Citation
The American cyprinid fishes of the sub-genus Luxilus (genus Notropis)

Material Information

Title:
The American cyprinid fishes of the sub-genus Luxilus (genus Notropis)
Series Title:
Bulletin of the Florida State Museum
Creator:
Gilbert, Carter Rowell, 1930-
Place of Publication:
Gainesville
Publisher:
University of Florida
Publication Date:
Language:
English
Physical Description:
96-194 p. : illus., maps, tables. ; 23 cm.

Subjects

Subjects / Keywords:
Freshwater fishes -- North America ( lcsh )
Notropis ( lcsh )
White River ( flgeo )
Fish ( jstor )
Fish scales ( jstor )
Stripes ( jstor )
Genre:
bibliography ( marcgt )
government publication (state, provincial, terriorial, dependent) ( marcgt )

Notes

Bibliography:
"References cited:" p. 185-194.
General Note:
Cover title.
Statement of Responsibility:
[by] Carter R. Gilbert.

Record Information

Source Institution:
University of Florida
Rights Management:
Copyright held by the Florida Museum of Natural History, University of Florida. All rights reserved. Text, images and other media are for nonprofit, educational, and personal use of students, scholars, and the public. Any commercial use or republication by printed or electronic media is strictly prohibited without written permission of the museum. For permission or additional information, please contact the current editor of the Bulletin at bulletin@flmnh.ufl.edu.
Resource Identifier:
AAA0820 ( NOTIS )
05067509 ( OCLC )
a 64007719 ( LCCN )

Related Items

Related Item:
PALMM Version

Downloads

This item has the following downloads:


Full Text



BULLETIN

OF THE

FLORIDA STATE MUSEUM

BIOLOGICAL SCIENCES


Volume 8


Number 2


THE AMERICAN CYPRINID FISHES OF THE
SUBGENUS LUXILUS (GENUS NOTROPIS)

Carter R. Gilbert


UNIVERSITY


OF FLORIDA


Gainesville

DIGITIZED964







Numbers of the BULLETIN OF THE FLORIDA STATE MUSEUM are pub-
lished at irregular intervals. Volumes contain about 300 pages and are not
necessarily completed in any one calendar year.

















OLIVER L. AUSTIN, JR., Editor
WALTER AUFFENBERG, Managing Editor




Consultant for this issue:
Donn Eric Rosen




















Communications concermng purchase or exchange of the publication and all
manuscripts should be addressed to the Managing Editor of the Bulletin, Florida
State Museum, Seagle Building, Gainesville, Florida.


Price for this issue $1.45


Published 2 June 1964








THE AMERICAN CYPRINID FISHES OF THE SUBGENUS
LUXILUS (GENUS NOTROPIS)


CARTER R. GILBERT 1


SYNOPSIS: The cyprinid fishes of the subgenus Luxilus (genus Notropis) comprise
three well-defined species groups of eight species and one subspecies. They
occupy clear, small to medium-sized streams over most of temperate North Amer-
ica east of the Rockies, being absent only from areas southwest of the Red River
drainage in Texas, the upper Missouri basin, the southern Piedmont plain, and
peninsular Florida. The present study defines the subgenus, reviews its nomen-
clatural history, explains its patterns of variation, describes the various forms and
their distributions, and discusses their probable zoogeographic and phylogenetic
histories.


TABLE OF CONTENTS


Introduction ----
Acknowledgements ---
Materials --
Methods of Counting and
Measuring
Hybridization --- ----
Zoogeography and Phylogeny -
Generic and Subgeneric
Nomenclatural History ..
Subgeneric Diagnosis and
Specific Relationships ----_--
Key to Species and Subspecies
of the Subgenus Luxilus -...
Species and Subspecies Accounts
Notropis coccogenis ..--...


Notropis zonistius -- 126
Notropis zonatus --- 129
Notropis pilsbryi ----- 133
Notropis cerasinus --- 136
Notropis cornutus --- 140
Notropis albeolus ____--151
Notropis chrysocephalus ------ 157
Notropis chrysocephalus
chrysocephalus -- 158
Notropis chrysocephalus:
chrysocephalus X isolepis 166
Notropis chrysocephalus
isolepis ...- 167
Tables 173
References Cited ...----------..- 185


1 This paper is a revision of a dissertation submitted to the University of Michi-
gan in 1960 in partial fulfillment of the Ph.D. degree. The work was carried
out while the author served as a teaching fellow in the Department of Zoology
and as a research assistant in the Division of Fishes of the Museum of Zoology.
Now at the University of Florida, he is Assistant Professor of Biological Sciences
and Assistant Curator of Natural Sciences in the Florida State Museum. Manu-
script received 5 July 1963.-ED.


Gilbert, Carter R. 1964. The American cyprinid fishes of the subgenus Luxilus
(Genus Notropis). Bull. Florida State Mus., vol. 8, no. 2, pp. 95-194.








96 BULLETIN FLORIDA STATE MUSEUM Vol. 8


INTRODUCTION
Of the many species of cyprinid fishes inhabiting eastern North
America, few are better known to the layman than the Common Shiner,
Notropis cornutus. This is due to its abundance, wide distribution,
large size, and striking coloration. Nevertheless the systematic re-
lationships of the forms referred to this and to closely related species
have long perplexed students of North American freshwater fishes.
The group that includes Notropis cornutus and its most closely
related forms (here referred to as the subgenus Luxilus) is in a rel-
atively early stage of evolution. Within the Luxilus complex may be
found all stages of speciation, and in at least two cases considerable
subjective judgment is needed to determine whether forms are in-
cipient or actual species. The Luxilus group thus furnishes an ex-
cellent opportunity for observing and understanding the basic proc-
esses of change which occur during evolutionary history.
The members of the subgenus Luxilus have had an unstable no-
menclatural history. Their recent evolution and consequent close
inter-relationships have made the arrangement of species difficult.
D. S. Jordan, who was mainly responsible for previous attempts to
classify these groups, moved species back and forth among genera
and subgenera, usually with little apparent reason.
Although morphological criteria have been the primary bases for
the resolution of various taxonomic problems encountered in the pres-
ent study, zoogeography, ecology, and clinal variation also have been
taken into account. This has resulted in the recognition of 8 species
and 1 subspecies instead of 6 species and 4 subspecies as in the cur-
rently accepted classification. Both arrangements are listed below.

Current classification Proposed classification
coccogenis group
Notropis coccogenis (Cope) Notropis coccogenis (Cope)
Notropis zonistius (Jordan) Notropis zonistius (Jordan)

zonatus group
Notropis zonatus zonatus (Agassiz) Notropis zonatus (Agassiz)
Notropis zonatus pilsbryi Fowler Notropis pilsbryi Fowler

cornutus group
Notropis cerasinus (Cope) Notropis cerasinus (Cope)
Notropis cornutus cornutus (Mitchill) Notropis cornutus (Mitchill)







GILBERT: FISHES OF THE SUBGENUS LUXILUS


Current classification Proposed classification
cornutus group (continued)
Notropis cornutus frontalis (Agassiz)
Notropis albeolus Jordan Notropis albeolus Jordan
Notropis cornutus chrysocephalus Notropis chrysocephalus chrysoceph-
(Rafinesque) alus (Rafinesque)
Notropis cornutus isolepis Notropis chrysocephalus isolepis
Hubbs and Brown Hubbs and Brown

The subgenus Luxilus is one of the widest ranging of the minor
cyprinid groups in North America. Its species occur in fresh waters
from the Gulf Coast north to 520 N. latitude in Manitoba and from
Nova Scotia to the eastern slope of the Rocky Mountains. The only
major area east of the Mississippi River from which it is absent is the
southeastern Atlantic coastal plain from central North Carolina south-
ward through peninsular Florida. To the southwest it does not oc-
cur beyond the Red River drainage.
All species ordinarily inhabit rather clear, small or medium-sized
streams with gravel to rocky bottoms and moderate to swift grad-
ients. Although some populations may be found living in extremely
turbid waters, this is abnormal and usually temporary.
Hybridization between species of the Luxilus complex is frequent
and occurs in many combinations both within the subgenus and with
species of other subgenera and genera. These are reviewed, and
the possible causes and effects of hybridization are discussed.

ACKNOWLEDGMENTS
This study was made possible by the aid and cooperation of many
people. I am especially grateful to Dr. Reeve M. Bailey for his ad-
vice, criticism, and professional guidance throughout the course of
the study; to my wife, Nancy, for her constant encouragement and
invaluable help in the preparation and completion of this thesis; to
Mrs. Elizabeth M. Anthony who prepared the drawings as well as
the letters and margins for the distribution maps; to Dr. Ernest A.
Lachner for much helpful advice regarding the preparation of this
manuscript; and to Dr. Clarence L. Smith for field assistance and for
most of the photographs included in this paper. I am indebted to
the following persons for the use of specimens and equipment under
their care: Dr. Edward C. Raney, Cornell University; Drs. Leonard
P. Schultz, Ernest A. Lachner and William Ralph Taylor, United States
National Museum; Dr. James E. Bohlke, Academy of Natural Sci-






BULLETIN FLORIDA STATE MUSEUM


ences of Philadelphia; Dr. Milton B. Trautman, Ohio State Archeo-
logical Museum; Dr. Royal D. Suttkus, Tulane University; Dr. Philip
W. Smith, Illinois Natural History Survey; and Dr. Shelby D. Gerking,
Indiana University. Sincere thanks are extended to Mrs. H. Williams
Diamond, Jr., who helped type the manuscript; to Dr. Edward J.
Crossman, Royal Ontario Museum of Zoology, for supplying certain
locality records for Canada; to Dr. George A. Moore, Oklahoma State
University, for several Oklahoma records; to Dr. Clark Hubbs, Uni-
versity of Texas, for all Texas records; to Dr. Frank B. Cross, Univer-
sity of Kansas, for several Kansas records; and to Dr. Ralph W. Yer-
ger, Florida State University, for several Florida records. Apprecia-
tion is extended to Dr. Robert R. Miller, University of Michigan,
for helpful advice and for the use of an unpublished list of hybrid
fish combinations; to Dr. Karl F. Lagler, University of Michigan, for
the use of equipment under his care; to Dr. Louella Cable, for the
use of facilities at the U. S. government fish hatchery, Northville,
Michigan; to Dr. Robert D. Ross, Virginia Polytechnic Institute, for
supplying color notes for Notropis albeolus; and to Dr. William
Ralph Taylor, for advice regarding the study. Finally, I am grateful
to my fellow graduate students, Thomas L. Poulson, Teruya Uyeno,
R. Jack Schultz and Howard F. Huddle for their help in the field
and laboratory.
MATERIALS
The collections examined were selected to include samples from
all parts of the ranges of the various forms, with special emphasis
being placed on series from critical areas. Specimens have been
examined or recorded from the following museum collections: Acad-
emy of Natural Sciences of Philadelphia (ANSP); Chicago Natural
History Museum (CNHM); Cornell University (CU); Illinois Natural
History Survey (INHS); Indiana University (IU); Kansas University
(KU); Museum of Comparative Zoology, Harvard University (MCZ);
Nova Scotia Provincial Museum (NSPM); Ohio State University (OSU);
Royal Ontario Museum (ROM); Stanford University (SU); Texas
Natural History Collection, University of Texas (TNHC); Tulane
University (TU); University of Florida (UF); University of Michigan,
Museum of Zoology (UMMZ); United States National Museum
(USNM).
Locality records of specimens examined may be found in the
original dissertation.


Vol. 8






GILBERT: FISHES OF THE SUBGENUS LUXILUS


METHODS OF COUNTING AND MEASURING

To reduce sample bias usually no more than 10 specimens were
counted from a single series, except in lots from certain critical or
poorly collected areas, where larger numbers were used. As certain
relative body proportions change with increase in size, body measure-
ments were made only on specimens falling within certain size ranges
in each species. For the larger forms (N. cornutus, N. albeolus, N.
chrysocephalus, and N. coccogenis) specimens measuring approxi-
mately 60 to 80 mm in standard length were used; for the smaller
species (N. zonistius, N. cerasinus, N. zonatus, and N. pilsbryi) the
limits were 55 and 75 mm.
Counts and measurements were taken as outlined by Hubbs and
Lagler (1947: 8-15) except that the mouth gape was determined by
the transverse distance between the posterior angles of the mouth,
instead of the greatest transverse distance across it. To eliminate
difficulties in counting fine-scaled species, a new method was de-
vised for taking scale counts on the anterior dorsal part of the body.
The usual predorsal scale count on the dorsal midline directly in front
of the dorsal fin is entirely satisfactory for coarse-scaled species, but
overlaps and irregularities in fine-scaled forms make accurate deter-
minations difficult. The scale rows along the side of the body are
more regular than the predorsal scales, and provide a more accurate
and effective assessment of differences in scale size among popula-
tions.
Anterior dorsal-lateral scale counts were made along the longi-
tudinal row of scales lying approximately three-fourths of the distance
from the lateral line to the mid-dorsal line. This varied from the
third to the sixth row above the lateral line, depending on the size
of the scales. The scales were counted from a point directly under
the origin of the dorsal fin forward to their termination. The sum
of this count and the body circumferential scale count is of great
value in interpreting differences between the various species, par-
ticularly those in the N. cornutus species group.
Scale counts for both the circumferential and the caudal peduncle
series are usually expressed as total numbers, but may also be writ-
ten to indicate the relative disposition of scales above and below
the lateral line. Thus a circumferential count written as 14-2-10
signifies 14 scales between and above and 10 scales between and
below the lateral lines on each side of the body.


1964







BULLETIN FLORIDA STATE MUSEUM


Proportional measurements are expressed in thousandths of the
standard length (shortened hereafter to SL). These measurements
were tabulated initially in the form of frequency distributions by ma-
jor drainage areas for each member of the subgenus except N.
cerasinus. When inspection revealed no consistent geographical var-
iation in morphology except in N. albeolus, the proportional measure-
ments for all the others were combined in the final tabulations.
Certain meristic counts for N. cornutus, N. albeolus, N. c. chryso-
cephalus, and N. c. isolepis were separated in the tables by major
drainage areas. This was done because the wide ranges of these
forms increases the possibilities for clinal or genetic variation. The
better to show the overall differences between N. cornutus living
east and west of the Appalachians, all populations in each of these
two areas were combined, with three exceptions: Those in the upper
Susquehanna and the St. Lawrence river systems are on or near the
drainage divide between the Atlantic Ocean and the Mississippi;
specimens from the streams in Maine and the Maritime provinces of
Canada show rather unusual meristic characters. Counts from all
areas were combined in the overall comparison of the Luxilus group.
Vertebral counts were determined from X-ray photographs. The
procedures follow the methods discussed by Bailey and Gosline
(1955). The Weberian apparatus, regarded as containing four fused
vertebra, and the hypural plate were included in the counts. All
photographs were made with type M film and a "hard ray" machine
(Miller, 1957).
The angle of the mouth to the top of the head was determined
with the aid of a special ocular as described by Smith (1956: 250).
With the pointer set at a 00 reading, the fish is placed on the stage
so that one crosshair (arbitrarily called the horizontal hair) in the
ocular lies directly over the length of the lateral opening of the mouth,
with the intersection of the crosshairs slightly anterior to the tip of
the snout. The ocular is then rotated clockwise until the vertical
crosshair touches the head at the posterior edge of the occiput. The
angle is then read in degrees to the nearest half degree.
Inspection revealed that usually about 90 percent of the counts
showed relatively little deviation from the mean of a normal distri-
bution, while the extreme values dropped away rather sharply. As
it is manifestly desirable to show this condition in a species account,
the method shown in the following hypothetical examples was adopted
to express it: A frequency distribution expressed as (22) 25 to 29 (38)
indicates that 90 percent of the counts fall between 25 and 29, with


Vol. 8






GILBERT: FISHES OF THE SUBGENUS LUXILUS


the extremes 22 and 33; a distribution expressed as (8) 9 or 10,
usually 9, shows that less than 10 percent of the counts number 8
and, while a count of 10 is not particularly rare (more than 10 percent)
9 is a much more common value.
To determine possible sexual variation, approximately equal num-
bers of males and females were measured for the various morpho-
metric characters. In most cases the means and range of variation
differed so little and the overlap was so great that measurements were
lumped in the final tabulations. Tables and lists of locality records
in the original dissertation not regarded as essential have been omit-
ted. Only new names or new combinations of names have been re-
tained in the synonymy.

HYBRIDIZATION
Numerous hybrid combinations involving members of the Luxilus
complex are known. Three of these, N. c. chrysocephalus X N. cor-
nutus, N. c. chrysocephalus X N. rubellus, and N. cornutus X N.
rubellus, are the most common cyprinid hybrids in eastern North
America. The cornutus X rubellus hybrid was described by Jordan
(1889a: 354) as a new species, Notropis macdonaldi, and the chryso-
cephalus X rubellus cross was erroneously recorded by Forbes (1909:
386) and by Forbes and Richardson (1920: 149) as Notropis pilsbryi.
A number of papers have appeared dealing entirely or in part
with various aspects of hybridization in certain species of Luxilus.
Among the more comprehensive and important of these are Raney's
(1940b) account of the comparative spawning behavior of N. cornutus
and N. rubellus in New York, and Miller's (1962; 1963) papers deal-
ing with the comparative morphology, sexual development, and
hermaphroditism of N. cornutus X N. rubellus. In my own (1961b)
treatment of the relationships of N. cornutus and N. c. chrysocephalus
I list the arguments for and against specific differentiation of cornutus
and chrysocephalus, and favor their recognition as species, although
the two forms often act as subspecies. This evidence is as follows:

1. Intergrading populations of subspecies normally show a per-
fect blending of morphological characters. Although this is ap-
proached in some associations containing both cornutus and chryso-
cephalus, in other populations the two forms remain morphologically
distinct, with few apparent intermediates. In addition many popu-
lations within the zone of "intergradation" contain only one form.


1964







BULLETIN FLORIDA STATE MUSEUM


This mosaic pattern is not concordant with the normal expectation
for subspecies.

2. Competition is, in effect, nonexistent between subspecies, and
where one has moved into a region formerly occupied solely by an-
other, intergrading populations occur. The presence of isolated
relict populations of cornutus surrounded by populations of chryso-
cephalus far to the south of the principal range, together with the
observed gradual replacement of cornutus by chrysocephalus in parts
of Ohio, indicate that competition exists where these fishes occur
together.

3. Chrysocephalus apparently has disappeared from several mixed
populations in southern Michigan following construction of dams
near the mouths of the creeks where it formerly occurred. If cor-
nutus and chrysocephalus were only subspecifically distinct, the phe-
notypic character of the populations would be expected to remain
more or less constant.

4. The experimentally-verified differences in temperature toler-
ance between cornutus and chrysocephalus in the northern limits of
their respective ranges are greater than one might normally expect
in subspecies. The apparent preference of cornutus for headwaters
and of chrysocephalus for the lower reaches of streams furnishes fur-
ther evidence of this difference.

5. The distributional patterns of these forms suggest a long sepa-
ration of cornutus and chrysocephalus, ample to permit evolution to
the species level, and possibly dating back to late Pliocene.

Raney's paper includes a discussion of the morphology, fertility,
hybrid vigor, and sex ratio of the cornutus X rubellus hybrids. He
decided, as did Hubbs and Brown (1929: 36-37), that hybridization
between these species is accidental, and usually results from simul-
taneous spawning over the nests of other species, such as Hybopsis
micropogon, Semotilus atromaculatus, and Exoglossum maxillingua.
In the absence of such nests rubellus and cornutus spawn in different
parts of the same riffle, rubellus in mid-water near the head, and
cornutus on the bottom toward the base. Milt from rubellus males
is washed downstream by the current, where it unites with eggs of
cornutus. The latter observations are similar to those of Moore and
Paden (1950: 92) for Notropis pilsbryi and Dionda nubila. Hybrids
of this combination apparently arise as a result of the simultaneous


Vol. 8







GILBERT: FISHES OF THE SUBGENUS LUXILUS


spawning of Notropis pilsbryi in riffles and Dionda nubila in pools at
the bases.
The following is a list of the known hybrids involving members
of the subgenus Luxilus. Some have been reported in the literature
and are so indicated; those which have not yet appeared in the pub-
lished record are indicated by catalog numbers only. I have made
no attempt to verify the accuracy of all these identifications, nor have
I included all literature references. Current nomenclature is used
in all cases.

Notropis albeolus X Notropis rubellus (UMMZ 174754)
Notropis cerasinus X Chrosomus oreas (Raney and Lachner, 1946: 226)
Notropis chrysocephalus chrysocephalus X Campostoma anomalum (Trautman,
1957: 114)
Notropis chrysocephalus chrysocephalus X Hybognathus nuchalis (Luce, 1933:
116; O'Donnell, 1935: 482)
Notropis chrysocephalus chrysocephalus X Hybopsis biguttata (Luce, 1933: 115;
O'Donnell, 1935: 482)
Notropis chrysocephalus chrysocephalus X Hybopsis micropogon (Trautman,
1957: 114)
Notropis chrysocephalus chrysocephalus X Notropis chrosomus (UMMZ 174833,
UMMZ 175614)
Notropis chrysocephalus chrysocephalus X Notropis cornutus (Gilbert, 1961b:
181-192)
Notropis chrysocephalus chrysocephalus X Notropis leuciodus (UMMZ 175194)
Notropis chrysocephalus chrysocephalus X Notropis rubellus (Forbes, 1909: 386
[as N. pilsbryi]; Forbes and Richardson, 1920: 149 [as N. pilsbryi]; Hubbs,
1955: 10; Hubbs and Brown, 1929: 35-36; Hubbs and Moore, 1940: 96
[identification of N. pilsbryi]; Murray, 1938:84; O'Donnell, 1935:482;
Trautman, 1948: 171 [in part]; Trautman, 1957: 114-115 [in part])
Notropis cornutus X Campostoma anomalum (Trautman, 1957: 114)
Notropis cornutus X Chrosomus eos (UMMZ 78827)
Notropis cornutus X Chrosomus oreas (Raney, 1950: 167)
Notropis cornutus X Clinostomus elongatus (Greeley, 1938: 51; Greene, 1935:
88-89; Koster, 1939: 207; Raney, 1940a: 7)
Notropis cornutus X Clinostomus funduloides (Raney, 1950: 167)
Notropis cornutus X Hybopsis biguttata (Taylor, 1954: 43)
Notropis cornutus X Hybopsis micropogon (Greeley, 1938: 51; Raney, 1940a:
7; Raney, 1940c: 135; Trautman, 1957: 114)
Notropis cornutus X Notropis photogenis (Greeley, 1938: 51)
Notropis cornutus X Notropis rubellus (Goldsborough and Clark, 1908: 35 [as
N. macdonaldi]; Greeley, 1938: 51; Hubbs, 1955: 10; Hubbs and Brown,
1929: 35-36; Hubbs and Moore, 1940: 96 [identification of N. macdonaldi];
Jordan, 1889a: 354 [original description of N. macdonaldi]; Lachner, 1952:
436; Raney, 1940a: 7; Raney, 1940b: 361-367; Raney, 1940c: 135; Raney,


1964






BULLETIN FLORIDA STATE MUSEUM


1950: 165, 167; Taylor, 1954: 43; Trautman, 1948: 171 [in part]; Trautman,
1957: 114-115 [in part])
Notropis cornutus X Semotilus atromaculatus (Bailey and Oliver, 1939: 172;
Greeley, 1938: 51-52; Koster, 1939:207; Raney, 1940a: 7; Raney, 1940c:
135; Simon, 1946: 91; Taylor, 1954: 42)
Notropis cornutus X Semotilus corporalis (Bailey, 1938: 150-151; Greeley, 1938:
52; Raney, 1940a: 7)
Notropis pilsbryi X Dionda nubila (Moore and Paden, 1950: 92)
Notropis pilsbryi X Notropis rubellus (Cross, 1954: 313; Moore and Paden,
1950: 92)

ZOOGEOGRAPHY AND PHYLOGENY

The zo6geographic and phylogenetic history of a group of animals,
often the subject of speculation, is usually reconstructed only with
difficulty. Because the genus Notropis has evolved so recently, most
of the forms linking the included natural groups are still extant and
allow a relatively clear interpretation of the phylogenetic relation-
ships. On the other hand, the high degree of speciation in the genus
makes the zo6geographical relationships extremely complex.
The subgenus Luxilus probably evolved in the Mississippi basin,
perhaps during the early to middle Pliocene. This hypothesis is
based first on the present geographical distribution of the Luxilus
complex and, second, on the assumption of a Miocene invasion of
North America by the family Cyprinidae (Romer, 1945: 583).
The primary division within the subgenus most likely involved
the precursor of Notropis coccogenis and Notropis zonistius. Judging
from its present distribution, the coccogenis group probably evolved
in the Tennessee River system. The marked differentiation of cocco-
genis and zonistius from the other species of Luxilus indicates that
isolation from stocks in the Mississippi Valley lasted for a consider-
able time. The reasons for isolation, however, are unknown at pres-
ent. Later, presumably still during the Pliocene, a segment of this
stock reached the Chattahoochee River basin, there evolving into
N. zonistius, while the stock remaining in the Tennessee system gave
rise to N. coccogenis. Presumably this transfer was effected through
a localized stream capture rather than by direct connection of the
Chattahoochee with the Tennessee. No evidence indicates such a
connection, and these river systems show little close faunal similarity.
Also during the Pliocene another segment of the Mississippi Val-
ley stock presumably became isolated in the Ozark region. This
eventually gave rise to the zonatus group, which includes Notropis


Vol. 8






GILBERT: FISHES OF THE SUBGENUS LUXILUS


zonatus and Notropis pilsbryi. The cause of this isolation also is un-
known, although it probably is related to the fact that the Ozarks
contain clear, cool, rather swift-flowing streams. Unlike some other
Ozarkian endemics, Notropis zonatus and Notropis pilsbryi have no
immediate relatives living in upland areas east of the Mississippi
River. Consequently these species are believed to be autochthonous
for the Ozark region.
The distributions of N. zonatus and N. pilsbryi suggest a long sep-
aration between the Missouri and White river systems that has pre-
vented gene interchange. This interpretation is supported by the
parallel distribution patterns of other species pairs, such as Etheo-
stoma euzonum-E. tetrazonum. The presence of N. zonatus in the
Black River system strongly suggests that this river at one time was
tributary to the Missouri, and that a reversal of flow has occurred
rather recently, long after zonatus and pilsbryi separated. Records
of zonatus from the headwaters of the St. Francis and Little rivers
of southeastern Missouri are best explained on the basis of stream
captures, although the possibility exists that these streams also once
flowed to the north. Probably the most perplexing aspect of the
zoogeography of these species has to do with the presence of pilsbryi
in the Arkansas and Red river drainages. The distribution of sev-
eral species (Notropis spilopterus and Lepomis megalotis) indicates a
definite faunal relationship between the Illinois and Neosho rivers
(tributaries of the Arkansas) and the Ozarkian tributaries of the Mis-
souri River. The absence of Notropis zonatus from the Illinois and
Neosho is an apparent contradiction, inexplicable save for the possi-
bility that if zonatus did invade these river systems it could not com-
pete with the well-established population of pilsbryi. Hubbs and
Moore (1940: 94) have suggested the isolated populations of pilsbryi
in the Red River system may have been introduced. It is equally
possible that they represent relict populations, especially as they have
been recorded from three separate localities in this drainage.
Prior to the Pleistocene epoch the drainage pattern of the Ohio
Valley differed considerably from that of today. The Ohio River
extended little east of Cincinnati, and what is now the upper Ohio
flowed northward, probably to a river in the area of the present Lake
Erie basin. Larger than either of these streams was the Teays River,
a major prolongation of the Kanawha River. The Teays flowed north-
westward across Ohio, Indiana, and apparently Illinois, and thence
southward to the present Mississippi Valley. It occupied a wide, deep
trench comparable in size with that of the lower Ohio River today
(Flint, 1947: 166).







