TGl-

TULANE STUDIES IN
ZOOLOGY AND BOTANY

MCZ

library

,.^c 0 6 1999

' i i- ’■"'VARD

Volume 31, Number 1

1 November 1999

OBSERVATIONS ON THE NINE-BANDED ARMADILLO, DASYPUS
NOVEMCINCTUS, IN SOUTHERN LOUISIANA
ROYAL D. SUTTKUS

Tulane University Museum of Natural History
Belle Chasse, Louisiana 70037

AND

CLYDE JONES

Department of Biological Sciences and
Museum of Texas Tech University, Lubbock, Texas 79409

SYSTEMATIC REVIEW OF SUBGENUS FUSCATELUM OF
ETHEOSTOMA WITH DESCRIPTION OF A NEW SPECIES FROM
THE UPPER BLACK WARRIOR RIVER SYSTEM, ALABAMA
HENRY L. BART, JR.

AND

MICHAEL S. TAYLOR
Tulane University Museum of Natural History
Belle Chasse, Louisiana 70037

FUSCONAIA APALACHICOLA, A NEW SPECIES OF FRESHWATER
MUSSEL (BIVALVIA: UNIONIDAE) FROM PRECOLUMBIAN
ARCHAEOLOGICAL SITES IN THE APALACHICOLA BASIN OF
ALABAMA, FLORIDA, AND GEORGIA
JAMES D. WILLIAMS

U.S. Geological Survey, Biological Resources Division
7920 NW 71st Street, Gainesville, Florida 32653

AND

ARLENE FRADKIN
Department of Anthropology
Florida Atlantic University, Boca Raton, Florida 33431

Tulane
University
New Orleans

TULANE STUDIES IN ZOOLOGY AND BOTANY

ISSN 0082-6782

Tulane University Museum of Natural History
Building A-3, Wild Boar Road
Belle Chasse, Louisiana 70037

Henry L. Bart, Jr.

Editor

editor@museum.tulane.edu

INFORMATION FOR AUTHORS

Manuscripts dealing with all aspects of ecology, evolution, and systematic biology
are encouraged. Manuscript style should follow the latest edition of the Council of
Biology Editors Style Manual. Manuscript (with abstract) must be printed on good
quality 8.5 x 11 inch paper, double-spaced throughout (including Literature Cited,
figure captions, footnotes, and tables). All pages must have one inch margins on all
sides. Do not hyphenate words at ends of lines. Last name(s) of author(s) should
appear on each page with page number; this includes tables and figure captions.
Lettering used in figures must be easily read after reduction to a page size of 5 x 7.75
inches. Photographs (halftones) should be made for no or little reduction. Prospec-
tive authors should request a copy of the detailed guidelines for manuscript prepa-
ration (including footnotes, typefaces, literature and specimen citation, and figure
dimensions) from the Editor at the above address.

The original and three copies of the manuscript, along with figures, should be
submitted. All manuscripts are peer-reviewed. Following review, authors are re-
quired to submit revised manuscripts (text, tables, and figures, if possible) in elec-
tronic form, either on floppy or Zip™ disk, or by FTP. Consult the Editor for
acceptable formats and modes of electronic submission.

Page charges are assessed at the rate of $50 per page. Charges may be reduced or
waived if authors are unable to meet these charges.

EXCHANGES AND STANDING ORDERS

Tulane Studies in Zoology and Botany is available for exchange with institutions
publishing comparable series. Otherwise, standing orders are accepted, with is-
sues priced individually (foreign postage extra). A list of back issues with prices
is avail-able at http://www.museum.tulane.edu/pubs/or on request. Remittances
should be made payable to Tulane University.

MCZ

library

UEC 0 6 1999

OBSERVATIONS ON THE NINE-BANDED ARMADILLci,'^ R □
DASYPUS NOVEMCINCTUS, IN SOUTHERN LOUISIAN A' --O^l I Y

Royal D. Suttkus

Tulane University Museum of Natural History
Belle Chasse, Louisiana 70037
AND

Clyde Jones

Department of Biological Sciences and
Museum of Texas Tech University, Lubbock, Texas 79409

Abstract

From 1970 to 1995, observations were made on the nine-banded armadillo,
Dasypus novemcinctus, in southern Louisiana at the Riverside Campus of Tulane
University. Observations and corresponding data were obtained on natural history
in general, with emphasis on home range, as well as development and growth of
young, in particular.

Introduction

Although a considerable amount of information has been amassed on the
ecology and life history of Dasypus novemcinctus (McBee and Baker, 1982), we
present herein some additional data from a long-term study of the species con-
ducted in southern Louisiana.

