen, AMXH 62118, UMMZ 105200-205, 105206 (5), 105207 (4).
Hidalgo Co.: Bentsen-Rio Grande State Park, UMMZ 114378; 6 mi. S McAllen,
TXHC 7136-39; Santa Ana Refuge, TCWC 13495-96; Weslaco, TCWC 17658-
60.

MEXICO, Nuevo Leon: Salto Cola de Caballo, AMXH 57953-54, FMXH
30644-45, 37169-70; Monterrev, UIMXH 13324; 40 km. SE Monterrey,
UIMXH 3686. Tamaulipas: 80 km. Matamoros, FMNH 27150 (13).

University of Kansas Publs., Mus. Nat. Hist.

Fig. 4: Sijrrhophiis ctjstignothoides campi (left, TCWC 13490) and S. c. cysti-
gnathoides (right, KU 105500). Dorsal views X2, sides of heads X3.

Intergrades [S. c. ci/stignathoides X S. c. campi (88)] MEXICO, San Luis
Potosi: 5 km. E Cuidad del Maiz, UMMZ 106435; 16 km. W Naranjo, FMNH
104584; Salto de Agiia, 34 km. WSW Antigua Morelos, TCWC 6980. Tamauli-
pas: 5 km. W Acuna, 1060 m., UMMZ 101172, 101173 (16), 101174-76,
101177 (6); 14.5 km. NNW Chamal, 430 m., UMMZ 111337 (2); 20 km.
NNW Chamal, 700 m., UMMZ 111338 (11); 8 km. N Gomez Farias, 450 m.,
UMMZ 101165; 8 km. NE Gomez Farias, Pano Avuctle, UMMZ 102264,
102924 (6); 8 km. NW Gomez Farias, 1060 m., LSUMZ 11084, UMMZ 101199,
102928 (5), 102929-32. 110124 (3); Rio Guayala, near Magiscatzin, MCZ
24138-42, 85071-81, UMMZ 88242 (2); Magiscatzin, TCWC 6981; Las Yucas,
north of Aldama, MCZ 29665-68; 16 km. NE Zamorina, UMMZ 101124.

Syrrhophus cystignathoides cystignathoides (Cope),
New combination

Phyllobates cystignathoides Cope, 1877:89-90 [Syntypes.— USNM 32402-32409,
from Potrero, near Cordoba, Veracruz, Me.xico, collected by Francis Sumi-
chrast]. Boulenger, 1882:196.

Syrrhophus cystignathoides: Cope, 1879:268. Kellogg, 1932: 126-27. Taylor
and Smith, 1945: 582-83. Smith and Taylor, 1948:50. Martin, 1958:49.

Syrrhaplius cystignathoides: Giinther, 1900:218.

Syrraphus cystignathoides: Diaz de Leon, 1904:10.

Sijrrhopus cystignathoides: Barbour and Loveridge, 1946-170.

Lei'todactylid Frog Genus Syrrhophus

Diagnosis. — Venter areolate; interorlMtal light and dark Inirs present; ground
color yellow to brownish-xellow in life (Fig. 4h).

Remarks. — Firschein (1954) briefly considered the status of Peters' (1871)
PhijUobates veiruculatiis and noted that if it was a Sijrrliopliiis it would prob-
ably be referrable to S. cystignathoides. Peters' (1871) original description
corresponds well with S. cystignathoidcs, and the type-locality ("Huanusco"=
Huatusco ) is within the range of that species. Firschein ( 1954 ) expressed

100°

98°

96'

26°

24'

I. ~— '

100 200

<ILOMETERS

500

26'

24'

100'

98=

96°

Fig. 5: Distribution of Syrrhophus cystignathoides campi (solid symbols) and
the nominate subspecies ( open symbols ) .

18 University of Kansas Publs., Mus. Nat. Hist.

doulit that vernicidatiis was a Sijrrhophiis, because Peters placed it in another
genus. Ho\ve\er, Peters described verniculatus a decade before Cope diagnosed
the genus Sijnhophus. Most frogs now called Stjnhophus, plus a number of
lower Central American frogs now placed in a variety of genera were placed
in PliyUobates by Boulenger, Cope, and Peters.

The types of PJiyUohatcs verniculatus were destroyed dining World War II
(Giinther Peters, in litt.); the specimens subsequently assigned to the taxon
by Kellogg (1932) are Sijnhophus cijstignathoides. Because the type specimens
are lost and because the name antedates the more established name, cysti-
gnathoidcs, I favor retaining PhyUohates verniculatus Peters as a nomen duhium.

Smith and Taylor (1948) reported S. verniculatus from Tianguistengo,
Hidalgo, Mexico. These specimens are examples of vernicipes. Smith (1947)
reported a specimen of verniculatus from San Lorenzo, Veracruz. Firschein
(1954) referred it to cystignathoides, and Duellman (1960) concluded that
both authors were in error and that the specimen (USNM 123530) was a
leprus.

Etymology. — The trixial name is the diminutive of Cystignathus, a once-
used generic name for several leptodactylid frogs.

Distribution. — Low and moderate elevations in the foothills along the
Sierra Madre Oriental from eastern San Luis Potosi to Central Veracruz,
Mexico (Fig. 5).

Specimens examined.— (130), MEXICO, Puebla: Necaxa, UMMZ 69519-20.
San Luis Potosi: 5 km. W Aguismon, LSUMZ 4962-63; along Rio Axtla, road
to Xilitia, UMMZ 105500; Tamazunchale, UIMNH 3199; 6.5 km. N Tama-
zunchale, UMMZ 104039; 8 km. N Tamazunchale, UMMZ 119490. Veracruz:
Coatepec, 1210 m., FMNH 704966-67; 11 km. SE Coatepec, 850 m., FMNH
70468-70; below Cordoba, FMNH 104588, UIMNH 13321; Cuautlapam, 1000
m., FMNH 106477-80, KU 100364, UIMNH 58200-03, UMMZ 105392; Fortin
de las Flores, UIMNH 13322, 13339; 1.6 km. N Fortin de las Flores, UIMNH
42799-808, UMMZ 105389; 3.2 km. N Fortin de las Flores, UIMNH 26633-35;
4.8 km. N Fortin de las Flores, UIMNH 71967-68; 3.2 km. W Fortin de las
Flores (Barranca Metlac), 910 m., UIMNH 49294-95, UMMZ 115444-46,
118221, 119893 (2); Huatu.sco, KU 100363; jalapa, 1400 m., FMNH 70440,
70443-51, 70454-65; 16 km. NE Talapa, 1300 m., FMNH 70452-53; 8 km.
E Talapa, UIMNH 13338; 9.5 km. "S jalapa, UMMZ 122083 (2); Mirador, KU
23967; Paraja Nuevo, El Suchil, UMMZ 85490(7), 85491(2), 90315; La Passa,
UIMNH 49293, 49297; 1 km. E Plan del Rio, 240 m., UMMZ 102067 (2);
Potrero Viejo, FMNH 104583, 104586, 105326-27, KU 26789, 100357-62,
UIMNH 13323, 13340-43; USNM 32402 (lectotype), 32403-04, 32406-09; 9.6
km. S Santa Rosa, TCWC 12785; 24 km. NE Tezuitlan (Puebla), UMMZ
105388; Teocelo, FMNH 70437-38, KU 26080, 26790; 3.2 km. N Teocelo,
FMNH 704.39, 70441-42; 9.6 km. NW Tihuatlan, UIMNH 3684-85; 15 km.
ENE Tlacotepec, KU 23966; 26 km. NW Tuxpan, UMMZ 126419.

Syrrhophus leprus Cope

Syrrhophus leprus Cope, 1879:268-69 [Holotype.— USNM 10040, from Santa
Efigena, Oaxaca, Mexico, Francis Suniichrast collector]. Kellogg, 1932:124-5,
128. Taylor and Smith, 1945:582. Smith and Tavlor, 1948:50-51.
Duellman, 1958:8, pi. 1, Fig. 2; 1960:56-57. Gorham, 1966:165.

Syrrhaphus leprus: Giinther, 1900:217.
Syrrltophus leprus leprus: Neill, 1965:85-86.

Syrrhophus leprus cholorum Neill, 1965:85-86 [Holotype. — Wilfred T. Neill
collection 1525, from 3.9 mi. N San Antonio, Toledo District, British Hon-

Leptodactylid Fhoc Genus Syrriiophus

duras, collected October 28, 1959, by R. A. Allen, T. C. Allen, and W. T.
NeillJ.

Diagnosis. — Medium-sized frogs, males 20.5-26.5 mm. in snout-vent, fe-
males 22.0-29.3 mm. in snout-vent length; vocal slits present in males; tips of
fingers dilated slightly; first finger longer than second; inner metatarsal tubercle
twice size of small, conical outer metatarsal tubercle; skin of dorsum pustular,
that of venter smooth; snout sulxicuminate; diameter of tympanum 47.5-62.5
per cent of eye in males, 38.6-57.9 per cent in females; dorsum yellowish-green
with chocolate brown blotches or spots forming reticulations in most specimens;
venter white to gray; flanks brown, spotted with white or not; limbs banded;
interorl)ital liar obsciued by dorsal pattern.

Fig. 6: Dorsal views of SyrilwpJius Icprtis showing variation in dorsal pattern
(left, UMMZ 121244, x2; right, KU 26106, xl.7). Side of head (UIMNH

42726, x7).

University o¥ Kaxsas Publs., Mus. Nat. Hist.

Fig. 7: Distribution of three species of eastern complex SyrrJiopJitts: lepras
(circles), ruhrimaculatus (triangles), and vernicipes (squares).

Remarks. — M\' distribution map (Fig. 7) differs somewhat from that of
Duellman (1958), who was unaware of specimens reported by Taylor and
Smith (1945) from central Veracruz, Mexico.

Duellman (1958, 1960) regarded S. leprus as having a gray venter. Neill
(1965) characterized his new subspecies on the basis of \\'hite venter and
spots on the dorsum. Some specimens from throughout the range have only
small round spots, instead of \ermiculations (Fig. 6). The gray ventral colora-
tion is largely restricted to the population in Los Tuxtlas, Veracruz, but only
about 80 per cent of the specimens from the Los Tuxtlas ha\e gray venters,
whereas specimens from Guatemala, Oaxaca, Tabasco, and central Veracruz,
Mexico, have white \'enters (rarely gray). Since the specimens from British
Honduras are not distinct from specimens throughout most of the range, there
is no reason to recognize them as a subspecies.

Etymology. — Greek, lepra, leprosy, in reference to the mottled color pattern.

Distribution. — Discontinuous; central Veracruz to British Honduras to low
elevations in the foothills of the Sierra Madre Oriental, Los Tuxtlas, Sierra
Madre de Chiapas (Isthmus of Tehuantepec (Fig. 7) ).

Specimens examined. — (84). GUATEMALA, Alta Verapaz: Chinaja, KU
55961-62. EI Peten: 15 km. NW Chinaja, KU 55963; Piedras \egras, USNM
114085-92; Tikal, UMMZ 117035; Uaxacti'm, ANLXH 55121-22.

MEXICO, Oaxaca: Cerro San Pedro del Isthmo, UIMNH 35510; Finca La
Gloria, USXM 114093; 30.5 km. \ Matias Romero, UIMXH 39459, 71969;
Santa Efigenia, USNM 10040 (holotvpe). Tabasco: Teapa, UMMZ 113799-800;
13.5 km. W Teapa, UMMZ 120253. Veracruz: 27.5 km. N Acavucan, UIMNH
42726; Atovac, UINLNH 13331, 49296; 3.2 km. N Catemaco, UIMNH 71976-77;
Covame, UIMNH 3S995, 38998, 40342; Dos Amates, TCW'C 21211; Fortin de
Las Flores, FMNH 113751, 113753; Paraja Nuevo, El Suchil, UMMZ 90315;
Potrero Viejo, FMNH 113743-50, 126114-18, KU 26104-06, UI^LNH 13332-37,
UMMZ 88S37; San Andres Tuxtla, UINLNH 27123-31, 28611, 71975, UMMZ
115450 (5); San Lorenzo, USNM 123530; 4.5 km. NW Santiago Tuxtla, JDL
992 (skeleton), UIMNH 27122; 32 bn. S Sayula. EAL 1696; Tepalapan, 1.6

Leptodactylid Frog Genus Syurhophus

kin. S Catemaco, UMMZ 118222 (2); Volcan San Martin, south slope, UMMZ
118223; Volcan San Martin, Rancho El Tular, UIMNH 35399-400, 40340-41.

Syrrhophus rubrimaculatus Taylor and Smith

Syrrliophus ruljiimaciilatus Taylor and Smith, 1945:583-85 [Holotvpe. — USNM
114070, from La Esperanza, near Escuintla, Chiapas, Mexico, collected May
13, 1940, bv H. M. and R. Smith]. Duellman, 1958:1-4, 7, 12, 14. Gorham,
1966:167.

Syrrhophus ruhrimaculata: Smith and Taylor, 1948:48-49.

Diafinosis. — Small frogs, males 18.2-23.5 mm. snoiit-\ent, females 19.0-22.5
mm. snoiit-\ent length (small sample); vocal slits in males; digital tips scarcely
expanded (Fig. 1); first finger shorter than second; outer palmar tubercle re-
duced in size; inner metatarsal tubercle elongate, twice the size of small, conical
outer metatarsal tubercle; diameter of tympanum 35.5-46.5 per cent that of

Fig. 8: Si/rrho))hus riihrimaciiJatus (upper right, KU 58911, Xl.6; lower right,
KU 58910, X4) and S. verrucipes (upper left, UIMNH 15995, Xl.6; lower left,

UIMNH 15989, x3.7).

22 University of Kansas Publs., Mus. Nat. Hist.

eye in both sexes; dorsum brown with small pale spots (red in life); venter
gray.

Remarks. — Previous authors who treated Syrrliophiis placed this species in
the western complex, because it occurs on the Pacific \'ersant and has a reduced
outer palmar tubercle. Duellman (1958) placed rubrimaculatus apart from
the other western species, because of its relati^'ely unexpanded digital tips and
coloration. The digital tips are like those in leprus, which nibrimacultmis
reseml:)les. Except for the reduction of the outer pahnar tubercle, ruhrimaciiJa-
tus could be a member of the leprus group.

Syrrhophus rubrimaculatus is probably best treated as a Pacific derivative
of the leprus group, even though the pahnar tubercles do not agree. The re-
moval of rubrimaculatus from the western complex results in a more homogene-
ous remainder and does not greatly increase the heterogeneity of the eastern
complex.

Etymology. — Latin, meaning spotted with red; in reference to the colors in
life.

Distribution. — Low to moderate ele\'ations on the Pacific versant of south-
eastern Chiapas, Mexico ( Fig. 7 ) ; probably extending into adjacent Guatemala.

Specimens examined.— (48) MEXICO, Chiapas: Escuinda, UMMZ 88283;
6 km. NE Escuintla, UMMZ 87876-80; La Esperanza, UIMiXH 13285, UMMZ
88496-97, USNM 114070 (holotype), 114054-69, 114072; Monte Cristo, UMMZ
88353; 1.3 km. N Puerto Madero, KU 58910-11; Finca San Jeronimo, 600-650
m., UIMNH 55299-312, 55313-16 (cleared and stained).

Syrrhophus guttilatiis (Cope)

Malachylocles guttilatus Cope, 1879:264 [Holotype.— USNM 9888, from Guana-
juato, Guanajuato, Mexico; collected in 1877 by Alfredo Duges].

Syrrhopus guttulatus: Boulenger, 1888:204-06.

Syrrhaphus guttulatus: Giinther, 1900:317.

Syrraphus guttulatus: Diaz de Leon, 1904:11.

Syrrhophus guttilatus: Nieden, 1923:399-400. Kellogg, 1932: 125, 127-28.
Smith and Taylor, 1948:49, 51. Firschein, 1954:52-54. Gorham, 1966:164.

Syrrhophus .imithi Taylor, 1940b: 43-45, pi. 1 [Holotype.— USNM 108594, from
15 mi. SW Galeana, Nuevo Leon, Mexico, 1575 m.; collected on October 13,
1939, by Hobart M. Smithl. Smith and Taylor, 1948:49, 51. Firschein,
1954:54-55. Martin, 1958:50. Gorham, 1966:167.

Syrrhophus gaigeae Schmidt and Smith, 1944:80 [Holotype.— FMNH 27361,
from the Basin, Chisos Moimtains, Brewster Co., Texas; collected on Julv
24, 1937, by Walter L. Neckerl.

Syrrhophus petrophilus Firschein, 1954:50-52 [Holotype.— UIMNH 7807, from
5 km. SW San Luis Potosi, San Luis Potosi, Mexico; collected on July 18,
1949, by David Langebartel]. Gorham, 1966:166.

Syrrhophus marnocki: Milstead, Mechani, and McClintock, 1950: 548 (in part).

Diagnosis. — Medium-sized frogs, males 20.6-29.0 mm. snout-vent, females
25.7-31.0 mm. snout-vent length; vocal slits in males; digital tips slightly ex-
panded (Fig. 1); first and second fingers equal; skin of dorsum smooth to
moderately pustular, that of \enter smooth; snout blunt; diameter of tympanum
55.1-75.7 per cent that of eye in males, 47.6-.61.7 in females; dorsum and flanks
cream to gray with light broun to black flecking and \ermiculations; thighs
usually not banded; interorbital bar present (Fig. 8).

Leptodactylid Frog Genus Syhhiiopiius

Fig. 9: Si/nhophus puttilatus (upper left, UIMNH 55519, Xl.4; lower left,
UIMNH 55519, x2.3) and S. marnockii (upper right, TCWC 9317, xl.4;

lower right, TCWC 13510, X2.1 ).

Remarks. — Cope (1879) distinguished Malachtjlodes from Synlwphu.s on
the basis of the presence of a frontoparietal fontanelle in the holotype of
giittilatus. The holotype is a juvenile female and as is the case in the juveniles
of nearly all leptodactylids, a frontoparietal fontanelle is present. Firschein
(1954) used the presence of the fontanelle to distinguish giittilatus from his
petwphiltts.

As is clearly evident from the length of tlie synonymy, I consider a number
of currently used names to he synonymous with giittilatus. I have seen the
holotypes of all four names and am unable to recognize more than a single
species. The holotype of petrophilus is a male, whereas that of smithi is a
female. The supposed differences are a reflection of sexual dimorphism in the
size of the eye (Table 5). The two holotypes, as well as those of gaigeae and
Malaclujlodes giittilatus agree in color pattern.

Schmidt and Smith (1944) named Syrrhophus gaigeae from the Chisos
Mountains of the Big Bend region of Texas and compared it only with S.
marnockii. Milstead, Mecham and McClintock (1950) synonymized gaigeae
and marnockii because they were, unable to \erify the characters Wright and
Wright (1949) used to separate them. Specimens from the Big Bend region
differ from those of the Ed\\'ard and Stockton Plateaus in having a vermiculate

Unr'ersity of Kansas Publs., Mus. Nat. Hist.

pattern, an interorl)ital bar, and a supra tympanic stripe. In these respects they
agree with specimens from northern Mexico. Based on limited observations, the
Mexican population is yellowish to brownish in life whereas the central Texas
population is green in life. Lacking evidence of genetic exchange, the t\vo are
held to be specifically distinct.

28'

102'

100=

98=

100 200 300

KILOMETERS

Fig. 10: Distribution of SyrrJiophus guttilatus.

Leptodactylid Frog Genus Syrriiophus 25

Nearly every specimen examined was infested with chiggers of the genus
Hanncmania. The greatest concentrations are on the venter, in the groin, and on
the thighs. Man>' specimens ha\'e chiggers on the digits and tarsi. The same,
or a related, chigger was found on many specimens of Sijirliophus marnockii
and a few S. vernicipes, but on no other species of the genus. Mr. Willy Wrenn
told me that he has seen heavy infestations of Hannemania on Sijrrhophus paJli-
diis. Infestation b>' Hannemania probably reflects similar ecologies rather than
close relationships.

Etymology. — Latin, gttttula, meaning spotting or flecking, in reference to
the color pattern.

Distribution. — Moderate to intermediate elevations (600 to 2000 m.) along
the Sierra Madre Oriental from the Big Bend Region of Texas to Guanajuato,
Mexico (Fig. 10).

Specimens examined. — (32) TEXAS, Brewster Co.: Juniper Canvon, Chisos
Mts., FMXH 27361 (holot^■pe of S. gaigcae), 27360, 27362-63, MCZ 1.5346,
27801, UMMZ 66080, 66082, 66085-91, USNM 76876; Upper Green Gulch,
TCWC 15943.

MEXICO: Coahuila: 8 km. S Saltillo, UIMNH 55518-21. Guanajuato:
Guanajuato, USNM 9888 (holotvpe of Malachulodes guttilatus); 8 km. E
Guanajuato, AMNH 73425; Cerro Cubilete, AMNH 73424. Nuevo Leon: 3 km.
S Galeana, IDL 1215 (skeleton), UIMNH 58204; 24 km. SW Galeana. 1575 m.,
USNM 108594 (holotvpe of Syrrhoohus .smitJii). San Luis Potosi: 5 km. SW
San Luis Potosi, UIMNH 7807 (holotype of S. petrophihis) . Tamatdipas: 1.6
km. NW La Joya de Salas, 1530 m., UMMZ 110736 (4).

Syrrhophus marnockii Cope

Syrrhophus marnockii Cope, 1878:253 [Syntvpes. — ANSP 10765-68, from "near
San Antonio," Bexar Co., Texas; collected by G. W. Marnock].

Syrrhophus marnocki: Yarrow, 1882:24, 193. Milstead, Mecham, and Mc-
Clintock, 1950:550.

Diagnosis. — Medium-sized frogs, males 18.4-28.9 mm. snout-\ent, females
20.4-35.4 mm. snout-vent length: vocal slits in males: digital tips widened (Fig.
1); first and second fingers equal; skin of dorsum smooth to weakly pustular,
that of venter smooth; snout blunt, rounded; diameter of tympanum 47.2-68.3
per cent that of eye in males, 45.8-73.3 in females; dorsum tan to light brown
in preser\'ati\ e with rusty-brown flecks, \enter white; groimd color green in life;
thighs banded; interorbital bar absent.

Remarks. — Specimens from the southern edge of the Ed\\'ards Plateau and
the eastern edge of the Stockton Plateau ha\e larger flecks on the back that
tend to form a \ermiculate pattern like that of S. guttilatus. The \ermiculation
is never well developed (see plate 38 in Conant, 1958). Most of the specimens
from the Edwards Plateau ha\'e a punctate pattern (Fig. 9).

Fossils are known from the Sangamon interglacial deposits in Foard and
Knox Counties, Te.xas (Lynch, 1964; Tihen, 1960).

Etymology. — A patronym for the collector of the type specimens.

Di.strihiition. — The Edwards Plateau and the extreme eastern edge of the
Stockton Plateau in Texas (Fig. 11). The fossil records lie some 200 miles
to the north. Two specimens (FMNH 103216-17) from Browns\ille, Cameron
Co., Texas, were formerly in the EHT-HMS collection (nos. 31348-49). Data
given in Taylor's field catalogue (housed in the Di\ision of Reptiles, Field
Museum) are "Brownsville, A. J- K.irn collector, April 15, 1934." Until verifi-

University of Kansas Publs., Mus. Nat. Hist.

cation by recently collected material is available, this record must be dis-
regarded.

Specimens examined.— (103) TEXAS, Bandera Co.: 10 mi. SW Medina,
TCWC 13508-10; S mi. W Medina, KU 60243; 13 mi. W Medina, KU 60242,
TCWC 13506-07. Bexar Co.: UIMNH 34694; Classen ranch, near San Antonio.
UMMZ 98891; Helotes, EAL 1560. MCZ 11837 (2), UMMZ 64045. USNM
13635; 2 mi. N Helotes, TCWC 9234-35; 3.5 mi. N Helotes, LSUMZ 10363;
8 mi. N Helotes, TCWC 1549, 4364; San Antonio, FMNH 15553-56, TCWC
13497-99. Blanco Co.: 8 mi. NE Blanco, TCWC 4782. Comal Co.: New Braun-
fels, TCWC 13500-05; 5 mi. NE New Braunfels, UMMZ 71016 (10). Hay.s Co.:
San Marcos, AMNH 22661-64, 32700, FMNH 15245-46, 26250, 26253-57,
37617, 37665, MCZ 15649-50, 23268-69; 6 mi. SW San Marcos, TCWC 5070-71.
7140, 9232-33, 9236, 9316-17, 9320. Kendall Co.: 11 mi. E Boerne, AMNH
54660-61, 54662 (2); 10 mi. W Boerne, KU 18441; Kendalia, UIMNH 21434.
Kerr Co.: Kerr W. M. Area, TCWC 15859; 40 mi. NW Kerrville, TCWC 6555.
Medina Co.: UIMNH 13287-88; 12 mi. N Castroville, UIMNH 21423; 14 mi.
N Castroville, UIMNH 21424-25; 16 mi. N Castroville, UIMNH 21421-22; 17
mi. N Castroville, UIMNH 21428-29; 18 mi. N Castroville, UIMNH 21426-27,
214.30-33; 6.5 mi. NW Rio Medina, KU 18440. Real Co.: Rio Frio, FMNH
55156-57. Travis Co.: Austin, AMNH 44221-22; Mount Bonnell, 5 mi. S Austin,
UMMZ 101453 (10). Uvalde Co.: 13 mi. from Uvalde, UIMNH 62322. Val-
Verde Co.: 40 mi. N Del Rio, JDL 214 (skeleton).

102°

100° 98° 95°

1 ^ 1 '

34°

32°

0 50 100 200

32°

MILES

•

• •

30°

/
/
/

•

^ • • •

■• • •

30°

102°

100° 98° 96°

Fig. 11: Distribution of StjrrhopJttis marnockii (circles). Starred localities are

late Pleistocene records.

Lkptodactylid Frog Genus Syrriiophus 27

Syrrhophus verrucipes Cope

Syniiopliiis vcnuciiics Cope, 1885:383 [Holotype. — ANSP 11325, from near
Zaciialtipan, Hidalgo, Mexico ( 1800 feet lower in a rocky gorge of a stream
near its jimction with the Rio San Miguel), collected bv Dr. Santiago
Bernard]. Kellogg, 1932:126-29. Smith and Taylor, 1948:52-53. Firschein,
1954:55-57. Gorham, 1966:167.

Synlui})1itis vernicipes: Giinther, 1900:216-17.

Toiuodacti/Ius macroti/niixinum Taylor, 1940e:496-99, pi. 55, figs. 2a-h. [Holo-
tvpe.— FMNH 100049 (formerly EHT-HMS 6838), from La Placita, 8 km.
S lacala, Hidalgo, Mexico, 1850 m.; collected on Tulv 2, 1936, by Edward H.
Taylor]. Smith and Taylor, 1948:47-48.

SynJwpliiis macrotijmpanum: Dixon, 1957:384. Gorham, 1966:165.

Diagnosis. — Medium-sized frogs, males 17.5-26.1 mm. snout-vent, females
28.0-31.7 mm. snout-\ent length; vocal slits in males; digital tips slightly ex-
panded; first finger shorter than second; skin of dorsum pustular, that of venter
areolate; snout elongate, subacuminate; diameter of tympanum 56.1-76.7 per
cent that of eye in males, 54.3-56.8 in females; in preservative, dorsum reddish
brown with numerous small black or dark brown spots (Fig. 8); venter white
to cream; in life dorsum green with darker green spots, belly white; iris gold
above, bronze below.

Remarks. — Cope's (1885) original description was not sufficiently clear to
enable subsequent authors to recognize this species. Taylor (1940e) described
it as a Tomodactyhis, but Dixon (1957) pointed out that T. macrotympauum
differed from the other species of the genus in having a poorly developed
lumbo-inguinal (inguinal) gland, and placed the species in the genus Syrr-
hophus. Comparison of the holotypes of S. verrucipes and T. macrotympauum
lea\es no doul^t in my mind that a single species is involved. This same species
was reported by Smith and Taylor (1948) as S. verruculatus.

Syrrhophus verrucipes bears resemblence to members of both the leprus
and maruockii groups. In snout shape it is closer to the leprus group, whereas
in digital pad, the shape of the general body form, and contiguity of habitat
it is most similar to the marnockii g^-oup (S. guttilatus).

Etymology. — Latin, meaning warty foot, probably in reference to the
numerous plantar supernumerary tubercles.

Distribution. — Moderate elexations in southeastern San Luis Potosi, Quere-
taro, and northwestern Hidalgo, Me.xico (Fig. 7).

Specimens examined— (43) MEXICO, Hidalgo: lacala, UMMZ 106434;
9.6 km. XE Jacala, Puerto de la Zorra, 1820 m., KU 60240-41, TCWC 11090,
11147; 8 km. S lacala. La Placita. 1850 m., FMNH 100049 (holotvpe of
Tomodactyhis macrotiimvanum). 100791-803, 105334-35, 114287, UIMNH
15989-92. 15995-96, UMMZ 117252, USNM 137202; Tianguistengo. FMNH
113705-09, UIMNH 13328-30; near Zacualtipan, ANSP 11325 (holotype of
Syrrhoplius verrucipes) . Qucrctaro: 3. .5 km. S San luan del Rio, EAL 1343.
San Luis Potosi: 9.6 km. W Ahuacatlan, LSUMZ 4968-70.

Syrrhophus dennisi new species

Syrrhoj)]ius hitodactyhis: Martin, 1958:49 (in part).

Holotype. — UMMZ 101121, adrdt male from a cave near EI Pachon, 8 km.
N Antiguo Morelos, Tamaulipas, Mexico, 250 m., collected on March 13, 1949,
bv Paul S. Martin.

28 University of Kansas Publs., Mus. Nat. Hist.

Paratopotijpes.— (26). UMMZ 101122 (10), 101123 (2), 101126, 126993
(12).

Diagnosis. — Medium-sized frogs, males 22.8-28.4 mm. snout-\'ent, females
25.9-32.0 mm. snout-\ent; vocal slits in males; digital tips greatly expanded,
more than t^vice width of digit; first finger shorter than second; skin of dorsum
shagreened to pustular, that of venter weakly to moderately areolate; toes
webbed basally; dorsiun light brown to tan with brown \ermiculations; \enter
wliite; diameter of tympanum 53.9 to 64.2 per cent that of eye in males, 50.6 to
58.7 per cent in females.

Description and variation. — (Fig. 12). Head wider than body; head as
wide or wider than long in males, sometimes longer than wide in females;
snout acuminate in dorsal view, elongate and rounded in lateral profile; canthus
rostralis rounded but distinct; loreal region slightly concave, sloping abruptly
to lip; lips not flared; eyelid about two-thirds interorbital distance; length of
eye less than distance between eye and nostril; diameter of tympanum 53.9
to 64.2 per cent that of eye in males, 50.6 to 58.7 per cent in females; tympanum
round and distinct in both sexes; supratympanic fold moderately distinct;
choanae within border of jaws, completely \isible from directly below, rounded
to slightly oval; dentigerous processes of pre\omers and teeth absent; tongue
free for posterior one-half, generally oval in outline; vocal slits present in males.

Many scattered pustules on dorsum; flanks areolate; skin of venter areolate
or not ( xariability may be due to difterences in preservation ) ; ventral disc
distinct on chest and lower abdomen; inguinal gland present or not, when
present varying from very large and distinct to poorly defined; axillary gland
absent.

First finger shorter than second; all fingers bearing truncate tips with pads,
each pad having a terminal groo\e; fingers fringed; fingers three and four having
dilated pads t\\o to three times width of digit; subarticular tubercles large,
conical, rounded, simple; supernumerary tubercles nimierous on thenar surface,
none on digits; three palmar tubercles, outer slightly smaller than largest super-
numerary tubercles; row of tubercles on outer edge of forearm \ariable, weak
to very distinct; tips of toes wider than digits, rounded to truncate at tips, each
pad ha\ing terminal groo\e; toes ha\ing lateral fringes, bases of toes united by
web, web not extending to basal subarticular tubercle; subarticular tubercles
smaller than those of hand, round, conical, simple; supernumerary tubercles
numerous on plantar surfaces, extending bet\veen metatarsal tubercles, present
on toes between basal two subarticular tubercles in some specimens; outer
metatarsal tubercle round, conical, one-half as large as ovoid, non-compressed
inner metatarsal tubercle; tarsal tubercles or folds absent.

Ground color pale reddish-brown to tan dorsally, creamy on flanks; dorsal
pattern consisting of reddish-brown to brown vermiculations extending onto
flanks; distinct interorbital light bar present; loreal region darker than snout,
reddish-brown compared to tan or pale reddish-brown; arms colored like dor-
sum; thighs banded, unicolor brown on posterior surfaces; shanks and tarsi
banded; \'enter white to cream punctated with brown in some specimens.

The variation in proportions is summarized in Table 5.

Remarks. — Martin (1958) expressed some doubt that this series of 26
specimens was identical with "S. latodachjlus." My stiidy indicates that the
specimens from El Pachon represent a distinctive but allied species. Males of
the two species can be readily separated by the relative sizes of the tympani,

Leptodactylid Fhog Genus Syrrhophus

'W^

|U^.:.;....,.. ^.■.;,.:^;■■..- ■-■■ ■■'v■■-■■^■ .-.■---.■■■.■:' ■■"'■■.y ••D;>.;>'y-;r ' T?''^; •■-

^i.. . ■ "■■ " i"i I III II I ■^■- 1. -.■,,-.•...■.. ■ . ". 1.*.. -A'jiaf* '

Fig. 12: Syrrhophus dennisi sp. nov., holotype, UMMZ 101121 (dorsum Xl.S,

side of head X6.1 ).

30 University of Kansas Publs., Mus. Nat. Hist.

presence or absence of vocal slits, and color pattern. Females of the two species
can be separated by color pattern. Within the type-series, the pattern varies
from \\'eakly to strongly \ermicnlate but is al\va\"s recognizable as vermiculate
rather than spotted as in S. longipes (=S. Jatodactyhis of Taylor and Martin).

Etymology. — The specific name is a patron\in for Da\id M. Dennis, whose
drawings greatK' enhance the worth of this paper.

Distribution. — Known only from the t>'pe series.

Syrrhophus longipes (Baird), New combination

Batiachi/la longipes Baird, 1859:35, pi. 37, fig. 1-3 [Holotvpe. — apparently
USNM 3237 (cited as 3207 by Cope, 1887:16), now lost, from 40 Leagues
from (probably north) Mexico City; collected by John Potts]. Kellogg,
1932:107.

Epirlicxi.s longipes: Cope, 1866:96.

Eleutherodactybis longipes: Kellogg, 1932:107 (part). Smith and Taylor,
1948:61. Lynch, 1963:580-581. Gorham, 1966:82.

Syrrhophus latoclaciylus Tavlor, 1940d:396-401, pi. 43, figs. A-F, text fig. 7
[Holotype.— FMNH 100063 (formerly EHT-HMS 6807), from Huasteca
Canvon, 15 km. W Monterrey, Nue\'o Leon, Mexico, 680 m.; collected on
June 20, 1936, by Edward H. Ta\lor]. Smith and Taylor, 1948:50-52.
Martin, 1958:48-50. Gorham, 1966:165.

Diagnosis. — Large frogs, males 22.1-33.2 mm. snout-vent, females 26.8-39.6
mm. snout-vent length; vocal slits lacking in males; digital tips greatly expanded
(more than twice the width of digit); first finger shorter than second; skin of
dorsum pustular, that of \ enter smooth; diameter of tympanum in males 61.1-
87.2 per cent that of eye, 49.5-72.1 per cent in females; dorsum tan with large
or small spots and blotches; limbs banded; interorbital bar or triangle present.

Remarks. — I ha\e applied Baird's Batrachyla longipes to the frog Taylor
(1940d) called Syrrhophus latodactylus because the color pattern (Fig. 13)
predominant in the southern part of the range agrees with that described
(figured) for Batrachyla longipes.

The color pattern of indi\ iduals in the southern part of the range of this
species consists of large spots or blotches, whereas in the northwestern part
the pattern is made up of smaller spots. In the northeastern part of the range,
the pattern is more reduced and tends to consist of hea\y flecking. The inter-
orbital bar is narrower in specimens from Nuevo Leon and TamauHpas and is
triangular in specimens from Hidalgo and Queretaro.

The status of the name Batrachyla longipes is currently that of a nomen
duhium (Lynch, 1963). At that time, I was unaware of the geographic varia-
tion in color pattern in Syrrhophus latodactylus.

The exact type-locality of Batrachyla longipes is not known. If it is 40
Leagues north of Mexico City, the locality would be in an area where the
species has a blotched instead of a flecked or spotted pattern. No justifiable
evidence was presented to place Batrachyla longipes in Eleutherodactylus in-
stead of Syrrhophus. Barbour (1923) and Kellogg (1932) associated another
species (E. batrachylus) with longipes. Taylor (1940a) noted this as a case of
misidentification and corrected the error but left longipes in the genus Eleii-
therodactyltis. Lynch (1963) noted se\eral points of morphological agreement
between Syrrhophus and B. longipes but did not place longipes in Syrrhophus.

Baird's (1859) figures of the holotype do not illustrate prevomerine teeth,
but according to Cope (1866) they were present in the holotype. The digital

Lei'todactylid Fhog Genus Syrriiophus

;^l

tips of the trotr in the figure are somewliat narrower than those typically seen
in S. hitodactyhts. If the specimen was slightly desiccated, as possibly was the
case, the digits \\ould appear narrower. There is no evidence contrary to
placing Synlioplut.s latodachjlus in the synonymy of Batrachijla longipes.

Application of Baird's name Batrachijla longipes to the species of frog
heretofore called Synluij^Jitis latodactyhis poses one serious problem. Batrachyla

Fig. 13: Dorsal views of Syrrhophiis longipes illustrating geographic variation
in pattern (left, TCWC 12179, xl.5; right, KU 92572, Xl.8); side of head

(TCWC 10966, x6).

University of Kansas Publs., Mus. Nat. Hist.

longipes is the type-species (by original designation) of the genus Epirhexis
Cope, 1866, which has priority over Synliophits Cope, 1878. If Batrachyla
longipes is left in the status of a nomen dubiiim, Epirhexis can be forgotten,
for the two names are tied together. However, since it seems almost certain
that BatracJujla longipes and Syrrhophiis latodactyhis are conspecific, the former
name should not be left as a nomen duhiiim. Epirhexis never came into general
usage ( Cope cited the name four times, but no one else has used it ) , whereas
Syrrhophiis is well established in the zoological literature. It would serve only
to confuse the Hterature to adhere strictly to the Law of Priority and replace
Syrrhophus with Epirhexis. Therefore, Syrrhophus is used in this paper, even
though Epirhexis has priority. A request for the suppression of Epirhexis Cope,
1866, has been submitted to the International Commission of Zoological
Nomenclature (Lynch, 1967).

Etymology. — Latin, meaning long-footed; Taylor's latodoctylus refers to
the wide digital pads.

96'

26'

100 200 300

KILOMETERS

24'

22'

96'

Fig. 14: Diatrihution oi Syr rhopJnis dennisi (triangle) and S. longipes (circles).

Leptodactylid Fhog Genus Syrriiopiius 33

Distribution. — Moderate elevations (650 to 2000 meters) along the Sierra
Madre Oriental frt^n central Nuevo Leon to northern Hidalgo, Mexico (Fig. 14).

Specimens examined. — (122) MfiXICO, Hidalgo: 3 km. NE Jacala,
AMNH 52977; 9.6 km. NE Jacala, 1800 m., TCWC 10966-70, 12179; 8 km.
S Tacala, La Placita, 1850 m., FMNH 100266-68, 103244, UIMNH 13291,
13327. Nuevo Leon: Salto Cola de Caballo, KU 92572; Huasteca Canyon, 15
km. W Monterrev, 680 m., FMNH 100063 (holotype of S. latodactijlus),
UIMNH 13290; 6.5 km. N Pablillo, EAL 1319; Sabinas Hidalgo, USNM
139728. Qucretaw: Cueva de los Riscos, 8 km. SW Jalpan, KU 106300. San
Luis Potosi: 13 km. E Santa Barberita, LSUMZ 2295; second camp, San Luis
Potosi road, UIMNH 13326; Xilitla, Cueva sin nombre, UMMZ 125892.
Tamaulipas: 4 km. W El Carrizo, 500 m., UMMZ 111343 (31); 8 km. N
Chamal, Bee Cave, KU 106299; 14.5 km. NNW Chamal, 420 m., UMMZ
111339-40, 111342 (4), 111344 (11); 19 km. NNW Chamal, 700 m., UMMZ
111341 (3); El Chihue, 1880 m., UMMZ 111289 (4); 11 km. N Gomez Farias,
1060 m., UMMZ 101166; 11 km. WNW Gomez Farias, 1800 m., UMMZ 108507
(3); 8 km. NW Gomez Farias, 1060-1400 m., LSUMZ 11085, UMMZ 101167
(3), 101168 (4), 101169 (2), 101170 (3), 101171 (2), 101360-61, 102860,
102933 (4), 102934 (2), 102935-38, 102939 (2), 102940-43, 108800 (3),
110735, 111345-46.

Syrrhophus pipilans Taylor

Syrrhophus pipilans Taylor, 1940c:95-97, pi. 1 [Holotype.— FMNH 100072
(formerly EHT-HMS 6843), 14.6 km. S Mazatlan, Guerrero, Mexico; col-
lected on July 22, 1936, by Edward H. Taylor].

Diagnosis. — Medium sized frogs, males 22.6-28.5 mm. snout-vent, females
21.1-29.4 mm. snout-vent length; vocal slits present in males; finger tips slightly
expanded, truncate in outline; inner metatarsal tubercle less than twice the
size of outer; skin of dorsum smooth to shagreened, that of venter smooth;
t\anpanimi 36.5-54.0 per cent diameter of eye; dorsum dark brown with large
or small light brown, orange-brown, or yellowish spots or blotches; limbs
banded; interorbital bar absent.

Remarks. — Two subspecies were recognized by Duellman (1958). Pre-
\iously both had been treated as species. The two populations were dis-
tinguished on the basis of color pattern and the size of the tympanum.
Measurements of 17 males of S. p. nebidosus from central Chiapas and 18 males
of S. p. pipilans from south-central Oaxaca and Guerrero, Mexico, demonstrates
that the supposed difference in tympanum size is not significant (Fig. 15).

nebulosus

400 450 500

Fig. 15; Dicegrams of ear size relati\e to eye diameter in the two subspecies
of Syrrhophus pipilans. Ni=17 in nebulosus, 18 in pipilans.

University of Kansas Publs., Mus. Nat. Hist.

Fig. 16: Syrrhophus pipilans nehulosus (left, KU 58908) and S. p. pipilans

(right, KU 86885). X2.7.

There is, howexer, a tendency for the western population of S. pipilans to have
larger tympani. Based on the present examination of 112 specimens of this
species the t\vo populations are held to be sufficiently distinct to warrant
taxonomic recognition as subspecies (Fig. 16).

The parotoid glands attributed to this species by Taylor (1940c:95) are
merely the superficial expression of the m. depressor mandibitlae and scapula.
No true glands are present in the parotoid region.

Syrrhophus pipilans nebulosus Taylor

Sinrhophus nehulosus Taylor, 1943:353-55, pi. 27, figs. 3-5 [Holotype.— FMNH
100095 (formerly EHT-HMS 3774), near Tonola, Chiapas, Mexico; col-
lected on August 27, 1935, by Hobart M. Smith and Edward H. Taylor].
Smith and Taylor, 1948:49, 51.

Syrrhophus pipilans nebulosus: Duellman, 1958:2-4, 9, 12, 14. Stuart, 1963:32-
33. Gorham, 1966:166-67.

Diagnosis. — Diameter of tympanum 36.6-47.8 per cent that of eye; dorsum
dark brown with numerous small light brown to \e!lowish spots.

Remarks. — The distribution of this subspecies is adequately described by
Duellman (1958). Fouquette (1960) described the vocalization of this frog.

Etymology. — Latin, nebula, in reference to the clouded dorsal pattern.

Distribution. — Low to moderate elevations along the Pacific \ersant of
Chiapas and in the Grijalva valley of Chiapas and Guatemala (Fig. 17).

Specimens examined. — (54) GUATEMALA, Huehuetenango: facaltenango,
UMMZ 117036; 35 km. SE La Mesilla, TNHC 29652. MEXICO, Chiapas: 11.2
km. N Arriaga, 300 m., UMMZ 125891; 11.8 km. N Arriaga, UMMZ 117279;

Leptodactylid Frog Genus Syrrhophus

12.8 km. N Arriaga, UMMZ 117280; 17.,5 km. S Arriaga, UIMNH 57108-109;
1.5 km. S Bochil, 1250 m., KU 58898-908; Ceno Hueco, 7 km. S Tuxtla
Gutierrez, UMMZ 123007; 3.2 km. S Ixtapa, UMMZ 124000; Linda Vi.sta, ca.
2 km. N\V Pueblo Xuexo Solistahuacan, KU 58897; Hda. Mon.serrate, 40 km.
NW Arriaga, UMMZ 102258; near San Ricardo, FMNH 100720; Tapachula,
FMNH 75792, 103242, 100695-96, UIMNH 13292; 56 km. E Tapanatepec,
Oaxaca, TXHC 26942, Tonola, FMNH 100095 (holotype), 100686-92, UIMNH
1.3293-95; Tuxtla Gutierrez, FMNH 100693-94, UIMNH 13297; 19 km. N
Tuxtla Gutierrez, TNHC 25229-30; 15.5 km. NE Tuxtla Gutierrez, UMMZ
119892 (3); 19 km. NE Tuxtla Gutierrez, UMMZ 119891 (3); 8 km. NNW
Tuxtla Gutierrez, KU 37809; Union de Juarez, FMNH 105294.

Syrrhophus pipilans pipilans Taylor

PSyrrhopiis veiniculatus: Gadovv, 1905:194.

Syrrhophus pipilans Taylor, 1940c: 95-97, pi. 1 [Holotype.— FMNH 100072
(formerly EHT-HMS 6843), from 14.6 km. S Mazatlan, Guerrero, Mexico;
collected on July 22, 1936, by Edward H. Taylor]. Taylor and Smith,
1945:581-82. Smith and Taylor, 1948:49, 50-51.

Syrrhophus pipilans pipilans: Duellman, 1958:1-4, 8-9, 13-14, pi. 2, fig. 1.
Gorham, 1966:166.

Diagnosis. — Diameter of tympanum 40.6-54.0 per cent that of eye; dorsum
dark brow n w ith large light spots or blotches.

Remarks. — Duellman's (1958) synopsis of this subspecies is adequate; the
distribution has not been extended, but several records are now a\ ailable which
fill in gaps.

Gadow's (1905) record of S. verructilatus from "Buena Vista, S. Guerrero"
is most likely applicable to this species. Gadow simply included the name in
a list of the species he had collected during liis trip in Mexico (1902-04); no

13'

KILOMETERS

103 =

99'

95"

91 «

Fig. 17: Distribution of Syrrhophus pipilans: nebulosus (open circles) and

pipilans ( solid circles ) .

University of Kansas Publs., Mus. Nat. Hist.

further comment \\as made on this species although references to Syrrhopus
(sic) appear in se\eral places in the paper and would appear to apply to the
species he had.

Etymology. — Latin, pipilo, chirping, peeping, in reference to the call of the
male.

Distribution. — Sea level to about 1800 meters along the Pacific versant of
western Mexico from central Guerrero to the Isthmus of Tehuantepec (Fig. 17).

Specimens examined. — (62). MEXICO, Guerrero: Acapulco, UMMZ
110125; 6.4 km. N Acapulco, FMXH 100389, 100525; Agua del Obispo, 980-
1000 m., FMNH 75791, 100518-21, 100526, KU 86884-86, UIMXH 13315,
UMMZ 119152, 125890 (4); 13.3 km. XW Covuca, UIMXH 38367, 71982-83:
14.5 km. S Mazatlan, FMXH 100072 (holotvpe), 100408, 100511-17, UIMXH
13302-309; Tierra Colorado, 300 m., KU 67961, UIMXH 13313-14; near El
Treinte, FMXH 126639; Xaltinanguis, FMXH 100522-24, 126640. Oaxaca:
Cacahuatepec, UIMXH 52853; 8 km. XW Rio Canoa, 53 km. ESE Cuajini-
cuilapa, UIMXH 52852; 6.4 km. X El Candelaria, UIMXH 9501; 11.2 km.
S EI Candelaria, UIMXH 9502; 17 km. XE luchatengo, 1600 m., KU 86887;
31.5 km. X Pochutla, UMMZ 123999 (2); 32.9 km. N Pochutla, 850 m., UMMZ
123996; 37.1 km. X Pochutla, UMMZ 123998 (2); 41.4 km. X Pochutla, UMMZ
123997 (2); Cerro Quiengola, FMXH 105653; 3.8 km. N Santiago Chi%ela,
UMMZ 115449; 14.5 km. W Tehuantepec, UMMZ 115448 (2).

Syrrhophus interorbitalis Langebartel and Shannon

Syrrhophus interorbitalis Langebartel and Shannon, 1956: 161-65, figs. 1-2
[Holotype.— UIMNH 67061 (formerly FAS 9378), 36 mi. X Mazatlan,
Sinaloa, Mexico, collected on Xoxember 17, 1955, bv E. C. Bay, J. C.
Schaffner, and D. A. Langebartel]. Duellman, 1958:1-4, 10, 12, 14.
Gorham, 1966:164-65.

Syrrhophis interorbitalis: Campbell and Simmons, 1962:194, fig. 1.

Diagnosis. — Medium sized frogs, only known male 25.6 mm. snout-\ent,
females 20.0-26.7 mm. snout-vent length (small sample); vocal slits in males;

Fig. 18: Left to right. Syrrhophus interorbitalis UIMXH 38095, Xl.5),
nivocolimae (LACM 3203, Xl.3), and S. teretistes (KU 75263, xl.5).

Leptodactylid Frog Genus Syrrhophus 37

finger tips expanded; first finger shorter than second; outer metatarsal tubercle
one-third size of inner; skin of dorsum shagreened, that of \enter smooth;
diameter of t>inpaniun 37.7-42.4 per cent that of eye in both se.xes; pale yellow-
brown ground color mottled with brown; limb bands broad, much wider than
narrow light interspaces; interorbital bar very long, edged with dark brown to
black (Fig. 18).

Remarks. — Duellman's (1958) measinements and proportions of S. inter-
orhitalis were based exclusi\ely on the type series, which is composed of only
females; therefore his interorbitalis data are not comparable with the data for
tlie other species in his talkie. Campbell and Simmons ( 1962 ) collected the
onl\ known male. The type series \\ as collected beneath rocks in a stream bed;
the collectors heard calling frogs in the bushes but were imable to obtain speci-
mens (Langebartel and Shannon, 1956). Campbell and Simmons (1962)
reported that their specimen had a poorly developed interorbital bar in life;
in preser\ati\e the bar compares faxorably with the bar in the female (Fig. 18).

Etymology. — Latin, in reference to the pale interocular band.

Distribution. — Pacific lowlands of Sinaloa, Mexico (Fig. 20).

Specimens examined. — (10). MEXICO, Sinaloa: 36 mi. N Mazatlan
UIMNH 38094-96, 67061 (holotype), 71970-74; 65 mi. N Mazatlan, LACM
13773.

Syrrhophus modestus Taylor

Syrrhophus modestus Taylor, 1942:304-06, pi. 29 [Holotype.— FMNH 100048
(formerly EHT-HMS 3756), from Hacienda Paso del Rio, Colima, Mexico;
collected on July 8, 1935, by Hobart M. Smith]. Smith and Tavlor,
1948:49-50.

Sijrrhophus modestus modestus: Duellman, 1958:2-5, 7, 14, pi. 1, fig. 1.
Gorham, 1966:166.

Diagnosis. — Small frogs, males 15.8-20.1 mm. snout-\ent length, single
female 18.5 mm.; \'ocal slits present in males; finger tips widely expanded;
first finger shorter than second; inner metatarsal tubercle about three times size
of outer; skin of dorsum shagreened, that of \enter smooth; tympanum con-
cealed; pale cream in preser\'ati\ e with dark brown spots; limbs banded;
bands on forearm and thigh poorly developed or absent; interorbital bar absent.

Remarks. — The tympanum is concealed in S. modestus, S. nivocolimae, S.
pallidus, S. teretistes, and to a lesser degree in S. interorbitalis. However, if
tlie specimen is permitted to dry slightly, the annulus tympanicus becomes
visible through the skin and a tympanum /eye ratio can be computed.

One of the few cases of sympatry within the genus Syrrhophus involves
this species; modestus and nivocolimae are known to be sympatric at one
locality in southwestern Jalisco, Mexico.

Duellman (1958) used the trinomial for this population and named a new
subspecies, pallidus, from Nayarit. I consider pallidus to be specifically distinct
from modestus because there is no evidence of genetic exchange, and there is
no overlap in the distinguishing morphological features. I do consider the two
populations to be closely related Init feel the inter-relationships betsveen
modestus, pallidus, nivocolimae, and teretistes are more complex than would be
indicated by the use of trinomials. The sympatric occurrence of modestus and
nivocolimae is significant; morphologically, they might otherwise be regarded
as subspecies. Although allopatric, similar arguments could be advanced for

University of Kansas Publs., Mus. Nat. Hist.

Fig. 19: Synliophus modcstus [left, UMMZ 115447 (WED 11155)] and S.
pallidus (right, UMMZ 115453). x2.2.

the morphologically similar pallidus and teretistes. The four are here afforded
species rank since morphological similarity and allopatry are not sufficient
grounds for the assumption of genetic exchange.

Etymology. — Latin, meaning unassuming, modest, in reference to the small
size of the species.

Distribution. — Low elevations (up to 700 meters) in the lowlands and
foothills of Colima and southwestern Jalisco, Mexico (Fig. 20).

Specimens examined. — (14). MEXICO, Colima: Hda. Paso del Rio, FMNH
100048 (holotype), 100167, 100299, UIMNH 13300, UMMZ 110877 (2),
USNM 139729; 7.2 km. SW Tecolapa, UMMZ 115477 (4); Jalisco: 17.6 km.
SW Audan, 606 m., KU 102627; 3.2 km. N La Resolana, UMMZ 102100;
Bahia Tenacatita, UMMZ 84264.

Syrrhophus nivocolimae Dixon and Webb

Syrrhophus nivocolimae Dixon and Webb, 1966:1-4, Fig. 1 [Holotype.^ — LACM
3200, from Nexado de Colima (6 airline miles west of Atenquique), Jalisco,
Mexico, 7800 feet; collected on July 20, 1964, by Robert G. Webb].

Diagnosis. — Small frogs, males 18.5-21.1 mm. snout-\ent length, only known
female 24.1 mm. snout-\ent; vocal slits present in males; finger tips widely
expanded; first finger shorter than second; inner metatarsal tubercle about three
times size of outer; skin of dorsum warty, that of xenter smooth; tympanum
concealed, its diameter 30.0-39.3 per cent that of eye in males; mid-dorsal
brown band from interorbital bar to anus; bands on limbs narrow, dark bands

Leptodactylid Frog Genus Syrrhophus

less than one-half widtli of light bands, npper arm not banded; narrow inter-
orbital light bar.

Remarks. — This species is closely related to S. niodesttis and differs in color
pattern and degree of \\'artiness of the skin. Dixon and Webb (1966) held
that nivocoliinae had no close relatives, but the condition of the tympanum,
size, natine of the outer palmar tubercle, relative sizes of the metatarsal
tubercles, and shape and size of the digital pads all point to a close relationship
between S. modestus, S. nivocolimac, and S. paUidus.

Dixon and Webb (1966) reported that S. nivocolimae has a large tym-
panum (50.0-59.0 per cent diameter of eye). However, my examination of
the type series and several other specimens from Jalisco reveals that the

24'

22"

20"

108*

106'

104"

Fig. 20: Distribution of the species of the modestus group: interorhitalis (open
circles), teretistes (sohd circles), modestus (open triangles), pallidus (solid
triangles) and nivocolimae (square). Arrow indicates locality of sympatry be-
tween modestus and nivocolimae. Solid line about the localities for interorhitalis
is a range estimate based on call records and specimens examined.

40 University of Kansas Publs., Mus. Nat. Hist.

largest tympanum/eye ratio is 39.3 per cent. Therefore, the tympanum/eye
ratio in S. uivocolimae is in agreement with those for S. modestus, S. pallidiis,
and S. teretistes (Table 6).

Etymology. — niv, Latin, and Colima ( Nevado de), meaning high on the
volcano, in reference to the higher distribution of this species (around 2000
meters) than other members of the group.

Distribution. — Known from southwestern Jalisco, Mexico, at moderate to
high elevations ( 600-2400 meters ) .

Specimens examined. — (48) MEXICO, ]aiisco: 17.6 km. SW Autlan, 606
m., KU 102626, 102631; 6.4 km. W Atenquique, 2060 m., KU 102628-30,
102632; 8 km. W Atenquique, 1970 m., LACM 3210-12; 9.6 km. W Atenquique.
2360 m., LACM 3200 (holotvpe), 3201-09; 14.5 km. W Atenquique, 2000 m.,
LACM 25424-36, 25439-41, 25446; 15 km. W Atenquique, LACM 37044-46,
37244-47; 16 km. W Atenquique, 2105 m., LACM 25443-45; 17 km. W Aten-
quique, 2180 m., LACM 25442.

Syrrhophus pallidus DuelLman, New combination

Synliopluis modestus: Davis and Dixon, 1957:146.

Syrrhophus modestus pallidus Duellman, 1958:2-3, 5-7, 14, pi. 3 [Holotype. —
UMMZ 115452, from San Bias, Nayarit, Mexico, sea level; collected on
August 13, 1956, by William E. and Ann S. Duellman]. Zweifel, 1960:86-88,
91, 93-94, 118, 120-22. Gorham, 1966:166.

Syrrhopliis modeistus pallidus: Campbell and Simmons, 1962:194.

Diagnosis. — Small frogs, males 17.9-19.3 mm. snout-vent lengtli; vocal slits
in males; finger tips widely expanded; first finger shorter than second; inner
metatarsal tubercle about three times size of outer; skin of dorsvun shagreened,
that of venter smooth; tympanum concealed, its diameter 27.0-35.6 per cent of
eye in males; ground color cream vermiculated with brown, upper arm and
tiiigh lacking, or with few, indistinct, bands; interorbital bar absent.

Remarks. — Considerable debate has been waged relative to the value of
subspecies and to tlie reasons for recognizing distinct disjunct populations as
species versus subspecies. Lacking evidence of genetic exchange, I prefer to
retain disjunct populations that are distinctive as species.

All known specimens of paUidus can be separated from tliose of modestus
by color pattern. The tvvo nominal species exhibit overlap in proportions but
the same can be said about nearly every species of Syrrlioplnis; therefore,
overlap in proportions can be disregarded in assessing specific versus sub-
specific rank. Until contrary exidence is forthcoming, I consider the disjunct
populations heretofore held to be subspecies of modestus to be specifically
distinct. The specimens of the disjunct population of paUidus on the Tres
Marias do not differ from the mainland population in Nayarit. This e\idence,
though perhaps secondary, supports my contention that two species should be
recognized.

Etymology. — Latin, in reference to the pale ground color in comparison
with that of S. modestus.

Distribution. — Low elevations in coastal Nayarit and on Islas Tres Marias
(Fig. 20).

Specimens examined. — (12) MEXICO, Nayarit: 18.8 mi. NW Ahuacatlan,
UIMNH 7808; San Bias, UMMZ 1154.52 (holotvpe), 115453-57; 17 km. NE
San Bias, 150 m., MSU 5085; 12.8 km. E San Bias, UIMNH 71979; 31 km.
E San Bias, UIMNH 71978; 13.5 km. N Tepic, UIMNH 71980-81.

Leptodactylid Frog Genus Syrhhophus 41

Syrrhophus teretistes Duellman

Sijirhophus teretistes Duellman, 1958:2-3, 10-14, pi. 2, fig. 2 [Holotype. —
UMMZ 1 15451, from 4.8 km. NW Tepic, Navarit, Me.xico, 840 m.; collected
on August 12, 1956, by William E. Duellman]. Gorham, 1966:167.

Diapnosis. — Medium-sized frogs, males 19.2-23.2 mm. snout-vent length,
single known female 24.8 mm. snout-\ent; vocal slits in males; finger tips
widely expanded; first finger shorter than second; inner metatarsal tubercle
about three times size of outer; skin of dorsum shagreened, that of venter
smooth; t\mpaninn partially concealed, its diameter 28.6-43.8 per cent of eye
in males; ground color brown vermiculated with dark brown to nearly black;
upper arm and thigh banded; interorbital light bar absent.

Remarks. — S. teretistes appears to be most closely related to S. paUidus;
I consider it to be an upland derivative of pallidus. Morphologically, the differ-
ences bet\veen the two are few, but lacking evidence of genetic exchange they
are retained as species.

Etymology. — Greek, in reference to the whistle-like nature of the call.

Distribution. — Moderate ele\'ations (840-1200 meters) in the Sierra Occi-
dental of Nayarit, Sinaloa, and Durango, Mexico (Fig. 20).

Specimens examined. — (13) MEXICO, Nayarit: 4.8 km. NW Tepic, 840
m., UMMZ 115451 (holotype). Sinaloa: Santa Lucia, 1090 m., KU 75263-72;
1 km. NE Santa Lucia, 1156 m., KU 78257; 2.2 km. NE Santa Lucia, 1156 m.,
KU 78258.

Discussion

There are relatively few clear-cut morphological differences
among the fourteen species now assigned to Syrrlwpluis. The
majority of the species are allopatric and differ primarily in color
patterns. Sympatric occurrence serves as an indicator of specific
distinctness and is one of the more practical tests of species validity
when cross-breeding experiments are not possible. Two cases of
sympatric occurrence are known for the species of Sijrrhophus in
western Mexico: modestus and nivocoUmae are sympatric in south-
ern Jalisco and piuilam nehuhsus and riihrimaculatus are sympatric
in southeastern Chiapas. In eastern Mexico, longipes and verrucipes
are sympatric in southern Hidalgo, and longipes is sympatric with
cystia^nathoides, dennisi, and ^utiilatus in southern Tamaulipas.
Syrrhophus cystignathoides and hprtis are apparently sympatric in
central Veracruz.

Subspecific assignments have been made only when there is evi-
dence of intergradation. The sympatric occurrence of morpho-
logically similar species in this genus has led me to adopt a con-
servative approach to the degree of difference philosoohv. I have
therefore recognized all morphologically distinct allopatric popula-
tions as species.

University of Kansas Publs., Mus. Nat. Hist.

Fig. 21: Generic distrilmtions of Sijrrhophiis (stipple) and T omodachjlus
(hatcliing). Black areas are zones of intergeneric sympatry.

Syrrhophus is closely allied to another Mexican leptodactylid
genus, Tomoclactijhis, which was revised by Dixon (1957), who
along with numerous other authors noted the close relationship
between the two genera. There is an almost complete lack of
sympatry between the two genera; in \'ery few places in Mexico do
they coexist (Fig. 21). Tomodactylus has its greatest diversity in
the Cordillera Volcanica and Sierra Madre del Sur, whereas Syrr-
Jiophus reaches its greatest div^ersity in the Sierra Madre Oriental
and eastern foothills. The species of both genera are about the
same size and presumably have similar requirements insofar as
food, breeding sites, and habitat selection.

Four cases of inter-generic sympatry are known for the two
genera : 1 ) the Chilpancingo region of Guerrero, 2 ) the lowlands of
Colima and the mountains just inland in Jalisco, 3) the lowlands of
central Nayarit, and 4) the Sierra Madre Occidental on the Dur-
ango-Sinaloan border. The apparent sympatry in the Chilpancingo
region involves four species: S. pipilom, T. aJboIohris, T. (hiatus, and
T. nitidus. Of the four, T. dilatus appears to be completely allopatric
in that it occurs at higher altitudes (above 2000 meters), whereas
the other three occur below 1800 meters in the region (Davis and
Dixon, 1965). In the Colima-Jalisco region, Tomodactyhis tends

Leptodactylid Frog Genus Syrriiophus

3000m

2000m

to occur higher (Dixon and Webb, 1966) than some of the Sijrr-
JwpJiUs, but one subspecies of Tomodactijlus nitichis is a lowland
frog, occurring sympatrically with the lowland SyrrJiophus modestus.
A similar situation is observed in Nayarit; the lowland Tomodactijlus
occurs sympatrically with the small Syrriiophus paUidus. In both
cases the Syrriiophus is smaller than the Tomodactijlus.

Frogs of the genus Syrr-
iiophus tend to occur at lower
elevations than do their close
relatives of the genus Tomo-
dactijlus (Fig. 22). This gen-
eralization is complicated by the
occurrence in the Sierra Madre
Oriental in relatively high alti-
tude Syrriiophus (up to 2000
m.) and the occurrence in
M i c h o a c a n of low altitude
Tomodactijlus (to sea level).
There are no Tomodactijlus in
the Sierra Madre Oriental,
whereas the genus Syrriiophus
is represented in the lowlands
of western Mexico ( modestus
group). Syrriiophus and Tomo-
dactijlus exhibit essentially para-
patric distributions. The two
genera as now composed can be
characterized as low to moderate
elevation frogs (Syrriiophus) and
moderate to intermediate eleva-
tion frogs (Tomodactijlus).

lOOOm. _

Fig. 22: Altitudinal distributions of
Syrriiophus and Tomodactijlus. Widths
of the columns are proportional to the
numbers of species at a given alti-
tude; narrowest width equals one
species.

Literature Cited

Baird, S. F.

1859. Reptiles of the Boundary. United States and Mexican Boundary
Survey, pp. 1-35, pis. 1-41.
Barbour, T.

1923. The reappearance of Batrachyla longipes. Proc. New England Zool.
Club, 8:81-83.
Barbour, T., and A. Loveridge

1946. Typical reptiles and amphibians; supplement. Bull. Mus. Comp.
Zool., 96:59-214.

BOULEXGER, G. A.

1882. Catalogue of the Batrachia Salientia . . . British Museum., 2nd ed.
1888. Note on the classification of the Ranidae. Proc. Zool. Soc. London,
1888, pt. 2:204-06.

44 University of Kansas Publs., Mus. Nat. Hist.

Campbell, H. W., and R. S. Simmons

1962. Notes on some reptiles and amphiliians from western Mexico. Bull.
So. California Acad. Sci., 61:193-203.

CONAXT, R.

1958. A field j^uide to reptiles and amphibians. Hougliton-MifHin Co.
Boston. 366 pp.
Cope, E. D.

1866. On the structm-es and distribution of the genera of the arciferous
Aniira. J. Acad. Nat. Sci. Philadelphia, n. ser., 6:67-112.

1877. Tenth contribution to the herpetolog}' of tropical America. Proc.
Anier. Philos. Soc, 17:85-98.

1878. New genus of Cystignathidae from Texas. Amer. Nat., 12:252-53.

1879. Eleventh contribution to the herpetology of tropical America. Proc.
Amer. Philos. Soc, 18:261-77.

1885. A contribution to the herpetology of Mexico. Ibid., 22:379-404.
Davis, W. B., and J. R. Dixox

1957. Notes on Mexican amphibians, with description of a new Micro-
hatrachylus. Heipetologica, 13:145-47.

1965. Amphibians of the Chilpancingo Region, Me.xico. Ihicl., 20:225-33.
Diaz de Leox, T.

1904. Indice de los Batracios que se enquentran en la Republica Me.xicana.
Imprenta de Ricardo Rodriquez Romo. Aguascalientes. 40 pp.

Dixon , T. R.

1957. Geographic xariation and distribution of the genus Tomodactvlus
in Mexico. Texas T- Sci.. 9:379-409.
Dixov. T. R., and R. C. Webb

1966. A new Syrrhophus from Mexico (Amphibia: Leptodactylidae).
Cont. Sc, Los Angeles Co. Mus., 102:1-5.

DUELLMAN, W. E.

19.58. A review of the frogs of the genus SynJioj>Jiiis in western Mexico.

Occ. Pap. Mus. Zool. Univ. Michigan, 594:1-15.
1960. A distributional study of the amphibians of the Isthmus of Tehuan-

tepec, Mexico. Univ. Kansas Pubis. Mus. Nat. Hist., 13:19-72.
FiRSCHEIX, I. L.

1954. Definition of some little-imderstood members of the leptodactylid

genus StirrJiophiis, with a description of a new species. Copeia,

(l):48-58.

FOUOUETTE, M. J.

1960. Call structure in frogs of the familv Leptodactvlidae. Texas T. Sci.,
12:201-15.
Gadow, H.

1905. The distribution of Mexican amphibians and reptiles. Proc. Zool.
Soc. London, 1905, pt. 2:191-244.

COKHAM, S. W.

1966. Liste der rezenten Amphibien und Reptilien. . . . Das Tierreich.
Lief, 85:1-222.

GUXTHFR. A. C. L. G.

1885-1902. Biologia Centrali-Americana. Reotilia and Batrachia. 326 pp .
76 pis. Syrrhophus section dated 1900.
Kellogg, R.

1932. Mexican tailless amphibians in the United States National Museum.
Bull. U.S. Natl. Mus., 160.
Langfkartel, D. a., and F A. Shaxxox

1956. A new frog (Syrrhophus) from the Sinoloan lowlands of Mexico.
Herpetologica, 12: 161-65.
Lynch, J- D.

1963. The status of Elptithnnchictiihis loiigipes CRaird) of Mexico
(Amphibia: Leptodactvlidae).' Copeia, (3):580-81.

1964. .additional hylid and leptodact>lid remains from the Pleistocene of
Texas and Florida Hcpetoloeica, 20:141-42.

1967. Epiihc'xis Cope, 1866 (.\mphiliia: Salientia): recjuest for suppres-
.sion under the plenarv powers. I. N. (S). Bull. Zool. Nomencl.,
24:313-15.

Leptodactyliu 1^'iiOG Genus SynRnopiius 45

Lynch, J. D.

1968. Genera of leptodactvlid frogs in Mexico. Univ. Kansas Pubis., Mus
Nat. Hist., 17:503-15.
Martin, P. S.

1958. A biogeography of reptiles and ampliibians in the Gomez Farias
region, Taniaulipas, Me.xico. Misc. Pubis. Mus. Zool. Univ. Michi-
gan, 101: 1-102.
MiLSTEAD, W. M., J. S. Mecham, and H. McClintock

19.50. The amphibians and reptiles of the Stockton Plateau in northern
Terrell Gounty, Texas. Texas J. Sci., 2:543-62.
Xeill, W. T.

1965. New and noteworthy amphibians and reptiles from British Hon-
duras. Bull. Florida State Mus., 9:77-130.

NiEDEX, F.

1923. Anura I ... Das Tierreich. Lief., 46:1-584.
Peters, W.

1871. iJber neue Amphibien . . . des Konigl. Zoologischen Museums.
Monatsb. k. k. Preuss. Akad. Wiss. Berlin, 1870:641-52.
Schmidt, K. P., and T. F. Smith

1944. Ampliibians and reptiles of the Big Bend Region of Texas. Field
Mus. Nat. Hist., zool. ser., 29:75-96.

Smith, H. M.

1947. Notes on Mexican amphibians and reptiles. J. Washington Acad.
Sci., 37:408-12.

Smith, H. M., and E. H. Taylor

1948. An annotated checklist and kev to the Amphibia of Mexico. Bull
U.S. Natl. Mus., 194:1-118.

Stejneger, L.

1915. A new species of tailless batrachian from North America. Proc. Biol
Soc. Washington, 28:131-32.
Taylor, E. H.

1940a. A new eleutherodactylid frog from Mexico. Proc. New England

Zool. Glub, 18:13-16.
1940b. Two new anuran amphibians from Mexico. Proc. U.S. Natl. Mus.,

89:43-47, 1 pi.
1940c. A new Syrrhophus from Guerrero, Mexico. Proc. Biol. Soc. Wash-
ington, 53:95-98, 1 pi.
1940d. New species of Mexican Anura. Univ. Kansas Sci. Bull, 26:. 385-405.
1940e. Herpetological miscellany no. I. Ibid., 26:489-571.

1942. New Caudata and Salientia from Mexico. Ibid., 28:295-323.

1943. Herpetological novelties from Mexico. Ibid., 29:343-61.

1952. A review of the frogs and toads of Costa Rica. Ibid., 35:577-942.
Taylor, E. H., and H. M. Smith

1945. Summary of the collections of amphibians made in Mexico under
tlie Walter Rathbone Bacon Traveling Scholarship. Proc. U.S.
Natl. Mus., 95:521-613.

TlHEX, J. A.

1960. Notes on Late Cenozoic hylid and leptodactylid frogs from Kansas,
Oklahoma and Texas. Southwest. Nat., 5:66-70.
Wright, A. H., and A. A. Wright

1949. Handbook of frogs and toads. 3rd ed. Gomstock. 640 pp.
Yarrow, H. G.

1882. Checklist of North American Reptilia and Batrachia, with catalogue
of specimens in U.S. National Museum. Bull. U.S. Natl. Mus.,
24:1-249.

ZWEIFEL, R. G.

1960. Results of the Puritan-American Museum of Natural History Ex-
pedition to Western Mexico. 9. Herpetology of the Tres Marias
Islands. Bull. Amer. Mus. Nat. Hist., 119:81-128.

University of Kansas Publications
Museum of Natural History

Volume 20, Number 2, pp. 47-223, 38 figs.
Deeember 23, 1970

A Revision of Colubrid Snakes
of the Subfamily Homalopsinae

BY
KG KG GYI

University of Kansas

Lawrence

1970

University of Kansas Publications, Museum of Natural History
Editors of this number: Frank B. Cross and William E. Duellman

Volume 20, Number 2, pp. 47-223, 38 figs.
Published December 23, 1970

University of Kansas
Lawrence, Kansas

printed by

THE university OF KANSAS PRINTING SERVICE

LAWRENCE, KANSAS

1970

CONTENTS

Introduction 51

Acknowledgments 51

Materials and Methods - 52

Subfamily Homalopsinae 56

Definition 56

Nomenclatinal History 57

Key to the Genera - 61

Genus Etiliijdris Sonnini and Latreille 61

Description and Variation 62

Key to the Species 71

The Enlujdris phtmbea Gronp 75

Enhijdris indica (Gray) 75

Enhijdris alternans (Renss) 77

Enhydiis pJumbea (Boie) 82

The Enhijdris cidujdris Gronp 86

Enhydris culu/dris (Schneider) 86

Enhijdris jagorii Peters 92

Enhijdris innominata (Morice) 95

Enhijdris smitlii (Smith) 98

The Enlujdris chinensis Group 100

Enhijdris chinensis (Gray) 100

Enhijdris hennetti (Gray) 103

Enhijdris longicaiida (Boiirret) 105

The Enlujdris macleaiji Group 107

Enhijdris macleaiji (Ogilby) 108

Enhijdris polijlepis (Fischer) 110

Enhijdris matannensis (Boulenger) 112

The Enhijdris maculosa Group — 113

Enhijdris macidosa (Blanford) 114

Enlujdris pahangensis Tweedie 116

The Enlujdris sieholdi Group 118

Enlujdris sieholdi (Schlegel) 118

Enlujdris dussumieri (Dumeril and Bibron) 121

The Enlujdris hocourti Group 122

Enlujdris hocourti (Jan) 122

Enhijdris alhomaculata (Dumeril and Bibron) 125

The Enhydris punctata Group 127

Enlujdris punctata (Gray) 127

Enlujdris doriae (Peters) 129

The Enhydris pakistanica Group — - 132

Enlujdris pakistanica Mertens 132

(49)

Genus Homalopsis Kuhl and Hasselt 135

Homalopsis buccata (Linnaeus) 136

Genus Cerberus Cuvier 146

Description and Variation 147

Key to the Species and Subspecies 153

Cerberus microlepis Boulenger 154

Cerberus australis (Gray) 157

Cerberus rynchops ( Schneider ) 159

Cerberus rynchops rtjnchops (Schneider) 160

Cerberus rynchops novaeguineae Loveridge 169

Genus Heurnia Jong 170

Heurnia ventromaciilata Jong 170

Genus Myron Gray 172

Myron richardsonii Gray 172

Genus Gerarda Gray 174

Gerarda prevostiana (Eydoux and Gervais) 174

Genus Fordonia Gray - 177

Fordonia leucobalia (Schlegel) 177

Genus Cantoria Girard 182

Key to the Species 182

Cantoria violacea Girard 183

Cantoria annulata (Jong) 186

Genus Bitia Gray 187

Bitia hydroides Gray 187

Genus Erpeton Lacepede 189

Erpeton tentactdatus Lacepede 190

Osteology 193

Nasal region 193

Cranium and associated elements 194

Maxillo-palatal-pterygoid arch 196

Mandible 197

Dentition 198

Vertebrae 198

Intergeneric Variation in the Skulls 199

Life History — 203

Phylogenetic Relationships 205

Interspecific Relationships 205

Intergeneric Relationships 207

Cenozoic History of Southeastern Asia and the Indo-Australian

Archipelago 210

Evolution and Zoogeography of the subfamily Homalopsinae 211

Summary and Gonclusions 216

Literature Cited 219

(50)

Introduction

The rich herpetofauna of southeastern Asia and the Indo-
Austrahan Archipelago offers many opportunities for the study of
reptiles and amphibians, their relationships with one another and
with their environment, their Hfe histories and their distribution.
However, many of these studies cannot be carried out effectively
until a series of thorough taxonomic investigations have been com-
pleted. The present account of colubrid snakes of the subfamily
Homalopsinae is the first of several projected taxonomic studies.

The objectives of this revision are determination of the valid
species of homalopsine snakes, and description of their geographic
and individual variation so far as practicable. Because many spe-
cies are represented by few specimens and others by insufficient
specimens from throughout the range of the species, my emphasis
is placed on description of the variation itself and not on nomen-
clatorial recognition of infraspecific variation; hence my taxonomic
conclusions may be conservative.

Although my study was mainly morphological, I include avail-
able information on habits, habitat, life history, and other data to
give a broader and more sound basis to the systematic arrangement.
The study was completed in May, 1965.

Acknowledgments

For permission to examine specimens, and for information concerning
specimens in their care, I am grateful to Charles M. Bogert and Richard G.
Zweifel, American Museum of Natural History; James E. Bohlke, Academy of
Natural Sciences, Philadelphia; L. D. Brongersma and H. E. Muller, Rijksmu-
seum van Natimrlijke Historic, Leiden; the late Doris M. Cochran and James
A. Peters, United States National Museum; M. E. Dottrens, Naturhistorisches
Museum, Geneva; Lothar Forcart, Naturhistorisches Museum, Basel; Alice G.
C. Grandison, British Museum (Natural History); Jean Guibe, Museum National
d'Histoire Naturelle; D. Hillenius and P. J. H. van Bree, Zoologisch Museum,
Amsterdam; Robert F. Inger, Chicago Natural History Museum; Alan E.
Leviton, California Academy of Sciences and Stanford University Natural
History Museum; Robert Mertens and Konrad Klemmer, Senckenbergische
Naturforschende Museum, Frankfurt; Charles F. Walker, Museum of Zoology,
University of Michigan; Ernest E. Williams, Museum of Comparative Zoology,
Harvard.

William E. Duellman offered invaluable information, guidance, and critical
reading of the manuscript. A. Byron Leonard and George W. Byers provided
encouragement, advice, and read the manuscript. Edward H. Taylor allowed
me to study specimens in his care and gave help in obtaining pertinent litera-
ture and information on localities; he also helped in solving some taxonomic
problems and gave advice concerning photographic techniques.

Special thanks are due to Alice G. C. Grandison and H. E. Goto (Imperial
College, London) for help while I was in London and for part of the photo-

52 Unr'ersity of Kansas Publs., Mus. Nat. Hist.

graphic work. Thanks are also due to M. E. Dottrens, Lothar Forcart, P. J. H.
van Bree, H. E. MuUer and Konrad Klemmer for help while I was working at
their respecti\'e institutions. Veronica Gyi typed part of the manuscript and
assisted me in every phase of the study.

The present study was carried out in the Division of Herpetology of the
Museum of Natural History, University of Kansas, where I was supported by a
three-year Fellowship from UNESCO.

Materials and Methods

Scale counts, measurements, and notes on color pattern were made on 983
specimens of the 1038 specimens of the subfamily Homalopsinae available to
me. My methods of counting scales and taking measurements are described
below.

The dorsal scales were counted at three places: The first count is opposite
the first complete ventral, starting from the dorsal scale adjacent to the first
ventral, and counting obliquely backward to the \'ertebral row and then
forward to the dorsal scale adjacent to the first ventral on the other side of the
body. By this method the count is repeatable. The customary way is to count
at "one head-length" or "two head-lengths" behind the head, or "in the
anterior region of the body." By these methods two persons may not be able to
make the same count on one animal. The second count is around the midbody,
half-way between the first ventral and the vent, counting obliquely forward and
turning backward at the vertebral row. This method results in repeatable
counts. I chose a point at half the length of the body instead of half the
number of ventrals, because the anterior ventrals are usually narrower than
those behind; consequendy the midbody count taken at the location of the
median ventral plate would be anterior to the middle of the body. The final
count is taken at a le\'el equal to the third from the last ventral, counting
obliquely forward and turning backward at the vertebral row. Occasionally the
last two ventrals are narrow, incomplete plates bordered on each side by one
extra dorsal scale.

The formula used for showing reduction in the rows of dorsal scales is that
proposed by Dowling (1951a). The antero-posterior position where reduction
occurs in the number of the dorsal scale rows is designated by citing the
number of the ventral scale directly beneath that place.

Ventrals were counted following the system proposed by Dowling (1951b).
I omitted incomplete ventrals; therefore, my counts are slighdy lower than
those of other workers. The anal plate was not included. The terminal spine
was not included in the count of subcaudals.

The counts of gulars are the number of giilars that are arranged longitudi-
nally, adjacent to the mental groove, and extending from the posterior edge of
the anterior chin-shields to the first complete ventral. The number or range (in
parentheses) of gulars plus the number of incomplete \'entrals are of taxonomic
value at the generic level in homalopsine snakes. These characters are ex-
pressed as 6-1-3 or (6 — 8)-f 3.

The umbilical scar or "mnbilicus" is a longitudinal slit-like scar near the
vent, usually present in one, two, or three ventrals. In young specimens the
scar is distinct; in older individuals it can be discerned imder a binocular
dissecting microscope in only about 60 per cent of the specimens. The ventrals
in all homalopsine snakes (except in Bitia and Erpeton) are rounded. The
umbilical scar is a A-shaped indented area, located medially along the margin

CoLUBRiD Snakes, Subfamily Homalopsinae 53

of the ventrals; it is more pigmented than the surrounding area (Fig. 1). The
position of the umbilical scar is expressed by the number of ventrals (counting
anteriorly from the vent) included in the scar. This character has taxonomic
\alue in homalopsine snakes. Beddard (1907) pointed out that in the Viperidae
the position of the "umbilicus" seems to have taxonomic value.

Another taxonomic character used in this work is the ratio of the width of
the ventral to the width of the adjacent dorsal scale.

Measurements of head scales (Fig. 2) were made with calipers under a
dissecting microscope as follows: (1) greatest width and height of rostral;
(2) width of prefrontals and internasals; (3) length of prefrontals and inter-
nasals along their median sutures (therefore, the measurement is not always
an expression of greatest length); (4) length of frontal along middorsal line;
(5) width of frontal at the broadest point; (6) greatest length of parietal,
regardless of shape; (7) greatest height and length of loreal, regardless of
shape; and (8) greatest length and width of anterior and posterior chin-shields,
regardless of shape.

The shape and size of the frontal is expressed for each taxon by the follow-
ing ratios: the width of the frontal to its length; the width of the frontal to
the width of the supraocular; the length of the frontal as compared with the
distance from the anterior margin of the frontal to the tip of the snout; and the
length of the frontal to the length of the parietal.

Some terminology of the head scales requires additional clarification. First,
if the nasal cleft passes through the nasal opening and separates the scale into
two parts, the nasal is referred to as "divided"; if the cleft is completely absent,
it is taken as "entire" or "single"; if the suture extends only to one side, leaving
the scale complete on the opposite side of the nostril, it is considered as a
"semi-divided" nasal. Second, the azygous shield refers to the shield between
the paired prefrontals. The suboculars are those plates that are below the eye;
plates pardy below the eye are considered suboculars if they are longer than
they are high. Any labial which partly touches the loreal or the chin-shield, or
partly enters the eye, is considered to be a full scale.

In the description of color pattern, some terms or words have been used to
shorten the ordinarily long descriptive phrases that are used repeatedly. The
term rostral-orbital -postorbital stripe refers to the stripe that originates from
the rostral, passes through the eye, and continues posteriorly on the side of the
head. The median nape stripe refers to the longitudinal, middorsal stripe
extending posteriorly from the frontal or median suture of the parietals. The
dorsolateral nape stripe refers to the longitudinal stripe extending posteriorly
from the dorsolateral corner of the parietal. The ventrolateral transverse bar
refers to the transverse bar that originates on the ventrals or on the margin of the
ventrals and extends laterally onto the first few rows of dorsal scales. The
width of the bar or blotch is its transverse extent expressed in the number of
dorsal scale rows it covers. The length of the bar or blotch is its longitudinal
extent, and is given as the number of dorsal scales it covers in an antero-
posterior direction.

In the present work, some words are used in a special way. Oliver's (1948:
166) definirion and use of the terms "normally" and "usually" are used as
quantitative expressions for conditions existing in 75 per cent or more of a
sample. "Rarely" is used to mean conditions existing in less than 10 per cent
of a population.

In accounts of the species and subspecies, characters having little variation,

University of Kansas Publs., Mus. Nat. Hist.

Fig. 1. Umbilical scars in Homalopsis biiccata. Left, adult male (KU 92425)
showing scar on ventrals 135-138; right, juvenile male ( KKG 252) showing scar

on ventrals 126-130.

Prefrontal width

Internasal
width

Internasal length

Prefrontal length

Frontal width
Frontal length

Parietal length

TS.

Fig. 2. Diagrams of the dorsal aspect of the head of Enhydris maculosa ( ANSP
5096 ) showing places of measuring scutes.

CoLUBRiD Snakes, Subfamily Homalopsinae 55

such as head scales, have been presented in frequency distriljution tables.
Characters having a wide range of variation and based on large numbers of
observations have been analyzed statistically. The observed range of variation
is followed by the mean (M) in parentheses; in some instances the mean is
followed by the standard deviation (S.D.), also in parentheses. An example is
120-134 (128.2ih3.21). The coefficient of variability (C.V.) is given. For those
characters showing sexual dimorphism, or geographical variation, or onto-
genetic variation, the coefficient of difference (CD.) between the sexes or
between populations has been calculated. The statistical terms and formulas
have been adopted from Mayr, Linsley, and Usinger (1953).

Each synonymy includes, so far as I have found, all the generic and
specific combinations, that have been used for that taxon. Aside from these
references 1 have also included catalogues and check lists.

The maps show the localities where the specimens examined were
collected. Definite locality data were not available for many specimens. The
maps are intended only to present a general picture of the relation of the
localities to each other. Hollow symbols are literature records; these have
been used only when the identification of the specimen is known to be correct.
The National Geographic Society Maps, National Geographic Atlas of the
World and Webster's Geographical Dictionary have been the primary sources,
but other maps and atlases have been consulted. In the list of specimens
examined, the localities and specimens are given in the following alphabetical
order: country, state in each country, locality in each state. Localities that
have not been assigned to a state follow the country. Specimens bearing data
giving only the country or state are listed first in that political unit under "no
definite locality." Museum abbreviations (as listed below) are in alphabetical
order; museum numbers are in numerical order after the abbreviations. When
more than one specimen is included on a single number, the number of
specimens is given in parentheses following the museum number.

The abbreviations for the museum collections are:

AMNH American Museum of Natural History

ANSP Academy of Natural Sciences of Philadelphia

BMNH British Museum (Natural History)

CAS California Academy of Sciences

EHT Edward H. Taylor, Lawrence, Kansas

FMNH Field Museum of Natural History

KKG Ko Ko Gyi, Rangoon, Burma

MCZ Museum of Comparative Zoology

MNHN Museum National d'Histoire Naturelle, Paris

NMB Museum of Natural History, Basel, Switzerland

NMG Museum of Natural History, Geneva, Switzerland

RNHL Rijksmuseum van Naturrlijke Historic, Leiden

SMF Senckenbergische Museum, Frankfurt

SU Stanford University Natural History Museum

KU University of Kansas Museum of Natural History

UMMZ University of Michigan Museum of Zoology

USNM United States National Museum

ZMA Zoologisch Museum, Amsterdam

The keys in this work are intended only as aids to species identification
and are not designed to show phylogenetic relationships. Identifications made

56 University of Kansas Publs., Mus. Nat. Hist.

by use of these keys should be checked against the diagnosis provided for each
species.

Subfamily HOMALOPSINAE Jan

Hydiophidie Boie, (in part), Isis, p. 510, 1827.

Hijdridac Gray, (in part), Catalogue snakes British Museum, p. 35, 1849.
Anisodontiens Dumeril, (in part), Mem. Acad. Sci. France, 23:427, 1854.
Plattjrhiniens Dumeril and Bibron, firpetologie generale . . . reptiles. Paris,

7:796, 1854.
Homalopsinae Tan, (in part), Elenco systematico degli ofidi, p. 74, 1863. Cope,

(in part), Proc. Amer. Philos. Soc, 23:484, 1886; (in part). Trans. Amer.

Philos. Soc, 18:209, 1895. Boulenger, Fauna of British India . . . Reptilia

and Batrachia, p. 372, 1890; Catalogue snakes British Museum, 3:1, 1896.

Taylor, Philippine Sci. Bull., 16:110, 1922. Werner, Arch. Naturg. Berlin,

89(8): 158, 1923. Smith, Proc. Zool. Soc. London, p. 398, 1931. Fauna

of British India . . . Reptilia and Amphibia, 3:379, 1943.
Hornalopsidae Giinther, Reptiles of British India, p. 275, 1864.
Homalopsidinae Forcart, Verb. Naturf. Ges. Basel, 64(2):382, 1953.

Definition. — Body of moderate length; adults ranging approxi-
mately from 300 mm. to 1000 mm. in total length; body cylindrical
or slightly compressed; head rather thick, broad, not very distinct
from neck; tail stout, of moderate length, tapering, more or less
prehensile and compressed proximally; scales subequal in size, not
noticeably imbricate; ventrals relatively narrow, usually smooth but
bicarinate in Erpeton and Bitia; anal divided; subcaudals paired;
eye small; pupil elliptical; nostrils anterodorsally on head, small,
valvular; nasals usually larger than internasal; vertebral hypapo-
physes extending throughout column; pair of grooved teeth on
posterior extremity of maxillaries; hemipenes forked; viviparous.

Remarks. — The snakes of this subfamily are aquatic, but often
they are found on land in the vicinity of water; several species enter
the sea. In many morphological characters the homalopsine snakes
are similar to sea snakes; these similarities caused Gray (1849) to
associate the two groups in his system of classification.

The anterodorsal position of the nostrils enables snakes to
breathe by raising only a small part of their heads out of the water;
this is the same arrangement as that found in the crocodiles, sea
snakes, and other aquatic animals. The nostril is a lunate slit; the
nasal pad projects from the posterior margin of the nasal cavity.
The opening into the interior of the pad is large, and directed
anteriorly. The whole pad can be distended and thus forms an
effective valve. The glottis which fits into the internal nares supple-
ments the eflFective closure of the nasal cavity. As a contrivance for
the complete closure of the mouth, the rostral is never deeply exca-
vated, as in most of the Colubrinae. Furthermore, the snakes are
provided with a distinctly downward projecting tongue, which is

CoLUBRiD Snakes, Subfamily Homalopsinae 57

best developed in those genera that lead an entirely aquatic
existence, such as, Cantoria and Bitia.

The hemipenis is forked, and except for small variations in
detail, does not differ throughout the subfamily. The distal end is
finely cah^culate, the cups are shallow \\'ith small blunt spines; the
calyces and spines are progressively larger proximally. In the re-
gion of bifurcation large flat triangular papilla-like processes are
arranged in longitudinal series, each one ending in a small spine.
Proximal to this region, the hemipenis is naked. The sulcus usually
is forked, in Enhijdris chinensis two sulci are present.

Distribution. — Snakes of this subfamily range from the delta
region of Indus to the southeastern part of China, as far north as
Nanking, southward through all the islands of the Indo-Australian
Archipelago, New Guinea, and the northern coast of Australia. The
distribution of the Homalopsinae is similar to that of the sea snakes
(Smith, 1943:379).

Nomenclattiral history. — Boie (1827) combined most of the
aquatic snakes into his family Hydrophidae; this was the first at-
tempt to group the aquatic snakes as an entity distinct from all other
snakes. The taxonomic arrangement was based on external mor-
phological characters exhibited by these snakes in response to the
aquatic habitat. The homalopsine genus, Erpeton, was considered
by Boie as belonging to his family, Colubrini.

Gray (1849) recognized the family Hydridae consisting of 38
genera, a large and cumbersome grouping. This family contained
most of the genera now recognized for the subfamily Homalop-
sinae, and several other genera. His grouping, like Boie's, was based
on external morphological characters exhibited as aquatic adapta-
tions.

Dumeril and Bibron ( 1854 ) set aside as a separate group the
family Platyrhiniens, which included all the aquatic snakes that
ha\'e flat snouts and opisthogh'phous dentition. In this group, they
recognized seven genera — Erpeton, Homalopsis, Cerhere, Hypsir-
hine, Campylodon, Euroste, and Trigonure. The last four are now
treated as synonyms.

Jan (1863) renamed the Platyrhiniens as the subfamily Homa-
lopsinae, the name now in use. The name of the subfamily is based
on the generic name, Homalopsis Kuhl (1826). In this subfamily,
Jan included most of the genera that are now considered to be
homalopsine snakes.

Giinther ( 1864 ) elevated the subfamily Homalopsinae to family
level because he found that some of the head shields frequently

University of Kansas Publs., Mus. Nat. Hist.

deviated from the arrangement typically found in the colubrine
snakes. According to him, homalopsine species inhabiting British
India could be grouped into eight genera — Fordonia, Contoria,
Cerberus, Hypsirhina, Ferania, Homalopsis, Hipistes and Herpeton.

Giinther's genus Ferania is a synonym of his other genus Hijp-
sirhina; thus, only seven of these genera are recognized today.
Other genera of the subfamily which were not included in his work
are Gerarda, which is found in India, Burma and Thailand, and
Myron and Heurnia, which are found in northern Australia and
New Guinea.

Cope (1886) considered those snakes possessing spinous hypa-
pophyses to the caudal region and lacking enlarged anterior maxil-
lary teeth to belong to the subfamily Homalopsinae. Thus, his
grouping became large and heterogenous and contained 30 genera,
some of which are aquatic, some terrestrial, some aglyphous, and
others opisthoglyphous.

Boulenger (1890 and 1896) reaffirmed Giinther's taxonomic ar-
rangement of the genera included in the subfamily, and he included
two other genera— Gerarda Gray (1849) and Myron Gray (1849).

Jong (1926) added a tenth genus to the subfamily; this new
genus, Heurnia, inhabits New Guinea.

In the present study, the Homalopsinae is considered to contain
the seven genera proposed by Giinther (1864) and three other gen-
era— Myron Gray (1849), Gerarda Gray (1849) and Heurnia Jong
(1926).

All the trivial names proposed for the genera of Homalopsinae
and other names that are now referred to the genera are here listed,
with their nomenclatural status as recognized in this paper:

acutus (Cerberus) Gray, 1849
aer (Homalopsis) Boie, 1826
albolincata (Hypsirhina) Morice, 1875
albomaculatus (Homalopsis) Dunieril

and Bibron, 1854
alternans (Brachyorrhos) Reuss, 1834
annulata (Cantoria) Jong, 1926
atrocaeruleus (Hydras) Shaw, 1802
australis (Homalopsis) Gray, 1842

bennetti (Hypsirhina) Gray, 1842
bicolor (Gerarda) Gray, 1849
bilineata (Hurria) Daudin, 1803
blanfordi (Hypsirhina) Boulenger,

1890
boaeformis (Elaps) Schneider, 1801
bocourti (Hypsirhina) Jan, 1865

zCerbenis rynchops rynchops
-Enhydris cnhydris
zEnhydris enhydris

-Enhydris alhomactdata
-Enhydris alternans
-Cantoria annulata
zEnliydris cnJiydris
zCerherus australis

zEnhydris bennetti
-Gerarda prevostiana
zCerberus rynchops rynchops

^Enhydris maculosa
zCerberus rynchops rynchops
zEnhydris bocourti

CoLUBRiD Snakes, Subfamily Homalopsinae

hocourti soctrangcnsis (Hijpsirhina)

Bonnet, 1936
borneensis (Pythonopsis) Peters, 1872
buccattis (Coluber) Linnaeus, 1754

caenilea (EnJiydris) Sonnini and

Latreille, 1802
chahjhacus (Hemiodontiis) Jan, 1863
cerberus (Coluber) Daudin, 1803
chinensis (Hijpsirhina) Gray, 1842
cincreus (Hydras) Shaw, 1802

dayana (Cantoria) Stoliczka, 1870
decussata (Homalopsis) Schlegel, 1837
doriae (Homalopsis) Peters, 1871
doriae \ar. (Homalopsis) Steindachner,

1887
dussumieri (Eurostus) Dumeril and

Bibron, 1854

elapiformis (Hydrodipsas) Peters, 1859
elongata (Canioria) Giinther, 1864
enhydris (Hydrus) Schneider, 1799
enhydris bilineata (Hypsirhina)

Lampe, 1902
enhydris suhiaeniata (Hypsirhina)

Bourret, 1934
enhydris var. maculata (Hypsirhina)

Jan, 1868

fasciatus (Hipistes) Gray, 1849
flavescens (Heleophis) Muller, 1885
furcata (Hypsirhina) Gray, 1842

gigantea (Hypsirhina) Werner, 1923
grantii (Cerberus) Cantor, 1836

hageni (Hypsirhina) v. Lidth de
Jeude, 1890

hardwickii (Hypsirhina) Gray, 1834
hedemanni (T achyplotus) Reinhardt,

1866
herpeton (Homalopsis) Schlegel, 1837
heteraspis (Eurostus) Bleeker, 1859
horridus (Coluber) Daudin, 1803
hydrina (Homalopsis) Cantor, 1847
hydroides (Bitia) Gray, 1849

indica (Raelitia) Gray, 1842
innominata (Hypsirhina) Morice, 1875

jagorii (Hypsirhina) (Eurostus) Peters,

1863
jamnaeticus (Feranioides) Carlleyle,

1869

-Enliydris hocourti
-Enhydris punctata
^Homalopsis buccata

zEnhydris enhydris
-Fordonia leucobalia
zCerberus rynchops rynchops
-Enhydris chinensis
zCerberus rynchops rynchops

zCantoria violacea
-Enhydris alternans
zEnhydris doriae

zEnhydris punctata

zEnhydris dussumieri

zCantoria violacea
zCantoria violacea
zEnhydris enhydris

zEnhydris enhydris

zEnhydris jagorii

zEnhydris bennetti

zBitia hydroides
zGerarda prevostiana
zEnhydris enhydris

zEnhydris bocourti
zCerberus rynchops rynchops

zEnhydris punctata
zEnhydris plumbea

zEnhydris punctata
zErpeton tentaculatus
zErpeton punctata
zHomalopsis buccata
zBitia hydroides
zBitia hydroides

zEnhydris indica
zEnhydris innominata

zEnhydris jagorii
zEnhydris sieboldi

University of Kansas Publs., Mus. Nat. Hist.

kentii (Neospades) De Vis, 1889

leucobalia (Homalopsis) Schlegel,

1837
longicauda (Hypsirhina) Bourret, 1934
lusingtonii (Poiamophis) Cantor, 1836

macleayi (Pseudoferania) Ogilby, 1890
macidosa (Hypsirhina) Blanford, 1881
maculaia (Hypsirhina) Diimeril and

Bibron, 1854
maculata (Hypsirhina) (non Dumeiil

and Bibron, 1854) (Blanford, 1879)
malaharica (Hypsirhina) Werner, 1913
matanncnsis (Hypsirliina) Boulenger,

1897
microlepis (Cerberus) Boulenger, 1896
mohirus (Homalopsis) Boie, 1826
moniUs (Coluber) Linnaeus, 1758
multilineata (Hypsirhina) Tirant, 1885

obtusatus (Coluber) Reinwardt in

Schlegel, 1837
olivaceous (Homalopsis) Cantor, 1839

pahangensis (Enhydris) Tweedie, 1946
pakistanica (Enhydris) Martens, 1959
papuensis (Fordonia) Macleay, 1877
plumbea (Homalopsis) Boie, 1827
polylepis (Hypsirhina) Fischer, 1886
prevostianus (Coluber) Eydoux and

Genais, 1837
punctata (Phytolopsis) Gray, 1849
pythonissa (Coluber) Daudin, 1803

rhynchops (Python) Merreni, 1820
richardsonii (Myron) Gray, 1849
russelli (Cerberus) Cuvier, 1837
rynchops australis (Cerberus) Love-
ridge, 1948
rynchops (Hydrus) Schneider, 1799
rynchops novaeguincae (Cerberus)
Loveridge, 1948

schneideriana (Hurria) Daudin, 1803
semizonata (Homalopsis) Blyth, 1855
sieboldii (Homalopsis) Schlegel, 1837
smithi (Hy))sirhina) Boulenger, 1914
sidjalbidus (Coluber) Gmelin, 1788

tentacidatus (Erpeton) Lacepede,

1800
trilineata (Hypsirhina) Gray, 1842

=Mi/roM richardsoni

=^Fordonia leucobalia
^Enhydris longicauda
=zEnhydris enhydris

^^Enhijdris macleayi
^^Enhydris maculosa

=:Enhydris bcnnetti

^Enhydris maculosa
^Enhydris dussumieri

=Enhydris matannensis
z=Cerberus microlepis
zzzHomalopsis buccata
^Homalopsis buccata
=:Enhydris bocourti

=Cerberus rynchops rynchops
^Enhydris enhydris

=^Enhydris pahangensis
^Enhydris pakistanica
=zFordonia leucobalia
=:Enhydris plumbea
^Enhydris polylepis

=zGerarda prevostiana
^EnJiydris punctata
^^Enhydris enhydris

=zCerberus rynchops rynchops
=^Myron richardsoni
=iCerberus rynchops rynchops

^Cerberus australis
:=Cerberus rynchops rynchops

^^Cerberus rynchops novacguineae

^Cerberus rynchops rynchops
:=Homalo))sis buccata
^Enhydris sicboldi
^Enhydris smithi
^Homalopsis buccata

-Erpeton tentacidatus
-Enhydris enhydris

CoLUBRiD Snakes, Subfamily Homalopsinae 61

unicolor (Cerberus) Gray, 1849 =^Homalopsis buccata

iinicohr (Fordania) Gray, 1849 =Fordonia leucobalia

variabilis (Fordonia) Macleay, 1878 =Fordonia leucobalia

ventromacidata (Heurnia) Jong, 1926 ^^Heurnia ventrotnaculata
violacea (Cantoria) Girard, 1857 =Catitoria violacea

Key to the Genera

1. Ventrals broad, not keeled 2

Ventrals narrow, bicarinate 9

2. Nasals in contact with one another 3

Nasals separated by internasal 5

3. Dorsal scales smooth; parietals entire Enhijdris

Dorsal scales keeled; parietals entire or fragmented 4

4. Parietals fragmented; dorsal scales in 21-29 rows Cerberus

Parietals entire; dorsal scales in 39-47 rows Homalopsis

5. Nasal cleft absent 6

Nasal cleft present 7

6. Anterior mandibular teeth longest; body extremely

elongated _- Cantoria

Mandibular teeth subequal; body moderately long Gerarda

7. Dorsal scales keeled, in 19-21 rows Myron

Dorsal scales smooth, in 25-29 rows 8

8. Nasal semi-dixided, nasal cleft extending to prefrontal;

dorsal scales in 25-29 rows Fordonia

Nasal completely divided; dorsal scales in 27 rows Heurnia

9. Nasal completely divided; dorsal scales smooth; tail

feebly compressed; no rostral appendages Bitia

Nasal semi-divided; dorsal scales keeled; tail ovoid;

rostral appendages present Erpeton

Genus Enhydris Sonnini and Latreille

Hijdrus Schneider, (in part), Historiae Amphibiorum, 1:233, 1799.

Enhydris Sonnini and Latreille, Histoire natnrelle des reptiles, 4:200, 1802
[Type species. — Enhydris caendea = Hydrus enhydris Schneider (not of
Merrem, 1820)]. Smith, Fauna of British India . . . Reptilia and Amphibia
3:380, 1943.

Hypsirhina Wagler, Natiirliches System der Amphibien, pp. 132, 169, 1830
[Type species. — Homalopsis aer Boie, 1827]. Gray, Zoological miscellany,
p. 66, 1842; Catalogue snakes British Museum, p. 71, 1849. Dumeril and
Bibron, (in part), Mem. Acad. Sci. France, 23:495, 945, 1853. Giinther,
Reptiles of British India, p. 280, 1864. Jan (in part), Arch. Zool. Anat.

62 University of Kansas Publs., Mus. Nat. Hist.

Phys., 3:258, 1865. Boulenger, Fauna of British India . . . Reptilia and
Batrachia, p. 375, 1890; Catalogue snakes British Museum, 3:2, 1896.
Rooij, Reptiles Indo-Australian Archipelago, 2:179, 1917. Bourret, Serpents
del'Indochine, 1:275, 1936.
Potamophis Cantor, Trans Med. Phys. Soc. Calcutta, 8:139, 1836 [Type species.
— Potamophis lusingtorii Cantor, 1836].

Homalopsis Schlegel, (in part), Essai sur la physionomie des serpents, 2:332,

1837.
Ferania Gray, Zoological miscellany, p. 67, 1842 [Type species. — Homalopsis

sieboldii Schlegel, 1837].
RacUtia Gray, Zoological micellany, p. 67, 1842 [Type species. — Raclitia indica

Gray, 1842].
Miralia Gray, Zoological miscellany, p. 68, 1842 [Type species. — Brachyorros

alternans Reuss, 1834].
Hypsiscopus Leopoldo, Systema Reptilium, 1:25, 1843 [Type species. —

Homalopsis plumbea Boie, 1927]. Gray, Catalogue snakes British Museum,

p. 72, 1849.
Pelophis Leopoldo, Systema Reptilium, 1:25, 1843 [Type species. — Brachyor-

rhos alternans Reuss, 1834].
Pythomorphus Fitzinger, Systema Reptilium, p. 25, 1834 [Type species. —

Homalopsis sieboldii Schlegel, 1837].

Phijtolopsis Gray, Catalogue snakes British Museum, p. 67, 1849 [Type species.
— Phytolopsis punctata Gray, 1849].

Eurosttis Dumeril, Prodome classification ophidiens, p. 498, 1852 [Type species.
— Eurostus dussumierii Dumeril 1852]. Dumeril and Bibron, (in part),
Mem. Acad. Sci. France, 23:498, 951, 1853.

Trigonurus Dumeril, Prodrome classification ophidiens, p. 499, 1852 [Type
species. — Homalopsis sieboldii Dumeril, 1852]. Dumeril and Bibron, Mem.
Acad. Sci. France, 18:498, 1853; firpetologie generale . . . reptiles. Paris,
7:959, 1854.

Tachyplotus Reinhardt, Vidensk. Meddel., p. 151, 1866 [Type species. —
Tachyplotus hedemanni Reinhardt, 1866].

Feraniodes Carlleyle, Jour. Asiatic Soc. Bengal, 38:192, 196, 1869 [Type
species. — Feranioides jamnaetica Carlleyle, 1869].

Pythonopsis Peters, Monats. Akad. Wiss. Berlin, p. 576, 1871 [Type species. —
Pythonopsis borneensis Peters, 187 l=Phytolopsis punctata Gray, 1849].

Homalophis Peters, Monats. Akad. Wiss. Berlin, p. 577, 1871 [Type species. —
Homalophis doriae Peters, 1871].

Pseudoferania Ogilby, Proc. Linn. Soc. New South Wales, ser. 2, 5:51, 1890
[Type species. — Pseudoferania macleayi Ogilby, 1890].

Dieurostus Berg, Com. Mus Nac. Buenos Aires, p. 290, 1901 [Substitute name
for Eurostus Dumeril, 1853, preoccupied].

Diagnosis. — Body usually cylindrical; head depressed, slightly
distinct from neck; eye small, pupil vertically elliptical; head shields
large; nasals in contact behind rostral, semi-divided; nasal cleft
extending from nostril to first or second labial; internasal single or
divided; loreal present; parietals entire; ventrals not keeled; tail
moderately long or short; scales smooth, in 19-33 rows at midbody;
ventrals rounded; maxillary teeth 10-16, followed by interspace and
two enlarged, grooved teeth.

Description and variation. — The small eyes, situated close to
each other on top of the head, have vertically elliptical pupils. The

CoLUBRiD Snakes, Subfamily Homalopsinae

rostral is as broad as, or broader than, it is high. The nasals form a
median suture behind the rostral. The nostril is directed upwards,
crescent-shaped and con\'ex anteriorly. The nasal cleft extends lat-
erally to the first or second upper labial; in Enhydris dussumieri and
E. pokistonica the nasal cleft extends to the internasal. The nasal is
usually semi-divided, but in E. doriae, it is completely divided. The
internasal is situated behind the nasals. The presence of a single or
divided internasal is a diagnostic specific character, except in E.
madeaiji and E. poJylepis in which the internasal is single in some
specimens and divided in others. Enhydris enhydris and E. bocourti
usually ha\e a divided internasal, but in some specimens the inter-
nasal is entile. For each, whether or not the internasal is in contact
with the loreal is a diagnostic specific character, except in E.
bocourti, E. cdbomaculata, and E. pakistanica, in which both condi-
tions exist. The prefrontals form a median suture behind the inter-
nasal and are not in contact with the upper labials, except in some
specimens of E. polylepis, in which the prefrontals extend laterally
between the loreal and preocular to the labials. The loreal is single
in all species; rarely it is divided, a condition which can be consid-
ered as abnormal. The loreal is in contact with two or three upper

Table 1. — Variation in the Number of Upper Labials Entering Orbit in the

Species of Enhydris

4 4 4-5 4 5 5 5-6

4 4-5 4-5 5 5 5-6 5-6 Other Total

Species

E. indica 2

E. altcnians 8

E. phunbca 14

E. jagorii 16

E. enhydris 140

E. innominata - 4

E. smitJii 3

E. longicauda __ 2

E. bennetti 7

E. chinensis 20

E. macleayi 2

E. pohjlepis —

E. maianensis - —

E. maculosa 16

E. bocourti 14

£. dussumieri __

E. sieboldi

£. albomaculata
E. pahangensis ..
E. pakistanica -

E. punctata

E. doriae

2
7
2
1
37

1
15

— 3

— 1

— 2

1
4

140

(
24

4
10

1
16
17

8
14

1
52

7
12

6/6=4; 6/7=1; 7/7=2.

7/7=2; 7/8=2; 8/8=5; 8-9/8-9=3.

—

140

. — .

. —

— .

—

14**

64 Unr'ersity of Kansas Publs., Mus. Nat. Hist.

Table 2. — Variation in tlie Number of Upper Labials in the
Species of Enhtjdris

Species 7-8 8-8 8-9 9-9 Other Total

E. indica — 2 — — — 2

E. alternans — 10 — —

E. plumhea — 40 — —

E. jagorii — 16 — —

E. enhydris _. — 140 — —

E. innominata — 4 — —

E. smithi — 2 11

E. lotigicaiida — 2 — 1

E. bennetti — 7 — —

E. chinensis — 24 1 —

E. macleayi — 3 — 1

E. polylepis — 4 — 5

E. matanensis — 1 — —

E. maculosa 1 15 — —

E. bocourti 1 14 — 2 —

E. dusstimieri — 2 — —

E. sieboldi — 8 — —

E. albomaculata — 2 2 9

E. pahangensis — 1 — —

£. pakistanica — 36 9 8

E. punctata — — — —

E. doriae — — — —

' 11/11=1; 12/12=3; 13/13=2.

** 13/13=1; 13/14=1; 14/14=5; 14/16=1; 1.5/15=5; 15/16=1.

Table 3. — Variation in the Number of Labials in Contact with the Loreal in

the Species of Enhydris

1-2 1-3 2-3 1-4 2-4 1-5

Species 1-2 1-3 2-3 1-4 2-4 1-5 Other Total

E. indica — — 2 — — — — 2

E. alternans — — 10 — — — — 10

E. plumbea — 5 17 — — — 1 23

E. jagorii — 16 — — — — — 16

E. enhydris — 130 — — — — — 130

E. innominata — 4^ — - — — — — 4

E. smithi — 4 — — — — — 4

E. longicauda — 3 — — — — — 3

E. bennetti — 7 — — — — — 7

E. chinensis — 20 — — — — 1 21

E. macleayi — — 2 — 2 — — 4

E. polylepis — — 3 — 6 — — 9

E. matanensis — — 1 — — — — 1

E. maculosa — 16 — — — — — 16

E. bocourti 7 7 — — — — — 14

E. dusstimieri — 2 — — — — — 2

E. sieboldi — 8 — — — — — 8

E. alhomacidata .... — — 5 — 7 — — 12

E. pahangensis — — 1 — — — — 1

E. pakistanica 3 50 — — — — — 53

E. punctata — — — 4 — 2 — 6

E. doriae — — — 2 — 6 — 8

CoLUBRiD Snakes, Subfamily Homalopsinae 65

Table 4. — Variation in the Number of Lower Labials in the Species of

EnJujdris

f— 1

>—(

1—!

i-H

1—i
1

I-H

1-H

1—1

•4-'
O

Species

1— H

1 — 1

I-H

£. indica 2 — — — — — — — — — 2

E. altcmans 1 2 5 — — — — ___ g

E. plumhca — 1 20 12 — — — — — 24

E. ja'fioiii — — 1 — 6 — — — — 1 8

£. cuhydris — — 17 1 52 4 63 — — — 137

E. innominata — — 2 — — — — — — — 2

jE. smithi — — — — 2 — 2 — — — 4

E. longicauda — — — — 2 1 — — — — 3

E. hcnnetn — 4 2 — — — — — — 1 7

E. chincnsis — 1 11 1 5 — — — — — 18

E. macJcai/i — — — — — — 4 — — — 4

£. polylepis — — — — 4 2 2 — — 1 9

E. nmlanensis — — — — — — — - — — — —

£. maculosa — 110 2 1 — ___ — 14

E. hocourti ____ 2 2 2 — — 7° 13

E. dussumieri — — — — 1 — — — 1 — 2

E. sieholdi — — — — 1 — 5 — 1 — 7

E. alhomacidata ____i_227 1 13

E. pahangensis — — — — 1 — — — — — 1

E. pakistanica ____ 9 821 — — 3 41

£. punctata ____ i_ 3_ 2— 6

E. doriae _________ 14»* 14

* 14/14=2; 15/15=3; 16/16=2.

*' 15/15=1; 15/16=1; 16/16=1; 16/18=1; 17/17=4; 17/18=2; 18/18=4.

Table 5. — Variation in the Number of Labials in Contact with the Anterior
Chin-shields in the Species of Enhtjdiis

1-2 1-3 2-4 1-4 1-4 1-5 1-5 1-6 2-5 2-6

Species 1-2 1-3 2-4 1-4 1-5 1-5 1-6 1-6 2-5 2-6 Total

£. iiidica — — — 2 — — — — — — 2

£. altcmans — — — — — 10- — — — — 10

E. plumhea _ _ _ 2 1 21 — 1 — — 25

E. jagorii — 2 — 14 — — — — — — 16

£. cnht/dris _ 86 — 44 — — — — — — 130

E. innominata — — — 2 — 2 — — — — 4

E. smithi — — — 1— 3 — — — — 4

E. longicauda — — — 3 — — — — — — 3

£. bennetti — 1 — 6 — — — ■ — — — 7

£. chincnsis — — — 19 — 2 — — — — 21

E. macleayi 2 2 — — — — — — — — 4

E. polylepis — 8 — 3 — — — — — — 11

E. matanensis — — — — — 1 — — — — 1

£. maculosa — 3 — 13 — — — — — — 16

£. hocourti _ _ _ _ 1 13 1 1 — — 16

E. dussumieri — — — 2 — — — — — — 2

E. sieholdi — 1 — 7 — — — — — — 8

£. alhomacidata — — — — — 4 1 9 — — 14

E. pahangensis — — — 1 — — — — — — 1

£. pakistanica 2 20 — — — — — — — — 22

£. punctata — — — 4 — — — — 3 — 7

E. doriae — — 1 — — — — — 7 — 8

University of Kansas Publs., Mus. Nat. Hist.

I>3

o
Q

g
Z

•43

o
H

CO
CM

»n

(M t-

rJH

-t

-H in

1 — 1

i-H

C<I o co^^^

-*-*< cot-\^o

1 — 1

CO 03(^00

•-H ^^^

1^-^

.— ( CO CO ^

i—(

I — 1 I — 1

»o

i-H

,-HV©

1— 1

I 00 CO 1 CO

CO CO I (M '

i(MintM05^'*ico i^rrioo

,— ( ^ Tfi I <M

CO in

oq o 00

—I CO

o o Tfi ^ CO t^ 1— I I in '

CM -r in

C-l

tt CO

OOOl^OC'f^'tCOi— I

t^ ^ 1 m

X m '

CO — I "* in

-H I CO

.—I I lO 1— I

O I CO

CO I

2 «

o r- =

5 ? 5 IS "^ ~

tj(j'^ 5 s ~ 5j

a ^ =^ t~ o _!n

Co •:» 00 ~ O -..a

2 a ~ ^ c i c
;;* ^ ^ r- '^ •— ^

-?2 2
o 5 ~

g"Q 2 ^5 a

?> S o

?3 a =

~ s "— ^

":? 3 S- ::x s.'tS

bjbqi4)tqtqijqtqtqtqtq[x:]bqijqtqbj(t^bqbjtq[ij[]qtq

CoLUBRiD Snakes, Subfamily Homalopsinae

-s3

C/3

a
en

Li
c

c
o

oi
>

Q
u

(72

c
cs

1 — 1

CO ^

o o

cc to

CM CD O

00 00 ro

cot-t-

(-0

i>- 00 ■-; i> q t-; o

in CO o i> 00 1^ --!

CO CM CO in o '— < CO

in t^

t- I CO

t>pCD

-t< in -t*

in CO

^ o

—Id

00 CM

oy o

t- fM

t~ OJ O

in I oi I in

o I CO I in

>-i d d

C5 CD
00 p

CO 05

CO
(M

cq (M ' ^ CM

O-l p p p 05

^ 00 in d oi

-f oi -p in -*

in .—I

p I "*
i-i I in

r-(Ot>

ooi-iin

inrf tP

m ^ 05 00 o '
t^ in oi CO t^ '

I — ( -t CO CO oi CO
I oi CO in in in in

in 1—1 1- CD t^

in CO "f in 'f

^ cDin
o m in

C5 oQ 05 -f in m 00

Ol — I r-H CO O — I OI

05 00 — ^
CO -f f

CO -f COCO o

OI OI -t"* "*

cr>-r Oi
in CO CO

( Tfi OI t^ in oq -f OI .

CO -*

<cOOIt^005X01inoOOt^^l:^
^H o^

M
C

oq ' oi

CO
CO

1— I i> -^ p o

-t< CD oi — < -f
Tfi OI CO CD ^

CO "^i OI CD OI CD
t^ CD CO '^ t- ^

.— ( I— t 1— I 1— I 1 1 T— I

I r I I I

I t- -t — I 05 OO

I OI -H OI 't* O

—I o
in cj5

i-i CO

o in
-t i>.

oi in

o in

00 Ol

^ oi

CD 05
p 00

oi oi

CD — < ^t-

in Ol

t-CD

O CD 05 Ol

CD -t --P -t

1 t-'o6

' Tf' Ol

t-'o6

-t ^ O 00 l^ — < Ol

CD in in -f CO in CO

00 CO 00 r-
in CO ^ CO

CO CO

t Ol

Ol ' Ol

in
oi

■^_ in oo
05 oid

in ^ 'i^

CO ■— I CO Ol CD CD

in m Ol CD "^ ^

'i* CO

in c:5i:^
in CO CO

icot-oocooiooin

* CO CO I

iCDoocoGO^coo^in

^ Ol

C/2

w 6? ^ ■^^ . — -?-. ^

c4l*JbJliJWWWWWI*JI*JWIiJWWI*lW«WKll*lW

University of Kansas Publs., Mus. Nat. Hist.

'£

"Ta

lii

ca
u

C/3

■c

CO CD (M

CCOOJ

122

1?:

o
1 in

O-HO

I— 1

--^O

lOCOO 1 1

1 ^■

OC 00
1 CCI>

-t CO

a> o-i CO ' '

1 lO

' ^1>

t> TT

inco C75 1

00
1 "*

CDt-
00 t>

CD t-

1 °°

CD .-( 't' '

1 oq

'^'oi

CO '-^

' 1—1

in ^ Tf 00 in o in

00 in o

CD CD

^ 1^

.-ir-oo

t- CO ^ -H Tf Tf t^

oi CO in CD T}H in CO

^ 00 --1
^CO^

i-iin

'^CO

in CO

Tf<-t CO
J>CO^

0OCD0OCOl>CDCD-rint-^l>

(Mcocoint^^int-^niTfiTti

CO 05 o o CO oq .— I
oq cq in in '^ in CD

I Tf ^ in in oi CO CM

CO '^

X CD CD
CO COCO

I I

c<i oq

coco

CD O

m '^

CD CD

"* CO

t> a> ^

i>coin

I I 1

Ol^ r-H

< CO oq t^ I t^ CO

I m 00 ' (M^f CO
cq

C/D

C5 ■

i CO
CO

cox t^

o CD oq
X G^ oq

oq^^ in

CD oq
oq oi

CO 'i^
coco

oq
oq

mm o c:2 o

CO oq t- CO

m i-H

\r^M^

CD00500

CO ^ mi> m

oq t^ CO ^

m -T ^ '^

oq r-H

.-hX

m'*

X
X

Tf 05 com X oq

CO co^ CO t^m

I I I < I

I m m -T m X

I coco m m "^

CD -^ CO CO CO 05 1>

m m -^ 'I' 't' m M^

o CO oq o

m ^ ^ ^

xt-

'^CO

oq o 1-H oq ^ CO
mmm otj mm

^ CD \ ^ I i-H

m^ I X I m

^HCocDt^cDoqcom^oqco^^cDX

I— ( CD "—I

oqx.

I oq ^ CO
oq

2 «

•4-j
2

■*^

•2 "5

i ?

S*j

*«*

iculata —

^cnsis

nira

a i

[4jbqi*jtijtqtt3Kl>iJiJJWiJJWiiJiJJiJJi*iWbqbjtqiJji*i

CoLUBRiD Snakes, Subfamily Homalopsinae

tn
O

C/3

a
>

o
c

G
o

Q
U

c
a

CM in 05

-* 0000 1

946

(MCO

COO

05T}<

ooo

^ cq

0 0(M

CO as ^

00 ^ X

o -r '-t^
CO CD en

^H oi rH

CD
05

t^05

—1 00

|§

TfM

CDO

1 CO

■^ 00 ^ 05 (M CT) p

^ 'i^' c<i CO -r CO 'H
^ --H cq CM cq c<i CO

t>; in M o 00 C7; o 00

r-H ^ t-- lO t^ CD lO CD
^ -H — I Ol CM CM (M CO

I -r c>i a> CD in t— cm

I -H CO I> C5 — < CM in
1—1 I— I 1—1 f— I (M CM CM

inooc:5
in CO CD

cDi>:in

^^ ^^ ^^
xqci

(>i CD CO

00 -^

in CO

CM d

05 -t

■^ CM 00
(M ■— I I ^

^ in

dco

CM— I

t-r-co

CM CM CD

CO >— I 1—1

I m m CM O CM 05 05

CO in CM ^ '*

CM ^H -H CM -H

CM CD

in CO

t-^co

'-P CM 00

CO ^H •— I
I I I

O -H 00

ci ci -f

CM ^ --H

iCMCMin^CMfCMOCOCMX

CM f -H

l^l>-HCOt:^ ' 05 "* CD

0)
bO
C
a
P5

"* O CD
05 05 05

^ CD c:5
in CO CD

^H in CM

^ p CO p o:;

t" CD CD 00 in
^H ^H CM CM CM

t> Ol p p t>; CO

CO in X 05 -t t-'

1— I ^^ —I CM CO CM
t I I I I

I -f" in -H o] in
I oi CO CO oi t'

1— I ^ CM CM CM

CM —I

CMp

00 I>^

coin

CO CO
X CM

■^in
in in

in o ' in CO

-t CO CM co_

— < X 05 05

Cq ^H ^H ^H

CO in

-r 05

Tt< o

'c:5

Xlr-

•-^ CM

CO -*

! 05

CM.

00 X -f p ^ P p

CO d 05 -H CD CD d
CM -H ^H CM CO CM CM

in 05 ^ "i^.
d in d CD

00 ^_

CM r^

X p p CM p f _

cm' d co' •^' d CO

CM CM CM "* ^ CM

II I I

O CM I p , '-P

dt-^ I 00 I CO

CM ^ CO CM

faqbjbjbjtijhjlij^tijWIiJtjqyfaqWWWWttJfaJWKi

70 University of Kansas Publs., Mus. Nat. Hist.

labials, except in E. punctata and E. doriae, in which the loreal is in
contact with more than three upper labials ( the exact ordinal num-
bers of labials in contact with the loreal, are given in the account of
each species). The preocular is single (Tables 1-6). Except for E.
doriae, suboculars are absent. Usually two postoculars are present,
except in E. indica which has only one. The numbers of temporals
usually are one, two, and three in the primary, secondary, and terti-
ary rows. Variations exist in which the primary row has two tem-
porals, the secondary row one or three, and the tertiary row two,
four, or none. Usually the primary temporal is larger than the oth-
ers. The upper secondary or tertiary temporal occasionally is an
elongated plate. Of the 22 species in the genus, 20 have 7-9 (8.2)
upper labials. The sixth and/ or seventh are the largest and usually
are undivided, except in E. madeayi, E. bocourti and E. sieholdi,
in which the next to last upper labial is divided. Enlujdris punctata
and E. doriae have 12-15 (13.2) upper labials. The anterior labials
are narrow; those labials posterior to, and including the labial
ventral to the eye, are divided into two or three tiers. The anterior
chin-shields are in contact with the first three to six lower labials,
except in E. punctata and E. doriae in which the chin-shields are in
contact with second to fifth or sixth lower labials. In these two
species the first and second pairs of lower labials are in contact and
form a median suture. The anterior chin-shields are usually larger
than posterior chin-shields. The posterior chin-shields are separated
from each other by a pair of small scales, except in E. polylepis, in
which they are separated by the anterior chin-shields. The variation
in the number of gulars and incomplete ventrals that separate the
chin-shields from the first complete ventral are given in the account
of each species. The number of ventrals varies from 105 to 175 in
the genus. The species having the lowest number of ventrals are E.
innominata (105-116) and E. smithi (118-127); the species having
the highest numbers arc E. indica (173-175) and E. pakistanica
(153-162) (Table 7). The number of subcaudals varies from 23 to
82 in the genus. The species having the lowest numbers of sub-
caudals are E. alternans (23-29) and E. indica (31-38); the species
having the highest numbers are E. longicauda (61-74) and E.
pakistanica (70-92) (Table 8). The umbilical scar is situated ante-
rior to the vent, on or between the 13th to 29th ventral. The dorsal
scales are smooth, lacking pits and in 19-33 rows; the scales in the
first row are usually larger than the others.

The maxillary teeth vary in number from 10-16, followed by an

CoLUBRiD Snakes, Subfamily Homalopsinae 71

interspace and two slightly cnlargcxl, grooved teeth posteriorly. The
19-25 mandibular teeth slightly decrease in size posteriorly.

The length of the hemipenis extends to the level of the 7th to
14th subcaudal; the point of bifurcation varies from the level of the
2nd to 11th subcaudal. The reti-actor penis magnus muscle origi-
nates on a caudal vertebra between the 20th and 35th subcaudal.

Males have more subcaudals and a longer tail/ snout- vent ratio
than females; usually the coefficient of difference between the pair
of means in each species is not significant ( Table 9) .

Intrageneric variation of head scutellation and dorsal scale row
counts of Enlujdris are summarized in tables 1-6.

Distribution. — The western end of the range of the genus is
West Pakistan and the eastern end is Queensland, Australia. The
genus is not represented from the Philippines, although it occurs in
Formosa and as far north as Nanking on the mainland of China.
Fifteen of the 22 species belonging to this genus occur in the south-
east corner of the Asiatic continent, including Malaya, Thailand,
Cambodia, Laos, South and North Vietnam.

Remarks. — The nomenclatural status of the genus Enhydris was
discussed by Stejneger ( 1907), who stated:

The authors who employ the generic name, "Enliydris, Merrem, 1820," for
two of the Hydrid snakes which should correctly stand as Lapemis, seem to
have overlooked entirely the fact that Latreille, as early as 1802, established
the genus Enhydris for an entirely different set of snakes, embracing all
Schneider's species of Hydrus, from H. caspius on. This name, which can not
be discarded, as it is not a synonym of any name pre\'iously giNen, must have
for t>pe one of the six species enumerated by Latreille, and as neither he nor
any of his successors have designated a type for the name, we are obliged to
resort to the method of elimination. The last species mentioned l)y Latreille,
Enhydris dorsalis, is of dubious application. The third Enhydris rynchops, is
tlie type of Daudin's Hurria (1803), Goldfusz' Strephon (1820), and Cuvier's
Cerberus (1829), while three other species, viz., E. caspius, piscotor, and
pahtstris belong to the genus Natrix (Tropidonotus). A single species is thus
left as type, viz., Enhydris caerulea, which is Schneider's Hijdnis enhydris.
This species, or rather a synonym of this species, was afterwards (1830) made
the type of Hypsirhina, and there can be no doubt but that the latter name
v\'ill have to give way to Enliydris Latreille. The same result is obtained by
those who would fix the t>'pe by employing the "principle of tautonymy."

Key to the Species

1. Only first pair of lower labials forming median suture;

upper labials posterior to eye not divided 2

First and second pairs of lower labials forming median

suture; upper labials posterior to eye divided 21

72 Unr'ersity of Kansas Publs., Mus. Nat. Hist.

2. Dorsal scales in 19 rows; loreal in contact with second

and third upper labials 3

Dorsal scales in more than 19 rows; loreal in contact

with first three upper labials 5

3. Internasal divided; prefrontals larger than nasals; ven-
trals less than three times width of adjacent dorsal

scale of first row; one or two postoculars 4

Internasal single; prefrontals smaller than nasals; ven-
trals more than three times width of adjacent dorsal
scale of first row; two postoculars E. plumhea

4. Nasal cleft extending to first upper labial; postocular
single; anterior chin-shields in contact with first four
lower labials; nine upper labials; umbilical scar on, or
between, ventrals 13-14 anterior to vent; ventrals 173-

175; subcaudals 31-38 E. indica

Nasal cleft extending to second upper labial; two post-
oculars; anterior chin-shields in contact with first four
or five lower labials; 10 or 11 upper labials; umbilical
scar on, or between, ventrals 21-24 anterior to vent;
ventrals 125-164; subcaudals 23-39 E. alternans

5. Dorsal scales in 21-23 rows 6

Dorsal scales in 25-33 rows 15

6. Internasal in contact with loreal 7

Internasal not in contact with loreal 11

7. Fourth upper labial entering orbit; loreal in contact
with first to third upper labials; posterior chin-shields

separated by pair of small scales 8

Fourth and fifth, or fifth and sixth upper labial enter-
ing orbit; loreal in contact \\'ith second and third, or
second to fourth, upper labials; posterior chin-shields
separated by anterior chin-shields -E. pohjlepis

8. Body increasing in size posteriori)- to le\el of \'ent; dor-
sum and sides having black spots or blotches; anterior
chin-shields larger than posterior chin-shields and in

contact with first four or five lower labials 9

Body almost cylindrical; dorsum and sides having black
longitudinal stripes; anterior chin-shields smaller than

or same size as posterior chin-shields, and in contact

with first three or four lower labials E. enhydris

CoLUBRiD Snakes, Subfamily Homalopsinae 73

9. Large blotches at least three scales long and five scales

wide on sides of body and extending onto ventrals . 10

Small spots no larger than two scales in length and
three scales in width on sides of body; third and fourth
rows of dorsal scales and ventrals yellow; a black lon-
gitudinal band on first and second rows of dorsal scales
and lateral edges of ventrals E. jagorii

10. A middorsal and a pair of dorsolateral rows of black
spots on dorsum; anterior chin-shields in contact with
first five lower labials; prefrontal slightly broader than
supraocular; loreal single, a little longer than high; one
high preocular, not extending below eye; ventrals 105-

116; subcaudals 40-52 E. innominata

A single row of large middorsal spots on anterior third
of body; yellow annuli forming festoons; anterior chin-
shields in contact with first four or five lower labials;
prefrontal not much broader than supraocular; loreal
single or divided, as long as high; one (rarely two) pre-
ocular partly extending below eye; ventrals 118-127;
subcaudals 54-56 E. smithi

11. Fourth upper labial entering orbit; loreal in contact

with first three upper labials 12

Fourth and fifth, or fifth and sixth, upper labials enter-
ing orbit; loreal in contact with second and third upper
labials 14

12. Internasal divided; subcaudals 61 or more E. longicauda

Internasal single; subcaudals 56 or less 13

13. Dorsal scales in 21 rows; ventrals 158 or more E. hennetti

Dorsal scales in 23 (rarely 25) rows; ventrals 153 or

less E. chinensis

14. Dorsal scales in 21 rows; anterior chin-shields in con-
tact with first five lower labials; ten lower labials, none

of upper labials divided E. mantannensis

Dorsal scales in 21 or 23 rows; anterior chin-shields in
contact with first three lower labials; twelve lower labi-
als, second to last upper labial di\'ided E. macleayi

15. Dorsal scales in 25 rows; internasal single 16

Dorsal scales in more than 25 rows; internasal single or
divided 17

74 Unr'ersity of Kansas Publs., Mus. Nat. Hist.

16. Nasal cleft extending to first upper labial; seven (rarely
eight) upper labials, the first four or five in contact with
anterior chin-shields E. maculosa

Nasal cleft extending to second upper labial; nine up-
per labials, first four in contact with anterior chin-
shields E. pahangensis

17. Nasal cleft extending to first upper labial 18

Nasal cleft extending to internasal 19

18. Chin-shields present; ventrals more than two times
width of adjacent dorsal scale of first row; dorsal scales

in 27 rows E. dussumieri

Chin-shields absent; ventrals less than two times width
of adjacent dorsal scale of first row; dorsal scales in 29
rows E. pakistanica

19. Ventrals about five times width of adjacent dorsal
scale of first row; loreal not in contact with the first
labial; prefrontal as large as nasal; dorsal scale in 27

rows E. albomaculata

Ventrals about three times width of adjacent dorsal
scale of first row; loreal in contact with the first
labial; prefrontal larger than nasal; dorsal scales in 27
or 29 rows 20

20. Internasal single, rarely divided; fifth (rarely fourth)
upper labial entering orbit; anterior chin-shields in
contact with first five lower labials; loreal in contact
with first two or three upper labials; ventrals less than

136 E. bocourti

Internasal divided; fourth (rarely third) upper labial
entering orbit; anterior chin-shields in contact with
first four lower labials; loreal in contact with first three
upper labials; ventrals more than 143 E. sieholdi

21. Dorsal scales in 25 rows; nasal larger than prefrontal,
partly divided by nasal cleft; internasal single, not in
contact with loreal; loreal in contact with first upper

labial E. punctata

Dorsal scales in 31-33 rows; prefrontal larger than
nasal and completely divided by nasal cleft; internasal
divided, in contact with loreal; loreal not in contact
with first upper labial E. doriae

CoLUBRiD Snakes, Subfamily Homalopsinae 75

The Enhydris plumbea Group

This group, composed of three species — plumbea, alternans and
indica — is characterized by a short, rounded snout; indistinct head;
loreal in contact with second and third upper labials; eight upper
labials, fourth entering orbit; dorsal scales in 19 rows at midbody;
dorsum dark brown or black, venter brown or pale black, dorsum
and venter having narrow yellow transverse bands or not.

Members of this group range from Burma eastward to the south-
eastern part of China, including the continental islands of Hainan
and Formosa, and southeastward through the Greater Sunda Islands
to Celebes.

Enhydris indica (Gray)

Raclitia indica Gray, Zoological miscellany, p. 67, 1842 [Svntvpes. — BMNH
1946.1.2.36, 1946.1.2.46, from Malay Peninsula; Gen. Har'dwicke collector];
Catalogue snakes British Museum, p. 79, 1849.

Hypsirhina indica; Boulenger, Catalogue snakes British Museum, p. 4, 1896.
Flower, Proc. Zool. Soc. London, p. 887, 1896; Proc. Zool. Soc. London,
p. 675, 1899.

Enhydris indica; Tvveedie, Snakes of Malaya, p. 89, 1957.

Dio gnosis. — Nasal cleft extending from nostril to first labial; two
internasals, in contact with each other and with loreal; fourth upper
labial entering orbit; one postocular; anterior pair of chin-shields in
contact with first four lower labials; dorsal scales in 19 rows at mid-
body; ventrals 173-175.

Description and variation. — The rostral is broader than high and
is narrowly visible from above. The nasal is almost hexagonal, a
little larger than the internasal. The latter is divided, and lies
behind the nasals. The prefrontals are in contact with each other
and are larger than the nasals. The frontal is a hexagonal plate
having parallel fronto-supraocular sutures. The parietals are about
twice as long and not quite two times as broad as the frontal. The
loreal is single and is in contact with the second and third upper
labials. A large supraocular covers almost half the orbit of the eye
and is not nearly as broad as the frontal. There is one small
preocular; one large postocular, which is twice the size of the
preocular, is twice as long as high. One large primary temporal is
followed by two small secondary temporals (Fig. 3). Of the eight
upper labials, the fourth enters the orbit, and the seventh is the
largest; and none of them is divided. Of the nine lower labials, the
first four are in contact with the anterior pair of chin-shields, which
are about twice as long and broad as the posterior pair. The
ventrals are narrow with rounded edges. The umbilical scar is

UXR'ERSITY OF KANSAS PUBLS., MuS. NaT. HiST.

situated on \ entrals 13 and 14 anterior to the \'ent in the male
specimen and on ventral 16 in the female specimen. One male
(BMNH 1946.1.2.46) and one female (BMNH 1946.1.2.36) have 173
and 175 ventrals and 34 and 28 subcaudals, respectively. Addi-
tional records (Smith, 1930:61) of two males from Taiping (Selangor
Museum) and from Singapore (Dublin Museum) have 165 and 175
ventrals and 31 and 34 subcaudals, respectively. The body at its
greatest width is 8 mm. from side to side. The dorsal scales are in
19 rows, and sometimes reduce to 17 rows just before the vent. The
dorsal scale reduction for one female (BMNH 1946.1.2.36) is,

3+4(4)

21 19(178).

3+4(4)

There are 12 maxillary teeth, of which the last two are enlarged
and grooved and are separated from the others by an interspace.

The dorsum is purplish brown with 65 to 75 short, irregular,
yellow crossbars nearly one scale in length. The occiput has one
or two narrow yellow crossbars, which are sometimes interrupted.
The venter and tail are purplish brown with irregularly arranged
yellow spots.

The males have more subcaudals than the female, \\'hich has
more ventrals than the males. The body and tail length of a male
(BMNH 1946.1.2.46) are 303 mm. and 42 mm. and of a female

■V--

Fig. 3. Ventral, lateral, and dorsal views of the head of a synt\pe of Enhydris

indica (BMNH 1946.1.2.36), Xl.5.

COLUBRID SXAKES, SUBFAMILY HOMALOPSINAE 77

(BMNH 1946.1.2.36) are 290 mm. and 34 mm., respectively. The
chin-shields and guhirs of the male specimen are tuberculate.

Remarks.— Gray (1842:67, 1849:79) listed India as the locality
of collection for the two specimens presented by Gen. Hardwicke
to the British Museum. Boulenger (1896:4), and Flower (1898:887)
referred to the locality data, of Gen. Hardwicke's specimens as
"Maly Peninsula?" In the register at the British Museum, these
species are entered as "Malay Peninsula?" Tweedie (1957:89) men-
tioned that the species was described in 1842 from specimens "sup-
posed to be from the Malay Peninsula." In 1930 two more speci-
mens were reported, one in the Selanger Museum from Taiping in
Perak and one in the Dublin Museum said to be from Singapore.
Nothing is known of this species besides these four specimens.

Distribution. — Known only from Malay Peninsula (Fig. 4), speci-
mens examined, two, as follows.

Malaya: No definite locality, BMNH 1946. 2.36, 1946.1.2.46.

Enhydris alternans (Reuss)

Biachijorrhos alternans Reuss, Mus. Senckenb., 1:155, pi. 9, Fig. 3, 1834 [Holo-
type. — SMF 19465 from "Java"; Dr. Peitsch collector].

Homalopsis decussata Schlegel, Essai sur la physionomie des serpents, 2:344,
pi. 13, figs. 14-16, 1837 [Holot>'pe.— RMHL 1157 from "Java"; Reinwardt
collector].

MiraJia alternans; Gray, Zoological miscellany, p. 68, 1842. Catalogue snakes
British Museum, p. 79, 1849. Giinther, Catalogue of the colubrine snakes
in the . . . British Museum, p. 277, 1858; Proc. Zool. Soc. London, p. 590,
1872.