BULLETIN FLORIDA STATE MUSEUM


A number of species that are completely or largely restricted to
the Roanoke River system, which drains into the Atlantic Ocean, have
affinities not with other Atlantic coastal forms, but with species in
the Ohio Valley. The headwaters of the Kanawha River are narrowly
separated from those of the Roanoke over a fairly extensive area in
western Virginia. Much of the present Roanoke fauna, including
Notropis cerasinus and, possibly, N. albeolus, almost certainly reached
this drainage by headwater transfer from the old Teays system dur-
ing preglacial time (Wright, 1934: 65), though whether the invasions
by these species occurred simultaneously cannot be determined.
Evolution of the common N. cornutus-N. chrysocephalus stock
probably took place in the Mississippi Valley. This hypothesis is sup-
ported by the distribution pattern of these two species, which cen-
ters in this region. Some of the common cornutus-chrysocephalus
stock is believed to have become separated from the main popula-
tion in late Pliocene in one of the following areas: the Atlantic coast,
the upper Mississippi Valley, the Great Plains, or the Great Lakes.
The relative distributions of the two species suggest this segment
evolved into N. cornutus, while that part of the population to the
south became N. chrysocephalus.
There is little evidence that N. cornutus evolved in either the Mis-
sissippi Valley or on the Atlantic coast, except for its presence there
today. Geological evidence (Flint, 1947: 163-167; and 1957: 168-170)
indicates that before the Pleistocene the middle and upper parts of
the Missouri did not flow into the Mississippi River as they do today,
but instead "flowed north into the Souris-Assiniboine system, perhaps
discharging via the Lake Winnipeg depression, the Nelson, and the
Hudson Bay region. The advent of the ice sheet flowing from the
northeast blocked all this drainage and detoured the Missouri along
the ice margin." Contact was then established with the Kansas River
below Kansas City, thus effecting a new outlet for the upper Missouri
River. The upper Missouri therefore provided an isolated region
where evolution of Notropis cornutus could have occurred, but its
apparent absence from the upper Missouri drainage today argues
strongly against this area as a place of origin.
Radforth (1944: 10-11) describes and pictures a hypothetical course
for the Laurentian River, which drained the present Great Lakes re-
gion (with the possible exception of Lake Superior) in pre-Pleistocene
time. Possibly a segment of the cornutus-chrysocephalus ancestral
stock entered this river system during late Pliocene and there evolved
into N. cornutus. This Laurentian center of origin for N. cornutus


Vol. 8







GILBERT: FISHES OF THE SUBGENUS LUXILUS


appears to be the best of the several possibilities. In addition to other
factors cited previously, spread to such distant areas as the Atlantic
coast and the Rocky Mountains could have been accomplished most
readily from a geographically intermediate area.
Following its initial invasion of the Atlantic coast the species
spread southward, possibly to the Roanoke River. The population
may then have become isolated from the remaining cornutus stock
farther north, resulting in the eventual evolution of a new form,
Notropis albeolus. The presence of other closely related species and
subspecies pairs in the James and Roanoke systems, such as Etheo-
stoma longimanum-Etheostoma podostemone, Moxostoma rhothoecum-
Moxostoma hamiltoni, and Notropis procne procne-Notropis procne
longiceps, tends to confirm this idea. It is equally plausible, as in-
dicated earlier, that albeolus reached the Atlantic coast via the Teays
River. This is substantiated by the similarity in distribution pattern
of this form with other species, such as Notropis ardens and N. ma-
tutinus, which are restricted to the Roanoke and neighboring systems
on the east coast, but whose affinities are with the Ohio Valley fauna.
The differentiation of albeolus into two rather distinct forms, one
occurring in the Roanoke and New drainages, the other in the Neuse,
Tar, and Cape Fear systems, suggests that albeolus may even have
evolved in some way from a union of stocks coming both from the
Atlantic coast and the Ohio Valley.
The first glacial advance completely inundated the Laurentian
River and thus split the range of cornutus into two major segments,
one along the Atlantic coast, the other west of the glacier. Whether
cornutus reinvaded the glaciated area during all the interglacial pe-
riods, which is a possibility, cannot be determined. Some remixing
of segments of the Atlantic coastal and western populations of cor-
nutus probably did occur during the Pleistocene, for no taxonomic
separation can be made between the eastern and western popula-
tions of cornutus at the subspecific level. If these populations did
not come into contact throughout the entire Pleistocene, the long pe-
riod of isolation should have produced more pronounced morpho-
logical differences. Furthermore cornutus is so widespread and adapt-
able that it would be among the most likely species to expand into
newly available territory.
Dispersal of N. chrysocephalus to various parts of the Mississippi
Valley evidently occurred throughout the Pleistocene. The Missis-
sippi River was then slightly cooler, less turbid, and less sluggish,
and probably did not pose the barrier to small-stream fishes that it







BULLETIN FLORIDA STATE MUSEUM


does today. Certain parts of the Mississippi could have once acted
as "filter bridges" for their movement.
Lowered ocean levels during the glacial periods permitted many
freshwater streams to join before reaching salt water, for example
those draining into the Gulf of Mexico from the Alabama system west
to the Mississippi. During one of the glacial advances some chryso-
cephalus stock in the Mississippi Valley apparently moved through
the lower reaches of the Mississippi into some of the Gulf coastal
streams. When, with the retreat of the glacier, the sea level rose
and restored the Mississippi to its former proportions, the chryso-
cephalus stock isolated in the coastal streams evolved into a new form,
isolepis, here regarded as a well-defined subspecies of chrysocephalus.
One can only speculate when the original invasion of the coastal
streams by chrysocephalus occurred, but the estimated time neces-
sary for a new form of this complex to evolve suggests it was probably
during one of the first two glacial periods. Much later, after the
morphological separation of chrysocephalus and isolepis was com-
plete, the two forms came together again when the Alabama-Coosa
system captured certain tributaries of the Tennessee. Van der Schalie
(1945: 357-358) discusses the geological and malacological evidence
showing the relationships of these two systems.
Notropis cornutus either survived in the upper Ohio Valley
throughout the Pleistocene or reached there during one of the inter-
glacial periods. Absence of this species from the upper Kanawha
River system suggests a fairly recent invasion, either from head\\ater
capture or through overflow streams from ponded lakes formed by
the advancing Wisconsin glacier. The further advance of the Wis-
consin ice sheet (Flint, 1947: 283) isolated the Ohio Valley popula-
tions of cornutus from both the Atlantic coast and the western popu-
lations.
As the Wisconsin ice sheet retreated, meltwater lakes formed at
the glacial margin. Because the glacier prevented northward flow,
the rising lakes began to cut outlets to the south. One of the largest,
Lake Agassiz, covered an extensive area and drained into the Mis-
sissippi Valley by means of the Warren River outlet (Radforth, 1944:
13), near the present southern extremity of the Red River of the
North. As the glacier retreated farther, a channel finally opened
to the northeast, the origin of the Nelson River. This stream was
situated at a lower elevation than the older Warren River, and thus
formed a new outlet for Lake Agassiz, which consequently decreased


Vol. 8






GILBERT: FISHES OF THE SUBGENUS LUXILUS


rapidly in area and was superseded by Lake of the Woods, Lake Win-
nipeg, and many lesser lakes.
Soon after formation of Lake Agassiz, there appeared farther to
the east glacial lakes Duluth, Chicago, and Maumee, the precursors,
respectively, of lakes Superior, Michigan, and Erie. These bodies,
also plugged at their northern extremities, drained into the Missis-
sippi Valley by way of the St. Croix outlet (Lake Duluth), the Fox and
Chicago outlets (Lake Chicago) and the Fort Wayne outlet (Lake
Maumee). With further glacial retreat changes continued until the
present-day Great Lakes were eventually formed.
N. cornutus probably moved northward via the Warren River
outlet into Lake Agassiz, and reinvaded the newly-formed Great
Lakes from the south through glacial channels such as the St. Croix
and Fox rivers (Greene, 1935: 15) and from the north via glacial Lake
Agassiz through the Ft. William outlet in western Ontario (Underhill,
1957: 31). Notropis cornutus apparently was able to follow the re-
treating glacier more closely than many other species and, as a result,
was again able to invade the east coast, this time via the Horseheads
outlet (in the Lake Ontario drainage) into the Upper Susquehanna
system (Bailey, 1945: 125-126). There it again mixed with the popu-
lation of cornutus which had remained on the Atlantic coast presum-
ably throughout the Pleistocene.
At this time ocean levels still were much lower than at present.
The shallow bottom of Chesapeake Bay was above water, and all
streams from the Susquehanna south to and including the James were
interconnected. Although influence of the cornutus stock that crossed
onto the Atlantic coast may have extended into the Chesapeake Bay
area, the bulk of the western gene pool probably remained near the
point of original entry. Proof that the preceding sequence of events
occurred depends in large part on the usual absence of pigmentation
on the chin and gular region in cornutus from the upper Susquehanna.
This character, while not consistent enough to be taxonomically name-
worthy, nevertheless is of value in the determination of zobgeograph-
ical relationships among populations. Pigment usually is lacking
west of the Alleghenies, but not in coastal streams.
N. cornutus may possibly have utilized a temporary outlet between
the forerunner of Lake Ontario and the Mohawk River. Apparently
some species did enter the Mohawk in this way, but the history is
blurred by the construction of the Mohawk-Erie canal across the
drainage divide in the early 1800's. Notropis cornutus seems to have
moved a considerable distance eastward from its western glacial refu-


1964






BULLETIN FLORIDA STATE MUSEUM


gium, and this stock may have given rise, at least in part, to the present
population in the Canadian maritime provinces.
With the openings of the Chicago and later the Maumee outlet
into glacial lakes Chicago and Maumee, N. chrysocephalus also
reached the Great Lakes drainage, there to live sympatrically with
Notropis cornutus, possibly for the first time since the stocks separated
in the Pliocene. Apparently chrysocephalus did not enter the Lake
Ontario drainage till well after cornutus. By this time the Horse-
heads connection was broken, which prevented chrysocephalus reach-
ing the Atlantic coast via the upper Susquehanna.
The general warming of streams in North America since the re-
treat of the Wisconsin ice sheet, and particularly with man's activities
during the past century or so, has wrought significant changes in the
relative distributions of N. cornutus and N. chrysocephalus which are
still continuing. In the Ohio Valley at the height of the Wisconsin
advance, cornutus was apparently the dominant, if not the sole repre-
sentative of the Luxilus complex above roughly the mouth of the Wa-
bash River. As stream conditions changed chrysocephalus invaded
this area, and the resulting competition virtually eliminated cornutus
from the lower Ohio Valley, though isolated relict populations re-
main in the lower Kanawha River system in southern West Virginia
and in the White River system in southern Indiana. The gradual re-
placement of cornutus by chrysocephalus in eastern Ohio during the
past 35 years has been well documented by Trautman (1957: 357).

GENERIC AND SUBGENERIC NOMENCLATURAL HISTORY

Luxilus Rafinesque, 1820a: 47-48 (original description of genus and subgenus;
type species, Luxilus chrysocephalus Rafinesque, by subsequent designation
of Jordan and Gilbert, 1877). Girard, 1856:202-203 (name mistakenly
allied with Notemigonus). Jordan, 1876a: 94 (Luxilus chrysocephalus syn-
onymized with Luxilus cornutus). Jordan, 1876b: 286-287 (Hypsilepis syn-
onymized with Luxilus). Jordan and Copeland, 1876: 134, 153 (type species
said to be Luxilus cornutus (Mitchill), but that species was not among those
included in Luxilus by Rafinesque; Hypsilepis a synonym). Jordan, 1877: 12,
28-31 (Luxilus chrysocephalus a synonym of Luxilus cornutus; review of Rafin-
esque's species; type species, Luxilus cornutus). Jordan and Gilbert, 1877: 86
(type species, Luxilus chrysocephalus Rafinesque). Jordan, 1882: 852-854
(Luxilus divided into three subgenera: Luxilus, Coccotis, and Alburnops).
Jordan, 1885: 814 (subgenus of Notropis). Jordan, 1929: 82 (Luxilus restored
to generic status).
Hypsolepis Agassiz, 1854:359 (original description of genus Hypsolepis; type
species, by original designation, Leuciscus cornutus; name attributed to


Vol. 8







GILBERT: FISHES OF THE SUBGENUS LUXILUS


Baird). Cope, 1864: 279 (spelling emended to Hypsilepis). Jordan, 1876b:
283 (Hypsilepis synonymized with Luxilus). Jordan and Copeland, 1876: 153
(Hypsilepis synonymized with Luxilus).
Plargyrus Rafinesque, 1820a: 50-51 (substitute name for Rutilus if included spe-
cies are not in that genus; type species, by absolute tautonymy, Rutilus plar-
gyrus). Kirtland, 1845: 26 (refers to species Leuciscus plargyrus). Storer,
1846: 158 (refers to species Leuciscus plargyrus). Girard, 1856: 196 (Plar-
gyrus accorded generic status). Jordan, 1876a: 94 (Rutilus plargyrus Rafines-
que wrongly synonymized with Luxilus kentuckiensis). Jordan, 1876b: 281
(incorrectly allied with part of present subgenus Cyprinella). Jordan, 1877:
28-29 (Plargyrus Girard [not of Rafinesque] placed in synonymy of Luxilus;
review of Rafinesque's species).
Coccotis Jordan, 1882: 852 (proposed as a subgenus of Luxilus; type species,
by original designation, Luxilus coccogenis). Jordan, Evermann and Clark,
1930: 127 (generic status).

Coccogenia Cockerell and Callaway, 1909: 190-191 (proposed as a subgenus of
Notropis on basis of scale characters; type species by original designation,
Notropis coccogenis).

The genus Notropis Rafinesque (1818) is a large and diverse group,
containing numerous closely related, morphologically similar forms.
At various times in the past its principal components have been re-
garded as distinct, though poorly-defined genera. Gilbert (1884b)
was the first to propose that a number of these be combined under
the name Notropis. Jordan (1885b) relegated these groups to sub-
generic status within the broad genus Notropis, where they have
remained ever since despite sporadic efforts (Jordan, 1929; Jordan,
Evermann and Clark, 1930) to separate them.
Although some of the chief subgroups of Notropis are difficult to
separate and the relationships of some are not yet clear, several
clusters of species share morphological characters and obviously con-
stitute natural units. The subgenus Luxilus Rafinesque (1820a) is one
such group. The species of Luxilus are characterized by common
diagnostic features, but the diversity of their past assignment to
genera shows that ichthyologists have not always recognized their
close interrelationship. Some have been moved indiscriminantly
from group to group without apparent reason.
Part of the difficulty regarding the status of Luxilus stems from
the original description (Rafinesque, 1820a: 47-48) which, as is often
true of Rafinesque's accounts, is quite general and vague. Of the
four species he included in this group, only one (L. chrysocephalus)
is now assigned to Luxilus as presently defined.


1964






BULLETIN FLORIDA STATE MUSEUM


His description of Rutilus plargyrus in the same paper shows this
species probably is identical to L. chrysocephalus, despite its assign-
ment to a different genus. The description is accompanied by the
following statement: "I call this genus Rutilus in the supposition that
Cyprinus rutilus may be the type of it; if it should be otherwise, it
may be called Plargyrus."
In 1817 Mitchill described Cyprinus cornutus from the Hudson
River drainage in New York. This species is not closely related to
the genus Cyprinus, a fact that was soon recognized, and it was as-
signed to the old-world genus, Leuciscus. Agassiz (1854), realizing
that the affinities of cornutus were not with Leuciscus either, erected
a new genus, Hypsolepis (this name has been attributed to Baird),
with cornutus as type species.
Plargyrus Rafinesque was employed by Girard (1856: 31-32) to
include the species cornutus. Girard states: "Since Rafinesque's genera
are to be restored, his name Plargyrus is to take the precedence over
the genus Hypsolepis of more modern coining. The name of Plargyrus
was provided for in the Ichthyologia Ohiensis, to replace that of
Rutilus in the eventuality that Cyprinus rutilus of Europe, which was
the type of the genus Rutilus, should prove generically distinct from
Rutilus plargyrus and similar American species, and which is the
case." Girard evidently misunderstood the true relationships of Lux-
ilus chrysocephalus, for he regarded this species as a close relative of
Notemigonus crysoleucas, a species with no close relatives among the
North American Cyprinidae.
Apparently Girard was alone in his use of the name Plargyrus, for
subsequent investigators such as Putnam (1863), Cope (1864, 1867,
1869a, 1869b, 1870, and 1871), Abbott (1870 and 1874), and Jordan
(1875) continued to use Hypsolepis (spelling changed to Hypsilepis
by Cope, 1864) for cornutus and its related forms.
Jordan (1876a: 94), in his first review of Rafinesque's Ichthyologia
Ohiensis, concluded that both Luxilus chrysocephalus and Rutilus
plargyrus of Rafinesque were synonyms of Hypsilepis cornutus. The
genera Luxilus and Rutilus both were described prior to Hypsilepis,
however, and as Rutilus contains only old-world species not closely
related to N. cornutus, Hypsilepis must obviously be placed in the
synonymy of Luxilus. He made no mention of Girard's (1856) paper
which resurrects Plargyrus as a genus for cornutus.
Jordan (1882: 852) redefined the genus Luxilus and included in
it "three well-marked subgenera": Luxilus, Coccotis, and Alburnops.
He defined Luxilus as "large species, with the scales very closely im-
bricated, and much deeper than long; the dorsal over the ventrals


Vol. 8






GILBERT: FISHES OF THE SUBGENUS LUXILUS


and jaws even; type L. cornutus." In Coccotis he proposed to include
"rather large species, with the scales less closely imbricated, but still
deeper than long; dorsal behind ventrals and lower jaw projecting;
type L. coccogenis." The third subgenus, Alburnops, he described
as "small species with the scales normal; the jaws equal and the dor-
sal somewhat behind the ventrals; type L. blennius."
The name Coccotis was not used for many years, and its existence
was overlooked by Cockerell and Callaway (1909) when they erected
the subgenus Coccogenia for the sole reception of N. coccogenis,
based entirely on a supposed difference in the character of its scales.
The name has not been used since.
Jordan and Gilbert (1883) combined all the forms now considered
to belong to Notropis and which have 2,4-4,2 teeth into the genus
Minnilus. As extensive variation in dental arrangement exists among
closely related species of Cyprinidae (Gibbs, 1957: 186), this scheme
of classification was unrealistic and resulted in the splitting of a
number of natural groups.
Jordan (1884: 284) again referred cornutus and its related forms,
including coccogenis, to Luxilus. Gilbert (1884b: 210) proposed that
Notropis Rafinesque (1818) replace Minnilus as the generic name for
a number of species previously referred to many separate genera.
In this Jordan (1885b: 810-814) concurred and published a subgeneric
classification to distinguish the various groups within Notropis.
This system was followed with minor changes by virtually all
ichthyologists until 1929, when Jordan again split Notropis and ele-
vated its numerous subgenera to full generic status. He again re-
garded Luxilus as a genus for cornutus and its closest relatives, but
he placed N. coccogenis, solely on the basis of scale characters, in the
genus Hydrophlox, an artificial group containing a number of unre-
lated species.
Jordan, Evermann and Clark (1930) retained Luxilus as a genus.
They also resurrected as a genus Jordan's Coccotis, which had been
proposed as a subgenus 48 years earlier, and placed in it the species
coccogenis, zonistius, zonatus, brimleyi, and macdonaldi. No reasons
were given for setting up this group.
Few compilers and no practicing ichthyologists have followed
either Jordan's or Jordan, Evermann and Clark's classification con-
sistently. No structural or other basis has been discovered for a
meaningful division at the generic level of this young and cohesive
group. The current trend is to defer judgment on the relationships
between the various subgenera of Notropis until a thorough system-
atic study is made of the entire genus.






BULLETIN FLORIDA STATE MUSEUM


SUBGENERIC DIAGNOSIS AND SPECIFIC
RELATIONSHIPS

The subgenus Luxilus is characterized by the following diagnostic
features: anal rays modally 9 (10 in specimens of N. zonistius from
southern part of range); pectoral rays 14 to 17; lateral-line scales
usually 39 to 41; caudal peduncle scales usually 14 to 16 (7-2-5 to 7);
peritoneum black or heavily speckled with black; pharyngeal teeth
2,4-4,2; dorsal fin inserted directly above or slightly behind anterior
insertion of pelvic fin; anterior lateral-line scales and adjacent scales
taller than wide, notably so in some forms; dark pigmentation present
on scale pockets along side of body in most species, resulting in the
appearance of black crescent-shaped vertical streaks; mouth large,
terminal and oblique; size relatively large; body form in most forms
more or less compressed.

SPECIES SUBSPECIES
N.cornutus


N. albeolus


N.chrysocephalus
chrysocephalus
N. chrysocephalus isolepis

N. cerasinus

N. zonatus


N. pilsbryi



N. coccogenis


N. zonistius

Figure 1. Hypothetical Phylogeny of the Subgenus Luxilus.


Vol. 8







GILBERT: FISHES OF THE SUBGENUS LUXILUS


Three well-defined species groups are recognizable within the
subgenus: The coccogenis group contains two species (N. coccogenis
and N. zonistius), as does the zonatus group (N. zonatus and N. pils-
bryi); the cornutus group consists of five forms (N. cornutus, N. albe-
olus, N. chrysocephalus chrysocephalus, N. chrysocephalus isolepis,
and N. cerasinus). The first two species pairs are completely allo-
patric, but cornutus and its immediate relatives exhibit variable geo-
graphic relationships.
The coccogenis group is defined as follows: body strongly com-
pressed or not; peritoneum heavily speckled with black pigment but
never uniformly black; dorsal fin inserted slightly behind pelvic; scales
along anterior-lateral part of body somewhat taller than wide; poste-
rior borders of scales along side of body smooth and without scallops;
nuptial tubercles weakly developed, nearly granular in some areas;
tubercles present in two or three rows along lower jaw, present on
tip of chin, on upper part of head concentrated mainly along side
and tip of snout; tubercles either present or absent on top of head or
back; tubercles present in both sexes, though poorly developed in
females; crescentic vertical streaks moderately developed, not evi-
dent in some individuals; a red bar present on cheek or on margin of
opercle and preopercle; caudal spot present or absent; a consistent
pattern of dark pigmentation in the dorsal fin, either present or absent
in the caudal fin; circumferential scales usually 13 to 16-2-10 to 12;
caudal peduncle scales normally 7-2-5 or 6; no black lateral stripe
on side of body.
The zonatus group has the following characters: body terete, not
strongly compressed; peritoneum uniformly black; dorsal fin inserted
slightly behind pelvic; scales along anterior-lateral part of body only
slightly taller than wide; posterior edge of scales along side of body
each with four or five definite scallops; nuptial tubercles weakly de-
veloped; tubercles granular, present in a single row along lower jaw,
absent from tip of chin, present on top of head; tubercles pres-
ent in both sexes, though weakly developed in females; crescentic
vertical streaks absent from side of body; no red bar present on cheek
or on margin of opercle and preopercle; caudal spot absent; no per-
sistent pattern of dark pigmentation in either dorsal or caudal fin;
circumferential scales usually 12 or 13-2-12 or 13; caudal peduncle
scales normally 7-2-5 or 6; a distinct black lateral band present along
side of body and encircling snout; a narrow, short secondary stripe
running parallel and slightly dorsad to main lateral band.
The cornutus group may be diagnosed as follows: body deep and
compressed, this character being quite variable among different popu-


1964






BULLETIN FLORIDA STATE MUSEUM


lations of albeolus; peritoneum uniformly black; dorsal fin inserted
directly above pelvic; scales along anterior-lateral part of body much
taller than wide, becoming more pronounced with age; posterior edges
of scales along side of body smooth and without scallops; nuptial
tubercles well developed, the tips usually straight, sometimes curved
forward; tubercles present in a single row along lower jaw, absent
from tip of chin, present over entire head and along back as far as,
and including, anterior part of dorsal fin; tubercles normally present
only in breeding males (weakly developed in females of N. cerasinus);
crescentic vertical streaks usually well developed along side of body,
but either absent or weakly developed in N. albeolus; no red bar pres-
ent on cheek or on margin of opercle and preopercle; caudal spot ab-
sent; no consistent pattern of dark pigmentation present in either the
dorsal or caudal fins; circumferential scales usually 12 to 17-2-11 to
15; caudal peduncle scales usually 7-2-7 (7-2-5 or 6 in N. cerasinus);
a black lateral stripe sometimes present on body of young but not
consistently present throughout life.
The three species groups do not seem extremely close to one an-
other. Their morphological differences are fairly sharp, and no con-
necting links exist between them. The relationships within each
group, on the other hand, appear generally quite close, particularly
in the zonatus and, to a lesser extent, the coccogenis groups. With
one exception this holds true for the cornutus group also. N. cerasi-
nus, although obviously much more closely related to N. cornutus and
N. chrysocephalus than to any of the other species of Luxilus, differs
notably from them in the character of the nuptial tubercles and their
normal presence in females, in the slight difference in appearance of
the scales, and in the number of caudal peduncle scales.
The closest relatives of Luxilus are those species of Notropis that
share with it the largest number of similar or identical morphological
characters. To base a relationship on only one or two shared features
can be misleading. Some of the most constant diagnostic characters
in the forms of Luxilus-pharyngeal tooth counts, anal ray counts,
color of peritoneum, shape of scales along side of body, size of eye,
and relative position of dorsal and anal fins-vary considerably among
other closely related Notropis species.
Notropis ariommus seems rather close to the subgenus Luxilus.
This species has in common with Luxilus: 2,4-4,2 teeth, a black peri-
toneum, large eye, dorsal fin inserted slightly posterior to the pelvics,
and crooked parallel stripes running along the upper side of the back.
In addition ariommus lives under ecological conditions quite similar
to those preferred by all forms of Luxilus, particularly N. zonatus and


Vol. 8







GILBERT: FISHES OF THE SUBGENUS LUXILUS


N. pilsbryi, which it resembles superficially so closely that it has been
confused with zonatus in the literature (Evermann and Hildebrand,
1916: 444).
N. ariommus differs from typical forms of Luxilus in the form of
the scales on the side of the body, which are normal in shape instead
of high and diamond-shaped. From N. zonatus and N. pilsbryi it
also differs in lacking prominent scallops on the posterior edge of the
lateral scales and the black stripe along the side of the body.
While no other Notropis species shares so many features with
Luxilus as ariommus does, Notropis pyrrhomelas of the headwater
areas of the southern Piedmont Plain, which Gibbs (1957) placed in
the subgenus Cyprinella, shares the following characteristics with N.
coccogenis and N. zonistius: Dark pigmentation on the dorsal fin; a
dark crescent-shaped blotch of pigment in the fork of the caudal; high
diamond-shaped scales on the side of the body; a large eye; a prom-
inent dark scapular bar; a peritoneum heavily flecked with black; and
10 anal rays. The last character is found in only a few populations
of zonistius. N. pyrrhomelas differs from coccogenis and zonistius in
having a 1,4-4,1 dental formula. The similarities suggest fairly close
ties between the subgenera Cyprinella and Luxilus.