Lowery (1936) recorded range extensions of D. novemcinctus within Louisiana.
Most of his records were from west of the Mississippi River. However, he men-
tioned that in late December 1935, two armadillos were captured south of Baton
Rouge east of the Mississippi River. Later, Lowery (1943) stated that armadillos
were abundant in southwestern Louisiana and that they were common in most of
the western portions of the state, but that these mammals were encountered
infrequently east of the Mississippi River in the Florida parishes. Fitch et al. (1952)
presented many distributional records for the armadillo in Louisiana, Arkansas,
and Mississippi, but they had no records for either the Louisiana coastal marshes or
the lower Mississippi River delta area. Lowery (1974) included a distribution map
based on museum specimens; his record for the delta region just southeast of New
Orleans was based on specimens in the Museum of Natural History of Tulane
University. For their distribution map of the armadillo, Choate et al. (1994) included
all of coastal Louisiana, except for the extreme lower end of the Mississippi River
delta and adjacent marshes on both sides. We postulate that D. novemcinctus may
have gained early access to the lower Mississippi River delta by moving down along
the Mississippi River levees.

It is of interest that the earliest specimen of D. novemcinctus deposited in the
collection of mammals of Tulane University was obtained on 14 January 1966.
However, based on our observations, as well as those of other personnel of the

Tulane Studies in Zoology and Botany 31: 1-22. 1999.

1

2

Tulane Studies in Zoology and Botany

[VoL 31, No. 1

University, armadillos were reported as common at the Riverside Campus since
1964.

The purpose of this paper is to present summaries of our observations of D.
novemcinctus in southern Louisiana. Published literature pertinent to our findings
is reviewed. However, for a more extensive list of the literature on this species, see
the work by McBee and Baker (1982).

STUDY AREA. The Riverside Campus of Tulane University is a 520-acre tract of land
at the northern end of Plaquemines Parish, several miles southeast of New Orleans,
Louisiana (Figure 1). The area is west of the Mississippi River at English Turn Bend
(river mile 78). The Mississippi River forms the southern boundary of the Riverside
Campus. More than 1,000 acres of bottomland hardwoods are adjacent to the
eastern and northern boundaries of the property.

During World War II, ammunition storage bunkers were constructed through-
out the central part of the current Riverside Campus. Throughout the entire area,
inclusive of the adjacent acreage, a system of drainage ditches was excavated, and
the soil was piled along the ditches. Many of the ditches date back to the early
plantation days typical of the area. In subsequent years, trees (Celtis laevigata, Acer
negundo, A. rubrum, Carya illinoensis, Salix nigra, Ulmus americana, Populus deltoides,
Quercus virginiana, Myrica cerifera, and Sambucus canadensis, to name some of the
common species) became established along and on these spoil ridges alongside the
drainage ditches. In the adjacent wooded areas, there are stands of scattered
individuals of Taxodium distichum, Quercus nuttali, Q. nigra. Morns rubra. Magnolia
grandiflora, Liquidambar styraciflua, Platanus occidentalis, Diospyrus virginiana, Gleditsia
triacanthos, Salix interior, Cephalanthus occidentalis, Nyssa aquatica, and Fraxinus
pennsylvanica.

Some of the excavated dirt was piled on and around the ammunition storage
bunkers for camouflage; this resulted in the formation of extensive series of ponds
(Figure 1-B). The elevated spoil banks along ditches and the elevated soil slopes
around the bunkers have become ideal burrowing sites for armadillos. Moreover,
the drainage system that was developed during the plantation days and enhanced
further by development of the ammunition storage facility resulted in the lowering
of the water table and the eventual collapse of soil around the bases of large trees,
thus providing additional sites for burrowing by armadillos.

A detailed description of the ammunition storage bunkers was provided by J ones
and Pagels (1968). Warkentin (1968) briefly described the common vegetation
around the ponds (borrow pits). Suttkus and Jones (1991) described the Riverside
Campus of Tulane University and adjacent wooded areas, and provided a list of the
common trees on the area. During about the last five years, two invading plants have
become abundant in the area. The wind-blown seeds of the Chinese tallow (Sapium
sebiferum) have germinated on the bare, moist soils along the edges of the ponds
during low water levels. It seems that birds have spread the seeds of the Chinese
privet {Ligustrum sinense) into the area; now most of the margins of the wooded areas
are crowded with privet.

During the early to late 1960s, the railroad tracks along the fronts of the
ammunition storage bunkers were removed. The railroad track right-of-ways were
converted to shell roadways. In addition, most of the brush and trees from on top

1999]

Observations on Nine-Banded Armadillo

3

Figure 1. Location of the study area (Riverside Campus of Tulane University) in
southern Louisiana. A. General location; B. Enlargement of Riverside tract.

4

Tulane Studies in Zoology and Botany

[Vol. 31, No. 1

of and around the bunkers were removed. The resultant, open, mowed areas
became foraging grounds for armadillos, in addition to the diversity of foraging
areas in the adjacent wooded areas.