Eurosttis alternans; Dumeril and Bihron, Erpetologie generale . . . reptiles.
Paris, 7:957, 1854.

HypsirJuna alternans; Jan, Elenco systematico degli ofidi, p. 78, 1863; Arch.
Zool. Anat. Phvs., 3:262, 1865. Jan and Sordelli, Iconographie generale des
Ophidiens, liv.' 30, pi. 6, figs. 1-2, 1868. Boettger, Ofi^enb. Verh. Naturk.,
p. 133, 1892. Boulenger, Catalogue snakes British Museiun, p. 4, 1896.
Rooij, Reptiles Indo-Australian Archipelago, 2:180, 1917. Brongersma,
Zool. Meded., 16:23, 1933.

Enhydris alternans; Haas, Treubia, 20(3):575, 1950.

Diagnosis. — Nasal cleft extending from nostril to second, or
suture between first and second, upper labial; internasals two, in
contact with each other and with loreal; fourth upper labial enter-
ing orbit; two postoculars; anterior pair of chin-shields in contact
with first four or five lower labials; dorsal scales in 19 rows at mid-
body; ventrals 125-164 (Fig. 5).

Description and variation. — The rostral is broader than high and
is narrowly visible from above. The nasal is a pentagonal plate and
is larger than the internasal. The latter is double, broader than long,
and lies behind the nasals. The prefrontals are larger than the
nasals and are broadlv in contact with each other. The frontal is a

University of Kansas Publs., Mus. Nat. Hist.

CoLUBRiD Snakes, Subfamily Homalopsinae 79

pentagonal plate having a parallel fronto-supraocular suture. It is
as long as, or longer than, its distance from the snout. The parietals
are nearly twice as long and as broad as the frontal. The loreal is
single and is in contact with the second and third upper labials.
The supraocular is a narrow rectangular plate about one fourth the
width of the frontal. There is one high preocular and two (very
rarely one) postoculars, of \\'hich the upper is nearly twice as high
as long, and the lower nearly twice as long as high. The temporals
are arranged one, two, and three in the primary, secondary and
tertiary rows, respectively. Of the eight upper labials, the fourth
enters the orbit, and the sixth is the largest; none is divided. There
are 10 or 11 lower labials, of which the first four or five are in
contact with the anterior chin-shields. The anterior chin-shields

Fig. 5. Lateral view of the head of Enhydris altemans (USNM 56022), Xl.5.

are in contact with each other and are almost twice as long as the
posterior chin-shields, which are separated by a pair of scales. Six
to eight gulars plus two or three incomplete ventrals separate the
chin-shields from the first complete ventral. The position of the
umbilical scar varies from 21 to 24 ventrals anterior to the vent.
The ventrals are narrow, about three times the width of the adjacent
dorsal scales on the first row. Six males have 127-164 (144.1) ven-
trals and 35-39 (36.5) subcaudals; four females have 129-151 (135.7)
ventrals and 23-36 (27.5) subcaudals. The body at its maximum
width is 13 mm. The dorsal scales are in 19 rows and reduce to 17
ro\\'s just before the vent. The dorsal scale reductions for one male
(CNHM 11092) and one female (BMNH 63.12.4.22) are, respectively:

3+4(4) 4+5(135)

21 19 17 (145).

3+4(4) 3+4(134)

4+5(4) 4+5(142)
21 19 17 (150).

3+4(5) 4+5(148)

80 University of Kansas Publs., Mus. Nat. Hist.

There are 10 or 11 maxillary teeth which are separated by an
interspace from the two enlarged posterior grooved teeth.

The hemipenis extends to the level of the 11th or 12th sub-
caudal, and the point of bifurcation is at the level of the 5th sub-
caudal. The retractor penis magnus muscle originates on the caudal
vertebra at the level of the 26th or 27th subcaudal.

The general ground-color is purplish brown. Dorsally the body
is traversed by 40-45, and the tail by 8-9, indistinct yellow cross-
bars, sometimes broken into spots, less than one scale in length.
The first crossbar on the occiput forms a complete ring and is two
scales in length. The venter is purplish brown, traversed by 39-61
(48.1) yellow transverse bars which are two ventrals long. Each
yellow transverse bar typically originates near the midventral line
and extends laterally onto the fourth row of dorsal scales; the first
yellow transverse bar is between the angles of the mouth. The
subcaudals are variegated with purplish brown and yellow.

Males have more subcaudals and a longer tail than females.
The tail/ snout-vent ratio of six males is 12.4-15.2 (14.1) and of two
females is 11.4-11.5 (11.45). The chin-shields and gulars of the male
specimens are heavily tuberculate.

Remarks. — On the criteria of size, shape and coloration of the
body, the snakes of this species resemble those of E. indica; the two
species are probably closely related. Their geographical ranges
indicate that one could be derived from the other. Gray (1849:79),
in fact, suggested that E. alternans might be an abnonnal variety of
E. indica. Enhijdris alternans differs from E. indica in having a
longer tail/ snout-vent ratio in both males and females, more distinct
and broader yellow crossbars, nasal clefts extending to second, or
interlabial suture between the first and the second upper labials, the
divided condition of the internasal, and a lower number of ventrals.
Because of these differences and others, which are discussed in the
species account, the recognition of two species seems to be justified.

Distribution. — The snakes of this species occur (Fig. 6) only in
Sumatra, Bangka, Belitung, Java, and Borneo. Specimens exam-
ined, 15, as follows.

Borneo: no definite locality, BMNH 55.4.13.8, 55.4.13.8. British Borneo:
North Borneo — no definite locality, FMNH 121429. Indonesia: Belitung — no
definite locality, RNHL 5703. Java— no definite locality, RNHL 1157, SMF
19465-9. Sumatra— no definite locality-, USNM 56022; Defi, ZMA no number
(deBussy leg.); Inderagiri River, NMB 2537; Ogan River, FMNH 11092;
Palembang, RNHL 8769.

CoLUBRiD Snakes, Subfamily Homalopsinae

82 University of Kansas Publs., Mus. Nat. Hist.

Enhydris plumbea (Boie)

Homalopsis plumbea Boie, Isis, p. 560, 1827 [Holotype. — RNHL 1163 from
Java; Kiihl and Hasselt collectors]. Schlegel, Essai sur la physionomie des
serpents, 2:346, pi. 13, figs. 12-13, 1837. Cantor, Catalogue Malay Reptiles,
p. 101, 1847.

Hijpsirhina haidwickii Gray, Illus. Indian Zool. vol. 2, pi. 87, fig. 1, 1834
[Holotype. — BMNH 1946.1.2.44 from Penang; presented by Gen. Hard-
wicke]; Catalogue snakes British Museum, p. 72, 1849.

Cohtber plumbeus; Eydou.x and Gervais, Guerin Mag. Zool. Club, vol. 3, pi.
16, fig. 1, 1837.

Hijpsirhina plumbea; Grav, Zoological miscellany, p. 66, 1842. Giinther, Rep-
tiles of British India, 'p. 280, 1864. Jan, Arch. Zool. Anat. Phys. 3:261,
1865. Jan and Sordelli, Iconographie generale des Ophidiens, liv. 30, pi. 5,
figs. 2-3, 1868. Morice, Sur la Faune de la Cochinchine francaise. Lyon,
p. 58, 1875. Theobald, Catalogue reptiles British India, p. 182, 1876.
Tirant, Excursions et Reconnaissance, 20:402, 1885. Boettger, Offenb. Verh.
Naturk., 24/25:123, 1885; Offenb. Verh. Natink., p. 33, 1888. Boulenger,
Fauna of British India . . . Reptilia and Batrachia, p. 376, 1890. Sclater,
List of snakes in Indian Museum, p. 54, 1891. Boettger, Senck. Nat. Ges.
Hainan, p. 134, 1894. Muller, Verh. Naturf. Ges. Basel, 10:831, 1895.
Werner, Verh. Ges. Vien., 46:20, 1896. Flower, Proc. Zool. Soc. London,
p. 887, 1896. Catalogue snakes British Museum, 3:5, 1896. Boulenger,
Proc. Zool. Soc. London, 13:201, 1897. Boettger, Katalog Schlangen Muse-
um Senckenberg, p. 51, 1898. Flower, Proc. Zool. Soc London, p. 675,
1899. Werner, Zool. lahrb. Svst., 13:490, 1900. Boettger, Senck. Nat.
Gesel. Berlin, p. 51, 1901. Schenkel, Verh. Nat. Ges. Basel, 13:165, 1901.
Laidlaw, Proc. Zool. Soc. London, 3:578, 1901. Lampe, Jahrb. Nassau
Veriens, 55:30, 1902. Wall, Proc. Zool. Soc. London, p. 91, 1903. Werner,
Abh. Bayer. Akad. Wissen, ser. 22, 2:366, 1903. Boulenger, Fasciculi
Malayenses Zoology, 1:164, 1903. Roux, Verh. Nat. Ges. Basel, 15:429,
1904. Mocquard, Mission Pavie Indochine, 3:482, 1904. Vaillant, Bull.
Mus. Paris, p. 298, 1904. Boulenger, A verte]:)rate fauna of the Malay
Peninsula . . . Reptilia and Batrachia, p. 160, 1912. Vogt, Sitzber. Ges.
France, p. 227, 1913. Smith, Jour. Nat. Hist. Siam, 1(3):136, 213, 1914;
Jour. Nat. Hist. Siam, 2:162, 1916. Gyldenstolpe, Svensk. vet. Akad.
Handl., 55(3) :17, 1916. Annandale, Jour. Nat. Hist. Siam, 2(2) :91, 1916.
Rooij, Reptiles Indo-Australian Archipelago, 2:181, 1917, Robinson and
Kloss, Jour. Federated Malay States Museum, 7(2):303, 1920. Vogt, Arch.
Natiu-g., 88A(10):141, 1922. Phisalix, Animaux venimeux et venins 2:303,
1922. Mell, Arch. Naturg., 88A(10):122, 1922. Werner Arch. Naturg.,
89A(8):160, 1923. Wall, Jour. Bombay Nat. Hist. Soc, 29:866, 1924.
Lonnberg and Rendahl, Akad. Zool. Stockliolm, 17A(23):2, 1925. Parker,
Ann. Mag. Nat. Hist., 15(9):300, 306, 1925. Bourret, Invent. Gen. Indo-
chine, 3:241, 1927. Scortecci, Atti. Soc. Italy, 68(1 ):74, 1929. Brongensma,
Treubia, p. 67, 1929. Mell, Jour. Lingnan Univ. Sci., p. 31, 1929. Kop-
stein, Weltevreden, p. 131, 1930. Bourret, Bull. Instr. Pub., p. 174, May,
1934; Bull. Instr. Pub. p. 19, September, 1934; Bull. Instr. Pub., p. 80,
December, 1934; Serpents de I'lndochine, p. 276, 1936.

Enwstus plumbeus; Dumeril and Bibron, Erpetologie generale . . . reptiles.
Paris, 7:955, pi. 84. fig. 2, 1854. Boucourt, Nouv. Arch. Mus., p. 9, 1886.

Enhiidris plumbea; Stejneger, Jour. Coll. Tokyo, 12(3) :222, 1898; Proc. U.S.
Natl. Mus., 58:300, 1907; Bull. U.S. Natl. Mus., 38:105, 1910. O.shima,
Annot. Zool. Japonen.sis, 7:190, 1908. Barbour, Proc. New England Zool.,
4:67, 1909. Smith, Jour. Nat. Hist. Siam, 4(2): 205, 1923. Stejneger,
Proc. U.S. Natl. Mus., 66:80, 1925. Schmidt, Bull. Amer. Mus., 54(3):448,
1927; Bull. Amer. Mus., 54(4) :538, 1927. Pope, Bull. Amer. Mus.,
58(8):455, 1929; Proc. Soc. Fukien, 2:22, 1929. Mell, Jour. Lingnan
Univ. Sci., 8:10, 1929. Mertens, Zool. Anz., 86:66, 1929. Cochran, Proc.
U.S. Natl. Mus., 77:31, 1930. Smith, Bull. Raffles Mus., 3:60, 1930. Gee,

CoLUBRiD Snakes, Subfamily Homalopsinae 83

Bull. Biol. Yenching Univ., p. 72, 1930. Herklots, Hongkong Nat., 4:119,
1933. Taylor, Tour. Lingnan Univ. Sci., 13(2) :299, 1934. Pope, Amer.
Mus. Novit, 733:10, 1934; Amer. Mus. Nat. Hist., 10:315, 1935.

Dioii,nosis. — Nasal cleft extending from nostril to first or second
labial; internasal single, not touching loreal; fourth, or fourth and
fifth, upper labials entering orbit; two postoculars; anterior pair of
chin-shields in contact with first five or six lower labials; dorsal
scales in 19 rows at midbody; ventrals 112-139.

Description and variation. — The body is cylindrical or oval. The
tail is short and slightly compressed. The head is not distinct from
the neck. The rostral is broader than high and is visible from above.
The nasals are large and are narrowly in contact behind the rostral.
The internasal is single, triangular, broader than long, and wedged
between the nasals posteriorly. The prefrontals are as large as, or
smaller than, the nasals and are in contact with each other. The
frontal is a pentagonal plate and has parallel fronto-supraocular
sutures. It is as long as its distance from the tip of the snout and
shorter than the parietals. A single loreal is as long as high and in
contact with the second and third (rarely with the first) upper
labials. The supraocular is a rectangular plate; it is broadest pos-
teriorly. There is one high preocular, and two postoculars of nearly
equal size. Temporals are one, two, and three in the primary,
secondary and tertiary rows, respectively. None of the eight upper
labials is divided; the fourth, or the fourth and fifth, enter the orbit,
and the sixth, or seventh is the largest. There are 10 or 11 lower
labials, of which the first five (rarely six) are in contact with the
anterior chin-shields. The anterior chin-shields are in contact with
each other and are almost twice as long as the posterior chin-shields
which are separated by a pair of scales. Four or five gulars and one
incomplete ventral separate the chin-shields from the first complete
ventral. The position of the umbilical scar varies from 19-25 ven-
trals anterior to the vent. The ventrals are wide, nearly four times
the width of the adjacent dorsal scales on the first row. Seventeen
males and 32 females have 114-132 (126.7) and 112-129 (123.8) ven-
trals; sixteen males and 31 females have 35-43 (40.5) and 29-38 (33.4)
subcaudals; ten males and 22 females have tail /snout- vent ratios of
13.5-18.6 (16.0) and 13.2-17.2 (14.8), respectively. The coefficients of
the differences between three pairs of means are 0.41, 0.76, and 0.43,
respectively; none is significant. The dorsal scales are in 19 rows
and reduce to 17 or 15 rows just before the vent. The dorsal scale
reductions for two males (MCZ 7419 and MCZ 7420) and two fe-
males (BMNH 1946.1.2.44 and RNHL 1163) are, respectively:

84 University of Kansas Publs., Mus. Nat. Hist.

3+4(4) 4+5(84) 5+6(117)
21 19 17 15(129).

3+4(4) .3+4(90) 5+6(114)

3+4(3) 4+5(91) 3+4(121) 2+3(126)

21 19 17 16 15(129).

3+4(3) 3+4(89)

4+5(5) 4+5(103)

21 19 17(123).

4+5(5) 4+5(106)

5+6(4) 3+4(84)

21 19 17(121).

4+5(3) 3+4(87)

The 11 to 13 maxillary teeth are followed by an interspace and
two enlarged grooved teeth.

The hemipenis extends to the level of the 7th or 8th subcaudal
and the point of bifurcation is at the level of the 2nd subcaudal.
The retractor penis magnus muscle originates on the caudal verte-
bra at the level of the 21st or 22nd subcaudal.

The dorsum is brown or grayish olive with deep brown or black
edges. The upper lip and venter are white or pale yellow. The
ventrals have black dots medially and the subcaudals have a mid-
ventral black line.

The snakes from Taiwan have more ventrals, more subcaudals
and relatively longer tails than snakes from other parts of the range.
Seven males and eight females have 126-129 (127.7) and 124-126
(125.5) ventrals; five males and seven females have 35-42 (40.6) and
32-38 (35.1) subcaudals; the tail/ snout-vent ratios for four males and
six females are 13.5-17.3 (15.3) and 12.7-16.3 (14.5), respectively.
The coefficients of the differences between three pairs of means are
0.53, 0.99 and 0.28, respectively; none is significant.

Remarks. — Three species — E. indica, E. altenmns, and E. phnn-
hea — are closely related and form a group within the genus Enhij-
dris. These species can be distinguished from other members of the
genus in having the following combination of characters: 19 rows
of dorsal scales, loreal in contact with second and third upper
labials, maxillary teeth 10-13 in number, and almost uniform width
of body from head to vent.

Distribution. — The snakes of this species are found from Burma
to southeastern China as far north as Nanking (Fig. 7). Also, they
range from Sumatra to Bali and Celebes. Specimens examined, 74,
as follows.

CoLUBRiD Snakes, Subfamily Homalopsinae

Annam: no locality, RNHL 4754. Borneo: no locality, BMNH 111.25. l.a,
64.9.2.63-5, RNHL 1164. Burma: Mandalmi D/u/s;o)i— Mandalav, BMNH
1925.5.25.14. China: Fukien Province— Amoy, BMNH 99.8.31.9, 1932.3.11.33,
RNHL 695. Kwangtung P/ou()ice— Canton, SU 74506-7; Foochow, USNM
66451; Hainan, Hoihow, BMNH 93.1.3.14-15; Tingan, MCZ 7420(2). Kiangsu
p,-or/,ift'— Nanking, SU 71955. Cochin China: no locality, MCZ 4073, 5977.
Formosa: no definite locality', BMNH 62.12.16.82-83, FMNH 127281, 127291;
Bankoro, MCF 7419; Hokuto, SU 14975, 19029; Kagi, SU 14976; Taipa,
AMNH 76125, USNM 36513; Yang-Ming Shan Mt., AMNH 84535, 84537-8,
FMNH 127292-3; Polisia, SU 19030. Hong Kong: BMNH 56.11.17.64. Indo-
china: no locality, MCZ 45270. Indonesia: Bali — no definite locality, RNHL
8166. Bangka— no definite locality, RNHL 5982. Borneo— Pasii, MCZ 22653;
Rantal, RNHL 7578, 8194. Celebes— Kema, BMNH 96.12.9.78; Menado,
RNHL 1165. Java— no definite locality, BMNH 79.10.20.22, NMG 67-28,
842-14, RNHL 1163, 3798, 10311. Sumatra— no definite locality, UMMZ
67225; Deli, RNHL 6631, 8644; Langhart, MCZ 3774; Medan, UMMZ 65456;
Tebing Tinggi, USNM 82231. Laos: no locality, BMNH 1912.5.11.2.
Malaya: jahore, BMNH 1903.4.13.75; Kedah, BMNH 98.9.22.43; Kelantan,

100

Fig. 7. Locality records for Enhydris plumbea.

86 University of Kansas Publs., Mus. Nat. Hist.

BMNH 1912.2.22.15-16, 1929.4.29.7; Penang, BMNH 1946.1.2.40, 60.3.19.
1209, SU 13081, 13688; Perak, Taiping, BMNH 97.2.3.8; Tapah Fisheries
Station SU 13081, 13103, 13688. North Vietnam: Ton/c/)!— Cjialam, RNHL
6509, Thai-Nien, BMNH 1924.12.9.198. Thailand: Noiili rhaihnd—Chhmg
Mai, RNHL 10377; Den Chai, AMNH 8831, 8832.

The Enhydris enhydris Group

This group, composed of four species — enhydris; jagori, inno-
minata and smithi — is characterized by the nasal cleft extending to
first upper labial; usually a single nasal in contact with loreal; loreal
in contact with first three upper labials; eight upper labials, fourth
entering orbit; dorsal scales in 21 (rarely 23) rows at midbody;
dorsum brown with dark brown longitudinal stripes or with lateral
blotches, venter pale yellow having a series of dark brown spots or
not.

Members of this group range from India eastward to the south-
eastern part of China, and southeastward to Borneo.

Enhydris enhydris (Schneider)

Hydras enhydris Schneider, Historiae Amphibioium, 1:245, 1799 [From

"Indiae orientalis"; ( "Mntta Pam," Russell, Indian Serpents, p. 35, pi. 30,

1796)].
Hydriis atrocaeruleus Shaw, General Zoology Amphibia, 3:567, 1802 ["Mutta

Pam," Russell, Indian Serpents, p. 35, pi. 30, 1796].
Enhydris caerulca Sonnini and Latreille, Histoire naturclle des reptiles, 4:202,

1802 ["Mutta Pam," Russell, Indian serpents, p. 35, pi. 30, 1796].
Coluber pythonisso Daudin, Histoire naturelle des reptiles, 7:107, 1803 ["Mutta

Pam," Russell, Indian serpents, p. 35, pi. 30, 1796].

HomaJopsis aer Boie, Isis, p. 214, 1826; Isis, p. 560, 1827 [Holotype. — Paris

Museum from "Java"; M. Oppel collector]. Schlegel, Essai sur la physio-

nomie des serpents, 2:347, pi. 13, figs. 10-11, 1837.
Potamophis hisingtonii Cantor, Trans. Med Phys. Soc. Calcutta, 7:139, 1836

\fide Smith, 1943:383; type-locality "India"].
Cohiber aer; Eydoux and Gervais, Guerin Mag. Zool. Club, 3, pi. 16, figs. 2-3,

1837.
Homalopsis ohvaceus Cantor, Proc. Zool. Soc. London, p. 55, 1839 [Holotype. —

Depository imknown from Bengal; colored sketch in Bodleian Library].
Hypsirhina triJineata Gray, Zoological miscellany, p. 66, 1842 [Holotype. —

BMNH 1946.1.7.22 from "India"; collector unknown].
Hypsirhina furcata Gray, Zoological miscellany, p. 66, 1842 [Syntypes. — BMNH

1946.1.21.67-68 from "India"; collector unknown].
Hypsirhina hiUncaia Grav, Zoological miscellany, p. 66, 1842 [Holotype. —

■ BMNH 1946.1.5.31 from "China"; presented by W. Lindsay]; Catalogue
snakes British Museum, p. 73, 1849.

Hypsirhina aer; Gray, Catalogue snakes British Museum, p. 72, 1849.
Homalopsis enhydris; Cantor, Catalogue Malay Reptiles, p. 99, 1847.
Hypsirhina enhydris; Dumeril and Bibron, Erpetologie generale . . . reptiles.

■ Paris, 7:946, 1854. Giinther, Reptiles of British India, p. 281, pi. 22, fig.
K, 1864. Jan, Arch. Zool. Anat. Phys., 3:261, 1865. Jan and Sordelli,
Iconographie generale des Ophidiens, liv. 30, pi. 3, fig. 2, & pi. 5, fig. 1,
1868. Nicholson, Indian snakes. Madras, p. 62, 1874. Morice, Sur la
Faune de la Cochinchine francaise. Lyon, p. 58, 1875. Theobald, Cata-
logue reptiles British India, p. 183, 1876. Tirant, E.xcursions et Reconnais-

CoLUBHin Snakes, Subfamily Homalopsinae 87

sauces, 20:403, 1885. Bofonrt, Noiiv. Arch. Mus., p. 5, 1886. Boiilenger,
Fauna of British India . . . Rcptilia and Batrachia, p. 376, 1890. Sclatcr,
List of snakes in Indian Museum, p. 54, 1891. Werner, Verh. Ges. Wien,
46:20, 1896. Flower, Proc. Zool. Soc, London, p. 887, 1896. Boulenger,
Catalouue snakes British Museum, 3:6, 1896; Proc. Zool. Soc. London,
13:201, 226, 1897. Boettger, Katalog Sehlangen Museum Senckenberg, p.
87, 1898. Flower, Proc. Zool. Soc. London, p. 676, 1899. Wall and E\ans,
Jour. Bomba%- Nat. Hist. Soc, 13:348, 1900; Jour. Bombay Nat. Hist. Soc,
15. 616, 1901. Lampe, Jahrb. Nassau Veriens, 55:30, 1902. Brown, Proc.
\cad. Philadelphia, 11:180, 1902. Boulenger, Fasciculi Malayenses, Zool-
()g^•, 1:175, 1903. Wall, Proc. Zool. Soc. London, p. 94, 1903. Werner,
Abh. Bayer, Akad. Wissen, ser. 22, 2:366, 376, 1903. Mocquard, Mis.sion
Pa\ie Indochine, 3:482, 1904; Revue Coloniale, p. 51, 1907. WaM, Jour.
Bombay Nat. Hist. Soc, 17:323, 1908; Jour. Bombay Nat. Hist. Soc,
19:831, 845, 1910; Jour. Bombay Nat. Hist. Soc, 21:1017, 1912. Boulen-
ger, A \ertebrate fauna of the Ma'av Peninsula . . . Reptilia and Batrachia,
p. 160, 1912, D'Abreu, Jour. Bombay Nat. Hist. Soc, 22:203, 1913. Aber-
crombv, Spolia Zeyl, p. 4, 1913. Smith, Jour. Nat. Hist. Siam, 1(1):100,
1914; 'jour. Nat. Hist. Siam, 1(2):126, 1914; Jour. Nat. Hist. Siam,
1(3):2"13, 1914; Jour. Nat. Hist. Siam, 2:162, 1916. Gyldenstolpe, Svensk.
vet. Akad. Handl., 55(3) :18, 1916. Rooij, Reptiles Indo-Australian Archi-
pelago, 2:181, 1917. Fea, Ann. Mus. Genova, ser. 2, 17:475, 1917.
Holzinger-Tenever, Arch. Naturg., 85A(11):109, 1919. Robinson and
Kloss, Jour. Federated Malay States Museum, 7(2):303, 1920. Inglis,
Traver.s, O'Donal, and Shebbeare, Jour. Bombay Nat. Hist. Soc, 27:159,
1920. Mell, Arch. Naturg., 88A:122, 1922. Phisalix, Animaux venimeux et
venins 2:285, 1922. Werner, Arch. Natin-g., 89A(8):161, 1923. Wall,
Jour. Bombay Nat. Hist. Soc, 29:866, 1924; Jour. Bombay Nat. Hist. Soc,
30:817, 1925. Angel, Bull. Mus. Paris, 2:496, 1927. Bourret, Invent. Gen
Indochine, 3:241, 1927. Brongersma, Treubia, p. 67, 1929. Mell, Jour.
Lingnan Univ. Sci., p. 31, 1929. Gee, Bull. Biol. Yenching Univ., p. 73,
1930. Kopstein, Treubia, 11(3) :307, 1930; Weltevreden, p. 137, 1930;
Treubia, 12(3/4) :274, 1930. Brongersma, Zool. Med. Rejks. Mus., 16:15,
1933. Bourret, Bull. Instr. Pub., 9:19, September, 1934; Serpents de ITndo-
chine, p. 280, 1936.

Eiirostus dussiimicri Dumeril and Bibron, Erpetologie generale . . . reptiles.
Paris, 7(2):953, pi. 84, 1854 [Holotype.— MNHN 5751 from Malabar,
India; collector unknown].

Hypsirhiua alholineata Morice, Sur la faune de la Cochinchine francaise. Lyon,
p. 58, 1875 [Holotype. — Lyons Museum; type locality and collector un-
kno\\n].

Hypsirhina cnliydiis hilineata; Lampe, Jahrb. Nassau Viriens, 55:31, 1902.
Volz, Zool. Jahrb. Syst., 20:498, 1904.

Hdicops indicu.s Annandale, Jour. Asiatic Soc Bengal, p. 211, 1905 [Holotype.
— Calcutta Museum from "Bengal"; collector imknown].

Euht/dris euhydris; Barbour, Mem. Mus. Comp. Zool., 44(1):122, 1912.
Sworder, Singapore Nat., 3/4:66, 1924. Mertens, Senckenberg, 2:31, 1927.
Shaw and Shebbeare, Jour. Darjeeling Nat. Hist. Soc, 4:54, 1929. Mell,
Jour. Lingnan Univ. Sci., 8:10, 212, 1929. Cochran, Proc. U.S. Natl. Mus.,
77(11):30, 1930. Smedley, Bull. Raffles Mus., 7:15, 1932. Pope, Amer.
Mus. Novitates, 733:10, 1934; Reptiles China, p. 314, pi. 13, figs. D-I, 1935.
Smith, Fauna of British India . . . Reptilia and Amphibia, 3:383, 1943.

Diagnosis. — Nasal cleft extending from nostril to first upper
labial; internasal single (rarely divided), touching loreal; fourth
upper labial entering eye; two postoculars (rarely one); anterior
pair of chin-shields in contact with first three or four lower labials;
dorsal scales in 21 (rarely 23) rows at midbody; ventrals 141-174.

Description and variation.— The body is cyHndrical, and the

88 University of Kansas Publs., Mus. Nat. Hist.

snout is narrowly rounded anteriorly. The rostral is nearly twice as
broad as high and is just visible from above. The nasals are large
and are broadly in contact with each other behind the rostral. The
internasal is diamond-shaped, more than twice as broad as long, in
contact with the loreal, and usually single, but divided in about 20
per cent of the Indian and Burmese specimens. The prefrontals
form a median suture and are smaller than the nasals. The frontal
is slightly longer than broad, broader than the supraocular, shorter
than its distance from the end of the snout, and shorter than the
parietals. A single loreal is as high or slightly higher than long, and
is in contact with the first three upper labials. The supraocular is
broadest posteriorly. The preocular is twice as high as long. The
two (rarely one) postoculars are subequal, and the lower scale is
partly under the eye. The temporals are one in the primary, two in
the secondary, and three (or four) in the tertiary rows, respectively;
the tertiary row is absent in some specimens. None of the eight
upper labials is divided; the fourth upper labial enters the orbit,
and the sixth is the largest. Twenty per cent of the specimens have
10 lower labials; 39 per cent have 11, and 41 per cent have 12.
Specimens from Burma have the first three lower labials in contact
with the anterior pair of chin-shields. Specimens from Thailand
have the first four (25 per cent have the first three) lower labials in
contact with the anterior pair of chin-shields. Specimens from the
other parts of the range have the first three or four (approximately
in equal number) lower labials in contact with the anterior pair of
chin-shields. Of the two pairs of chin-shields, the posterior pair is
longer than the anterior pair. The posterior chin-shields are sepa-
rated by a pair of small scales. The number of gulars and incom-
plete ventrals between the chin-shields and the first complete ven-
tral are, 7-14 (10.2) and 0-4 (2.0), respectively. The narrow ventrals
are nearly three times the width of the adjacent dorsal scale on the
first row. The position of the umbilical scar varies between 17 and
25 ventrals anterior to the vent (Fig. 3). Seventy-three males and 45
females have 149-172 (161.6) and 134-170 (157.8) ventrals, respec-
tively; 62 males and 45 females have 55-78 (70,9) and 50-77 (61.8)
subcaudals, respectively. The coefficients of the differences be-
tween the two pairs of means are 0.31 and 0.89 and are not signifi-
cant. The numbers of dorsal scale rows at the level of the first
complete ventral are as follows: 29 rows in 5.7 per cent of the
specimens, 27 rows in 60.3 per cent, and 25 rows in 34.0 per cent.
At midbody the dorsal scales are arranged in 21 rows in 93.7 per
cent of the specimens and in 23 rows in 6.3 per cent. Just anterior

5+6(21)

5+6(22)

5+6(4)

6 + 7(5)

4+5(7)

CoLUBRiD Snakes, Subfamily Homalopsinae 89

to the vent there are 17 rows in 5.3 per cent of the specimens, 19
rows in 86.0 per cent, and 21 rows in 8.7 per cent of the specimens.
The dorsal scale reductions in two males (ANSP 5094, BMNH 1946.
1.5.31) and one female (BMNH 1946.1.7.22) are, respectively:

3+4(45)
23 21(163).

3+4(42)

7+8(8) 4+5(22) 3+4(144)
25 23 21 19(160).

5+6(10) 3+4(27) 3+4(145)

5+6(16) 3+4(54)
25 23 21(155).

4+5(7) 5+6(22) 3+4(65)

The 20-22 maxillary teeth are followed by an interspace and a
pair of slightly enlarged grooved teeth. The 20-22 mandibular teeth
decrease in size posteriorly.

The hemipenis extends to the level of the 7th subcaudal, and
the point of bifurcation is at the level of the 4th subcaudal. The
retractor penis magnus muscle originates on a caudal vertebra be-
t\\'een the 26th and 30th subcaudal. The proximal four or five
whorls of spines just distal to the naked basal portion are the largest.
Distally, the spines are shorter, blunt, and more numerous. The
sulcus is forked.

The dorsum of the body and the tail is brown, gray, or olive with
one middorsal and a pair of lateral dark brown or dark gray stripes.
In \'oung specimens the middorsal stripe is paired due to the devel-
opment of a vertebral white line. Each lateral stripe occupies the
fourth to the seventh (or eighth) scale rows. In some specimens
both edges of the lateral stripe are black; in others the edges of the
scales within the lateral stripe are black, whereas in some others
the anterior edges of the scales within the lateral stripe are black
and form a series of spots. The first three rows of dorsal scales are
white, yellow, buff, or red; the colors sometimes alternate. Usually
the first row is white or yellow, and the second and third rows are
buff or red. The ventrals are yellow or white and margined laterally
by brown or gray and by a median series of brown or gray spots. In
some specimens the median spots form a continuous midventral line
or an interrupted series of spots. The head is brown or gray above,
indistinctly variegated with dark brown or gray. In young, the pale
yellow dorsolateral stripes extend anteriorly across the parietals to
the nasals. These two pale stripes meet between the nasals and
extend anteriorly across the rostral. An indistinct black rostral-

90 University of Kansas Publs., Mus. Nat. Hist.

orbital stripe is more pronounced in young than in the adults. The
subcaudals are white or yellow with three brown or gray stripes,
which are fomied by the brown or gray lateral edges and by the
brown or gray median edges of the subcaudals.

Males have more subcaudals than females. The tail/ snout-vent
ratios in 27 males and 41 females are 22.0-30.3 (28.0) and 19.6-27.8
(23.9), respectively. The coefficients of the differences between the
means is 0.88 and is not significant (Fig. 8).

Remarks. — The subdivision of Enhydris enhijdris by Bourret
(1936:281) into E. e. enhydris and E. e. suhtaeniata on the basis of
the anterior chin-shields in contact with four, or three, lower labials,
respectively, is unjustified. Other characters suggest that E. e. suh-
taeniata is conspecific with E. jagorii.

Distribution. — The snakes of this species are found in river sys-
tems of central and eastern India, East Pakistan, central and lower
Burma, Malay Peninsula, Thailand, Cambodia, Laos, Vietnam, and
the southeastern part of China on the Asiatic mainland (Fig. 9).
They also inhabit river systems on the Sunda Islands as far east as
Borneo. Specimens examined, 131, as follows.

Burma: Irrawadchj Division — Mvaungmya, KU 92527-92548. Mandalaij
D/u/.s/on— Mandalay, BMNH 1925.12.22.8-14. Pegu Division— Pegu, UMMZ
65658; Rangoon, MCZ 3681, SU 13062; Twante, near Rangoon, KU 92470-
92492, 92551-92563, KKG 115, 118, 119, 122, 124, 126, 127, 129, 132, 138,
139, 143, 145-148, 155-157. Borneo: no definite locality, BMNH 111.25.3.a.
Ceylox: no definite locality, CAS 12767. China: no definite locality, BMNH
1946.1.5.31. CocHix China: no definite locality, MNHN c3458.14; tay Ninh,
MCZ 4075. East Pakistan: Baromchal, MCZ 58253; Burgoalai, MCZ 58246-
52; Decca, AMNH 89297. French Inuo-china: no definite locality, FMNH
105561-2. India: no definite locafity, ANSP 5091-2, BMNH 10.52.21.4,
52.10.4.,57, 1946.1.7.22, 1946.1.21.67-8. Assam— Darjeefing, BMNH 72.4.17.
343; Dibrugarh, BMNH 1908.6.23.06. Bengal— no definite locality, ANSP

• • • m "

»• • • • o o

«• . ° o

o o^
o

•

-1 1 1 ] I 1 J L.

'^° '90 "0 270 310 350 390 410 470

BODY LENGTH (mm)

510 550 590 630 670 710

Fig. 8. Relationship between body length and tail length of males ( closed cir-
cles) and females (open circles) of Enhydris enhydris.

CoLUBRiD Snakes, Subfamily Homalopsinae

o
y

92 University of Kansas Publs., Mus. Nat. Hist.

5093; Calcutta, MCZ 3192, SU 13063. B/7iflr— Lake Puiula, BMNH
1940.3.4.43. Uttar Pradesh— Gonda, BMNH 1937.0.7.6. Indonesia: Bangka—
no definite locality, BMNH 1929.12.4.6. West Borneo— Knpuas River, ANSP
2304, 2561, 26401-2. Java— no definite locafity, AMNH 62665, FMNH 105560,
MCZ 1976, 5602; Buitenzorg, MCZ 7559. Sumatra— no definite locality,
BMNH 660.69. Malaya: Kedah—Ahr Star, BMNH 98.9.22.53; Kuala Kele-
tan, CAS 16852; Penang, BMNH 60.3.19.11-12; Singapore, BMNH 160.68.
Thailand: no definite locality, ANSP 5094, BMNH 62.11.1.247, 65.4.28.7-16,
FMNH 123493. South Thailand— Bangkok, AMNH 85648, 92287-93, BMNH
91.6.25.9, 97.10.8.30, 1956.1.12.93(2), MCZ 5672. 20398, UMMZ 65337(2),
USNM 83431, 100993. South Vietnam: Saigon, NMG 743.64.

Enhydris jagorii Peters

Hi/psirhina (Eurostus) jagorii Peters, Monats. Akad. Wiss. Berlin, p. 245, 1863
[Holotype. — Berlin Museum from Thailand; collector unknown]. Giinther,
Reptiles of British India, p. 282, 1864. Morice, Sur la France de la Cochin-
chine francaise. Lyon, p. 58, 1875. Tirant, Notes sur les Reptiles de Co-
chinchine et Cambodge, p. 1, 1885; Excursions et Reconnaissances, 20:403,
1885. Boulenger, Catalogue snakes British Museum, 3:6, 1896. Flower,
Proc. Zool Soc. London, p. 676, 1899. Mocquard, Revue Coloniale, p. 51,
1907. Smith, Jour. Nat. Hist. Siam, 1(1):101, 1914; Join-. Nat. Hist. Siam,
1(3):213, 1914. Phisali.x, Animaux venimeux et venins, 2:285, 1922.
Werner, Arch. Naturg., 89A(8):160, 1923. Bourret, Invent. Gen. Indo-
chine, 3:241, 1927.

Enhydris jagorii; Cochran, Proc. U.S. Natl. Mus., 77(11):31, 1930. Smith,
Fauna of British India . . . Reptifia and Amphibia, 3:384, 1943.

Hijpsirhina enht/dris subtacniata Bourret, Bull. Gen. Instruct. Pub. Hanoi, p. 9,
March, 1934 [Holotype.— MNHN 358 from Soc Trang, Cochinchine;
Bourret collector]; Serpents de ITndochine, p. 282, 1936.

Diagnosis. — Nasal cleft extending from nostril to first labial;
internasal single in contact with loreal; fourth upper labial entering
orbit; two postoculars; anterior pair of chin-shields usually in con-
tact with four (occasionally three) lower labials; dorsal scales in 21
rows at midbody; ventrals 108-146.

Description and variation. — The head is only slightly distinct
from the narrow neck. Posterior to the neck, the oval body progres-
sively increases in size. The tail is short and flattened laterally.
The rostral is nearly twice as broad as high and is narrowly visible
from above. The nasals are large and are broadly in contact with
each other behind the rostral. The internasal is single, more than
twice as broad as long. The prefrontals form a median suture and
are as large as or slightly smaller than, the nasals. The frontal is an
elongated, shield-shaped plate, a little broader than the supraocu-
lars, as long as its distance from the end of the snout, and slightly
shorter than the parietals. A single loreal, as long as high, is in
contact with the first three upper labials. The supraocular is a
rectangular plate which is slightly broader at the posterior end.
There is one high preocular and two subequal postoculars. Tem-
porals are one, two and three in the primary, secondary and tertiary
rows, respectively. None of the eight upper labials is divided; the

CoLUBRiD Snakes, Subfamily Homalopsinae 93

fourth enters the orbit and the sixth is the largest. There are 11
(rarely 10) lower labials, of which the first four (67 per cent) or three
(33 per cent) are in contact with the anterior chin-shields. The
anterior chin-shields are in contact with each other and are wider
and longer than the posterior pair of chin-shields, which are sepa-
rated by a pair of scales. Seven to 10 gulars plus one or two incom-
plete ventrals separate the chin-shields from the first complete
\entral. The ventrals are narrow, not (|uite three times the width
of the adjacent dorsal scales on the first row, and the edges are
rounded. Seven males have 108-146 (109.3) and six females have
111-138 (121.2) ventrals; five males have 48-62 (55.4) subcaudals and
four females have 42-52 (46.7) subcaudals. The dorsal scales are in
21 rows and reduce to 19 or 17 rows just anterior to the vent. The
dorsal scale reduction in one female (BMNH 65.4.28.5) is,

4+5(10) 4+5(109)
23 21 19.

4+5(7) 3+4(108)

Males have more subcaudals and significantly longer tails than
females. The tail /snout- vent ratio for six males and six females are
23.0-29.6 (26.2) and 11.1-25.0 (19.7). The coefficient of difference is
1.51.

The 12 to 13 maxillary teeth are followed by an interspace and
two enlarged, grooved teeth.

The hemipenis extends to the level of the 8th subcaudal, and the
point of bifurcation is at the level of the 4th subcaudal. The re-
tractor penis magnus muscle originates on the caudal vertebra at
the level of the 20th or 21st subcaudals. The basal part of the
hemipenis is naked, followed by four or five tiers of moderately
large spines that gradually become smaller distally.

The head is brown with a rostral-orbital-postorbital stripe,
which is continuous with the lateral row of dark brown angular
flower-shaped spots, which extend to the tip of the tail. These spots
are one to two scales in length and width; they occupy the fourth to
the sixth rows of dorsal scales, and are separated by a distance of
two scales. The median stripe of the nape originates from the
posterior region of the frontal, extends posteriorly through the
occipital region, and becomes continuous with the vertebral row of
dark brown spots, which are usually arranged in pairs. These spots
are one or two scales in length and width, and are separated by a
distance of three or four scales. The lower half of the upper labials,
and the lower labials are yellow; the latter have dark brown sutures.
The ventrals are yellowish white with brown lateral edges. The

University of Kansas Publs., Mus. Nat. Hist.

first two rows of the dorsals are yellow and heavily mottled with
brown; the scales in the third row have dark brown edges dorsally.
The subcaudals are brown with yellow posterior edges (Fig. 10).

Remarks. — Peters (1863:245) stated that the holotype had ten
maxillary teeth plus two enlarged posterior teeth; Giinther (1864:
282) stated that his specimen had seven maxillary teeth, plus one
enlarged posterior tooth. The six specimens that I examined have
12 or 13 teeth.

Peters (1863:246) described the coloration of the holotype, as
follows:

"Was die Farbung anbelangt, so sincl die Korperseiten mit bieiten blaii-
schwarzen Binden bedeckt, die mit schnialen, kaiim eine Schuppe breiten
gelben Binden abwechseln, welche letztere auf den unteren Schuppenreihen
etwas breiter werden und entweder mit denen der andern Seite auf der
Bauchseite sich vereinigen oder mit ihnen abwechseln; auf dem Nacken
beginnt eine Reihe von grossen rhomboidalen Flecken, welche alier sehr oft
mit den schwarzen Binden der einen oder anderen Seite zusammenffiessen, so
dass dann der sie umgebende briiunlichgraue, stauformig mit Schwarz he-
sprengte Grund eine gezackte Riickenlinie bildet. . . ."

I have not seen the holotype, nor have I seen a specimen that
fits the description of the holotype. The description fits well with
the pattern of E. innominata, but the ventral and subcaudal counts
of the holotype (128 and 66, respectively) are too high for E. inno-
minata. Presently, I would agree with Giinther (1864:282), who
wrote: "Although our specimen differs in the coloration from that
described by Peters, I have no doubt that both belong to the same
species."

«i....5|,.^.

'^^,

Fig. 10. Lateral \iew of the body of Enhydm jagoiii (UNMZ 65336), Xl.8.

CoLUBRiD Snakes, Subfamily Homalopsinae

96 Unwersity of Kansas Publs., Mus. Nat. Hist.

Distribution. — The snakes of this species occur (Fig. 11) on the
plains of Thailand, Cambodia, Laos and South \'ietnam. Specimens
examined, 18, as follows.

Cambodl\: Kompong Speu, NMB 12438. Laos: no locality, BMNH
1912.5.11.3. South Vietnam: Kontum, BMNH 1927.5.20.14; Soc Trang,
xMNHN C3459.1-3. Thailand: no locality, BMNH 59.7.1.9, 65.4.28.5,
78.2.14.14, NMG 985.37, USNM 67516. East Thailand— near Korat, BMNH
1938.8.7.21. South Thailand— Bangkok, BMNH 98.11.8.30-31, 98.29, MCZ
16645, UMMZ 65336; Tahkamen, BMNH 97.10.8.28.

Enhydris innominata (Morice)

Hypsirhina innominata Morice, Sur la Faune de la Cochincliine francaise, p.
58, 1875 [Holotype. — Lyon Museum from Tay Minh, Cochin China; collec-
tor unknown]; Tirant, Excursions et Reconnaissances, 20:403, 1885; Bourret,
Invent. Gen. Indochine, 3:241, 1927.

Enhydris innominata; Smith, Jour. Nat. Hist. Soc. Siam, 8:49, 1929; Fauna of
British India . . . Reptilia and Amphibia, 3:385, 1943.

Diagnosis. — Nasal cleft extending from nostril to first upper
labial; internasal single, in contact with loreal; fourth upper labial
entering orbit; two postoculars; first five lower labials in contact
with anterior pair of chin-shields; dorsal scales in 21 rows at mid-
body; ventrals 105-116; subcaudals 40-52.

Description and variation. — The head is small, depressed, and
scarcely distinct from the narrow neck. The body is stout and
cylindrical, and the tail is ovoid. The rostral is twice as broad as
high. A single internasal is twice as broad as long. The prefrontals
are in contact with each other and are smaller than the nasals. The
frontal is as long as its distance from the end of the snout, a little
broader than the supraocular, and as long as the parietals. The
loreal is a little broader than high and is in contact with the first
three upper labials. One high preocular and two postoculars are
present; the lower postocular is smaller than the upper and is partly
below the eye. The temporals are usually one in the primary, two
in the secondary, and three in the tertiary rows. Of the eight upper
labials, the fourth enters the orbit. The last upper labial is divided
in some specimens. Of the 10 lower labials, the first five (rarely
four) are in contact with the anterior chin-shields, which are much
larger than the posterior pair. The posterior chin-shields are sepa-
rated by a pair of small scales. There are nine or ten gulars and two
or three incomplete ventrals between the chin-shields and the first
complete ventral.

Two males have 108-116 (112.0) ventrals, 48-52 (50.0) subcaudals,
and total lengths of 354-492 (423) mm. Two females have 105-111
(108) ventrals, 43-46 (44.5) subcaudals, and total lengths of 198-231
(214.5) mm. The ventrals are narrow, not quite three times the

CoLUBRiD Snakes, Subfamily Homalopsinae 97

width of the adjacent dorsal scales, which are not enlarged on the
first row. The dorsal scales are in 21 or 23 rows, which reduce to 19
rows just before the vent.

The 10 to 12 maxillary teeth are followed by an interspace and
t\\ o slightly enlarged, grooved teeth.

The hemipenis extends to the level of the 7th subcaudal, and the
point of bifurcation is at the level of the 4th subcaudal. The re-
tractor penis magnus muscle originates on the caudal vertebra adja-
cent to the 23rd subcaudal.

The dorsum is brownish gray with small black spots arranged in
three regular longitudinal rows. The sides and venter are yellowish
w hite with closely set, black vertical blotches (five scales wide and
three scales long; interspaces one scale in width), which extend
onto the ventrals. There are 38 to 39 such lateral black blotches on
the body. The tail is alternately banded with long, black and short,
yellow rings.

The males have a divided last ventral with an extra half scale.
Males have more subcaudals than females; the tail/ snout-vent
ratios for two males and two females are 24.5-27.3 (25.9) and 21.5-
29.9 (24.2), respectively.

Remarks. — Tirant (1885:41) pointed out, "La description de YH.
innominata est tout a fait rudimentaire et insuffisante. II es impossi-
ble de reconnaitre et de conseruer sous ce nom cette espece du a
Morice, si espece il y a."

This species resembles E. smithi, and there is no difference in
head scutellation. The holotype, a female, has 23 rows of dorsal
scales around the body; five other specimens in the Paris Museum
have 21 rows of dorsal scales around the body (Smith, 1943:385).
Enhydrls smithi also has 21 rows of dorsal scales around the body.
Thus, there is a little difference between E. smithi and E. innomi-
nata in dorsal scutellation, but E. innominata differs from E. smithi
in having lower ventrals 105-116 and subcaudals 40-52, as com-
pared with 118-127 ventrals and 54-56 subcaudals in E. smithi. The
average total length of E. innominata is less than that of E. smithi.
Enhydris innominata lacks the yellow festoon pattern typical of E.
smitJii.

Distribution. — The snakes of this species occur (Fig. 6) in Cochin
China and Thailand. Specimens examined, four, as follows.

Cochin China: no definite locality, BMNH 1938.1.31.1, NMB
1736, 5812.

98 University of Kansas Publs., Mus. Nat. Hist.

Enhydris smithi (Boulenger)

Hypsirhina smithi Boulenger, Tour. Nat. Hist. Soc. Siam, 1:69, 1914 [Holo-
tvpe.— BMNH 1948.1.7.27 from Mivam Choa Phraya, Bangkok, Thailand;
collector Malcolm Smith]; Smith, Jour. Nat. Hist. Siam, 1(2): 101, 1914,
Jour. Nat. Hist. Siam, 1(3):213, 1914; Werner, Arch. Nating., 89A(8):161,
1923.

Enhydris smithi; Smith, Jour. Nat. Hist. Soc. Siam, 8:50, 1929.

Diagnosis. — Nasal cleft extending from nostril to first upper
labial; internasal single, in contact with loreal; fourth upper labial
entering orbit; two (rarely three) postoculars; first four or five lower
labials in contact with anterior pair of chin-shields; dorsal scales in
21 (rarely 23) rows at midbody; ventrals 118-127; subcaudals 54-56.

Description and variation. — The head is small, depressed, and
scarcely distinct from the narrow neck. The body is stout and
cylindrical and the tail is ovoid. The rostral is twice as broad as
high. The nasals are pentagonal plates fomiing a median suture.
The nostril is a lunate slit with the cleft extending to the first upper
labial. The internasal is twice as broad as long and is in contact
with the loreal. The prefrontals form a median suture and are
smaller than the nasals. The frontal is not much broader than the
supraocular. It is as long as its distance to the end of the snout, and
is a little shorter than the parietals. The loreal is single or divided,
as high as broad, and in contact with the first three upper labials.
There is one (rarely two) high preocular which partly extends below
the eye. Of the two (rarely three) postoculars, the upper is higher
than long; the lower is longer than high and is partly below the eye.
The temporals are one in the primary, two or three in the secondary,
and three or four in the tertiary rows, respectively. Of the eight
(rarely nine) upper labials, the fourth (rarely fifth) enters the orbit;
none is divided. The number of lower labials varies from 10-12.
The first four or five are in contact with the anterior chin-shields,
which are much larger than the posterior pair. The posterior chin-
shields are separated by a pair of small scales. Eight or nine gulars
plus one or two incomplete ventrals separate the chin-shields from
the first complete ventral. The umbilical scar in BMNH 1930.1.1.3
is situated on the 18th and 19th ventrals anterior to the vent. Four
females have 115-121 (118.5) ventrals, 52-56 (54.0) subcaudals, and
total lengths of 503-691 (596.3) mm. The dorsal scales are in 21
rows, which reduce to 19 rows just before the vent. The dorsal scale
reduction in one female (BMNH 1948.1.7.27) is,

6+7(2) 5+6(11) 3+4(111)

25 23 21 19(120).

6+7(3) 5+6(13) 3+4(111)

CoLUBRiD Snakes, Subfamily Homalopsinae

The 11 maxillary teeth are followed by an interspace and two
slightly enlarged, grooved teeth.

The dorsum and sides are black, and the venter is dark gray.
There are 38-39 narrow (one scale long), more or less complete
annuli, which are pink on the dorsum and yellow on the sides and
venter. The yellow annuli are linked together on the back at
midbody so as to fonn festoons. The head is black with pink or
pale gray markings, which essentially constitute the middorsal,
spear-shaped, longitudinal stripe. In the region between the eye
and the occiput two pairs of yellow transverse stripes extend out-
\\'ard and backward from the middorsal stripe. The tail has alter-
nate long black and narrow yellow rings.

One specimen (EHT 677) has 33 large irregular black dorsal
spots on a grayish brown ground color, which is variegated with
irregular small spots. The black color pattern on the head is broken
into asymmetrical scattered spots (Figs. 12-13).

Remarks. — The color varies considerably in this species. The
basic color pattern on the body consists of yellow festoons and two
pairs of diverging yellow stripes, but this pattern is sometimes inter-
rupted by irregular black spots. I have examined no males.

m-^^

Fig. 12. Dor. sal \ie\\ of the head of

Enhijdris smithi (BMXH 1930.1.1.3),

Xl.2.

Fig. 13. Lateral view of the head of
Enhydris smithi (EHT 677), Xl.2.

100 Unr'ersity of Kansas Publs., Mus. Nat. Hist.

Distribution. — Snakes of this species occur (Fig. 4) only in Thai-
land. Specimens examined, four, as follows.

Thailand: Gulf of Siam—Uua. Hin Beach, BMNH 1928.12.18.1.
South Thailand— Bangkok, BMiNH 1930.1.1.3, 1948.1.7.27, EHT 676-
77, SMF 56011.

The Enhydris chinensis Group

This group, composed of three species (chinensis, bennetti, and
longicauda), is characterized by internasal not in contact with
loreal; fourth upper labial entering orbit; loreal in contact with first
four upper labials; first four lower labials in contact with anterior
pair of chin-shields; dorsal scales in 21 or 23 rows at midbody;
dorsum having dark brown or black vertebral and dorsolateral
series of spots, venter yellow or light brown, having series of black
spots or not.

Members of this group are found from Thailand eastward to the
southeastern part of China and the continental islands of Hainan
and Formosa, and southeastward to Java.

Enhydris chinensis (Gray)

Hypsiriiina cliinensis Gray, Zoological miscellany, p. 66, 1842 [Holotype. —
BMNH 1946.1.2.42 from "China"; T- R. Reeves collector]; Catalogue snakes
British Museum, p. 73. 1849. Gimther, Reptiles of British India, p. 283,
1864. Boettger, Offenb. Verh. Natmk., p. 123, 1885. Boulenger, Catalogue
snakes British Museum, 3:8, pi. 1, fig. 2, 1896. Boettger, Abh. Mus. Dres-
den, 7:88, 1898; Katalog. Schlangen Museum Senckenberg, p. 88, 1898.
Flower, Proc. Zool. Soc. London, p. 676, 1899. Wall, Proc. Zool. Soc. Lon-
don, p. 94, 1903. Werner, Abh. Bayer Akad. Wissen, 2:366, 1903.
Mocquard, Revue Coloniale, p. 51, 1907. Vogt, Sitzber. Ces. Nat. France,
p. 227, 1913. Smith, Jour. Nat. Hist. Siam, 1(1):101, 1914; Jour. Nat.
Hist. Siam, 1(3):213, 1914. Mell, Arch. Natiug., 88A(10):122, 1922.
Werner, Arch. Naturg. 89A(8):161, 1923. Bourret, Invent. Cen. Indo-
chine, 3:241, 1927; Bull. Instr. Pub., p. 80, December, 1934; Serpents da
I'Indochine, p. 287, 1936.

Eiihtfdris chinensis; Smith, Jour. Nat. Hist. Soc. Siam, 6:203, 1923. Stejneger,
Proc. U.S. Natl. Mus., 2562:79. 1925. Schmidt, BuH. Amer. Mus.,
54(3) :449, 462, .539, 1927. Mell, Jour. Lingnan Univ. Sci., 8:3, 212,
1929. Pope, Bull. Amer. Mus., 58(8) :456, 1929; Proc. Soc. Fukien, 2:22,
1929; Amer. Mus. Novitates, 620:10, 1934. Cee, Bull. Biol. Yenching. Univ.,
p. 72, 1930. Ping, Contrib. Biol. Lab. China, 7(4): 195, 1931. Chang and
Fang, Peking Nat. Hist. Bull., 7(8):263, 1931. Chang, The China Jour.,
21(5):252, 1934; Peking Nat. Hist. Bull., 9(2):143, 1935. Pope, Reptiles
China, p. 311, pi. 13, figs. A, B, C, 1935. Taylor, Jour. Lingnan Univ.
Sci., 13(2):299, 1936.

Diagnosis. — Nasal cleft extending from nostril to first labial;
internasal single, not touching loreal; fourth upper labial entering
orbit; two postoculars; anterior pair of chin-shields in contact with
first four (rarely five) lower labials; dorsal scales in 23 rows at mid-
body; ventrals 133-153.

CoLUBKiD Snakes, Subfamily Homalopsinae 101

Description and variation. — The body is cylindrical, and the tail
is short and ovoid. The head is depressed, small, and slightly dis-
tinct from neck. The rostral is broader than high and narrowly
visible from above. The nasals are large, pentagonal, and broadly
in contact with each other behind the rostral. The internasal is
single, small, broader than long, and rarely in contact with the
loreal. The prefrontals form a median suture and are smaller than
the nasals. The frontal is an elongated, shield-shaped plate, which is
about two times broader than the supraocular, almost twice as long
as broad, as long as its distance from the tip of the snout, and a
little shorter than the parietals. The loreal is single, as long as deep,
and is in contact with the first three upper labials. The supraocular
is almost rectangular and broadest posteriorly. There is one high
preocular and two postoculars; the upper postocular is higher than
long, and the lower, longer than high. There is one primary, two
secondary, and three tertiary temporals. None of the eight (rarely
nine) upper labials is divided; the fourth enters the orbit, the sev-
enth is the largest, and the eighth is the smallest. Of the 10 (rarely
11) lower labials, the first four (rarely five) are in contact with the
anterior chin-shields. The anterior chin-shields are twice as broad
as, and the same length as the posterior pair. The latter is separated
by a pair of small scales. Six or seven gulars and one or two incom-
plete ventrals separate the chin-shields from the first complete
ventral. The umbilical scar is situated on or between 20-25 ventrals
anterior to the vent. The wide ventrals are nearly four times the
width of the adjacent dorsal scales. Eleven males and 13 females
have 133-151 (146.1) and 139-153 (142.7) ventrals; and 43-51 (47.2)
and 38-47 (41.8) subcaudals, respectively. The coefficient of the
differences between the means are 0.357 and 1.122; neither is sig-
nificantly different. The dorsal scales are in 23 rows at midbody and
reduce to 19 (rarely 17) rows just before the vent. The dorsal scale
reduction for two males (BMNH 1946.1.2.42; BMNH 93.1.3.16) and
one female (MCZ 95967) are, respectively:

5+6(88) 4+5(119)
23 21 19(149).

4+5(88) 4+5(122)

44.5(91) 4+5(121)
23 21 19(149).

4+5(90) 4+5(122)

4+5(8) 4 + 5(92) 4+5(123)
25 23 21 19(139).

4+5(14) 4+5(96) 4+5(126)

102 Unr'ersity of Kansas Publs., Mus. Nat. Hist.

The 12 or 13 maxillary teeth are followed by an interspace and
two enlarged, grooved teeth. The 21-22 mandibular teeth decrease
in size posteriorly.

The hemipenis extends to the 9th or 10th subcaudal, and the
point of bifurcation is at the level of the 4th subcaudal. The re-
tractor penis magnus muscle originates on the caudal vertebra at
the level of the 25th or 26th subcaudal. The basal part of the hemi-
penis is naked; the median part is adorned with curved spines,
which gradually change to papilla-like structures distally without
reduction in size or number. The sulcus is single and moderately
prominent.

The dorsum is brown or black. On the head is a rostral-orbital-
postorbital stripe, which is continuous with a lateral row of black
spots. A black nape stripe about four scales wide, originates on the
posterior end of the parietals and continues for one or two head
lengths posteriorly before it joins with the vertebral row of spots.
The anterior upper and lower labials are brown or black. The chin
is variegated with yellow and brown. On the body and tail are a
pair of paravertebral rows and lateral rows of black spots all of
which continue to the tip of the tail. The spots on the paravertebral
rows are one or two scales wide and long and separated by three or
four scales. In some specimens the spots are connected on the inter-
vertebral line. The black spots in the lateral rows are three scales
wide and t\\'o or three scales long and are separated by one or two
scales. The first row of dorsal scales is black; the second and adja-
cent half of the third row are yellow. The ventrals and subcaudals
are black, edged posteriorly with yellow.

Males have relatively longer tails than females. The tail/ snout-
vent ratios for ten males and 13 females are 15.9-19.8 (18.3) and 12.8-
17.6 (15.5), respectively. The coefficient of the difference between
the means is 0.945 and is not significant. In males a series of five to
seven tuberculate postanal scales are present on each half of the
posterior margin of the vent.

Remarks. — Pope (1935:311) gave the range of ventrals for 26
males as 135-147 and for 20 females as 134-141, and that of sub-
caudals for 26 males as 40-52 and for 19 females as 35-43.

Distribution. — This species is found in Tonkin, Hainan, and
southern China to Taiwan (Fig. 6). They are common in irrigated
fields, ponds, and canals in Tonkin (Bourret, 1936:288) and in the
lowlands of southern China (Pope, 1935:312). Pope noted that this
species is found also at considerable altitudes on the plateaus of

CoLUBRiD Snakes, Subfamily Homalopsinae 103

southern China. Smith (1943:387) obtained specimens at sea in the
straits of Hainan. Specimens examined, 31, as follows.

China: no deBnite locality, BMNH 84.2.26.87, 1946.1.2.42. Fiikicn Prov-
ince—no definite loca!it\', AMNH 33873, BMNH 1910.9.6.13-15; Amoy,
BMXH 1932.3.11.32, CAS 74515; Fuelling Hsien, AMNH 33884; Yeaping,
MCZ 45967-8. Hontimg Ptol/jicc— Tsins^tau, SMF 19495. Ktcanfitimp Prov-
ince—Canton, MCZ 33514, Hainan, no definite locality, SMF 19487-8; Hoi-
how, BMNH 93.1.3.16, 1932.12.15.1. South Hupeh Province— ichang, BMNH
70.1.14.3. Formosa: no definite locality, FMNH 127280, 127970, USNM
133998-9; Yang Ming Shan, Yung Foh Lee, SMF 56959, USNM 142762.
North Vietnam: Tonkin— no definite locality, MNHN c3453, NMG 762.61.

Enhydris bennetti (Gray)

Ht/psirJuna bennetti Gray, Zoological miscellany, p. 67, 1842 [Holotype. —
' BMNH 1946.1.2.49 from "Cliina"; G. Bennett collector]; Catalogue snakes
British Museum, p. 74, 1849. Gimther, Catalogue of the colubrine snakes
in tlie . . . British Museum, p. 283, 1864. Boettger, Offenb. Verh. Naturk.,
p. 151, 1885. Sclater, List snakes Indian Museum, p. 55, 1891. Boettger,
Senck. Natmf. Ges. Hainan, p. 134, 1894. Boulenger, Catalogue snakes
British Mu.seum, 3:8, 1896. Wall, Proc. Zool. Soc. London, p. 94, 1903.
Werner, Abh. Bayer. Akad. Wissen, ser. 22, 2:366, 1903. Mocquard, Revue
Coloniale, p. 51, 1907. Phisalix, Animau.x venimeux et venins, 2:285, 1922.
Mell, Arch. Naturg., 88A:122, 1922. Bourret, Invent. Gen. Indochine,
3:241, 1927; Serpents de I'lndochine, 1:286, 1936.

Hypsirhina macnlata Dumeril and Bibron, Erpetologie generale . . . reptiles.
Paris, 7(2):950, 1854 [Holotype.— MNHN c3453(2) from "China"; col-
lector unknown].

Hypsirhina enhydris var. macuhta; Tan and Sordelli, Iconographie generale
des Ophidiens, liv. 30, pi. 4, fig. 1,1868.

Enhydris bennetti; Stejneger, Bull. U.S. Natl. Mus., 58:302, 1907. Oshima,
Annot. Zool. Japonenses, 7:190, 1908. Stejneger, Proc. U.S. Natl. Mus.,
38:105, 1910. Smith, Jour. Nat. Hist. Soc. Siam, 6:203, 1923. Stejneger,
Proc. U.S. Natl. Mus., 66:80, 1925. Schmidt, Bull. Amer. Mus. Nat. Hist.,
54(3) :462, 1927. Werner, Zool. Jahrb. Tena, 57:185, 1928. Mell, Jour.
Lingnan Univ. Sci., p. 28, 1929. Gee, 1930, Bull. Biol. Yenching Univ., p.
72, 1930. Pope, Amer. Mus. Novitates, 620:10, 1934; Reptiles China, p.
309, 1935. Smith, Fauna of British India . . . Reptilia and Amphibia,
3:386, 1943.

Diagnosis. — Nasal cleft extending from nostril to first labial;
internasal single, widely separated from loreal; fourth upper labial
entering orbit; two postoculars; anterior pair of chin-shields in
contact with first four (rarely three) upper labials; dorsal scales in 21
rows at midbody; ventrals 158-163.

Description and variation. — The body is cylindrical and almost
the same diameter throughout its length; the tail is oval. The head
is slightly depressed and the snout is broadly rounded. The rostral
is broader than high. The nasals are large and broadly in contact
with each other behind the rostral. The internasal is single, small,
and surrounded completely by the nasals and prefrontals. The
latter plates form a median suture and are smaller than the nasals.
The frontal is an elongated hexagonal plate, broader than the supra-
ocular, not quite twice as long as broad, a little longer than the

104 University of Kansas Publs., Mus. Nat. Hist.

distance from the end of the snout, and as long as, or a little shorter
than, the parietals. A single loreal is as long as deep and is in
contact with the first three upper labials. The supraocular is a
rectangular plate. One high preocular and two subequal postocu-
lars are present. There are one, two, and three temporals in the
primary, secondary, and tertiary rows, respectively. None of the
eight upper labials is divided, and the fourth enters the orbit.
There are nine or ten lower labials, of which the first four (rarely
three) are in contact with the anterior pair of chin-shields, which
are almost twice the length of the posterior pair. The latter shields
are in contact with each other or separated by a pair of small scales.
Four or five gulars and one or two incomplete ventrals separate the
chin-shields from the first complete ventral. The umbilical scar is
situated on or between 17-22 ventrals anterior to the vent. The
ventrals are slightly more than thrice the width of the adjacent
dorsal scales in the first row and have rounded edges. Six males and
one female have 158-164 (161.0) and 153 ventrals; four males and
one female have 50-56 (53.2) and 45 subcaudals. The dorsal scale
reductions in two males (BMNH 1932.12.152; 111.25.6.b) are, re-
spectively:

3+4(102) 3+4(130)

21 20 18 17(158).

3+4(131) 3+4(147)

3+4(107) 3+4(140)
21 19 17(160).

4+5(123) 3+4(132)

The 13-15 maxillary teeth are followed by an interspace and two
enlarged grooved teeth. The 20 to 22 mandibular teeth gradually
increase in length anteriorly.

The hemipenis extends to the 10th subcaudal, and the point of
bifurcation is at the 5th subcaudal. The retractor penis magnus
muscle originates on the caudal vertebra adjacent to 29th or 30th
subcaudal. The basal portion of the hemipenis is naked, and the
rest is spinous. Proximally the spines are large and curved; they are
smaller distally and gradually form papilla-like processes. The
sulcus is forked.

The dorsum is grayish olive or brown. The head has a black
nape stripe which is one or two scales broad and extending poste-
riorly for nearly two head lengths. The body has two series of large
(two scales length and breadth) black spots, sometimes connected
with one another on the ventral line. The upper lip, the sides of the
body, the scales on rows two to four, and the venter are yellowish

CoLUBHiD Snakes, Subfamily Homalopsinae 105

white. The \entrals, subcaudals, and the first row of dorsal scales
are heavily edged with gray or brown.

Remarks. — Stejneger (1907:304) noted a specimen in the Indian
Museum (No. 12693) from Formosa. Oshima (1908:191) stated that
he had not found this species on Formosa. The reported presence of
E. hennetti on Formosa may now be discredited (Maki, 1931:67).

This species resembles E. chinensis in head scutelation, general
coloration, and size but it differs in having 21 rows of dorsal scales,
instead of 23 and in having more ventrals (158-163) and subcaudals
(45-56) than E. chinensis, which has 133-153 ventrals and 38-51
subcaudals.

Distribution. — This species occurs (Fig. 11) in the southeastern
part of China, its coastal islands, and Java. They have been caught
at sea in the straits of Hainan (Smith, 1943:387). Specimens exam-
ined, seven, as follows.

China: no locality, BMNH 1946.1.2.49, 111.25.6.6. MNHN c.34,53.2.
Kwangtung Province— Hainan, Hoihow, BMNH 1932.12.1.52, SMF 19486.
HoxG Kong: USNM 22128. Indonesia; Java — ZMA no number.

Enhydris longicauda (Bourret)

Ht/psirhina longicauda Bourret, Bull. Instr. Pub. Gen. Hanoi, p. 20, September,
' 1934 [Svntvpes.— MXHN 48.95, 48.96, 38.143; collector Bourret]; Serpents

de rindochine, 1:284, 1936.
Enhydris longicauda; Smith, Fauna of British India . . . Reptilia and Amphibia,

3:386, 1943.

Diagnosis. — Nasal cleft extending from nostril to first upper
labial; internasal divided, touching loreal; fourth, or fourth and
fifth, upper labials entering eye; two postoculars; anterior pair of
chin-shields in contact with first four or five lower labials; dorsal
scales in 21 rows at midbody; ventrals 129-137.

Description and variation. — The rostral is nearly twice as broad
as long and is visible from above. Three pairs of shields — nasals,
intemasals, and prefrontals — form a straight median suture. The
nasals are small. The intemasals are twice as broad as long. The
prefrontals are pentagonal and smaller than the nasals. The frontal
is an elongated hexagonal plate, broader than the supraocular, as
long as the parietals and its distance from the end of the snout. The
loreal is single or divided and touches the first three upper labials.
One high preocular and two subequal postoculars are present. The
temporals are one primary, one or two secondary, and two or three
tertiary. There are eight or nine undivided upper labials. The
fourth, or fourth and fifth, upper labials enter the orbit; the seventh
or eighth is the largest. There are 12 lower labials, of which the
first four or five are in contact with the anterior pair of chin-shields.

106 University of Kansas Publs., Mus. Nat. Hist.

The posterior pair of chin-shields is much smaller than the anterior
pair. The scales of the posterior pair of chin-shields are separated
by a pair of small scales. Eleven or 12 gulars and one or two incom-
plete ventrals separate the chin-shields from the first complete
ventral. The dorsal scales are in 21 rows at midbody and reduce to
19 rows just anterior to the vent. The broad ventrals are more than
four times the width of the adjacent dorsal scales on the first row;
the edges of the ventrals are rounded. Two females have 128 and
134 ventrals and 61 and 74 subcaudals. Their total lengths are 675
mm. and 230 mm., and their tail lengths are 145 mm. and 62 mm.
The tail/snout-vent ratios are 36.8 and 25.2, respectively.

The 12 maxillary teeth are followed by an interspace and two
enlarged, grooved teeth.

In adults the dorsum is grayish brown with a vertebral series of
large, elongated, dark brown or black spots (about nine scales in
length) and two similar dark dorsolateral stripes, or an indistinct
series of dark spots. The venter is pale brown with small white
spots, one series of which forms a midventral line. The color of
the dorsum extends to the belly as indistinct V-shaped marks. A
series of pale brown chevron-shaped marks are present on the tail.

In young the dorsum is dark brown with three black longitudinal
series (one vertebral and two dorsolateral) of rounded black spots.
The vertebral spots, which are larger, extend onto the tail. The
series of 39 smaller dorsolateral spots (three to four scales in length)
terminate at the vent. The venter is black; this color is separated
from the dark brown color of the dorsum by a narrow zigzag white
line, the angles of which correspond to the dorsal spots. A median
series of pale brown or \\'hitish brown, transversely arranged spots
are connected with the angles of the zigzag lines. The tail has pale
brown transverse lines. The head is dark brown with black and
white markings. The chin and throat are white.

Remarks. — This species is known from three specimens: They
resemble E. bennetti in having 21 dorsal scale rows and a narrow
internasal not reaching the loreal, but they apparently constitute a
distinct species in having fewer ventrals (129-137) and more sub-
caudals (61-74) than E. J)ennetti.

Distribution. — All specimens of this species are from Cambodia
(Fig. 14). Specimens examined, three, as follows.

Cambodia: Tonle Sap, MNHN 48:96; near Tonle Sap, MNHN 48.95,
38.143.

CoLUBRiD Snakes, Subfamily Homalopsinae

107

0 100

Ml II

100

120

Fig. 14. Locality records for Eiihydiis bocourti (circles), Enhydris longicauda
( stars ) , and Enhydris maculosa ( triangles ) .

The Enhydris macleayi Group

This group, composed of three species — macleayi, polylepis and
matanensis — is characterized by the nasal cleft extending to the
second upper labial or to the interlabial suture between first and
second upper labials; fourth and fifth, or fifth and sixth upper labials
entering orbit; loreal in contact with second and third or second to
fourth upper labials; dorsal scales in 21 or 23 rows at midbody.

Members of this group are found in Celebes, New Guinea and
northern Australia.

108 University of Kansas Publs., Mus. Nat. Hist.

Enhydris macleayi (Ogilby)
(new combination)

Pseudoferania macleayi Ogill^y, Proc. Linnean Soc. New South Wales, ser. 2,
5:51, 1891 [Holotype. — Australian Museum from Herbert River, Queens-
land, Australia; J- A. Boyd collector].

HypsirJiina macleayi; Boulenger, Catalogue snakes British Museimi, 3:9, 1896.

Diagnosis. — Nasal cleft extending from nostril to first labial;
internasal single, or divided, widely separated from loreal; fourth
and fifth or fifth and sixth, upper labials entering orbit; one or two
postoculars; anterior pair of chin-shields in contact with first three
lower labials; dorsal scales in 21 or 23 rows at midbody; ventrals
147-152.

Description and variation. — The body is slightly compressed and
broader at midbody. The head is depressed and the snout is nar-
rowly pointed. The rostral is twice as broad as high and is barely
visible from above. The nasals are large and are broadly in contact
with each other behind the rostral. The internasal is single or
divided, and is almost twice as broad as long. The prefrontals form
a median suture and are larger than the nasals. The frontal is
lanceolate having the apex wedged between the parietals, broader
than the supraocular, nearly twice as long as broad, as long as its
distance from the end of the snout, and shorter than the parietals.
The single loreal is a triangular plate, which is broadest anteriorly,
twice as long as broad, and in contact with the second and third
upper labials. The supraocular is a rectangular plate, which is
slightly broadest posteriorly. One high preocular and two subequal
postoculars are present. There are one primary, two secondary and
three tertiary temporals. None of the eight or nine upper labials is
divided. In specimens having eight upper labials, the fourth and
fifth enter the orbit, and in specimens having nine, the fifth and
sixth enter the orbit. The penultimate upper labial is the largest,
and the last is the smallest. There are usually twelve lower labials,
of which the first three are in contact with the anterior pair of chin-
shields. The two pairs of chin-shields are of equal size, and the
scales of the posterior pair are separated by small scales. Nine or
ten gulars and one to four incomplete ventrals separate the chin-
shields from the first complete ventral. The umbilical scar is
situated on or between 13-15 ventrals anterior to the vent. The
narrow ventrals are not quite three times the width of the adjacent
dorsal scales on the first row; the edges of the ventrals are rounded.
Three males and two females have 148-150 (149.3) and 148-152
(150.0) ventrals, and 42-46 (43.7) and 36-41 (38.5) subcaudals, re-

CoLUBRiD Snakes, Subfamily Homalopsinae 109

specti^■ely. The dorsal scales are slightly keeled on the posterior
third of the body in both males and females. There are 23 (rarely
21) rows at midbody which reduce to 19 rows just before the vent.
The dorsal scale reductions in one male (BMNH 93.3.14.5) and one
female (MCZ 35067) are, respectively:

4+5(5) 4+5(78) 4+5(123)
25 23 21 19(148).

4+5(5) 4+5(76) 3+4(121)

5+6(10) 5+6(98) 4+5(136)

25 23 21 19(152).

4+5(11) 4+5(96) 3+4(136)

The 15 or 16 maxillary teeth are followed by an interspace and |

two enlarged, grooved teeth. The 22 or 23 mandibular teeth de-
crease in size posteriorly.

The hemipenis extends to the 9th subcaudal, and the point of
bifurcation is at the 5th subcaudal. The retractor penis magnus
muscle originates on the caudal vertebra adjacent to the 25th or
26th subcaudal. The basal part of the hemipenis is naked; medially
the organ bears strong curved spines, which gradually change into
papilla-like structures distally without altering their size or number.
The sulcus is simple.

The dorsum is gray or brown. A rostral-orbital-postorbital stripe
continues onto the body on the fourth or fifth row of dorsal scales.
The nape stripe originates on the occiput and extends posteriorly as
a broad (four scales wide) line for a head length, and then continues
as a pair of narrow paravertebral lines on the anterior third of the
body. The remaining two thirds of the body and the tail have black
wavy crossbars, one to two scales in length, separated by inter-
spaces two to three scales in length. A black streak extends between
the ventrals and the first row of dorsal scales; the streak is broken
into spots on the posterior third of the body. A broad black stripe
is present on the midventral surface of the tail.

Males have relatively longer tails than females. The tail /snout-
vent ratios for two males are 19.1 and 19.4, and for two females are
16.0 and 17.7. In males a series of five to six tuberculate postanal
scales is present on each side of the posterior margin of the vent.

Remarks. — The three specimens (holotype in the Australian
Museum, and two other specimens in the Macleay Museum at Syd-
ney University), on which Ogilby based his description, have 21
rows of dorsal scales, two preoculars and one postocular, and pre-
frontals having an azygous shield.

University of Kansas Publs., Mus. Nat. Hist.

Distribution. — Snakes of this species are known only from
Queensland, Australia (Fig. 15). Specimens examined, five as fol-
lows.

Australia: Queemland—E.erhert River, BMNH 93.3.14.4-6,
SMF 19480; Cape York, MCZ 35067.

Enhydris polylepis (Fischer)

Hypsirhina polylepis Fischer, Abh. Nat. Ges. Hamburg, 9:14, 1886 [Holotype.
— Dresden Museum; collector unknown]. Boulenger, Catalogue snakes
British Museum, 3:9, 1896.

Enhydris polylepis; Kinghorn, Snakes of Australia, p. 86, 1929. Worrell, Rep-
tiles of Australia, p. 106, 1963.

Diagnosis. — Nasal cleft extending from nostril to first (rarely
second) labial; intemasal single or divided, touching loreal; fifth, or
fifth and sixth, upper labials entering orbit; two postoculars; first
three (rarely four) lower labials in contact with anterior pair of
chin-shields; dorsal scales in 21 or 23 rows at midbody; ventrals 137-
158.

Description and variation. — The body is cylindrical, and the

Fig. 15. Locality records for Enhydris maclcayi (closed stars), Enhydris poly-
lepis (open star in circle), Heurnia ventroniaculata (triangle), and Myron rich-
ardsoni (closed circles for specimens examined; open circle for literature record).

CoLUBRiD Snakes, Subfamily Homalopsinae 111

head is slightly depressed. The rostral is broader than deep and is
narrowly visible from above. The nasals are large and broadly in
contact \\'ith each other behind the rostral. The internasal is single
or divided, usually twice as broad as long, and in contact with the
loreal. The prefrontals form a median suture and are larger than
the nasals; occasionally the prefrontals extend between the loreal
and preocular to meet with the upper labial, as in AMNH 69309,
82443, USNM 128450; or have an azygous shield in between the
prefrontals as in AMNH 82443, BMNH 86.5.20.20. The frontal is
lanceolate, broader than the supraocular, twice as long as broad,
slightly longer than its distance from the end of the snout, and
slightly shorter than the parietals. The single loreal is longer than
hish and in contact with the second and third, or the second to the
fourth, upper labials. In 20 per cent of the specimens, the loreal is
separated from the preocular by the prefrontal, which is in contact
with the upper labials. The supraocular is a rectangular plate,
which is slightly broadest posteriorly. There is one high (rarely
two) preocular and two subequal postoculars. There are one pri-
mary, two (rarely three) secondary, and three or four tertiary tem-
porals. None of the eight (rarely nine) upper labials is divided;
usually the fifth, or the fifth and sixth, and rarely the fourth and
fifth, enter the orbit. Of the 11 or 12 lower labials the first three
(rarely four) are in contact with the anterior pair of chin-shields,
which separates the posterior chin-shields. The chin-shields are
separated from the first complete ventral by 6-10 (7.2) gulars and
2-4 (3.2) incomplete ventrals. The narrow ventrals are not quite
thrice the width of the adjacent dorsal scales in the first row; the
edges of the ventrals are rounded. The umbihcal scar is situated on
or between 17-20 ventrals anterior to the vent. Three males and
seven females have 137-148 (142.7) and 141-158 (147.6) ventrals, and
40-43 (41.3) and 36-47 (41.0) subcaudals. The dorsal scales are in 21
or 23 rows and reduce to 19 (rarely 17) rows just before the vent.
The dorsal scale reduction of one male (USNM 128474) and one
female (AMNH 97265) are, respectively:

5+6(6) 4+5(19) 3+4(98) 3+4(132)

25 23 -— 21 19 17(137).

4+5(3) 4+5(21) 3+4(10.3) 3+4(133)

4+5(2) 4+5(19) +4(29) 4+5(74) 4+5(117)

25 23 22 23 21

4+5(3) 4+5(68) 3+4(113)

19 20 19(145).

+4(118) 3+4(121)

112 Unwersity of Kansas Publs., Mus. Nat. Hist.

The 13 to 16 maxillary teeth are followed by an interspace and
two enlarged, grooved teeth.

The hemipenis extends to the level of the 8th or 9th subcaudal,
and the point of bifurcation is at the level of the 3rd subcaudal.
The retractor penis magnus muscle originates on a vertebra at the
level of the 31st or 32nd subcaudal. Distal to the naked basal region
of the hemipenis is a spinous region having curved spines, which
distally grade into papilla-like structures without change in size or
number.

The dorsum and sides are dark olive or black. The second and
third rows of dorsal scales are variegated with yellow and black.
The venter is yellow. The ventrals on the anterior third of the body
and the lateral edges of the ventrals on the rest of the body are
mottled with black. The tail is black and the subcaudals are edged
posteriorly with yellow. From the angle of the mouth a yellow
stripe extends posteriorly along the side of the body on the third
row of dorsal scales. The head is black, and the chin and throat are
variegated with dark brown and black.

Males have longer tails than females. The tail/ snout-vent ratios
for three males and eight females are 16.4-21.0 (19.3) and 16.0-17.5
(16.9), respectively.

Remarks. — This species is said to be strictly a fresh-water spe-
cies in the Fly River and Lorentz River in southeastern New Guinea.
In Australia, it is confined to rivers in the northern part of Queens-
land and Groote Eylandt (Fig. 15).

Distribution. — This species is found in northern Australia and
southeastern New Guinea. Specimens examined, 12, as follows.

Australia: Northern Territon/ — Groote Eylandt, Point Langdon, USNM
128474; Umba Kuml)a, USNM 128450. Queensland — no definite locality,
BMNH 1903.10.19.33, 1911.4.1.8, FMNH 97265, MNHN c3462; Cairns,
AMNH 82443; Tozer Range, AMNH 69306. New Guinea: Papua— ¥\y River,
AMNH 57518, BMNH 86.5.20.20, 86.5.20.22 (2).

Enhydris matannensis (Boulenger)

Hijpsirhina matannensis Boulenger, Proc. Zool. Soc. London, p. 225, pi. 15,
fig. 1, 1897 [Holotype.— NMB 1735 from Lake Matanna, Celebes; P. and
F. Sarasin collectors]. Schenken, \'erh. Natur. Ges. Basel, 13:116, 1901.
Rooij, Reptiles Indo-Australian Archipelago, 2:182, 1917.

Enhydris matannensis; Hass, Truebia, 20(3) : 576, 1950.

Diagnosis. — Nasal cleft extending from nostril to first labial;
internasal divided, not in contact with loreal; fourth and fifth upper
labials entering orbit; one postocular; anterior pair of chin-shields
in contact with first five lower labials; dorsal scales in 21 rows at
midbody; ventrals 137.

CoLUBRiD Snakes, Subfamily Homalopsinae 113

Description. — The rostral is broader than high. The nasals are
large and form a median suture. The internasal is divided. The
frontal is broader than the supraocular, twice as long as broad,
longer than its distance from the end of the snout, and shorter than
the parietals. The single loreal is slightly longer than high and in
contact with the second and third upper lal)ials. There is one
preocular and one postocular. There are one primary, two secon-
dar\', and three tertiary temporals. Of the eight upper labials, the
fourth and fifth enter the orbit. Of the ten lower labials, the sixth
is largest and the first five are in contact with the anterior pair of
chin-shields. The posterior pair of chin-shields are smaller and are
separated by small scales. The only known specimen, a male, has
137 ventrals, 43 subcaudals, a total length of 240 mm., and a tail
length of 65 mm. The dorsal scales are in 21 rows at midbody and
reduce to 19 rows just before vent. The scale reduction in one male
(NMB 1735) is,

3+4(93)
21 19(137).

3+4(96)

The dorsum is dark olive brown with an indistinct darker line
along the vertebral scale row. The throat is yellowish white with
large transverse olive brown spots. The ventrals at the middle of
the body are almost entirely olive brown; the ventrals posteriorly
are olive brown at their bases and edged with yellowish white. On
the tail is an olive brown midventral line. The lower labials, chin,
and throat are heavily tuberculate in a male specimen (NMB 1735).

Remarks. — The description is based on one immature male. It
resembles E. longicoucla and E. bennetti in having 21 dorsal scale
rows around the midbody and a narrow internasal, which does not
reach the loreal, but it differs from those species by having the
loreal in contact with the second and third labials.

Distribution. — Known only from the southeastern Celebes (Fig.
4). Specimen examined, one (holotype), from Celebes, Lake Ma-
tanna, NMB 1735.

The Enhydris maculosa Group

This group, composed of two species — maculosa and palwngen-
sis — is characterized by a single internasal not in contact with
loreal; fourth upper labial entering orbit; usually first four lower
labials in contact with the anterior pair of chin-shields; dorsal scales
in 25 rows at midbody.

114 University of Kansas Publs., Mus. Nat. Hist.

Both species are rare and have restricted distribution; E. macu-
losa is known only from Bunna, and E. pahangensis occurs only in
Malaya.

Enhydris maculosa (Blanford)

Hypsirhina inactilaia (non Dumeril and Bibron) Blanford, Jour. Asiatic Soc.
Bengal, 48:130, 1879 [Holotype. — Calcutta Museum from Pegu District,
Burma].

Hypsirhina maculosa; Blanford, Proc. Zool. Soc. London, p. 226, 1881 [Substi-
tute name for H. maculata Blanford, preoccupied by H. maculata Dumeril
and Bibron, 1854].

Hypsirhina blanfordi; Boulenger, Fauna of British India . . . Reptilia and
Batrachia, p. 377, 1890 [Substitute name]. Sclater, Jour. Asiatic Soc.
Bengal, p. 244, 1891. Boulenger, Catalogue snakes British Museum, 3:10,
1896. Mocquard, Revue Coloniale, p. 51, 1907. Phisalix, Animaux veni-
meux et venins, 2:285, 1922. Werner, Arch. Natiug., 89A(8):161, 1923.
Wall, How to identify the snakes of India, p. 37, 1923; Jour. Bombay, Nat.
Hist., 29:966, 1924. Bourret, Invent. Gen. Indochine, 3:241, 1927; Ser-
pentes de I'lndocliine, p. 289, 1936.

Enhydris maculosa; Smith, Fauna of British India . . . Reptilia and Amphibia,
3:387, 1943.

Diagnosis. — Nasal cleft extending from nostril to first labial;
internasal single, separated from loreal; fourth upper labial entering
orbit; two postoculars; anterior pair of chin-shields in contact with
first three or four lower labials; dorsal scales in 25 (rarely 23) rows
at midbody; ventrals 123-157.

Description and variation. — The body and tail are oval, and the
head is depressed and slightly distinct from the neck. The rostral is
nearly twice as broad as high. The nasals are large and form a
median suture. The internasal is single, small, and separated from
the loreal. The prefrontals form a median suture and are smaller
than the nasals. The frontal is broader than the supraocular, not
quite twice as long as broad, shorter than its distance from the
end of the snout, and shorter than the parietals. The loreal is about
as long as high and is in contact with the first three upper labials.
The supraocular is a rectangular plate slightly broadest posteriorly.
One high preocular and two postoculars are present. Of the seven
(rarely eight) undivided upper labials, the fourth enters the orbit.
Usually ten, rarely nine or eleven, lower labials are present; the
first four (rarely three) lower labials are in contact with the anterior
pair of chin-shields. The posterior pair of chin-shields is almost the
same size as the adjacent scales. There are six to eight gulars and
two or three incomplete ventrals between the chin-shields and the
first complete ventral. The ventrals are approximately thrice the
width of the adjacent dorsal scales in the first row; the edges of the
ventrals are round. The umbilical scars on three specimens are on
18-21, 20-22, and 21-24 ventrals anterior to the vent. Six males and

CoLUBRiD Snakes, Subfamily Homalopsinae 115

nine females have 143-151 (147.5) and 123-155 (141.2) ventrals, and
six males and seven females have 48-59 (52.5) and 32-47 (41.6) sub-
caudals. The coefficients of the differences between the means are
1.15 and 1.26 and are not significant. The dorsal scales are in 25
(rarely 23) rows at midbody and reduce to 21 (rarely 19) rows just
before the vent. The dorsal scale reductions for two males (KU
92467, 92464) and one female (KU 92468) are, respectively:

6+7(5) 5+6(16) 4+5(109) 3+4(135)

29 27 25 23 21(151).

6+7(4) 4+5(39) 5+6(109) 3+4(135)

7+8(4) 5+6(13) 4+5(96) 4+5(126) 3+4(143)

29 27 25 23 21

6+7(4) 5+6(15) 4+5(95) 4+5(128) 3+4(142)

19(148).

7+8(2) 7+8(5) 5+6(124) 3+4(130)

29 27 25 23 21(142).

4+5(2) 4+5(5) 5+6(122) 4+5(133)

The 13 maxillary teeth are followed by an interspace and two
enlarged, grooved teeth. The 20 to 22 mandibular teeth decrease in
size posteriorly.

The hemipenis extends to the 8th subcaudal, and the point of
bifurcation is at the 5th subcaudal. The retractor penis magnus
muscle originates on the caudal vertebra adjacent to the 22nd sub-
caudal. The basal part of the hemipenis is naked. Proximal to the
point of bifurcation strong curved spines are present. Distally on
the branches of the hemipenis, the spines are smaller but without
change in size and number.

The dorsum is ashy-black. A black median nape stripe extends
from the parietals to the occipital region where it continues as an
interrupted vertebral stripe about three scales wide. The inter-

FiG. 16. Lateral view of the body of Enhijdris maculosa (MCZ 18390), Xl.5.

116 University of Kansas Publs., Mus. Nat. Hist.

rupted lateral stripe originates from the anterior upper labials and
continues along the side of the body; the stripe is about three scales
wide and occupies the fifth, sixth, and seventh rows of dorsal scales.
A white or yellow line with zigzag edges extends posteriorly from
the angle of the mouth, along the side of the body; the line occupies
the third and fourth rows of dorsal scales. The lateral edges of the
ventrals and the first and second rows of dorsal scales form a black
line which continues anteriorly to the mental. The midventral yel-
low line resembles a pile of cones one on top of the other. The tail
is black. Yellow spots on the first and second rows of caudal scales
and on the subcaudals are in four longitudinal series (Fig. 16).

Males have longer tails than females. The tail/snout-vent ratios
for six males and seven females are 22.8-26.6 (24.4) and 15.2-23.1
(17.9); the coefficient of the difference between the two means is
1.62, which is significantly different. Males have a series of five or
six tuberculate postanal scales on each side of the posterior margin
of the vent.

Remarks.— According to Sclater (1891:245) the type specimen is
lost. The only specimen in the British Museum was presented by
Loveridge and lacks locality data.

Distribution. — This species is known (Fig. 14) from Burma,
China and Sumatra. Specimens examined, 18, as follows.

No locality, BMNH 1913.6.12.1. Burma: Inawaddij D/t/.s/on— Maubin,
KKG 93-94, KU 92395, 92397, 92464-69. China: no locality, MNHN
c3453(2); Canton River, ANSP 5095-8. Indonesia: Nais Island— no definite
locality, MCZ 18390.

Enhydris pahangensis Tweedie

Enhydris pahangensis Tweedie, Ann. Mag. Nat. Hist., ser. 11, 13:142-144,
1946 [Holotype.— BMNH 1947.1.1.70 from Pahang, Malaya; collector
Tweedie]; The snakes of Malaya, pp. 88-89, 1957.

Diagnosis. — Nasal cleft extending from nostril to second labial;
internasal single, not in contact with loreal; fourth upper labial
entering orbit; t\\o postoculars; anterior pair of chin-shields in
contact with first four lower labials; dorsal scales in 25 rows at
midbody; ventrals 126 (Fig. 17).

Description. — The rostral is as broad as high. The nasals are
large and form a median suture. The internasal is single, twice as
broad as long, and not as long as the prefrontals. The frontal is
about twice as long as broad, as long as its distance from the end of
the snout, and as long as the parietals. The single loreal is in con-
tact with the second and third upper labials. One high preocular
and two subequal postoculars are present. One primary, two sec-

CoLUBRiD Snakes, Subfamily Homalopsinae

117

ondary, and three tertiary temporals are present. Of the eight
undi\ ided upper labials, the fourth enters the orbit, and the seventh
is the largest. Of the eleven lower labials, the sixth is the largest,
and the first five are in contact with the anterior pair of chin-shields.
The posterior pair of chin-shields are small and are separated by a
pair of small scales. One male has 126 ventrals, 55 subcaudals, a
total length of 220 mm., and a tail length of 40 mm. The dorsal
scales are in 25 rows at midbody and reduce to 21 rows just before
the vent. The scale reduction in one male (BMNH 1947.7.1.70) is,

5+6(8) 5+6(91) 4+5(108)

27 25 23 21(126).

5+6(8) 5+6(91) 4+5(108)

The dorsum is grayish brown with small scattered dark spots on
each scale. A broad pale yellow band occupies the first four rows of
dorsal scales and is bordered above by a line of dark dots and below
by a zigzag dark line extending onto the ventrals. A broad white
lateral stripe on each side of the head meets a poorly defined dark
mark extending posteriorly from each corner of the mouth. The
venter is yellow, mottled with brown.

Remarks. — Only one immature male specimen of this snake is
known. It was collected in April, 1940 at Kuala Tahang in the King

Fig. 17. Ventral, lateral, and dorsal \-ie\\s of the head of the holotype of En-
hydris pahangensis (BMNH 1947.1.1.70), Xl.5.

118 Unr^ersity of Kansas Publs., Mus. Nat. Hist.

George V National Park, between 500 and 1000 feet altitude, and
over 70 miles from the east coast of the Malay Peninsula.

Tweedie (1946:144) pointed out that this species was most
closely related to E. cJunemis. I think that E. pahangensis more
closely resembles E. maculosa than E. chinensis. Enhijdris pahang-
ensis and E. maculosa have the following in common: A single
internasal posterior to the nasal and not touching the loreal; the
fourth upper labial entering the orbit; 25 rows of dorsal scales; eight
undivided upper labials, of which the seventh is the largest. Enhij-
dris pahangensis differs from E. maculosa in having the first labial
not touching the loreal, fewer ventrals and more subcaudals (as
compared with male E. maculosa), and in the pale yellow lateral
stripe being four scales wide instead of only two.

Distribution. — Known only from Malaya (Fig. 11). Specimen
examined, one (holotype), from Malaya, Pahang, Kuala Tahang,
BMNH 1947.1.1.70.

The Enhydris sieboldi Group

This group, composed of two species — sieboldi and dussumieri
— is characterized by two internasals not in contact with loreal;
fourth upper labial entering eye; loreal in contact with first three
upper labials; dorsal scales in 27 or 29 rows at midbody.

Members of this group range from India to Malaya.

Enhydris sieboldi (Schlegel)

Homalopsis sieboldi i Schlegel, Essai sur la physionomie des serpents, 2:349,
pi. 13, figs. 4-5, 1837 [Holotype.— RNHL 1168 from Bengal; collector
unknown]. Cantor, Catalogue Malay reptiles, p. 98, 1847.

Ferania sieboldii; Gray, Zoological miscellany, p. 67, 1842; Catalogue snakes
British Museum, p. 66, 1849. Giinther, Reptiles of British India, p. 284,
1864. Anderson, Proc. Zool. Soc. London, p. 180, 1871. Theobald, Cata-
logue reptiles British India, p. 184, 1876. Murray, Jour. Bombay Nat.
Hist. Soc. 1:219, 1886.

Trigonurus sieboldii; Dumcril and Bibron, Erpetologie generale . . . reptiles.
Paris, 7:960, 1854.

Hypsirhina sieboldi; Jan, Elenco s>stematico degli ofidi, p. 78, 1863; Arch.
Zool. Anat. Phys., 3:260, 1865. Jan and Sordelli, Iconographie generale des
Ophidiens, liw 30, pi. 4, fig. 2, 1868. Boulenger, Fauna of British India . . .
Reptilia and Batrachia, p. .377, 1890. Sclater, Jour. Asiatic Soc. Bengal,
60:245, 1891. Boulenger, Catalogue snakes British Museum, p. 11, 1896.
Flower, Proc. Zool. Soc. London, p. 887, 1896. Wall, Tour. Bombay Nat.
Hist. Soc, 11:732, 1897. Flower, Proc. Zool. Soc. London, p. 677, 1899.
Werner, Abh. Bayer, Akad. Wissen, ser. 22, 2:367, 1903. Boulenger,
Fasciculi Malayenses. Zoology, 1:175, 1903. Wall, Jour. Bombay Nat.
Hi.st. Soc, 17:3, 1906; Jour. Bombay Nat. Hist. Soc," 18:103, 117, 920,
1907. Boulenger, A vertebrate fauna of the Malay Peninsula . . . Reptilia
and Batrachia, p. 161, 1912. Phisalix, Animaux venimeux et venins, 2:285,
1922. Werner, Arch. Naturg. 89A(8):161, 1923. Wall, Jour. Bombay
Nat. Hist. Soc, 29:866, 1924.

CoLUBRiD Snakes, Subfamily Homalopsinae 119

Fciaui()i(h\ iaiinuuiictis Carlleyle, joiir. Asiatic Soc. Bengal, 38:196, 1869
[H()l()t\po. — In Calcutta Museum lioni juinnia River, near vVgra; collector
inikn()\\n|.

Euhydris sichohli; Mcll, Jour. Lingnan Uni\-. Sci., pp. 253, 264, 1929. Gee,
Bull. Biol. Yenehing Uni\'., p. 72, 1930. Smith, Fauna of Briti.sh India . . .
Reptilia and Amphibia, 3:389, 1943.

Didiinosis. — Nasal cleft extending from nostril to first labial;
intcrnasal divided, not in contact with loreal; fourth upper labial
entering orbit; two postoculars; first four lower labials in contact
with anterior pair of chin-shields; dorsal scales in 27 or 29 (rarely
33) rows at midbody; ventrals 143-156.

Descriptioti and variation. — The body and tail are ovoid, and
the head is slightly depressed. The rostral is nearly as deep as
broad. The nasals are large and broadly in contact with each other
behind the rostral. The internasal is divided, twice as broad as long,
and rarely in contact with the loreal. The prefrontals usually form
a median suture (sometimes an azygous shield is present between
them), and are larger than the nasals. The frontal is pentagonal,
broader than the supraocular, as long as its distance from the end of
the snout, and as long as the parietals. The loreal is single, as long as
deep, and in contact with the first three upper labials. The supra-
ocular is a rectangular plate, broadest posteriorly. One high pre-
ocular and two subequal postoculars are present; sometimes there is
a small subocular below the eye. Usually eight (rarely seven) upper
labials are present; of these the fourth (rarely third) enters the eye,
and the sixth (rarely fifth) is the largest; the last is the smallest, and
the seventh (rarely sixth) is divided. Of the 11 or 12 lower labials,
the first four are in contact with the anterior pair of chin-shields.
The anterior chin-shields are about twice the size of the posterior
pair, and the scales of the latter are separated by a pair of small
scales. Nine or ten gulars and two or three incomplete ventrals
separate the chin-shields from the first complete ventral. The ven-
trals are almost three times the width of the adjacent dorsal row of
scales. The edges of the ventrals are round. In two specimens the
umbilical scars are situated on or between 29-32 ventrals anterior
to the vent. Three males and five females have 144-153 (147.7) and
143-156 (150.0) ventrals, and two males and five females have 51-52
(51.5) and 46-56 (50.4) subcaudals. The dorsal scales are in 27 or 29
(rarely 33) rows at midbody and reduce to 21 rows just anterior to
the vent. The dorsal scale reductions in one male (BMNH 51.5.9.2)
and one female (MCZ 22384) are, respectively:

120 University of Kansas Publs., Mus. Nat. Hist.

5+6(4) 7+8(59) 5+6(86) 5+6(90) 5+6(133)

31 29 27 25 23

5+6(5) 5+6(73) 5+6(84) 4+5(107) 4+5(130)

21(144).

5+6(4) 8+9(84) 8+9(91) 5+6(130)

29 27 25 23 21(148).

5+6(5) 7+8(83) 6+7(93) 4+5(128)

The 11 or 12 maxillary teeth are followed by an interspace and
two enlarged grooved teeth. The 18 or 19 mandibular teeth de-
crease in size posteriorly.

The hemipenis extends to the 14th subcaudal, and the point of
bifurcation is at the 11th subcaudal. The retractor penis magnus
muscle originates on the caudal vertebra adjacent to the 32nd sub-
caudal. The naked basal part of the hemipenis extends to about the
4th subcaudal. The median part bears straight spines, which be-
come shorter and eventually small and numerous distally. The sul-
cus is forked.

The dorsum is yellow or pale brown. The body is marked by
20-26 black-edged, elliptical or rhomboidal, dark brown transverse
blotches, five to eight scales in length, separated by interspaces
about two scales in length. A series of dark brown, round spots
(two scales in length and breadth) on each side alternate with the
dorsal blotches. Dark brown rings four or five scales in length
occur on the tail. The head is marked by three dark brown, longi-
tudinal stripes narrowly separated by yellow interspaces. The ven-
ter is variegated with yellow and dark brown spots.

Males have relatively longer tails than females. The tail/ snout-
vent ratios for six males and six females are 20.0-22.8 (21.3) and
17.7-21.1 (20.1), respectively. The coefficient of the difference be-
tween the two means is 0.49, \\'hich is not significant.

Re/?3fl;7vs.— Anderson (1871:181) and Gimther (1864:284) stated
that this species had seven prediastemal maxillary teeth and one
posterior tooth.

Carlleyle (1869) named and described Feranioides iammaeticus.
In a footnote, the editor remarked, "The snakes principally differ
from Ferania [^lEnlujdris] by its round pupil, and is in this respect
one of the rare instances recorded among the Homalopsidae, most
of which have a narrow vertical pupil of the eye. The dentition
would also appear to be peculiar. . . ."

Anderson (1871:181) pointed out that these remarks were
founded on an imperfect drawing, and therefore their inaccuracy
was not surprising. All the specimens of E. sieboldi, that I have

CoLUBRiD Snakes, Subfamily Homalopsinae 121

examined ha\e \ertical pupils and dentition characteristic of the
genus EnJiydris.

Distribution. — The snakes of this species are found in India as
far west as Bombay and eastward to Malaya (Fig. 6). Specimens
examined, 14, as follows.

East Pakistan: Dacca Division — Mymensingh, BMNH 29.149.51. India:
Bengal— no definite locality, MXHN c3463, RNHL 1168. Dc//i/— BMNH
190S.1.23.67-6cS. .A/fl/;flra.s7i/ra— Bombav, BMNH 87.12.23.1. Uttar Pradesh—
BMNH 1907.2.14.32-37, MCZ 22384. Malaya: Penang, BMNH 60.3.19.118.

Enhydris dussumieri (Dumeril and Bibron)

Enrostus dussumieri Dumeril and Billion, firpetologie generale . . . reptiles.
Paris, 7(2)953, 1854 [Holotype.— MNHN c3458 from Bengal; collector
unknown]. Boulenger, Catalogue snakes British Museum, 3:19, 1896.
Mocquard, Re\ ue Coloniale, p. 51, 1907. Phisalix, Animaux venimeux et
venins, p. 285, 1922. Werner, Arch. Naturg., 89A(8):163, 1923. Bourret,
Invent. Gen. Indochine, 3:241, 1927; Serpents de I'lndochine 1:298, 1936.

Hypsirhina dussumieri; Jan, Elenco systematico degli ofidi, p. 78, 1863; Arch.

Zool. .\nat. Ph\s., 3:260, 1865; Iconographie generale des Ophidiens, liv.

30, pi. 3, fig. 1, 1868. Morice, Sur la Faune de la Cochinchine francaise.

Lyon, p. 58, 1875. Tirant, Excursions et Reconnaissance, 20:403, 1885.
Hypsirhina meIaJ)arica Werner, Jahrb. Wiss. Anst. Hamburg, 30(2):26, 1913

[Ho]ot\pe. — Hamburg Museum from Cochin, Malabar coast, India; collector

unknown].

Enlit/dris dussumieri; Smith, Fauna of British India . . . Reptilia and Amphibia,
3:389, 1943.

Diagnosis. — Xasal cleft extending from nostril to internasal;
internasal divided, narrowly separated from the loreal; fourth upper
labial entering orbit; two postoculars; anterior pair of chin-shields
in contact with first four or five lower labials; dorsal scales in 27
rows; ventrals 143-147.

Description and variation. — The body is cylindrical. The head
is small, scarcely distinct from the neck. The snout is truncate, and
the rostral is broader than high. The nasals are semi-divided and
are in contact with each other behind the rostral. The internasal is
divided and nearly twice as broad as long. The prefrontals form a
median suture behind the internasals. The frontal is nearly as broad
as the supraocular, not quite twice as long as broad, as long as its
distance from the end of the snout, and slightly shorter than the
parietals. The single loreal is as long as high and is in contact with
the first three upper labials. One high preocular and two subequal
postoculars are present. There are one primary, two secondary, and
three tertiary temporals. None of the eight upper labials is divided;
the fourth enters the orbit, and the seventh is the largest. Of the 12
or 13 lower labials, the sixth is the largest, and the first four or five
are in contact with the anterior pair of chin-shields. Both pairs of
chin-shields are small, but the anterior pair is slightly larger than

122 University of Kansas Publs., Mus. Nat. Hist.

the posterior pair, the scales of which are separated by smaller
scales. Two females have 143 and 147 ventrals; 28 and 34 sub-
caudals. The total length of MNHN c3458 is 665 mm. and the tail
length is 74 mm. The dorsal scales are in 27 rows and reduce to 23
or 21 rows just before the vent.

The 13 or 14 maxillary teeth are followed by an interspace and
two enlarged, grooved teeth.

The dorsum is brown with a vertebral and a pair of dorsolateral
black stripes. The first three rows of dorsal scales, ventrals, and
subcaudals are white with their lateral edges spotted with brown.
A median series of small black spots may be confluent to fonn a
black line.

Remarks. — P. de Grys compared Smith's description of the type
of E. dussnmieri with the type of Hypsirliina malaharica in the
Hamburg Museum. Both workers agreed that the species are con-
specific (Smith, 1943:389). Since I have not seen the type of
Hypsirhina malaharica, I follow the conclusions reached by these
workers.

Distribution. — This species is known only from the southwestern
coast of India and from Bengal (Fig. 4). Specimens examined, two,
as follows.

India: Bengal— no definite locality, MNHN c3458. Kerala — Malabar,
MNHN 5751.

The Enhydris bocourti Group

This group, composed of two species — bocourti and albomacu-
lata is characterized by a nasal cleft extending to first labial; fourth
or fifth upper labial entering orbit; first five or six lower labials in
contact with anterior pair of chin-shields.

Enhydris bocourti is known from Malaya to Vietnam and E.
albomaculata is known from Sumatra to Java.

Enhydris bocourti (Jan)

Hypsirhina bocourti Jan, Arch. Zool. Anat. Ph>.s., 3:258, 1865 [Holohpe. —
Paris Museum, from Bangkok, Thailand; collector unknown]; Iconographie
generale des Ophidiens, liv. 28, pi. 5, fig. 2, 1868. Morice, Sur la Faune
de la Cochinchine francaise, p. 58, 1875. Tirant, Excursions et Reconnais-
sances, 20:403, 1885. Bocourt, Nouw Arch. Mus., p. 9, 1886. Boulenger,
Catalogue snakes British Museum, 3:10, 1896. Flower, Proc. Zool. Soc.
London, p. 676, 1899. Laidlaw, Proc. Zool. Soc. London, p. 578, 1901.
Boulenger, Fasciculi Malavensis. Zoologv', 1:164, 1903. Mocquard, Mis-
sion Pavie Indochine, 3:482, 1904; Revue Coloniale, p. 51, 1907. Boulen-
ger, A vertebrate fauna of the Malav Peninsula . . . Reptilia and Batrachia,
p. 161, 1912. Smith, Jour. Nat. Hist. Soc. Siam, 1:100, 1914. Gyldenstolpe,
Svensk. vent. Akad. Handl., 55(3):18, 1916. Phisalix, Animaux venimeux
et venins, 2:285, 1922. Werner, Arch. Naturg., 89A(8):151, 1923. Bourret,
Invent. Gen. Indochine, 3:241, 1927.

CoLUBRiD Snakes, Subfamily Homalopsinae 123

Ferania sieboklii (non Schlegel); Giinther, Ann. Mag. Nat. Hist., ser. 3, 18:28,
1866.

Hijpsirliina vuiJtilincata Tirant, Excursions et Reconnaissances, 20:403, 1885
[Molotvpe. — Paris Mnsenm from Cochinchina; collector nnknown]; Mission
Pavie indochine, 3:484, 1904. Werner, Arch. Naturg., 89A(8):160, 1923.

Hijpsirhina gigantca Werner, Ann. Natinhist. Mns. Wien, 36:163, 1923 [Holo-
tvpe in Vienna Mnsenm; tvpe-locality and collector unknown]; Zool. lahrb.
Jena, 57:165, 1928. Smith, Ann. Mag. Nat. Hist., ser. 10, 1:495, 1928.

Enliydris hocourti; Smith, Bull. Raffles Museum, 3:61, 1930; Fauna of British
India . . . Reptiha and Amphibia, 3:388, 1943.

Ht/psirliina ])ocourti soctrangensis Bourret, Serpents de I'lndochine, 1:291,
1936 [Holotype.— MNHN 38144 from Soc Trang; collector Bomret|.

Diagnosis. — Nasal cleft extending from nostril to first labial;
internasal single (rarely divided), touching or just separated from
loreal; fourth (rarely fifth) upper labial entering orbit; two post-
oculars; first five lower labials in contact with the anterior pair of
chin-shields; dorsal scales in 27 (rarely 29) rows at midbody; ven-
trals 120-136.

Description and variation. — The body is stout and cylindrical,
and the tail is short. The head is deep and distinct from neck. The
rostral is broader than high and narrowly visible from above. The
two large nasals form a median suture behind the rostral. The
internasal is single (about 90%) or rarely divided (about 10%), twice
as broad as long, and in contact with, or just separated from, the
loreal. The prefrontals form a median suture and are larger than
the nasals. The frontal is an elongate plate, which is narrower than
the supraocular, at least twice as long as broad, as long as its dis-
tance from the end of the snout, and a little shorter than the
parietals. The loreal is single, slightly longer than high, and in
contact with the first two or three upper labials. The supraocular
is a triangular plate, broadest posteriorly. There is one high pre-
ocular and two subequal postoculars; the upper postocular is about
twice as large as the lower one. There are eight or nine upper
labials. In specimens having eight, the fourth enters the orbit, and
the seventh is horizontally divided; in specimens having nine, the
fifth enters the orbit, and the eighth is horizontally divided. Of the
11-15 (12.2) lower labials, the first five (rarely first four) are in con-
tact with the anterior pair of chin-shields, which are pear-shaped
and much larger than the posterior pair. The posterior chin-shields
are separated by a pair of small scales. Nine or ten gulars and two
or three incomplete ventrals separate the chin-shields from the first
complete ventral. The relatively narrow ventrals are not quite three
times the width of the adjacent dorsal scales in the first row. The
edges of the ventrals are round. The umbilical scar is situated on or
between 13-15 ventrals anterior to the vent. Ten males and eight

124 University of Kansas Publs., Mus. Nat. Hist.

females have 123-132 (128.2) and 124-131 (128.0) ventrals and eight
males and females have 37-47 (41.1) and 32-40 (35.6) subcaudals.
The eoefficients of the differences are 0.04 and 0.95 and are not sig-
nificant. The dorsal scales are usually in 27 (rarely in 29) rows at
midbody and reduce to 21 (occasionally 23 to 19) rows just before
the vent. The dorsal scale reduction for one male ( AMNH 92295) is,

6+7(6) 5+6(24) 4+5(100) 1+2(118)
31 29 27 ■ — 25 23(126).

6+7(8) 5+6(28) 4+5(99) 1+2(118)

The 10 or 11 maxillary teeth are followed by an interspace and
t\\o enlarged grooved teeth. The 21 or 22 mandibular teeth de-
crease in size posteriorly.

The hemipenis extends to the 12th or 13th subcaudal, and the
point of bifurcation is at the 7th or 8th subcaudal. The retractor
penis magnus muscle originates on the caudal vertebra adjacent to
the 34th or 35th subcaudal. The basal part of the hemipenis is
naked; medially large curved spines are present, and distally the
spines become smaller and gradually more numerous. The sulcus
is forked.

The dorsum of the body and tail is olive brown with yellow
crossbars [30-40 (31.4) on the body, 7-13 (10.8) on the tail] or a
series of yellow spots one scale in length. Small, median spots are
sometimes present on the intervening scales; in which case, distinct,
longitudinal lines are formed. The venter is yellow. The dark color
of the dorsum tapers into vertical bars on the sides of the body and
forms complete or interrupted rings across the belly. The head is
variegated with pale olive brown and dark olive brown. A yellow
nape stripe originates on the parietals and extends posteriorly about
six or seven scales. Of the two yellow crossbars on the nape, one
crosses near to the posterior edges of the parietals, and the other
crosses along the posterior edges of the occipital scales. The upper
and lower labials, chin, and throat are yellow. The upper labials
and anterior chin-shields are edged with dark olive brown.

Males have longer tails than females. The tail/ snout-vent ratios
for eight males and seven females are 17.1-20.0 (19.0) and 13.6-15.5
(14.4). The coefficient of the difference between the two means is
2.0; the difference is significant.

Remarks. — Enliydris ])ocotuti is the largest species, both in
length and girth in the genus. The largest known specimen, a fe-
male from Bangkok (USNM 70302), has a total length of 1375 mm.
and a tail length of 275 mm. The recognition of the subspecies E. h.
soct range nsis is unwarranted. The holotype of E. 1). soctrangensis is

CoLUBRiD Snakes, Subfamily Homalopsinae 125

like E. bocourti except that it has 29 rows of dorsal scales. This
number of dorsal rows occurs in low frequency throughout the
range of the species.

Distribution. — The snakes of this species are known (Fig. 14)
from Thailand as far north as Paknampo, Cambodia, Cochin China,
and the Malay Peninsula as far south as Kedah. Specimens exam-
ined, nine, as follows.

Cambodia: no definite locality, SU 7296. Cochin China: no definite
locality, FMNH 11549-50, MCZ 5969, USNM 20410; Soc Trang, MNHN
38.144. Malaya: Kcdah—BMNH 98.9.22.44; Alor star, BMNH 98.9.22.44.52,
1903.4.13.76-77. Thailand: no definite locality, BMNH 85.4.28.2; South
Tliailand— Bangkok, AMNH 92295; Trang, AMNH 36278, USNM 22885
Thailand-Cambodia Border: Sre Umliel, BMNH 1938.8.7.19.

Enhydris albomaculata (Dumeril and Bibron)

Homalopsis albomaculatiis Dumeril and Bibron, Erpetologie generale . . . rep-
tiles. Paris, 7(2):974, 1854 [Holotype.— MNHN c3452 from Padang,
Sumatra; collector Kunhardt].

Hypsirhina albomaculata; Jan, Elenco systematico degli ofidi, p. 77, 1863;
Arch. Zool. Anat. Phys., 3:259, 1865; Iconographie generale des Ophidiens,
liv. 28, pi. 5, fig. 1, 1868. Boulenger, Catalogue snakes British Museum,
3:11, 1896. Rooij, Reptiles Indo-Australian Archipelago, 2:183, 1917.

Enhydris albomaculata; Haas, Treubia, 20(3) :575, 1950.

Diagnosis. — Nasal cleft extending from nostril to first labial;
internasal divided, in contact with, or narrowly separated from,
loreal; fourth or fifth labial entering orbit; one or two postoculars;
first six, rarely five, lower labials in contact with anterior pair of
chin-shields; dorsal scales in 27 rows; ventrals 140-151.

Description and variation. — The body and tail are cylindrical;
the head is slightly depressed and is distinct from the body. The
pupil is round. The rostral is nearly twice as broad as high. The
nasals, internasals, and prefrontals form a straight median suture.
The internasals are small, twice as broad as long. The nasals are as
large as the prefrontals. The frontal is broader than the supraocu-
lar, not quite twice as long as broad, as long as its distance from the
end of the snout, and slightly shorter than the parietals. The single
loreal is longer than high and is in contact with the second, third,
and fourth upper labials. The supraocular is nearly rectangular,
slightly broader posteriorly. One high preocular, and one (42%) or
two (58%) postoculars are present. There are eight (40%) or nine
(60%) upper labials; usually none is divided but in some specimens
the last one or two are divided. In specimens having eight upper
labials, the fourth enters the orbit; in those having nine, the fifth
enters the orbit. There are 12 (22%), 13 (66%) or 14 (11%) lower
labials, of which the first six (rarely five) are in contact with the

126 University of Kansas Publs., Mus. Nat. Hist.

anterior pair of chin-shields. The posterior pair of chin-shields is
smaller than the anterior pair, and the scales in the posterior pair
are separated by two small scales. There are six to eight gulars
and one to three incomplete ventrals between the chin-shields and
the first complete ventral. The relatively wide ventrals are about
five times the width of the adjacent dorsal scales in the first row.
The edges of the ventrals are round. On three specimens the
umbilical scar is situated on 24 and 25 ventrals anterior to the vent.
Eight males and eight females have 143-151 (148.6) and 140-147
(142.9) ventrals, and 46-50 (48.5) and 36-40 (37.4) subcaudals re-
spectively. The coefficient of the difterences are 1.05 and 2.74; the
difference in the number of subcaudals is significant. The dorsal
scales are in 27 rows at midbody and reduce to 25 or 23 (rarely 21)
rows just anterior to vent. The dorsal scale reductions in one male
(BMNH 84.12.31.5) and one female (BMNH 84.12.31.6) are, re-
spectively:

6+7(5) 6+7(80) 6+7(131)

29 27 25 23(143).

6+7(6) 4+5(71) 4+5(88)

6+7(4) 5+6(89) 6+7(115)

29 27 25 23 22(140).

6+7(7) 4+5(86) 5+6(114) 5+6(128)

The 13 maxillary teeth are followed by an interspace and two
enlarged grooved teeth; 18-20 mandibular teeth, 17 pterygoid, and
7 or 8 palatine teeth are present.

The hemipenis extends to the 12th subcaudal, and the point of
bifurcation is at the 5th subcaudal. The retractor penis magnus
muscle originates on the caudal vertebra adjacent to the 29th sub-
caudal. The basal part of the hemipenis is naked. The median part
bears spines, which become gradually smaller and more numerous
distallv. The sulcus is forked.

The dorsum is olive brown or black with small yellow or orange
spots. A yellow nape stripe and one or two indistinct yellow cross-
bars are present on the occiput. The chin, throat, first three rows of
dorsal scales, and subcaudals are variegated with \'ello\v and olive
brown or black.

Males have significantly longer tails than females. The tail/
snout-vent ratios for seven males and seven females are 17.2-20.9
(19.4) and 13.0-14.0 (13.5), the coefficient of the difference between
the two means is 4.24 and is significant.

Distri])iition. — The snakes of this species are known from Su-

CoLUBRiD Snakes, Subfamily Homalopsinae 127

inatra, Java and the neighbouring small islands (Fig. 4). Specimens
examined, 21, as follows.

Indonesia: Java — SMF 19465-9. Nia.s Island — no definite locality, MCZ
27104; Piilo, BMNH 84.12.31.5-7. Simculuc Island— no definite locaUty,
I'SNM 30760-7, 30769, 35776, 35778, 35782-3; Siljigo, RNHL 5189; Sinal)ang,
RXHL 4714(2), ZMA d. v. 1913, no number. Sumatra — no definite locality,
MNHN c3452.

The Enhydris punctata Group

This group, composed of two species — punctata and doriae — is
characterized by two tiers of upper labials; sixth upper labial below
eye; second to fifth (or sixth) in contact with anterior pair of chin-
shields; 12 or more upper labials.

Members of this group are found from India to Malaya and
southeastward to Borneo.

Enhydris punctata (Gray)

Pliytolopsis punctata Grav, Catalogue snakes British Museum, p. 68, 1849
[Holotype.— BMNH 1946.1.2.37 from "India"; Wanvick's collection].

Euiostus hctcraspis Bleeker, Nat. Tijdsehr. Nederl. Ind., 14:440, 1859 [Holo-
type.— BMNH 1946.1.2.39 from Sinkawang, Borneo; collector Dr. Bleeker].

Tachyplotus hedcmanni Reinhardt, Vidensk. Meddel., p. 151, 1866 [fide
Boulenger, 1896:12].

PytJionopsis horneensis Peters, Monats. Akad. Wiss. Berlin, p. 576, 1871 [Holo-
type.— Berlin Museum? from Sarawak; from the collection of Marquis
Don a].

PytJionopsis punctata; Peters, Ann. Mus. Genova, 3:37, 1872. Giinther, Proc.

Zool. Soc. London, p. 590, 1872.
Hypsirhina Jiageni v. Lidth de Jeude, Notes Leyden Mus., 12:20, pi. 1, 1890

[Holotype.— RNHL 306 from Delhi, India; collector Dr. B. Hagen].

Hypsirhina punctata; Boulenger, Catalogue snakes British Museum, 3:12, 1896.

Rooij, Reptiles Indo-Australian Archipelago, 2:184, 1917. Westermann,

Treubia, 18:616, 1942.
Enhydris punctata; Kinghorn, Snakes of Australia, p. 89, 1929. Haas, Treubia,

20(3) :577, 1950. Worrell, Reptiles of Australia, p. 106, 1963.

Diagnosis. — Nasal cleft extending from nostril to first labial;
internasal single, widely separated from loreal; usually sixth
(rarely seventh) labial entering orbit; two postoculars; anterior
pair of chin-shields in contact with second to fifth or sixth lower
labials; dorsal scales in 25 (rarely 23 or 27) rows at midbody; ven-
trals 135-156.

Description and variation. — The body is slightly depressed ante-
riorly, and the tail is cylindrical. The head is depressed and is
distinguishable from the body. The rostral is slightly broader than
high. A small single internasal is completely surrounded by nasals
and prefrontals. The latter form a median suture and are smaller
than the nasals. The frontal is an elongate shield, as broad as the
supraocular, as long as its distance from the end of the snout, and

128 University of Kansas Publs., Mus. Nat. Hist.

slightly shorter than the parietals. The loreal is single or double, a
little longer than high, and is in contact with the first four or five
upper labials. The supraocular is a rectangular plate, slightly
broadest posteriorly. There is one high preocular and two subequal
postoculars. The one primary, two secondary, and three tertiary
temporals resemble the occipital scales. The first five or six upper
labials are not divided; the sixth and the seventh, which enter the or-
bit, are divided longitudinally into two plates. The upper labials,
from the seventh or eighth to the last are divided longitudinally into
three tiers. The upper margin of the mouth forms an angle at the
tenth labial. There are 14 to 16 lower labials, of which the second
to fifth or sixth are in contact with the anterior chin-shields and
with each other. Of the two pairs of chin-shields, the anterior pair
is petal-shaped and much larger than the posterior pair. Eight
gulars and two or three incomplete ventrals separate the chin-
shields from the first complete ventral. The wide ventrals are al-
most four times the width of the adjacent dorsal scales in the first
row. The edges of the ventrals are round. Two males and four
females have 139-146 (142.5) and 134-156 (147.0) ventrals; one male
and four females have 51 and 27-39 (34.7) subcaudals. The dorsal
scales are in 25 or 23 rows and reduce to 21 or 19 rows just before
the vent. The dorsal scale reductions in two females (BMNH 1946.
1.2.39 and MCZ 5165) are, respectively:

4+5(107) 6+7(124) 3+4(130)
25 23 21 19(155).

3+4(112) 6+7(125) 4+5(140)

6+7(7) 6+7(10) 5+6(119) 4+5(143)
29 27 25 23

5+6(8) 6+7(13) 5+6(133) 4+5(150)

4+5(151)
21 19(156).

The 14 to 16 maxillary teeth are followed by an interspace and
two enlarged, grooved teeth.

The dorsum is brown. A yellow crossbar, one scale in length, is
present in the occipital region. The body has numerous transverse
rows of yellow spots; six or seven anterior rows of spots are continu-
ous and form transverse bands not quite one scale in length and two
to four scale lengths apart. Each caudal scale has a yellow spot.
The venter and first three rows of dorsal scales are yellow. The
median edges of the subcaudals are brown. The one male has a
longer tail than the females; the tail /snout-vent ratio for one male
is 19.6, and for four females, 12.1-12.9 (12.7).

CoLUBRiD Snakes, Subfamily Homalopsinae

129

130 University of Kansas Publs., Mus. Nat. Hist.

Distribution. — This species is known from India, Malaya, Su-
matra and Borneo (Fig. 18). Specimens examined, eight, as follows.

No locality: BMNH 1940.2.2.8. India: no locality, BMNH 1946.1.2.37.
British Borneo: Sarawak — no definite locality, MCZ 5165. Indonesia: West
Bonit'o— Kuala Kapuas, NMB 12027; Sinkawang, BMNH 1946.1.2.39. Sumatra
— Djabura, NMB 1750. Malaya: J ohore—Gunong Pulai, BMNH 1938.9.7.2;
Kemaman, BMNH 98.11.29.9.

Enhydris doriae (Peters)

Homalophis doriae Peters, Monats. Akad. Wiss. Berlin, p. 577, 1871 [Syntypes.

— Two specimens in Museum, Geneva, Italy, from Sarawak, British Borneo;

collector Marquis G. Doria]; Ann. Mus. Genova, 3:38, pi. 5, fig. 2, 1872.
Homalophis doriae; var. Steindachner, Sitz. Akad. Wiss. Wein, 96(1); 71,

1887 [Holotype. — Museum, Wein, from Nangabadau, Borneo; collector

unknown].
Htjpsirhina doriae; Boulenger, Catalogue snakes British Museum, 3:13, 1896.

Rooij, Reptiles Indo-Australian Archipelago, 2:185, 1917.
Enhydris doriae; Haas, Treubia, 20(3):576, 1950.

Diagnosis. — Nasal completely divided; nasal cleft extending lat-
erally to first or second labial and medially to internasal; internasal
divided, in contact with loreal; seven to nine upper labials below
eye, separated by suboculars; two or three postoculars; anterior
chin-shields in contact with second to sixth lower labials; dorsal
scales in 31 or 33 rows at midbody; ventrals 137-149 (Fig. 19).

Description and variation. — The body is cylindrical, and the tail
is short and oval. The head is deep and distinct from neck. The
rostral is higher than broad and is visible from above. The nasals
are large and broadly in contact with each other behind the rostral.
The internasal is divided into two triangular plates (ANSP 2311 has
three plates). The prefrontals form a median suture and are larger
than the nasals. The frontal is an elongate, shield-shaped plate
about twice as broad as the supraocular, shorter than its distance
from the tip of the snout, and as long as the parietals or a little
shorter. The loreal is single (80'; o), or divided (20%), two times
longer than high, and in contact with the second through fifth or
sixth upper labials. There are nine or ten oculars, composed of
three (rarely two) supraoculars, one high preocular, two or three
suboculars, and two or three postoculars. The temporals resemble
the small occipital scales. There are 13-16 upper labials, of which
the first six or seven are narrow and high; the posterior labials are
divided in three tiers. There is a deep groove along the dorsal
edges of the upper labials. The upper margin of the mouth forms
an angle usually at the eleventh or twelfth upper labial. There are
16-18 lower labials, of which the first two pairs are in contact with
each other, and the second to fifth are in contact with the anterior

CoLUBRiD Snakes, Subfamily Homalopsinae

131

chin-shields; the anterior pair is subtriangular and is three to four
times larger than the posterior pair. The scales in the latter pair are
separated by small scales. Nine or ten gulars and one or two incom-
plete ventrals separate the chin-shields from the first complete ven-
tral. The umbilical scar is situated on or between 13-15 ventrals
anterior to the vent. The ventrals are nearly four times the width
of the adjacent dorsal scales in the first row. The edges of the
ventrals are round. Five males and seven females have 137-146
(140.8) and 139-149 (145.7) ventrals; three males and six females
have 51-56 (54.0) and 31-51 (43.8) subcaudals, respectively. The
dorsal scales are in 31 or 33 rows at midbody and reduce to 25 or 23
rows just before the vent. The dorsal scale reductions for one male
(MCZ 11277) and one female (FMNH 131727) are, respectively:

6+7(21) 6+7(86) 6+7(105) 5+6(130)

5+6(20)
8+9(11)

7+8(72)

7+8(95)

7+8(103)
6+7(113)

5+6(129)

5+6(134)

25(142).

25(143).

7+8(10) 7+8(96) 6+7(113) 4+5(129)

The 15 to 17 maxillary teeth are followed by an interspace and
t\\'0 enlarged, grooved teeth. The 21-22 mandibular teeth decrease
in size posteriorly.

The hemipenis extends to the level of the 12th subcaudal, and
the point of bifurcation is at the level of the 4th subcaudal. The
retractor penis magnus muscle originates on the caudal vertebra at
the level of the 28th or 29th subcaudal. The basal part of the hemi-
penis is naked, and the median part bears curved spines, which dis-
tally become papilla-like in form, but are as large as the spines. The
papillae are present on the distal end where they are smaller and
more numerous.

Fig. 19. Lateral view of the head of Enhijdris doiiae (MCZ 11277), Xl.2.

132 University of Kansas Publs., Mus. Nat. Hist.

The dorsum is olive brown; the venter and the first to five or six
rows of dorsal scales are bright yellow, which is either uniform or
spotted with black. The anterior upper and lower labials, and
anterior pair of chin-shields are black. The subcaudals are olive
brown with yellow posterior edges.

Males have more subcaudals and longer tails. The tail/snout-
vent ratios for three males and six females are 23.4-26.4 (24.7) and
13.5-18.9 (16.3) respectively.

Remarks. — Enhydrls doriae resembles E. punctata more than
any other species in the genus. In both species, none of the upper
labials enter the orbit and the first two pairs of lower labials form a
median suture. There are 12 or more upper labials and 14 or more
lower labials. These two species can be separated from each other
on the basis of the number of rows of dorsal scales, nature of the
nasals, and the loreal in relation to upper labials.

Distribution. — This species is known only from Borneo (Fig. 4).
Specimens examined, 14, as follows.

British Borxeo: North Borneo — no definite locality, SMF 19496; Baram
district, BMNH 97.3.4.10; Baram River, BMNH 1933.6.20.28; jandakan Bay,
RNHL no number. Sarawak— no definite locality, BMNH 93.3.6.55, FMNH
131724, 131727-29; Limburg, MCZ 11277; Upper Mujong River, Rejang,
BMNH 1929.12.2.1. Borneo: no localit>', MCZ 5240, UMMZ 65864. Indo.nesia:
West Borneo— Kapuas River, ANSP 26411.

The Enhydris pakistanica Group

Enhijdris pakistanica differs from other Enlujdris by having 29
rows of dorsal scales; more subcaudals; nasal cleft extending to
internasal; two internasals; no definite chin-shields; and very nar-
row ventrals.

This species is known only from West Pakistan.

Enhydris pakistanica Mertens

Enhydris pakistanica Mertens, Senck. biol., 40(3/4): 117, 1959 [Holotype.—
SMF 56340 from Jati, Sind, West Pakistan; M. G. Konieczny collector].

Diagnosis. — Nasal cleft extending from nostril to internasal;
internasal divided, in contact with or narrowly separated from
loreal; fourth, or fourth and fifth, upper labials entering orbit; two
postoculars; chin-shields absent; dorsal scales in 29 rows at mid-
body; ventrals 153-162.

Description and variation. — The body is cylindrical, and the tail
is ovoid. The head is slightly distinct from the neck. The rostral is
slightly broader than high and is visible from above. The larger
semicircular nasal and smaller semicircular internasal together
form a circular structure. The internasal is three times as broad as
long and is usually in contact with, or rarely narrowly separated

COLUBRID SXAKES, SUBFAMILY HOMALOPSINAE 133

hoiii, the loreal. The prefrontals are arched over the internasals
and are larger than the nasals. The frontal is slightly longer than
broad, slighth- broader than the supraocular, shorter than its dis-
tance from the end of the snout, and shorter than the parietals. The
loreal is a small plate, slightly longer than high, in contact with the
first three upper labials. The supraocular is broadest posteriorly.
There is one high preocular about twice as high as long. The two
postoculars are subequal, and the lower is partly under the eye.
The temporals are variable; there are usually one or two in the
primary row, three in the secondary row and four or five in the
tertiary row. Some specimens lack the tertiary row. There are
eight (77%) or nine (23%) upper labials. In specimens having eight
upper labials, the fourth enters the orbit, the sixth and seventh are
largest, and the eighth is the smallest; in specimens having nine,
the fourth and fifth enter the orbit, the seventh and eighth are
largest, and the ninth is the smallest. The fourth upper labial in the
series of eight is about the same size as the fourth and fifth together
in the series of nine. This suggests the fourth and fifth labials are
formed by a transverse division from a single plate. The lower ^

labials range from 10-14 (12.1). The first pair of lower labials is
narrow and long. Chin-shields are absent. The gulars are arranged
in three longitudinal rows on each side of the mental groove. There
are 10-14 (12.4) gulars and incomplete ventrals between the first
pair of lower labials and the first complete ventral; the incomplete
ventrals are scarcely distinguishable from the gulars. The narrow
\entrals are about twice the width of the adjacent dorsal scales in
the first row. The edges of the ventrals are round. The umbilical
scar is situated between ventrals 20 and 29 anterior from the vent.
Twenty-three males and 27 females have 155-162 (159.4) and 153-
161 (158.1) ventrals, respectively; 22 males and 22 females have
84-92 (88.6) and 70-77 (74.1) subcaudals, respectively. The coeffi-
cients of the differences are 0.322 and 3.551, respectively. The dif-
ference in subcaudals between males and females is significant.
The maximum number of dorsal scale rows in the anterior region is
33 or 31. At midbody there are 29 rows, just anterior to the vent
there are 23 or 21 rows. The dorsal scale reductions in one male
(AMNH 89296) and one female (AMNH 93157) are, respectively:

8+9(1) 7+8(30) 5+6(122) 6+7(143)

33 31 29 27 25

7+8(4) 6+7(39) 5+6(139) 4+5(147)

3+4(156)

23(159).

4+5(155).

134

University of Kansas Publs., Mus. Nat. Hist.

8+9(11) 7+8(23) 6+7(60)
31 29 27

6+7(110) 5+6(133)
25

8+9(2) 7+8(23)

4+5(149)

21(159).

6+7(64) 6+7(114) 5+6(135)

5+6(147)

The 13-14 maxillary teeth are followed by an interspace and a
pair of shghtly enlarged, grooved teeth. The 19-21 mandibular
teeth decrease in size posteriorly.

The hemipenis extends to the level of the 9th or 10th subcaudal,
and the point of bifurcation is at the level of the 5th subcaudal.
The retractor penis magnus muscle originates on a caudal vertebra
adjacent to the 29th or 30th subcaudal. The proximal part of the
hemipenis is naked. The distal half of the proximal stem of the
hemipenis is spinous, having five or six whorls of curved spines;
the proximal whorl has the largest spines. The branches of the
hemipenis are beset with minute blunt spines, equal in size and
density. The sulcus is forked.

A black or dark brown middorsal stripe originates on the frontal

BODY LENGTH (mm)

Fig. 20. Relationship behveen body length and tail length of males ( closed cn-

cles for adults; dots for juveniles) and females (open circles for adults; lines

for ju\eniles) of Euhijchis pakistanica.

CoLUBRiD Snakes, Subfamily Homalopsinae 135

between the eyes and continues posteriorly to the tip of the tail.
On the body this stripe is about five scales wide, whereas on the
tail it is about two scales wide. The stripe is bordered on either
side by a broad olive or brown lateral stripe, which is only about
one scale wide on the tail. Ventral to the olive or brown lateral
stripe is a black or dark brown stripe, which is about three scales
\\ide on the body and two scales wide on the tail. This lateral
black stripe extends anteriorly along the upper edges of the upper
labials to the rostral. Ventrally, there are two yellow stripes, one
occupies the first and part of the second row and the other part of
the fifth and the sixth rows of scales. The area between the two
\'ellow stripes is pale brown. The venter is yellow with a broad
midventral black or dark brown line extending the length of the
body. In the anterior region, where the ventrals are narrow, the
midventral black or dark brown line completely covers the ventrals.
The coloration in young is the same as in adults, except that the
stripes are more distinct.

Males have more subcaudals and fewer ventrals than females.
The tail/ snout-vent ratios for 21 males and 19 females are 38.6-44.2
(40.0) and 29.0-34.2 (32.7), respectively. The coefficient of difference
between the means is 4.24 and is significant (Fig. 20).

The tail/ snout-vent ratios for four adult males and 17 young
males are 39.5-44.2 (41.6) and 38.5-41.1 (39.7); the coefficient of
difference between the means is 0.864 and is not significant.

Distribution. — This species is known only from West Pakistan
(Fig. 4). Specimens examined, 61, as follows.

West Pakistan: no definite locality (shipped from Karachi), AMNH
93154-81 + 27 untagged specimens. Tatta District — near Jati, AMNH 89296,
MNHN C3458.19, SMF 56340-1, 56567, UMMZ 123843.

Genus Homalopsis Kuhl and Hasselt

Homalopsis Kuhl and Hasselt, Alg. Konst. Lett. Bode, 1(7):101, 1822; Isis,
p. 474, 1822 [Type species. — Coluber horridiis Daudin, 1803]. Schlegel,
(in part), Essai sur la physionomie des seipents, 2:332, 1837. Gray,
Zoological miscellany, p. 64, 1842; Catalogue snakes British Museum, p.
66, 1849. Dumeril and Bibron, Erpetologie generale . . . reptiles. Paris,
7:967, 1864. Gunther, Reptiles of British India, p. 285, 1864. Jan, (in
part). Arch. Zool. Anat. Phys., 3:256, 1865. Boulenger, Fauna of British
India . . . Reptilia and Batrachia, p. 373, 1890; Catalogue snakes British
Museum, 3:13, 1896.

Pijthonia Blyth, Jour. Asiatic Soc. Bengal, 28:297, 1859 [Type species. —
Pythonia scmizonata Blyth, 1859].

Diagnosis. — Head distinct from neck, head having large shields;
nasals in contact behind rostral; internasal single or divided behind
nasals; one or more loreals present; dorsal scales small, distinctly
striated and keeled, in 39 to 43 rows at midbody; ventrals broad,

136 University of Kansas Publs., Mus. Nat. Hist.

not keeled; maxillary teeth 11 to 13, followed by an interspace and
a pair of slightly enlarged, grooved teeth.

Homalopsis buccata (Linnaeus)

Coluber huccatus Linnaeus, Museum S.R.M. Adolphi Friderici Regis, p. 29,
pi. 19, fig. 3, 1754; Systema naturae, ed. 10, p. 217, 1758 [Holotype.— Lost
( Andersson, 1899), from "India"].

Coluber monilis Linnaeus, Systema naturae, ed. 10, p. 221, 1758 [Type
locality. — "America"]. Daudin, (in part), Histoire naturelle des reptiles,
7:59, 1803. Andersson, Bih. Svens. vet. Akad. Stockholm, 24(4) :34, 1899.

Coluber stibalbidus Gmelin, Systema naturae, p. 1103, 1788 [Based on Seba,
vol. 2, pi. 21, fig. 3].

Vipera buccata; Daudin, Histoire naturelle des reptiles, 6:220, 1803.

Coluber horridus Daudin, Histoire naturelle des reptiles, 7:71, 1803 [Based on
Seba, Vol. 2, pi. 21, fig. 3].