KEY TO SPECIES AND SUBSPECIES OF THE
SUBGENUS LUXILUS

la. A dark band on dorsal fin (may be orange-red in young). Peri-
toneum heavily speckled with dark melanophores, but not uni-
formly black. A red stripe usually present in life on cheek or
on margin of opercle and preopercle. Nuptial tubercles in two
or three rows along lower jaw extending to tip of chin. Dorso-
lateral scales heavily margined with dark. Circumferential scales
usually 13 to 15-2-10 to 12 2---_ -------
2a. Black band on posterior half of dorsal fin almost parallel to
distal margin. A wide black border in fork area of caudal
fin. No caudal spot. Prominent red bar at margin of opercle
and preopercle. Body more slender and terete. No nuptial
tubercles on top of head, back, or dorsal fin. Angle between
mouth and top of head usually 50 to 520. Vertebrae 40 to
42, usually 41 _____------__________Notropis coccogenis
Tennessee, Savannah, Catawba and New drainages; Ala-
bama, Tennessee, Georgia, South Carolina, North Carolina,
and Virginia.


1964








BULLETIN FLORIDA STATE MUSEUM


2b. Dark diagonal band (orange-red in young individuals, black
in adults) on dorsal fin, not paralleling distal margin, but
beginning on lower third of anterior margin and terminating
at middle of posterior margin. No black border in fork area
of caudal fin. Caudal spot present, about equal to diameter
of eye. Faint red bar sometimes present on cheek. Body
deeper and more compressed. Nuptial tubercles present on
top of head, back, and dorsal fin. Angle between mouth and
top of head usually 57 to 610. Vertebrae 38 to 40, usually
39. Anal rays usually 10 in specimen from southern part
of range. ----------- Notropis zonistius
Chattahoochee and upper Savannah drainages of Alabama,
Georgia, and Florida.

lb. No dark band on dorsal fin. Peritoneum uniformly black. No
red stripe on cheek or on margin of opercle and preopercle.
Nuptial tubercles in single row along lower jaw not extending
to tip of chin. Dorso-lateral scales usually not broadly margined
with dark pigment. Circumferential scales usually 12 to 17-
2-11 to 15 .-______ ..____- ... ...... .3___. .. 3
3a. A prominent black stripe along length of body and encircling
snout; a thin but distinct black stripe parallel to and slightly
above this stripe on body. Body slender and terete. No
black crescent-shaped bars on side of body. Insertion of
dorsal fin slightly behind insertion of pelvic. Nuptial tu-
bercles weakly developed, granular. Lateral scales with
4 or 5 prominent scallops on posterior margin _____--- 4
4a. No black pigment on posterior-dorsal margin of opercle
or on area bounded by or between opercle and anterior
part of lateral line. Pigment covering cleithrum usually
very dark in adults. Area between lateral line and
lateral stripe on middle part of body unpigmented. Pig-
ment unequally distributed on caudal peduncle, very
little present below lateral line. Fairly dense pigment
around edges of dorso-lateral scales, forming parallel
crooked lines that run length of back. Breeding colors
usually more pronounced __- Notropis zonatus
Missouri, Black, St. Francis, and Mississippi drainages;
Missouri and Arkansas.
4b. Black pigment present on posterior-dorsal margin of
opercle and on area between opercle and anterior part


Vol. 8







GILBERT: FISHES OF THE SUBGENUS LUXILUS


of lateral line. Pigment present only on upper half of
cleithrum, usually relatively light in adults. Area be-
tween lateral line and lateral stripe on middle part of
body pigmented. Pigment equally distributed above
and below lateral line on caudal peduncle. Little pig-
ment present around edges of dorso-lateral scales, with
no parallel crooked lines along back. Breeding colors
usually more subdued ------ Notropis pilsbryi
Red, Arkansas and White river systems excluding the
Black River; Arkansas, Missouri, eastern Kansas, and
eastern Oklahoma.
3b. No prominent black stripe along length of body and head
(some darkly pigmented juveniles have an indistinct dark
stripe). Body usually deep and compressed (often slender
and terete in N. albeolus). Black crescent-shaped bars usu-
ally present on side of body (ordinarily absent in N. albeo-
lus). Insertion of dorsal fin directly above insertion of pel-
vic. Nuptial tubercles well developed, not granular. Lat-
eral scales without prominent scallops on posterior margin ___5
5a. Black, crescent-shaped marks on side of body very prom-
inent, numerous, and distinct. Dark upper half of head
sharply contrasting with light lower half, particularly in
breeding males. Nuptial tubercles present in both sexes
though much better developed in males; tips of tu-
bercles in breeding males curved forward at tips.
Scales on dorso-lateral part of body with a fine line about
a quarter of the way from and paralleling outer edge.
Caudal peduncle scales usually 14 or 15 (7-2-5 or 6).
Size relatively small, not exceeding 90 mm. ---
--------------- Notropis cerasinus
Roanoke and New river systems in Virginia and North
Carolina.
5b. Black, crescent-shaped marks on side of body usually
prominent and distinct, though less so than in N. cerasi-
nus (marks ordinarily absent in N. albeolus). Dark up-
per half of head not contrasting so sharply with lighter
lower half. Nuptial tubercles normally present only in
males; tubercles not curved forward at tips. Scales on
dorso-lateral part of body without line paralleling outer
edge. Caudal peduncle scales usually 16 (7-2-7). Size
larger, reaching 175 mm. 6


1964







BULLETIN FLORIDA STATE MUSEUM


6a. Scales relatively large, the -anterior dorso-lateral
scales usually 13 to 16, circumferential scales usu-
ally 24 to 29. Scale pockets in dorso-lateral area
with considerable dark pigment, forming three or
four prominent straight or crooked parallel stripes
along length of back; these meet posterior to the
dorsal fin, forming distinct "V's." No broad, light
area lateral to mid-dorsal stripe in breeding male.
Mid-dorsal stripe not attaining a bright golden hue
in live breeding males; narrow longitudinal stripe
not present beneath parallel stripes on side of back.
Pigment usually present on. chin and anterior part
of gular area ---__7----------
7a. Parallel dark lines along dorso-lateral part of
back crooked, sometimes fairly indistinct. Pre-
dorsal and anterior dorso-lateral scale rows
always crooked. Anterior dorso-lateral scales
(13) 14 to 16 (19); circumferential scales (23) 26
to 29 (32) -
-----.- Notropis chrysocephalus chrysocephalus
Northeastern Oklahoma east to northern Geor-
gia and north to southeastern Wisconsin and
western New York. Intergrading with isolepis
in central Alabama.
7b. Parallel dark lines along dorso-lateral part of
back straight, always distinct. Predorsal and
anterior dorso-lateral scale rows straight and
even, rarely crooked. Anterior dorso-lateral
scales (12) 13 or 14 (15); circumferential scales
(23) 24 to 28 -....---- ------
-____----_- ..-Notropis chrysocephalus isolepis
Southeastern'Oklahoma and northeastern Texas
to west-central Georgia and the Gulf coast. In-
tergrading with chrysocephalus in central Ala-
bama.
6b. Scales relatively small, the anterior dorso-lateral
scales usually 17 to 25, circumferential scales usu-
ally 26 to 35. Scale pockets in dorso-lateral area
with little or no dark pigment; no parallel longitudi-
nal stripes running along back. A broad, light
stripe lateral to mid-dorsal stripe, particularly evi-


Vol. 8








1964


GILBERT: FISHES OF THE SUBGENUS LUXILUS 121

dent in breeding male. Mid-dorsal and narrow
longitudinal stripes (the latter just below broad,
light stripe) attaining a brilliant golden hue in live
breeding males, disappearing soon after death (not
yet observed in N. albeolus). Pigment either pres-
ent or absent on chin and anterior part of gular
area -_-------------- 8
8a. Scales smaller; those in anterior dorso-lateral
region (16) 19 to 24 (30), those in circumferen-
tial series (26) 30 to 35 (39). Crescent-shaped
bars usually present on side of body. General
tone of body silvery-bronze. Dorsal region
without a definite greenish cast in life. Breed-
ing males with prominent, broad, dorso-lateral
light stripe and much red on fins, less on body.
Body usually deep and compressed
---._- ------__-___- .-Notropis cornutus
Northern United States and southern Canada
from northeastern Colorado to southeastern
Saskatchewan and the Atlantic coast as far
south as southern Virginia.
8b. Scales larger; those in anterior-dorso region
(15) 17 to 19 (23), those in circumferential series
(24) 26 to 80 (33). Crescent-shaped bars usu-
ally absent on side of body. General tone of
body silvery with a definite greenish caste in
life. Breeding males with the broad dorso-
lateral light stripe not particularly evident; lit-
tle or no red on fins, none on body except oc-
casionally on tip of snout. Body usually slen-
der and terete, more so in specimens from Roa-
noke and New river systems; less so in speci-
mens -from Neuse, Tar and Cape Fear sys-
tems --____________------- Notropis albeolus
Roanoke, New, Neuse, Tar and Cape Fear
drainages in North Carolina, Virginia, and West
Virginia.








BULLETIN FLORIDA STATE MUSEUM


Notropis coccogenis (Cope)

Warpaint Shiner, figures 2, 14D, map 1.










Figure 2. Adult male Notropis coccogenus, UMMZ 131488, 113 mm SL, from
Cosby Creek, Cocke County, Tennessee.

Hypsilepis coccogenis Cope, 1867: 160 (original description; type locality Holston
River, Virginia).
Leuciscus coccogenis, Giinther, 1868: 253 (description; specimens from Vir-
ginia in British Museum).
Luxilus coccogenis, Jordan and Copeland, 1876: 153 (upper Tennessee
River).
Minnilus coccogenis, Jordan and Gilbert, 1883: 188 (description; range).
Notropis coccogenis, Jordan, 1885b: 814 (range).
Hydrophlox coccogenis, Jordan, 1929: 83 (description; range).
Coccotis coccogenis, Jordan, Evermann and Clark, 1930: 127 (reference to
original description; range).
Notropis brimleyi Bean, 1903: 913-914 (original description; Cane River, North
Carolina; based on atypical specimen of N. coccogenis).
Hydrophlox brimleyi, Jordan, 1929: 83-84 (description; range).
Coccotis brimleyi, Jordan, Evermann and Clark, 1930: 127 (reference to orig-
inal description).

TYPE. A series of 100 specimens (ANSP 3561-3660) bearing the
data "Holston River, Virginia, collected by E. D. Cope," have been
assumed to represent the syntypes of Hypsilepis coccogenis Cope
(Fowler, 1910: 285). Jordan and Evermann (1896b: 285) listed as the
type a specimen (USNM 36849), also collected by Cope, from the
"Holston River, North Carolina." As the Holston River closely ap-
proaches but does not enter North Carolina, the label on the latter
specimen is assumed to be partly in error and should probably read
Virginia. Neither the U. S. National Museum nor Academy of Nat-
ural Sciences of Philadelphia has any record showing how or when
this specimen was transferred, although judging from the entry in the
National Museum catalogue Cope himself probably sent it late in
1884. The readily explained difference in locality data, the fact that
Jordan and Evermann presumably had a valid reason for labeling the


Vol. 8







GILBERT: FISHES OF THE SUBGENUS LUXILUS


specimen as "type," and the similarity in quality of preservation of
specimens from the ANSP and USNM series strongly indicate that
they were collected at the same time. Therefore, I think it logical to
assume that the National Museum specimen is from Cope's original
type series, and I hereby designate it as a lectotype of Hypsilepis
coccogenis. The lectotype specimen is well preserved and is typical
of N. coccogenis. Following are the more important meristic counts:
anterior dorso-lateral scales 16; circumferential scales 14-2-10; lateral-
line scales 39; caudal peduncle scales 7-2-6; pectoral rays 15 on each
side; anal rays 9; pharyngeal teeth 2,4-4,2.
An examination of the syntypes in the Academy of Natural Sciences
revealed they actually comprise five species. In addition to 90 para-
types of Hypsilepis coccogenis which have been recatalogued as ANSP
3561-3650, the following are also present: Notropis ariommus (6 speci-
mens)-ANSP 3651-3656; Notropis rubricroceus (2 specimens)-
ANSP 3657-3658; Notropis chrysocephalus chrysocephalus (1 speci-
men)-ANSP 3659; Notropis galacturus (1 specimen)-ANSP 8660.
DIAGNOSIS. An attenuate and terete, moderately large species of
Luxilus, reaching 116 mm SL (Fowler, 1936: 192). Notropis cocco-
genis differs from other members of its subgenus, except N. zonistius,
in the following characters: 2 to 3 rows of nuptial tubercles on lower
jaw, extending to tip of chin (figure 14D); dorso-lateral scales out-
lined with dark pigment; peritoneum speckled with dark pigment, not
appearing uniformly black. N. coccogenis is unique among forms of
Luxilus in having: outer third of dorsal fin with a moderately oblique
band which nearly parallels distal edge of fin; lower two-thirds of
dorsal fin yellowish; a wide black border on posterior edge of caudal
fin; a red maxillary spot; a prominent red bar at margin of opercle
and preopercle; nuptial tubercles present on side and tip of snout,
absent from top of head and back; an extremely well-developed black
scapular bar; lower jaw usually projecting slightly.
A comparison of N. coccogenis and N. zonistius is presented in
table 1.
DESCRIPTION. Fin ray and scale row counts and proportional
measurements are listed in tables 4 and 9 to 15. Characters men-
tioned in the diagnosis are not repeated in the following description:
Angle of mouth to top of head usually 500 to 520 (range 46 to
550); anterior dorso-lateral scales 15 to 18 (19); circumferential scales
25 to 29 (32) (relative distribution of circumferential scales usually 13
to 15-2-10 to 12; range 13-2-10 to 17-2-13); sum of anterior dorso-
lateral and circumferential scales (40) 41 to 46 (50); caudal peduncle
scales 14 to 16, usually 15; anal rays (8) 9 or 10, usually 9; pectoral


1964







BULLETIN FLORIDA STATE MUSEUM


rays 14 to 16 (17), usually 15; vertebrae 40 to 42, usually 41; back
dark olive; pockets of some scales along side of body with dark pig-
ment, forming a few vertical crescent-shaped bars; breast, belly, and
side metallic-silvery, with a rosy hue in breeding males; prominent
gray-blue mid-dorsal stripe present; snout and upper lip red in breed-
ing males; lower fins white; nuptial tubercles weakly developed, pres-
ent in both sexes, but better developed in males; tubercles present
on upper surface of pectoral fin and on side and tip of snout, absent
from rest of head and back; females metallic-silvery, with only traces
of red markings.
VARIATION. Hildebrand (1932: 63-64) compared the holotype of
Notropis brimleyi Bean (USNM 50601) with specimens of N. cocco-
genis, and concluded that differences in appearance of the lower jaw
in these two forms (projecting in coccogenis, included in brimleyi)
warrant specific separation. His conclusions are not substantiated
by the present study; examination of many individuals of coccogenis
reveals that the supposed differences fall within the normal range of
variation of the species.
Sexual dimorphism is less pronounced than in some Luxilus forms.
Males are usually large, with most individuals over 100 mm long
belonging to this sex, although one of the largest examples reported
by Outten (1957: 75) was a female. During the breeding season males
are more brightly colored than females, particularly as regards the
red on the snout and side of the body. Although nuptial tubercles
are present in both sexes, they are somewhat better developed in
adult males.
DISTRIBUTION. The center of distribution of Notropis coccogenis
is in the upper half of the Tennessee River system in eastern Ten-
nessee, western North Carolina, northern Georgia and western Vir-
ginia, with a relict population present in the Tennessee system of
western Alabama. It occurs in the headwaters of the Savannah River
in North and South Carolina (Pickens, 1928: 30), of the Catawba
River in North Carolina, and (one record) of the New River in north-
western North Carolina (USNM 162281). Its presence in the Catawba
and New systems may possibly be due to introduction. Its apparent
absence from the upper Elk drainage in south-central Tennessee is
possibly due to inadequate collecting.
Contrary to numerous accounts (Jordan and Evermann, 1896a;
Jordan and Evermann, 1896b), N. coccogenis is not known from the
Cumberland River system in Kentucky. Specimens recorded by Wool-
man (1892: 283) from the Big Sandy River of Kentucky have not been
located, and from the known distribution this record is probably in


Vol. 8








GILBERT: FISHES OF THE SUBGENUS LUXILUS


error. A reexamination of material recorded by Fowler (1922: 12;
1923: 11) from the Dan River, Virginia, and from the Neuse River
system, North Carolina, shows these specimens are Notropis albeolus.
LIFE HISTORY AND ECOLOGY. Most of the following information
has been taken from Outten's (1957: 68-84) detailed study of this spe-
cies' life history.
Notropis coccogenis is an inhabitant of the swift riffles and flow-
ing pools of medium to large upland streams. It occurs character-
istically in the middle to upper levels of water and prefers a bottom
composed of rocks and gravel.
Mostindividuals attain maturity when 2 years old. At this time
they usually are about 75 mm SL, with the male averaging slightly
larger than the female.
Outten observed spawning activity in the French Broad River
system of North Carolina from early June to middle July when the
water temperature was 680 or higher. This is somewhat later than
the period of spawning for N. cornutus and N. chrysocephalus in
Michigan. As the temperatures that N. coccogenis requires for breed-
ing do not appear appreciably different from those for cornutus and
chrysocephalus, the late spawning of coccogenis may be due to the
cooler stream temperatures at the high elevations of the Tennessee
uplands.
Spawning takes place in rapids 6 inches or more in depth or in
more slowly flowing water up to 4 feet deep. In the shallow rapids
small groups of up to 8 or 10 males were observed to hold territories
over Hybopsis micropogon nests 21/2 feet long by 2 feet wide, with the
largest male occupying the upstream position. Occasionally one of
the females lingering in a group farther back in the pools moved for-
ward near a male. After remaining side by side a short time, they
settled to the bottom where, with considerable vibration of their
bodies, they spawned. This behavior is similar to that Raney (1940a:
7-8) observed for Notropis cornutas in New York.
Age-growth studies show the main increase in length to take place
in the first and second years of life. During the first year males and
females show little size difference, but in the second and third years
males grow considerably more than females. Apparently little or
no difference in growth rates occurs during the fourth year.
Counts of males and females in each age group show equal num-
bers of each sex present during the first year of life, but a higher
proportion of females among older fish (66.2 percent in the second
year of life, 76 percent in the third). This is paralleled by my ob-
servations on breeding populations of N. cornutus and N. chryso-


1964









BULLETIN FLORIDA STATE MUSEUM


cephalus in Michigan and by those of Pfeiffer (1955: 101) on N. ru-
bellus, in which females appear always to dominate markedly.
N. coccogenis feeds on terrestrial and aquatic animals, principally
insects, with occasional small amounts of vegetable matter, usually
taken at the surface or in the upper water levels. Food seems se-
lected largely according to its relative abundance, with little or
no species-preference evident.


91" 8




,--- .'"




S* -
,I-s

f' -

P/'

J


9' 87*
I9 T


S-u


Io N. :' is: : .:is
0 N. zonistius


85' 83'
J Ir .




1-









,: A
k'
*w ** '


*.^y
^ "-


25 0 25 0 100 0oo
SCALE OF MILES


91' 89" 87' 85' 83* 8J* 79'

Map 1.


Notropis zonistius (Jordan)

Bandfin Shiner, figures 3, 14B, map 1.


44^
t i


Figure 3. Adult male Notropis zonistius, UMMZ 175580, 85 mm SL, from
Cane Creek, Dahlonega, Lumpkin County, Georgia.


Vol. 8


81" 79* 77'


I







GILBERT: FISHES OF THE SUBGENUS LUXILUS


Codoma eurystoma Jordan and Brayton, 1878: 42, 84-85 (partim, type series con-
tained specimens of N. zonistius and C. eurystoma [= N. venustus].
Luxilus zonistius Jordan, 1880:239-240 (original description; type series consists
of young specimens forming part of Jordan and Brayton's type series of
Codoma eurystoma; holotype USNM 23452, 68.5 SL; type locality Su-
wannee Creek, Gwinnett County, Georgia).
Minnilus zonistius, Jordan and Gilbert, 1883: 188-189 (description; range).
Notropis zonistius, Jordan, 1885b: 814 (list; range).
Coccotis zonistius, Jordan, Evermann and Clark, 1930: 128 (reference to orig-
inal description; range).
Hydrophlox zonistius, Schrenkeisen, 1938: 127 (characters; related to Hydro-
phlox coccogenis; northern Georgia).
DIAGNOSIS. A deep, compressed, small species of Luxilus, the
males rarely exceeding 85 mm SL. Notropis zonistius differs from
other members of its subgenus, except N. coccogenis, in the following
characters: 2 to 3 rows of weakly-developed nuptial tubercles on lower
jaw, extending to tip of chin (figure 14B); dorso-lateral scales outlined
with dark pigment; peritoneum speckled with dark pigment, not ap-
pearing uniformly black. N. zonistius is unique among forms of
Luxilus in having: dorsal fin with a strongly oblique black band
(orange-red in younger specimens) beginning on lower third of an-
terior margin and terminating at middle of posterior margin; a distinct
caudal spot, about equal to diameter of eye; a faint red bar along
cheek.
N. zonistius and N. coccogenis are compared in table 1.
DESCRIPTION. Fin ray and scale row counts and proportional
measurements are listed in tables 4 and 9 to 15. Characters men-
tioned in the diagnosis are not repeated in the following description:
Body deep and compressed; mouth large, terminal, and oblique,
usually forming an angle of 570 to 610 with top of head (range 510
to 61.50); anterior dorso-lateral scales 15 to 18 (19); circumferential
scales (25) 27 to 30 (relative distribution of circumferential scales
usually 15 or 16-2-11 or 12; range 13-2-10 to 16-2-13); sum of anterior
dorso-lateral and circumferential scales (40) 43 to 48 (49); caudal ped-
uncle scales 13 to 15, usually 14; anal rays (8) 9 or 10, usually 9;
pectoral rays 14 to 16, usually 15; vertebrae 38 to 40, usually 39; color
of body olivaceous to steel-blue above; side of body with consider-
able coppery luster; pockets of some scales along side of body with
dark pigment, forming a few vertical crescent-shaped bars; prominent
gray-blue mid-dorsal stripe; tubercles on upper part of head best
developed on side and tip of snout, granular on top of head, back,
anterior part of dorsal fin, and upper surface of pectoral fin; tubercles
present on posterior edges of scales on upper half of body.


1964







BULLETIN FLORIDA STATE MUSEUM


The following color description by Reeve M. Bailey 'is based on
live specimens (UMMZ 157882) collected 23 August 1939 from a
tributary of the Chattahoochee River 2.5 miles southwest of Chipley,
Harris County, Georgia: side silvery, tinged with pinkish-orange on
lower parts in large adults; upper parts dark olivaceous; lower fins
white; dorsal fin white, with a band in middle, the band pink in
young, reddish-orange in small adults, becoming entirely black in
large adults; narrow reddish edge at base of caudal fin in smaller
breeding males; tail bright red in large adult, with a whitish base and
distal margin; lowermost ray of caudal fin whitish; reddish-orange bar
extending upward and downward from base of pectoral, this bar
black in breeding males; cheek pale silvery blue.
VARIATION. Nuptial tubercles are present in both males and fe-
males, but they are better developed in the males. Though no de-
tailed comparison of breeding colors has been made between the
sexes, one may assume that the male colors are more pronounced, as
in other species of Luxilus.
Notropis zonistius apparently is the only species of Luxilus with
a gradient in anal ray count (table 13). Specimens from the southern
part of the range have a modal count of 10, as compared to 9 in
specimens from farther north. This situation is comparable to that
found in Notropis cummingsae, in which individuals from upland
areas usually have 9 anal rays and those from lowland areas 10 or 11
(Hubbs and Raney, 1951: 18), and in Notemigonus crysoleucas, in
which specimens from farther south have a greater modal number of
anal rays than those farther north (Schultz, 1927).
Martin (1949: 65) has shown that "a fast growing group of indi-
viduals may have relatively large body parts, such as head and fins."
Thus in individuals developing in a warm environment, the anal fin
grows more rapidly into a longer fin with more fin rays than in indi-
viduals from colder waters. As the variation in anal-ray counts in
N. zonistius probably has an environmental rather than a genetic
basis, the populations differing in this character have not been ac-
corded subspecific status.
DISTRIBUTION. Notropis zonistius is restricted to small tributaries
of the Chattahoochee River system in Georgia, Alabama, and Florida
and the upper Savannah River system in Georgia. The single record
from the Savannah system (CU 25170) is probably the result of stream
capture, although the species may have been introduced there.
LIFE HISTORY AND ECOLOGY. Nothing has appeared in the litera-
ture regarding the life history of N. zonistius. In view of its close


Vol. 8








GILBERT: FISHES OF THE SUBGENUS LUXILUS


taxonomic relationship to N. coccogenis, it is probably similar or
identical to that species in many aspects of its life history.
N. zonistius inhabits clear water with a rather swift current and
apparently prefers pools at the base of riffles to the swifter areas.
Various field records and my limited field experience with it suggest
it also prefers slightly smaller streams than does N. coccogenis.