Methods

Our methods of capture and marking armadillos were similar to those used by
Clark (1951), Fitch et al. (1952), and Breece and Dusi (1985), except that instead of
numerals, we sprayed a polka-dot pattern on the animals. We captured the animals
either by hand or more often with a deep, large-mesh, nylon, long-handled net. We
made stencils of cardboard with holes (quarter size) that we used to transfer a polka-
dot pattern to the scapular or pelvic shield (or both) with different colors of fluorescent
spray paint. The same patterns, numbers, and colors were painted on the right and left
side of each animal. During the course of the study, some armadillos were repainted
when either one side or the other started to fade. In addition to the spray-paint spots,
each animal was tagged on the left ear pinna with a small fisheries opercular strap tag.
For nearly all of the armadillos, total length in millimeters (mm), weight in kilograms
(kg), and sex were recorded at the time of capture, marking, and tagging.

Armadillos captured initially near the ammunition storage bunkers were marked
accordingly. Animals captured near bunker A-9 were marked with yellow; those
from the bunker A-10 area were marked with purple; armadillos from around
bunker A-11 were marked with green; those from near bunker A-12 were marked
with red; animals captured near bunker A-13 were marked with blue; those from
around bunker A-14 were marked with gold; and those from the areas adjacent to
bunkers A-1, A-2, and A-3 were marked with white.

Animals were marked, tagged, and the corresponding data were recorded at the
site of capture; upon completion of the process, the armadillos were released. Most
of the Dasypus were captured initially and marked on the following dates (numbers
of animals in parentheses): 14 February 1970 (17); 17 February 1970 (3); 18 February
1970 (8); 22 February 1970 (6). No new armadillos were captured and marked after
22 February 1970. Approximately half of the animals (23) used in the computations
of home ranges were recaptured and remarked during the aforementioned period
of observations. Identity of each individual remarked was verified by the number
of the ear tag.

Some observations of animals were made with the use of field binoculars.
However, most of the spray-paint markings could be observed easily with the naked
eye. Each observation of a marked armadillo was plotted on a ground chart of
known scale. In addition, the time of day, air temperature, and presence and
location of other marked, as well as unmarked, animals were recorded. After 14
February 1970, observations were continued periodically throughout March and
most of April of 1970.

In this study, minimum home ranges of D. novemcinctus were calculated in the
manner described by Clark (1951). Each observation of a marked animal was plotted
on a known-scale ground chart. The outermost points of the observed ranges of the
animals were connected by straight lines; thus, the minimum home range was
circumscribed. The minimum home range in acres was determined by placement

1999]

Observations on Nine-Banded Armadillo

5

of a grid over the known-scale ground chart, counting the number of whole squares
within the range diagram, and adding the whole-square equivalent to all of the
partial enclosed squares.

Materials for the study of development were obtained by salvage of animals dead
on the roads or from females captured in the area. All embryos were deposited in
the Tulane University Museum of Natural History as were the salvageable females
(mothers). Measurements to the nearest 0.1 mm were taken with a dial caliper.

Additional observations on various aspects of life history and other activities of
D. novemcinctus in the study area and adjacent regions were carried out sporadically
from February 1970 through the fall of 1995.

Results

Home Range and Movements. — A total of 34 armadillos (20 males, 14 females) was
captured and marked during this study. After the initial capture, three males and
two females were not seen again. One male and one female were observed only one
time after being marked. One male and three females were observed two times after
the initial capture. Information from these 11 individuals was not used in the
determinations of home ranges. Throughout most of the study, especially in the
1970s and 1980s, some armadillos were found dead from apparent gunshot wounds.
For example, three animals were found dead after 12, 14, and 12 days of observa-
tions, respectively. Minimum home ranges were determined for 14 males and nine
females (Table 1). Home range overlap was common; overlaps of ranges of marked
animals are presented in Table 2. There were fifteen periods of observations during
17-28 February 1970. Beginning times of observation periods ranged from 1010 to
1722 hours, air temperatures during these observation periods ranged from 14°C to
23 °C, with a mean of 17.9°C. There were 22 periods of observations during 1-28
March 1970. Beginning times of observation periods ranged from 1015 to 1800
hours, with air temperatures that ranged from 12°C to 23.5°C with a mean of 17.9°C.
Whether the mean temperature of 17.9°C for the 15 observation periods in February
and the 22 observation periods in March was coincidental or whether the animals
were highly sensitive to air temperatures and regulated their above ground activi-
ties accordingly we are not prepared to say. There were two observation periods in
April (12 April 1970; 1445 hrs.; 17 April 1970; 1735 hrs.). Air temperature was 25°C
during both observation periods.

Development. — Data on seven litters of armadillos from five different years were
obtained from the study area. Three of the females carrying fetuses were found dead
on roads; four pregnant females were caught by hand. The aforementioned female
armadillos were obtained on 15 April 1970, 10 March 1982, 11 February 1983, 14
February 1986, 7 March 1986, 28 February 1987, and 15 March 1987. As expected, all
litters included four fetuses. For convenience of presentation, the data presented in
Table 3 are arranged by average sizes of fetuses rather than chronology of capture.