Homalopsis molurus Boie, Isis, p. 213, 1826 [Based on illustration in Russell's
Indian seipents, vol. 2, pi. 40].

Homalopsis monilis; Boie, Isis, p. 521, 1827.

Homalopsis buccata; Schlegel, Essai sur la phvsionomie des serpents, vol. 2,
pi. 13, figs. 1-5, 1837. Cantor, Catalogue Ma'ay Reptiles, p.^ 96, 1847. Gray,
Catalogue snakes British Museum, p. 67, 1849. Diuneril and Bibron,
Erpetologie generale . . . Reptiles. Paris, 7:968, 1854. Giinther, Reptiles
of British India, p. 285, 1864. Morice, Sur la Faune de la Cochinchine
Francai.se. Lvon, p. 58, 1875. Tirant, Notes sin* les Reptiles de Cochinchine
et du Cambodge, 3(20) :402, 1885. Boulenger, Fauna of British India . . .
Reptilia and Batrachia, p. 374, fig. 109, 1890. Sclater, List of snakes in
Indian Museum, p. 53, 1891. Boulenger, Ann. Mus. Cenova, 13:327, 1893.
Miiller, Verb. Ges. Basel, 10:205, 1895. Flower, Proc. Zool. Soc. London,
p. 887, 1896. Werner, Verb. Ges. Wien, 46:19, 1896. Boulenger, Catalogue
snakes British Museum, 3:14, 1896. Boettger, Katolog der Reptilien . . .
Schlangen, 2:88, 1898. Flower, Proc. Zool. Soc. London, p. 677, 1899.
Werner, Zool. Tahrb. Svst. 13:490, 1900. Laidlaw, Proc. Zool. Soc. London,
3:578, 1901. Wall and Evans, Jour. Bombay Nat. Hist. Soc, 13:349, 616,
1901. Lampe, Tahrb. Nassau Veriens, 555:31, 1902. Wall, Proc. Zool. Soc.
London, p. 94", 1903; Jour. Bombay Nat. Hist. Soc, 17(2) :388, 1905.
Boulenger, Fasciculi Malayenses. Zoology, 1:175, 1903. Werner, Abb.
Baver Akad. Wiss., 12(2): 366, 376, 1903. Rosen, Ann. Mag. Nat. Hist., ser.
7, 15:175, 1905. Mocquard, Revue Coloniale, p. 51, 1907. Boulenger, A
xertebrate faima of the Malay Peninsula . . . Reptilia and Batrachia, p. 162,
1912. Despax, Bull. Mus. Paris, 18:199, 1912. Barbour, Mem. Comp. Zool.,
44(1):123, 1912. Smith, Jour. Nat. Hist. Soc. Siam, 1(1):101, 1914;
Tour. Nat. Hist. Soc. Siam, 1(3):213, 1914; Tour. Nat. Hist. Soc. Siam,
2:162, 1916. Gyldenstolpe, Svensk. vet. Akad. Handl., 55(3): 19, 1916.
Rooi), Reptiles Indo-Australian Archipe'ago, 2:186, 1917. Fea, Ann. Mus.
Genova, ser. 2, 17:475, 1917. Robinson and Kloss, Jour. Federated Malav
States Mu.seum, 7(2):303, 1920. Mell, Arch. Naturg., 88A:123, 1923.
Phisali.x, Animaux venimeux et venins, 2:285, 296, 1922. Werner, Arch.
Naturg., 89A(8):162, 1923. Wall, How to identify the snakes of India,
p. 38, 1923. Sworder, Singapore Nat., 2:66, 1923. Wall, Tour. Bombav
Nat. Hist. Soc, 29:867, 1924; Jour. Bonilxiy Nat. Hist. Soc, 30:817, 1925.
Dammerman, Treubia, 8:323, 1926. Dunn, Amer. Mus. Novit., 144:4,
1927. Bourret, Invent. Gen. Indochine, 3:241, 1927. Mertens, Zool. Anz.,
86:66, 1929. Brongersma, Treubia, p. 67, 1927. Cochran, Proc. U.S. Natl.
Mus., 77(2):31, 1930. Kopstein, Weltevreden, 138:136, 1930; Treubia,
12(3/4):274, 1930. Smith, Bull. Raffles Museum, 3:61, 1930. Bourret, Bull.
Instr. Pub., p. 22, 1934; Bull. Instr. Pul)., p. 80, 1934. Mertens, Archi\-.
fur Hydrologie, 12(4) :695, 1934. Bourret, Serpents de ITndochine, p. 293,
1936.

CoLUBRiD Snakes, Subfamily Homalopsinae 137

Homalopsis liardwickii Gray, Zoological miscellany, p. 65, 1842 [Holotype. —
BMNH 1946.1.7.26 from "India"; collector unknown].

Homalopsis scinizonata Blyth, Jour. Asiatic Soc. Bengal, 24:187, 1855 [Holo-
type.— Calcutta Museum from Martalnui; collector unknown].

Ptjtlioma semizonata; Blyth, Jour. Asiatic Soc. Bengal, 28:297, 1859.

Diagnosis. — Characters of genus (Tables 10-16).

Description and variation. — The body is stout and cyhndrical.
The snout is broadly rounded anteriorly. The rostral is broader
than high and is evident from above. A pair of large nasals lie
completely on the dorsal side of the head; they are partly divided
and broadly in contact mesially. The nostril is connected by a
suture to the first labial. The internasal is usually single but is
divided by a longitudinal suture in some specimens. The azygous
shield is usually present in Burmese specimens; the specimens from
the other parts of the range do not have the azygous shield. The
frontal is usually fragmented into two or more scales; the anterior
half of the frontal is entire and is as broad as, or slightly broader,
and longer than the supraocular, shorter than its distance from the
tip of the snout, and considerably longer than the parietals. The
latter shields are short, about as broad as long, and usually entire.
The loreal is an elongate shield, usually single, twice as long as
wide, in contact with the first four (rarely three) upper labials, and
narrowly separated from the internasal. In specimens from Burma
the loreal is divided into two or more scales. The eye is small and
completely surrounded by one or two preoculars, two to four nar-
row suboculars, and one or two postoculars. The temporals are
scale-like and are arranged as one or two scales in the primary row
and two to four in the secondary row. The scales in the tertiary
row are not distinguishable from those of the fragmented posterior
upper labials. Of the ten to fourteen upper labials, the first six or
seven are vertically elongate; the posterior several labials are
longitudinally divided into three tiers. The fifth or sixth upper
labial is situated below the eye. The upper margin of the mouth
forms an angle at the tenth and/or eleventh upper labials. Of the
15 to 19 lower labials, the first three (rarely two or four) are in con-
tact with the anterior chin-shield. There are three or four pairs of
chin-shields, which are progressively smaller posteriorly. The an-
terior pair have a median suture, and the scutes in posterior pairs
are separated medially by small scales. The broad ventrals are
about four times the width of the adjacent dorsal scales in the first
row. There are nine to 13 gulars and two to four incomplete ven-
trals between the chin-shields and the first complete ventral. The
umbilical scar is situated between 20 to 25 ventrals anterior to the

138 UxR'ERSiTY OF Kansas Publs., Mus. Nat. Hist.

vent. The variation in numbers of ventrals and subcaudals and in
the tail /snout-vent ratio is given in tables 13, 14, and 15. The dorsal
scales are small, striated, keeled, and in 39 to 45 rows anteriorly,
37-41 rows at midbod\', and 27 to 31 rows just anterior to the vent
(Table 12).

The dorsal scale reductions for four females (KU 92382, 92383,
92385 and UMMZ 55330) and for t^^ o males (KU 92384 and UMMZ
55325) are, respectively,

+ 10(58) 10+11(63) 10+11(71) 8+9(96)

43 45 43 41 39

+13(52) 10+11(60) 10+11(64) 8+9(85)

6+7(109) 7+8(115) 7+8(119) 6+7(128)

37 35 33 31

10+11(101) 9+10(117) 8+9(121) 8+9(125)

5+6(133)

29(159).

5+6(141)

+8(74) +7(78) 8+9(81) 8+9(95)

43 42 43 41 39

9+10(69) 8+9(81) 8+9(81) 9+10(99)

5+6(108) 7+8(120) 3+4(136) 6+7(146)

35 33 31(160).

5+6(119) 8+9(121) 4+5(134) 6+7(144)

9+10(68) 9+10(84) 9+10(110) 5+6(116)

41 39 37 35 33

8+9(70) 6+7(94) 7+8(104) 8+9(112)

6+7(136)

31(159).

7+8(136)

7+8(73) 8+9(90) 5+6(103) 6+7(118)
39 37 35 33 31

7+8(83) 7+8(93) 5+6(106) 6+7(116)

4+5(123) 5+6(143) 4+5(152)

29 27 25(157).

6+7(128) 5+6(141) 4+5(152)

9+10(84) 5+6(96) 8+9(103) 5+6(114)

41 39 37 35 33

6+7(93) 5+6(99) 9+10(106) 5+6(116)

5+6(122) 4+5(135) 5+6(144)

31 29 27(159).

6+7(132) 5+6(136) 5+6(141)

COLUBRID SXAKES, SuBFAMILY HOMALOPSINAE 139

7+8(78) 7+8(94) 5+6(106) 6+7(113)

39 37 35 33 31

7+8(83) 7+8(98) 5+6(108) 6+7(116)

5+6(128) 6+7(141) 4+5(155)

29 27 25(161).

5+6(129) 5+6(142) 4+5(153)

The 11-14 maxillary teeth are followed by a pair of slightly
enlarged, grooved teeth. The 18 to 20 mandibular teeth decrease in
size posteriorly.

The hemipenis extends to the level of the 12th to 14th subcaudal,
and the point of bifurcation is at the level of the 5th or 6th sub-
caudal. The retractor penis magnus muscle originates on a caudal
vertebra at the level of the 32nd to 35th subcaudal.

The head is pale brown or pale gray with a V-shaped dark
brown mark on the snout, and a A-shaped mark on the top of the
head, which laterally joins a postorbital stripe. The latter actually
originates anterior to the eye, passes through the eye, and continues
posteriorly to join the first black crossbar on the neck, or termi-
nates just posterior to the angle of the mouth. The first black
crossbar bears a narrow, anterior projection along the vertebral
line. The dorsum and sides of the body are black, gray, or pale
brown. There are 19 to 51 narrow, black-edged, yellow crossbars
(Table 16) on the body. In Burmese specimens the crossbars are
three to four scales in length, straight edged, and usually complete.
In other specimens the crossbars are one to two scales in length,
wavy edged, and complete or broken into two or three parts. From
12 to 16 similar yellow crossbars, one or two scales in length, are on
the tail. The venter is white or yellow with a series of small black
spots on the lateral edges of the ventrals. Snakes from Thailand
have 17-43 (35.3) spots; other specimens have 35-62 (53.5) spots.
The white belly gradually changes laterally to a rich lemon-yellow
on the first three rows of dorsal scales, where the black dorsal
markings abruptly commence. The ventral surface of the tail is
mottled with yellow and dark brown.

The young are marked like adults, but the crossbars are darker,
the interspaces are white or yellow-ochre, and the entire ventral
surface is white or lemon-yellow.

The adults have spinous keels on the first three or four rows of
dorsal scales to a point 14 to 18 ventrals anterior to the vent. The
upper and lower surfaces of the head are tuberculate. The tail/
snout-vent ratios of adults and young of both sexes are given in
Table 17. Adult males have longer tails than adult females, and the

140 University of Kansas Publs., Mus. Nat. Hist.

coefficient of difference is highly significant. The greater tail length
of the adult male is presumably an adaptation to accommodate the
hemipenis. This difference between the sexes is graphically illus-
trated in Fig. 21, which shows that the difference is greatest in
larger snakes. The variation in numbers of ventrals and subcaudals
is summarized in Tables 13-14. The adult females possess slightly
more ventrals than the adult males. This is due to a slightly longer
body, which gives additional space for the development of young.
Young and adult males have significantly more subcaudals than do
females (Table 17). The higher number of subcaudals is correlated
with the longer tail of the males. Comparison of the numbers of
subcaudals between adults and young of the same sex and from the
same locality shows no significant difference. Males have more
crossbars on the body and on the tail than do the females, but the
difference is not significant (Table 16). No significant difference
was noted in the number of crossbars of young and adults of the
same sex from the same locality (Fig. 22).

Remarks. — There is no specimen of Coluber huccatns in the
Royal Museum at Stockholm (Andersson, 1899). According to An-
dersson (1899:34) there are four specimens in the Royal Museum
identified as Coluber monolis; one of them agrees completely with
Linnaeus' description, and this probably is Linnaeus' type of Colu-
ber buccatus.

Distribution. — This species is common in Burma and Indo-
China, where it is found in rivers, canals, and ponds. The snake also
is known from India, Thailand, Cambodia, Cochin China, China,

TABLE 10. — Variation in the Number of Upper Labials in Homalopsis buccata

'^C-ltN CO CO ^ ^ ^. '-'

I— i^rH rHi— ii— ii— i]i;«

Locality :^ :^ ^ ^ '^ ^ ± B ^

Burma 2 2 13 17 24 10

Thailand 1 -

Malaya - -

Sumatra 1 —

Java - 1

TABLE 11. — Variation in the Number of Upper Labials below Eye in

Homalopsis buccata

Locality 5-5 5-6 6-6 6-7 7-7 Total

Burma 1 5 69 3 1 79

Thailand - - 19 - - 19

Malaya — _ 1 - 5 _ _ 6

Sumatra - 1 7 _ _ 8

Java 4 1 1 _ _ 6

—

CoLUBRiD Snakes, SuBFAAriLV Homalopsinae

141

5
a:

"IS
o

C/2

o
Q

o
o

i-H

c
c

•5b

(25

CO Ot-- CD lO

b- C<] -H Tfl

CqcM

CO I CM

- I CO

lO O CD t^ lO

CO CO

oi I I in -H

Tf cq oq

<M I C\I

t- s- •„ ._ .

o _c ^2 c S

e c c c "^
<<<<<

'^ ^

J I K H^co .

>. X >. >> >,

O C C O c
^ ^ ^ ^ ^

CO (M

-H ■* C-l

^ ^ ^ C-l

O I -t ^ ^

i-HCO

o o o c o

'C 'n 'c 'c 'c

0) OJ 4i OJ 0)

*J .w -w +j ^

c/^ v: v^ y; c/3

O C C O O

Ph fin pLi Oh p^

'5 «

S~ rt 5 -

— r- Cw " >

m H S c^ hS,

142 University of Kansas Publs., Mus. Nat. Hist.

TABLE 13. — Variation in the Number of Ventrals in Homalopsis huccata from

Five Localities

Locality Sex Number Range Mean S.D. C.V. CD.

Burma $ 46 155-163 158.8 2.10 1.32

0.01

9 30 154-170 158.9 3.72 2.34

Thailand $ 7 169-174 171.7 2.57 1.50

0.06

$ 13 161-171 166.4 2.47 1.48

Malaya S 3 161-167

5 4 155-163

Java S 6 156-165 161.8 3.65 2.26

Sumatra $ 7 157-164 160.1 2.67 1.67

TABLE 14. — Variation in the Number of Subcaudals in Hotnalopsis huccata

from Five Localities

Locality Sex Number Range Mean S.D. C.V. CD.

Burma $ 42 77-89 81.3 2.37 2.91

2.51

9 28 68-75 71.1 1.88 2.64

Thailand $ 6 95-99 97.0 1.55 1.60

1.41

9 12 81-97 88.2 4.69 5.32

Malaya $ 3 84-101

9 3 73-84

Java $ 5 72-84 81.0 5.10 6.30

Sumatra 9 6 76-81 78.2 1.94 2.48

TABLE 15. — Variation in Tail/Snout-vent Ratio in Hotnalopsis huccata from

Four Localities

Locality Sex Number Range Mean S.D. C.V. CD.

Burma S 37 29.3-34.0 31.6 1.23 3.90

2.45
9 24 23.7-28.7 25.9 1.10 4.25

Thailand S 5 30.7-33.3 .32.4 1.08 3.33

0.19
9 11 28.0-.35.1 31.8 2.35 7.39

Java S 5 26.5-33.5 30.7 2.79 9.08

Sumatra 9 6 27.3-31.6 29.6 2.32 7.85

CoLUBRiD Snakes, Subfamily Homalopsinae

143

TABLE 16. — \ aviation in the Number of Dorsal Transverse Bars on the Body
in Homalopsis buccata from Four Localities

Locality Sex Number Range Mean S.D. C.V. CD.

Burma $ 26 19-24 21.8 1.29 5.90

0.01
9 29 19-25 21.8 2.22 10.17

Thailand $ 5 26-44 36.6 6.62 18.09

0.56
2 11 32-51 43.7 6.15 14.06

lava S 5 22-26 23.4 1.67

Sumatra 9 7 23-29 26.0 2.08

TABLE 17. — Comparison of Adults and Young of Both Sexes in
Homalopsis buccata from Burma

Character

Sex Age No. Range Mean S.D. C.V. CD.

Ventrals $ Adult 46 155-163 158.8 2.10 1.32

Young 15
9 Adult 30

152-160 156.1
154-170 158.9

2.36 1.51
3.72 2.34

Young 15 153-161 156.0 2.36 1.51

0.61

0.48

Subcaudals

$ Adult 42 77-89 81.3 2.37 2.91

Young 15
9 Adult 28

81-86 83.6

68-75 71.1

1.80 2.15
1.88 2.64

Young 15 72-76 73.9 1.16 1.57

Young 15
9 Adult 24

31.8-35.4 33.5
23.7-28.7 25.9

2.70 8.06
1.10 4.25

Young 15 29.1-32.8 30.8 1.15 3.73

Young 15
9 Adult 29

20-25 23.1

19-25 21.8

1.60 6.91
2.22 10.17

Young 15 20-25 21.9 1.41 6.45

Young 15
9 Adult 29

12-16 14.3

10-13 11.7

1.29 9.00
1.02 8.72

Young 15 11-14 12.2 1.08 8.85

-0.54*

-0.91

Tail/snout- vent -- $ Adult 37 29.3-34.0 31.6 1.23 3.90

-0.49

-2.20

Body Bars $ Adult 26 19-24 21.8 1.29 5.90

-0.44

-0.01

Tail Bars $ Adult 24 11-16 14.3 1.22 8.56

1.14

0.94

Negative CD. means the young have higher mean value than the adults.

144

University of Kansas Publs., Mus. Nat. Hist.

260

240

220

200

180

X I 60

_i 140

I 20

I 00

:»

oo
o

•• •••

o o

•.V

o " o

o
o o o

• • o o
•• o

• o
„ o

,••

140 220 300 380 460 540 620

BODY LENGTH (mm)

700

780

860

Fig. 21. Relationship between body length and tail length of males (closed
circles ) and females ( open circles ) of Homalopsis buccata.

Fig. 22. Geographical variation in the color pattern of Homalopsis buccata,

represented by two jmeniles. Left: KKG 252 from Twante, near Rangoon,

Irrawaddy Division, Burma; right: SU 8538 from Singapore, Malaya.

CoLUBRH) Snakes, Subfamily Homalopsinae

145

,1=4

V-i

tn
C

to
O

_ o

HI 0)
to s-t

^ <U

0-1
•**

o
y

146 University of Kansas Publs., Mus. Nat. Hist.

Malaya, Java, Sumatra (Fig. 23). Specimens examined, 173, as
follows.

Burma: Irawaddij Division— Mauhin, KKG 65, 67, 189, KU 92347-61.
Pegu Division— Rangoon, SU 12392; Twante near Rangoon, KKG 252-262,
KU 92362-92394, 92396, 92398-463. China: no definite locality, UMMZ
61289. Cochin China: no definite locality', MCZ 5961. India: no definite
locality, BMNH 1946.1.7.26. Indonesia: 'yaua— Buitenzorg, MCZ 7625(8).
Sumatra — Kampong Silau Mardja Asaham, UMMZ 66326-31; Little Siak
River, USNM 37813. Malaya: Johore, 45 mi. north Kota Tinggi, SU 13064-5;
Kuala Lumpur, USNM 142401; Malacca, UMMZ 65842; Perak, Tapah Fisher-
ies Station, SU 13066; Singapore, CAS 16769, MCZ 5219, SU 8537-8. Thai-
land: no definite locaUtv, ANSP 5116-7, USNM 56451. South Thailand-
Bangkok, AMNH 92296-301, CAS 85302, UMMZ 65369(4), 65370(2),
65371, 65372(2), 96276, MCZ 3093; Trang, AMNH 36274.

Additional records. — Indonesia: Banga (De Rooij, 1917:187). Borneo —
Bulangan River; Buntal; Kuching; Montrado; Rejang River; Sebruan Valley;
Singkawang (De Rooij, 1917:187). Java — Depok; Preanger (De Rooij,
1917:187). Sumatra — Indragiri; Labuan; Langkat; Laut Tador; Ringgat;
Stabat (De Rooij, 1917:187).

Genus Cerberus Cuvier

Hijdrus Schneider (in part), Historiae Amphibiorum, 1:246, 1799 [Based on
Russell, Indian serpents, 1:23, pi. 17, 1796. Type species. — Hijdrus rijn-
chops from Ganjam, India].

Elaps Schneider, (in part), Historiae Amphibiorum, 2:301, 1801 [Type species.
— Elaps hoaeformis Schneider, 1801].

Hurria Daudin, Mag. Encyclop. An. 8, 5:434, 1802 [Type species. — Hurria
hilineata Daudin, 1802]. Bull. Soc. Pliilom. Paris, 3(72): 187, 1803;
Histoire naturelle des reptiles, 5:281, 1803. Stejneger, Bull. U.S. Natl.
Mus., 58:304, 1907. Wall, Jour. Bombay Nat. Hist. Soc, 29:867, 1924.

Coluber (non Linne, 1766) Daudin, Histoire naturelle des reptiles, 7:167,
1803.

Strephon Goldfusz, Handb. Zool., 2:151, 1820 [Type species. — Hurria hilineata
Daudin, 1802].

Python (non Daudin, 1803), Merrem, Tentamen Systematicis Amphibiorum,
p. 89, 1820.

Cerberus Cuvier, Regne Animal ed. 2, 2:81, 1829 [Type species. — Coluber
cerberus^^Htjdrus njnchops Schneider]. Gray, Zoolological miscellany, p. 64,
1842; Catalogue snakes British Museum, p. 63, 1849. Dumeril and Bibron,
firpetologie generale . . . reptiles, 7:977, 1854. Giinther, Reptiles of
British India, p. 278, 1864. Smith, Bull. Raffles Museum, 3:61, 1930.
Boulenger, Fauna of British India . . . Reptilia and Batrachia, p. 374, 1890;
Catalogue snakes British Museum, 3:15, 1896. Wall, Jour. Bombay Nat.
Hist. Soc, 26:89, 1900.

Homalopsis Schlegel, (in part), Essai sur la physionomie des serpents, 2:332,
1837. Jan (in part), Arch. Zool. Anat. Phys., 3:256, 1865.

Historical summary. — The snakes of this genus were first made
known by Russell (1796) who illustrated a specimen from Ganjam,
India; in 1801 he again illustrated a specimen from "India."
Schneider (1799:246) proposed the generic name Hydrus with Hy-
drus rynchops as the type species. Schneider (1801:301) named the

CoLUBRiD Snakes, Subfamily Homalopsinae 147

genus Elaps with Elaps boaeformis as the type species. Daudin
(1802:434) proposed the generic name Hurria for an illustration in
Russell (1796: pi. 40) of a snake from Hyderabad. The specimen
shown in this color plate has two white dorsolateral lines extending
the length of the body and has a small, depressed head. The snake
does not resemble members of the genus Cerberus. Cuvier (1829:
81) proposed the generic name Cerberus with Coluber cerberus
(^Hijdrus njnchops) as the type species. Stejneger (1907:304)
revived the name Hurria and stated: "The name Cerberus, by
which this genus is generally known, must give way to Hurria of
Daudin, 1803, the exclusixe type of which is Russell's, Indian Ser-
pents, pi. xl, which Doctor Boulenger identifies as H. njnchops."
Malcolm Smith (1930:61) pointed out that Stejneger was confused
with respect to the names and stated: "Stejneger has revived Dau-
din's name Hurria for this genus on the grounds that 'Doctor
Boulenger identifies Russell's plate XL H. njnchops'. This is not so.
The plate XL to which Boulenger refers (Cat. Sn. Brit. Mus., 1896,
vol. Ill, p. 16) is in vol. II of Russell, whereas plate XL to which
Daudin refers is in vol. I of Russell. Daudin refers to this snake
again in his Hist. Nat. Rept., 1803, vol. V, p. 284 under the name of
Hurria hilineata. The 'Hurriali of Russell, which was made by Dau-
din the exclusive type of his genus Hurria (Mag. Encyclop. An. 8 V,
1803, p. 434), has never, as far as I am aware, been identified by any-
one, and it certainly is not Schneider's njnchops. It was described
and figured by Russell from a sketch of a head, neck and tail and a
description, sent him by a correspondent. With its unusual sub-
caudal scutellation — the anterior plates being single and the poste-
rior double — it should not be difficult to recognize if met with again.
The name Hurria therefore based exclusively upon this specimen,
cannot stand for the genus now under discussion. Strephon Gold-
fusz 1820, based upon the same specimen, stands on the same posi-
tion. Cerberus Cuvier (Regne Animal, 2nd edit., 1829, vol. 2, p. 81)
type Cohiljer cerberus is correct."

I have seen the three color plates mentioned above, one of which
Boulenger identified as H. njnchops at the British Museum ( Natural
History); this is plate XL in volume 2, of Indian Serpents 1801.

Diagnosis. — Head barely distinct from neck, head with small
shields; nasals in contact behind rostral; internasal single or divided;
loreal present; body cylindrical, stout; scales keeled, in 21 to 29
rows; ventrals well developed, not keeled; maxillary teeth 12 to 19,
followed by an interspace and a pair of enlarged grooved teeth.

Description and variation. — The small eyes are situated close

148 University of Kansas Publs., Mus. Nat. Hist.

together on top of the head and have vertically elliptical pupils.
The rostral is as broad as, or broader than, high. The nasals are in
contact. The nostril is directed upwards, crescent-shaped, and con-
vex anteriorly. The nasal cleft usually extends outwards to the first
labial in most populations, except in C. njnchops novaeguineae, in
which the nasal cleft extends to the second labial in about 20 per
cent of the specimens (Loveridge, 1948:3(89). The internasals are
paired and situated posterior to the nasals. The prefrontals are in
contact. The frontal is a small shield which is partly or completely
fragmented. The parietals are fragmented into small scales. The
loreal is usually single; only four per cent of the total specimens of
C njnchops have a divided loreal. The loreal is in contact with the
first to fourth upper labials, except in Burma where in 41 per cent of
the males and in 74 per cent of the females the loreal is in contact
with the first to third. In C. njnchops novaeguineae the loreal is in
contact with the second to fourth labials and in C australis, the
second and third upper labials. The supraocular is relatively large.
There is a single (rarely two) preocular, and single (rarely two)
(Tables 18-31) postocular. The two (rarely one or three) subocu-
lars separate the eye from the labials. The nine to twelve upper
labials (usually nine to ten) are higher than long. The fifth or the
sixth upper labial is situated below the eye; the seventh or eighth is
the largest and is usually in contact with the postoculars. Specimens
having the sixth upper labial below the eye have ten upper labials,
whereas specimens ha\'ing the fifth upper labial below the eye have
nine upper labials, except in C. njnchops novaeguineae, in which
reduction in the number of upper labials takes place both anterior
to and posterior to the eye. The reduction in the number of labials
seem to be directly correlated with the shortening of the head. The
anterior labials are narrow and long and gradually become larger
posteriorly to the seventh labial, which is usually the largest, except
in C. australis in which the eighth is largest. All labials posterior to
the largest one are broader than long. The three pairs of chin-
shields are subequal in length; the last two pairs are separated by
small scales. The body is cylindrical or slightly compressed. Except
for the scales in the first row, the dorsal scales are strongly keeled,
and longitudinally striated. The number of rows varies from 21 to
29 at midbody. The dorsal scales of the first row are one and one-
half times larger than the scales in the other rows. The ventrals are
broad. The tail is moderately short, 23-26 per cent of the body
length. There are 12 to 19 maxillary teeth plus two enlarged,
grooved teeth. The mandibular teeth decrease in length poste-

CoLUBRiD Snakes, Subfamily Homalopsinae 149

TABLE 18. — Variation in the Number of Upper Labials in the Species and

Subspecies of Cerberus

Taxon 8^9 9^9 9^10 10-10 10-11 11-11 Total

C.r. ri/nchops — — — — —

India — 9 5 11 — 1 26

Bumia _ 20 5 11 2 2 40

Thailand — — 3 11 — 1 15

Malaya _ _ _ 11 5 1 17

Java — — — 1 — — 1

Borneo _ _ _ 2 — 2 4

Philippines 1 4 2 27 2 3 39

Palau Islands _ _ _ 3 1 — 4

C.r. novae gtiineae — 2 2 — — — 4

C. austraUs 1 — 2 — — — 3

C microlepis — — — 3 1 — 4

TABLE 19. — Variation in the Number of Upper Labials below the Eye in the

Species and Subspecies of Cerberus

Taxon

5-5

6-5

5-6

5+6-

5+6

6-6

Others Total

C.r. rtjnchops

India

Burma —

24
30

2
2

2
2
1

3
5

2
1

36
39

Thailand

Malaya

1 18

Java

Borneo

Philippines

— 41

Palau Islands — .

C.r. novaeguineae ..
C australis —.

5
1 3

C. microlepis

TABLE 20. — Variation in the Number of Upper Labials in Contact with the

Loreal in the Species and Subspecies of Cerberus

Taxon 1-3 1, 3, 4 1-4 2-3 2-4 Others Total

C.r. rht/nchops

India 16 5 25 4 6 2 58

Burma 51 1 30 — 2 — 84

Thailand 2 — 10 — 12 — 24

Malaya _ _ 34 — 2 — 36

Java _____ 2 2

Borneo — — 8 — — — 8

Phihppines — 10 59 — 7 2 78

Palau Islands 1 — 5 — 2 — 8

C.r. novaeguineae — — — 3 5 — 8

C. australis 1 — — 5 — — 6

C microlepis 3 — 1 — — — 4

150

University of Kansas Publs., Mus, Nat. Hist.

TABLE 21. — Variation in the Number of Lower Labials in the Species and

Subspecies of Cerberus

Taxon 12-12 12-13 13-13 13-14 14-14 Others Total

C.r. rynchops

India 3 2 8 3 7 2 25

Burma 2 5 15 5 9 2 38

Thailand — 1 4 1 6 — 12

Malaya — 1 5 5 5 — 16

Java _ _ _ 1 _ _ 1

Borneo — 1 — 12 — 4

Philippines 4 2 13 8 9 2 38

Palau Islands — — 12 1 — 4

C.r. novaeguineae — 1 3 ^ — ■ — — 4

C. australis — — 11 — 13

C microlepis 1 — 1 — 114

TABLE 22. — Variation in the Number of Preocular and Postocular Scales in the

Species and Subspecies of Cerberus

Preoculars

Postoci

liars

Taxon

1-1

1-2

2-2

1-1

1-2

2-2

Total

C.r. rynchops
India _

31
9

18
1
2

36
5
4
2
1

1
1

2
2

1
1

3
4

3
11

Burma

Thailand

Malaya

lava - -

Borneo _._. .

Philippines

Palau Islands

C.r. novaeguineae

C australis

C microlepis

36
5
4
3
2

TABLE 23. — Variation in the Number of Subocular Scales in the Species and

Subspecies of Cerberus

Taxon 0-0 1-1 1-2 2-2 2-3 3-3 Total

C.r. rynchops

India 2 7 1 8 1 4 23

Burma — — — 30 1 — 31

Thailand 11 1 2 1 — — 15

Malaya 1 2 — 13 3 — 19

Java — — — 1 — — 1

Borneo — — — 3 — — 3

Philippines 2 2 4 19 7 2 36

Palau Islands — — — 4 1 — 5

C.r. novaeguineae — — — 2 — 2 4

C. australis — 2 1 — — — 3

C microlepis — — — 2 — — 2

CoLUBRiD Snakes, Subfamily Homalopsinae

151

TABLE 24. — X'ariation in die Number of Rows of Dorsal Scales in the Species

and Subspecies of Cerberus

Midbody

Posterior

Taxon

Total

C.r. n/nchops

India

2
1

15
5
9

29
6

3
14

1
1

3
6

Burma

Thailand

Malaga

Java

Borneo

Philippines

Palau Islands

C.r. novacguincae

C. australis

C microlepis — ..

42
7
4
3
4

TABLE 25. — Variation in the Number of Ventrals in the Species and

Subspecies of Cerberus

Taxon

Sex No. Range Mean S.D. C.V. CD.

C.r. rynchops
India

9
Burma 6

$
Thailand $

9
Malaya S

9
Java S

9
Borneo $

9
Philippines S

9
Palau Islands $

9
C.r. nuvaeguineae $

9
C. australis 9

C. microlepis $

18 141-151 145.2 2.70 1.85

0.105

139-149

144.6

3.08

2.13

139-148

143.4

2.47

1.72

0.564

136-147

140.6

2.48

1.72

145-152

149.2

2.71

1.82

0.390

143-149

146.4

4.28

2.92

142-148

144.6

1.96

1.35

0.032

142-147

144.5

2.08

1.44

151-153

152.0

—

143

145-150

147.5

—

143-148

145.0

2.16

1.49

136-157

151.1

4.23

2.17

0.053

142-156

151.1

3.28

2.80

144-151

147.7

—

143-147

145.3

2.00

1.38

145-150

148.3

147

— 159

—

144-151

147.7

—

159

159.0

152 University of Kansas Publs., Mus. Nat. Hist.

TABLE 26. — Variation in the Number of Subcaudals in the Species and

Subspecies of Cerberus

Taxon Sex No. Range Mean S.D. C.V. CD.

C.r. rynchops

India $ 16 58-68 64.1 2.79 4.35

1.426
9 16 51-62 56.4 2.60 4.61

Burma S 17 50-63 59.9 3.26 5.44

1.210
9 23 50-58 53.1 2.35 4.43

Thailand $ 5 55-66 61.1 4.10 6.65

0.593
9 6 55-66 57.1 3.37 5.89

Malaya 6 13 52-63 58.2 3.10 5.33

1.162
5 6 50-54 52.5 1.76 3.35

Java S 2 64-66 65.0 — —

9 1 —55 — — —

Borneo S 2—63 63.0 — —

9 4 44-57 51.3 5.44 —

Philippines $ 20 51-69 61.6 4.60 7.47

0.351

9 22 52-68 58.7 3.58 6.10
Palau Islands S 3 52-61 57.3 — —

9 3 52-54 53.0 — —

C.r. novaeguineae $ 3 44-49 43.7 — —

9 1 —43 — — —

C. australis 9 3 40-49 46.0 — —

C. microlepis S 2 66-67 66.5 — —

9 2 58-62 60.0 — —

riorly. The palatine is short, weak, usually bearing 10 teeth; the
pterygoid is free from the quadrate and usually has 15 teeth.

The hemipenis is deeply bilobed and clavate. The extreme basal
area of the undivided part is naked. Near, or at the place of bifurca-
tion of the sulcus, triangular papilla-like structures are arranged in
longitudinal series; each ends in a small curved spine. The distal
end of each branch of the hemipenis is finely calyculate and has
small blunt spines. The distal ends gradually merge into a median
area which have larger calyces and spines. The length of the
hemipenis is 12-16 (13.8) subcaudals; the length of the organ from
the vent to the point of bifurcation is 7-10 (9.1) subcaudals. The
retractor penis magnus muscle originates on the caudal vertebra
adjacent to subcaudals 36-41 ( 37.0) .

Distribution. — Snakes of the genus Cerberus range through
tropical coastal waters in southeastern Asia, the Indo-Australian
Archipelago and northwestern Australia, where they usually occur
in, or near, mouths of rivers and frequently in fresh water ( Fig. 24 ) .

COLUBRID SjVAKES, SUBFAMILY HOMALOPSINAE 153

TABLE 27. — Variation in Tail/Snout- vent Ratio (in per cent) in the Species

and Subspecies of Cerberus

Taxon

Sex

No.

Range

Mean

S.D.

C.V.

CD.

C.r. rhijnchops

India

23.4-31.1

27.2

2.25

8.26

0.915

20.9-27.4

23.6

1.67

7.07

Burma

24.1-31.7

26.0

1.81

6.95

1.062

18.9-24.2

22.5

1.55

6.70

Thailand

23.2-27.8

21.5

1.83

8.50

0.394

21.4-24.2

22.7

1.25

5.50

Malava _

20.8-31.5

27.9

2.40

8.61

1.290

21.8-24.8

23.2

1.22

5.26

Java

23.1

—

Borneo

— 23.8

18.5-24.0

22.1

—

Philippines

21.4-29.1

25.9

1.97

7.60

0.684

20.1-26.0

23.4

1.71

7.30

Palau Islands

23.1-25.9

24.5

24.9

—

C.r. novaeguineae

20.0-20.8

20.3

—

18.9

C. australis

9
S

2
2

21.3-22.8
25.0-25.7

22.0
25.3

—

C. microlepis

20.3

—

Key to the Species and Subspecies of Cerberus

1. Dorsal scales at midbody in 21-25 rows; ventrals 158 or

less 2

Dorsal scales at midbody in 27-29 rows; ventrals 159 or

more C. microlepis

2. Scales ovate, strongly keeled; nasal cleft extending to
first (rarely second) upper labial; first to fourth (rarely
third) lower labials in contact with anterior chin-
shields; no black diagonal streaks on chin and no black

stripe from angle of mouth to side of neck 3

Scales lanceolate, feebly keeled; nasal cleft extending to
second upper labial; first to third lower labials in con-
tact with anterior chin-shields; diagonal black chin
streaks and black stripe present from angle of mouth to
side of neck C australis

154 University of Kansas Publs., Mus. Nat. Hist.

3. Loreal in contact with first to third or fourth upper
labials; width of ventral more than three times width
of adjacent dorsal scale in first row; subcaudals 49-
72 C. rynchops rynchops

Loreal in contact with second to third or fourth upper
labials; width of ventral less than three times width
of adjacent dorsal scale in first row; subcaudals 43-
51 C. rynchops novaeguineae

Cerberus microlepis Boulenger

Cerberus cinercus Gray, (in part). Catalogue snakes British Museum, p. 64,
1849.

Cerberus microlepis Boulenger, Catalogue snakes British Museum, 3:18, 1896
[Syntypes. — BMNH 1946.1.7.24-25, adult females (types of Cerberus
cinereus Gray, (in part), 1849:64) from "Philippines"; H. Cuming collector].
Loveridge, Bull. Mus. Comp. Zool. 101:389, 1948.

Hurria microlepis; Taylor, The snakes of the Philippines Islands, Manila, p.
114, 1922.

Diagnosis. — Nasal cleft extending to first upper labial; loreal not
in contact with internasal, in contact with first three (rarely four)
upper labials; anterior chin-shields in contact with first three lower
labials; dorsal scales small, feebly keeled, arranged in 27 (rarely 29)
rows; ventrals 159 or more.

Description and variation. — The rostral is slightly broader than
high and is visible from above. The nasals are larger and form a
median suture. The nostril is a lunate slit; its cleft extends to the
first labial. The internasals are small, as long as broad, and sepa-
rated from the loreal. The prefrontals form a median suture and
are smaller than the nasals. The frontal and parietals are irregu-
larly fragmented. The loreal is single or divided, higher than long,
and in contact with the first three (rarely four) upper labials. There
is one high preocular (rarely divided), two suboculars and one or
two postoculars. The supraocular is broadest posteriorly and is
narrower than the frontal. Temporals usually consist of one pri-
mary, two secondary and three tertiary scales. The ten (rarely nine)
upper labials are higher than long; the fifth, or the fifth and sixth
(MCZ 25683, right side), or the sixth (BMNH 1946.1.7.25, right side),
is below the eye; the seventh or the eighth is the largest. The one or
two upper labials posterior to the eye are divided; and the eighth
or the ninth forms the angle of the upper margin of the mouth.
Thirteen to 15 lower labials are present of which the first three
(rarely four) are in contact with the anterior chin-shields; the sev-
enth is the largest and the tenth or eleventh forms the angle of the
lower margin of the mouth. The anterior pair of chin-shields is

CoLUBRiD Snakes, Subfamily Homalopsinae

155

- CO I

!« ft. D

°1 2f
y u i=

■'" g 13

0) (U
•- CO *-■

*-> *.
c8 la Si

— ^^

o -.i-

>- '" S

>« a; o

<u o a; "u

5 M

w 2 C

^ -y; S:

r/i

•^

a !»

Oi^

^
g

■^

r~"

-a

"Tt

-H

"-^

■2

■^

■j-i

'-!.>

M-i

O
O

■ — "

-2

y
'y

»S1

1-^

feU-

156 University of Kansas Publs., Mus. Nat. Hist.

wider and longer than the two posterior pairs; the latter pairs are
separated by small scales. The mental is small and triangular. Nine
to ten scales separate the chin-shields from the first complete ven-
tral. Males have 160 to 165 (162.5) ventrals, and two females have
159. Males have 66 or 67, and females 58 to 62 subcaudals. The
dorsal scales are small, weakly keeled, and arranged in 27 (rarely
29) rows which reduce to 19 rows just before the vent. Scale reduc-
tion in two females (BMNH 1946.1.7.24-25) and one male (CAS
60947 ) are, respectively,

5+6(114) 4+5(128) 4+5(135) 4+5(142)

29 27 25 23 21

6+7(112) 5+6(128) ^ 4+5(135) 4+5(143)

5+6(153)

19(159).

5+6(153)

6+7(97) 6+7(116) 4+5(132)

29 28 26 24 22

6+7(83) 4+5(94) 5+6(116) 4+5(135)

3+4(147)

20(159).

5+6(22) 5+6(27) 4+5(108) 4+5(121)

31 29 27 25 23

6+7(16) 6+7(31) 6+7(116) 5+6(131)

4+5(135) 3+4(148)

21 19(160).

5+6(139) 2+3(153)

The 15 to 17 maxillary teeth are followed by an interspace and
two enlarged, grooved teeth.

The hemipenis extends to the level of the 12th to 16th sub-
caudal; the point of bifurcation is at the level of the 7th to 10th
subcaudal. The retractor penis magnus muscle originates on the
caudal vertebra at the level of the 41st to 43rd subcaudal.

The dorsal ground color is reddish brown. The rostral-orbital-
postorbital stripe is continuous or broken on the side of the anterior
part of the body. A single median nape stripe, one scale wide,
extends posteriorly one head length from the posterior edges of the
parietals. A pair of dorsolateral median stripes, two scales apart,
extend parallel to the median nape stripe for the length of the body.
The three outer rows of scales are yellow, which gradually merges
into the reddish brown of the dorsum. There are 35 to 40 black

CoLUBRiD Snakes, Subfamily Homalopsinae 157

crossbars or series of spots, two scales in length on the body, and 14
to 16 similar crossbars or spots on the tail. The venter is yellow with
reddish brown blotches and a broad mid-ventral reddish brown line
extending from the base to the tip of the tail.

The body and total lengths are 633-658 (645.5 mm.) and 791-827
(804 mm.) respectively in the males; 582 mm. and 700 mm. respec-
tively in one female. Males have more ventrals (160-165) and sub-
caudals (66-67) than females, which have 159 and 58-62 respectively.
The tail/ snout-vent ratios in males are 25.0 to 25.7 (25.3); a female
has a ratio of 20.3. In males the entire body is heavily tuberculate;
at least five papillae are arranged in a row on the posterior edge of
each dorsal scale, and about 60 are scattered at random on each
ventral.

Remarks. — Gray's syntypes of Cerberus cinereus are BMNH
1946.1.7.24-25 from "Philippines" collected by H. Cuming and
BMNH 1946.1.21.55-57 from "India" collected by Gen. Hardwicke.
Those collected by H. Cuming from the "Philippines" are Cerberus
microlepis Boulenger and those collected by Gen. Hardwicke from
"India" are Cerberus rynchops njnchops (Schneider).

Distribution. — This species is known only from Luzon, Philip-
pine Islands ( Fig. 24 ) . Specimens examined, four, as follows.

Philippines: Luzon — Camarines Province: Lake Buhi, CAS 60947, MCZ
25683. No specific locality, BMNH 1946.1.7.24-25.

Cerberus australis (Gray)

Homalopsis australis Gray, Zoological miscellany, p. 65, 1842 [Holotype. —
BMNH 1946.1.2.40, from Port Essington, northwestern Australia; John
Gilbert collector].

Cerberus australis; Gray, Catalogue snakes British Museum, p. 65, 1849.
Boulenger, Catalogue snakes British Museum, 4:18, 1896. Glauret, A
handbook of the snakes of western Australia, p. 20, 1950. Kinghorn, The
snakes of Australia, p. 90, 1956.

Cerberus njnchops australis; Loveridge, Bull. Mus. Comp. Zool., 101:389,
1948.

Diagnosis. — Nasal cleft extending to second upper labial; loreal
not in contact with internasal, in contact with second and third or
second to fourth upper labials; anterior chin-shields in contact with
first three lower labials; dorsal scales in 23 or 25 rows, not so
strongly keeled as in C. rijncliops; ventrals 158 or less; subcaudals
40-49 (Fig. 25).

Description and variation. — The rostral is as high as, or slightly
higher than broad, and is visible from above. The nasals are large
and form a median suture. The nostril is a transverse slit, crescent-
shaped, and convex anteriorly. There are two nasal clefts; one

158

University of Kansas Publs., Mus. Nat. Hist.

Fig. 25. Lateral view of the head of Cerberus australis (MCZ 48862), xl.2.

extends to the second labial and the other extends to the internasal,
which is divided by longitudinal suture into two parts and separated
from the loreal. The prefrontals form a median suture and are
smaller than the nasals. The frontal and the parietals are irregu-
larly fragmented. The loreal is single and is in contact with the
second and third upper labials. The ocular scales overlap one
another. There is usually one high postocular (BMNH 1946.1.2.40
has two; in MCZ 48862 the preocular is fused with the subocular on
the left side), one subocular, and two postoculars (MCZ 48862 has
three). The temporals are scale-like. Of the nine or 10 upper
labials, usually the fifth, occasionally the fifth and sixth, are below
the eye. The ninth is the largest, and the last two are divided. The
labial border angles dorsally at the eighth or ninth upper labial.
Usually 13, occasionally 14 or 15, lower labials are present; the first
three (rarely four) border the anterior chin-shields; the seventh is
the largest, and the labial border angles dorsally at the tenth or
eleventh lower labial. The anterior pair of chin-shields is wider and
longer than the posterior pairs; the scales in each of the latter pairs
are separated by small scales. The mental is small and triangular.
Nine to 10 gulars and four or five incomplete ventrals separate the
chin-shields from the first complete ventral. The narrow ventrals
are about three times the width of the adjacent dorsal scales in the
first row. No males are available. Females have 140-146 (143.3)
ventrals and 40-49 (45.3) subcaudals. The dorsal scales are small,
lanceolate, feebly keeled, and in 23 rows, which reduce to 17 rows
just before the vent. Scale reduction for one female (MCZ 48862)

IS,

6+7(7) 6+7(59) 6+7(98)
27 25 23 21

7+8(9)
4+5(134)

4+5(133)

5+6(68) 6+7(96)
17(144).

5+6(117)
4+5(117)

CoLUBRiD Snakes, Subfamily Homalopsinae 159

The 15 or 16 maxillary teeth are followed by an interspace and
two enlarged, groo\'ed teeth.

The bod\- and tail lengths of two females are 285 mm. and 494
mm., and 65 mm. and 105 mm., respectively.

The head is gray with a rostral-orbital-postorbital stripe which
extends about two or three head lengths posteriorly from the level
on the seventh row of dorsals. In the occipital region is a short
median nape stripe not quite one scale wide. The dorsum of the
body is slate gray with 26 to 32 irregular short, dark brown or black
crossbars, which are one to two scales in length and do not extend
below the fifth dorsal scale row. One to three outer rows of dorsals
are vellow. A black line, about two scales wide, extends from the
angle of the mouth posteriorly on the neck along the second and
third rows of dorsals. The upper labials, the mental, the first pair of
lower labials, and the margins between the anterior and posterior
chin-shields are black; the remaining lower labials and the chin are
yellow, spotted with black. The ventrals are yellowish-white. The
transverse black bars which are about two ventrals in length offset
at mid-line, form a checker-board pattern on the venter. The black
ventral bars, 31 to 42 in number on each side, extend to the level of
the third row of dorsal scales. The tail is yellow midventrally and
bordered by black lines which unite distally to form a black tip.

Remarks. — This species is found in the sea, in estuaries, and in
fresh water streams. Like other species of Cerberus, this snake
basks on mud flats. Although the snake is forbidding in appearance,
it is harmless and has a quiet disposition. Kinghorn (1956:90) re-
ported that it emits a disagreeable odor when unduly excited. Like
Cerberus rijnchops, when progressing along the ground C. austraUs
throws a coil of its body in advance of the head.

Distribution. — Specimens have been collected (Fig. 24) at Drys-
dale River, Kimberleys, Darwin, and at Port Essington in northern
Australia. Specimens examined, three, as follows.

Australia: Northern Territory — Darwin, MCZ 48846, 48862; Port Essing-
ton BMXH 1946.1.2.40.

Cerberus rynchops (Schneider)

Hydriis rijnchops Schneider, Historiae Amphibiorum, 1:246, 1799 [Based on
illustration in Russell's Indian serpents, 1:23, pi. 17, 1796. Type species. —
Hydrus rynchops from Ganjam, India].

Diagnosis. — Nasal cleft extending to first upper labial; loreal
usually in contact with internasal, in contact with first four (or
three) upper labials; anterior chin-shields in contact with first four

160 University of Kansas Publs., Mus. Nat. Hist.

(rarely three) lower labials; dorsal scales larger, ovate, strongly
keeled, arranged in 23 or 25 rows; ventrals 158 or less.

Description and variation. — The rostral is as broad as high and
is visible from above. The nasals are large and form a median
suture. The nasal cleft extends to the first (rarely second) labial.
The intemasal is divided and is in contact with, or narrowly sepa-
rated from, the loreal. The prefrontals form a median suture. The
parietals are fragmented into small scales. Usually there is one
preocular, two suboculars, and one postocular. The temporals are
scale-like. There are nine or ten (rarely 11) upper labials.

The 12 to 19 maxillary teeth are followed by an interspace and
a pair of enlarged, grooved teeth.

Cerberus rynchops rynchops (Schneider)

Hydriis rynchops rynchops Schneider, Historiae Amphibioruni, 1:246, 1799
[Based on illustration in Russell's Indian serpents, 1:23, pi. 17, 1796. Type
species. — Hydrus rynchops from Ganjarn, India]. Loveridge, Bull. Mus.
Comp. Zool., 101:389, 1948. Haas, Treubia, 20(3) :577, 1950.

Elaps hoaeformis Schneider, Historiae Amphibiorum, 2:301, 1801 [Holotype. —
Mus. Universite, Halle; locality and collector unknown].

Hydrus cinereus Shaw, General Zoology, 3:567, 1802 [Based on illustration in
Russell's Indian serpents, 1:23, pi. 17, 1796]. Cantor, Proc. Zool. Soc. Lon-
don, p. 54, 1839.

Hurria schneideriana Daudin, Histoire naturelle des reptiles, 5:281, 1803
[Based on Schneider's Elaps hoaeformis, 1801].

Hurria bilineata Daudin, Histoire natiuelle des reptiles, 5:284, 1803 [Based on
illustration in Russell's Indian serpents, vol. 1, pi. 40; type locality,
Hyderabad].

Cohiber cerberus Daudin, Histoire naturelle des reptiles, 7:167, 1803 [Based
on illustration in Russell's Indian serpents, 1:23, pi. 17, 1796].

Python rhynchops; Merrem, Tentamen systematis amphibiorum, p. 90, 1820.

Homalopsis molurus Boie, Isis, p. 213, 1826 [Based on illustration in Russell's
Indian seipents, vol. 2, pi. 40].

Cerberus grantii Cantor, Trans. Med. Phvs. Soc. Calcutta, 8:135, 1836 \fide
Smith, 1943:393; type-locality "India"].

Coluber obtusatus Reinwart, in Schlegel, Essai sur la phvsionomie des serpents,
2:341, 1837.

Cerberus cinereus; Cantor, Proc. Zool. Soc. London, p. 54, 1839. Gray, (in
part), Catalogue snakes British Museum, p. 64, 1849.

Homalopsis rhynchops; Cantor, Catalogue Malay reptiles, p. 94, 1847.

Cerberus unicolor Gray, Catalogue snakes British Museum, p. 65, 1849 [Holo-
type.— BMXH 1946.1.2.45 from "Philippines"; H. Cuming collector],

Cerberus acutus Gray, Catalogue snakes British Museum, p. 65, 1849 [Holo-
t>pe. — BMXH 1946.1.2.34 from Borneo; presented by Leyden Museum].

Cerberus hoaeformis; Dumeril and Bil^ron, Erpetologie generale . . . reptiles,
7:978, 1854. Bocourt, Nouv. Arch. Mus., p. 9, 1886. Bourret, Invent.
Gen. Indochine, 3:241, 1927.

Honialopsis hoaeformis; Jan, Elenco systematico degli ofidi, p. 77, 1863; Arch.
Zool. Anat. Phys., 3:257, 1865.

Cerberus rhynchops; Giinther, Reptiles of British India, p. 279, 1864. Ander-
son, Proc. Zool. Soc. London, p. 179, 1871. Nicholson, Indian snakes, p.
62, 1874. Morice, Sur la Fauna de la Cochinchine francaise, Lyon, p. 58,

CoLUBRiD Snakes, Subfamily Homalopsinae 161

1875. Theobald, Catalogue reptiles British Lndia, p. 185, 1876. Murray,
N'ertebratc Zool. Sind, p. 381, 1884. Keswal, Jour. Bombay Nat. Hist. See,
1:173, 1886. Sclater, List of snakes in Indian Museum, p. 54, 1891. Muller,
\'erh. Ges. Basel, 10:831, 1895. Boettger, Prof. Dr. W. Kukenthal's Zool.
Anz., 18:132, 1895. Elera, Catalogo sistematico . . . Ophidiens. Manila,
p. 431, 1895. Ferguson, Jour. Bombay Nat. Hist. Soc., 10:74, 1896.
Flower, Proc. Zool. Soc. London, p. 888, 1896. Boulenger, Catalogue
snakes British Museum, 3:16, 1896; Proc. Zool. Soc. London, 13:201,
1897. Bethencourt-Ferreira, Jour. Sci. Lisboa, 4:229, 1897. Boulenger,
Proc. Zool. Soc. London, 13:507, 1897. Boettger, Katalog der Rep-
tilien . . . Schlangen, pt. 2, p. 88, 1898. Peracca, Rev. Suisse Zool.,
7:327, 1899. Flower, Proc. Zool. Soc. London, p. 679, 1899. Wall and
Evans, Jour. Bombay Nat. Hist. Soc, 13:345, 1900. Schenkel, Verb.
Naturf. Ges. Basel, 13:166, 1901. Boettger, Abh. Senckenb. Ges., 25:326,
1901. Laidlaw, Proc. Zool. Soc. London, 3:578, 1901. Wall and Evans,
Jour. Bombay Nat. Hist. Soc, 13:612, 1901. Alcock and Rogers, Proc Roy.
Soc. London, p. 449, 1902. Lampe, Jahrb. Nassau Veriens, 55:31, 1902.
Boulenger, FascicuU Malayensis Zoology, 1:175, 1903. Volz, Zool. Jahrb.
Syst., 20:498, 1904. Mocquard, Mission Pavie Indocliine, 3:482, 1904.
Annandale, Jour. Asiatic Soc. Bengal, p. 176, 1905; Mem. Indian Mus.,
5:170, 1905. Wall, Jour. Bombay Nat. Hist. Soc, 16(2) :307, 1905. Rosen,
Ami. Mag. Nat. Hist., ser. 7, 15:175, 1905. Annandale, Record Indian Mus.,
1317, 1907. Mocquard, Revue Coloniale, p. 51, 1907. Bedot, Rev. Suisse
Zool., 17:147, 1909. Roux, Zool. Jahrb. Jena, 5:502, 1911. Boulenger, A
vertebrate fauna of the Malay Peninsula . . . Reptilia and Batracliia, p.
163, 1912. Abercromby, Spolia Zeyl, 8:304, 1913. Smith, Jour. Nat. Hist.
Siam, 1(1):102, 1914; Jour. Nat. Hist. Siam, 1(3):213, 1914. Annandale,
Record Indian Mus., 2:96, 1915. Smith, Jour. Nat. Hist. Siam, 1(3):246,
1915. Gyldenstolpe, Svensk. vet. Akad. Handl., 53(3) :19, 1916. Hol-
zinger-Tenever, Mitt. Zool. Mus. Berlin, 8:440, 1917. Rooij, Reptiles
Indo-Australian Archipelago, 2:187, 1917. Wall, Jour. Bombay Nat. Hist.
Soc, 26:89, 1919. Robinson and Kloss, Jour. Federated Malay States Mus,,
7(12):303, 1920. Wall, Snakes of Ceylon. Colombo, p. 257, 1921. Annan-
dale, Record Indian Mus., 22(4) :332, 1921. Rooij, Zool. Meded. Leiden,
6:218, 1922. Pliisalix, Animaux venimeux et venins, 2:285, 1922. Werner,
Arch. Naturg., 89A(8):162, 1923. Roux, Rev. Suisse, Zool., 32(20):319,

1925. Kopstein, Zool. Med. Rijks Mus., 9:109, 1926. Jong, Treubia, 7:87,

1926. Dammenuan, Treubia, 3:323, 1926. Bourret, Invent. Gen. Indo-
chine, 3:241, 1927. Kopstein, Treubia, 9(4):443, 1927. Werner, Zool.
Jahrb., Jena, 57:185, 1928. Jong, Treubia, 10:151, 1928. Scirteccu, Atti.
Soc. Italy, 68(1 ):74, 1929. Brongersma, Treubia, p. 67, 1929. Kopstein,
Weltevreden, 138:131, 1930; Treubia, 13(1 ):1, 1931. Smedley, Bull.
Raffles Mus., 5:104, 1931. Brongersma, Zool. Med. Rijks. Mus., 16:3, 1933.
Prater, Jour. Bombay Nat. Hist. Soc, 36(2) :393, 1933. Bourret, Bull.
Instr. Pub., p. 22, 1934.^ Brongersma, Zool. Med. Rijks. Mus., 17:200, 1934.
Bourret, Serpents de ITndochine, 2:295, 1936. Smith, Fauna of British
India . . . Reptilia and Ampliibia, 3:393, 1943.

Hurria njnchops; Stejneger, Bull. U.S. xNad. Mus., 58:304, 1907. Osliima,
Annat. Zool. Japonensis, 7(2): 191, 1908. Griffin, Jour. Philippine Sci.,
4:599, 1909. Stejneger, Proc. U.S. Natl. Mus., 38:105, 1910. Barbour,
Mem. Mus. Comp. Zool., 44:123, 1912. Wall, Snakes of Ceylon, Colombo,
p. 257, 1921. Talyor, The snakes of the Philippine Islands, p. Ill, 1922.
Sworder, Singapore Nat., 2:66, 1923. Prater, Jour. Bombay Nat. Hist. Soc,
30:171, 1923. Wall, How to identify the snakes of India, p. 36, 1923.
Sworder, Singapore Nat., 3/4:20, 1924. Wall, Jour. Bombay Nat. Hist.
Soc, 29:867, 1924. Prater, Jour. Bombay Nat. Hist. Soc, 30:171, 1924.
Smith, Sarawak Museum Jour., 3(2):5, 1925. Wall, Jour. Bombay Nat.
Hist. Soc, 30:817, 1925. Raj, Bull. Madras Govt. Mus., 1(1):183, 1927.
Gee, Bull. Biol. Yenching Univ., p 73, 1930. Mertens, Abh. Senckenb.
Naturf. Ges., 42(3/4) :310, 1930. Cochran, Proc U.S. Natl. Mus.,
77(11):31, 1930.

Cerberus rynchops; Smith, Bull. Raffles Mus., 3:61, 1930.

162 University of Kansas Publs., Mus. Nat. Hist.

Diagnosis. — Loreal in contact with first to third, or fourth, upper
labials; subcaudals 49-72; postoculars not horizontally divided.

Description and variation. — The nasal cleft extends to the first
labial. The internasal is usually divided unequally; in some speci-
mens (KU 92510 and 92515; ANSP 11997-9) it is not divided. In one
specimen (MCZ 25692) the left internasal is fused with the azygous
shield. In two specimens from Ernakulam, Cochine State, India
(SU 13073, 12257), a small scale lies between the internasal and
loreal on each side. The prefrontals usually form a median suture,
but about 40 per cent of the Philippine and Malayan specimens and
about 4 per cent of the Indian and Burmese specimens have azygous
shields. The frontal is usually partly fragmented into one large
anterior piece, and several small posterior pieces. In 33 per cent of
the specimens the anterior piece extends beyond the supraoculars.
Sometimes the whole plate is broken up into small scales, which
resemble those in the occipital region. The loreal is usually single;
in about 6 per cent of the specimens, it is divided vertically. In
some specimens the loreal is divided only on one side. In speci-
mens having a divided loreal, the posterior one is smaller. The
supraocular is usually of uniform width, although sometimes the
posterior end is wider than the anterior end. The arrangement of
ocular scales (one preocular, one postocular, and two suboculars)
occurs fairly constantly throughout the geographical range of this
subspecies.

Some specimens have one or three suboculars instead of the
normal two. In Thai specimens, and in some Indian and Malayan
specimens one high preocular extends below the eye and is in con-
tact with the postocular (Table 23). Usually there are 10 (rarely
nine) upper labials. The snakes from Burma have the fifth upper
labial below the eye; others have the sixth, or fifth and sixth below
the eye. In Burmese and Indian snakes the sixth labial is the larg-
est; in other specimens the eighth upper labial is the largest, and
the last two or three are divided (MCZ 25690 has none divided).
The margin of the upper lip forms an angle at the eighth or ninth
upper labial. Twelve to 14 lower labials are present; the first four
(rarely three) are in contact with the anterior chin-shields. The
seventh is the largest and usually the tenth (occasionally ninth or
eleventh) forms an angle at the margin of the mouth. The anterior
pair of chin-shields is wider and longer than the posterior pair, the
scales of which are separated by a pair of small scales. Seven to
nine gulars and one to three incomplete ventrals separate the chin-
shields from the first complete ventral. The ranges and means of

CoLUBRiD Snakes, Subfamily Homalopsinae

163

trt

>■

4-*

•^H

(ZJ

;->

119

».

55.

-C

!»

-O

<»

t-M

d
Z

4.^

■n

<-4-l

;-!

't-

;-.

(—•

(-•

4-*

•rH

Tt^

t--

»o

1— (

1— H

t-cq

00 CD
T (M

coin

CO 05

CD 00

ini>

C:5CD

1 — 1

■-i"*'

^'od

»— I I— i

^'ic

OCD

doi

oit-^

let-'

CO CD

^H 00

oi oi

f>ioi

doi

coco

O-h'

CO m CO

lot- Tft- Tf t-

t o

t-O

■>-' Bf-" Bf ■"

ci:

5t3

c'5

a
>

3
C/D

Cfl

4-1

•»-<

C/]

ft
ft

r£3

t- 00

c
>

-t CO 05 O 05 1- (N

^_ fM p CO *n p (N

d ^' CO '^' CO ^ cm'

ccd

in-^

005

m o

05 >— <

cop

l:-0^

CO'^'

TT CD

CD C<1

coco"

-r oJ
^ in

Oi CD

in CM

cMin

cm' — <

in CD

din

CD (M

'^ 30
— 1 CO

1— 1

CO ' — '
CO CO

■27.8
■148

-^ CO

^ CO

o .-<'
coco

in in

CM^

inOT

rHCO

00 05
CO CM

ooo

'^ in

i-H

^CO-
CM

COC^l

in^

^ m
1—1

05 '—<

in^-

in 6

CM^

00 -H

Tt^in
■—1

00 4

1— (

tc^ Wlj^ tc.^ M^^ tc^ tt^ be

164 University of Kansas Publs., Mus. Nat. Hist.

TABLE 29. — Geographical Variation in the Number of Ventrals in Cerberus
njnchops rijnchops (Males and Females).

Locality Number Range Mean S.D.

Bombay 6 142-151 147.0 5.47

Calcutta 10 142-148 145.4 1.78

Madras 6 143-147 145.2 1.84

Ceylon 2 137-139 138.0 —

Andamans 1 — 142 — —

Bunna 42 136-148 142.0 2.47

Thailand 15 143-152 147.8 3.49

Penang 11 141-149 145.5 2.77

Singapore 12 141-147 143.8 1.64

Java 1 — 143 — —

Bali 1 — 146 — —

Philippines 42 136-157 151.1 3.75

Luzon 6 147-157 150.8 3.54

Table 4 150-154 151.8 1.71

Cebu 9 149-155 151.6 2.18

Negros 7 151-153 152.1 0.28

Mindanao 11 138-160 152.3 6.11

ventrals, subcaudals, and tail/ snout-vent ratios are summarized in
Tables 25-27. The ventrals usually are entire but in a few speci-
mens some ventrals are divided; frequently the last ventral is di-
vided. The last two ventrals are always narrower and longer than
the preceding ventrals and are bordered by an extra dorsal scale on
each side; thus a dorsal scale count taken at the level of the last two
ventrals results in a count higher by two scales than if the count
were taken at the third or fourth ventral anterior to the vent. The
umbilical scar seems to leave a permanent marking in many adult
specimens of this species. The correlation of the number or num-
bers of ventrals on which the scar occurs with the total num-
bers of ventrals in males and females of Burmese and Philippine
specimens of C. r. njnchops is shown in Fig. 26. The location of the
"umbilicus" in relation to the number of ventrals constantly lies
between 20-25 ventrals anterior to the vent in both specimens of
both sexes from Burma and the Philippines. The \'entrals on which
the scars occur in Burmese and Philippines specimens arc different,
for Burmese samples ha\'e lower ordinal nmnbers on which the scars
occur, whereas the Philippine samples have more ventrals and thus
higher ordinal number on which the scars occur. At midbody the
dorsal scales are arranged in 23 or 25 rows; these numbers occur in
approximately equal proportions in Indian and Philippine specimens,
whereas in Burma about 85 per cent of the specimens have 25 rows.
The Malayan and Thai specimens have a higher percentage of in-
dividuals having 23 rows. The scale reduction for one male (BMNH
1946.1.2.45), and one female (KU 92516) are, respectively,

CoLUBRiD Snakes, Subfamily Homalopsinae 165

TABLE 30. — Ceoj^iapliical Variation in the Number of Subeaiidals in Cerberus
njnchops rynchops ( Males and Females ) .

Locality Number Range Mean S.D.

Bombay 4 51-67 60.5 6.81

Calcutta 9 54-67 60.3 4.71

Madras 6 54-64 58.0 4.00

Ceylon .- 1 53-55 54.0 —

Andamans -— 1 — 63 — —

Burma 40 50-63 56.5 2.80

Thailand 11 55-66 59.4 3.73

Penang 12 50-63 57.3 4.01

Singapore -- 9 52-61 55.9 2.85

Java 1 —55 — —

Bali - 1 —62 — —

Philippines 42 51-69 60.1 4.09

Luzon 6 52-63 57.2 4.02

Table 4 56-64 61.8 3.86

Cebu 9 58-68 62.6 3.36

Negros 6 57-62 58.7 1.75

Mindanao 10 54-69 61.6 4.40

6+7(9) 4+5(140) 3+4(149)
25 23 21 injured 17 35(153).

6+7(11) 4+5(137) 3+4(148)

7+8(52) 5+6(90) 5+6(107) 4+5(123)
27 25 23 21 19(142).

6+7(84) 6+7(88) 4+5(99) 4+5(123)

The hemipenis extends to the level of the 12th to 15th sub-
caudal, and the point of bifurcation is at the level of the 8th to 10th
subcaudal. The retractor penis magnus muscle originates on a
caudal vertebra adjacent to the 36th to 41st subcaudal.

The coloration is variable. Usually the dorsum is slate gray or
dark brown. A dark interorbital stripe, a pair of rostral-orbital-
postorbital stripes, a median nape stripe and a pair of dorsolateral
nape stripes, are present. The 28-45 crossbars are broken into dots
or dashes in some specimens. The ventrals and subcaudals are
variegated or marked by a broad midventral dark hne. Indistinct
crossbars, dots or dashes are present on the dorsal surface of the
tail. Specimens from India and Burma can be distinguished by the
presence of fewer black crossbars— 27-34 (31.6) and 25-38 (28.8), re-
spectively. The upper and lower labials, chin-shields and gulars are
usually devoid of black spots. The black dorsal crossbars and the
black ventrolateral transverse bars have parallel margins and are
two or three scales in length; they extend laterally onto two or
three rows of dorsal scales. In the Burmese specimens the black
ventrolateral transverse bars are dome shaped and the bases are
broad, four to six scales in length. The black ventrolateral bars

166 University of Kansas Publs., Mus. Nat. Hist.

TABLE 31. — Geographical Variation in tlie Ratio of Tail/Snout-vent Length in

Cerberus rynchops njnchops (Samples Males and Females).

Locality Number Range Mean S.D.

Calcutta 9 22.88-31.14 25.5 2.38

Andamans 1 — 27.73 — —

Burma 42 18.88-31.67 24.3 1.68

Thailand 10 21.77-27.84 22.1 1.54

Penang 11 21.77-26.08 24.5 1.18

Singapore 7 20.78-27.49 25.3 2.53

Java 1 — 23.05 — —

Bali 1 — 25.89 — —

Pliilippines 38 20.07-29.14 24.7 1.84

Luzon 6 21.43-25.91 23.1 1.68

Table 4 21.63-28.16 25.9 2.92

Cebu 9 22.97-29.14 26.3 1.86

Negros 6 22.96-24.83 23.9 .22

Mindanao 9 19.81-27.45 24.1 2.72

extend laterally onto two to five dorsal rows of scales and often are
confluent \\ ith the dorsal crossbars. In Indian specimens the post-
orbital stripe is broad and long, whereas in Burmese specimens it
is narrow and short.

Snakes from Thailand, Malaya, and the Philippines have more
dorsal crossbars. The crossbars are usually distinct and one or al-
most two scales in length. The margins of the crossbars are zigzag,
or the crossbars are broken into short zigzag streaks, or represented
merely by dots and dashes. The black ventrolateral transverse bars
of the body seldom extend above the first dorsal row of scales; in
some specimens one or two ventrolateral transverse bars extend to
the second row of scales. The upper and lower labials, chin-shields
and gulars are spotted. The subcaudals are usually marked by a
broad brown midventral line. Their lateral edges and those of the
first caudal row are white or pale brown, or spotted by white. In
some specimens the subcaudals are variegated with black and
white, and the black ventrolateral transverse bars extend to the first
row of dorsal scales.

The interorbital stripe in Thai and Malayan specimens is usually
continuous but in some specimens the stripe is interrupted medially.
The rostral-orbital-postorbital stripes and nape stripe are present,
but dorsolateral stripes are absent. The margin of the dorsal cross-
bars are zigzag or straight. Usually the subcaudals and the first
dorsal row of scales on the tail are variegated with more black,
brown, or gray than white (Fig. 27).

Males have more subcaudals and longer tails than females
(Tables 26-27), but the differences are not statistically significant.

Ontogenetic variation. — An attempt was made to determine if

CoLUBRiD Snakes, Subfamily Homalopsinae

167

120 125

.130

,135

Position of umbilicus -

-ventral

115 +

140

145

Fig. 26. Relationship between position of uml)ilical scar and total number of
\entrals in Cerberus njnchops rijnchops from Burma (circles) and the Philip-
pines (squares). Circles and squares represent the ordinal numbers of the
scales involved with a scar. Each isolated symljol (closed for males; open for
females) or group of symbols connected by a line represents a single scar. The
diagonal lines represent the position of the 20th ventral (right) and the 25th
ventral (left) counted anteriad from the anal plate.

168

University of Kansas Publs., Mus. Nat. Hist.

Fig. 27. Lateral view of tlie body of Cerberus rynchops rynchops (MCZ 15212),

Xl.l.

"selection of juveniles," as suggested by Dunn (1942), Inger (1943),
and Gans (1949), occurs in this species. The differences in ventrals,
caudals, and tail/ snout-vent ratios of juveniles and adults from
three localities are summarized in Table 28. Although juveniles are
more brightly colored than adults, there is no ontogenetic change in
pattern.

Remarks. — This snake is usually found in the brackish waters of
tidal rivers, creeks, and estuaries, but is equally at home in fresh
water in the upper reaches of the rivers, and in salt waters along the
coasts. Wall (1919:89) stated that it was a powerful swimmer and
that the snakes often anchor themselves by the tail to some sub-
merged object and swing with the current.

Boulenger (1896:17) reported a specimen having 27 rows of dor-
sal scales from Trevandum, India, preserved in the Travancore
Museum.

Distribution. — This subspecies is found (Fig. 24) almost
throughout the range of the subfamily Homalopsinae. The snakes
have been recorded from India as far west as Sind, Burma, Malay
Peninsula, Thailand, Philippines, and the Sunda Islands as far east
as Ceram; they range into the Pacific to the Palau Islands. Speci-
mens examined, 142, as follows.

Borneo: no locality, BMNH 1946.1.2.34, UMMZ 65671. British Borneo:
North Borneo — Ranau Mt. wooded area, FMNH 121434. Saraivak — Kiiching,
Matang Mt. MCZ 15212-3. Burma: Pegu Division — Hlaing River near Insein,
KKG 100, 102-104, 109, KU 92496-507, 92518-26; Rangoon, MCZ 3696;
Twante near Rangoon, 92493-95, 92508-17. East Pakistan: Bnngaalii, MCZ
58256. India: no locality, ANSP 11997-99, 17037, BMNH 1946.1.21.55-57;
Krusadai Island, SU 13073-4. Kcra/o— Ernakulum, SU 12257, 12374.
Andamans — no definite locality, SU 12379. Madras — Madras, SU 12256;
Mannallapurnni, 55 mi. sonth of Madras, SU 12381; Pondicherv, MCZ 2229;
Tambarau, UMMZ 113469. A/a/iara.s/i/ra— Thana district, Gholvad, UMMZ
94949-50. West Bengal— no definite locality, AMNH 3837-9, ANSP 5118-9;
Calcutta, ANSP 5122-3, MCZ .5501, 5589. Indonesia: Ba//— Boedeleng, MCZ
7527. /a L-a— Djakarta, MCZ 3208. Malaya: Jahore state— AMNH 58371;

CoLUBRiD Snakes, Subfamily Homalopsinae 169

Penang, MCZ 940(7), 5596(2), UMMZ 65657, 65862; Singapnir, MCZ 936,
5120, 58879; Iiirong River, MCZ 5880. Palau Island: no definite locality,
MCZ 1299(3), UMMZ 65857; Koror, AMXII 70651. Philippines: no definite
localit>-, BMXH 1946.1.2.45. Bantai/an—SV 16842. Cehu—SU 13075-6
12369, 12371-73, 12380. JoIo—SV 12255. Lu^on— Manila, MCZ 25692; near
Manila, AMXH 75487-8, SU 6258, 13077. Mindanao— Ka^amhugan, Lenao,
SU 13067; Tahnan, Cotabato, SU 25684-5, UMMZ 65432; Zamhoanga SU
13104-5, 13068. A'c^ms— Dumaguete, SU 8578, 8704, 8706, 8719, 12254,
12354. 12375, 13080. Pfl»fl!/— Aliminos, MCZ 25690; Capiz, SU 13079; Iloilo,
SU 8696-7. South Vietnam: Cape St. Jacques, AMNH 14188-90, MCZ
12543-5. Thailand: Soidh Thailand—Bangkok, MCZ 20351-8; 38 km. from
Bangkok, AMNH 92286.

Cerberus rynchops novaeguineae Loveridge

Ccibcnts n/nchops novaeguineae Loveridge, Bull. Mus. Comp. Zool., 101: 388,
1948 [Holotype.— MCZ 22818, an adult male from Merauke, Dutch New
Guinea; P. T. L. Putman collector].

Diagnosis. — Loreal in contact with second and third, or second
to fourth upper labials; subcaudals 43 to 51; two postoculars, one
horizontally divided.

Description and variation. — The nasal cleft extends to the first
(rarely second) upper labial. The prefrontals form a median suture
(separated from azygous shield in MCZ 22818). The frontal is en-
tire, half again as broad and as long as the supraocular. The latter
is fragmented. A single loreal is in contact with second to fourth,
or second and third, upper labials. Of the eight or nine upper
labials, the fifth or sixth is the largest and is below the eye; those
labials behind the fifth or sixth are divided. The margin of the
upper lip forms an angle at the eighth or ninth upper labial.
Usually 13 (occasionally 12) lower labials are present. The first to
fourth are in contact with the anterior chin-shields; the seventh is
the largest, and the tenth forms the angle of the lower margin of
the mouth. The anterior pair of chin-shields is wider and longer
than the posterior pair; the scales in the latter are separated by a
pair of small scales. Eight or nine gulars and two or three incom-
plete ventrals separate the chin-shields from the first complete
ventral. The dorsal scales are large, ovate, keeled, and arranged in
23 or 25 rows, which reduce to 17 rows just before the vent (Table
24). Two males have 145 and 150 ventrals, and 49 and 48 sub-
caudals; two females have 147 and 150 ventrals, and 43 and 44 sub-
caudals respectively. The narrow ventrals are about three times
the \\'idth of the adjacent dorsal scales in the first row at midbody.
The scale reduction in one male (MCZ 22818) and one female (MCZ
22820) are, respectively,

4+5(5) 4+5(53) 4+5(136)
27 25 23 21 injured (145).

3+4(5) 4+5(55) 4+5(132)

170 University of Kansas Publs., Mus. Nat. Hist.

6+7(53) 5+6(70) 6+7(95) 4+5(115)
25 23 21 19 17.

3+4(4) ^ 8+9(23) 8+9(27) 4+5(119)

The hemipenis extends to the level of the thirteenth subcaudal,
and the point of bifurcation is at the level of the seventh subcaudal.
The retractor penis magnus muscle originates on the caudal verte-
bra at the level of the 34th subcaudal.

The dorsum is dirty brown with about 31 black transverse bars
on the body and 13 on the tail. The first three rows of dorsal scales
are pale brown. The venter is pale yellow with 48 to 55 black
patches, which extend across the belly to the first row of dorsal
scales. In the proximal region of the tail black spots on the outer
edges of the subcaudals merge to become a continuous black line
distally.

The head and chin of males are heavily tuberculate. Males have
more subcaudals ( 48-49 ) than females ( 43-44 ) . The body and total
lengths are 682 mm. and 690 mm., and 827 mm. and 824 mm. in two
males; 656 mm. and 626 mm., and 770 mm. and 751 mm. in two
females, respectively.

Remarks. — The snakes of this subspecies are robust and have a
relatively large head and short tail, which is slightly compressed
laterally. The number of ventrals does not exceed 150, whereas the
total range in the species is 122-160. The number of subcaudals is
the lowest in the species.

Distribution. — New Guinea (Fig. 24). Specimens examined, five,
as follows.

Indonesia: West New Guinea — Merauke, MCZ 22818-21 plus one head.

Genus Heurnia Jong

Hcumia Jong, Zoologische Anzeiger, 67( 12/13) :302, 1926 [Type-species. —
Heurnia ventromacuJata Jong, 1926]. Smith, Fauna of British India, Rep-
tilia and Ampliibia, 3:379, 1943.

Diagnosis. — Head distinct from body; head shields large; nasals
completely divided; internasal single, separating nasals; dorsal
scales smooth, in 27 rows; ventrals broad, not keeled; maxillary
teeth eleven, increasing in length posteriorly, last two grooved.

Heurnia ventromaculata Jong

Heurnia ventromaculata Jong, Zoologische Anzeiger, 67( 12/13) :302-3, 1926
[Holotype. — ZMA 11066, a male, from Pionierbivak, Mamberano River,
North New Guinea; Heurn collector].

Diagnosis. — As in genus.

Description. — The head is rather large, clearly distinct fiom the
body. The snout is as long as the parietal. The rostral is wider than

CoLUBRm Snakes, Subfamily Homalopsinae 171

high and visible from above. The large internasal completely sepa-
lates the smaller nasals, and is as long as the median suture of the
parietals. The nasals lie completely on the dorsal side of the head
and are completely divided. The nasal cleft extends laterally to the
first labial and medially to the internasal. A loreal is present only
on the left side (on the right it is fused to the prefrontal) and is
longer than high and does not border the internasal. The prefron-
tals are in contact with one another anteriorly for a short distance.
The left prefrontal is as long as wide; the right is fused with the
loreal, and in contact with the second, third and fourth labials.
There is one preocular on each side. The left side has three, and
the right has two postoculars. The frontal is long, hexagonal, wider
than the supraocular, as long as its chstance from the tip of the snout
and as long as the parietals. There are one primary and two sec-
ondary temporals. Of the seven upper labials, the fourth (fifth on
right side) enters the orbit. On the right side there are eight upper
labials, of which the sixth is divided longitudinally into three small
scales. Three lower labials are in contact with the anterior chin-
shields, which are longer than the posterior chin-shields; the scales
in the latter pair are completely separated by two pairs of smaller
scales. Of the 13 lower labials, the seventh is the largest. There are
160 complete ventrals and 62 subcaudals. The umbilical scar is
situated on the 21st ventral anterior to the vent. The type-specimen,
a male, has a total length of 730 mm. and a tail length of 145 mm.
The dorsal scales are smooth and in 27 rows. The scale reduction in
the type specimen is,

5+6(6) 6+7(98)

29 27 26 25 24

6+7(4) 6+7(30) 6+7(78) 6+7(91) 4+5(104)

21(160).

4+5(127)

The dorsum and sides are uniform grayish black. The first and
second rows of scales are yellow. The ventral surface is pale yel-
lo\\'ish white with dull grayish-black flecks. The upper lips are
white posteriorly, and the lower lips are white with black flecks.
The color immediately after capture was described by Mr. van
Heurn as follows: "Oberflache und Seiten uniform grauschwarz bis
bleigrau, vielleicht am besten zu definieren als schlammfarbig.
Unterseite elfenbein weiss mit matten grauschwarzen Flecken.
Oberlippe hinten weiss, Unterlippe weiss mit schwarzen Flecken."

Remarks. — This snake resembles the snakes of the genus Enhy-
dris in many characters, and the diagnostic character that separates

172 Unr'ersity of Kansas Publs., Mus. Nat. Hist.

it from Enhydris is its large internasal, which Hes between the
smaller nasals.

Distribution. — This species is known from New Guinea (Fig.
15). Specimen examined, one, from Indonesia, West New Guinea,
Mamberamo River, ZMA 11066.

Genus Myron Gray

Myron Gray, (in part). Catalogue snakes British Museum, p. 70, 1849 [Type-
species. — Myron richardsonii Gray, 1849]. Boulenger, Catalogue snakes
British Museum, 3:19, 1896. Rooij, Reptiles Indo-Australian Archipelago,
2:191, 1917. Worrell, Reptiles of Australia, p. 107, 1963.

Neospades (non Blackburn, 1887) De \'is, Proc. Roy. Soc. Queensland, 4:238,
1889 [T\-pe-species. — Neospades kcntii De \'is, 1889].

Diagnosis. — Head slightly distinct from neck, bearing large
shields; parietals entire; internasal single, separating nasals; loreal
present; dorsal scales slightly keeled, in 19 to 21 rows at midbody;
ventrals broad, not keeled; maxillary teeth about 10, followed by a
short interspace and a pair of enlarged, grooved teeth.

Myron richardsoni Gray

Myron richardsonii Gray, Catalogue snakes British Museum, p. 70, 1849 [Holo-
t\pe. — BMNH 1946.1.2.43 from North West Australia; collector Sir J.
Richardson]. Boulenger, Catalogue snakes British Museum, 3:20, 1896.
Loveridge, Bull. Mus. Comp. Zool. Harvard, 101:389, 1948. Glauert, A
handbook of the snakes of western Australia, p. 21, 1950.

Neospades kenlii De Vis, Proc. Roy. Soc. Queensland, 4:238, pi. 14, 1889

[Holot\Tpe. — Depository unknown from Cambridge Gulf, N. W. Australia;

collected by W. H. Saville-Kent].
Myron richardsoni; Rooij, Reptiles Indo-Australian Archipelago, 2:192, figs.

73-74, 1917. Kinghorn, The snakes of Australia, p. 91, 19.56. Worrell,

Reptiles of Austraha, p. 107, 1963.

Diagnosis. — Characters of genus.

Description and variation. — The snakes of this species are small;
the average total length of eight adult specimens is 401.6 mm. The
body is cylindrical but the tail is feebly compressed and short. The
head is slightly distinguishable from the body. The dome-shaped
rostral is as high as broad. The nasals are ovoid, semi-divided, and
smaller than the internasal. The nostril is a transverse slit; the nasal
cleft extending to the first or second labials. The internasal is a
single triangular plate (MCZ 38963 divided), lying between the
nasals, the apex in contact with the rostral, and the base in contact
with the prefrontals. The loreal is an elongate plate, which is in
contact \\ith the second and third (seven of twelve specimens) or
with the second to fourth (five of twelve specimens) labials. One
high or two small preoculars and two subequal postoculars are
present. The lower postocular extends half way below the eye.

CoLUBRiD Snakes, Subfamily Homalopsinae

173

The prefrontals are large, sciuarish plates, \\'hieh are in contaet with
each other (in RNHL 320 an azygous shield separates the prefron-
tals). The frontal is a pentagonal plate, narrowest anteriorly, about
twice as broad as the supraocular, as long as or a httle longer than
broad, as long as its distance from the snout, and shorter than the
parietals. A pair of postparietal scales are about four to six times
the size of the occipital scales. One primary, two (rarely three) sec-
ondary, and three (rarely four) tertiary temporals are present. The
upper secondary plate (Fig. 28) extends backwards as a long rectan-
gular shield and is fused with the postparietal on the right side in
BMXH 1935.7.6.2. Of the eight or nine undivided upper labials,
part of the fourth (or part of the fourth and fifth) enters the orbit;
usually the sixth is the largest. There are three pairs of chin-shields;
the first two pairs are almost the same size and are separated by a
deep mental groove. The third pair is smaller and separated by a
pair of small scales. The anterior pair is in contact with the first
three lower labials. Of the 11 lower labials (rarely 10 or 12), the
fifth (or sixth) is the largest. There are four or five gulars and one
(rarely two) incomplete ventrals between the chin-shields and the
first complete ventral. Four males have 132-139 (135.5) ventrals and
38-39 (38.5) subcaudals; eight females have 130-138 (135.6) ventrals
and 31-36 (.35.4) subcaudals. The anal plate is one and a half to two
times the length of a ventral. The umbilical scar is situated on or
between ventrals 20-25 anterior to the vent.

The dorsal scales are ovate, equal, striated, feebly keeled, and at
midbody in 21 or 19 rows, which reduce to 17 or 15 just anterior to
the vent. The dorsal scale reduction in one female (BMNH 1935.7.
6.2) is.

2+3(85)

19 i:

2+3(87)

2+3(135)
2+3(133)

15(137).

Fig. 28. Dorsal view of the head of Myron richardsoni (BMNH 1935.7.6.2),

Xl.2.

174 University of Kansas Publs., Mus. Nat. Hist.

The tail/ snout-vent ratios in four males are 18.6-19.7 (19.26) and
in eight females 12.9-18.5 (16.27).

The ground-color of the dorsum and sides is olive-brown or
yellow. The head has a pair of black rostral-orbital-postorbital
stripes, which extend from the rostral to the occipital region or
continue posteriorly for one or two head lengths. There are 33-47
(43) similar crossbars on the tail. Two or three outer rows of scales
and the venter are pale brown or yellow. The anterior edges of the
ventrals and subcaudals are black; these marks form a distinct
dusky median streak on the ventrals and a black streak on the
subcaudals.

In adult males the first three to four rows of dorsal scales are
prominently keeled in the anal region; the keeled scales extend
anteriorly from the vent for a distance of about 15-19 ventrals. The
males have slightly longer tails than the females.

Distribution. — This snake is confined to the southeastern limits
of the range of the subfamily along the coasts and rivers of New
Guinea and northern Australia (Fig. 15). Specimens examined, 12,
as follows.

Australia: North Australia, no definite locality, BMNH 1935.7.6.2.
Northern Territory — no definite locality, FMNH 97649; Karumba, 45 mi.
north of Newton, AMNH 86236; Port Essington, BMNH 55.10.16. Western
Australia — no definite locality, BMNH 1946.1.2.43. Indoxesia: Am — Kobroor,
Seltutti, SMF 19569. West Nciv Guinea— Wknaar, RNHL 4856; Birak Id.,
MCZ 38963, RNHL 320; North River, RNHL 320 (2). New Guinea: no
definite locality, FMNH 116769.

Genus Gerarda Gray

Gcrarda Gray, Catalogue snakes British Museum, p. 77, 1849 [Type-species. —
Gerarda bicolor Gray, 1849].

Campijlodon (not of Cuvier, 1832) Dumeril, Mem. Acad. Sci. France, 23:499,
1853. Dumeril and Bihron, Erpetologie generale . . . reptiles, Paris, 7:963,
1854 [Type-species. — Campijlodon prevostianum Dumeril and Bibron,
1854]. Jan, Arch. Zool. Anat. Phys., 3:263, 1865.

Heleophis Miiller, Verh. Natmf. Ges. Basel, 7:286, 1884 [Type-species. —
Heleophis flavescens Miiller, 1884].

Gerardia Boulenger, Fauna of British India . . . Reptilia and Batracliia, p.
379, 1890; Catalogue snakes British Museum, 3:20, 1896 [emendation].

Diagnosis. — Head slightly distinct from neck, bearing large
shields; nasals separated by internasal; loreal present; body moder-
ately long; scales smooth, in 17 rows at midbody; ventrals broad, not
keeled; maxillary teeth 10-12, small, followed by a pair of enlarged,
grooved teeth.

Gerarda prevostiana (Eydoux and Gervais)

Coluber (Homalopsis) prcvostianus Eydoux and Gervais, Geurin Mag. Zool.
Club, 3:5, pi. 15, 1837 [Holotype. — Depository unknown, from "Manila";
collector unknown].

CoLUBRiD Snakes, Subfamily Homalopsinae 175

Gciaida bicolor Gray, Catalogue snakes British Museum, p. 77, 1849 [Holo-
t>pe.— BMNH 1946.1.2.33, from "West Indies"; Mr. Luna's collection.
Locality is an error for "East Indies," Theobald (1868:58)]. Giinther,
Ann. Mag. Nat. Hist., ser. 4, 1:421. Theobald, Jour. Linn. Soc. Zool.,
10(41):58, 1868; Catalogue reptiles British India, p. 180, 1876.

C(H)ipijIodon picvostiaiuim; Dumeril and Bibron, Erpetologie gcnerale . . .
reptiles. Paris, 7:964, 1854. Jan, Arch. Zool. Anat. Phys., 7:263, 1865.
Jan and Sordelli, Iconographie generale des Ophidiens, liv. 30, pi. 6, fig. 3,
1868. Elera, Catalogo sistematico . . . Ophidiens, p. 432, 1895.

Hdiophis jiavescens Miiller, Verh. Naturf. Ges. Basel, 7:286, 1884 [Holotype. —
NMB 1771 from "East Indies"; G. and F. Miiller collectors].

Gcrardia prevostiaiia; Boulenger, Fauna of British India . . . Reptilia and
Batrachia, p. 379, 1890; Catalogue snakes British Museum, 3:20, 1896.
Haly, Jour. Asiatic Soc. Ceylon, 11:197, 1892. Phipson, Jour. Bombay Nat.
Hist. Soc, 9:486, 1895. Wall and Evans, Jour. Bombay Nat. Hist. Soc,
13:816. Wall, Jour. Bombay Nat. Hist. Soc, 16:307, 1905. Rosen, Ann.
Nhig. Nat. Hist. ser. 7, 15:175, 1905. Wall, Ophidia Taprobanica, p. 262,
1921; Jour. Bombay Nat. Hist. Soc, 30:171, 1923.

Gcioida prevostiana Smith, Bull. Raffles Mus., 3:62, 1930.

Diagnosis. — Characters of genus.

Description and variation. — The rostral is broader than high,
narrowly visible from above, and its posterior border is indented
medially. It is in contact with nasals and internasals posteriorly.
The nasals are ovoid, entire, and smaller than the internasal, which
is somewhat constricted in the middle and lies between the nasals.
The single loreal is slightly longer than high and slightly smaller
than the nasal; the loreal usually is in contact with the first three
upper labials. In EHT 33567 and 33559, each loreal is divided
longitudinally into two scales. The loreal is in contact with the
first two upper labials on one side in EHT 33554 and 33562 and
with the second and third upper labials in EHT 33568. In the
ocular region, the narrow supraocular is half the width of the
frontal; one high preocular, and usually two postoculars, the lower
of which extends half way below the orbit, are present. In BMNH
1937.9.7.1 and EHT 33567 there are three postoculars on the right
side. The prefrontals are large, slightly broader than long, form a
median suture, and in contact with the loreal, preocular and supra-
ocular. The holotype of Heleopliis flavescens (NMB 1771), a female,
has an extra pair of very small scales between the internasal and
prefrontals. The frontal is a pentagonal plate, a little longer than
broad, as long as or shorter than its distance from the end of the
snout. The frontal is shorter than the parietals. The number of
temporal plates varies from three to five on each side. Of 21 speci-
mens (42 sides) two primary and two secondary temporals occur on
36 sides; the remaining sides have one primary and three secondary
temporals. Of the 23 specimens examined, there are seven upper
labials in 65 per cent (30 sides). One specimen (BMNH 1908.6.23.73)

176 University of Kansas Publs., Mus. Nat. Hist.

has six upper labials on the right side. Usually no labials are di-
vided; the fourth upper labial enters the orbit. In those specimens
having eight upper labials the sixth is largest, the seventh the long-
est, and the eighth the smallest. The first four lower labials are in
contact with the anterior chin-shields, except the type (BMNH 1946.
1.2.33, a male) has the first three labials in contact with the anterior
chin-shields on both sides; one other specimen (EHT 33554, a male)
has the first five labials in contact with the anterior chin-shields on
the right side. There are two pairs of chin-shields; the anterior pair
is twice as long, and as broad as, the posterior pair. The anterior
pair of chin-shields form a median suture, and the scales of the
posterior pair are separated by a pair of small scales. Of the 23
specimens examined, there are eight lower labials in 67 per cent
(sides) and nine in 33 per cent (sides). There are four or five gular
scales and one or two incomplete ventrals between the chin-shields
and the first complete ventral. The umbilical scar is situated on or
between ventrals 22-25 anterior to the vent. Males have 145-154
(150.3) ventrals; females, 144-157 (148.0). Males have 32-36 (33.7)
subcaudals, and females have 29-34 (31.1). The dorsal scales are
smooth, ovate, and subequal; the maximum number of rows at the
level of the first ventral is 19 to 21, at midbody 17, and the minimum
number at the level of the third ventral anterior to the vent is 15 to
13. The dorsal scale reduction in one male (SU 12391) and in one
female ( ANSP 5090 ) are, respectively,

3+4(8) 4+5(102)
19 17 15(153).

3+4(8) 4+5(102)

4+5(6) 3+4(103) 6+7(130)

19 17 15 13(148).

3+4(8) 3+4(103) 6+7(136)

The 10-12 small maxillary teeth are followed by a pair of en-
larged, grooved teeth. The 20 mandibular teeth are subequal in
length; usually the anterior ones are the longest.

The hemipenis extends to the level of the 10th or 11th subcaudal,
and the point of bifurcation is at the level of the 5th or 6th sub-
caudal. The retractor penis magnus muscle originates on a caudal
vertebra at the level of the 24th or 25th subcaudal. The spines in
the area just above the naked basal region of the hemipenis are
small, few in number, and are not widely spaced.

The dorsal ground-color of the body is uniform dark olive or
gray. The three outer rows of dorsal scales are yellow or white
with olive or gray edges. The upper labials are yellow or white

CoLUBRiD Snakes, Subfamily Homalopsinae 177

with dark olive sutures; the lower labials, ehin, throat, and sides of
the neek are yellow or white. The rostral is dark olive or gray. The
\entrals and subeaudals are gray; each of the former is edged later-
all)- \\'ith \\ hite, whereas each of the latter is spotted centrally with
white.

The numbers of ventrals and caudals are about the same in
males and females, but the tail/ snout-vent ratio of males is higher
than that of the females. The tail/ snout-vent ratios in 12 males are
14.8-16.8 (16.0) and in nine females 13.9-15.6 (14.8); the coefficient
of difference between the means is 0.567 and is not significant.

Remarks. — The type locality of Coluber (Homalopsis) prevosti-
ana E^doux and Gervais (1837:5) is "Manila." One specimen
(ANSP 5090) supposedly came from "Philippines." The collectors
for these two specimens are unknown. No authentic records from
the Philippines are known.

Distribution. — This species is known (Fig. 11) from the coast of
India, Burma, Ceylon, Perak, and Thailand. The Thai specimens
were collected from mangrove swamps at Ang Hin (Taylor, 1965).
Specimens examined, 23, as follows.

No locality data, BMNH 1946.1.2.33. Burma: Inawaddij Division— Ran-
goon, BMNH 1908.6.23.73. Pegu Division— Vegu, BMNH 68.4.3.24, 68.4.9.6.
Ceylox: no specific locality, FMNH 121532. East Indies: no specific local-
ity, NMB 1771. India: Krusadai Island, SU 12.391. Maharashtra— A\ihag,
B'MNH 1937.9.7.1; Bandora, BMNH 95.1.3.1, 1956.1.31.1. Kerala State—
Malabar, Cananore, BMNH 1904.10.18.7. Thailand: Chan Biiri— Ang Hin,
EHT 335.54, 33557, 33559, 33561-2, 33564-8. Philippines: no specific local-
ity, ANSP 5090.

Genus Fordonia Gray

Homalopsis Schlegel, (in part), Essai sur la physionomie des serpents, 2:332,
1837.

Fordonia Gray, Zoological miscellany, p. 67, 1842 [Type-species. — Fordonia
leucobaha^Homalopsis leucohaUa Schlegel, 1837]; Catalogue snakes British
Museum, p. 76, 1849. Giinther, Reptiles of British India . . . Reptilia and
Batraclria, p. 378, 1890; Catalogue snakes British Museum, 3:21, 1896.

Hydropsis; Fitzinger, Systema Reptiliimi, p. 25, 1843.

Hemiodontus Dumeril and Bibron, Mem. Acad. Sci. France, 23:494, 1853;
Erpetologie generale . . . reptiles, 7:882, 1854 [Type-species. — Hemiodontus
leucohaUa, 1854]. Jan, (in part). Arch. Zool. Anat. Phys., 263, 1865.

Diagnosis. — Body stout; head not distinct from body; head
shields large; nasals separated by an internasal; loreal absent; scales
smooth, in 25-29 rows at midbody; ventrals broad, not keeled;
maxillary teeth six to eight.

Fordonia leucobalia (Schlegel)

Homalopsis leucohaUa Schlegel, Essai sur la physionomie des serpents, 2:345,
pi. 13, figs. 8-9, 1837 [Holotype.— RNHL 1161 from Timer; collector un-

178 University of Kansas Publs., Mus. Nat. Hist.

known]. Schlegel and Miiller, Verh. Nat. Nederl. Overz. Bezitt., p. 61, pi.
7, 1844. Cantor, Catalogue Malay reptiles, p. 102, pi. 40, fig. 5, 1847.

Fordonia leucohalia; Gray, Zoological miscellany, p. 67, 1842. Catalogue
snakes British Museum, p. 77, 1846. Boulenger, Fauna of British India . . .
Reptilia and Batrachia, p. 378, 1890; Catalogue snakes British Museum,
3:21, 1896; A vertebrate fauna of the Malay Peninsula . . . Reptilia and
Batrachia, p. 164, 1912. Sclater, Jour. Asiatic Soc. Bengal, 60:245, 1891;
List of snakes in Indian Museum, p. 55, 1891. Flower, Proc. Zool. Soc.
London, p. 888, 1896; Proc. Zool. Soc. London, p. 679, 1899. Boettger,
Katalog der Reptilien . , . Schlangen, 2:89, 1898. Wall and Evans, Jour.
Bombay Nat. Hist. Soc, 13:347, 1900. Lampe, Jahrb. Nassau Veriens, p.
31, 1902. Bordenger, Fasciculi Malayenses. Zoology^ 1:175, 1903. Annan-
dale, Jour. Asiatic Soc. Bengal, p. 176, 1905. Mocquard, Re\ue Coloniale,
p. 51," 1907. Barbour, Mem. Mus. Comp. Zool., 64(1):124, 1912. Despax,
Bull. Mus. Paris, 18:200, 1912. Boulenger, Trans. Zool. Soc. London,
20:264, 1914. Rooij, Reptiles Indo-Australian Archipelago, 2:189, fig. 71,
1917. Holzinger-Tenever, Arch. Naturg., 85A(2):104, 1919. Kloss, Jour.
Federated Malay States Mus., p. 201, 1921. Phisalix, Animaux venimeux
et \enins, 2:285, 1922. Taylor, The Snakes of the Philippine Islands, p.
115, 1922. Werner, Arch. Naturg., 89A(8):63. 1923. Rol:)inson and Kloss,
Jour. Federated Malay States Mus., p. 364, 1923. Sworder, Singapore Nat.,
2:66, 1923. Wall, Jour. Bombay Nat. Hist. Soc, 29:868, 1924; How to
identifv the snakes of India, p. 37, 1924. Smitli, Sarawak Museum Jour.,
3(2) :51, 1925; Bull. Raffles Mus., 3:62, 1930. Jong, Treubia, 7:87, 1926.
Bourret, Invent. Gen. Indocliine, 3:241, 1927. Brongersma, Treubia, 11:67,
1929. Kopstein, Treubia, 13(1 ):1, 1931; Treubia, 14(1 ):80, 1932.
Bourret, Serpents de I'lndochine, 2:299, 1936.

Fordonia tinicolor Gray, Catalogue snakes British Museum, p. 77, 1949 [Syn-
types. — BMNH 11 1.22.2. a, from Borneo?; Sir Belcher, collector; 111.22.2.b,
from Borneo; Lowe's collection]. Giinther, Reptiles of British India, p.
277, 1864; Zool. Record, p. 154, 1865. Theobald, Catalogue reptiles British
India, p. 182, 1876. Peters and Doria, Ann. Mus. Genova, 13:389, 1878.

Hemiodontus leucohalia; Dumeril and Bibron, firpetologie generale . . . rep-
tiles, Paris, 7:884, 1854. Jan, Arch. Zool. Anat. Phys., 3:264, 1865. Jan
and Sordelli, Iconographie generale des Ophidiens, liv. 28, pi. 6, fig. 1,
1868.

Hemiodontus chahjhacus Jan, Elenco systematico degli ofidi, p. 79, 1863. Jan
and Sordelli, Iconographie generale des Ophidiens, Hv. 28, pi. 6, fig. 3,
1868 [Holotype. — Milan Museum from Singapore; collector unknown].

Fordonia bicolor Theobald, Jour. Linn. Soc. Zool., 10:56, 1868 [Holotype. —
Depository unknown; from near Rangoon; collector unknown]; Catalogue
reptiles British India, p. 181, 1876.

Fordonia papuensis Macleay, Proc. Linn. Soc. New South Wales, 2:35, 1877
[Holotype. — Australian \Iuseum?, from Kataw, New Guinea; collector un-
known].

Fordonia variabilis Macleay, Proc. Linn. Soc. New South Wales, 2:219, 1878
[Syntypes. — Macleay Museum?, Sydney, three specimens from Port Danvin,
Australia; collector Mr. Spalding].

Diagnosis. — Characters of genus.

Description and variation. — The head is short and broad and has
a rounded snout. The mouth is subterminal. The rostral is broader
than high, and its posterior margin is arched outwards and is nar-
rowly visible from above. The nasal is oval, twice as long as the
internasal. The nostril, which is a lunate slit, has a nasal cleft
extending to the prefrontal. The internasal is single and lies
between the nasals. There is one preocular and two postoculars;

CoLUBRiD Snakes, Subfamily Homalopsinae 179

the lower extends partly under the eye. The prefrontals are small
and form a median suture. The frontal is pentagonal, narrow an-
teriorly, about four times as broad as the supraocular, as long as, or
a little longer than broad, as long as its distance from snout, but
shorter than the parietals. The temporal region is variable, but
usually four plates are present; there are one to two plates in the
primar\' row and one to three in the secondary row. Usually none
of the five upper labials is divided; the fifth is the largest. The third
upper labial enters the orbit, but in two specimens (AMNH 58418
and BMNH 1913.10.31.217-218) the second labial also enters the
orbit on both sides. The first three lower labials are in contact with
the first pair of chin-shields, which are small, subquadrangular, and
a little longer than the second pair. Both pairs of chin-shields form
median sutures. Four to seven gulars and one (rarely two) incom-
plete ventral separates the chin-shields from the first complete
ventral. The umbilical scar is situated on or between 20-23 ventrals
anterior to the vent. The ventrals are broad and entire, but in some
specimens (10 of 45) a few ventrals are divided longitudinally. Fif-
teen males have 141-153 (145.9) ventrals, and 22 females have 137-
159 (149.3). The subcaudals are paired; in some specimens (12 of 45)
a few subcaudals are entire. Fourteen males have 27-43 (39.4) sub-
caudals, and 22 females have 27-42 (32.2). The variation in the

TABLE 32. — Geographic Variation in Fordonia leticobalia
Locality Character No. Range Mean S.D.

Malaya Ventrals 4 146-150 148.0 —

Subcaudals 4 31-35 32.5 —

Tail/snout-vent 4 12.5-14.6 13.6 —

Cochin China — ._ Ventrals 1 — 150 — —

Subcaudals 1 — 32 — —

Tail/snout-vent 1 — 13.3 — —

Philippines Ventrals 2 145-148 146.5 —

Subcaudals 2 31-37 34.0 —

Tail/snout-vent 2 12.3-16.4 14.5 —

Borneo Ventrals 5 145-149 151.8 3.77

Subcaudals 4 34-42 36.7 3.78

Tail/snout-vent 4 12.8-16.4 14.1 1.54

Ceram Ventrals 3 141-146 143.7 —

Subcaudals 3 32-38 34.2 —

Tail/snout-vent 3 12.6-18.0 14.6 —

Timor Ventrals 2 — 151 151.0 —

Subcaudals 2 30-32 31.0 —

Tail/snout-vent 2 12.3-16.4 14.5 —

New Guinea Ventrals 12 143-155 149.6 3.87

Subcaudals 12 28-34 30.9 2.39

Tail/snout-vent 12 11.2-14.5 11.5 3.90

Australia Ventrals 5 143-147 145.2 1.79

Subcaudals 5 27-40 34.6 6.11

Tail/snout-vent 5 13.5-17.3 15.8 1.86

180

Unr^rsity of Kansas Publs., Mus. Nat. Hist.

^..iSiA. .,_;.jA_

jiH (ifclr -1*1

Fig. 29. Lateral view of the body of Fordonia leucobalia ( MCZ 22813), xO.7.

number of ventrals, subcaudals, and the tail/ snout-vent ratio for
eight locaHties is given in Table 32. The dorsal scales are smooth,
entirely attached to the interstitial skin, in 25-27 (rarely 29) rows
and reduce to 21 or 19 rows just anterior to the vent. The scale
reductions in one male (MCZ 49999) and two females (MCZ 22813
and MCZ 22815) are, respectively,

4+5(81) 5+6(116) 4+5(143)
25 23 21 19(153).

3+4(91) 3+4(108)

3 + 4(144)

4+5(88) 5+6(128)
25 23 21(150).

6+7(87) 5+6(123)

6+7(6)
25 27

5+6(120)

6+7(9)
5+6(154)

5+6(152)

5+6(127)
21(156).

5+6(131)

5+6(149)

The maxillary has an edentulous space anteriorly and extends
posteriorly beyond the palatine. The six to eight small maxillary
teeth are followed by a short interspace and two enlarged, grooved
teeth. The mandibular teeth are subequal, short, and blunt, widely
separated, and 10 to 13 in number. The hemipenis (in 7 specimens)
extends to a level between the 9th and 12th subcaudal, and the
point of bifurcation is at the level of the 4th or 5th subcaudal. The
retractor penis magnus muscle originates on a caudal vertebra at
the level of the 24th to 26th. The spines above the naked region of
the hemipenis are large and gradually become smaller distally
without differentiating into definite zones.

The color of the dorsum is variable, but the venter is uniformly

CoLUBRLD Snakes, Subfamily Homalopsinae 181

vellow or pinkish white. Five color patterns can be recognized from
the preserved specimens examined:

A. — Upper part of head and dorsal scales uniformly reddish brown;
one to four outer rows of scales pink or variegated with pink
and reddish brown; lower surface of head, ventrals, and sub-
caudals pink.

B. — Upper part of head dark brown, with variegated dark red and
pink blotches on occiput and lateral part of body to tip of tail;
one to four outer rows of scales, ventrals, subcaudals and lower
surface of head pale pink (Fig. 29).

C. — Upper part of head and dorsal rows of scales pale brown or
yellow; lower surface of head, one to four outer rows of scales,
ventrals, and subcaudals uniformly pale yellow or pale brown.

D. — Like pattern "C" above, but with dots or dashes across body.

E. — Upper part of head black, lower surface white; three to five
middorsal rows of scales black with irregularly dispersed
white spots; one to four outer rows of scales, ventrals, sub-
caudals white; lateral rows of scales variegated with black and
white.

Some specimens from Timor, northwestern Australia, and New
Guinea are variegated (pattern B); no specimens having this pattern
are known from north of the Equator. Except for the variegated
coloration, these specimens morphologically resemble other speci-
mens.

The numbers of ventrals and subcaudals do not differ signifi-
cantly in males and females. Males have slightly longer and more
compressed tails than females (Table 32). In males the tail is pro-
vided with a dorsal hump extending from the level of the fourth to
about the 26th subcaudal. In females the tail gradually tapers to a
point.

Remarks. — The holotype of Hemiodontus chahjbaeus Jan, 1863,
is probably based on an abnormal specimen, lacking an internasal
(Boulenger, 1890).

The holotv'pe of Fordonia papuemls Macleay, 1877, differs essen-
tially from Fordonia leucohalia in having 22 rows of dorsal scales
and no labials entering the orbit, which is completely surrounded
by the anterior and two posterior ocular shields. Macleay men-
tioned that several specimens were obtained at Katow, but no
variation was given in his account. The intervention of oculars be-
tween the eye and the labials probably is an individual anomaly,
and the count of 22 rows of dorsal scales may be an error, or the

182 UxrvTERSiTY OF Kansas Publs., Mus. Nat. Hist.

count might have been taken near to the \'ent, \\'here the ro\\'s re-
duce to 21 or 19.

Distribution. — The snakes of this species are known from
Burma, Cochin China, the PhiHppine Islands, Borneo, Ceram,
Timor, New Guinea and northern Austraha (Fig. 18). Specimens
examined, 45, as follows.

Australia: no definite localit>% AMNH 86239, 86241, BMXH 82.3.27.13-
14, 85.6.20.6. Northern Territory— 5 mi. west of Darwin, USNM 128234.
Queensland — Kaiumba, lower Norman Ri\er, AMXH 82442. British Borneo:
Sarawak— no definite localit>-, FMXH 67271, 71591-2; Niah, BMXH 92.9.2.3;
Sebnven, MCZ 18391. British New Guinea: Papua — Binaturi River, MCZ
49999; Dagwa, near Darn, AMNH 57952; Flv River, BMNH 86.5.20.21; 5 mi.
below Palmen junction, AMXH 57503; Port Moresby, FXHM 67271; Tarara,
AMXH 58417-8, 62991. Borneo: no definite locality, BMXH 111.22.2.a-b.
Cochin* China: no definite localit>', BMXH 85.3.3.11. Indonesia: Ceram —
no definite localit^^ BMNH 63.2.28.25-26; Piru, USNM 103524. West Borneo
— SMF 19571. West Neiv G(//)!Cfl— Miinika River, BMNH 1913.10.31.217-
218; Meranke, MCZ 22813-7; Turama River, UMMZ 80487. Malaya: Penang,
BMNH 60.3.19.1175, MCZ 941 (2), 15217. Philippines: Mindanao— no
definite locality', SMF 19570. Timor: no definite locality, RXHL 1161;
Atapupu, BxMNH 97.12.30.37.

Genus Cantoria Girard

Cantoria Girard, Proc. Acad. Philadelphia, p. 182, 1857 [Type-species. —
Cantoria violacea Girard, 1857]; U.S. Explor. Exped., Herpetologica, p. 156,
pi. 11, figs. 7-10, 1858. Giinther, Reptiles of British India, p. 278, 1864.
Boulenger, Fauna of British India, p. 380, 1890; Catalogue snakes British
Museum, 3:23, 1896.

Htjdrodipsas Peters, Monats. Akad. Wiss. Berlin, p. 270, 1859 [Type-species. —
Hi/drodipsas elapifonn is] .

Diognosis. — Body compressed, elongated; head depressed not
distinct from body; head shields large; nasals separated by inter-
nasal; loreal present; dorsal scales smooth, in 19 or 21 rows at mid-
body; ventrals narrow and not keeled; maxillary teeth nine to
eleven, followed by a pair of enlarged, grooved teeth.

Remarks. — Specimens of Cantoria violacea have more ventrals
(243-276), and subcaudals (56-69), as against one specimen of C.
anniilata which has 182 ventrals and 44 subcaudals. This is an
added example of reduction in the number of \entrals and sub-
caudals from west to east along the Sunda Arch for species and
subspecies within the same genus.

Distribution. — Cantoria violacea is kno\\'n from Burma, Malaya,
Sumatra, Borneo and Timor and C. annulata is known only from
West New Guinea.

Key to the species

Dorsal scales in 19 rows at midbody; prefrontals broadly in
contact; two yellow crossbars on head C. violacea

CoLUBRiD Snakes, Subfamily Homalopsinae 183

Dorsal scales in 21 rows at niidboch'; internasals, prefron-
tals and frontal meeting in a point in middle of snout; three
yellow crossbars on head C annulata

Cantoria violacea Girard

Cantoria violacea Girard, Proc. Acad. Pliiladelphia, p. 182, 1857 [Holotype. —
USNM 5523 from neighborhood of Singapore; collected by the U.S. Ex-
ploring Expedition]; U.S. Explor. Exped., Herpetologica, p. 156, pi. 11,
figs. 7-10, 1858. Cope, Proc. Acad. Philadelphia, p. 312, 1866. Lutk.,
\'idensk. Meddel. p. 151, 1866. Boulenger, Fauna of British India . . .
Reptilia and Batrachia, p. 380, fig., 1890. Sclater, List of snakes British
Museum, p. 55, 1891. Boulenger, Catalogue snakes British Museum, 3:23,
1896. Flower, Proc. Zool. Soc. London, p. 888, 1896; Proc. Zool. Soc.
London, p. 676, 1899. Wall and Evans, Jour. Bombay Nat. Hist. Soc,
13:612, 1901. Boulenger, Fasciculi Malayenses. Zoology, 1:175, 1903. Volz,
Zool. Jahrb. syst., 20:498, 1904. Boulenger, A vertebrate fauna of the
Malay Peninsula . . . Reptilia and Batrachia, p. 165, 1912. Wall, Jour.
Bombay Nat. Hist. Soc, 23:166, 1914. Rooij, Reptiles Indo-Australian
Archipelago, 2:191, fig. 72, 1917. Robinson and Kloss, Jour. Federated
Malay States Museum, p. 303, 1920. Phisalix, Animaux \enimeux et
venins, 2:285, 1922. Werner, Arch. Naturg., 89A(8):163, 1923. Wall,
How to identify the Snakes of India, p. 34, 1923. Sworder, Singapore Nat.,
2:66, 1923. Wall, Jour. Bombay Nat. Hist. Soc, 29:868, 1924. Bourret,
Invent. Gen. Indochine, 3:241, 1927. Smith, Bull. Raffles Museum, 3:62,
1930.

Hydrodipsas elapiformis Peters, Monats. Akad. Wiss. Berlin, p. 270, fig. 1,
1859 [Holotype. — Berlin Musermi from Sarawak; Haupt collector].

Hemiodontus elapiformis; Jan, Elenco systematico degli ofidi, p. 79, 1863;
Arch. Zool. Anat. Phys., 3:265, 1865. Jan and Sordelli, Iconographie
generale des Ophidiens, liv. 28, pi. 6, fig. 2, 1868.

Cantoria elongata Giinther, Reptiles of British India, p. 277, 1864 [Based on
Girard's specimen].

Cantoria elapiformis; Giinther, Zoological Record, p. 124, 1868.

Cantoria daijana Stoliczka, Jour. Asiatic Soc. Bengal, 39:208, pi. 11, fig. 5,
1870 [Holotype.— lost (Smith, 1943:.398); type locality', near Amherst,
mouth of Moulmein River, Burma]. Anderson, Proc Zool. Soc. London, p.
178, 1871. Theobald, Catalogue reptiles British India, p. 181, 1876.

Diagnosis. — Prefrontals broadly in contact; dorsal scales in 19
rows at midbodv; ventrals 243-276; two yellow crossbars on head
(Fig. 30).

Description and variation. — The body is compressed, and ex-
tremely elongate, and has a short, rounded snout and a short tail.
The mouth is subterminal. The rostral is broader than high, nar-
rowly visible from above; its posterior margin is pointed above; the
upper lateral margins are concave. The nasals lie on the dorsal
side of the head. The nasals are elongate plates and have a lunate
nostril in the posterior region. The nasal cleft is absent. The inter-
nasal is a single elongate plate lying between the nasals. The loreal
is squarish and is well separated from the internasal. There is one
high preocular, one elongate subocular which extends partly behind
the eye, and one small postocular. The prefrontals are large, as
large as the nasals, and form a median suture. The frontal is

184

Unr'ersity of Kansas Publs., Mus. Nat. Hist.

-^4*

Fig. 30. Dorsal view of the head of the holot>'pe of Cantoria violacea (USNM

5523), Xl.5.

hexagonal, broad anteriorly, and about four times as broad as the
supraocular, as long as broad, as long as its distance from snout, but
shorter than the parietals. There is one long primary temporal and
two secondary temporals. None of the five upper labials is divided;
the fourth and the fifth are the largest. The third and fourth upper
labials lie below the eye. There are eight lower labials; the anterior
pair forms a suture behind the mental. The two pairs of chin-
shields form a straight median suture; the anterior pair is as large
as the posterior pair, and is in contact with the first three or four
lower labials. The ventrals are smooth, narrow, and not quite three
times the width of a dorsal scale; the last ventral is divided. There
are four or five gulars and one incomplete ventral between the chin-
shields and the first complete ventral. The umbilical scar is situated
between ventrals number 47-49 anterior to the vent (BMNH 61.10.
11.53). Two males have 243 and 276 ventrals, and 69 and 63 sub-
caudals, respectively; one female has 267 ventrals and 56 sub-
caudals. The anal plate is about twice the length of the ventral.
The dorsal scales are smooth, entirely attached to the interstitial
skin, in 19 rows at midbody, and reduce to 17 rows just anterior to
the vent. The dorsal scale reduction in one female (BMNH 61.10.
11.53) and one male (USNM 5523) are, respectively,

3+4(258)

3+4(231)
3+4(231)

17 (266).

17 (275).

3+4(232)
The maxillary projects beyond the palatine and has nine to

CoLUBRiD Snakes, Subfamily Homalopsinae 185

eleven teeth follow ed b\ a pair of enlarged, grooved teeth. There
are six or seven palatine teeth, 14 to 15 pterygoid teeth, and 15 or 16
mandibular teeth, decreasing in size posteriorly.

The hemipenis extends to the level of the 12th to 15th subcaudal,
and the point of bifurcation is at the level of the 8th to 11th sub-
caudal. The retractor penis magnus muscle originates on a caudal
vertebra at the level of the 28th to 44th subcaudal. The stem of the
hemipenis is formed by a naked proximal part and distal spinous
part of equal length. The spinous part has six or seven rows of
spines which are progressively smaller distally. The two branches
of the hemipenis are beset with fine spinules of equal size through-
out the length of the branch. The sulcus is forked.

The general color is dull yellow (Fig. 31). The tip of the snout
is bluish black followed by a narrow yellow crossbar anterior to the
eye. Immediately behind this is the bluish black crossbar (including
the eyes, frontal and fourth upper labials), which is followed by an
interrupted yellow crossbar. From behind the angle of the mouth
across the occiput is another black crossbar. There are 49 to 60
yellow crossbars on the body which are separated on the back by
narrow interspaces becoming wider on the sides. On the posterior
part of the body some of the bars are confluent. Twelve to 18
black rings encircle the tail and occasionally are fused above and
below. The throat and venter are yellow with faint black bands
corresponding to the dorsal bands. There is another color phase
which is dark brown above, with narrow white crossbars; the outer
rows of scales and belly are white; the head has white spots. The
above coloration of Hemiodontus elapifonnis resembles Jan and
Sordelli (1868: pi. 6, fig. 2).

The last two ventrals in males are shorter, and the last ventral
in females is longer, than the other ventrals. Two males (BMNH
1937.9.7.5, USNM 5523) have more subcaudals and less ventrals

Fig. 31. Lateral view of the head of the holot>'pe of Cantoria violacea (USNM

5523), XO.9.

186 University of Kansas Publs., Mus. Nat. Hist.

than the female specimen (BMNH 61.10.11.53). The tail/ snout-vent
ratio in two males and one female are 16.7, 19.9 and 14.1, respec-
tivelv.

Dlstribtition. — This species is found in tidal rivers and coasts of
Burma and the Malay Peninsula (Fig. 18) and also in Andaman
Islands, Sumatra, Borneo and Timor. Specimens examined, three,
as follows.

No locality, BMNH 61.10.11.53. Malaya: Singapore — neighborhood of
Singapore, USNM 5523. India: Andamans— Ron Island, BMNH 1937.9.7.5.

Cantoria annulata (Jong)

Cantoria annulata Jong, Zool. Anz., 67( 12/13) :304, 1926 [Holotype.— ZMA
11065 a female, from Prins Hendrik-eiland, New Guinea; Heiirn collector].

Diagnosis. — Prefrontals, internasal, and frontal meeting in a
point in middle of snout; dorsal scales in 21 rows at midbody; ven-
trals 182; three vellow crossbars on head.

Description. — The body is compressed, elongate, and has a
short, round snout and a short tail. The mouth is subterminal. The
rostral is a little broader than high and is evident from above. The
nasals lie on the dorsal side of the head and are not completely
divided. The nasal cleft begins at the internasal; the latter com-
pletely separates both nasals. The prefrontals are short, hexagonal,
just as wide as long. The frontal is longer than broad, not so long as
its distance to the tip of the snout, and shorter than the parietals.
Both prefrontals, the internasal, and the frontal join together in a
point in the middle of the snout. The loreal is just as long as high
and contacts the first, second, and third upper labials, the nasal, the
prefrontal, and two preoculars. The eye is small and has a vertically
elliptical pupil. The eye is bordered by two preoculars and two
postoculars; the lower postocular on the right extends under the
eye. There are three temporals, one anterior and two posterior;
the anterior is not in contact with the supraocular, and the upper
posterior is twice as high as the lower one. The supraocular is more
than half as broad as the frontal. None of the eight upper labials
are divided, and the fourth narrowly enters the orbit. Of the nine
lower labials, the first three are in contact with the anterior pair of
chin-shields. The posterior chin-shields are larger than the anterior
and are in contact with one another. The dorsal scales are smooth,
entirely attached to the interstitial skin, in 21 rows at midbody,
and reduce to 17 just anterior to the vent. The ventrals are smooth,
narrow, and are not quite three times the \\ idth of a dorsal scale in
the first row. The umbilical scar is situated on the 78th ventral
anterior to the vent. The anal plate is divided. The holotype (ZMA

CoLUBRiD Snakes, Subfamily Homalopsinae 187

11065), a female, has 182 ventrals and 44 subcaudals, a total length
of 625 mm., and a tail length of 75 mm.

The dorsum and sides are brown. Three yellow crossbars are
present on the head, all originate from the angle of the jaw. The
first traverses the upper labials and continues dorsally across the
nasal region to meet its fellow at the rostral; the second extends
anteriorly, diagonally over the temporals. The third crossbar ex-
tends straight across the occipital region. There are 61 yellow
crossbars on the body and 14 on the tail. These crossbars extend to
the second or third outer rows of scales and are not completely one
scale in length. The second and third outer rows of scales are pale
brown; the color merges into the yellow of the ventrals.

There are 11 maxillary teeth; the last two are enlarged and
grooved.

Distribution. — This species is known only from Prins Hendrik-
eiland. Specimen examined (Fig. 18), one, from Indonesia, West
New Guinea, Prins Hendrik-eiland, ZMA 11065.

Genus Bitia Gray

Bitia Gray, Zoological miscellany, p. 64, 1842 [Type-species, Bitia liydroides
Gray, 1842].

Hipistes Gray, Gatalogue snakes British Museum, p. 77, 1849 [Type-species,
Hipistes fasciatus Gray, 1849].

Diagnosis. — Head not distinguishable from neck; parietals frag-
mented; head, neck, and anterior third of body narrow, gradually
becoming larger towards vent; internasal single, separating nasals;
loreal present; dorsal scales smooth, in 37-43 rows at midbody;
ventrals narrow, about half width of body, having two strong lateral
keels; maxillary teeth 11-13, followed by short diastema and pair of
enlarged, grooved teeth.

Bitia hydroides Gray

Bitia hydroides Gray, Zoological miscellany, p. 64, 1842; Catalogue snakes
British Museum, p. 63 [Holotype. — BMNH lll.lS.l.a; t>'pe locality un-
known, collector unknown].

Homalopsis hijdrina Cantor, Jour. Asiatic Soc. Bengal, 16(2): 104, pi. 40,
fig. 4, 1847 [Holotype. — Calcutta Museum?, from sea ofF Kedah Coast,
Malay Peninsula; collector unknown].

Hipistes fasciatus Gray, Catalogue snakes British Museum, p. 78, 1849 [Holo-
type.— BMNH 111. 32. La, from East Indies; collector unknown].

Hipistes hijdrinus, Giinther, Reptiles of British India, p. 287, 1864. Stoliczka,
Jour. Asiatic Soc. Bengal, 29:207, 1870. Anderson, Proc. Zool. Soc. Lon-
don, p. 181, 1871. Theobald, Catalogue reptiles British India, p. 184, 1876.
Boulenger, Fauna of British India . . . Reptilia and Batrachia, p. 382, 1890;
Catalogue snakes British Museum, 3:24, 1896; A vertebrate fauna of the
Malay Peninsula . . . Reptilia and Batrachia, p. 166, 1912. Sclater, List
of Snakes in Indian Museum, p. 56, 1891. Flower, Proc. Zool. Soc. Lon-
don, p. 888, 1896; Proc. Zool. Soc. London, p. 679, 1899. Wall and Evans,

188 University of Kansas Publs., Mus. Nat. Hist.

Jour. Bombay Nat. Hist. Soc, 13:347, 616, 1900. Boiilenger, Fasciculi
Malayensis Zoology, 1:175, 1903. Wall, Jour. Bombav Nat. Hist. Soc,
17:54, 1906; jour. Boml^ay Nat. Hist. Soc, 18:230, 1908' Smith, Jour. Nat.
Hist. Siam, 1(1):103, 1914; Jour. Nat. Hist. Siam, 1(3):214, 1914. Rooij,
Reptiles Indo-Australian Archipelago, 2:194, 1917. Phisalix, Animaux veni-
meux et venins, 2:23, 1922. Robinson and Kloss, Jour. Federated Malay
States Museum, p. 364, 1923. Wall, How to identify the snakes of India,
p. 38, 1923. Sworder, Singapore Nat., 2:66, 1923. Werner, Arch. Natiug.,
89A(8):164, 1923. Wall, Jour. Bombay Nat. Hist. Soc, 29:868, 1924.
Smith, Bull. Raffles Museum, 3:62, 1930. Smedlev, Bull. Raffles Museum,
5:53, 1931.

Diagnosis. — As in genus.

Description and variation. — The head is small; the eyes are
dorsal and the mouth is subterminal. The rostral is a dome-shape
plate a little broader than high, and slightly visible from above.
The nasal is almost, or completely, divided transversely into ante-
rior and posterior parts by a nasal cleft; the external cleft extends to
the loreal and the internal cleft to the internasal. The two nasals
are separated from each other by an internasal. The single loreal
is in contact with the first three (rarely with second and third) upper
labials. The supraocular is narrower than the frontal. There is one
high preocular which extends half way below the eye of two post-
oculars, the upper is much smaller than the lower, which extends
below the eye and meets the preocular. The prefrontals form a
median suture. The frontal is an elongate plate about two to two
and one-half times longer than broad, as long as, or a little longer,
than its distance from the end of the snout. The parietals are
fragmented into small plates. The single primary temporal is
followed by two secondary and three tertiary temporals. Of the
seven upper labials, the fourth is below the eye, and the fifth and
sixth are the largest. There are two pairs of chin-shields; the ante-
rior pair is about two to two and one-half times longer than the
posterior pair. The anterior chin-shields are in contact with each
other and with the first five lower labials. The scales in the posterior
pair of chin-shields are separated by a pair of small scales. The
number of lower labials ranges from 10 to 12; the sixth is the largest.
There are eight or nine gulars and one or two incomplete ventrals
between the chin-shields and the first complete ventral. The dorsal
scales are smooth, elongate, entirely attached to the interstitial
skin, and have a gap between the base of the scale and the apex of
the preceding scale. A similar condition is found in the Xenoder-
minae (Smith, 1943:400). The maximum number of rows of dorsal
scales (37 to 43) occur at midbody; the minimum number (31 to 33)
is just anterior to the vent. Six males have 157-165 (160.5) ventrals
and 31-34 (32.7) subcaudals, and six females have 160-163 (160.5)

CoLUBRm Snakes, Subfamily Homalopsinae 189

and 24-27 subcaudals. The anal is divided and is about twice the
length of a ventral.

The tail/ body ratios for two males are 11.3 and 13.2 and for
three females, 8.5, 8.6, and 8.8.

The ground-color is pale brownish or grayish yellow above and
pale yellow below. There are 40 to 42 dark gray or black crossbars
having parallel margins on the body, and eight or nine similar
crossbars on the tail. The crossbars on the body are three scales in
length and extend laterally to the seventh or eighth row of scales.
The crossbars are separated by two or three scales. The head is
dark brown or gray with one or two small dark brown circular spots
on each shield. The upper labials, and the scales in the occipital
region and on the neck are likewise spotted. The lower surface of
the head, chin, first seven rows of scales, ventrals, and subcaudals
are pale yellow. Wall and Evans (1900) described the snake in life
as having ". . . alternate yellow and black dorsal bars, the belly buff.
The colors of the back are bright and the scales glazed like enamel."

Remarks. — The small head; narrow anterior third of the body;
narrow ventrals and subcaudals; and compressed tail resemble the
characteristics of members of the family Hydrophidae.

Distribution. — Specimens are known (Fig. 18) from mouths of
rivers and coasts of lower Burma, Thailand, and the Malav Penin-
sula. Specimens examined, five, as follows.

Xo specific locality: BMXH 1946.1.2.38, 1946.1.2.50. East Indies: no
definite locality, NMB 1774. Malaya: Singapore — Nee Soon Jungle, MCZ
8984, 18392.

Additional records. — Burma: Pegu Division — Pegu (Smith, 1943:25).
Malaya: Penang (Smith, 1943:25). Thailand: South Thailand — Bangkok
(Smith, 1943:25).

Genus Erpeton Lacepede

Erpeton Lacepede, Bull. Sci. Soc. Philom. Paris, 2(46) :169, 1800 [Type-species
Erpeton tentaculatus Lacepede, 1800]; Ann. Mus. Hist. Nat. Paris, 2:280,
1803. Gray, Catalogue snakes British Museum, p. 62, 1849. Dumeril and
Biliron, Erpetologie generale . . . reptiles. Paris, 7:983, 1854.

Rhinopirus Merrem, Tentamen systematis amphibiorum, p. 81, 1820 [substitute
name for Erpeton].

Homalopsis Schlegel, (in part), Essai sur la physionomie des serpents, 2:332,
1837.

Herpeton; Wagler, Natiirliches system der amphil)ien, p. 169, 1830 [Emenda-
tion]. Giinther, Proc. Zool. Soc. London, p. 115, 1860; Ann. Mag. Nat.
Hist., ser. 3, 7:266, 1861; Reptiles of British India, p. 288, 1864. Jan,
Arch. Zool. Anat. Phys., 3:255, 1865. Boulenger, Catalogue snakes British
Museum, 3:25, 1896.

Diagnosis. — Head small, compressed with a pair of long scaly,
tentacle-like, rostral appendages; head distinct from narrow neck;
head shields large, intercalated with small scales; nasals separated

190 University of Kansas Publs., Mus. Nat. Hist.

by a series of small scales; loreal region with many small scales;
body narrow anteriorly, gradually becoming larger posteriorly,
tapering just anterior to vent towards tip of tail; dorsal scales
strongly keeled in 35 or 37 rows at midbody; ventrals narrow with
two closely set keels; subcaudals not well differentiated; maxillary
teeth 12 to 14, followed by a short interspace and a pair of slightly
enlarged, grooved teeth.

Erpeton tentaculatus Lacepede

Erpeton tentaculatus Lacepede, Bull. Sci. Soc. Philom. Paris, 2 (46): 169, 1800
[Holotype. — Paris Museum; from the old Dutch collection; locality un-
known]; Ann. Mus. Hist. Nat. Paris, 2:280, 1803. Daudin, Histoire naturelle
generale . . . reptiles, 7:246-253, pi. 86, 1803. Gray, Catalogue snakes
British Museum, p. 63, 1849. Dumeril and Bibron, Erpetologie generale
. . . reptiles. Paris, 7:984, 1854. Cochran, Proc. U.S. Nat.^ Mus., 77:31,
1930. Bourret, Bull. Instr. Pub., p 80, 1934; Serpents de I'lndochine, p.
305, fig. 123, 1936.

RJii)iO])irus tentaculatus; Merrem, Tentamen svstematis amphibiorum, p. 82,
1820.

Homalopsis hcrpeton; Schlegel, Essai sur la physionomie des serpents, 2:359,
1837.

Herpcton tentaculatiim; Jan and Sordelli, Iconographic generale des Ophidiens,
p. 1, pi. 1, 1860. Giinther, Proc. Zool. Soc. London, p. 114, color pi. 23,
1860. CornaHa, Rev. and Mag. Zool. ser. 2, 13:145, 1861. Peters, Monats.
Akad. Wiss. Berlin, p. 902, 1861; Monats. Akad. Wiss. Berlin, p. 247, 1863.
Gunther, Reptiles of British India, p. 288, 1864. Bocourt, N. Arch. Mus.
Bull., 2:6, 1866. Morice, Ann. Sci. Nat. Paris, ser. 6, vol. 2, art. 5, pi. 20,
1875; Sur les habitudes du remarquable Serpent de la Cochinchine, I'Her-
peton tentaculatum. Lyon, 1875; Jour. Bombay Nat. Hist., 15:59, 1875.
Peters, Sitzb. Ges. Naturf. France, p. 74, 1882. Tirant, Notes sur les Rep-
tiles de Cochinchine et du Cambodge, 3(20);404, 1885. Boulenger, Cata-
logue snakes British Museum, 3:25, 1896. Flower, Proc. Zool. Soc. London,
p. 680, 1899. Mocquard, Revue Coloniale, p. 51, 1907. Ditmars, Reptiles
of the World. London, p. 280, 1910. Smith, lour. Nat. Hist. Siam,
1(2):103, 1914; Jour. Nat. Hist. Siam, 1(3): 187, 1915. Annandale, Jour.
Nat. Hist. Siam, 2(2):91, 1916. Gyldenstolpe, Kungl. Vet. Akad. Stock-
holm, 55:19, 1916. Kloss, Jour. Federated Malay States Museum, p. 201,
1921. Phisalix, Animau.x venimeux et \enins, 2:285, 1922. Werner, Arch.
Naturg., 89A(8):164, 1923. Bourret, Invent. Gen. Indochine, 3:241, 1927.
Smith, Bull. Raffles Mus., 3:63, 1930. Bourret, Bull. Instr. Pub., p. 23,
September, 1934. Smith, Faima of British India . . . Reptilia and Am-
phibia, 3:401, fig. 130, 1943.

Diagnosis. — Characters of genus (Fig. 32).

Description and variation. — The head is small, narrow and angu-
lar anteriorly. The rostral is nearly twice as broad as high, and the
anterior and posterior margins are notched mesially. Behind the
rostral is a median hexagonal or pentagonal postrostral scale which
is flanked on either side by a single or double row of five or six small
scales, completely separating the rostral from the nasals. The latter
plates are ovoid, large, and raised, with a lunate slit — the nostril.
The nasals are broadly in contact (BMNH 1921.4.1.37 & MCZ
8887) with each other or partly (MCZ 4070, EHT 111281, USNM

CoLUBRiD Snakes, Subfamily Homalopsinae

191

67500) or completely separated (USNM 6016) by a series of small
scales. The nasal is partially divided by a nasal cleft which extends
laterally from the nostril to the edge of small scale. The internasal
is di\ided longitudinalh' into two equal scales. An azygous scale
lies completeh' between them and the prefrontals, which are as
large as the nasals. The prefrontals form a median suture and are
separated from the supraocular by a large scale which in some is
di\"ided into two scales. The frontal is a large hexagonal plate,
slighth' longer than broad, shorter than the parietals, and much
broader than the supraocular. The frontal is separated from the
supraocular by a row of two to four small scales. The parietals are
entire and form a median suture immediately posterior to the
frontal and narrowly bordering the supraocular. The loreal region
has four to eight small scales. The eyes are small, raised, and have
vertically elliptical pupils. They are usually bordered by one high
preocular and two postoculars, the upper of which is half the size
of the lower. Four or five suboculars, together with the scales
anterior and posterior to these, form a series of scales parallel to and
above the upper labials. The temporals are small, scale-like, and
strongly keeled. There are 13 to 15 upper labials, the eighth and
ninth, or eighth to tenth, upper labials lie below the eye. The
margin of the upper lip forms an angle at the second or third to the
last labial. Three pairs of chin-shields, progressively larger laterally,
are situated in transverse rows; the chin-shields are separated from
the lower labials by a row of small scales parallel to the lower
labials. The dorsal scales are strongly keeled and usually in 35 or
37 (rarely 39) rows, which reduce to 29 or 27 rows posteriorly. The
ventrals are narrow, about twice as wide as the adjacent scales and

Fig. .32. Dorsal view oi Erpeton tentaculattis (BMNH 1921.4.1.37), XO.5.

192 Unwersity of Kansas Publs., Mus. Nat. Hist.

are bicarinated. The last ventral is about twice as broad as the pre-
ceding one. In one specimen (BMNH 1921.4.1.37) the last ventral
has two pairs of keels. Six males have 108-114 (109.8) and three fe-
males 111-133 (123.3) ventrals. The subcaudals resemble the caudal
scales; five males have 94-126 ( 102.2) and two females have 87 and
94 subcaudals. The anal plate is almost three times broader than the
second to the last ventral, and is bilobed (trilobed in USNM 6016)
and feebly keeled with a row of small scales on the posterior rim
and two to three rows on the lateral rim of the vent. The rostral
appendage is covered by ten or eleven rows of scales in its length
and seven or eight row^s of scales around its base, fewer distally.
The umbilical scar, which has the appearance of a pair of dorsal
scales occurs on 13 to 15 (EHT 111281), 16 to 18 (MCZ 4070, USNM
67500), 18 to 20 (BMNH 1921.4.1.37, MCZ 8387, USNM 6016, 53442)
ventrals anterior to the vent.

The maxillary bone does not extend as far anteriorly as the
palatine, and has 12 to 14 teeth followed by a short interspace and a
pair of slightly enlarged, grooved teeth; the mandibular teeth are
subequal.

The hemipenis (six specimens) extends to the level of the 10th
or 11th subcaudal, and the point of bifurcation is at the level of the
5th subcaudal. The retractor penis magnus muscle originates on a
caudal vertebra at the level of the 33rd to 35th subcaudal.

There are two color phases:

A. — Brown striped form: The ground-color is brown with dark
longitudinal stripes occupying at midbody the following rows
of scales: first to part of fourth, brown; fourth to part of sixth,
dark brown or black; sixth and part of seventh, brown; seventh
to part of tenth, dark brown or black; tenth to part ot four-
teenth, brown; fourteenth to part of sixteenth, dark brown or
black and sixteenth to vertebral row of scales, biown. Ten to
13 pairs of white and black cross streaks about one to two
scales in length extend from the lateral edges of ventrals to the
fifth or sixth row of dorsal scales. The streaks are five to seven
ventrals apart, distinct on the anterior half of body, and faded
posteriorly. A distinct black stripe extends from the rostral
appendage to the eye and posteriorly on to the neck.

B. — Black, variegated form: The whole animal is black, variegated
and mottled lavender brown, and lacks longitudinal stripes. A
vertebral series of large, and lateral series of smaller, gray-
lavender blotches are present. Eleven to thirteen pairs of white

COLUBRID SXAKES, SUBFAMILY HOMALOPSINAE 193

and black cross streaks are present, followed by 10 to 12 pairs
of white and black spots to the vent. A black stripe extends
from the base of the rostral appendage to the eye and posteri-
orly on to the neck. White flecks are present on the anterior
angles of the lower lip and just below the angle of the mouth.