Notropis zonatus (Agassiz)
Bleeding Shiner, figures 4, 5A, 14A, map 2











Figure 4. Adult Notropis zonatus, UMMZ 148446, 82 mm SL, from Huzzah
Creek, Crawford County, Missouri.

Alburnus zonatus Agassiz, in Putnam, 1863: 9 (original description; type locality
Osage River, presumably in Missouri).
Leuciscus zonatus, Giinther, 1868: 253-254 (description; Osage River).
Minnilus zonatus, Jordan and Copeland, 1876: 154 (Osage River).
Cliola zonata, Jordan and Gilbert, 1883: 183 (in part; Osage River; descrip-
tion based on Notropis dorsalis piptolepis).
Notropis zonatus, Jordan, 1885b: 814 (in part; description; range).
Hydrophlox zonatus, Jordan, 1929: 83 (in part; description; Ozark region).
Coccotis zonatus, Jordan, Evermann and Clark, 1930: 128 (range; synonymy).
Notropis zonatus zonatus, Hubbs and Moore, 1940: 91-99, pl. 1, fig. C (gen-
eral account; comparisons; distribution; synonymy).

TYPE. No holotype was designated for this species, but 13 syn-
types exist, collected in 1854 by George Stolley from the Osage River,
presumably in Missouri. One of these, an adult 65 mm SL (UMMZ
174594) I designate as lectotype of Alburnus zonatus. The specimen
is typical of the species and shows all the distinguishing pigmentary
characters plainly. Three of the paratypes are in the University of
Michigan collections (UMMZ 86915), the other nine at Harvard Uni-
versity (MCZ 1914).
NOMENCLATURE. The authorship of Alburnus zonatus, as well as
of several other species of North American freshwater fishes originally







BULLETIN FLORIDA STATE MUSEUM


described in a paper written by F. W. Putnam in 1863,- has been the
subject of considerable debate. As each of the new names is followed
by "Agassiz, MS. (Nov. sp.)," most succeeding references have cited
Agassiz as their describer. Some still contend that Putnam is the
authority, as he wrote the paper and at no place in the text is it
stated clearly that Agassiz was responsible for anything other than
coining the new specific names. In view of the disagreement the
authorship of these species is best cited as "Agassiz, in Putnam."






A
2 -

















Figure 5. Pigmentation patterns of: A. Adult N. zonatus, UMMZ 152697, 68
mm SL, from Barren Fork, Tuscumbia, Miller County, Missouri. B. Adult N.
pilsbryi, UMMZ 80987, 65.5 mm SL, from Elk River, Grove, Delaware County,
Oklahoma.

Past references to N. zonatus often apply entirely, or in part, to
another species, N. pilsbryi. These forms were regarded as identical
(Hubbs and Brown, 1929:36; Hubbs and Ortenburger, 1929:81-82)
until 1940, when Hubbs and Moore pointed out several well-defined
morphological differences in addition to differences in range. They
regarded the two forms as only subspecifically distinct, but presented
evidence indicating that this was possibly a conservative separation.
DIAGNOSIS. A small species of Luxilus, rarely exceeding 85 mm
SL. Distinguished from other species of Luxilus (except N. pilsbryi)
by: A prominent black lateral stripe that encircles snout and is present
at all ages; a well-developed, narrow, secondary lateral stripe situ-


Vol. 8







GILBERT: FISHES OF THE SUBGENUS LUXILUS


ated parallel and dorsad to the main lateral stripe; a jet-black predorsal
streak; few apical radii on scales; four or five prominent scallops on
posterior margin of those scales situated in or close to the lateral line;
weakly developed nuptial tubercles on top of head and back, present
in one row along lower jaw and absent from tip of chin; body slender
and terete at all sizes.
Distinguished from N. pilsbryi by the following pigmentary char-
acters (see figure 5A and table 2): Lateral stripe on posterior part of
opercle curved sharply upward, not bordering upper half of posterior
margin; lateral dark stripe extending little or not at all below lateral
line posterior to head and anterior to caudal peduncle, and absent
from a small area above middle part of lateral line; black pigment
present over all of cleithral area dorsad to pectoral insertion, becoming
jet-black in breeding individuals; pigment in lateral stripe on caudal
peduncle unequally distributed around lateral line, more present
above than below; pigment more dense around edges of dorso-lateral
scale pockets, forming crooked parallel lines along side of back, the
stripes converging posteriorly with those from opposite side of body;
red and black pigment usually more pronounced, particularly in
breeding males.
DESCRIPTION. Counts and measurements are given in tables 5
and 9 through 15. All distinctive pigmentary characters of the spe-
cies are analyzed in table 2 and are shown in figure 5A. Characters
mentioned in the above diagnosis are not repeated in the following
paragraph:
Anterior dorso-lateral scales 14.to 16 (18); circumferential scales
(25) 26 to 28 (29); sum of anterior dorso-lateral and circumferential
scales (39) 40 to 44 (46); an olive area about 21/ times as wide as mid-
dorsal stripe (not so sharply defined as in N. cornutus) extending along
body on either side of mid-dorsal stripe; side of body silvery with an
intense black stripe running from snout to end of caudal peduncle,
this stripe often overlain with a plumbeous coloration which disap-
pears in preservative; mid-dorsal stripe black, slightly more than half
as wide as eye, and extending around base of dorsal fin; breast, belly,
and lower side of head silvery, becoming flame-red in breeding males;
back olivaceous; lips, preopercular bar, and pectoral axil bright blood-
red in breeding males; top of head dark gray to black, especially dark
in breeding individuals; branchiostegals ordinarily colorless, becom-
ing deep red in breeding males; fins ordinarily pale, becoming orange-
red in breeding males except for gray borders; red most pronounced
in dorsal fin, with some green beneath; red often more reduced in
area and/or intensity in other fins, and in low breeding males largely


1964







BULLETIN FLORIDA STATE MUSEUM


confined to the submarginal bands; nuptial tubercles weakly devel-
oped, granular in all but highest breeding males, in which they pro-
ject slightly; four or five tubercles present in single row on lower jaw
and absent from chin; tubercles present on snout and upper part of
head, but apparently absent from predorsal area of back and dorsal fin.


Map 2.


DISTRIBUTION. Restricted to tributaries of the lower Missouri
River in Missouri, to the Black, St. Francis, and Little river systems
in Missouri and Arkansas, and to a few small tributaries of the Mis-
sissippi River in eastern Missouri.
The Bleeding Shiner is common in the lower (eastern) half of the
Osage system, and is generally distributed throughout the Gasconade
and Meramec drainages. It has crossed the lower Missouri in sev-
eral places and has become established in at least four small tribu-
taries flowing into this river from the north.
The record of N. zonatus for eastern Tennessee (Evermann and
Hildebrand, 1916: 444) was based on specimens of N. ariommus.
Specimens recorded as N. zonatus by Cockerell (1908: 170) from
Boulder, Colorado, are N. cornutus.


Vol. 8








GILBERT: FISHES OF THE SUBGENUS LUXILUS


LIFE HISTORY AND ECOLOGY. N. zonatus is typically an inhabitant
of small to medium-sized Ozarkian streams with a gravel or rubble
bottom and clear, rapidly-flowing water. In these situations it often
is the most abundant fish species present. Hubbs and Moore (1940)
reported this species, as well as N. pilsbryi, to be most common in
moderately swift pools, but my observations indicate it is equally at
home in or near the base of swift riffles, where it often is associated
with N. ariommus telescopus. This habitat is somewhat different
from that preferred by either N. chrysocephalus or N. cornutus, both
of which characteristically live slightly farther away from the bases
of riffles and rarely occur on the riffles except during spawning time.
This apparent preference of zonatus for swifter water is reflected in
its more streamlined form, as contrasted to the deeper, more com-
pressed appearance of chrysocephalus and cornutus.
N. zonatus commonly lives with N. chrysocephalus chrysocephalus,
but has been found only once with N. cornutus. At no time have
N. zonatus and N. pilsbryi been found together, though the two occur
within a few miles of each other in a number of places.
No accounts of either spawning or food habits in N. zonatus have
been published. Its breeding habits are presumably similar to those
of the closely related N. pilsbryi in the Illinois River system of Okla-
homa, which Moore and Paden (1950: 92) have reported on.

Notropis pilsbryi Fowler

Duskystripe Shiner, figures 5B, 6, map 2.





y '" ,, Al .. ..''




Figure 6. Adult male Notropis Ili'.l.ii,. USNM 161632, 74.5 mm SL, from
Spring Creek, Cherokee County, Oklahoma.

Notropis zonatus, Jordan, 1885b: 814 misidentificationn in part; description; range).
Hydrophlox zonatus, Jordan, 1929: 83 misidentificationn in part; description;
Ozark region).
Coccotis zonatus, Driver, 1942: 276 misidentificationn in part; key).


1964








BULLETIN FLORIDA STATE MUSEUM


Notropis pilsbryi Fowler, 1904: 245-246, fig. (original description; compared with
N. leuciodus; holotype, ANSP 24514, 72.5 mm SL; type locality, White
River system at Rogers, Arkansas).
Notropis zonatus pilsbryi, Hubbs and Moore, 1940: 91-99, pl. 1, figs. A and
B (general account; comparisons; distribution; synonymy).
Notropis zonata pilsbryi, Cross, 1954:309 (Cedar Creek and Cottonwood
River, Kansas; Jennings [1942] record for N. cornutus frontalis based on
this species).
Notropis cornutus frontalis, Jennings, 1942: 365 (Rock Creek, Chase County, Kan-
sas).
NOMENCLATURE. Although adequately described by Fowler (1904:
245-246) the taxonomic relationships of Notropis pilsbryi have often
been misunderstood. This is evident from Fowler's original descrip-
tion, which compares the species with Notropis leuciodus, and also
from the identification by Forbes and Richardson (1920: 149) of the
hybrid combination Notropis cornutus (chrysocephalus) X Notropis
rubellus as "Notropis pilsbryi." Hubbs and Ortenburger (1929: 81-
82) and Hubbs and Brown (1929: 36) relegated pilsbryi to the syn-
onymy of N. zonatus, a decision with which Fowler concurred. Al-
though Jordan, Evermann and Clark (1930: 124) and Schrenkeisen
(1938: 123) list pilsbryi as a full species, their works were merely com-
pilations and were not based on original information. Hubbs and
Moore (1940) correctly demonstrated the distinctness of pilsbryi and
zonatus, and their findings have been followed since.
DIAGNOSIS. A small species of Luxilus, rarely exceeding 85 mm
SL. Distinguished from other species of its subgenus (except N.
zonatus) by: A prominent dusky-black lateral stripe that encircles
snout and is present at all ages; a well-developed narrow secondary
lateral stripe situated parallel and dorsad to main lateral stripe; a jet-
black predorsal streak; few apical radii on scales; four or five prom-
inent scallops on posterior margin of those scales situated in or close
to lateral line; weakly developed nuptial tubercles on top of head
and back, present in one row along lower jaw, and absent from tip
of chin; body slender and terete at all sizes.
Distinguished from N. zonatus by the following pigmentary char-
acters (see figure 5B and table 2): Lateral stripe on posterior part of
opercle not curved sharply upward, extending straight across and
bordering upper half of posterior margin; lateral dark stripe extend-
ing below lateral line for entire distance posterior to head; black
pigment in cleithral area extending ventrally only about halfway to
pectoral insertion; pigment in lateral stripe on caudal peduncle equally
distributed above and below lateral line; pigment around edges of
dorso-lateral scales not particularly dense, not forming parallel lines


Vol. 8








GILBERT: FISHES OF THE SUBGENUS LUXILUS


running along dorso-lateral part of back; red and black pigment not
especially pronounced in breeding specimens.
DESCRIPTION. Counts and measurements are given in tables 5 and
9 through 15. All distinctive pigmentary characters of the species
are analyzed in table 2 and are shown in figure 5B. Characters men-
tioned in the above diagnosis are not repeated in the following para-
graph:
Body slender and terete at all sizes; anterior dorso-lateral scales
(14) 15 to 17 (18); circumferential scales (24) 25 to 29 (30): sum of
anterior dorso-lateral and circumferential scales (39) 41 to 45 (48);
an olive area about 21/z times as wide as mid-dorsal stripe (not so
sharply defined as in N. cornutus) extending along body on either side
of mid-dorsal stripe; side of body silvery, with a dusky-black stripe
running from snout to end of caudal peduncle, this stripe often over-
lain with a plumbeous coloration which disappears in preservative;
mid-dorsal stripe black, slightly more than half as wide as eye, and
extending around base of dorsal fin; breast, belly, and lower side of
head silvery, becoming reddish in breeding males; back olivaceous;
lips, preopercular bar, and pectoral axil reddish in breeding males;
top of head dark gray to black; branchiostegals ordinarily colorless,
becoming reddish in breeding males except for gray borders; nuptial
tubercles weakly developed, granular in all but highest breeding
males in which they project slightly; four or five tubercles present
in single row on lower jaw and absent from chin; tubercles present
on snout and upper part of head, but apparently absent from pre-
dorsal area of back and dorsal fin.
DISTRIBUTION, In the White River system (excluding the Black
River) in Arkansas and Missouri and in a few tributaries of the Ar-
kansas and Red rivers in Arkansas, Kansas, and Oklahoma.
N. pilsbryi probably is the most abundant cyprinid in the White
River system, particularly in clear, upland tributaries. It is also
common in many streams draining into the Illinois and Neosho rivers,
tributaries of the Arkansas, although it seems rather scarce in some
parts of the Neosho system. It is found in the lower reaches of the
Illinois River and has been reported from the Arkansas River itself
(Hubbs and Moore, 1940: 94). Its presence in the Red River system
has been suggested by Hubbs and Moore (1940: 94) to be due to in-
troduction, but its occurrence in three widely separated localities in
this drainage suggests natural distribution.
LIFE HISTORY AND ECOLOGY. Notropis pilsbryi occupies the same
type of habitat as the closely related N. zonatus. Its occurrence in


1964







BULLETIN FLORIDA STATE MUSEUM


the turbid Arkansas River may be explained by the presence of springs
in the river bed (Hubbs and Moore, 1940: 94).
Moore and Paden (1950: 92) published the following observations
of spawnitig behavior in "N. zonatus pilsbryi" in the Illinois River
system in Oklahoma, together with an account of possible hybridiza-
tion with Dionda nubila:
"May 3, 1947, was one of those beautiful balmy spring days (air
temperature 22'C., water 200C.) with a perfectly clear sky and no
wind. As we approached a long stretch of riffle, shortly after mid-
day, a considerable area of water appeared to reflect a beautiful
crimson. Approaching closer, to about 20 feet, a most unusual sight
greeted our eyes. An area of about 10 by 6 feet in gently flowing
water about 1 or 11/ inches deep was occupied by thousands of male
zonatus lying so close to each other that it seemed there was no room
for more. They were so beautiful that the eye at first failed to detect
the occasional less brilliant female, and surely there were not very
many lying between the males. Their activities made the surface
of the water quite rough and so the whole process of spawning
escaped us. The males were rolling from side to side so that their
brilliant crimson underparts made the water appear to be on fire.
When we stepped into the riffle the minnows scattered. By means
of a screen we were able to collect eggs, presumed to be of this
species, from the redd.
"Below the riffles occupied by Notropis zonatus was a deeper
area (about 2 by 21/2 feet) from which our seines took a considerable
number of Dionda in breeding color. No evidence of the spawning
of the Ozark minnow was obtained but it seems quite possible that
spermatozoa of Notropis zonatus could be carried by the current into
the spawning area of Dionda nubila where they might fertilize the
eggs of that species."

Notropis cerasinus (Cope)

Crescent Shiner, figures 7, 14C, map 3.
Hypsilepis cornutus cerasinus Cope, 1867: 159 (original description; type locality
headwaters of Roanoke River, Virginia).
Minnilus cornutus cerasinus, Jordan and Gilbert, 1883: 187 (description;
range).
Notropis megalops cerasinus, Jordan, 1889b: 121-123 (description; compari-
son with N. megalops albeolus; abundant in Roanoke River system).
Notropis cerasinus, Jordan and Evermann, 1896a: 256 (range; Synonymy).
Notropis cornutus cerasinus, Fowler, 1910: 284, pl. 18, fig. 31 (comparisons;
status; types in ANSP).


Vol. 8








GILBERT: FISHES OF THE SUBGENUS LUXILUS


Luxilus cornutus cerasinus, Jordan, 1929: 82 (description; range).
Luxilus cerasinus, Jordan, Evermann and Clark, 1930: 128 (reference to orig-
inal description; range).



*.. C.

4. 4 .






Figure 7. Adult male Notropis cerasinus, UMMZ 174710, 81 mm SL, from
Cub Creek, Appomatox, Appomatox County, Virginia.

TYPE. Cope (1867: 159) did not designate a holotype for Hyp-
silepis cornutus cerasinus. The syntypic series (ANSP 3791-3833)
consists of 43 specimens, of which I designate an adult male, 73 mm
SL, as lectotype; this specimen is now catalogued as ANSP 3791.
The remainder of the series contains 8 paratypes of Notropis cerasinus
(ANSP 3792-3799) and 34 specimens of Notropis albeolus (ANSP
3800-3833).
DIAGNOSIS. A small species of Luxilus, not exceeding 87.5 mm
SL, distinguished by: Relatively large scales, the sum of anterior
dorso-lateral and circumferential scales usually 41 to 47; caudal
peduncle scales (12) 13 to 15 (16), usually 14 (range 5-2-5 to 7-2-7);
body deep and compressed; upper half of head darkly pigmented,
contrasting sharply with lighter lower half, particularly in breeding
males; considerable red pigment on body and fins of nuptial males,
possibly more than in any other form of Luxilus; scale pockets on side
of body heavily pigmented, particularly in breeding males, with
many scales appearing entirely black, forming vertical crescent-shaped
bars which are more pronounced than in any other member of the
subgenus; scales in dorsal region with a thin black line parallel to
and approximately one-fifth of way in from posterior edge; nuptial
tubercles present in both sexes, but less well developed in females;
four or five tubercles present in a single row on lower jaw, not reach-
ing tip of chin (see figure 14C); tubercles covering top and side of
head, back and anterior part of dorsal fin; tubercles on head curved
forward at tips.


1964







BULLETIN FLORIDA STATE MUSEUM


DESCRIPTION. Fin ray and scale row counts and body proportions
appear in tables 6 and 9 through 15. A comparison of N. cerasinus
with other members of the Notropis cornutus species group is pre-
sented in table 3. Characters mentioned in the diagnosis are not re-
peated in the following paragraph:
Lateral-line scales and those in surrounding area much taller than
wide, this character more pronounced in older fish, particularly in
adult males; anterior dorso-lateral scales (14) 15 to 20 (22); circum-
ferential scales (23) 24 to 28 (30); sum of anterior dorso-lateral and
circumferential scales (37) 41 to 47 (51); vertebrae (38) 39 (40); angle
of mouth to top of head averaging 57.50 (range 52.5-62.5); color of
dorsal region olive-brown, becoming steel-blue in breeding males;
mid-dorsal stripe gray-blue, slightly more than half as wide as eye
and extending around base of dorsal fin; red on cheek, opercle, lips
and snout of breeding males.
VARIATION. Raney (1947: 126) described the sexual dimorphism
in breeding specimens of N. cerasinus. The following is a summary
of his findings:
1. Males are more brilliantly colored than females, both with
regard to red pigment on fins and body and other dark colors on the
body.
2. Nuptial tubercles are better developed in males.
3. Males have a relatively small, non-protruding anal papilla;
in females this structure is quite swollen and protrudes posteriorly.
4. Males have longer pectoral and pelvic fins than females.
DISTRIBUTION. Limited to the upper part of the Roanoke River
drainage in Virginia and North Carolina and to a few eastern tribu-
taries of the New River system in Virginia, N. cerasinus has the most
restricted range of any Luxilus species. Unlike its near relative,
N. albeolus, it is usually confined to headwater streams. Fowler's
records (1922: 9, 12; 1923: 10; and 1945: 29) of N. cerasinus from the
James River system are based on specimens of N. cornutus.
LIFE HISTORY AND ECOLOGY. Raney's (1947) paper is the only
published work on the life history or ecology of this species. Con-
cerned also with several other species of Cyprinidae, it is quoted in
part here: "On June 21, 1946, in Otter River, Roanoke River system,
4 miles northeast of Bedford, Bedford County, Virginia, three min-
nows, Notropis cerasinus, Notropis ardens ardens and Chrosomus
oreas, were observed over a Nocomis leptocephalus nest. The stream
was a rather fast falling, clear mountain brook, about 75 feet wide


Vol. 8







GILBERT: FISHES OF THE SUBGENUS LUXILUS


on the average, with long, moderately fast riffles and short, shallow
pools. The temperature of the water was 70F., and air 750F., at
12:30 p.m.


Map 3.


"The nest was located near the head of a riffle in moderately swift
water 18 inches deep, about 20 feet from the bank .... A large
male Notropis cerasinus was observed at the head of the nest .
Other male cerasinus soon moved in over the nest and established
territories. They darted nervously about, attempting to drive away
the smaller males of Chrosomus, which were well established over
the nest .... On the other hand, cerasinus males did not often clash
with male gardens, which were conspicuous just over them. .
Female cerasinus occasionally appeared about the periphery of the
nest, but spawning was not seen. The behavior over the nest was in
general much like that of a related species, Notropis cornutus, as
reported by Raney (1940a: 8)."
Actual spawning had not yet taken place when the observations
ended. From all indications the spawning act probably is similar to
that of N. cornutus.


1964









BULLETIN FLORIDA STATE MUSEUM


Breeding individuals of N. cerasinus are usually found from middle
to late June, those of the closely related N. albeolus from 2 to 3 weeks
earlier. This may in part be accounted for by the cooler temperatures
of the headwater streams cerasinus inhabits. The difference in spawn-
ing time may be one explanation for the apparent lack of hybrids
between the two species.

Notropis cornutus (Mitchill)

Common Shiner, figures 8, 9A, 14F, 15D, 15E, map 4.












Figure 8. Adult male Notropis cornutus, UMMZ 121835, 106 mm SL, from
Lost Creek, Defiance County, Ohio.

Cyprinus cornutus Mitchill,1817: 289 (original description [July 1817]; type
locality Wallkill River, New York).
Figure 8. Adult male Notropis cornutus, UMMZ 121835, 106 mm SL, from
Lost Creek, Defiance County, Ohio.

Cyprinus cornutus Mitchill,1817: 289 (original description [July 1817]; type
locality Wallkill River, New York).
Leuciscus cornutus, DeKay, 1842: 207-208, pl. 29, fig. 92 (description; ecol-
ogy; New York).
Hypsolepis cornutus, Storer, 1855: 284-285, pl. 2, fig. 3 (description; Massa-
chusetts).
Plargyrus cornutus, Girard, 1856: 196 (in part; in list of species; Argyreus
rubripinnis a synonym).
Hi,,1-,.. I, cornutus, Cope, 1864: 279 (comparison of specimens from Michi-
gan with those from Susquehanna and Delaware drainages).
Hypsilepis cornutus cornutus, Cope, 1867: 159-160 (description; Susquehanna
River drainage, Pennsylvania).
Luxilus cornutus, Jordan, 1876b: 286-287 (in part; description; range; syn-
onymy).
Minnilus cornutus cornutus, Jordan and Gilbert, 1888: 187 (description; range).
Minnilus cornutus, Gilbert, 1884a: 14 (Ellis, Kansas).
Notropis cornutus, Jordan and Evermann, 1896a: 256 (in part; range; refer-
ence to original description).
Nototropis cornutus, Ruthven, 1906:108 (Porcupine Mountains, Michigan).
Notropis cornutus cornutus, Hubbs, 1926: 46 (intermediate in appearance be-
tween N. cornutus chrysocephalus and N. cornutus frontalis; Atlantic
coast drainage).


Vol. 8









GILBERT: FISHES OF THE SUBGENUS LUXILUS


Luxilus cornutus cornutus, Jordan, 1929: 82 (range; intermediate in appear-
ance between L. cornutus frontalis and L. cornutus chrysocephalus).
Notropis cornuta, Hubbs, 1955: 10 (in part; hybridization with N. rubella).
Cyprinus megalops Rafinesque, 1817: 121 (original description [December 1817];
type locality Hudson River system, New York).
Notropis megalops, Jordan, 1885a: 123 (in part; Plargyrus bowmani and
Plargyrus argentatus are synonyms).
Cyprinus melanurus Rafinesque, 1817: 121 (original description; type locality Hud-
son River drainage, New York).
Cyprinus haematopterus Rafinesque, 1820b: 6 (original description; type locality
Hudson River system, New York).
Cyprinus trivittatus Rafinesque, 1820b: 6 (original description; type locality Fish
and Wallkill creeks, New York).
Leuciscus vittatus DeKay, 1842: 212, pl. 34, fig. 108 (original description; type
locality Mohawk River system, New York).
Argyreus rubripinnis Heckel, 1843: 50 (original description; pharyngeal teeth fig-
ured).
Leuciscus spirlingus Valenciennes, in Cuvier and Valenciennes, 1844: 321-323
(original description; type locality New Jersey.).
Leuciscus plargyrus, Kirtland, 1845: 26, pl. 8, fig. 2 misidentificationn in part; de-
scription; Ohio).
L'euciscus diplemia, Kirtland, 1846: 276, pl. 22, fig. 3 misidentificationn in part;
description; ecology; Ohio).
Leuciscus frontalis Agassiz, 1850: 368 (original description; comparison with Leu-
ciscus cornutus and Leuciscus gracilis; type locality Montreal River, Kee-
weenaw Point, Lake Superior, Michigan).
Plargyrus frontalis, Girard, 1856: 196 (Lake Superior).
Hypsolepis frontalis, Putnam, 1863: 7 (in Harvard collection).
Notropis cornutus frontalis, Evermann and Goldsborough, 1907: 96 (Lake
Superior region, Canada).
Luxilus cornutus frontalis, Jordan, 1929: 82 (description; range).
Notropis cornuta frontalis, Schelske, 1957: 38 (lists probably erroneous record
from Verdigris River system, Kansas).
Leuciscus gracilis Agassiz, 1850: 370 (original description; comparison with Leu-
ciscus cornutus and Leuciscus frontalis; type locality Lake Huron).
Plargyrus gracilis, Girard, 1856: 196 (Lake Huron).
Plargyrus typicus Girard, 1856: 196 (in part; name replaces Rafinesque's Rutilus
plargyrus).
Plargyrus bowmani Girard, 1856: 196 (original description; type locality Sweet-
water River, Nebraska).
Plargyrus argentatus Girard, 1856: 212 (original description; type locality James
River system, Virginia).
Hypsilepis cornutus cyaneus Cope, 1867: 160 (original description; type locality
Montreal River, Keeweenaw Point, Lake Superior, Michigan).
Minnilus cornutus cyaneus, Jordan and Gilbert, 1883: 188 (description; range).
Notropis megalops cyaneus, Jordan, 1885b: 814 (in list; synonymy).
Notropis cornutus cyaneus, Jordan and Evermann, 1896b: 283 (description;
reference to original description; northern Michigan).
Notropis zonatus, Cockerell, 1908: 170 misidentificationn; Boulder, Colorado).