Discussion

Home Range and Movements. — Some early work on home range of D. novemcinctus
was conducted in southern Texas by Clark (1951). Between 18 October 1947 and 15

6

Tulane Studies in Zoology and Botany

[Vol. 31, No. 1

Table 1. Minimum home ranges and related data for 23 armadillos in southern
Louisiana.

Tag

Number

Sex

Weight (kg)

Total Length
(mm)

Number of
Observations

Days of
Observations

Minimum Home Range
Acres Hectares

428

Male

4.2

730

8

35

0.89

0.36

460

Male

2.75

710

5

35

0.59

0.24

461

Male

2.75

710

4

12

0.39

0.16

432

Male

4.5

775

14

22

1.67

0.68

434

Male

4.5

735

25

39

2.86

1.16

435

Male

—

—

7

23

0.74

0.30

459

Male

5.0

840

12

55

1.15

0.46

464

Male

3.4

712

6

15

1.08

0.44

441

Male

4.25

720

10

14

1.33

0.54

442

Male

—

—

10

16

2.31

0.93

447

Male

—

725

14

12

1.47

0.59

448

Male

—

660

15

18

2.15

0.87

449

Male

4.0

740

19

21

3.58

1.45

465

Male

3.25

760

6

8

0.39

0.16

457

Female

3.25

635

9

17

1.74

0.70

429

Female

5.5

835

9

12

0.64

0.26

. 431

Female

4.5

750

11

43

1.92

0.78

433

Female

5.0

790

23

63

2.32

0.94

454

Female

3.5

727

8

19

0.57

0.23

436

Female

3.5

705

7

9

0.54

0.22

439

Female

2.0

—

12

25

1.94

0.78

450

Female

4.5

750

11

21

1.99

0.80

451

Female

3.5

735

4

25

3.32

1.34

February 1948, surveys to observe armadillos were made 30 times. Four animals
(one juvenile, three adults) were observed on more than four occasions. The longest
period of observation was 28 days. It was estimated that the juvenile armadillo had
a home range of 0.69 acres (0.3 hectares, ha); the three adults had an average home
range of 8.5 acres (3.5 ha).

From February to June of 1948, Fitch et al. (1952) studied home ranges of
armadillos in Natchitoches Parish, Louisiana. Based on 32 observations of 13
different animals, they concluded that the usual home range of an individual was
approximately 50 acres (20.3 ha).

Layne and Glover (1977) marked and released 47 armadillos in Highlands
County, Florida. They presented estimates of minimum home ranges for eight of 12
animals that were observed four or more times between July 1968 and September
1974. Their estimates of home ranges of D. novemcinctus varied from 2.72 acres (1.1
ha) to 34.10 acres (13.8 ha), with an average of 14.08 acres (5.7 ha). In addition, Layne

1999]

Observations on Nine-Banded Armadillo

7

Table 2. Tag numbers and sexes of free-ranging armadillos with overlapping home
ranges in southern Louisiana.

Number, Sex Number, Sex of Armadillos with which Home Ranges Overlapped

428, male

429, female

431, female

432, male

433, female

434, male

435, male

436, female
439, female

441, male

442, male

447, male

448, male

449, male

450, female

451, female
454, female
457, female

459, male

460, male

461, male

464, male

465, male

457, female; 460, male; 461, male
457, female; 434, male; 461, male

432, male; 434, male; 433, female; 435, male; 454, female; 459, male
431, female; 434, male; 433, female; 435, male; 454, female; 459, male
431, female; 432, male; 434, male; 435, male; 454, female; 459, male

429, female; 431, female; 432, male; 433, female; 435, male; 454, female; 459,
male

431, female; 432, male; 434, male; 454, female; 459, male; 441, male
459, male; 464, male; 441, male; 442, male
441, male; 442, male

459, male; 436, female; 439, female; 442, male
439, female; 441, male
449, male, 451, female
449, male; 450, female

447, male; 448, male; 450, female

448, male; 449, male
447, male

431, female; 432, male; 434, male; 433, female; 435, male; 459, male
428, male; 460, male; 461, male; 429, female

433, female; 435, male; 454, female; 464, male; 436, female; 441, male
457, female; 428, male; 461, male

457, female; 428, male; 460, male
435, male; 454, female; 459, male

No overlapping home ranges with any marked armadillos

and Glover (1977) provided information on sizes of animals. The mean weight for
12 armadillos was 2.7 kg; the largest male weighed 3.4 kg and the largest female
weighed 3.8 kg.

Thomas (1980) fitted a transmitter to an adult male D. novemcinctus in Bosque
County, Texas. The animal weighed 5.0 kg when captured; the weight was 4.7 kg
at the time of release five days later. The animal was released on 4 June 1978 and
located 43 times during a period of 55 days. Based on the minimum polygon
method, the home range was estimated at 20.01 acres (8.1 ha).