Females have more ventrals, and males have more subcaudals
and longer tails. The tail/ body ratio in seven males is 31.7-45.3
(41.1) and in five females is 27.2-39.2 (34.3), but the differences are
not statistically significant.

Remarks. — This species is unique among the fresh water snakes
in ha\ing the greatly reduced ventrals and a pair of rostral ap-
pendages. On the posterior part of the body one ventral may
correspond to one and a part of another or two rows of the dorsal
scales, which is contrary to the colubrid pattern of one ventral to
one dorsal row of scales.

Distribution. — This species is not uncommon in ponds and slug-
gish waters in Thailand around Bangkok. Specimens have been
collected from Cambodia and Cochin, China (Fig. 6). Annandale
obtained it in the inland sea at Siggora; this is the southern-most
locality. Specimens examined, 17, as follows.

Cochin- China: Long Xuyen, BMNH 1920.1.20.2644; Tayninh, MCZ
4070. Thailand: no definite locality, AMNH 8850, BMNH 1920.1.20.1876,
BMXH 59.1.1.11-12, USNM 6016. South Thailand— Bangkok, BMNH
1915.5.13.9-10, 1912.4.1.37, 1921.5.4.10, 97.10.8.31, 1921.5.4.10; SMF 32662,
59353; USNM 67500, 53442, 6016.

Osteology

For osteological study I have had available 20 skulls and six sets
of dentigerous bones representing eight species of four genera. The
terminology of the cranial elements is that used by Parker ( 1878 ) ,
Szunyoghy (1932), Radovanovic (1937), Duellman (1958) and
Wellman (1963); for vertebral terminology I have followed Gadow
( 1933 ) , Sood ( 1948 ) , and Romer ( 1956 ) .

The nominate genus, Homalopsis, was selected for a detailed
study of the osteology of the skull; the skulls of other genera are
compared with H. huccata. The following description is based pri-
marily on the skull of a female specimen (KU 92355) of H. huccata.

The elements are discussed in the following order: nasal region,
cranium and associated elements, maxillo-palatal-pterygoid arch,
mandible, dentition, and vertebrae.

Nasal region. — This region consists of the fused premaxillaries
and paired septomaxillaries, vomers and nasals. All are firmly joined
to each other.

194 University of Kansas Publs., Mus. Nat. Hist.

The premaxillary is relatively heavy, and has two anterodorsal
concavities and a postero\entral concaxity. A foramen passes
through the posteroventral concavity to the medial anterodorsal
concavity. The lateral processes of the premaxillaries slope slightly
upward and curve posteriorly. The dorsal spine is concave and
extends to the common suture between the nasals. The posterior
processes of the premaxillaries are short and bifurcated; thus, the
line of fusion with the nasals and septomaxillaries is broad. The an-
terior processes of the septomaxillaries are broad and lie below the
nasals and border the ventral edge of the nasal septum for its entire
length. A lateral wing extends in an arch from the lateral edge of
the septomaxillary to a point lateral to the nasal plate. The broad
medial part of the septomaxillary forms the roof and anterior border
of the vacuity of Jacobson's organ. Vomers form the posterior and
lateral boundaries of the vacuity. The thin postero-inferior process
of the \'omer encloses the vomerine fenestra. The postero-superior
process of the vomer is a thin sheet of bone lying medially to the
posterior process of the septomaxillary. The nasals are relatively
flat above, each curves slightlv downward mediallv and fuses into
the medial nasal septum. Laterally, each nasal is deflected down-
ward and forms a small dorsal shield over the nasal cavity; anteri-
orly, the nasals fuse with the dorsal spine of the premaxillary. The
nasal septum is short and limited to the posterior part of the nasal
area.

Cranium and associated elements. — The frontal is about one and
one-half times as long as it is wide; it is flat above with an emargin-
ate dorsolateral margin that forms the upper limit of the optic cap-
sule. Ventrally the frontal is concave and forms the median limits of
the optic cavity. Farther ventrally and medially the frontal articu-
lates with the parasphenoid. The frontal forms the anterior border
of the optic foramen, and the roof of the frontal and its septomaxil-
lary process surround the olfactory fenestra. The prefrontal is
sutured to the anterolateral corner of the frontal; it provides partial
roofing for the nasal ca\ ity and forms the posterior boundary of that
cavity and the anterior edge of the orbit. A curved nasal process
consisting of a narrow spine of bone extends dorsomedially from the
anteroventral edge of the prefrontal. The orbital-nasalis foramen
is located in the anterior surface of the prefrontal. The parietals are
fused into a large bone that forms the roof and sides of the middle
part of the cranial cavity. The dorsal surface of the parietal is flat;
this area is delimited by the parietal crests which extend from near
the anterolateral corners to a median point near the posterior mar-

CoLUBRiD Snakes, Subfamily Homalopsinae 195

gin of the parietal. A posterior extension ot the parietal erests forms
the suprateniporal crest on the posterior part of the parietal and on
the supraoecipital. The postfrontals are broad, flat, triangular plates
sutured to the anterolateral processes of the parietal. Together the
anterior surfaces of these two bones form the posterior rim of the
orbit. Dorsally to \cntrally the posterior edge of the parietal is
sutured to the supraoecipital, the prootic, and the basisphenoid.
Anteromedially, the parietal forms the roof and posterior margin of
the optic foramen. The basisphenoid, which is fused with the para-
sphenoid, also forms a small part of the posteroventral margin of
the optic foramen. The basisphenoid forms the floor of the middle
part of the cranial cavity and the ventromedial depression that con-
tains the pituitary body. Posterolateral to the parietal and dorsal to
the posterior part of the basisphenoid is the prootic. This bone is
deeply emarginate posteriorly, and forms a large part of the otic
notch, through which the columella passes. The columella is a long,
thin bony rod that terminates posteriorly in cartilage. There are
se\eral foramina on the lateral surface of the prootic. On the
anterolateral surface of the prootic, branches of the trigeminal
ner\'e pass through three foramina whereas the facial nerve passes
through the single posterior foramen near the otic notch. The squa-
mosal is heavy, flat and broad anteriorly, and curves slightly in a
dorsal direction throughout its length; it becomes thinner and nar-
rower posteriorly. It is attached dorsoventrally to the posterior part
of the parietal and to the lateral part of the prootic. At this place of
attachment there is on the prootic a broad depression that forms an
articulating base. The quadrate articulates with the dorsolateral
surface of the posterior part of the squamosal. The quadrate is a
thin, flat, rectangular bone. About there-fourths of the way down
its medial surface is the columellar process, the point of fusion of
the columella. Dorsal to this process is a supracolumellar crest, and
v^entral to the process is an infracolumellar crest. Both crests are on
the medial side of the quadrate. A diagonal medial ridge passes
from the columellar process dorsally across the slightly depressed
medial surface of the quadrate and terminates at the articular sur-
face at the dorsal end. This diagonal ridge is presumably to
strengthen the quadrate. The posterior part of the cranium consists
of a median ventral basioccipital, a median dorsal supraoecipital,
and paired dorsolateral exoccipitals. Prominent on the ventral sur-
face of the basioccipital is a pair of posteriorly projecting tubercles,
the pterygoid processes. Posteriorly the basioccipital forms the
ventral part of the occipital condyle. The rest of the condyle, on

196 Unrtersity of Kansas Publs., Mus. Nat. Hist.

each side, is formed by the exoccipitals. The exoccipitals extend
around the sides of the foramen magnum and meet dorsally. They
bear moderately high occipital crests that extend posterolaterally
across the supraoccipital as branches from the supratemporal crest.
The supraoccipital also has a medial crest that extends a short dis-
tance posteriorly from the supratemporal crest onto the exoccipitals
at their dorsal line of fusion ( Fig. 33) .

Maxillo-paJatal-ptenjgoid arch. — Anteriorly the maxillary begins
just posterior to the tip of the lateral process of the premaxillary.
The maxillary is a stout bone which forms an arch, convex laterally.
On the anterior third of the lateral surface, adjacent to the region
where the broad prefrontal comes in contact with the maxillary, are
two fonnamina. Issuing from the dorsomedian surface at about one-
third of its length from the anterior end is a median process, some-
what like a backwardly projecting spine with a flat, thin, broad base.
This median process extends medially but does not articulate with
the lateral process of the palatine. The maxillary teeth, which are
progressively larger posteriorly, are set in sockets on the ventral
edge of the bone. The curvature of the maxillary bone is greatest at
the level of the third or fourth to the last prediastemal tooth. The
posterior part of the maxillary is curved medially. The maxillary
expands into a flat knob posterior to the last prediastemal tooth.
From the posteroventral surface of the knob issue two posteriorly
projecting, enlarged, groo\ed teeth, behind which are two or there
small, groo\'ed replacement teeth. The posterodorsal surface of the
knob articulates with the anteroventral surface of the transpalatine,
which is a flat bone. The anterior knob of the transpalatine is much

Fig. 33. Dorsal \ie\v of the skull of Homalopsis hiiccata (BMXH
f 920. 1.20.2542) showing lateral view of left mandible, Xl.l.

CoLUBRiD Snakes, Subfamily Homalopsinae 197

broack'i" than the posterior knob of tht> maxillary and therefore,
covers the posterior knob of the maxillary. The transpalatine is
sutured to the lateral surface of the pterygoid just at the place
where the latter expands posteriorly to form the pterygoid groove.
The palatine is rounded and extends anteriorly to the level of the
anterior edge of the \acuity of Jacobson's organ. Posteriorly, the
palatine is forked into two lateral projections which fit around the
anterior end of the pterygoid. About midway in the length of the
palatine is a lateral process, which extends posterolaterally to termi-
nate in a spine. The large palatal foramen is located at the base of
this process. The sphenoid process is a large plate with a long base,
which extends from the dorsomedian surface of the palatine to the
anterior region of the parasphenoid, and forms a dorsal arch over
the nasopharyngeal duct. The palatine teeth are set in deep sockets
on the xentromedian edge of the bone; the nine teeth are curved
posteriorly and increase in length anteriorly. The replacement teeth
arise from the lateral margin of the palatine, at, or just above, the
bases of the sockets. Posterior to the transpalatine-pterygoid suture
is a deep groove in the dorsolateral edge of the pterygoid. The
dorsal surface of the pterygoid is rounded anteriorly and flattened
posteriorly. Anterior to the lateral process articulating with the
transpalatine, the pterygoid is rounded and has a slight lateral shelf
along the dorsolateral edge near the transpalatine-pterygoid suture.
The 21 backwardly projecting teeth decrease in size posteriorly; the
teeth are situated in sockets on a high median crest, which is a
posterior extension of the rounded anterior part of the pteiygoid.
Lateral to this crest is a shallow, thin plate, bounded by a lateral
pterygoid crest extending posteriorly from the transpalatine suture
to the posterior tip of the pterygoid.

Mandil)le. — The dentary is laterally compressed, rounded be-
low, and has a narrow dorsolateral crest bearing teeth on the medial
side of the bone. The posterior half of the dentary overlies the
anterior part of the fused surangular-prearticular part of the articu-
lar. In the middle of the broad lateral surface of the dentary is the
anteriorly directed, large mental foramen. Ventrally, the posterior
part of the dentary underlies the splenial, which is set in a median
trench within the dentary. Near the posterior end of their common
dorsal suture is the large anterior inferior alveolar foramen; lying
completely within the splenial and ventral to the inferior alveolar
foramen is the anterior mylohyoid foramen. Posterior to the splenial
and also forming a part of the ventral surface of the mandible is the
wedge-shaped angular, which lies directly beneath the fused sur-

198 University of Kansas Publs., Mus. Nat. Hist.

angular-prearticular. Near the anterior edge of the angular is the
small posterior mylohyoid foramen. The articular, prearticular and
surangular are fused. The prearticular portion forms the high
median wall of Meckel's canal. In the surangular part, immediately
posterior to the end of the dentary, is the large surangular foramen.
Meckel's canal separates the surangular and prearticular through
the posterior two-thirds of their lengths. The canal terminates at
the articular condyle, which is followed by a short tympanic crest of
the articular. The articulation of the articular with the quadrate is
formed by two lateral flanges of the quadrate that fit over a medial
ridge formed by the articular.

Dentition. — The 12 prediastemal maxillary teeth decrease in
size posteriorly. The teeth are directed \ entrally from the socket
and then curved sharply backwards. On the medial surface of the
bone at the base of each tooth is a small replacement tooth in its
own small socket. Posterior to the diastema is a pair of slightly en-
larged and backwardly projecting, grooved teeth. On the posterior
side of the base of each enlarged tooth is a replacement tooth in its
own socket. The replacement teeth are directed posteriorly. There
are nine teeth on the palatine, 21 on the pterygoid and 18 on the
dentary (=mandible). The teeth on these bones decrease in size
posteriorly and each has at its base a replacement tooth in its own
socket. The replacement teeth on the palatine and pterygoid form
a row of teeth on the lateral surface of the bones; those of the den-
tary form a row of teeth on the medial surface of the bone.

Vertebrae. — Hypapophyses are present throughout the vertebral
column including the atlas and axis. The fiftieth vertebra is chosen
for description. The neural spine is elongate, thin, and has almost
parallel anterior and posterior edges. The zygosphene projects an-
teriorly, with the dorsally and laterally facing surfaces forming an
angle. The stout, spine-like prezygapophyses extend laterally be-
yond the terminal articulating processes of the flat and broad post-
zygapophyses. The articular surfaces of the prezygapophyses and
postzygapophyses are o\al. The zygantrum is deep, narrow, and
its ventral and medial surfaces form an angle. Immediately ventral
to the prezygapophyses is the oblong parapophysis. It is formed by
the ventrolateral extension of the centrum. The antero ventral edge
of the parapophysis has a flattened articulating surface. On the
ventral surface of the centrum is a stout backwardK^ projecting
longitudinal h^papophysis, which extends beyond the ball and
terminates in a flattened, blunt spine. The anteroventral articulat-

CoLUBHiD Snakes, Subfamily ITomaloi'Sixae 199

ing surface at the base of the prezygapophysis is rounded and low.
This is joined by a ridge to the posterodorsal articulating surface on
the parapophysis which is rounded and high. The capitulum and
tuberculum of the rib are fused to form an articulating depression.

Intergeneric \'ariation in the Skulls

The skull of Enhi/dris enhijdris (KU 92352) differs strikingly
from that of Homalopsis huccata. In the latter species the premaxil-
lary is relati\ ely thin and lacks concavities. The lateral processes of
the premaxillaries slope downward; the dorsal spine is convex and
cur\es posterodorsally to the common suture between the nasals.
The anterior process of the septomaxillary is relati\ ely narrow, elon-
gate, and thin. The lateral wing of the septomaxillary is relatively
shorter and has a narrower base than in H. huccata. The postero-
inferior process of the vomer is thin and encloses a relatively small
and nearly round \omerine fenestra. The shape of the frontal of E.
enhijdris is almost rectangular (Fig. 34); it is about two times as
long as wide. The postfrontal is narrow, relatively small, and has an
upturned edge. The squamosals are narrower and more rounded
than in H. huccata (Fig. 35). About half way down the medial
surface of the quadrate is the columellar process. The quadrate is
slightly depressed on the medial surface. There is no medial
diagonal ridge, but the supracolumellar and infracolumellar crest
are present; the former is the more pronounced. Approximately at
the level of the fifteenth maxillary tooth the maxillary bone is
sharply indented posteriorly on its dorsal surface. The posterior
part of the maxillary does not curve medially as in H. huccata; the
posterior knob is relatively thick and small. There are 20 maxillary
teeth, which are progressi\'ely larger posteriorly, followed by a
diastema and two backw^ardly projecting, enlarged, grooved teeth.
There are nine teeth on the palatine and 18 on the pterygoid,
approximately the same numbers as in //. huccata. The replacement
teeth and their sockets are arranged as in H. huccata. The trans-
palatine is relatively short, thin, and more convex dorsally than that
of U. J)iiccata.

The skull of Cerherus njnchops (KU 92353) is notably different
from that of Homalopsis huccata. In the former species the pre-
maxillary is relatively thin and has a posteroventral concavity, but
lacks anterodorsal concavities. The lateral processes of the pre-
maxillary slope downward; the dorsal spine is convex and curves
posterodorsally to the common suture of the nasals. The septo-
maxillarv is broad and short anteriorly; the lateral wing of the

200

University of Kansas Publs., Mus. Nat. Hist.

Fig. 34. Dorsal \iew of the skull of
Enhijchis enhydris (KU 92352) show-
ing lateral \ iew of the right mandible,
Xl.2.

Fig. 35. Ventral \ie\\' of the skull of
EnJiydris enliydris (KU 92528) .show-
ing lingual \'ie\\' of the left mandible,
Xl.2.

septoma.xillary is thin, shorter, and has a narrower base than in H.
huccata. There is no postero-inferior process of the vomer. The
shape of the frontal of C. rynchops is ahnost rectangular (Fig. 36).
It is about twice as long as wide. The postfrontal is narrow, rela-
tively small, and has a rounded edge. The squamosals are narrower
and more rounded than in H. lyuccaia. About two-thirds of the way
down the medial surface of the quadrate is the o\'oid coluniellar
process. The quadrate is depressed, with the diagonal ridge on the
medial surface as in H. huccata. Both supracolumellar and infra-
columellar crests are present; the latter is more pronounced. Ap-
proximately at the level of the twelfth maxillary tooth the maxillary
bone is acutely indented posteriorly on the dorsal surface of the
bone (Fig. 37). The posterior part of the maxillary does not curve
medially as in H. huccata; the posterior knob is relatively thick and
broad. There are 15 maxillary teeth, which are progressively larger
posteriorly, followed by a short diastema and two backwardly
projecting enlarged, grooved teeth. There are nine teeth on the
palatine and 25 teeth on the pterygoid. The replacement teeth and
their sockets are arranged as in H. huccata. The transpalatine is
relatively short and less convex dorsally than that of H. huccata.

The skull of Fordonia leucohalia (BMNH 97.12.30.37) shows
notable differences in comparison with that of Homalopsis huccata.
In the former species the premaxillary is relatively thin, has a
posteroventral concavity, and lacks anterodorsal concavities. The
lateral processes of the premaxillary are heavy at the base; the
anterior edge slopes posteriorly, and the posterior edge slopes ante-

CoLUBRiD Snakes, Subfamily Homalopsinae

201

Fig. 36. Dorsal view of the skull of Ccrhcitis rtjncliops ryncliojit; (BMNH, no

number given ) , Xl.3.

riorly. Distally the process is short and stout; the dorsal spine is
convex and curves posterodorsally to the common suture of the
nasals. The septomaxillary is broad and short anteriorly; the lateral
wing of the septomaxillary is thin and shorter than in H. huccata.
The postero-inferior process of the vomer is relatively thick and
encloses a small \'omerine fenestra. The shape of the frontal of F.
leucobalia is rhomboidal with wavy edges (Fig. 38). It is as long
as wide. The postfrontal is relatively large and has an expanded
dorsal edge, which is sutured to the anterolateral process of the

Fig. 37. \'entral \iew of the skull of Cerberus njnchops njnchops (BMXH, no

number given ) , Xl-3.

202 University of Kansas Publs., Mus. Nat. Hist.

parietal; ventrally the postfrontal is expanded into a rounded knob.
The parasphenoid is broad and flat without a groo\'e and midxentral
ridge. The squamosal is flat, straight and short. About two-thirds of
the way down the medial surface of the quadrate is the o\oid
columellar process. The quadrate has a broad dorsal edge which
articulates with the squamosal. The supracolumellar crest is thin
and high and the infracolumellar crest is stout and low. Ventrally
there is a narrow neck in the quadrate before it expands into the
articulating process. There is no medial diagonal ridge, and the
medial surface of the quadrate is slightly depressed. The anterior
third of the maxillary is edentulous. The median process of the
maxillary is thin, broad, long, and perpendicular to the anterior
half of the maxillary. The posterior knob is relatively thick. The
transpalatine is thin, fan-shaped and short; the broad end covers
the posterior knob of the maxillary dorsally, and the narrow,
tapered, posterior end is attached dorsally to the transpalatine
process of the pterygoid. The sphenoid process of the palatine is
broad, stout and extends toward the midline; the two sphenoid
processes of the palatines are in contact at the midline where they
cur\e \entrally; their medial edges project into the buccal ca\ity.
This modification presumably enables the snake to crush crabs.
There are seven or eight subequal maxillary teeth followed by a
relatively long diastema and two posteriorly projecting, enlarged,
grooved teeth. There are four or fixe teeth on the palatine and eight
or nine teeth on the pterygoid. The replacement teeth and their
sockets are arranged as in H. huccata.

Fig. 38. Dorsal \iew of the skull of Fordonia leucobalki (BMXH 97.12.30.37),

Xl.5.

CoLUBiUD Snakes, Subfamily Homalopsinae 203

Life History

Little is known about the life histories of the species of the
subfamily Homalopsinae. The data presented here represent a com-
pilation of all known literature records, in addition to my personal
obserxations.

The snakes of the subfamily Homalopsinae are o\ovi\iparous.
The actual process of birth was recorded by Cantor (1847:948) for
Enhijdris enlujdris. He stated, "A large female, after having been
confined upwards of six months in a glass vessel filled with water,
brought forth eleven young ones in the manner noted above under
Acrochordus javanicus. During the process she lay motionless on
the bottom of the \'essel, the anterior part of the abdomen was
retracted towards the vertebral column, while the muscles of the
posterior part were in activity. Shortly after the parturition she
expired under a few spasmodic movements, and also two of the
\oung ones died in the course of about two hours, after having, like
the rest, shed the integument."

Cantor (1847) reported that in the valleys of Penang the female
of Homalopsls huccata brought forth six to eight living young at a
time, each between 7 and 8 inches in length; he also reported that in
a young female of Fordonia leucolmJia the oviduct enclosed four
white cylindrical eggs, which contained much yolk, and each of
which measured "about an inch in length."

Giinther (1864) reported that the female of Cerberus rynclwps
brought forth eight living young, 7 to 7*2 inches in length.

Wall ( 1921 ) recorded that a gravid female Enhijdris sieholdi
collected in Bengal in July contained five embryos and seven non-
fertile eggs; the length of one embryo was 7/2 inches. According to
Wall (1921), the young of Cerberus rijnchops are born in May,
June and July; the period of gestation probably is about six months
because he found eggs, but no embryos, in a gravid female from
Rangoon, in February.

Schmidt (1927) recorded 4 to 11 eggs in a series of Enhijdris
phimbea from Hainan.

Pope (1929) recorded in EnJiydris chinetisis, 3 to 13 young
(average 5.7) in 17 females from Fukien, 3 young from a female
from Foochow, and 7 and 8 young from two female specimens from
Kuliang.

Maki (1931) recorded a Formosan specimen of Enlujdris chinen-
sis giving birth to 13 young in August, and of E. pJumbea giving
birth to 10-15 young in August.

204 University of Kansas Publs., Mus. Nat. Hist.

According to Smith (1943), 6-18 young are produced at a time
in Enhijdris enlujdris, 3 to 12 in E. cJiinensis, 9-21 in Homolopsis
biiccata, S to 26 in Cerberus rynchops, and 9 to 13 in Erpeton ten-
tacuhtuni.

Kinghorn ( 1956 ) reported that as many as 12 eggs occur in
females of Enhijdris pohjiepis.

Tweedie (1957) reported that a specimen of Erdnjdris hocourti
from Kedah ga\e birth to 17 young in capti\'ity.

In female Enhijdris enhijdris I have observed 9 to 11 eggs having
an average length of 22 mm. and width of 12 mm., and in female
Homalopsis huccota 9 to 11 eggs having an average length of 32
mm. and width of 15 mm. Another specimen of H. hiiccata (KU
92351 ) collected near Rangoon in early February had 6 eggs having
an average length of 23 mm. and width of 9 mm.

Four females (AMNH 93155-58) shipped alive from Karachi to
the American Museum of Natural History gave birth to 49 young
between July 20 and 30, 1964. Another of the females (AMNH
93154) has 16 embryos.

One specimen of Gerardo prevosfiona (MCZ 74092) from Thai-
land, collected on November 9, 1957, has 5 eggs, each measuring
about 12 mm. in length and 5 mm. in width.

One specimen of Fordonia Jeucohalio (BMNH 63.2.28.5) has 13
eggs; the date of collection is unknown.

The only available record of the growth and longevity of these
snakes is that of Wall ( 1921), who reported that two small Burmese
specimens of Cerberus rynchops, which were brought to him mea-
suring "7f8 and 7% inches" in May and July, respectively, doubled
their lengths in the first vear of life, and had about trebled their
lengths by the end of the second year.

Although possessing venom and enlarged posterior maxillary
teeth, snakes of the subfamily Homalopsinae are usually docile
creatures; the) seldom, if ever, attempt to bite when captured or
handled. Only two species — Enhydris bocourti and E. plundjea —
are known to bite savagely, if carelessly handled or frightened.

D'Abreu (1913) reported on the efi^ect of a bite from Enhydris
enhydris. The subject, a healthy man about 36 years old, was bitten
on the back of the right hand. There were t\\'o distinct punctures.
The region of the bite inflamed immediately. Fifteen minutes later,
the hand began to throb; the pain lasted for an hour, after \\'hich no
ill effects were felt.

Flowers (1899) reported that Homalopsis buccata, which were
kept as pets by him in Thailand, seemed to have more intelligence

CoLUBiui) Snakes, Subfamily Homalopsixae 205

than most snakes and appreciated being petted. He stated that
w'henever he went near the tank where they were kept, the snakes
would crawl up on him or curl up in a pocket, but they resented
being touched by anyone else.

Most homalopsine snakes are sluggish in their movements in the
water, and on land they are extremely awkward. The mode of
progression of Cerberus on land is unusual. The body is thrown
forward in a curve in adxance of the head, and the head subse-
quently, in a "side- winding" manner (Wall, 1919).

Most of the species live on fish or frogs, and their predation may
be detrimental in ponds where fish are cultured. Schmidt (1927),
and Pope (1929) reported that stomachs of Enhijdris plumhea con-
tained only amphibians and that the stomachs of E. chinensis con-
tained only fish. Bourret (1912) stated that Bltia lives entirely on
fish. Fordonia leucohaUa lives along the coasts and in tidal rivers.
It reportedly inhabits crab-holes and feeds on crabs. Cantor ( 1847)
reported that F. leucohalia lives only on fishes and crustaceans. One
specimen of F. leucohaUa (BMNH 63.2.28.5) has in its stomach a
crab measuring 40 mm. in breadth.

Phylogenetic Relationships

The following interpretation of the interrelationships of the
genera and species of the subfamily Homalopsinae is based on the
data given in the sections on systematics, morphology, and ecology
and of corresponding situations that have been studied in other
groups of animals.

Interspecific Relationships

On the basis of morphological characters, such as scutellation,
proportions, coloration and dentition, the 22 species of Enhijdris can
l)e di\ided into nine species groups. Enhijdris pakistanica and the
E. punctata group apparently are early divergences from the main
stock of EnJnjdris. Enhijdris pakistanica differs from all other
members of the genus in having 29 rows of dorsal scales, more
subcaudals, nasal cleft extending to internasal, two internasals, and
no definable chin-shields. In coloration it resembles E. enhijdris in
having dark brown longitudinal stripes alternating with pale brown
stripes on the body. It is found in the delta region of Indus, the
western end of the range of the genus, to the exclusion of all other
members of the genus. The species in the E. punctata group have
more upper labials anterior to the orbit and more upper and lower
labials than any other members of the genus.

206 University of Kansas Publs., Mus. Nat. Hist.

Of the seven remaining groups, I consider the E. phnnhca group
to be an early divergence of the main evolutionary stem of Enliydris.
This group consists of small snakes having short heads, square
frontal bones, dorsal scales in 19 rows and almost uniform brown
or black colors.

The other six groups have comparatively longer heads and
rectangular frontal bones, which are longer than wide. Three of
these groups — enJujclris, cliinensis and maculosa — are alike in
coloration and in having the fourth upper labial entering the orbit
and the first three labials in contact with the loreal. I think that the
E. cJiinensis group differentiated from the E. enhydiis group and
that the E. maculosa group, in turn, differentiated from the E.
chinensis group. The snakes in the £. enliydris group have 21 rows
of dorsal scales at midbody and the loreal in contact with the
internasal and with the first three upper labials. Members of the E.
chinensis group have 21 or 23 rows of dorsal scales and the loreal
not in contact with the internasal. The snakes in the E. maculosa
group have 25 rows of dorsal scales and the loreal not in contact
with the internasal.

The E. macleayi group must have diverged from the main
Enliydris stock before the differentiation of the stem giving rise to
the E. enliydris, chinensis and maculosa groups. The E. macleayi
group is confined to Celebes, New Guinea and northern Australia.
The species in this group have 21 or 23 rows of dorsal scales, the
fourth and fifth or the fifth and sixth upper labials entering the orbit,
and the loreal in contact with the second and third upper labials.
Enliydris matannensis differentiated in Celebes by developing a
large anterior pair of chin-shields, the scales of which are in contact
with the first five lower labials. The differentiation of Enliydris
macleayi in northern Australia resulted in the second to the last
upper labial being divided horizontally and the first three lower
labials being in contact with the anterior pair of chin-shields. The
third species of this group, E. polylepis, is found in New Guinea and
northern Australia and differs from the other two members of the
group in coloration and in having the loreal in contact with the
internasal.

Prior to the differentiation of the E. macleayi group, the E.
siel)oldi and E. l)ocourti groups must have diverged as a single stem
from the main Enliydris stock and later differentiated to the E.
sieholdi and E. bocourti groups. Both groups have 27 or 29 rows of
dorsal scales and nine or ten gulars and two or three incomplete
ventrals between the chin-shields and the first complete ventral.

CoLUBHiD Snakes, Subfamily Homalopsinae 207

The E. siehokU group differs from the E. hocourti group in having
the loreal in contact with the first three upper labials, the first four
lower labials in contact with the anterior pair of chin-shields, and in
having a different color pattern.

The species and subspecies of the genus Cerberus are related in
the follo\\'ing way. The genus ranges from India to northern Aus-
tralia and northward to the Philippines. The snakes of the genus
inhabit fresh, brackish, and salt water. The earlier invaders of the
original stock to the Philippines, New Guinea and Australia differ-
entiated into C. microlepis, C. njnchops novaeguinae and C. aiis-
tralis in their respective geographic areas. Cerberus r. njnchops,
which presumably is closest to the original stock, has an extensive
range and lives sympatrically with C microlepis.

The two species of Cantoria inhabit brackish and salt water.
Cantoria violacea is the western species found in the tidal rivers
and coast of Burma and the Malav Peninsula, the Andamans, Su-
matra and Borneo. Cantoria annulata occurs in New Guinea.

Intergeneric Relationships

Morphological characters useful in showing relationships in one
group of snakes may be of no significance in another. Oliver
(1948) found the presence or absence of a loreal in Leptopljis
(=:T]mlerop1iis) to be significant only at the specific level, whereas
this character is used to separate certain genera of Homalopsinae.
A single or di\ ided anal plate is a constant characteristic of many
genera of snakes, but in the Homalopsinae, a divided plate is a
character of the subfamily. The presence of absence of grooves in
the posterior maxillary teeth has been used as a single criterion
for the separation of many genera of colubrid snakes and also has
been the basis for subfamilial groupings. Such examples serve to
illustrate the possible danger in separating genera on the basis of
only one or two morphological characters, for in so doing the actual
generic relationships can be obscured.

The characters utilized in separating genera within the sub-
family Homalopsinae are: head scutellation, such as nature of
parietals, relation of nasals to each other, nature of nasal cleft,
presence or absence of loreal, arrangement and number of chin-
shields; nature of ventral and dorsal scales; length of hemipenis and
of its bifurcated parts; nature and number of maxillary teeth.

The 10 genera of the Homalopsinae form three distinct groups —
Enhydris, Homalopsis and Cerberus; Heurnia and Myron; and
Gerarda, Fordonia and Cantoria. Bitia and Erpefon stand apart.

208 University of Kansas Publs., Mus. Nat. Hist.

The Enliydris-Homolopsis-Cerbems group can be distinguished
from the other genera by the following combination of characters:
distinct head; semi-divided nasals forming a median suture; loreal
present; anterior mandibular teeth longest; posterior pair of chin-
shields separated by small scales; more than fixe upper labials;
ventrals smooth and rounded; maxillarv teeth more than 10, fol-
lowed by an interspace and two posterior, grooved teeth.

The members of this group differ among themselves in the
nature of parietals and dorsal scales, arrangement and number of
chin-shields, number of dorsal scale rows and minor differences in
skull elements.

The genus Cerberus has the most extensive geographical range,
from Sind (India) to Palau Islands (Pacific Ocean) and northern
Australia. This wide geographical distribution can be attributed to
its long and efficient existence in both fresh and marine waters.
The genus Enlnjclris has a less extensive geographical range than
that of Cerberus. Enhydris occurs from the Indus delta region of
West Pakistan to Formosa and southward to northern Australia; it
is not found on the Philippines and Palau Islands. The differentia-
tion into 22 species in this genus and its present distribution suggest
that it has been in existence longer than Cerberus (Darlington,
1948) and that it has been more confined to fresh-water. Enhydris
bennetti and E. chinensis are known to enter into the sea, but the
other species of Enhydris are apparently strictly fresh-water inhabi-
tants.

The genus Homalopsls has the least extensive geographical range
of the group. It occurs from India to southeastern China; on the
Sunda Islands as far east as Java and northward into Borneo. The
genus is monotypic and inhabits strictly fresh-water. In Burma
these snakes are abundant in "inns" (natural fish ponds) and are a
menace to fisheries. The present geographical distribution and
habitat of this monotypic genus suggests that, in comparison with
Cerberus and EnJujdris, its origin is relatively recent.

Myron and Heurnia can be distinguished from the other genera
by the following combination of characters: distinct head; parietals
entire; nasals separated by an internasal, nasals with nasal cleft;
anterior mandibular teeth longest; one preocular and two postocu-
lars; more than five upper labials; ventrals smooth and rounded;
maxillary teeth more than 10, plus two posterior grooved teeth.

Myron resembles Cerberus by having keeled dorsal scales, and
small size. Heurnia resembles the Enhydris-Cerherus-Homalopsis
group by having smooth dorsal scales, depressed head, and small

CoLUBRiD Snakes, Subfamily Homalopsinae 209

size. The species in this group differ in the nature of the nasal cleft,
the number and arrangement of the chin-shields, the nature of the
dorsal scales and the number of the dorsal scale rows.

Both genera are found in the southeastern limits of the geo-
graphical range of the subfamily. Myron extends from northern
Australia through New Guinea to Ceram, whereas Heurnia is known
only from New Guinea.

The Fordonia-Gerarda-Cantoria group can be distinguished from
the other genera by the following combination of characters: head
not distinct from neck; parietals entire; nasals entire, separated by
an internasal; two pairs of chin-shields; dorsal scales smooth; ven-
trals smooth with rounded edges.

Fordonia differs from Gerarda by lacking a loreal, having both
pairs of chin-shields forming a median suture, having 25-29 dorsal
scale rows as compared with 17, having six to eight maxillary teeth
as compared with 11 to 13, having an edentulous anterior end of
maxillary and palatal, and in minor differences in the maxillo-
palatine-pterygoid arch.

Fordonia differs from Cantoria by having blunt, subequal man-
dibular teeth, two postoculars as compared with a single postocular
and 25-29 maxillary teeth as compared with 19.

Cantoria differs from Gerarda by having the mandibular teeth
increasing in size anteriorly, a single postocular as compared with
two, both pairs of chin-shields forming a median suture, differences
in the number of dorsal scale rows and maxillary teeth, and by
having a relatively elongated body. The elongate and slender body
distinguishes Cantoria from all other genera in the subfamily.

Of the three genera, Fordonia has the most extensive geographi-
cal range, from Burma eastward to northern Australia, and it has
been reported as far north as Sulu Islands. The unusual nature of
the skull elements can be correlated with its crab-eating habits.
Two color phases are known from specimens collected south of the
equator.

Fordonia and Gerarda are monotypic, and Cantoria has two
species. C. violacea with two color phases, is known from the
coastal areas of Burma and Malaya and C. annidata is known only
from New Guinea. Both species are thoroughly marine and feed
mainly on fish.

Gerarda is found from Thailand through Burma to India. It is
found in mangrove swamps in Thailand.

Bitia and Erpeton stand alone with little resemblance to the
other genera in the subfamily. Bitia resembles Erpeton by having

210 University of Kansas Publs., Mus. Nat. Hist.

narrow, keeled ventrals. It resembles the Enhydris-Cerberus-Homa-
lopsls group by having a loreal, nasals with nasal cleft, mandibular
teeth increasing in size anteriorly, and posterior chin-shields sepa-
rated by small scales. It resembles the Mijron-Huernia group by
having an internasal between the nasals, loreal, mandibular teeth
increasing in size anteriorly, nasals with nasal cleft, and two post-
oculars. It resembles the Fordonia-Gerarda-Cantoria group by hav-
ing an internasal between the nasals; two pairs of chin-shields, and
smooth dorsal scales.

Bitia resembles the sea-snakes in having a small head and a
narrow neck with compressed body and tail. Although marine, its
geographical range is rather limited; it is known only from the
coastal areas of Burma, Thailand and Malaya.

The unique, paired, mobile rostral appendages of Erpeton dis-
tinguish the snake from other genera of the subfamily, for that
matter, from all other snakes, although the head scutellation is
discernible into different component scutes. Each scute usually is
surrounded by a series of small scales. It is a fresh-water snake
known only from Thailand.

Cenozoic History of Southeast Asia and
THE Indo-Australian Archipelago

This summary of the history is based on the work of Dickerson
(1924), Umbgrove (1938, 1942), Mayr (1944), Hess (1948), van Bem-
melen (1949), De Beaufort (1951) and Inger (1954).

As early as 1854, the British geologist Earle pointed out that
geologically the Malay Archipelago consists of three parts — a
western one comprising the Greater Sunda Islands and the adjoin-
ing parts of Asia, and an eastern one consisting of New Guinea and
Australia, both of which were very stable during the Tertiary, and
the unstable islands in between.

According to Dickerson and van Bemmelen, the Eocene or pos-
sibly early Oligocene was a period of land emergence. The Indian
peninsula before the upheaval of the Himalayas was temporarily
separated from the Asiatic continent by the Tethys Sea, the great
sea which stretched from the Mediterranean eastward, and opened
into the Bay of Bengal. The first uplift of the Himalayas occurred
in the Middle Eocene, and uplift continued until the Pleistocene.
The vast amount of sediment eroded from this uplifted region was
deposited in what is now the plain of the Ganges. These deposits,
known as the Siwalik deposits, range from the Lower Miocene to
the Lower Pleistocene and attain at some places a thickness of

CoLUBRiD Snakes, Subfamily Homalopsinae 211

16,000 feet (De Beaufort, 1951). The peninsula of India was also
greatly disturbed. The Deccan volcanic action was of exceptional
violence and lasted from the end of the Cretaceous to the Middle
Eocene. The then Malay Peninsula, including islands of the Sunda
shelf, was a region of uplands in the Eocene. According to Inger
(1954), northern Luzon, Mindoro, the Calamians, Palawan, and the
Zamboanga Peninsula may also have been exposed. The Sangihe-
Minahasa ridge of van Bemmclen joined the northern arm of the
Celebes and the Talaud-Maju ridge of van Bemmelen joined the
eastern arm of the Celebes during the Eocene (Umbgrove, 1938,
fig. 8).

Subsidence of the land in the Lower Miocene was widespread.
During the Upper Miocene period, uplift and folding affected the
Luzon Arc, the Samar Arc, and the Celebes-Timor geosyncline. Ac-
cording to Vening Meinesz (1934), the folding was so violent that it
resulted in the widespread overthrusting of the older strata over
the younger ones. The outer Banda Arc, consisting of the islands
Sumba, Timor, Barbar, Timor Laut, Kei, Ceram, and Burn, was
formed along part of this fold. Slightly later, but still in the Mio-
cene, the second fold was formed consisting of parts of Sumatra
and Java, as well as the Inner Banda Arc (Bali, Lombok, Sumbawa,
Flores, Alor, Wetar, Dammer, and Banda).

Umbgrove (1938) and van Bemmelen (1949) disagree on the
Pliocene history of Borneo and Celebes; the former suggested that
the islands were largely emergent, whereas the latter stated that
they were covered by shallow seas.

Smit Sibinga (1947) gave an account, based on geological evi-
dence, of the vicissitudes of this part of the world from the Late
Pliocene to the Recent. In the Lower Pliocene western Java formed
the end of a peninsula. The rest of Java was mostly below the sea.
During the Pleistocene these islands were joined to western Java
and only during the third and again during the fourth glaciation
was the region of the Java sea dry land (De Beaufort, 1951, figs.
7a-e).

Evolution and Zoogeography of the
Subfamily Homalopsinae

The reconstruction of the evolutionary history of the subfamily
is based on the assumed phylogenetic relationships of the extant
species; no fossil record exists. Dating of different events is not
possible because of lack of knowledge concerning the rates of evo-
lution. Although I am aware of the absence of essential information

212 University of Kansas Publs., Mus. Nat. Hist.

for the interpretation of the evolution of the species and subspecies
of Homalopsinae, I feel that some speculation is in order.

The subfamily Homalopsinae belongs to the Old World group
of colubrids that evolved on the southeastern part of the Asiatic
continent, according to clues that are supposed to show the places
of origin and directions of dispersal of groups of animals (Darling-
ton, 1948). The probable evolutionary history of this group is based
primarily on the morphology and distribution of the living species
and correlated with what is known of the morphology and distribu-
tion of the other living groups in this region and the past geology of
this region. In the discussion of folding and emergence of land
masses with their accompanying river systems, the importance of
the possibility of saltatory mode of dispersal is kept in mind.

Within the Enhydris-Homalopsis-Cerberus generic complex, the
monotypic genus Homalopsis is strictly fresh-water in habitat. The
snakes of this genus are larger than other snakes of the subfamily,
and feed mainly on fish. The authentic range is from Burma, south
of 17° lat. (Smith, 1943) to southeastern China and southeastward
to Borneo. The present distributional pattern of the genus and its
fresh-water habitat suggest that it was already differentiated from
the common ancestral stock before the emergence of the Pleistocene
Sundaland. The distribution pattern conforms with the view of
Molengraaff and Weber (1919), and Smit Sibinga (1947), that during
the Pleistocene glacial periods the shallow Java sea was dry land
and that the islands were united. On the other hand this distribu-
tion could have existed since the Miocene. Since the snakes are well
represented in the eastern parts of Borneo and Java but are absent
in the western parts of those islands, theoretical Pleistocene disper-
sal conforms with Weber's views on the distribution of the fresh-
water fishes in the Kapuas and Mahakam rivers. Paucity of these
snakes in India and China and their absence in the Andamans and
Nicobars, Simeulue, Nias, Siberut and Mentawai support an as-
sumption of dispersal to the Sunda Islands by way of the Malayan
route.

Enhydris, having 22 species, is the dominant genus in the sub-
family. Although most of the species are fresh-water, some are
inhabitants of brackish-water, and some even venture into the sea.
This distribution closely resembles that of Cerberus, which is found
in fresh, brackish and salt water. Enhydris probably is the oldest
genus of the Enhydris-Homalopsis-Cerherus stem. Four of the
species (E. indica, innominata, longicauda, and pahangensis) are
endemic in the center, whereas E. matanensis and E. macleayi are

CoLUBRiD Snakes, Subfamily Homalopsinae 213

endemic on the periphery of the generic range. Endemism within
the center of origin suggests that the ranges of some species are re-
ceding and that their niches are becoming narrower through com-
petition with other species.

Enliydris indica of the E. plumhea group is endemic to the
Malay Peninsula and probably is a receding species. Enliydris
alternans of the same group replaces E. imlica in the Greater Sunda
Islands. The presence of E. alternans on Belitung and North Bor-
neo suggests that it might have been a widespread species during
the Pleistocene regression of the sea. The third species of this group,
E. plumhea, ranges from Burma to southeastern China, including
Hainan and Formosa, and southward along the Sunda Islands to
Bali and Celebes. There are authentic records of snakes of this
species in fresh-water streams in hilly districts in Burma, Cambodia
and Formosa, and Smith (1943) reported a specimen caught in a
fishing-net at Ko Lak in the Gulf of Siam. Evidently this species
inhabits both fresh and salt water. Its probable occurrence in
Makassar and Tempe (Rooij, 1917) suggests its penetration from
Java and the Lesser Sunda Islands into the southwestern arm of
Celebes when the Java Sea was at its lowest in the Pliocene and
Pleistocene.

Among the four species of the E. enhydris group, E. enhydris
has the most extensive range. The snakes are common in ponds,
irrigated fields and sluggish waters of Burma, Siam and Cambodia.
Dispersal southeastward suggests probable invasion of the Sunda
Islands during pre-Pleistocene time. The penetration of this species
into India may have been by the upper reaches of the Brahmaputra
River because the snakes are known as far north as the foothills of
the Himalayas. The course of the river was previously through
southeastern China (Hora, 1938). Once this species was established
in the Brahmaputra River it could have spread westward along the
Ganges River. It is known as far inland as Gonda in Uttar Predesh,
and southward into Bihar. Rooij (1917) listed Ceylon, Celebes and
New Guinea in her locality records; I have not seen specimens from
those areas. Enhydris jagorii, E. innominata and E. smithi are
closely related; the ancestral stock must have had a wider distribu-
tion when the rivers of southeastern China had an easterly course
discharging into the Pacific during Miocene time. Later these rivers
were succeeded by rivers flowing to the west and south, thereby
enabling the fauna to spread (Hora, 1937). Enhydris smithi differen-
tiated in the Chao Phraya Basin and E. innominata in the Mekong
Basin.

214 University of Kansas Publs., Mus. Nat. Hist.

Enhydris chinensis flourishes in southeast China and becomes
rare in the northern part. Specimens are known as far inland as
Ichang. It must have populated the Yang-tse-kiang when it origi-
nally drained southward into the Red River, but later turned north-
eastward across central China to enter the Pacific (Mori, 1936) —
hence the remarkable fact that the upper course of this great river
has an Oriental species, which reaches a lower Palaearctic one.
Enhydris hennetti differentiated on the coastal areas of southeast
China and on the continental islands of Hong Kong and Hainan,
when these islands were connected to the mainland. Specimens
have been caught from the Strait of Hainan, thereby suggesting
tolerance of salt water. This is the only species of Enhydris known
to occur in the sea. Enhydris longicauda differentiated from the
main ancestral stock of the E. chinensis group in the Tonle Sap.

The ancestral stock of the E. rnacleayi group must have been
pushed by competing species as far east as northern Australia dur-
ing the Pliocene, or earlier in the Miocene. The E. madeayi stock
differentiated into E. matannensis in Celebes, E. madeayi in north-
ern Australia, and E. polylepis in New Guinea and northern Aus-
tralia.

The ancestral stock of the E. maculosa group presumably was
widespread in Burma and Malaya. Later it differentiated into E.
maculosa in Bumia and E. pahangensis in Malaya.

The ancestral stocks of the E. sieholdi group and E. pakistanica
spread through India from Malaya during the Miocene when the
Indobrahm, also known as the Sawalik River, drained westward into
the Arabian Sea.

The ancestral stock of the E. hocourti group possibly was wide-
spread from Malaya to Vietnam and the Greater Sunda Islands,
during Pliocene and Pleistocene times. Later, due to regression of
the sea, populations isolated in Sumatra and Java diff^erentiated into
E. alhomaculata, and the mainland population into E. hocourti.

The ancestral stock of the E. punctata group may have had an
extensive range from India via Malaya to Borneo in Miocene time.
A part of the population isolated in Borneo during regression of the
sea differentiated into E. doriae, and the mainland population into
E. punctata which spread again into Borneo during Pliocene and
Pleistocene times.

The distributional pattern of Cerberus, which is conspicuously
absent from the southeastern part of China and its coastal islands,
negates the possibility of its dispersal across the China Sea as
postulated by von Koenigswald (1939), who attempted to explain

CoLUBRiD Snakes, Subfamily Homalopsinae 215

the presence of the fresh-water cyprinoid fishes Elephas and
Stcgodon in the PhiHppines. The dispersal route taken by Cerberus
to enter the Phihppines was via the Malay Peninsula. It entered the
Pliilippines twice; the earlier invasion probably took place during
the Miocene via the Greater Sunda Islands across the Palawan-
Calamians route to Luzon, where it differentiated into C. micro-
lepis. The second invasion probably occurred in Pliocene or Pleisto-
cene times, when the snakes entered the Philippines via the Greater
Sunda Islands through the Palawan-Calamians route to Luzon, and
the Sulu Islands route to Mindanao. The southeastern dispersal
of the original stock probably took place in the Pleistocene when
the regression of the sea was maximal. Populations were isolated
in northern Australia and New Guinea, and differentiated into C.
australis and C. rijnchops novaeguineae, respectively.

Like Cerberus, Fordonia lives in both fresh and salt waters. It
is a monotypic genus that does not occur as far west or north as
does Cerberus. Fordonia must have been well differentiated from
the ancestral stock prior to the Pleistocene when it dispersed.

The ancestral stock of Cantoria probably had a wider distribu-
tion than the genus now has. During the Pleistocene it occupied
the tidal rivers and coastal areas of Burma, Malaya and the Sunda
Islands and New Guinea. The populations in New Guinea differen-
tiated into C. annulata, and those in Burma and Malaya differen-
tiated into C. violacea. The populations in the intervening areas
have become extinct.

Gerarda, unlike its close relatives Fordonia and Cantoria, inhab-
its mangrove swamps and seldom ventures into the sea. Probably it
was well differentiated before the Pleistocene. Its distribution is
westward from Thailand through Burma to India.

Myron, Heurnia, Bitia and Erpeton are monotypic genera hav-
ing restricted distributions. The ancestral stock of Myron and
Heurnia were probably pushed eastward by competing genera as
far east as northern Australia and New Guinea in the Pliocene and
Pleistocene. In northwestern New Guinea the stock differentiated
into Heurnia ventromaculata. In northern Australia and on the
islands of the Arafura Sea the ancestral stock of Myron-Heurnia
group differentiated into Myron richardsoni.

Bitia and Erpeton are early divergences from the main Homa-
lopsine stem; Bitia evolved into a marine littoral inhabitant in
Burma, Malaya and Thailand, and Erpeton differentiated by taking
refuge in the rivers and ponds of Thailand.

216 University of Kansas Publs., Mus. Nat. Hist.

Summary and Conclusions

The snakes of the genera Homalopsis, Enhydris, Cerberus,
Heurnia, Myron, Fordonia, Gerarda, Cantoria, Bitia and Erpefon
are members of a group of aquatic snakes forming a subfamilial unit,
the Homalopsinae. Seven genera are monotypic; of the remaining
three, the genus Enhydris embraces 22 species, Cerberus three spe-
cies, and Cantoria two species.

The subfamily is characterized by maxillary teeth that increase
in size posteriorly, followed by a diastema and two enlarged,
grooved teeth. The eyes are relatively small and have elliptical
pupils; the nasals usually are larger than the internasal, and the
nostrils are small, valvular and situated anterodorsally on the head.
The ventrals are relatively narrow, usually smooth, but bicarinate
in Erpeton and Bitia; the anal plate is divided, and the subcaudals
are paired. Vertebral hypapophyses are present throughout the
length of the column. The hemipenes are forked. So far as is
known, all species are ovo viviparous.

The characters that are useful in separating genera are the
nature of the dentition, the parietal scutes, the nasal scutes, and the
nature and number of ventrals and rows of dorsal scales. Other
generic characters are the position of the internasal relative to the
nasals and the position of the nasal cleft to the surrounding scutes.

The genus Enlnjdris contains nine species-groups that are distin-
guished by the number of rows of dorsal scales, the relations of the
orbit to the upper labials, the loreal and nasal cleft to their sur-
rounding scutes, and the anterior pair of chin-shields to the lower
labials. Additional distinguishing characters are number of upper
and lower labials, and the color pattern of the body and head.

Analysis of numerous clines in morphological characters shows
that parallel clines exist in two or more species in certain characters,
such as reduction in the number of ventrals and subcaudals from
north to south and from west to east. Color pattern usually shows
discontinuous clinal variation.

Studies of the skulls of Homalopsis, Enhydris, Cerberus, and
Fordonia show that the skulls are of normal colubrid type but have
differences in the shape and nature of the premaxillary, maxillary,
palatine, vomerine fenestra, quadrate, frontal and the dentition.

The distribution of the subfamily is from the Indus Basin in
West Pakistan to China as far north as Nanking in Kiangsu Prov-
ince, the Indo-Austrahan Archipelago, and northern Australia.
Most of the species inhabit fresh water; some live in brackish water,
and some are marine. Only a few species occur far inland in the

CoLUBRiD Snakes, Subfamily Homalopsinae 217

upper reaches of river systems. The degree of aquatic adaptation
in different taxa is demonstrated in the nature of ventrals, dorsals,
and lateral compression of the tail, and in the size of the head and
neck in relation to the body. Other characters modified for aquatic
existence, such as the nature of nostrils and eyes, and the position of
mouth and tongue are nearly alike in all the taxa.

Snakes of this subfamily feed primarily on fish and frogs; some
lizards are included in the diet. Fordonia leucobalia lives on fish
and crabs; the peculiar nature of the maxillo-palatal-pterygoid arch
correlates well with this habit. So far as is known all species give
birth to three to 21 young at a time.

The hemipenis, except for small variations in detail, is uniform
throughout the subfamily. The organ is comparatively short and is
forked for about half its length. The distal end is finely calyculate;
the low lips of the cups are stiffened with small blunt spines. Proxi-
mally these small spines merge into a median region where the
calyces and spine are larger. Near the point of bifurcation there is
an area beset with large, flat, triangular, papilla-like processes ar-
ranged in longitudinal rows, each ending in a spine. The basal
region is naked. The sulcus is forked.

Based on the Cenozoic history of southeastern Asia and the
Indo- Australian Archipelago, coupled with studies of the morphol-
ogy and distribution of these snakes and other vertebrates, hypoth-
eses have been formed as to the evolution and the dispersal routes
of the homalopsine snakes to various islands. Generic differentia-
tion from a common ancestral stock probably occurred in Miocene,
in southeastern Asia especially around the Choa Phray, Mekong,
Si and Peh drainages. The fresh water taxa in the Sind area are
probably the result of a Miocene dispersal through the Siwalik
River system. The presence of southeastern Asiatic species in the
upper part of the Ganges supports Hora's view that the Brahma-
putra River previously had its origin in southern China. The pres-
ence of Enhijdris chinensis in Ichang possibly is due to southward
drainage of the Yang-tse-kiang (Mori, 1936). The dispersal of the
snakes into the Indo-Australian Archipelago took place in the Plio-
cene and Pleistocene by several waves of migration when the Java
Sea was at its lowest. The migratory route to the Philippines was
via the Malay Peninsula, through the Sunda Islands across the
Palawan-Calamians and/ or Sulu chains, and not across the South
China Sea. The genus Myron, two species of Enhtjdris and one of
Cerberus, which are not represented in Asia, occur in northern
Australia. The paucity of southeastern Asiatic snakes in Australia

218 University of Kansas Publs., Mus. Nat. Hist.

is due to age of water gaps between the islands and the snail size
of the islands themselves. Once established in Australia, the spread
of the animals is checked by geographical and climatic barriers
(Storr, 1964).

CoLUBRiD Snakes, Subfamily Homalopsinae 219

Literature Cited

Anderson, J.

1871. On some Indian reptiles. Proc. Zool. Soc. London, 1871, pp. 148-211.
AXDERSSON, L. G.

1899. Catalogue of Linnean type-specimens of snakes in the Royal Mu-
seum in Stockholm. Bih. K. Svenska Vet. Akad. Hand., 24, Afd. 4,
No. 6:34-35.

BeMMELEN, R. W. VAX

1949. The geology of Indonesia. lA:xxiii-|-732 pp., 375 figs.

BOULEXGER, G. A.

1890. The fauna of British India, including Ceylon and Burma. Reptilia

and Batrachia. London, 541 pp., text-figs.
1896. Catalogue of the snakes in the British Museum (Natural History).

London, 3:xiv+l-721, figs. 1-37, pis. 1-25.
1912. A \ertebrate fauna of the Malay Peninsula from the Isthmus of Kra

to Singapore including the adjacent islands. Reptilia and Batrachia.

London, 297 pp., figs., pis., map.

BOURRET, ReXE

1936. Les Serpents de I'lndochine. Tome II. Etudes sur la Fauna. |505
pp., text-figs.

Broxgersma, L. D.

1929. On the zoogeography of Java. Treubia, supplement, 1929:64-68.

Caxtor, Th.

1847. Catalogue of reptiles inhabiting the Malayan Peninsula and Islands.
Jour. Asiatic. Soc. Bengal, 16(2) :897-952, 1026-78.

Carlleyle, a. C. L.

1869. Description of two new species belonging to the genera Varamts and
Fcranioidcs respectiveely from near Agra. Jour. Asiatic Soc. Bengal,
38:192-200.

Cope, E. D.

1886. An analvtical table of the genera of snakes. Proc. Amer. Phil. Soc,

23:479-499.
1895. The classification of the Ophidia. Trans. Amer. Phil. Soc, 18(2):
186-219.

Darlixgtox, p. J., Jr.

1948. The geographical distribution of cold-blooded vertebrates. Quar-
terly Re\ie\v Biol., 23:1-26, 105-123.

Daudix, F. M.

1803. Histoire naturelle, generale at particuliere des reptiles. Paris, 7:1-
436.

DiCKERSOx, R. E., et al.

1928. Distribution of life in the Philippines. Phil. Bur. Sci., Manila,
Monogr. 21, 322 pp.

DowLixG, H. G.

1951a. A proposed standard system of counting ventrals in snakes. British

Jour. Herpetology, l(5):97-99, fig. 1.
1951b. A proposed method of expressing scale reductions in snakes. Copeia,
1951(2):131-134.

Dowlixg, H. G., and Savage, J. M.

1960. A guide to the snake hemipenis: A survey of basic structure and
systematic characteristics. Zoologica, 45( 1 ) : 17-28.

DuNiERiL, A. M. C, and Bibrox, G.

1854. firpetologie generale ou histoire naturelle complete des reptiles.
Paris, 8(2) :xii+781-1536.

Fitzixger, L. J. F. J.

1843. Systema reptilium. Vindobonnae, viii-|- 1-106 pp.

220 University of Kansas Publs., Mus. Nat. Hist.

Flower, S. S.

1896. Notes on a collection of Reptiles and Batiachians made in the Malay
Peninsula in 1895-1896, with a list of the species from that region.
Proc. Zool. Soc. London, pp. 856-914.
1899. Notes on a second collection of Reptiles made in the Malay Penin-
sula of Siam, from Nov. 1896-Sept. 1898. Proc. Zool. Soc. London,
pp. 674-681.
Gans, C.

1959. A taxonomic revision of the African snake genus Dasypeltis, ix+237
pp., pis. 1-13.
Glauert, L.

1950. A handbook of the snakes of Western Australia. Perth, 50 pp.,
fig. 1.
Gray, J. E.

1849. Catalogue of the specimens of snakes in the collection of the British
Museum. London, xv4- 1-125 pp.

GtJNTHER, A. C. L. G.

1800. On the Reptiles of Siam. Proc. Zool. Soc. London, pp. 113-117.
1858. Catalogue of colubrine snakes in the collection of the British

Museum. London, x\i+ 1-281 pp.
1864. The reptiles of British India. The Ray Soc, xxvii-|- 1-443.
Haas, C. P. J. De

1950. Checklist of the snakes of the Indo-Australian Archipelago. Treubia,

20:511-62.5.

Hora, S. L.

1937. Distribution of Himalayan fishes and its bearing on certain palaeo-
graphical problems. Rec. Indian Mus., 39(3);251-9.

1938. Changes in the drainage of India, as evidenced by the distribution
of freshwater fishes. Proc. National Inst. Sci. India, 4(4):39.5-409.

Inger, R. F.

1954. Svstematics and zoogeographv of Philippine Amphibia. Fieldiana:
Zoology, 33(4) :183-531.

Jan, G.

1863. Elenco sistematico degli ofidi, descritti e disegnati per I'iconografia
generale. Milan: Lombardi, vii-|- 1-143 pp.

Jan, G., and Sordelli, F.

1860. Iconographie generale des ophidiens, Milan and Paris, 100 pp.

Jong, J. K. De

1926. Heur7ua ventro-mactdata n. g. n. sp. und Cautoria annulata n. sp.,
zwei neue Schlangen \'on Neu-Guinea. Zoolog. Anzeiger, Bd.
LXVII, Heft 11/12, 20.7.26. Akad. Verlagsgessellschaft m. b. H.,
Leipzig.

Keast, a.

1959. The reptiles of Australia. Monogr. Biol., 8:115-135, 3 figs.

KiNGHORN, J. R.

1956. The snakes of Australia. Sydney, vii-f 198 pp.

Koenigswald, G. H. R. von

1939. Das Pleistocan Javas. Quartar, 2:28-53.

Krefft, G.

1869. The snakes of Australia. Sydney, 100 pp., 12 pis.

Loveridge, a.

1934. Australian reptiles in the Museum of Comparative Zoolog>', Cam-
bridge, Massachusetts. Bull. Mus. Comp. Zool. Harvard, 77:243-
383, pis.

1948. New Guinean reptiles and amphibians in the Museum of Compara-
tive Zoology and United States National Museum. Bull. Mus.
Comp. Zool., 101:305-430.

CoLUBRiD Snakes, Subfamily Homalopsinae 221

Maki, M.

1931. A monograph of the snakes of Japan. Dai-Ichi Shobo, Tokyo,
vii+240 pp., 158 text-figs.

Mayr, E.

1944. Wallace's line in the light of recent zoogeographic studies. Quart.
Rev. Biol., 19(1):1-14.

Mayr, E., Linsley, E. G., and Usixger, R. L.

1953. Methods and Principles of Systematic Zoology. New York, ix+328
pp., 45 figs.

Merrem, Blasius

1820. Tentamen systematic amphibionmi, 80-90 pp.

Mertexs, R.

1930. Die Aniphibien und Reptilien der Inseln Bali, Lombok, Sumbawa
und Flores. Frankfurt, 344 pp., 9 pis.

1959. Enhydris pakisianica, sp. nov. Type loc. Jati, Indus region, Sind,
W. Pakistan. Note on E. diissumierii. Senck. biol. 40, 3-4:117-120,
2 photos.

Molengraaff, G. a. F.

1922. Geologie in: De Zeen van Nederlandsch Oost-Indie. Leiden, pp.

272-357.

OSHIMA, M.

1916. Key to the snakes of Formosa and the Riu-Kius. JAP. Transactions
of the Natural Hist. Soc. of Formosa, VI 23:1-12.

Parker, W. K.

1878. On the structure and development of the skull in the common

snake (Tropidonotus natrix). Phil. Trans. Roy. Soc. London, pt.

2:385-417, pp., pis. 27-33.
Peters, J. A.

1960. The snakes of the subfamily Dipsadinae. Mus. Zool., Univ. of
Michigan, pp. 1-224, 8 pis.

Peters, W. C. H.

1863. Uber die \on Hrn. Dr. F. Jagor in Siam gesamnulten Amphiliien.
Monats. Akad. Wiss. Berlin, pp. 245-247.

Pope, C. H.

1929. A list of reptiles known to occiu- in Fukien Province, China. Proc.

Soc. Fukien, 2:20-22.
1935. The reptiles of China. Turtles, crocodiles, snakes, lizards. Nat.
Hist, of Central Asia. New York, 604 pp., text-figs., pis.

Radovanovic, M.

1937. Osteologie des Schlangenkopfs. Jenaische Zeitschr. Naturw.,
71 (2): 179-312.

Raven, H. C.

1935. Wallace's Line and the distribution of Indo-Australian mammals.
Bull. Amer. Mus. Nat. Hist., 68:179-293., 10 pis.

Romer, a. S.

1956. Osteology of the reptiles. Chicago, pp. xxi+ 1-772, figs. 1-248.

RooiT, N. De

1917. The reptiles of the Indo-Australian Archipelago. Vol. 2, Ophidia,
xiv+334 pp., 117 figs.

Russell, P.

1796. An account of Indian serpents collected on the coast of Coromandel;
containing descriptions and drawings of each species, together with
experiments and remarks on their several poisons. London, 90 pp.,
pis.

1801-9. A continuation of an account of Indian serpents; containing descrip-
tions and figures from specimens and drawings, transmitted from
various parts of India. London, 53 pp., pis. 1-41.

SCHLEGEL, H.

1837. Eassai sur la physionomie des serpents, 2:340-361.

222 University of Kansas Publs., Mus. Nat. Hist.

Schneider, I. G.

1799. Historiae Amphibiorum Classis Genus II. Hydrus. Angium cogna-
tum, pp. 235-251.
Smith, M. A.

1914. Snakes of Bangkok. Jour. Nat. Hist. Soc. Siam, 1(1) :5-18, 2 pis.
1926. Monograph of the sea-snakes ( Hydrophiidae ) . British Museum
Natural Hist., London, .\ix+ 1-130 pp.

1929. Remarks on three rare reptiles from the Indo-Chinese region. Jour.
Nat. Hist. Siam, 8:49-50.

1930. The Reptilia and Amphil^ia of the Malay Peninsula. A supplement
to G. A. Boulenger's Reptilia and Batrachia, 1912. Bull. Raffles
Mus., 3:1-149.

1943. The fauna of British India, Ceylon and Burma, including the whole
of the Indo-Chinese sub-region. Reptilia and Amphibia, 3:.\ii-f583
pp., 166 figs.

SooD, M. S.

1948. The anatomy of the vertebral column in serpents. Proc. Indian
Acad. Sci., 28:1-26, figs. 1-10.

Stejneger, L.

1907. Herpetology of Japan and adjacent territory. Smithsonian Institu-
tion, U.S. Nat. Mus. Bull., 58:299-307.

Storr, G. M.

1964. Some aspects of the geography of Australian reptiles. Senck. biol.,
45 (3/5): 577-589.

SZUNYOGHY, J.

1932. Beitrage zur xergleichenden Formenlehre des Colubridenschiidels,
nebst einer Kraniologischen Synopsis der fossilen Schlangen
Ungarns. Acta Zool., 13:1-56.

Taylor, E. H.

1922. The snakes of the Philippine Islands. Phil. Bur. Sci., Manila,
Monogr. 16, 1-312 pp., 37 pis.

Theobald, W.

1868. Catalogue of the reptiles of British Burma, emliracing the pro\inces
of Pegu, Martaban and Tenasserim. Jour. Linn. Soc. London,
10(41):l-68.

TiRAXT, G.

1885. Notes sur les reptiles et les batraciens de la Cochinchine et du
Cambodge. Saigon, pp. 38-43.

Tweedie, M. W. F.

1946. A new snake from the Malay Peninsula [Enhydris pahangensis].

Ann. Mag. Nat. Hist, ser. 11, 13:242-244, fig.
1954. The snakes of Malaya, Singapore. Rafiles Museum, 1954:\i4-139
pp., 12 pis., 26 figs.

Umbgrove, J. H. F.

1929. The amount of maximal lowering of sea level in the Pleistocene.

Proc. 4th Pacific Science Congress (Java), 2 A: 105-113, 3 figs.
1938. Geologic historv of the East Indies. Bull. Amer. Assoc. Petrol. Geol.,

22:1-70, 1 pi., 24 text-figs.
1942. The pulse of the earth. The Hague, Martinus Nijhofi", .\xii+358 pp.,

11 pis., 204 text-figs.

Wall, F.

1914. Occurrence of Cantor's watersnake (Cantoria violacea) in the

Andamans. Jour. Bombay Natl. Hist. Soc, 23:166-167.
1919. A popular treatise on the common Indian snakes. Jour. Bombay

Nad. Hist. Soc, 26:88-93.
1921a. Ophidia Trabrobani or Snakes of Ceylon. Colombo. 254-262 pp.
1921b. Notes on a gravid female of Siebold's watersnake (H. sieboldii).

Jour. Bombay Natl. Hist. Soc, 18:920.

CoLUBRiD Snakes, Subfamily Homalopsinae 223

Wellmax, J.

1963. A revision of snakes of the genus Conophis (Family Colubridae,
from Middle America). Univ. Kansas Sci. Bull., 15( 6) :251-295,
9 figs.

The manuscript for this publication was completed in its present form in
Nhi>-, 1965; soon tliereafter the author joined the faculty of the Department of
Biology, Uni\ersity of Rangoon, Rangoon, Burma. He has not had the oppor-
timity to include literature references subsequent to his retiun to Burma.

University of Kansas Publications
Museum of Natural History

Vol. 20, No. 3, pp. 225-570, 15 figs., 193 maps
February 16, 1971

A Distributional Study of
Missouri Fishes

WILLIAM L. PFLIEGER

University of Kansas

Lawrence

1971

University of Kansas Publications, Museum of Natural History

Editors of this number: Frank B. Cross, Philip S. Humphrey,
William E. Duellman

Volume 20, No. 3, pp. 225-570, 15 figs., 193 maps
February 16, 1971

University of Kansas
Lawrence, Kansas

Published in cooperation with the
Missouri Department of Conservation

PRINTED BY

THE UNIVERSITY OF KANSAS PRINTING SERVICE

LAWRENCE, KANSAS

1971

CONTENTS

Introduction 231

Historical Review of Ichthyology in Missouri 232

The Environmental Setting 238

General Physical Features of Missouri 238

Southeastern Lowlands 241

Ozark Uplands 244

Dissected Till Plains 250

Osage Plains 253

Missouri and Mississippi Rivers 255

Fish Faunae Regions of Missouri 258

Range Limit Analysis 258

Species Composition Analysis 261

Fish Faunal Regions 264

Distribution of Fishes in Relation to Environment 270

Ozark Faunal Group 271

Lowland Faunal Group 274

Prairie Faunal Group 275

Big River Faunal Group 276

Ozark-lowland Faunal Group 278

Ozark-prairie Faunal Group 278

Wide-ranging Faunal Group 279

Competition as a Limiting Factor 280

Climate as a Limiting Factor 282

Variations in Faunal Diversity 282

Cenozoic Events Affecting Distribution of Missouri Fishes 285

Physiographic Changes 285

Drainage Evolution of Mississippi River System 286

(227)

Climates of the Past 291

Developmental History of the Fish Fauna of Missouri

AND THE Central Mississippi Valley 293

Pre-Pleistocene Fish Fauna 294

Pleistocene Changes in Fish Fauna 295

Post- Wisconsin Changes in Fish Fauna 305

Changes in Fish Fauna Within Historic Time 306

Accounts of Species 309

Explanation of Accounts 309

Petromyzontidae 310

Acipenseridae 314

Polyodontidae 316

Lepisosteidae 317 ™

Amiidae 319

Anguillidae 320

Clupeidae 320

Hiodontidae 323

Salmonidae 324

Esocidae 324

Cyprinidae 326

Catostomidae 378

Ictaluridae 390

Percopsidae 398

Aphredoderidae 399

Amblyopsidae 399

Cadidae 401

Cyprinodontidae 401

Poecihidae 406

(228)

Atherinidae 407

Percichth>idae 408

Centrarchidae 409

Percidae 420

Sciaenidae 442

Cottidae 443

Summary and Conclusions 444

Acknowledgments 447

Literature Cited 450

Distribution Maps 462

Lndex to Common and Scientific Names 561

(229)

A DISTRIBUTIONAL STUDY OF
MISSOURI FISHES

BY
WILLIAM L. PFLIEGER

Introduction

The fish fauna of Missouri is perhaps more diverse than that
of any other inland area of comparable dimensions in North Amer-
ica. But, unlike many states that have a rich history of ichthyo-
logical investigations going back nearly to the time when the
country was settled, systematic fish collections were not made in
Missouri until the second quarter of the present century, and to
date no general summary and synthesis of distributional information
has been attempted. The objectives of this report are to summarize
current knowledge concerning the distribution of Missouri fishes,
and to interpret the distribution patterns in terms of the environ-
mental conditions and historical events that have determined their
de\ elopment. Because the natural areas of the state extend beyond
the political boundaries, and the character of the fish fauna has
been determined in part by events that occurred far from Missouri,
an attempt will be made to relate the distribution of fishes in Mis-
souri to distribution in the Mississippi Valley as a whole. The un-
resolved biogeographic and taxonomic problems relating to the
fishes of this region are many, and the conclusions reached will
probaljly be altered by future investigations. If I succeed in casting
some light on the subject, and pointing out areas where further
research will prove fruitful, this report will have fulfilled its major
objecti\'es.

So that the report will be most useful to the general reader, I
have departed from common practice by using vernacular names
rather than scientific names wherever possible in the sections pre-
ceding the species accounts. If questions arise about the scientific
name of the species being referred to, the scientific and vernacular
names are associated on the pages indicated by bold face type in
the index.

This report is written as a companion volume for a handbook
to Missouri fishes still in preparation. Publication of a separate
report on fish distribution permits more thorough treatment of the
subject and inclusion of much material too technical for a popular-
ized handbook. Emphasis in the handbook will be on identification,
life history, and importance to man. Keys for identification of Mis-

(231)

232 University of Kansas Publs., Mus. Nat. Hist.

souri fishes have been dupHcated for distribution, and may be ob-
tained on request from the Fisheries Research Section, Missouri
Department of Conservation, Columbia, Missouri.

Historical Review of Ichthyology in Missouri

The hterature on Missouri fishes is scattered and fragmentary,
and a vast store of unpubHshed distributional information has ac-
cumulated through the efforts of many indi\'iduals. The following
brief review is presented for its historical interest, and to provide a
frame of reference within which to interpret the remainder of this
report. Knowledge about Missouri fishes has not accumulated at a
uniform rate. Based on this changing rate, we may logically divide
the history of ichth\ology in Missouri into three periods of unequal
length. These are: 1820-1905, 1905-1930, and 1930 to present.

The scientific study of Missouri fishes seems to have begun with
the description of Procerus maculatus (=Pohjodon spathuJa) by
Rafinesque (1820:87) in his pioneering "Ichthyologia Ohiensis."
Rafinesque indicated that this fish "lives in the Mississippi and is
sometimes caught at Ste. Gene\'ie\'e in the State of Missouri." He
had not seen specimens, but based his description on a communica-
tion "by Mr. M of Ste. Genevieve."

In April of the year in which the above publication appeared,
the French naturalist C. A. Lesueur made an excursion to the
"Lamotte mine in Missouri" near the present town of Mine La
Motte, Madison County. He obtained se\en brook lamprey's from
"Wilkinson Cave," and these served as the basis for the description
by Lesueur (1827:9-11, pi. 6) of Petromyzon Lamottenii (=^Lom-
petra Jamottei) in "one of the rarest works on American ichthyol-
ogy" (Hubbs and Trautman, 1937:22). The plate of Petromijzon
Lamottenii was long ago reproduced by DeKay (1842:382, pi. 79,
fig. 249), and Hubbs and Trautman {op. cit.) reproduced the long-
lost original description.

Louis Agassiz described nine species of fishes from Missouri, of
which only three currently stand as valid. The descriptions of
Poecilichthijs punctulatus (=EtJieostoma punciulatum) and Poecil-
ichthys spectahilis {^^EtJwostoma spectahile) were published by
Agassiz (1854:304); the description of Alburnus zonatus (z=No-
tropis zonatus) appeared in a paper by F. W. Putnam (1863:9),
but was credited to Agassiz. All were collected from the Osage
River by a Mr. Stolley.

The governmental surveys of railroad routes and territories con-
tributed little to knowledge of Missouri fishes. Girard (1858) listed

Missouri Fishes 233

three species (none new) from St. Louis, and Cope (1871) listed
nine species (se\'en new) from near St. Joseph. Of the names pro-
posed by Cope, only Hybopsis n}issurien.sis {=:Notropis stramineus
niissuriensis) is in current use.

S. E. Meek was the single greatest contributor to knowledge
of Missouri fishes before 1900. In company with D. S. Jordan he
made a series of collections in Missouri and Iowa in August 1884
(Jordan and Meek, 1885). They obtained 48 species from seven
localities in western Missouri. Specimens collected from the Mis-
souri Ri\er near St. Joseph later served as the type material for
Hybopsis meeki (Jordan and Evermann, 1896:317). In the same
summer. Meek and C. H. Gilbert made collections in southwestern
Missouri, and obtained the types of Etheostoma cymatotaenia
{=Percina cymatotaenia) and Etheostoma nianguae (Gilbert, 1888).
Meek (1891) returned to the Ozark region in July and August of
1889 and spent six weeks exploring the streams of Missouri and Ar-
kansas. He obtained 70 species of fishes in 20 collections from Mis-
souri. Three species from Missouri were described as new, of which
Notropis ozarcanus from the North Fork of White River and Etheo-
stoma jiiliae from the James River are still considered valid.

Call (1887) made nine fish collections in and near the Ozark
region of Missouri in the summer and autumn of 1886, obtaining 38
species of fishes. Evermann and Kendall (1895) collected 34 spe-
cies in October of 1891 from Indian Creek and the spring branch
of the U. S. fish hatchery near Neosho in Newton County. Carman
( 1889) reported on ca\'e fishes from a well in Jasper County. These
represented the then undescribed Amblyopsis rosae, later named by
Eigenmann (1898) from Day's Cave, Jasper County. Evermann
and Cox (1896) made no original investigations in Missouri, but
listed 77 species from the Missouri portion of the Missouri River
system in their synopsis of the fishes then known from that basin.
The excellent study of Illinois fishes by Forbes and Richardson
(1909) was not published during this period, but should be men-
tioned here, because it is based on collections made before 1905.
This report dealt only incidentally with Missouri fishes but included
records for 65 species from the Mississippi River where it forms the
common boundary of Missouri and Illinois.

In summary, a number of new species were described from the
state, and the general nature of the fish fauna of south-central and
western Missouri was indicated as a result of studies during this
period. But no effort was made to collect in a systematic fashion,
and large areas remained virtually unexplored.

234 University of Kansas Publs., Mus. Nat. Hist.

The second period (1905-1930) was characterized by Httle ac-
tivity in the study of Missouri fishes. Only two pubHshed references
to fishes collected in Missouri during this period ha\'e come to my
attention. Fowler (1921) listed 13 species collected near St. Louis
by a Mr. Julius Hurter in 1912, and Coker (1930) added a few new
records for the Mississippi Rixer below Keokuk, Iowa.

The early part of the third period ( 1930 to the present ) was
characterized bv renewed interest in the Missouri fish fauna. E. P.
Greaser, whose primary interest was crayfish, made 42 fish collec-
tions in the Ozark region in 1931 and 1932. A. Hugh Denny ob-
tained 32 collections from the Current and Eleven Point systems
in 193S while employed by the U. S. Forest Service. One stimulus
to the study of Missouri fishes was the establishment of the Missouri
Department of Conservation in 1936. This agency put two survey
crews in the field during the summers of 1941 and 1942 and ob-
tained 206 fish collections from the southern half of the state.
Leaders of these crews were Aden C. Bauman and Carl B. Obrecht
in 1941; in 1942 W. C. Frohne, who was in charge of the survey,
led one crew, and Bauman led the other. Their activities were
coordinated with those of George V. Harry, a graduate student at
the Uni\'ersit}' of Michigan working under the direction of Carl
L. Hubbs. During the summers of 1940 and 1941 Harry made 345
collections of fishes, covering all the principal stream systems of
the state. All collections, including those made by Department
personnel, were taken to the University of Michigan Museum of
Zoology. There they were sorted and identified by Harry, with the
assistance of Dr. Hubbs and Reeve M. Bailey.

The study of Missouri fishes was interrupted by World War H,
but was resumed soon thereafter and has continued sporadically
since. Population studies of selected streams by Department biolo-
gists since 1945 have contributed additional distribution records,
and have provided much valuable information concerning the
abundance of the larger species. In the 1950's Perry E. Robinson
made numerous collections, mainly from streams north of the
Missouri Ri\'er. My association with the study dates from 1961,
when I was employed by the Department of Conservation to pre-
pare a report on Missouri fishes. Since 1961 I ha^•e \'isited most
areas of the state and have obtained about 700 collections from
more than 500 localities. These have served primarily to pro\ide
additional information on areas not adequately sampled in earlier
surveys, and to document recent changes in fish distribution and
abundance.

Missouri Fishes 235

All of the published reports resulting from investigations since
1930 have dealt with limited areas or with one or a few species.
In the former category are the papers by Barnickol and Starrett
( 1951 ) on the commercial and sport fishes of the Mississippi River,
Funk et al. ( 1953 ) on the fishes of the Black River, and Fisher
( 1962) on the fishes of the Missouri River. Papers in the latter cate-
gory are too numerous to mention here, but are cited in the ac-
counts of the appropriate species.

Primarily as a result of collections made since 1930, extensive
series of Missouri fishes are available for study at several museums.
B>' far the largest holdings are at the University of Michigan Mu-
seum of Zoology. These include most of the collections made from
1930 to 1942, as well as some earlier material. Significant holdings
of material collected in the early 1940's, as well as some material
collected since that time, are housed at the University of Missouri.
Most of the material collected before 1900 is in the United States
National Museum or the Field Museum of Natural History, Chi-
cago. Most of my collections are at the University of Kansas Mu-
seum of Natural History. Lesser holdings of Missouri fishes are
housed at other museums listed below. The following abbrevia-
tions are used in this report in referring to museums where Mis-
souri fishes are housed: CAS — Chicago Academy of Science; CU —
Cornell University, Ithaca, New York; FMNH — Field Museum of
Natural History, Chicago; INHS — Illinois Natural History Survey,
Urbana; KU — University of Kansas Museum of Natural History,
Lawrence; MCZ — Museum of Comparative Zoology, Harvard Uni-
\ersity; MU — University of Missouri, Columbia; ANSP — Academy
of Natural Sciences of Philadelphia; UMMZ — University of Michi-
gan Museum of Zoolog)', Ann Arbor; USNM — United States Na-
tional Museum, Washington, D.C.

In spite of the fragmentary nature of the literature on Missouri
fishes, a surprising number of original descriptions are based on
specimens collected in the state (Table 1). In all, 39 binomials
have been proposed, of which 15 are in use for species currently
recognized as valid, and one has recently been applied to a sub-
species. Six trinomials are based on specimens collected in Mis-
souri. The most active period in the naming of Missouri fishes was
1820-1905. During that period 34 binomials were proposed, and
11 of these are currently in use.

236

University of Kansas Publs., Mus. Nat. Hist.

3
<

O en
•J3 ID

En"*-
•-< O

c
o

•43

a
<j

c
-a

o
p
a

g
3

00 00

■ ■-<

"rt

Zi2

«3

»o

i-H

1— (

i-H

1 — 1

I— ( e"

»— H

#—

;Z^

00 •~ 5

■^

C/5

c
U

0;
>

1— 1
a

pe, 1

assiz

Puti

I— (
a

;-i

o oc

Q E

ffi

U <

CT>

lOTf

I— ( 1—1

a5C35

cz:

CO-H

i-H

f-H

=On

<D
c
o

00 -S g 00 00

>— I _N ^ ^ i— I

a cb'^ a^ a

o W) iS o

c:5

W5 C3

CX)

in
in

u <:

3m 15 pa

c:5
a

in
t-

CO
O

I-H

a^ -S • >

c _oo P5 ^^ p:; g

.1-^0 ao ar",

iX -Si-J .2"^ .9-

Si, O -^ .;2 r"^ .- -^ O

CS «5 en

o
U

IH
U

3 «

■35 c/:i J -tl

^ o

PSU

0; •-!

r a

Lr .^

._ c ~

a
O

°C/2

gJ5

r a r a

W en PQ en ^

•d 1^ "d h^ "o* cc

P S 0) "^

•t: r a

o
U

c
H

4J o;

S s ^ d z S

1—1
CO

1—1

N
U

o
U

'^S "^
Spa M

en rt

C3 en

u o

t»

«a

.CJ

•«*

s..

c
s

•s;

.IS

■to

•^-

.Co

.Co
.Co

.Co
'co

l>5

3
l»
S

"co

2
3

.CO

J
3

CO
3

3
3

2
1^

<-N

"a.

•^^

'a.

"a.

i-S

^-*.

i-C

.Co

<**.

j..^

'cj

C/3

-5*

5:1
S

^ a:

:5'

1^ :^

;?:

12:

■<S

!**

i-O

•**

c
0

<-*

•**
l*c

0
5

1^
0

1
"5

iJ2

3
5

3
3

-2

.Co

3
t(

.Co

-S

Co
35

-Si

-S-
<

.a.

3
CJ

Cijlindnx
Htjhopsis

a.
0

.05

a.
0

C/2

2
S

■S

S
.3

Htjhopsis
Alhurntis

.Co

2
J

0
0

-5

i-«i

Missouri Fishes

237

IT Tf

00 OC ^

I— I 1— I 1>
„ ^ °0

N N ^

C
■J3

C
CS

XA ^ ■-
tc o

o
00
00

05 G5

CO CD ^^

O 05 C5

^ ^ 00

--H O K5

^ O CD

N 00 00

^ s

-H CO

Tf< CO

CD lO

•-H ^H

05 '—1

CO CO

O p^T

C^ CO

n '^

N N

U U

2 S

:s^

< ^

.«»

i — i

r-» ^

5iU

3, s

o c

55 u

5J ^

c -^ ^

< < u H w m

O M

H H S

<; O O O

CO in

C , r- 00

r Oi c

O
05

^ _, —' TJ ^ T'

C - C7>
rt-^ 00
Oj I— I

-;;« -T^ffl ?^2

^■^■-B r^'^.^^^.D

—I CO

» CQ

—I CO

I— I ^H

in in

in CD ^

T-^ CO if^

CD 00 '^

COo '^

•^ CO >^

2 CD ,«s

Z^ Z

C/D CD

O
CO

I— I
N

CD
CO

C/1

C/2

■^ 00

'En ■/

< S

00
CD
05

« -73 ■=!
DC •_r' *j-i

CD -rH

CO -^ CO 12 ^

:s p :^ w

o
U

o ^

a; <U .^ rt „, O

" Q- r. rt fc 3 ^

^ (^ .22 L_i « Cli cs •-

J Ci-J ccf " P5 .^-D .2 .2 d .S cs wi •-■ •= <aS ^S rt-£ cri a^ ai W ffi

r: .;2 .-t; ii U _a>

c 5

c c

o
m

o
U

.„• o ■- ^

:./:-Kc/2.-I^-?^U;>jj cJ cO oj cP rcc; c> ^U m.c to «^ 5^ m
g- .;£ :5 « 5 "^. . 2 .2 -^ .2 5 .2 S ?^ 5, ^ '^ -Bf ^
-^ ^ES ^ *^ ^" ^' ^ ^ " ^ " ^--^ " - ^

Q 2

>i;

•^

1»

r-S

•^

•2

_.^

*»•

— ^

c/} c/:

H Z U

^. O c/D O ►::5

r-*

.a.

■2

•2

iter

ater

adieus

■2

•2

'5)
0

-5 g §

r^ Q '^

= '^ 2

^ 05 ^

1-^

,co

1>5

n»j
2

^ j^ *-

—

= = "S

!»

5»

i-»i

0 0 3

•♦—

.♦^

.^.i

.1^

-K*

•*«

■s^

;^ ;?: [t.

fei

fcq

ttl

(il

»>

"«*

.K>

3
cr

c
3

■^^

-Q

2
5

•**

ft)

C? CO

"§

niipta
vicauda

0
S

-a

•2

s
s

>agei

s spectab

ft)

00
3

iodcs hi
lostomu

Morone intc\
Pomoxys bre

s
0

S
0

CO
0

stoma p
ilichthy

s
0

a. ci

"«™

ft)

ft) CJ

ft)

^ ii

.«.

••^

n.-

t£

ft)

t;^

ft)

-s ft)

►C

5= c
0 (in

(X,

•♦»*

Co •-: -i 3

? a-r; .-; -c:

2.38

University of Kansas Publs., Mus. Nat. Hist.

The Environmental Setting
General Physical Features of Missouri
Missouri is located in the Mississippi Basin somewhat east and
a little north of the center of the continental United States ( Fig. 1 ) .

Figure 1. Geographic position and drainage relationships of Missouri.

Figure 2 (facing page). Streams and reservoirs of Missouri, with special
reference to those mentioned in this report. Numbers and letters refer to the

streams and reserxoirs listed below.
Streams

1. Tarkio River

2. Nodaway River

3. Platte River

a. Hundred and Two River

4. Big Blue River

5. Fishing River

6. Tabo Creek

7. Grand River

a. Thompson River

b. Medicine Greek

8. Ghariton River

9. Lamine River

a. Blackwater River

b. Flat Greek

10. Perche Greek

a. Hinkson Greek

11. Moniteau Greek

12. Moreau River

13. Osage River

a. Maries River

b. Niangua River

c. Tebo Greek

Missouri Fishes

239

d. South Grand River

e. Pomme de Terre River

f. Sac River

14. Auxvasse Creek

15. Gasconade River

a. Little Piney Creek

b. Big Piney River

c. Jones Creek

d. Osage Fork

e. Lick Fork

16. Loutre River

17. Missouri River

18. Des Moines River

19. Fox River

20. Wyaconda River

21. Fabius River

22. North River

23. Salt River

24. Cuivre River

25. Meramec River

a. Fox Creek

b. Big River

c. Bourbeuse River

d. Huzzah Creek

Reservoirs

A. Thomas Hill

B. Lake of the Ozarks

C. Pomme de Terre

D. Stockton

E. Table Rock

e. Dry Fork

26. Headwater Diversion

a. Whitewater River

b. Castor River

27. Mississippi River

28. Spring River

a. Muddy Fork

29. Shoal Creek

a. Hickory Creek

30. Elk River

a. Indian Creek

31. White River

a. Flat Creek

b. James River

c. Bryant Creek

d. North Fork

32. Eleven Point River

33. Current River

a. Jacks Fork

b. Spring Valley Creek

34. Black River

a. West Fork

b. Toms Creek

35. St. Francis River

36. Little Ri\ er Drainagevvay

F. Taneycomo

G. Bull Shoals
H. Norfork

L Clearwater
J. Wappapello

240 Unwersity of Kansas Publs., Mus. Nat. Hist.

Its maximum width is about 280 miles, its maximum length is about
310 miles, and it has a total area of nearly 70,000 square miles.

The Mississippi River forms all of the eastern boundary of Mis-
souri except for a small segment in the extreme northeast, and it
receives numerous direct tributaries from Missouri (Fig. 2). The
Missouri River forms the western boundarx^ of the state from the
Iowa border south to Kansas City, flowing thence in an easterly
direction across the state to join the Mississippi near St. Louis.
Most of the streams of northern and central Missouri empty into the
Missouri River. Lesser areas in southern Missouri drain into the
Arkansas, White, and St. Francis rivers, which join the Mississippi
beyond the borders of the state.

Because of the inland location of Missouri, its climate is of the
continental type, with marked extremes of temperature and precipi-
tation in some years (Yearbook of Agriculture, 1941:953-954).
Winters are seldom severe, but there are occasional short periods of
extreme cold. Summers are generally warm, but excessively hot
summers are exceptional. The average January temperature ranges
from 38 °F in the southeast to 24 °F in the north, and the average
July temperature varies from 80°F in the southeast to 76° F in the
north. The number of days a year when the temperature drops
below 32 °F averages 65 in the southeast and 105 in the north.
Precipitation decreases from south to north and from east to west.
The average annual precipitation \'aries from about 48 inches in
the southeast to about 32 inches in the northwest. Most precipita-
tion comes in the spring and early summer. Severe droughts oc-
curred in Missouri in the mid-1930's and mid-1950's, causing the
lowest flows ever recorded at stream gaging stations (Skelton,
1966:13). These droughts resulted from a combination of de-
ficiency and unfavorable distribution of precipitation, accompanied
by high temperatures, winds, and low humidity.

Missouri is physiographically diverse. Four distinct physio-
graphic regions may be recognized ( Fig. 3 ) : ( 1 ) the Southeastern
Lowlands, a small but well-defined area in southeastern Missouri;
(2) the Dissected Till Plains, including nearly ah of Missouri north
of the Missouri River; (3) the Osage Plains, a small wedge-shaped
area in western Missouri south of the Missouri River; and (4) the
Ozark Uplands, occupying the remainder of the state. These re-
gions are remarkably difterent in their geologic, topographic, and
edaphic features, and these differences are reflected in the distri-
butional relationships of theii" biotas.

Missouri Fishes

241

S j Di ssected Ti I I Plain

r-\ I

I 1 Tertiary-Quofernary

^^ Cretaceous

\ _\ Penn sy I van i on

^j Mijsissippian

S JIu r i an-Devonlon

O r d o V i c i a n

C a m b r ion
Pr ec ombr t on

v/////M^/^. i Ms.-: f L.._.,i L

Soulhsaslefn Lowlands

!f±}

Figure 3. Physiographic regions and bedrock geology of Missouri. Heavy
black lines indicate the Isoundaries between physiographic regions.

Southeastern Lowlands

Location and ])onndaries. — The Southeastern Lowlands of Mis-
souri represent the northern extremity of the Mississippi Embay-
ment, a vast alluvial plain extending from near Cape Girardeau,
Missouri southward to the Gulf of Mexico. The Southeastern
Lowlands are the most clearly defined physiographic region of
Missouri. They are set off from the Ozark Uplands to the north
and northwest by a steep rocky bluff extending in a curved line
from the Mississippi River just south of Cape Girardeau westward
and southwestward through Poplar Bluff to the state line in south-
eastern Ripley County. This bluff line constitutes a sharp rise of
50 to 250 feet from the level of the lowlands to the level of the
adjacent uplands of the Ozarks (Marbut, 1902:1-2). The South-
eastern Lowlands occupy an area of about 5500 square miles.

Topography. — The region as a whole is a broad plain with a
gentle slope to the south. For the most part, the surface relief is
less than 10 feet, but two prominent topographic features known
as Benton Hills and Crowley's Ridge rise from 50 to 150 feet above
the general level of the plain (Grohskopf, 1955:11). The Benton
Hills are near the Mississippi River in northern Scott County, and

242 University of Kansas Publs., Mus. Nat. Hist.

Crowley's Ridge runs parallel to, and somewhat south of, the bluff
Ime previously described, from Benton Hills southwestward to the
state line. Crowley's Ridge is not continuous but is cut by several
gaps where it has been breached by streams. The most prominent
of these gaps is located about 15 miles west of the Mississippi
River and is about 9 miles wide. A narrower gap at the state line
is occupied by the St. Francis River. Sikeston Ridge, a flat-topped
terrace standing about 25 feet about the general level of the plain,
extends south from Benton Hills to the town of New Madrid, and
a similar low ridge known as Maiden Ridge extends from Crowley's
Ridge in Dunklin County south into Arkansas. Lowland areas
lying between these ridges are: (1) Advance Lowland between
Crowley's Ridge and the Ozark Uplands; (2) Morehouse Lowland,
\\diich leaves the Advance Lowland at the northern gap in Crow-
ley's Ridge and extends southward between Sikeston Ridge and
Maiden Ridge; and (3) Charleston Lowland, lying between Sike-
ston Ridge and the Mississippi River. The surface altitude of the
Southeastern Lowlands varies from 335 feet in the northern part
to 240 feet in the southern part (Grohskopf, 1955:11).

Geology and soils. — Cretaceous and Tertiary sediments, con-
sisting mostly of unconsolidated clays, sands, and gravels, lie be-
neath the Southeastern Lowlands, but these are exposed at the sur-
face only at scattered localities in Crowley's Ridge and the margin
of the adjacent Ozark Uplands. The thickness of these deposits
ranges from a few feet in the outcrop area to more than 2700 feet
in the extreme southeast corner of the state (Grohskopf, 1955:13).
In the lowlands the older deposits are overlain by alluvium ranging
in depth from a few feet to 200 feet, and in the uplands they are
largely obscured by a thick mantle of loess. The allu\ium and
loess were laid down during Pleistocene and Recent times.

The soils of the region are nearly all alluvial in origin, but
because of differences in age and the stream conditions under
which soil materials were laid down, they exhibit considerable
variation in their physical properties (Miller and Krusekopf, 1929:
95). The oldest soils, derived primarily from loess washed down
from the adjacent uplands, are found in the Advance Lowland
west of Crowley's Ridge. The Morehouse Lowland is occupied by
a belt of heavy soils apparently derived from material deposited
in an old channel of the Mississippi River after it was closed at its
upper end, permitting deposition of fine sediments. Extensive
areas of sandy soils bordering this belt of heavy soils on both sides
represent old river banks subject to frequent o\'erflow accom-

Missouri Fishes 243

panicd by swift cunt'iits. The soils of Crowley's Ridge are derived
priniarih' from loess and are similar to those found along the
bluffs of the Missouri and Mississippi rivers farther north.

Drainage features. — Before settlement much of the Southeastern
Lowlands consisted of swamp, and during times of flood the area
was swept by the waters of the Mississippi River as far west as the
St. Francis River. Several Ozark streams penetrated the lowlands
from the north and pursued meandering courses southward to the
Mississippi i^iver. Two of these, the Castor and Whitewater rivers,
joined to form Little River, w^iich drained most of the interior of
the lowlands. Modification of the Little River Valley began in
19L3 and 1914 with the construction of a headwater diversion
channel that shunted the flow of the upper Castor and Whitewater
rivers, as well as a number of minor streams, eastward along the
northern border of the lowlands to the Mississippi River south of
Cape Girardeau. The drainage of Little River was thus effectively
disrupted, and the extensive swamps that formerly occupied its
\alley were eliminated by the construction of an extensive network
of ditches that now drain Little River Valley south into Big Lake,
Arkansas. Similar drainage projects in other parts of the lowlands
have almost eliminated swamplands.

Natural vegetation and land use. — In its original condition the
Southeastern Lowlands was one of the most heavily timbered re-
gions of the state, with cypress, ash, and gum predominating in the
swampy areas, and elm, hickory, oak, and catalpa predominating
on the high ground. The better-drained sites were settled early,
but serious efforts to clear and drain the swamplands were not
initiated until the early part of the present century. At that time
the Advanced Lowland from Advance westward to Puxico was
described by Marbut (1902:6) as ". . . argillaceous and very flat.
Water spreads over the greater part of it, which, owing to the
clay soil and the heavy timber, remains during a greater part of
the year. . . ." And the Morehouse Lowland east of Dexter he
described (p. 9) as "... a series of low ridges with swampy
cypress sloughs between. . . . East of Kennett it is under a few
inches of water during ordinary seasons, and a strip about half a
mile wide along Little River is permanently submerged."

Clearing of timber in the central trough of Little River began
in the lS90's, but did not reach its peak until the first two decades
of the 1900's (Keefe, 1955:6). The area made available to agri-
culture by elimination of the swamps constitutes the richest and
most intensively cultivated section of the state, and it is about 97%

244 Unrersity of Kansas Publs., Mus. Nat. Hist.

in cultivation (Keefe, 1955:7). The only extensive areas of stand-
ing timber and swamps that remain are the 6000-acre Duck Creek
Wildlife Area and the 22,600-acre Mingo National Wildlife Refuge,
two adjacent tracts in Bollinger, Stoddard, and Wayne counties.
Even here water levels are managed artificially for waterfowl pro-
duction.

Aquatic habitats. — The principal remaining habitat for fishes
and other aquatic life in the lowlands is the approximately 1200
miles of drainage ditches. These vary considerably in \'olume of
flow, turbidity, current, bottom type, aquatic vegetation, and shad-
ing by riparian vegetation. Some of the major ditches, which are
large enough to be designated as small rivers, are wide and shallow
with considerable current throughout. Submerged aquatic vege-
tation is generally sparse, but water willow (Justicia) is often dense
along the shores. Cover is scarce and is confined mostly to the
margins where the current has undercut the roots of shoreline
vegetation. Fine shifting sand is the principal bottom type, with
occasional areas of small gravel or silt. The smaller ditches are
more variable in character. Some have no perceptible current
at normal flow; others are fairly swift. Those without perceptible
current usually have a silt bottom, but those with current have
bottoms composed mostly of sand and small gravel. Ditches that
have been dredged recently or that are heavily shaded by small
trees and shrubs have little aquatic vegetation, but most of the
more open ditches are choked with submergent aquatic vegeta-
tion, especially coontail (CeratophyJhim), water milfoil {Mijrio-
pJujUum), and various pondweeds (Potamogeton). The sands and
gravels underlying the lowlands are excellent aquifers, providing
an abundant and reliable source of water for the ditches. In
spite of intensixe cultivation of their watersheds, the ditches run
clear most of the time, because erosion is minimized by low relief.

Although the swamplands have been largely eliminated, some
lentic habitats remain. These consist principally of natural lakes
scoured out by the Mississippi River in time of flood, the aban-
doned channels of Little River and other streams, and borrow pits
where fill for roads was removed. These are generally fairly clear,
with dense growths of the same aquatic vegetation found in the
quieter ditches.

Ozark Uplands
Location and boundaries. — The Ozark Plateaus Physiographic
Province of Fenneman (1938:631-662) is an area of approximately
40,000 square miles of elevated country lying mostly in southern

Missouri Fishes 245

Missouri and northern Arkansas, but also including parts of south-
eastern Kansas, northeastern Oklahoma, and southern Illinois. To-
gether with the Ouachita Mountains, it forms the most extensive
tract of elevated land between the Appalachian Uplift and the
Rocky Mountains (Sauer, 1920:3). The Ozark Plateaus Province
consists of two rather distinct parts, a northern dissected plateau
comprising all of the area here referred to as the Ozark Uplands,
and a southern slightly higher and more completely dissected area,
the Boston Mountains. The boundary between the Ozark Uplands
and the Southeastern Lowlands is narrow and sharply defined,
consisting of the distinct bluff line already described. The boundary
between the Ozark Uplands and the plains to the north and west
is broad and less sharply defined, in many places consisting of a
transition zone many miles in width. From the state line in Jasper
County northeastward to the Missouri River, the boundary is
generally placed at the contact zone betsveen the Mississippian
rocks underlying the Ozark Uplands and the Pennsylvanian sand-
stones and shales underlying the plains regions. North of the
Missouri River the boundary is placed near the heads of a series
of short, direct tributaries of the Missouri River, and includes in
the Ozarks the narrow belt of country intricately dissected by
these streams.

Three subdivisions of the Ozark Uplands can be recognized
in Missouri: the St. Francois Mountains, the Salem Plateau, and
the Springfield Plateau. The St. Francois Mountains, located not
far from the Mississippi River in east-central Missouri, cover a
70-square-mile tract of closely assembled igneous peaks or knobs
that are the exhumed portions of a pre-Cambrian landscape, still
largely buried beneath a thick series of Paleozoic marine sediments.
The Salem Plateau, by far the largest subdivision of the Ozark
Uplands, surrounds the St. Francois Mountains and is coextensive
with the outcropping of Ordovician and older sedimentary rocks.
The Springfield Plateau lies west of the Salem Plateau and is co-
extensive with the area of outcrop of Mississippian rocks in south-
western Missouri. It is separated from the Salem Plateau by the
Eureka Springs Escarpment, which marks the retreating edge of
the Mississippian formations.

Topography. — The highest elevations in the Ozarks (and Mis-
souri) are on some of the granite knobs in the St. Francois Moun-
tains. Many of these are near or above 1700 feet, and the highest,
Tom Sauk Mountain, is 1772 feet above sea level. The highest
elevations on sedimentary rocks are found in Wright and Webster

246 Unwersity of Kansas Publs., Mus. Nat. Hist.

counties near the western extremity of the structural dome. Here
elevations of 1740 feet are reached (Bretz, 1965:100). The main
divide, which runs for approximately 100 miles between these
two points, seldom falls much below 1400 feet (Fenneman, 1938:
637). The elevation drops off on all sides from the dome, more
abruptly on the short southern and eastern slopes than on the
northern and western slopes. Most of the uplands of the Ozarks
are at elevations above 1000 feet, although appreciable areas below
this level are found near the borders of the region.

Because the present topography of the Ozark Uplands is largely
erosional, the greatest relief is along the major streams where
dissection has proceeded most rapidly. The principal streams oc-
cupy narrow, steep-sided valleys that are bordered by belts of
rough country cut by tributary streams. Where small streams are
numerous, dissection is often so complete that no upland flats re-
main, and the surface features consist of narrow, deep valleys
separated by sharp ridges. The southern slope of the Ozarks is
generally more deeply and intricately dissected than the northern
slope. Local relief in the St. Francois Mountains is commonly as
much as 500 to SOO feet. Unlike the ridges and hills of the remain-
der of the Ozarks, the crests of the St. Francois Mountains ha\c
no common level. The area drained by White River is nearly
as rugged, having ridge-tops often standing 300 to 500 feet above
the valley floors. The country bordering the major streams on the
northern slope is prominently dissected, but the relief seldom
exceeds 350 feet. The most extensive area of undissected uplands
in the Ozarks is found west of the St. Francois Mountains along
the major east-west divide, with important outliers along other
divides to the north, south, and west. Here the surface is undu-
lating rather than hilly, and local relief seldom exceeds 100 feet.
The valleys are relatively broad and shallow, and are occupied by
surface streams only after hard rains. Large areas are uncut by
any stream, and all drainage is subsurface.

Geology and soils. — The pre-Cambrian crystalline rocks exposed
in the St. Francois Mountains are the oldest and geologically the
lowest of the bedrocks underlying the Ozark Uplands. Arranged
outward from this crystalline core, in irregular and locally inter-
rupted belts, are successively younger series of sedimentary rocks
ranging in age from Cambrian through Mississippian. Remnants
of Pennsylvanian rocks cap some upland di\ides on the northern
slope. The pre-Cambrian rocks are mainly granites and porphyries.
The sedimentary rocks consist of a large series of limestones, sand-

Missouri Fishes 247

stones, and shales, but doloniitie limestones containing large quan-
tities of chert predominate. The chert is resistant to weathering,
and when the surrounding rock is dissoKed away, chert remains
as thick beds of coarse, angular rock fragments that mantle the
uplands and fill the stream \'alleys to great depths. The chert is
exposed at the surface in the stream beds, in road cuts, and over
large areas on the upland slopes; the litter of chert is one of the
characteristic features of the Ozark landscape.

Weathering of the highly soluble limestones that are the im-
portant soil forming rocks of the Ozark leaves little behind except
cliert, and soil formation is exceedingly slow ( Miller and Krusekopf,
1929:65). The residue from which the soils are formed is high
in iron, and this oxidizes on exposure, giving a red color to the
soils. The soils are loose and porous, and heavy clay subsoils are
absent. This facilitates the rapid filtration of water and causes ex-
tensive leaching. As a result the soils are infertile, low in organic
matter, and acid in reaction. The content of chert gravel in the
soils \'aries greatly, with the more le\'el areas being least stony.
In general, the soils of the central Ozarks are exceptionally stony;
those of the border areas are less stony.

Drainage features. — The long axis of the structural dome ap-
proximates the main drainage di\'ide separating north-flowing
streams that empty into the Missouri River (through the Osage
and Gasconade rivers) or into the Mississippi (through the Mera-
mec River) from south-flowing streams emptying into the White
River (through the James, North Fork, Eleven Point, Current, and
Black rivers ) and into the Mississippi Ri\'er ( through the St. Francis
and Headwater Diversion ) . A narrow belt along the eastern border
is drained by short, direct tributaries of the Mississippi. Drainage
of the southwestern slope is into the Arkansas River by way of the
Spring and Elk rivers.

During the period preceding the last uplift of the Ozarks an
extensi\'e network of subterranean drainage channels was formed
in the highly soluble limestone rocks. Erosion following uplift
has partially exposed this subterranean network, and it manifests
itself at the surface in the form of numerous sinkholes, caves and
springs. Sinkholes are best developed on the undissected uplands.
A notable concentration of sinkholes is found not far from the
Mississippi River in Ste. Genevieve and Perry counties. Caves and
springs occur where streams have encountered subterranean chan-
nels during downcutting. Missouri has more than 1450 catalogued
caves (Bretz, 1965:12), of which the majority are in the Ozarks.

248 University of Kansas Publs., Mus. Nat. Hist.

The Ozarks Uplands are noted for their many springs. Some of
these are among the largest in the United States. Twelve Missouri
springs each have an average daily flow of 65,000,000 gallons or
more (Beckman and Hinchey, 1944:17). Most springs are along
the valleys of the major streams, but few parts of the Ozarks are
completely without springs.

Another surface manifestation of the extensive subterranean
drainage is the numerous dry valleys. Some valleys that are a quar-
ter of a mile wide carry surface water only after a hard rain. In
some cases sti-eams that are permanent in their upper reaches
disappear downstream as a result of subterranean stream piracy.
Dry Fork of Meramec River in Dent and Phelps counties and
Roubidoux Creek in Texas and Pulaski counties are examples of
such streams. Dry Fork is thought to contribute its flow to Mera-
mec Spring, which emerges in the xalley of the Meramec River
many miles north of the place where the upper Dry Fork disap-
pears into the ground.

Natural vegetation and land use. — Timber covers most of the
dissected part of the Ozark Uplands and much of the undissected
part. Extensive stands of short-leaf pine mixed with various hard-
woods are found in the southeastern Ozarks, but mixed oaks make
up the bulk of the timber. White, black, red, post, and black-jack
are the common oaks of the Ozark Uplands. Red cedar forms
extensive glades on limestone ledges that are thinly covered with
soil. Various hardwoods are found in the stream valleys, with
sycamore, sugar maple, walnut, butternut, hackberry, and tulip
tree being common. Fingers of prairie formerly extended into the
Ozarks from the west along the broad upland divides, with isolated
pockets of prairie farther east. Scattered timber was present over
part of this, lending a park-like aspect to the landscape. Most
of the prairie is now gone, but some remains in the western Ozarks
where it is maintained for prairie hay.

Because of its thin, unproductive soils and rugged terrain, the
Ozark Uplands is the least intensively cultivated section of the
state. Extensive areas have never been under culti\'ation or have
been allowed to revert to a natural state.

Aquatic habitats. — The typical Ozark stream descends abruptly
from undissected upland flats into a narrow, steep-sided valley,
and by the time it is large enough to maintain a permanent flow
it is deeply incised and bordered by rugged hills. The stream
itself consists of a rapid succession of short pools and well defined
riffles, and is floored primarily by chert fragments that are little

Missouri Fishes 249

loss angular than those on the surrounding uplands. The chert is
thrown up into huge bars during times of flood, and these bars
are the dams behind which the stream pools are ponded. Riffles oc-
cur where the stream breaks across a bar to descend into the next
pool. The chert bars are often unstable, and the configuration of
the stream bottom undergoes marked changes at every major flood.
Sand and silt occupy only small areas of the stream bottom, mainly
in the deeper pools and protected backwaters. Chert accumulates
in the stream valleys to such depths that bedrock is exposed in the
channel only where the stream encounters the valley wall. Here
limestone bluffs are formed by imdercutting, and the channel is
often partly blocked by coarse rubble and large limestone blocks
that have fallen following undercutting. In late summer and
autumn many small Ozark streams are reduced to a series of iso-
lated pools, maintained by seepage through the bars of chert that
separate them. Where pools are poorly developed, the stream bed
may be dry for long distances, although water continues to flow
beneath the surface.

Farther downstream, the valley broadens somewhat but is
e\en more deeply incised, and is often bordered by sheer limestone
blufi^s that rise 150 feet or more above the stream bed. The pools
are proportionately longer and deeper, and silt and sand bottoms
are more prevalent. The riffles are better defined, and are floored
by rounded gravel, rubble, and small boulders. In general, the
bottoms of the larger Ozark streams are more stable than those
of the smaller streams. The largest springs in the Ozarks emerge
along the major streams. Some of these have a flow ecjual to that
of a small river, and have a marked effect on the flow character-
istics and temperature of the streams they enter. The flow of a
stream may double below the outlet of such a spring, and the
water remains cool enough for a considerable distance to support
trout and other coldwater fish.

Ozark streams are unusually clear because the thin, stony soils
of the uplands do not contribute many fine sediments to surface
runoff, and much of the water entering the streams is first filtered
through beds of chert. Because their base flow is maintained almost
entirely by springs, Ozark streams have the highest, most well-
sustained base flows of any Missouri streams (Skelton, 1966:25).
The high relief of most Ozark watersheds results in rapid runoff
during periods of heavy precipitation, and at such times the streams
undergo sudden drastic fluctuations in water level.

Higher aquatic plants are sparse in small Ozark streams. Water

250 University of Kansas Publs., Mus. Nat. Hist.

cress ( Nasturtium ) is present at the entrances of many springs and
along the margins of streams that are strongly influenced by springs.
Submergent vegetation, mostly coontail and water milfoil, is con-
fined to the larger spring pools and occasional natural or manmade
cut-offs. Vascular plants are more abundant in large streams. Ex-
tensive beds of water willow border most of the riffles and many
pools. Submergent aquatics, mostly coontail, water milfoil, and
pondweeds abound along the margins of many large pools.

The streams that flow over igneous rocks in the St. Francois
Mountains present an interesting departure from the "typical"
Ozark stream. These streams are reexcavating old pre-Cambrian
valleys, and are floored for the most part by Cambrian sedimentar)'
rocks. But some streams transgress former divides and flow for
short distances over the highly resistant igneous rocks. Here they
have cut steep-sided gorges that are not much wider than the
stream channel. The gradient in these sections is steep, and the
stream descends across the igneous barrier in a series of turbulent
rapids. Such barriers are called "shut-ins" and they form a tempo-
rarv base level above which the streams flow in broad vallevs. A
classic example of such a stream is Stouts Creek in Iron County.

Streams along the northern and western border of the Ozark
Uplands likewise depart from my general description. These
streams are broadly transitional between typical Ozark streams and
those typical of the plains regions to the north and west, and exhibit
all degrees of intermediacy between the t\\'o basic stream types.

Dissected Till Plains

Location and boundaries. — The Dissected Till Plains, a subdi-
vision of the Central Lo\\'lands Physiographic Province, has been
characterized by Fenneman (1938:589) as ". . . a nearly flat till
plain, submature to mature in its erosion cycle, with a present
relief of 100 feet to 300 feet or even more; covered by loess, gen-
erally a few feet deep, but increasing to 30, 50, or even, in one
locality, 90 feet near the large bordering rivers." This region is
essentially coextensi\'e with the area covered by exposed Kansan
drift. It is bordered on the east by the Mississippi River and in-
cludes much of northern Missouri, southern Iowa, and eastern
Kansas and Nebraska. Its southern boundary approximates the
maximum limit of glaciation in Missouri, and it includes nearly
all of the state north of the Missouri River, as well as parts of
Jackson, Lafayette, Saline, and St. Louis counties south of the ri\er.

Topography. — That part of the state north of the Missouri

Missouri Fishes 251

River can be broadly characterized as a le\ el to rolling plain, with
a gentle slope to the south and east. The elevation ranges from
about 1200 feet in the extreme northwest to about SOO feet in the
uplands fronting on the Mississippi River. The flattest topography
in the region extends along the divide separating the Mississippi
and Missouri drainages, from near the town of Mexico northward
to the state line. This is the area referred to by Collier (1955:383)
as the Audrain Prairies. Local relief is seldom more than 100 feet,
and is usually much less. This flat area is an undissected remnant
of the Kansan Till Plain that formerly extended over the whole
region. To the east and south along the Mississippi and Missouri
rivers, and to the west along the Chariton River, this level topog-
raphy breaks off abruptly into belts of rolling to hilly country where
dissection has been extensive. Here the local relief is commonly
between 200 and 300 feet along the larger streams. Most of north-
western Missouri has a gently rolling to undulating surface that has
resulted from the deposition of loess.

Geology and soils. — The rocks underlying this region are mostly
Pennsylvanian in age, consisting mainly of alternating series of
shales and thin limestones. Mississippian and older rocks, consisting
mainly of limestones, occur at the surface along the Mississippi and
lower Missouri rivers. The general dip of the strata is to the west,
and east-facing escarpments tend to form on the more resistant
limestones. These escarpments are largely obscured by glacial drift,
which is 50 or more feet thick in many places. In general, the
drift thins southward; it is thickest in old preglacial valleys that
have not been re-excavated. Loess deposits reach their maximum
thickness along the two large boundary rivers, thinning rapidly
away from them. In places along the Missouri River these deposits
are at least 150 feet thick. Loess deposits are most extensive in
northwest Missouri.

The soils are entirely glacial and loessal in origin (Miller and
Krusekopf, 1929:26). Loess soils occur in broad bands along the
two large boundary rivers; the remainder of the region has soils
of mixed glacial and loessal origin. The particles in loess soils are
generally within the size range classified as silt and fine sand
rather than clay, and gravel or coarse sand is rarely present. The
soils of mixed origin contain considerable c^uantities of clay, as
well as coarse sand and gravel derived from glacial till. The mixed
soils often have heavy clay subsoils that are not readily penetrated
by water, but the loess soils are quite porous and readily free them-
selves of excess water.

252 University of Kansas Publs., Mus. Nat. Hist.

Drainage features. — Missouri north of the Missouri River is
drained by nearly parallel streams that flow eastward into the
Mississippi River or southward into the Missouri River. About the
eastern one-fourth of the region drains into the Mississippi by
way of the Des Moines, Fox, Wyaconda, Fabius, North, Salt, and
Cuivre rivers. All major streams in the Missouri drainage head
in southern Iowa and flow southward. From west to east the
principal streams are the Nishnabotna, Tarkio, Nodaway, Platte,
Grand, and Chariton rivers. From Boone County eastward, the
Missouri River receives only short, high-gradient streams from the
north, and only the extreme headwaters of these are in the Till
Plains. That part of the dissected Till Plains south of the Missouri
River is likewise drained only by small streams.

Natural vegetation and land use. — This region originally was
covered by prairie grasses, broken by bands of timber in the dis-
sected country along the major streams. Even the broad flood
plains of many of the larger streams were covered by wet prairies.
Wet prairies were especially extensive along the Grand and Chari-
ton rivers. The forests were similar in composition to those of the
Ozark Uplands, with less diversification of species (Steyermark,
1963:xxii). Nearly all areas formerly covered by native grasses are
now cultivated. Much of the timber has also been removed, but
large stands remain on some of the more deeply dissected areas.
Except for the Southeastern Lowlands, the Till Plains region is the
most extensively culti\'ated part of the state.

Aquatic Imhitats. — Generally, there is a gradient from west to
east and from north to south in the physical characteristics of the
streams of this region. The streams of extreme northwest Missouri
resemble those of the High Plains. Their waters are habitually
turbid except during periods of low flow, and their bottoms are
composed predominantly of fine shifting sand and silt. Streams
farther east are clearer, with a greater predominance of graxel and
rubble bottoms. In the Mississippi drainage a similar gradient is
apparent from north to south, with some tributaries of the Cuivre
River approaching Ozark streams in their characteristics.

The major streams of north Missouri occupy broad, flat \alleys
that in many places slope gradually and almost imperceptibly into
the surrounding uplands. Some of these, such as Grand and Chari-
ton rivers, occupy valleys that were excavated by much larger
streams in Pleistocene and pre-Pleistocene times. The valley of
Grand Ri\ er is three to five miles wide in places, and the valley of
Chariton River averages two miles in width. In their original con-

Missouri Fishes 253

dition these streams occupied extremely tortuous channels, and
in shifting back and forth across their broad valleys, they created
many sloughs and oxbow lakes. Nearly all of the streams have
been at least partially straightened, and most sloughs and oxbow
lakes have been drained. The streams now flow in channels that
are much shorter than formerly. Streams that formerly consisted of
a series of long, deep pools separated by short but well defined
riffles now are nearly uniform in depth, with little cover for fishes
and other aquatic life.

Silt, sand, and fine gravel are the predominant bottom types in
streams of the Dissected Till Plains. Bedrock is exposed in the
stream channels only in some upland tributaries that have cut
completely through the thick mantle of glacial debris, and in some
larger streams that transgress divides of the preglacial drainage.
The shales and heavy clay subsoils that underlie most of the region
are poor aquifers, and as a result, the low-flow potential of the
streams is poor. Only a few highly mineralized springs are present.
Erosion of the deep, fine-textured soils contributes large quantities
of fine sediments to runoff^, and the streams are habitually turbid.

Beds of water willow are found wherever rocky riffles provide
a stable subtrate, but otherwise higher aquatic plants are sparsely
developed. Many streams are bordered by heavy stands of maple
and other bottomland trees that form a nearly continuous canopy
over the stream. In late fall, large quantities of leaves are dropped
into these streams, staining the water and causing fish kills as a
result of local deficiencies in dissolved oxygen. This problem is
especially acute in years when stream levels are low and no heavy
rains come to flush the leaves away before the streams freeze over.

Osage Plains

Location and boundaries. — The Osage Section of the Central
Lowlands is an unglaciated plain of low relief stretching from Mis-
souri and Kansas south to Texas (Fenneman, 1938:605). It is the
smallest physiographic subdivision of Missouri, consisting of a
wedge-shaped area in western Missouri south of the Dissected Till
Plains and west of the Ozark Uplands. Its eastern boundary is
approximately at the edge of the Pennsylvanian rocks, and its
northern boundary marks the maximum southward penetration of
glacial ice into Missouri. The developmental history of this region
was similar to that of the Dissected Till Plains except for glaciation,
and the two regions have much in common.

Topograpluj. — This region is one of the smoothest in the state.

254 Unwersity of Kansas Publs.^ Mus. Nat. Hist.

Its most distinctive topographic feature is the Cherokee Lowland, a
flat, northeastward-trending plain extending into Missouri from
Kansas and Oklahoma. The elevation of the Osage Plains is about
the same as that of north Missouri, with a gentle slope to the south-
east. A maximum elevation of about 1150 feet is reached in Jackson
and Cass counties. The streams occupy broad, shallow valleys that
slope gently and in places almost imperceptibly into the flat to
gently rolling uplands. Where east-flowing streams penetrate the
Ozarks, some stream valleys change from three or four miles wide
and 50 feet deep to half a mile wide and 200 feet deep within a
few miles (Branson, 1944:355).

Geology and soils. — The rocks underlying the region are nearly
all Pennsylvanian in age, consisting of alternating beds of shales,
sandstones, and limestones having a gentle dip to the northwest.
From this structure a series of flat plains separated by east-facing
escarpments has been formed, with the plains underlain by soft,
easily erodable shales, and the escarpments marking the location
of the more resistant limestones and sandstones.

The soils are largely residual in origin, similar in character to
some of the soils of the Dissected Till Plains, but generally less
fertile because of their greater age and consequent greater leaching
(Miller and Krusekopf, 1929:52). For the most part, these soils
are fine textured and deep, with heavy clay subsoils that are quite
impervious to water.

Drainage features. — Drainage of the region is eastward into the
Missouri by way of the Blackwater and Osage rivers, and southwest-
ward into the Arkansas by way of the North Fork of Spring River.
Most of the central part is drained by the Osage River and two of
its tributaries, the South Grand and Marmaton rivers.

Natural vegetation and land use. — Where the prairie sod has
not been broken, the level uplands of the Osage Plains are occupied
by a prairie flora similar to that which formerly covered the uplands
of the Dissected Till Plains, but with an increasing proportion of
southwestern species to the south (Steyermark, 1963:xxiii). Wet
meadows and bottom prairies are found in undisturbed areas on
the flood plains of the major streams. The prairie flora has been
better preserved in this region than in any other section of the
state due to the maintenance of large tracts for hay production
and grazing. The prairie flora merges with the flora of the Ozarks
in a broad transition zone, with the prairie flora extending east-
ward along the upland di\ides, and the Ozark flora extending
westward in the belts of hilly countr\' along the major streams.

Missouri Fishes 255

Aquatic hahitats. — Sloughs and oxbow lakes abound on the
broad, flat flood plains of the major streams, and the streams them-
selves flow in narrow, steep-banked channels that are deeply in-
cised in the alluvium. The meandering character of the streams
is well shown by the fact that construction of a 20-mile ditch
eliminated meander detours aggregating 50 miles in one section
of Osage River (Bretz, 1965:107). Channelization in this region
has been as extensi\'e as that north of the Missouri River.

The clays and shales underlying the region are poor aquifiers,
and only a few highly mineralized springs are present. The streams
are as a consequence intermittent and turbid, but perhaps are
slightly less turbid than those of the glaciated region to the north.
The principal bottom types are silt and sand in the pools, and flat,
slate-like pieces of shale and sandstone on most of the riffles. The
pools are long and sluggish, and the riffles are poorly defined. The
streams are on the whole much like those of the Dissected Till
Plains.

The Missouri and Mississippi Rivers

The character of these two large streams as a habitat for fishes
is not determined to any marked degree by local conditions, but
rather is a reflection of conditions over all of the areas that they
drain before they enter Missouri. For this reason it is convenient
to consider them in a separate section.

The Missouri River originates at the confluence of the Gallatin,
Madison, and Jefferson rivers in southwestern Montana and flows
for approximately 2500 miles in a easterly and southeasterly di-
rection, joining the Mississippi River just above St. Louis, Missouri.
It is the longest river on the North American Continent, and drains
an area of about 525,000 square miles, or about one-sixth of the
continental United States.

At its headwaters the Missouri River is a clear mountain stream.
As it flows eastward and southeastward it receives silt-laden tribu-
taries from the badlands of Wyoming and the Dakotas, and farther
downstream its silt load is augmented by drainage received from
the intensi\ely cultivated prairie regions of the Dakotas, Nebraska,
Iowa, Kansas, and Missouri. As a consequence it is one of the most
turbid large streams on the continent. Most of the region drained
by the Missouri River is arid or semiarid, and consequently it con-
tributes only about 12% of the total flow of the Mississippi, al-
though it drains about one-third of the Mississippi Basin. The Mis-
souri River was formerly subject to wide seasonal fluctuations in
volume of flow. Two periods of high water occurred each year

256 University of Kansas Publs., Mus. Nat. Hist.

(Neel et al, 1963:7). One of these came in early spring and re-
sulted from the melting of snow in the prairie region; the other
came in June and resulted from the melting of snow in the moun-
tains and rains in the prairie region. Construction of several large
navigation and flood control reservoirs on the upper Missouri has
reduced peak discharges and stabilized the flow during the navi-
gation season.

The lower 570 miles of the Missouri River lie entirely within,
or form the western boundary of, Missouri. From the Iowa state
line downstream to Glasgow, the river flows through an area of
easily-erodable shales, and it occupies a valley ranging from four
to 10 miles in width. This broad valley is bordered mostly by
rounded loess-capped hills with little bluff formation. At Glasgow
the valley narrows abruptly to a width of about 2.5 miles, and from
that point to its mouth the river occupies a deep trench carved
from more resistant limestone. In this section the valley is bordered
by steep hills and sheer bluffs, capped in places by deep deposits
of loess.

In its original condition the Missouri River occupied one or
more wide, shallow channels. The location of these channels
changed with every major flood, and bank erosion occurred con-
tinuously. Wing dikes and revetments have been installed in recent
years, and channel shifting and bank erosion have been almost
eliminated. The river is now confined to a single relatively deep
channel, and most of the quiet backwater areas that formerly were
present have been eliminated. Fine shifting sand is the principal
bottom type in the river, with fine gravel in sections of the main
channel where the current is strong, and deep deposits of flocculent
silt in protected areas along the shore.

The volume of flow of the Missouri River more than doubles as
it passes through Missouri. The average discharge at St. Joseph
for the 34-year period from 1928 to 1962 was 37,050 cubic feet per
second (cfs) (U. S. Geol. Surv., 1962:32), only Al% of the average
discharge at Hermann, 350 miles downstream. At Hermann, the
average discharge during the 65-year period from 1897 to 1962
was 79,570 cfs. The peak discharge during this period was 676,000
cfs, and the minimum was 4200 cfs. Berner (1951:4) reported
turbidity readings as high as 8000 parts per million (ppm) during
high water periods along the lower Missouri River in 1945, and
he found that the average turbidity in late summer and autumn
was 1700 ppm. Large reservoirs constructed in recent years on the
upper Mis.souri are effecting marked reductions in turbidity all the

Missouri Fishes 257

way to the river mouth (Neel et al., 1963:31 and table 7). Prior
to 1953, average annual turbidities at Kansas City were seldom less
than 1500 ppm and often exceeded 2500. Since 1953, average an-
nual turbidities at this point on the river have not exceeded 900.
Current velocities in the main channel during periods of normal
water level range from two to five miles per hour ( U. S. Corps of
Engineers, 1957). The average gradient in the Missouri section
of the river is 0.88 feet per mile, a rather high gradient for so large
a stream.

The Mississippi River originates in northern Minnesota and
flows approximately 2400 miles in a southerly direction to the Gulf
of Mexico. It drains an area of about 1,245,000 square miles. It
forms the entire eastern border of Missouri, a distance of about
494 miles. From the Iowa state line to Cape Girardeau it occupies
a \alley similar to that of the lower Missouri River. At Cape Gi-
rardeau it enters a broad allu\ial plain which extends all the way
to the Gulf of Mexico. Along the eastern border of Missouri the
Mississippi is joined by its two major tributaries, the Missouri and
Ohio rixers. These two tributaries have a marked effect on the
character of the main stream.

Above the mouth of the Missouri, the Mississippi is a relatively
clear stream. In its original condition, it consisted of a series of
deep pools separated by shallow bars and rapids ( Carlander, 1954 :
20), and it was a much more placid stream than the Missouri. The
contrast between the two streams has been heightened by the con-
struction of a series of navigation dams along the upper Mississippi
River. Six such dams are found in the Missouri section of the
river. These have converted the river into a series of lakes, but
areas with considerable current still occur immediately below the
dams. Such areas have a relatively silt-free sand, gravel or rubble
bottom, while the bottoms of the navigation pools are largely silt.
Extensive growths of aquatic vegetation are found in the numerous
backwaters of the pools.

It is noteworthy that the upper Mississippi, with a drainage
area only 32% as large as that of the Missouri, has an 18% greater
average discharge. The total drainage area of the upper Mississippi
River is 171,500 square miles. The discharge for the 35-year period
from 1927 to 1962 ranged from 37,400 cfs to 337,000 cfs, and aver-
aged 93,910 cfs (U. S. Geol. Surv., 1962:28).

Below the mouth of the Missouri, the Mississippi River takes on
much of the character of the Missouri River, but is a much larger
stream. Like the Missouri, it has not been impounded and has a

258 Unr'ersity of Kansas Publs., Mus. Nat. Hist.

swift current. The turbidity is high, but is somewhat less than
that of the Missouri. Plattner (1946:16) reported an average tur-
bidity of 1880 ppm in this section, whereas the turbidity above the
mouth of the Missouri averaged only 300. Fine sand and gravel
are the predominant bottom types, but rubble is not uncommon
where the stream encounters the valley wall. Bedrock and shingle
bottoms occur just below Cape Girardeau. Silt is the most common
bottom t}'pe in backwater areas.

The Ohio River is even clearer than the Mississippi River above
its junction with the Missouri. Although the Ohio drains only
214,000 square miles, 30% as large an area as the Mississippi River
at the mouth of the Ohio, its average discharge is 50% greater.
Thus, the flow of the Mississippi River more than doubles at the
mouth of the Ohio Ri\'er, and dilution by the clear waters of the
Ohio greatly reduces the turbidity. On August 13, 1963, I obtained
a turbidity reading of only 12 ppm in the Ohio River at its mouth,
whereas the turbidity of the Mississippi River immediately up-
stream was 130. A reading of 80 was obtained in the Mississippi
Ri\er 33 miles below the mouth of the Ohio, and at Cottonwood
Point, 129 miles below the mouth of the Ohio, the turbidity was
only 20. In the Mississippi River downstream from the mouth of
the Ohio River, sand and gravel are the predominant bottom types
in the main channel, with silt in the backwater areas. Bottom ma-
terials larger than coarse gravel seem to be entirely absent.

The Fish Faunal Regions of Missouri

Few Missouri fishes are statewide in distribution, and none is
uniformly distributed over the state. The distributions of many
species are similar, and certain distribution patterns are repeated
over and over with slight modification. Conversely, many species
have distributions that are largely or entirely complementary.
Given these basic facts, it should be possible to delimit major fish
faunal regions in Missouri, to indicate the kinds of fishes that char-
acterize these regions, and to make generalizations concerning the
factors responsible for the distribution patterns of the fishes. In
the present section an attempt will be made to delimit fish faunal
regions using two relatively independent types of analysis. The
species characterizing these regions and the factors responsible for
their distribution patterns will be treated in succeeding sections.

Range Limit Analysis

This analysis is based on the tenet that faunal regions are areas
where few range limits occur, whereas the boundaries between

Missouri Fishes 259

these regions are marked b}' zones where many speeies reaeli some
hmit of then- natural range. For this analysis, spot distribution
maps w ere used. Range limits were indicated on each species map
by lines drawn to enclose all known records. Where a species
clearly existed as two or more broadly disjunct populations, this
was indicated by drawing lines around the separate areas of occur-
rence. A transparent oxerlay marked off into quadrats correspond-
ing to an area of eight square miles was placed over the species
maps, one at a time, and the number of range limits occurring in
each quadrat was determined. A map was then prepared, using
various kinds of shading to correspond to the number of range
limits occurring in a given quadrat ( Fig. 4 ) . Faunal boundaries,
as indicated by high concentrations of range limits, were fitted
using a map on which all range limits had been drawn as a guide.
This is essentially the method used by Hagmeier and Stults ( 1964 )
for delimiting the mammal provinces of North America, except
that they employed larger quadrats and plotted an "index of fau-
nistic change" rather than the actual numbers of range limits. The
"index of faunistie change" corresponds to the percentage of spe-
cies occurring within a given quadrat whose ranges end there.
I did not use this index because I believe that the actual number
of range limits occurring in a given area is itself a valuable piece
of information, and estimating the number of species occurring in
each of the 934 quadrats used in the present analysis would have
carried with it an element of uncertainty that might supersede any
added precision resulting from use of the index.

The faunal boundaries indicated by this method are of three
kinds. One kind is relatively independent of drainage patterns and
corresponds rather closely to the boundaries between major physio-
graphic regions. Two such boundaries are indicated. Tlie shortest
and sharpest of these corresponds closely to the physiographic
boundary between the Ozark Uplands and the Southeastern Low-
lands. This boundary breaks off more sharply on the side towards
the low^lands, and is broadest along the major Ozark streams enter-
ing the lowlands. An upstream gradient indicating a gradual reduc-
tion in the number of range limits per quadrat occurs along these
streams. This reflects the penetration of lowland species into the
uplands along the major streams, and indicates only a slight down-
stream penetration of upland species. The second boundary of
this kind approximates the physiographic boundary separating the
Ozark Uplands from the Osage Plains and Dissected Till Plains.
From the Kansas state line in northern Jasper County northeast-

260

University of Kansas Publs., Mus. Nat. Hist.

Rangelimits Oyodrat

FiGXJRE 4. Boundaries of fish fauna! regions, as indicated by range-limit
analysis (number of range-limits in quadrats 8 square miles in area).

ward to the Osage Rixer in central Benton County this boundary
is fairly narrow and sharp, being marked by a zone in which most
quadrats had 20-29 range limits. From Benton County northeast-
ward to the Missouri River in Moniteau County it is broader, and
is marked by quadrats having 10-19 range limits. From south-
eastern Boone County to the point where it recrosses the Missouri
Ri\er in southeastern Warren County, this boundary is near, and
parallels, the Missouri River.

A second kind of faunal boundary indicated by this method
follows the courses of major ri\ers, most notably the Missouri and
Mississippi. This kind of boundary results from the occurrence
in these large rivers of certain fishes that rarely penetrate into
smaller streams. The zones of highest concentration of range limits
occur along the lower one-third of the Missouri River, and along
the Mississippi River from just abo\'e the mouth of the Meramec
River downstream to southern Cape Cirardeau County. These high
values are due not to a greater concentration of big river fishes in
these sections, but rather to a reinforcing effect resulting from

Missouri Fishes 261

superimposing the range limits of fishes characteristic of major
physiographic regions o\'er those of the big ri\'er species. The
lower Missouri River is close to the boundary between the Ozark
Uplands and Dissected Till Plains, and several species characteristic
of one or the other of these physiographic regions reach the north-
ern or southern limit of their distribution near the river. The re-
inforcing effect along the Mississippi River results because certain
species that are otherwise confined to the plains or lowland physio-
graphic regions penetrate north or south along the river. The
boundary paralleling the Osage River may be accounted for in a
similar manner. The fewest range limits are along the upper Mis-
souri and the lower Mississippi rivers, where species otherwise
confined to large rivers penetrate into the plains and lowland
physiographic regions.

The third kind of faunal boundary indicated by this method
follows major drainage di\'ides. Only one principal boundary of
this kind is indicated. From southwestern Dade County eastward
to central Texas County it follows the divide separating streams
draining the northern and southern slopes of the Ozark Uplands.
In Texas County it turns northeastward and approximates the di-
vide separating the Gasconade and Meramec stream systems.

Species Composition Analysis

For this analysis, species were assigned to groups based on their
distribution patterns and centers of abundance in Missouri (Table
2). Four primary groups were recognized: "Ozark," "lowland,"
"prairie," and "big river." Species of these groups have their distri-
butions centered in either the Ozark Uplands, the Southeastern
Lowlands, the Osage and Dissected Till Plains, or the Missouri
and Mississippi rivers, respectively. Two secondary groups ("Ozark-
lowland" and "Ozark-prairie") were recognized for species that
were approximately equally abundant and widely distributed in
two of the primary areas. A "wide-ranging" group \\'as recognized
for species more widespread than those in the other groups listed
abo\'e. Few species in the wide-ranging group are truly state-wide
in distribution, but most occur at least occasionally in all sections
of the state. Four species (northern pike, goldfish, pumpkinseed,
and yellow perch) were left unclassified, because they are ex-
tremely rare in Missouri, and did not occur in the collections used
in the analysis. All other species of Missouri fishes were included in
one of the seven groups listed in Table 2.

The criteria for determining centers of abundance require some

262

Unwersity of Kansas Publs., Mus. Nat. Hist.

Table 2. — A Classification of Missouri Fishes Based on Distribution Patterns

and Centers of Abundance

Northern brook lamprey
Southern brook lamprey
American brook lamprey
Least brook lamprey
Alabama shad
Rainbow trout
Chain pickerel
Southern red]:;elly dace
Hornyhead chub
Bigeye chul)
Streamline chub
Gravel chub
Rosyface shiner
Telescope shiner
Bleeding shiner
Duskystripe shiner
Striped shiner
Wedgespot shiner
Bigeye shiner
Steel color shiner
Whitetail sliiner

Spotted gar
Bowfin

Pugnose minnow
Riblion shiner
Ironcolor shiner
Weed shiner
Pallid shiner
Blacktail shiner
Sabine shiner
Taillight shiner
Cypress minnow
Central silvery minnow
Bullhead minnow

Ozark

Spotfin shiner
Bluntface shiner
Blacknose shiner
Ozark shiner
Ozark minnow
Silverjaw minnow
Slim minnow
Largescale stoneroUer
Highfin carpsucker
Northern hog sucker
Black redhorse
Golden redhorse
Silver redhorse
River redhorse
Creek chubsucker
Slender madtom
Ozark madtom
Checkered madtom
Ozark cavefish
Southern cavefish
Northern studfish
Plains topminnow

Lowland

Spotted sucker
Lake chubsucker
Brown bullhead
Tadpole madtom
Freckled madtom
Brindled madtom
Pirate perch
Golden topminnow
Starhead topminnow
Mosquitofish
Warmouth
Bantam sunfish
Spotted sunfish

Smallmouth bass
Redear sunfish
Rock bass
Bluestripe darter
Longnose darter
Logperch
Gilt darter
Channel darter
Missouri saddled darter
Arkansas saddled darter
Banded darter
Greenside darter
Niangua darter
Yoke darter
Stippled darter
Rainbow darter
Arkansas darter
Orangethroat darter
Fantail darter
Least darter
Mottled sculpin
Banded sculpin

Flier

Banded pygmy sunfish
Dusky darter
Stargazing darter
Crystal darter
Western sand darter
Scaly sand darter
Bluntnose darter
Harlequin darter
Mud darter
Slough darter
Cypress darter

Suckemiouth minnow
Common shiner
Bigmouth shiner
Red shiner
Sand shiner
Topeka shiner

Prairie

Ghost shiner
Brassy minnow
Fathead minnow
Quilll)ack
River caipsucker
Neosho madtom

Stonecat
Trout perch
Plains killifish
Orangespotted sunfish
Blackside darter
Redfin darter

Missouri Fishes

263

Sil\er lamprey
Lake stui"<i;e()n
Shovelnose sturgeon
Pallid sturgeon
Paddlefish
Alligator gar
Shortnose gar
Skipjack herring
Threadfin shad
Goldeye
Siher chub

Grass pickerel
Northern mimic shiner
Blackspotted topminnow

Creek chub
Central stoneroller

Chestnut lamprey
Longnose gar
Gizzard shad
Mooneye
Carp

Golden shiner
Redfin shiner
Bluntnose minnow

Table 2. — Concluded

Big River

Speckled chub
Flathead chub
Sturgeon chub
Sicklefin chub
Emerald shiner
SiKerband shiner
Spottail shiner
River sliiner
Channel mimic shiner
Plains minnow

Ozark-lowland

Brook siherside
Spotted bass

Ozark-prairie

White sucker
Northern redhorse

Wide-ranging

Bigmouth buffalo
Smallmouth buffalo
Black buffalo
Black bullhead
Yellow bullhead
Channel catfish
Flathead catfish
American eel

Western sihery minnow-
Blue sucker
Blue catfish
Burbot
White bass
Yellow bass
Sanger
River darter
Drum
Mississippi silverside

Longear sunfish
Speckled darter

Slenderhead darter
Johnny darter

Blackstripe topminnow
Largemouth bass
Green sunfish
Bluegill
Black crappie
White crappie
Walleye

explanation. Little information is available on the absolute abun-
dance of fishes in various parts of the state, so species were as-
signed in part on the basis of frequency of occurrence in collec-
tions from each of the primary areas, and in part on the basis of
a subjective appraisal of abundance in the four areas.