1964








BULLETIN FLORIDA STATE MUSEUM


Notropis universitatis Evermann and Cockerell, 1909: 187 (original description;
type locality Boulder, Colorado).
Hydrophlox universitatis, Schrenkeisen, 1938: 127 (related to Hydrophlox
coccogenis; Colorado).
Notropis cerasinus, Fowler, 1922: 9 misidentificationn in part; James River at Mid-
way Mills, Virginia).
Notropis cornutus cerasinus, Fowler, 1923: 10 misidentificationn in part; James
River system, at Kyle, Virginia).
Notropis cornutus chrysocephalus, Jones, 1929: 32 misidentificationn; Iowa).
Notropis cornutus: chrysocephalus X frontalis, Hubbs and Brown, 1929: 36-38 (in
part; discussion of intergradation; hybridization with Notropis rubellus;
records for Ontario).

TYPE. Mitchill did not designate a type specimen for this species.
As Notropis cornutus has often been confused with N. chrysocephalus,
particularly in areas where the two species occur sympatrically, I
feel that selection of a neotype is desirable. I therefore designate
as neotype of Cyprinus cornutus Mitchill (1817) an adult female
(UMMZ 174540), 69 mm SL, collected in the Wallkill River, 3 miles
southwest of New Paltz, Ulster County, New York, 2 June 1936 by
John R. Greeley and Reeve M. Bailey. Five other topotypes (UMMZ
114107) were collected with the neotype.
The more important meristic counts for the neotype are: Anterior
dorso-lateral scales 20, circumferential scales 34, lateral-line scales 40,
caudal-peduncle scales 16, pectoral rays 16-16, anal rays 9.
NOMENCLATURE. Notropis cornutus has been described as new
and named eleven times. Types are extant for only five of these nom-
inal species; I have examined them as follows:
Leuciscus frontalis Agassiz (MCZ 1751)-holotype
Leuciscus gracilis Agassiz (MCZ 1752)-two syntypes
Plargyrus bowmani Girard (ANSP 3236)-holotype
Hypsilepis cornutus cyaneus Cope (ANSP 3950-3959)-ten syntypes
Notropis universitatis Evermann and Cockerell (USNM 64151)-holo-
type
From 1885 to 1896 the specific name megalops was used in place
of cornutus. The change was instituted by Jordan on the grounds
that Mitchill's (1817: 289) brief diagnosis of Cyprinus cornutus, which
was published in July or August, 1817, did not constitute an adequate
description. Though Mitchill published a detailed description of the
species 6 months later, Rafinesque in the meantime had described the
same form as Cyprinus megalops. Jordan (in Jordan and Evermann,
1896b: 281) later reversed himself, and in 1916 the matter was finally
resolved by the International Commission of Zo6logical Nomenclature,


Vol. 8







GILBERT: FISHES OF THE SUBGENUS LUXILUS


which ruled (Jordan, 1916: 28) that a diagnosis, no matter how brief,
constitutes a description providing there is no reasonable doubt as
to the identity of the species involved.
DIAGNOSIS. A species of Luxilus distinguished from other mem-
bers of the subgenus by the following characters: Anterior dorso-
lateral scales relatively small, numbering (16) 18 to 24 (30); circum-
ferential scales (26) 30 to 35 (39); sum of anterior dorso-lateral and
circumferential scales (42) 48 to 59 (69); mid-dorsal stripe gray-blue,
turning to brilliant golden in live breeding males; a light olive stripe,
slightly wider than dorsal stripe extending along body on either side
of dorsal stripe (figure 15D); below this is another dark stripe, the
same width as the dorsal stripe, which also becomes brilliant golden
in breeding males; pigment usually absent from chin and gular region
(figure 9A), particularly on specimens from west of the Alleghenies.
Further distinguished from the closely related N. albeolus by the
following characters: Anterior dorso-lateral and circumferential scales
smaller (for counts see preceding paragraph, table 7, and tables 9
through 11); considerable red pigment always present on body and
fins of nuptial males, usually absent or subdued in breeding females;
color of dorsal region olive-brown; scale pockets along side of body
usually with considerable pigment, forming prominent crescent-shaped
bars, particularly in breeding males; side of body with a bronzy
aspect in life, as contrasted to the more silvery aspect of N. albeolus;
body deep and compressed; maximum size apparently somewhat
larger.
DESCRIPTION. Fin ray and scale row counts and body propor-
tions were taken on varying numbers of specimens; these appear in
tables 6, 7, and 9 through 15. Characters mentioned in the diagnosis
are not repeated in the following paragraph:
One of the largest forms of Notropis, the males sometimes attain-
ing a length of 175 mm or more; sides of body silvery-bronze, with a
pink wash in breeding males; breast and belly silvery, with little or
no pink in breeding males; pigment wholly or partly absent on chin
and gular region in most populations west of the Appalachian Moun-
tains, variable in most populations east of the mountains; top of head
blue-brown, becoming deep blue-gray in breeding males; sides of
head lighter gray than top; branchiostegals ordinarily colorless, but
light rosy in breeding males; dorsal fin slightly dusky, often becoming
notably darkened in breeding males; other fins normally plain; breed-
ing males pink or red on distal third or fourth of all fins.


1964







BULLETIN FLORIDA STATE MUSEUM


VARIATION. Geographical variation is noteworthy only in number
of anal rays, pigmentation on chin and gular area, and in size of an-
terior dorso-lateral scales. No where is variation consistent enough
to be deemed worthy of taxonomic
Recognition.
S A Populations in Maine, Nova
Scotia, and, presumably, New
Brunswick often have many indi-
viduals (usually 40 to 60 percent)
with 8 anal rays, a variation found
only occasionally in other parts of
the range (table 13). An analysis
of other morphological characters
from the Maine-New Brunswick-
Nova Scotia area reveals surpris-
ingly low anterior dorso-lateral and
circumferential scale counts (table
7), though, as discussed below, they
fall within the overall variation of
this form. In no other way do
SB these fish differ from typical cor-
nutus.
) The chin and anterior part of
the gular area are unpigmented in
populations west of the Appala-
chian Mountains, but this character
shows little apparent consistency in
populations east of the Appala-
chians. Only in the upper Sus-
quehanna River system is pigment
consistently absent and, as dis-
cussed in the zoogeographical sec-
Figure 9. A. Notropis cornutus. In- tion, this is of value in determining
dividuals living west of the Alle- phylogenetic relationships. There
ghenies usually lack pigment on the c b l d t t
gular areas and most of the chin; can be little doubt that this charac-
those from the Atlantic side often ter is largely under genetic control,
have pigment here. B. Notropis a premise substantiated by the fact
chrysocephalus chrysocephalus. This that this part of the head usually is
pattern remains rather consistent pigmented in specimens of N.
throughout most of the range of this chrysocephalus living sympatrically
form. It is often lacking in speci-
mens from the Tennessee River drain- with N. cornutus. In many popu-
age. lations over the northern part of the


Vol. 8








GILBERT: FISHES OF THE SUBGENUS LUXILUS


range specimens of cornutus are heavily pigmented not only on the
underside of the head, but over the entire body. This heavy pigmen-
tation, often so intense that individuals are almost black, is without
doubt a direct result of environmental conditions. Many northern
waters are clear but heavily stained or tea-colored, which apparently
causes an expansion and multiplication of melanophores over the
entire body.
The size of the predorsal scales shows the greatest geographic
variation of any character encountered in this species. Within a
given population scale size is usually more or less uniform. This is
not invariable, and in several series scale size has been found to
vary considerably. There may even be great variation between popu-
lations in adjacent streams. Scale size generally decreases gradually
from south to north, a gradient that is particularly noticeable in
Atlantic coast populations. Northeastern populations, however, show
an abrupt increase in average scale size (a decrease in scale counts),
which reverses the trend seen in other coastal populations of cornutus.
Individuals on the Atlantic coast have larger anterior dorso-lateral
and circumferential scales, on the average, than those living west of
the Appalachians. In the upper Susquehanna River system scale size
averages somewhat smaller than for other coastal populations, a fur-
ther indication of a post-Wisconsin invasion of the Susquehanna by
western populations of cornutus. Difference in scale size has been
the basis for a subspecific separation of populations from opposite
sides of the Appalachians (Hubbs, 1926: 45-46). The desirability of
maintaining this separation, however, has been questioned by some
authors (Hubbs and Brown, 1929: 39; Greeley, 1938: 67), and their
observations are confirmed by the present study.
Sexual dimorphism is pronounced in this species, as in most other
nest-building or territorial cyprinids, such as Semotilus atromaculatus,
Campostoma anomalum, and Hybopsis biguttata. Some differences
are not evident until the breeding season, when males develop a
deep red coloration on body and fins, as well as prominent nuptial
tubercles on the head, back, anterior part of the dorsal fin, and upper
surface of the pectoral fin. Females, in contrast, have a more sub-
dued coloration and usually lack tubercles. Other sexual differences
are present throughout the year and become pronounced in males
only during the breeding season. This is particularly true of the
black crescent-shaped markings on the side of the body, which be-
come notably darker and more prominent. Also the mid-dorsal stripe
and the narrow, light, parallel, longitudinal stripes on the lower part
of the back become relatively lighter, contrasting markedly with the








BULLETIN FLORIDA STATE MUSEUM


rest of the body, and present a definite golden caste in life (Raney,
1940a: pls. 1-4). The golden color disappears shortly after death,
and the stripes again appear darker than the surrounding areas. Sim-
ilar changes have been noted by Lachner (1952: 438) for Hybopsis
biguttata. Other sexual differences are present at all seasons: The
adult male has a deeper and more compressed head and body, longer
pectoral and pelvic fins (table 6), a greater amount of fatty tissue on
the fins, and reaches a larger maximum size (Raney, 1940a: 5).
DISTIBUTION. Occurs over a large part of the northern half of
the United States and southern Canada from northeastern Colorado,
southeastern Wyoming, eastern parts of North and South Dakota and
southeastern Saskatchewan eastward in the southern parts of the Hud-
son Bay drainage, the upper Mississippi and Ohio valleys, and the
Great Lakes-St. Lawrence drainage to the Atlantic coast as far south
as south-central Virginia.
On the Atlantic slope it ranges from the Gasp6 Peninsula and
Nova Scotia as far south as the James River system in Virginia; it is
common in the St. Lawrence and Great Lakes drainages and occurs
in the southern part of the Hudson Bay drainage. The northern-
most record is from a tributary of the Red River of the North in Man-
itoba (Keleher, 1956: 265); the westernmost records are from Wyoming
and Colorado. As indicated by Radforth (1944: 89) the northern
limits of its range coincide closely with the 650 F. July isotherm.
It occurs in the upper part of the Mississippi River drainage, including
parts of the Missouri and Ohio systems, but is not recorded from the
Missouri River system in Montana, and is known from this system
in North and South Dakota only from the James River and other
eastern tributaries. It occurs sparingly in the lower Missouri system
of Nebraska, Iowa, and Missouri, with most records in central Mis-
souri. In the Platte system it is known mostly from upland tributaries
in northeastern Colorado, southeastern Wyoming, and western Ne-
braska, with a few scattered records from the lowest parts of this
drainage. In the Kansas River system it is abundant in places, the
westernmost records being from isolated spring-fed streams. It is
absent from the greater part of the Ohio River system, but is found
in headwater tributaries in Indiana, Ohio, Pennsylvania and New
York, being especially common and widespread in the Allegheny and
Shenango drainages of New York and Pennsylvania. Isolated relict
populations occur to the south in the lower Kanawha River system of
West Virginia (Paint Creek, Fayette County), and in a tributary of
the middle part of the White River in Indiana (Mill Creek, Hendricks
and Morgan counties).


Vol. 8








~B-rlq,~ "4[y:~ J:


0 5
.,%i7. .- T' .-.
43'



'vi to %
W.- q AV's 0s
5 -4 -- 3 .0.

5** 5- 0 '..' ,.. ~ 4


,*Nil v.-

61 o,.~ p ~ f
*e 0: -
4~~9 M~ s5-

seAll* 0
*t7- --i p .tT Z-
.. L*. -'* *


0 37'





I I 33 tJ

Map 4 .
.1i --f -,.
II **' --..I .
-I-
Map 4.


;----r-~

*p,/







BULLETIN FLORIDA STATE MUSEUM


A specimen of N. cornutus in the United States National Museum
(USNM 86211) bears the locality "vicinity of Medicine Hat, Alberta".
The presence of several eastern species such as Semotilus corporalis
and Pimephales notatus in the same collection suggests the locality
data may have been transposed.
A single specimen supposedly collected in the Black River, Clark
National Forest, Missouri (UMMZ 117557) also is regarded as suspect.
There are no other records of this species from the Black River basin,
and the next closest populations are in the Missouri River system ap-
proximately 150 miles to the north. N. chrysocephalus, which was
taken in the same collection, is common throughout the Black drain-
age and, from the known relative distribution of these two forms it
is rather unlikely that they should occur sympatrically so far south.
This record may have resulted either from an introduction or, more
likely, from an accidental transfer of specimens in the laboratory.
Two records of N. cornutus from the Arkansas River drainage in
Kansas (UMMZ 122075, Winfield, Cowley County; KU, Rock Creek,
near lola, Allen County) are also thought to be due to transposed
data. Other collections from Kansas have been found with faulty
labeling, and until these records are substantiated it seems best to
treat them as erroneous.
The populations in the headwaters of the Platte and Kansas rivers,
those from the middle part of the White River, Indiana, and in the
lower Kanawha River system, West Virginia, are the most interesting
and significant ones in the range. Apparently N. cornutus once was
much more widely distributed than at present, and its disappearance
from large areas of the Great Plains can be attributed to the drying
this area has undergone since the Wisconsin glacier retreated north-
ward.
The absence of N. cornutus from a large part of the Ohio River
system is due partly to natural conditions and partly to man's
influence. Trautman (1957:357) has shown that cornutus is now
missing from areas where man's activities have caused the streams
to warm and silt and the flowing springs to disappear. Commonly
it has been replaced by N. chrysocephalus, which reflects the two
species' close ecological relationship and the tolerance of chryso-
cephalus for warmer, more turbid conditions. Early collections
from the Ohio River system show that chrysocephalus has long been
the dominant form and evidently started to replace cornutus before
man started altering the landscape. Apparently a warming of the
streams occurred, and though cornutus probably could have survived
had no other factor been involved, chrysocephalus was favored by


Vol. 8







GILBERT: FISHES OF THE SUBGENUS LUXILUS


the warmer conditions and largely dispossessed it in the southern
part of its range. As these species occupy similar ecological niches,
competition apparently extirpated cornutus from most of the Ohio
Valley.
The presence of isolated populations of cornutus in the middle
part of the White River system in south-central Indiana and in the
lower Kanawha River system of West Virginia strongly supports this
view. Although these populations may have been introduced by
man, the evidence strongly favors their presence as natural. Though
populations of chrysocephalus live only short distances away (see
maps 4 and 5), there is no evidence that it has ever occurred in these
pockets occupied by cornutus. I collected in these areas in August
and September 1958, some 25 years after the cornutus colonies were
discovered, and failed to find chrysocephalus. Apparently some bar-
rier has prevented its moving in here as it has elsewhere.
In the Indiana population the barrier is almost certainly eco-
logical. Mill Creek is unusual in having a shifting-sand bottom
almost throughout its length, with short stretches of rubble and gravel
riffle in a few places where sand has not covered the bottom. N.
cornutus undoubtedly spawns in these limited stretches and has been
able to survive by this thin margin. Apparently the stream is fed by
a number of small springs, for on 26 August 1958 the water tempera-
ture in a shallow, unshaded section of the stream was only 700 F.,
much lower than might be expected considering the weather at the
time. This stream is also one of the few places in southern Indiana
where Rhinichthys atratulus can still be found. The habitat in Mill
Creek is marginal for cornutus and has not allowed the population to
reach large numbers. Paradoxically it has also permitted the species
to survive, for the long stretches of shifting sand bottom have appar-
ently effectively barred the movement of chrysocephalus into the
area.
The reasons for chrysocephalus' absence from Paint Creek in the
lower Kanawha system are less obvious. The area where cornutus
was collected appears suitable for chrysocephalus, which has been
collected near the mouth of Paint Creek. Thus some barrier, either
physical or ecological, must exist in the lower half of the stream. In
its lower reaches Paint Creek is a gravel-bottomed stream consisting
almost entirely of shallow riffles with few or no deep pools. Perhaps
this habitat is unsuitable for chrysocephalus.
LIFE HISTORY AND ECOLOGY. N. cornutus prefers small to medium-
sized streams with clear, cool, weedless water, a moderate to swift
current, and alternating pools and riffles, the latter with a gravel or








BULLETIN FLORIDA STATE MUSEUM


rubble bottom. Though the species is confined to this type of habitat
over most of its range, populations in prairie and plains areas, particu-
larly in Missouri, Iowa, and Nebraska, tolerate extreme turbidity
and seasonally warm water provided a rubble or gravel bottom and
a good current are still present. These factors seem more critical
for spawning than clarity of water. Two other fishes, Rhinichthys
cataractae and Hybognathus hankinsoni, which are commonly thought
to be limited to clear, cool waters, are also able to survive under
these same conditions of turbidity and high temperatures.
N. cornutus is usually found at the head or foot of riffles, most
often the latter, but it tends to avoid the actual riffle except at spawn-
ing time. It likewise avoids still waters, though individuals may
occasionally be found in them, especially if a slight current is present.
Although this species is generally a stream fish, it is also common
in lakes in the northern part of its range. The apparent geographical
bias probably has several causes. Lakes are relatively few in the
southern part of its range and ecologically unsuitable. The northern
lakes differ in usually having clean, weedless, wave-washed shores
with gravel or rubble bottoms, thus in part duplicating the preferred
stream conditions. The species is not known to spawn in lakes, and
apparently enters them from feeder streams. Strangely neither cor-
nutus nor chrysocephalus occurs, except as strays, in western Lake
Erie where ecological conditions similar to those of inland streams
permit the presence of such usually stream-inhabiting fishes as
Noturus flavus, Etheostoma blennioides, and Etheostoma flabellare.
Because cornutus is so common and so easily observed, much has
been written about its ecology and life history. Numerous food
study accounts exist, many of which Adams and Hankinson (1928)
reviewed. Largely carnivorous, it apparently prefers aquatic insects
and their larvae. The diet is controlled largely by food availability,
and plant material is not uncommon fare. This shiner usually feeds
at the surface or in the upper water levels, but it may also take food
on the bottom. This versatility in feeding habits probably partly
explains its great success as a species.
Numerous accounts of its spawning behavior have appeared.
Raney (1940a) reviewed all prior literature and also presented de-
tailed new observations. The species may spawn from mid-May to
mid-July, the time depending mainly on the water temperature; a
minimum of at least 640 F. apparently must be attained before
spawning can occur (Raney 1940a: 4). Although spawning has been
observed in water as warm as 830 F., this is unusual, for spawning is
normally completed before such temperatures are reached.


Vol. 8








GILBERT: FISHES OF THE SUBGENUS LUXILUS


Spawning takes place usually in shallow to moderately deep riffles
(3 to 8 inches) with a gravel or rubble bottom. The fish may excavate
small depressions in gravel or sand in running water, or utilize the
nests built by other species, even when these are in pools. It appears
to prefer the nests of other species when available, with those of such
nest-building forms as Exoglossum maxillingua, Semotilus atromacu-
latus, Semotilus corporalis, Campostoma anomalum, and species of
Hybopsis (subgenus Nocomis) most frequently utilized.
Males arrive first on the riffles, sometimes in large numbers (Fow-
ler, 1909; 540; Raney 1940a: 5). There each individual attempts to
hold a small territory. Females meanwhile remain a short distance
downstream, usually in the first pool. When ready to breed the
females move upstream, take a position over a male, and shortly
thereafter the spawning act is consummated.

Notropis albeolus Jordan

White Shiner, Figures 10, 11, 15C, 15F, map 4












Figure 10. Adult male Notropis albeolus, CU 20322, 120 mm SL, from South
Fork, Roanoke River, Allegheny Springs, Roanoke County, Virginia. (This speci-
men is typical of those in the Neuse, Tar, and Cape Fear drainages.)

Hypsilepis cornutus cornutus, Cope, 1870:459 misidentificationn; Neuse River,
North Carolina).
Luxilus cornutus, Jordan, 1876b: 286-287 misidentificationn in part; range).
Notropis cornutus, Goldsborough and Clark, 1908: 35 misidentificationn; Horse-
pen Creek, Virginia).
Notropis megalops, Jordan, 1888: 58-59 misidentificationn in part; description;
range).
Notropis megalops albeolus Jordan, 1889b: 123, 125, 129 (original description;
comparison with N. megalops; type locality Roanoke River, near Roa-
noke, Virginia; also in Tar and Neuse river drainages, North Carolina).
Notropis albeolus, Evermann and Cox, 1896: 305, 309 (synonymy; Neuse River
system, North Carolina).


1964







BULLETIN FLORIDA STATE MUSEUM


Luxilus cornutus albeolus, Jordan, 1929: 82 (description; range).
Luxilus albeolus, Jordan, Evermann and Clark, 1930: 129 (range; reference
to original description).
Notropis cornutus albeolus, Fowler, 1945:29 (Roanoke and Neuse rivers,
North Carolina).
Notropis coccogenis, Fowler, 1922: 12 misidentificationn; tributary of Dan River,
Virginia).

TYPE. Notropis megalops albeolus was described on the basis of
specimens David Starr Jordan collected in the Roanoke River at
Roanoke, Virginia, 31 July-2 August 1888. Although no types were
designated, the United States National Museum has a series of 16
specimens with the following data: "Roanoke River, Roanoke, Vir-
ginia, 1888, U. S. Fish Commission, D. S. Jordan;" the catalogue entry
is dated 8 April 1889. As these seem to comprise at least part of the
syntypic material of N. albeolus, I hereby designate as lectotype of
Notropis megalops albeolus an adult, 76 mm SL, which retains the
original catalogue number (USNM 40177). Fourteen other specimens
of the original syntypic series now become paratypes and are recata-
logued as USNM 177840. The remaining individual is Notropis
ardens, and has been recatalogued under that name.
DIAGNOSIS. A species of Luxilus distinguished from other mem-
bers of the subgenus by the following characters: Red pigment usually
lacking on body and fins of nuptial males, always absent in females;
dark pigmentation absent from dorsal and caudal fins; scale pockets
along side of body with little or no pigment, resulting in few or no
crescent-shaped bars; side of body silvery in life; no prominent, par-
allel, longitudinal stripes running along length of dorso-lateral region
of body (figure 15C); a light olive stripe, slightly wider than dorsal
stripe, extending along length of body on either side of mid-dorsal
stripe.
N. albeolus is further distinguished from the closely related N.
cornutus by the following characters: scales relatively large, those in
the anterior dorso-lateral series (15) 17 to 20 (23); those in the circum-
ferential series (24) 26 to 30 (33); sum of anterior dorso-lateral and
circumferential scales (41) 43 to 50 (55); color of dorsal region olive-
green; side of body silvery in life, as contrasted to the more bronzy
aspect of N. cornutus; broad light stripes on either side of mid-dorsal
stripe less evident in breeding males; body usually slender and terete
in northern populations, body form in southern populations not ap-
preciably different from N. cornutus; maximum size apparently some-
what smaller, adult males attaining a maximum length of about 130
mm.


Vol. 8







GILBERT: FISHES OF THE SUBGENUS LUXILUS


A comparison of N. albeolus and N. cornutus is presented in tables
3, 6, 7, and 9 through 15, and in figures 15C through 15F.
DESCRIPTION. Fin ray and scale row counts and proportional
measurements are listed in tables 6 through 15. Characters men-
tioned in the diagnosis are not repeated in the following paragraph:
Mid-dorsal stripe grayish-black, slightly more than half as wide as
eye, and extending around base of dorsal fin; a light streak about the
same width as dorsal streak extending along body on either side of
dorsal streak (this character not nearly so clearcut and distinct as in
N. cornutus); below each of these light streaks is another dark streak
of the same width and slightly lighter color than the dorsal streak
(this and the preceding character evident only in breeding males);
pigment on chin and anterior part of gular region either present or
absent; pinks and reds usually absent on body, rarely present on snout
of breeding males; pink or red coloration usually absent from all fins
at all ages, occasionally present on dorsal, caudal, anal, and pelvic
fins of breeding males; dorsal fin slightly dusky, but never greatly
darkened in breeding males; other fins plain; nuptial tubercles present
over head, the tips pointing straight upward, possibly not as well
developed as in N, cornutus and N. chrysocephalus (relatively few
tuberculate specimens of albeolus have been examined).
VARIATION. Notropis albeolus exhibits the most interesting mor-
phological variation found in any form of Luxilus. Differences in
body proportions in particular contribute to a rather perplexing prob-
lem involving its correct taxonomic status. In general specimens from
the New and Roanoke drainages seem similar in appearance, as do
those from the Neuse, Tar, and Cape Fear systems. This similarity
is particularly noticeable in body shape; specimens from the New
and Roanoke tend to be more slender and terete, those from the
other drainages are usually deeper and more compressed. Propor-
tional measurements for individuals from the New and Roanoke aver-




e.




Figure 11. Adult Notropis albeolus, USNM 101339, 76.5 mm SL, from Waqua
Creek at highway 140, Brunswick County, Virginia.