Galbreath (1982) reported information on home ranges of two adult armadillos
(one male, one female) at the Archibold Biological Station, Highlands County,
Florida. The size of the home range of the male was estimated at 26.69 acres (10.8 ha);
that of the female was 18.78 acres (7.6 ha). Galbreath (1982) also included some
information on estimated sizes of home ranges of 8.15 acres (3.3 ha) for males and
8.40 acres (3.4 ha) for female armadillos at an unspecified site in Mississippi.

Table 3. Morphometries (mm) of seven sets of fetuses of armadillos from southern Louisiana. Means are in parentheses. See
text for dates of collections and descriptions of fetuses.

Tail Hind Head Scapular Scapular Pelvic Pelvic Head

Tulane Studies in Zoology and Botany

[Vol. 31, No. 1

lO lO IN. oq
id Lfi id ^ LfS

(N in 00 CN

K [n! vd K K

(N CN CN CN (N

lO LO ID CM ^

00 in p lo lo

id id id id id

CM p tJH LfO
CO CM cd cd

CM CM CM CM

t-h p CO in in
vO ^ ^

t-h cN 00 cq p

On 00 00 00

00 00 ^ O

00 VO in tx ^
K K K VO K

p in in p

cN ni cd cd CM

vo CO 00 m

cd vd VO

in ^ o CO

in! !>; ts!

rH tN ^ 00 m
CD CD o o cd

^ p p p

cd vD rjH id

CM p p IN.

00 00 00 00 00

IX in T-i in CM
cd ^ ^

CO O IX CO

^ ^ cd ^

p p p p p
^ ^ ^ ^ ^

p p p p
OV OV OV Gv Ov

p p p
CM CM CM CM

O CO T— I O T— (
CM CM CM CM CM

p p p p ^

cd cd cd cd cd

o o o o
^ ^ ^

p p ^ p ^
cd cd cd cd cd

p p p p p

vd in! vd in! in!

p p p p
ON Ov Ov Ov

p ^ p p
CM CM CM CM

p p p p p
VO ^ ^ NO N^

o o o o

O O 00 o o

ppppp ppp
rjH'^cd'^rt^^^^ooNd

-MCMCMCMCMCMpCOCOCOCOCO
0) ^ — O) ' '

cn cn

00 00 00 00

(86.0) (29.0) (9.1) (4.0) (12.3) (14.1) (15.2) (15.3) (11.8) (23.2) (36.5)

7 March 1986

87.5 27.5 10.1 3.7 13.3 — 13.9 — 12.6 22.8 39.3

88.6 29.1 10.7 3.8 12.8 — 13.9 — 13.1 23.7 39.5

1999]

Observations on Nine-Banded Armadillo

CO LO CTs

T—l ^ 0^

CO CO

V.D LO 00 CO 00

O t-h O O O

Lo Lo lo Lo in

^ (N O CO O

lo m

O vd <N

ON ON Os

NO CO ON CM O

t-h t-h CO CN On

K in! NO NO

CO CO CO CO CO

lO CO t-h

c6 c6 CO

00 NO LO 00 IN.
00 00 00 00 00

LO <N ^ LO IN.

id

LO O LO 00 o

in! in! NO NO in!

(N (N (N (N fN

IN. (Nj 00 p

00 o ON In! ON

CN CO (N (N <N

On (N (N

^ ^ ^

IN (N (N (N (N
id id id id LO

O O O t-h p
On 00 00 00 00

IN. p p p p

d NO NO id

(N 04 (N CM (N

CO O O CO IN

CM NO CM
CO ^ Csi
CO CO CO

P IN. CM
CO CM CO

O CM NO T-i

In! NO in! in! in!

p CM p CM p
d 00 On ON ON

00 m 00

^ id ^

Ovl CM tN

ppp ppppp

COCOCO 0606060606

00000

ppp

ON ON On

oocom 00000

odd iNdiNiNiN!

ppppp
CM d d T— I r— I
CM CM CM CM CM

ppppp

^ ^

00000

CO ^ CM CO

10 10 uo m LO

00000

00000

NDND^ , . . _ .

10 LO LO LO LO

03 CM CM CM

242.0 101.0 35.0 19.0 24.5 32.2 25.6 34.2 32.2 52.5 93.5

(241.0) (102.0) (36.0) (19.0) (24.9) (33.1) (25.0) (35.3) (32.8) (51.8) (93.3)

10

Tulane Studies in Zoology and Botany

[VoL 31, No. 1

Figure 2. Enlargement of area around bunkers A-10, A-11, A-12 showing bunkers
in relation to shell road (heavy black line) and long pond. A. Home ranges of female,

number 433 ( ) and male, number 434 ( ); B. Home ranges of female,

number 431 ( — ) and male, number 432 ( ).