After the group assignments were made, the number of species
belonging to each group was determined for representative fish
collections from all parts of Missouri. The numbers thus obtained
were converted to percentages of the total species in the collections,
and the results are depicted in Figs. 5-11 (see pp. 265-268). A single
faunal group is illustrated on each map. The areas of the circles
are proportional to the percentage that group comprised of the

264 University of Kansas Publs., Mus. Nat. Hist.

total species in collections from each of the 599 localities used in
the analysis.

One point nicely illustrated by this method is the largely com-
plementary distribution patterns of the species that characterize
the four primary faunal groupings (Figs. 5-8). Another is the
rather close correspondence between the faunal boundaries as indi-
cated by this method and those indicated by range limit analysis
(Fig. 4). No faunal boundary is indicated along major stream
divides in the Ozark Uplands by the present analysis, but this is
not surprising, since no distinctions were made in the faunal group-
ings that would permit such a separation. One fact illustrated by
this method but not by range limit analysis is the manner in which
the species characteristic of a particular faunal area are distributed
elsewhere in the state. The fauna characteristic of the Ozark Up-
lands, for example, comprises a small but distinct element of the
fauna in the eastern part of the Dissected Till Plains. Some ele-
ments of the lowland fauna occur also in a broad southwest to
northeast-trending band from the Neosho River system into the
Salt River and other streams in northeastern Missouri. The prairie
fauna exhibits a distinct faunal break that nicely complements that
shown by the Ozark fauna, but certain prairie species penetrate
well into the Ozark Uplands along its western and northern border.
Some elements of the prairie fauna also occur in the Southeastern
Lowlands. The big river fauna likewise penetrates into the South-
eastern Lowlands, and makes up a significant proportion of the
total species in tributary streams of the northwestern Dissected
Till Plains. These distribution patterns are indicative of the in-
fluence of certain enxironmental factors on fish distribution, and
their significance will be discussed in the section on environmental
factors.

The Ozark-lowland faunal group recognizes approximately the
same boundary along the northern and western periphery of the
Ozarks as the Ozark and prairie species, but continues without a
break into the lowlands. Species of the Ozark-prairie faunal group
occurred only rarely in the Southeastern Lowlands, but are rather
uniformly distributed over the Ozark and plains physiographic
regions.

Fish Faunal Regions

Recognition of four primary faunal regions (Ozark, lowland,
prairie and big river) permits generalizations concerning the fac-
tors responsible for the distribution patterns of Missouri fishes, but
the fixing of precise boundaries between these regions is in some

Missouri Fishes

265

FIG. 5 OZARK FAUNAL GROUP

FIG. 6 LOWLAND FAUNAL GROUP

266

University of Kansas Publs., Mus. Nat. Hist.

FIG.7 PRAIRIE FAUNAL GROUP

Missouri Fishes

267

respects undesirable. To do so obscures certain of the interesting
faunal relationships touched on in the preceding section. Further-
more, the faunal regions are separated for the most part by transi-
tion zones rather than by sharp boundaries. Nevertheless, there
are areas of rather abrupt faunal change, and these deserve to be
recognized. Fig. 12 illustrates the more important aspects of all
these interesting relationships. The transition zones are indicated
by an "Ozark border" where fishes characterizing the Ozark, low-
land and prairie faunal regions meet and mix in complex fashion.
The outer limits of the Ozark border encompass the area in which
Ozark species comprise more than 10 per cent of the total species
present in most collections. Conversely, the inner limits of the
Ozark border define the areas in which prairie or lowland species,
or both, comprise more than 10 per cent of the species in most
collections. The actual species composition of the Ozark border
varies markedly from one area to another, and it is best thought
of as a broad ecotone rather than a distinct faunal region. The
primary faunal regions likewise exhibit some internal variability
in species composition, but each is characterized by widespread
and abundant species that lend a fundamental unity not found in

268

University of Kansas Publs., Mus. Nat. Hist.

Missouri Fishes

269

Figure 12. Fish faunal regions of Missouri. Transition zones are indicated
by an "Ozark l)order"; "lines of best fit" separate areas in which species of a

given faunal region predominate.

the Ozark border. This point will be discussed in more detail in
the section on environmental factors.

At any given locality within the Ozark border, species charac-
teristic of one of the primary faunal regions usually predominate,
and this has been indicated in Fig. 12 by "lines of best fit." Within
these lines, species characteristic of the Ozark faunal group usually
predominate, both in terms of numbers of species and numbers of
indi\'iduals; beyond these lines, species characteristic of the prairie
or lowland faunal groups usually predominate. The "lines of best
fit" approximate the faunal boundaries indicated by range limit
analysis and the zones of abrupt faunal change indicated by species
composition analysis.

No attempt has been made to indicate transition zones between
the big river and other faunal regions in Fig. 12. Transition zones
where the Missouri and Mississippi rivers receive no major tribu-
taries are generally narrow and abrupt, whereas those between
the two rivers and their larger tributaries are broad and gradual.

270 University of Kansas Publs., Mus. Nat. Hist.

It would have been desirable to illustrate this point in Fig. 12,
but the added complexity would have obscured the relationships
previously discussed. The limits of the big river faunal region as
shown in Fig. 12 therefore represent "lines of best fit" within
which big river species comprise 25 per cent or more of the total
species in most collections.

Two previous attempts have been made to recognize biogeo-
graphic regions in Missouri. Bennett and Nagel (1937:15-21)
recognized four zoogeographic regions (Northern Glacial, Western
Prairie, Ozark Highland, and Mississippi Lowland), which are co-
extensive with the major ph)'siographic regions of the state. They
divided these into a number of "game ranges." They recognized
a "Western Ozark Border" and a "Northern and Eastern Ozark
Border" within the Ozark Highlands, but these bear little relation
to the Ozark border recognized in the present study. Steyermark
(1963:xviii-xxiv) recognized three primary plant regions (Prairie,
Ozark, and Southeastern Lowlands), the boundaries of which cor-
respond in a general way with the prairie, Ozark, and lowland
fish faunal regions of the present study. The latter three fish faunal
regions correspond rather closely to the Missouri portions of the
Illinoian, Carolinian, and Austroriparian biotic pro\'ices of Dice
(1943). Biotic provinces were delimited primarily on the basis of
terrestrial biotas, but the present study suggests that they apply as
well to many fishes.

Distribution of Missouri Fishes in Relation to Environment

The fish fauna of Missouri is diverse, including nearly 200
species. This diversity stems in part from the variety of aquatic
environments within the state, and in part from exents that occurred
during the developmental histoiy of the fauna. The environmental
and historical factors are not entirely independent, but for purposes
of discussion it is convenient to consider them separately. Environ-
mental factors will be discussed in the present section; historical
factors will be discussed in the sections that follow.

The discussion of fish distribution in relation to the environment
is hampered by a lack of specific information concerning the en-
vironmental tolerances and responses of \arious fishes, and of the
range of environmental conditions occurring in different habitats.
Furthermore, the environmental factors are not independent \ari-
ables, and where the patterns of variation for two or more factors
are correlated, it is rarely possible to demonstrate that one factor
is of overriding importance in controlling the distribution of a given

Missouri Fishes 271

species. No doubt for most species many physical and biological
factors are in\olved, interacting in a complex fashion. Because of
these difficulties, suffice it in the present discussion to point out
correlations between the distribution patterns of the fishes and
patterns of variation in the more obvious features of the environ-
ment, realizing that in some instances more subtle and less readily
obser\ed factors may have greater importance.

The manner in which the range limits of numerous fishes cut
across major stream systems and parallel the boundaries of physio-
graphic regions is the most noteworthy feature of fish distribution
in Missouri. The physiographic boundaries separate areas in which
the aquatic habitats are quite different; probably the distribution
patterns of these fishes represent true responses to environmental
gradients. The occurrence of certain fishes in two habitat areas
(but not in others) suggests fundamental similarities in the habi-
tat areas having species in common. Analysis of the relationships
outlined above provides a means for gaining some insight into the
role of environmental factors in controlling fish distribution. For
this analysis it is convenient to consider species with similar distri-
butions together, and the classification adopted in Table 2 and
the maps resulting from the numerical analysis of the preceding
section serve this purpose very well.

Ozark Faunal Group

The largest and most distinctive of the faunal assemblages is
that having its distiibution centered in the Ozark Uplands ( Fig. 5 ) .
Sixty-five species, just over a third of the total fauna, are included.
Fifty-two of these rarely occur more than a few miles beyond the
boundaries of the Ozarks in Missouri, and 13 are endemic to the
Ozark Uplands of Missouri and adjacent states. Regional differen-
tiation exists in the fish fauna of the Ozark Uplands, but its funda-
mental unity is indicated by the distribution patterns of such species
as the redbelly dace, rosyface shiner, Ozark minnow, northern hog
sucker, black redhorse, banded darter, greenside darter, and banded
sculpin. \\'ithin Missouri, these species occur throughout the Ozark
Uplands, but rarely occur beyond its boundaries. Other character-
istic and abundant Ozark species are nearly as widespread, being
absent only from one or two of the principal stream systems. The
distribution patterns of these species are likewise strongly corre-
lated with the boundaries of the Ozark Uplands. Included are the
bigeye chub, largescale stoneroller, northern studfish, rainbow
darter, and mottled sculpin. An excellent example is presented by

272 University of Kansas Publs., Mus. Nat. Hist.

the closely related or conspecific bleeding and duskystripe shiners,
two of the most common and characteristic Ozark fishes. Neither
occurs over all of the Ozark Uplands, but one or the other occurs
in each of the principal stream systems, and together their distri-
bution marks the boundaries of the region with considerable pre-
cision.

A strong correlation exists between the distribution of fishes
of the Ozark faunal group and bedrock geology; these fishes are
confined primarily to portions of Missouri underlain by rocks that
are Mississippian in age or older. Most show a definite break in
distribution along the contact zone between the Pennsylvanian
shales and sandstones underlying the Osage Plains and the Missis-
sippian limestones underlying the southwestern Ozarks. This cor-
relation gives some insight into the factors controlling the distribu-
tion of fishes in the Ozark faunal group. The sandstones and shales
are poor aquifers, and the soils derived from them are predomi-
nantly fine-textured silt loams with heavy clay subsoils. In contrast,
the highly soluble limestone bedrocks are good aquifers, and the
soils derived from them are predominantly coarse-textured gravelly
or stony loams without heavy clay subsoils. Reflecting these dif-
ferences in bedrock and soils, streams draining the limestone re-
gions are clearer and maintain a stronger base flow than those
draining the shale regions. Also, the large quantities of coarse,
angular chert derived from the limestone bedrocks results in a
bottom type quite different from that derived from the laminated
shales. It is logical to conclude that as a group the Ozark species
are quite intolerant of siltation and continuous high turbidity, and
require streams having permanent flow and a predominance of
coarse gravel or rock bottoms.

Some Ozark species are less tolerant than others with regard
to these factors, avoiding drainages where shales and sandstones
occur in the undissected uplands, even though the streams them-
selves have cut down into the underlying limestones. This rela-
tionship can be demonstrated by comparing the fish fauna of the
Bourbeuse River with that of the other two subsystems (Meramec
main stem and Big River) of the Meramec Basin. Although the
fish fauna of the Bourbeuse drainage is Ozarkian in character, it
has peculiarities that correlate with the occurrence of Pennsyl-
vanian shales under much of the undissected uplands within its
watershed (Fig. 3). Three minnows (bleeding shiner, wedgespot
shiner and Ozark minnow) that are among the most abundant and
ubiquitous fishes elsewhere in the Meramec system are entirely

Missouri Fishes 273

absent from the Bourbeuse drainage. The blaekspotted topniinnow
is abundant over most of the Meramec system, but is replaced in
the Bourbeuse drainage by the blaekstriped topminnow. Several
minnows whose distributions suggest a preference for warm, quiet
waters and a tolerance for moderate amounts of turbidity (redfin
shiner, bigeye shiner and bluntnose minnow) are much more
abundant in the Bourbeuse drainage than elsewhere in the Mera-
mec system.

The only significant area of occurrence for Ozark species out-
side of the Ozark Uplands is in the eastern part of the Dissected
Till Plains. Here direct tributaries of the Mississippi and lower
Missouri rivers have cut through the Pennsylvanian shales and
sandstones into the underlying limestones. Extensive areas in the
undissected uplands are still underlain by Pennsylvanian shales,
and the streams as a result are more turbid and have lower base
flows than those of the Ozark Uplands. Ozark species occurring
in the eastern part of the Dissected Till Plains, and therefore
thought to have broader environmental tolerances than their asso-
ciates, include the bigeye shiner, golden redhorse, slender madtom,
smallmouth bass, logperch, fantail darter, and orangethroat darter.
The occurrence of these and other Ozark species in this area, along
with the more abundant prairie species, is responsible for its in-
clusion in the Ozark border (Fig. 12).

No species of the Ozark faunal group occurs frequently in the
Southeastern Lowlands. The rock bass has a wider distribution
there than other Ozark species, but even it is uncommon. Bottom
type and current velocity are probably important factors controlling
the distribution of Ozark species in the lowlands. The largescale
stoneroller and banded darter, for example, are known in the low-
lands only from a single rubble riflfle of a swift-flowing ditch at the
foot of Crowley's Ridge.

The faunas of various stream systems in the Ozark Uplands are
more strongly differentiated than those elsewhere in Missouri.
Many fishes are widespread and abundant in certain drainages
but are absent from others. Examples are: streamline chub, tele-
scope shiner, whitetail shiner, bluntface shiner, Ozark shiner, Ozark
madtom, checkered madtom, bluestripe darter, Missouri saddled
darter, Arkansas saddled darter, and yoke darter. Other species
that are more widespread are represented by phenotypically dis-
tinguishable populations in different stream systems. Among these
are the rosyface shiner, slim minnow, slender madtom, rock bass,

274 Unwersity of Kansas Publs., Mus. Nat. Hist.

longear sunfish, greenside darter, fantail darter, and orangethroat
darter.

The rather marked regional differentiation characterizing the
faunas of the principal Ozark stream systems is due in large part
to their remoteness with respect to avenues of dispersal for fishes.
But more than mere distance is invoKed. It seems likely that the
large, connecting ri\ ers that could potentially serve as avenues for
dispersal are instead effective barriers for species adapted to Hfe
in clear, upland streams. Where upland tributaries enter the large
rivers with sufficient frequency, waif dispersal from one habitable
tributary to another is possible. But where the large rivers flow
for long distances through lowland areas, the potential for dispersal
in this manner is virtually nonexistent. The lower reaches of all
the principal streams on the southern slope of the Ozark Uplands
are in the lowlands of the Mississippi Embay ment, and at present
must be effective barriers to upland fishes. This may account for
the stronger differentiation between the faunas of various stream
systems on the southern slope of the Ozarks compared to those of
the northern slope. Thus, the Arkansas saddled darter, fantail
darter, and orangethroat darter are represented by phenotypically
distinguishable populations in the White and Black river systems,
whereas these species or their counterparts in the Meramec and
Gasconade systems do not exhibit a similar degree of differentiation.

Lowland Faunal Group

Thirty-eight species, or approximately 20 per cent of the Mis-
souri fish fauna, have their distribution centered in the Southeastern
Lowlands ( Fig. 6 ) . Twenty-five of these are either confined to the
lowlands, or occur elsewhere in Missouri only as strays. Many
reach the northern limit of their range in southeastern Missouri,
occurring southward in the Mississippi Embayment to the Gulf of
Mexico. Some are rare and localized in distribution in Missouri,
but the fundamental unity of the lowlands as a faunal region is
indicated by the distribution of such common and generally dis-
tributed species as the ribbon shiner, weed shiner, blacktail shiner,
pirate perch, dusky darter, and cypress darter.

As a group, lowland species resemble Ozark species in being
rather intolerant of siltation and continuous high turbidity. All
are characteristic inhabitants of standing waters or of sluggish
streams and ditches having bottoms composed mostly of sand,
fine gravel, and organic debris. Most are associated with, or at
least tolerant of, dense growths of submcrgent aquatic vegetation.

Missouri Fishes 275

The distribution of U)\\ land fishes is perhaps more elosely eorrclatcd
\\ ith topographie rehet and stream gradient than with other obvious
physical features of the environment. One indication of this is the
penetration of the lowland fauna into the Ozark Uplands along
the low-gradient lower sections of major streams entering the
lowlands. Another is the presence of certain species characteristic
of the lowlands (bullhead minnow, tadpole madtom, freckled
madtom, mosquitofish, bluntnose darter, and slough darter) north-
ward along the floodplain of the Mississippi River and in a broad
northeastward-trending band across central Missouri from the
Neosho system on the southwest to the Salt River and other nearby
streams on the northeast. The latter area includes the Cherokee
Lowland and the Audrain Prairies, two of the most extensive areas
of nearly level terrain outside of the Southeastern Lowlands. Per-
haps these species would be more widespread westward in the
Dissected Till Plains if the streams were less silty and turbid. Pos-
sibly some lowland species were more widespread in that region
before the prairie sod was broken and the streams were straight-
ened.

The fish fauna of the Southeastern Lowlands must have been
profoundly altered when the extensive swamps of that area were
ditched and drained. Probably those species characteristic of
standing waters (pugnose minnow, ironcolor shiner, pirate perch,
pygmy sunfish, flier, and slough darter) were formerly more abun-
dant than they are today. Other species that are now widespread
in the flowing waters of the ditches ( emerald shiner, ribbon shiner,
blacktail shiner, bullhead minnow, brindled madtom, spotted bass,
stargazing darter, and dusky darter) were originally confined to
the main channels of the Little, St. Francis, and Black rivers.

Prairie Faunal Group

Prairie species comprise the smallest of the primary assemblages
in the Missouri fish fauna (Fig. 7). Only 18 species, or approxi-
mately 9 per cent of the total fauna, are included. Ten of these
are confined almost entirely to the plains physiographic regions,
occurring elsewhere in Missouri only as strays. Few distinctions
exist between the fish faunas of the Osage Plains and Dissected
Till Plains, so the two physiographic regions are not distinguishable
faunistically. The bigmouth shiner is the only fish whose distri-
bution seems to be correlated with the glacial boundary. This fish
occurs only rarely south of the Missouri River, which marks the
approximate southern limit of glaciation in Missouri. The funda-

276 University of Kansas Publs., Mus. Nat. Hist.

mental unity of the two plains regions as a faunal region is indicated
by the distribution of such species as the suckermouth minnow,
red shiner, sand shiner, fathead minnow, and river carpsucker.

As a group, the prairie species seem to have broader ecological
tolerances than those in the preceding two groups. Some of the
more abundant of them (suckermouth minnow, red shiner, sand
shiner, quillback carpsucker, and orangespotted sunfish) penetrate
well into the Ozark Uplands along its northern and western border.
Species of this group also make up a significant proportion of the
total species in the Missouri and Mississippi rivers, and comprise
a lesser proportion of the fishes in the Southeastern Lowlands. For
the most part the prairie species are rare or absent in streams with
high gradients and a continuous strong flow of extremely clear or
cool waters.

Big River Faunal Group

The larger rivers of Missouri, in particular the Missouri and
Mississippi, are inhabited by a distinctive assemblage of fishes that
sets them apart as a separate faunal region (Fig. 8). In all, 30
species and one subspecies, or about 16 per cent of the total Mis-
souri fauna, occur mainly in these two large rivers. Ten species
(silver lamprey, shovelnose sturgeon, pallid sturgeon, sicklefin
chub, sturgeon chub, spottail shiner, silverband shiner, burbot,
threadfin shad, and Mississippi silverside) are known in Missouri
only from the Missouri and Mississippi rivers and their backwaters.
(Occurrences in impoundments, often resulting from introductions,
are discounted.) The channel mimic shiner occurs (or intergrades)
with the northern mimic shiner in the lower reaches of some tribu-
tary streams, but is otherwise confined to the two large rivers.

The species composition in different sections of the Missouri
and Mississippi rivers varies considerably but the fundamental
unity of the big river faunal region is indicated by the distribution
of fishes such as the shovelnose sturgeon, emerald shiner, river
shiner, silverband shiner, silver chub, blue sucker, and sauger. The
fish fauna of the Missouri and Mississippi rivers is characterized
as much by tributary species that avoid them as by fishes confined
to them. Virtually all tributary species must occur in these large
connecting rivers at least occasionally, but more than 40 have not
been collected there. An additional 23 are so rare in the Missouri
and Mississippi rivers that they can only be classed as strays.

The correlation between the distribution of certain big river
species and environmental gradients within the Missouri and Mis-
sissippi rivers provides som{> insight into the factors controlling

Missouri Fishes 277

their distribution. Although maii\ t-inironnicntal factors are in-
\oh'ed, three (bottom type, current \elocity, and turbidity) seem
to be of fundamental importance. As a fish habitat, the Missouri
Ri\er is cliaracterized by continuous high turbidity, a swift current,
a scarcity of quiet backwaters, and an unstable sand and silt
bottom. In comparison, the Mississippi River above the mouth of
the Missouri is a much clearer stream, with less current, an abun-
dance of quiet backwaters, and a bottom in the swifter sections
composed of coarse gravel and rubble. The most characteristic
fishes in the Missouri Ri\er are the barbeled minnows (Hyhopsis)
and the silvery minnows {Hyhognathus) . In a series of drag seine
collections made in 1963, these two groups comprised 51 per cent
by number of all fishes taken in the Missouri River, whereas they
comprised only 2.2 per cent of those taken in the upper Mississippi
Rixer. Five of these minnows (sicklefin chub, sturgeon chub, flat-
head chub, plains minnow, and western silvery minnow) do not
ascend the upper Mississippi Ri\er. There they are replaced by
the emerald and river shiners, which together comprised 70 per
cent of the fishes in the 1963 collections from the upper Mississippi.
Two madtoms (stonecat and freckled) and four darters (slender-
head, logperch, river and western sand) occur in significant num-
bers in the upper Mississippi River, but none of these except pos-
sibly the stonecat could be classed as anything but strays in the
Missouri River. Bottom type seems to be the principal factor con-
trolling the distribution of the madtoms and darters. The western
sand darter generally occurs over silt-free sandy bottoms, and the
others are usually found over coarse gravel or rubble. In summary,
the fish fauna of the upper Mississippi River is much more complex
than that of the Missouri River, including significant numbers of
Ozark, prairie, and lowland species, as well as the characteristic
big river species. This complexity is perhaps a reflection of the
greater diversity and stability of aquatic habitats and the less
extreme environmental conditions prevailing in the upper Mis-
sissippi River.

The Mississippi River below the mouth of the Missouri River
is intermediate in character between the upper Mississippi and
Missouri rivers, and this is reflected in its fish fauna. Virtually all
of the fishes known from the Missouri River also occur in the lower
Mississippi, but the barbeled and silvery minnows are decidedly
less abundant. Most of the species of the upper Mississippi River
likewise occur below the mouth of the Missouri. Certain species
find the clear waters below the mouth of the Ohio River more to

278 Unwersity of Kansas Publs., Mus. Nat. Hist.

their liking than the more turbid waters aboxe. These include the
skipjack herring, threadfin shad, central silver)' minnow, and the
Mississippi silverside. Darters and madtoms are unknown from the
Mississippi River below the mouth of the Ohio, and this may be
attributed to the absence of rubble bottoms in that stream section.
Apart from the mainstreams of the Missouri and Mississippi
rivers, the most significant area of occurrence for big river species
is in the northwestern part of the Dissected Till Plains. Here certain
species (Hathead chub, river shiner, and plains minnow) that are
largely confined elsewhere in the state to large rivers penetrate
into rather small streams. Possibly these species are responding to
a release from competition, because the streams of northwestern
Missouri have a depauperate fish fauna. Another significant area
of occurrence for certain big river fishes (e.g., emerald shiner,
speckled chub, and ri\'er darter) is in the lowland ditches of the
southeast. These ditches resemble major rivers in having low
gradients and continuous strong flow, and possibly these are the
factors to which the big river species are responding.

Ozark-lowland Faunal Group

This group includes six species and one subspecies, of which
three ( blackspotted topminnow, longear sunfish, and brook silver-
side) are fairly uniformly distributed o\er the Ozark Uplands and
Southeastern Lowlands. The other four (grass pickerel, northern
mimic shiner, spotted bass, and speckled darter) are generally
distributed in the lowlands, but tend to be peripheral in distribu-
tion in the uplands. Except for a single record for the brook silver-
side along the upper Mississippi River, no member of this group
has been collected more than a few miles north or west of the
boundary between the Ozark Uplands and the plains physiographic
regions (Fig. 9). A consideration of the habitats characteristic of
all Ozark-lowland species gives some insight into the factors con-
trolling their distribution. All are typically found in the quiet
pools of moderately clear, permanent-flowing streams and ditches.
Most tend to avoid streams with high gradients, such as those of
the central Ozarks, and seem to be limited in distribution in the
plains regions of north and west Missouri by siltation and continu-
ous high turbidity.

Ozark-prairie Faunal Group

Six species are about equally abundant in the Ozark Uplands
and the plains physiographic regions, but exhibit a definite faunal

Missouri Fishes 279

break along the boundaiN' l)('t\\c'cn the Ozark Upkmds and the
Southeastern Lowlands (Fig. 10). Within their area of oeeurrence
the Ozark-prairie species are least abundant in the western parts
of the Osage and Disseeted Till Plains. The factors controlling the
distribution of Ozark-prairie species may not be the same for all.
Four of them (creek chub, central stoneroller, white sucker, and
johnny darter) are characteristic inhabitants of moderately small
upland creeks \\'ith intermittent or low flow during dry seasons.
The northern redhorse and slenderhead darter are most often
found in larger streams \\'ith permanent flow. All are most often
found over coarse gravel or rocky bottoms, and this may be the
principal factor controlling their distribution in the lowlands. It is
significant in this regard that the creek chub is known in the low-
lands only from a few gravelly creeks draining Crowley's Ridge.

Wide-ranging Faunal Group

This group includes 23 species, or about 12 per cent of the
Missouri fish fauna. Among them are the most ubiquitous Missouri
fishes, such as the gizzard shad, carp, bluntnose minnow, and
green sunfish. There is little common pattern to the distribution
of the wide-ranging fishes, except that many are less abundant in
the central Ozarks than elsewhere (Fig. 11). In discussing the
distribution of species in the other faunal groups, I have attempted
to account for their distribution patterns in terms of environmental
gradients across the boundaries of physiographic regions; in dis-
cussing the distribution of the wide-ranging species, it is appro-
priate to consider why they are not similarly restricted. It is logical
to assume: (1) that the wide-ranging species have broader environ-
mental tolerances than species of more restricted distribution, and
(2) they are adapted to some widespread habitat or habitats. With
respect to the common requirements of these species, it is perhaps
significant that all are characteristic of lentic environments and the
quiet pools and backwaters of streams. Three of them (black buf-
falo, channel catfish, and flathead catfish) seem to be more tolerant
of current than the others, but even they are more often found in
pools. Pool-type habitats occur in most parts of Missouri, and
species with broad tolerances for such factors as turbidity, tempera-
ture, and bottom type are thus provided with a habitat that is not
limited to any one physiographic region. Perhaps the scarcity of
wide-ranging species in the central Ozarks is to be explained by
the fact that pools and backwaters are less developed in the high-
gradient streams of that region than in streams elsewhere in the

280 University of Kansas Publs., Mus. Nat. Hist.

state. Another factor that may be significant in accounting for the
widespread distribution of these species is that their habitat is
nearly continuous in, or adjacent to, large rivers, and they can
readily disperse from one area to another. The parasitic chestnut
lamprey may owe its widespread distribution to passive transport
while attached to other fishes that are wide-ranging. Most species
of this group have been favored by man's activities, and are per-
haps more widespread than they were under pristine conditions.
The carp is an introduced species and thus owes its presence in
Missouri to man. The bluegill, largemouth bass, and channel cat-
fish are widely stocked in small, artificial impoundments, and are
certainly more abundant, if not more widely distributed, than
was formerly the case. The construction of reservoirs along the
major streams has increased the habitat available to species that
thrive in lentic environments. Many of these large reservoirs are
in the Ozark Uplands where lentic environments were formerly
less common than elsewhere in the state. The American eel is al-
most certainly less abundant in the Ozark Uplands than formerly.
The same reservoirs that have favored other wide-ranging species
now block the upstream migrations of the eel.

Competition as a Limiting Factor

The role of competition in controlling the distribution of fishes
has been alluded to but not discussed in detail in the preceding
sections. Its action is more subtle than that of the more readily
observed physical factors, but it must have a profound efi^ect on
the distribution of certain fishes. One indication of its importance
comes from the complementary distributions of certain species.
The blackstripe and blackspotted topminnows provide a striking
example. Although these two species are broadly sympatric in
Missouri and elsewhere, they are rarely syntopic. Their habitat
requirements are similar, and their complementary distributions do
not seem to be consistently correlated with any obvious variations
in environmental conditions. Presumably they do differ slightly
in their environmental tolerances, and at any given locality one
species is favored and eventually replaces the other. The few
syntopic populations that do occur may result from a continued
influx of one or both species from a nearby area where conditions
are in its favor, or from fluctuations in environmental conditions
that favor first one species and then the other. Other pairs whose
largely complementary distributions may be due at least in part
to competition are: silver and chestnut lampreys, least and Ameri-

MissouHi Fishes 281

can brook lampreys, grass and cliain pickerels, streaiuline and
gra\ (-1 chnbs, dnskystripe and bleeding shiners, common and striped
shiners, bigmouth shiner and siKerjaw minnow, western and cen-
tral siKery minnows, creek and lake chubsuckers, tadpole and
freckled madtoms, smallmouth and spotted basses, river and star-
gazing darters, bluntnose and johnny darters, Arkansas and stippled
darters, orangethroat and rainbow darters, striped and barred
fantail darters, and least and cypress darters.

Certain pairs that seem to have similar requirements are note-
worthy for their ability to li\'e together with little indication of
competition. Perhaps both members of these pairs are limited
bv factors unrelated to their common requirements, or are able;
to utilize common resources in ways detrimental to neither. The
largescale and central stonerollers provide an example. These two
species appear to have similar requirements, but occur together in
dense populations over much of the Ozark Uplands. Perhaps there
is little competition between them for at least two important re-
sources— space for spawning, and food. The extensive gravelly and
rocky bottoms of Ozark streams provide an abundant substrate for
spawning, and for growth of the diatoms and other attached algae
that are the principal food of stonerollers. Other factors (preda-
tion, for example) may keep the two species within the carrying
capacity of the environment, and thus prevent effective competition
between them. The greenside and banded darters pro\'ide another
example. These two species are almost invariably found together
in Missouri. Adults of both are typically found on rocky riffles
and utilize filamentous algae for spawning. Their food habits are
not known to me, but it is likely that both are carnivorous. Traut-
man (1957:573) indicated that competition for spawning sites is
not great, inasmuch as the greenside spawns earher and utilizes
deeper and swifter water than the banded darter. Perhaps a de-
tailed comparative hfe history study would reveal other differences
that limit competition between them.

Another indication of the role of competition in controlling fish
distribution is the occurrence of certain species outside of their
usual habitat in some sections of the state; these occurrences may
be due to a release from competition. Populations of the flathead
chub and the river shiner in the faunally depauperate small streams
of northwestern Missouri exemplify this phenomenon. The distri-
butional relationships of the spotted bass and smallmouth bass
provide another example. Prior to the introduction of the spotted
bass into the Missouri River system, the smallmouth occupied
some central Missouri streams that may have been marginal for it.

282 University of Kansas Publs., Mus. Nat. Hist.

The spotted bass is now invading these streams and in some in-
stances is replacing the smalhnouth bass. One indication of com-
petition bet^veen them is the extensive hybridization that occurs
wherever the spotted bass has recently invaded streams formerly
occupied only by the smallmouth. This hybridization suggests
competition for spawning sites or for spawning mates. It also sug-
gests that the two species are so closely related that competition for
other requirements is likely.

Certain fishes that have broad tolerances for physical factors
are much more abundant where few other species are present than
elsewhere, suggesting an inability to compete in faunally rich
habitats. Examples include: creek chub, redbelly dace, fathead
minnow, black bullhead, and green sunfish.

Climate as a Limiting Factor

Certain meteorological conditions, most notably temperature
and precipitation, vary considerably over Missouri. These varia-
tions probably influence the distribution of certain fishes through
their effects on the temperature and flow regimens of the streams,
but their action is largely obscured by the effects of other factors.
The more widespread occurrence of certain fishes in the recent
geological past, as indicated by fossil evidence and disjunct popu-
lations, attests to the importance of climate in controlling fish dis-
tribution. Examples are cited in the section on historical factors.
Many fishes reach some limit of their natural range in Missouri.
The range limits of some of these do not seem to be correlated with
gradients in other physical factors, and for these the effects of cli-
mate may be critical. Examples include: silver lamprey, northern
pike, alligator gar, threadfin shad, spottail shiner, brassy minnow,
white sucker, burbot, trout perch, pumpkinseed sunfish, and yellow-
perch.

Variations in Faunal Diversity

Although the fish fauna of Missouri as a whole is quite rich,
the same cannot be said for all sections of the state. Since difl^cr-
ences in faunal diversity may result from local variations in the
environment, it is appropriate to consider them in this section.
One approach to the study of variations in faunal diversity is to
compare the number of species known from various stream systems
(Fig. 13). In most streams, there is a longitudinal succession of
fishes, with each stream section being occupied by a characteristic
faunal assemblage (Shelford, 1911; Thompson and Hunt, 1930).

Missouri Fishes

283

10 0 10 30 50
1 1 1 I I I I

MILES

Figure 13. Number of fish species from selected stream systems in Missouri.
Underscored figures indicate number of species from stream systems liounded
by solid lines; circled figures indicate number of species from subsystems set

off by dashed lines.

Therefore, a general correlation is to be expected between drainage
area and species diversity, with large stream systems having more
species than small stream systems. It is evident from Fig. 13 that
this general correlation does not hold for all Missouri streams.
Stream systems in the prairie faunal region have fewer species
than those of comparable or even of smaller size in the Ozark
and lowland faunal regions. Also, stream systems that span the
boundary between faunal regions have more species than those of
comparable size that lie entirely within a single faunal region.

Both of these conclusions are demonstrated by comparing the
fish faunas of the Lamine and Blackwater rivers in west-central
Missouri. These adjacent subsystems join a few miles above the
point where their combined waters empty into the Missouri River.
Yet the Lamine, with a drainage area of 1090 square miles, has
more than twice as many species as the Blackwater River, with a
drainage area of 1550 square miles. The Blackwater River lies

284 University of Kansas Publs., Mus. Nat. Hist.

entirely within the prairie faunal region, whereas the Lamine
River spans the boundary between the prairie and Ozark faunal
regions ( Fig. 12 ) . The same phenomenon is evident when various
subsystems of the Osage River are compared. South Grand River,
lying entirely in the prairie faunal region ( if the "lines of best fit"
in Fig. 12 are taken as the boundary), has the fewest species (37).
As one proceeds in a counterclockwise direction toward the Ni-
angua River and into the Ozark faunal region, the fauna becomes
richer. In the southern part of the state, stream systems that head
in the Ozark faunal region and flow into the lowland faunal region
have more species than those of comparable size that lie entirely
within the Ozarks. Thus, more species are known from the Head-
water Diversion than from the much larger Missouri portion of the
White River system. Among streams lying entirely within the
prairie faunal region, those draining eastward into the Mississippi
River have more species than those draining southward into the
Missouri River. In the Missouri River portion of the prairie faunal
region, there appears to be an increase in the number of species
per stream system from west to east, but this difference largely
disappears if the portions of these stream systems lying in Iowa
are included. Thus, the total numbers of species known from the
entire Nodaway and Platte systems are 48 and 41, respectively,
whereas the total species from the Grand and Chariton systems
remain unchanged (49 in each).

The increased faunal diversity of streams that span the bound-
aries between faunal regions is readily explained. Each faunal
region is inhabited by a characteristic assemblage of fishes. Stream
systems that span the boundaries between faunal regions are in-
habited by species characteristic of two faunal regions, whereas
stream systems lying entirely within a single faunal region are
inhabited only by species of one faunal assemblage.

The low faunal diversity of the prairie faunal region is less
readily explained, but is perhaps a reflection of instability in the
aquatic environments. Lotic habitats of the prairie region present
a less stable environment than those of the Ozark and lowland
faunal regions with respect to such factors as volume of flow,
turbidity, temperature, and dissolved oxygen. Survival in the un-
stable prairie environments places a premium on adaptability and
generalized requirements, whereas the stable environments of the
Ozark and lowland faunal regions place less of a premium on
adaptability, and permit greater speciaHzation. Specialization leads
to a proliferation of ecological "niches" and enrichment of the

Missouri Fishes 285

fauna. A similar explanation has been put forward by some work-
ers to explain the greater diversity of tropical biotas as compared
w ith those of temperate regions. That the prairie species may in-
deed be more adaptable than the Ozark and lowland species is
suggested by their more widespread occurrence outsid(> of tlieir
primary distribution center (compare Figs. 5, 6, and 7).

Cenozoic Events Affecting the Distribution of Missouri Fishes

The character of the Missouri fish fauna has been affected by
changes in physiography, drainage relationships, and climate in the
recent past. These changes must be considered before turning to
a discussion of the developmental history of the fauna.

Physiographic Changes

All of Missouri except the Southeastern Lowlands was elevated
above the sea during the Appalachian Revolution near the close of
the Paleozoic Era, and has been a land area continuously since that
time. The Ozark Uplands have had a more complex geomorphic
history than the plains regions to the north and west. Since initial
elevation above the sea, the Ozark Uplands have been subjected
to three major uplifts (Bretz, 1965:21). The last of these occurred
near the close of the Tertiary, and the cycle of erosion thus initiated
is responsible for the rugged topography that characterizes the
region today. This uplift was preceded by a long period of tectonic
quiet during which the region was worn down to a low plain
drained by sluggish streams. According to Bretz (1965:133), the
streams at that time occupied wide, gently sloping valleys, with
local relief of as much as 100 feet. The record of this old plain is
preserved in the general accordance of summit levels in the Ozarks.
Following the last major unlift, there was a long pause during
which the major streams reduced their beds to base level and
developed a meandering character. This was followed by two
minor uplifts that rejuvenated the streams, resulting in rapid down-
cutting and development of the entrenched meanders for which
Ozark streams are famous.

The plains regions were little affected by this series of upHfts,
and at the beginning of the Pleistocene Epoch the area as a whole
was a low plain drained by sluggish streams. The Nebraskan ice
sheet entered Missouri from the north, but it apparently did not
advance as far south as the Missouri River. It disrupted the late-
Tertiary drainage patterns and spread a thick layer of glacial
debris over the landscape. The record of this early ice advance was

286 University of Kansas Publs., Mus. Nat. Hist.

largely destroyed by the Kansan ice sheet, which advanced to, and
in places slightly beyond, the Missouri River. Upon \\'ithdra\val,
the Kansan ice sheet left a level till plain that was largely destroyed
by subsequent erosion. Illinoian ice lay briefly along the bluffs of
the Mississippi River south of St. Louis and impounded a pro-
glacial lake to the north, but this ice advance did not penetrate
Missouri. During earlier ice advances, but especially during Wis-
consin time (Branson, 1944:344), loess was spread over all of north
Missouri and for a considerable distance to the south.

The Mississippi Embayment, of which the Southeastern Low-
lands of Missouri represent the northern extremity, has been con-
tinuously above the sea only since the close of the Tertiary. At
that time an ancestral Mississippi River flowed southward to the
Gulf of Mexico on a surface that was almost even with that of the
adjacent uplands (Fenneman, 1938:84). Crowley's Ridge and
Benton Hills are the dissected remnants of the upland plain from
which the present trough was carved. Although the trough is
largely erosional, its most recent history seems to have been one
of filling rather than downcutting. From Cairo to Memphis the
depth of the alknium below low water is 150 to 200 feet, although
the river is not known to scour its channel to depths greater than
100 feet (Fenneman, 1938:85).

Drainage Evolution of the Mississippi River System

The late-Tertiary drainage patterns of the central United States
were quite different than they are today, and an extensive literature
has developed on the evolution of the present drainage patterns.
Recent summaries for all or parts of the Mississippi River system
have been presented by Horbcrg and Anderson (1956:103-107, fig.
2), Flint (1957:168-171, fig. 10-2), Thornbury (1965:214-216, fig.
12.3), and Metcalf (1966:64-81, figs. 3 and 4). A hypothetical
reconstruction of the preglacial drainage patterns is presented in
Fig. 14; in that figure and parts of the subsequent discussion I
follow Metcalf (he. cit.).

Most significant from the standpoint of fish distribution is that
glaciation was accompanied by deflection into the Mississippi Ri\er
system of streams that formerly drained northward into Hudson
Bay and eastward into the Atlantic Ocean. The preglacial Missouri
and Ohio rivers were relatively minor streams that had their drain-
age areas considerably enlarged by these events and by deflection
to them of streams that formerly pursued independent courses to
the ancestral Mississippi River. Also, there have been significant

Missouri Fishes

287

(

'n.^

DRAINAGE

Hudson Boy

Laurentton

Plaint

Teays-Miss

SS i ppi

105

Figure 14. A hypothetical reconstruction of the preglacial drainage patterns

of tlie central United States.

changes, only indirectly related to glaciation, in drainage relation-
ships at the head of the Mississippi Embayment and in the central
plains.

The major features of the preglacial drainage relationships in
the Missouri Basin were worked out by Todd (1914), who first
pointed out that the ancestral upper Missouri River formerly dis-
charged northeastward into Hudson Bay. At that time the Gulf-
Arctic divide was located not far north of Pierre, South Dakota,
and the ancestral Cheyenne River and all streams north of it were
in the Arctic drainage (Lemke et al, 1965:15). Farther south,
the ancestral Bad, White, and Niobrara rivers continued eastward
across the present course of the Missouri River to join the ancestral
Iowa River, which discharged southeastward across Iowa into the
ancestral Mississippi. The preglacial valley of the Missouri River

288 UNrvnERSiTY OF Kansas Fuels., Mus. Nat. Hist.

in Missouri was then an extension of the Kansas River Valley
(Greene and Trowbridge, 1935:3; Heim and Howe, 1963: 385).
The ancestral Kansas River headed on the eastern slope of the
FHnt Hills in Kansas (Frye and Leonard, 1952:184), and in Mis-
souri followed closely the present valley of the Missouri River ex-
cept for a minor detour into Salt Fork of Blackwater River at Malta
Bend. It re-entered the present Missouri River valley at the moutli
of Lamine River, where it was joined from the northwest by another
large stream that roughly paralleled the present valley of Grand
Ri\er from Gentry County downstream.

The number of diversions inxohed in the establishment of the
present course of the Missouri River and the dates of these diver-
sions are not definitely known (Thornbury, 1965:248). Deflection
southward of Arctic drainage occurred when an early ice sheet
blocked the lower courses of streams that formerly flowed east or
northeast. As each ponded stream rose, it found a new outlet
into the next major stream to the south, and integration of these
new outlets formed the present channel of the Missouri River.
In Missouri, the present north-south portion of the Missouri Ri\er
between St. Joseph and Kansas City probably developed during
and after the retreat of the Kansan ice sheet (Heim and Howe,
1963:386). The principal features of the drainage in north Missouri
were developed at about the same time and have not been appreci-
ably altered since.

The late-Tertiary drainage patterns of the central plains are
not known in detail, but Frye and Leonard (1952:180-199 and figs.
12-15) indicated that the Flint Hills of Kansas, which now are
breached by tributaries of the Kansas River, were then a major
divide. Streams west of this divide discharged southward into
Oklahoma. The drainage relationships in this region remained
essentially unchanged until Kansan time, when a western lobe of
the Kansan ice sheet crossed the valle\' of the Kansas River that
then headed east of the Flint Hills, and a temporary meltwater
outlet was established into streams west of the divide. By late
Illinoian time all drainage from the Saline to the Republican rivers
had been captured by the Kansas River, and the Smoky Hill River
was diverted to the Kansas River drainage by early Wisconsin
time. Bayne and Fent (1963) presented a somewhat different in-
terpretation of drainage changes in this region. These authors
indicated that the Solomon River was a major through-flowing
stream of the Kansas River system throughout the Pleistocene, and
they do not recognize a meltwater connection between the eastern

Missouri Fishes 289

and wc'slcni drainages during Kansan time. The extent oi the
western drainage north and south of Kansas is uncertain. Frye
and Leonard {o}). cit.) suggested that it discharged southward
across Oklahoma into the ancestral Red River rather than into the
lower Arkansas River. Quinn (1958:40-42 and fig. 1) suggested
that piracy of western drainage by the lower Arkansas occurred
during the Pleistocene some time prior to the Sangamon Inter-
glacial. Frye and Leonard indicated that the headwaters of the
western drainage were in northern Kansas, but Metcalf (1966:66),
after re\iewing e\idence presented by Lugn (1935:35-37) and
Lueninghoener (1947:60), suggested that streams as far north as
the upper Platte and upper Loup rivers may also have been part
of this drainage, for \\hich he coined the name "Preglacial Plains
Stream."

The pre-Pleistocene Ohio River was a relatixely minor stream
that probabh' headed southwest of Madison, Indiana (Wayne,
1952:576). The present upper Ohio (Monongahela- Allegheny)
drainage at that time flowed north, probably into a river that con-
tinued northeastward through the basin of Lake Erie (Flint, 1957:
171; Thornbury, 1965:207). This river was probably a part of the
much larger Laurentian stream system that occupied the present
basins of the Great Lakes and drained eastward into the Atlantic
Ocean. The pre-Pleistocene master stream of most of the Ohio
Basin was the Teays River (Fidler, 1943; Horberg, 1945; Wayne,
1952). This stream headed near the eastern scarp of the Blue
Ridge in North Carolina, flowed northwestward to central Ohio,
and then westward across Indiana and Illinois to the ancestral
Mississippi River. The latter stream at that time occupied the
present \ alley of the Illinois River in central and southern Illinois.
An early glacier, probably the Kansan (Flint, 1957:171) blocked
the lower Teays and Monongahela-Allegheny systems and diverted
them from their preglacial courses into the ancestral lower Ohio
River.

During Pleistocene and Recent time significant changes oc-
curred in drainage relationships at the head of the Mississippi
Embayment ( Marbut, 1902; Matthes, 1933; Fenneman, 1938:87-90;
Fisk, 1944). In late-Tertiary time the Ohio and Mississippi rivers
flowed on opposite sides of Crowley's Ridge and did not unite
until some point below Helena, Arkansas, more than 200 miles
below their present confluence. At that time the Mississippi River
occupied the Advance Lowland west of Crowley's Ridge, and the
Ozark streams that now enter the lowlands from the north were

290 Unr'ersity of Kansas Publs., Mus. Nat. Hist.

its direct tributaries. The course of the Ohio River was not that
of the present Mississippi River below Cairo, but rather was at the
eastern edge of Crowley's Ridge. The Ohio was then flowing across
southern Illinois in the Cache Lo\\'land, from five to 15 miles north
of its present course. The course of the Tennessee Ri\er in late-
Tertiary time is uncertain. It may have joined the Ohio River
before the latter stream entered the Cache Lowland, but more
likely the Tennessee followed a course from its present mouth
down the present Ohio and Mississippi, joining the Ohio near the
southern extremity of Crowley's Ridge.

During all subsequent time the junction of the Ohio and Mis-
sissippi rivers has shifted progressively north\\'ard. After a number
of shifts farther south, the Mississippi Ri\er broke through into the
Morehouse Lowland near the northern extremity of Crowley's
Ridge. Matthes (1933) and Marbut (1902) considered this diver-
sion to have taken place during the Pleistocene, but Fisk (1940:40)
thought that it was post-Pleistocene. Di\ersion of the Mississippi
River into its present course at Thebes Gap has been so recent that
the river has not had time to adapt the narrow gap to its size. The
causes of these changes are not definitely known. Fisk (1940:41)
attributed the Thebes diversion to beheading of a north-flowing
tributary of the Mississippi by the Ohio Ri\'er.

When the Mississippi River first abandoned the Advance Low-
land, its direct tributaries from the W^hitewater Ri\'er south to
Black River were left flowing in the old channel (Fisk, 1940:46).
Subsequently the Castor, Whitewater, and St. Francis rivers found
independent courses to the Mississippi River through Crowley's
Ridge, and Black River was diverted into the White River system
along a course formerly followed only by the Current, Spring,
Eleven Point, and Strawbcrr\' rivers, which then became tribu-
taries of Black Ri\'er.

The drainage patterns of the Ozark Uplands are consequent to
the surface that was formed by the late Paleozoic uplift, and they
have remained essentially unchanged since that time (Fenneman,
1938:641). However, during the long period of downcutting that
followed uplift, many captures ha\'e undoubtedly occurred, and
streams have shifted their courses to adjust their drainage patterns
to minor irregularities in the surface that originated with uplift.
A stream capture that may be of significance in the dispersal of
certain fishes was noted by Bretz (1965:46). A stream-cut notch
or windgap about 15 miles southeast of Marshfield in Webster
County indicates the capture of a former James Ri\'er tributary by

Missouri Fishes 291

a high-gradient tributary of the Gasconade Ri\er. Drainage pat-
terns suggesting stream piracy are afforded by Flat Creek, tribu-
tary to James River in Barry County, and Muddy Fork of Spring
Ri\er in Barton County. Fhit Creek flows north-by-northeast for
about 15 miles before making an abrupt turn to the east and south-
east near McDowelk Muddy Fork flows northwest for more than
20 miles before turning abruptly southward near Lamaar. Bretz
(1965:97) found little evidence that would indicate that these
aberrant drainage patterns indicated stream captures.

Climates of the Past

Because climates as such are not subject to preservation, the
nature of past chmates can be inferred only indirectly from studies
of plant and animal fossils, sediment types, and physical conditions
of past geologic ages (Dorf, 1959:182-185). This evidence has not
been interpreted in the same way by all workers, but there is
general agreement that the Cenozoic Era has been characterized
by marked climatic fluctuations.

At the end of the Eocene, some 40 million years ago, there were
no western mountains to block the flow of warm, moist air from
the Pacific Ocean, and an arm of the Atlantic Ocean extended up
the Mississippi Embayment far into the interior of North America.
These two factors combined to maintain a uniform oceanic climate
o\'er much of North America. Subtropical conditions existed as far
north as the Dakotas, and cool-temperate conditions were found
nearly to the North Pole (Dorf, 1959: map 1). Fossil remains of
tropical forests such as those that now characterize the lowlands
of northern South America and coastal Mexico occurred as far
north as Tennessee and Missouri (Dorf, 1959:185). With elevation
of the Rocky Mountains and gradual withdrawal of the sea from
the Mississippi Embayment during the middle and late Tertiary,
there was a change toward cooler conditions, accompanied by
increasing aridity in the Great Plains. As a result of these climatic
changes, there was a southward shift of the vegetational zones and
the development of grasslands in the Great Plains. By the end of
the Pliocene, about two million years ago, there was an approach
toward modern climatic conditions, and the vegetation was similar
in distribution and aspect to that of today (Dorf, 1959:191).

The cooling trend that began in the Oligocene culminated in
the severe glaciation of the Pleistocene, a period characterized by
a series of cyclic changes in climate that marked the advance and
retreat of the continental ice sheets. The advances apparently

292 University of Kansas Publs., Mus. Nat. Hist.

resulted from a combination of cooler temperatures and increased
precipitation, and the interglacial periods marked a return to
warmer and drier conditions.

The equivocal nature of the evidence concerning climatic con-
ditions in the southern United States during glacial periods has
provoked a lively controversy among biogeographers. The extreme
viewpoints were expressed by Deevey (1949:1375) and Blair
(1965:543), who expressed the opinion that glacial cooling was so
severe that warmth-loving plants and animals could have persisted
only in peninsular Florida and Mexico, and by Braun (1951:142)
and Thomas (1951:155) who presented evidence that they inter-
preted as indicating little southward displacement of biotas during
glacial maxima. Proponents of extreme cooling cite the presence
of fossil pollen of spruce and fir and the remains of such mammals
as musk oxen and mastodon in deposits as far south as Florida,
Louisiana, and Texas, whereas proponents of the opposite view
cite the close correlation between the distribution of many southern
species and the glacial boundary, and the presence of disjunct
populations of other southern species close to the glacial boundary
but north of their main areas of distribution. The strongest argu-
ment against extreme southward displacement would seem to be
that the biotas now occupying the ecok)gically diverse Appalachian
Plateaus and Interior Highlands would have found conditions un-
suitable for survival in the topographically monotonous lowlands
of peninsular Florida or the plains that separate the upland areas
from a Mexican refugium. The most plausible explanation for
these seemingly conflicting lines of evidence is that glaciation was
accompanied by some cooling and a marked steepening of the
north-south temperature gradients, permitting a mixing of northern
and southern biotas, with only slight or moderate displacement of
the latter.

At least some of the interglacial periods were apparently char-
acterized by warmer temperatures than those that prevail today.
A fossil biota from the Sangamon Interglacial Period indicates a
mean temperature at Toronto, Ontario, that was 2 to 3°C higher
than at present (Fhnt, 1957:340).

From a nearly complete sequence of molluscan and mammalian
faunas of the High Plains, Taylor (1965:603) concluded that all
pre-Wisconsin glacial and interglacial changes were broadly simi-
lar, and glacial climates during the Wisconsin were more severe
than in any previous period of the Pleistocene. He further con-
cluded that aridity and strong seasonal contrast are a late Wisconsin

Missouri Fishes 293

or p()st-\\^i,sconsin dcxclopmc'iit, and none of the prc-Wisconsin
faunas could li\e in the combination of hot smnnicrs and bitterly
cold spells that characterize the climate of the High Plains today.
From analysis of fossil pollens, Deevey (1949:1356) recognized
fi\e post-Pleistocene climatic phases in eastern North America.
These were: (1) a cool period characterized by spruce and fir
pollen, (2) a warmer and drier period, with a predominance of
pine pollen, (3) a still warmer but moist period, often referred to
as the Climatic Optimum, characterized by pollen of oaks and
such mesophytes as beech and hemlock, (4) a warm, dry period,
often referred to as the Xerothermic Period, with a predominance
of oak and hickory pollen, and (5) the cooler, moister climate of
today.

Developmental History of the Fish Fauna of
Missouri and the Central Mississippi Valley

Fossil evidence that would have a bearing on the problem at
hand is distressingly fragmentary. Conclusions concerning the de-
velopmental history of the Missouri fish fauna must therefore be
based primarily on the present-day distribution patterns of the
fishes, and what can be deduced concerning environmental condi-
tions and drainage relationships in eastern North America during
the period under consideration. These conclusions are, to say the
least, highly conjectural. The evidence is sometimes subject to more
than one interpretation, and there is often little basis for deciding
^^'hich interpretation is the correct one.

A basic tenet of the approach used in this discussion is that
the habitat preferences and environmental tolerances of present-
day species do not differ significantly from those of their immediate
ancestral stocks. Since most if not all fishes known as fossils from
Pleistocene deposits are morphologically indistinguishable from
existing species (Miller, 1965:571), there would seem to be some
basis for concluding that their requirements have likewise under-
gone little change during the same period. From this it follows
that we can draw inferences concerning the fish faunas of past
geologic ages from what is kno\\'n of environmental conditions that
existed at that time. Another tenet of this approach is that the
theory of geographic speciation, as propounded by Mayr (1942)
and others, is a valid working hypothesis. Thus, the distributional
relationships of subspecies and closely related species in most in-
stances reflect the distributional relationships of their immediate

294 Unwersity of Kansas Publs., Mus. Nat. Hist.

ancestral stocks, and can be used in deducing the place of origin
for the forms in\olved.

In the following discussion, specific examples will be cited, but
not all will be discussed in detail. For a more detailed discussion
of these examples, the appropriate species accounts should be
consulted.

Pre-Pleistocene Fish Fauna

The origin of most elements in the freshwater fish fauna of
North America barely antedates the Cenozoic Era, and about 60
per cent of the living fauna may be no older than late Miocene or
early Pliocene (Miller, 1965:569). For the present discussion, the
late Tertiary will be taken as the point of departure.

Environmental conditions in Missouri during the late Tertiary
probably resembled those found today on the coastal plain of the
southeastern United States, and it is likely that the fish fauna had
similar affinities. At that time, Missouri as a whole was a region
of low relief drained by sluggish streams, and its climate was warm-
er and more humid than that of today. Fossil fishes from regions
west of Missouri pro\'ide the only definite evidence on which to
base conclusions about the late Tertiary fauna. Hubbs (1942)
reported a species of Menidio from Pliocene deposits in Oklahoma.
C. L. Smith (1962) listed 20 species of freshwater fishes from
Pliocene deposits in Kansas, Oklahoma, and Nebraska. Only two
of these (green sunfish and channel catfish) occur today near the
fossil localities, and most of the other species suggest a fauna much
like that now inhabiting the lower Mississippi Valley and Gulf
coastal plain. Species tentatively identified from the fossil deposits
included the bowfin, alligator gar, a pike (Esox), smallmouth
buffalo, brown bullhead, redear sunfish, and rock bass.

Lack of topographic diversity probably resulted in rather uni-
form ecological conditions over Missouri in late-Tertiary time, and
the strong regional differentiation that characterizes the fish fauna
today was little dexeloped. In particular, the highly distinctive
fauna now occupying the Ozark Uplands may be a relatively recent
development. This is suggested by the absence among Ozarkian
fishes of any extensive lineages that seem to be autochthonous to
the region. Many Ozark endemics are represented elsewhere by
forms so closely related as to suggest that their common ancestral
stocks had a widespread preglacial distribution, or dispersed into
the Ozark Uplands during the Pleistocene. Thus, the Ozark shiner,
Ozark cavefish, longnose darter, Missouri saddled darter, and

Missouri Fishes 295

Niaiigua darter arc all represented b)' geminate forms in uplands
east of the Mississippi Ri\'er.

On the other hand, a few Ozark endemics are strongly differ-
entiated from their closest relatives elsewhere, and these may
represent a faunal element that was already localized in the Ozark
Uplands by late-Tertiary time. Early localization of these species
in the Ozarks may have resulted from regional differences in bed-
rock geology. Pennsylvanian rocks like those now exposed in the
plains regions of north and west Missouri formerly covered the
Ozark Uplands, but erosion initiated b}' pre\ious uplifts of the
Ozarks had probably stripped away the Pennsylvanian rocks by
late-Tertiary time (Bretz, 1965:132). The resulting differences in
bedrocks and soils may have been reflected by regional differences
in the late-Tertiary stream environments. Gilbert (1964:104) in-
dicated that the common ancestral stock of the bleeding shiner
and duskystripe shiner became isolated in the Ozarks during the
Pliocene. It is perhaps significant that these two species avoid
sections of the Ozark Uplands where remnants of Pennsylvanian
shales remain, suggesting that localization in the Ozarks could
have resulted from an affinitv for streams draining limestone re-
gions. Other Ozark endemics that are strongly differentiated from
their closest relatives, and which may also have been localized in
the Ozark region by late-Tertiary time, are the wedgespot shiner,
Arkansas saddled darter, and yoke darter.

Renewed uplift of the Ozark Uplands, withdrawal of the sea
from the Mississippi Embayment, and a shift towards a cooler and
less humid climate toward the close of the Tertiary Period resulted
in changing ecological conditions in Missouri. Many southern spe-
cies withdrew from Missouri or became localized in the now-
emerging lowlands of the southeast. Fishes that formerly ranged
to the north and west of Missouri became established in the state,
and the fauna as a whole began to take on a more modern char-
acter.

Pleistocene Changes in the Fish Fauna

The intrusion of continental ice sheets into the central United
States on at least four separate occasions during the Pleistocene
Epoch profoundly affected the fish fauna. Species formerly local-
ized in drainages north of the Mississippi Valley dispersed south-
ward through connectives that developed with glaciation, and the
lower Mississippi Valley served as an important refugium for
northern fishes displaced by glaciation. Certain areas were so modi-
fied by glaciation as to be unsuited for fishes that occurred there

296 Unwersity of Kansas Publs., Mus. Nat. Hist.

preglacially. Changing ecological conditions in the Mississippi
Embay ment favored the alternate dispersal and isolation of up-
land fishes. Stream captures in the central plains provided oppor-
tunity for faunal exchange with western drainages.

Noiihern component: southward extensions of ran<i,e induced
by glaciotion. — Preglacial drainages north of the Mississippi Valley
(Hudson Bay and Laurentian) probabh' were inhabited by species
that were absent from the Mississippi \^alley. With the advance
of Pleistocene ice sheets, some of these species dispersed southward
and became a permanent part of the Mississippi Valley fish fauna.

Certainly any fishes that persisted in the unglaciated upper
Missouri River became part of the Mississippi fauna when that
stream was di\erted southward into its present course. Baile\' and
Allum (1962:119-120) listed 11 northern species (only one of
which now occurs in Missouri) that the\^ thought persisted during
the last glaciation in the upper Missouri system, and Metcalf (1966:
74-75) added 13 more to the list. As Metcalf (1966:75) pointed
out, the \irtual restriction of three of these (pallid sturgeon, stur-
geon chub, and sicklefin chub) to the Missouri system suggests
an origin in the preglacial northward-flowing upper Missouri.

Several other species ha\"e distributions suggesting that they are
Pleistocene invaders of the Mississippi Valley from northern drain-
ages. All are widespread in regions drained by the preglacial Hud-
son Bay or Laurentian stream systems. None now occurs widely
in the Mississippi Valley, where they are represented by closely
related forms that seem to be autochthonous to the region. Fishes
in this category include: silver lampre\', northern brook lamprey,
goldeye, northern pike, northern flathead chub {Hijhopsis g. gra-
cilis), common shiner, western sibery minnow, brassy minnow,
smoothlip northern redhorse (Moxoslonia nh macrolepidotum),
pumpkinseed, northern logperch (Percina caprodes semifasciata) ,
and least darter.

Some other Missovu'i fishes are primariK" northern in distribu-
tion, but cannot be definitely paired with southern counterparts.
These fishes ^vere probably present in the preglacial Hudson Bay
or Laurentian systems, and may have entered the Mississippi Valley
during the Pleistocene: American brook lamprey, lake sturgeon,
spottail shiner, blacknose shiner, burbot, trout perch, and yellow
perch.

That glaciation was indeed accompanied b>' southward dis-
placement of fishes is indicated by the fossil record. C. L. Smith
(1954, 1958) and C. R. Smith 0963) listed fossil fishes from

Missouri Fishes 297

Plcistocc'iu> (Illinoian) deposits in southwestern Kansas and north-
western Oklahoma. Many of the species known as fossils do not
now ocein- near the fossil localities, but are widespread in the
upper Mississippi Valley and adjacent Great Lakes drainage. One
species (muskellunge) identified in tlie fossil fauna is not native
to Missouri, but its occurrence as fossils southwest of Missouri
suggests that it was present here during the Pleistocene. Another
species (yellow perch) now occurs in Missouri only as strays and
introduced populations, but it must ha\'e been more widespread
in the past.

Occurrence oi relict populations of northern animals in the
Ozark Uplands pro\'ides additional evidence for southward dis-
placement. Steep-sided \'alleys and abundant springs in the Ozarks
have provided conditions that iavor the long-term survival of
northern animals displaced by glaciation. Ross ( 1965 ) interpreted
broadly disjunct populations of boreal caddisflies and stoneflies
in the Ozark Uplands as Pleistocene relicts. Ozark and boreal dis-
tribution patterns in three species of amphibians have been in-
terpreted similarly (P. W. Smith, 1957:207).

Five species of Missouri fishes (northern brook lamprey, spot-
fin shiner, channel darter, least darter, and mottled sculpin) having
the main body of their ranges in the area covered by the Wisconsin
ice sheet have outlying populations in the Ozark Uplands. Perhaps
these outlying populations date from southwestward displacement
during the last ice advance. Two other species (rosyface shiner
and rainbow darter) are similarly distributed, but morphological
differentiation of populations in various stream systems of the
Ozarks suggests that these two have resided there for a longer
period than species of the first group. Possibly Ozark populations
of the rosyface shiner and rainbow darter date from more than one
ice ad^•ance, or are remnants of a more \\'idespread preglacial
distribution. A disjunct population of the rainbow darter in the
state of Mississippi almost certainly is a Pleistocene relict.

The distribution of some Missouri fishes pro\'ides evidence that
the influence of the glacial climate on fish distribution extended
far beyond the glaciated regions. Two species (redbelly dace and
creek chub) occur as disjunct populations in the foothills of the
Rocky Mountains in northeastern New Mexico. That these popu-
lations may date from one of the glacial periods is indicated by
the presence of the creek chub in late-Illinoian fossil deposits from
southwestern Kansas (G. R. Smith, 1965:280). Also present in this
fossil fauna were the Ozark minnow and black redhorse, t\\'0 fishes

298 Unwersity of Kansas Publs., Mus. Nat. Hist.

that now occur no closer to southwestern Kansas than the Ozark
Uplands.

Probable restrictions of range induced by glaciation. — Although
glaciation appears to have provided for extension of range in some
species, it may have restricted the range of others. Some areas
appear to have been so modified by glaciation as to be unsuited
for species that formerly occurred there. Ross (1965:586) indi-
cated that the leveling action of the glaciers produced conditions
in at least Ohio, Indiana, Illinois, and Iowa that have prevented
some insects from recolonizing these areas. Probably many streams
in glaciated areas now carry much more finely divided material
than they did preglacially. Forbes and Richardson ( 1920:civ-cvii)
produced an impressive list of fish that avoid turbid streams in the
area covered by the Illinoian ice sheet in southern Illinois.

Reference has been made to two fishes (rosyface shiner and
rainbow darter) whose ranges may have been restricted by glacia-
tion. These species seem to have reinvaded areas covered by the
Wisconsin ice sheet, but not those covered by earlier ice sheets.
The gilt darter and banded darter also exhibit this type of distri-
bution. Possibly this relationship is due to the fact that Wisconsin
drift in the main consists of coarser material than the drift of earlier
ice sheets (Flint, 1957:341), and thus contributes less clay and silt
to the streams. Some fishes scarcely penetrate glaciated areas at
all, but are distributed in relation to the glacial border in ways
that suggest they enjoyed a more widespread distribution in these
areas at times in the past. The Ozark minnow and largescale stone-
roller occur in the Driftless Area (of Wisconsin and adjoining
states ) and in the Ozark Uplands, but are absent from the glaciated
area between. Possibly these two species reinvaded the Driftless
Area postglacially from the Ozark Uplands, but more likely they
survived the last ice advance in both the Ozark Uplands and the
Driftless Area, and have failed to redisperse into the glaciated
areas. In any case, these two fishes must have been more wide-
spread in the upper Mississippi Valley in preglacial or inter glacial
times than they are today.

Certain Ozark fishes are represented in the upper Ohio Valley
by closely related species that may have been derived from a
common ancestral stock that had a widespread preglacial distri-
bution in glaciated regions. The ancestral species probably became
localized in the Ozark Uplands and unglaciated portions of the
upper Ohio Valley during an early ice adxance, and gave rise to

Missouri Fishes 299

geminate forms aftc>r failing to rcoccupy much of their former
range.

The Missouri saddled darter and its sister species the variegate
darter (Ethcostoma variaium) provide a striking example. Both
are abundant within their areas of occurrence, and approach the
glacial boundary without penetrating far into glaciated regions.
The Missouri saddled darter is endemic to the northern slope of
the Ozark Uplands, whereas the variegate darter inhabits un-
glaciated streams that were tributary to the preglacial Teays River.
The variegate darter occurs also in the Green River and a few other
streams that were tributary to the preglacial Ohio River, but these
populations could have resulted from slight downstream movement
since deflection of the upper Teays into the Ohio system. Absence
of the variegate darter or any near-relative in the lower Ohio or
Tennessee systems argues against the occurrence of the common
ancestral stock in the preglacial Ohio River. More plausibly, the
common stock occurred throughout the preglacial Teays River
system, and in parts of the preglacial upper Mississippi, Iowa, and
lower Missouri systems.

The longnose darter and the closely related sharpnose darter
(Percina oxyrhijncha) may have had a distributional history simi-
lar to that of the Missouri saddled and variegate darters. However,
the longnose and sharpnose darters are more localized in distribu-
tion, and do not now occur in proximity to the glacial border. The
sharpnose darter is localized in the Cheat and New rivers of Vir-
ginia and West Virginia, suggesting that the common ancestral
stock of these two species may have inhabited the ancestral Teays
and other streams of glaciated regions.

Dispersal and isolation of upland fishes as a residt of changing
base levels in the lower Mississippi Valley. — Many Ozark fishes in
addition to those cited in the previous section are represented by
disjunct populations of the same or a closely related species in
unglaciated uplands east of the Mississippi River. However, these
fishes differ from the former group in that the eastern relative is
widespread in the area drained by the ancestral Ohio River, sug-
gesting that these species or their ancestral stocks could have had
a continuous distribution by way of the lower Ohio and Mississippi
rivers, rather than by way of the preglacial Teays. It is unlikely
that all of the distribution patterns to be discussed are of the same
age. Perhaps those involving geminate species date from pre-
glacial times or dispersal early in the Pleistocene, while those in-

300 Unwersity of Kansas Publs., Mus. Nat. Hist.

volving disjunct populations of the same species may date no
further back than the Wisconsin ice advance.

All of these fishes inhabit upland streams, and the broad low-
lands of the Mississippi Embayment are now an important barrier
to their dispersal. To account for their present distribution, we
must assume that conditions in the Mississippi Embayment were
better suited for upland fishes in the past, or that these fishes cir-
cumvented this barrier. Ross (1965:588-589) proposed a mecha-
nism by which certain aquatic insects could ha\'e dispersed bet\\'een
the Ozark and Appalachian uplands. According to Ross, dispersal
occurred through the hilly country around the head of the Missis-
sippi Embayment. He suggested (p. 589) "that changes in the
ecological conditions of the Illinois Ozarks associated with Pleisto-
cene events alternately permitted the dispersal of ancestral forms
between eastern and western areas and then broke the ranges at
this point and produced geographic isolation between the eastern
and western populations of the species." Ross further suggested
that for insects the dispersal had all been from east to west, be-
cause no extensive lineages in these groups have evolved in the
Ozarks, and he dated the initiation of dispersal as "not earlier than
late Pliocene or early Pleistocene."

This explanation is appealing, and may account for the dispersal
of aquatic insects and other animals that can move overland; how-
ever, fishes must have continuous waterways for their dispersal.
Because the juncture of the Ohio and Mississippi rivers was far
south of its present location through much of the Pleistocene, east-
west dispersal of fishes through the hills around the head of the
Mississippi Embayment is unlikely. Geological evidence presented
by Fisk (1944:5) may have an important bearing on the problem
of fish dispersal. He indicated that

"During the last glacial stage, when sea levels were sev-
eral hundred feet lower than at present, the MississiDpi
River valley became deeply incised within the coastal plain
sediments. Slow aggradation during and subsequent to the
period of rising sea level, when the glacial ice masses were
melting, incompletely filled this entrenched valley system.
. . . The topography buried under the alhnium is rugged,
and in many places the bottoms of the trenches extend far-
ther below the olain than the exposed \alley walls rise abo\e
it. The alknial plains of tributary \'al]eys also overlie deep
trenches and are bordered by steep ^'alley walls."

Fisk indicated further ( 1944:11) :

"The entrenched streams were incised in steep-wallcd
valleys whose sides were sculptured by many small tribu-

Missouri Fishes 301

taries . . . The streams had stiaighter courses and steeper
gradients than the present ones and were incised in generally
little resistant sedimentary rocks."

Environmental conditions in the Mississippi Embayment may
not ha\e been inimical to upland fishes at the time of lowest base-
leN el. Perhaps many small, direct tributaries of the Mississippi and
Ohio rivers in the Embayment area provided habitats like those
now foimd in small streams along the eastern margin of the Ozarks.
If so, upland fishes could ha\e dispersed through the Mississippi
and Ohio ri\ers in stepwise fashion from one habitable tributary
to another. Aggradation subsequent to the last glacial stage pro-
duced the environmental conditions now prevailing in the Embay-
ment, restricting further dispersal by upland fishes. All of the
glacial and interglacial periods probably were accompanied by al-
ternate entrenchment and aggradation in the Mississippi Embay-
ment, and this would seem to provide an adequate mechanism for
the alternate dispersal and isolation of populations east and west
of the Embayment.

Species whose disjunct distribution may be explained in this
manner are: least brook lamprey, telescope shiner, whitetail shiner,
bluntface shiner, slender madtom, northern studfish, bluestripe
darter, greenside darter, and banded darter. In some the eastern
and western populations are morphologically distinguishable, but
in others they are not. Some show evidence of having more than
one wave of dispersal. Miller (1968:17) envisioned two separate
westward movements of a subspecies of greenside darter (Etlieo-
stoma hlennioides newmani) from the Tennessee and Cumberland
stream systems into the southern Ozarks. Possibly the occurrence
of disjunct populations of the barred fantail darter (£. /. jiaheUare)
in the Black River system, and of a closely allied but phenotypically
distinct form in the White River system can best be explained in
this manner.

Two Ozark species (Ozark shiner and Niangua darter) are rep-
resented east of the Mississippi Embayment by related species
( mirror shiner and arrow darter ) that may have been derived from
ancestral stocks localized in the region drained by the preglacial
Ohio or Tennessee rivers. It is uncertain whether the present
distribution results from fragmentation of a more widespread pre-
glacial distribution, or from east-west dispersal during the Pleisto-
cene. In either case, dispersal appears to have been across the
Mississippi Embayment by way of the lower Ohio and Mississippi
rivers.

302 University of Kansas Publs., Mus. Nat. Hist.

The dispersal of amblyopsid cavefish pro\ides a special case,
inasmuch as they seem dependent on subterranean channels
(Woods and Inger, 1957:253). Two species occur in Missouri.
One (Ozark cavefish) is endemic to the Ozark Uplands; its closest
relative is the northern cavefish {Ambh/opsis spelaea DeKay) of
the Interior Low Plateaus. The other Missouri species (southern
cavefish) occurs disjunctly in the Ozark Uplands and Interior Low
Plateaus. Woods and Inger (1957:254) suggested that amblyopsid
cavefish have undergone more than one wave of dispersal, and
indicated that fluctuations in the water table have influenced dis-
persal. It is likely that the alternate cycles of erosion and sedi-
mentation in the Mississippi Embayment during the Pleistocene
were accompanied by fluctuations in the ground-water table, ac-
counting for the dispersal and present east-west disjunction in the
distribution of cavefishes.

Western component: faunal exchange hetxceen central Missis-
sippi Valley and icestern drainages. — During late Tertiary and
early Pleistocene times, drainage of much of the central and south-
ern plains west of the Flint Hills and Ozark-Ouachita uplands was
southward, perhaps into the ancestral Red River (Frye and Leon-
ard, 1952:199; Metcalf, 1966:66). Probably this stream system
(Preglacial Plains Stream of Metcalf) was sufficiently isolated from
the central Mississippi Valley to permit the development of a dis-
tinctive fish fauna. Diversion of much of the western drainage
eastward into the Missouri and lower Arkansas rivers during the
Pleistocene has permitted mixing of eastern and western faunas,
blurring the distinctions that formerly prevailed.

Metcalf ( 1966:64-71) reviewed the geological and zoogeographi-
cal evidence for the "Preglacial Plains Stream" system and pre-
sented a list of fishes that could have originated there. These fishes
are all widespread in the now-separate drainages thought to have
been a part of this stream system. They either fail to range very
far northward or eastward into the upper Missouri and Mississippi
River systems, or appear to have invaded those systems secondarily.
Most species range southwestward into coastal streams of Texas
and northeastern Mexico, or are represented there by closely re-
lated forms. Some are also represented by close relatives in the
central Mississippi or upper Missouri systems. Fishes so designated
by Metcalf are: southern flathead chub (Hijhopsis gracilis gulonel-
lus), suckermouth minnow, red shiner (Notropis I. lutrensis), plains
sand shiner (N. stramineus missuriensis) , Topeka shiner, plains
minnow, northern river carpsucker (Carpiodes c. carpio), plains

Missouri Fishes 303

killifish, plains topminnow, green sunfish, orangespottcd sunfish,
and plains orangothroat darter ( Etiieostoma spectahile pulcliel-
luni ) . I think it equally plausible that the Topeka shiner evolved
in the preglacial Iowa or lower Missouri drainage, and achieved
its present distribution in the Arkansas River system through stream
connections that existed between the Kansas and Arkansas systems
at \arious times during the Pleistocene.

Except for the Topeka shiner and plains topminnow, all the
species listed by Metcalf now inhabit turbid plains streams, sug-
gesting that they CNolved in an environment of this type. Metcalf
(1966:64) \isualized late Pliocene streams of the Great Plains as
"shallow, sediment-laden, at least periodically, and meandering
over a terrain with little gradient," and this would seem to provide
conditions fa\'orable for these species. But if his "Preglacial Plains
Stream" received drainage from the Flint Hills and the western
slope of the Ozark-Ouachita uplands, it may have had tributaries
of a clear, upland type. Since these tributaries had no direct con-
nections with streams of other upland areas, conditions should have
been fa\orable for the development of a distinctive upland fauna
in this basin.

Seven kinds of Missouri fishes that arc characteristic of clear,
upland streams are restricted to the Flint Hills and western slope
of the Ozark-Ouachita uplands or have limited distribution else-
where. All are represented eastward in the Mississippi Valley by
related species or subspecies that could have evolved from a com-
mon ancestral stock.

Fishes exhibiting this distributional relationship are listed below.

Western Form Eastern Counterpart

Duskystripe shiner Bleeding shiner

Bluntface shiner Whitetail shiner

Southwestern mimic sliiner Northern mimic shiner

( Notropis volucelhis ssp. ) ( Notropis v. vohiceUiis)

Western slim minnow Eastern slim minnow

{Pimephales tenellus tenellus) {Pimcphales t. parviceps)

Xeosho madtom Northern madtom

(Notunis placidus) (Noturus stigmosus)

Arkansas Ri\er scaly orangethroat Northern orangethroat darter

( Etiieostoma spectahile squamosum ) ( Etiieostoma s. spectahile )

Stream connections between the upper Arkansas and middle
Missouri systems appear to have been important in the southwest-
ward dispersal of certain fishes. These fishes are all primarily
northern and eastern in distribution, and occur disjunctly in the
upper Arkansas system. Dispersal by way of the lower Arkansas
River cannot be ruled out, but most of these species have such
a limited distribution in the lower Mississippi Valley that this seems

304 Unwersity of Kansas Publs., Mus. Nat. Hist.

unlikely. Stream connections between the Kansas and Arkansas
rivers in mid-Pleistocene, and possibly as recently as early Wiscon-
sin time, provide the most likely avenue for dispersal. However,
dispersal by way of a more eastern connection seems probable for
some species. Perhaps these species dispersed southwestward by
way of a stream capture between the Osage and Spring river sys-
tems, but there is little geological evidence for such a capture
(Bretz, 1965:97). Fishes that may have dispersed into the upper
Arkansas system by way of stream connections with the middle
Missouri system are: spotfin shiner, blacknose shiner, Topeka shin-
er, stonecat, slenderhead darter, channel darter, johnny darter, and
least darter.

Movement in the opposite direction (Arkansas system to Mis-
souri system) also may have occurred. The bluntnosc darter and
slough darter occur in the Arkansas River system and disjunctly
in tributaries of the upper Osage Ri\'er in the Missouri River sys-
tem. Possibly the Osage River populations owe their present iso-
lated position to dispersal by way of connections with the Arkansas.

Importance of local stream captures in the dispersal of fishes
within Missouri. — Headwater transfers across major di\ides prob-
ably ha\'e been important in the dispersal of fishes in the Ozark
Uplands. The occurrence of many upland species in all or most
of the principal drainages of the region is strong evidence for this.
l\. V. Miller (1968:29) presented e\idence for the movement of a
subspecies of the greenside darter into the upper Gasconade River
by way of a connection with the Current (or possibly White)
River. Variational studies of other species may disclose more
examples of connections important in the dispersal of Ozark fishes.

Minor stream connections have probably had less importance
in the dispersal of fishes in northern Missouri. Most fishes of this
region are quite tolerant of conditions in large rivers, and move
readily from tributary to tributary through the Missouri and Mis-
sissippi rivers. However, populations of the redfin shiner in the
Salt River system are phenotypically more like those in the Missouri
River system than those in tributaries of the Mississippi Ri\'er
adjacent to the Salt River. Possibly occurrence of the Missouri
River form in Salt River is the result of stream capture. Evidence
for such a capture is afforded by drainage patterns in northeastern
Boone County. These patterns suggest that Silver Fork, a high-
gradient tributary of Perche Creek in the Missouri system, may have
beheaded Goodwater Creek, a tributary of the Salt River. The up-
per Silver Fork flows north for se\ eral miles on a direct line t(n\'ard

Missouri Fishes 305

the licatlw atcis of Goodwator Creek before turning abruptly west-
ward and then southwestward.

Post- Wisconsin Changes in the Fish Fauna

Several Ihies of evidence suggest that the major chmatic se-
([uence since retreat of the last ice sheet was cool- warm-cool, and
the warm phase consisted of a warm-humid period (Climatic
Optimum) and a warm-dry period (Xerothermic Interval). P. W.
Smitli (1957) analyzed the distribution patterns of terrestrial verte-
brates in relation to this chronology, and concluded that post- Wis-
consin climatic fluctuations played a significant role in determining
distribution patterns of animals in the Prairie Peninsula (midwest-
ern and northeastern United States). He cited relict populations
of 10 species north of the main body of their ranges, and suggested
tliat these relicts dated from the Climatic Optimum. Smith indi-
cated that the Xerothermic Interval caused fragmentation of the
ranges of many forest animals and eastward extension of the ranges
of grassland species. With the return of the cooler and moister
conditions of today, eastern plants and animals again came into
ascendency, and elements of the grassland fauna have persisted
only as localized relicts.

Four species of fishes (ironcolor shiner, weed shiner, bantam
sunfish, and spotted sunfish) that are primarily southern in distri-
bution occur as disjunct populations in the southwestern Great
Lakes drainage or adjacent parts of the upper Mississippi River
system or both. Possibly the disjunct populations are relicts from
the Climatic Optimum. All four species are lowland forms that
might be expected to disperse under the mesic conditions of that
period. Also, isolated populations of the bluntnose darter and
slough darter in the upper Osage drainage of Missouri may result
from westward movement by these species into the Missouri River
system during the Climatic Optimum.

Gerking ( 1947 ) suggested that the ironcolor shiner took ad-
\ antage of minor postglacial connections in its northward dispersal.
These connections presumably were early post- Wisconsin, and were
of short duration. If Gerking's interpretation is correct, this species
and by analogy others, may have dispersed northward soon after
retreat of the Wisconsin ice sheet rather than during the Climatic
Optimum. It is also possible that these range disjunctions are
c^uite recent, having resulted from environmental changes brought
on by man's activities within historic time.

Few fish have distribution patterns that indicate eastward range

306 Unwersity of Kansas Publs., Mus. Nat. Hist.

extension during the Xerothermic Interval. The bigmouth shiner
is the only noteworthy example to come to my attention. This
species has several disjunct populations in the Great Lakes Region
that indicate a more widespread eastern distribution in the past.
The habitat of the bigmouth shiner suggests that it would be
favored by climatic conditions ascribed to the Xerothermic Inter-
val, and perhaps the isolated populations are relicts from that
period (Trautman, 1957:376). P. W. Smith (1965:634) interpreted
the disjunct populations of many fishes and amphibians in the
Interior Low Plateaus and Ozark Uplands as relicts of the Xero-
thermic Interval. I have indicated elsewhere ( p. 299 ) that for fishes
these east-west disjunctions are more likely the result of rising sea
levels and aggradation in the Mississippi Embayment as the Pleisto-
cene ice sheets melted. Although this process may have continued
into the Xerothermic Interval, it was initiated much earlier, and
the resulting range disjunctions are not directly related to the
occurrence of a warm-dry period.

Changes in the Fish Fauna Within Historic Time

The short span of time since European man first came to Mis-
souri has seen changes in the fish fauna as rapid and dramatic as
any that have occurred since retreat of the Wisconsin ice sheet.
Unfortunately, the full impact of these changes is difficult to assess
because of the paucity of early fish collections. Probably these
changes were already underway by the time the first serious col-
lecting was done in Missouri in the late 1800's, and in any case
these collections are too few and are not properly distributed to
give an adequate picture of the fish fauna under pristine conditions.
Nevertheless, collections made before 1900 indicate rather marked
changes in the distribution of some species over the past 75 years,
and trends that can be discerned from the extensive series of col-
lections made since 1930 reveal that these changes are continuing.
Many species are less abundant or have more restricted distribu-
tions than formerly. Others are more abundant or have extended
their ranges, and a few exotics have been added to the fauna.
Except for the introduction of exotics and overfishing, these changes
are the result of habitat modifications. Chief among modifications
of the aquatic environment are: increased turbidity and siltation,
changes in flow regimens, stream channelization, drainage of
swamps and other natural standing-water habitats, release of in-
dustrial and domestic pollutants into streams, widespread applica-
tion of pesticides, and construction of impoundments.

Missouri Fishes 307

Table 3. — Exotic Fishes Stocked in Public Waters by the Missouri Fish Com-
mission and tlie Missouri Game and Fish Department, 1879-1933. Figures

taken from biennial reports.

Approximate Period

Numbers Stocked When Stocked

American shad, Alosa sapidissima (Wilson) _. 1,040,000 1879-1884

Pacific salmon, Oncorhyndius sp. 645,000 1878-1886

Adantic salmon, Salnio salar Linnaeus 28,000 1881-1882

Riiinbow trout, Salmo gairdneri Richardson. 3,135,000 1879-1933

Brown trout, Salmo trutta Linnaeus 260,000 1927-1933

Brook trout, Salvelintis fontinalis (Mitchill) 608,000 1879-1914

Crayling, riiymallus aicticus (Pallas) 300,000 1902-1908

Tench, Tinea tinea (Linnaeus) 425 1893-1894

Carp, Cypriniis caipio Linnaeus 870,000 1881-1896

"Yellow perch, Peica ftavescens (Mitchill).... 5,000,000 1916-1933

* This species is probably native to Missouri, but is included here because
of its rarity in natural waters.

Of numerous exotics that were early introduced into Missouri
(Table 3), only the carp has been notably successful. A few local-
ized, self-sustaining populations of the rainbow trout occur in the
Ozarks, but the long-term survival of these is doubtful. The gold-
fish has been released into natural waters on numerous occasions,
but few if any self-sustaining populations have been established.
Experimental introductions of kokanee (Oncorhynchus nerka),
brown trout {Salmo trutta), muskellunge {Esox masquinongij) , and
striped bass (Morone saxatilis) have been made in recent years by
the Missouri Department of Conservation, but the long-term
success of these cannot yet be determined.

Many advertent and inadvertent introductions have been made
of native species outside their native Missouri ranges, and some of
these have been successful. Included are: threadfin shad, studfish,
blackstripe topminnow, mosquitofish, spotted bass, smallmouth
bass, and rock bass. Recent records for the bowfin in the Missouri
River system and for the fathead minnow in the southern Ozarks
are probably the result of introductions, but it is doubtful if these
represent self-sustaining populations. The northern pike recently
has been introduced into a few reservoirs, but there is little evi-
dence as yet of natural reproduction.

It cannot be stated with certainty that any fish species has been
extirpated from Missouri as a result of man's activities, but several
that were already rare and highly localized in distribution have
not been found for many years. The longnose darter ^ was known
in Missouri only from the section of White River now covered by

^ After this report went to press, two specimens of the longnose darter were
collected from the St. Francis River, Madison County.

308 Unwersity of Kansas Publs., Mus. Nat. Hist.

Table Rock Reservoir; it has not been collected since impoundment.
The taillight shiner, cypress minnow and golden topminnow orig-
inally occurred in the Southeastern Lowlands, but they have not
been collected there since the 1940's. Recent efforts to take the
redfin darter, known only from North Fork of Spring River in
Jasper County, have been unsuccessful. The pallid shiner was
widespread in eastern and southeastern Missouri in the early 1940's
but has not been collected since 1954. The harelip sucker ( La<io-
chila lacera Jordan and Brayton) formerly occurred in the White
River near Eureka Springs, Arkansas, and must have occurred also
in that stream system in Missouri. This species probably is ex-
tinct, because it has not been collected since about 1900.

Reductions in the abundance and distribution of many Missouri
fishes are well documented. The lake sturgeon was an important
commercial fish before 1900, but it is now so rare that the taking
of one is worthy of note in newspapers. The blacknose shiner is
exceedingly rare over most of its former Missouri range. Other
species that are less abundant and widely distributed than formerly
are: paddlefish, skipjack herring, Alabama shad, pugnose minnow,
hornyhead chub, bigeye chub, striped shiner, common shiner,
Ozark shiner, central silvery minnow, slim minnow, blue sucker,
blue catfish, American eel, bluestripe darter, l)lacksidc darter, lilunt-
nose darter, and harlequin darter.

Not all species have been adversely affected by habitat modifica-
tions. The bigmouth shiner has extended its Missouri range con-
siderably within the last 25 or 30 years. The construction of large
impoundments has favored several species; notable among these
are: longnose gar, channel catfisli, white bass, largemouth bass,
spotted bass, bluegill, white crappie, black crappie, and walleye.
A few range extensions that have occurred cannot be attributed
definitely to man's activities. The threadfin shad and Mississippi
silverside were not collected in Missouri until 1963. If they were
formerly as abundant in the lower Mississippi Ri\'er as they are
now, it is unlikel)' that they woidd have been overlooked by early
collectors. Possibly the same is true for occurrence of the Sabine
shiner in the Black River, but it is so restricted in distribution that
it may have been overlooked. Recent range extensions of the mos-
quitofish and the studfish are partly the result of introductions, but
cannot be accounted for entirely in this manner. The mosquitofish
has long had access to the Neosho River system, and its absence
there until recent years indicates changes in environmental condi-
tions favorable to it. Recent establishment of the studfish in streams

Missouri Fishes 309

aloDg the wostcni and iiortlu-ni periphery of its Missouri range
probably results liom natural extensions, as well as from introduc-
tion.

Accounts of Species

Explanation of Accounts and Maps

The material in the accounts follows a uniform arrangement,
and includes for each species a discussion of distribution and
habitat in Missouri, comments on zoogeography, and references.
If the taxonomic treatment adopted here is not in accord with that
of other workers, the reasons are given under the heading "taxo-
nomic considerations." The sequence in which the families of fishes
are arranged follows that proposed by Greenwood et al. (1966).
With few exceptions, the common and scientific names used are
those in "A list of common and scientific names of fishes from the
United States and Camula" (Amer. Fish. Soc, Spec. Publ. 2, 3rd
ed., 1970).

In order to provide a nearly complete index to the literature on
Missouri fishes, I have cited in the references many papers that are
not strictly distributional or taxonomic in nature. Only papers that
make reference to fishes in Missouri are cited; original descriptions
are cited only if they are based on specimens from Missouri. If I
did not examine specimens, allocation of names and name combina-
tions was based on: (1) a knowledge of the name-changes that
various species have undergone, (2) an evaluation of descriptions
accompanying the name or name combinations, and (3) the known
distribution of species to which the name could refer. Using these
three criteria, most records could be allocated with a high degree
of confidence; in the few cases where doubt remained about the
allocation, the name or citation is preceded by a question mark.
The distributions of all species are shown on maps that coiiclude
this report.

Specimens of all species were examined at least superficially
in an attempt to discern variational trends, but changes are pro-
posed only in silvery minnows {Hijhognathiis) and stonerollers
(Campastoma). In the accounts of some additional species I have
described known variation, gi\ing results of my preliminary in-
vestigations. Unless otherwise indicated, counts and measurements
were made according to the methods outlined by Hubbs and Lag-
ler (1947:8-15).

310 Unwersity of Kansas Publs., Mus. Nat. Hist.

Petromyzontidae
Ichthyomyzon unicuspis Hiibbs and Trautman — silver lamprey

(?) Ichthyomyzon concolor: Fowler, 1921:398 (Fox Cr., tiib. of Meramec
R. ). Luce, 1933:87,95 (Mississippi R. and Kaskaskia R. near their con-
fluence, 111. ) .

Ichthyomyzon unicuspis: Hubbs and Trautman, 1937:47,60 (Mississippi R.,
Chester 111.; Mississippi R., St. Louis). Starrett, Harth and Smith, 1960
(char.; abund.; Mississippi R. local, mapped).

Distribution and habitat. — The silver lamprey is known in Mis-
souri only from the Mississippi River (Map 3). The specimen re-
ported by Fowler (1921:398) from "Fox Creek, a tributary of the
Meramec River, at a point about 26 miles from St. Louis" may have
comes from the Mississippi River, or at least from a large river. The
collection on which this record is based includes an unusual as-
semblage of big-river and small-creek fishes ( paddlefish, shovelnose
sturgeon, redbelly dace, and mottled sculpin), stiongly suggesting
a mi.xture of collections from two localities. At Fox Creek I have
collected most of the small-creek species reported by Fowler but
none of the big-river species.

In the Missouri section of the Mississippi River the silver lam-
prey is far less common than the chestnut lamprey. Starrett ef al.
(1960:table 1) recorded four silver lampreys from the Missouri
section of the river, onlv one of which was taken below the mouth
of the Missouri River. Since 1961 I have examined numerous
lampreys from the Mississippi River between the mouth of the
Missouri River and the mouth of the Ohio Rixer, and all except one
were chestnut lampreys. The reports by Bailey (1959:163) of a
silver lamprey from the Missouri River in South Dakota, and by
Cross and Metcalf (1963:187) of another presumably taken in the
Missouri River at Omaha, Nebraska, suggest that this species occurs
in the Missouri section of the Missouri River.

Adults of the silver lamprey are found only in large rivers and
lakes, except when spawning. Spawning and larval development
take place in streams of moderate size (Trautman, 1957:133).
Spawning may occur but has not been observed in some direct
tributaries of the Mississippi River in Missouri.

Zoogeography. — Although their original ranges probably have
been altered by glaciation, the silver lamprey and its nonparasitic
deriv^ative (7. fossor) are clearly northern in their affinities. Possi-
bly they were confined preglacially to the Hudson Bay or Lauren-
tian systems and gained access to the Mississippi Valley through
connectives that developed as the ice sheets moved southward. The

Missouri Fishes 311

occunciicf ot related kinds (7. casianeus and /. '^a<^ei) that seem
to be autoehthonous to the Mississippi Valley adds credence to this
explanation.

Ichthyomyzon fossor Reighard and Cummins — northern brook

lamprey

(?) Ammococtes branchialis: Meek, 1891:116 (Meramec R. near Meramec
Spring).

Distribution and habitat. — The northern brook lamprey is wide-
1\' distributed on the northern slope of the Ozark Uplands from the
Niangua Ri\er eastward to the Meramec system (Map 4). In this
area it is by far the most common brook lamprey.

Both larvae and adults of the northern brook lamprey occur in
medium-sized streams, avoiding both the smallest headwater creeks
and the larger rivers. As is the case with other brook lampreys,
this species requires clear, permanent-flowing streams having clean,
gravelly riffles for spawning, and stable beds of silt, sand, and
organic debris for larval development.

Zoogeography. — Populations of this species in the Ozark Up-
lands are broadly disjunct from the main range, which lies within
the area covered by the Wisconsin ice sheet. The record nearest
to Missouri is from the Kankakee River in Illinois (Smith, 1965b:5).
Probably the Missouri populations are a glacial relict. The Ozark
Uplands may have served as an important refugium (perhaps the
only refugium) for this species during the Wisconsin ice advance,
and perhaps also during earlier ice advances.

Ichthyomyzon castaneus Girard — chestnut lamprey

Ichthtjomijzon castaneus: Hubbs and Traiitman, 1937:44,76 (Mississippi R.,
Chester, III.; Mississippi R., St. Louis). Martin and Campbell, 1953:46
(abnnd.; Black R.). Bailey, 1959:163 (Missouri R. at Hermann, Gas-
conade Co.). Starrett, Harth and Smith, 1960 (char.; abund.; Mississippi
R. local, mapped). Fisher, 1962:426-427 (Missouri R. local.). Cross and
Metcalf, 1963:187 (Missouri R. near St. Joseph). Hanson and Campbell,
1963 (hnear distr.; Perche Cr., Boone Co.). Cross, 1967:30 (Missouri R.
local, mapped).

Distribution and habitat. — The chestnut lamprey is the most
abundant and widely distributed lamprey in Missouri (Map 5).
According to reports of commercial fishermen, parasitic lampreys
(presumably all of this species) are common in the lower Missouri
River. Abundance of parasitic lampreys decreases upstream, and
above St. Joseph they are rarely encountered. In the Mississippi
River the chestnut lamprey is common between the mouths of the
Ohio and Missouri rivers but is uncommon above the town of Rock-
port, Illinois (Starrett et al., 1960: table 1). Northward in the

312 University of Kansas Publs., Mus. Nat. Hist.

Mississippi River, /. castanetis is replaced by the silver lamprey.
Adults of the chestnut lamprey are rather common in large streams
and reservoirs of the Ozarks, and larvae tentatively identified as
this species have been collected at many localities in that region.

In its parasitic phase, the chestnut lamprey inhabits large
streams and reservoirs, where there is an abundance of the large
fishes on which it feeds. Spawning adults and larvae are found
in medium-sized creeks to moderately large rivers. The larvae
require clear, permanent-flowing streams having stable bars of silt,
sand, and organic debris in which to complete their development.
In Missouri, streams suitable for larval development are found only
in the Ozark Uplands, accounting for the progressive decrease in
abundance of the chestnut lamprey in the large rivers of Missouri
away from that region.

Zoogeography. — The present distribution of /. castanetis and its
non-parasitic derivative (I. gagei) suggests that they may have
inhabited the Mississippi Valley continuously since before the
earliest ice advance of the Pleistocene. Possibly they were not
present in the preglacial Teays, since they are now represented
eastward in the Ohio Valley by the related /. hdcUmm (Jordan),
and /. greeleiji Hubbs and Trautman.

Ichthyomyzon gagei Hubbs and Trautman — southern brook

lamprey

Distribution and liabitat. — The only specimens of the southern
brook lamprey known from Missouri are a larva tentatively identi-
fied as this species and two adults. One adult and the larva
(UMMZ 173427) were collected in the Gasconade River east of
Hartville in April 1952; the other adult was taken in the Elk River
at Pineville in May 1965 (Map 6). The habitat reciuirements of
this species are much like those of the northern brook lamprey.

Zoogeograpliy. — The Gasconade River specimens are the only
ones known from north of the Arkansas Ri\'er system in north-
eastern Oklahoma and southwestern Missouri. Perhaps the Gas-
conade River population owes its present isolated position to a
stream capture across the divide between the Gasconade and
White rivers (physiographic evidence for such a capture was given
by Bretz, 1965:46). Alternatively, 7. gagei may have had a more
widespread preglacial or interglacial distribution northward in the
Mississippi Valley, and was replaced over parts of this range by
its northern counterpart (7. fossor) during the Wisconsin or an
earlier ice advance of the Pleistocene.

Missouri Fishes 313

Lampetra lamottei (Lcsucur) — Anu'iicaii brook lamprty

I'ctroniyzoii Lamotteuii: Lesueur, 1827:9-11, pi. 6 (oiig. descr.; types toll, al
W'ilkin.son Ca\e, presumably near Mine La Motte, Madi.son Co.).

EntospJicnus hnnottoiH: Hubbs and Trantinan, 1937:22-24 (reprod. ol orig.
descr.; type local, erroneously listed in Jackson Co.).

Taxonomic considerations. — The troubled nonienclatorial his-
tory of the two eastern North Anieriean species now included in the
genus Lampetra was reviewed and brought to its present state by
Hubbs and Trautnian (1937:22-24); they shifted Petromyzon
Lamottenii Lesueur from the species now known as L. aepyptera
to the species formerly known as L. tcilderi or L. appendix. Al-
though the basis for this change has been weakened by the dis-
covery of L. aepyptera in the St. Francis River system, from which
Lesueur presumably obtained his type specimens, there is no
justification for further nomenclatural changes without additional
evidence. I have tried without success to locate Wilkinson Cave,
and I have been unable to collect any brook lampreys in the
\'icinity of Mine La Motte. The only brook lampreys known to me
from the St. Francis River system are five ammocoetes collected
in a tributary of Big Creek, six miles southeast of Ironton in Iron
County. I have identified these as L. aepyptera on the basis of
myomere count.

Distribution and habitat. — The only recent records for the
American brook lamprey in Missouri are from the Current River
system (two larvae and one adult) and a series of three larvae and
17 newly transformed adults from Castor River, Bollinger County
(Map 7).

In Missouri the American brook lamprey inhabits medium-
sized to moderately large streams, leaving the smaller creeks to
the least brook lamprey. The two have not been taken in the same
collections, although they occur in close proximity in the Current
River system.

Zoogeography. — The main body of the range of L. lamottei is
in the upper Mississippi, Ohio, and Great Lakes drainages. North-
westward into Alaska it is replaced by other species of Lampetra.
Perhaps L. lamottei was confined preglacially to the Laurentian
system or other northern drainages and entered the Mississippi
Valley by way of glacial connectives that developed during the
Pleistocene.

Lampetra aepyptera (Abbott) — least brook lamprey

Distribution and habitat. — The least brook lamprey is common
in the upper Current and Eleven Point systems and is known from

314 University of Kansas Publs., Mus. Nat. Hist.

a few localities in the White, St. Francis and Meramec systems
(Map 8). It probably has a more widespread distribution in the
Ozarks than these records indicate. The least brook lamprey is
decidedly a small-stream fish, occurring most abundantly, both as
adults and as larvae, in clear, permanent-flowing headwater creeks
and spring branches.

Zoogeography. — Lampetra aepyptera is probably autochtho-
nous to the preglacial Teays-Mississippi system. Ozarkian popula-
tions are broadly disjunct from the main body of the range east of
the Mississippi Rixer. Perhaps Ozarkian populations owe their
isolated position to fragmentation of a more widespread preglacial
range in the central Mississippi Valley. Alternatively, this species
may have invaded the Ozark Uplands during one of the Pleistocene
ice advances, when conditions in the Mississippi Embayment were
more favorable to the east-west dispersal of upland fishes.

Acipenseridae
Acipenser fulvescens Rafinesque — lake sturgeon

Acipenser rauchi: Dumeril, 1870:105,118 (orig. descr.; type local. Osage R.).

Acipenser anasimos: Dumeril, 1870:105,122 (orig. descr.; t>'pe local. Missouri
R. near St. Louis).

Acipenser ruhicundus: Forbes and Richardson, 1920:25 (decline in abund.;

Mississippi R. at Alton, 111.). Coker, 1930:150-152 (Mississippi R. below

mouth Des Moines R.; Mississippi R., Canton).
Acipenser fulvescens: Barnickol and Starrett, 1951:288-290 (utiliz.; decline in

abund. in Mississippi R.). Fisher, 1962:426 (Missouri R. near Lisbon,

Howard Co.). Cross, 1967:33-34 (Missouri R., Rule, Nebr.; Osage R.,

Mo.).

Distribution and habitat. — Before 1900 the lake sturgeon was
common in Missouri. The commercial fishing report of the U.S.
Commissioner of Fish and Fisheries for 1894 shows that Missouri
fishermen harvested 50,000 pounds of lake sturgeon in that vear.
By 1908, Forbes and Richardson (1920:25) reported that this spe-
cies was steadily decreasing in abundance and was only rarely
taken in the Illinois section of the Mississippi River. At present the
lake sturgeon is extremely rare in Missouri waters, and apparently
exists primarily as occasional stray individuals from farther north
(Map 9). A farmer along the lower Osage River informed me that
large sturgeons were taken from that stream until Bagnell Dam
was constructed. If these were A. fulvescens, reproducing popula-
tions may ha\'e persisted in Missouri at least until 1930. I have
examined the head of a specimen taken from the Missouri Rixcr
near Easley, Boone County, in a hoop net on 25 July 1967. It re-
portedly was 5 feet in length and weighed approximately 65 pounds.

Missouri Fishes 315

Also, I have seen a recognizable photograph of a specimen, re-
ported to ha\'e weighed 97 pounds, taken in the Missouri River
near Lupus, Moniteau County, in 1938. Another recent report is of
a 32-pound specimen taken in the Mississippi River near Ellsbery,
Lincoln County, in August 1966. Elsewhere in its range this fish
inhabits lakes as well as streams, but in Missouri it is known only
from large rivers. The marked decline in abundance which it has
undergone could have resulted from overfishing, increased silta-
tion, and/or the construction of dams which block its movements
and destroy its habitat.

Zoogeography. — This ancient species is northern in its affinities,
and may have long occupied the preglacial Hudson Bay, Lauren-
tian, and/or Teays-Mississippi drainages. It undoubtedly survived
the Wisconsin, and probably also earlier ice advances of the Pleisto-
cene, in the Mississippi Valley.

Scaphirhynchus platorynchus (Rafinesque) — shovelnose sturgeon

Scaphirhynchtis platorynchus: Fowler, 1921:398 (young example; Fox Cr.,
trilx Meramec R.). Barnickol and Starrett, 1951:288-290 (iitiliz.; hab.;
abund.; Mississippi R. local.). Bailey and Cross, 1954:169-199 (char.;
compar.; hab.; Missouri local, compiled and mapped). Fisher, 1962:427
(Missouri R. local.). Cross, 1967:34 (Missouri R. local, mapped).

Distribution and habitat. — The shovelnose sturgeon is the most
abundant sturgeon in the Missouri and Mississippi mainstreams but
is rare elsewhere in the state (Map 10). The record by Fowler
(1921:398) is doubted for reasons given in the account of /. uni-
cuspis. In the Michigan collections there are five specimens
(UMMZ 177858) with a label indicating that they were taken in
the Niangua River drainage by L. C. Salyer on August 27-29, 1931.
Bailey and Cross ( 1954 ) made no mention of these specimens and
did not include this locality on their distribution map. This record,
if authentic, indicates that the shovelnose sturgeon was present in
the Osage River and its major tributaries before construction of
Bagnell Dam. I have reliable reports of shovelnose sturgeon taken
by fishermen in the lower Current River in recent years.

The shovelnose sturgeon inhabits the open channels of large
rivers, and is usually found in a strong current over a firm sand
or gravel bottom.

Zoogeography.— Metcali (1966:94-95) suggested that the shov-
elnose sturgeon may be autochthonous to the preglacial Teays-
Mississippi system, perhaps occupying that system at a time when
ancestral stocks of the closely related S. albiis were localized in the
Hudson Bav svstem.

316 University of Kansas Publs., Mus. Nat. Hist.

Scaphirhynchus albus (Forbes and Richardson) — pallid sturgeon

Vara.scaphirluimhiis alhus: Forlies and Richardson, 1905:37-44 (orif^. descr.;
type local. Mississippi R. at Grafton, 111.; also in lower Missouri R.). Forbes
and Richardson, 1920:29 (Mississippi R. at Grafton and Alton, 111.).
Rarnickol and Starrett, 1951:290 (abund.; Mississippi R. abo\e mouth
Missouri R.).

Scaphirhynchus album: Bailey and Cross, 1954:169-190,199-202 (char.; corn-
par.; hal).; Missouri R. at Rocheport, and just above confl. with Mississippi
R.).

Scaphiih/nchus aJhus: Fisher, 1962:427 (Missouri R. at Easley. Rocheport,
Lexington, and St. Joseph). Cross, 1967:37 (Missouri R. local, mapped).

Distribution and hahitat. — The pallid sturgeon occurs in the
Missouri and lower Mississippi rivers, and ascends the Mississippi
River upstream from the mouth of the Missouri for only a few
miles (Map 11). It seems to be extremely rare in Missouri. The
habitat of the pallid sturgeon seems to be much like that of the
shovelnose sturgeon.

Zoogeography. — If the pallid sturgeon evoKed in the preglacial
Hudson Bay system as suggested by Mctcalf (1966:95), its presence
in the Mississippi Valley dates from the time of the southward
deflection of the upper Missouri Rixer during the Pleistocene.

Polyodontidae
Folyodon spathula (Walbaum) — paddlefish

Procerus maculatus: Rafinesque, 1820:87 (orig. descr.; type local. Ste. Gene-
vieve, Mo.).

Poh/oclon spatliula: Carman, 1890:148 ( Mi.ssis.sippi R. near Quincy, 111.).
Ellis, 1916:93-94 (Dry Wood Cr. near Nevada, Vernon Co.). Forbes and
Richardson, 1920:17 (abund.; bavous Mississippi R. about Alton). Fowler,
1921:398 (young; Fox Cr., trib. \Ieramec R.). Borges, 1950:16-33 (Nian-
gua Arm, Lake Ozark). Rarnickol and Starrett, 1951:290-291 (utiliz.;
abund.; hab.; Mississippi R. local.). Wilson, 1956 (para.sites; Missouri R.
near St. Joseph). Purkett, 1961 (reprod. and early develop.; Osage R.).
Fi.sher, 1962:427 (Mis.souri R. local.). Hanson and Campbell, 1963 (linear
distr.; Perche Cr.). Purkett, 1963 (artif. prop.; Osage R.). Needham, 1965
(artif. induct, spawning; Osage R.). Ballard and Needham, 1964 (em-
bryol.; Osage R. ). Cross, 1967:38 (Missouri R. local, mapped).

Paddlefish: Fry, 1962 (harv.; Black R. below Clearwater Res.).

Distribution and haJ)ifat. — The paddlefish probably occurs in
all the major rivers of the state where its movements are not blocked
by dams (Map 12). It now seems less abundant in the Mississippi
River than formerly (Rarnickol and Starrett, 1951:291). Destruc-
tion of spawning grounds and blocking of movements by dams,
along with overfishing, are probably the major factors responsible
for this decline. A similar decline has occurred in the Missouri
River, perhaps as a result of channelization and the consequent
elimination of backwaters. Construction of Kaysinger Bluff Dam
on the Osage River near Warsaw uill probably destroy the princi-

Missouri Fishes 317

pal spawning grounds for the large population of paddlefish pres-
ently found in Lake of the Ozarks. Since no suitable spawning
areas are present above the reservoir to be created by this dam,
a marked decline in abundance of the paddlefish in the Osage River
can be expected. This fish inhabits quiet pools and backwaters of
large rivers. It thrives in large man-made impoundments if they
have large tributaries that are suitable for spawning. Construction
of Lake of the Ozarks has provided these conditions.

ZoogeograpJiy. — This ancient fish probably has long inhabited
the Mississippi Valley.

Lepisosteidae

Lepisosteus spatula Lacepede — alligator gar

Lcpisostcus spatula: Barnickol and Stanctt, 1951:320-321 (utiliz.; abund.;
Mississippi R. local.).

Distribution and Habitat. — The alligator gar is known in Mis-
souri only from the Mississippi River, where it is rare (Map 13).
Forbes and Richardson (1920:35) reported that this species ascends
the Mississippi River above St. Louis, but they hsted no specific
localities, and gave no indication of abundance. Barnickol and Star-
rett (1951:320) reported the alligator gar from five stations in the
Missouri section of the Mississippi River; 80 of their 85 specimens
were taken at Cairo. The only recent reports known to me from
the Mississippi River where it bounds Missouri are of two speci-
mens taken by fishermen in 1965. One specimen, weighing 110
pounds, was taken near Chester, Illinois; the other, weighing 130
pounds, was taken near Cairo. These records, along with photo-
graphs of the specimens, were furnished by Philip W. Smith.

The alligator gar inhabits the sluggish pools and overflow waters
of large rivers.

Zoogeography. — Lepisosteus spatula reaches the northern limit
of its range in the Mississippi River near the mouth of the Missouri.
It probably has long inhabited the Mississippi Valley.

Leijisosteus platostomus Rafinesque — shortnose gar

Cijlindrosteus agassizii: Dunieril, 1870:351 (orig. descr.; St. Louis).

Lepisosteus platostomus: Forbes and Richardson, 1920:map VI (Mississippi
R. local.). Fowler, 1921:398 (young; Fox Cr., trib. Meramec R.). Bar-
nickol and Starrett, 1951:320-321 (utiliz.; abund.; Mississippi R. local.).
Patriarche, 1953 (abund.; import.; Lake Wappapello ) . Fisher, 1962:427
(Missouri R. local.). Han.son and Campbell, 1963 (linear distr.; Perche
Cr. ). Cross, 1967:41 (Mi.ssouri R. local, mapped).

Distribution and habitat. — The shortnose gar is the most abun-
dant gar over all of Missouri except the Ozark Uplands (Map 14).
It is most often found along major rivers, in quiet pools, backwaters.

318 University of Kansas Publs., Mus. Nat. Hist.

and oxbow lakes. It seems to be more tolerant of high turbidity than
the longnose and spotted gars.

Zoogeography. — The preference by L. platostomus for large
turbid rivers and its distributional relationships with the related
L. ocidatus suggest that L. plotostomtis may have occurred pre-
glacially in the Hudson Bay system and entered the Mississippi
Valley when the upper Missouri River was deflected southward
during the Pleistocene.

Lepisosteus oculatus (Winchell) — spotted gar

Lepisosteiis pioductus: Funk and Campbell, 1953:79 (abund.; Black R.).
Patriarche, 1953 (abund.; import.; Lake Wappapello ) .

Distribution and habitat. — The spotted gar is rather common
in the ditches of the lowlands (Map 15). It was not reported by
Barnickol and Starrett (1951) from the Mississippi River, but the
records of George V. Harry indicate that Barnickol and Shoemaker
took a specimen near Grafton in 1944. The spotted gar is known
from the Neosho River in southeastern Kansas (Branson and Hart-
mann, 1963:591), and therefore is to be expected in southwestern
Missouri. In most Missouri collections, the spotted gar has been
outnumbered by the shortnose or longnose gars. It seems less
tolerant of turbidity, and shows a greater affinity for aquatic vege-
tation than the other gars; it attains its greatest abundance in
quiet, clear waters having considerable aquatic vegetation or stand-
ing timber.

Zoogeography. — The widespread distribution of L. oculatus in
Gulf coastal drainages, and presence of the closely related L. platy-
rhinciis DeKay in Florida, suggest that this species may have
long occupied the Mississippi Valley.

Lepisosteus osseus (Linnaeus) — longnose gar

LepUlosteus osseus: Jordan and Meek, 1885:13,16 (Missouri R. at St. Joseph;
Grand R. at Clinton and/or Tebo Cr. at Calhoun).

Lepisosteus osseus: Forbes and Richardson, 1920:map V (Mississippi R. lo-
cal.). Borges, 1950 ( Niangua Arm, Lake Ozark). Cleary, 1956:map 5
(Des Moines R. local.). Fisher, 1962:427 (Missouri R. local.). Netsch and
Witt, 1962 (life history; Osage and Pomme de Terre ri\ers, Benton Co.;
Perche Cr., Boone Co.; Loose Cr., Osage Co.). Hanson and Campbell,
1963 (linear distr.; Perche Cr. ). Cross, 1967:43 (Missouri R. local,
mapped ) .

Lepisosteus osseus oxyurus: Barnickol and Starrett, 1951:320-321 (utiliz.;
abund.; Mississippi R. local.). Martin and Campbell, 1953:46 (abund.;
Black R.).

Distribution and habitat. — The longnose gar is nearly statewide
in distribution and occurs in all the major stream systems (Map 16).
It is the only gar in the clear, high-gradient streams of the Ozarks

Missouri Fishes 319

and reaches its greatest abundance in large reservoirs of that re-
gion. Elsewhere it is less abundant at most localities than the
shortnose gar.

The longnose gar typically inhabits the sluggish pools, back-
waters, and oxbows along large, moderately clear streams. Adults
usually are found in large, deep pools; young occur in shallow
backwaters, often in thick beds of aquatic vegetation. Spawning
is preceded by upstream migrations into smaller streams than those
inhabited at other times, and the young usually remain in these
smaller streams during the first summer of life.

Zoogeography. — Lepisostetis osseus was likely an inhabitant of
the Teays-Mississippi system but may have occupied other pre-
glacial drainages as well.

Amiidae
Amia calva Linnaeus — bowfin

Amia calta: Forbes and Richardson, 1920:map VII (Mississippi R. local.).

Barnickol and Starrett, 1951:321-322 (utiliz.; abund.; Mississippi R. local).

Funk and Campbell, 1953:72-81 (abund.; Black R. local.). Patriarche,

1953 (abund.; import.; Lake Wappapello ) . Fisher, 1962:427 (Missouri

R. at St. Charles).
Amiatus calvtis: Fowler, 1921:398 (Fox Cr., trib. Meramec R.).
Bowfin: Fry, 1962 (harv.; Black R. below Clearwater Res.).

Distribution and habitat. — The bowfin occurs in the lowlands
and northward along the Mississippi River (Map 17). Barnickol
and Starrett (1951:321) found this fish to be more common in the
Mississippi abo\e the mouth of the Missouri River than below.
It shows an affinity for clear, quiet water; this may account for the
decrease in abundance in the swifter, more turbid water below the
mouth of the Missouri River. Fisher (1962:427) took only a single
specimen near St. Charles in his extensive survey of the Missouri
River. The only other record known to me from the Missouri River
system in Missouri is of a 4/2 pound specimen taken from Fishing
River in Clay County in June, 1966. Cross (1967:46-47) examined
a specimen caught by a fisherman in Independence Creek, Atchison
County, Kansas, in 1965, and heard reports of others caught in
oxbows and backwaters of the Missouri River \\'here it borders
Kansas. Except for the specimen reported by Fisher from near
St. Charles not far above the mouth of the Missouri, all recent
records for the Missouri system in Missouri may be attributable
to introductions. Bailey and Allum (1962:30) reported that they
knew of no firm evidence for the native occurrence of the bowfin
in the Missouri River system of South Dakota.

320 Unr'ersity of Kansas Publs., Mus. Nat. Hist.

In the lowlands the bowfin is found in a variety of habitats,
including swamps, sloughs, borrow pits, ditches, abandoned stream
channels, and the sluggish lower reaches of Ozark streams that
enter from the north. In the ditches it is most abundant in sections
having thick beds of aquatic xegetation or other cover. Along the
Mississippi River the bowfin is more often found in backwaters and
oxbows than in the main channel.

Zoogeography. — This ancient species probably has long inhab-
ited the Mississippi Valley. It was reported from Pliocene fossil
deposits in Nebraska by C. L. Smith ( 1962:506).

AnguiUidae
Anguilla rostrata (Lesueur) — American eel

Anguilla rostrata: Coker, 1930:171-173 (al:)und.; nat. hist.; Mississippi R. at
Canton and Alexandria). Funk and Campbell, 1953 (alnmd.; Black R.
local.).

Anguilla hostoniensis: Barnickol and Starrett, 1951:291-292 (abund.; Missis-
sippi R. local.). Berner, 1951:table 5 (comnier. catch; Missouri and Mis-
sissippi rivers ) .

Distrihution and habitat. — The eel probably occurs occasionally
in every large stream in the state where its movements are not im-
peded by dams (Map 18). Its exact distribution and abundance
are difficult to assess, because it is not readily captured by the kind
of gear ordinarily used in fish surveys. It seems to be more abun-
dant in the southern Ozarks and the lowlands than elsewere. Coker
(1930:173) noted that the eel was declining in abundance in the
Mississippi River by the early 1900's. A decline has also occurred
in the Ozark Uplands, where high dams deny the eel access to
large areas. This fish is still present in the upper Mississippi River,
in spite of several navigation dams (Barnickol and Starrett, 1951:
291-292). The eel occurs in a variety of stream types but is most
abundant in medium-sized or large streams with continuous flow
and moderately clear water. It is most often found in the deeper
pools near logs, boulders, or other cover.

Zoogeography. — This catadromous species has probably long
occupied the Mississippi Valley.

Clupeidae
Alosa chrysochloris (Rafinesque) — skipjack herring

Poinolohus chrysocJiloris: Forbes and Richardson, 1920:49 (Mississippi R. at
Alton). Coker, 1930:165-169 (Mississippi R. below Warsaw, III; Missis-
.sippi R. at Canton, Mo.). Barnickol and Starrett, 1951:323 (abund.;
utiliz.; Mis.sissippi R. local.).

Distribution and liabitat. — The highly vagile skipjack herring

Missouri Fishes 321

must occur at least occasionalK in most large rivers of Missouri
(Map 19). In the Mississippi it was formerly abundant above the
moutli ol tlic Missouri Ri\cr but has undergone a marked decline
since that section of the ri\er was iiupounded (Coker, 1930:169;
Barnickol and Starrctt, 1951:323). It is rare between the mouths
of the Missouri and Ohio ri\'ers but is abundant in the Mississippi
Ri\cr below- the mouth of the Ohio. Elsewhere only occasional in-
di\iduals have been found. The skipjack herring inhabits the open
waters of large rivers. It seems intolerant of extreme turbidity, as
indicated by the paucity of records for the Missouri River and the
Mississippi Ri\'er downstream from the mouth of the Missouri.

Zoogeography. — Alosa chrysochloris probably was derived from
marine stocks that became adapted to life in the larger inland
streams of the Mississippi River system and other Gulf coastal
drainages.

Alosa alabamae Jordan and Evennann — Alabama shad

Distribution and habitat. — The Alabama shad was not collected
in Missouri until 1940. Since that time it has been taken at least
eight times, from the Meramec, Gasconade, and Osage rivers ( Map
20). Previously it had been reported in the Mississippi River sys-
tem (as Alosa ohiensis Evermann) only from the Ohio River at
Louisville (Evermann, 1902), from the Mississippi River near Keo-
kuk, Iowa (Coker, 1930:169-171), and from the Poteau River,
Oklahoma (Hutchins and Hall, 1951:83-84). Evidently it has de-
clined in abundance in the Mississippi River system since the early
1900's, because both Evermann and Coker judged it to be common
enough to support a limited commercial fishery.

That the Alabama shad is anadromous in the Mississippi River
system, as elsewhere in its range, was suggested by Coker (1930:
171), who noted that it was present near Keokuk only from early
May to the middle or latter part of July, and that all fish were in
a spawning condition. From this he concluded that its appearance
at Keokuk coincided with a spawning migration. Data from the
Missouri collections support this conclusion. The only adult speci-
men was collected in July; all of the remaining 90-plus specimens
are young-of-the-year collected between late July and early October.
The scarcity of adults suggests that they are present only briefly,
and the occurrence of young only in late summer and early fall
suggests a movement into other waters before the second summer
of life.

Zoogeography. — The distribution of A. alabamae suggests an

322 University of Kansas Publs., Mus. Nat. Hist.

origin in the eastern part of the Gulf of Mexico. It seems to be most
closely allied to the American shad, Alosa sapidissima (Wilson),
and may have been derived from a common ancestral stock that
invaded the Gulf of Mexico at a later time than the skipjack herring.

Dorosoma cepedianum (Lesueur) — gizzard shad

Dorosoma cepedianum: Jordan and Meek, 1885:14,17 (Missouri R. at St.
Joseph; Grand R. at Clinton and/or Tebo Cr. at Calhoun). Meek, 1891:
122 (Little Piney Cr. near Arlington). Forbes and Richardson, 1920:map
VIII (Mississippi R. local.). Borges, 1950 ( Niangua Arm, Lake Ozark).
Barnickol and Starrett, 1951:323 (abund.; Mississippi R. local.). Patri-
arche, 1953 (abund.; import.; growth; Lake Wappapello). Martin and
Campbell, 1953:46-64 (abund.; hab.; Black R.). Funk and Campbell,
1953:72-81 (abund.; Black R. local.). Cleary, 1956:map 10 (Des Moines
R. local.). Purkett, 1958a: 12.5-126 (growth; import.; Salt R.). Patriarche
and Campbell, 1958:251-252 (Clearwater Lake; abund.; growth; import.).
Fisher, 1962:427 (Missouri R. local.). Hanson and Campbell, 1963 (linear
distr.; Perche Cr. ).

Gizzard shad: Purkett, 1958b:4,21,43 (growth; length-weight relationship;
Missouri streams). Fry, 1962 (harv.; Black R. below Clearwater Res.).
Hanson, 1962 (abund.; utiliz. by largemouth bass; Table Rock Res.).

Distribution and habitat. — The gizzard shad is one of the most
widely distributed of the larger fishes in Missouri and occurs at
least occasionally in every principal stream in the state (Map 21).
It is most abundant along the Missouri and Mississippi rivers and
in large reservoirs. This fish inhabits lowland lakes and ponds,
man-made impoundments, and the pools and backwaters of streams.
It occurs in both extremely clear and very turbid waters, and ex-
hibits no obvious preference for a particular bottom type. D. cepe-
dianum avoids streams with extremely high gradients and those
which lack large, permanent pools.

Zoogeography. — This wide-ranging species was probably pres-
ent in the Teays-Mississippi system, as well as other preglacial
drainages of eastern North America. It is essentially southern in
distribution and perhaps was not present preglacially in the Hudson
Bay or Laurentian systems.

Dorosoma petenense (Giinther) — threadfin shad

Distribution and habitat. — The threadfin shad seems to be more
common than the gizzard shad in the Mississippi Ri\'er downstream
from the mouth of the Ohio River ( Map 22) . There it outnumbered
the gizzard shad 2:1 in a series of collections made in 1963. It is
rare above the mouth of the Ohio, and apparently reaches the
upstream limit of its range in southern Perry County. I have col-
lected this species from only one locality in the ditches of the
lowlands. The threadfin has been stocked in Bull Shoals and Table
Rock reservoirs, and is now well established there. It has also

Missouri Fishes 323

been introduced into Montrose Lake in Henry County, and spawned
successfull)' in 1965; survival there is presumably dependent on
continuous warm-water discharge by a power plant. Efforts are
being made to establish the threadfin shad in Thomas Hill Reservoir,
where there is similar warm-water discharge; the status of this
stock is not yet known. The habitat of D. petenen.se is similar to
that of the gizzard shad, except that the threadfin is more often
found in strong current.

Zoogeography. — Dorosoma petenen.se probably inhabited the
preglacial Teays-Mississippi system. It is probably native to the
lower Mississipi and Ohio rivers but seems to have increased in
abundance there over the last few decades. Recent changes in the
distribution of this species were discussed by Minckley and Krum-
holz, 1960:176-178.

Hiodontidae
Hiodon alosoides (Rafinesque) — goldeye

Hyodon tergisus: Girard, 1858:333, plate LXXV, figs. 1-4 (St. Louis, Mo.).

Hiodon alosoides: Jordan and Meek, 1885:14 (Missouri R. at St. Joseph; Tabo
Cr. 6 mi. E Le.xington). Fisher, 1962:427 (Missouri R. local.). Hanson
and Campbell, 1963 (linear distr.; Perche Cr.). Cross, 1967:56 (Missouri
R. local, mapped).

Amphiodon alosoides: Borges, 1950 (Niangua Arm, Lake Ozark). Barnickol
and Starrett, 1951:322 (abund.; Mississippi R. local.).

Distribution and habitat.— The goldeye inhabits the Missouri
and Mississippi rivers and the larger prairie streams of the Missouri
system (Map 23). In the Mississippi River it is much more com-
mon below the mouth of the Missouri River than above. This fish
is found most often in the open waters of large rivers, where it
frequents areas with strong current as well as quieter waters. It
occasionally occurs in deep pools of small rivers and creeks, where
these are adjacent to large rivers. The goldeye is more tolerant
of continuous high turbidity than the related mooneye.

Zoogeography. — The habitat preferences of H. alosoides, and
its distributional relationships with the related H. tergisus, suggest
an origin for H. alosoides in the preglacial Hudson Bav system
(Metcalf, 1966:97).

Hiodon tergisus Lesueur — mooneye

Hiodon tergisus: Borges, 1950 (Niangua Ann, Lake Ozark). Barnickol and
Starrett, 1951:322 (abund.; Mississippi R. local.). Martin and Campbell,
1953:46 (abund.; Black R.). Funk and Campbell, 1953 (abund.; Black
R. local.). Fisher, 1962:427 (Missouri R. local.).

Distribution and habitat. — The mooneye is nowhere abundant
in Missouri, but it occurs frequently in the Mississippi River and

324 Unwersity of Kansas Publs., Mus. Nat. Hist.

in the larger streams and reservoirs of the Ozarks (Map 24). It is
rare in the Missouri River and the larger prairie streams, where it
is largely replaced by the goldeye. The habitat of the mooneye is
much like that of the goldeye, except that it generally occurs in
clearer and quieter waters.

Zoogeography. — The distributional relationships of this fish and
the closely related H. alosoides suggest an origin for H. tergistis in
the preglacial Teays-Mississippi River system (Metcalf, 1966:97).

Salmonidae
Salmo gairdneri Richardson — rainbow trout

Rainbow trout: Ma\nard, 1889:55-56 (growth and reprod. after stocking;

spring at head of Spring R., Lawrence Co.). Fry, 1962 (harv.; taihvaters

of Table Rock and Taneycomo reservoirs).
Salmo iridens: Meek, 1891:118 ( Meramec Spring).

Distribution and habitat. — The rainbow trout is not nati\'e to
Missouri; it has had only limited success in maintaining wild popu-
lations. Small, self-sustaining populations occur in a few cold-
water streams and spring branches in the Ozarks, but the popula-
tions at most localities (Map 25) are maintained by continuous
stocking.

In recent years, experimental stockings have been made in Mis-
souri of two other salmonids. The brown trout, Salmo triitta Lin-
naeus, was stocked in some Ozark streams, and the Kokanee,
Oncorhynchus nerka (Walbaum), was stocked in Lake Taneycomo.
There is no evidence as yet of natural reproduction of either of
these species.

Esocidae
Esox americanus Gmelin — grass pickerel

Esox venniculatiis: Call, 1887:77 (in part (?), Sinking Cr., Spring Valley Cr.,
and Jacks Fork, Shannon Co.). Barnickol and Starrett, 1951:314 (Mis-
sissippi R. at Grafton, 111.). Martin and Campbell, 1953 (abund.; hah.;
Black R.). Funk and Campbell, 1953 (abund.; Black R. local.).

Lucius vermiculatus: Meek, 1891:119,130 (char.; abund.; Meramec Spring
and Meramec R. near St. James; North Fork White R., S Cabool).

Esox americanus vermiculatus: Crossman, 1966 (char.; syn.; Missouri local,
mapped ) .

Distribution and habitat. — The grass pickerel is the most abun-
dant and widely distributed pike in Missouri (Map 26). It is
common in the lowlands, and occurs sparingly in o\erflow waters
of the Mississippi River northward at least as far as St. Charles
Count)'. It is locally abundant in streams draining the eastern and
southern slopes of the Ozark LTplands but is absent from the Mis-

Missouri Fishes 325

souri and Neosho systems. In the Blaek and St. Francis systems
its distribntion largely complements that of E. nig,er with one or
the other predominating at any locality. In the lowlands the grass
pickerel frequents natural lakes, sloughs, borrow pits, and the
sluggish sections of ditches, often around thick beds of submergent
aquatic vegetation. In the Ozarks it inhabits pools of small creeks
draining undissected uplands, as well as spring pools, protected
inlets, and o\erflow waters along major streams.

Zoogeograpluj. — Grossman (1966:fig. 1) reported the grass
pickerel from the Missouri River at St. Joseph. This record is based
on a collection at the U.S. National Museum (USNM 6833) having
a label stating only "St. Joseph's, Bridger." Cross (1967:63-64)
discussed this record and suggested that the "St. Joseph's" intended
is elsewhere than in Missouri, perhaps the St. Joseph's River in
Indiana.

At present the grass pickerel occurs in the Missouri Basin only
in eastern Nebraska. Grossman (1966:18) suggested that this popu-
lation is a remnant of a more widespread distribution that was re-
duced by a recent drying trend or changes in flow patterns. It is
also possible that the Nebraska records are the result of introduc-
tions (Gross, 1967:64). If the populations are natural, it is likely
that they date from pre-Pleistocene or Pleistocene times, when the
drainage of this area was eastward through central Iowa to the
Mississippi River. If the grass pickerel formerly had a continuous
distribution downstream to the mouth of the Missouri, it should
now be present in the Osage and Gasconade systems of central
Missouri, where there appears to be an abundance of suitable habi-
tat, and where conditions for long-term survival would seem as
fa\ orable as those in Nebraska.

Occurrence of a well defined subspecies (E. a. vermicidotus
Lesueur) in the Mississippi Valley suggests that the grass pickerel
has long inhabited this region. Probably the grass pickerel has
been present in the Mississippi Valley since preglacial time.

Esox niger Lesueur — chain pickerel

Esox vermictilattis: Call, 1887:77 (at least in part; Sinking Cr., Spring Valley

Cr., and Jacks Fork, Shannon Co.).
Lucius reticuJatus: Meek, 1889:140 (reident. of a spec, recorded by Call,

1887:77 as E. vermiculatus; Spring Valley Cr., Shannon Co.).

Distribution and habitat. — The chain pickerel occurs in streams
on the southeastern slope of the Ozarks from Eleven Point River
east to the St. Francis River (Map 27). In the Eleven Point and
Gurrent rivers it is far more common than the grass pickerel. The

326 Unr'ersity of Kansas Publs., Mus. Nat. Hist.

complementary distribution patterns of the chain pickerel and grass
pickerel are perhaps the result of competition. Esox niger is found
along the major streams of the Ozarks in spring pools and marginal
waters having thick growths of submergent aquatic vegetation and
no perceptible current. Unlike the grass pickerel, it does not pene-
trate into small headwater creeks, and avoids the warmer lowland
rivers and ditches.

ZoogeograpJiy. — Esox niger has only a limited distribution in
the Mississippi Valley and may be a recent invader from the east.
Possibly it was localized prcglacially on the Atlantic Slope and
dispersed westward along the Gulf Coast into the lower Mississippi
Valley, when the lowering of sea levels accompanying Pleistocene
glaciation created stream connectives that facilitated such dispersal.

Esox lucius Linnaeus — northern pike
Esox lucius: Cross, 1967:63 (Fishing R., Ray Co.).

Distribution and habitat. — At present there seem to be no
established populations of northern pike in Missouri. All localities
plotted for this species (Map 28) are based on one or a few speci-
mens taken by fishermen. Most reports are from the lower Osage
Ri\'er, where a few northern pike are taken every year. Possibly
there is a self-sustaining population in the Osage River, but most
specimens taken in Missouri are probably strays that enter Missouri
from farther north along the Missouri and Mississippi rivers. The
northern pike has been stocked in a few reser\oirs by the Missouri
Department of Conservation in recent years, but there is little evi-
dence as yet of natural reproduction.

The muskellungc {Esox masquinongy Mitchill), a pike not
native to Missouri, is also being stocked on an experimental basis.

Zoogeography. — The widespread northern distribution of E.
lucius and its distributional relationships with other Esox suggest
that it may have been localized preglacially in drainages north of
the Mississippi Valley, and dispersed southward through glacial
connectives that de\'eloped during the Pleistocene.

Cyprinidae

Cyprinus carpio Linnaeus — carp

Cijprinus carpio: Steedman, 1884:64 (intiod. into Missonri). Borges, 1950
(Niangua Ann, Lake Ozark). Bevner, 1951:9,10 (food; Missonri R.).
Barnickol and Starrett, 1951:298-303 (utiliz.; alnnid.; Mississippi R. local.).
Patriarche, 1953:242-254 (abnnd.; import.; growth; Lake Wappapello ) .
Martin and Campbell, 1953 (abnnd.; hab.; Black R.). Fnnk and Camp-
bell, 1953 (abnnd.; B'ack R. local.). Fnnk, 1957:39-57 (movements;
Missouri streams). Purkett, 1958a: 122-123 (growth; import.; Salt R.).

Missouri Fishes 327

Patriarche and Campbell, 1958:253-254 (abiind.; import.; Clearwater
Lake). Fisher, 1962:427 (Mis.souri R. local.). Hanson and Camplx^l,
1963 (linear distr.; Perche Cr.). Cross, 1967:72 (Missouri R. local,
mapped ) .
Carp: Pin-kett, 1958b: 10,30,43 (growth; length-weight relationship; Mis.souri
streams). Fry, 1962 (harv.; tailwaters of Table Rock, Taneycomo, and
Clearwater res.).

Distribution and habitat. — A summary of early introductions of
carp into Missouri is given by Steedman ( 1884:64), and information
on subsequent stocking can be gleaned from biennial reports of the
Missouri State Fish Commission. The first carp were brought into
Missouri in the autumn of 1879, when the U.S. Fish Commission
supplied several persons with consignments of 20 carp each for
stocking in private ponds. At the same time, the Missouri Fish
Commission received 240 young carp, most of which were lost, and
the following spring they received another consignment of 500.
Between 1880 and 1885, the Missouri Fish Commission reared over
80,000 carp in their hatcheries at Forest Park, St. Louis, and St.
Joseph, and these were stocked in private and public waters
throughout the state. The program was discontinued in 1895,
largely because the carp had lost its early popularity, but by that
time it was well established in Missouri waters.

The carp is one of the most abundant and widespread large
fishes in the state (Map 29). In the Missouri and Mississippi rivers
the total poundage taken by commercial fishermen consistently
exceeds that of any other species. Carp are likewise abundant in
the prairie streams of north and west Missouri, the drainage ditches
of the lowlands, and in man-made impoundments and natural
lowland lakes throughout the state. C. carpio has been least suc-
cessful in the clear, high-gradient streams of the Ozarks, but even
there it is abundant locally in warm backwaters and in streams
polluted by domestic wastes.

The invasion of the Current and Eleven Point rivers, two of the
clearest and least polluted Ozark streams, by carp in 1957 demon-
strates the adaptability of this species In previous years carp had
made up only a very minor part of the fish population, but in 1957
they appeared in large numbers. These carp were of uniform age,
and their sudden appearance seems to have been the result of
population pressure caused by production of an unusually large
year-class in the lowland sections of these streams in Arkansas.
Carp have remained an important element in the fish populations
of Current and Eleven Point rivers since that time, but their num-
bers have progressively diminished. A few smaller carp were found

328 University of Kansas Publs., Mus. Nat. Hist.

in 1965 and 1966, indicating limited spawning success or additional
recruitment from downstream.

The carp is most abundant in large streams, natural lakes, and
impoundments that are highly productive as a result of natural
fertility or organic pollutants. It is often found in rather small
creeks if these have large, permanent pools. The carp is tolerant
of a wide range of turbidities, bottom types, and temperatures.
When not feeding, adult carp are usually found in the deeper parts
of pools, often associated with piles of drift, logs, or other sub-
merged cover.

Carassius auratus (Linnaeus) — goldfish

Goldfish: Fry, 1962 (haiv.; Clearwater Res. tailwater).

Distribution and habitat. — The goldfish has proved to be far
less adaptable and successful in Missouri waters than the carp
(Map 30). No intensive efforts have been made to establish C.
auratus, but the innumerable goldfish that are released annually
from hatcheries, bait buckets, home aquaria and artificial lakes
have afforded ample opportunity for its establishment. In spite of
this, there seem to be no self-sustaining populations of goldfish in
natural waters of the state. Goldfish are sometimes common in
streams below hatcheries, where their numbers are maintained by
continuous escape, and in large impoundments, where they are
used for bait. Otherwise there is no pattern to the distribution of
this species, and individuals may be encountered anywhere in the
state.

Notemigonus crysoleucas (Mitchill) — golden shiner

Notemigonus americanus crysoleucas: Jordan and Meek, 1885:15 (B)ack\vater
R. at Brownsville).

Nofeiiiigomis crysoleucas: Meek, 1891:118,125 (Little Dry Fork near Rol'a;
Maries R. near Dixon). Patriarche and Campbell, 1958:255-256 (abund.;
Cleanvater Lake). Fisher, 1962:427 (Missouri R. trib., Rocheport). Han-
son and Campbell, 1963 (linear distr.; Perche Cr. ).

Notetnigonits crysoleucas auratus: Barnickol and Starrett, 1951:322 (Mississippi
R. at Grafton, 111.). Martin and Campbell, 1953 (abund.; hab.; Blaek R.).

Distribution and habitat. — The golden shiner is most abundant
in the prairie and Ozark border streams of west-central and north-
eastern Missouri, and in the drainage ditches, sloughs and borrow
pits of the lowlands (Map 31). Distribution in Ozark streams is
sporadic, but N. crysoleucas is locally abundant in backwaters of
large Ozark reservoirs. The sandy, highly tmbid streams of north-
western Missouri do not provide suitable habitat for this species;
there the golden shiner occurs only as strays or highly localized

Missouri Fishes 329

populations. In Missouri, the golden shiner seldom attains the
extreme abundance that has bc^en attributed to it elsewhere in its
range, and many distributional rceords are based on one or a few
specimens.

The golden shiner is characteristic of standing-water habitats.
It is most often found in sloughs, ponds, lakes, impoundments, the
quiet pools of low-gradient streams and ditches, and the permanent
pools of small, intermittent upland creeks. It occurs only as strays
in the swifter sections of streams. The golden shiner is tolerant
of moderate amounts of turbidity, but thrives best in clear, heavily
vegetated habitats.

Zoogeographij. — The golden shiner is widespread in the eastern
United States and southern Canada. Some workers recognize three
subspecies, but Bailey, Winn and Smith (1954:12.3-124) indicated
that this is unwise, because the characters used are in part clinal
and discordant in distribution. Perhaps this pattern of variation has
resulted from mixing of stocks that were localized in different pre-
glacial drainages. If so, N. crysoleucas may have been as wide-
spread then as it is today.