1964







BULLETIN FLORIDA STATE MUSEUM


age less for a number of characters, particularly depth of body, depth
of head, and distance from dorsum to lateral line (table 6). Also
depth of caudal peduncle, length of head, mouth gape, width of body,
width of head, width of eye, length of upper jaw, and all fin lengths
average proportionately slightly less; the caudal peduncle, on the
other hand, averages slightly longer. Although these differences
are usually rather marked, atypical specimens can be found in both
the northern and southern populations (figure 10).
Morphological differences of the above magnitude exist among
individuals of N. cornutus. In that form these differences can always
be explained by environment, slender-bodied fish usually being found
where a shortage of food exists. As food is apt to be scarcer in cool,
relatively sterile northern streams, slender-bodied individuals of cor-
nutus tend to occur to the north. The morphological differences in
albeolus cannot be explained by environment. While streams of the
Roanoke and New River systems are in general clearer, swifter, and
presumably somewhat more sterile than those of the Neuse, Tar, and
Cape Fear drainages, this can hardly explain such sharp differences in
body proportions. In addition the closely related Notropis cerasinus,
a common species in the headwaters of the Roanoke, is characterized
by a body form fully as deep and compressed as either that of N.
cornutus or of the southern population of N. albeolus. This strongly
suggests that the differences in body form in albeolus have a genetic
rather than an environmental basis. Consequently N. albeolus may
best be regarded as containing two distinct races, one inhabiting the
New and Roanoke drainages, the other the Neuse, Tar, and Cape Fear
systems.
The slight difference in scale counts between the two races is sub-
ject to considerable overlap. Table 7 shows specimens from the
New and Roanoke drainages average about one scale more in the
dorso-lateral and circumferential scale series than do those from the
other systems. The upper limit (55) for this count in specimens from
the more northern drainages is 4 more than for specimens from the
Neuse, Tar, and Cape Fear; the lower limit (42) is 1 more.
The crescent-shaped markings that occasionally appear on the
side of the body are more common in specimens from the southern
drainages. Other kinds of variation apparently occur equally in
both races.
The lack of red pigment on both the body and fins, including the
breeding males, is one of this fish's most distinctive characters.
Though this feature holds true most of the time, Robert D. Ross in-
forms me in a letter that occasionally breeding males from both the


Vol. 8








GILBERT: FISHES OF THE SUBGENUS LUXILUS


New and Roanoke show some red on the fins or body. Large adult
males taken in the Tar River 20 August 1946 by Reeve M. and Joseph
R. Bailey (UMMZ 147583) also showed some red on the caudal fin.
Red pigment when present occurs only on the tips of the fins, not
including the pectorals. I have never seen specimens of albeolus so
colored, so I cannot assess possible differences from cornutus in distri-
bution of red on the fins. Apparently the only part of the body on
which red pigment appears is the snout.
Some individuals, particularly those from the Neuse, Tar, and
Cape Fear systems, have a few vertical crescent-shaped bars along
the side similar to those in cornutus, cerasinus, and chrysocephalus.
Such specimens are extremely difficult to distinguish in preservative
from sparsely pigmented individuals of cornutus.
Analysis of vertebral variation in albeolus (table 8) shows that
specimens from the Roanoke and New river systems usually have 40
vertebrae, individuals from the Neuse generally 39. Specimens from
the Tar drainage have either 39 or 40 vertebrae with equal frequency.
The Cape Fear population could not be analyzed because of lack of
material. At first glance the differences in vertebral counts appear
to be directly correlated with variation in body depth, though the
presence of either 39 or 40 vertebrae in Tar River specimens indi-
cates otherwise. Possibly these counts reflect ecological differences
between the several streams.
RELATIONSHIPS. The forms albeolus and cornutus are so close to
the species-subspecies line in their degree of evolution that it is diffi-
cult to decide their proper taxonomic relationship. In the past al-
beolus has been treated both as a subspecies and, in recent years, as
a species, which Moore (1957: 126) has questioned.
I myself have had doubts about the correct taxonomic status of
albeolus. My first inclination (1961a: 2411) was to regard it as a sub-
species of N. cornutus, primarily because not all preserved specimens
of the two forms can be distinguished. This situation is by no means
unique, and may be noted in other North American Cyprinidae, for
example in species of the genus Hybopsis (subgenus Nocomis). Mem-
bers of this group are very similar in over-all appearance, and several
forms are positively distinguishable only by means of the number and
arrangement of the nuptial tubercles on the heads of breeding males.
Thus though only a small percentage of individuals of these species
can be- told apart, the constant differences in the character of the
tubercles leaves little doubt that their relationships are at the spe-
cific level. Many other pertinent examples exist in. fishes, as well
as in other groups of animals.


1964






BULLETIN FLORIDA STATE MUSEUM


The fact that some individuals of cornutus and albeolus cannot
be distinguished does not disprove their specific distinction. Differ-
ences in color and in other features of body pigmentation, in size of
the anterior dorso-lateral scales, and in body shape all argue for
specific recognition.
DISTRmBUTION. Restricted to the Roanoke River system (including
the Chowan) in Virginia and North Carolina, parts of the New River
system in Virginia and West Virginia, and to the upper parts of the
Neuse, Tar, and Cape Fear systems in North Carolina.
The apparent scarcity of albeolus from the more lowland tribu-
taries of the Neuse, Tar, and Cape Fear drainages probably is due
mainly to lack of collecting. Most of the specimens examined from
these systems were taken during the spring when the streams are
usually high, particularly those in the area below the Fall line. Re-
cent autumn collections from the lowland areas indicate that this
species is, in fact, fairly common there.
In recent years N. albeolus apparently has come to occupy a much
more extensive area of the New River system than formerly. Jordan
(1889b: 140-141) did not record it in his 1888 collections from Reed
Creek, near Wytheville, Virginia, yet the species is plentiful there
today. Carl L. Hubbs failed to take it in Wolf Creek, Bland County,
Virginia in 1936, where it was collected both in 1953 and 1956 (UMMZ
169176 and UMMZ 174433). A series of 25 specimens (USNM 177876),
Frank J. Schwartz collected in the Greenbrier River near Caldwell,
Greenbrier County, West Virginia 13 October 1956 also represents
a record for an area where this form was not previously recorded,
despite rather intensive collecting both by Hubbs and John Addair
during the 1930's. Although these range extensions may have resulted
from human introduction, the fish could equally plausibly have found
their way without help.
LIFE HISTORY AND ECOLOGY. No detailed work on the ecology
or life history of N. albeolus has been published. The only data that
have appeared consist of general habitat descriptions or collection
sites.
This species seems to prefer the same habitat as N. cornutus,
cool, clear, medium-sized streams with alternating riffles, flowing
pools, and a rubble to rocky bottom. Although it is usually collected
at the foot of riffles in midwater, it may on occasion be found in
quite dissimilar situations. Some 28 specimens (UMMZ 174749) I
collected in the Otter River, Campbell County, Virginia on 12 June
1956 were living in very warm, turbid, swift water on a bottom com-


Vol. 8









GILBERT: FISHES OF THE SUBGENUS LUXILUS


posed entirely of shifting sand. Interestingly I also found here the
closely related Notropis cerasinus, which is ordinarily restricted to
headwaters of the Roanoke and New systems. Possibly both species
had temporarily moved or been washed out of a nearby tributary
where conditions were more to their liking.
Populations of albeolus in the Neuse, Tar, and Cape Fear drain-
ages are found more frequently in turbid situations than are those in
the New and Roanoke systems. This probably reflects ecological
differences rather than racial dissimilarities in physiological toler-
ances, which are nevertheless possible, especially in view of the
rather marked morphological differences between the two races.
N. albeolus probably breeds in late April or early May, an estimate
based largely on circumstantial evidence. All spring collections I
examined were taken either in early April, late May, or early June,
and none contained gravid females or high tuberculate males.

Notropis chrysocephalus

Striped Shiner.
DIAGNOSIS. A species of Luxilus distinguished from the closely
related Notropis cornutus by: larger scales, those in the anterior dorso-
lateral series (12) 13 to 16 (19); circumferential scales (23) 24 to 29
(32); sum of anterior dorso-lateral and circumferential scales (36) 37
to 45 (48); dorso-lateral scale pockets darkest around edges, which,
with the intensification of pigment caused by the overlapping scales,
results in three parallel stripes running lengthwise along dorso-lateral
part of back and meeting posterior to dorsal fin, thus forming distinct
"V's" when seen from above; pigment usually present on chin and
gular region; mid-dorsal stripe not turning brilliant golden in breed-
ing males; dorso-lateral part of back without a light olive stripe ex-
tending along back just lateral to dorsal stripe.
DESCRIPTION. A large species of Luxilus, the males sometimes
attaining a length of 175 mm or more; body and head deep and com-
pressed, especially in adult males; pectoral fin rays 15 to 17 (18);
lateral-line scales (36) 37 to 40 (42); dorsal region brown; scale pockets
along side of body with considerable pigment on some scales forming
a number of vertical crescent-shaped bars; side of body metallic-
bronze in life; mid-dorsal stripe gray-black, slightly more than half
as wide as eye, and extending around base of dorsal fin.


1964









BULLETIN FLORIDA STATE MUSEUM


Notropis chrysocephalus chrysocephalus (Rafinesque)

Northern Striped Shiner, figures 9B, 12, 15B, map 5.












Figure 12. Adult male Notropis chrysocephalus chrysocephalus, UMMZ 154507,
140 mm SL, from Clear Creek, Wildie, Rockcastle County, Kentucky.

Luxilus chrysocephalus Rafinesque, 1820a: 48 (original description; type locality
Kentucky).
Notropis cornutus chrysocephalus, Hubbs, 1926: 46 (recognized as a distinct
form; intergradation with N. c. frontalis; name frontalis applied to chry-
socephalus by Jordan and Evermann; nominal forms plumbeolus and
gibbus of Cope are synonyms; Great Lakes drainage).
Luxilus cornutus chrysocephalus, Jordan, 1929: 82 (description; range).
Notropis chrysocephalus chrysocephalus, Gilbert, 1961a: 2411-2412 (general
account).
Notropis chrysocephalus, Larimore and Smith, 1963: 823, 328, 332-333, 335,
338-339, 343, 345, 358, fig. 31 (Champaign County, Illinois).
?Rutilus plargyrus Rafinesque, 1820a: 50-51 (original description; type locality
Kentucky).
Leuciscus plargyrus, Kirtland, 1845: 26-27, pl. 8, fig. 2 (in part; description;
range; Ohio).
Luxilus plargyrus, Jordan, 1876a: 94 (review of Rafinesque's species).
?Leuciscus gibbosus, Storer, 1846: 166 (in part; description; Huntsville, Alabama).
?Hypsolepis gibbosus, Agassiz, 1854: 359 (Storer's Leuciscus gibbosus placed
in genus Hypsolepis).
?Plargyrus gibbosus, Girard, 1856: 196 (in part; list of related species; Ala-
bama).
Plargyrus typicus Girard, 1856: 196 (in part; name replaces Rafinesque's Rutilus
plargyrus).
Plargyrus cornutus, Girard, 1856: 196 misidentificationn in part; list of species).
Hypsolepis cornutus, Putnam, 1863: 7 misidentificationn in part; in Harvard
collection).
Leuciscus cornutus, Giinther, 1868: 249 misidentificationn in part; specimens
in British Museum from Michigan, Lake Erie, and Miami River).
Hypsilepis cornutus, Cope, 1869b: 372-373 misidentificationn in part; descrip-
tion; variation; range; Pennsylvania).
Luxilus cornutus, Jordan, 1876a: 94 misidentificationn; review of Rafinesque's


Vol. 8









GILBERT: FISHES OF THE SUBGENUS LUXILUS


species; Luxilus chrysocephalus Rafinesque synonymized with Cyprinus
cornutus Mitchill).
Minnilus cornutus, Jordan and Swain, 1888:249 misidentificationn; Clear
Fork of Cumberland River, Kentucky).
Notropis cornutus, Gilbert, 1884b: 202 misidentificationn; restoration of genus
name Notropis; East Fork of White River, Indiana).
?Cyprinus cornutus, Call, 1896: 15 misidentificationn; Falls of the Ohio, Louis-
ville, Ky.; possibly refers to another species since Notropis megalops also
listed).
Notropis cornutus cornutus, Knapp, 1958: 61 misidentificationn; replaces N.
cornutus isolepis immediately to north).
Notropis cornuta, Hubbs, 1955: 10 misidentificationn in part; hybridization
with N. rubella).
Alburnops plumbeolus Cope, 1864: 282 (original description; type locality Saginaw
River, Flint, Michigan).
Hybopsis plumbeolus, Jordan, 1876b: 283 (description; range).
Minnilus plumbeolus, Jordan and Gilbert, 1883: 192 (description; range).
Luxilus plumbeolus, Jordan, 1884: 294 (description; range; synonymy).
Hypsilepis frontalis, Cope, 1864: 279 misidentificationn; description; southeastern
Michigan).
Hypsilepis cornutus frontalis, Cope, 1867: 159 misidentificationn; description;
Holston River, Virginia).
Minnilus cornutus frontalis, Jordan and Gilbert, 1883: 187 misidentificationn;
description; range).
Notropis megalops frontalis, Jordan, 1885b: 814 misidentificationn; in list;
synonymy).
Notropis cornutus frontalis, Jordan and Evermann, 1896a: 256 (misidentifica-
tion; range; reference to original description).
Hypsilepis cornutus gibbus Cope, 1867: 158 (original description; type locality
southeastern Michigan).
Minnilus cornutus gibbus, Jordan and Gilbert, 1883: 187 (description; range).
Hybopsis lacertosus Cope, 1869a: 280-232 (original description; type locality Bear
Creek, tributary to Middle Fork of Holston River, Virginia).
Alburnops lacertosus, Jordan, 1878: 420 (range).
Hydrophlox lacertosus, Jordan and Brayton, 1878: 64, 84-85 (Holston River,
Virginia).
Minnilus lacertosus, Jordan and Gilbert, 1883: 192 (description; headwaters
of Tennessee River).
Luxilus lacertosus, Jordan, 1884: 294 (description; range; synonymy).
Notropis lacertosus, Jordan, 1885b: 814 (range).
Luxilus cornutus lacertosus, Schrenkeisen, 1938: 129 (headwaters of Tennes-
see River).
Notropis megalops, Jordan, 1885a: 123 misidentificationn in part; Plargyrus typi-
cus a synonym).
Notropis cornutus: chrysocephalus X frontalis, Hubbs and Brown, 1929: 36-39 (in
part; discussion of intergradation; hybridization; records for Ontario).

TYPE. Rafinesque designated no type specimen for this species.
As N. chrysocephalus has often been confused with N. cornutus, par-


1964







BULLETIN FLORIDA STATE MUSEUM


ticularly in areas where the two species occur sympatrically, I feel
the selection of a neotype is desirable. I therefore designate as neo-
type of Luxilus chrysocephalus Rafinesque (1820) an adult male
(UMMZ 174539) 99 mm SL, collected in a creek about 6 miles SSW
of Danville, off highway 35, Lincoln County, Kentucky 5 April 1953
by R. M. Bailey and party. Six other topotypes (UMMZ 165232)
were collected with the neotype.
The more important meristic counts for the neotype are: Anterior
dorso-lateral scales 14, circumferential scales 28, lateral-line scales
38, caudal-peduncle scales 16, pectoral rays 17-17, anal rays 9.
NOMENCLATURE. The "large-scaled form" of N. cornutus has long
been recognized by ichthyologists. Many chose to regard it merely
as a variant of cornutus, although some separated it subspecifically.
Consequently the identification of N. chrysocephalus in the literature
often must be resolved by geography. Cope (1864: 279) believed
Agassiz's description of Hypsilepis frontalis was based on this spe-
cies, and as a result the name frontalis was incorrectly applied for
more than 50 years. Not until 1926 did Hubbs point out that frontalis
represents the fine-scaled northern form, and that the name chryso-
cephalus, which Jordan (1876a: 94) had placed in the synonymy of
Cyprinus cornutus Mitchill, applied to the coarse-scaled southern
form.
Cope's (1867: 158) description of Hypsilepis cornutus gibbus pos-
sibly was based on the hybrid Notropis cornutus X Notropis chryso-
cephalus. Although Cope recorded only 16 predorsal scales for this
form, individuals of the cross sometimes have almost as large scales
as does chrysocephalus. Unfortunately the types of gibbus appar-
ently no longer exist so this question is impossible to resolve.
An examination of the seven syntypes of Alburnops plumbeolus
Cope (ANSP 2055-2061) shows that six specimens are N. chrysoceph-
alus, the other Notropis heterodon. To fix the status of the specific
name plumbeolus and to prevent possible confusion with Notropis
heterodon I designate as lectotype of Alburnops plumbeolus a speci-
men 40.5 mm SL, catalogued as ANSP 2055. Five specimens become
paratypes (ANSP 2056-2060), while the last is recatalogued as Notropis
heterodon (ANSP 2061).
Cope (1869a: 230-232) described Hybopsis lacertosus on the basis
of five specimens from Bear Creek, a tributary of the Middle Fork of
the Holston River, Virginia. Nothing in the original description indi-
cates how this form differs from Notropis c. chrysocephalus or No-
tropis coccogenis, the species of Luxilus occurring in this area, al-
though the name "lacertosus" signifies a slender body. Unfortunately,


Vol. 8








GILBERT: FISHES OF THE SUBGENUS LUXILUS


the types (ANSP 2835-2839) are in extremely poor condition, the
bodies having become too decomposed as to be of aid in recognition.
The heads, including the pharyngeal teeth are still intact, as are most
of the fins. The head is deep in all specimens, in no case does the
lower jaw project notably beyond the upper, and there is no sign of
a black band on any dorsal fin. As all the above characters are indica-
tive of Notropis chrysocephalus, I regard Hybopsis lacertosus Cope
as a synonym of that species.
DIAGNOSIS. A subspecies of N. chrysocephalus distinguished from
N. c. isolepis by: Slightly smaller scales, those in anterior dorso-lateral
series (13) 14 to 16 (19); circumferential scales (23) 26 to 29 (32); sum
of anterior dorso-lateral and circumferential scales (38) 40 to 45 (48);
predorsal and anterior dorso-lateral scale rows more crooked and un-
even; anterior dorso-lateral stripes crooked, more poorly defined, and
not uniformly parallel (figure 15B); pigment usually present on chin
and gular region (figure 9B often absent in specimens from Tennessee
drainage).
DESCRIPTION. Fin ray and scale row counts and body proportions
were taken on varying numbers of specimens. These appear in tables
6 and 9 through 15. Characters mentioned in the description of
Notropis chrysocephalus and in the diagnosis are not repeated in the
following paragraph:
Arrangement of circumferential scales above and below lateral
line (11-2-10) 12-2-12 to 14-2-13 (15-2-15); dorso-lateral scale pockets
darkest around edges which, with the intensification of pigment
caused by the overlapping scales, results in three crooked, fairly
well-defined parallel stripes running lengthwise along dorso-lateral
part of back; side of body silvery-bronze with a deep rose-red in breed-
ing males; breast and belly silvery with little or no pink in breeding
males; branchiostegals colorless, with a decided rosy wash under-
lain with gray in breeding males; distal third or fourth of dorsal, cau-
dal, anal, and pelvic fins pink in breeding females, deep pink to red
in breeding males, except for extreme distal edges which are white;
outer fifth of pectoral fin rosy, distal edge white in nuptial males;
basal two-thirds of dorsal fin normally pallid, with a pinkish blush in
nuptial males; basal two-thirds to three-fourths of anal, pelvic, pec-
toral, and caudal fins pallid.
VARIATION. Relatively little morphological variation occurs in this
subspecies. Populations containing individuals with slightly smaller
scales are found throughout the range, but the resultant high counts
fall well within the normal range of variation for the species and pose
no nomenclatural problem. A series of 35 specimens from the Coosa


1964









162 BULLETIN FLORIDA STATE MUSEUM Vol. 8


K.







GILBERT: FISHES OF THE SUBGENUS LUXILUS


River system, Alabama (UMMZ 175746) which showed the highest
counts for any individuals of this form exhibited a number of abnor-
malities such as deformed fins and "bulldog" snout. These anomalies
were also found in other species from the same locality, suggesting
that some factor such as the chemical composition of the water was
responsible, probably during early embryological development.
Occasionally the ordinary dark pigment on the anterior dorso-
lateral scale pockets is either faint or virtually absent. In such cases
chrysocephalus approaches cornutus in appearance and looks like a
hybrid between these two species. The resemblance is merely super-
ficial, for such individuals are found in areas where it seems unlikely
that cornutus has ever occurred. As these are typical chrysocephalus
in all other respects, the suppression of pigment along the upper side
of the body apparently is of no special significance.
Individuals occasionally lack pigment on the chin and gular re-
gion. This is particularly common in the Tennessee River system,
where the pigment is wholly or partially absent in approximately half
the individuals.
DISTRIBUTION. Found throughout a large part of east-central
United States from the upper Coosa River system in Alabama and
Georgia and the lower Mississippi River drainage in Alabama, north-
eastern Mississippi, Tennessee, northern Arkansas, and northeastern
Oklahoma, northward to the lower Great Lakes region from south-
eastern Wisconsin to western New York.
Common in the upper Coosa River system of Georgia and Ala-
bama, but becomes increasingly less common in the lower parts of
that drainage. It is abruptly replaced by N. c. isolepis a short dis-
tance north of the confluence of the Coosa and Tallapoosa rivers, with
no apparent evidence of intergradation. It also occurs in the eastern-
most tributaries of the Cahaba and Black Warrior rivers and is re-
placed by isolepis in the lower parts of these drainages also. Collec-
tions are at present not extensive enough to permit an accurate pic-
ture of the relationships of the two forms, although intergradation is
proved by a series of 16 specimens (UMMZ 168613) from the head-
waters of the Black Warrior system.
Found in the headwaters of the Illinois and Neosho rivers (trib-
utaries of the Arkansas) in northeastern Oklahoma, northwestern Ar-
kansas, and southwestern Missouri. It has not yet been recorded from
Kansas, but should occur in the extreme southeast corner of that state.
To the east chrysocephalus is common throughout most of the White
and Black river systems of northern Arkansas and southern Missouri,
but occurs sparingly in the St. Francis River system and other small


1964







BULLETIN FLORIDA STATE MUSEUM


tributaries of the Mississippi River in eastern Arkansas and Missouri.
Apparently absent from western Tennessee west of the Tennessee
River system, although it is quite common throughout most of the
Tennessee drainage itself in Tennessee, extreme northeastern Missis-
sippi, northern Alabama, northwestern Georgia and southwestern Vir-
ginia. It has not yet been reported from North Carolina.
To the north it is one of the most abundant cyprinids in the Ohio
River system, except for the upper Kanawha (New) River drainage,
where it is practically absent-the single record for this drainage
(USNM 177878) is believed an introduction.
Common in tributaries of the Gasconade and Meramec rivers
in Missouri, it is rare in the Osage drainage of that state. It is found
in some of the smaller tributaries of the Mississippi River in eastern
Missouri both north and south of the Missouri River, and occurs in a
few scattered localities in northeastern Missouri. There are as yet
no records for Iowa. It is common in the Illinois River system in
Illinois and in suitable habitats of the more eastern and southern
parts of the state, but it has not yet been found west of the Illinois
drainage.
It occurs throughout most of the lower Great Lakes region and
is entirely absent only from the Lake Superior drainage. Its appar-
ent absence from the more northern tributaries of the other Great
Lakes seems to be related primarily to temperature, as its northern dis-
tributional limit coincides closely with the 70'F July isotherm (Rad-
forth, 1944: 89). In the Lake Michigan basin it is known from ex-
treme southeastern Wisconsin, northeastern Illinois, northern Indiana,
and from western Michigan. It occurs sparingly in the Lake Huron
basin, having been recorded only from certain tributaries of Saginaw
Bay in Michigan and from a few localities in Ontario. It is wide-
spread and common throughout most of the Lake Erie drainage. In
the Lake Ontario basin its occurrence is spotty, with all known records
coming from streams flowing into the south side of the lake.
The lack of records from any streams draining into the Atlantic
Ocean indicates that this species probably was a late entrant into the
Great Lakes drainage and failed to reach the Horseheads Outlet into
the upper Susquehanna River system (Bailey, 1945: 125-126) before
this connection was eliminated.
In recent times N. chrysocephalus seems to have replaced the
closely related N. cornutus throughout much of the latter's former
range. This is substantiated by the presence of isolated populations
of cornutus in West Virginia and Indiana, and by Trautman's (1957)
observations over the past 35 years on the relative distribution of


Vol. 8







GILBERT: FISHES OF THE SUBGENUS LUXILUS


these two species in Ohio. A more complete discussion of the rela-
tionships of the two forms appears in the account of cornutus and in
the section on zoogeography and phylogeny.
LIFE HISTORY AND ECOLOGY. N. c. chrysocephalus prefers an en-
vironment similar to that of N. cornutus. Both are ordinarily found
in small to medium-sized streams having clear, weedless water, a
moderate to swift current, and alternate pools and riffles, the latter
with a gravel and/or rubble bottom. Individuals usually are found
at the head or foot of riffles, more often the latter, but tend to avoid
the riffles themselves except at spawning time. They likewise avoid
quiet water, but individuals occasionally may be found there espe-
cially if a slight current is present.
As a rule chrysocephalus seems somewhat more tolerant than N.
cornutus of warm, turbid conditions and has replaced it in a number
of places where the environment has been so modified. The apparent
preference of chrysocephahus for warmer waters is shown by the dis-
tribution of the two species when both occupy the same stream; the
main body of the cornutus population is more apt to occur toward the
headwaters, chrysocephalus usually is more common in the lower parts
(Trautman, 1939: 285). When an obstruction was built across one
such stream (Gilbert, 1961b: 188) N. cornutus took over completely,
apparently because that segment of the population of chrysocephalus
occurring toward the headwaters could not compete successfully un-
less continually reinforced from the main population downstream.
N. chrysocephalus is rarely found in lakes, partly because they are
scarce throughout most of the species' range. Only to the north
is chrysocephalus found with any consistency in standing waters, and
here clear, gravel or rubble-bottomed, wave-washed shores are usually
present. As in N. cornutus, spawning probably does not occur in the
lakes themselves, but in tributary streams.
References to food habit studies of N. cornutus listed by Adams
and Hankinson (1928) refer in part to N. chrysocephalus. Further ac-
counts of the food habits of chrysocephalus are discussed by Langlois
(1954). These studies reveal no apparent differences in food habits
between cornutus and chrysocephalus.
Raney (1940a) described in detail the spawning behavior of N.
cornutus and assembled the literature on breeding both in this form
and N. chrysocephalus. This shows no essential differences in spawn-
ing behavior between the species, a finding supported by my own
observations and those of others (Hankinson, 1932; Lachner, 1952).
Whatever differences exist in spawning requirements most likely in-
volve optimum temperatures and/or ranges of spawning temperature.


165


1964







166 BULLETIN FLORIDA STATE MUSEUM Vol. 8

The northern limits of distribution of the two species seem to indicate
that such differences do exist. Both species hybridize with other
species with roughly equal frequency.