Breece and Duci (1985) observed 26 adult and 26 juvenile, marked armadillos at
the Fred T. Stimpson Sanctuary in Clarke County, Alabama, from July 1972 through
August 1973. Average minimum sizes of home ranges for 21 adults seen three or
more times each were estimated at 8.67 acres (3.5 ha), whereas those of 15 juveniles
were 3.63 acres (1.5 ha).

Based on our observations of D. novemcinctus in southern Louisiana, estimates of
sizes of home ranges were smaller than those presented in the aforementioned
studies. Our estimates were derived by using the minimum polygon method for
data obtained on 23 of the original 34 armadillos that were marked and released
during the study. Number of observations per individual ranged from four to 25;
duration of observations per individual ranged from eight to 63 days (Table 1).
Estimated minimum home ranges for 14 males varied from 0.39 acres (0.2 ha) to 3.58
acres (1.4 ha), with a mean of 1.47 acres (0.6 ha). Similar estimates for nine females
varied from 0.57 acres (0.2 ha) to 3.32 acres (1.3 ha), with a mean of 1.66 acres (0.7 ha).
Analyses of combined data for 14 males and nine females resulted in average.

1999]

Observations on Nine-Banded Armadillo

11

estimated, minimum home ranges of 1.55 acres (0.6 ha).

There were no apparent correlations between sizes of armadillos and estimated
sizes of home ranges. For example, the smallest, marked animal in our study was
a female (tag no. 439; 2.0 kg) that was observed on 12 occasions over a period of 25
days; she had an estimated home range of 1 .94 acres (0.8 ha). A large, marked female
(tag no. 431; 4.5 kg) that was observed 11 times during a period of 43 days had an
estimated home range of 1.92 acres (0.8 ha). Figure 2-A depicts the home range of
a female, tag no. 433; 5.0 kg; 790 mm total length (TL), based on 23 observations
during a period of 63 days and a male, tag no. 434; 4.5 kg; 735 mm TL, based on 25
observations during a 39 day period.

Figure 2-B depicts the home range of a female, tag no. 431; 4.5 kg; 750 mm TL,
based on 11 observations during a period of 43 days and a male, tag no. 432; 4.5 kg;
775 mm TL, based on 14 observations during a period of 22 days (Tables 1 and 2).

Obviously the home ranges of the four armadillos cited above overlapped
considerably and so it was impractical to place all four on the same diagram.
Moreover, there were two additional females and two additional males that had
overlapping home ranges with one or more of the four foregoing listed animals
(Table 2). These latter four animals also had elongate home range patterns similar
to those shown in Figure 2-A and 2-B.

The small sizes of our estimated home ranges of armadillos may be correlated
with the topography of the study area, as described previously (Figure 1). The
abundance of relatively well-drained burrowing sites seemingly was important to
the animals in the area. Hard-packed roads may have served as partial deterrents
to movements of animals, especially when foraging. However, it was not uncom-
mon to see armadillos crossing roads. During this study, marked animals frequently
were observed foraging in open areas on several occasions (one to four times) during
a span of time (one to five hours) daily at relatively short distances between the
points of observations. Therefore, it seems logical to assume that there was a
plentiful supply of food in the areas frequented by the armadillos.

Layne and Glover (1977) indicated that there were no overlaps in home ranges
of adult females. However, several other authors (Clark, 1951; Fitch et al., 1952;
Galbreath, 1982; Breece and Dusi, 1985) reported that overlaps of home ranges of
armadillos frequently were conspicuous. We observed numerous instances of
overlaps of home ranges between juveniles and adults of both sexes (Table 2). Only
one individual (adult male) was determined to not overlap with home ranges of
other animals. Overlaps of home ranges and closeness of armadillos frequently
observed by us may have been related to pairing behavior of D. novemcinctus as
discussed by McDonough (1997).

There are relatively few published reports of densities of armadillos. Galbreath
(1982) provided an estimated density of these mammals as less than one animal per
hectare. Kalmbach (1943) attempted to estimate densities of armadillos based on
questionnaire surveys; he reported a high density of 0.82 animals per acre and a low
density of 0.21 armadillos per acre. Taylor (1946) gave some estimates of densities
of D. novemcinctus based on strip counts of active dens. However, he expressed
uncertainty with regard to the direct correlation between the numbers of active dens
and the numbers of animals occupying the burrows; as many as five armadillos were

12

Tulane Studies in Zoology and Botany

[Vol. 31, No. 1

found in one den. Taber (1945) attempted to correlate numbers of animals with
numbers of dens; he found one armadillo per 4.5 dens as a result of one study and
one animal per 8.5 dens in another observation.

Random strip surveys were conducted to estimate the number of armadillos in
the study area. A road 1 .9 kilometers long, with an adjacent observable, open, grassy
area estimated to be 20 acres (8.1 ha), was selected for these surveys. The road was
traveled in 10 to 15 minutes and the numbers of armadillos observed were recorded.
The results of the strip surveys are depicted in Table 4.