Semotilus atromaculatus (Mitchill) — creek chub

SemotiJu.s atromaculatus: Jordan and Meek, 1885:12,14,16,17 (Hundred and
Two R. at Maryville; Tabo Cr. 6 miles E Lexington; Flat Cr. near Sedalia
and/or Blackwater R. at Browns\i]]e; Grand R. at Clinton and/or Tebo Cr.
at Calhoun). Call, 1887:77,80 (West Fork Black R., Tom's Cr., and
Barren Fork, Reynolds Co.; Spring Valley Cr., Shannon Co.; Big Cr.,
Texas Co.; Meramec R., Dent Co.; Bear Cr., Boone Co.). Meek, 1891:
118,122,125,126,130 (Meramec Spring, 5 mi. SE St. James; Big Dry Fork
near Meramec Spring; Little Dry Fork near Rolla; scarce in all collections,
Gasconade system; Niangua R. near Marshfield; Sac R. near Springfield;
Maries R. near Dixon; one spec, from Neosho system; Bryants Cr. near Mans-
field). Martin and Campbell, 1953:46 (abund.; Black R.). Fisher, 1962:
427 (Missouri R. local.). Hanson and Campbell, 1963 (linear distr.; Perche
Cr. ). Metcalf, 1966:tables 7 and 8 (merisHcs; Moniteau Cr., Moniteau Co.;
Richland Cr., Morgan Co.; trib. Eleven Point R., Ripley Co.). Cross, 1967:
79 ( Missoiui R. local, mapped).

Distrihiition and Imhitat. — The creek chub is one of the most
widely distributed Missouri minnows (Map 32). Its main area of
abundance is in the smaller prairie creeks of central and east-
central Missouri. Westward, it becomes increasingly spotty in dis-
tribution, but where it does occur it is often abundant. It is present
in \'irtually all small creeks and spring branches of the Ozarks but
is seldom abundant. In the lowlands it is known only from a few
small creeks in Crowley's Ridge.

The creek chub is well named, for it is most abundant in small
headwater creeks where few other fishes are present. These small
creeks often cease to flow in dry weather, and the creek chub

330 University of Kansas Publs., Mus. Nat. Hist.

survives in isolated pools, or repopiilates from below when the
creeks again begin to Row. The creek chub requires flowing water
for spawning, but does not thrive in streams with a continuous
strong flow. Such streams usually harbor numerous fish species,
and in this situation the rather generalized creek chub seems unable
to compete successfully. In many Ozark streams its niche seems
to be largely filled by the hornyhead chub, and competition with
that species may be an important factor in limiting the abundance
of the creek chub in these streams. Semotihis atwmactilatus can
tolerate moderately high turbidity, as long as the stream gradient
is sufficient to create stretches of coarse gravel for spawning. The
scarcity of this bottom type in many small streams of northern
and western Missouri and in the ditches of the lowlands may be
the most important factor limiting the distribution and abundance
of the creek chub in those regions.

Zoogeograpliy. — The creek chub occurs widely in the eastern
United States and may have been present preglacially in all the
principal stream systems. Occurrence of a disjunct population in
the upper Canadian and Pecos stream systems indicates a more
extensive southwestern distribution in the past. Perhaps this dis-
junction dates from one of the ice advances of the Pleistocene,
when the climate in the southern plains was more favorable for the
creek chub than it is today. Semotihis otromacuhtus is known from
Illinoian fossil deposits of southwestern Kansas and the Oklahoma
panhandle (Smith, C. L., 1958:177; Smith, G. R., 1963:279).

Opsopoeodus emiliae Hay — pugnose minnow

Distribution and Jtahitat. — The pugnose minnow is largely con-
fined in Missouri to the lowlands, where it is rare (Map 33). It
was formerly more abundant there, as indicated by its greater
frequency in collections made before 1945. It formerly occurred
along the upper Mississippi River, and in the Neosho system of
southwestern Missouri, but there are no recent records from these
areas. In the Neosho system of Kansas it is known only from a
single collection made in 1931 (Cross, 1967:83-84). A single speci-
men (MU 5190) was taken from the Gasconade River east of Rich
Fountain by an ichthyology class from the University of Missouri
in 1961. Dr. Arthur Witt informed me that annual seining at that
locality since 1961 has yielded no more specimens. Trautman
(1957:335-337) noted a drastic reduction in abundance of the
pugnose minnow in Ohio, attributing the decline to increased tur-
bidity and siltation, and to the disappearance of aquatic vegetation.

Missoulu Fishes 331

Similar changes may have caused a decline of this species in Mis-
souri, but habitats with clear water and an abundance of aciuatic
vegetation are still common in southeastern Missouri.

The pugnose minnow thrives only in clear waters where there
is an abundance of aquatic vegetation and no noticeable current.
Usually it is found in lentic environments or the c^uiet pools and
backwaters of low-gradient streams.

Zoogeoiiraphij. — Opsopoeodus emiliae probably inhabited the
preglacial Teays-Mississippi system. Redispersal into glaciated re-
gions undoubtedly took place solely from a refugium in the lower
Mississippi Valley.

Phoxinus erythrogaster (Rafinesque) — southern redbelly dace

Chrosomus enjihwgaster: Agassiz, 1854:359 (Osage R.). Call, 1887:75 (West
Fork Black R. and Toms Cr., Reynolds Co.; Sinking Cr., Spring Valley
Cr., and Jacks Fork, Shannon Co.). Meek, 1891:117,121,124 ( Meramec
Spring and Big Dr>- Fork near St. James; Osage Fork SE of Marshfield;
Big Piney R. near Cabool; Jones Cr. and Maries R. near Dixon; Niangua R.
near Marshfield; Sac R. near Springfield; Bryants Cr. near Mansfield).
Evermann and Kendall, 1895:470 (spring branch at Neosho). Fowler,
1921:398 (Fox Cr., trib. Meramec R.). Hubbs and Ortenburger, 1929:89
(t^vo mi. S DeSoto). Martin and Campbell, 19.53:46 (abund.; Black R.).

Distribution and habitat. — The southern redbelly dace is vir-
tually restricted to the Ozarks, occurring elsewhere only in a few
spring-fed creeks in Lincoln County (Map 34). Populations of the
redbelly dace are found mainly in small streams where springs
maintain permanent flow. Along the larger creeks and rixers of the
Ozarks it occurs only as strays, or as highly localized populations
in spring pools away from the main channel.

Zoogeography. — Isolated populations of P. erythrogaster in the
Flint Hills of Kansas (Metcalf, 1966:102-103; Cross, 1967:82) may
date from Pleistocene time, when cooler and moister conditions
permitted westward dispersal from the Ozark Uplands. Disjunct
populations in the upper Canadian River of northeastern New
Mexico (Koster, 1957:59) may likewise be Pleistocene relicts. Per-
haps P. erythrogaster has occupied the Ozark Uplands continuously
since late-Tertiary times.

Nocomis biguttatus (Kirtland) — hornyhead chub

Hybopsis biguttatus: Jordan and Meek, 1885:12,16 (Hundred and Two R. at
Maryville; Flat Cr. near Sedalia and/or Blackwater R. at Brownsville).
Martin and Campbell, 1953:46 (abund.; Black R.). Patriarche and Camp-
bell, 19.58:25.5-2.56 (alnmd.; Clearwater Res.).

Hijbopsis kenttickiensis: Meek, 1891:118,122,125,126,130 (Meramec R. and
Big Dry Fork near St. James; Little Dry Fork near Rolla; "abundant in
all streams examined," Gasconade and Osage systems; "common in all
streams examined," Neosho system; James R. near Springfield; North Fork

332 University of Kansas Publs., Mus. Nat. Hist.

White R. S Cabool). Evermann and Kendall, 1895:471 (Indian Cr. near
Neosho ) .

Disirihution and Jiabitat. — The hornyhead chub occurs through-
out the Ozarks, where it is one of the common and characteristic
stream fishes (Map 35). However, it never attains the abundance
of some of the other common minnows of the region. Scattered
early records document a former rather widespread distribution
for N. biguttattis in the Till Plains. These populations are now
largely or entirely extirpated. Listing of the hornyhead chub from
the Hundred and Two River in northwest Missouri by Jordan and
Meek (1885:12) is far removed from any recent records, and this
species has not been collected in northeastern Missouri since the
early 1940's. Even in the Ozarks N. higiittatus seems to be less
abundant and widespread. There are no recent records for the
Bourbeuse system, for example, although that area has been quite
thoroughly sampled. Similar but more dramatic changes in the
distribution of the hornyhead chub have been noted in Ohio by
Trautman (1957:294) and in Kansas by Metcalf (1966:105-106)
and Cross (1967:87). Increased siltation and intermittent flow
resulting from destruction of vegetative cover and intensive culti-
vation may be major factors in elimination of this chub from parts
of Missouri.

The hornyhead chub inhabits clear streams with permanent flow
and a predominance of clean gravel or rubble bottoms. Adults are
most often found near riffles, but not in the swifter currents. The
young frequently occur where there is no current, among growths
of algae or higher aquatic plants.

Zoogeography. — The range of N. biguttattis is principally north
and west of other Nocomis. Perhaps the ancestral stock of N.
biguttatus occupied most of the preglacial upper Mississippi or
Laurentian systems, with ancestral stocks of other Nocomis occu-
pying the preglacial Ohio and Teays stream systems. Occurrence
of the closely related N. effiisus Lachner and Jenkins in the Cum-
berland, Greene, and lower Tennessee drainages indicate that N.
biguttatus was not present in the preglacial Ohio River system.
That N. biguttatus may have long occupied the Ozark Uplands is
indicated by the occurrence there of an undescribed form ( Lachner
and Jenkins, 1967:558). Disjunct populations of N. biguttatus in
Colorado and Wyoming may be glacial relicts.

Hybopsis storeriana (Kirtland) — silver chub

Htjbopsis stoierianns: Jordan and Meek, 1885:17 (Grand R. at Clinton and/or
Tebo Cr. at Calhoun). Forbes and Richardson, 1920: map L (Missouri R.
local.).

Missouri Fishes 333

Hybopsis storeiiaiui: Fisher, 1962:127 (Missouri R. local.). Cross, 1967:92
(Mis.sonri H. local, mapped).

Distribution and habitat. — The silver chub is one of the common
minnows in the Missouri and Mississippi rivers (Map 36). It also
occurs in other large streams of the state but is seldom abundant.
The paucity of recent records from the Grand and upper Osage
rivers is probably due to inadequate sampling rather than to a
recent restriction in distribution. Uijbopsis storeriana is an inhabi-
tant of large streams and occurs most abundantly in quiet pools and
backwaters.

Zoogeography. — The present distribution of H. storeriana sug-
gests that it was an inhabitant of the preglacial Teays-Mississippi
system.

Hybopsis amblops (Rafinesque) — bigeye chub

Hyhopsis amblops: Meek, 1891 (Big Dry Fork near St. James; Shoal Cr. and
Hickory Cr. near Neosho; Janies R. near Springfield; North Fork White R.
S Cabool). Evermann and Kendall, 1895:471 (Indian Cr. near Neosho).
Hubhs and Ortenburger, 1929:66 (Sarcoxie). Martin and Campbell, 1953
(abund.; Black R.). Patriarche and Campbell, 1958:255-256 (abimd.;
Clearwater Res.).

Distribution and habitat. — The bigeye chub is confined to the
Ozarks, where it occurs in all the principal stream systems except
the Osage and Gasconade (Map 37). It is still common in the
White, Current, and Black river systems, but seems to have de-
clined in abundance in the Meramec and Neosho systems since
the early 1940's. This suggests a pattern of extirpation like that
reported by Trautman (1957:301-303) in Ohio. The bigeye chub
is an inhabitant of clear streams having permanent flow and silt-
free gravel or rubble bottoms. It often occurs near riffles, but not
in the main current. Rather, it is found at the foot of the riffle,
where the current slackens, or in quiet pools with no current.

Zoogeography. — Hybopsis amblops was probably an inhabitant
of the preglacial Teays-Mississippi system. Its present distribution
suggests that it may be a recent invader of the Ozark Uplands. The
presence of closely related or conspecific forms in eastern Gulf
coastal drainages suggests that this species and its ancestral stocks
have long occupied uplands east of the Mississippi River.

Hybopsis dissimilis (Kirtland) — streamline chub

Hybopsis dissimilis: Martin and Campbell, 1953 (abund.; hab.; Black R. ).

Hybopsis dissimilis Jwrnji: Hublis and Crowe, 1956:2-4,6 (orig. descr.; type
local. White R., 3 mi. SE of Mano, Barry Co.).

Erimystax dissimilis: Patriarche and Campbell, 1958:255-256 (abund.; Clear-
water Res.).

Distribution and habitat. — The streamline chub is confined to

334 UNrs'ERSiTY OF Kansas Publs., Mus. Nat. Hist

the southern slope of the Ozark Uplands (Map 38). It is the most
abundant chub in many of the larger streams of that region. This
minnow inhabits moderately large, clear streams with continuous
strong flow and clean gravelly or rocky bottoms. It is most often
found just below riffles or in pools with a noticeable current. The
distribution of this species complements that of the related gravel
chub in Missouri; the two have been taken together only at a single
locality on lower Current River. Trautman (1957:309) indicated
that the two are competitive and tend to occupy different habitats
where they occur together in Ohio. Possibly competition is respon-
sible for the complementary distribution of the two species in
Missouri.

Zoogeo^rapJiy. — Populations of the streamline chub in the
Ozarks are broadly disjunct from those east of the Mississippi Em-
bayment. Possibly disjunction resulted from fragmentation of a
continuous preglacial distribution, but could also indicate a Pleisto-
cene invasion of the Ozarks. The streamline chub may have had
its origin south and east of the area occupied by ancestral stocks
of the gravel chub. Perhaps stocks of H. dbsimilis occupied the
preglacial Teays and/ or Ohio systems, at a time when H. x-punc-
tata occupied the northern and western part of the Mississippi
system, and stocks of the closely related H. imignis occupied the
Tennessee system.

Hybopsis x-punctata Hubbs and Crowe — gravel chub

Hijhopsis dissimilis: Meek, 1891:122 (Gasconade R. at Arlington; Little Piney
Cr. at Newburg and/or Arlington ) .

Hybopsis x-punctata x-punctata: Hnbbs and Crowe, 1956:2-4,7 (orig. descr.;
type local. Gasconade R. at Starks Ford, 8 mi. S Richland, Pulaski Go.).

DistriJmtion and habitat. — The gravel chub is rather common
and generally distributed in the northern and western Ozarks ( Map
39). It is otherwise known only from the Salt River system of north-
eastern Missouri, and from single localities on the lower Mississippi
and lower Current rivers. Hubbs and Crowe (1956:7) recognized
two subspecies of the gra\'el chub, and described the nominate
.subspecies from Missouri. They erronously listed the type locality
as "Starks Fork of Gasconade River." Examination of field data
sheets indicates that this should read "Gasconade River at Starks
Ford."

The gravel chub inhabits clear to moderately turbid streams
with permanent flow and well defined gravel riffles. In the Ozarks
it tends to be most abundant in the downstream sections of the
larger streams, where the gradient is less, and the water is warmer

Missouri Fishes 335

and less clear than in the headwaters. This minnow is most often
found in slight to moderate current, over a silt-free gravel or rubble
bottom.

Zoogeograpluj. — The range of the gravel chub is primarily
north and west of that of other members of the subgenus Erimysiax,
suggesting an origin in the pre-glacial lower Missouri, Iowa, or
upper Mississippi systems. This species undoubtedly survived the
Wisconsin ice advance in the Ozark Uplands, and perhaps also
in the Driftless Area of Wisconsin.

Hybopsis aestivalis (Girard) — speckled chub

Flyhopsis aestivalis: Fisher, 1962:427 (Missouri R. local.). Cleary, 1956: map
35 (Des Moines R. local.). Cross, 1967:96 (Missouri R. local, mapped).

Distribution and habitat. — The speckled chub is most abundant
in the Missouri and Mississippi rivers, where it is one of the char-
acteristic minnows (Map 40). It is also common in the main
ditches of the lowlands and some of the larger prairie streams of
the till plains. It seems to have only recently invaded the lowland
ditches, since it did not occur in collections made there in the
early 1940's.

Hij])opsis aestivalis inhabits the main channels of large, low
gradient streams. It is found over a sand or fine gravel bottom,
often in a moderate to strong current.

Zoogeography. — Variation and zoogeography of the speckled
chub were discussed at length by Metcalf (1966:108-110). The
widespread distribution and strong regional differentiation of the
speckled chub in the Mississippi River system and other Gulf coastal
drainages indicate that it has long resided there. Several well
marked forms are recognized, but until variation in this species
has been thoroughly studied, it would be unwise to assign Missouri
populations to one of these.

Hybopsis gracilis (Richardson) — flathead chub

Flatijgobio gracilis: Jordan and Meek, 1885:13 (Missouri R. at St. Joseph).

Berner, 1951:9 (abund.; Missouri R.). Shoemaker, Pickering and Durham,

1951 (Mississippi R. at Gates, Lake Co., Tenn.).
Hybopsis gracilis gracilis: Olund and Cross, 1961 (char.; syn.; Missouri local.

comp. and mapped). Cross, 1967:84 (Missouri R. local, mapped).
Hybopsis gracilis: Fisher, 1962:427 (Missouri R. local.).

Taxonomic considerations. — Olund and Cross (1961) found sta-
tistically significant differences in several meristic and proportional
characters among populations of H. gracilis. They recognized a
northern and eastern subspecies (H. g. gracilis) which characteristi-
cally inhabits large rivers, and a southern and western subspecies

336 University of Kansas Publs., Mus. Nat. Hist.

(H. g. guIoneUus) that characteristically inhabits small streams.
Bailey and Allum (1962:44-45) suggested that the differences may
be phenotypic, resulting from differing temperatures during devel-
opment. They hypothesized that the big river form undergoes
development at lower temperatures than the creek form, because
the larger streams are fed by cooler water from mountain tribu-
taries. This argument loses force when it is applied to the popula-
tions of H. g. gracilis in the lower Missouri and Mississippi rivers,
which are not under the influence of cool mountain streams. I
have recorded a temperature of 80 °F in the lower Missouri River
as early as mid-June. I therefore concur with Metcalf (1966:103-
104) that the differences between the two forms have a genetic
basis.

Distrihuiion and habitat. — The flathead chub is one of the most
abundant minnows in the Missouri and lower Mississippi rivers
( Map 41 ) . It does not ascend the Mississippi River above the
mouth of the Missouri, and enters tributary streams only in the
extreme northwestern part of the state. Possibly competition is an
important factor limiting the distribution of the flathead chub,
and its occurrence in the faunally depauperate streams of northwest
Missouri reflects a release from competition. Olund and Cross
(1961:341) noted the absence of this species from creeks in eastern
Kansas and Missouri that appeared to have suitable habitat and
suggested that the flathead chub was excluded from these creeks
by competition from the creek chub. I subsequently found both
species in Mill Creek and Rock Creek of northwestern Missouri,
where they were taken in the same seine hauls.

This minnow is most abundant in the main channels of large
silty rivers, where it lives in a strong current over a firm, sandy
bottom. Where it enters tributary streams in Missouri it inhabits
the pools of small creeks with little current and bottoms composed
mostly of gravel or bedrock. These small creeks are turbid after
heavy rains, but are clear during periods of little rainfall.

Variation and zoogeograpliy. — To determine whether or not
the flathead chubs from tributary streams of northwestern Missouri
are the creek subspecies (//. g. gidoneUiis), counts of lateral line
scales and pectoral rays were made on a sample from Mill Creek,
Atchison County, and another from the Missouri River, Platte
County (Table 4). Counts of lateral line scales did not differ in
the two samples, but they were significantly lower than counts
reported by Olund and Cross (1961: fig. 1) for H. g. gracilis from
the Missouri River below the Platte River, Nebraska. Counts of

Missouri Fishes 337

Table 4. — Frequency
Fin-ray Coiuits i

Distribution of Lateral Line Scale Counts and Pectoral
n Two Samples of Hybopsis gracilis from Missouri.

Locality

44 45

46 47

Lateral line scales
48 49 50 51 52 53 X S.D.

Mill Cr., Atchison Co.
Missouii R., Platte Co

-.-.- 30
. -.. 29

1 1

6 5
2 4

7 3 4 4 1 48.3 1.79
9 4 6 1 1 48.3 1.83

Locality

Pectoral fin rays
17 18 19 20 X S.D.

Mill Cr.. Atchison Co.

.... 30

12 1 16.5 0 56

Missouri R., Platte Co.

-___ 29

17 7 1 17.2 0.82

pectoral rays differed slightly in the two samples from Missouri;
the mean ray-count of Missouri 1-liver specimens was close to that
reported by Olund and Cross (1961:fig. 1) for H. g. gracilis from
the Missouri River below the Platte River, Nebraska, and the mean
count for Mill Creek specimens was close to that reported by Olund
and Cross for intergrades from the Republican River in Nebraska.
The small size of the Mill Creek .specimens (42.8-58.4 mm. standard
length, mean 50.8 mm.) precluded a comparison of proportional
measurements but the pectoral fins in the Mill Creek specimens are
less falcate than those of specimens of comparable size from the
Missouri River. Thus, specimens from Mill Creek are morpho-
logically intermediate between the two subspecies. This is con-
sistent with the report by Olund and Cross (1961: plate 21) of in-
tergrades from other nearby tributary streams entering the Missouri
River from Nebraska and Kansas. The low number of lateral
line scales in specimens from the Missouri River in Missouri are
suggestive of intergradation, but the ray count and shape of the
pectoral fin indicate that the Missouri River population is closer
to H. g. gracilis than is the Mill Creek population.

Metcalf (1966:104) suggested that H. g. gracilis had its origin
in the ancestral north-flowing upper Missouri River, whereas H. g.
gidonellus had its origin in the preglacial drainage of the central
and southern plains. If this interpretation is correct, mixing of the
two stocks and penetration of H. g. gracilis into Missouri resulted
from drainage diversions during the Pleistocene. Partial segrega-
tion of the two forms evidently occurs in northwest Missouri, where
the species inhabits creeks as well as the mainstream of the Missouri
River.

Hybopsis gelida (Girard) — sturgeon chub

Hijhopsis gelida: Fisher, 1962:427 (Missouri R. local.). Bailey and Allum,
1962:46 and fig. 4 (Missouri local, compiled and mapped). Cross, 1967:97
( Mi.ssouri R. local, mapped).

338 University of Kansas Publs., Mus. Nat. Hist.

Distribution and habitat. — The sturgeon chub occurs only in the
Missouri and lower Mississippi rivers (Map 42). Like several other
fishes of the upper Missouri system, the sturgeon chub ascends
neither the Ohio Ri\'er nor the Mississippi River above the mouth
of the Missouri. This chub inhabits the main channels of large
silty rivers and occurs in swift current over a bottom of sand or
fine gravel. According to Bailey and Allum (1962:46), the sturgeon
chub is most often found over gravel, and this may account for its
rarity in the lower Missouri Rixer, where sand is the principal
bottom type.

Zoogeography. — Metcalf (1966:11) suggested that H. gelida
evolved in the preglacial Hudson Bay system, perhaps from the
same stock that gave rise to H. aestivalis in the preglacial Missis-
sippi River system. If this is correct, the presence of H. gelida in
the Mississippi Valley dates from diversion of the upper Missouri
into its present course during the Pleistocene.

Hybopsis meeki Jordan and Evermann — sicklefin chub

Hybopsis gelidus: (misident. ) Jordan and Meek, 1885:13 (Missouri R. at

St. Joseph).
Hybopsis meeki: Jordan and Evermann, 1896:317 (orig. descr.; type local.

Missouri R. at St. Joseph). Fisher, 1962:427 (Missouri R. local.). Bailey

and Allum, 1962:48 and fig. 5 (Missouri local, compiled and mapped).

Cross, 1967:95 (Missouri R. local, mapped).

Distribution and habitat. — The distribution of the sicklefin chub
in Missouri is similar to that of the sturgeon chub, but the sicklefin
chub is decidedly more abundant (Map 43). The sicklefin chub
seems to increase in abundance in the Missouri Rixer towards its
mouth. Its habitat in Missouri is not notably different than that of
the sturgeon chub.

Zoogeography.— MeicaM (1966:107-108) suggested that H.
meeki originated in the preglacial upper Missouri River system,
possibly from the same stock that gave rise to //. storeriana in the
Mississippi River system.

Phenacobius mirabilis (Girard) — suckermouth minnow

Sarcidium scopiferum: Cope, 1871:440-441 (orig. descr.; type local. Missouri
R. near St. Joseph ) .

Plienacobius mirabilis: Jordan and Meek, 1885:12,14,16,17 (Hundred and
Two R. at Maryville; Tabo Cr. 6 mi. E Lexington; Flat Cr. near Sedalia
and/or Blackwater R. at Bro\vns\ille; Grand R. at Clinton and/or Tebo Cr.
at Calhoun). Fisher, 1962:427 (Missouri R. local.). Hanson and Camp-
bell, 1963 (linear distr.; Perche Cr.). Cross, 1967:100 (Missouri R. local,
mapped ) .

Phenacobius mirabilis scopifer: Call, 1887:80 (Bear Cr. at ColumlMa).

Distribution and habitat. — The suckermouth minnow is abun-

Missouri Fishes 339

dant and w idcspicad in the prairie region and oeeurs oecasionally
in lowland ditches of the southeast (Map 44). It is rather com-
mon in the warmer and more turbid streams of the Ozark border,
but a\oids the clear streams of the central Ozarks. This minnow
is found in streams of all sizes, but avoids those with intermittent
flow or cold, clear water. It is tolerant of turbidity and siltation,
as long as there is enough current to keep the riffle areas free of
silt. The suckermouth minnow is strictly a riffle fish, and reaches
its greatest abundance on the sluggish riffles of warm, moderately
turbid streams with low gradient.

Zoogeograpliy. — Metcalf (1966:111) suggested that P. mirabilis
originated in the central and southern plains and gained access to
the central Mississippi Valley when part of the plains drainage was
diverted eastward. Eastward dispersal of this species into Ohio
within historic time was documented by Trautman (1957:323-324).
This range expansion is presumably correlated with increased sil-
tation and turbidity of the streams as a result of the removal of
vegetative cover and intensive cultivation.

Notropis atherinoides Rafinesque — emerald shiner

AlhurneUtis percobromus: Cope, 1871:440 (orig. descr.; type local. St. Joseph,
Mo.).

Notropis dUectus: Jordan and Meek, 1885:16 (Grand R. and/or Tebo Cr. at
Clinton).

Notropis atherinoides: Carman, 1890:142 (Mississippi R. near Quincy, 111.).
Forbes and Richardson, 1920:map XLIII (Mississippi R. local.). Patri-
arche, 1953:247 (Lake Wappapello). Cleary, 1956:map 41 (Des Moines
R. local.). Fisher, 1962:427 (Missouri R. local). Cross, 1967:102 (Mis-
souri R. local, mapped).

Notropis ruhrifrons: Fowler, 1910:290, plate 21, fig. 50 (in part.; char.; type
of Alburnellus percobromus figured).

Notropis percobromus: Hubbs, 1945:16-17 (not the southern representative
of N. rubellus; compar. with N. atherinoides atherinoides). Cross and
Minckley, 1958:104 (char.; Missouri R., Atchison Co., Kansas).

Distribution and habitat. — The emerald shiner is the most abun-
dant fish in the Missouri and Mississippi rivers and is common in
large streams elsewhere in the state (Map 45). It penetrates the
Ozarks only along a few of the largest streams. In Missouri, the
emerald shiner lives primarily in the open waters of large, perma-
ment-flowing streams with moderate or low gradients. It is tolerant
of a wide range of turbidities, bottom types, and current velocities.
Elsewhere in its range the emerald shiner is often abundant in
large lakes and reservoirs, but it is not especially abundant in such
habitats in Missouri.

Taxonomic considerations. — Since its original description by
Cope (1871:440) from specimens collected at St. Joseph, Missouri,

340 University of Kansas Publs., Mus. Nat. Hist.

AlburneUus percobwmus has had a compHcated nomenclatural
history. At various times it has been treated as a distinct species
(e.g., Jordan and Gilbert, 1883:202; Hubbs, 1945:16-17; Hubbs and
Bonham, 1951:93), as a synonym of N. rubifrons {=N. riibeUus)
{e.g., Jordan and Evermann, 1896:295; Fowler, 1910:290), and as
the southwestern representative of N. nibellus (Hubbs and Orten-
burger, 1929:83-85). In the most recent treatment of this form,
Bailey and Allum (1962:56-60) placed it in the synonymy of N.
atlierinoides. They indicated that the supposed differences between
N. percobromus and N. atherinoides vary locally and are not geo-
graphically consistent, and suggested that such differences as do
exist represent phenotypic responses to varying environmental
conditions. I follow Bailey and Allum in considering the two forms
to be conspecific, but I doubt that there is no genetic basis for the
differences.

Variation and zoogeographij. — The emerald shiner exhibits con-
siderable variability in body proportions both within collections
from one locality and between localities. In Missouri there seems
to be a character gradient from northwest to southeast; populations
in the Missouri Ri\'er and its tributaries of northwestern Missouri
are more chubby than those from the lower Mississippi River and
the Southeastern Lowlands.

Notropis atherinoides probably occurred in the preglacial Teays-
Mississippi system, as well as in other ancient drainages. If there
is a genetic basis for the percobromus phenotype, this form may
have occupied the preglacial drainage of the central and southern
plains (Metcalf, 1966:113-114).

Notropis rubellus (Agassiz) — rosyface shiner

Notropis ruhrifwns: Meek, 1891:118,122,125,126,129 ( Meramec R. and Big
Dry Fork near St. James; Little Dry Fork near Rolla; Osage Fork SE
Marshfield; Lick Fork and Brjants Cr. near Mansfield; Little Piney Cr. at
Newburg and Arlington; Sac R. and James R. near Springfield; Shoal Cr.
and Hickory Cr. near Neosho). Jordan and Meek, 1885:15 (Flat Cr. near
Sedalia and/or Blackwater R. at Brownsville). Evermann and Kendall,
1895:470 (Indian Cr. near Neosho). Fowler, 1910:290 (in part.; char.;
Carthage, Mo. ) .

Notropis dilecfus: Fowler, 1910:290 (Creenfield, Mo).

Notroj)is percobromus: Hubbs and Ortenbmger, 1929:83 (Sac R.).

Notropis rubellus: Martin and Campbell, 1953 (abimd.; hab.; Black R.). Met-
calf, 1966: table 10 (char.; Richland Cr., Morgan Co.).

Distribution and haJyitat. — The rosyface shiner is confined to the
Ozarks, where it is one of the most abundant and generally distrib-
uted minnows (Map 46). The rosyface shiner is intolerant of con-
tinuous high turbidity or siltation, and is most abundant in streams

Missouri Fishes 341

of iiiedium size having well defined gravel or rnbble riffles. It is
generally found in a moderate to swift current, either just below
riffles or in rocky pools where riffles and pools alternate in rapid
succession.

Zoogeography. — Ozark populations of N. rubelhis are disjunct
from the main body of the range, which is north and east of Mis-
souri. The distributional hiatus between Missouri populations and
those in Illinois and Iowa is at least 100 miles wide at its narrowest
point. That this species has long resided in the Ozark Uplands is
suggested by the presence of marked regional differentiation in
body form, number of fin rays and pigmentation. Possibly N.
ruheUus invaded the Ozark Uplands during the Pleistocene Epoch,
but more likely it has occurred there continously since preglacial
time.

Notropis telescopus (Cope) — telescope shiner

(?) Notropis micropteryx: Call, 1887:76 (Spring Valley Cr., Shannon Co.).
Meek, 1891:129 (James R.).

Notropis arionimus: Martin and Canipl^ell, 1953 (abund.; halx; Black R.).
Patriarche and Campbell, 1958:255-256 (abund.; Clearwater Res.).

Notropis telescopus: Gilbert, 1969:474-492, fig. 4 (syn.; char.; compar.; Mis-
souri local, mapped ) .

Distrihtition and habitat. — The telescope shiner is one of the
must abundant minnows in streams draining the southern slope of
the Ozark Uplands (Map 47). It is most often found in medium-
sized to moderately large streams, occurring only rarely in head-
water creeks. The telescope shiner is usually found near riffles,
where there is a moderately swift current, and a bottom composed
of gravel, rubble, or boulders.

Gilbert (1969:485) has called attention to two specimens
(UMMZ 187280) of a related species, Notropis ariommus Cope,
that were presumably collected by S. E. Meek in 1889 from Big
and Little Dry Forks, Phelps County, Missouri. Since Meek (1891)
did not mention N. ariommus, and this species has not been other-
wise collected in Missouri or elsewhere west of the Mississippi
River, Gilbert doubted the validity of the record.

Variation and Zoogeography. — N. telescopus from different
stream systems of the Ozarks exhibit differences in counts of anal
fin rax's. Specimens from the St. Francis River and Headwater
Diversion have 9 and 10 anal rays in about equal numbers; only
a few specimens have 11. Those from the Black and White river
systems have modally 10 anal rays and far more specimens have
11 rays than 9 rays.

Notropis telescopus is one of several species that occur dis-

342 Unr'ersity of Kansas Publs., Mus. Nat. Hist,

junctly in the two main uplands of the southcentral United States
east and west of the Mississippi Embayment. Possibly N. telescopiis
evolved in the Ozark Uplands from the common ancestral stock that
gave rise to N. ariommus east of the Mississippi Embayment. At
some time during the Pleistocene, N. telescopiis may have mo\'ed
across the embayment, replacing N. ariommus in part of its pre-
glacial range.

Notropis umbratilis (Girard) — redfin shiner

Notropis umbratilis: Jordan and Meek, 1885:11,15,17 (Hundred and Two R.
at Maryville; Flat Cr. near Sedalia and/or Blackwater R. at Brownsville;
Grand R. at Clinton and/or Tebo Cr. at Calhoun). Call, 1887:76 (Spring
Valley Cr.. Shannon Co.). Meek, 1891:125 (Maries R. near Dixon; Sac
R. near Springfield). Fowler, 1910:291-292 ( Marshfield and Sedalia).
Martin and Campbell, 1953 (abund.; hab.; Black R.). Fisher, 1962:427
(Missouri R. tribs.). Hanson and Campbell, 1963 (linear distr.; Perche
Cr.).

Notropis umbratilis cyanocephalis: Meek, 1891:118,121 (Big Dry Fork near
St. James; Little Dry Fork near RoUa; Osage Fork SE Marshfield; Lick
Fork at Mansfield ) .

Distribution and habitat. — The redfin shiner is nearly statewide
in distribution, but it is absent from the White River system of the
southern Ozarks, and is scarce in the northwestern till plains ( Map
48). It is most abundant in streams of the northern and western
Ozark border. Notropis umbratiJis is common in ditches of the
lowlands and adjacent streams of the southeastern Ozarks.

The redfin shiner occurs in a variety of habitats. In the prairie
region it occurs most abundant!}' in rocky or gra\'elly creeks with
high gradient and low or intermittent flow. In the lowlands IV.
umbratiJis occurs in quiet ditches with an abundance of submerged
aquatic vegetation. Along the cool, spring-fed streams of the
Ozarks it is almost invariably found in protected backwaters and
overflow pools that are several degrees warmer than the main
stream. The common denominators of all these habitats are rela-
tively clear, warm water and the absence of strong current.

Variaton and zoogeographij. — Notropis umbratilis, as currently
recognized, is an extremely variable species, and probably consists
of a complex of recognizable subspecies, or perhaps of two or more
species. Some workers recognize two subspecies, referring popu-
lations from the Ohio, Great Lakes and Mississippi systems to N. u.
cijanocephahis (Copeland), and referring populations from the
lower Mississippi and western Gulf coastal drainages to N . u. um-
bratilis ( Girard ) .

At least two well marked forms occur in Missouri. Populations
in the Missouri River system have modally 10 anal rays and ha\e

Missouri Fishes 343

the dark spot at the anterior base of the dorsal fin weakly devel-
oped. Breeding males of this form have tubercles over the opercu-
lum and predominantly black fins. An isolated population of this
form occurs in the Salt River drainage of northeastern Missouri,
and a similar form but with modally 11 anal rays occurs in the
Neosho Rixer system of southwestern Missouri. In all direct tribu-
taries of the Mississippi Ri\er, except for Salt River, in the South-
eastern Lowlands, and in streams draining the southeastern slope
of the Ozark Uplands, N. umhraiilis is represented by a form having
modalK' 11 anal rays and a prominent black spot at the anterior
base of the dorsal fin. Breeding males in this form lack tubercles
on the operculum and have predominantly red fins. All Missouri
populations are assignable to one form or the other, but those from
Salt River and adjacent tributaries of the upper Mississippi River
show evidence of intergradation or introgression. The distribu-
tional relationship of the two suggests an origin for the red-finned
form in the preglacial Teays-Mississippi system and an origin for
the black-finned form in the preglacial drainage of the central and
southern plains. Diff^erences between populations of the black-
finned form in the Missouri and Arkansas systems suggest a rather
long separation.

Notropis fumeus Evermann — ribbon shiner

Distribution and habitat. — The ribbon shiner is common in the
lowlands (Map 49), where it is confined primarily to open water
of the larger, swifter, and more sparsely vegetated ditches. It gen-
erally occurs over a sandy bottom in a slight to moderate current.

Zoop,eofi,raphy. — The ribbon shiner is southern in its affinities,
reaching the northern limit of its range in southeastern Missouri and
southern Illinois. It may have occupied the lower Mississippi Valley
continuously since preglacial times, or entered the Mississippi Valley
from western Gulf coastal drainages at a more recent date.

Notropis shumardi (Girard) — silverband shiner

Notropis illecehrosus: Cross and Minckley, 1958:104-105 (descr.; ecol.; Mis-
souri R., Atchison Co., Kans.). Fisher, 1962:427 (Missouri R. at St.
Charles). Bailey and Allum, 1962:61 (Missouri local, compiled).

Notropis shumardi: Gilbert and Bailey, 1962 (char.; syn.; Missoini local, com-
piled and mapped). Cross, 1967:107 (Missouri R. local, mapped).

Distri])ution and liabitat. — The silverband shiner is a charac-
teristic element in the fauna of the Missouri and Mississippi rivers,
penetrating only rarely into the lower sections of tributaries (Map
50). In the Mississippi it has never been collected above Pike

344 University of Kansas Publs., Mus. Nat. Hist.

County. The silverband shiner is fairly common in the lower
Mississippi River, but it is rare elsewhere in its Missouri range.
Notropls shumardi inhabits the open channels of large rivers, where
there is strong current and a firm sand or gravel bottom.

Zoogeography. — Probably N. shumardi had its origin in the
Mississippi Valley (Gilbert and Bailey, 1962:817). These authors
suggested that N. shumardi gained access to the Alabama River
through stream connections created by lowering of sea levels dur-
ing the Pleistocene, and may have reached western Gulf coastal
drainages in the same manner or as a result of stream transfer
between the upper Red and Brazos rivers.

Notropis zonatus (Agassiz) — bleeding shiner

Alburnus zonatus: Agassiz in Putnam, 1863:9 (orig. descr.; type local. Osage
River ) .

Notropis zonatus: Call, 1887:76 (West Fork Black R., Reynolds Co.; Jacks
Fork, Shannon Co.; Piney R., Texas Co.; Meramec R., Dent Co.). Meek,
1891:118,121,125 (Meramec R. and Meramec Spring near St. James;
Big Piney R. at Cabool; Little Piney Cr. at Newburg and Arlington; Gas-
conade R. at Arlington; Lick Fork at Mansfield; Osage Fork SE Marsh-
field; Jones Cr. and Maries R. near Dixon; Niangiia R. near Marshfield;
Sac R. near Springfield). Fowler, 1921:399 (Fox Cr., trib. Meramec R.).
Hubbs and Ortenburger, 1929:81 (Big R. trib. 4 mi. S Potosi). Patriarche
and Campbell, 1958:255-256 (Clearwater Res.). Gilbert, 1964:95-121,
129-133, map 2 (char.; compar.; syn.; Missouri local, mapped).

Notropis zonatus zonatus: Hublis and Moore, 1940 (char.; compar.; syn.; Mis-
souri local, mapped). Martin and Campbell, 1953 (abund.; hab.; Black
R.).

Distribution and habitat. — The bleeding shiner is endemic to
the Ozark Uplands of Missouri and northeastern Arkansas (Map
51). In Missouri it occurs in all major drainages not occupied by
the closely related duskystripe shiner. Within its area of occurrence
N. zonatus is one of the most abundant minnows. The bleeding
shiner inhabits clear, small to medium-sized streams with continuous
strong flow and clean gravel or rubble bottoms. It avoids sections
of the Ozarks that are underlain by shale bedrock, probably because
of increased turbidity and intermittency of the streams. Notropis
zonatus is most often found in a moderate to swift current near
riffles, or in small pools where riffles and pools alternate in rapid
succession. The young tend to occupy quieter water than the adults.

Zoogeography. — Gilbert (1964:104-105) suggested that the
common zonatus-pih])ryi stock became isolated in the Ozark Up-
lands during the Pliocene. He further indicated that the present
distribution of the two species suggests a long separation between
the Missouri and White river systems and that the presence of
N. zonatus in the Black River system indicates that this stream

Missouri Fishes 345

was formerly a tribiitar)' of the Missouri. According to Gilbert,
the Black Ri\'er underwent a reversal of flow at a recent date,
after N. zon(ltt^s and N. pilshn/i had difi^erentiated. I doubt this
explanation. There is no geological evidence to indicate that the
Black Ri\er was ever a tributary of the Missouri River, and the
distribution of N. zonatus can be accounted for without such a
Inpothesis. This species has a continuous distribution from the
Missouri to the Black by way of the St. Francis and Little rivers
and direct tributaries of the Mississippi Ri\'er, and there is no
reason to believe that this has not long been the case. All of the
major streams of the southeastern Ozarks were direct tributaries
of the Mississippi River at the time the latter stream flowed west
of Crowley's Ridge; thus, there was ample opportunity for waif
dispersal westward to the Black River through the Mississippi
Ri\er. Before drainage of the northern and central plains was
diverted to the lower Missouri River, it and the Mississippi River
may have been less of a barrier to dispersal by upland species
such as N. zonatus than now.

Notropis pilsbryi Fowler — duskystripe shiner

Notropis zonatus: Meek, 1891:126-129 (Shoal Cr. and Hickory Cr. near
Neosho; James R. near Springfield; Bryants Cr. near Mansfield; North
Fork White R. S Cabool). Evermann and Kendall, 1895:470 (Indian Cr.
S Neosho; spring branch at Neosho). Fowler, 1910:285-286 (Marshfield?;
Carthage; and James R.). Hubbs and Ortenburger, 1929:81 ( Sarcoxie).

Notropis zonatus pilshnji: Hubbs and Moore, 1940 (char.; compar.; syn.;
Missouri local, mapped).

Notropis pihhnji: Gilbert, 1964:95-121,13.3-136, map 2 (char.; compar.; syn.;
Missouri local, mapped).

Distribution and habitat. — The duskystripe shiner replaces the
closely related bleeding shiner in the Neosho and upper White
river systems of the southwestern Ozarks (Map 52). In that re-
gion it is one of the most abundant fishes in small and medium-sized
streams. Its habitat is like that of the bleeding shiner.

Zoogeography. — N. pilsbryi probably originated in clear upland
streams of the Flint Hills and western slope of the Ozark-Ouachita
uplands. It likely attained its present distribution in the White
River system by stream capture between the upper White and
Neosho systems (Hubbs and Moore, 1940:94). The argument for
a western origin of this species is strengthened by the occurrence
of disjunct populations in the Red River system, which presumably
had connections with the preglacial upper Arkansas River. Hubbs
and Moore (1940:94) suggested that presence of the duskystripe
shiner in the Red River system was perhaps due to "an accidental

346 Unwersity of Kansas Publs., Mus. Nat. Hist.

transfer, either by data or of the Hving fish" because of the isolated
position of the records. However, occurrence at three locaHties
suggests natural distribution (Gilbert, 1964:135).

Notropis cornutus (Mitchell) — common shiner

Notropis megalops: Jordan and Meek, 1885:12,15 (Hundred and Two R. at
Maryville; Flat Cr. near Sedalia and/or Blackwater R. at Brownsville).

Notropis cornutus: Gilbert, 1961 (char.; compar.; Missouri local, mapped).
Fisher, 1962:427 (Moniteau Cr. near Rocheport). Hanson and Campbell,
1963 (linear distr.; Perche Cr.). Gilbert, 1964:95-121,140-151 (char.;
compar.; syn.; Missouri local, mapped). Cross, 1967:113 (Missouri R.
local mapped ) .

Distribution and habitat. — The present center of abundance for
the common shiner is in short, direct tributaries of the Missouri
River in central and west-central Missouri with isolated populations
in a few tributaries of the upper Chariton River (Map 53). An
early report for the Hundred and Two River near Maryville (Jordan
and Meek, 1885:12) suggests a former more widespread distribu-
tion in northwestern Missouri. The common shiner is one of the
most abundant minnows in the creeks of central Missouri, and
seems not to have suffered any further restriction in distribution
since the early 1940's. It inhabits small, moderately clear streams
having high gi-adients and a predominance of gravel, rubble and
bedrock pools. These streams lack permanent strong How and are
frequently reduced to a series of isolated pools by late summer
and early autumn.

Taxonomic considerations and zoogeography. — The taxonomic
and distributional relationships of N. cornutus and N. chrysocepha-
lus are complex. Over much of their extensive ranges they are
allopatric, but they are broadly sympatric in the upper Ohio, Mis-
sissippi, and Great Lakes systems. There they exhibit all degrees
of relationship. In places only one form occurs; in others they act
as subspecies and at some localities they occur together and remain
distinct. The two were long treated as subspecies, but Gilbert
( 1961 ) subjected them to a thorough analysis and considered them
to be valid species. Notropis cornutus and N. chrysocephahis are
allopatric in Missouri, although they occur in adjacent tributaries
of the Missouri River in southern Warren County. Gilbert (1961:
189) suggested that the distributional relationship of these two
species in Missouri indicates that they have come together at cer-
tain times since they invaded the area, and the present allopatric
relationship is most likely due to competition. Missouri populations
of N. cornutus and N. chrysocephahis exhibit no evidence of intro-
gression or intergradation.

Missoulu Fishes 347

Gilbert (1964:106-107) favored an origin for N. cormitus in
the pre-Pleistocene Laurentian system. If this hypothesis is correct,
occurrence of N. cormitus in Missouri dates from one of the Pleisto-
cene ice advances. This species, as well as many other northern
species, may have invaded Missouri during more than one ice
ad\ance, only to be eliminated with the warming trend that oc-
curred during the interglacial periods. The present populations of
N. cormitus in Missouri may date no farther back than the Wiscon-
sin glacial period.

Notropis chrysocephalus (Rafinesque) — striped shiner

Nofro))is mcgalops: Jordan and Meek, 1885:17 (Grand R. at Clinton and/or
Tebo Cr. at Calhoun). Call, 1887:76 (West Fork Black R., Reynolds Co.;
jacks Fork, Shannon Co.; Pinev R., Texas Co.; Meramec R., Dent Co.).
"Meek, 1891:117,121,126,129 (Big Piney R. at Cabool; Lick Fork at Mans-
field; Shoal Cr. near Neosho; James R. near Springfield; Bryants Cr. near
Mansfield; Xorth Fork White R." S Cabool).

Notropis comtitiis: Fowler, 1921:399 (Fox Cr., trib. Meramec R.). Patriarche
and Campbell, 1958 (abund.; Clearwater Res.).

Notropis cormitus chrysocephalus: Martin and Campbell, 1953:46 (abund.;
Black R.).

Notropis chrysocephalus chrysocephahis: Gilbert, 1961 (char.; compar.; Mis-
souri local, mapped). Gilbert, 1964:95-121,157-166, map 5 (char.; com-
par.; syn.; Missouri local, mapped).

Distribution and habitat. — The striped shiner occurs over most
of the Ozarks, and at scattered localities in the eastern till plains
(Map 54). It is not uniformly abundant over its Missouri range,
and in some streams it seems to have declined in the past 80 years.
At present, it is one of the most abundant minnows in the eastern
and southern Ozarks. It is now rare in the Gasconade system,
although it was reported "common" by early collectors (Call, 1887:
76; Meek, 1891:121). No specimens have been taken in Big Piney
River in spite of almost annual seining since 1950. In the Spring
River drainage of southwestern Missouri it was reported as "very
common" by Meek (1891:126), and it occurred in several collec-
tions made there in 1940 and 1941. However, only a single speci-
men occurred in numerous collections made in that stream system
in 1964. A similar decline may have occurred in the Osage system.
At present the striped shiner is confined to the southern part of the
Osage system, but Jordan and Meek (1885:17) reported it from
South Grand River many miles north of recent records. The
reasons for decline of the striped shiner are not apparent. The
present scarcity of this fish in the Gasconade system is particularly
puzzling, because N!. chrysocephahis remains abundant just across
the dix'ide in the Meramec system, where stream conditions seem
to be similar.

348 University of Kansas Publs., Mus. Nat. Hist.

In Missouri the striped shiner is most abundant in clear, perma-
nent-flowing streams with clean gravel or rubble bottoms. It fre-
quently occurs just below riffles in a slight to moderate current
but is more often found in nearby backwaters or short, rocky pools
with little or no current. Gilbert (1964:185) stated that N. chryso-
cephahis is more tolerant of warm, turbid conditions than N. cornu-
tus, but the reverse relationship seems to exist in Missouri, where
N. cornutus occupies the warmer, more turbid prairie streams, and
N. chnjsocephalus inhabits the cooler and clearer streams of the
Ozark Uplands.

Zoogeography. — Gilbert (1964:106) suggested that N. chryso-
cephahis had its origin in the Mississippi system and was brought
into contact with N. cormiius when the latter species gained access
to the Mississippi system during the Pleistocene. If this explanation
is correct, N. chrysocephalus or its progenitor long may have occu-
pied the Ozark Uplands.

Notropis chalybaeus (Cope) — ironcolor shiner

Distribution and habitat. — The ironcolor shiner is known in
Missouri only from the lowlands (Map 55). It is not widespread,
but where it occurs it is usually abundant. Because it was formerly
present in eastern Iowa (Harlan and Speaker, 1956:95), it may
also have occurred along the Mississippi River in Missouri. It has
not been taken in Iowa for more than 50 years. Notropis chaly-
baeus is found only in the clearest ditches, where there is little or
no current and an abundance of submergent aquatic ^'egetation.

Zoogeography. — Notropis chahjbaeus may have inhabited the
Mississippi Valley in preglacial time or entered from the east by
way of stream connecti\'es formed with eastern Gulf coastal drain-
ages during the Pleistocene. The broadly disjunct populations
around the lower end of Lake Michigan indicate a more widespread
northern distribution in the past; possibK' these populations are
southern relicts that date from the post-glacial Climatic Optimum.
However, Gerking (1947) suggested that N. chahjbaeus followed
the Wisconsin ice sheet into northern Indiana and southwestern
Michigan soon enough to take advantage of minor glacial connec-
tives in its dispersal. If this interpretation is correct, N. chahjbaeus
may have achieved its present northern distribution before the
Climatic Optimum. It is also possible that this disjunct range
results from the extirpation of intervening populations within his-
toric time.

Missouri Fishes 349

Notropis texanus (Girard) — weed shiner

Dislrihiitiun diul Itahitat. — The weed shiner is widespread in
the lowlands, and penetrates into adjacent sections of the Ozarks
along the larger streams (Map 56). It is more widespread in the
lowlands than the ironcolor shiner, but does not seem to attain as
high a population density. Like N. chahjbaeiis, the weed shiner
occurs northward in the Mississippi Basin to Iowa and southern
Wisconsin (Suttkus and Raney, 1955:24) and should occur along
the Mississippi River in northeastern Missouri. An old record for
St. Louis suggests that formerly this may have been the case.
Harlan and Speaker (1956:95) reported it from two localities on
the Mississippi River in northeastern Iowa and predicted that it
"probably occurs throughout the downstream slough areas of the
Mississippi River." The weed shiner occurs most abundantly in
large ditches and lowland rivers having noticeable current, a sandy
bottom, and little or no aquatic vegetation. Therefore, it is sepa-
rated ecologically from the ironcolor shiner.

Zoogeography. — The present distribution of N. texanus suggests
that it inhabited the Teays-Mississippi system in preglacial time.
This species and N. chalybaeus seem to have had a similar history
of postglacial redispersal into the upper Mississippi Valley and
southwestern Great Lakes Region; the northern disjunct popula-
tions of both are perhaps subject to the same explanation.

Notropis hudsonius (Clinton) — spottail shiner

Notropis hudsonius: Forbes and Richardson, 1920:141-143, map XXXVIII
(abund.; Mississippi R. local.).

Distribution and habitat. — The spottail shiner is uncommon
but generally distributed in the Mississippi River above the mouth
of the Missouri River (Map 57). In the lower Mississippi it is
known only from a few scattered localities. It is not known from
the Missouri River system in Missouri but occurs farther upstream
in the Missouri system of northwestern Iowa (Cleary, 1956:map
45), southwestern Minnesota (Underbill, 1957:map 16), and South
Dakota (Bailey and Allum, 1962:62). Notropis hudsonius inhabits
moderately clear waters, over a bottom of sand, gravel or rubble.
It seldom occurs where there is a strong current. In Missouri it
is strictly a fish of the larger rivers, but elsewhere it commonly
occurs in lakes.

7,00 geography. — This northern species may have been localized
in northern drainages in preglacial time, gaining access to the
Mississippi Valley by way of connectives that developed during

350 University of Kansas Publs., Mus. Nat. Hist.

the Pleistocene. Bailey and Alluni (1962:123) indicated that N.
hudsonius and a number of other species probably gained access
to the upper Missouri system by way of headwater stream connec-
tions with the upper Mississippi. All species so considered by
Bailey and Allum occur in the Missouri Ri\'er system of South
Dakota but have a restricted distribution elsewhere in the Missouri
Basin.

Notropis blennius (Girard) — river shiner

Notropis jejunus: Forbes and Richardson, 1920:map XLII (Mississippi R.
local.).

Notropis blennius: Cleary, 1956:298, map 46 ( Des Moines R. local.). Fisher,
1962:427 (Missouri R. local). Metcalf, 1966:tab]e 11 (char.; Missouri R.,
Platte Co.; Mississippi R., Grand Tower, 111.). Cross, 1967:117 (Missouri
R. local, mapped).

Distribution and habitat. — The river shiner is widespread in the
Missouri and Mississippi rivers, but it occurs only occasionally in
tributary streams (Map 58). In the Mississippi River, it is exceeded
in abundance only by the emerald shiner. In the Missouri River
it is common from the northern border of the state downstream to
Lexington but is known from only one collection between Lexington
and the river mouth. In a series of collections I made in 1963, this
species occurred abundantly in a collection made near the mouth
of the Missouri River but was not taken again below Lexington, 322
river miles upstream. From Lexington upstream to the state line
it occurred in every collection, making up 6 per cent by number
of all fishes in drag seine hauls. That this distribution pattern has
persisted for 20 years or more is indicated by the fact that Fisher
(1962: table 1) reported the river shiner from only two Missouri
River collections, both above Kansas City. I can think of no
plausible explanation for this unusual distribution. Notropis blen-
nius inhabits the main channels of large rivers. It avoids strong
currents and occurs over all types of bottom. It is tolerant of con-
tinuous high turbidity but reaches its greatest abundance in mod-
erately clear waters.

Variation and zoogeography. — Hubbs and Bonham (1951:103-
107) tentatively recognized two subspecies of N. blennius. They
reported the nominate subspecies from the Arkansas (except for
the Neosho and Illinois drainages) and Red River systems in Okla-
homa and from the Arkansas and Missouri systems in Kansas. The
wide-ranging subspecies (N. b. jejunus) was reported to occupy
most of the remainder of the range, with populations from Ne-
braska being intermediate, in some respects, between the two sub-
species. Metcalf (1966:120) reported intermediate specimens from

Missouri Fishes 351

the Kansas River system. I lia\'e not attempted a variational analy-
sis of A^ blenniiis in Missouri, but a superficial comparison of speci-
mens from the Missouri Ri\er in northwestern Missouri with those
from the Mississippi River suggests slight phenotypic diderences.
Metcalf (1966:122) suggested that the "N. /;. blennim" pheno-
type may have differentiated in the preglacial drainage of the
central and southern plains and subsequently came into contact
with eastern populations when these streams were deflected into
the central Mississippi Valley.

Notropis greenei Hubbs and Ortenburger — wedgespot shiner

Notropis hoops: Meek, 1891:121 (char.; Gasconade R. and Little Piney Cr.
near Arlington; Jones Cr. near Dixon).

Notropis greenei: Hubbs and Ortenburger, 1929:78-81 (orig. descr.; para-
types from trib. of Big R., 6 mi. S Potosi). Martin and Campliell, 1953
(abund.; hab.; Black R.). Patriarche and Campbell, 1958:255-256 (abund.;
Clearwater Res. ).

Distribution and habitat. — The wedgespot shiner is confined to
the Ozarks and reaches its greatest abundance in the Gasconade
and Meramec systems (Map 59). It is common in most streams of
the southern Ozarks but is not known from Eleven Point River.
It is rare and highly localized in distribution in the Neosho and
Osage systems. The reference by Meek (1891:121) to the wedge-
shaped spot at the base of the caudal, the smaller eye, and more
posterior position of the dorsal fin in his specimens of "Notropis
boops' from the Gasconade system, compared with specimens from
the Meramec system, strongly suggests that he had this species.
Notropis greenei is most abundant in clear streams with permanent
strong flow and gravelly or rocky bottoms. It never occurs in small
headwater streams. The wedgespot shiner is usually found near
riffles in a slight to moderate current, where the bottom is mostly
sand and fine gravel.

Zoogeography. — The wedgespot shiner is endemic to, and may
have evolved in, the Ozark Uplands. Unlike some other Ozark
endemics, it seems to have no near-relatives in uplands east of the
Mississippi River.

Notropis boops Gilbert — bigeye shiner

Notropis scahriceps: Call, 1887:76 (West Fork Black R. and Barren Fork,
Reynolds Co.; Jacks Fork and Spring Valley Cr., Shannon Co.).

Notropis boops: Meek, 1891:117 (Meramec R. and Big Dry Fork near St.
James; Little Dry Fork near Rolla). Martin and Campbell, 1953 (abrmd.;
hab.; Black R.). Patriarche and Campbell, 1958:255-256 (abund.; Clear-
water Res.).

Notropis illecehrosus: Evermann and Kendall, 1895:470 (spring branch at
Neosho ) .

352 Unr'ersity of Kansas Publs., Mus. Nat. Hist.

Notropis shumardi: Evermann and Kendall, 1895:470 (Indian Cr. near Neo-
sho ) .

Distribution and habitat. — The bigeye shiner is present in most
parts of the Ozarks and northeastern Ozark border (Map 60). It
is one of the most abundant minnows in the Meramec and Neosho
systems. The most unusual feature of its distribution is its absence
from the Osage system and extreme rarity in the Gasconade system.
Both of these streams appear to contain an abundance of habitat
of the type occupied by N. hoops elsewhere. In this respect, the
distribution of the bigeye shiner is like that of the striped shiner,
a species with which the bigeye shiner is commonly associated.
Notropis hoops is characteristic of quiet pools having clear, warm
water, a firm bottom that is relatively free of silt, and much aquatic
vegetation. Such habitat is abundant in small creeks draining un-
dissected uplands of the central Ozarks. These streams have not
cut down into the rock strata from which the major springs of the
Ozarks originate; hence they tend to be warmer and more inter-
mittent than deeply incised Ozark streams. Along larger and cooler
streams, the bigeye shiner invariably is found in quiet backwaters
and overflow pools that are several degrees warmer than the main
stream.

Zoogeography. — Notropis hoops was probably present in the
preglacial Teays- Mississippi system; it may have long inhabited
the Ozark Uplands.

Notropis dorsalis (Agassiz) — bigmouth shiner

(?) Notropis gilberti: Jordan and Meek, 1885:16 (Grand R. at Clinton and/or
Tebo Cr. at Calhoim ) .

Notropis dorsalis: Cleary, 1956:niap 47 ( Des Moines R. local.). Fisher, 1962:
427 (Missouri R. local.). Hanson and Campbell, 1963 (linear distr.;
Perche Cr. ). Cross, 1967:130 (Missouri R. local, mapped).

Distribution and habitat. — The bigmouth shiner is widespread
and abundant in the till plains (Map 61). It is the only Missouri
fish with a distribution that seems to be correlated with the glacial
border. Except for one doubtful early report for the South Grand
River near Clinton (Jordan and Meek, 1885:16), N. dorsalis has
never been taken more than a few miles south of the Missouri
River, which approximates the southern limit of glaciation in Mis-
souri. Failure of early collectors to report the bigmouth shiner from
Missouri (except as noted above), and its absence from collections
made in northwestern Missouri in the early 1940's, suggests that
this species is expanding its range in Missouri. The bigmouth shiner
occurs most abundantly in small streams with permanent flow and

Missouri Fishes 353

unstable sandy bottoms. It is usually found in open, shallow chan-
nels ha\ing slight current. It is rare or absent in larger streams,
where it is replaced In- the sand shiner. Extensive channelization
of prairie streams in the till plains since the turn of the century
created conditions favorable to the bigmouth shiner.

Zoogeography. — Western populations of N. dorsalis are some-
times referred to a distinct subspecies (N. d. pipfolepis), which
differs from the wide-ranging eastern subspecies {N. d. dorsalis)
primarih- in having the nape naked or with minute and embedded
scales. All xMissouri specimens have the nape well scaled; thus
they are referable to the nominate subspecies. This pattern of vari-
ation suggests that N. dorsalis was separated into eastern and west-
ern populations at some time in the past. Perhaps ancestral stocks
of N. d. pipfolepis survived the last glacial period in the unglaciated
western tributaries of the Missouri Ri\er, while ancestral stocks
of N. d. dorsalis survived not far south of the glacial front in the
central or eastern part of the Mississippi system. Trautman (1957:
376) attributed disjunct eastern populations of N. dorsalis to range
adjustments during and subsequent to the Xerothermic Interval.

Notropis amnis Hubbs and Greene — pallid shiner

Notropis amnis: Hubbs and Greene in Hubbs, 1951 (orig. descr.; hab.; Mis-
souri local, mapped ) .

Distribution and habitat. — As recently as 1941 the pallid shiner
occurred at many localities in eastern Missouri (Map 62). Its main
areas of occurrence were streams and ditches of the lowlands, and
the Salt and Meramec stream systems. Recent efforts to collect
N. amnis in these areas have not been successful, indicating a
marked decline in abundance or complete extirpation from Missouri.
The pallid shiner inhabits streams of medium to large size. It
seems intolerant of siltation and turbidity, and it avoids strong
currents. The reasons for its decline in Missouri are not known.

Zoogeography. — Probably N. amnis has inhabited the Missis-
sippi Valley continuously since preglacial time. Hubbs (1951)
recognized a southwestern subspecies {N. a. pinnosus) and a
northern subspecies ( 2V. a. amnis ) with a broad zone of intergrada-
tion in the central Mississippi Valley. Perhaps ancestral stocks of
iV. a. pinnosus occupied the ancestral Red or other southwestern
drainages, while those of N. a. amnis occupied the central and
northern part of the Mississippi Valley. The presence of N. a. pin-
nosus in Gulf coastal streams of Texas provides further evidence
for former connections between these streams and the Red River.

354 University of Kansas Publs., Mus. Nat. Hist.

Such a connection was suggested by Gilbert and Bailey (1962:817)
to explain a similar distribution pattern in N. shtimardi.

Notropis whipplei (Girard) — steelcolor shiner

Nofropis tchipplii: Meek, 1891:117 (in part (?); Big Dry Fork near St.
James). Gibbs, 1963:511-520 (syn.; char.; compar.; liab.; Missouri local,
mapped ) .

Distribution and habitat. — The steelcolor shiner is common and
widespread in the Meramec, Headwater Diversion, and St. Francis
stream systems and occurs occasionally in the southern Ozarks as
far west as the upper White River (Map 63). North of the Missouri
River it occurs only in the Cui\'re River system. The habitat re-
quirements of this minnow are similar to those of the spotfin shiner;
the two are often taken together in the eastern Ozarks. The steel-
color shiner seems less tolerant of high turbidity than the spotfin
shiner, and avoids the Mississippi and Missouri rivers, where the
spotfin shiner occurs in small numbers.

Zoogeography. — Gibbs (1963:525-526) suggested that the com-
mon ancestral stock of N. ichipplei and N. analostanus (Girard)
differed little from the present N. wlupplei and had its center of
dispersal in the Mississippi Valley. If this is the case, N. ichipphi
or its ancestral stock may have occupied the Ozark Uplands con-
tinuously since at least the late Tertiary.

Notropis spilopterus (Cope) — spotfin shiner

Notropis whipplei: Meek, 1891:117,121 (in part (?) Big Dry Fork near St.
James; Gasconade R. near Arlington; Osage Fork 6 mi. SE Marshfield).

Notropis notatus: Call, 1887:76 (Piney R., Te.xas Co.).

Notropis ichippJii: Fowler, 1910:282 (Carthage, Mo.).

Notropis spiloptertis hypsisomatiis: Gibbs, 1957b: 195-198 (syn.; char.; com-
par.; Missouri local, mapped).

Distribution and habitat. — The spotfin shiner is common in the
Gasconade and Meramec systems, and occurs in lesser numbers
in the upper Mississippi River and in the Spring River system of
southwestern Missouri (Map 64). Strays occur in the lower Mis-
souri River, and in the lower Mississippi River as far south as Cape
Girardeau County. Notropis spilopterus inhabits moderately clear,
permanent-flowing streams ranging in size from small creeks to
large rivers. It is most often found near riffles o\er a clean gravel
or rubble bottom.

Zoogeography. — Gibbs (1957b) recognized two subspecies
within the rather broad range of this species. The subspecies to
which Missouri populations are assigned [N. s. hypsisomatus Gibbs)
occupies tributaries of the Mississippi River north of the Ohio

Missouri Fishes 355

River, with disjunct populations in the upper Arkansas system.
Gibbs (1957b:204) suggested that subspeeifie differentiation was
initiated w hen the Nebraskan or Kansan ice sheet spHt the range
into western and eastern segments and that the Ozark region was
the most Hkely center of origin for the western subspecies. He
further suggested that disjunct populations in the Arkansas are the
result of stream capture from the Osage or White river systems.
Notropis spiloptenis is found in neither of the latter two stream
systems today, so this explanation assumes a former more wide-
spread distribution in the Ozark Uplands. I can think of no plausi-
ble reason for the persistence of N. spiloptenis in the upper Arkan-
sas system, if it was eliminated from the White and Osage systems.
Perhaps N. spiloptenis was never present in the Osage or White
ri\'er systems, but instead dispersed into the upper Arkansas by
way of the lower Mississippi and lower Arkansas rivers. The Wis-
consin glacial period, with the consequent southward displacement
of northern species and initiation of an erosional cycle in the Mis-
sissippi Embayment, would have provided an opportunity for such
dispersal. Climatic changes and alluviation subsequent to retreat
of the Wisconsin ice sheet would have favored the related species
N. venustus in the lower Mississippi Valley, permitting it to replace
its northern counterpart in that area.

Notropis venustus (Girard) — blaektail shiner

(?) Notropis notatus: Call, 1887:76 (very abund.; West Fork Black R. and
Toms Cr., Reynolds Co.; Jacks Fork, Spring Valley Cr., Barren Cr., and
Sinking Cr., Shannon Co.).

Notropis venustus: Patriarche, 1953:247 ( Lake Wappapello ) .

Notropis venustus venustus: Gibbs, 1957a: 175-189 (syn.; char.; compar.; hab.;
Missouri local, mapped).

Distribution and Jmbitat.— The blaektail shiner is the most
abundant minnow in the lowlands and penetrates for considerable
distances into the Ozarks along the larger streams (Map 65). In
Ozark streams it is much less abundant than the closely related
whitetail and steelcolor shiners. Notropis venustus occupies a va-
riety of habitats but shows a decided preference for flowing waters.
It is most abundant in the large, sparsely vegetated lowland ditches,
where there is a strong current and a firm sand or gravel bottom.

Zoogeo^rapliy. — This species is most closely allied to N. spilop-
terus, and the distributional relationship of the two suggests an
origin for N. venustus in Gulf coastal drainages (Gibbs, 1957a:
195-200).

356 University of Kansas Publs., Mus. Nat. Hist.

Notropis galacturus (Cope) — whitetail shiner

Notropis galacturus: Call, 1887:76 (Jacks Fork and Spring Valley Cr., Shan-
non Co.). Meek, 1891:129 (Tames R. near Springfield). Martin and
Campbell, 1953 (abund.; hab.; Black R.). Patriarche and Campbell, 1958:
255-256 (abund.; Cleanvater Res.). Gibbs, 1961:338-343 (syn.; char.; corn-
par.; hab.; Missouri local, mapped).

Distribution and habitat. — The whitetail shiner is one of the
most abundant minnows in streams draining the southern slope
of the Ozarks (Map 66). Notropis galacturus inhabits the swifter
sections of clear, high-gradient streams with permanent strong
flow and firm, silt-free bottoms. It avoids small headwater creeks.

Zoogeography. — Notropis galacturus occurs disjunctly in the
Ozark and Appalachian uplands. A report of this species from
Wyatt, Missouri, about halfway between the eastern and western
segments of the range (Gibbs, 1961:342), seems to be based on
faulty locality data. Gibbs subsequently informed me that the jar
containing this collection (USNM 63029) bears an external label
and a catalog entry reading "Wyatt, Mo.," but the label in the jar
reads "Wyatt Cr., Wyatt, Ark." According to Gibbs, the letters
interpreted as "W " on the pencil label in the jar could as well be
interpreted as "M," and the locality becomes "Myatt Cr., Myatt,
Ark." in the White River system, well within the known range of
N. galacturus. Gibbs (1961:351) suggested that N. galacturus dif-
ferentiated in the Tennessee system, and then dispersed westward
into the Ozarks, probably during one of the glacial periods of the
Pleistocene.

Notropis camurus (Jordan and Meek)— bluntface shiner

Notropis galacturus: Meek, 1891:126 (Shoal Cr. near Neosho).
Notropis camurus: Gibbs, 1961:343-351 (syn.; char.; compar.; hab.; Missouri
local, mapped).

Distribution and habitat. — The bluntface shiner is confined to
the Neosho system on the southwestern slope of the Ozark Uplands
(Map 67). It has been reported from the upper White River sys-
tem (Gibbs, 1961:350), but this report is probably based on faulty
locality data. The collection was made by personnel of the Missouri
Department of Conservation in 1942. An examination of the field
sheet for this collection reveals that the locality was originally
recorded as "Shoal Creek-White R. near Protem, Tanev Co., near
Ark. state line.' It was later changed to read "Shoal Cr.-( Spring R.
watershed) near Kansas state line." Also included in this collection
was N. lutrensis, a fish known in Missouri from the Spring Rixer
but not from the White River. Notropis camurus inhabits moder-

Missouri Fishes 357

ately clear, permanent-flowing streams with strong current and
clean gravel or rubble bottoms. It is most often found near riffles
in a moderate current.

Zoogeog.rop]uj. — According to Gibbs (1961:350), IV. camunis
was derix'cd from a western stock of the same ancestral form that
gave rise to N. galacttinis in the Tennessee River system. Isolated
populations in the lower Mississippi Valley indicate a former more
widespread distribution in that region for N. camurus. Possibly
this species was localized preglacially on the western slope of the
Ozark-Ouachita uplands; probably it gained access to the lower
Mississippi \^illey when the lower Arkansas River breached the
Ozark-Ouachita divide and captured upper Arkansas drainage.
Perhaps N. camurus was then able to cross the lower Mississippi
Embayment during one of the erosional cycles that accompanied
glaciation, only to be isolated in eastern drainages during the
subsequent period of allu\ iation.

Notropis lutrensis (Baird and Girard) — red shiner

Cypiinella Billingsiana: Cope, 1871:439 (orig. descr.; type local. Missouri R.
at St. Joseph).

Moniana Jugalis: Cope, 1871:439-440 (orig. descr.; type local. Missouri R. at
St. Joseph).

Notropis lutrenm: Meek, 1891: 11 8,121,125,129 (Big Dry Fork near St. James;
Little Piney Cr. near Arlington; Maries R. near Dixon; (?) James R. near
Springfield). Jordan and Sleek, 1885:11,14,15,17 (Hundred and Two R.
at Maryville; Talio Cr. 6 mi. E Lexington; Flat Cr. near Sedalia and/or
Blackwater R. at Brownsville; Grand R. at Clinton and/or Tebo Cr. at
CaUioun). Fowler, 1910:279-280, figs. 19 and 20 (syn.; char.; Clinton;
Osage R.; Browns\ille; Sedalia; Marshfield; Greenfield; t>'pes of Cypiinella
Billingsiana and Moniana Jugalis figured). Forbes and Richardson, 1920:
map XXXIX (Mississippi R. local.). Fisher, 1962:427 (Missouri R. local.).
Hanson and Campbell, 1963 (linear distr.; Perche Cr.). Cross, 1967:126
Missouri R. local, mapped).

Distribution and habitat. — The red shiner is the most abundant
and widely distributed minnow in the prairie region (Map 68). It
is common in most streams of the Ozark border and occurs occa-
sionally in the lowlands. Its distribution largely complements that
of its close relatives (whitetail, bluntface, spotfin, and steelcolor
shiners), and competition from these species may be an important
factor in controlling its distribution. In the Neosho system, for
example, it is confined primarily to the warmer, more turbid North
Fork of Spring River and the lower main stream of Spring River,
leaving the cooler, clearer Ozark streams to the bluntface shiner.
Notropis lutrensis scarcely enters lowland ditches, where the black-
tail shiner is abundant, but penetrates the short tributaries of the
Mississippi River immediatch' to the north, where no other species

358 University of Kansas Publs., Mus. Nat. Hist.

of this group occurs. Along the northern Ozark border it is found
throughout most of the Osage system, where no other species of the
Cyprinella group occurs, but scarcely enters the Gasconade and
Meramec systems, where N. spilopterus and N. wliipplei are found.

Notropis lutrensis occurs in streams of all sizes, but it is most
abundant in large creeks and rivers. It inhabits a variety of habi-
tats, including quiet pools and backwaters as well as riffles. Al-
though it sometimes thrives when introduced into impoundments,
in natural waters it is abundant only in streams. The red shiner
is tolerant of high turbidity and siltation but avoids waters that are
continuously clear and cool.

Zoogeography. — Metcalf (1966:118-119) suggested a southwest-
ern origin for N. hitrerisis, because of its distribution and habitat
preferences, and the presence of closely related forms in north-
eastern Mexico.

Notropis sablnae Jordan and Gilbert — Sabine shiner

Distribution and habitat. — The Sabine shiner has been collected
in Missouri only from Black River where it descends into the low-
lands (Map 69). In 1964, the Sabine shiner was found to be fairly
common a few miles below Poplar Bluff, but only one specimen
was contained in a collection made several miles farther upstream.
None occurred in collections from Black River above and below
these localities. Black River at the point where the Sabine shiner
was collected is clear, narrow, and deep, with a bottom composed
entirely of fine, silt-free sand. The Sabine shiner was taken over
sand bars in slight to moderate current.

Zoogeography. — This southern species may have inhabited the
lower Mississippi Valley in preglacial time or dispersed into the
Mississippi from western Gulf coastal drainages during the Pleisto-
cene. Disjunct populations in the White River system may indicate
a more widespread northern distribution in the past, perhaps during
the postglacial Climatic Optimum.

Notropis stramineus (Cope) — sand shiner

Hyhopsis rnisstiriensis: Cope, 1871:437-38 (orig. descr.; types from "near St.
Joseph, Missouri.")

Notropis deliciosus: Jordan and Meek, 1885:11,13,14,15,16 (Hundred and
Two R. at Mary\iile; Missouri R. at St. Joseph; Talx) Cr. 6 mi. E Lexing-
ton; Flat Cr. near SedaMa and/or Blaekwater R. at Brownsxille; Grand R.
at Clinton and/or Tebo Cr. at Calhoun). Call, 1887:76 ( Pinev R., Texas
Co., rare). Meek, 1891:124 (Sac R. near Springfield). Hanson and
Campbell, 1963 (linear distr.; Perche Cr.).

Notropis deliciosus missouriensis: Call, 1887:80 (Bear Cr., Boone Co.).

Missouri Fishes 359

Nofiopis stiatniueus: Fisher, 1962:426 (Missouri R. local.). Bailey and Alliim,
1962:tal)le 5 (var.; Missouri local.). Metcalf, 1966:table 12 (var.; Missouri
local.). Cross, 1967:133 (Missouri R. local, mapped).

Distribution and habitat. — The sand shiner is widespread in the
prairie region (Map 70). This species and the red shiner comprise
the bulk of the minnows in many streams of the prairie region. The
only real difference in the distribution of these two species in
Missouri is the absence of the sand shiner from the Neosho system
and its greater abundance and more widespread distribution in the
Gasconade and Meramec systems. As its name implies, N. stra-
mineus shows a strong affinity for sandy bottoms. It is tolerant of
a wide range of turbidities and occurs in streams of all sizes. How-
ever, it is seldom abundant in the largest rivers, and it is replaced
towards the headwaters of many streams by the bigmouth shiner,
another minnow exhibiting an affinity for sandy bottoms. The
sand shiner is most abundant in the shallow, sandy pools of mod-
erately large creeks having permanent flow, moderately clear water,
and low or moderate gradient.

Variation and zoogeography. — Bailey and Allum (1962:64)
recognized two subspecies of the sand shiner in the northern plains,
distinguishing them primarily on the basis of the number of cir-
cumferential scales. They reported the coarse-scaled N. .s. stra-
mineus from most of eastern South Dakota and northwestern Iowa,
whereas the small-scaled N. s. missuriensis (Cope) occupies the
remainder of the Missouri system as far downstream as the Nio-
brara River in Nebraska. They found little evidence of intergrada-
tion in South Dakota, but reported intergrades in southeastern
Nebraska and northwestern Missouri. They also noted (p. 68)
that the type locality for Hyhopsis missuriensis near St. Joseph,
Missouri, was within the zone of intergradation. Metcalf (1966:
124-131) extended this comparison to include most of Kansas and
Missouri. He concluded that intergradation in the Kansas River
system was "slight and mosaic in nature."

Since the Arkansas River system is occupied by N. s. missurien-
sis almost exclusively, Metcalf (1966:125) concluded that this sub-
species had its origin in the central plains and utilized former con-
nections between the Arkansas and Kansas systems in attaining its
present distribution. Isolated populations of N. s. stramineus in
the Missouri River system of eastern South Dakota have apparently
resulted from westward dispersal by way of connections between
the Missouri and the Minnesota and/or Des Moines rivers (Bailey
and Allum, 1962:64).

360 University of Kansas Publs., Mus. Nat. Hist.

Notropis topeka Gilbert — Topeka shiner

Notropis topeka: Jordan and Meek, 1885:11 (Hundred and Two R. at Maiy-
\dlle). Fisher, 1962:427 (Missouri R. tribs. near Hermann and Rocheport).
Bailey and Allum, 1962:69-70 and fig. 6 (Missomi local, compiled and
mapped). Hanson and Campbell, 1963 (linear distr.; Percbe Cr. ).

Distribution and habitat. — The main area of occurrence for the
Topeka shiner is in small, direct tributaries of the Missouri River
in central Missouri, \\ith scattered populations northwestward in
the prairie region (Map 71). An early report for the Hundred and
Two River near Maryville (Jordan and Meek, 1885:11) indicates
a possible former more widespread distribution in northwestern Mis-
souri. The Topeka shiner is nowhere abundant, but is a character-
istic element in the fish fauna of small streams in central Missouri.
The Topeka shiner inhabits quiet pools of small, clear upland creeks,
having bottoms composed mostly of sand, gravel, or rubble. These
streams often cease to flow during dry seasons, but permanent pools
are maintained by the movement of ground water through deep
beds of sand and gravel. Increased siltation as a result of intensive
culti\'ation may have reduced the abundance of the Topeka shiner
in Missouri. At present this minnow is largely restricted to direct
tributaries of the Missouri River that have sufficient gradient to
prevent extensive deposition of silt.

Zoogeography. — Metcalf (1966:132-133) suggested an origin for
N. topeka in the "Pleistocene south-draining stream of the Great
Plains." I think it is more likely that this species had its origin in
the western Mississippi Valley and that it achieved its present
highly localized distribution in the Arkansas Ri\'er system by way
of connectives that existed between the Arkansas and Kansas river
systems one or more times during the Pleistocene.

Notropis maculatus (Hay) — taillight shiner

Distribution and habitat. — The taillight shiner reaches the north-
ern limits of its range in southeastern Missouri, where it has not
been collected since 1941 (Map 72). Its habitat is the sluggish
sections of lowland rivers and creeks.

Zoogeography. — Perhaps N. maculatus has inhabited the lower
Mississippi Valley continuously since preglacial time. Alternatively,
it may have entered the Mississippi Valley from eastern Gulf coastal
drainages by way of stream connections that developed during the
Pleistocene.

Missouri Fishes 361

Notropis heterolepis Eigenmann and Eigenmann — blacknose

shiner

Xotro))is caijuga: Meek, 1891:117,121,124 (char.; abund.; Meramec R. near
St. James; Big Piney R. at Ca1)ool; Osage Fork 6 mi. SE Marshfield; Lick
Fork at Mansfield; Niangua R. near Marshfield).

Distrihtition and habitat. — The blacknose shiner is nowhere
abnndant in Missouri, where it is confined to the northern Ozarks
and adjacent Ozark border (Map 73). Although the blacknose
shiner was reported as "scarce in southern Missouri" by one early
collector (Meek, 1891:121) it occurred with much greater fre-
quency in collections then than it does today. Even some streams
where N. heterolepis occurred in the early 1940's no longer seem
to support populations. The few remaining populations in Missouri
seem on the verge of extinction; if the present trend continues N.
heterolepis will soon be eliminated from the state. The blacknose
shiner is characteristic of quiet pools of small, clear prairie streams
ha\'ing bottoms composed mostly of muck and organic debris
(often overlying sand or gravel), and moderate to large amounts
of aquatic vegetation. A secondary area of occurrence is in the
quiet, weedy backwaters of large Ozark streams. Continuous tur-
bidity and the covering of organic sediments by inorganic silt
seem to be especially unfavorable to this species, and may be the
most important factors in its declining abundance. The largest
remaining populations are in streams of the Ozark border that
drain level uplands underlain by thin, rocky soils. These soils are
not conducive to intensive cultivation, and large areas of native
grasses that retard siltation are being maintained for hay produc-
tion and pasture.

Zoogeography. — Perhaps this northern species was localized
preglacially in the Laurentian or other northern drainages, and
entered the Mississippi Valley by way of glacial connectives. Oc-
currence in the upper Arkansas system (Cross, 1967:142) may have
resulted from southwestward dispersal by way of Pleistocene
connections between the Kansas and Arkansas systems.

Notropis ozarcanus Meek — Ozark shiner

Notropis ozarcanus: Meek, 1891:129 (orig. descr.; type local. North Fork
White R. S Cabool). Martin and Campbell, 1953 (abund.; hab.; Black
R.).

Distribution and habitat. — The Ozark shiner is endemic to the
southern Ozark Uplands (Map 74). It is common in Current
River and North Fork of White Ri\'er; elsewhere it seems to be
rather rare. It was formerly abundant in the section of White River

362 University of Kansas Publs., Mus. Nat. Hist.

now inundated by Bull Shoals and Table Rock reservoirs. The
Ozark shiner inhabits large, clear streams having high gradients
and permanent strong flow. It occurs most abundantly near rifl^les
in a slight to moderate current, over a firm, silt-free bottom.

Zoogeography. — Notropis ozarcanus is represented in the Ten-
nessee system by the related mirror shiner, IV. spectrtinciihis
(Cope). Perhaps their common ancestral stock had a continuous
preglacial distribution, and became localized in the two uplands
during an early ice advance of the Pleistocene. Alternatively, east-
west dispersal of the common stock may have occurred during the
Pleistocene, by way of the lower Ohio and Mississippi rivers.

Notropis volucellus (Cope) — mimic shiner

Notropis volucellus wickliffi: Traiitman, 1931 (orig. descr.; hah.; St. Louis,
Missouri ) .

Distribution and habitat. — Three well-marked forms of the
mimic shiner occur in Missouri (Map 75). One of these, 'N. v.
icickliffi Trautman, is confined to the Missouri and Mississippi
rivers and the lower few miles of their major tributaries. Another,
N. V. vohicelhis (Cope), occurs abundantly in the larger lowland
ditches, in major Ozark streams draining into the lowlands, and
in the Osage and Meramec systems of east-central Missouri. The
two are usually isolated ecologically; N. v. vohicellus inhabits clear
streams ranging in size from medium-sized creeks to rather large
rivers, and N. v. icickliffi inhabits the main channels of very large,
moderately clear to highly turbid rivers. The zone of intergrada-
tion between the two is exceedingly narrow and is limited to the
lower reaches of larger tributaries of the Missouri and Mississippi
rivers. A third form has no available name and occurs in the
Neosho system of southwestern Missouri. This is the "unnamed
creek subspecies" referred to by Hubbs and Bonham (1951:103).
As indicated by these authors, it differs from typical N. vohiceUus
in the only moderate elevation of the lateral line scales, the large
eye, the high dorsal fin, the chubb\' body, and the smaller size.

Zoogeography. — Until a thorough study has been made, con-
clusions concerning the distributional history of this complex must
be tentative. The distributional relationship of the unnamed south-
western subspecies and N. v. vohicelhis suggests that the former
had its origin in streams draining the southern plains, at a time
when N. v. vohicelhis was localized in the preglacial Teays-Missis-
sippi or Laurentian systems. Notropis v. icicMiffi seems to have
evolved from stock that became specialized for life in large rivers
of the Teays-Mississippi system.

Missoulu Fishes 363

Notropis buchanani Meek — ghost shiner

N()ti(>]>i.\ voluccllus: Fisher, 1962:427 (Missouri R. local.).

Notropis buchanani: Cross, 1967:139 (Missouri R. local, mapped).

Distribution and habitat. — The ghost shiner reaches its greatest
abundance in the prairie and Ozark border streams of west-central,
central and northeastern Missouri (Map 76). It is common in the
Missouri River as far upstream as Boone County but is rare in the
upper Missouri and in the Mississippi River. Notropis buchanani
inhabits the low-gradient sections of large creeks and rivers having
permanent flow and moderately clear water. It is a quiet-water
species, inhabiting the larger pools and the lower reaches of tribu-
taries or other protected backwaters, where there is no noticeable
current.

Zoogeography. — Notropis buchanani is most closely allied to
N. Dohicellus, with which it was long considered to be conspecific.
The widespread southwestern distribution of N. buchanani sug-
gests an origin in Gulf coastal drainages west of the Mississippi
River.

Dionda nubila (Forbes) — Ozark minnow

Dionda nubila: Call, 1887:75 (Toms Cr., Reynolds Co.; Piney R., Texas Co.).
Hubbs and Ortenburger, 1929:90 (trib. Big R. S Potosi; Sarcoxie). Mar-
tin and Campbell, 1953 (abund.; hab.; Black R.).

(?) Hijbognathus meeki: Call, 1887:75 (West Fork Black R., Reynolds Co.;
Jacks Fork, Shannon Co. ) .

Hybognafhus nubila: Meek, 1891:117,121,124,126,129 (Meramec R. near St.
James; Little Dry Fork near Rolla; Little Piney Cr. near Arlington and/or
Newburg; Jones Cr. and Maries R. near Dixon; Lick Fork and Bryants Cr.
near Mansfield; Osage Fork and Xiangua R. near Marshfield; Sac R. and
James R. near Springfield; Shoal Cr. near Neosho; North Fork White R.
S Cabool). Evermann and Kendall, 1895:470 (Indian Cr. near Neosho).

Distribution and habitat. — With respect to its distribution in
Missouri the Ozark minnow is well named, for it occurs only in
the Ozark Uplands, where it is one of the most abundant fishes
(Map 77). Although it is widespread elsewhere in the Meramec
system, D. nubila is absent from the Bourbeuse River and the upper
Dry Fork. It occurs north of the Missouri River only in a few small
streams in southern Montgomery and Warren counties, and it is
not found in any tributary of the Missouri River above the mouth
of the Osage River. The Ozark minnow is almost invariably found
in association with either the bleeding shiner or the duskystripe
shiner, and its distribution in the state parallels the combined dis-
tribution of these two closely related species to a remarkable degree.
The Ozark minnow inhabits streams with silt-free bottoms and a
permanent strong flow of clear, cool water. It is usually found in

364 University of Kansas Publs., Mus. Nat. Hist.

protected backwaters near riffles, or in pools just below riffles where
the current slackens.

Zoogeography. — Dionda mihila seems to be autochthonous to
the Mississippi Valley, and has probably long occupied the Ozark
Uplands. Disjunct populations in the Driftless Area may have been
established by dispersal from an Ozark refugium postglacially, but
it is equally plausible that D. nubila was present in or near the
Driftless Area throughout the Wisconsin glacial period. G. R. Smith
(1963:285) reported this species in a late-Illinoian fauna in south-
western Kansas, thus providing evidence of a more western distri-
bution during glaciation.

Ericymba buccata Cope — silverjaw minnow

Distribufion and habitat. — The silverjaw minnow occurs in the
Meramec River and other tributaries of the Mississippi River south-
ward to the Headwater Diversion (Map 78). Although not widely
distributed, it is common at most localities where it occurs. Ericymba
buccata inhabits the sandy stretches of small, clear, permanent-
flowing streams. It seems to be the ecological counterpart of the
plains-inhabiting N. dorsalis in the clearer and more stable streams
in the eastern and southern parts of the Mississippi Valley. Traut-
man (1957:376) noted the similarity in requirements of these two
species and presented data indicating interspecific competition be-
tween them in Ohio. I have found them together only in a single
collection from the lower Meramec River, where both were rare.

Zoogeography. — The distribution of E. buccata suggests an
origin in the preglacial Teays-Mississippi system. It probably sur-
vived the last ice advance in the lower Mississippi Valley and re-
dispersed northward solely from that area. Trautman (1957:376)
indicated that this species is still extending its range northeastward.

Hybognathus hankinsoni Hubbs — brassy minnow

Hijhognathus hankinsoni: Bailey, 1954 (Missouri local, mapped). Cross, 1967:
144 (Missouri R. local, mapped).

Distribution and habitat. — The brassy minnow is common in
small tributaries of the upper Chariton River and occurs rarely in
the Missouri River and its tributaries westward in the Dissected
Till Plains (Map 79). It inhabits small, moderately clear, low-
gradient streams with permanent pools and bottoms of sand or fine
gravel.

Zoogeography. — Possibly this northern species was localized
preglacially in the Hudson Bay or Laurentian systems. Bailey

Missouri Fishes 365

(1954:291) concluded that //. Jiankimoni is "of Mississippi deriva-
tion" and "survived the Wisconsin glaciation in the Missouri and
Upper Mississippi drainages." That it has had a more southerly
distribution in the past is indicated by its presence in Illinoian
fossil deposits from southwestern Kansas (G. R. Smith, 1963:279).

Hybognathus nuchalis Agassiz — central silvery minnow

Hybognatluis niiclialis: Agassiz, 1855:224 (orig. descr.; type local. Quincy, 111.;
St. Louis, Mo.). Garnian, 1890:143 ( Mi,ssis.sippi R. near Quincy, 111.).
Hubbs, 1951:17 (Mississippi R. at St. Louis). Forbes and Richard,son,
1920: map XXV (Mississippi R. local.).

Taxonomic considerations. — At present, most workers recognize
four species of Hybognatluis, all of which occur in Missouri. Two
of these (//. Imnkinsoni and H. hayi) appear to present no taxo-
nomic difficulties. The other two {H. michaJis and H. placitus) are
superficially similar, and until recently there was no consensus re-
garding their taxonomic relationship. At different times and by
different workers they have been treated as distinct species, as sub-
species, and as environmental variants. The discovery of striking
differences in the pharyngeal apparatus of the two forms (Niazi
and Moore, 1962; Bailey and Allum, 1962:72 and plate I) seems to
leave little doubt that they are distinct species.

Hi)])ognathus nuchalis is commonly divided into two subspecies:
H. n. regius Girard of the Atlantic Slope, and N. n. nuchalis of the
Mississippi Valley. The two were formerly thought to intergrade
along the eastern Gulf Coast (Hubbs and Lagler, 1947:68), but
they are now known to be entirely allopatric (Bailey, 1954:291).
In addition to the slight differences in external morphology used
to separate the two forms in the past, they are now known to differ
in the shape of the basioccipital process; the process of regius is
most nearly like that of placitus (Al-Rawi and Cross, 1964). Con-
sidering the nature of the differences and the absence of opportu-
nity for gene exchange, nuchalis and regius could be considered as
different species.

In examining numerous series of nuchalis from the Mississippi
Valley, I find that the "H. n. nuchalis" of previous workers is a
complex of two forms. One of these occurs sympatrically with
H. placitus in the Missouri River system; the other occupies the
remainder of the nuchalis range in the Mississippi Valley and Gulf
coastal drainages. Both occur in the Mississippi River from the
mouth of the Missouri River downstream to the mouth of the Ohio
River. There seems to be no lessening of differences between the
two forms in the zone of sympatric occurrence, and I tentatively

366 University of Kansas Publs., Mus. Nat. Hist.

consider them to be distinct species. I resurrect the name Hijhog-
natlnis argyritis Girard for the form in the Missouri River system,
and restrict the name Hijhognathus niichalis Agassiz to the form in
the central Mississippi Valley. The types of argyritis consist of
seven specimens (USNM 87) from Milk River, Montana, and one
specimen (MCZ 1788) from Arkansas River near Fort Smith
(Girard, 1856). Hybognathus argyritis seems to be the only name
available for a species of Hybognathus from the Missouri River
system that is not applicable to H. placitus. The specimen from
Fort Smith is undoubtedly nucJialis; H. argyritis is not known from
the Arkansas River. The types of H. nuchalis are from Quincy,
Illinois. Fingermann and Suttkus ( 1961 ) designated a lectotype
(MCZ 1926); other syntypes originally sharing that number were
recatalogued (MCZ 40697).

The only truly diagnostic character for separating nuchalis and
argyritis is the shape of the basioccipital process. That of nuchaUs
is greatly expanded posteriorly, and the posterior margin is deeply
emarginate (see Niazi and Moore, 1962:fig. 21). The process of
argyritis is less expanded posteriorly, and the posterior margin is
truncate or only shallowly emarginate (see Bailey and Allum, 1962:
plate ID). In nuchalis the greatest width of the process exceeds its
length ( measured from its base to the center of the emargination ) ;
in argyritis the width is less than the length (Fig. 15). Specimens
from the zone of sympatric occurrence are as different in this
character as specimens from allopatric populations. External dif-
ferences between nuchalis and argyritis are slight, but a practiced
observer can identify most specimens without dissection. These
differences involve eye length (greater in nuchalis than in argy-
ritis), interorbital width and gape width (less in nuchalis than in
argyritis), and pigmentation (dark margins of scale pockets more
pronounced in nuchalis than in argyritis).

The differences in external morphology of the two forms need
to be quantified, and more material of argyritis is needed from the
zone of sympatric occurrence; I have examined many nuchalis but
only 14 specimens of argyritis ( mostly young of the year ) from that
area. Both species were found in collections from four localities.

Distribution and habitat. — Hyljognathus nuchalis, as here re-
stricted, is abundant in the lower Current, Black, and St. Francis
rivers, and the lower sections of some of the larger ditches of the
lowlands (Map 80). It occurs along the full length of the Missis-
sippi River, but is far more common below the mouth of the Ohio
River than above. It now seems to be rare in the upper Mississippi

Missouri Fishes

367

5.0-

4.0-

5
S

3.0

2.0

nuc ha I i s
□ I lopa t r ic O

sym po t r i c -O-

o rg y r i t IS

-O-

o^o o

-o-o-yoo o o
o o-s^c$tp o

o o dfcoo
(9 o- -#
o o-§- -o-
o
-oo

• •••

1.0

2.0

3.0

LENGTH (MM)

4.0

5.0

Figure 15. Relationship of basioccipital process width to basioccipital process
lengtli in Hijhognatlius nuchalis and HtjhognatJius argyiiiis. The graph is based
on data from the following specimens :

Hyhognatlius nuchalis. — Wisconsin: Chippewa R., Eaii Claire Co. ( KU 2012,
10 spec). Illinois: Mississippi R. at Quincy (UMMZ 150029, 3 spec); Piasa
Cr., Madison and Jersey Co. (uncat., 13 spec); Mississippi R., Grand Tower
(UMMZ 111579, 5 spec). Missouri: Sugar Cr., Lewis Co. (UMMZ 149435,
1 spec); Missouri R. at its mouth, St. Charles Co. ( KU 9697, 2 spec);
Meraniec R. 1 mi. S Pacific, Jefferson Co. (MU 1307, 1 spec); Mississippi R.
at Claryville, Perry Co. (uncat., 1 spec); Apple Cr., Perry Co. (UMMZ
152946, 1 spec); Brazeau Cr., Perry Co. (UMMZ 149878, 1 spec); Current
R., Ripley Co. (uncat., 10 spec). Indiana: Wabash R., New Harmony
(UMMZ 81316, 5 spec). Alabama: Town Cr.. Wilson L. (UMMZ 122851,
5 spec). Texas: Neches R. above Beaumont, Hardin Co. (UMMZ 166490,
1 spec).

Hijbognathus argyritis. — Alberta: Milk R. 3 mi. S Groton PO, 8 mi. N and
1 mi. W Alden PO (National Museum Canada 66-431, 1 spec). North
Dakota: Little Missouri R. below Marmarthen (UMMZ 94800, 3 spec).
South Dakota: Moreau R. 14 mi. N Eagle Butte, Dewey Co. (UMMZ 178956,
3 spec). Nebraska: Elkhorn R. 1 mi. N Winslow, Dodge Co. (UMMZ
135765, 2 spec); Platte R., Bellwood, Butler Co. (UMMZ 134702, 2 spec).
Missouri: Tarkio R. 4 mi. S Tarkio, Atchison Co. (uncat., 20 spec); Mis.souri
R. at Gasconade, Gasconade Co. (KU 9627, 11 spec); Meramec R. 1 mi. S
Pacific, Jefferson Co. (MU 1307, 1 spec); Mississippi R. at Claryville, Perry
Co. (uncat., 1 spec).

368 University of Kansas Publs., Mus. Nat. Hist.

River and its tributaries, but it was common in collections made
in that area in the early 1940's. The reasons for this decline are
not known. The central silvery minnow is most abundant in the
low-gradient sections of clear, moderately large streams. It seems
to be rather intolerant of continuous high turbidity, as indicated
by marked reduction in abundance in the Mississippi River above
the point where that stream receives the clear waters of the Ohio
River. The central silvery minnow avoids strong currents and
occurs most abundantly in pools and backwaters over a silt or sand
bottom.

Zoogeography. — Hyhognathus michalis, H. argyritis, and //.
placittis constitute a close-knit group having recent distributional
relationships that seem to reflect their distribution in preglacial
time. Each species has its distribution centered in an area drained
in preglacial time by one of the principal stream systems that were
modified during the Pleistocene to form the present Mississippi
River system. Hyhognathus nuchaUs is widespread in the central
Mississippi Valley; probably it inhabited the preglacial Teays-
Mississippi system. Hyhognathus argyritis has its distribution cen-
tered in the upper Missouri system, which drained in preglacial
time into Hudson Bay. Hyhognathus placitus is widespread in the
central and southern plains, and probably inhabited the preglacial
drainage of that region. The distributional relationships of these
three species still can be discerned, because they seem to have
made only modest invasions of each other's range since disruption
of the preglacial drainage patterns. The most overlap is between
argyritis and placitus. Hyhognathus nuchaUs is represented south-
westward into the Pecos and Rio Grande drainages by another
nominal form {Hyhognathus amarus Girard) of uncertain rela-
tionship. Some workers have treated this form as a subspecies of
H. placitus, but the shape of the basioccipital process suggests that
its relationships are with H. nuchaUs.

Hyhognathus argyritis Girard — western silvery minnow

HybognatJius nuchaUs: Jordan and Meek, 1885:11 (Hundred and Two R. at
Maryville). Call, 1887:75 (in part (?); "t\vo specimens only"; Piney R.,
Texas Co.; Meramec R., Dent Co.). Fisher, 1962:427 (Missouri R. local.).
Cross, 1967:148 (Missouri R. local, mapped).

Hyhognathus placita: Hanson and Campbell, 1963 (in part (?); linear distr.;
PercheCr.).

Distri])ution and hahitat. — The western silvery minnow is re-
stricted to the Missouri system and the Mississippi River from the
mouth of the Missouri River downstream to about Scott County

Missouri Fishes 369

(Map 81). It is abundant in the Missouri River and some large
prairie streams of the till plains but is rare in the Mississippi River.
In a series of collections made in 1963, it occurred at every locality
studied in the Missouri Ri\er, but it was less abundant than H.
plucitus. In tributaries of the Missouri River from the Grand River
system westward it is generally less abundant than H. placitus, but
it is more abundant than that species in collections from the clearer
streams of central Missouri. The specimen reported by Call (1887:
75) from Big Piney River is probably this species, whereas that
from the Meramec River could be this species or H. nuchalis. The
presence of any species of Hijhognathus at either of the localities
reported by Call is surprising, because they are far removed from
any other Missouri localities for Hijhognathus. The habitat re-
quirements of this minnow are similar to those of the central silvery
minnow, except that it seems more tolerant of high turbidity. It is
generally found over a silt or sand bottom in the quiet backwaters
and pools of large streams, and in the ponded lower reaches of
smaller tributaries.

Hybognathus placitus Girard — plains minnow

Hijhognathus nuchalis placita: Jordan and Meek, 1885:13 (Missouri R. at St.
Joseph ) .

Hijhognathus placita: Hanson and Campbell, 1963 (in part (?); linear distr.;

Perche Cr.). Fisher, 1962:427 (Missouri R. local.).
Hijhognathus placitus: Al-Ra\vi and Cross, 1964 (char.; var.; Missouri local.

mapped). Cross, 1967:146 (Missouri R. local, mapped).

Distribution and habitat. — The distribution of the plains minnow
(Map 82) is similar to that of the western silvery minnow. The
plains minnow is abundant in the Missouri River and in prairie
streams of north Missouri from Grand River westward. Its abun-
dance declines in the Missouri River toward its mouth, and H.
placitus is uncommon in the lower Mississippi River. Like several
other species inhabiting turbid plains streams, H. placitus occurs
downstream in the Mississippi River from the mouth of the Missouri
but not upstream. It is far more abundant than the western silvery
minnow in the Missouri River and most of its tributaries in the
till plains of northwestern Missouri from Grand River westward.
Although the two species occur at the same locality, they tend to
be segregated ecologically; the plains minnow predominates in the
main channel where there is a sandy bottom and some current,
whereas the western silvery minnow is more abundant in the
mouths of tributary streams and other protected backwaters where
there is little or no current and a silt bottom.

370 University of Kansas Publs., Mus. Nat. Hist.

Hybognathus hayi Jordan — cypress minnow

Distribution and habitat. — In the early 1940's the cypress min-
now was common in collections made in the lower Black and St.
Francis rivers, but it has not been taken since anywhere in Missouri
(Map 83). The cypress minnow often occurs with the central sil-
very minnow, and its habitat requirements are probably similar.

Zoogeography. — This southern species may have had its origin
in the lower Mississippi Valley, because it has only a limited distri-
bution in other Gulf drainages.

Pimephales vigilax (Baird and Girard) — bullhead minnow

Ceratichythtjs vigilax: Fowler, 1924:404 (char.; Carthage, Mo.).

Ceratichthtjs perspicuus: Hubbs and Black, 1947:5-31; map 1 (char.; syn.;
Missouri local, mapped ) .

Pimephales perspicuus: Patriarche, 1953:247 (abiind.; Lake Wappapello ) .

Distribution and hahiiat. — The bullhead minnow is widespread
in the lowlands, and occurs northward in the Mississippi River and
its direct tributaries to the Iowa line (Map 84). In the lowlands
it is second in abundance only to Notropis venustus. It is common
in the sluggish prairie streams of northeastern Missouri and in the
upper Mississippi River, but it is scarce in the swifter and more
turbid Mississippi River below the mouth of the Missouri. The
bullhead minnow has not occurred in recent collections from the
Neosho River system, but an old literature record indicates its
former presence there. It still occurs in the Neosho system of Kan-
sas. P. vigilax was reported from South Grand River and Tabo
Creek as Cliola vigilax by Jordan and Meek (1885:16), but the
specimens on which this report is based are Pimephales notatus
(Hubbs and Black, 1947:21). Records for this minnow plotted
by Hubbs and Black (1947: map 1) in the Missouri River system
of Iowa, Nebraska, and South Dakota are based on erroneous
locality data or misidentifications (Bailey and Allum, 1962:109).
The bullhead minnow inhabits sluggish pools and backwaters of
medium-sized to large streams with continuous flow and low to
moderate gradients. It avoids strong current but is rather tolerant
of turbidity and siltation.

Zoogeography. — Hubbs and Black (1947:33) recognized two
subspecies of P. vigilax, with the nominate subspecies occupying
that portion of the species range from the upper Red River south-
westward. These authors suggested that P. v. perspicuus (Girard)
had its origin in the Mississippi Valley and that it has replaced
P. V. vigilax in much of the Red River system and in other coastal
streams of Texas as far west as the Trinity River.

Missouri Fishes 371

Pimephales tenellus (Girard) — slim minnow

Cciuticlitlujs tenellus tenellus: Hubbs and Black, 1947:35-38, map 2 (char.;

syn.; Missouri local, mapped).
Ccmtichthys tenellus pcuvieeps: Hubbs and Black, 1947:36-43, map 2 (orig.

disci-.; t>pe local. Wliite R. 3 mi. S Kissee Mills, Taney Co.; Missouri local.

mapped ) .

Ceratichthys callarchus: Hubbs and Black, 1947:43-47, map 2 (orig. descr.;
only tlie type known; Castor R., Bollinger Co.).

Taxonomic considerations. — Two well marked subspecies of the
slim minnow occur in Missouri. These are P. t. tenellus (Girard)
and P. t. parviceps (Hubbs and Black). Hubbs and Black (1947:
37) regarded populations of P. t. parviceps in the St. Francis and
Black river systems as "nongeographic intcrgrades," because they
are in some respects intermediate between this subspecies and the
nominate subspecies. Hubbs and Black (1947:43-47) described
Ceratichthys callarchus, which they regarded as a species closely
related to P. tenellus, from a single specimen collected in the
Gastor River, Bollinger County, Missouri. A series of 9 specimens
that I collected recently from Castor River differ in no important
respect from P. t. parviceps from the adjacent St. Francis and
Black river systems, thereby suggesting that the type of callarchus
is merely an aberrant specimen of P. tenellus, as suggested by
Moore (1957:136).

Distribution and habitat. — Pimephales t. tenellus is confined to
the Neosho River system, and P. t. parviceps occurs disjunctly in
the upper White, Black, St. Francis, and Gastor rivers (Map 85).
Pimephales t. tenellus is fairly common within its area of occurrence,
but P. t. parviceps is rare. The latter subspecies now may be ab-
sent from the upper White River, since most localities from which
it has been recorded are covered by Bull Shoals and Table Rock
reservoirs.

The streams in which P. t. tenellus occurs are warmer and more
sluggish than typical Ozark streams. Its habitat preferences seem
to be much like those of the bullhead minnow, but it occupies
smaller streams and is less tolerant of turbidity. Pimephales t. parvi-
ceps inhabits clearer and higher gradient streams than P. t. tenellus,
suggesting that the two are differentiated ecologically as well as
morphologically.

Zoogeography. — At the time the two subspecies of P. tenellus
were diverging, P. t. tenellus may have been localized in streams
draining the Flint Hills of Kansas and the western slope of the
Ozark-Ouachita uplands, while P. t. parviceps was localized far-
ther east in the White River and other streams draining the

372 University of Kansas Publs., Mus. Nat. Hist.

southern and southeastern Ozark Uplands. If this theory is cor-
rect, breaching of the Ozark-Ouachita divide by the lower Arkan-
sas River during the Pleistocene (Quinn, 1958) brought the two
subspecies into contact, but their original distributional relation-
ship has remained essentially intact.

Pimephales notatus (Rafinesque) — bluntnose minnow

Pimephales notatus: Jordan and Meek, 1885:11,15,16 (Hundred and Two R.
at Maryville; Flat Cr. near Sedalia and/or Blackwater R. at Brownsville;
Grand R. at Clinton and/or Tebo Cr. at Calhoun). Call, 1887:80 (Bear
Cr., Boone Co.). Meek, 1891:117,121,124,126,129 (Meramec R. and Big
Dry Fork near St. James; Lick Fork near Mansfield; Big Piney R. and
Nordi Fork White R. near Cabool; Niangua R. and Osage Fork near
Marshfield; James R. and Sac R. near Springfield; Shoal Cr. and Hickory
Cr. near Neosho; Maries R. near Dixon; Little Piney Cr. at Newburg).
Evermann and Kendall, 1895:470 (Indian Cr. near Neosho). Fowler,
1921:399 (St. Louis). Martin and Campbell, 1953 (Black R.). Patri-
arche and Campbell, 1958:255-256 (Clearwater Res.). Fisher, 1962:427
(Missouri R. local.). Hanson and Campbell, 1963 (linear distr.; Perche
Cr.).

Cliola vigilax: Jordan and Meek, 1885:16 (Grand R. at Clinton and/or Tebo
Cr. at Calhoun ) .

Hyborhijnclitis notatus: Hubbs and Black, 1947:21,23 (reident. of spec, re-
ported by Jordan and Meek, 1885:16 as Cliola vigilax).

Distribution and Imhitat. — The bluntnose minnow is virtually
statewide in distribution, but it is rare in the northwestern part
of the prairie region (Map 86). Only the ubiquitous green sunfish
has occurred in more collections. P. notatus is common over most
of its Missouri range, but reaches its greatest abundance in and
adjacent to the Ozark border. This minnow occurs in a variety of
habitats but is most numerous in the quiet pools and backwaters
of medium-sized to moderately large streams having clear, warm
waters, permanent flow, and moderate amounts of aquatic vegeta-
tion. In the cooler Ozark streams it is most often found in back-
waters that are a few degrees warmer than the main channel. In
the more turbid and intermittent prairie streams it is largely re-
placed by the fathead minnow, and in the larger rivers and low-
land ditches it is outnumbered by the bullhead minnow.

Zoogeography. — In preglacial time, this wide-ranging species
probably inhabited the preglacial Teays-Mississippi system, and
perhaps also the Laurentian system.

Pimephales promelas Rafinesque — fathead minnow

Pimephales promelas: Agassiz, 1855:221 (smaller brooks west of St. Louis).
Smiley, 1885:347 (salt springs, Saline Co.). Jordan and Meek, 1885:11
(Hundred and Two R. at Maryville). CaH, 1887:80 (Bear Cr., Boone Co.).
Coker, 1930:201 (Mississippi R. near Warsaw, 111.). Fisher, 1962:427
(Missouri R. local.). Hanson and Campl:)ell, 1963 (linear distr.; Perche
Cr.). Cross, 1967:150 (Missouri R. local, mapped).

Missouri Fishes 373

Colisctis parietalis: Cope, 1871:437 (orig. descr.; Missouri R. near St. Joseph).
Pinicpltch's promelas promeJas: Martin and Campliell, 1953:47 (abund.;
Black R.).

Distribution and habitat. — The fathead minnow is abundant
and widespread in the prairie region (Map 87). Although it is
propagated at sexeral hatcheries and is commonly used as bait in
Ozark reser\oirs, the fathead minnow is rare in natural waters of
the Ozarks. Pimephales promelas occurs occasionally in streams
of all sizes but is abundant only in the pools of small, intermittent
headwater creeks. Because of its tolerance of high temperature,
extreme turbidity, and low oxygen, the fathead minnow is well
suited for survival in the stagnant pools that provide the only
refuge for fish in many small prairie streams during extended dry
periods. In such situations the fathead minnow, along with a few
other hardy species such as the creek chub, black bullhead, and
green sunfish, usually comprise the entire fish population. The
fathead minnow seems intolerant of competition and is seldom
abundant in habitats that support a variety of other fishes.

Zoogeographij. — This wide-ranging species exhibits a trend
from north to southwest in several characters, and Metcalf (1966:
136-137) concluded that these trends could be environmentally
induced, or could have arisen in ancestral stocks that were present
in all three of the principal drainages that were integrated to form
the present Mississippi River system.

Campostoma oligolepis Hubbs and Greene — largescale stoneroller
Taxonomic considerations. — Most workers recognize three sub-
species of the wide-ranging Campostoma anomaltim. These are:
C. a. anomaltim (Rafinesque) of the Ohio River system; C. a. oli-
golepis Hubbs and Greene, occurring disjunctly in the Driftless
Area of Wisconsin and Ozark Uplands; and the wide-ranging C. a.
pulliim (Agassiz). A fourth subspecies, C. a. plumbeum (Girard)
of the western plains is recognized by some workers. Subspecific
allocation has been made principally on the basis of the number
of scales, supplemented by difi^erences in morphometries and colora-
tion.

Two of these forms (ptdlum and oligolepis) occur in Missouri,
and their taxonomic relationship is the subject of the present discus-
sion. Hubbs and Greene (1935) considered pulltim and oligolepis
to be conspecific, although they found little evidence of intergrada-
tion where the two forms occur together in Wisconsin. Their de-
cision was based on the fact that subspecies ano7nalum is inter-
mediate in certain characters (most notably scutellation) between

374 University of Kansas Publs., Mus. Nat. Hist.

ptiUimi and oligolepis. Known at the time of its original description
only from Wisconsin, oligolepis has since been found to be wide-
spread in the Ozark Uplands. Here as in Wisconsin, oligolepis re-
mains distinct although it is sympatric with puUum. In Wisconsin
there is only moderate overlap in the ranges of the two forms; in
the Ozark Uplands oligolepis is wholly sympatric with pulliim.
Bailey (1956:334) pointed out the inconsistency of treating sym-
patric forms as subspecies, and suggested that oligolepis is either
a full species or an environmental variant. He favored the latter
explanation, pointing out that oligolepis usually is found in regions
having numerous springs, and that fishes spawned in the relatively
uniform temperatures of streams emerging from large springs
might be structurally different from those exposed to wider fluctua-
tions in environmental conditions.

Scale counts of the two forms in Missouri are presented and
compared with Wisconsin populations in Table 5. Counts for
oligolepis from the northern and southeastern Ozarks (Osage, Gas-
conade, Meramec and Headwater Diversion systems) are similar
to those of Wisconsin populations, but are consistently lower than
those for populations from the southern Ozarks (White, Black, and
St. Francis systems). Variation is not clinal; scale counts for the
Headwater Diversion are as low as those for drainages in the
northern Ozarks, in spite of the proximity of populations with
high scale counts in the adjacent St. Francis system. Scale counts
for pulltim vary from stream system to stream system with no
definite pattern. The number of circumferential scales varies more
than the number of lateral line scales, and variation of the two
counts is discordant. Counts for Wisconsin pulhim are comparable
with those of Missouri populations. The relatively high number of
scales in oligolepis in the southern Ozarks could result from present
or past gene exchange with piilhim, but this seems unlikely, be-
cause there is no increased variability (as indicated by the co-
efficient of variation) in scale counts of populations from that
region.

In other characters Missouri populations of pulltim and oli-
golepis agree rather well with the list of differential characters
given by Hubbs and Greene (1935: table I). Hubbs and Greene
did not note differences between breeding males of the two forms,
but Missouri specimens differ consistently in tubercle pattern and
fin color. Males of piillum typically have a crescent-shaped row of
1-3 tubercles mesial to each nostril and a prominent black band in

Missouri Fishes

375

>■

S U

rt .

• C/3

+ C

^- IX

C/3

2 IX

•z

•c

3
O

CO
(V)

(-«

-Iq

c^^

M-H

l§i

^ 5

4-'

JIX

•43

CL,

CO U

4^^

"Sq

llJ

C/D

,— ^

c So

<4-»

(L> r-

3
O

■— 3

-2

.^\x

CL,

COCO
coco

CD 00
ICi— 1

OCO

CD M
CD CM

cqin

OCO
00l>

O5 00

coco

qoq

cooq

coco

CO (N

COCO

C0(>1

cocq

'^ (N

OCD

1— 1 C5

locq

CM CD

00^

CDi-i

)0 CD

coin

^00

CDoq

coco

<0(M

CO CM

CO -H

COCsi

CO (n'

co'c<i

c<ioi

cooq

cooi

cocq

00 05
O5 0O

OOCTi

c:5oo

1— 1

o ^
1—1 ro

C35CD
05 00

05CD
O5 00

CO
oin

i-H 00

0-*

.-^00

Oin

.-HOO

in OS

00 t^

ini>

00 l>

00 00

in in
00 1>

in4

00 1>

00 CD
00 I>

1 1

00 t^

CDcq
00 t>

TJ<0

iqrf

C<lTf<

in CD

COOT

oin

oqin

inrf

O5 00

CJJOO

^H OJ

C:5I>

.-HOO

05t-

COO

O5 00

0105
OiI>

-^'od

05t>

COO

CD 00

OCO

t 00

int-

ino

CD CO

COCTs

coin

min

in CO
coco

>— 1 (M

in CO
t-o

in t— 1

O5 00

CO CD
00 t^

Cni05

1 — 1 r— (

CTiCD
'^ (M

05c::5

CT50

coco

M^'co

rf Csi

-^'cO

-:fCO

^'co

coco

00 '^
00 CO

cool

-f CO

t- 05

-f CO

00 .-1

Tt<00
COrf

OOO

pco

1— ( i-H

c<ii— i

Oi r-H

(M" ^'

oii-H

c<i--(

oq—H

'^O

in in

ino
in in

in in

t-c:5

in ~v

int-

in'^

00 00

in-t^

in 00
in-h

co4

4co

CD in

in -H

CO ^

CDr^

C<IC<I

00 Tf

COf

t--cq

I-HTJH

OCD

oV^

in^f

00 CD

oV;

in rf

oi^

<Nin

inTjH

CDTt<

in'^

coo

CD 00

OCO

■^ 00

rf r-H

ini-

in o

coco

C0 05

CD in

in in

OOO
00 05

coin
CO 00

ino5

CO-*
,-hO

(MOl

t-co
in CO

1— 1 1— 1

(MOO

^'co

in'^f

in CO

in'*

T^'rjJ

■>*TP

in CO

COCO

o_Tr

05in
oqi>

OOO
OtP

(MCD

1— ICO
pin

OCO
(Mp

(M'l-i

CQ r-i

CN't-H

cq^'

i-i i—i

oi'-H

cq 1— I

t^O

t-o

C^O

in CO

Tt<co

CD 05
Tf< CO

O5 00
-*C0

CO 00
M<CO

00 CO

coco

00 cq

COCO

coco

00 cq

coco

CJcq

coco

1 1

00 (M

COCO

oqoo

t-o

C0 05

in CD

pco

OJp

(NO)

c<icD

'^co

c4cD

TfCO

coco

'^CO

^ CO

.-HTf

'^CO

COTtn"
TPCO

^co

t^o

CD 00

-HCO

"-r 00
i>i>

int>

ino
coco

C0 05

coin

^CO

— lO

^'co

mt>

I-H CO

cq'i-H

00 05

poi

cQin

m^f

t^ 05

05Tfi
^ (M

mco

CO 00
CO-*

1—1 oq

00 p

OOO

mco

OOC35
CO<M

COCO

(NCO

T}<CO

00 00

com

--KM

t_,

•2

3
O

•2

5 ii

Co •-.
Co (-"

§1

a o

«*

.:s

a o

C/D ffi

a o

O 03

Si. U
? <5i >

a. o

a.
to

'■-^
c

~ O

~ OC

3;s

a o

3
CD
<U

3
03

>.
^

3
3
O
CJ

-M

3
ft

•43
03

3
O
O

376 University of Kansas Publs., Mus. Nat. Hist.

the anal fin; in males of oligolepis these characters are seldom
developed.

I doubt that oligolepis is an environmental variant as proposed
by Bailey. Largescale stonerollers occur in Missouri at localities
so remote from springs that dispersal there after hatching is un-
likely. The differences between oligolepis and pnlliim are slight,
but involve so many characters that they could hardly all be the
result of differing conditions during development (in particular,
I doubt that conditions during embryonic development determine
the distinctive characters of breeding males). I have stripped and
artificially fertilized eggs of oligolepis, hatched them in the labora-
tory, and reared the offspring in ponds. Scale counts of pond-reared
oligolepis were within the range of those for naturally occurring
populations.

The widespread sympatric occurrence of oligolepis and puUum
with no indication of intergradation leaves little doubt that they
are reproductively isolated. If it were not for the supposed poten-
tial for gene exchange between oligolepis and anomalum, there
would be no basis for treating oligolepis and ptillum as anything
but distinct species. Evidence of reproductive compatability be-
tween oligolepis and anomalum is lacking, because the two forms
are allopatric. In the absence of such evidence it seems more
realistic to recognize reproductive isolation between oligolepis and
pullum by treating them as different species than to recognize an
undemonstrated potential for gene exchange between anomalum
and oligolepis by treating them as conspecific.

Distribution and habitat. — The largescale stoneroller occurs
throughout the Ozarks except for the Neosho River system (Map
88). The habitats of the largescale and central stonerollers seem
to be similar, except for the preference of the latter species for
smaller streams. The largescale stoneroller is the more abundant
of the two species in most large Ozark streams, but often there is
a shift in abundance in favor of the central stoneroller toward the
headwaters. I have found no localities where the largescale stone-
roller occurred to the exclusion of the central stoneroller.

Zoogeography. — The present distribution of the largescale
stoneroller suggests that it inhabited the Mississippi Valley in pre-
glacial time and has long inhabited the Ozark Uplands. Perhaps
the disjunct occurrence of C. oligolepis in the Ozark Uplands and
the Driftless Area of Wisconsin resulted from the fragmentation of
a widespread preglacial distribution. The lack of strong dilferen-

Missouri Fishes 377

tiation between populations of the Driftless Area and the northern
Ozark Uplands suggests that this range disjunction is quite recent.

Campostoma anomalum pullum (Agassiz) — central stoneroller

Cmnpostoma anomalum: Jordan and Meek, 1885:15,16 (Flat Cr. near Sedalia
and/or Blackwater R. at Brownsville; Grand R. at Clinton and/or Tebo
Cr. at Calhoun). Call, 1887:74,75,79 (West Fork Black R., Reynolds Co.;
Tacks Fork, Shannon Co.; Piney R., Texas Co.; Meramec R., Dent Co.;
Bear Cr. and Hinkson Cr., Boone Co.). Meek, 1891:117,121,124,126,129
Meramec R., Meramec Spring and Big Dry Fork near St. James; Little Dry
Fork near Rolla; Jones Cr. near Dixon; Big Piney R. near Cabool; Lick Fork
at Mansfield; Osage Fork SE Marshfield; Niangiia R. near Marshfield; Sac
R. and James R. near Springfield; Maries R. near Dixon; Shoal Cr. near
Neo.sho; North Fork White R. S Cabool). Evemiann and Kendall, 1895:
470 (Indian Cr. and spring branch near Neosho). Fowler, 1921:398 (St.
Louis). Hubl^s and Ortenburger, 1929:95 (creeks 2 mi. W Shepard, 2 mi.
S DeSoto, and 4 mi. S Potosi). Martin and Campbell, 1953 (abund.; hab.;
Black R.). Fisher, 1962:428 (Missouri R. local.). Hanson and Campbell,
1963 (linear distr.; Perche Cr.).

Distribution and habitat. — The central stoneroller occurs in all
of Missouri except for the lowlands and the extreme northwestern
part of the prairie region (Map 89). It is most abundant in the
Ozarks and northern Ozark border. It is generally less abundant
than the largescale stoneroller in the Ozarks but in some places
occurs to the exclusion of that species in small headwater streams.
The central stoneroller inhabits the rocky pools and riffles of clear,
permanent-flowing streams with moderate or high gradients. It is
tolerant of moderate turbidity, if there is sufficient current to keep
the bottom mostly free of silt.

Previous workers did not distinguish the central and largescale
stonerollers, and some published records cited above may apply to
either or both forms.

Zoogeography. — Metcalf (1966:139) suggested that C. a. ptdhun
had a preglacial origin west, south or southwest of the central
Teays-Mississippi system and cited the occurrence of the small-
scaled Campostoma ornatum Girard in southwestern drainages as
proof that fine-scaled stonerollers have long inhabited that region.
However, Ross (1958:18) suggested that occurrence of pxdhim
in the upper Allegheny River, a former tributary of the preglacial
Laurentian system, may date from pre-Pleistocene times. If this
is correct, pidhim must have had an eastern or a mixed origin. The
presence of an east-west cline in scale size (Metcalf 1966:138) may
have resulted from intergradation between partially differentiated
forms that were localized preglacially in the central Mississippi
Valley and in plains drainages that were later diverted into the
central Mississippi Valley.

378 University of Kansas Publs., Mus. Nat. Hist.

Catostomidae
Cycleptus elongatus (Lesueur) — blue sucker

Ctjcleptus elongatus: Forbes and Richardson, 1920:65 (Mississippi R. at Cairo
and Grafton, III). Bamickol and Starrett, 1951:292-293 (abund.; utiliz.;
Mississippi R. local.). Fisher, 1962:428 (Missouri R. local.). Cross, 1967:
166 (Missouri R. local, mapped).

Distribution and habitat. — The blue sucker is scarce but widely
distributed in the Missouri and Mississippi rivers and their largest
tributaries (Map 90). It seems to be less abundant now than in
the early 1900's. Probably it is more generally distributed than our
records indicate, because it is difficult to capture with the gear
used in making most of the collections. This sucker inhabits deep,
swift channels. It is tolerant of turbidity if current is sufficient to
prevent deposition of silt on the firm sand, gravel, and rubble bot-
toms over which the blue sucker is usually found. Construction of
dams, with the attendant decrease in current velocity and increase
in siltation, is unfavorable to the blue sucker. Coker, 1930:183
noted the decline of C. elongatus in the upper Mississippi River
following the construction of a dam at Keokuk, Iowa.

Zoogeographtj. — Perhaps C. elongatus has inhabited the Mis-
sissippi Valley continuously since preglacial time. It is also possible
that it entered the Mississippi Valley during the Pleistocene, when
the lowering of sea levels that accompanied glaciation may have
created connections with Gulf coastal drainages to the west.

Ictiobus cyprinellus (Valenciennes) — bigmouth buffalo

Ichthyobus stolleyi: Agassiz, 1855:81 (orig. descr.; Osage R.).

Ictiobus cyprinella: Carman, 1890:145 (Mississippi R. near Quincy, 111.).
Jordan and Meek, 1885:13 (Missouri R. at St. Joseph). Meek, 1891:124
(Maries R. near Dixon). Forbes and Richardson, 1920:map IX (Missis-
sippi R. local.). Cleary, 1956:map 17 (Des Moines R. local.). Bemer,
1951:9,10 (commer. catch; food; Missouri and Mississippi R.).

Megastomatobus cyprinella: Bamickol and Starrett, 1951:293-298 (abund.;
utiliz.; Mississippi R. local.). Patriarche, 1953 (alwnd.; Lake Wappapello ) .

Ictiobus cyprinellus: Funk and Campbell, 1953 (abund.; Black R. local.).
Fisher, 1962:428 (Missouri R. local.). Hanson and Campbell, 1963 (linear
distr.; Perche Cr. ). Cross, 1967:169 (Missouri R. local, mapped).

Distribution and habitat. — The bigmouth buffalo occurs in large
streams and reservoirs over most of the state, but it is most abun-
dant in the Missouri and Mississippi rivers and the larger streams
of the till plains (Map 91). In the Missouri River and its major
tributaries the bigmouth far outnumbers the other two species of
buffalofishes. In collections made by Fisher (1962: table 1), from
the Missouri River, /. cyprinellus outnumbered the smallmouth
buffalo about three to one. Long-term population inventories con-

Missouri Fishes 379

ducted by the Missouri Department of Conservation reveal that the
bigmouth buffalo is the prevalent buffalofish in the Grand River
watershed. Similar inventories in Salt River (tributary of the upper
Mississippi River) indicate that the bigmouth buffalo is slightly
more abundant than the black buffalo there but is far less abundant
than the smallmouth buffalo. In the Mississippi River Barnickol
and Starrett (1951:294-296) ranked the three species of buffalo-
fishes in the same order as indicated above for the Salt River; this
has also been the case in our collections.

The habitat requirements of the three species of buffalofishes
are similar, but their distributional relationships in Missouri and
elsewhere suggest that the bigmouth buffalo is more tolerant of
high turbidity than the other species. Ictiohus cyprineUtis is pri-
marily an inhabitant of quiet waters, especially the pools of large
streams, natural lowland lakes, and impoundments. The young
are sometimes found in rather small creeks where these flow di-
rectly into large rivers.

Zoogeography. — The preglacial distributional relationships of
the buffalofishes probably have been profoundly altered by events
during the Pleistocene, making their distributional history difficult
to discern. Large rivers are ready avenues for dispersal by these
species; they probably were quick to take advantage of the stream
connections made available to them by glaciation. The bigmouth
buffalofish is less widespread in Gulf coastal drainages and occurs
farther north than the other buffalofishes. Perhaps it invaded the
Mississippi Valley from northern drainages during the Pleistocene,
and has had less time to disperse along the Gulf coast.

Ictiobus niger (Rafinesque) — black buffalo

BulmJichihys honastis: Agassiz, 1855:80 (orig. descr.; Osage R.).

Ictiohus urtis: Jordan and Meek, 1885:13 (Missouri R. at St. Joseph). Gar-
man, 1890:145 (Mississippi R. near Quincy, 111.). Forbes and Richardson,
1920:map X (Mississippi R. local.).

Ictiohus niger: Borges, 1950 (vert, distr.; Niangiia Arm, Lake Ozark). Bar-
nickol and Starrett, 1951:293-298 (abund.; utiliz.; Mississippi R. local.).
Bemer, 1951: table 5 (commer. catch; Missouri and Mississippi R.). Funk
and Campbell, 1953 (abund.; Black R. local.). Patriarche, 1953 (abund.;
Lake Wappapello ) . Fisher, 1962:428 (Missouri R. local.). Hanson and
Campbell, 1963 (linear distr.; Perche Cr.). Cross, 1967:171 (Missouri R.
local, mapped).

Distrihtifion and habitat. — This species is less common in Mis-
souri than the other buffalofishes (Map 92). In our collections the
black buffalo outnumbered the other two species only in the lower
Current River. It is slightly more abundant than the bigmouth
buffalo in the streams and ditches of the lowlands but is less abun-

380 University of Kansas Publs., Mus. Nat. Hist.

dant there than the smalhnouth buffalo. In tlie Mississippi River
it is more abundant in tlie unimpounded section below the mouth
of the Missouri River than in the impounded section above (Bar-
nickol and Starrett, 1951:294). The habitat requirements of /. niger
are much like those of the other buffalofishes, but it more often
occurs in strong current ( Cross, 1967: 172) .

Ictiobus bubalus (Rafinesque) — smalhnouth buffalo

Ictiobiis huhalus: Jordan and Meek, 1885:13 (Missouri R. at St. Joseph ).
Garman, 1890:144-145 (Mississippi R. near Quincy, 111.). Meek, 1891:124
(Maries R. near Dixon). Forbes and Richardson, 1920:niap XI (Mississippi
R. locaL). Borges, 1950 (vert, distr.; Niangua Arm, Lake Ozark). Bar-
nickol and Starrett, 1951:292-293 (abund.; iitiliz.; Mississippi R. local.).
Berner, 1951:9-10 (conimer. catch; food; Missouri and Mississippi R.).
Martin and Campbell, 1953:46 (abund.; Black R.). Funk and Campbell,
1953 (abund.; Black R. local.). Patriarche, 1953 (abund.; Lake Wappa-
pello). Purkett, 1958a: 123-124 (growth; Salt. R.). Fisher, 1962:428 (Mis-
souri R. local.). Hanson and Campbell, 1963 (linear distr.; Perche Cr. ).
Cross, 1967:173 (Missouri R. local, mapped).

Distribution and habitat. — The smallmouth buffalo is nearly as
widespread in Missouri as the bigmouth buffalo (Map 93). It is
the most abundant buffalofish in the Mississippi River and its major
tributaries in northeast Missouri, in the lowland ditches of the
southeast, and in large reservoirs of the Ozarks. Elsewhere in the
state it is usually less prevalent than the bigmouth buffalo but is
more abundant than the black buffalo. Ictiobus bubalus seems to
require clearer waters than 7. cyprineUus, and it is less often found
in strong currents than I. niger.

Zoogeographij. — The smallmouth and black buffalofishes are
more southern and southwestern in distribution than the bigmouth
buffalo. It seems likely that they inhabited the preglacial Teays-
Mississippi system and perhaps also other Gulf coastal drainages
to the west. C. L. Smith (1962:509) tentatively identified I. bubalus
in Pliocene fossil deposits from Oklahoma and suggested that this
species has long inhabited the Mississippi Valley.

Carpiodes cyprinus (Lesueur) — quillback

(?) Ictiobus velifer: Jordan and Meek, 1885:11,15,16 (Hundred and Two R.
at Maryville; Flat Cr. and/or Blackvvater R. at Brownsville; Grand R. at
Clinton and/or Tebo Cr. at Calhoim ) .

Carpiodes cyprinus hcinei: Trautman, 1956 (orig. descr.; Missouri local,
mapped ) .

Carpiodes cyprinus: Bamickol and Starrett, 1951:292-293 (abimd.; utiliz.;
Mississippi R.). Hanson and Campbell, 1963 (linear distr.; Perche Cr. ).

Carpiodes forhesi: Trautman, 1956 (Missouri local, mapped). Hanson and
Campbell, 1963 (linear distr.; Perche Cr. ).

Quillback: Fiy, 1962 (han^.; Taneycomo Res. tailwater).

Missouri Fishes 381

Taxonomic coiisideratiotis. — Until recently many workers have
recognized Carpiodes forhcsi Hubbs as a distinct species closely
allied to C. cyprinus. However, Bailey and Allum (1962:81) placed
C. for])esi in the synonymy of C. cyprinus, suggesting that "the
slender fish with a low dorsal fin are likely the product of their
en^'ironment." This conclusion was based on the assumption that
the C. forbesi phcnotype "is found chiefly in prairie and plains areas
where high turbidity and scanty food supplies in the rivers are
characteristic" (Bailey and Allum, op. cit.). I do not agree with
this assumption, because I have encountered the C. forbesi pheno-
type in the Ozark Uplands of Missouri (Big Piney River, Texas
Count)^ and Lake Taneycomo, Taney County), as well as in the
prairie regions. Furthermore, the conclusion that prairie and plains
streams are unproductive is unwarranted, because these streams
drain areas with highly productive soils and support large popula-
tions of other fishes that do not differ phenotypically from popula-
tions in streams of other areas. However, I concur that the evidence
is insufficient to retain C. forbesi as a valid species. Among Missouri
specimens I have noted all degrees of variation in development
of the anterior dorsal rays with no apparent bimodality that would
permit the recognition of two types. Furthermore, there seems to
be no obvious correlation between body depth and development
of the anterior dorsal ray in Missouri specimens. It is possible that
C. forbesi and C. cyprinus are sibling species, but if this is the case,
additional characters will have to be found if they are to be recog-
nized.

Distribution and Jmbitat. — The quillback is most abundant in
the Ozark border and in the clearer prairie streams of central and
northeastern Missouri (Map 94). It is fairly common in the upper
Mississippi River, but it is rare elsewhere in the Mississippi River
and in the Missouri River. The quillback is taken frequently in the
large reservoirs of the central Ozarks, but it is seldom encountered
in the streams of that region. Carpiodes cyprinus is characteristic
of moderately clear streams having low or moderate gradients, well
defined pools and riffles, and stable bottoms of gravel and other
coarse material. Like other carpsuckers, the quillback inhabits
quiet water, except when spawning. It is most abundant in mod-
erately large streams but also occurs in creeks if large, permanent
pools are present.

Zoogeography. — Until the C. cyprinus complex has been sub-
jected to a thorough variational analysis, and the basis for this
variation has been established, little can be said concerning its

382 University of Kansas Publs., Mus. Nat. Hist,

zoogeographic histoiy. The widespread distribution of C. cyprimis
suggests that it could have been present preglacially in all the
principal drainages that were integrated to form the present Mis-
sissippi River system, with the possible exception of the ancestral
Hudson Bay system.

Carpiodes carpio (Rafinesque) — river carpsucker

Carpiodes bison: Agassiz, 1854:356 (orig. descr.; type local. Osage R., Mo.).
Ictiobus carpio: Jordan and Meek, 1885:13 (Missouri R. at St. Joseph).

Carpiodes carpio: Forbes and Richardson, 1920:map XII (Missisippi R. local.).

Funk and Campbell, 1953 (abund.; Black R. local.). Purkett, 1958a: 121

(growth; Salt R.). Hanson and Campbell, 1963 (linear distr.; Perche Cr. ).

Cross, 1967:177 (Missouri R. local, mapped).
Carpiodes carpio carpio: Bamickol and Starrett, 1951:292-293 (abund.; iitiliz.;

Mississippi R. local.). Martin and Campbell, 1953:46 (abund.; Black R.).

Brezner, 1958 (food habits; Lake Ozark).
River carpsucker: Purkett, 1958b: 5,23 (growth; Missouri streams).

Distribution and habitat. — The river carpsucker is by far the
most abundant and widely distributed carpsucker in Missouri ( Map
95). In the larger prairie streams of north and west Missouri it is
the must abundant large fish at many localities or is second in
abundance only to the carp or gizzard shad. It rarely occurs in the
lowland and Ozark regions. The preferred habitat of the river
carpsucker is the quiet, silt-bottomed pools, backwaters, and ox-
bows of large streams having moderate or low gradients. It usually
thrives in impoundments. Carpiodes carpio is tolerant of high tur-
bidity and is replaced in clearer waters by the quillback or highfin
carpsucker.

Zoogeography. — Carpiodes carpio may have inhabited the
Teays-Mississippi system, as well as other preglacial drainages to
the southwest. The presence of a distinct subspecies (C. c. elonga-
tus Meek) in Gulf coastal drainages from eastern Texas to north-
western Me.xico suggests that ancestral stocks of this species have
long been present in southwestern drainages.

Carpiodes velifer (Rafinesque) — highfin carpsucker

Carpiodes difformis: Forbes and Richardson, 1920:map XIII (Mississippi R.
local. ) .

Carpiodes velifer: Barnickol and Starrett, 1951:292-293 (abund.; utiliz.; Mis-
sissippi R.). Cleary, 1956:map 23 (Des Moines R. local.).

Distribution and habitat. — The highfin carpsucker is rare in
Missouri. At present it seems to be confined to the Ozarks, where
it is more common in large reservoirs than in streams (Map 96).
Formerly it may have been more widespread and abundant in the
state. Cross (1967:180-181) noted a marked decline in abundance

Missouri Fishes 383

of the highfin carpsucker in Kansas. The habitat requirements of
the highfin carpsucker are similar to those of the river carpsucker,
but the highfin is generally found in clearer waters and over firmer
bottoms. It is much less tolerant of high turbidity and siltation than
other carpsuckers, and this intolerance probably explains its re-
stricted distribution in Missouri.

Zoogeography. — The present distribution of C. velifer suggests
that it was present in the preglacial Teays-Mississippi system ( Met-
calf, 1966:142).

Catostomus commersoni (Lacepede) — white sucker

Catostomus teres: Jordan and Meek, 1885:11,15,16 (Hundred and Two R. at
Mary\ille; Flat Cr. near Sedalia and/or Blackwater R. at Brownsville;
Grand R. at Clinton and/or Tebo Cr. at Calhoun). Call, 1887:74,79 (Jacks
Fork, Shannon Co.; Toms Cr., Reynolds Co.; Hinkson Cr., Boone Co.).
Meek, 1891:117,121,124,126 (Meramec Spring and Meramec R. near St.
James; Little Dry Fork near Rolla; Big Piney R. near Cabool; Niangua R.
near Marshfield; Maries R. near Dixon; Shoal Cr. near Neosho). Evermann
and Kendall, 1895:470 (Indian Cr. and spring branch near Neosho).

Catostomus commersoni: Purkett, 1958a:125 (growth; Salt R.). Fisher, 1962:
428 (Missouri R. local.). Hanson and Campbell, 1963 (linear distr.;
Perche Cr. ) .

Distrihuiion and habitat. — The white sucker is common in the
clearer prairie streams of central and northeastern Missouri but
becomes increasingly rare toward the northwest (Map 97). It is
locally common but spotty in distribution in the Ozarks and is ab-
sent from the lowlands. The white sucker is decidedly a small-
creek fish, occurring only rarely in the main channels of large
rivers. In the Ozarks C. commersoni is generally found in spring
branches and in clear, heavily vegetated spring pools along large
streams. In the prairie region it is most abundant in high-gradient
headwater creeks having gravelly or rocky bottoms and well de-
fined riffles. These creeks tend to be intermittent, but permanent
pools are maintained by seepage through gravel of the stream bed.
The habitats in which the white sucker is abundant are largely
devoid of other suckers, suggesting that competition with these
may be a factor in limiting its distribution. Its requirements are
most like those of the creek chub, a minnow with which it is com-
monly associated.