Notropis chrysocephalus: chrysocephalus X isolepis
Notropis megalops, Gilbert, 1891: 157 (not of Rafinesque; in part; collections from
area of intergradation in Black Warrior River system, Alabama).
The only collection I have seen that contains positively identified
intergrades (UMMZ 168613) is from Blackburn Fork (9 miles south-
west of Oneonta, Alabama), a headwater tributary of the Black War-
rior River. Further collecting will probably reveal intergrading pop-
ulations in a number of localities, most likely in other headwater trib-
utaries of the Black Warrior and Cahaba systems. Other areas where
intergrades might occur are the lower Coosa drainage where the two
subspecies appear to be allopatric, the headwaters of the Tallapoosa
River, the more westerly headwater gulf tributaries bordering the
Tennessee River drainage, and the Arkansas River system in Arkansas.
The chief basis for identification of UMMZ 168613 as an inter-
grading population is the presence of typical individuals of both sub-
species together with intermediates. Several specimens have pro-
nounced, straight, dorso-lateral stripes running along the side of the
back, while in others these stripes are crooked and poorly defined.
The predorsal and anterior dorso-lateral scale rows in the former
(isolepis type) individuals are regular and even, while these rows are
irregular and uneven in the latter (chrysocephalus type) specimens.
The subspecies chrysocephalus and isolepis are readily distin-
guished not only on the basis of the criteria discussed above, but also
by notable differences in both the anterior dorso-lateral and circum-
ferential scale counts (tables 9 through 11). These differences by
themselves are not sufficiently great to permit identification of small
series of intergrades, such as the above.
The apparent scarcity of intergrading populations of chrysocepha-
lus and isolepis, together with the rather unusual distribution of these
forms in the Coosa system, suggest that these forms are close to the
specific level of separation. I maintain them as subspecies because,
in addition to the intergrading population previously cited, specimens
of isolepis living well out of the area of potential intergradation occa-
sionally have rather uneven anterior dorso-lateral stripes, causing the
individual to appear more or less intermediate. Also past collections
from near the known area of intergradation usually contain too few
specimens or the individuals are too small to permit accurate identi-








GILBERT: FISHES OF THE SUBGENUS LUXILUS


fiction. Hence some series of specimens identified to subspecies may
actually consist of intergrades.

Notropis chrysocephalus isolepis Hubbs and Brown

Southern Striped Shiner, figures 13, 14E, 15A, map 5.













Figure 13. Adult male Notropis chrysocephalus isolepis, UMMZ 161262, 95
mm SL, from Vernon, Jackson Parish, Louisiana.

?Leuciscus gibbosus Storer, 1845: 48 (original description inadequate for positive
identification; no type specimens; type locality Tuscaloosa, Alabama).
?Plargyrus gibbosus, Girard, 1856: 196 (in part; list of related species; Ala-
bama).
Luxilus cornutus, Jordan, 1876b: 286-287 misidentificationn in part; description;
range; synonymy).
Notropis cornutus, Jordan and Evermann, 1896a: 256 misidentificationn in
part; range; reference to original description).
Notropis megalops, Gilbert, 1891: 157 misidentificationn in part; Black Warrior
and North rivers, Alabama).
Notropis cornutus isolepis Hubbs and Brown, in Ortenburger and Hubbs, 1927:
129-131 (original description; holotype, UMMZ 73090, 101 mm SL;
four paratopotypes, UMMZ 73091; type locality Mountain Fork R.,
10 mi. southeast of Broken Bow, McCurtain Co., Oklahoma; comparison
with related forms; counts; range).
Luxilus cornutus isolepis, Jordan, Evermann and Clark, 1930: 128 (range;
synonymy).
Notropis cornuta isolepis, Hubbs, 1951: 490 (first records for Texas).
Notropis chrysocephalus isolepis, Gilbert, 1961a: 2411-2412 (general account).

DIAGNOSIS. A subspecies of Notropis chrysocephalus distinguished
from N. c. chrysocephalus by: Slightly larger scales, those in anterior
dorso-lateral series (12) 13 or 14 (15); circumferential series (23) 24 to
28; sum of anterior dorso-lateral and circumferential scales (36) 37
to 41 (43); predorsal and anterior dorso-lateral scale rows very straight
and even; anterior dorso-lateral stripes usually straight, well-defined,


1964









BULLETIN FLORIDA STATE MUSEUM


Figure 14. Tubercle patterns in Luxilus: A. Notropis zonatus, UMMZ 116373;
B. N. zonistius, UMMZ 157882; C. N. cerasinus, UMMZ 174710; D. N. cocco-
genis, CU 18558; E. N. chrysocephalus isolepis, UMMZ 161262; F. N. cornutus,
UMMZ 95626.


Vol. 8







GILBERT: FISHES OF THE SUBGENUS LUXILUS


and uniformly parallel; pigment always present on chin and gular
region, often with a streak of black running posteriorly down gular
membrane.
A comparison of N. c. isolepis and N. c. chrysocephalus is presented
in table 3.
DESCRIPTION. Body proportions and counts of fin rays and scale
rows are listed in tables 6 and 9 through 15. Characters mentioned in
the description of Notropis chrysocephalus and in the above diagnosis
are not repeated in the following paragraph.
Arrangement of circumferential scales above and below lateral
line (10-2-11 or 11-2-10) 11-2-11 to 13-2-13 (15-2-13); dorso-lateral scale
pockets darkest around edges which, with the intensification of pig-
ment caused by the overlapping scales, results in three straight, well-
defined, parallel stripes running lengthwise along dorso-lateral part
of back, the stripes converging posteriorly with those from opposite
side to give the appearance of long, narrow, parallel "Vs" when viewed
from above (figure 15A); pigment always present on chin and gular
area.
The following color description is based on an adult male (UMMZ
161262) collected on 6 June 1949 from a tributary of the Ouachita
River 0.3 mi. southeast of Vernon, Jackson Parish, Louisiana: snout,
branchiostegals, lower surface of head, and belly rosy, shading to
scarlet along sides of body; dorsal and caudal fins of breeding males
washed with yellowish, outer margin clear, bordered inside by a nar-
row band of crimson; distal half of pectoral, pelvic, and anal fins
brilliant scarlet, except for the edges which are clear; proximal half
of anal fin cream, distal half a brilliant scarlet except for the narrow
border which is colorless; proximal half of pectoral and pelvic fins
slaty gray, distal half scarlet, with the border colorless.
VARIATION. This form shows relatively little geographic variation.
Occasionally an unevenness is evident in the arrangement of the pre-
dorsal and anterior dorso-lateral scales, but this is slight and never
approaches the usual situation in N. c. chrysocephalus. Anal ray
counts vary to some extent, the extremes in this subspecies being the
greatest encountered in any form of Luxilus. A specimen with 7
anal rays (UMMZ 128116) was found from the Red River drainage of
Arkansas, and one with 12 anal rays (UMMZ 161396) from a tributary
of the Mississippi River in northern Mississippi; other specimens in
these series have normal counts. Occasional populations contain an
unusually high percentage of individuals with an anal ray count
other than 9, a condition also encountered in certain populations of
Notropis cornutus. Counts of 31 specimens of isolepis (UMMZ 157780)


1964







BULLETIN FLORIDA STATE MUSEUM


F ~,





Figure 15. Pigmentation patterns and arrangement of dorso-lateral scales in
Luxilus. A. Notropis chrysocephalus isolepis, UMMZ 161262, Red River drain-
age, Louisiana. B. N. c. chrysocephalus, UMMZ 100877, Wabash River drainage,
Indiana. C. N. albeolus, CU 20322, Roanoke River drainage, Virginia. D. N.
cornutus, UMMZ 9526, St. Lawrence River drainage, New York. E. N. cornutus,
UMMZ 174813, James River drainage, Virginia. F. N. albeolus, UMMZ 174749,
Roanoke River drainage, Virginia.


Vol. 8






GILBERT: FISHES OF THE SUBGENUS LUXILUS


from the Chickasawhay drainage in Mississippi show 14 with 9 anal
rays and 17 with 10.
Sexual variation is pronounced in N. c. isolepis, as in the closely
related N. cornutus and N. c. chrysocephalus. Differences presumably
are identical to those occurring in the latter form and very similar to
those in the former. Breeding males develop a deep red coloration
on body and fins, the black crescent-shaped markings on the side
of the body become notably darker and more pronounced, and prom-
inent nuptial tubercles develop on the head, back, anterior part of the
dorsal and upper surface of the pectoral fins. Females have a more
subdued coloration and usually lack tubercles. Adult males also de-
velop a deeper and more compressed head, have slightly longer pec-
toral and pelvic fins, and an increased amount of adipose tissue sur-
rounding the fin rays, but apparently do not differ in other respects
from the females (table 6).
RELATIONSHIPS. N. c. chrysocephalus is the only form of Luxilus
that occurs with N. c. isolepis in the Gulf drainage. Both are found
in the Coosa River system and, although they occur in adjacent tribu-
tary streams a few miles apart, they have not yet been found to-
gether. Specimens from one collection taken in an eastern tributary
of the neighboring Black Warrior system have been identified as in-
tergrades, and several other collections in the same drainage have
been identified as N. c. chrysocephalus. Perhaps transfer of chryso-
cephalus into the Black Warrior has been effected through a localized
stream capture. The apparent lack of intergradation of chryso-
cephalus and isolepis in the Coosa basin is difficult to explain, though
this may in some way be related to the fact that, in this area at least,
the ranges of the two forms appear to be separated by the so-called
Fall Line which marks the transition from an upland to a lowland
environment (Fenneman, 1938: 126-131). On the other hand these
forms may be incipient species which, because of ecological similari-
ties, have each prevented the other from moving into the areas it
occupies. Should this explanation be correct the intergradation in
the Black Warrior drainage would be difficult to explain.
DISTIBUTION. Confined to tributaries of the Gulf of Mexico from
the eastern half of the Red River system (including the Ouachita
River) in Texas, Louisiana, Arkansas and Oklahoma, eastward to the
Tallapoosa drainage in western Georgia (UF 9563).
Its range west of the Mississippi River is, with one exception, re-
stricted to the. Red River system. A single record from the Arkansas
drainage in Arkansas (USNM 165878) is interpreted as the result of
either human introduction or localized stream capture. This form


1964







172 BULLETIN FLORIDA STATE MUSEUM Vol. 8

apparently is fairly common in favorable habitats in southern Ar-
kansas, extreme southeastern Oklahoma, and extreme northeastern
Texas, but farther west is known only from the Blue River system of
south-central Oklahoma. To the east it occurs in a number of smaller
tributaries of the Mississippi River in western Mississippi, and re-
cently has been recorded as far north as southwestern Tennessee
(USNM 179780-179781). It probably does not normally occur in any
independent river system east of Mobile Bay, although there are un-
published records (not verified by me) from the Escambia and Choc-
tawhatchee drainages in Alabama. In the Coosa River system isolepis
occurs as far north as Chilton County, Alabama; here its range dove-
tails with that of N. c. chrysocephalus. It is one of the most common
fishes in upland tributaries to the Gulf of Mexico.
LIFE HISTORY AND ECOLOGY. No studies dealing with the life
history and ecology of N. c. isolepis have been made. The form
probably differs little, if at all, from N. c. chrysocephalus in these
regards.








GILBERT: FISHES OF THE SUBGENUS LUXILUS


TABLE 1

Comparison of Notropis coccogenis and Notropis zonistius

Character Notropis coccogenis Notropis zonistius


Position of band
on dorsal fin



Color of dorsal
band

Caudal spot

Black border on
distal margin of
caudal fin

Red bar on side
of head


Nuptial tubercles
on top of head
and back

Body form



Angle of mouth to
top of head

Number of
vertebrae

Maximum length
of males


Confined to posterior part
of fin; almost parallels distal
margin


Black at all ages


Absent

Present



Prominent; situated at mar-
gin of opercle and pre-
opercle
Absent


Attenuate and terete



Usually 49 to 538


40 to 42; usually 41


Not confined to posterior
part of fin; slopes obliquely
forward and intersects lower
third of anterior margin

Orange-red in young; black
in adults

Present

Absent



Not prominent (often ab-
sent); situated on cheek


Present


Deep and compressed



Usually 57 to 61"


38 to 40; usually 89


Ca. 115 mm.


Ca. 85 mm.









BULLETIN FLORIDA STATE MUSEUM


TABLE 2

Comparison of pigmentary characters in
Notropis zonatus and Notropis pilsbryi

Character Notropis zonatus Notropis pilsbryi

Pigment on posterior- Not bordering posterior- Bordering posterior-
dorsal part of opercle dorsal margin dorsal margin
Pigment on cleithral Extending ventrally to Extending ventrally about
area pectoral insertion; be- halfway to pectoral in-
coming quite dark sertion; not becoming
particularly dark

Relation of lateral stripe Not extending below Extending below lateral
to lateral line (anterior lateral line line
to caudal peduncle)
Distribution of pigment Unequally distributed; Equally distributed
on caudal peduncle more present above lat- above and below lateral
eral line than below line

Dorso-lateral stripes Present Absent
on back
Red pigmentation on Usually more Usually less pronounced
body and fins pronounced
Black pigmentation More intense; appearing Less intense; appearing
on body deep black more dusky


Vol. 8





TABLE 3
Comparison between forms of the Notropis cornutus species group


Character


Anterior dorso-lateral
scales
Circumferential scales
Sum of anterior dorso-
lateral and circumfer-
ential scales
Caudal peduncle
scales
Parallel stripes on
anterior dorso-lateral
part of back
Crescent-shaped bars
on side of body
Reds on body and fins
of breeding males

Color of dorsum
Nuptial tubercles
present in:
Appearance of nuptial
tubercles on upper
part of head
Body form


N. cornutus


(16) 18 to 24 (30)

(26) 30 to 35 (39)
(42) 48 to 59 (67)


(13) 15 or 16 (18);
usually 16
Absent


Present

Present and
widespread

Olive-brown
Males only

Straight at tips


Deep and
compressed


Maximum size of males Ca. 175 mm.


N. albeolus


(15) 17 to 20 (23)

(24) 26 to 30 (33)
(41) 43 to 50 (55)


(13) 15 or 16 (17)

Absent


Usually absent

Usually absent;
occasional on snout
and fins
Olive-green
Males only

Straight at tips


Usually slender and
terete in north-
ern race; usually
deep and compressed
in southern race
Ca. 130 mm.


N. c. chryso-
cephalus


(13) 14 to 17 (19)

(23) 26 to 29 (32)
(38) 40 to 44 (48)


(12) 15 or 16 (17);
usually 16
Present;
crooked


Present


Present and
widespread

Olive-brown
Males only

Straight at tips


Deep and
compressed



Ca. 175 mm.


N. c. isolepis


(12) 13 or 14 (15)

(23) 24 to 28
(36) 37 to 41 (43)


(14) 15 or 16;
usually 16
Present;
straight


Present


Present and
widespread

Olive-brown
Males only

Straight at tips


Deep and
compressed


Ca. 175 mm. Ca. 90 mm.


N. cerasinus

(14) 15 to 19 (22)

(23) 25 to 28 (30)
(37) 41 to 47 (51)


(12) 13 to 15 (16);
usually 14
Absent


Present

Present and
widespread

Olive-brown
Both sexes

Curved forward
at tips

Deep and
compressed








BULLETIN FLORIDA STATE MUSEUM


TABLE 4. COMPARISON OF PROPORTIONAL MEASUREMENTS
(EXPRESSED IN THOUSANDTHS OF STANDARD LENGTH)
FOR Notropis coccogenis AND Notropis zonistius


Notropis coccogenis Notropis zonistius
Sex No. Range Mean Sex No. Range Mean


Standard length

Length of pectoral
fin

Length of pelvic fin


Length of head
Depth of head
Width of head
Interorbital width
Mouth gape
Diameter of eye
Length of snout
Length of upper jaw
Distance from mid-
dorsum to lateral
line
Depth of body
Width of body
Length of caudal
peduncle
Depth of caudal
peduncle
Predorsal length
Prepelvic length
Preanal length
Length of dorsal fin
Length of anal fin
Length of caudal fin
Angle of mouth to
top of head


61-84 mm.

M 19 182-201 193
F 21 177-210 188
* 66 174-210 190
M 19 139-161 153
F 21 134-169 154
* 66 134-169 153
* 35 241-281 263
* 35 153-190 163
* 35 128-146 136
* 35 82-95 88
S 35 43-59 52
* 35 66-81 74
* 35 66-85 76
* 35 96-115 106
* 35 131-168 149


* 40 192-244 215
* 35 119-159 134
* 35 181-224 204

* 35 85-102 95

S 35 497-536 520
a 35 472-521 503
* 35 659-704 665
* 35 200-233 216
* 35 174-208 188
* 29 266-300 279

* 58 46-54 50.5"


57.5-75 mm.


M 8
F 6
* 37
M 8
F 6
* 37
* 24
* 24
* 24
* 24
* 24
* 24
* 24
* 24
* 28


* 24
* 24
* 24


197-224 204
193-218 200
187-234 204
159-186 169
159-168 164
156-186 166
259-285 271
177-204 189
136-161 152
92-104 96
57-73 67
68-82 74
67-79 73
93-112 103
162-203 180


217-284 256
140-168 153
173-231 206


* 24 104-119 109


* 24
* 24
* 24
* 24
* 24
* 22


494-528 516
494-515 502
655-684 675
220-264 239
187-242 208
279-315 296


* 40 50-61.5 58


* Summation of sexed males and females and additional unsexed adult specimens.


Vol. 8






GILBERT: FISHES OF THE SUBGENUS LUXILUS


TABLE 5. COMPARISON OF PROPORTIONAL MEASUREMENTS
(EXPRESSED IN THOUSANDTHS OF STANDARD LENGTH)
FOR Notropis zonatus AND Notropis pilsbryi


Standard length

Length of pectoral
fin

Length of pelvic fin


Length of head
Depth of head
Width of head
Interorbital width
Mouth gape
Diameter of eye
Length of snout
Length of upper jaw
Distance from mid-
dorsum to lateral
line
Depth of body
Width of body
Length of caudal
peduncle
Depth of caudal
peduncle
Predorsal length
Prepelvic length
Preanal length
Length of dorsal fin
Length of anal fin
Length of caudal fin
Angle of mouth to
top of head


Notropis zonatus
Sex No. Range Mean

54-76 mm.
M 13 200-225 208
F 13 183-216 208
40 183-225 205
M 13 160-182 171
F 13 158-179 171
40 153-182 170
38 259-284 269
0 51 161-186 174
32 116-142 133
32 80-89 84
32 47-63 54
32 70-82 77
40 70-80 74
32 80-94 88
32 181-158 146


51 199-260 229
32 115-145 131
82 192-245 229

40 86-108 96

32 460-505 488
32 464-508 487
32 643-674 662
32 200-255 240
32 178-208 198
32 240-315 290

39 46.5-56 51,


Notropis pilsbryi
Sex No. Range Mean


55-74 mm.
M 14 190-216
F 13 184-210
* 42 181-216
M 14 159-196
F 13 151-174
* 42 146-196
* 36 251-284
* 49 154-183
* 34 129-136
* 34 75-86
* 34 50-56
* 34 71-82
* 43 65-81
* 84 78-90
* 34 133-154


49 174-273
34 126-157
36 199-239


* 40 85-111


34 470-504
34 475-497
34 659-683
34 226-253
34 175-208
34 262-321


* 40 46-56'


203
198
199
168
166
167
265
168
184
80
53
78
71
88
140


221
135
220

95

493
487
672
236
194
290

50.5*


* Summation of sexed males and females and additional unsexed, adult specimens.


1964








TABLE 6. COMPARISON OF PROPORTIONAL MEASUREMENTS (EXPRESSED IN THOUSANDTHS OF STANDARD LENGTH)
BETWEEN MEMBERS OF THE Notropis cornutus SPECIES GROUP


N. cornutus
Sex No. Range Mean


N. albeolus
(southern race)
Sex No. Range Mean


N. albeolus
(northern race)
Sex No. Range Mean


N. chrysocephalus N. chrysocephalus
chrysocephalus isolepis
Sex No. Range Mean Sex No. Range Mean


N. cerasimus
Sex No. Range Mean


Standard length
(in mm.)
Length of pectoral fin


Length of pelvic fin


Length of head
Depth of head
Width of head
Interorbital width
Mouth gape
Diameter of eye
Length of snout
Length of upper jaw
Distance from mid-
dorsum to lateral
line
Depth of body
Width of body
Length of caudal
peduncle
Depth of caudal
peduncle
Predorsal length
Prepelvic length
Preanal length
Length of dorsal fin
Length of anal fin
Length of caudal fin
Angle of mouth to ton
of head (in de-
grees)


60-82

170-236
176-220
164-240
157-206
152-185
144-211
225-291
155-215
119-155
75-106
43-78
61-86
59-79
73-96
135-193


* 87 210-295 260
* 42 122-170 145
* 75 171-257 222

* 79 82-120 101

* 36 471-554 496
* 32 457-508 486
* 31 647-693 665
* 57 201-280 241
* 50 164-221 197
* 52 238-339 299
* 67 53-63.5 57.7


58-81

183-231
182-223
170-231
160-192
155-187
145-192
232-287
160-202
126-155
74-95
50-80
60-94
60-83
70-101
155-207


* 51 235-316 269
* 15 130-161 141
* 52 182-238 206

* 48 95-116 105

* 13 464-517 487
* 5 460-485 475
* 5 655-675 670
* 48 205-283 246
* 40 170-232 203
* 30 265-337 306
* 30 51.5-61.5 57.3


59-80

166-211
163-198
159-213
158-188
155-181
147-192
230-285
147-200
109-140
72-104
31-61
60-87
59-76
68-95
126-185


* 74 189-256 230
* 32 113-158 128
* 61 195-241 214

* 62 79-114 96

* 29 465-511 487
* 12 445-480 464
* 14 636-695 658
* 32 196-266 238
* 22 160-220 192
* 25 248-320 286
* 41 47-58 53.1


60-84

176-230
174-218
172-237
149-197
151-190
142-206
251-286
151-197
124-163
90-103
46-70
65-94
66-82
84-107
137-196


* 67 213-305 263
* 45 109-139 126
* 62 172-244 217

* 62 90-117 100

* 30 464-515 494
* 30 464-509 495
* 30 640-718 692
* 62 205-269 249
* 62 159-215 194
* 62 248-342 295
* 36 53.5-65 59.0


60-80

174-234
177-225
175-236
158-198
150-191
150-199
249-292
167-197
130-153
86-104
47-71
65-90
63-86
83-105
143-199


* 58 225-309 261
* 46 116-171 133
* 57 168-244 211

* 59 94-123 106

* 31 450-512 294
* 27 453-519 493
* 27 633-709 688
* 54 222-278 254
* 49 174-214 198
* 49 282-364 302
* 33 52.5-64 58.0


53-78

193-233
187-223
187-233
164-204
184-204
158-204
238-292
169-205
129-169
77-101
45-77
63-89
54-82
73-102
150-204


* 41 238-330 274
* 33 118-175 143
* 42 190-250 216

* 43 97-123 110

* 41 456-516 493
* 41 448-520 485
* 41 652-708 672
* 44 240-287 266
* 41 171-244 208
* 35 274-331 299
* 27 52.5-62.5 57.


* Summation of sexed males and females and additional unsexed adult specimens.