Some seasonal changes were noted in connection with patterns of activities of D.
novemcinctus in the study area. During the winter and early spring, most of the
foraging by armadillos was diurnal. However, as the nights became warmer, the
animals began to emerge and forage increasingly later in the afternoons. During the
summer, most of the foraging activities of armadillos were nocturnal. Other than
periods of inclement weather, such as hurricanes, September and October were the
driest periods in the study area. The days were warm and the nights were cool;
frequently there were ground fogs and heavy dews in the early evening. Although
the ground frequently was dry and cracked, the open, grassy areas were frequented
by armadillos during this season. McDonough and Loughry (1997a) reported on

Table 4. Observations of armadillos during strip surveys along 1.9 kilometers of
road and adjacent areas of an estimated 20 acres (8.09 ha) in southern Louisiana.

Dates

Times of Observations

Numbers of
Armadillos

24 October 1987

1630-1640

14

25 October 1987

1500-1510

10

25 October 1987

1755-1805

8

26 October 1987

1535-1545

4

30 September 1990

42

1 October 1990

41

2 October 1990

1720-1730

29

15 October 1990

1805-1815

51

17 October 1990

1705-1720

60

20 October 1990

1755-1810

54

22 October 1990

18

23 October 1990

16

28 October 1990

1120-1130

18

28 October 1990

1620-1630

31

29 October 1990

1117-1127

7

30 October 1990

1120-1130

17

30 October 1990

1620-1630

29

11 September 1995

1845-1856

6

26 September 1995

1900-1905

5

29 September 1995

1819-1833

8

1 November 1995

1657-1703

10

1999]

Observations on Nine-Banded Armadillo

13

activity periods of armadillos in northern Florida. They observed considerable
flexibility in the timing of activity of these mammals. However, activity was
correlated positively with cloud cover during the day and with warm and dry
conditions at night.

A summary of patterns of mortality in D. novemcinctus was presented by
McDonough and Loughry (1997b). They found significantly higher incidences of
mortality caused by predators in juveniles as compared to adult armadillos. How-
ever, mortality during a prolonged period of drought was greater for adults than for
juvenile animals. Armadillos frequently are killed on roads (Loughry and
McDonough, 1996). However, reliable estimates of the impacts of road kills on
populations are difficult to determine (Case, 1978). As implied previously, mortality
of armadillos in the study area was attributed mostly to interactions with humans;
a few animals were found dead on roads and some armadillos apparently were shot.
Newman (1913) provided information on the slaughter of armadillos for the use of
the armor in making baskets. For example, he stated that one dealer claimed to have
shipped about 40,000 baskets made from armadillos in a period of six years.
Incidentally, ornamental items, such as baskets and purses, as well as stuffed
animals, still are available in many markets.

Development. — There have been some conflicting reports in the literature with
regard to positions of D. novemcinctus during copulation. For example, Newman
(1913) and other authors reported that copulation occurred with the female turned
on her back. However, Galbreath (1982) observed copulation by the male mounting
the female from the back. This position of animals during copulation was correlated
with studies of the shape of the penis of the male with the structure of the female
urogenital sinus (Newfang, 1947; Galbreath, 1982).

One of us (RDS) observed mounting behavior by armadillos in or near the study
area. On 6 September 1984, two armadillos were observed for a considerable period
of time. One animal was foraging, and was followed by the other which attempted
periodically to mount the foraging animal. Each time the rear animal would try to
mount, the foraging armadillo would take a few rapid steps forward, disengage, and
begin foraging again. The process was repeated numerous times over a distance of
several hundred feet. Although copulation was not observed, the actions of this pair
of animals could not be described as aggressive behavior as discussed by Denson
(1979) and McDonough (1994). On 25 November 1984, an armadillo was observed
mounted on the back of another animal at 2200 hours. Based on the movements of
the animals, it was assumed that copulation occurred. Some detailed descriptions of
pairing behavior of armadillos were presented by McDonough (1997).

Information on the reproductive cycle of D. novemcinctus has been provided by
numerous authors. Patterson (1913) reported that old females bred first, mating in
most cases before 15 October, while the second-year females continued to breed for
some time after this period. He indicated a period of quiescence of about three
weeks, with a gestation period of about an additional 140 days, with the majority of
young born in March and April. Hamlett (1933) suggested a quiescence period of
about 14 weeks in duration. Based on examination of over 2,000 female armadillos
killed at intervals throughout the year, Hamlett (1935) reported a total length of
gestation of nearly eight months, made up of a "Tree-vesicle stage" of three and a half

14

Tulane Studies in Zoology and Botany

[VoL 31, No. 1

months and an actual development period of about four months. A gestation period
of eight to nine months, with an embryonic diapause of three to four months, was
suggested by Enders (1966). A comprehensive summary of reproductive strategy
in D. novemcinctus was presented by Storrs et al. (1989). Although the gestation
period of armadillos has been regarded generally as eight to nine months, including
a three to four month period of embryonic diapause, captive females not exposed to
males gave birth to young 13 to 24 months after capture (Storrs et al., 1989).