Zoogeography. — Preglacially this widespread northern species
may have been confined to drainages north of the Mississippi
Valley. However, Metcalf (1966:144-147) discussed a trend in eye
size and number of rows of lip papillae from east to west and in
scale size from north to south in western populations of C, com-
mersoni; he suggested that this fish was represented preglacially

384 University of Kansas Publs., Mus. Nat. Hist.

by three geographically separated and morphologically different
stocks that occupied the Hudson Bay, Teays-Mississippi and "An-
cestral Plains" systems. Mixing of these three stocks as a result
of drainage alterations and distributional changes during the Pleisto-
cene presumably resulted in the present variational patterns.

Hypentelium nigricans (Lesueur) — northern hog sucker

Catostomus nigricans: Agassiz, 1855:77,92 (Osage R., Missouri). Meek, 1891:
117,121,124,126,129 (Meramec R., Meramec Spring and Big Dry Fork
near St. James; Little Piney Cr. at Newburg and Arlington; Osage Fork SE
Marshfield; Lick Fork at XIansfield; Jones Cr. and Maries R. near Dixon;
Niangua R. near Marshfield; Sac R. and James R. near Springfield; Shoal
Cr. near Neosho; North Fork White R. S Cabool). Evermann and Kendall,
1895:470 (Indian Cr. and Shoal Cr. near Neosho).

Hypentelium nigricans: Call, 1887:74 (Toms Cr., Reynolds Co.; Meramec R.,
Dent Co.). Coker, 1930:186 (Mississippi R. near Warsaw, 111.). Martin
and Campbell, 1953 (abund.; hab.; Black R.). Fvmk and Campbell, 1953
(abund.; Black R. local.). Metcalf, 1966:163 (Richland Cr., Morgan Co.).

Hog sucker: Purkett, 1958b:6,24,43 (growth; length-weight relat.; Missouri
streams ) .

Distribution and habitat. — The hog sucker is one of the most
abundant stream fishes in the Ozarks (Map 98). Among suckers
of that region it is exceeded in abundance only by the black and
golden redhorses. Its distribution in Missouri is remarkably similar
to that of the black redhorse. Hypentelium nigricans is an inhabi-
tant of clear streams with clean gravelly or rocky bottoms and per-
manent flow. It is most often found in or near riffles, or in pools
where there is noticeable current.

Zoogeography. — Hypentelium nigricans probably was an in-
habitant of the preglacial Teays-Mississippi system and has long
inhabited the Ozark Uplands. Its only close relatives are H. roano-
kense Raney and Lachner from the Roanoke River in Virginia and
H. etowanum (Jordan) from the Alabama River system.

Moxostoma duquesnei (Lesueur) — black redhorse

Moxostoma duquesnei: (?) Meek, 1891:117,121,126,129 (Meramec Spring,
Meramec R. and Big Dry Fork near St. James; Little Dry Fork near Rolla;
Osage Fork SE Marshfield; Lick Fork at Mansfield; Little Piney Cr. at New-
burg; Gasconade R. at Arlington; Shoal Cr. near Neosho; James R. near
Springfield; North Fork White R. S Cabool). (?) Evermann and Kendall,
1895:470 (Indian Cr. near Neosho). Patriarche, 1953 (abund.; Lake Wap-
papello). Funk, 1957 (mov.; Missouri streams). Patriarche and Campbell,
1958:251 (abund.; growth; Clearwater Res. ) .

Moxostoma duquesnei duquesnei: Martin and Campbell, 1953 (abund.; hab.;
Black R.).

Black redhorse: Purkett, 1958b:7,25,43 (growth; length-weight relat.; Missouri
streams ) .

Distribution and liabitat. — The black redhorse is restricted in
Missouri to the Ozark Uplands (Map 99). In many Ozark streams

Missouri Fishes 385

the total poundage of this fish and the golden redhoise equals that
of all other large fishes eonibined. The species of Moxostoma were
confused by workers prior to 1900, and literature records for that
period ha\e been plotted only where specimens were examined.
The black rcdhorse inhabits clear streams having permanent
flow and clean gravelly or rocky bottoms. It is most abundant in
streams of medium size, being replaced by the white sucker in
headwater creeks and spring branches and by the northern redhorse
and golden redhorse in the larger rivers. Generally it is more abun-
dant than the golden redhorse in the cooler and swifter streams.
Where the two occur together, the black redhorse tends to occupy
short, rocky pools with considerable current, whereas the golden
redhorse is most abundant in larger pools and backwaters with
little or no current.

Zoogeography. — Moxostoma duquesnei probably inhabited the
preglacial Teays-Mississippi system. It may have survived the
Wisconsin ice advance in both the Ozark and Appalachian uplands
and possibly also in the Driftless Area. Moxostoma duquesnei is
known from late-Illinoian fossil deposits in southwestern Kansas
(G. R. Smith, 1963:281), indicating a more western distribution in
the past.

Moxostoma erythrurum (Rafinesque) — golden redhorse

Ptychostomus bucco: Cope, 1871:437 (orig. descr.; St. Joseph, Mo.).

Moxostoma macrolepidotum duquesnei: (?) Jordan and Meek, 1885:15 (Flat
Cr. near Sedalia and/or Blackwater R. at Brownsville). (?) Meek, 1891
(Niangiia R. near Marshfield; Maries R. near Dixon).

Moxostoma macrolepidotum: (?) Call, 1887:74,80 (West Fork Black R., Rey-
nolds Co.; Jacks Fork, Shannon Co.; Meramec R., Dent Co.; Big Cr. and
Piney R., Texas Co.; Hinkson Cr., Boone Co.).

Moxostoma erythrurum: Martin and Campbell, 1953 (abund.; hab.; Black R.).
Patriarche, 1953 (alnmd.; Lake Wappapello). Funk, 1957 (mov.; Missouri
streams). Patriarche and Campbell, 1958:251 (abund.; growth; Clear-
water Res.). Fisher, 1962:428 (Missouri R. local.). Hanson and Campbell,
1963 (hnear distr.; Perche Cr. ).

Golden Redhorse: Purkett, 1958b:8,26,43 (growth; length-weight relat.; Mis-
souri streams ) .

Distribution and habitat. — The golden redhorse occurs through-
out the Ozarks and northeastern Ozark border (Map 100). It is
rare in the western part of the prairie region, where it may have
declined in abundance since settlement. It is the most abundant
redhorse in many Missouri streams. The habitat requirements of
this fish are similar to those of the black redhorse, except that M.
erythrurum prefers slightly warmer waters and less current, and is
more tolerant of turbidity and intermittent flow. It reaches its

386 University of Kansas Publs., Mus. Nat. Hist.

greatest abundance in moderately clear, unpolluted streams having
large, permanent pools and well defined gravelly or rocky riffles.

Zoogeography. — Except for a more widespread occurrence west
of the Mississippi River, M. erytlirurtim is distributed much like
M. dtiquesnei, and the two probably had a similar distributional
history during the Pleistocene.

Moxostoma anisurum (Rafinesque) — silver redhorse

Catostomtis anisiinis: Agassiz, 1854:87 (St. Louis).
Silver redhorse: Purkett, 1958b:27 (growth; Missouri streams).

Distribution and habitat. — The silver redhorse is widespread
but seldom abundant in the northern and eastern Ozarks (Map
101). I have found this species to be most abundant in the Salt
River of northeastern Missouri. There it is the most abundant red-
horse. It may also be abundant in other streams of the northeastern
Ozark border, but most of these have not been sampled with the
kind of collecting gear that is effective for sampling redhorse. The
silver redhorse inhabits the larger and deeper pools of medium
to large streams having moderately clear water, rocky or gravelly
bottoms, and permanent flow. It avoids swift-flowing streams hav-
ing high gradients and those that are excessively turbid.

Zoogeography. — Moxostoma anisurum is northern and eastern
in distribution. Perhaps it was localized in the Hudson Bay or
Laurentian systems preglacially, and entered the Mississippi Valley
by way of stream connections that developed during the Pleisto-
cene.

Moxostoma macrolepidotum (Lesueur) — northern redhorse

Moxostoma breviceps: Forbes and Richardson, 1920: map XXII (Mississippi
R. local. ) .

Moxostoma aureolum: Hubbs, 1930:30 (Osage R.). Barnickol and Starrett,
1951:292-293 (abund.; utiliz.; Mississippi R. local.). Purkett, 1958a: 124-
125 (growth; SaltR.).

Moxostoma aureolum pisolahrum: Trautman and Martin, 1951 (orig. descr.;
type local. Coon Cr., Jasper Co.; Missouri local, mapped). Martin and
Campbell, 1953:46 (abund.; Black R.). Patriarche, 1953 (abund.; Lake
Wappapello ) .

Moxostoma macrolepidotum: Hanson and Campbell, 1963 (linear distr.; Perche
Cr. ). Cross, 1967:189 (Missouri R. local, mapped).

Northern redhorse: Purkett, 1958b:9,28,43 (growth; length-weight relat.;
Missouri streams).

Taxonomic considerations. — Trautman and Martin (1951) rec-
ognized Moxostoma macrolepidotum pisolabrum from the Ozark
Uplands and adjacent parts of the Arkansas and Osage river systems
in Kansas and Oklahoma. They indicated that M. m. pisolahrum

Missouri Fishes 387

differed notabh' from tlie nominate subspecies only in the enlarge-
ment of the uppc>r lip, w hich in adults is developed into a prominent
bulbous knob. They treated all populations from Missouri south
of the Missouri River as typieal M. in. pisolahrum and restricted
the zone of intergradation between the two subspecies to Missouri
north of the Missouri River. Minckley and Cross (1960) subse-
quenth' extended the zone of intergradation to include all of the
Missouri Ri\er system in Kansas, except for the Osage drainage.
Since 1961 I have examined much additional material of M. macro-
lepidotum from Missouri, and I can see no basis for treating popu-
lations in Ozarkian tributaries of the Missouri River as typical
M. 71}. pisolahrum, while treating populations in the same drainage
system north of the Missouri River as intergi-ades. Specimens I
have examined from the Osage, Gasconade and Meramec systems
south of the Missouri River seem to exhibit as much variability in
the development of the bulbous knob as do specimens from Chari-
ton River, Perche Creek and Loutre River to the north. Specimens
from Mississippi River tributaries north of the Missouri River are
likewise quite variable in this character but are much closer to the
nominate subspecies than populations in the Missouri River sys-
tem. Thus, it would appear that the zone of intergradation is much
broader than that indicated by Trautman and Martin and perhaps
includes all of the Missouri River system in Missouri. Probably
populations in tributaries of the Mississippi River from Salt River
northward are typical M. m. macrolepidotum.

Distribution and habitat. — The northern redhorse is widespread
in the Ozarks, but is rare in the White River system (Map 102). It
is abundant in the northeastern Ozark border but becomes increas-
ingly scarce and spotty in distribution westward in the prairie
region. Moxostoma macrolepidotum inhabits streams of all sizes,
but it achieves its greatest abundance in moderately large rivers.
It is the most abundant redhorse in the lower sections of large
Ozark streams. In the prairie region it frequents small streams to
a much greater extent than in the Ozarks. In rivers the northern
redhorse is most abundant in deep, swift chutes; in small streams it
is sometimes found in pools with no noticeable current. No other
Missouri redhorse seems as plastic in its habitat requirements as
the northern redhorse.

Zoogeography.— Metcali (1966:144) suggested that M. m.
macrolepidotum may have been localized in the pregacial Hudson
Bay system and came into contact with M. m. pisolahrum after de-
flection of the upper Missouri River southward into the Mississippi

388 University of Kansas Publs., Mus. Nat. Hist.

Valley. Moxostoma m. pisolahrum has probably long inhabited the
Ozark Uplands. Perhaps it had a widespread preglacial distribu-
tion in the northern and western part of the Mississippi Valley.
Following retreat of the Wisconsin ice sheet, the upper Mississippi
and Great Lakes drainages may have been invaded by M. m.
macrolepidotum from a glacial refugium in the upper Missouri
River by way of temporary connections between the upper Mis-
sissippi and middle Missouri systems (Bailey and Allum, 1962:113-
114). Moxostoma m. pisolahrum and its close relative M. m. brevi-
ceps represent another example of the geminate forms inhabiting
the Ozark and Appalachian uplands. Moxostoma m. hreviceps
was probably localized preglacially in the Teays, Ohio, and/ or
Tennessee systems.

Moxostoma carinatum (Cope) — river redhorse

Placopharynx carinatus: Patriarche, 1953 (abund.; Lake Wappapello ) .
Moxostoma carinatum: Martin and Campl)ell, 1953:46 (abund.; Black R.).
River redhorse: Purkett, 1958b:29, 43 (growth; length-weight relat.; Missouri
streams ) .

Distribution and habitat. — The river redhorse occurs throughout
the Ozarks but is nowhere abundant (Map 103). In most collec-
tions it is represented by one or a few individuals. Moxostoma car-
inatum has declined markedly over much of its range within the
last century, but there seems to have been no changes in distribu-
tion in Missouri for at least the last 30 years. This sucker inhabits
the pools of clear, medium-sized or large streams with permanent
strong flow and clean gravelly or rocky bottoms. It seems to be less
tolerant of turbidity, siltation, and intermittent flow than any other
Missouri redhorse.

Zoogeography. — The distribution of M. carinatum suggests that
it inhabited the preglacial Teays-Mississippi system, and has long
been present in the Ozark Uplands. Disjunct populations in south-
ern Michigan and in the St. Lawrence River may date from a more
widespread distribution to the northeast during the Climatic Opti-
mum. Alternatively, the intervening populations may have been
extirpated by man's activities within historic time.

Minytrema melanops (Rafinesque) — spotted sucker

Minytrema melanops: Meek, 1891:117 ( Meramec Spring near St. James).
Patriarche, 1953 (abund.; Lake Wappapello). Martin and Campbell, 1953:
46 (abund.; Black R.). Funk and Campbell, 1953 (Black R. local).

Distribution and habitat. — The spotted sucker occurs mainly in
the lowlands with scattered populations in the eastern and south-

Missouri Fishes 389

western Ozarks (Map 104). It is known from the upper Mississippi
River and the Salt River system but has not been taken there for
nearly 30 years. Cross (1967:182) mapped early records of the
spotted sucker in the Missouri drainage of Kansas, but this species
has not been reported anywhere else in the Missouri system. If the
spotted sucker is native to the Missouri system, it is surprising that
it does not occur in the Gasconade and Osage drainages of Missouri,
where there seems to be an abundance of habitat suitable for this
species. In the lowlands the spotted sucker is most often found in
clear ditches ha^■ing little or no current and with thick growths of
submergent aquatic vegetation. In the Ozarks it inhabits the slug-
gish pools of small creeks draining level areas along the major
divides, and the quiet backwaters and overflow pools of larger
streams.

Zoogeographij. — Minytrema melanops is widespread in the cen-
tral and southeastern United States. Perhaps it has inhabited the
lower Mississippi Valley continuously since preglacial time.

Erimyzon sucetta Lacepede — lake chubsucker
Erimyzon sucetta kenneriyi: Martin and Campbell, 1953 (abund.; Black R.).

Dlstrihuiion and habitat. — The lake chubsucker occurs in the
lowlands, and has isolated populations along the Current and upper
Meramec rivers in the Ozarks (Map 105). Specimens collected
near St. Louis in 1853 suggest a former more widespread distribu-
tion along the Mississippi River. This fish is nowhere abundant
in Missouri, and may have declined in recent years. The lake
chubsucker is most often found in clear, quiet pools having a thick
growth of submerged aquatic vegetation and bottoms composed of
sand or silt mixed with organic debris. In the lowlands it is found
in the smaller and more sluggish ditches, and in the Ozarks it oc-
curs in overflow pools and quiet backwaters along the larger
streams. It is likely that the habitat of this species was more preva-
lent in the lowlands before the swamps were drained than is now
the case.

Zoogeograpluj. — The range and requirements of E. sucetta and
Minytrema melanops are much alike, suggesting a similar distribu-
tional history for the two species.

Erimyzon oblongus (Mitchill) — creek chubsucker

(?) Erimyzon sucetta: Meek, 1891:129 (North Fork White R. S Cabool).
Erimyzon oblongus claviformis: Martin and Campbell, 1953:46 (abund.; Black
R.). Funk and Campbell, 1953 (Black R. local.).

Distribution and habitat. — The creek chubsucker is widespread

390 University of Kansas Publs., Mus. Nat. Hist.

in the southern and eastern Ozarks but is absent from the Neosho
and Missouri stream systems (Map 106). It is nowhere abundant
and tends to occur as small, widely scattered populations.

The habitat of E. ohlongus is similar to that of the lake chub-
sucker, but it is less strictly associated with submergent vegetation,
and occurs more frequently in the main channels of flowing streams.
The two species may be highly competitive, as suggested by their
complementary distribution patterns in Missouri.

Zoogeography. — Two subspecies of E. ohlongus are recognized.
Erimijzon o. ohlongus (Mitchill) occupies the eastern part of the
Lake Ontario Basin and Atlantic coastal drainages as far south as
Virginia (Hubbs and Lagler, 1947:51), and E. o. claviformis
(Girard) occupies the remainder of the range as mapped (Map
106). Intergrades occur in Atlantic coastal drainages of Georgia
(Bailey, Winn and Smith, 1954:123). The distributional relation-
ships of these two subspecies are similar to those of the subspecies
of Esox americanus, and perhaps are subject to the same explana-
tion. Erimyzon o. claviformis and E. o. ohlongus probably evolved
from ancesti'al stocks that were localized in the Mississippi Valley
and Atlantic coastal drainages respectively. Possibly the progenitor
of E. o. ohlongus occupied the preglacial Laurentian system. Gulf
and Atlantic coastal drainages of the southeastern United States
were probably occupied at that time by ancestral stocks of Eri-
myzon tenuis (Agassiz). The lowering of sea levels during Pleisto-
cene ice advances may have created stream connections that per-
mitted eastward movement of E. o. claviformis along the Gulf Goast
and southward movement of E. o. ohlongus along the Atlantic
Goast, bringing the three previously isolated forms into contact.
After retreat of Wisconsin ice E. o. claviformis entered the Great
Lakes system through the outlets of glacial lakes Ghicago and
Maumee (Greene, 1935:63), and came into contact with E. o.
ohlongus in the Lake Ontario Basin.

't^"-

Ictaluridae
Ictalurus melas (Rafinesque) — black bullhead

Ameiunis melas: Jordan and Meek, 1885:10,13,14,15,16 (Hundred and Two
R. at Maryville; Missouri R. at St. Joseph; Tabo Cr. 6 mi. E Lexington;
Flat Cr. at Sedalia and/or Blackwater R. at Brownsville; Grand R. at
Clinton and/or Tebo Cr. at Calhoun). Meek, 1891:120,124 (Big Piney R.
at Cabool; Maries R. near Dixon). Fowler, 1915:208 (St. Joseph). Forbes
and Richardson, 1920: map LV (Mississippi R. local.). Patriarche and
Campbell, 1958 (abund.; Clearwater Res.).

Ameiunis melas melas: Barnickol and Starrett, 1951:303-311 (abund.; Missis-
sippi R. local.). Martin and Campbell, 1953 (abund.; Black R.). Funk and
Campbell, 1953 (abund.; Black R. local.).

Missouri Fishes 391

Ictalurus nichi.s: Fislier, 1962:428 (Missouri R. local.). Hanson and Campbell,
1963 (linear distr.; Perehe Cr.). Cross, 1967:198 (Missouri R. local,
mapped ) .

Distribution and habitat. — The black bullhead probably occurs
at least occasionally in ever)- principal stream system of Missouri
(Map 107). It is the most abundant bullhead in the prairie region.
In the Ozark and lowland regions it is usually less abundant than
the yellow bullhead. Ictalurus melas has broad environmental tol-
erances. The habitats in which it is abundant are characterized
by turbid water, a silt bottom, the absence of noticeable current
or strong flow, and a lack of di\'ersity in the fish fauna. Especially
fa\'orable are the permanent pools of small, low-gradient creeks
haN'ing intermittent flow, and the muddy oxbows and backwaters
of large streams in the prairie region. In these habitats the black
bullhead, fathead minnow, green sunfish, and golden shiner often
comprise the bulk of the fishes present.

Zoogeography. — Ictalurus melas is wide-ranging, but is more
western in distribution than the other bullheads. This, along with
its tolerance of extremes of temperature, turbidity, and flow, sug-
gests an origin in streams of the Great Plains. Some workers have
recognized a northern and a southern subspecies. Metcalf (1966:
149) suggested that the subspecies may have resulted from initial
isolation of ancestral stocks in the southwestern part of the Hudson
Bay system and in preglacial drainage of the central and southern
plains. Subsequent mixing of the two stocks and the formation
of a broad zone of intergradation occurred when part of the Hud-
son Bay drainage was diverted into the Mississippi Valley. The
black bullhead has been reported from Illinoian fossil deposits
from southwestern Kansas and northwestern Oklahoma (C. L.
Smith, 1954:285; 1958:178; G. R. Smith, 1963:281).

Ictalurus nebulosus (Lesueur) — brown bullhead

Amehinis nebulosus: Meek, 1891:120 (Osage Fork Casconade R. 6 mi. SE
Marshfield). Fowler, 1915:207-208 ("Paw Paw, Mo.").

Ictalurus nebulosus maimoratus: Barnickol and Starrett, 1951:303-311 (abund.;
Mississippi R. local.).

Ictalurus nebulosus: Fisher, 1962:428 (Moniteau Cr. near Rocheport).

Distribution and habitat. — I have confirmed locality records for
the brown bullhead in Missouri only from Duck Creek Wildlife
Area in Bollinger County, where this fish is common, and a road-
side ditch in Stoddard County, where a single specimen was taken
in 1946 ( Map 108 ) . I have not reexamined specimens for the
references cited above; any or all of them may be based on mis-
identifications, because the species of bullheads have often been

392 University of Kansas Publs., Mus. Nat. Hist.

confused. The "Paw Paw, Mo." referred to by Fowler (1915:207-
208) mav be Paw Paw Creek in Sullivan Countv. The brown
bullhead has been stocked in a few ponds in Missouri, but none of
these stockings is known to have resulted in the establishment of
populations in natural waters. The habitat requirements of the
brown bullhead are similar to those of the yellow bullhead, except
that nehulosus is less often found in flowing waters.

Zoogeography. — Ictalurus nehulosus is eastern in distribution,
reaching the western limit of its present range in Missouri. C. L.
Smith (1962:512) tentati\'ely identified this species from Middle
Pliocene deposits in Logan County, Kansas, thereby indicating a
former more westerly distribution.

Ictalurus natalis (Lesueur) — yellow bullhead

Ameiurus natalis: Jordan and Meek, 1885:13 (Missouri R. at St. Joseph).
Meek, 1891:116 (Little Dry Fork near Rolla). Fowler, 1915:207 "(Marsh-
field). Martin and Campbell, 1953 (abund.; Black R.). Funk and Camp-
bell, 1953 (abund.; Black R. local.). Fimk, 1957 (mov.; Missouri streams).
Patriarche and Campbell, 1958 (a]:)und.; Clearwater Res.).

Ameiurus natalis natalis: Barnickol and Starrett, 1951:303-311 (abund.; Mis-
sissippi R. local.).

Ictalurus natalis: Fisher, 1962:428 (Missouri R. local.). Hanson and Camp-
bell, 1963 (linear distr.; Perche Cr.).

Distribution ami habitat. — The yellow bullhead is nearly as
widespread in Missouri as the black bullhead (Map 109). It is the
commonest bullhead in the Ozark and lowland regions, and it is
nearly as common as the black bullhead in the clearer prairie
streams of central and northeastern Missouri. The yellow bullhead
prefers clearer waters than the black bullhead and is usually found
in or near streams with permanent flow. Like the other bullheads,
this fish avoids strong currents. In the Ozarks it is almost invariably
found in quiet, heavily vegetated backwaters and overflow pools.
Elsewhere it is less restricted and often occurs in the open pools
of the stream channel.

Zoogeography. — This species possibly had a more eastern origin
than /. melas, but its present widespread distribution provides no
clues to the drainages in which its ancestral stocks were localized.

Ictalurus punctatus (Rafinesque) — channel catfish

Ictalurus punctatus: Jordan and Meek, 1885:11,13,14,16 (Hundred and Two
R. at Marvville; Missouri R. at St. loseph; Tabo Cr. at Lexington; Grand R.
at Clinton and/or Tebo Cr. at Calhoim). Call, 1887:74 (Piney R., Texas
Co.). Meek, 1891:120 (Gasconade R. near Newburg). Fowler, 1915:206
(Brownsville and St. Louis). Foibes and Richardson, 1920:map LII (Mis-
sissippi R. local.). Coker, 1930:174-177 (alnuid.; life hist.; Mi.ssissippi R. at
Canton). Martin and Campbell, 1953:47 (abund.; Black R.). Funk and
Campbell, 1953 (abund.; Black R. local.). Marzolf, 1957 (reprod.; Mis-
souri ponds). Funk, 1957 (mov.; Missouri streams). Purkett, 1958a: 127-

Missouri Fishes 393

129 (mowth; import.; Salt R.). Fisher, 1962:428 (Missouri R. local.).
Biirress, 1962 (abiind.; Bull Shoals Res.). Hanson and Campbell, 1963
(linear distr.; Perche Cr. ). Morris, 1967 (prop.; Missouri ponds). Cross,
1967:206 (Missouri R. local, mapped).

Ictahinis (iii^tiilla: Forbes and Richardson, 1920:179-180 (Mississippi R. at
Alton and Ciralton, 111.).

Ictalurus lactistris punctatus: Borges, 1950 (Niangua Arm, Lake Ozark).
Barnickol and Starrett, 1951:303-311 (abund.; growth; Mississippi R.
local.). Marzolf, 1955 (growth; Niangua Arm, Lake Ozark).

Ictalurus lacustris: Bemer, 1951:9,10 (food; Missouri and Mississippi R.).

Patriarche, 1953 (abund.; Lake Wappapello).
Channel catfish: Purkett, 1958b:31,43 (growth; length-weight relat.; Missouri

streams). Fry, 1962 (harv.; Table Rock, Taneycomo, and Clearwater Res.

taiK\'aters ) .

Distrihution and Jiabitaf. — In Missouri the channel catfish is the
most abundant and widely distributed of the larger catfishes; it
occurs o\'er all of the state except for the central Ozarks (Map
110). It is found in a variety of habitats but attains its greatest
abundance in large streams and impoundments. It avoids clear
or cool streams, and those with high gradients. Adults are found
in the larger pools, in deep water or near cover. The young often
occur in riffles and in shallow pools.

Zoogeography. — Ictalurus punctatus ranges widely in the cen-
tral United States, and could have been present in all of the princi-
pal preglacial drainages of the region. It is known from Pliocene
(C. L. Smith, 1962:507) and Pleistocene (C. L. Smith, 1954:285;
1958:179 and G. R. Smith, 1963:281) fossil deposits in the southern
plains, thereby indicating that it has long occurred in or near
Missouri.

Ictalurus furcatus (Lesueur) — blue catfish

Ameiurus ponderosus: Bean, 1880a:286-290 (orig. descr.; Mississippi R. near

St. Louis).
Ictalurus furcatus: Jordan and Meek, 1885:13 (Missouri R. at St. Joseph).

Forbes and Richardson, 1920:179 (abund.; Mississippi R. at Alton, 111.).

Coker, 1930:174-175 (Mississippi R. at Warsaw, 111.). Berner, 1951:table

5 (commer. catch; Missouri and Mississippi rivers). Fisher, 1962:428

(Missouri R. local.). Cross, 1967:209-213 (size records; Missouri R. local.

mapped ) .
Ictalurus f. furcatus: Borges, 1950 (vert, distr.; Niangua Arm, Lake Ozark).

Barnickol and Starrett, 1951:303-311 (abund.; Mississippi R. local.).

Distribution and habitat. — In Missouri the blue catfish is known
only from the Missouri, Mississippi, Osage and Grand rivers (Map
111). In these streams it is much less common than the other large
catfishes. Ictahirus furcatus is principally an inhabitant of swift
chutes and of pools having noticeable current. It also occurs in
the open waters of large reservoirs.

Zoogeography. — An extinct species (Ictalurus lambda Hubbs

394 University of Kansas Publs., Mus. Nat. Hist.

and Hibbard) related to I. furcatus has been reported from early
Pliocene fossil deposits in Kansas (Hubbs and Hibbard, 1951:8-14;
C. L. Smith, 1962:510). This suggests that /. furcatus or its an-
cestral stocks long have been present in or near Missouri.

Noturus gyrinus (Mitchill) — tadpole madtom

Noturus gyrinus: Hanson and Campbell, 1963 (linear distr.; Perche Cr. ). Tay-
lor, 1969:35-54, map 1: (syn.; char.; compar.; Missouri local, mapped).

Distribution and habitat. — The tadpole madtom occurs in the
lowlands and in a broad northeastward-trending zone from Spring
River in southwestern Missouri into tributaries of the Mississippi
River in northeastern Missouri (Map 112). Noturus gyrinus may
have been more abundant in Missouri before settlement than it is
today, particularly in the prairie streams of west-central Missouri.
Trautman (1957:444-445) noted a decrease in the abundance of
this species in Ohio as a result of the drainage of marshes, stream
channelization, and increased turbidity and siltation; similar en-
vironmental changes have occurred in Missouri. The tadpole mad-
tom inhabits clear to moderately turbid waters with little current
and an abundance of cover, such as thick growths of submergent
aquatic vegetation or accumulations of organic debris. In the low-
lands N. gyrinus is most abundant in ditches having no noticeable
current and along the vegetated margins of ditches with current.
In the prairie region it inhabits quiet pools of small, sluggish creeks
draining flat upland areas, and backwaters and overflow pools
along the larger streams.

Zoogeography. — The present distribution of this wide-ranging
species provides few clues to its place of origin. Perhaps it has
inhabited the Mississippi Valley continuously since preglacial time.

Noturus exilis Nelson — slender madtom

Noturus exilis: Bean, 1880:112 (South Grand R.). Call, 1887:79 (Hinkson
Cr., Boone Co.). Meek, 1891:117,120,124,126,129 (Little Dry Fork near
Rolla; Jones Cr. near Dixon; Little Piney Cr. near Arlington and/or New-
burg; Niangua R. near Marshfield; Shoal Cr. near Neosho; James R. near
Springfield). E\ermann and Kendall, 1895:470 (Indian Cr. near Neosho).
Metcalf, 1966:151 (Moniteau Cr., Moniteau Co.; Moreau Cr., Morgan Co.).
Taylor, 1969:57-68, map 2 (syn.; char.; compar.; Missouri local, mapped).

Sdiilbeodes exilis: Jordan and E\ermann, 1900:3236, fig. 65 (sic.) (Ozark
Fork Gasconade R., Marshfield).

Schilbeodes insigtiis: Fowler, 1915:209 (Carthage). Hubbs and Raney, 1944:
18-24 (char.; svn.; compar.; Mi.ssouri local, mapped). (?) Martin and
Campbell, 1953 '(abund.; Black R.).

Distribution and habitat. — The slender madtom is common over
most of the Ozarks, and it occurs at scattered localities in the north-
eastern Ozark border (Map 113). Elsewhere in the state it is

Missouri Fishes 395

kno\\'n onl)' from the Blue River in Jackson County, and tributaries
of the Osage River in Vernon County. Over much of the Ozarks
it is the most abundant madtom. However, it is scarce in the south-
eastern Ozarks, perhaps as a result of competition from the Ozark
madtom, a species with similar habits. The slender madtom is char-
acteristic of small to medium-sized streams with moderate or high
gradients and permanent flow of clear water. It is usually found on
rocky riffles, or in rocky pools where there is sufficient current to
keep the bottom free of silt.

Zoo^eograpliy. — The present distribution of N. exilis suggests
that its origin was in the central Mississippi Valley, and that it
probably has long occupied the Ozark Uplands. Disjunct popula-
tions in the Tennessee and Cumberland basins (Taylor, 1969:64)
may owe their isolated position to eastward dispersal during one
of the Pleistocene ice advances, when the lowering of sea levels
produced conditions in the Mississippi Embayment more suitable
for upland fishes.

Noturus nocturnus Jordan and Cilbert — freckled madtom

Noturus nocturnus: Taylor, 1969:74-83; map 4 (syn.; char.; compar.; Missouri
local, mapped).

Distribution and habitat. — The distribution of the freckled
madtom in Missouri is similar to that of the tadpole madtom, but
the two are seldom found together (Map 114). Noturus nocturnus
is nowhere abundant, but it is frequently found in the upper Osage
and Mississippi rivers, as well as streams and ditches of the low-
lands. It inhabits clear to moderately turbid streams having per-
manent flow and low or moderate gradients. Noturus nocturnus oc-
cupies sluggish riffles over a gravel or rubble bottom. Cross (1967:
219) reported it from among leaves and other woody material over
a muddy bottom in Kansas, but I have not found it in such habitats
in Missouri.

Zoogeography. — Noturus nocturnus is primarily southern in
distribution and perhaps is autochthonous to the lower Mississippi
Valley.

Noturus flavus Rafinesque — stonecat

Noturus flavus: Jordan and Meek, 1885:10,15,16 (Hundred and Two R. at
Maryville; Flat Cr. near Seda'ia and/or Blackwater R. at Brownsville;
Grand R. at Clinton and/or Tebo Cr. at Calhoun). Fowler, 1915:208
(Clinton). Taylor, 1969:111-128, map 8 (syn.; char.; var.; compar.; Mis-
souri local, mapped).

Distribution and habitat. — The stonecat occurs over most of the
northern two-thirds of Missouri, but is most abundant in the north-

396 University of Kansas Publs., Mus. Nat. Hist.

ern Ozark border and the prairie region (Map 115). It is common
in the upper Mississippi, but it is rare in the lower Mississippi and
Missouri rivers. Notiinis flavus occurs in varied stream types but
avoids those with intermittent flow or extremely high gradients.
It usually occupies rocky riffles in swift current. In the Missouri
River I have found it over sandy bottoms in swift current.

Variation and zoogeography. — Stonecats from the Missouri and
lower Mississippi rivers have the eyes much reduced in size, but
otherwise appear to be typical of flamis (Taylor, 1969:123). These
differences may be environmentally induced, but it is also possible
that the Missouri-lower-Mississippi population is a genetic \'ariant
adapted for survival in the distincti\'e habitat of large, continuously
turbid streams. iV. jiavus has a widespread northern distribution
and could have been present preglacially in all of the principal
drainages that were integrated to form the present Mississippi
system.

Noturus albater Taylor — Ozark madtom

(?) Noturus flavus: Martin and Campbell, 1953:47 (abimd.; Black R.).

Noturus albater: Taylor, 1969:144-150, map 10 (orig. descr.; type local. Wliite
R., Taney Co.; Missouri local, mapped).

Distribution and habitat. — The Ozark madtom is the most abun-
dant small catfish in the southern Ozarks (Map 116). It inhabits
riffles and rocky pools of clear, high-gradient streams with perma-
nent strong flow.

Zoogeography. — The Ozark madtom is endemic to the southern
Ozarks, and probably had its origin in or near its present area of
occurrence. It is a distinctive species without close affinity to other
species of the subgenus Rabida (Taylor, 1969:149).

Noturus placidus Taylor — Neosho madtom

Distribution and habitat. — The Neosho madtom was included
only in a collection from the Spring River just upstream from the
Kansas state line (Map 117). This collection contained 60 slender
madtoms and 23 stonecats, but only two Neosho madtoms. Col-
lections elsewhere in the Spring River in Missouri have yielded
an abundance of other madtoms, but not the Neosho madtom;
Cross (1967: 233) reported it from the Spring River in Kansas.
N. placidus inhabits gravelly or rocky riffles of medium-sized to
moderately large streams with permanent flow, moderate gradients,
and fairly clear water. Deacon ( 1961:396) noted a marked decrease
in the abundance of the Neosho madtom in Kansas during a period

Missouri Fishes 397

of extreme drought, when the Neosho River became intermittent
for the first time since flow records were kept.

Zoog.eograpJiy. — The Neosho madtom is endemic to the Neosho
and iHinois stream systems of northeastern Oklahoma and adjoining
states. It seems to be most closely allied to madtoms of the furiosus
species group (Taylor, 1969:172) and may have evolved from a
common ancestral stock that has long been isolated west of the
Mississippi Embayment. No other species of this group occurs
west of the Mississippi River.

Noturus miurus Jordan — brindled madtom

Schilhcodes miurus: Martin and Campbell, 1953 (abund.; Black R.).
Noturus miurus: Taylor, 1969:190-201, map 14 (syn.; char.; compar.; Mis-
souri local, mapped).

DistriI)iition and habitat. — The brindled madtom occurs in the
lowland ditches and adjacent portions of the southeastern Ozarks,
and in the Spring River of southwestern Missouri (Map 118). It
is much more common in the lowlands than in Spring River. Other
observers (Cross, 1967:225; Trautman, 1957:441) reported the
brindled madtom to be primarily an inhabitant of pools with little
current and an abundance of organic debris, but in Missouri N.
miurus is most often found on gravelly or rocky riffles in a slight
or moderate current. The streams where it occurs are characterized
by moderate or low gradients, permanent flow, and bottoms that
are mostly free of silt.

Zoogeographij. — The origin of the miurus species group (IV.
miurus, N. fiavater and N. flavipinnis) was probably in the central
Mississippi Valley, where all are found today (Taylor, 1969:189).

Noturus flavater Taylor — checkered madtom

Noturus flavater: Taylor, 1969:204-208, map 14 (orig. descr.; type local. Flat
Cr., Barry Co.; Missouri local, mapped).

Distribution and habitat. — The checkered madtom occurs on the
southern slope of the Ozark Uplands, where it is similar in distribu-
tion to but generally less abundant than the Ozark madtom. The
principal difl^erence in their ranges is the absence of the checkered
madtom from the St. Francis and Black rivers (Map 119). The
checkered madtom inhabits moderately large, clear streams having
high gradients and permanent strong flow. It occurs in quiet pools
or backwaters where the bottom is composed of gravel and rocks
overlain by thick deposits of sticks, leaves, and other organic debris.

Zoogeography. — The checkered madtom is endemic to the
White and Black stream systems of the southern Ozarks. It is most

398 University of Kansas Publs., Mus. Nat. Hist.

closely allied to N. miunis and N. flavipinnis (Taylor, 1969:188),
and probably evolved from a common ancestral stock that has long
been localized in its present area of occurrence.

Pylodictis olivaris (Rafinesque) — flathead catfish

Leptops olivaris: Jordan and Meek, 1885:13,16 (Missouri R. at St. Joseph;
Grand R. at Clinton and/or Tebo Cr. at Calhoun). Meek, 1891:120
( Gasconade R. near Newburg ) .

Pylodictis olivaris: Borges, 1950 (vert, distr.; Niangua Arm, Lake Ozark).
Barnickol and Starrett, 1951:303-311 (abund.; growth; Mississippi R.
local.). Berner, 1951: table 5 (commer. catch; Missouri and Mississippi
rivers). Patriarche, 1953 (abund.; Lake Wappapello ) . Martin and Camp-
bell, 1953:47 (abund.; Black R.). Funk and Campbell, 1953 (abund.;
local.; Black R.). Patriarche and Campbell, 1958 (abund.; Cleanvater
Res.). Fisher, 1962:428 (Missouri R. local). Burress, 1962 (abund.;
harv.; Bull Shoals Res.). Hanson and Campbell, 1963 (Hnear distr.; Perche
Cr.). Cross, 1967:213 (Missouri R. local, mapped).

Flathead catfish: Purkett, 1958b:31 (growth; Missouri streams). Fry, 1962
(harv.; Taneycomo and Clearwater res. tailwaters).

D'lsirihution and habitat. — The flathead catfish occurs in most
of the large streams of Missouri (Map 120). It is most abundant
in the Missouri and Mississippi rivers and their major tributaries
in the prairie and Ozark border regions. It is common in large
reservoirs of the Ozarks, but is rare or absent in many streams of
that region. Barnickol and Starrett (1951:303) reported that the
total poundage of flatheads in their collections from the Mississippi
River exceeded that of any other species of catfish. Flathead cat-
fishes inhabit medium to large streams having moderate or low
gradients and permanent flow. The young are often found among
rocks in swift water, occupying essentially the same habitats as the
stonecat and other riffle-inhabiting madtoms. Adults occur in pools,
almost invariably near submerged logs, piles of drift, or other cover.

Zoogeograpluj.— The present distribution of P. olivaris sug-
gests an origin in the Mississippi Valley, or in other Gulf drainages
to the southwest.

Percopsidae
Percopsis omiscomaycus (Walbaum) — trout perch

Percopsis omiscomaycus: Hanson and Campbell, 1963 (linear dist.; Perche

Cr.).

Distribution and habitat. — The trout perch is widespread in
the Grand and Chariton stream systems, and occurs eastward along
the Missouri and Mississippi rivers to St. Charles and Perry coun-
ties (Map 121). It is fairly common in parts of the Chariton,
Lamine, Petite Saline and Perche stream systems, but it is rare
elsewhere in the state. Its distribution and abundance seem to

Missouri Fishes 399

have remained unchanged since at least the early 1940's. In Mis-
souri the trout perch is common only in the unstraightened sections
of prairie streams having permanent flow and deep pools floored
b\' sand and fine gravel. Possibly the trout perch was more wide-
spread in northern and western Missouri before extensive chan-
nelization of the prairie streams.

Zoogeograpliy. — The trout perch is widespread in the northern
United States and Canada, and could have been present preglacially
in all the principal drainages of that region. Perhaps it was absent
from the Mississippi Valley in pre-Pleistocene time, and dispersed
southward by way of stream connections that developed with
glaciation.

Aphredoderidae
Aphredoderus sayanus (Gilliams) — pirate perch

Distribution and habitat. — The pirate perch is widely distributed
in the lowlands and adjacent sections of the southeastern Ozark
border (Map 122). It is also known from Peruque Creek in St.
Charles County, and probably occurs elsewhere along the Missis-
sippi River. In spite of its general occurrence in the lowlands, A.
sayanus is seldom abundant. The pirate perch inhabits bottomland
lakes, o\'erflow ponds, and the quiet pools and backwaters of low-
gradient streams and ditches. The habitats it occupies are char-
acterized by clear water, absence of current, and an abundance of
aquatic vegetation, organic debris, or other cover.

Zoogeography. — The pirate perch is the only living species of
its family. It may have occupied the lowlands of the Mississippi
Valley throughout the Cenozoic Era.

Amblyopsidae
Amblyopsis rosae (Eigenmann) — Ozark cavefish

TypJilichtJujs subterraneous: Carman, 1889:226 (wells in Jasper Co.; Wilson's

and Day's caves, Jasper Co. ) .
TyphlicJithys rosae: Eigenmann, 1898:231 (original descr.; Day's Cave, Jasper

Co.). Ibid., 1899:247-251 (descr.; compar.).

Troglichthijs rosae: Cox, 1905 (subterranean waters in southern Missouri,
northern Arkansas, and probably eastern Kansas ) .

Amblyopsis rosae: Woods and Inger, 1957:245-247 (syn.; char.; Missouri
local, compiled and mapped).

Distribution and habitat. — The Ozark cavefish is known at pres-
ent only from the Ozark Uplands of southwestern Missouri (Map
123), but it also may occur in portions of adjoining states. No ele-
ment in the Missouri fish fauna is less well known than the cave-
fishes, and future efforts directed at correcting this deficiency will

400 Unr'ersity of Kansas Publs., Mus. Nat. Hist.

probably reveal a more widespread distribution for this and the
following species. Amhhjopsis rosae is confined to unglaciated
uplands underlain by limestone bedrocks that were not extensively
deformed by internal stresses during uplift. The highly soluble
limestone bedrocks are honeycombed by subsurface drainage ways.
The Ozark cavefish ventures rarely, if ever, into surface waters;
it has been collected only from caves, wells, and the outlets of
springs.

Zoogeography. — AmhJijopsis rosae is well isolated from its clos-
est relative, A. spelaea DeKay, which occurs east of the Mississippi
Embayment. Woods and Inger (1957:252-255) suggested that the
common ancestor of the family Amblyopsidae entered subterranean
waters of the limestone plateaus of the central United States during
the Tertiary. They envisioned two waves of dispersal. The first
wave involved Amhhjopsis, which spread through the limestone
pleateaus and was then split into two isolated populations by the
more competitive TijphUchtlnjs. One of these isolated populations
was localized in the limestone regions of the Ozark Uplands, where
it gave rise to A. rosae. Amblyopsids presumably disperse primarily
through subterranean channels which are largely independent of
surface drainage. Consequently, they may disperse across major
drainage divides, and perhaps also beneath the channels of large
streams, such as the Mississippi River. Perhaps the successive
waves of dispersal envisioned by Woods and Inger are related to
the repeated cycles of erosion and sedimentation in the Mississippi
Embayment that resulted from fluctuations in sea level during the
Pleistocene. Such changes might create ecological conditions in
subterranean waters that would favor alternate periods of dispersal
and isolation of cavefishes east and west of the Mississippi Embay-
ment.

Typhylichthys subterraneus Girard — southern cavefish

Tijphlichthijs eigenmanni: Hubbs, 1938:265 (nomen nudum, central Missouri).

Typhlichthi/s subterraneus: Woods and Inger, 1957:240-243, fig. 10 (syn.;
char.; Nlissouri local, compiled and mapped).

Distribution and habitat. — The southern cavefish inhabits sub-
terranean waters of the central and southeastern Ozarks ( Map
124). It has not been found in the same caves as the Ozark cave-
fish, but both species occur in Greene County. The habitat of T.
subterraneus seems to be like that of A. rosae.

Zoogeography. — Populations of T. subterraneus in the Ozark
Uplands are broadly disjunct from those east of the Mississippi

Missouri Fishes 401

River. The potential therefore exists for the divergence of two
kinds, as has oeeurred in the genus A)nbU/opsis. Probably T. siibter-
raneus is a recent in\'ader of the Ozark Uplands, and may have dis-
persed westward during the erosional cycle that accompanied the
Wisconsin ice advance.

Gadidae

Lota Iota (Linnaeus) — burbot

Lota lota: Gilbert, 1886:210 (Missouri R. at Leavenworth, Kans.). Fisher,
1962 (Missouri R. local.). Metcalf, 1966:153 (Missouri R. 8 mi. S Atchi-
son, Kans.). Cross, 1967:228 (Missouri R. local, mapped).

Distribution and habitat. — All records for the burbot in Mis-
souri are based on single specimens, usually adults, that probably
strayed into Missouri from the north along the Missouri and Mis-
sissippi rixers (Map 125). Most have been caught by commercial
fishermen. In some years there are many reports, whereas in other
years none are reported. This suggests that population pressure
or other factors result in occasional influx of this fish into Missouri
waters. I have examined a small burbot reported by Dr. Leonard
Durham to have been taken by him and his students at Duck
Creek Refuge, Bollinger County, Missouri, on May 10 or 11, 1964.
Durham told me that the specimen, a small individual, was not
noticed at the time the collection was made, but was later found
when the collection was being sorted. This record is in need of
further substantiation, since it is far south of any previous Missouri
locality, and is the only report outside of the mainstream of the
Missouri and Mississippi rivers.

Zoogeography. — This holarctic species reaches the southern
limit of its range in Missouri. Its presence in the Mississippi Valley
may date from the Pleistocene, when drainage derangements that
accompanied glaciation permitted faunal exchange with other
stream systems to the north.

Cyprinodontidae
Fundulus catenatus (Storer) — northern studfish

Fundulus catenatus: Call, 1887:77 (West Fork of Black R. and tribs., Rey-
nolds Co.; Jacks Fork and tribs.. Shannon Co.). Meek, 1891:118,122,130
(Meramec R. and Big Dry Fork near St. James; Jones Cr. near Dixon;
Gasconade R. and Little Piney Cr. near Arlington; James R. near Spring-
field). Martin and Campbell, 1953 (abund.; Black R. ). Patriarche and
Campbell, 1958 (abund.; Clear\vater Res.). Smith, 1965:11 (St. Louis,
Ste. Genevieve, Perry, and Cape Girardeau counties). Branson, 1964:746
Shoal Cr., Newton Co.).

Distribution and habitat. — The northern studfish is one of the
most common and widely distributed Ozark stream fishes (Map

402 University of Kansas Publs., Mus. Nat. Hist.

126). Occurrence of F. catenatus in Elk River of extreme south-
western Missouri is presumably the result of introduction. All
records are recent, and Hall (1956:20-21) suggested that occur-
rence of the northern studfish in the Neosho system of Oklahoma
is the result of introduction. Fundulus catenatus inhabits streams
of all sizes, having moderate or high gradients, permanent flow
of clear water, and bottoms composed mostly of silt-free sand,
gravel, or rubble. Fundulus catenatus is most often found in shal-
low water, along the margins of pools and riffles.

ZoogeograpJuj. — Fundulus catenatus occurs disjunctly in the
Ozark and Appalachian uplands. Presence of the closely related or
conspecific F. steUifer (Jordan) in the Alabama River system sug-
gests that F. catenatus has long occupied the uplands east of the
Mississippi River. Perhaps F. catenatus was localized preglacially
east of the Mississippi River, and dispersed westward into the
Ozark Uplands during one of the Pleistocene ice advances. Absence
of F. catenatus from Ozarkian tributaries of the Arkansas River
until its introduction by man (Hall, 1956:20-21) suggests a recent
invasion of the Ozark Uplands. The occurrence of other disjunct
populations of F. catenatus both north and south of its two main
areas of occurrence indicates a former more widespread distribu-
tion. Possibly these isolated populations have resulted from range
adjustments that accompanied climatic fluctuations during late-
Pleistocene and Recent time.

Fundulus kansae Carman — plains killifish

Fundulus kansae: Miller, 1955:1 (Clay Co.; Boone's Lick Spring, Howard
Co.). Fisher, 1962:428 (Missouri R. local.). Cross, 1967:235 (Missouri
R. local, mapped).

Distribution and habitat. — In recent collections the plains killi-
fish has occurred abundantly only in Salt Creek near Boone's Lick
Spring, Howard County (Map 127), where it is the predominant
fish. In the early 1940's, Ceorge V. Harry collected many speci-
mens in Clear Creek, Clay County, and in 1945 Fisher (1962: table
1) found it in the Missouri River as far downstream as Franklin
County. No specimens were collected by me during a rather
extensive survey of the Missouri River in 1963, suggesting that this
species is not now as abundant there as it was in 1945. No collec-
tions have been made in Clear Creek in recent years, so the present
status of the plains killifish in that stream is not known. Fundulus
kansae normally inhabits streams with alkaline or saline waters
and few other kinds of fish. Its distribution may be limited by a
requirement for high salinity, or by inability to compete in the

Missouri Fishes 403

more di\cnse fisli populations that occur in most Missouri streams.
Where it is abundant, this topminnow is found in all types of habi-
tats, from pools and backwaters to shallow sandy areas with con-
siderable current.

Zoogeography. — Metcalf (1966:152) suggested that F. kansae
originated in preglacial drainages of the central and southern
plains. This is suggested by its present widespread distribution
there and occurrence of the closely related or conspecific F. ze-
hrinus Jordan and Gilbert in adjacent southwestern drainages.
Fundiihis kansae now occurs, presumably as the result of introduc-
tion, in parts of South Dakota (Miller, 1955:11-12), and Montana
(C. J. D. Brown, pers. comm. to F. B. Cross).

Fundulus chrysotus (Giinther) — golden topminnow

Distribution and habitat. — The golden topminnow is known in
Missouri from only five specimens collected at two localities in the
lowlands in 1944 and 1946 (Map 128). Fiindtdus chrysotus did not
occur in collections made at these localities in 1964. The habitat
of the golden topminnow is much like that of the starhead top-
minnow.

Zoogeography. — Ftinduhis chrysotus is southeastern in its affini-
ties. Perhaps it has occupied the lower Mississippi Valley continu-
ously since preglacial time, or entered from Gulf coastal drainages
to the east when lower sea levels created stream connections that
favored dispersal.

Fundulus sciadicus Cope — plains topminnow

Zygonectes macdonaldi: Meek, 1891:122,126-127 (orig. descr.; Jones Cr.
near Dixon; Osage Fork Gasconade R. 6 mi. SE Marshfield; spring branch
at Neosho). Evermann and Kendall, 1895:471 (spring branch at Neosho).

Fundulus sciadicus: Gosline, 1949 (sensory canals; Miller Co.). Metcalf, 1966:
153 (Missouri R. tribs. in central Missouri).

Distribution and liabitat. — The plains topminnow occurs dis-
junctly in the Neosho and Missouri river systems of the north-
central and southwestern Ozarks (Map 129). It is locally abundant
but tends to occur as scattered, isolated populations. Fundulus
sciadicus inhabits quiet pools of small creeks, and the backwaters
and overflow pools of larger streams. It is invariably found where
the water is clear and without noticeable current, and there are
dense growths of aquatic vegetation. The plains topminnow seems
to have declined in abundance or disappeared from parts of the
Neosho and Osage stream systems since the early 1940's.

Zoogeography.— The distribution of F. sciadicus suggests that pre-

404 University of Kansas Publs., Mus. Nat. Hist.

glacially it may have inhabited either the Iowa and lower Missouri
systems, or the Ancestral Plains system. Occurrence in the Neosho
system may date from preglacial time, but could as plausibly have
resulted from southward dispersal through Pleistocene connections
between the Arkansas and middle Missouri systems.

Fundulus notti (Agassiz) — starhead topminnow

Distrihiition and habitat. — The starhead topminnow is known
only from the lowlands and from one locality in St. Charles County
(Map 130). Although limited in distribution, F. notti is usually
abundant where it occurs. This fish is an inhabitant of quiet,
weedy backwaters and oxbows along the courses of major streams.

Zoogeography. — Two subspecies of F. notti are recognized:
F. n. dispar, occurring from the southwestern Great Lakes south-
ward in the Mississippi Valley to northeastern Arkansas and west-
ern Tennessee, and F. /}. notti, occurring southward in the Missis-
sippi Valley and along the Gulf coastal plain from western Florida
west to Louisiana and eastern Texas. A distinct but closely related
species (F. lineolatiis) occurs along the Atlantic Coast, coming into
contact with F. notti in northwestern Florida. The presence of a
distinct subspecies of F. notti in the Mississippi Valley suggests
that it is not a recent invader from the south, but rather has long
been established there.

Fundulus olivaceus (Storer) — blackspotted topminnow

Zygonectes notatits: Call, 1887:77 (West Fork of Black R., Reynolds Co.;
Jacks Fork, Shannon Co.). Meek, 1891:130 (James R. near Springfield;
North Fork White R. S Cabool).

Fundulus olivaceus: Gosline, 1949 (sens, can.; Ozark Co.). Martin and Camp-
bell, 1953 (abund.; Black R.). Brown, 1956 (char.; compar.; Little
Gravois Cr. 2^2 mi. XE Gravois Mills, Morgan Co.; Little Black R. 2.4 mi.
E Fairdealing, Butler Co.). Patriarche and Campbell, 1958:255-256
(abund.; Clearwater Res.). Braasch and Smith, 1965 (char.; hab.; relat.;
Missouri local, mapped). Thomerson, 1966 (distr.; var.; ecol.; Missouri
local, mapped). Setzer, 1970:141 (karyotype; Xiangua R.).

Distribution and habitat. — The blackspotted topminnow is abun-
dant and widely distributed in the Ozark and lowland regions
(Map 131). In the lowlands it is the most abundant topminnow,
and in the Ozarks it is second in abundance to the studfish. Fundu-
lus olivaceus inhabits clear streams and ditches having permanent
strong flow, clean sand, gravel, or rocky bottoms, and thick stands
of water willow or other aquatic vegetation.

In Missouri, as elsewhere in the central and lower Mississippi
Valley, F. olivaceus and its close relative, F. notatus, tend to ex-
clude each other microgeographically and ecologically. In Missouri

Missouri Fishes 405

the\ ha\e occurred together only in collections from the lowlands,
and from Clear Creek, a tributary of the Osage River in St. Clair
County. In a series of collections made in the lowlands in 1964,
both species occurred in nine of 36 collections in which one or the
other occurred. In eight of the nine collections F. olivaceus out-
numbered F. notatus, often by a significant margin. In the Clear
Creek collection, F. notatus predominated. The largely allotopic
distribution of these two closely related topminnows may result
from competition. Their requirements seem to be similar, but they
may differ to the extent that, at any given locality, conditions favor
one species over the other.

The distributional relationships of F. olivaceus and F. notatus
and the factors determining these relationships in the area where
their ranges overlap have been the subject of two recent studies
(Braasch and Smith, 1965; Thomerson, 1966). Braasch and Smith
concluded that F. notatus occurred in "low-gradient streams and
sloughs in relatively level areas," and that F. olivaceus inhabited
"fast gravelly streams in upland areas." They also indicated that
the species have occurred together for 60 years or more in several
stream systems in southern Illinois. Thomerson studied the two
species over a wider area and indicated that their ecological prefer-
ences are not uniform throughout their ranges, but F. notatus
seldom occurs in "blackwater." He also concluded that syntopic
associations at a particular locality are "unstable and transient"
rather than relati\'ely stable, as indicated by Braasch and Smith.

My observations on the ecological preferences of the two species
are in general agreement with Braasch and Smith, except that
F. olivaceus is not confined to upland areas in Missouri. There F.
olivaceus seems to occupy clearer and perhaps cooler waters than
F. notatus. This relationship is apparent in the Meramec system,
where F. notatus is confined to the upper portions of the Bourbeuse
and Meramec mainstem, whereas F. olivaceus occupies most of the
remainder of the Meramec system. The Bourbeuse and upper
Meramec drain undissected uplands, underlain principally by
shales and sandstones, whereas the remainder of the Meramec sys-
tem drains uplands that are intricately dissected and underlain by
cherty dolomites and limestones. Springs are virtually absent from
the upper Bourbeuse and Meramec drainages, and the streams
are more turbid and ha\'e lower base flows than those elsewhere
in the Meramec system. In the lowlands, F. notatus is confined
principally to the larger, more turbid ditches, whereas F. olivaceus
is more widespread.

406 Unrtrsity of Kansas Publs., Mus. Nat. Hist.

Concerning the stability of syntopic populations, it appears that
the area of syntopic occurrence in southeastern Missouri has re-
mained stable for at least 25 years. Perhaps syntopy is maintained
there by influx of F. notatus from the Mississippi River and adja-
cent waters, where this species apparently occurs to the exclusion
of F. olivaceus. The same explanation could account for the syn-
topic occurrence of the two species in the upper Osage system.
Fiindiihis nofotiis occurs to the exclusion of F. olivaceus in the
prairie streams to the west of Clear Creek, whereas F. olivaceus
occurs in streams immediately to the east.

Fundulus notatus (Rafinesque) — blackstripe topminnow

(?) Zygonectes zonahis: Agassiz, 1854:353 (orig. descr.; St. Louis).

(?) Zygonectes notatus: Meek, 1891 (Big Dry Fork near St. James).

Fundulus notatus: Forbes and Richardson, 1920: map XLV (Mississippi R.
local.). Brown, 1956 (char.; compar.; Nleramec R. 2 mi. XW Short Bend,
Dent Co.). Braasch and Smith, 1965 (char.; hab.; relat.; Missouri local,
mapped). Thomerson, 1966 (distr.; var.; ecol.; Missouri local, mapped).

Distribution and habitat. — The blackstripe topminnow is the
most abundant and widely distributed topminnow in the Neosho
system of the southwestern Ozarks and in the northeastern Ozark
border (Map 132). In lowland ditches of the southeast it is less
abundant than the blackspotted topminnow. The blackstripe top-
minnow was not collected in the Missouri River system before 1950,
and its presence there may be due to introduction. The population
in Perche Creek, Boone County, is definitely known to have been
introduced since 1963. The habitat requirements of F. notatus
are like those of the blackspotted topminnow except that it pre-
fers slightly warmer and more turbid waters. The blackstripe top-
minnow avoids streams where the flow is strongly influenced by
springs, and is most often found along large, lowland ri\'ers and
in the isolated pools of small, intermittent creeks draining level
uplands.

Zoogeography. — The present distribution of F. notatus and F.
olivaceus proxides few clues to their distributional relationship in
preglacial time. Probably one or both species inhabited the pre-
glacial Teays-Mississippi system.

Poeciliidae

Gambusia affinis (Baird and Girard) — mosquitofish

Gamhusia affiinis affinis: Martin and Campbell, 1953 (abund.; hab.; Black R.).
Fisher, 1962:428 (Missouri R. local.).

Gambusia affinis: Patriarche and Campbell, 1958 (abund.; Clearwater Res.).

Missouri Fishes 407

Distribution cniil luihiUit. — In the early 1940's the inos(iuitofish
was eonfined to the lowlands and the Mississippi flood plain north-
ward from the lowlands to Pike County (Map 133). More recent
collections re\eal that this fish is now widely distributed in south-
western Missouri and occurs at scattered localities in central Mis-
souri. These range extensions may result in part from stockings
for mosquito control. Gambusia affinis is one of the most abundant
fishes in the lowlands. Backwaters and adjacent oxbows of slug-
gish lowland streams are its preferred habitat. In such situations
it is most abundant in shallow, marginal areas, where the water
is clear and warm and there is considerable aquatic vegetation
or other cover.

Zoogeography. — The widespread southern distribution of G.
affinis suggests that it has long inhabited the lower Mississippi
Valley and coastal drainages of the southeastern United States.

Atherinidae

Labidesthes sicculus (Cope) — brook silverside

Lahidcsthc'.s siccuJus: Jordan and Meek, 1885:16 (Flat Cr. near Sedalia and/or
Blackwater R. at Brownsville). Meek, 1891:119,122,125,126 (Big Dry
Fork near St. James; Osage Fork and Niangua R. near Marshfield; Big
Piney R. at Cabool; Maries R. near Dixon; Hickory or Shoal Cr. near
Neosho). E\-ermann and Kendall, 1895:471 (Indian Cr. near Neosho).
Patriarche, 1953:247 (abund.; Lake W'appapello ) . Patriarche and Camp-
bell, 1958:255-256 (abmid.; Clearwater Res.).

Labidesthes sicculus siccidus: Martin and Campbell, 1953 (abund.; hab.;
Black R.).

Distribution and habitat. — The brook silverside is widely dis-
tributed in the Ozark and lowland regions (Map 134). Labidesthes
siccidus seems to be intolerant of siltation and continuous high
turbidity; it is confined to clear, warm waters without noticeable
current. In the Ozarks it is most often found in the permanent
pools of small headwater creeks that receive little spring flow,
and the warm backwaters and overflow pools of larger streams.
It is one of the most abundant small fishes in Ozark reservoirs.

Zoogeography. — Labidesthes sicculus was probably derived from
an atherinid stock that invaded freshwaters of eastern North Amer-
ica at an early date; probably it has inhabited the Mississippi Valley
continuously since preglacial time.

Menidia audens Hay — Mississippi silverside

Menidia audens: Smith, 1965:12 (Mississippi Co.).

Distribution and habitat. — The Mississippi silverside is common
in the Mississippi Ri\er downstream from the mouth of the Ohio

408 University of Kansas Publs., Mus. Nat. Hist.

River (Map 135). It was not collected in Missouri until 1963.
Menidia audens inhabits the open waters of large, moderately
clear rivers. It is most readily seined at night, suggesting that it
moves inshore during the hours of darkness and returns to deeper
water in the daytime.

Zoogeography. — According to Goshne (1948:311), M. audens
developed through invasion of the Mississippi River by M. herylUna
(Cope) stock, and by subsequent isolation and differentiation in
the lower Mississippi Valley.

Percichthyidae ^
Morone chrysops (Rafinesque) — white bass

Lepihema chrysops: Boiges, 1950 (vert, distr.; Niangua Ann, Lake Ozark).

Bamickol and Starrett, 1951:319-320 (abund.; utiliz.; Mississippi R. local.).
Morone chrijsops: Patriarche, 1953 (growth; abund.; Lake Wappapello).
Rocciis chrysops: Cleary, 1956:map 73 ( Des Moines R. local.). Burress, 1962

(abund.; Bull Shoals Res.). Cross, 1967:243 (Missouri R. local, mapped).
White bass: Hanson, 1962 (abund.; harv.; Bull Shoals Res.). Fry, 1962

(harv.; tailwaters of Table Rock, Taneycomo and Clearwater res.).

Distribution and habitat. — The centers of abundance for the
white bass in Missouri are in the Mississippi River and the large
reservoirs of the Ozarks (Map 136). It is seasonally abundant in
major tributaries of these waters during its spring spawning migra-
tions. In the Mississippi River M. chrysops is more abundant above
the mouth of the Missouri River than below ( Barnickol and Starrett,
1951:319). White bass are now fairly common in the Missouri
River, but they were not found in 1945 by Fisher (1962). Perhaps
the recent appearance of the white bass in the Missouri River
reflects changes in turbidity and other factors resulting from im-
poundment of the river upstream from Missouri. Probably the
white bass was absent from most parts of the Ozarks until con-
struction of large reservoirs. The white bass inhabits the deeper
pools of streams and the open waters of lakes and reservoirs. It
tends to avoid waters that are continuously turbid and is most
often found over a firm sandy or rocky bottom.

Zoogeography. — This species and M. mississippiensis were prob-
ably derived from marine stocks that invaded the Mississippi River
system in Tertiary time.

Morone mississippiensis Jordan and Eigenmann — yellow bass
Lahrax chrysops: Girard, 1858:31 (St. Louis).

^ Gosline (1966) has recently restricted the family Serranidae to exclude
the genus Morone, as well as some other groups, and he elevated the subfamily
Percichthyinae to family rank.

Missouri Fishes 409

Moioiic iiitcirti})t(i: Gill, 1860:118-119 (orig. descr.; St. Louis; New Orleans).
Bamickol and Starrett, 1951:319-320 (aliund.; utiliz.; Mississippi R. local.).

Distribution and habitat. — In Missouri the yellow bass is known
only from the Mississippi River and its overflow waters (Map 137),
It is nowhere abundant but seems to be more common above the
mouth of the Missouri River than below. Morone mississippiensis
is typically an inhabitant of lakes and the quiet pools and back-
waters of large rivers.

Morone saxatiJis (Walbaum), a species of this family not native
to Missouri, has been stocked in recent years in Lake of the Ozarks
and in Taum Sauk Reservoir on upper Black River. There is no
evidence as yet of natural reproduction. Individuals stocked in
Taum Sauk Reservoir have been caught below Clearwater Reser-
\"oir, many miles downstream.

Centrarchidae
Micropterus punctulatus (Rafinesque) — spotted bass

Micropterus punctulatus piinctulaius: Hubbs and Bailey, 1940 (char.; syn.;
hybrid.; Missouri local, mapped). Barnickol and Starrett, 1951:317 (Mis-
sissippi R. at Caruthersville).

Micropterus punctulatus: Funk and Campbell, 1953 (abund.; Black R.). Pa-
triarche, 1953 (abund.; growth; Lake Wappapello ) . Funk, 1957 (mov.;
Missouri streams). Burress, 1962 (abund.; Bull Shoals Res.).

Spotted bass: Purkett, 1958b: 11,33 (growth; Missouri streams). Fry, 1962
(harv.; tailwaters of Table Rock, Taneycomo and Cleanvater res.). Hanson,
1962 (abund.; harv.; Bull Shoals Res.).

Distribution and habitat. — The spotted bass has two distribution
centers in Missouri ( Map 138 ) : ( 1 ) lowland ditches and the
larger streams of the southeastern Ozarks; (2) the western periph-
ery of the Ozarks, including parts of the White, Neosho, and Mis-
souri systems. In the lowlands it is the most abundant species of
Micropterus; elsewhere in its Missouri range it is the most abundant
black bass in the larger streams. In the cool, spring- fed streams of
the central Ozarks it is replaced by the smallmouth bass; in reser-
voirs and in the prairie streams it is replaced by the largemouth.
Occurrence of the spotted bass in the Missouri River system seems
to be due to introduction into the Osage drainage at a relatively
recent date. Hubbs and Bailey (1940: map 1) plotted no records
for this species anywhere in the Missouri River system. By the early
1940's, spotted bass were rather generally distributed in the upper
Osage drainage, but it was not until some time later that it entered
the Moreau River, which empties into the Missouri River not far
upstream from the mouth of the Osage River. Micropterus punc-
tulatus is now common in the Moreau, where it hybridizes e.xten-

410 University of Kansas Publs., Mus. Nat. Hist,

sively with the native smahmouth bass. I have seen no spotted bass
from Moniteau Creek, the next tributary of the Missouri westward,
but I have seen several hybrids taken there since 1962.

Spotted bass have been stocked by Otto Fajen of the Missouri
Department of Conservation in the Lamine, Grand, Chariton,
Perche, Loutre, and Salt drainages since 1962. Reproduction has
occurred in all of these streams. A single juvenile specimen was
taken in a collection from the junction of the Missouri and Missis-
sippi rivers in September, 1969. This locality is far removed from
any streams where the spotted bass has been stocked, but the speci-
men is presumed to be a downstream straggler from one of these
streams. Probably the spotted bass will eventually become estab-
lished in all suitable streams of central Missouri.

The spotted bass generally occupies flowing waters that are
warmer and slightly more turbid than those inhabited by the
smallmouth bass. In the main channels of large rivers, the spotted
bass commonlv occurs almost to the exclusion of other black basses.
It occurs abundantly in smaller streams onh' if these are not strongly
influenced by springs.

Zoogeograpliy. — Micropfenis pnnchdatus has probably long in-
habited the lower Mississippi Valley. After initial divergence from
a common ancestral stock, M. punctuhtus invaded the range of the
closely related M. coosae in coastal drainages to the east, and there
gave rise to subspecies M. p. Iwnslialli (Hubbs and Bailey, 1940:
24).

Micropterus dolomieui Lacepede — smallmouth bass

Microptcnis dolomieu: Call, 1887:78 (West Fork of Black R. and tribs., Rey-
nolds Co.; Jacks Fork and tribs., Shannon Co.; Piney R., Texas Co.; Mera-
niec R., Dent Co.). Meek, 1891 ( Meramec R., Meramec Spring, and Big
Dry Fork near St. James; Gasconade R. and Little Piney Cr. near Arlington;
Osage Fork 6 mi. SE Marshfield; Maries R. near Dixon; James R. near
Springfield; North Fork White R. S Cabool). Borges, 1950 (vert, distr.;
Niangua Arm, Lake Ozark).

Micwptenis dolomieti dolomieu: Hnbbs and Bailey, 1940 (char.; syn.; Missouri
local, mapped). Barnickol and Starrett, 1951:318 (Mississippi R. at Clary-
ville).

Micropterus dolomieu velox: Hnbbs and Bailey, 1940 (char.; syn.; hybrid.;
Missouri local, mapped).

Micropterus dolomieui: Martin and Campbell, 1953 (abund.; hab.; Black R.).
Funk and Campbell, 1953 (abund.; Black R. local.). Funk, 1957 (mov.;
Missouri streams). Patriarche and Campbell, 1958 (abund.; growth; Clear-
water Res.). Burress, 1962 (abund.; harv.; Bull Shoals Res.). Fajen, 1962
(homing; mov.; Little Saline Cr., Miller Co.; Big Buffalo Cr., Morgan Co.).
Pflieger, 1966a ( orangethroat darters on nest; Little Saline Cr., Miller Co.).
Pflieger, 1966b (reprod.; Little Saline Cr.).

Smallmouth bass: Purkett, 1958b: 12,34,43 (growth; length-weight relat.; Mis-
souri streams). Fry, 1962 (harv.; tailwaters of Table Rock, Taneycomo,
and Clearwater res.).

Missouri Fishes 411

Distribution and Jiabitat. — The sinallmouth bass is abundant
throughout the Ozarks and occurs sparingly in the northeastern
Ozark border (Map 139). Only one specimen is known from the
Missouri River, but the smallmouth is fairly common in swift water
below na\igation dams of the upper Mississippi. Its distribution
seems to have remained essentially unchanged since the early
1940's, except for introduction into the Lamine River in 1952, and
perhaps a reduction in abundance in streams of the Ozark border.

The smallmouth bass is the ecological replacement of the spotted
bass and largemouth bass in the clear, cool, permanent-flowing
streams of the Ozarks. It tolerates only moderate turbidity and
siltation, and occurs only in streams that maintain flow in all but
the most severe drouths. It is not usually found in swift current but
is most abundant near riffles and in short pools where there is
sufficient current to keep the bottom largely free of silt. Microp-
teriis dolomieui is often associated with thick beds of water willow
(Justicia aniericana), and prefers a coarse gravel, rubble, or
boulder-strewn bottom. It also occurs along the rocky, wave-swept
shores of large Ozark reservoirs.

Zoogeography. — The smallmouth bass is more northerly in dis-
tribution than the spotted bass, and may have been localized in
the preglacial Teays or Laurentian systems. Invasion of the Ozark
Uplands by the smallmouth bass may date from one of the ice
advances of the Pleistocene, when climatic changes and drainage
modifications provided conditions suitable for dispersal from the
northeast. Hubbs and Bailey (1940) recognized two subspecies of
the smallmouth bass in Missouri, but Bailey (per. comm.) no
longer considers these to be valid.

Micropterus salmoides (Lacepede) — largemouth bass

Microptents salmoides: Jordan and Meek, 1885:14,16,17 (Missouri R. at St.
Joseph; Flat Cr. near Sedalia and/or Blackwater R. at Brownsville; Grand
R. at Clinton and/or Tebo Cr. at Calhoun). Meek, 1891 ( Meramec R.
near St. James; Piney R. at Arlington and Newburg; Osage Fork 6 mi. SE
Marshfield; Lick Fork at Mansfield). Forbes and Richardson, 1920: map
LXXXI (Mississippi R. local.). Borges, 1950 (vert, distr.; Niangua Arm,
Lake Ozark). Barnickol and Starrett, 1951:315-319 (abund.; Mississippi
R. local.). Martin and Campbell, 1953 (abund.; hab.; Black R. local.).
Funk and Campbell, 1953 (abund.; Black R. local.). Patriarche and Camp-
bell, 1953 (abund.; growth; Clearwater Res.). Patriarche, 1953 (abund.;
growth; Lake Wappapello ) . Funk, 1957 (mov.; Missouri streams). Bur-
ress, 1962 (abund.; harv.; Bull Shoals Res.). Hanson and Campbell, 1963
(linear distr.; Perche Cr. ).

Micropterus salmoides salmoides: Fisher, 1962:428 (Missouri R. local.).

Largemouth bass: Purkett, 1958b: 13,35,43 (growth; length-weight relat.;
Missouri streams). Hanson, 1962 (abund.; harv.; Bull Shoals Res.). Fry,
1962 (harv.; Table Rock, Taneycomo, and Clearwater res. tailwaters).

412 University of Kansas Publs., Mus. Nat. Hist.

Distribution and habitat. — The largemouth bass is nearly state-
wide in distribution (Map 140). In the Ozarks it is common in the
quiet backwaters and overflow pools of large streams but attains
its greatest abundance in large reservoirs. In spite of widespread
stocking in ponds throughout the state, the largemouth is rare in
natural waters of northwestern Missouri. It is common in the
lowlands, but drainage of the numerous swamps, sloughs, and
natural lakes of that region has probably diminished its abundance.

The largemouth bass tolerates varied environmental conditions,
but it is more characteristic of lentic habitats than of streams. It is
intolerant of excessive turbidity and siltation; it is replaced by one
of the other black basses in streams with high gradients and con-
tinuous strong flow. The largemouth is especially characteristic of
backwaters and oxbows along major streams, permanent pools of
intermittent upland creeks, and artificial impoundments.

Zoogeograpluj. — The largemouth bass is widely distributed in
the eastern United States, and may have been present in all or most
of the major preglacial drainages of that region.

Lepomis gulosus (Cuvier) — wamiouth

Chaenohnjttus gulosus: Forbes and Richardson, 1920:map 74 (Mississippi R.
local.). Hanson and Campbell, 1963 (linear distr.; Perche Cr. ).

Chacnohrtjitus coronarius: Barnickol and Starrett, 1951:319 (abund.; Missis-
sippi R. local). Martin and Campbell, 1953:47 (abund.; Black R.). Funk
and Campbell, 1953:72-81 (abund.; Black R. local.). Patriarche and
Campbell, 1953 (abund.; growth; Clearwater Res.).

Warmoutii: Fry, 1962 (harv.; Clearwater Res. tail\\'ater ) .

Distribution and habitat. — The warmouth is common and wide-
spread in the lowlands, and it penetrates into the southeastern
Ozarks along the major streams (Map 141). Elsewhere in the state
it is known only from scattered localities, mostly along the Missis-
sippi and lower Missouri rivers. The warmouth inhabits the back-
waters of streams but is most abundant in adjacent o.xbows. It is
occasionally found in turbid, weedless waters, but the largest popu-
lations occur in clear waters having thick growths of submergcnt
aquatic plants. In rivers it avoids areas with noticeable current.

Zoogeograpluj. — The present distribution and ecological prefer-
ences of L. gulosus suggest a wide preglacial distribution in the
lower Mississippi Valley and other stream systems of the Gulf and
Atlantic coastal plains.

Lepomis cyanellus Rafinesque — green sunfish

Lepomis cyanellus: Jordan and Meek, 1885:12,14,16,17 (Hundred and Two
R. at Maryville; Missouri R. at St. Joseph; Tabo Cr. 6 mi. E Lexington;
Flat Cr. near Sedalia and/or Blackwater R. at Brownsville; Grand R. at

Missouri Fishes 413

Clinton and/or Tebo Cr. at Calhoun). Call, 1887:79 (Bear Cr. and Mink-
son Cv., Boone CoJ.NIeek, 1891:119,122,125,130 (Big Dry Fork near
St. James; Little Dry Fork near Holla; Jones C]r. and Nlaries R. near Dixon;
Gasconade R. and Little Piney Cr. near Arlington; Niangua R. near Marsh-
field; Big Piney R. and North Fork White R. near Cahool). Everniann
and Kendall, 1895:471 (Indian Cr. and spring branch at Neosho). Borges,
1950 (vert, distr.; Niangua Arm, Lake Ozark). Barnickol and Starrett,
1951:318 (abund.; Mississippi R. local.). Martin and Campbell, 195.3
(abund.; hab.; Black R.). Funk and Campbell, 1953 (abund.; Black R.
local.). Funk, 1957 (mov.; Missouri streams). Patriarche and Campbell,
1958 (abund.; growth; Clearwater Res.). Purkett, 1958a: 126-127 (growth;
Salt R.). Burress, 1962 (abund.; harv.; Bull Shoals Res.). Fisher, 1962:
428 (Missouri R. local.). Hanson and Campbell, 1963 ( Hnear distr.; Perche
Cr.). Cross, 1967:258 (Missouri R. local, mapped).

Green sunfish: Purkett, 1958b: 14,36,43 (growth; length-weight relat.; Mis-
souri streams). Fry, 1962 (harv.; tailwaters of Taneycomo and Clearwater
res.).

Distribution and habitat. — The green sunfish is statewide in
distribution, and has occurred in more collections than any other
Missouri fish (Map 142). It is the most abundant centrarchid in
streams of the prairie region and Ozark border. In the Ozark and
lowland regions it is less abundant than the longear sunfish in most
streams. It occurs only as strays in the open channels of the Mis-
souri and Mississippi rivers, but is sometimes abundant in muddy
ponds and ditches on their flood plains. The green sunfish toler-
ates a wide range of environmental conditions, but it thrives best
where few other sunfishes occur. It tolerates extremes of turbidity,
dissolved o.xygen, temperature, and flow, and is well suited for sur-
vival in the fluctuating environment of small prairie streams. By
late summer and fall these small streams often consist of a series
of isolated, stagnant pools. Green sunfish commonly share this
habitat with the creek chub, fathead minnow, and black bullhead.

Zoogeography. — I concur with Metcalf (1966:155-156) who
suggested a western origin for the green sunfish. Its present distri-
bution and ecological preferences indicate that it evolved under
conditions not unlike those found on the Great Plains today. Fossil
material referable to the green sunfish is known from late Pliocene
deposits in the central plains (C. L. Smith, 1962:516),

Lepomis symmetricus Forbes — bantam sunfish

Distribution and habitat. — The bantam sunfish is known in Mis-
souri only from the Duck Creek Wildlife Area in Bollinger County
(Map 143), where it is common. Perhaps it was more widespread
in the lowlands before the swamps were ditched and drained. At
Duck Creek the bantam sunfish inhabits clear, quiet water having
much submerged aquatic vegetation and standing timber.

Zoogeography. — The bantam sunfish seems to be autochthonous

414 University of Kansas Publs., Mus. Nat. Hist.

to the lower Mississippi Valley. Disjunct populations along the
lower Illinois River may date from range adjustments during, and
subsequent to, the postglacial Climatic Optimum.

Lepomis punctatus (Valenciennes) — spotted sunfish

Lepomis punctatus miniatus: Martin and Campbell, 1953:47 (abund.; Black
R. ). Funk and Campbell, 1953 (abund.; Black R. local.). Patriarche and
Campbell, 1958 (abund.; Clearwater Res.).

Distribution and liabitat. — The spotted sunfish is common in the
lowlands and penetrates into the southeastern Ozarks along the
major streams (Map 144). In the lowlands L. punctatus is most
often found in clear, heavily \egetated ditches; in Ozark streams
it commonly occurs in quiet water near boulders and submerged
logs.

Zoogeograpliy. — Occurrence of a distinct subspecies (L. punc-
tatus miniatus) in the Mississippi Valley suggests a rather long
occupancy of the region. The presence of disjunct populations
along the Illinois River perhaps result from a postglacial distribu-
tional history in the Mississippi Valley similar to that of L. sijm-
metricus.

Lepomis gibbosus (Linnaeus) — pumpkinseed

Distribution and habitat. — Only two records are known for the
pumpkinseed from streams in Missouri (Map 145). A single sub-
adult was collected in 1963 from an oxerflow pool of the Meramec
River in Crawford County, and five subadults were taken in 1967
from Salt Creek, a small tributary of Grand River in Chariton
County. Perhaps these records are the result of escapement from
near-by ponds, but there is no evidence for this. The pumpkinseed
has been stocked in a few Missouri ponds, but none is known to
have been stocked near either of the two localities cited above. I
suspect that these records represent natural occurrences. Elsewhere
in its range the pumpkinseed is reported to inhabit clear, quiet
waters with dense aquatic vegetation; this is the habitat in which
it was found along the Meramec River. Salt Creek is a small,
turbid prairie stream with a silty bottom and no aquatic vegeta-
tion. Perhaps the pumpkinseeds collected in Salt Creek were strays
from nearby Swan Lake or other natural lakes along lower Grand
River. No fish collections are available from these lakes.

Zoogeography. — Missouri is on the southern edge of the range
of L. gibbosus in the Mississippi Valley. The distributional rela-
tionship of this species and the related L. microlophus suggests a

Missouri Fishes 415

northeastern origin for L. gibbostis, perhaps in the prcglacial Lau-
rentian svstem.

Lepomis microlophus (Giinther) — redear sunfish

(?) Xystwplites heros. Bean, 1880b:98 (St. Louis).

Distribution and habitat. — In natural waters, the redear sunfish
is confined to the southern half of the state, where it is rare (Map
146). Self-sustaining populations occur as a result of stocking in
many small artificial lakes and ponds elsewhere in Missouri. Like
the pumpkinseed, this fish prefers quiet, clear waters having con-
siderable aquatic vegetation.

Zoogeography. — The distributional relationship of this fish and
the closely related pumpkinseed suggests an origin for the redear
in the Mississippi Valley. Lepomis microlophus was tentatively
identified from Pliocene deposits in Nebraska (C. L. Smith, 1962:
507). If valid, this report establishes the early presence of the
redear in the Mississippi Valley, and indicates a more widespread
distribution there preglacially.

Lepomis humilis (Girard) — orangespotted sunfish

(?) Lepomis anagaUimis: Cope, 1871:440 (St. Joseph).

Lepomis humilis: Jordan and Meek, 1885:12,16,17 (Hundred and Two R. at
Maryville; Flat Cr. and/or Blackw ater R. at Brownsville; Grand R. at
Clinton and/or Tebo Cr. at Calhoun). Call, 1887:79 (Bear Cr.; Boone
Co.). Meek, 1891:125 (Maries R. near Dixon). Fowler, 1921:399 (St.
Louis). Borges, 1950 (vert, distr.; Niangua Arm, Lake Ozark). Barnickol
and Starrett, 1951:318 (abund.; Mississippi R. local.). Patriarche, 19.53
(abund.; Lake Wappapello ) . Fisher, 1962:428 (Missouri R. local.). Han-
son and Campbell, 1963 (linear distr.; Perche Cr. ). Cross, 1967:267 (Mis-
souri R. local, mapped).

Distribution and habitat. — The orangespotted sunfish occurs
over much of Missouri except for the central Ozarks (Map 147)
but is abundant only in the prairie region and Ozark border. It is
common in some silted ditches of the lowlands. Lepomis ]ui7nilis
is tolerant of siltation and high turbidity. It is commonly found in
streams with low or intermittent flow, but occurs less frequently
in the extreme headwaters of streams than does the green sunfish.
It avoids streams with high gradient, clear or cool water, and con-
tinuous strong flow.

Zoogeography. — The distribution and ecological preferences of
L. humiUs suggest a western origin (Metcalf, 1966:156-157). This
fish has extended its range eastward within historic time in response
to increased siltation associated with clearing and tilling of the
land (Trautman, 1957:506-508).

416 University of Kansas Publs., Mus. Nat. Hist.

Lepomis megalotis (Rafinesque) — longear sunfish

Lcpomis megalotis: Call, 1887:78 (Jacks Fork, Shannon Co.). Meek, 1891:
123,127,130 (Osage Fork Gasconade R. 6 mi. SE Marshfield; Shoal Cr.
near Neosho; James R. near Springfield; North Fork White R. S Cabool).
Evermann and Kendall, 1895:471 (Indian Cr. at Neosho). Patriarche,
1953 (abund.; Lake Wappapello ) . Funk, 1957 (mov.; Missouri streams).
Patriarche and Campliell, 1958:247-250 (abund.; growth; Clearwater Res.).
Burress, 1962 (abund.; harv.; Bull Shoals Res.).

Lepomis megalotis megalotis: Borges, 1950 (vert, distr.; Niangua Arm, Lake
Ozark). Martin and Campbell, 1953 (abund.; Black R.). Funk and
Campbell, 1953 (abund.; Black R. local.). Witt and Marzolf, 1954 (nest-
ing; mouth Little Niangua R.; Lake Ozark). Fisher, 1962:428 (Missouri
R. local.).

Longear: Purkett, 1958b: 15,37 (growth; Missouri streams). Fry, 1962 (harv.;
tailwater Lake Taneycomo ) .

Distrihution and habitat. — The longear svmfish is the most abun-
dant centiarchid over most of the southern half of Missouri ( Map
148). The northern and northwestern limits of its distribution
correspond closely to the boundary of the Ozark Uplands. North of
the Missouri River it occurs only in short, direct tributaries of the
Missouri River from Calloway County eastward. Lepomis mega-
lotis inhabits clear, permanent-flowing streams having bottoms
mostly of sand, gravel, or rubble. The habitats where it occurs
often have considerable aquatic vegetation, but this is not essential.
Longears are found in small headwater creeks and moderately
large rivers, but are most abundant in streams of medium size.
Lepomis megalotis also thrives in large Ozark reservoirs. Like most
sunfishes, the longear avoids strong current, being found most
commonly in pools, protected inlets, and overflow waters.

Variation and zoogeograpluj. — Lepomis megalotis exhibits con-
siderable variability in coloration and in the development and angle
of the opercular flap in different stream systems. Breeding males
from the Neosho system, Osage system, and small, direct tribu-
taries of the Missouri River in central Missouri have a broad, red-
dish stripe on the nape, whereas those from elsewhere in the state
lack a nuchal stripe. Throughout most of Missouri the opercular
flap is horizontal or is directed upward, but in populations from
the upper White River system the flap is decurved. In some popu-
lations the flap is strongly expanded posteriorly, while in others
it is nearly uniform in width. Individual variation in all these
characters is evident within populations. A different subspecies
(L. 7n. peltastes Cope) has been recognized in the Great Lakes and
northern Mississippi systems, but until variation in this species has
been thoroughly investigated, allocation of Missouri populations
to subspecies seems unwise.

Missouri Fishes 417

The strong regional differentiation of L. megalotls in the Ozark
Uplands suggests that it has long resided there. Probably it has
inhabited that region eontinuously since preglacial time.

Lepomis macrochirus Rafinesque — bluegill

Lepomis pallidus: Jordan and Meek, 1885:14 (Missouri R. at St. Joseph).
Garman, 1890:139 (Mississippi R. near Quincy, 111.). Meek, 1891:119,123
(Big Dry Fork near St. James; Little Dry Fork near Rolla; Gasconade R.
and Little Piney Cr. near Arlington; Big Piney R. at Caljool; Osage Fork
6 mi. SE NLirshfield). Evermann and Kendall, 1895:471 (spring branch
at Neosho). Forbes and Richardson, 1920:map LXXVIII (Mississippi R.
local.).

Lepomis macrochirus: Meek, 1891:122,123 (Osage Fork 6 mi. SE Marshfield;
Lick Fork at Mansfield). Borges, 1950 (vert, distr.; Niangua Arm, Lake
Ozark). Patriarche, 1953 (abund.; growth; Lake Wappapello ) . Patriarche
and Campbell, 1958 (abund.; growth; Clearwater Res.). Fisher, 1962:428
(Missouri R. local.). Burress, 1962 (abund.; Bull Shoals Res.). Hanson
and Campbell, 1963 (linear distr.; Perche Cr.). Cross, 1967:263 (Missouri
R. local, mapped).

Lepomis macrochirus macrochirus: Barnickol and Starrett, 1951:318 (abund.;

Mississippi R. local.). Martin and Campbell, 1953 (abund.; Black R.).

Funk and Campbell, 1953 (abund.; Black R. local.).
Bluegill: Purkett, 1958b: 16,38,43 (growth; length-weight relat.; Missouri

streams). Fry, 1962 (har\est, tailwaters of Table Rock, Taneycomo, and

Clearwater res.).

Distribution and habitat. — The bluegill is nearly statewide in
distribution (Map 149), but it is rare in the northwestern part of
the prairie region and in the central Ozarks. In natural waters, it
reaches its greatest abundance in oxbows along the major rivers,
and in streams of the Ozark border. Lepomis macrochirus is the
most abundant centrarchid in many impoundments. The distribu-
tion of the bluegill is strikingly similar to that of the largemouth
bass, reflecting similarities in their requirements. The bluegill is
intolerant of continuous high turbidity and siltation, and thrives
best in warm, clear waters having considerable aquatic vegetation.

Zoogeography. — The present distribution of the bluegill pro-
vides few clues concerning its preglacial distribution. Perhaps it
has had a widespread distribution in the eastern United States east
of the Great Plains since late-Tertiary time and has occurred in
all or most of the ancestral drainages that now form the Mississippi
River system.

Ambloplites rupestris (Rafinesque) — rock bass

AmhlopUtcs rupestris: Call, 1887:78 (West Fork Black R., Reynolds Co.;
Spring Valley Cr. and Jacks Fork, Shannon Co.). Meek, 1891:119,130
( Meramec R. near St. James; North Fork White R. S Cabool; James R.
near Springfield). Martin and Campbell, 19.53 (abund.; hab.; Black R.).
Funk and Campbell, 1953 (abimd.; Black R. local.). Funk, 1957 (mov.;
Missouri streams ) .

Rock bass: Purkett, 1958b: 17,.39,43 (growth; length-ueight relat.; Missouri
streams). Fry, 1962 (har\.; Taneycomo Res. tailwater).

418 Unwersity of Kansas Publs., Mus. Nat. Hist.

Distribution and habitat. — The rock bass is widespread in the
Ozarks, where it is one of the most abundant centrarchids (Map
150 ) . Some ditches of the lowlands have small populations of rock
bass, but it does not occur widely in that area. North of the Mis-
souri River it is known onlv from Lost Creek in Warren Countv.
George Morris, long-time employee of the Missouri Department
of Conservation, informed me that the rock bass was not present
in the Niangua River until its introduction in the 1930's. There are
no early records for this fish in the Missouri River system, so the
rock bass may not be native to the Osage or Gasconade systems.
At present the rock bass has only a limited distribution in the Osage
system, not including all the streams that appear suitable for it.
In Missouri the rock bass is primarily a stream fish; it constitutes
only a minor element in the fauna of large Ozark reservoirs. Un-
like some other centrarchids, it occurs only rarely in overflow pools
away from the stream channel. Permanent flow, \o\\ turbidity,
abundant cover, and silt-free bottoms characterize its habitat. In
Ozark streams it lives near boulders, logs, or dense beds of water
willow (Justicia americana), where there is a slight to moderate
current. In the lowlands it is found in the clearer ditches where
flow is strong, and dense beds of submergent aquatic vegetation
exist.

Variation and zoogeography. — Two subspecies of A. rupestris
are recognized. These are: A. rupestris ariommus Viosca of the
lower Mississippi Valley and adjacent Gulf coastal drainages, and
A. r. rupestris ( Rafinesque ) , occupying the remainder of the Mis-
sissippi Valley and the Great Lakes system. Inter grades occur in
southeastern Missouri (Bailey and Hubbs, 1949:12). Rock bass
from the upper White River system differ strikingly in color pattern
from those elsewhere in Missouri, and perhaps represent an as
yet undescribed subspecies. Occurrence of a northern and a south-
ern subspecies suggests former isolation of stocks. Perhaps an-
cestral stock of A. r. rupestris were localized in the preglacial
Laurentian system, at a time when A. r. ariommus had a more
widespread distribution in the Mississippi Valley. Drainage diver-
sions and climatic fluctuations accompanying glaciation may have
resulted in southward displacement of A. r. rupestris, and formation
of a zone of secondary intergradation.

Pomoxis nigromaculatus (Lesueur) — black crappie

Pomoxis aparoides: Forbes and Richardson, 1920:map LXXI (Mississippi R.
local.).

Pomoxis nigromaculatus: Barnickol and Starrett, 1951:315-319 (abund.; Mis-

Missouri Fishes 419

sissippi H. local.). Martin and Campl)cll, 1953 (abnnd.; Black R.). Funk
and Campliell, 1953 (alnnid.; Black R. local.). I'atriaiche, 1953 (ahund.;
Lake W'appapcllo). Patriaiche and Canipliell, 1958 (abund.; Clearwater
Res.). Fisher, 1962:428 ( Mi.ssonri R. local.). Burre.ss, 1962 (ahund.;
har\-.; Bull Shoals Res.). Hanson and Camphell, 1963 (linear distr •
PercheCr.).

Black crappie: Purkett, 1958b: 18,40 (growth; Missouri streams). Hanson
1962 (ahund.; harv.; Bull Shoals Res.).

Distribution and habitat. — The black crappie is widespread but
sporadic in distribution in Missouri (Map 151). It is most prevalent
in the large reservoirs of the Ozarks, navigation pools of the upper
Mississippi Ri\er, and natural lakes and borrow pits of the lowlands.
The black crappie is less abundant than the white crappie at most
locahties. Metcalf (1966:158) and Cross (1967:278) presented
evidence suggesting that the black crappie was introduced into
Kansas. Early collectors failed to report this fish anywhere in Mis-
souri, except from the Mississippi River, but it is likely that the
native distribution of the black crappie included most of the state.
Pomoxis nigromaculatus is less tolerant of turbidity and siltation
than the more adaptable white crappie. Clear water, absence of
noticeable current, and abundant cover in the form of submerged
timber or aquatic vegetation are the principal requirements of the
black crappie.

Zoogeography. — The present distribution and habitat require-
ments of the two crappies suggest an eastern origin for the black
crappie, probably in the preglacial Mississippi, Laurentian or south-
eastern Gulf and Atlantic coastal drainages.

Pomoxis annularis Rafinesque — white crappie

Pomoxijs hrevicauda: Gill, 1865:64-65 (orig. descr.; type local. North Grand
R., Li\ingston Co.).

Pomoxtjs annularis: Jordan and Meek, 1885:14,16 (Missouri R. at St. Joseph;
Flat Cr. near Sedalia and/or Black-water R. at Brownsville). Call, 1887:80
(Bear Cr., Boone Co.).

Pomoxis annularis: Borges, 1950 (vert, distr.; Niangua Arm, Lake Ozark).
Barnickol and Starrett, 1951:315-319 (abund.; Mississippi R. local.). Mar-
tin and Campbell, 1953 (abund.; Black R.). Funk and Campbell, 1953
(abund.; Black R. local.). Patriarche, 1953 (abund.; growth; Lake Wappa-
pello). Funk, 1957 (mov.; Missouri streams). Purkett, 1958a: 127 (growth;
Salt R.). Patriarche and Campbell, 1958:244-245 (abund.; growth; Clear-
water Res.). Burress, 1962 (abund.; Bull Shoals Res.). Fisher, 1962:428
(Missouri R. local.). Hanson and Campbell, 1963 (linear distr.; Perche
Cr.). Cross, 1967:274 (Missouri R. local, mapped).

White crappie: Purkett, 1958b: 19,41,43 (growth; length-weight relat.; Mis-
souri streams). Hanson, 1962 (abund.; harv.; Bull Shoals Res.).

Distribution and habitat. — The white crappie is nearly statewide
in distribution, but it is rare or absent in many streams of the cen-
tral Ozarks and northwestern part of the prairie region (Map 152).

420 University of Kansas Publs.^ Mus. Nat. Hist.

It is most abundant in impoundments and the navigation pools of
the upper Mississippi River. The habitat of this fish is much hke
that of the black crappie, except that P. annularis is more tolerant
of turbidity and siltation, and is less closely associated with dense
cover.

Zoogeography. — The distributional relationships and habitat
preferences of the two crappies suggest a more western origin for
F. annularis. However, fossil remains from Middle Pliocene beds
of Logan County, Kansas seem to be closer to P. nigromaculatus
than to P. annularis (C. L. Smith, 1962:514).

Centrarchus macropterus (Lacepede) — flier

Centrarchus macropterus: Barnickol and Starrett, 1951:318 (abund.; Missis-
sippi R. local.).

Distribution and habitat. — The flier is confined to the lowlands,
where it is uncommon and sporadic in distribution (Map 153). Its
preferred habitat is clear, hea\'ily \'egetated waters without notice-
able current. The swamps which formerly were present in the low-
lands probably provided a more desirable habitat than the ditches,
and it is likely that the flier is now less abundant than it was before
settlement.

Zoogeography. — Centrarchus macropterus may have occupied
the lower Mississippi Valley continuously since preglacial time or
invaded from the eastern Gulf Coast by way of temporary stream
connections formed during the Pleistocene.

Elassoma zonatum Jordan — banded pygmy sunfish

Distribution and habitat. — The pygmy sunfish is strictly a low-
land species (Map 154). It is sporadic in occurrence but is some-
times abundant. Like the flier, it inhabits quiet, clear waters with
thick growths of aquatic vegetation, and was probably more abun-
dant before the lowland swamps were ditched and drained.

Zoogeography. — The present distribution of this fish suggests
an origin in the lowlands of the Mississippi Valley or Gulf coastal
drainages to the east. The presence of related species (E. evier-
gladei Jordan and E. okefenokee Bohlke) in Georgia and Florida
indicates that stocks of this group ha\e long inhabited lowlands
of the eastern United States.

Percidae

Stizostedion vitreum (Mitchill) — walleye

Stizostedion vitreum vitreum: Borges, 1950 (\ert. distr.; Niangua Arm, Lake
Ozark). Barnickol and Starrett, 1951:314-315 (aliund.; Mississippi R.

Missouri Fishes 421

local.). Martin and Campbell, 1953 (al)und.; Black R.). Funk and Cainp-
hcll, 1953 (abund.; Black R. local.). Ki.sber, 1962:428 (Missouri R. local.).

Wallcxc: Purkett, 1958b:32 (.mowtb; Missouri streams). Fry, 1962 (harv.;
Table Rock, Tane>como and Clearwater res. tailwaters ) .

Distribution and habitat. — The walleye occurs at least occasion-
alK' in neail\- all large streams of the state (Map 155). It is most
abundant in navigation pools of the upper Mississippi River and
large streams and reservoirs of the Ozarks. An extensive spawning
run occurs each spring out of Lake of the Ozarks into the upper
Osage River. In the Missouri and lower Mississippi rivers the
walleye is far less abundant than the sauger, but replaces that spe-
cies in tributary streams. The walleye inhabits reservoirs and the
deeper pools of streams. Its requirements are much like those of
the sauger, except that the walleye is less tolerant of continuous
high turbidity, and is more successful than the sauger in impounded
waters.

Zoogeography. — This species may have been localized pregla-
cially in northern drainages, entering the Mississippi Valley by way
of stream connections that developed with glaciation. The occur-
rence of S. vitreum in late-Illinoian fossil deposits of central Kansas
(Semkin, 1966:137) documents its presence in the Mississippi
Valley by mid-Pleistocene time.

Stizostedion canadense (Smith) — sauger

Stizostedion canadense: Jordan and Meek, 1885:14 (Missouri R. at St. Joseph).
Barnickol and Starrett, 1951:314-315 (abund.; Mississippi R. local.). Mar-
tin and Campbell, 1953 (abund.; Black R.). Funk and Campbell, 1953
(abund.; Black R. local.). Fisher, 1962:428 (Missouri R. local.). Cross,
1967:285 (Missouri R. local, mapped).

Distribution and habitat. — The sauger is more restricted in dis-
tribution than the walleye, being confined almost entirely to large,
free-flowing streams (Map 156). The sauger is the more abundant
of the two species in the Missouri and Mississippi rivers; elsewhere
in Missouri the walleye generally predominates. The requirements
of the sauger are much like those of the walleye, except that it is
more tolerant of turbidity, and exhibits a greater predilection for
strong current.

Zoogeography. — This wide-ranging northern species may have
invaded the Mississippi Valley from more northern drainages dur-
ing Pleistocene time. Its predilection for large, turbid rivers sug-
gests an origin in the western part of the ancestral Hudson Bay
svstem.

422 Unr^ersity of Kansas Publs., Mus. Nat. Hist.

Perca flavescens (Mitchill) — yellow perch

Distribution and habitat. — I know of only one specimen of the
yellow perch from natural waters in Missouri (Map 157). This
specimen was taken by a fisherman in mid-April, 1962, from the
Salt River east of New London in Ralls County. No specimens were
taken by Barnickol and Starrett (1951) or Fisher (1962) in their
extensive surveys of the Mississippi and Missouri rivers, and none
has been captured by biologists of the Missouri Department of
Conservation during long-term surveys of the fish populations in
Salt River and other north Missouri streams. It therefore seems
certain that no self-sustaining populations of this species occur in
natural waters of Missouri, and occurrence here is limited to oc-
casional stragglers from farther north. I have unverified reports
of small self-sustaining populations in some artificial lakes in north
Missouri where this species was stocked. Extensive attempts were
made to establish the yellow perch in natural waters throughout
Missouri in the 1930's and before by the Missouri Fish and Game
Commission. These attempts are summarized in the annual report
of that organization for 1933 (p. 48) with the statement: "Still
another foreigner which was formerly introduced in Missouri with-
out appreciable results is the yellow perch. More than three million
have been planted in our state in the last five years without suc-
cess."

Xoogeographij. — Occurrence of this northern species in Pleisto-
cene (Illinoian?) fossil deposits from the Oklahoma panhandle
(C. L. Smith, 1954:288) documents its occurrence in the Missis-
sippi Valley by mid-Pleistocene time, and suggests a more wide-
spread southern distribution in the past. The disjunct populations
in the Apalachicola, Choctawatchee, and Mobile Bay drainages of
Alabama and Florida probably represent native occurrence ( Smith-
Vaniz, 1968:108), providing further evidence for a formerly more
widespread southern distribution.

Percina cymatotaenia (Gilbert and Meek) — bluestripe darter

Etheostoma (Hadroptents) cymatotaenia: Gilbert and Meek in Gilbert, 1888:
51-52 (oriff. descr.; types from Niangua R. and Osage Fork of Gasconade
R. near Marshfield, and Sac. R. near Greenfield). Meek, 1891:123-125
(Osage Fork 6 mi. SE Marshfield; Little Piney Cr. at Newburg; Maries R.
near Dixon ) .

Percina ci/mafotacnia: Bailcv and Gosline, 1955:table 1, p. 35 (vert, counts;
Missouri local.). Collette, 1965:575-576 (char.; Gasconade R.).

Distribution and habitat. — The bluestripe darter is confined to
the Osage and Gasconade systems (May 158), where it is rare. It

Missouri Fishes 423

ina\' have been more abundant and widely distributed before 1900
than it is today. At the tiiue of its original description it was re-
ported by Gilbert (1888:51) as "abundant in the Niangua, Osage
Fork and Sac river systems." Subsequently it has not been collected
in the Sac. The bluestripe darter inhabits clear, medium-sized
streams ha\ing permanent flow and bottoms that are mostly free
of silt. Except when spawning, it is found in quiet pools and back-
waters, often in accumulations of dead leaves and sticks.

ZoogeograpJuj. — Percina cipnatotaenia occurs disjunctly in the
Ozark Uplands and in uplands of western Kentucky. Absence of
any apparent differentiation between eastern and western popula-
tions suggests that the range disjunction is recent, possibly dating
no farther back than the Wisconsin glacial period. Perhaps the Wis-
consin ice advance resulted in the fragmentation of a continuous
distribution that existed during the preceding interglacial period;
alternati\ely the lowering of sea level and formation of steep-sided
N-alleys within the Mississippi Embayment during glaciation may
ha\e created stream conditions that permitted dispersal from one
upland area to the other b\' way of the Ohio and Mississippi rivers.

Percina maculata (Girard) — blackside darter
Percina maculata: Hanson and Campbell, 1963 (linear distr.; Perche Cr. ).

Distribution and habitat. — The blackside darter has two centers
of occurrence in Missouri (Map 159). One of these is in the ditches
and streams of the lowlands; the other is in the prairie and Ozark
border streams of central and northeastern Missouri. Five speci-
mens of P. maculata (MCZ 24539) were collected by P. R. Hoy
in Grand Ri\er near Chillicothe in 1854, suggesting a more wide-
spread distribution for this species in western Missouri before 1900.
This darter inhabits small to medium sized streams having low or
moderate gradients and permanent flow. It tolerates moderate
turbidity if there is enough current to keep the bottom mostly free
of silt. The blackside darter is usually found on gravel riflEles in a
slight current, or in short, gravelly pools where pools and rifHes
alternate in rapid succession.

ZoogeograpJuj. — The present distribution of P. maculata sug-
gests that it is autochthonous to the Mississippi Valley.

Percina sciera (Swain) — dusky darter

Distribution and habitat. — The dusky darter is one of the
commonest darters in the lowlands, and it penetrates into the
Ozarks for a considerable distance along the larger streams (Map

424 Unwersity of Kansas Publs., Mus. Nat. Hist.

160). Percina sclera is characteristic of clear, low-gradient streams
and ditches that have continuous strong flow and silt-free sand or
gravel bottoms. It is usually found on sluggish gravel-bottomed
riffles, but also occurs in quiet backwaters in accumulations of
leaves, sticks and other organic debris.

Zoogeography. — The present distribution of P. sclera suggests
an origin in the southern Mississippi Valley or adjacent Gulf coastal
drainages to the west. Clark Hubbs ( 1954 ) discussed variation and
nomenclature in this species and described P. s. apristis from the
Guadalupe River in Texas; Missouri specimens are referable to
P. s. sclera. Suttkus and Ramsey (1967:138) suggested that the
closely related Percina auroJineata evolved from a stock of P. sclera
that invaded the upper Alabama system from the Tennessee system.

Percina phoxocephala (Nelson) — slenderhead darter

Hadroptenis phoxocephalus: Jordan and Meek, 1885:17 (Grand R. at Clinton).

Percina phoxocephala: Bailey and Gosline, 1955:table 1, p. 36 (vert, counts.;
Salt R., Ralls Co.). Hanson and Campbell, 1963 (linear distr.; Perche Cr. ).

Distribution and habitat. — The slenderhead darter occurs along
the Ozark border from Spring River in Jasper County northeast\\'ard
into the upper Mississippi River and its tributaries (Map 161).
Within this area it is one of the commonest darters in large streams.
Percina phoxocephala inhabits medium-sized creeks to large rivers
that have moderately clear water and permanent flow. It is usu-
ally found on gravelly or rocky riffles in moderate to swift current.
It avoids silty or turbid streams, and those with extremely high
gradients and continuously cool water.

Zoogeography. — The presence of P. phoxocephala in the middle
part of the Arkansas system suggests either that it formerly ranged
more widely in the lower Mississippi Valley, or that it entered the
Arkansas system by way of former connections between the Arkan-
sas and Missouri systems. Either explanation seems plausible.
There are documented stream connections between the Arkansas
and Missouri systems in Kansas one or more times during the Pleis-
tocene (Frye and Leonard, 1952:180-199). Also, it seems likely
that changes in climate and stream environments in the lower
Mississippi Valley would have created conditions favorable to the
dispersal of upland fishes by way of the lower Mississippi and Ar-
kansas rivers during the Pleistocene ice advances. The present dis-
tribution of P. phoxocephala in the central Mississippi Valley sug-
gests that it would have been in a position to take ad\antage of
such conditions.

Missouri Fishes 425

Percina nasuta (Bailey) — longnose darter

Distribution and habitat. — The longnose darter is known in
Missouri only from the section of White River now covered by
Table Rock Reservoir (Map 162). It has not been collected since
impoundment, and may no longer occur in Missouri. This darter
inhabits clear, upland streams with permanent flow and well de-
fined riffles and pools. Bailey (1941:7) reported its habitat as quiet
backwaters having silt bottoms and an abundance of aquatic vege-
tation.

Zoogeography. — Percina nasuta is endemic to the White and
Arkansas systems of the Ozark and Ouachita uplands. Although
closely related to P. phoxocephala, its nearest relative seems to be
P. oxijrhijncha (Hubbs and Raney), a species known only from the
Cheat and New river systems of West Virginia and Virginia. Pos-
sibly P. nasuta and P. oxijrhijncha were derived independently from
P. phoxocephala stock. More likely they represent the remnants of
a common stock that had a wide preglacial distribution in the
Teays-Mississippi system. The range disjunction that initiated spe-
ciation probably dates from an early ice advance of the Pleisto-
cene. Perhaps the ice sheets so modified the glaciated territories
as to make them unsuitable for the common ancestral stock. It is
also possible that failure of this stock to reoccupy much of its
former range resulted from invasion of those areas by P. phoxo-
cephala from more northern or western drainages, but there is no
evidence that P. phoxocephala ever had a more northern or western
distribution than it has today.

After this report went to press. Dr. Jamie E. Thomerson called
my attention to two juvenile P. nasuta which he and his students
obtained from the St. Francis River approximately K mile upstream
from the mouth of Leatherwood Creek (T31N, R5E, SIO), Madison
County, Missouri on 2 August 1969. These specimens represent a
significant range extension for P. nasuta, and confirm that this
species is still present in Missouri waters.

I had previously examined two adult specimens supposedly
collected in the St. Francis system but doubted the authenticity of
the record. These specimens were in a collection labeled "Lake
Wappapello, 1953" (exact locality and collector unknown) at the
fisheries research laboratory of the Missouri Department of Con-
servation. Since Department biologists at that time were sampling
the White River where P. nasuta was known to occur, I suspected
a mixing of collections.

426 UNrv'ERSiTY OF Kansas Publs., Mus. Nat. Hist.

Percina caprodes (Rafinesque) — logperch

Etheostoma caprodes: Meek, 1891:123,130 (Osage Fork 6 mi. SE Marsh-
field; Lick Fork at Mansfield; James R. near Springfield). E\erniann and
Kendall, 1895:471 (Indian Cr. near Neosho).

Percina caprodes: Patriarche and Campbell, 1958 (abund.; Clearwater Res.).
Hanson and Campbell, 1963 (linear distr.; Perche Cr. ).

Percina caprodes carhonaria: Martin and Campbell, 1953 (abund.; Black R.).
Funk and Campbell, 1953 (abund.; Black R. local.). Collette, 1965:577-
578 (breeding tub.; Lamine R.).

Distribution and habitat. — The logperch is widespread in the
Ozarks and northern Ozark border (Map 163), where it is one of
the most common darters. It occurred in a few collections from
the lowland ditches in the early 1940's, but it was not collected
there during a survey made in 1964. Percina caprodes inhabits a
variety of stream types, but it does not penetrate into headwater
creeks unless they maintain large, permanent pools, and it avoids
streams that are continuously turbid, excessively silty, or that lack
well defined gravel or rubble riffles. It is most often found in the
deeper and more sluggish sections of riffles, but it also occurs in
pools if the bottom is mostly free of silt.

Variation and zoogeography. — The southern logperch, P. c. car-
honaria (Baird and Girard) occupies most of the range of this
species in Missouri, but specimens from the Mississippi River as
far downstream as Lincoln County seem to be intergrades between
this form and the northern logperch, P. c. semifasciata (DeKay).
The suspected intergradation is indicated by a more or less inter-
mediate color pattern and nearly or quite naked nape. Surprisingly,
specimens from tributaries of the upper Mississippi (Salt and
Cuivre rivers), where the logperch is rare, seem to be typical P. c.
carhonaria. The nominate subspecies, P. c. caprodes (Rafinesque)
occurs in the Ohio system and southern parts of the Great Lakes
basin. The distributional relationships of the three forms suggest
that P. c. carhonaria has long inhabited the central Mississippi
Valley. Probably P. c. semifasciata inhabited the preglacial Lau-
rentian or Hudson Bay systems, and was brought into contact with
the other subspecies as a result of drainage derangements that
accompanied glaciation. Percina c. caprodes probably inhabited
the preglacial Ohio or Teays.

Percina evides (Jordan and Gilbert) — gilt darter

(?) Etheostoma aspro: Meek, 1891:123 (Gasconade R. and Little Piney Cr.

near Arlington ) .
Hadropterus evides: Martin and Campbell, 1953 (abund.; Black R.).

Distribution and liahitat. — The gilt darter is common and wide-

Missouri Fishes 427

spread in the eastern and southern Ozarks (Map 164). It has oc-
curred in onK' a single collection from the Osage River system, and
it is not present in the Neosho system. Percina evides inhabits
clear, medium-sized to large streams with clean, silt-free bottoms
and continuous strong flow. It is most often found in the gravelly
sections of riffles and pools, in a sliglit or moderate current.

Zoogeograpliy. — Percina evides is probably autochthonous to
the Teays-Mississippi system. Perhaps the disjunctions in its pres-
ent range result from failure to reoccupy parts of its former range
after becoming localized in the Driftless Area and the Ozark and
Appalachian uplands during the Wisconsin ice advance. Frag-
mentation of its range could alternatively have resulted from the
extirpation of intervening populations by man's activities. Gerking
(1945:87) noted a recent decline in abundance of the gilt darter in
Indiana, and no specimens have been taken in Iowa since the
1890's (Harlan and Speaker, 1956:148).

Percina shuniardi (Girard)— river darter

Distribution and habitat. — The river darter is the most abundant
darter in the Mississippi River and is common in many of the
larger ditches and streams of the lowlands (Map 165). It is almost
invariably found in deep chutes and riffles where the current is
swift and the bottom is composed of coarse gravel or rubble. It
seems more tolerant of continuous high turbidity than most darters,
as indicated by its occurrence in the lower Mississippi River.

Zoogeograpliy. — The distributional relationships of this fish and
the related P. tiranidea suggest a northern origin for the river
darter, in the Teays, upper Mississippi, or Laurentian systems.

Percina uranidea (Jordan and Gilbert) — stargazing darter

Percina uranidea: Bailey and Gosline, 1955: table 37 (vert, counts; Little R.
Floodway, New Madrid Co.; ditch 1 mi. E Anniston, Mississippi Co.).
Smith, 1965:12 (Scott Co.).

Distribution and habitat. — The stargazing darter is confined to
the lowlands, where its distribution and abundance are much like
those of the related river darter (Map 166). Although the two
species seem similar in their recjuirements, both are seldom abun-
dant at the same locality. Possibly they compete intensively, and
tend to exclude each other microgeographically.

Zoogeography. — The distribution of P. uranidea suggests that
it is autochthonous to the Mississippi Valley.

428 University of Kansas Publs., Mus. Nat. Hist,

Percina copelandi (Jordan) — channel darter

Etheostoma copelandi: Meek, 1891:127 (scarce; Shoal Cr. and/or Hickory Cv.
near Neosho). Evermann and Kendall, 1895:471 (Shoal Cr. near Neosho).

Distribution and habitat. — The channel darter is common in the
larger streams of the Neosho system (Map 167), where it occurs
on sluggish riffles or in pools having enough current to create silt-
free rocky or gra\elly bottoms.

ZoogeogmpJiij. — Populations of the channel darter in the middle
Arkansas and Red river systems and in the Mobile Bay drainage
of Alabama are broadly disjunct from the main part of the range
of this northeastern species. Probably these populations are glacial
relicts, dating from southward displacement during one of the
Pleistocene ice advances. Dispersal into the Ai'kansas and Red
systems most likely was by way of the lower Mississippi Valley,
although utilization of temporary connections between the middle
Missouri (Kansas River) and upper Arkansas systems cannot be
ruled out.

Ammocr)T3ta asprella (Jordan) — crystal darter
Ammocrypta asprella: Smith, 1965:12 (St. Louis Co.).

Distribution and habitat. — The crystal darter is common in
lower Black and St. Francis rivers and the larger lowland ditches
( Map 168 ) . It is rare in the lower Meramec and Gasconade ri\ers.
Ammocrypta asprella inhabits open stretches of large, clear streams
with low or moderate gradients. It is usually found in a slight
current, on a bottom of sand or small gravel.

Zoogeographij. — This distinctixe darter is probably autochtho-
nous to the Mississippi \^alley.

Ammocrypta clara Jordan and Meek — western sand darter

Ammocrypta clara: Bailey and Gosline, 1955:table 1, p. 38 (vert, counts;
Salt R., Pike Co.; Mississippi R., Lincoln Co.). Linder, 1959 (char.;
compar.; Missouri local, mapped). Collette, 1965:583 (breeding tub.;
SaltR.).

Distribution and habitat. — The western sand darter is nowhere
abundant in Missouri, but it occurs most commonly in the upper
Mississippi Ri\er and in lowland ditches of the southeast (Map
169). In the lowlands it is less common than the scaly sand darter.
Ammocrypta clara is inxariabh' found on a bottom composed of
fine, silt-free sand. It avoids strong currents, occupying the c[uiet
margins of the channel and shallow backwaters.

Zoogeography. — This species is replaced east of the Mississippi
River by the closely related A. pellucida (Baird). The distribu-

Missouri Fishes 429

tional relationship of these two fishes suggests an origin for A.
pc'llticida in the preglacial Teays or the Laurentian system, at a
time when ancestral stocks of A. clara occupied the western or
lower parts of the Mississippi Valley.

Ammocrypta vivax Hay — scaly sand darter

Ammocrijpta vivax: Martin and Campbell, 1953 (abnnd.; Black R.). Bailey
and Gosline, 1955: table 1, p. 38 (vert, counts; Black R., Butler Co.).
Smith, 1965:12 (Cape Girardeau Co.).

Distribution and habitat. — The scaly sand darter is confined to
the lowland ditches and lower sections of large Ozark streams
entering the lowlands (Map 170). It is not abundant, but occurs
more commonly in the lowlands than the western sand darter. The
habitats of the two are much alike.

Zoogeography. — The present distribution of this species sug-
gests that it is autochthonous to the lower Mississippi Valley. It is
replaced in eastern Gulf coastal drainages by the closely related
A. beani Jordan.

Etheostoma nigrum Rafinesque — johnny darter

Bolesoma brevipinne: Cope, 1871:440 (St. Joseph).

Poecilichthijs beani: Jordan, 1885:479 (orig. descr.; type local. Tabo Cr. near
Lexington ) .

Bolesoma olmstedi maculatum: Jordan and Meek, 1885:12,16,17 (Hundred
and Two R. at Maryville; Flat Cr. near Sedalia and/or Blackwater R. at
Brownsville; Grand R. at Clinton and/or Tebo Cr. at Calhoun).

Bolesoma olmstedi ozarkaniim: Call, 1887:78,79 (nomen nudum; Jacks Fork,
Shannon Co.; Big Cr., Texas Co.; Bear Cr., Boone Co.).

Etheostoma nigrum: Meek, 1891:119,123,125,131 (Meramec R. and Big Dry
Fork near St. James; Lick Fork near Mansfield; Niangua R. near Marsh-
field; Maries R. near Dixon; North Fork White R. S Cabool). Evermann
and Kendal], 1895:471 (Indian Cr. near Neosho). Hanson and Campbell,
1963 (linear distr.; Perche Cr. ).

Distribution and liahitat. — The johnny darter is one of the
common Missouri darters, occurring over most of the state except
for the lowlands and the central Ozarks (Map 171). It is most
common in the prairie and Ozark border streams of central and
northeastern Missouri. Westward in the prairie region it becomes
progressively more scarce and spotty in distribution. Etlwostoma
nigrum occurs in streams of all sizes but is more abundant in creeks
than in rivers. It is more tolerant of turbidity than most darters,
but avoids streams that are excessively turbid and silty, and those
with high gradients or continuous strong flow of cool water. Un-
like most darters, it inhabits the quieter areas of streams rather
than riflBes. It is usually found over a sand, gravel, or rubble bot-

430 University of Kansas Publs., Mus. Nat. Hist.

torn. Perhaps the johnny darter is exckided from the lowlands by
competition from the related blnntnose and speckled darters.

Varkition and zoogeograplnj. — The johnny darter varies in squa-
mation of the nape, cheek, and breast in Missouri. Specimens from
the Neosho system and streams of the southeastern Ozarks have
these areas largely scaled, as is typical of E. n. eulepis (Hubbs and
Greene). Specimens from some streams of west-central Missouri,
especially the Lamine system, have less complete squamation, and
could be interpreted as intergrades between eulepis and nigrutn.
Elsewhere in Missouri, populations with squamation typical of the
latter subspecies occur.

Greene (1935:179) postulated a glacial refugium for E. n. eu-
lepis in the Driftless Area. If populations in southern Missouri are
properly referable to that subspecies, it also occupied a refugium
in or near the Ozark Uplands, and became isolated in these areas
by intervening populations of the typical subspecies. There is
little evidence for determining the place of origin for either form.
A related species (E. olmstcdi Storer) occurs on the Atlantic slope.

Etheostoma chlorosomum (Hay) — bluntnose darter

Etheostoma chlorosomum: Martin and Campbell, 1953 (alnind.; Black R.).

Distribution and habitat. — The bluntnose darter inhabits low-
land streams and ditches of the southeast and small prairie creeks
of northeastern Missouri (Map 172). It is otherwise known in the
state only from one locality in the Osage system and another in
the North Fork of Spring River. It seems to have declined in abun-
dance since the 1940's, especially in northeastern Missouri. Etlieo-
stoma chlorosomum occupies sluggish streams and ditches draining
lowlands and level, undissected uplands. It is found in pools and
backwaters without noticeable current, where the bottom is com-
posed of sand or organic debris. Increased siltation has probably
been a factor in the decline of this fish in the intensively cultivated
prairies of northeastern Missouri.

Zoogeography . — The present distribution of the bluntnose darter
suggests that it is autochthonous to the Mississippi Valley. The
presence of an isolated population in the upper Osage system could
have resulted from either: (1) dispersal from the Arkansas system
by way of a former stream connection, (2) range adjustments
during and subsequent to the warm, moist Climatic Optimum in-
volving westward dispersal into the lower Missouri system, or (3)
extirpation of populations elsewhere in the lower Missouri system
as a result of man's activities within historic time.

Missouri Fishes 431

Etheostoma stigmaeum (Jordan) — speckled darter

ElJico.stonui ,stif^niinu))i: Bailev and Cosliiic, 1955: table 1, p. 38 (vert, counts;
Black H., Butler Co.). Smith, 1965:12 (Cape Girardeau Co.).

Distribution and habitat. — The speckled darter has two distribu-
tion centers in the state (Map 173). One is in the lowlands and
adjacent sections of the Ozarks; the other is in the southwestern
Ozarks. The habitat of E. stigmaeum is much like that of the
johnny and bluntnose darters, except that the speckled darter oc-
curs more often in clear, high-gradient streams. It occupies slug-
gish riffles when spawning, but is otherwise confined to quiet pools
and backwaters with sandy or rocky bottoms.

Zoogeography. — Etheostoma stigmaeum has probably long oc-
cupied the lower Mississippi Valley or adjacent Gulf coastal drain-
ages. Its nearest relatives occur in the Tennessee River system.

Etheostoma tetrazonum (Hubbs and Black) — Missouri

saddled darter

Etheostoma uranidea: Meek, 1891:123 (Gasconade R. at Arlington; Little
Piney Cr. at Newbing).

Poecilichthijs variatus: Hubbs and Trautman, 1932:33 (Niangua R. near
Marshfield ) .

Poecilichthijs tctrazonus: Hubbs and Black, 1940:11-14, fig. 1 (orig. descr.;
type local. Big Niangua R. at mouth of Greasy Cr., 6 mi. SE Buftalo,
Dallas Co.).

Distribution and habitat. — The Missouri saddled darter is en-
demic to Missouri, occurring in streams draining the northern slope
of the Ozarks from the Moreau and Osage systems east to the
Meramec (Map 174). In this area it is one of the most abundant
darters. Etheostoma tetrazonum inhabits clear, high-gradient
streams with continuous strong flow. It is most often found in the
swifter riffles over a coarse gravel or rubble bottom.

Zoogeography. — Etheostoma tetrazonum is represented in the
upper Ohio River system by the closely related E. variatum (Kirt-
land). Morphological differences between the two are not great,
and they could be considered allopatric subspecies rather than
distinct species. Both occur along the glacial border without pene-
trating far into glaciated regions. Probably their common ancestral
stock had a widespread preglacial distribution from the upper
Teays system westward to the lower Missouri system, perhaps in-
cluding also the preglacial Iowa and upper Mississippi systems.
Absence of E. variatum from the Wabash, Green, and Tennessee
systems indicates that this ancestral stock was not present in the
preglacial Ohio or Tennessee systems, and did not utilize the lower

432 University of Kansas Publs., Mus. Nat. Hist.

Ohio River in its dispersal. Probably this stock had its range split
by an early ice advance of the Pleistocene, followed by differentia-
tion of E. variatum and E. tetrazonum in the Appalachian and
Ozark uplands respectively.

Etheostoma euzonum (Hubbs and Black) — Arkansas

saddled darter

Poecilichthijs eiizonus erizonus: Hubbs and Black, 1940:17-23 (orig. descr.;
types from Current R.).

Eiheosioma euzonum euzonum: Bailey and Gosline, 1955:table 1, p. 39 (vert,
counts; Little North Fork, Ozark Co.).

Eiheosioma euzonum erizonum: Bailey and Gosline, 1955:table 1, p. 39 (vert,
counts; Current R., Carter Co.).

Etheostoma euzonum: Collette, 1965:589-590 (breeding tub.; White and
Current rivers ) .

Distribution and habitat. — The Arkansas saddled darter occurs
on the southern slope of the Ozark Uplands, from the upper White
east to the Current River (Map 175). It replaces the closely related
Missouri saddled darter in streams of that region, and the habitat
of the two is not recognizably different.

Variation and zoogeography. — The Arkansas saddled darter is
endemic to the White Ri\'er system of Missouri and Arkansas. It is
quite distinct from E. tetrazonum and E. variatum, suggesting an
early separation from the stock ancestral to the latter two species.
Probably stocks of E. euzonum have inhabited the southern Ozarks
continuously since late-Tertiary time.

Two subspecies have been recognized (Hubbs and Black,
1940): E. e. euzonum (Hubbs and Black), in the White River
system upstream from Batcsville, Arkansas, and E. e. erizonum
(Hubbs and Black) in Current River farther east. Intergrades
occur in White River below Batesxille and in the lower Black
River. The two subspecies differ in squamation and color pattern.

Etheostoma histrio Jordan and Gilbert — harlequin darter

Etheostoma histrio: Bailey and Gosline, 1955: talile 1, p. 39 (vert, counts;
Black R., Butler Co.; Floodway ditches 4 mi. E Kennett, Dunklin Co.).
Tsai, 1968:fig. 1 (Missouri local, mapped).

Distri])ution and habitat. — The harlequin darter is known only
from the lowlands (Map 176). It is one of the rarest Missouri
fishes and seems to be less common and widely distributed now
than in the early 1940's. Etheostoma histrio is primarily an inhabi-
tant of moderately large, lowland streams. In Missouri it has been
found in quiet water among finely divided tree roots or in beds of
organic debris. Elsewhere it has been reported most often from
swift, gravelly riffles (Tsai, 1968).

Missouri Fishes 433

Y.ooii^co'^niphij. — The present distribution of this fish suggests
an origin in the lower Mississippi or Alabama river systems. Re-
lated species occur in the Tennessee River system and coastal drain-
ages eastward of the range of E. histrio.

Etheostoma zonale (Cope) — banded darter

Ethcostoma zonale arcansamnii: Jordan and Gilbert, 1886:5 (orig. descr.;
tvpes in part from Spring R. near Carthage and James R. near Springfield).
Martin and Campbell, 1953 (abund.; Black R.).

Etheostoma zonale: Meek, 1891:119,131 ( Meramec R. near St. James; James
R. near Springfield). Evermann and Kendall, 1895:471 (Indian Cr. near
Neosho ) .

Distribution and habitat. — In Missouri the banded darter is con-
fined to the Ozarks where it is one of the most abundant and wide-
spread darters (Map 177). It inhabits clear, high-gradient streams
having permanent strong flow. Adults are most often found on swift
rifHes over a gravel or rubble bottom, but the young commonly
occur in quiet water around aquatic vegetation or accumulations
of organic debris.

Zoogeography. — The range of the banded darter consists of
three disjunct segments, centering in the Ozark Uplands, Appalach-
ian Uplands, and the Driftless Area. Possibly this species survived
the last ice advance in all three of these areas, and failed to re-
occupy fully the glaciated regions. The Ozark isolate is sometimes
accorded subspecific status as E. z. arcansanum Jordan and Gilbert.

Etheostoma blennioides Rafinesque — greenside darter

Diplesion blennioides: Call, 1887:78 (Jacks Fork and Sinking Cr., Shannon
Co.; West Fork of Black R., Reynolds Co. ).

Etheostoma blennioides: Meek, 1891:119,123,125,130 (Meramec R. and Big
Dry Fork near St. James; Little Dry Fork near Rolla; Lick Fork near
Mansfield; Osage Fork 6 mi. SE Marshfield; Little Piney Cr. and Gasconade
R. near Arlington; Maries R. near Dixon; Sac R. and James R. near Spring-
field; North Fork White R. S Cabool). Evermann and Kendall, 1895:471
(Indian Cr. near Neosho). Martin and Campbell, 1953 (abund.; Black
R.). Patriarche and Campbell, 1958 (abund.; Clearwater Res.). Miller,
1968 (char.; syn.; subspecies; Missouri local, mapped).

Distribution and habitat. — The greenside darter is confined to
the Ozarks, where it is similar in distribution and abundance to
the banded darter (Map 178). No two Missouri fishes are more
closely associated than the greenside and banded darters; there
seems to be no significant difference in their habitat preferences
in Missouri.

Variation and zoogeography. — R. V. Miller (1968) recognized
two subspecies of the greenside darter in Misssouri. Etheostoma
b. newmani (Agassiz) occurs in streams draining the southern
slope of the Ozark Upands, whereas E. b. pholidotum Miller occurs

434 University of Kansas Publs., Mus. Nat. Hist.

in the Osage, Gasconade, and Meramec systems, as well as direct
tributaries of the Mississippi Ri\'er south of the Meramec. Inter-
grades between the two occur in the upper Gasconade system,
presumably as a result of stream capture involving headwaters of
the Current or White rivers. According to Miller, E. h. phoUdoium
evolved within its present area of occurrence in the Ozark Uplands.
He recognized a "Missouri race" which has long been present in
eastern Missouri. During one of the interglacial periods preceding
the Wisconsin, this race crossed the Missouri River and ga\'e rise
to a "Wabash-Great Lakes race." Etheostoma h. neicmani evolved in
the Tennessee and Cumberland systems. In late-Tertiary time the
"Tennessee River race" of E. h. newmani dispersed westward,
where it gave rise to an "Arkansas race," which now occupies the
Arkansas, Ouachita, Saline, and Little Red systems. The "Cumber-
land River race" moved westward at a later date and evolved into
subraces in the St. Francis, Black, and White systems. Perhaps the
successive waves of dispersal into the southern Ozarks postulated
by Miller were associated with the Pleistocene ice advances, when
erosional cycles associated with lowering of sea levels may have
created conditions favorable to the alternate dispersal and isolation
of upland fishes east and west of the Mississippi Embayment.

Etheostoma nianguae Gilbert and Meek — Niangua darter

Etheostoma niangue: Gilliert and Meek /;i Gilbert, 1888:52-53 (orig. descr.;
type local. Niangua R. near Marshfield). Meek, 1891:125 (Niangua R.
near Marshfield). Bailey and Gosline, 1955:table 1, p. 41 (vert, counts;
Barren Fork, Miller Co.). Kuhne and Bailey, 1961 (char.; Missouri local,
compiled and mapped). Gollette, 1965:595-596, fig. 6 (breeding tub.;
Big Tavern Gr., Miller Co.).

Poecilichthys nianguae: Bailey, 1948:79 (char.; range).

Distribution and habitat. — This distinctive darter is known only
from a few tributaries of the Osage River (Map 179), most com-
monly in the Maries River and Big Tavern Creek. There seem to
have been no noticeable changes in the abundance or distribution
of this rare and highly localized fish since the early 1940's. Its
status before that is unknown, but there is no indication that it was
more widespread or abundant before 1900 than it is today. Etheo-
stoma nianguae inhabits clear, high-gradient streams with perma-
nent flow. Except when spawning, it is found along the margins of
pools having clean gravel or rubble bottoms, often near water
willow or other aquatic plants. During spring, when spawning
occurs, the adults are found on riffles.

Zoogeograpluj. — The Niangua darter is closely related to the
arrow darter, E. sagitta (Jordan and Swain), which is confined to

Missouri Fishes 435

the upper Caimbt'iland and Kentucky systems east of the Mississippi
Embayinent. The aneestor of these two forms may have had a
wide preglaeial or interglacial distribution in the eentral Mississippi
Valley, in wliieh ease divergence of E. nianguae and E. sagitta
probably resulted from fragmentation of a continuous range by an
early ice advance of the Pleistocene. Alternatively, the ancestral
stock could have been localized preglacially east or west of the
Mississippi River, and dispersed across the Mississippi during one
of the ice advances of the Pleistocene. E. sagitta occurs in the
headwaters of both the Cumberland River, a tributary of the pre-
glaeial Ohio or Tennessee, and the Kentucky River, a tributary of
the preglaeial Teays. Kuehne and Bailey (1961) presented evi-
dence that this species entered the Kentucky system by stream cap-
ture. This suggests that the common nianguae-sagitta ancestral
stock did not inhabit the preglaeial Teays system, and enjoyed a
continuous east-west distribution at some time in the past by way
of the lower Ohio and Mississippi rivers.

Etheostoma whipplei (Girard) — redfin darter

Etheostoma tvJiipplci: Evermann and Kendall, 1895:471 (char.; Indian Cr
near Neosho). (?) Meek, 1891:131 (North Fork White R. S Cabool).

Poecilichtliijs ichipplii whipplii: Hubbs and Black, 1941:14-15, map 1 (syn.;
char.; Missouri local, mapped).

Distrihtition and habitat. — This darter is rare in Missouri, and
is known definitely only from the Neosho system (Map 180). Meek
(1891) reported it from the North Fork of White River, but his
record may have been based on a misidentification. Recent efforts
to find the redfin darter in southwestern Missouri have been un-
successful, and it may no longer be present in the state. According
to Cross (1967:309), E. whipplei inhabits the gravelly riffles of
streams with moderate or low gradients.

Zoogeography. — The closest relatives of E. ichipplei are E.
radiosum (Hubbs and Black) of the Red River system, and E.
artesiae (Hay) occurring disjunctly in southern tributaries of Red
River and in Gulf coastal streams east of the Mississippi River
(Moore and Rigney, 1952). The marked regional differentiation of
this complex indicates that it has long occupied the lower Missis-
sippi and adjacent Gulf drainages. Perhaps E. whipplei was derived
from stocks that were localized in the ancestral lower Arkansas or
White river systems, followed by dispersal into the upper Arkansas
system when the lower Arkansas breached the Ozark-Ouachita
divide. It is also possible that the isolation of stocks that gave
rise to E. whipplei in the Arkansas and E. radiosum in the Red is

436 Unr^rsity of Kansas Publs., Mus. Nat. Hist.

more recent, dating from diversion of upper Arkansas drainage from
the Red into the lower Arkansas during the Pleistocene. If this
explanation is correct, the common ichipplei-radiosum stock had a
preglacial distribution that included the ancestral drainage of the
central and southern plains.

Etheostoma asprigene (Forbes) — mud darter

Distribution and habitat. — The mud darter is locally common
at scattered localities in the lowlands and occurs occasionally in
the Mississippi River and the lower parts of some of its major tribu-
taries (Map 181). It inhabits lowland lakes and ponds, and the
sluggish riffles and pools of large, low-gradient streams.

Zoogeography. — Etheostoma asprigene seems to have originated
in lowlands of the lower Mississippi Valley or adjacent Gulf drain-
ages to the west.

Etheostoma juliae Meek — yoke darter

Etheostoma juliae: Meek, 1891:130-131 (orig. descr.; type local. James R.
near Springfield). Bailey and Gosline, 1955:41,42,table 1 (vert, counts;
James R., Webster Co.; North Fork Wliite R., Ozark Co.).

Distribution and habitat. — The yoke darter is endemic to the
White River system of Missouri and Arkansas (Map 182), where it
is one of the most abundant darters. It inhabits clear, high-gradient
streams with continuous strong flow, occuring only on riffles having
swift current and coarse gravel or rubble bottoms.

Zoogeography. — Unlike many Ozark endemics, E. juliae seems
to have no near-relati\'e in uplands east of the Mississippi Ri\ er or
elsewhere. Probably this distinctive darter has inhabited the
southern Ozarks continuously since preglacial time.

Etheostoma punctulatum (Agassiz) — stippled darter

Poccilichthijs punctulatus: Agassiz, 1854:304 (orig. descr.; tvpe local. Osage
R.).

Etheostoma punctulatum: Call, 1887:79 (trib. Big Cr., Texas Co.). Gilbert,
1888:60-61 (descr.; abund.; small streams, SW Missouri; Sac R. near
Greenfield; Niangua R., James R., and Osage Fork near Marshfield).
Meek, 1891:123,125 (Jones Cr. near Dixon; Big Piney R. at Cabool; Osage
Fork 6 mi. SE Marshfield; Lick Fork near Mansfield; Niangua R. near
Marshfield). Bailey and Gosline, 1955:42, table 1 (vert, counts; Shoal
Cr., Taney Co.). Collette, 1965:600 (breeding tub.; Osage R.).

Distribution and habitat. — The stippled darter is locally com-
mon and widely distributed over the western half of the Ozarks
(Map 183). It is very rare in the southeastern Ozarks, where it
may be less common than before 1900. Etheostoma punctuhtum is
most often found in small creeks and spring branches haxing clear

Missouri Fishes 437

water, pcniiancnt How, and silt-frcc bottoms. Except when spawn-
ing, it is found in quiet pools and backwaters, where it hides be-
neath large rocks or in accumulations of organic debris.

Zoogeography. — The stippled darter is most closely related to
E. cragini and E. paUididorsuni Distler and Metcalf, which are
endemic to the Arkansas and Caddo (Ouachita) systems, respec-
tively. Etheosioma ptinctuhtum is sympatric with E. cragini in the
Neosho system, but its range otherwise lies north and east of the
ranges of those two species. Perhaps E. punctidatwn evolved in
the Ozark Uplands east of the Arkansas River system, and was
brought into secondary contact with E. cragini by stream capture
between the Neosho system and the Osage or White systems. It
is also possible that it dispersed westward through the lower
Arkansas River after that stream breached the Ozark-Ouachita
uplands and captured the upper Arkansas.

Etheostoma cragini Gilbert — Arkansas darter

Etheostoma pagei: Meek, 1894:957 (ovig. descr.; type local, spring l)ranch
on grounds of U.S. Fish Hatchery at Neosho).

Etheostoma cragini: Bailev and Gosline, 1955:42, table 1 (vert, counts; Shoal
Cr., Barry Co.). Collette, 1965:599 (breeding tub.; Shoal Cr., Neosho R.).

Distribution and habitat. — The Arkansas darter is known in
Missouri only from the Neosho system of the southwestern Ozarks
(Map 184), where it is locally abundant. The habitat of E. cragini
is much like that described for the stippled darter. The Arkansas
darter occupies quiet pools of the smallest spring branches and
creeks, where the water is clear and cool, and there is an abundance
of watercress or other aquatic vegetation. The Arkansas and stip-
pled darters have never been taken together in Missouri. Where
both occur in the same area, the stippled darter tends to occupy
somewhat larger streams than the Arkansas darter.

Zoogeography. — Etheostoma cragini is endemic to the Arkansas
Ri\'er system, where it probably exolved from the same ancestral
stock that gave rise to E. punctidatum in the Ozark Uplands to the
northeast, and to E. pallididorsum in the Caddo (Ouachita) river
system to the southeast. Isolation from the stock which gave rise
to E. pallididorsum may date from the Pleistocene, when the an-
cestral lower Arkansas River cut through the Ozark-Ouachita Up-
lands (Quinn, 1958:42) and captured the present upper Arkansas
from the ancestral Red River.