TABLE 7. COMPARISON OF IMPORTANT AREAS OF DIFFERENTIATION FOR SUM OF ANTERIOR DORSO-LATERAL AND CIRCUMFERENTIAL SCALE COUNTS IN
Notropis cornutus and Notropis albeolus

Anterior dorso-lateral scales
Area 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 No. Mean

albeolus
New & Roanoke drs. 1 2 38 55 48 21 7 3 175 18.46
Tar, Neuse & Cape Fear drs. 2 6 39 63 45 15 6 1 177 18.28
cormutus
Atlantic coast* 3 29 84 137 159 139 103 43 34 13 2 1 1 748 20.40
Maine-Maritime dr. 1 9 16 35 34 34 14 6 4 3 2 158 20.16
St. Lawrence dr. 2 10 14 17 20 20 7 10 1 1 102 20.90
upper Susquehanna dr. 3 17 16 30 15 9 5 3 1 1 100 23.12
west of Alleghenies 5 27 100 211 287 333 213 142 74 28 12 4 1 1 1438 21.85


Circumferential scales
Area 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 No. Mean
albeolus
New & Roanoke drs. 2 12 31 47 39 36 7 1 175 28.43
Tar, Neuse & Cape Fear drs. 1 5 31 54 32 35 15 3 1 177 27.68
cornutus
Atlantic coast* 5 8 46 126 175 179 127 60 17 4 1 748 31.60
Maine-Maritime dr. 1 4 27 54 34 24 8 5 1 158 30.59
St. Lawrence dr. 18 18 20 23 9 11 2 1 102 32.32
upper Susquehanna dr. 13 13 21 24 13 12 1 1 2 100 32.70
west of Alleghenies 8 69 165 279 348 255 178 88 33 7 8 1438 33.17

(Continued)








TABLE 7 (continued)

Sum of above counts over
Area 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 66 60 No. Mean

albeolus
New & Roanoke drs. 2 4 15 22 36 28 30 11 19 3 1 2 1 1 175 47.01
Tar, Neuse &
Cape Fear drs. 2 4 13 26 26 39 34 18 11 2 2 177 45.88
cornutus
Atlantic coast* 1 4 20 56 53 83 106 125 102 73 53 23 26 12 7 3 1 748 52.00
Maine-Maritime dr. 1 1 17 18 21 20 25 18 11 12 3 4 3 2 1 1 158 50.74
St. Lawrence dr. 3 5 12 14 12 9 12 8 11 12 2 1 1 102 53.22
upper Susquehanna dr. 1 4 4 7 11 23 16 15 4 7 2 6 100 55.82
west of Alleghenies 1 5 13 35 73 120 187 209 224 160 149 123 57 35 47 1438 55.02
*south of Maine


TABLE 8. COMPARISON OF VERTEBRAL COUNTS IN Notropis albeolus
BY DRAINAGE SYSTEMS

Vertebrae

Drainage 37 38 39 40 41 No. Mean

Neuse 7 31 8 46 39.02
Tar 1 9 7 17 39.29
Cape Fear 1 1 39.00
Roanoke 4 23 6 33 40.06
New 1 16 3 20 40.10







TABLE 9. COMPARISON OF ANTERIOR DORSO-LATERAL SCALES IN THE FORMS OF Luxilus

Anterior dor-o-lateral scales
Form 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 No. Mean

N. cornutus 4 45 137 286 424 497 486 299 205 99 38 16 6 3 1 2546 21.33
N. albeolus 8 77 118 93 36 13 3 1 352 18.84
N. c. chrysocephalus 47 352 430 166 26 1 1 1023 14.78
N. c. isolepis 2 150 155 23 880 13.60
N. cerasinus 16 35 60 32 15 8 5 1 175 17.20
N. zonatus 33 -94 30 4 1 '-- 162 15.05
N. pilsbryi 6 60 81 23 7 177 15.80
N. coccogenis -.18 57 58 27 4- 157 16.64
N. zonistius 26 35 27 8 4 100 16.29


TABLE 10. COMPARISON OF CIRCUMFERENTIAL SCALES IN THE FORMS OF Luxilus

Circumferential scales
Form 2-3" 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 No. Mean

N. cornutus 1 5 12 81 280 405 523, 530 342 218 96 36 7 10 2546 32.50
N. albeolus 1 7 43 85 79 74 51 10 1 1 352 28.05
N. c. chrysocephalus 1 6 35 159 310 316 134 47 13 2 1023 27.52
N. c. isolepis 7 102 81 64 42 84 -- 330 25.87
N. cerasinUs 1 10 32 65 48 13 5 1 175 26.22
N. zonatus 13 45 51 42 10 1 162 26.96
N. pilsbryi 1 14' 33 54 44 -21 10 177 27.29
N. coccogenis 19 60 26 35 16 2 1 159 26.87
N. zonistius 1 1 10 43 27 18 100 28.48


















TABLE 11. COMPARISON OF SUM OF ANTERIOR DORSO-LATERAL AND CIRCUMFERENTIAL SCALES IN THE FORMS OF Luxilus



Sum. of anterior dorso-lateral and circumferential scale counts

Form 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 No. Mean


N. cornutus

N. albeolus

N. chrysocephalus
chrysocephalus ,

N. chrysocephalus isolepis

N. cerasinus

N. zonatus

N. pilsbryi

N. coccogenis

N. zonistius


1 1 5 38 82 92 151 222 279 316 317 311 214 205 143 73 41 27 16 4 2 4 1 1 2546 53.83

2 6 17 41 48 75 62 48 22 21 5 1 2 11 352 46.44


4 24117183 241 237 125 51 28 11 2 -

6 57 80 71 51 49 14 2 -

- 1 3 7 18 29 48 21 20 11 10 2 2 2 1 -

4 25 31 39 44 9 7' 3 -

2 9 25 34 32 36 27 4 7 1 -

4 20 32 32 20 22 20 7 1 1 -

1 1 4 13 31 23 11 6 8 2 -


- 1023 42.50

- 0 38.96

- 175 43.42

-- 162 42.01

- 177 43.09

159 43.51

- 100 44.77


- - - ---

- - - -

- - - -

- -- --- ---








GILBERT: FISHES OF THE SUBGENUS LUXILUS


TABLE 12. COMPARISON OF LATERAL-LINE SCALES IN THE FORMS OF LuxilUS

Lateral-line scales
Form 36 37 38 39 40 41 42 43 No. Mean

N. cornutus 1 24 165 326 234 53 5 2 810 39.18
N. albeolus 22 66 80 28 14 210 38.69
N. chrysocephalus
chrysocephalus 2 90 255 249 69 14 1 680 38.50
N. chysocephalus
isolepis 39 80 46 7 172 38.12
N. cerasinus 4 27 21 6 58 39.48
N. zonatus 9 18 31 19 2 4 83 39.99
N. pilsbryi 1 13 18 21 8 61 40.36
N. coccogenis 10 28 26 12 5 81 40.68
N. zonistius 6 17 33 16 12 2 1 87 39.24


TABLE 13.' COMPARISON OF ANAL RAYS IN THE FORMS OF LuxiluS

Anal rays
Form 7 8 9 10 11 12 No. Mean

N. cornutus 100 749 20 869 8.91
N. albeolus 83 4 1 88 9.07
N. chrysocephalus chrysocephalus 49 530 10 1 590 8.94
N. chrysocephalus isolepis 1 7 146 26* 1 181 9.11
N. cerasinus 2 70 72 8.97
N. zonatus 4 101 5 110 9.01
N. pilsbryi 3 67 4 74 9.01
N. coccogenis 4 98 13 115 9.08
N. zonistius** 5 110 88 203 9.41

17 specimens in one collection
** Of 62 specimens from Florida, 53 had ten anal rays.

TABLE 14. COMPARISON OF VERTEBRAE IN THE FORMS OF Luxilus

Vertebrae
Form 36 37 38 39 40 41 42 No. Mean

N. cornutus 2 2 36 136 55 6 237 40.08
N. albeolus 1 7 46 54 9 53 39.54
N. chrysocephalus
chrysocephalus 6 53 48 7 114 39.49
N. chrysocephalus isolepis 2 35 10 1 1 49 39.27
. cerasinus 3 22 8 33 39.15
.zonatus 4 22 2 28 39.93
pilsbryi 6 16 1 23 39.78
coccogenis 4 36 10 50 41.12
V. zonistius 5 32 9 46 39.09













TABLE 15. COMPARISON OF ANGLE OF SNOUT TO TOP OF HEAD
IN THE FORMS OF Luxilus


46' 48* 50 52 54 56 58 60 62*
Form I I I I I I I I I No. Mean
47.5* 49.5* 51.5* 53.5* 55.50 57.5" 59.5 61.5 63.5*

N. cornutus 10 12 15 11 10 9 67 57.65"
N. albeolus (southern race) 1 4 2 11 9 3 30 57.33"
N. albeolus (northern race) 1 5 9 11 6 8 1 41 53.13"
N. chrysocephalus chrysocephalus 5 6 8 7 6 4 36 59.06
N. chrysocephalus isolepis 3 5 7 7 6 5 33 57.85*
N. cerasinus 3 8 4 8 2 2 27 57.41"
N. zonatus 5 8 9 6 8 2 38 51.210
N. pilsbryi 5 10 8 8 4 1 36 50.70*
N. coccogenis 5 11 26 15 1 58 50.50*
N. zonistius 1 2 2 6 10 7 28 57.94*


w
td


0
r
m

z
r
0
w
B
^





c3








GILBERT: FISHES OF THE SUBGENUS LUXILUS


REFERENCES CITED

Abbott, Charles C.
1870. Notes on fresh-water fishes of New Jersey. Am. Naturalist, 4(2): 99-117.
1874. Notes on the cyprinoids of central New Jersey. Am. Naturalist, 8(6):
326-338.
Adams, Charles C., and Thomas L. Hankinson
1928. The ecology and economics of Oneida Lake fish. Roosevelt Wild Life
Ann., 1(3-4): 239-548, figs. 175-244, 4 pls., 1 map.
Agassiz, Louis
1850. Lake Superior: Its physical character, vegetation, and animals, compared
with those of other and similar regions .. with a narrative of the tour
by J. Elliot Cabot and contributions by other scientific gentlemen. Gould,
Kendall and Lincoln, Boston, xii + 48 pp., 8 pls., many figs.
1854. Notice on a collection of fishes from the southern bend of the Tennessee
River, Alabama. Am. Jour. Sci. and Arts, 17(2): 297-308 and 353-365.
Bailey, Joseph R., and James A. Oliver
1939. The fishes of the' Connecticut watershed. New Hampshire Fish and
Game Dept., Surv. Rept., 4: 150-189, figs. 57-80.
Bailey, Reeve M.
1938. The fishes of the Merrimack drainage. In: Biological survey of the Mer-
rimack watershed. New Hampshire Fish and Game Dept., Surv. Rept.,
3: 149-185, 12 figs.
1945. Review of: Some considerations on the distribution of fishes in Ontario.
By: Isobel Radforth. Copeia, 1945(2): 125-126.
Bailey, Reeve M., and William A. Gosline
1955. Variation and systematic significance of vertebral counts in the American
percid fishes of the family Percidae. Misc. Publ. Mus. Zool., Univ.
Mich., 93: 1-44.
Bean, Barton A.
1903. Notice on a collection of fishes made by H. H. Brimley in Cane River
and Bollings Creek, North Carolina, with a description of a new species
of Notropis (N. brimleyi). Proc. U. S. Nat. Mus., 26(1339): 918-914.

Beckman, William C.
1952. Guide to the fishes of Colorado. Univ. Colorado Mus., Leaflet 11: 1-
110, many figs.

Call, B. Ellsworth
1896. Fishes and shells of the Falls of the Ohio. Mem. Hist. Louisville, 1:
9-20.

Cockerell, T. D. A.
1908. The fishes of the Rocky Mountain region. Univ. Colorado Studies,
5(3): 159-178.


1964








BULLETIN FLORIDA STATE MUSEUM


Cockerell, T. D. A., and Otis Callaway
1909. Observations on the fishes of the genus Notropis. Proc. Biol. Soc.
Wash., 22: 189-196.
Cope, Edward Drinker
1864. Partial catalogue of the cold-blooded vertebrata of Michigan, Pt. I.
Proc. Acad. Nat. Sci. Phila., 16: 276-285.
1867. On the genera of fresh-water fishes Hypsilepis Baird and Photogenis
Cope, their species and distribution. Proc. Acad. Nat. Sci. Phila., 19:
158-166.
1869a. On the distribution of fresh-water fish in the Allegheny region of south-
western Virginia. Jour. Acad. Nat. Sci. Phila., Ser. 2, 6(5): 207-247,
pls. 22-24.
1869b. Synopsis of the Cyprinidae of Pennsylvania. Trans. Am. Phil. Soc., 13:
351-399, pls. 10-13.
1870. (reprinted in 1877). A partial synopsis of the fishes of the fresh waters
of North Carolina. Trans. Am. Phil. Soc., 14: 448-495.
1871. Recent reptiles and fishes. Report on the reptiles and fishes obtained by
the naturalists of the expedition. In: U. S. Geological Survey of Wyom-
ing and Continuous territory (1870), pp. 432-442.
Cross, Frank Bernard
1954. Fishes of Cedar Creek and the South Fork of the Cottonwood River,
Chase County, Kansas. Trans. Kans. Acad. Sci., 57(3): 303-314.
DeKay, James E.
1842. New York fauna. Part 1, Zoology; part 4, Fishes. Albany, 415 pp.,
79 pls.
Driver, Ernest C.
1942. Name that animal. Kraushar Press, Northampton, Mass., 527 pp.,
many pls.
Ellis, Max W.
1914. The fishes of Colorado. Univ. Colorado Stud., 11(1): 1-136, 12 pls.

Evermann, Barton W., and Theodore D. A. Cockerell
1909. Descriptions of three new species of Cyprinoid fishes. Proc. Biol. Soc.
Wash., 22: 185-188.
Evermann, Barton W., and Ulysses O. Cox
1896. The fishes of the Neuse River basin. Bull. U. S. Fish Comm. (1895),
15: 303-310.

Evermann, Barton W., and Edmund Lee Goldsborough
1907. A checklist of the freshwater fishes of Canada. Proc. Biol. Soc. Wash.,
20: 89-119.
Evermann, Barton W., and Samuel F. Hildebrand
1916. Notes on the fishes of east Tennessee. Bull. U. S. Bur. Fisheries (1914),
34: 431-457.


Vol, 8








GILBERT: FISHES OF THE SUBGENUS LUXILUS


Fenneman, Nevin M.
1938. Physiography of eastern United States. McGraw-Hill Book Co., New
York, v-xiii + 714 pp. and 196 figs.

Flint, Richard Foster
1947. Glacial geology and the Pleistocene Epoch. John Wiley and Sons,
New York, 589 pp., 88 figs., 30 tables, 6 pls.
1957. Glacial and Pleistocene geology. John Wiley and Sons, New York,
v-xiii + 553 pp., many figs. and tables, 5 pls.

Forbes, Stephen A.
1909. On the general and interior distribution of Illinois fishes. Bull. Illinois
St. Lab. Nat. Hist., 8: 381-437, maps 1-103.

Forbes, Stephen A., and Robert E. Richardson
1920. The fishes of Illinois. Second edition. Ill. St. Jour. Co., Springfield,
vol. 3, cxxxi + 357 pp., many figs. and color pls., separate vol. of 103
maps.

Fowler, Henry W.
1904. Notes on fishes from Arkansas, Indian Territory and Texas. Proc. Acad.
Nat. Sci. Phila., 56: 242-249, 1 fig.
1909. A synopsis of the Cyprinidae of Pennsylvania. Proc. Acad. Nat. Sci.
Phila. (1908), 60: 517-553.
1910. Notes on the variation of some species of the genus Notropis. Proc.
Acad. Nat. Sci. Phila., 62: 273-293, pls. 15-21.
1922. Records of fishes for the southern and eastern United States. Proc.
Acad. Nat. Sci. Phila., 74: 1-27, 2 pls.
1923. Records of fishes for the southern states. Proc. Biol. Soc. Wash., 36:
7-34.
1936. Fresh-water fishes obtained in North Carolina in 1930 and 1934. Fish
Culturist, 15(10): 192-194.
1945. A study of the fishes of the southern Piedmont and coastal plain. Monog.
Acad. Nat. Sci. Phila., 7: i-vi, 1-408, figs. 1-313.

Gibbs, Robert H., Jr.
1957. Cyprinid fishes of the subgenus Cyprinella of Notropis. I. Systematic
status of the subgenus Cyprinella, with a key to the species exclusive
of the lutrensis-ornatus complex. Copeia, 1957(3): 185-195.

Gilbert, Carter Rowell
1961a. A systematic and geographical study of the subgenus Luxilus (family
Cyprinidae). Dissertation Abstracts, 21(8): 2411-2412.
1961b. Hybridization versus intergradation: an inquiry into the relationship of
two cyprinid fishes. Copeia, 1961(2): 181-192.

Gilbert, Charles H.
1884a. Notes on the fishes of Kansas. Bull. Washburn Coll. Lab. Nat. Hist.,
1(1): 10-16.


1964








BULLETIN FLORIDA STATE MUSEUM


1884b. A list of fishes collected in the East Fork of White River, Indiana, with
descriptions of two new species. Proc. U. S. Nat. Mus., 7(423): 199-205.
1891. Notes on explorations made in Alabama during 1889, with notes on the
fishes of the Tennessee, Alabama, and Escambia rivers. Bull. U. S.
Fish Comm. (1889), 9: 143-159, pl. 43.

Girard, Charles
1856. Researches upon the Cyprinoid fishes inhabiting the fresh waters of
the United States, west of the Mississippi Valley from specimens in the
Museum of the Smithsonian Institution. Proc. Acad. Nat. Sci. Phila.,
8: 165-213.

Goldsborough, Edmund Lee, and H. Walton Clark
1908. Fishes of West Virginia. Bull. U. S. Bur. Fisheries (1907), 27: 29-39,
1 fig.

Greeley, John R.
1938. Fishes of the area with annotated list. In: A biological survey of the
Allegheny and Chemung watersheds. Suppl. 27th Ann. Rept. N. Y.
St. Cons. Dept. (1937): 48-73, col. pls. 1-2, 3 figs.

Greene, C. Willard
1935. The distribution of Wisconsin fishes. Wisconsin St. Cons. Comm.,
Madison, 235 pp., 96 maps.

Giinther, Albert
1868. Catalogue of the Physostomi Catalogue of the fishes in the British
Museum. Taylor and Francis, London, vol. 7, xx + 512 pp.

Hankinson, Thomas L.
1932. Observations on the breeding behavior and habits of fishes in southern
Michigan. Papers Mich. Acad. Sci. (1931), 14: 411-425.

Heckel, Joh. Jakob
1843. Abbildungen und Beschreibungen der Fische Syriens, nebst einer neuen
Classification und Characteristik Sammtlicher Gatlingen der Cyprinen.
Stuttgart, 258 pp., 23 pls.

Hildebrand, Samuel F.
1932. On a collection of fishes from the Tuckaseegee and upper Catawba river
basins, N. C., with a description of a new darter. Jour. Elisha Mitchell
Sci. Soc., 48(1): 50-82, pl. 3, figs. 1-2.

Hubbs, Carl L.
1926. A check-list of the fishes of the Great Lakes and tributary waters, with
nomenclatorial notes and analytical keys. Misc. Publ. Mus. Zool., Univ.
Mich., 15: 1-77, pls. 1-4.
1955. Hybridization between fish species in nature. Systematic Zool., 4(1):
1-20.


Vol. 8.








GILBERT: FISHES OF THE SUBGENUS LUXILUS


Hubbs, Carl L., and Dugald E. S. Brown
1929. Materials for a distributional study of Ontario fishes. Trans. Roy. Can.
Inst. (1929), 17: 1-56.
Hubbs, Carl L., and Karl F. Lagler
1947. Fishes of the Great Lakes Region. Bull. Cranbrook Inst. Sci., 26: i-xi,
1-109, many figs.
1958. Fishes of the Great Lakes Region. Bull. Cranbrook Inst. Sci., 26: i-xi,
1-213, many figs. 44 color pls. (revised edition).
Hubbs, Carl L., and George A. Moore
1940. The subspecies of Notropis zonatus, a cyprinid fish of the Ozark upland.
Copeia, 1940(2): 91-99.
Hubbs, Carl L., and A. I. Ortenburger
1929. Fishes collected in Oklahoma and Arkansas in 1927. Publ. Univ. Okla.
Biol. Surv., 1(3): 45-112, pls. 6-13.

Hubbs, Carl L., and Edward C. Raney
1951 Status, subspecies, and variation of Notropis cummingsae, a cyprinid
fish of the southeastern United States. Occ. Papers Mus. Zool., Univ.
Mich., 585: 1-25, 1 pl.

Hubbs, Clark
1951. Records from east Texas of three species of fish, Semotilus atromacu-
latus, Notropis cornuta, and Microperca proelearis. Texas Jour. Sci.,
3(3): 490.

Jennings, Dolph
1942. Kansas fish in the Kansas State College Museum at Manhattan. Trans.
Kans. Acad. Sci., 45: 863-366.

Jones, David T.
1929. Notes on fishes taken in the vicinity of Vinton, Iowa. Copeia, 171:
29-34.

Jordan, David Starr
1875. Synopsis of the genera of fishes to be looked for in Indiana. Ann. Rept.
Geol. Surv. Indiana (1874): 197-228.
1876a. Concerning the fishes of the Ichthyologia Ohiensis. Bull. Buffalo Soc.
Nat. Sci., 3: 91-97.
1876b. Manual of the vertebrates of the northern United States 1st ed.,
Jansen, McClurg and Co., Chicago, 342 pp.
1877. Contributions to North American Ichthyology. .. I. Review of Rafin-
esque's memoirs on North American fishes. Bull. U. S. Nat. Mus.,
9: 1-53.
1878. A catalogue of the fishes of the fresh waters of North America. Bull.
U. S. Geol. and Geog. Surv. Terr., 4: 407-442.
1880. Description of new species of North American fishes. Proc. U. S. Nat.
Mus. (1879), 2(84): 235-241.


1964








BULLETIN FLORIDA STATE MUSEUM


1882. Report on the fishes of Ohio. Section IV. Rept. Geol. Surv. Ohio,
4 (Zool. and Bot., Pt. I), pp. 737-1002.
1884. Manual of the vertebrates of the northern United States 4th ed.,
Jansen, McClurg and Co., Chicago, 406 pp.
1885a. Identification of the species of Cyprinidae and Catostomidae, described
by Dr. Charles Girard, in the Proceedings of the Academy of Natural
Sciences of Philadelphia for 1856. Proc. U. S. Nat. Mus., 8(500): 118-
127.
1885b. A catalogue of the fishes known to inhabit the waters of North America,
north of the Tropic of Cancer, with notes on the species discovered in
1883 and 1884. Ann. Rept. Comm. Fish and Fisheries (1885): 787-973.
1888. A manual of the vertebrate animals of the northern United States .
5th ed., A. C. McClurg and Co., Chicago, iii + 375 pp.
1889a. Descriptions of fourteen species of fresh-water fishes collected by the
U. S. Fish Commission in the summer of 1888. Proc. U. S. Nat. Mus.
(1888), 11(723): 351-362, pls. 43-45.
1889b. Report of explorations made during the summer and autumn of 1888,
in the Allegheny region of Virginia, North Carolina and Tennessee, and
in western Indiana, with an account of the fishes found in each of the
river basins of those regions. Bull. U. S. Fish Comm. (1888), 8: 97-
173, pls. 13-15.
1916. The nomenclature of American fishes as affected by the opinions of the
International Commission on Zoological Nomenclature. Copeia, 29:
25-28.
1929. Manual of the vertebrate animals of the northeastern United States in-
clusive of marine species 13th ed., World Book Co., Yonkers-on-
Hudson, N. Y., xxx + 446 pp., 1 map, 15 figs.
Jordan, David Starr, and Alembert W. Zrayton
1878. Contributions to North American ichthyology. Based primarily on the
collections of the United States National Museum. III. A. on the
distribution of the fishes of the Allegheny region of South Carolina,
Georgia, and Tennessee, with descriptions of new and little known
species. Bull. U. S. Nat. Mus., 12: 7-95.
Jordan, David Starr, and Herbert E. Copeland
1876. Check list of the fishes of the fresh waters of North America. Bull.
Buffalo Soc. Nat. Sci., 3: 133-164.
Jordan, David Starr, and Barton Warren Evermann
1896a. A check-list of the fishes and fish-like vertebrates of North and Middle
America. Rept. U. S. Comm. Fish and Fisheries (1895): 207-584.
1896b. The fishes of North and Middle America Bull. U. S. Nat. Mus.,
47(1): i-lx, 1-1240.

Jordan, David Starr, and Howard Walton Clark
1930. Check list of the fishes and fishlike vertebrates of North and Middle
America, north of the northern boundary of Venezuela and Colombia.
Rept. U. S. Fish Comm. (1928), 2: 1-670.


Vol. 8,








GILBERT: FISHES OF THE SUBGENUS LUXILUS


Jordan, David S., and Charles H. Gilbert
1877. On the genera of North American fresh-water fishes. Proc. Acad. Nat.
Sci. Phila., 29: 83-104.
1883. Synopsis of the fishes of North America. Bull. U. S. Nat. Mus., 16:
i-lvi, 1-1018.
Jordan, David S., and Joseph Swain
1883. List of fishes collected in the Clear Fork of the Cumberland, Whitley
County, Ky., with descriptions of three new species. Proc. U. S. Nat.
Mus., 6(378): 248-251.
Keleher, J. J.
1956. The northern limits of distribution in Manitoba for cyprinid fishes.
Canadian J. Zool., 34: 263-266.
Kirtland, Jared P.
1845. Descriptions of the fishes of the Ohio and its tributaries (continued).
Boston Jour. Nat. Hist., 5: 21-32, pls. 7-9.
1846. Descriptions of the fishes of Lake Erie, the Ohio River, and their tribu-
taries. Boston Jour. Nat. His., 5: 265-276.
Knapp, Frank T.
1953. Fishes found in the freshwaters of Texas. Ragland Studio and Litho.
Printing Co., Brunswick, Ga., vii + 165 pp., 190 figs.
Koster, William J.
1939. Some phases of the life history and relationships of the cyprinid, Clin-
ostomus elongatus (Kirtland). Copeia, 1939 (4): 201-208.
Lachner, Ernest A.
1952. Studies on the biology of the cyprinid fishes of the chub genus Nocomis
of northeastern United States. Am. Midland Naturalist, 48(2): 433-466.
Langlois, Thomas H.
1954. The western end of Lake Erie and its ecology. J. W. Edwards, Ann
Arbor, Michigan, 479 pp., 72 figs.
Larimore, R. Weldon, and Philip W. Smith
1963. The fishes of Champaign County, Illinois, as affected by 60 years of
stream changes. Bull. Ill. Nat. Hist. Surv., 28(2): 299-382, figs. 1-70.
Livingstone, D. A.
1953. The fresh water fishes of Nova Scotia. Proc. Nova Scotian Inst. Sci.,
23(1): 1-90, figs. 1-36.
Luce, Wilbur M.
1933. A survey of the fishery of the Kaskaskia River. Bull. Ill. Nat. Hist.
Surv., 20: i-iii, 71-123, figs. 1-12.
Martin, W. R.
1949. The mechanics of environmental control of body form in fishes. Univ.
Toronto Stud., Biol. Ser. 58, Publ. Ont. Fish. Res. Lab., 70: 1-91, figs.
1-24.


1964









BULLETIN FLORIDA STATE MUSEUM


Miller, Robert Rush
1957. Utilization of x-rays as a tool in systematic zoology. Systematic Zool.,
6(1): 29-40.
Miller, Rudolph J.
1962. Sexual development and hermaphroditism in the hybrid cyprinid, No-
tropis cornutus x N. rubellus. Copeia, 1962(2): 450-452.
1963. Comparative morphology of three cyprinid fishes: Notropis corutus,
Notropis rubellus, and the hybrid, Notropis cornutus x Notropis ru-
bellus. Am. Midland Naturalist, 69(1): 1-33.
Mitchill, Samuel L.
1817. Report on the ichthyology of the Walkill, from specimens of fishes pre-
sented to the Society (Lyceum of Natural History) by Dr. B.
Akerly . Am. Monthly Mag. and Crit. Review, vol. 1, no. 4,
pp. 289-290.
Moore, George A., and John M. Paden
1950. The fishes of the Illinois River in Oklahoma and Arkansas. Am. Mid-
land Naturalist, 44(1): 76-95.
Murray, Merritt J.
1938. Survey of some northern Indiana streams with special reference to
trout production. Invest. Indiana Lakes and Streams, Indiana Dept.
Cons., 1(7): 80-99, 7 figs.
O'Donnell, D. John
1935. Annotated list of the fishes of Illinois. Bull. Ill. Nat. Hist. Surv., 20:
473-491, figs. 1-2.
Ortenburger, A. I., and Carl L. Hubbs
1927. A report on the fishes of Oklahoma, with descriptions of new genera
and species. Proc. Okla. Acad. Sci. (1926), 6(19): 123-141.
Outten, L. M.
1957. A study of the life history of the cyprinid fish Notropis coccogenis.
Jour. Elisha Mitchill Sci. Soc., 73(1): 68-84;
Pfeiffer, Roman A.
1955. Studies on the life history of the Rosyface shiner, Notropis rubellus.
Copeia, 1955(2): 95-104.
Pickens, A. L.
1928. Fishes of upper South Carolina. Copeia, 167: 29-32.
Putnam, F. W.
1863. List of the fishes sent by the Museum to different institutions, in ex-
change for other specimens, with annotations. Bull. Mus. Comp. Zool.,
1(1): 1-16.
Radforth, Isobel
1944. Some considerations on the distribution of fishes in Ontario. Contr.
Roy. Ont. Mus. Zool., 25: 1-116, figs. 1-32.


Vol. 8