Etheostoma caeruleum Storer — rainbow darter

Poecilichthys versicolor: Agassiz, 1854:304 (orig. descr.; paratypes from

Osage R.).

438 University of Kansas Publs., Mus. Nat. Hist.

Etheostoma caeruleitm: Meek, 1891:119,123,125,131 (Little Dry Fork near
Rolla; Gasconades R. syst., "Found in all streams"; Osage R. syst., "few
specimens . . . taken from each stream"; James R. near Springfield;
Bryants Cr. near Mansfield). Martin and Campbell, 1953 (abund.; Black
R.). Collette, 1965:597 (breeding tub.; Madison Co.; Gasconade R.).

Poecilichthys caendeus: Fowler, 1921:399 (Fox Cr., trib. Meramec R.).

Distribution and habitat. — The rainbow darter is one of the
most abundant and characteristic darters of the Ozarks, where it
occurs in all the principal stream systems except for the Neosho
(Map 185). North of the Missouri River this fish is definitely
known only from Lost and Charrette creeks in Warren County.
Hanson and Campbell (1963) reported E. caeruhum from Perche
Creek in Boone County, and there are specimens at UMMZ that
were supposedly collected in Richland Creek, Calloway County.
Neither of these streams now harbors this species, nor do they
appear to be suitable for it. I suspect that the Perche Creek
record is a misidentification, and the Richland Creek record results
from mixing of collections. Etheostoma caendeum inhabits clear,
high-gradient streams with permanent strong flow. It is most often
found on the swifter riffles, over a coarse gravel or rubble bottom.

Zoogeography. — Knapp ( 1964 ) recognized three subspecies of
the rainbow darter — one in the White and Black river systems of
Missouri and Arkansas, another in the Homochitto River, Missis-
sippi, and the nominate subspecies elsewhere (Map 185). He sug-
gested that E. caeruleum originated in the Ozark Uplands. The
present distribution of this fish suggests that it survived the Wis-
consin ice advance in the Ozark and Appalachian uplands, and
perhaps also in the Driftless Area. The disjunct population in the
Homochitto drainage of Mississippi indicates a more widespread
southern distribution in the past, possibly during Wisconsin time.

Etheostoma spectabile (Agassiz) — orangethroat darter

Poecilichthys spectahilis: Agassiz, 1854:304 (orig. descr.; type local. Osage R.).

Etheostoma variatum spectabile: Jordan and Meek, 1885:16,17 (Flat Cr. near
Sedalia and/or Blackwater R. at Browns\ille; Grand R. at Clinton and/or
Tebo Cr. at Calhoun ) .

Etheostoma caeruleum spectabile: Call, 1887:79 (West Fork Black R. and
Barren Cr., Reynolds Co.; Sinking Cr. and Spring Valley Cr., Shannon Co.;
Big Cr., Texas Co.). Meek, 1891: 123,127,131^ ("found in all streams,^ but
in much larger numbers than E. caeruleum," Gasconade R. syst.; "\'ery
abundant," Neosho R. syst.; James R. near Springfield; Bryants Cr. near
Mansfield). Evermann and Kendall, 1895:472 (spring branch and Indian
Cr. near Neosho).

Etheostoma spectabile: Martin and Campbell, 1953 (abund.; Black R.).
Fisher, 1962:428 (Missouri R. trib.). Hanson and Campbell, 1963 (linear
distr.; Perche Cr. ). Pflieger, 1966:139 (fry in bass nests; Little Saline
Cr., Miller Co.).

Etheostoma spectabile spectabile: Bailey and Gosline, 1955:42, table 1 (vert.

Missouin Fishes 439

counts; Bois Brule Cr., Cole Co.). Collette, 1965:600-601 (hreeding tub •
0.sage R.). Di.stler, 1968:148-159, map 1 (char.; syn.; di.str.).

(?) EiJicostonia cacnilcuni: Hanson and Campliell, 1963 (linear distr.; Perche
Cr.).

Etheostoma specfahile unipoiuni: Distler, 1968:159-162, map 1 (orig. descr.;
type local. Pigeon Cr., Dent Co.; char.; distr. ).

Etheostoma spcctahilc pulchclhnn: Distler, 1968:165-173, map 1 (intergrades
with E. .v. S))cctabile; distr.).

Etheostoma spectabile squamosum: Distler, 1968:173-177, map 1 (orig.
descr.; type local. Indian Cr., Newton Co. (char.; syn.; distr.).

Distrihution and habitat. — The orangethroat darter i.s one of the
most abundant darters in the Ozarks and northern Ozark border,
and occurs at scattered localities westward along the Missouri
Rixer to Clay and Jackson counties (Map 186). It is typically an
inhabitant of small creeks and spring branches. It tolerates mod-
erate turbidity but is most abundant in clear streams with clean
sand and gravel bottoms. Etheostoma spectabile avoids streams
with continuous strong flow, where it is largely replaced by the
closely related E. caemleum. The orangethroat darter is most often
found on sluggish riffles, or in pools where there is sufficient cur-
rent to prevent extensive siltation.

Variation and zoogeography. — The orangethroat darter exhibits
extreme geographic variation. Distler (1968) recognized five sub-
species and treated the nominate subspecies as a complex of weakly
differentiated races. Four subspecies occur in Missouri. Distler
suggested that the center of origin for E. .spectabile is in the Ozark
Uplands, because it is there that it shows the greatest amount of
racial differentiation. He indicated that a stock of E. .spectabile
dispersed into the Missouri-Platte stream system early in the Pleisto-
cene, where it evolved into E. s. pulchellum, dispersing southward
over the plains through connectives that existed then and subse-
quently. Populations of this stock became isolated in streams drain-
ing the Springfield Plateau during the extensive pluvial periods of
the Pleistocene and evolved into E. s. squamosum. Another stock
of E. spectabile was isolated in Current River and other streams
of the Black River system in post-Nebraskan time when the Mis-
sissippi Rixer shifted its course eastward from the Ozark Uplands,
and gave rise to E. s. uniporum. Distler recognized two races of
the nominate subspecies in Missouri — the "northern Ozark" race in
tributaries of the Missouri River from Osage River eastward, and
in streams of the eastern and southeastern Ozarks west to the upper
Black River; and the "southern Ozark" race in the upper White
River system. The nominate subspecies intergrades with E. s.

440 Unr'ersity of Kansas Publs., Mus. Nat. Hist.

pulcheUum (Girard) in tributaries of the Missouri River westward
from the mouth of the Osage River.

Etheostoma flabellare Rafinesque — fantail darter

Etheostoiua flabellare: Meek, 1891:123,125,127 (Jones Cr. and Maries R. near
Dixon; Little Piney Cr. near Newburg; Osage Fork and Niangua R. near
Marshfield; Shoal Cr. and/or Hickory Cr. near Neosho).

Etheostoma ftahcUarc lineolatum: Martin and Campbell, 1953:47 (abund.;
Black R.). Bailey and Gosline, 1955:43, table 1 (vert, counts; Hickory Cr.,
Newton Co. ).

Distribution and habitat. — The fantail darter occurs over most
of the Ozarks and northeastern extension of the Ozark border
(Map 187). It is abundant over most of its Missouri range except
for the southern Ozarks. Etheostoma fabellare inhabits moderately
clear, permanent-flowing streams, where it is most often found on
the swifter riffles, over a bottom of gravel or rubble.

Variation and zoogeographij. — The fantail darter is represented
in Missouri by two, and perhaps by three distinct forms. The
nominate subspecies occurs in the upper two thirds of the Cur-
rent and Black river systems. Populations from the upper White
River and its tributaries are not definitely referable to either of the
currently recognized subspecies, and perhaps are worthy of sepa-
rate taxonomic recognition. The remainder of the Missouri range
is occupied by the wide-ranging E. f. lineolatum (Agassiz), pos-
sibly intergrading with the nominate subspecies in lower Current
River.

Populations of E. /. flabellare in the southeastern Ozarks are iso-
lated from the main range of the subspecies to the north and east.
Possibly these disjunct populations are glacial relicts, dating from
a westward dispersal across the Mississippi Embayment during the
Wisconsin ice advance, when conditions there were more favorable
for upland fishes. Perhaps the undescribed White River form re-
sulted from a similar movement during an earlier ice advance. An
alternate possibility is that ancestral stocks of the undescribed
form and the nominate subspecies were widespread in the pre-
glacial Mississippi River system. Perhaps the ancestral stock of
E. f. lineolatum then occupied the Neosho Ri\ er and other upland
streams draining the Flint Hills and western slope of the Ozark-
Ouachita uplands, because its present distribution and habitat sug-
gest a western or southwestern origin. Diversion of western drain-
ages into the middle Missouri and lower Arkansas systems late in
the Pleistocene would have permitted eastward dispersal of E. f.
lineolatutn, and partial replacement of E ./. flabellare and its White
River counterpart.

Missouri Fishes 441

Etheostoma gracile (Girard) — slough darter

Ethcostonui ^lacilc: Collette, 1962:132-149 (char.; syn.; relat.; Missouri local,
compiled and mapped).

Distribution and habitat. — The slough darter is common in the
lowlands, and occurs rarely in the southwestern and northeastern
Ozark border (Map 188). The distribution and habitat require-
ments of the slough darter are much like those of the bluntnose
darter. Both are confined to sluggish streams and ditches of low-
lands and le\'el, undissected uplands. As its name suggests, the
slough darter commonly occurs in oxbows and other overflow
waters away from the main stream. Increased siltation and the
drainage of lowland swamps and sloughs have probably been un-
favorable to the slough darter.

ZoogeograpJiy. — The present distribution of the slough darter
suggests an origin in the lower Mississippi Valley or adjacent Gulf
drainages. Eastward in the Alabama and Tombigbee rivers E.
gracile is replaced by the closely related E. zoniferum (Hubbs and
Cannon ) . Occurrence of an isolated population of E. gracile in the
upper Osage system is subject to the same explanations as similarly
isolated populations of E. chlorosomiini (see account of that spe-
cies ) .

Etheostoma proeh'are (Hay) — cypress darter

Etheostoma proeUarc: Bailey and Gosline, 1955:44, table 1 (vert, counts;
floodway ditches 4 mi. E Kennett, Dunklin Co.; drainage ditch 1 mi. E
Anniston, Mississippi Co. ) .

Distribution and habitat. — FJheostoma proeliare is the most
common darter in the lowlands (Map 189). It is characteristic
of clear, heavily vegetated ditches without noticeable current, oc-
curring along the swifter ditches only in quiet, marginal waters.

Zoogeography. — The distributional relationship of this species
and the closely related least darter (E. microperca) suggests that
the cypress darter had a southern origin, probably in the lowlands
of the Mississippi Valley or adjacent Gulf drainages.

Etheostoma microperca Jordan and Gilbert — least darter

Etheostoma microperca: Meek, 1891:123 (Jones Cr. near Dixon. Cross, 1967:
323 (Osage Fork Gasconade R., Webster Co. ).

Distri])ution and habitat. — The least darter is widely distributed
along the northern and western margin of the Ozarks (Map 190).
Because of its specialized requirements, it is sporadic in occurrence,
but it is often abundant where it occurs. Etheostoma microperca
replaces the cypress darter in upland streams; the habitats of the

442 UxVrvERSiTY of Kansas Publs., Mus. Nat. Hist,

two are otherwise similar. The least darter is invariably found in
clear, quiet, heavily vegetated waters: pools of small creeks with
permanent flow, and spring-pools and seeps along the flood plains
of larger streams.

ZoogeograpJiy. — The distributional relationship of this fish and
the cypress darter suggests a northern origin for the least darter;
perhaps E. microperca was localized preglacially in the Laurentian
system, and entered the Mississippi Valley by way of stream con-
nections that developed with glaciation. Ozark populations of the
least darter are broadly disjunct from the remainder of the range
and perhaps are glacial relicts. The Ozark Uplands probably
served as a glacial refugium for this fish during the Pleistocene.
The presence of isolated populations in the Neosho system pro-
\'ides further e\'idence for a former stream connection between the
Arkansas and middle Missouri systems.

Sciaenidae
Aplodinotus grunniens Rafinesque — freshwater drum

AmhJodon lineatus: Agassiz, 1854:307-308 (orig. descr.; Osage R.).

Amhlodun grunniens: Girard, 1858:98 (St. Louis).

Aplodinotus grunniens: Jordan and Meek, 1885:14,16,17 (Missouri R. at St.
Joseph; Flat Cr. near Sedalia and/or Blackwater R. at Brownsville; Grand
R. at Clinton and/or Tebo Cr. at Calhoun). Carman, 1890:136 (Mississippi
R. near Quincy, 111.). Borges, 1950 (\ert. distr.; Niangua Arm, Lake
Ozark). Berner, 1951: table 5 (commerc. catch; Missouri and Mississippi
R.). Barnickol and Starrett, 1951:311-312 (abund.; Mississippi R. local.).
Funk and Campbell, 1953 (abund.; Black R. local.). Patriarche, 1953
(abund.; growth; Lake Wappapello). Funk, 1957 (mov.; Missouri streams).
Purkett, 1958a: 119-121 (growth; import.; Salt R.). Witt, 1960 (length
and weight in relat. to otolith length and weight; Mississippi R. at Hanni-
bal). Fisher, 1962:428 (Missouri R. local.). Burress, 1962 (abund.; harv.;
Bull Shoals Res.). Hanson and Campbell, 1963 (linear distr.; Perche Cr. ).
Cross, 1967:324 (Missouri R. local, mapped).

Freshwater drum: Purkett, 1958b: 20,42,43 (growth; length-weight relat.;
Missouri streams). Fry, 1962 (harv.; Table Rock, Taneycomo, and Clear-
water res. tailwaters).

Distribution and habitat. — The freshwater drum occurs over
much of the state (Map 191), but it is most abundant in the Mis-
souri and Mississippi ri\ers and the downstream sections of their
larger tributaries. It is scarce in most Ozark streams, but is rather
common in the large reservoirs of that region. This fish is an in-
habitant of large lakes, impoundments, and the pools of large
streams.

Zoogeograpluj. — The present distribution of this \'agile, big-river
fish indicates an origin in the preglacial Mississippi system or other
Gulf drainages to the west. It is the sole freshwater representative

Missouri Fishes 443

of a large marine family, and probably was derived from marine
stock that invaded freshwater streams in pre-Pleistocene time.

Cottidae

Cottus bairdi Girard — mottled sculpin

(in part ?) Uranidca richardsom: Call, 1887:79 (West Fork of Black R. and
tribs., Reynolds Co.; Jacks Fork and tribs., Shannon Co.; Piney R., Texas
Co.; Meramec R., Dent Co.).

(in part?) Cottus bairdi: Meek, 1891:119,123,125,131 (Meramec Spring and
Meramec R. near St. James; Big Piney and Little Piney Cr. near Arlington;
Osage Fork and Xiangna R. near Marshfield; Jones Cr. and Maries R. near
Dixon; Lick Fork and Bryants Cr. near Mansfield; Sac R. and James R.
near Springfield). Fowler, 1921:399 (Fox Cr. near St. Louis).

Distribution and habitat. — The mottled sculpin is common and
widespread in the Ozarks, occurring in all the principal stream
systems except the Neosho (Map 192). North of the Missouri
River it is known only from Lost Creek in Warren County. Cottus
bairdi inhabits clear streams and spring branches with a permanent
strong flow of cool water. It is found in riffles as well as pools,
usually in association with thick growths of water cress or other
cover.

Zoogeography. — Populations of the mottled sculpin in the
Ozark Uplands are disjunct from the main range of the species to
the north and east. Ozarkian populations are probably glacial
relicts. Robins (1954:303) indicated that the bairdi group had as
its eastern center of dispersal the old Teays system and perhaps
other portions of the preglacial upper Mississippi system. Glacia-
tion permitted southwestward dispersal, followed by isolation
of Ozarkian populations when the ice sheets again retreated north-
ward. Robins (1954) recognized another species of this group in
the Ozark Uplands; this species is still technically undescribed.

Cottus carolinae (Gill) — banded sculpin

Cottus bairdi: Meek, 1891:127 (Hickory and Shoals crs. near Neosho). Ever-

mann and Kendall, 1895:472 (Indian Cr. near Neosho).
Cottus carolinae: Martin and Campbell, 1953 (abund.; Black R.).

Distribution and habitat. — The banded sculpin occurs through-
out the Ozarks, and is known from a few localities north of the
Missouri River in Lincoln and Pike counties (Map 193). In the
Ozarks it is about as common and widely distributed as the mottled
sculpin. The requirements of C caroUnae are much like those of
C. bairdi, and the two are often found together. Howe\'er, the
banded sculpin tends to occupy the larger and warmer Ozark
streams, the mottled sculpin the smaller creeks and spring branches.

444 University of Kansas Publs., Mus. Nat. Hist.

Zoogeography. — The distribution of the banded sculpin sug-
gests that it sul■^'i\'ed the Wisconsin ice advance in the Ozark and
Appalachian uplands, from which it has scarcely penetrated the
glaciated regions. According to Robins (1955:302-303), the Ten-
nessee River system was the dispersal center for the carolinae
group, and glaciation has had little effect on their dispersal and
distribution.

Summary and Conclusions

In this report, information about the distribution of Missouri
fishes is summarized, and the patterns of fish distribution in the
state are interpreted in terms of the environmental conditions and
historical factors that have determined their development.

The fish fauna of Missouri is diverse; 191 species are listed for
the state. From analyses of range limits and the species composi-
tion of collections, four fish faunal regions are recognized. Three
of these (Ozark, lowland, and prairie) correspond closely to major
physiographic subdivisions of Missouri (Ozark Uplands, South-
eastern Lowlands, and Osage and Dissected Till plains, respec-
tively). The fourth faunal region (big river) includes the Missouri
and Mississippi rivers and the down-stream portions of their major
tributaries. The faunal regions are separated by broad ecotones,
designated as an "Ozark border" where fishes characteristic of
different faunal regions meet and mix in complex fashion. "Lines
of best fit" are also designated; these indicate zones of abrupt
faunal change within the Ozark border.

Missouri fishes are classified into four primary and three secon-
dary faunal groups. Primary faunal groups include species ha\'ing
their distribution centered in a single faunal region; secondary
faunal groups include species that are more widespread. Four
rare species are left unclassified. The Ozark faunal group is the
largest of the primary faunal assemblages, including 65 species.
Thirteen of these are endemic to the Ozark Uplands. Fishes of the
Ozark group are confined primarily to regions underlain by lime-
stone bedrock, and inhabit clear streams ha\'ing permanent flow
and a predominance of coarse gravel or rock bottoms. Thirt)'-eight
species are in the lowland faunal group. Lowland species occur
in areas of low relief, inhabiting lentic environments and low-
gradient streams having clear water and bottoms composed mostly
of sand, fine gravel, and organic debris. The prairie faunal group
is the smallest of the primary faunal assemblages, including 18

Missouri Fishes 445

species. Prairie species inhabit streams subject to wide fluctuations
of turbidity, tcniperature, and volume of How. Thirty species and
one subspecies belong to the big river faunal group. The distribu-
tion of big river fishes is correlated with gradients in environmental
factors (notably turbidity, current velocity, and bottom type)
within the Missouri and Mississippi rivers. Certain fishes are nearly
equall)- abundant in two faunal regions. Seven of these belong to
the Ozark-lowland faunal group. All are characteristic of the quiet
pools of moderately clear, permanent-fiowing streams. Six species
belong to the Ozark-prairie faunal group. The requirement for
coarse gravel or rock bottoms seems to be the principal factor
controlling the distribution of fishes in this faunal group. Twenty-
tliree species are placed in a wide-ranging faunal group. All spe-
cies of this group are characteristic of lentic environments and
the quiet pools of streams, but otherwise have broad tolerances for
many environmental factors. Fishes of this group disperse readily
through big rivers, and most have been favored by man's activities.

Competition seems to be an important factor in controlling the
distribution of certain fishes, as indicated by: (1) the comple-
mentary distribution patterns of species that seem to have similar
requirements, (2) the occurrence of certain species outside of their
usual habitat in faunally depauperate areas, and (3) the avoidance
of faunally-rich habitats by certain species that otherwise seem to
have broad environmental tolerances. Climate is also thought to
limit the distribution of fishes in the state, but its effects are largely
obscured by gradients in other environmental factors.

Marked variations occur in faunal diversity within Missouri.
Stream systems in the prairie faunal region have the fewest species;
streams that span the boundaries between faunal regions are
faunally rich. Lack of faunal diversity in the prairie faunal region
seems to be related to the instability of aquatic environments there.

The character of the Missouri fish fauna has been profoundly
affected by changes in physiography, drainage relationships, and
climate since the late Tertiary. Meager fossil evidence and what
can be deduced concerning environmental condition suggests that
the late-Tertiary fish fauna of Missouri was much like that now
found on the coastal plain of the southeastern United States. The
strong regional differentiation that characterizes the present fauna
was probably little developed, although evidence is presented sug-
gesting that certain species were already localized in the Ozark
region by the late Tertiary.

Renewed uplift of the Ozark Uplands, withdrawal of the sea

446 University of Kansas Publs., Mus. Nat. Hist.

from the Mississippi Embayment, and a shift towards a cooler and
less humid climate near the close of the Tertiary resulted in chang-
ing environmental conditions in Missouri, and the fish fauna began
to take on a more modern character. The intrusion of continental
ice sheets into the central United States during the Pleistocene had
profound effects on the fish fauna, both within and beyond the
glaciated regions. Glaciation was accompanied by the southward
displacement of species ranges, and drainage changes that resulted
directly or indirectly from glaciation permitted a mixing of faunal
elements formerly localized in separate stream systems. Certain
Missouri fishes may have been localized preglacially in the Hudson
Bay or Laurentian systems north of the Mississippi Valley, dispers-
ing southward as avenues became available. Examples are: silver
lamprey, northern brook lamprey, pallid sturgeon, goldeye, northern
pike, sturgeon chub, sicklefin chub, northern flathead chub, com-
mon shiner, brassy minnow, western silvery minnow, northern
redhorse, pumpkinseed, northern logperch, and least darter. All
are primarily northern in distribution, and are represented south-
ward by forms that appear to be autochthonous to the Mississippi
Valley. Large springs in the Ozark Uplands have sustained ecologi-
cal conditions favorable to survival of northern species displaced
southward during the Pleistocene. Disjunct populations of five
fishes (northern brook lamprey, spotfin shiner, channel darter, least
darter, and mottled sculpin) in the Ozark Uplands are considered
to be glacial relicts.

Alternating cycles of erosion and aggradation in the lower Mis-
sissippi Valley as a result of sea level fluctuations during the Pleisto-
cene were accompanied by changing ecological conditions that pro-
vided for the alternate dispersal and isolation of fishes in uplands
east and west of the Mississippi Embayment. East-west disjuncts
whose present distributions may have resulted from this process
include: least brook lamprey, telescope shiner, whitetail shiner,
bluntface shiner, studfish, southern cavefish, bluestripe darter,
greenside darter, and banded darter.

Glaciation was accompanied by restriction of ranges as well as
by range extensions. Certain fishes that had a widespread pre-
glacial distribution seem to have given rise to geminate species
after becoming localized in uplands east and west of the Mississippi
River during an early ice advance of the Pleistocene. Ozarkian
species that may have been derived in this manner include: Ozark
shiner, Ozark cavefish, longnose darter, Missouri saddled darter,
and Niangua darter. Disjunct populations of the Ozark minnow and

Missouri Fishes 447

largescale stoneiollcr in the Ozark Uplands and the Driftless Area
of Wisconsin also may indicate a more widespread preglacial or
interglacial distribution.

Certain fishes formerly localized in the preglacial drainage of
the central and southern plains (including the Flint Hills and the
western slope of the Ozark-Ouachita uplands) have gained access
to the central Mississippi Valley when drainage of that region was
diverted eastward into the lower Arkansas and Missouri Rivers.
Examples are: plains killifish, plains minnow, duskystripe shiner,
bluntface shiner, \\'estern slim minnow, and Arkansas darter. All
members of this group are primarily western in distribution, and
most are represented eastward by forms that seem to be autochtho-
nous to the central Mississippi Valley. Other fishes that are primarily
northern or eastern in distribution occur disjunctly in the upper
Arkansas, and may have dispersed southwestward by way of stream
connections between the Arkansas and Missouri systems. Included
are: spotfin shiner, blacknose shiner, Topeka shiner, stonecat,
plains topminnow, slenderhead darter, channel darter, johnny
darter, and least darter. Dispersal of some of these by way of the
lower Mississippi and Arkansas rivers cannot be ruled out, although
none of them now occur in that area.

Climatic fluctuations since retreat of the Wisconsin ice sheet
have resulted in minor adjustments in the ranges of some fishes.
Disjunct northern populations of the ironcolor shiner, weed shiner,
bantam sunfish, and spotted sunfish are perhaps relicts from the
warm, moist Climatic Optimum. Disjunct eastern populations of
the bigmouth shiner may be relicts from the warm, dry Xerothermic
Interval.

Habitat modifications by man during the last two centuries
have resulted in rapid and dramatic changes in the distribution of
many fishes. Five species that occurred in the state as recently as
30 years ago may no longer be present. Numerous others are now
more restricted in distribution. Several species are more abundant
and widespread than formerly. One exotic (carp) introduced in
the late 19th century is now a dominant element in the Missouri
fish fauna.

Acknowledgments

Many individuals have contributed materially to this study since
its inception nearly 30 years ago. Much of the distributional ma-
terial here reported for the first time was acquired through the

448 University of Kansas Publs., Mus. Nat. Hist,

efforts of collectors mentioned in the historical review presented
elsewhere in this report. Deserving special mention is George V.
Harry, not only for his extensive efforts in the field, but also for
performing the laborious task of sorting and identifying the speci-
mens resulting from surveys of the 1930's and 1940's, and for com-
piling virtually all of the pertinent distributional data that had
accumulated up to about 1946. Harry was guided and materially
aided in his work by Carl L. Hubbs and Reeve M. Bailey. When
I first became associated with the study in the spring of 1961, Dr.
Bailey spent much time apprising me of taxonomic and distribu-
tional problems relating to fishes found in Missouri and collabo-
rated with me in working up a mimeographed list of the fishes
of the state.

The Missouri Department of Conservation has supplied funds,
equipment, and personnel for conducting field work since the study
began. The bulk of my studies have been carried out while em-
ployed by that agency, and have been financed in part through the
Dingell-Johnson Program (Project F-l-R). Dingell- Johnson funds
also helped to defray costs for publishing this report. My co-
workers with the Department have assisted me in many ways.
For aiding me in the field or for providing me with specimens
or records resulting from their studies I am indebted to Otto F.
Fajen, Thomas R. Russell, William H. Dieffenbach, John W. Robin-
son, George G. Fleener, James R. Whitley, James P. Fry, and Willis
D. Hanson. My many field assistants worked long hours and suf-
fered no little discomfort in helping me to obtain collections. John
Humpf helped draft the distribution maps for this report. John L.
Funk, Superintendent of Research, and Charles A. Purkett, Jr.,
Chief of Fisheries, have facilitated my studies in ways too numer-
ous to mention.

Frank B. Cross has adxised me in my studies since 1965, and
saw to it that facilities and funds for carrying on my research were
available while I was on the campus at the University of Kansas.
Marvin E. Braasch and Ronald Nolan assisted me in obtaining
additional fish collections from western Missouri in 1965.

The following individuals made specimens in their care avail-
able for study: John D. Black, Northeast Missouri State College;
Reeve M. Bailey, University of Michigan Museum of Zoology;
Arthur Witt, Jr., University of Missouri; Jamie Thomerson, South-
ern Illinois University; Leonard Durham, Eastern Illinois Univer-
sity; and Philip W. Smith, Illinois Natural History Survey. Dr.
Smith also supplied numerous locality records for fishes in the

Missouri Fishes 449

Mississippi Ri\cr where it forms the eoininon boundary of Missouri
and Ilhnois.

For critically reading the manuscript and providing many valu-
able suggestions I am indebted to Frank B. Cross, Gerald R. Smith,
J. Knox Jones, William E. Duellman, and Philip S. Humphrey.
Robert E. Jenkins made valuable suggestions on the range maps
for species of Moxostoma. Tom Swearingen lettered the distribution
maps and assisted with other illustrations.

My wife, Jo Ann, has been a constant source of encouragement,
and assisted in preparation of the manuscript.

450 University of Kansas Publs., Mus. Nat. Hist.

Literature Cited

Agassiz, L.

1854. Notice of a collection of fishes from the southern bend of the Ten-
nessee River, Alabama. Amer. Jour. Sci., Arts, 17 (2nd Ser. ):297-
308, 353-369.

1855. Synopsis of tlie ichthyological fauna of the Pacific Slope of North
America. . . . Amer. Jour. Sci., Arts, 19 (2nd Ser.):71-99, 215-231.

Al-Rawi, a. H., and F. B. Cross

1964. Variation in the plains minnow, Hybognathus placitus Girard. Trans.
Kansas Acad. Sci., 67(1) :154-168. June 12.
Bailey, R. M.

1941. Hadropicrus nastitus, a new darter from Arkansas. Mus. Zool., Univ.
Micliigan, Occ. Pap. 440:1-8. June 18.

1948. Status, relationships, and characters of the percid fish, PoecilichtJiys
sagitta Jordan and Swain. Copeia, 1948(2) :77-85. June 30.

1954. Distribution of the American cyprinid fish Hijhognathus hankinsoni

with comments on its original description. Copeia, 1954(4):289-

291. October 29.
1956. A re\ised fist of the fishes of Iowa, with keys for identification (pp.

326-377 in Iowa Fish and Fishing, by J. R. Harland and E. B.

Speaker). Iowa Cons. Comm., Des Moines.
1959. Parasitic lamprevs (Ichtlii/oiniizon) from the Missouri River, Missouri

and South Dakota. Copeia, 1959(2) : 162-163. July 24.
Bailey, R. M., and M. O. Allum

1962. The fishes of Soutli Dakota. Mus. Zool, Univ. Michigan, Misc.
Publ. 119:1-131. June 5.

Bailey, R. M., and F. B. Cross

1954. River sturgeons of the American genus Scaphirhynchus: characters,
distribution, and s\nonvmv. Papers Michigan Acad. Sci., Arts, Let.,
39:169-208.

Bailey, R. M., axd W. A. Goslixe

1955. Variation and systematic significance of vertebral counts in the
American fishes of the familv Percidae. Mus. Zool., Uniw Michigan,
Misc. Publ. 93:1-44. September 28.

Bailey, R. M., axd C. L. Hubbs

1949. The black basses (Micropteriis) of Florida, with description of a
new species. Nhis. Zool., Uniw Michigan, Misc. Publ. 516:1-40.
February 25.

Bailey, R. M., H. E. Wixx, and C. L. Smith

1954. Fishes from the Escambia Ri\er, .\labama and Florida, with ecologic
and taxonomic notes. Proc. Acad. Nat. Sci., Philadelphia, 106:109-
164. December 17.
Ballard, W. W., axd R. G. Needham

1964. Normal embryonic stages of Poh/odon spathula (Walbaum). Jour.
Morph., 114(3) :465-478. May."
Barnickol, P. G., axd R. S. Campbell

1952. Summarv of selected pond studies in Missouri. Jour. Wildl. Mgt.,
16(3):270-274. July.
Barxickol, p. G., and \V. C. Starrett

1951. Commercial and sport fishes of the Mississippi Ri\er between
Caruthersville, Missouri, and Dubuque, Iowa. Illinois Nat. Hist.
Surv., Bull. 25(5):267-350. September.

Bayxe, C. K., and O. S. Fent

1963. The drainage historv of the upper Kansas River Basin. Trans. Kan-
sas Acad. Sci., 66(3') :363-377. December 17.

Beax, T. H.

1880a. Description of a new .species of Amiurus (A. ponderosus) from the
Mi,ssi.ssippi River. Proc. U.S. Natl. Mus., 2( 1879) :286-290.
March 25.

1880b. Check-list of duplicates of North American fishes distributed by the
Smithsonian Institution in behalf of the United States National
Museum, 1877-1880. Proc. U.S. Natl. Mus., 3:75-116.

Missouri Fishes 451

Beckman, H. C, and N. S. Hinchey

1944. The large springs of Missouri. Mis.somi Ceol. Sur\. and Water
Res., 2nd Ser., 29:1-141.
Bennitt, R., and VV. O. Nagel

1937. A survey of the resident game and furl)earers of Missouri. Univ.
Mi.ssouri Studies, 12(2): 1-215.
Berner, L. M.

1951. Limnologx' of the lower Missouri River. Eeol., 32(1):1-12.
Borges, M. H.

1950. Fish distribution studies, Niangua Arm of the Lake of the Ozarks,
Mi.ssouri. Jour. Wild). Mgt., 14(l):16-33. January.

Braasch, M. E., and p. VV. Smith

1965. Relation,ships of the topminnows Fiindulus notatus and Ftuidulus

olivaceus in the upper Mississippi River Valley. Copeia, 1965(1):

46-53. March 18.
Branson, B. A.

1964. Additions to the known Kansas fish fauna. Trans. Kansas Acad. Sci.,
66(4):745-746. February 25.

Branson, B. A., and R. Hartmann

1963. Lepisosteus oculatus (Rafinesque) in Kansas. Copeia, 1963(3):
591. September 25.
Branson, E. B.

1944. The geology of Missouri. Univ. Missouri Studies, 19(3): 1-535.
Braun, E. L.

1951. Plant distribution in relation to the glacial boimdary. Ohio Jour.
Sci., 51 (3): 139-146. May.

Bretz, J. H.

1965. Geomoiphic history of the Ozarks of Missouri. Missouri Geol. Surv.
and Water Res., 2nd Ser., 41:1-147.

Brezner, J.

1958. Food habits of the northern river carpsucker in Missouri. Progr.
Fish-Cult., 20(4): 170-174.
Brown, J. L.

1956. Identification and geographic variation of the cyprinodont fishes
Funduhis olivaceus (Storer) and Fiindulus notatus (Rafinesque).
Tulane Stud. Zool., 4(3) : 119-134. February 3.

BURRESS, R. M.

1962. A quantitative creel census on two arms of Bull Shoals Reservoir,
Missouri. Proc. 16th Conf. Southeastern Assoc. Game and Fish
Comm.: 387-398.
Call, R. E.

1887. Memoranda on a collection of fishes from the Ozark Region of
Missouri. Proc. Davenport Acad. Nat. Sci., 5:73-80.
Carlander, H. B.

1954. A history of fish and fishing in the upper Mississippi River. Upper
Mississippi River Cons. Comm., 96 pp.

Cleary, R. E.

1956. The distribution of the fishes of Iowa (pp. 276-324 in Iowa Fish
and Fishing, by J. R. Harlan, and E. B. Speaker). Iowa Cons.
Comm., Des Moines.
Coker, R. E.

1930. Studies of common fishes of the Mississippi River at Keokuk. Bull.
U.S. Bur. Fi.sh., 45:141-225.
Collette, B. B.

1962. The swamp darters of the subgenus Hololepis (Pisces, Percidae).

Tulane Stud, in Zool., 9(4) :11.5-211. March 30.
1965. Systematic significance of breeding tubercles in fishes of the family
Percidae. Proc. U.S. Natl. Mus., 117(3518) :567-614.
Collier, J. E.

1955. Geographic regions of Missouri. Ann. Assoc. Amer. Geol., 55(4):
368-392.

Cope, E. D.

1871. Recent reptiles and fishes. Report on the reptiles and fishes obtained

452 University of Kansas Publs., Mus. Nat. Hist.

by the naUiralists of the expedition. In Preliminary report, U.S.
Geol. Survey Wyoming, 8:432-442.
Cox, U. O.

1905. A revision of the cave fishes of North America. Report U.S. Bur.
Fish., (1904): 377-393.
Cross, F. B.

1967. Handbook of fishes of Kansas. Mus. Nat. Hist., Univ. Kansas,
Misc. Publ. 45:1-357. April 24.
Cross, F. B., and A. L. Metcalf

1963. Records of three lamprevs (Ichtht/omtizon) from the Missouri Ri\er
system. Copeia, 1963(1 ):187. NIarch 30.
Cross, F. B., axd W. L. Mixckley

1958. New records of four fishes from Kansas. Trans. Kansas Acad. Sci.,
61(1):104-108. April 23.

Crossmax, E. J.

1966. A taxonomic study of Esox americanus and its subspecies in eastern
North America. Copeia, 1966( 1 ):l-20. March 22.
DeKay, J. E.

1842. Fishes. Zoology- of New- York or the New- York fauna, pt. 4:i-xv,
1-415.
Deacon, J. E.

1961. Fish populations, following a drought, in the Neosho and Marias
des Cygnes ri\'ers of Kansas. Mus. Nat. Hist., Univ. Kansas, Publ.
13(9):359-427. August 11.

Deevey, E. S., Jr.

1949. Biogeography of the Pleistocene. Bull. Geol. Soc. Amer., 60:1315-
1416. September.
Dice, L. R.

1943 The biotic provinces of North America. Univ. Michigan Press, Ann
Arbor, 78 pp.
DORF, W.

1959. Climatic changes of the past and present. Mus. Paleont., Univ.
Mich., Contr. 13:181-210. April 17.

DXINIERIL, A.

1870. Histoire naturelle des poissons, ou Ichthyologie generale. Tome II.
Ganoides, Dipnes, Lophobranches. Libraire Encyclopedique de
Roret, Paris, 623 pp. (Original not seen).
Eigenmann, C. H.

1898. A new blind fish. Proc. Indiana Acad. Sci., ( 1897) :231.

1899. A case of convergence. Proc. Indiana Acad. Sci., ( 1898) :247-251.
Ellis, M. M.

1916. Fohjodon in western Missouri. Copeia, 38:93-94. December 24.
Evermann, B. W.

1902. Description of a new species of .shad (Alosa ohicnsis) with notes on
other food fishes of the Ohio River. Rept. U.S. Fish Comm.,
(1901):273-288.
Evermann, B. W., and U. O. Cox

1896. A report upon the fishes of the Missouri River Basin. Rept. U.S.
Fish Comm., 20( 1894) :325-429. November 27.
Evermann, B. W., and W. C. Kendall

1895. A list of the species of fishes known from the vicinitv of Neosho,
Missouri. Bull. U.S. Fi.sh Comm., 14:469-472.
Fajen, O. F.

1962. The influence of stream stability on homing beha\ior of two small-
mouth bass populations. Trans. Amer. Fish. Soc, 91(4) :346-349.
October.

Fenneman, N. M.

1938. Physiography of Eastern United States. McGraw-Hill Book Co.,
New York, 714 pp.
Fidler, M. M.

1943. The preglacial Teays Valley in Indiana. Jour. Geol., 51:411-418.
Fingerman, S. W., and R. D. Suttkus

1961. Comparison of Ht/ho^uathus hai/i Jordan and Hybognathus nuchalis
Agassiz. Copeia, ' 1961(4) :462-"467. December 22.

Missouri Fishes 453

Fisher, H. J.

1962. Some fishes of the lower Missouri River. Amer. Midi. Nat., 68(2):
424-429. October.
FisK, H. M.

1944. Geological in\'estigation of the alluvial valley of the lower Missis-
sippi Ri\er. Mississippi River Comm. Report, 78 pp.

Flint, R. F.

1957. Glacial and Pleistocene geolog>'. John Wiley and Sons, New York,
553 pp.
Forbes, S. A., and R. E. Richardsox

1905. On a new sho\elnose sturgeon from the Mississippi River. Illinois

State Lab. Xat. Hist., Bull. 7:37-44.
1909 (and 1920). The fishes of Illinois. Illinois Nat. Hist. Surv., 3:1-357,
plus Atlas of 103 maps.
Fowler, H. W.

1910. Notes on the \ariation of some species of the genus Notropis. Proc.

Philadelphia Acad. Nat. Sci., 62:273-293.
1915. Notes on Nematognathous fishes. Proc. Philadelphia Acad. Nat.

Sci., 67:203-243.
1921. Description of a new cyprinoid fish (Notropis stonei) with notes on
other fishes obtained in the United States. Proc. Philadelphia Acad.
Nat. Sci., 72:385-402.
1924. Notes on North American cyprinoid fishes. Proc. Philadelphia Acad.
Nat. Sci., 76:389-416.
Fry, J. P.

1962. Harvest of fish from tailwaters of three large impoundments in Mis-
souri. Proc. 16th Conf. Southeastern Assoc. Game and Fish Comm.
: 405-411.
Frye, J. C., AND A. B. Leonard

1952. Pleistocene geology of Kansas. State Geol. Survey, Univ. Kansas,
Publ. 99:1-230. November.

Funk, J. L.

1957. Movements of stream fishes in Missouri. Trans. Amer. Fish. Soc,
85:39-57.

1958. Relati\e efficiency and selectivity of gear used in the study of
stream fish populations. Trans. 23rd North Amer. Wildl. Conf.
: 236-248.

Funk, J. L., and R. S. Campbell

1953. The population of larger fishes in Black River, Missouri (pp. 69-82
in The Black Ri\er Studies, bv J. L. Funk, et al.). LTniv. Missouri
Studies, 26(2): 1-1.36.

Funk. T. L., et al.

1953. The Black River Studies. Univ. Missouri Studies, 26(2): 1-136.
Garman, H.

1890. A preliminary report on the animals of the Mississippi bottoms near
Quincy, Illinois in August, 1888. Part I. Illinois State Lab. Nat.
Hist, Bull. 3:123-184.
Carman, S.

1889. Cave animals from southwestern Missouri. Harvard Mus. Compar.
Zool., Bull. 17(6):225-240.
Gerking, S. D.

1945. The distribution of the fishes of Indiana. Invest. Indiana Lakes
and Streams, 3(1):1-137.

1947. The use of minor postglacial drainage connections by fishes in
Indiana. Copeia, 1947(2) :89-91. June 30.
GiBBs, R. H., Jr.

1957a. Cyprinid fishes of the subgenus Cyprinella of Notropis. III. Varia-
tion and subspecies of Notropis vemistus (Girard). Tulane Stud,
in Zool, 5(8):175-203. August 7.

1957b. Cyprinid fishes of the subgenus Cyprinella of Notropis. II. Distri-
bution and \ariation of Notropis spilopterus, with the description of
a new subspecies. Lloydia, 20( 3) : 186-211. September.

1961. Cyprinid fishes of the subgenus Cyprinella of Notropis. IV. The

454

University of Kansas Publs., Mus. Nat. Hist,

Notropis galaciurus-camurus complex. Amer. Midi. Nat., 66(2):
337-354. October.

1963. Cyprinid fishes of the suligemis Cyphnella of Notropis. The Notropis
whipplei-analostanus-chloristius complex. Copeia, 1963(3) :511-
528. September 25.

Gilbert, C. R.

1961. Hyl^ridization \ersus intergradation: An inquiry into the relation-
ship of two cyprinid fishes. Copeia, 1961(2) : 181-192. June 19.

1964. The American cvprinid fishes of the subgenus Ltixilus (Genus
Notropis). Florida State Mus., Bull. 8(2):95-194. June 2.

1969. Systematics and distribution of the American cyprinid fishes No-
tropis ariommus and Notropis telescopiis. Copeia, 1969(3) :474-492.
August 29.
Gilbert, C. R., and R. M. Bailey

1962. Svnonvmv, characters, and distribution of the American cyprinid
fish Notropis shumardi. Copeia, 1962(4) :807-819. December 31.

Gilbert, C. H.

1886. Third series of notes on Kansas fishes. Bull. Washburn Lab. Nat.

Hi.st., 1:207-211. December 15.
1888. Descriptions of new and little known etheostomids. Proc. U.S. Natl.
Mus., 10:47-64. May 17.
Gill, T.

1860. Monograph of the genus Lahrax. Proc. Philadelphia Acad. Nat.

Sci., 12:108-119.
1865. Synopsis of the genus Pomxijs, Raf. Proc. Philadelphia Acad. Nat.
Sci., 17:64-66.

GiRARD, C.

1858. Fi.shes (in general report on the zoology). U.S. Pacific Railroad
Surveys, 10(4):l-400.

GOSLINE, W. A.

1948. Speciation in the fishes of the genus Mcnidia. Evol, 2(4) :.306-311.
December.

1949. The sensory canals of the head in some cyprinodont fishes, with
particular reference to the genus Fundtihis. Mus. Zool., Univ.
Michigan, Occ. Pap. 519:1-17. June.

1966. The limits of the fish family Serranidae, with notes on other lower
Percoids. Proc. California Acad. Sci., 33(6) :91-112. January 31.
Greene, C. W.

1935. The distribution of Wisconsin fishes. Wise. Cons. Comm., Madison,
235 pp.
Greene, F. C, and R. M. Trowbridge

1935. Pre-glacial drainage pattern of northwest Missouri. Missomi Geol.
Surv. and Water Res., 58th Bienn. Rept., Append. 7:3-7.
Greenwood, P. H., D. E. Rosen, S. H. Weitzman, and G. S. Myers

1966. Phyletic studies of teleostean fishes, with a proxisional classification
of living forms. Bull. Amer. Mus. Nat. Hist., 131(4) :339-456.
April 18.
Grohskopf, J. G.

1955. Subsmface geology of the Mississippi Embayment of southeast
Missouri. Missouri Geol. Surv. and Water Res., 2nd Ser., 37:1-133.

Hagmeier, E. M., and C. D. Stults

1964. A numerical analysis of the distribution patterns of North American
mammals. Syst. Zool., 13(3) :125-155. September 22.
Hall, G. E.

1956. Additions to the fish fauna of Oklahoma, with a summary of intro-
duced species. Southwest. Nat., l(l):16-26. lune 4.

Hanson, W. D.

1962. Dynamics of the largemouth bass population in Bull Shoals Reser-
voir, Missouri. Proc. 16th Conf. Southeastern .\ssoc. Game and
Fish Comm.: 398-404.

Hanson, W. D., and R. S. Campbell

1963. The effects of pool size and bea\er acti\ity on distribution and
abimdance of warm-water fishes in a north Missouri stream. Amer.
Midi. Nat., 69(1):136-149. January.

MissouHi Fishes 455

IIaHLAN, J. R., AM) E. li. Sl'EAKKH

1956. Iowa fish and fishing (3id ed. ). Iowa Cons. Conini., Des Moines,

377 pp.
Heim, G. E., and W. B. Howe

1963. Pleistocene drainage and depositional history in northwestern Mis-

sonri. Trans. Kansas Acad. Sei., 66(3) :378-392. December 17.

HORBERG, C. L.

1945. A major buried \alley in east-central Illinois and its regional rela-
tionships. Jour. Geol., 53:349-359.

HORBERG, C. L., AND R. C. ANDERSON

1956. Bedrock topography and Pleistocene glacial lobes in central United

States. lour. Geol., 64(2): 101-116. March.
Hov, P. R.

1872. Jom-nal of an exploration of western Missouri in 1854, under the

auspices of the Smithsonian Institution. Smithsonian Inst., Annual

Rept. for 1864:431-438.
HUBBS, C. L.

1930. Materials for a revision of the catostomid fishes of eastern North

America. Mus. Zool., Univ. Michigan, Misc. Publ. 20:1-47. April 30.
1938. Fishes from the caves of Yucatan. Carnegie Inst. Washington,

Publ. 491:261-295.
1942. An atherinid fish from the Pliocene of Oklahoma, jour. Paleont.,

16(3):399-400. May 6.
1945. Corrected distributional records for Minnesota fishes. Copeia, 1945

(l):13-22. March 31.
1951. Notio}))^ amnis, a ne\\- cyprinid fish of the Mississippi fauna, with

two subspecies. Mus. Zool., Univ. Michigan, Occ. Pap. 530:1-30.

March 14.
HuBBS, C. L., AND R. M. Bailey

1940. A revision of the black basses (Micropfeius and Hum) with de-
scriptions of four new forms. Mus. Zool., Univ. Michigan, Misc.

Publ. 48:1-51. July 27.
HuBBS, C. L., AND J. D. Black

1940. Percid fishes related to Poecilichthys vaiiatus, with descriptions of
three new forms. Mus. Zool., Univ. Michigan, Occ. Pap. 416:1-30.
June 26.

1941. The subspecies of the American percid fish, Poecilichtht/s whipplii.
Mus. Zool., Univ. Michigan, Occ. Pap. 429:1-27. May 26.

1947. Re\ision of Ceraiichthys, a genus of American cyprinid fishes. Mus.
Zool, Uni\-. Michigan, Misc. Publ. 66:1-56. January 27.

HuBBS, C. L., AND K. BONHAM

1951. New cvprinid fishes of the genus Notropis from Texas. Texas Jour.
Sci., 3(1):91-110. March 30.
HuBBS, C. L., AND W. R. Crowe

1956. Preliminary analysis of the American cyprinid fishes, seven new,
referred to the genus Hyhopsis. suligenus Eriun/stax. Mus. Zool.,
Univ. Michigan, Occ. Pap. 578:1-8. August 24. "
HuBBS, C. L., AND C. W. Greene

1935. Two new subspecies of fishes from Wisconsin. Trans. Wisconsin
Acad. Sci., Arts, and Letters, 29:89-101.

HuBBS, C. L., AND C. W. HiBBARD

1951. Ictahuus lambda, a new catfish, Ixised on a pectoral spine from the
Lower Pliocene of Kansas. Copeia, 1951( 1 ) :8-14. March 21.
HxiBBS, C. L., AND K. F. Lagler

1947. Fishes of the Great Lakes Region. Cranbrook Inst. Sci., 26:1-186.
HuBBS, C. L., AND G. A. Moore

1940. The subspecies of Notropis zonatus, a cyprinid fish of the Ozark
Upland. Copeia, 1940(2) :91-99. ]\\\y 28.
HuBBS, C. L., AND A. I. Ortenbxjbger

1929. Fishes collected in Oklahoma and Arkansas in 1927. Biol. Surv.,
Univ. Oklahoma, Publ. 1(3):45-112. April.
HxJBBS, C. L., AND E. C. Raney

1944. Systematic notes on North American siluroid fishes of the genus

456 University of Kansas Publs., Mus. Nat. Hist.

SchiJbeodes. Mus. Zool., Univ. Michigan, Occ. Pap. 487:1-36.
August.
Hltbbs, C. L., and M. B. Trautmax

1932. PoeciliclitJiys oshurni, a new darter from the upper Kanawha River
system in Virginia and West Virginia. Ohio Jour. Sci., 32( 1 ) : 31-38.
January.
1937. A revision of the lamprev genus Ichihifomyzon. Mus. Zool., Univ.
Michigan, Misc. Publ. 3.5:1-109. June 14.
HuBBS, Clark

1954. A new Texas subspecies, aprisies, of the darter Hadroptenis scienis,
with a discussion of variation within the species. Amer. Midi. Nat.,
52(1):21 1-220. June.
HuTCHEXs, L., AND G. E. Hall

1951. Occurrence of the Ohio shad, Alosa ohiensis Evermann, in eastern
Oklahoma. Copeia, 1951( 1 ):83-84. March 21.
Jordan, D. S.

1885. Descriptions of four new species of PoecilichtJiys in the United
States National Museum. Proc. U.S. Natl. Mus., 7:477-480. Sep-
tember 27.
Jordan, D. S., and B. W. Evermann

1896- The fishes of North and Middle America. Bull. U.S. Natl. Mus.,
1900. 47(1-4): 1-3313.
Jordan, D. S., and C. H. Gilbert

1882. Synopsis of the fishes of North America. Bull. U.S. Natl. Mus.,
16:1-1018. May 20.
Jordan, D. S., and S. E. NIeek

1885. List of fishes collected in Iowa and Missouri in August, 1884, with
descriptions of three new species. Proc. U.S. Natl. Mus., 8:1-17.
April 20.
Kathrein, J. W.

1953. An intensive creel census of Clearwater Lake, Missouri, during its
first four years of impoundment, 1949-1952. Trans. 18th North
Amer. Wildl. Conf.: 282-295.

Keep, J.

19.55. Lowdown on the Delta. Missouri Conservationist, 16(7):4-7; 16
(8): 1-4. July and August.
Knapp, L. W.

1964. Systematic studies of the rainbow darter, Eihcosioma cacnileum
(Storer), and the subgenus Hadroptenis (Pisces, Percidae). Diss.
Abstr., 25(3):2112-2113.

Koster, W. J.

1957. Guide to the fishes of New Mexico. Univ. New Mexico Press,
Albuquerque, 116 pp.
KuEHNE, R. A., and R. M. Bailey

1961. Stream capture and the distribution of the percid fish EtJieostoma
sagitta, with geologic and taxonomic considerations. Copeia, 1961
(1):1-18. March 17.
Lachner, E. a., and R. E. Jentcins

1967. Systematics, distribution, and e\'olution of the chub genus Nocomis
(Cyprinidae) in the southwestern Ohio Ri\er basin, with the de-
scription of a new species. Copeia, 1967(3) :557-580. September 1.
Lane, C. E., Jr.

1954. Age and growth of bluegills in a new impoundment. Jour. Wildl.
Mgt., 18(3):3.58-365. July.

Lemke, R. W., et al.

1965. Quaternary geology of the northern Great Plains (pp. 15-27 in The
Quaternary of the United States, By H. E. Wright, and J. C.
Frye). Princeton Univ. Press.

Lesueur, C. a.

1827. American ichthyology, or, natural history of the fishes of North
America: with coloured figures from drawings executed from nature.
New-Harmony, Indiana, 44 pp. (unnumbered). (Original not
seen. )

Missouri Fishes 457

LiNDER, A. D.

1959. The American percid fislies AniDiocrtipta clara Jordan and Meek
and Ammocrtji)ta pclhicida (Baird). Southwestern Nat., 4(4): 176-
184. Decenilier.

Luce, W. M.

1933. A survey of the fisheiy of the Kaskaskia River. Illinois Nat. Hist.
Surv., Bull. 20:71-123.

LUEXINGHOENER, G. C.

1947. The post-Kansan geologic history of the lower Platte Valley area.
Univ. Nebraska Studies, New Ser., 2:1-82. September.
LuGx, A. L.

1935. The Pleistocene geology of Nebraska. Nebraska Geol. Surv. Bull.,
2nd Ser., 10:1-223.

M.ARBUT, C. F.

1902. The e\olution of the northern part of the lowlands of southeastern
Missouri. Uni\ . Missouri Stud., l(3):l-63.
Martin, R. G., axd R. S. Campbell

1953. The small fishes of Black River and Clearwater Lake, Missouri (pp.
45-66 in The Black River studies, by John L. Funk, et al.). Univ.
Missouri Stud., 26(2): 1-136.
Marzolf, R. C.

1955. Use of pectoral spines and \ertebrae for determining age and rate
of growth of the channel catfish. Jour. Wildl. Mgt., 19(2):243-
249. April.
1957. The reproduction of channel catfish in Missouri ponds. Tour. Wildl.
Mgt., 21(l):22-28. January.
Matthes, F. E.

1933. The Pleistocene diversion of the Mississippi Ri\er across Crowley's
Ridge, southeast Missouri. Science (New Ser.), 77:459-460.
Mayxard, H. J.

1889. Rainbow trout in southwestern Missouri. Bull. U.S. Fish Comm.,
7:55-56.
Mayr, E.

1942. Svstematics and the origin of species. Columbia Univ. Press, N.Y.,
334 pp.
Meek, S. E.

1891. Report of explorations made in Missouri and Arkansas during 1889,
with an account of fishes observed in each of the river basins ex-
amined. Bull. U.S. Fish Comm., 9:113-141.
1894. A new Etheostoma from Arkansas. Amer. Nat., 28:957.
Metcalf, a. L.

1966. Fishes of the Kansas River system in relation to zoogeography of
the Great Plains. Mus. Nat. Hist., Univ. Kansas, Publ. 17(3):23-
189. March 24.
Miller, M. F., axd H. H. Krusekopf

1929. The soils of Missouri. Agric. E.xpt. Sta., Univ. Missouri, Bull.
264:1-120.
Miller, R. R.

1955. An annotated list of the American cyprinodont fishes of the genus
Fundulus, with the description of Fundttlus persimilis from Yucatan.
Mus. Zool., Univ. Michigan, Occ. Pap. 568:1-25. August 1.
1965. Quaternary freshwater fishes of North America (pp. 569-.581 //! The
Quaternary of the United States, by H. E. Wright, Jr., and D. G.
Frey). Princeton Univ. Press, Princeton.
Miller, R. V.

1968. A systematic study of the greenside darter, Etheostoma bleninoides
Rafinesque (Pisces :Percidae). Copeia, 1968( 1 ) :l-40. March 15.
Mixckley, W. L., and L. A. Kru.mholz

1960. Natural hybridization between the clupeid genera Dorosoma and
Si^nalosa, with a report on the distribution of S. petenen.se. Zoolog-
ica, 44(4): 171-180. December 31.

Moore, G. A.

1957. Fishes (pp. 31-210 in Vertebrates of the United States, by W. F.
Blair, et al. ) . McGraw-Hill Book Co.

458 University of Kansas Publs., Mus. Nat. Hist.

Moore, G. A., and C. C. Rigxey

1952. Taxonomic stahis of the percid fish, Poecilichthys radiosus in Okla-
homa and Arkansas, with the description of two new subspecies.
Copeia, 1952( 1 ) :7-15. June 2.

Morris, G. A.

1967. Production of channel catfish to creel size. Progr. Fish-Cult., 29(2):
84-86. April.
Needham, R. G.

1965. Spawning of paddlefish induced b\' means of pituitary material.
Progr. Fish-Cult., 27(1):13-19. January.
Neel, J. K., H. P. Nicholson, and A. Hirsch

1963. Main stem reservoir effects on water quality in the central Missouri
River. U.S. Dept. Health, Education, and Welfare Publ.: 1-112.
Netsch, N. F., ant) A. Witt, Jr.

1962. Contributions to the life history of the longnose gar, Lcpisostctis
osseus, in Missouri. Trans. Amer. Fish. Soc, 91( 3) :251-262. July.
Niazi, a. D., and G. a. Moore

1962. The weberian apparatus of Hybo^nathus placitus and H. uuchalis
( Cyprinidae ) . Southwestern Nat., 7(l):41-50. June 1.

Olund, L. J., AND F. B. Cross

1961. Geographic \ariation in the North American cyprinid fish, Hybopsis
gracilis. Mus. Nat. Hist., Univ. Kansas, Publ. 13(7) :323-348.
Febniary 10.
Patriarche, M. H.

1953. The fishery in Lake Wappapello, a flood-control reserxoir on the
St. Francis River, Missouri. Trans. Amer. Fish. Soc, 82:242-254.

Patriarche, M. H., and R. S. Camphell

1958. The development of the fish population in a new flood-control reser-
voir in Missouri, 1948-1954. Trans. Amer. Fish Soc, 87:240-258.

Pflieger, W. L.

1966a. Young of the orangethroat darter (Etheostoma spectabile) in nests

of the smallmouth bass (Micioi)tcnis doloiiucui). Copeia, 1966(1):

139-140. March 22.
1966b. Reproduction of the smallmouth bass (Micropteius dolomieiii) in a

small Ozark stream. Amer. Midi. Nat., 76(2):410-418. October.
Platner, W. S.

1946. Water quality studies of the Mississippi River. U.S. Fish and Wildl.

Serv., Spec. Sci. Rept. 30:1-77.
Purkett, C. a., Jr.

1958a. Growth of the fishes in the Salt Ri\er, Missouri. Trans. Amer. Fish.

Soc, 87:116-131.
1958b. Growth rates of Missouri stream fishes. Missouri Cons. Comm., Fish

and Game Div., D-J Ser. 1:1-46.
1961. Reproduction and earlv de\elopment of the paddlefish. Trans. Amer.

Fish. Soc, 90(2):125-129. April.

1963. Artificial propagation of paddlefish. Progr. Fish-Cult., 25(l):31-33.
January.

Putnam, F. W.

1863. List of the fishes sent by the Museimi to diff^erent institutions in
exchange for other specimens, with annotations. Bull. Mus.
Compar. Zool., Harvard, 1(1):2-16.
QUINN, J. H.

1958. Plateau surfaces of the Ozarks. Proc. Arkansas Acad. Sci., 11:36-43.
Rafinesque, C. S.

1820. Ichthyologia ohiensis, or natural history of the fishes inhabiting the
river Ohio and its tributary streams, preceded by a ph>'sical de-
scription of the Ohio and its branches. W. G. Hunt, Lexington,
90 pp.

Robins, C. R.

1954. A taxonomic revision of the Cottiis bairdi and Coitus caiolinae
species groups in eastern North America (Pisces, Cottidae). Diss.
Abstr., 15(2):302-303.

Missouri Fishes 459

Ross, H. H.

1965. Pleistocene events and insects (pp. 583-596 in The Quaternary of
the United States). Princeton Univ. Press, Princeton.

Ross, R. D.

1958. Races of the cyprinid fi.sh Canqyostonui anovialum jxiUuiii (Agassi/,)
in eastern United States. N'irginia Agric. Exper. Sta., Tech. Bull.
136:1-20.
Sauer, C. O.

1920. The geography of the Ozark Highland of Missouri. Univ. Chicago
Press. 245 pp.
Semkin, H. a. Jr.

1966. Stratigraphy and paleontology of McPherson Equs beds (Sandahl
local fauna, McPherson Countv, Kansas). Mus. Paleont., Univ.
Michigan, Contr. 20(6): 137. January 5.

Setzer, p. Y.

1970. An analysis of a natural hybrid swarm l)y means of chromosome
morphology. Trans. Amer. Fish. Soc, 99( 1 ) :139-146.
Shelford, V. E.

1911. Ecological succession: stream fishes and the method of physiographic
analysis. Biol. Bull., 21:9-34. June.
Shoemaker, H. H., Q. H. Pickering, and L. Durham

1951. The occurrence of the flathead chub, Platygohio gracilis, in Ten-
nessee. Jour. Tennessee Acad. Sci., 26(1 ):84.
Skeltox, J.

1966. Low-flow characteristics of Missouri streams. Missouri Geol. Surv.
and Water Res., Water Res. Rept. 20:1-95.

Smiley, C. W.

1885. Other fish mistaken for carp. Bull. U.S. Fish Comm., 5:347.
Smith, C. L.

1954. Pleistocene fishes of the Berends faiuia of Beaver Countv, Oklahoma.

Copeia, 1954(4) :282-289. October 29.
1958. Additional Pleistocene fishes from Kansas and Oklahoma. Copeia,
1958(3) :176-180. August 28.

1962. Some Pliocene fishes from Kansas, Oklahoma and Nebraska. Copeia,
1962(3):505-520. September 28.

Smith, G. R.

1963. A late Illinoian fish fauna from southwestern Kansas and its climatic
significance. Copeia, 1963(2 ) :278-285. June 14.

Smith, P. W.

1957. An analysis of post- Wisconsin biogeography of the Prairie Peninsula
region based on distributional phenomena among terrestrial \'erte-
brate populations. Ecol., 38( 2) :205-218.
1965a. Recent adjustments in animal ranges (pp. 633-642 in The Quater-
nary of the United States, by H. E. Wright, Jr., and D. G. Frey).
Princeton Univ. Press, Princeton.
1965b. A preliminary annotated list of the lampreys and fishes of Illinois.
Illinois Nat. Hist. Survey, Biol. Notes 54:1-12. June.
Smith-\'antz, W. F.

1968. Freshwater fishes of Alabama. Auburn University, Agric. E.xper.
Sta., 211 pp.
Starrett, W. C, W. J. Harth, and P. W. Smith

1960. Parasitic lampreys of the genus Ichthi/omyzon in the rivers of
Illinois. Copeia, 1960(4) :3.37-346. December 30.
Steedman, J. G. W.

1884. Carp and carp culture in Missouri, \\ith appendix on natixe fish.
Missouri State Fish Comm., Bienn. Rept. for 1883-84:59-116.

Steyermark, J. A.

1963. Flora of Missouri. Iowa Univ. Press, 1725 pp.
SuTTKUs, R. D., and J. S. Ramsey

1967. Peicina aiiwlineata, a new percid fish from the Alabama River sys-
tem and a discussion of ecology, distribution, and h\bridizati()n of
darters of the subgenus Hadioptenis. Tulane Stud, in Zool., 13(4):
129-145. February 24.

460 University of Kansas Publs., Mus. Nat. Hist.

SuTTKUS, R. D., AND E. C. Raney

1955. Notropis asperifrons, a new cyprinid fish from the Mobile Bay
drainage of Alal^ama and Georgia with studies of related species.
Tulane Stud, in Zool, 3(l):3-33. July 8.

Taylor, D. W.

1965. The study of Pleistocene nonmarine moUusks in North America (pp.
597-611 in The Quaternary of the United States, by H. E. Wright,
Jr., and D. G. Frey). Princeton Uniw Press, Princeton.

Taylor, W. R.

1969. Revision of the catfish genus Nottinis Rafinesque, with an analysis
of higher groups in the Ictaluridae. U.S. National Museum, Bull.
282:1-315.
Thomas, E. S.

1951. Distribution of Ohio animals. Ohio Jour. Sci., 51(4): 153-167. July.
Thomersox, J. E.

1966. A comparative biosystematic study of Fitndiihis noiatus and Fun-
dtihis olivaceiis ( Pisces :Cyprinodontidae). Tulane Stud, in Zool.,
13(l):29-47. March 17.

Thompson-, D. H., axd F. D. Hunt

1930. The fishes of Champaign County: a study of the distribution and
abundance of fishes in small streams. Illinois Nat. Hist. Survey,
Bull. 19:1-101.

Thornbury, W. D.

1965. Regional geomorphology of the United States. John Wiley and
Sons, N.Y., 609 pp.
Todd, J. E.

1914. Pleistocene history of the Missouri River. Science (New Ser. ),
39:263-274.
Traxttman, M. B.

1931. Notropis vohicelhis tcickliffi, a new subspecies of cyprinid fish from
the Ohio and upper Mississippi rivers. Ohio Jour. Sci., 31(6):468-
474. Novemlier.

1956. Carpiodcs cijprinus hinci, a new subspecies of carpsucker from the
Ohio and upper Mississippi river systems. Ohio Jour. Sci., 56(1):
33-40. Januarv.

1957. The fishes of Ohio. The Ohio State Univ. Press, Columbus, 638 pp.
Trautman, M. B., and R. G. Martin

1951. Moxostoma aurcohim pisolahrum, a new subspecies of sucker from
the Ozarkian streams of the Mississippi River system. Mus. Zool.,
Univ. Michigan, Occ. Pap. 534:1-10. November 12.

TsAi, C.

1968. Distribution of the harlequin darter, Etheostoma histrio. Copeia,
1968(1 ):178-181. March 15.
Underhill, J. C.

1957. The distribution of Minnesota minnows and darters in relation to
Pleistocene glaciation. Minnesota Mus. Nat. Hist., Occ. Pap. 7:1-45.
U. S. Corps of Engineers

1957. Navigation charts, Missouri River, Omaha, Nebraska to mouth.
U.S. Corps of Engineers, Missouri River Di\'., Omaha, Nebraska.
U. S. Department of Agriculture

1941. CHmate and man. Yearbook Agric, U.S. Dept. Agric, Washington,
D.C., 1248 pp.
U. S. Geological Survey

1962. Surface waters of Missouri. Water Res. Div., U.S. Geol. Survey,
161 pp.
Wayne, W. J.

1952. Pleistocene evolution of the Ohio and Wabash valleys. Jour. Geol.,
60:575-585.

Weyer, a. E.

1940. The Lake of the Ozarks. A problem in fishery management. Progr.
Fish-Cult., 51:1-10.
Wilson, W. D.

1956. Notes on cestodes in paddlefish, Pohjodon spathula (Walbaum),

Missouri Fishes 461

irom the Missouri Ri\er. Trans. Kansas Acad. Sci., 59(4) :459-
460. February 6.
Witt, A., Jr.

1960. Length and weight of ancient freshwater drum, A))lodinotus gruii-
niciis, calculated from otoliths foimd in Indian middens. Copeia,
1960(3) :181-185. September 26.
Witt, A., Jr., and R. C. Marzolf

1954. Spawning behaxior of the longear simfi.sh, Lepomis tnegalotis
mcgahiis. Copeia, 1954( 3) :188-190. July 29.
Woods, L. P., and R. F. Inger

1957. The ca\e, spring, and swamp fishes of the family Amblyopsidae of
central and eastern United States. Amer. Midi. Nat., 58(1):232-
256. July.

462

University of Kansas Publs., Mus. Nat. Hist.

Explanation of Maps

The maps that follow indicate, for each species, all locality rec-
ords that I consider to be valid. Not all museum holdings have
been canvassed, but I have examined virtually all catalogued ma-
terial at the University of Kansas Museum of Natural History and
the University of Missouri, and much of that at the University of
Michigan Museum of Zoology. The extensive collections made by
George V. Harry were reexamined insofar as questionable records
seemed to be involved; otherwise his identifications and localities
were accepted.

The following numbered list of the counties of Missouri corre-
sponds to the numbers on the maps.

1. Clark

2. Scotland

3. Schuyler

4. Putnam

5. Mercer

6. Harrison

7. Worth

8. Nodaway

9. Atchison

10. Holt

11. Andrew

12. Gentry
1.3. Dekalb
14. Daviess
1.5. Grundy

16. Sullivan

17. Adair

18. Knox

19. Lewis

20. Marion

21. Shelby

22. Macon

23. Linn

24. Livingston
2.5. Caldwell

26. Clinton

27. Buchanan

28. Platte

29. Clay

30. Ray

31. Carroll

32. Chariton

33. Randolph

34. Monroe
.35. Ralls

36. Pike

37. Audrain

38. Howard

39. Boone

40. Calloway

41. Montgomery

42. Lincoln

43. St. Charles

44. Warren

45. Saline

46. Lafayette

47. Jackson

48. Cass

49. Johnson
.50. Pettis

51. Cooper

52. Moniteau

53. Cole

54. Osage

.55. Gasconade

56. Franklin

57. St. Louis

58. Jefferson

.59.

Maries

88.

Texas

60.

Miller

89.

Shannon

61.

Morgan

90.

Reynolds

62.

Benton

91.

Wayne

63.

Henry

92.

Bollinger

64.

Bates'

93.

Cape Girardeau

65.

\^ernon

94.

Scott

66.

St. Clair

95.

Stoddard

67.

Hickory

96.

Butler

68.

Camden

97.

Carter

69.

Pulaski

98.

Douglas

70.

Phelps

99.

Christian

71.

Crawford

100.

Lawrence

72.

Washington

101.

Jasper

73.

St. Francois

102.

Newton

74.

Ste. Genevieve

103.

McDonald

75.

Perry

104.

Barry

76.

Madison

105.

Stone

77.

Iron

106.

Taney

78.

Dent

107.

Ozark

79.

Laclede

108.

Houell

80.

Dallas

109.

Oregon

81.

Polk

110.

Ripley

82.

Cedar

111.

Mississippi

83.

Barton

112.

New Madrid

84.

Dade

113.

Dunklin

85.

Greene

114.

Pemiscot

86.

Webster

87.

Wright

In order to depict recent changes in fish distribution, three
symbols are used. Large open circles represent records prior to
1905, small open circles correspond to collections made between
1905 and 1945, and solid black circles correspond to collections
made subsequent to 1945. The distribution of collection localities
is shown on maps 1 and 2.

On the small range-maps, only natural distribution has been
indicated except for the introduced carp and goldfish. The range-
maps were compiled almost entirely from the literature, and their
accuracy is proportional to the amount of published information
available on a given species.

Missouri Fishes

463

464 University of Kansas Publs., Mus. Nat. Hist.

Missouri Fishes

465

MAPS

CHESTNUT LAMPREY

ICHTHYOMYZON CASTANEUS

MAP 6
SOUTHERN BROOK LAMPREY
ICHTHYOMYZON GAGEI

466

University of Kansas Publs., Mus. Nat. Hist.

Missouri Fishes

467

468 University of Kansas Publs., Mus. Nat. Hist.

MAP 12

PADDLEFISH

POLYODON SPATHULA

Missouri Fishes

469

MAP 14

SHORTNOSE GAR

LEPISOSTEUS PLATOSTOMUS

470 University of Kansas Publs., Mus. Nat. Hist.

MAP 16

LONGNOSE GAR

LEPISOSTEUS OSSEUS

Missouri Fishes

471

MAP 18

AMERICAN EEL

ANGUILLA ROSTRATA

472 University of Kansas Publs., Mus. Nat. Hist.

MAP 19

SKIPJACK HERRING

ALOSA CHRYSOCHLORIS

Missouri Fishes

473

MAP 22
THREADFIN SHAD
DOROSOMA PETENENSE

474 University of Kansas Publs., Mus, Nat. Hist.

MOONEYE
HIODON TERGISUS

Missouri Fishes

475

476 University of Kansas Publs., Mus. Nat. Hist.

MAP 28

NORTHERN PIKE

ESOX LUCIUS

Missouri Fishes

477

478

University of Kansas Publs., Mus. Nat. Hist.

Missouri Fishes

479

480

Unr'ersity of Kansas Fuels., Mus. Nat. Hist.

MAP 36

SILVER CHUB

HYBOPSIS STORERIANA

Missouri Fishes

481

MAP 38

STREAMLINE CHUB
HYBOPSIS DISSIMILIS

482 University of Kansas Publs., Mus. Nat. Hist.

MAP 40

SPECKLED CHUB

HYBOPSIS AESTIVALIS

Missouri Fishes

483

MAP 42
STURGEON CHUB
HYBOPSIS GELIDA

484 University of Kansas Publs., Mus. Nat. Hist.

MAP 44
SUCKERMOUTH MINNOW
PHENACOBIUS MIRABILIS

Missouri Fishes

485

486 University of Kansas Publs., Mus. Nat. Hist.

MAP 48

REDFIN SHINER

NOTROPIS UMBRATILIS

Missouri Fishes

487

MAP 50
SILVERBAND SHINER
NOTROPIS SHUMARDI

488 University of Kansas Publs., Mus. Nat. Hist.

MAP 52

DUSKYSTRIPE SHINER

NOTROPIS PILSBRYl

Missouri Fishes

489

490 University of Kansas Publs., Mus. Nat. Hist.

Missouri Fishes

491

492

University of Kansas Publs., Mus. Nat. Hist.

MAP 60

BIGEYE SHINER

NOTROPIS BOOPS

Missouri Fishes

493

MAP 62

PALLID SHINER

NOTROPIS AMNIS

494 University of Kansas Publs., Mus. Nat. Hist.

SPOTFIN Sf
NOTROPIS SPILOPTERUS

Missouri Fishes

495

MAP 66

WHITETAIL SHINER

NOTROPIS GALACTURUS

496

University of Kansas Publs., Mus. Nat. Hist.

7/?

"7^

^^\3

■ k

' '' '®

■^ -"\|

jT; '^

4'f/

^^V\ vi

\5^

(,' 23

V v^oti

k-n,

K^-::^^

~'^^

^^v^^

' 25

' '■' 1

•^r' "^

- ^'- 1

32,

-1 -,

\ 34

2ej

v\-

r 1

' '\^'

ryy-^

\^ci '

..>fL^>^

'^ ^wp.

S?)

371 -^

<i -

' 70

o 10-

C I

;^°^

"ieTT

c r

tdrj:.

MAP 67
BLUNTFACE SHINER
NOTROPIS CAMURUS

Missouri Fishes

497

MAP 70

SAND SHINER

NOTROPIS STRAMINEUS

498

Unr'ersity of Kansas Publs., Mus. Nat. Hist.

Missouri Fishes

499

MAP 73

BLACKNOSE SHINER

NOTROPIS HETEROLEPIS

MAP 74

OZARK SHINER

NOTROPIS OZARCANUS

500

University of Kansas Publs., Mus. Nat. Hist.

MAP 75

MIMIC SHINER

NOTROPIS VOLUCELLUS

Missouri Fishes

501

502 University of Kansas Publs., Mus. Nat. Hist.

MAP 80

CENTRAL SILVERY MINNOW

HYBOGNATHUS NUCHALIS

Missouri Fishes

503

MAP 82
PLAINS MINNOW
HYBOGNATHUS PLACITUS

504 University of Kansas Publs., Mus. Nat. Hist.

Missouri Fishes

505

MAP 85

SLIM MINNOW

PIMEPHALES TENELLUS

506

University of Kansas Publs., Mus. Nat. Hist.

Missouri Fishes

507

MAP 90

BLUE SUCKER

CYCLEPTUS ELONGATUS

508

University of Kansas Publs., Mus, Nat. Hist.

MAP 92
BLACK BUFFALO
ICTIOBUS NIGER

Missouri Fishes

509

MAP 94

QUILLBACK

CARPIODES CYPRINUS

510 University of Kansas Publs., Mus. Nat. Hist.

MAP 96
HIGHFINCARPSUCKER
CARPIODES VELIFER

Missouri Fishes

511

MAP 98
HOG SUCKER
HYPENTELIUM NIGRICANS

512 Unr'ersity of Kansas Publs., Mus. Nat. Hist.

GOLDEN REDHORSE
MOXOSTOMA ERYTHRURUM

Missouri Fishes

513

MAP 102
NORTHERN REDHORSE
MOXOSTOMA MACROLEPIDOTUM

514 University of Kansas Publs., Mus. Nat. Hist.

MAP 104

SPOTTED SUCKER

MINYTREMA MELANOPS

Missouri Fishes

515

MAP 106
CREEK CHUBSUCKER
ERIMYZONOBLONGUS

516 University of Kansas Publs., Mus. Nat. Hist.

MAP 108
BROWN BULLHEAD
ICTALURUS NEBULOSUS

Missouri Fishes

517

MAP no

CHANNEL CATFISH
ICTALURUS PUNCTATUS

518 University of Kansas Publs., Mus. Nat. Hist.

MAP 1 12
TADPOLE MADTOM
NOTURUS GYRINUS

Missouri Fishes

519

520 University of Kansas Publs., Mus. Nat. Hist.

MAP 116

OZARK MADTOM

NOTURUS ALBATER

Missouri Fishes

521

MAP 118

BRINDLED MADTOM

NOTURUS MIURUS

522 Unwersity of Kansas Fuels., Mus. Nat. Hist.

MAP 120
FLATHEAD CATFISH
PYLODICTIS OLIVARIS

Missouri Fishes

523

MAP 121
TROUT PERCH
PERCOPSIS OMISCOMAYCUS

524 University of Kansas Publs., Mus. Nat. Hist.

MAP 124
SOUTHERN CAVEFISH
TYPHLICHTHYS SUBTERRANEUS

Missouri Fishes

525

University of Kansas Publs., Mus. Nat. Hist

Missouri Fishes

527

MAP 130
STARHEAD TOPMINNOW
FUNDULUS NOTTI

528 University of Kansas Publs., Mus. Nat. Hist.

MAP 131
BLACKSPOTTED TOPMINNOW
FUNDULUS OLIVACEUS

MAP 132
BLACKSTRIPE TOPMINNOW
FUNDULUS NOTATUS

Missouri Fishes

529

MAP 134

BROOK SILVERSIDE

LABIDESTHES SICCULUS

530 University of Kansas Publs., Mus. Nat. Hist,

MAP 136

WHITE BASS

MORONE CHRYSOPS

Missouri Fishes

531

( n

771

"^"?

"a"

[>—p

\ ,,

v«ll

\ \^"

•X' '*

~'f/¥

n'SI

\ f

,' '','

- 2S

1,' 23
' 1' '»

•5(^^ — r-f

Si"- V

""^Xr

si!

3r^

/^' i

''r A

T i'^/

•^'^

' '\^'

\ c^f^

•-,1^ ^^

^A\

.< -

r-s

A-c

0 10 30 50 MILES

,V'

. \

^^^

MAP 137
YELLOW BASS
MORONE MISSISSIPPIENSIS

:3:

MAP 138

SPOTTED BASS
MICROPTERUS PUNCTULATUS

532 University of Kansas Publs., Mus. Nat. Hist.

MAP 140

LARGEMOUTH BASS

MICROPTERUS SALMOIDES

Missouri Fishes

533

MAP 142

GREEN SUNFISH

LEPOMISCYANELLUS

534 Unwersity of Kansas Publs., Mus. Nat. Hist.

MAP 143

BANTAM SUNFISH

LEPOMIS SYMMETRICUS

Missouri Fishes

535

MAP 146

REDEAR SUNFISH

LEPOMIS MICROLOPHUS

536

University of Kansas Publs., Mus. Nat. Hist.

Missouri Fishes

537

538 University of Kansas Publs., Mus. Nat. Hist.

MAP 152
WHITE CRAPPIE
POMOXIS ANNULARIS

Missouri Fishes

539

540

University of Kansas Publs., Mus. Nat. Hist.

Missouri Fishes

541

542 University of Kansas Publs., Mus. Nat. Hist.

MAP 160

DUSKY DARTER

PERCINA SCIERA

Missouri Fishes

543

MAP 162

LONGNOSE DARTER

PERCINA NASUTA

544 University of Kansas Publs., Mus. Nat. Hist.

MAP 164

GILT DARTER

PERCINA EVIDES

Missouri Fishes

545

546 University of Kansas Publs., Mus. Nat. Hist.

MAP 168
CRYSTAL DARTER
AMMOCRYPTA ASPRELLA

Missouri Fishes

547

MAP 170
SCALY SAND DARTER
AMMOCRYPTA VIVAX

548 University of Kansas Publs., Mus. Nat. Hist.

Missouri Fishes

549

550 Unwersity of Kansas Publs., Mus. Nat. Hist.

MAP 176
HARLEQUIN DARTER
ETHEOSTOMA HISTRIO

Missouri Fishes

551

552 University of Kansas Publs., Mus. Nat, Hist.

MAP 180

REDFIN DARTER

ETHEOSTOMA WHIPPLEl

Missouri Fishes

553

MAP 181

MUD DARTER

ETHEOSTOMA ASPRIGENE

MAP 182
YOKE DARTER
ETHEOSTOMA JULIAE

554 University of Kansas Publs., Mus. Nat. Hist.

MAP 184
ARKANSAS DARTER
ETHEOSTOMA CRAGINI

Missouri Fishes

555

556 University of Kansas Publs., Mus. Nat. Hist.

MAP 188
SLOUGH DARTER
ETHEOSTOMA GRACILE

Missouri Fishes

557

'f^~''J^l

MAP 190
LEAST DARTER
ETHEOSTOMA MICROPERCA

558 University of Kansas Publs., Mus. Nat. Hist.

Missouri Fishes

559

1
\

> i

MAP 193

BANDED SCULPIN

COTTUS CAROLINAE

Missouri Fishes

561

Index to Common and Scientific Names
( Page numbers of prineipal aeeounts in boldface)

Acipenser

anasimos, 236, 314

fulvescens, 236, 314, 467

rauchi, 236, 314

nibicundus, 314
Alabama shad, 262, 308, 321, 472
Alburnellus

percobronius, 236, 339, 340
Alburnus

zonatiis, 232, 236, 344

lineolatiis, 236
Alligator gar, 263, 282, 294, 317, 469
Alosa

alabamae, 321, 472

chrysochloris, 320, 472

ohiensis, 321

sapidissima, 307, 322
Aniblodon

grunniens, 442

lineahis, 237
Ambloplites

rupestris, 417, 537

rupestris ariommus, 418

rupestris rupestris, 418
Amblyopsis

rosae, 233, 237, 399, 524

spelaea, 302, 400
Ameiurus

melas, 390

melas nielas, 390

natalis, 392

natalis natalis, 392

nebulosus, 391

ponderosus, 237, 393
American brook lamprey, 262, 280,

296, 313, 466
American eel, 263, 280, 308, 320, 471
American shad, 307, 322
Amia

calva, 319, 471
Amiatus

calvus, 319
Ammocoetes

branchialis, 311
Ammocrypta

asprella, 428, 546

beani, 429

clara, 428, 547

pellucida, 428

vivax, 429, 547
Amphiodon

alosoides, 323
Anguilla

bostoniensis, 320

rostrata, 320, 471
Aphredoderus

sayaniis, 399, 523
Aplodinotus

grunniens, 237, 442, 558
Arkansas darter, 262, 281, 437, 447,

554
Arkansas River scaly orangethroat, 303
Arkansas saddled darter, 262, 273,

274, 295, 432, 550
Arrow darter, 301, 434
Atlantic salmon, 307

Banded darter, 262, 271, 273, 281,

298, 301, 433, 446, 551
Banded pygmy sunfish, 262, 275, 420,

539
Banded sculpin, 262, 271, 443, 559
Bantam sunfish, 262, 305, 413, 447,

534
Barred fantail darter, 281, 301
Bigeye chub, 262, 271, 308, 333, 481
Bigeye shiner, 262, 273, 351, 492
Bigmouth buffalo, 263, 378, 379, 380,

508
Bigmouth shiner, 262, 275, 281, 306,

308, 352, 447, 493
Black buffalo, 263, 279, 379, 380, 508
Black bullhead, 263, 282, 373, 390,

413, 516
Black crappie, 263, 308, 418, 420, 538
Black redhorse, 262, 271, 297, 384,

385, 512
Blacknose shiner, 262, 296, 304, 308,

361, 447, 499
Blackside darter, 262, 308, 423, 542
Blackspotted topminnow, 263, 273,

278, 280, 404, 406, 528
Blackstripe topminnow, 263, 273, 280,

307, 406, 528

562

Unwersity of Kansas Publs., Mus. Nat. Hist.

Blacktail shiner, 262, 274, 275, 355,

357, 495
Bleeding shiner, 262, 272, 281, 295,

303, 344, 363, 488
Blue catfish, 263, 308, 393, 518
Blue sucker, 263, 276, 308, 378, 507
Bluegill, 263, 280, 308, 417, 537
Bluestripe darter, 262, 273, 301, 308,

422, 446, 541
Bluntface shiner, 262, 273, 301, 303,

356, 357, 446, 447, 496
Bluntnose darter, 262, 275, 281, 304,

305, 308, 430, 548
Bluntnose minnow, 263, 273, 279, 372,

505
Boleosoma

brevipinne, 429

olmstedi maculatum, 429

olmstedi ozarkanum, 429
Bowfin, 262, 294, 307, 319, 471
Brassy minnow, 262, 282, 296, 364,

446, 502
Brindled madtom, 262, 275, 397, 521
Brook silverside, 263, 278, 407, 529
Brook trout, 307

Brown bullhead, 262, 294, 391, 516
Brown trout, 307, 324
Bubalichthys

bonasus, 237, 379
Bullhead minnow, 262, 275, 370, 371,

372, 504
Burbot, 263, 276, 282, 296, 401, 525

Campostoma

anomalum, 373, 377, 507

anomalum anomalum, 373

anomalum plumbeum, 373

anomalum pulliuu, 373, 377, 507

oligolepis, 373, 506

ornatum, 377
Carassius

auratus, 328, 477
Carp, 263, 279, 280, 307, 326, 447,

477
Carpiodes

bison, 237, 382

carpio, 237, 382, 510

carpio carpio, 302, 382

carpio elongatus, 382

cyprinus, 380, 509

cyprinus heinei, 380

difformis, 382

forbesi, 380, 381

velifer, 382, 510
Catostomus

anisurus, 386

commersoni, 383, 511

nigricans, 384

teres, 383
Cenh-al silver\- minnow, 262, 278, 281,

308, 365, 370, 502
Central stoneroller, 263, 279, 281,

376, 377, 507
Centrarchus

macropterus, 420, 539
Ceratichthys

callarchus, 236, 371

perspicuus, 370

tenellus parviceps, 236, 371

tenellus tenellus, 371

vigilax, 370
Chaenobryttus

coronarius, 412

gulosus, 412
Chain pickerel, 262, 281. 325, 476
Channel catfish, 263, 279, 280, 294,

308, 392, 517
Channel darter, 262, 297, 304, 428,

446, 447, 546
Channel mimic shiner, 263, 276
Checkered madtom, 262, 273, 397,

522
Chestnut lamprey, 263, 280, 310, 311,

465
Chrosomus

erythrogaster, 331
Cliola

vigilax, 370, 372
Coliscus

parietalis, 236, 373
Common shiner, 262, 281, 296, 308,

346, 446, 489

Cottus

bairdi, 443, 558
carolinae, 443, 559

Creek chub, 263, 279, 282, 297, 329,
373, 413, 478

Creek chubsucker, 262, 281, 389, 515
Crystal darter, 262, 428, 546
Cycleptus

elongatus, 378, 507

Missouri Fishes

563

Cylindrosteus

agassizii, 236, 317
Cypress darter, 262, 274, 281, 441,

557
Cypress minnow, 262, 308, 370, 504
C>prinella

Billingsiana, 236, 357
C>prinus

caipio, 307, 326, 477

Dionda

nubila, 363, 501
Diplesion

blennioides, 433
Dorosoma

cepedianum, 322, 473

petenense, 322, 473
Drum, 263, 442, 558
Dusky darter, 262, 274, 275, 423, 542
Duskystripe shiner, 262, 272, 281,

295, 303, 344, 345, 363, 447, 488

Eastern slim minnow, 303
Elassoma

evergladei, 420

okefenokee, 420

zonatum, 420, 539
Emerald shiner, 263, 275, 276, 277,

278, 339, 350, 485
Entosphenus

lamottenii, 313
Ericymba

buccata, 364, 501
Erimystax, 335

dissimilis, 333
Eriniyzon

oblongus, 389, 515

oblongus claviformis, 389, 390

oblongus oblongus, 390

sucetta, 389, 515

sucetta kennerlyi, 389

tenuis, 390
Esox

americanus, 324, 390, 475

americanus vermiculatus, 324

lucius, 326, 476

masquinongy, 307, 326

niger, 325, 476

vermiculatus, 324, 325
Etheostoma

artesiae, 435

asprigene, 436, 553

aspro, 426

blennioides, 433, 551

blennioides newniani, 301, 433

blennioides pholidotum, 433

caeruleum, 437, 439, 555

caeruleum spectabile, 438

caprodes, 426

chlorosomum, 430, 441, 548

copelandi, 428

cragini, 237, 437, 554

cymatotaenia, 233, 237, 422

euzonum, 432, 550

euzonum erizonum, 237, 432

euzonum euzonum, 432

flabellare, 440, 556

flabellare flabellare, 301, 440

flabellare lineolatum, 440

gracile, 441, 556

histrio, 432, 550

juliae, 233, 237, 436, 553

microperca, 441, 557

nianguae, 233, 237, 434, 552

nigrum, 237, 429, 548

nigrum eulepis, 430

nigrum nigrum, 430

olmstedi, 430

pagei, 237, 437

pallididorsum, 437

proeliare, 441, 557

punctulatum, 232, 237, 436, 554

radiosum, 435

sagitta, 434

spectabfle, 232, 237, 438, 555

spectabfle pulchellum, 303, 439

spectabfle spectabfle, 303, 438

spectabfle squamosum, 237, 303,

439
spectabile uniporum, 237, 439
stigmaeum, 431, 549
tetrazonum, 237, 431, 432, 549
uranidea, 431
variatum, 299, 431, 432
variatum spectabfle, 438
whipplei, 435, 552
zonale, 433, 551
zonale arcansanum, 433
zoniferum, 441

Fantafl darter, 262, 273, 274, 440, 556

564

University of Kansas Publs., Mus. Nat. Hist.

Fathead minnow, 262, 276, 282, 307,

372, 391, 413, 506
Flathead catfish, 263, 279, 398, 522
Flathead chub, 263, 277, 278, 281,

335, 483
Flier, 262, 275, 420, 539
Freckled madtom, 262, 275, 277, 281,

395, 519
Freshwater drum, 263, 442, 558
Fundulus

catenatus, 401, 525

chrysotus, 403, 526

kansae, 402, 526

lineolatus, 404

notatus, 237, 404, 406, 528

notti, 404, 527

notti dispar, 404

notti notti, 404

olivaceus, 404, 528

sciadicus, 237, 403, 527

stellifer, 402

zebrinus, 403

Gambusia

affinis, 406, 529

afRnis affinis, 406
Ghost shiner, 262, 363, 500
Gilt darter, 262, 298, 426, 544
Gizzard shad, 263, 279, 322, 473
Goldeye, 263, 296, 323, 446, 474
Goldfish, 261, 307, 328, 477
Golden redhorse, 262, 273, 384, 385,

512
Golden shiner, 263, 328, 391, 478
Golden topminnow, 262, 308, 403, 526
Grass pickerel, 263, 278, 281, 324,

326, 475
Gravel chub, 262, 281, 334, 482

Grayling, 307

Green sunfish, 263, 279, 282, 294,

303, 372, 373, 391, 412, 415, 533
Greenside darter, 262, 271, 274, 281,

301, 304, 433, 446, 551

Hadropterus
evides, 426
phoxocephalus, 424

Harelip sucker, 308

Harlequin darter, 262, 308, 432, 550

Highfin carpsucker, 262, 382, 510
Hiodon

alosoides, 323, 324, 474

tergisus, 323, 474
Hog sucker, 262, 271, 384, 511
Hornyhead chub, 262, 308, 330, 331,

480
Hybognathus

amarus, 368

argyritis, 366, 368, 503

hankinsoni, 364, 365, 502

hayi, 365, 370, 504

meeki, 363

nubila, 363

nuchalis, 365, 368, 369, 502

nuchalis nuchalis, 365

nuchalis placita, 369

nuchalis regius, 365

placita, 368, 369

placitus, 365, 368, 369, 503

Hybopsis

aestivalis, 3.35, 338, 482
amblops, 333, 481
biguttatus, 331
dissimilis, 236, 333, 334, 481
dissimilis harryi, 236, 333
gelida, 337, 483
gelidus, 338
gracilis, 335, 483
gracilis gracilis, 296, 335
gracilis gulonellus, 302, 336
insignis, 334
kentuckiensis, 331
meeki, 233, 236, 338, 484
missuriensis, 233, 236, 358, 359
storeriana, 332, 338, 480
storerianus, 332
x-punctata, 236, 334, 482
x-punctata x-punctata, 236, 334

Hyborhynchus

notatus, 372
Hyodon

tergisus, 323
Hypentelium

etowanum, 384

nigricans, 384, 511

roanokense, 384

Ichthyobus

stolleyi, 237, 378

Missouri Fishes

565

Ichth\'oni\-zon

bdellium, 312

castaneus, 311, 465

concolor, 310

fossor, 310,311,312, 464

gagei, 311, 312,465

greeleyi, 312

imiciispis, 310, 315, 464
Ictalurus

anguilla, 393

furcatus, 237, 393, 518

furcatus furcatus, 393

lacustris, 393

lacustris punctatus, 393

lambda, 393

melas, 390, 392, 516

natalis, 392, 517

nebulosus, 391, 516

nebulosus marmoratus, 391

punctatus, 392, 517
Ictiobus

bubalus, 380, 509

carpio, 382

cyprinella, 378

cypriiiellus, 237, 378, 380, 508

niger, 237, 379, 380, 508

urus, 379

velifer, 380
Ironcolor shiner, 262, 275, 305, 348,

447, 490

Johnny darter, 263, 279, 281, 304,
429, 548

Kokanee, 307, 324

Labidesthes

sicculus, 407, 529

sicculus sicculus, 407
Labrax

chrysops, 408

Lagochila
lacera, 308

Lake chubsucker, 262, 281, 389, 515

Lake sturgeon, 263, 296, 308, 314,
467

Lampetra

aep>'ptera, 313, 466

appendix, 313

lamottei, 232, 236, 313, 466

wilderi, 313
Largemouth bass, 263, 280, 308, 409,

411, 417, 532
Largescale stoneroller, 262, 271, 273,

281, 298, 373, 377, 447, 506
Least brook lamprey, 262, 280, 301,

313, 446, 466
Least darter, 262, 281, 296, 297, 304,

441, 446, 447, 557
Lepibema

chrysops, 408
Lepidosteus

osseus, 318
Lepisosteus

oculatus, 318, 470

osseus, 318, 470

osseus oxyurus, 318

platostomus, 236, 317, 469

platyrhincus, 318

productus, 318

spatula, 317, 469
Lepomis

anagallinus, 415

cyanellus, 412, 533

gibbosus, 414, 535

gulosus, 412, 533

humilis, 415, 536

niacrochirus, 417, 537

macrochirus niacrochirus, 417

megalotis, 416, 536

megalotis megalotis, 416

megalotis peltastes, 416

microlophus, 414, 415, 535

pallidus, 417

punctatus, 414, 534

punctatus miniatus, 414

symmetricus, 413, 414, 534
Leptops

olivaris, 398
Logperch, 262, 273, 277, 426, 544
Longear sunfish, 263, 274, 278, 413,

416, 536
Longnose darter, 262, 294, 299, 307,

425, 446, 543

Longnose gar, 263, 308, 318, 470

Lota

lota, 401, 525

Lucius

reticulatus, 325
vermiculatus, 324

566

University of Kansas Publs., Mus. Nat. Hist.

Megastomatobus

cyprinella, 378
Menidia, 294

audens, 407, 530

beryllina, 408
Micropterus

coosae, 410

dolomieu, 410

dolomieu dolomieu, 410

dolomieu velox, 410

dolomieui, 410, 532

punctulatus, 409, 531

punctulatus henshalli, 410

punctulatus punctulatus, 409, 410

salmoides, 411, 532

salmoides salmoides, 411
Mimic shiner, 362, 500
Minytrema

melanops, 388, 389, 514
Mirror shiner, 301, 362
Mississippi silverside, 263, 276, 278,

308, 407, 530
Missouri saddled darter, 262, 273, 294,

299, 431, 432, 446, 549
Moniana

Jugalis, 236, 357

Mooneye, 263, 323, 474

Morone

chrysops, 408, 530

interrupta, 237, 409

mississippiensis, 237, 408, 531

saxatilis, 307, 409
Mosquitofish, 262, 275, 307, 308, 406,

529
Motded sculpin, 262, 271, 297, 310,

443, 446, 558
Moxostoma

anisurum, 386, 513

aureolum, 386

aureolum pisolabnun, 237, 386

breviceps, 386

carinatum, 388, 514

duquesnei, 384, 386, 512

duquesnei duquesnei, 384

erythrurum, 237, 385, 512

macrolepidotum, 385, 386, 513

macrolepidotum breviceps, 388

macrolepidotum duquesnei, 385

macrolepidotum macrolepidotum,
296, 387

macrolepidotum pisolabrum, 237,
386
Mud darter, 262, 436, 553
Muskellunge, 297, 307, 326

Neosho madtom, 262, 303, 396, 521
Niangua darter, 262, 295, 301, 434,

446, 552
Nocomis

biguttatus, 331, 480

effusus, 332
Northern brook lamprey, 262, 296,

297, 311, 446, 464
Northern cavefish, 302
Northern flathead chub, 296, 446
Northern hog sucker, 262, 271, 384,

511
Northern logperch, 296, 446
Northern madtom, 303
Northern mimic shiner, 263, 276, 278,

303
Northern orangethroat darter, 303
Northern pike, 261, 282, 296, 307,

326, 446, 476
Northern redhorse, 263, 279, 385,

386, 446, 513
Northern river carpsucker, 302
Northern studBsh, 262, 271, 301, 307,

308, 401, 446, 525
Notemigonus

americanus crysoleucas, 328

crysoleucas, 328, 478

crysoleucas auratus, 328
Notropis

amnis, 353, 493

amnis amnis, 353

amnis pinnosus, 353

analostanus, 354

ariommus, 341

atherinoides, 236, 339, 485

atherinoides atherinoides, 339

blennius, 350, 491

blennius blennius, 350

blennius jejunus, 350

boops, 351, 492

buchanani, 363, 500

camurus, 356, 496

cayuga, 361

chalybaeus, 348, 490

chrysocephalus, 347, 489

chrysocephalus chrysocephalus, 347

Missouri Fishes

567

cornutiis, 346, 347, 348, 489

coinutiis chrysocephalus, 347

deliciosus, 358

deliciosiis niissoiiriensis, 358

dilcctus, 339, 340

dorsalis, 352, 364, 493

dorsalis dorsalis, 353

dorsalis piptolepis, 353

fumeus, 343, 487

galacturus, 356, 357, 495

gilberti, 352

greenei, 351, 492

heterolepis, 361, 499

hudsoniiis, 349, 491

illecebrosus, 343, 351

jejunus, 350

luh-ensis, 236, 356, 357, 496

lutrensis lutrensis, 302

maculatiis, 360, 498

megalops, 346, 347

micropteryx, 341

notatus, 354, 355

ozarcanus, 233, 236, 361, 499

percobromus, 339, 340

pilsbryi, 344, 345, 488

rubellus, 339, 340, 485

rubrifrons, 339, 340

sabinae, 358, 497

scabriceps, 351

shumardi, 343, 352, 354, 487

spectrunculus, 362

spilopterus, 354, 355, 358, 494

spilopterus h\'psisomatiis, 354

stramineus, 236, 358, 497

stramineus missuriensis, 233, 302,

359
stramineus stramineus, 359
telescopus, 341, 486
texanus, 349, 490
topeka, 360, 498
umbratilis, 342, 486
imibratilis cyanocephalus, 342
umbratilis umbratilis, 342
venustus, 355, 370, 495
venustus venustus, 355
volucellus, 362, 363, 500
volucellus volucellus, 303, 362
volucellus ssp., 303
volucellus wicklifR, 362
whipplei, 354, 358, 494
whipplii, 354

zonatus, 232, 236, 344, 345, 488
zonatus pilsbryi, 345
zonatus zonatus, 344
Noturus

albater, 237, 396, 520
exilis, 394, 519
flavater, 237, 397, 522
flavipinnis, 397, 398
flavus, 395, 396, 520
gyrinus, 394, 518
miurus, 397, 521
nocturnus, 395, 519
placidus, 303, 396, 521
stigmosus, 303

Oncorhynchus

nerka, 307, 324

sp., 307
Opsopoeodus

emiliae, 330, 479
Orangespotted sunfish, 262, 276, 303,

415, 536
Orangethroat darter, 262, 273, 274,

281, 438, 555
Ozark cavefish, 262, 294, 302, 399,

446, 524
Ozark madtom, 262, 273, 396, 520
Ozark minnow, 262, 271, 272, 297,

298, 363, 446, 501
Ozark shiner, 262, 273, 294, 301, 308,

361, 446, 499

Pacific salmon, 307
Paddlefish, 263, 308, 310, 316, 468
Pallid shiner, 262, 308, 353, 493
Pallid sturgeon, 263, 276, 296, 316,

446, 468
Parascaphirhynchus

albus, 236, 316
Perca

flavescens, 307, 422, 541
Percina

aurolineata, 424

caprodes, 426, 544

caprodes caprodes, 426

caprodes carbonaria, 426

caprodes semifasciata, 296, 426

copelandi, 428, 546

cymatotaenia, 233, 237, 422, 541

evides, 426, 544

maculata, 423, 542

568

Unwersity of Kansas Publs., Mus. Nat. Hist.

nasuta, 425, 543

oxyrhv-ncha, 299, 425

phoxocephala, 424, 425, 543

sciera, 423, 542

sclera apristis, 424

sciera sciera, 424

shumardi, 427, 545

uranidea, 427, 545
Percopsis

omiscomaycus, 398, 523
Petromyzon

lamottenii, 232, 236, 313
Phenacobius

mirabilis, 236, 338, 484

mirabilis scopifer, 338
Phoxinus

erythrogaster, 331, 479
Pimephales

notatus, 370, 372, 505

perspicuus, 370

promelas, 236, 372, 506

promelas promelas, 373

tenellus, 371, 505

tenellus parviceps, 236, 303, 371

tenellus tenellus, 303, 371

vigilax, 370, 504

vigilax perspicuus, 370

vigilax vigilax, 370
Pirate perch, 262, 274, 275, 399, 523
Placophar>Tix

carinatus, 388
Plains killifish, 262, 302, 402, 447, Redbelly dace, 262, 271, 282, 297,

whipplii whipplii, 435
Polyodon

spathula, 232, 236, 316, 468
Pomolobus

chrysochloris, 320
Pomoxis

annularis, 237, 419, 538

nigromaculatus, 418, 420, 538

sparoides, 418
Pomoxys

annularis, 419

brevicauda, 237, 419
Procerus

maculatus, 232, 236, 316
Ptychostomus

bucco, 237, 385
Pugnose minnow, 262, 275, 308, 330,

479
Pumpkinseed, 261, 282, 296, 414, 415,

535
Pygmy sunfish, 262, 275, 420, 539
Pylodictis

olivaris, 398, 522

Quillback, 262, 276, 380, 509

Rainbow darter, 262, 271, 281, 297,

298, 437, 555
Rainbow trout, 262, 307, 324, 475
Red shiner, 262, 276, 278, 302, 357,

359, 496

526

310, .331, 479

Plains minnow, 263, 277, 278, 302, Redear sunfish, 262, 294, 415, 535

369, 447, 503
Plains orangethroat darter, 303
Plains sand shiner, 302
Plains topminnow, 262, 303, 403, 447,

527
Platygobio

gracilis, 335
Poecilichthys

beani, 237, 429

caeruleus, 438

euzonus erizonus, 237, 432

nianguae, 434

punctiilatus, 232, 237, 436

spectabilis, 232, 237, 438

tetrazonus, 237, 431

variatus, 431

versicolor, 437

Redfin darter, 262, 308, 435, 552
Redfin shiner, 263, 273, 304, 342, 486
Ribbon shiner, 262, 274, 275, 343, 487
River carpsucker, 262, 276, 382, 383,
510

River darter, 263, 277, 278, 281, 427,

545
River redhorse, 262, 388, 514

River shiner, 263, 276, 277, 281, 350,
491

Roccus

chrysops, 408

Rock bass, 262, 273, 294, 307, 417,
537

Rosyface shiner, 262, 271, 273, 297,
298, 340, 485

Missouri Fishes

569

Sabiiie shiner, 262, 308, 358, 497
Salnio

gairdneri, 307, 324, 475

iiideus, 324

salar, 307

ti-utta, 307, 324
Salvelinus

fontinalis, 307
Sand shiner, 262, 276, 358, 497
Sarcidium

scopiferum, 236, 338
Sanger, 263, 276, 421, 540
Scaly sand darter, 262, 429, 547
Scaphirhynchiis

album, 316

albus, 236, 315, 316, 468

platorynchus, 315, 467
Schilbeodes

exilis, 394

insignis, 394

miurus, 397
SemotHus

atromaculatus, 329, 478
Shaipnose darter, 299
Shortnose gar, 263, 317, 318, 319, 469
Shovelnose sturgeon, 263, 276, 310,

315, 467
Sicklefin chub, 263, 276, 277, 296,

338, 446, 484
Silver chub, 263, 276, 332, 480
Silver lamprey, 263, 276, 280, 282,

296, 310, 312, 446, 464
Silver redhorse, 262, 386, 513
Silverband shiner, 263, 276, 343, 487
Silverjaw minnow, 262, 281, 364, 501
Skipjack herring, 263, 278, 308, 320,

472
Slender madtom, 262, 273, 301, 394,

396, 519
Slenderhead darter, 263, 277, 304,

424, 447, 543
Slim minnow, 262, 273, 308, 371, 505
Slough darter, 262, 275, 304, 305,

441, 556
Smallmouth bass, 262, 273, 281, 307,

409, 410, 532
Smallmouth buffalo, 263, 294, 378,

380, 509
Smoothlip northern redhorse, 296

Southern brook lamprey, 262, 312,

465
Soudiern cavefish, 262, 302, 400, 446,

524 J

Southern flathead chub, 302
Southern redbelly dace, 262, 271, 282,

297, 310, 331, 479
Southwestern mimic shiner, 303
Speckled chub, 263, 278, 335, 482
Speckled darter, 263, 278, 430, 431,

549
Spotfin shiner, 262, 297, 304, 354,

357, 446, 447, 494
Spottail shiner, 263, 276, 282, 296,

349, 491
Spotted bass, 263, 275, 278, 281-82,

307, 308, 409, 411, 531
Spotted gar, 262, 318, 470
Spotted sucker, 262, 388, 514
Spotted sunfish, 262, 305, 414, 447,

534
Starhead topminnow, 262, 404, 527
Stargazing darter, 262, 275, 281, 427,

545
Steelcolor shiner, 262, 354, 355, 357,

494
Stippled darter, 262, 281, 436, 554
Stizostedion

canadense, 421, 540

vitreum, 420, 540

vitreum vitreum, 420
Stonecat, 262, 277, 304, 395, 396, 398,

447, 520
Streamline chub, 262, 273, 281, 333,

481
Striped bass, 307
Striped fantail darter, 281
Striped shiner, 262, 281, 308, 347,

352, 489
Studfish, 262, 271, 304, 307, 308, 401,

446, 525
Sturgeon chub, 263, 276, 277, 296,

337, 446,483

Suckermouth minnow, 262, 276, 302,

338, 484

Tadpole madtom, 262, 275, 281, 394,

518
Taillight shiner, 262, 308, 360, 498
Telescope shiner, 262, 273, 301, 341,

446, 486

570

University of Kansas Publs., Mus. Nat. Hist.

Tench, 307

Threadfin shad, 263, 276, 278, 282,

307, 308, 322, 473
Thymallus

arcticus, 307
Tinea

tinea, 307
Topeka shiner, 262, 302, 303, 304,

360, 447, 498
TrogHchthys

rosae, 399
Trout pereh, 262, 282, 296, 398, 523

T>phHehthys
eigenmanni, 400
rosae, 237, 399
subterraneus, 399, 400, 524

Uranidea

richardsoni, 443

Variegate darter, 299

Walleye, 263, 308, 420, 540

Warmouth, 262, 412, 533

Wedgespot shiner, 262, 272, 295, 351,
492

Weed shiner, 262, 274, 305, 349, 447,

490
Western sand darter, 262, 277, 428,

547
Western silvery minnow, 263, 277,

281, 296, 368, 369, 446, 503
Western slim mirmow, 303, 447
White bass, 263, 308, 408, 530
\\'hite crappie, 263, 308, 419, 538
White sucker, 263, 279, 282, 383,

385, 511
Whitetail shiner, 262, 273, 301, 303,

355, 356, 357, 446, 495

Xystroplites
her OS, 415

Yellow bass, 263, 408, 531

Yellow bullhead, 263, 391, 392, 517

Yellow perch, 261, 282, 296, 297, 307,

422, 541
Yoke darter, 262, 273, 295, 436, 553

Zygonectes

macdonaldi, 237, 403
notatus, 404, 406
zonatus, 237, 406

INDEX TO \^OLUME 20

New systematic names are in boldface type
See pp. 561-570 for index to "A distributional study of Missouri fishes"

Acrochordus javaniciis, 203
acutus, Cerberus, 58, 160
aer

Cohiber, 86

Homalopsis, 58, 61, 86

Hvpsirhina, 86
albolabris, Tomodactylus, 42
albolineata, Hypsirhina, 58, 87
albomaculata

Enhvdris, 58, 63, 64, 65, 66, 67, 68,
69, 74, 78, 122, 125, 214

Hypsirhina, 125
albomaculatus, Homalopsis, 58, 125
alternans

Brachyorrhos, 58, 62, 77

Enhvdris, 58, 59, 63, 64, 65, 66, 67,
68, 69, 70, 72, 75, 77, 79, 80, 81,
84, 213

Eurostus, 77

Miralia, 77
anderssoni, Eleutherodactylus, 4
Anisodontiens, 56
annulata, Cantoria, 58, 129, 182, 183,

186, 207, 209, 215
areolatus

Eleutherodactylus, 4

Syrrhophus, 4
atrocaeruleus, Hydrus, 58, 86
australis

Cerberus, 58, 60, 148, 149, 150,
151, 152, 143, 155, 157, 158,
159, 207, 215

Cerberus rynchops, 60, 157

Homalopsis, 58, 157

Batrachyla longipes, 4, 30, 31, 32
batrachylus, Eleutherodactylus, 30
bennetti

Enhydris, 58, 59, 60, 63, 64, 65, 66,
67, 68, 69, 73, 95, 100, 103, 105,
106, 113, 208, 214

Hypsirhina, 58, 103
bicolor

Fordonia, 178

Gerarda, 58, 174, 175
bilineata

Hurria, 58, 146, 147, 160

Hypsirhina, 86

Hypsirhina enhydris, 59, 87

Bitia, 52, 56, 57, 59, 61, 155, 187,
205, 207, 209, 210, 215, 216
hydroides, 59, 155, 187
blandfordi, Hypsirhina, 58, 114
boaeformis
Cerberus, 160
Elaps, 58, 146, 147, 160
Homalopsis, 160
bocourti

Enhydris, 58, 59, 60, 63, 64, 65,
66, 67, 68, 69, 70, 74, 107, 122,
123, 124, 125, 204, 206, 207, 214
Hypsirhina, 58, 122
borneensis, Pythonopsis, 59, 62, 127
Brachyorrhos alternans, 58, 62, 77
buccata

Homalopsis, 54, 59, 60, 61, 136,
140, 141, 142, 143, 144, 145,
193, 196, 199, 200, 201, 202,
204
Vipera, 136
buccatus. Coluber, 59, 136, 140

caerulea, Enhydris, 59, 61, 71, 86
calcaratus, Syrrhophus, 4
campi,

Syrrhophus, 4, 15

Syrrhophus cystignathoides, 4, 7,
12, 13, 15, 16, 17
Campyledon, 57, 174, 175

prevostianum, 174, 175
Cantoria, 57, 58, 59, 61, 129, 182,

183, 184, 185, 186, 207, 209,
210, 215, 216

annulata, 58, 129, 182, 183, 186,

207, 209, 215
dayana, 59, 183
elapiformis, 183
elongata, 59, 183
violacea, 59, 61, 129, 182, 183,

184, 185, 207, 209, 215
caryophyllaceus

Eleutherodactylus, 4

Syrrhophus, 4
caspius

Enhydris, 71

Hydrus, 71
Cerbere, 57
Cerberus, Coluber, 59, 146, 147, 160

-Univ. Kansas Publs. Mus. Nat. Hist., Vol. 20, ( 1970-1971 )

571

572

Unrtersity of Kansas Publs., Mus. Nat. Hist.

Cerberus, 58, 59, 60, 61, 71, 146,
147, 148, 149, 150, 151, 152,
153, 154, 155, 157, 158, 159,
160, 161, 163, 164, 165, 166,
167, 168, 169, 199, 200, 201,
203, 204, 205, 207, 208, 210,
212, 214, 215, 216, 217

aciitus, 58, 160

australis, 58, 60, 148, 149, 150, 151,

152, 153, 155, 157, 158, 159,
207, 215

boaeformis, 160

cinereus, 154, 157, 160

grantii, 59, 160

microlepis, 60, 149, 150, 151, 162,

153, 154, 155, 157, 207, 215
russelli, 60

r>aichops, 148, 157, 159, 160, 161,

199, 200, 203, 204
lynchops australis, 60, 157
rynchops novaeguineae, 60, 148,

149, 150, 151, 152, 153, 154,
155, 169, 207, 215

rynchops rynchops, 58, 59, 60, 149,

150, 151, 152, 153, 154, 155, 157,
160, 163, 164, 165, 166, 167, 168,
201, 207

unicolor, 61, 160
chalcaus

Eleutherodactylus, 4

Phyllobates, 4
chalybaeus, Hemiodontus, 59, 178,

181
chiensis, Hypsirhina, 59, 100
chinensis, Enhydris, 57, 59, 63, 64,

65, 66. 67, 68, 69, 73, 81, 100, 105,

118, 203, 204, 205, 206, 208, 214,

217
cholorum, Syrrhophus leprus, 4, 18
cinereus

Cerberus, 154, 157, 160

Hydrus, 59, 160
coeruleus

Eleutherodactykis, 4

Syrrhophus, 4
Colostethus pahnatus, 4
Coluber, 59, 60, 82, 86, 136, 140, 146,
147, 160, 174, 177

aer, 86

buccatus, 59, 136, 140

Cerberus, 59, 146, 147, 160

horridus, 59, 136

monilis, 60, 136, 140

obtusatus, 60, 160

plumbea, 82

prevostianus, 60, 174, 177

pythonissa, 60, 86

subalbidus, 60, 136
cystigathoides

Phyllobates, 4, 14, 16

Syrrhophus, 4, 6, 7, 10, 11, 12, 13,

Svrrhophus — Concluded
' 14, 15, 16, 17, 18, 41
Cystignathus, 18

dayana, Cantoria, 59, 183
decussata, Homalopsis, 59, 77
dennisi, Syrrhophus, 4, 5, 8, 9, 11, 13,

27, 29, 32, 41
diastema, Eleutherodactylus, 4
Dieurostus, 62
dilatus, Tomodactylus, 42
doriae

Enhydris, 59, 63, 64, 65, 66, 67, 68,
69, 70, 74, 78, 127, 130, 131,
132, 214

Homalopsis, 59, 62, 130

Hypsirhina, 130
dorsalis, Enhydris, 71
dussumieri

Enhydris, 59, 60, 63, 64, 65, 66, 67,
68, 69, 74, 78, 118, 121, 122

Eurostus, 59, 62, 87, 120

Hypsirhina, 121

elapiformis

Cantoria, 183

Hemiodontus, 183, 185

Hydrodipsas, 59, 182, 183
Elaps, 58, 146, 147, 160

boaeformis, 58, 146, 147, 160
Elaphas, 215
Eleutherodactylus, 4, 5, 12, 13, 30

anderssoni, 4

areolatus, 4

batrachylus, 30

carxophyllaceus, 4

chalceus, 4

coeruleus, 4

diastema, 4

hylaeformis, 4

longipes, 30

lutosus, 4

molinoi, 4

omiltemanus, 4

pardalis, 4

punctariolus, 4

rhodopis, 4, 12

ridens, 4
elongata, Cantoria, 59, 183
Epirhexus, 30, 32

longipes, 30
Erpeton, 52, 56, 57, 59, 60, 61, 81,
189, 190, 191, 207, 209, 210, 215,
216

punctata, 59

tentaculatum, 204

tentaculatus, 59, 60, 81, 189, 190,
191
enhydris

Enhydris, 58, 59, 60, 63, 64, 65, 66,
67, 68, 69, 72, 86, 87, 90, 91,
199, 200, 203, 204, 205, 206, 213

Index to Volume 20

573

enhydris — Concluded

Enhydris enhydris, 90

Homalopsis, 80

Hydius, 59, 61, 71, 86
Hypsirhina, 86
Enhvdris, 58, 59, 60, 61, 63, 64, 65,

66, 67, 68, 69, 70, 71, 72, 73, 74,
75, 76, 77, 78, 79, 80, 81, 82, 84,
85, 86, 87, 90, 91, 94, 95, 96, 97,
98, 99, 100, 103, 105, 106, 107,
110, 112, 113, 114, 115, 116, 117,
118, 119, 120, 121, 122, 123, 124,
125, 127, 129, 130, 131, 132, 134,
171, 172, 203, 204, 205, 206, 207,
208, 210, 212, 213, 214, 216, 217

albomaciilata, 58, 63, 64, 65, 66, 67,
68, 69, 74, 78, 122, 125, 214

alternans, 58, 59, 63, 64, 65, 66, 67,
68, 69, 70, 72, 75, 77, 79, 80, 81,
84, 213

bennetti, 58, 59, 60, 63, 64, 65, 66,

67, 68, 69, 73, 95, 100, 103, 105,
106, 113, 208, 214

bocourti, 58, 59, 60, 63, 64, 65, 66,

67, 68, 69, 70, 74, 107, 122, 123,
124, 125, 204, 206, 207, 214

bocourti soctrangensis, 124

caerulea, 59, 61, 71, 86

chinensis, 57, 59, 63, 64, 65, 66, 67.

68, 69, 73, 81, 100, 105, 118,
203, 204, 205, 206, 208, 214, 217

doriae, 59, 63, 64, 65, 66, 67, 68,

69, 70, 74, 78, 127, 130, 131, 132,
214

dorsalis, 71

diissiimieri, 59, 60, 63, 64, 65, 66,

67, 68, 69, 74, 78, 118, 121, 122

enhvdris, 58, 59, 60, 63, 64, 65, 66,

67, 68, 69, 72, 86, 87, 90, 91,
199, 200, 203, 204, 205, 206, 213

enhydris enhydris, 90

enhvdris subtaeniata, 90

indi'ca, 59, 63, 64, 65, 66, 67, 68,

69, 70, 72, 75, 76, 78, 80, 84,

212, 213
innominata, 59, 63, 64, 65, 66, 67,

68, 69, 70, 73, 81, 86, 94, 96, 97,
212, 213

jagorii, 59, 63, 64, 65, 66, 67, 68,

69, 73, 86, 90, 92, 94, 95, 213
longicauda, 60, 63, 64, 65, 66, 67,

68, 69, 70, 73, 100, 105, 107,
113,212,214

niacleayi, 60, 63, 64, 65, 66, 67, 68,

69, 70, 73, 107, 108, 110, 206,
212, 214

maculosa, 54, 58, 60, 63, 64, 65, 66,

67, 68, 69, 74, 107, 113, 114
115, 118,206, 214

matannensis, 60, 63, 64, 65, 66, 67,

68, 69, 73, 78, 107, 112, 206,

matannensis — Concluded

212, 214
pahangensis, 60, 63, 64, 65, 66, 67,

68, 69, 74, 95, 113, 114, 116,

117, 118,212,214
pakistanica, 60, 63, 64, 65, 66, 67,

68, 69, 70, 78, 132, 134, 205, 214
phuiibea, 59, 60, 63, 64, 65, 66, 67,

68, 69, 72, 75, 82, 84, 85, 203,

204, 205, 213

polylepis, 60, 63, 64, 65, 66, 67, 68,

69, 70, 72, 107, 110, 204, 214
punctata, 59, 60, 63, 64, 65, 66, 67,

68, 69, 70, 74, 127, 129, 132,

205, 214
rynchops, 71

sieboldi, 59, 60, 63, 64, 65, 66, 67,

68, 69, 70, 74, 81, 118, 119, 120,

203, 206, 207, 214

smithi, 60, 63, 64, 65, 66, 67, 68,

69, 70, 73, 78, 86, 97, 98, 99, 213
Enhydrus

caspius, 71

pahistris, 71

piscator, 71
Eupsophus juninensis, 4
Euroste, 57
Eurostus, 59, 62, 77, 82, 87, 92, 127

alternans, 77

dussumieri, 59, 62, 87, 120

heterapis, 59, 127

plumbeus, 82

fasciatus, Hipistes. 59, 187
Ferania, 58, 62, 118, 120, 123

sieboldi, 118, 123
Feranodies, 59, 62, 119, 120
Feranioides jamnaeticus, 59, 62, 119,

120
festae

Niceforonia, 4
Paludicola, 4
flavescens, Heleophis, 59, 174, 175
Fordonia, 58, 59, 60, 61, 129, 177,
178, 179, 180, 181, 200, 201,
202, 203, 204, 205, 207, 209,
210, 215, 216
bicolor, 178

leucobalia, 59, 60, 61, 129, 177,
178, 179, 180, 181, 200, 201,
202, 203, 204, 205, 217
papuensis, 60, 178, 181
unicolor, 61, 178
variabilis, 61. 178
furcata, Hypsirhina, 59, 86

gaigeae, Syrrhophus, 3, 4, 22, 23, 25
Gerarda, 58, 59, 60, 61, 95, 174, 175,

204, 207, 209, 210, 215, 216
bicolor, 58, 174, 175
prevostiana, 58, 59, 60, 95, 174,

204

574

University of Kansas Publs., Mus. Nat. Hist.

Gerardia, 174, 175

prevostiana, 175
gigantea, Hypsirhina, 59, 123
grantii, Cerberus, 59, 160
giittilatus

Malachvlodes, 4, 22, 23, 25

Svrrhophus, 4, 5, 7, 8, 9, 10, 11,
14, 22, 23, 24, 25, 27, 41
guttulatus

Syrrhaphus, 22

Syrrhopus, 22
Gyi, Ko Ko. A revision of Colubrid

snakes of the subfamily Homalop-

sinae, 47

hageni, Hypsirhina, 59, 127
Hannemania, 25
hardwickii

Homalopsis, 137

Hypsirhina, 59, 82
hedemanni, Tach\plotus, 59, 127
Heleophis, 59, 174. 175

flavescens, 59, 174, 175
Helicops indicus, 87
Hemiodontus, 59, 177, 178, 183, 185

chalybaeus, 59, 178. 181

elapiformis, 183, 185

leucobalia, 177, 178
Herpeton, 58, 189, 190

tentaculatum, 190
herpeton, Homalopsis, 59, 190
heterapis, Eurostus, 59, 127
Heurnia, 58, 61, 110, 170, 207, 208,
209. 210, 215, 216

ventromaculata. 61, 110, 170, 215
Hipistes, 58, 59, 187

fasciatus, 59. 187

hydrinus. 187
Homalopsis. 57. 58. 59, 60. 61. 62.82.
86. 118. 125. 130. 135. 136, 140,
141, 142, 143, 144, 145, 146, 157,
160, 174, 177. 187, 189, 190, 193,
196, 199, 200, 201, 202, 203, 204,
207. 208, 210, 212, 216

aer, 58, 61, 86

albomaculatus, 58, 125

austral is, 58, 157

boaeformis, 160

buccata, 54, 59, 60, 61. 136. 140,
141, 142, 143, 144, 145, 193, 196,
199, 200, 201. 202, 204

decussata, 59, 77

doriae, 59, 62, 130

enhydris, 86

hardwickii, 137

herpeton, 59, 190

hydrina, 59, 187

leucobalia, 60, 177

molurus, 60, 136, 160

monilis, 136

olivaceaus, 60, 86

Homalopsis — Concluded

plumbea, 60, 62, 82

rvnchops, 160

sieboldii, 60, 62, 118

semizonata, 60, 137
horridus. Coluber, 59, 136
Hurria, 58, 60, 71, 146, 147, 154, 160,
161

bilineata, 58, 146, 147, 160

microlepis, 154

rynchops, 161

schneideriana, 60, 160
hydrina. Homalopsis, 59, 187
hvdrinus, Hipistes, 187
Hvdrodipsas, 59, 182, 183. 177

elapiformis, 59, 182, 183
hydroides, Bitia, 59, 155, 187
Hydropsis, 177

Hvdrus, 58, 59, 60, 61, 71, 86, 97,
146, 147, 159. 160

atrocaeruleus, 58, 86

caspius, 71

cinereus, 59, 160

enhydris. 59. 61, 71, 86

innominata. 97

rynchops, 60, 146, 147, 159, 160

rynchops rynchops, 160
hylaeformis

Eleutherodactylus, 4

Phyllobates, 4
Hypsirhina. 58, 59. 60, 61, 71. 82. 86.
87. 92, 96. 98, 100. 103, 105,
108, 110, 112, 114, 118, 121, 122,
123. 125, 127, 130

aer, 86

albolineata, 58, 87

albomaculata. 125

bennetti, 58. 103

bilineata. 86

blanfordi, 58, 114

bocourti, 58, 122

bocourti soctrangensis, 59, 123

chiensis. 59, 100

doriae, 130

clussumieri. 121

enhydris, 86

enhydris bilineata, 59, 87

enhydris maculata, 59. 103

enhvdris subtaeniata, 59, 92

furcata, 59, 86

gigantea. 59, 123

hageni, 59, 127

hardwickii, 59, 82

indica, 75

innominata, 59, 96

jagorii, 59, 92

iongicauda. 60, 105

maclea\i, 108

maculata, 60, 103. 114

maculosa, 60, 114

malabarica, 60, 121, 122

Index to Volume 20

575

Hj'psiiliina — Concluded

matannensis, 60, 112

miiltilineata, 60

polylepis, 60, 110

plumbea, 82

punctata, 127

sieboldi. 118

smithi, 60, 98

tiilineata, 60, 86
H\psirhine, 57
Hysiscopiis, 62

indica

Enh^■dri.s, 59, 63, 64, 65, 66, 67, 68,
69, 70, 72, 75, 76, 78, 80, 84,
212, 213
Hypsiihina, 75
Raclitia, 59, 62, 75
indicus, Helicops, 87
ineptus, Syrrhophus, 4
innominata

Enhvdris, 59. 63, 64, 65, 66, 67, 68,
69, 70, 73, 81, 86, 94, 96, 97,
212, 213
Hvdru's, 97
HNpsirhina, 59, 96
interorbitalis

Syrrhophus, 4, 5, 8, 12, 14, 36, 37.

39
Syrrhopis, 36

jagorii

Enhydris, 59, 63, 64, 65, 66, 67, 68,
69, 73, 86, 90, 92, 94, 95, 213

Hypsiihina, 59, 92
jamnaeticus, Feranioides, 59, 62, 119,

120
javaniciis, Arcochordus, 203
jiminensis

Eupsophus, 4

Syrrhophus, 4

kentii, Neospades, 60, 172

Lapemis, 71

latodactyhis, Syrrhophus, 4, 27, 28,

30, 31, 32, 33
leprus, Syrrhophus, 3, 4, 5, 7, 10, 11,

13, 18, 19, 20, 22, 27, 41
Leptophis, 207
leucobalia

Fordonia, 59, 60, 61, 129, 177, 178,
179, 180, 181, 200, 201, 202, 203,
204, 205, 217
Hemiodontus, 177, 178
Homalopsis, 60, 177
longicauda

Enhvdris, 60. 63, 64, 65, 66, 67, 68,
69, 70, 73, 100, 105, 107, 113,
212, 214
Hypsirhina, 60, 105

Hingipes

Batrachyla, 4, 30, 31, 32

Eleutherodactyhis, 30

Epirhexis, 30

Syrrhophus, 4, 5, 6, 7, 8, 9, 11, 13,
30, 31, 32, 41
lusingtonii, Potaniophis, 60
hitosus

Eleutherodactyhis, 4

Syrrhophus, 4
Lynch. John D. A taxonomic revision

of the Leptodactylid frog genus

Syrrhophus Cope, 1

macleavi

Enhydris, 60, 63, 64, 65, 66, 67, 68,
69, 70, 73, 107, 108, 110, 206,
212, 214

Hypsirhina, 108

Pseudoferania, 60, 62, 108
macrotyanpanum

Syrrhophus, 27

Tomodactyhis, 3, 4, 27
maculata

Hypsirhina, 60, 103, 114

Hypsirhina enhydris, 103
maculosa

Enhvdris, 54, 58, 60, 63, 64, 65, 66,
67, 68, 69, 74, 107, 113, 114,
115, 118, 206, 214

Hypsirhina. 60, 114
malabarica. Hypsirhina, 60, 121, 122
Malachvlodes guttilatus, 4, 22, 23, 25
marnocld, Syrrophus, 25
marnockii. Syrrhophus, 3, 4, 5, 6, 8,

9, 10, 12, 13, 14, 22, 23, 25, 26, 27
matannensis

Enhvdris, 60, 63, 64. 65. «6, 67. 68,
69, 73, 78, 107, 112, 206, 212,
214

Hypsirhina, 60, 112
microlepis

Cerberus, 60, 149. 150. 151, 152,
153, 154, 155, 157, 207, £15

Hurria, 154
Miralia, 62, 77

alternans, 77
modestus

Svrrhophis, 37

Svrrhophus, 4, 5, 6. 7. 8, 10, 12, 14,
' 37, 38, 39, 40, 41, 43
modestus pallidus

S\rrhophis, 40

Syrrhophus, 4
niolinoi

Eleutherodactylus, 4

Syrrhophus, 4
molurus, Homalopsis, 60, 136, 160

monilis

Coluber, 60, 136, 140
Homalopsis, 136

576

University of Kansas Publs., Mus. Nat. Hist.

montia, Niceforonia, 4
montiiim, Syirhophus, 4
multilineata, Hypsirhina, 60
Myron, 58, 60, 61, 110, 172, 173, 207,
208, 209, 210, 215, 216, 217

richardsoni, 60, 110, 172, 173, 215

richardsonii, 60, 172
mystaceus, Syrrhophiis, 4

NatrLx, 71
nebulosus

Syirhophus, 4, 34

Syrrhophus pipilans, 4, 8, 12, 14,
33, 34, 35, 41
Neospades, 160, 172

kentii, 60, 172
Niceforonia

festae, 4

montia, 4
nitidus, Tomodactyhis, 42, 43
nivocolimae, Svrrhophiis, 4, 5, 6, 8,

10, 12, 14, 36, 37, 38, 39, 40, 41
novaeguineae, Cerberus rynchops, 60,

148, 149, 150, 151, 152, 153, 154,

155, 169, 207, 215

obesus, Syrrhophus, 4
obtusatus, Cohiber, 60, 160
olivaceaus, Homalopsis, 60, 86
omiltemanus

Eleutherodactylus, 4

Syrrhophus, 4

pahangensis. Enhvdris, 60, 63, 64, 65,

66,^67. 68. 69, 74, 95, 113, 114,

116, 117, 118,212,214
pakistanica, Enhydris, 60, 63, 64, 65,

66, 67, 68, 69, 70, 78, 132, 134,

205, 214
pallidus

Svrrhophis modestus, 40

Svrrhophus, 4, 5, 6, 8, 14, 25, 37,
38, 39,40,41,43

Synhophus modestus, 4
palmatus

Colostethus, 4

Phyllobates, 4
Pahidicola

festae, 4

simonsii. 4
palustris, Enhydrus, 71
papuensis, Fordonia, 60, 178, 181
pardahs

Elcutherodactyhis, 4

Syrrhophus, 4
Pelophis. 62

petrophihis, Syrrhophus, 4, 22, 23, 25
Pflieger, William L. A distributional

study of Missouri fishes, 225 (see

index pp. 561-570)
Phyllobates, 4, 14, 16, 17, 18

chalceus, 4

Phyllobates — Concluded

cystigathoides, 4, 14, 16

hylaeformis, 4

palmatus, 4

ridens, 4

verruculatus, 4, 17, 18
Phvtolopsis, 60, 62, 127

punctata, 60, 62, 127
pipilans

Syrrhohus, 4, 5, 7, 8, 10

Svrrhophus pipilans, 4, 8, 14, 33,

34, 35
piscator, Enhydrus, 71
Platyrhiniens, 56
plumbea

Coluber, 82

Enhydris, 59, 60, 63, 64, 65, 66, 67,

68, 69, 72, 75, 82, 85, 203, 204,
205, 213

Homalopsis, 60, 62, 82

HNTOsirhina, 82
plumbeus, Eurostus, 82
polvlepis

Enhvdris, 60, 63. 64, 65. 66, 67, 68,

69, 70,72, 107. 110,204,214
Hvpsirhina. 60, 110

Potamophis, 60, 62. 86

lusingtonii, 60, 62, 86
prevostiana

Gerarda, 58. 59, 60, 95, 174. 204

Gerardia, 175
prevostianum. Campylodon. 174. 175
prevostianus. Coluber, 60, 174, 177
Pseudoferania. 60, 62

macleayi, 60. 62. 108
punctariolus, Eleutherodactylus, 4
punctata

Enhvdris, 59. 60. 63, 64, 65. 66, 67,
68, 69, 70, 74, 127, 129, 132,
205, 214

Erpeton, 59

Hypsirhina, 127

Phytolopsis, 60. 62, 127

Pvthonopsis. 127
Python, 146, 160

rhynchops. 60. 160
Pythonia, 135, 137

semizonata, 135, 137
Pythomorphus, 62
pythonissa. Coluber, 60, 86
Pythonopsis, 59, 62

borneensis. 59, 62, 127

punctata, 127

Raclitia, 59, 62, 75

indica, 59, 62, 75
Rhinopirus, 189. 190

tentaculatus, 190
rhodopis, Eleutherodactylus, 4, 12
richardsoni, Myron, 60, 110, 172, 173,

215

Index to Volume 20

577

richardsonii, Myron, 60, 172
lidens

Eleutherodactylus, 4

Phyllobates, 4
nihriniaciilatus, Syrrhophus, 4, 5, 6,

7, 9, 10, 14, 20, 21, 22, 41
nisselli, Cerberus, 60
r>nchops

Cerberus, 148, 157, 159, 160, 161,
199, 200, 203, 204

Cerberus r\nchops, 58, 59, 60, 149,
150, 151, 152, 153, 154, 155, 157,
160, 163, 164, 165, 166, 167, 168,
201, 207

Enhydris, 71

Honialopsis, 160

Hurria, 161

Hydrus, 60, 146, 147, 159, 160

Hydrus rvnchops, 160

Python, 60, 160

schneideriana, Hurria, 60, 160
semizonata

Homalopsis, 60, 137

Pytlionia, 135, 137
sieboldi

Enhvdris, 59, 60, 63, 64, 65, 66, 67,

68, 69, 70, 74, 81, 118, 119, 120,
203, 206, 207, 214

Ferania, 118, 123

Hj'psirhina, 118

Trigonurus, 118
sieboldii, Homalopsis, 60, 62, 118
simonsii, Paludicola, 4
smithi

Enhydris, 60, 63, 64, 65, 66, 67, 68,

69, 70, 73, 78, 86, 97, 98, 99,
213

Hypsirhina, 60, 98

Syrrhophus, 4, 7, 22, 23, 25
soctrangensis

Enhydris bocourti, 124

H>psirhina bocourti, 59, 123
Stegodon, 215
Strephon, 71, 146, 147
subalbidus. Coluber, 60, 136
subtaeniata

Enhydris enhydris, 90

Hypsirhina enhydris, 59, 92
Syrraphus guttulatus, 22
Syrrhaphus

guttulatus, 22

verrucipes, 27
Syrrhophis modestus pallidus, 40
Syrrhophus, 3, 4, 5, 6, 7, 8, 9, 10, 11,
12, 13, 14, 15, 16, 17, 18, 19, 20.
21, 22, 23, 24, 25, 26, 27, 28, 29,
30, 31, 32, 33, 34, 35, 36, 37, 38,
39, 40, 41, 42, 43

areolatus, 4

calcaratus, 4

Syrrhophus — Concluded
campi, 4, 15
caryophyllaceus, 4
coeruleus, 4
cystignathoides, 6, 10, 14, 15, 16,

17, 18, 41
cystignathoides campi, 4, 7, 12, 13,

15, 16, 17
cystignathoides cystignathoides, 4,

7, 11, 12, 13, 15, 16
dennsi, 4, 5, 8, 9, 11, 13, 32, 41
dennisi, 4, 27, 29
gaigeae, 3, 4, 22, 23, 25
guttilatus, 4, 5, 7, 8, 9, 10, 11, 14,

22, 23, 24, 25, 27, 41
ineptus, 4
interorbitalis, 4, 5, 8, 12, 14, 36, 37,

39
juninensis, 4
latodactylus, 4, 27, 28, 30, 31, 32,

33
leprus, 3, 4, 5, 7, 10, 11, 13, 18, 19,

20, 22, 27, 41
leprus cholorum, 4, IS
leprus leprus, 18
longipes, 4, 5, 6, 7, 8, 9, 11, 13, 30,

31, 32, 41
lutosus, 4
marnockii, 3, 4, 5, 6, 8, 9, 10, 12,

13, 14, 22, 23, 25, 26, 27
macrotympanum, 27
modestus, 4, 5, 6, 7, 8, 10, 12, 14,

37, 38, 39, 40, 41, 43
modestus modestus, 37
modestus pallidus, 4, 40
molinoi, 4

montium, 4

mystaceus, 4

nebulosus, 4, 34

nivocolimae, 4, 5, 6, 8, 10, 12, 14,

36, 37, 38, 39, 40, 41
obesus, 4
omiltemanus, 4
pallidus, 4, 5, 6, 8, 14, 25, 37, 38,

39, 40, 41, 43
pardalis, 4

petrophilus, 4, 22, 23, 25
pipilans, 4, 5, 7, 8, 10, 33, 34, 35,

42
pipilans nebulosus, 4, 8, 12, 14, 33,

34, 35, 41
pipilans pipilans, 4, 8, 14, 33, 34,

35
rubrimaculatus, 4, 5, 6, 7, 9, 10, 14,

20, 21, 22, 41
smithi, 4, 7, 22, 23, 25
terestistes, 4, 5, 8, 10, 14, 36, 37,

38, 40, 41

verrucipes, 4, 5, 7, 8, 9, 10, 11, 12,

13, 18, 20, 21, 25, 27, 41
verruculatus, 3, 18, 27, 35

578

University of Kansas Publs., Mus. Nat. Hist.

Syrrhopis, interorbitalis, 36
Synhopus, guttulatus, 22
Syrrophus marnocki, 25

Tachyplotus, 59, 62, 127

hedemanni, 59, 127
tentaculatum

Erpeton, 204

Herpeton, 190
tentaculatus

Erpeton, 59, 60, 81, 189, 190, 181

Rhinopirus, 190
teretistes, Synhophus, 4, 5, 8, 10, 14,

36, 37, 38, 40, 41
Thalerophis, 207
Tomodactylus, 3, 12, 13, 27, 42, 43

albolabris, 42

dilatus, 42

macrotympanum, 3, 4, 27

nitidus, 42, 43
Trigonuie, 57

Trigoniirus, 62, 118

sieboldi, 118
tribneata, Hypsirhina, 60, 86
Tropidonotus, 71

unicolor

Cerberus, 60, 160
Fordonia, 61, 178

variabilis, Fordonia, 61, 178
ventromaculata, Heurnia, 61, 110,

170, 215
verrucipes

Syrrhaphus, 27

Syrrhophus, 4, 5, 7, 8, 9, 10, 11, 12,
13, 18, 20, 21, 25, 27, 41
verniciilatus

Phyllobates, 4, 17, 18

Syrrhophus, 3, 18, 27, 35
violacea, Cantoria, 59, 61, 129, 182,

183, 184, 185, 207, 209, 215
Vipera buccata, 136

Date Due

m — Hfsm

3 2044 093 361 533

Internet Archive Audio

Featured

Top

Images

Featured

Top

Software

Featured

Top

Books

Featured

Top

Video

Featured

Top

Mobile Apps

Browser Extensions

Archive-It Subscription

Save Page Now

Full text of "University of Kansas publications, Museum of Natural History"

See other formats