Talmage and Buchanan (1954) noted that female armadillos produced new
litters as early as February. Lowery (1974) reported that February was typical for
time of parturi-tion of armadillos in Louisiana. Enders (1966) found fetuses
approaching full-term size usually in late March and early April. During this study,
we found the largest fetuses in mid April. However, length of gestation and time
of parturition may be influenced by various stresses to which female armadillos are
subjected (Storrs et al., 1989).

According to Enders (1966), the time of parturition of armadillos in late March
and early April coincided with the presence of large numbers of lactating females.
In this study, a non-pregnant female captured on 28 February 1987 weighed 4.0 kg,
had nipples 9 mm in length, and a vestibular cleft 10 mm in length. A pregnant
female captured on the same day weighed 4.6 kg, had nipples 13 mm in length, and
a vestibular cleft 20.5 mm in length. A lactating female captured on 30 April 1986
weighed 4.6 kg and had nipples 20.5 mm in length. Another lactating female
captured on the same date weighed 4.5 kg and had similar nipple measurements.

^ On 21 April 1983, a female, dead-on-road, was examined and prepared as a
museum specimen. She weighed 5.0 kg, was not lactating and had not given birth
recently. There were unusual amounts of fat ever}rwhere, around the legs, at base
of tail and in neck and belly regions.

Illustrations of sets of fetuses of D. novemcinctus were provided by Newman and
Patterson (1910), Newman (1913), Patterson (1913), and Kalmbach (1943); however,
dates when fetuses were obtained and measurements of individuals were not
included. Newfang (1947) gave crown-rump measurements of 29 fetuses and the
dates that they were preserved. In addition to internal structures, she described and
illustrated the external genitalia of armadillos.

The following descriptions of external morphology of fetuses of D. novemcinctus
are based on sets of fetuses (Table 3). The sets of fetuses are listed in order of average
fetus size. Following the date of capture of female are the standard measurements
(total length, tail length, hind foot length, and ear length) and weight, when known,
of the female (mother). If the individual female had a stub tail, then the total length
and tail length measurements are enclosed by parentheses. The seven sets of fetuses,
with date of capture and measurements of female are as follows: Set 1,14 February
1986, (668)-(265)-100-37, unknown weight; Set 2, 28 February 1987, 768-337-96-38,
4.6 kg; Set 3, 11 February 1983, 760-320-70-44, 3.8 kg; Set 4, 15 March 1987, (718)-(302)-
96-39, 4.2 kg; Set 5, 10 March 1982, 788-353-91-41, 6.3 kg; Set 6, 15 April 1970, 700-323-
95-30, 4.9 kg. Data are reported for an additional set of fetuses obtained on 7 March
1986 (Table 3) from a road-killed female. Measurements for the female were 758-313-
100-40, unknown weight. The fetuses were blood-stained and damaged such that
accurate measurements could not be obtained for width of the scapular and pelvic

1999]

Observations on Nine-Banded Armadillo

15

shields. Thus, this set of fetuses was not used in the descriptions of development of
external morphology and pigmentation.

Rates of growth of armadillos are depicted in Figure 3. The head, trunk, hind foot,
and ear pinnae have a similar pattern of change in relative size during development
(Figure 3). The relative proportions of these structures are comparatively stable
during the course of early development of the fetuses. However, the tail increases
in length at a faster rate than the trunk after a total body length of about 80 mm
(Figure 3).

Examples of fetuses from Sets 1-6 are shown in Figures 4-6. In the early sets of
fetuses (Sets 1-4), there was little color (Figure 4). In some cases, the internal organs
could be seen through the translucent skin of the belly and the band region of the
carapace. In fetus Set 5, adult coloration was discernable and the areas between the
scutes were pigmented. However, pigmentation was most obvious on the posterior
part of the head, the pinnae, the dorsal part of the carapace, and the bases of the hind
feet (Figure 5). The fetuses of the Set 6 essentially were of the same coloration as the
adults; but the fetuses were lighter brown in color than the adults (Figure 6).

Progression of the development of the heads was obvious in the sets of fetuses
that we examined. In the 1st set of fetuses, the eyes were developed, the lenses could
be discerned, and the eye lids were open partially. In fetus Set 2, the eye lids were
fused together, but the lenses were obvious. However, in fetuses of Set 3, the eye lids
were thicker and opaque; the lenses were obscured. In fetus Set 5, there were narrow
slits between the eye lids. In fetuses of Set 6, the eyes were half open.

The nostrils appeared as small pits in the first set of fetuses studied. During
subsequent development, the pits were increasingly deeper and greater in diameter.
In fetus Set 4, fleshy folds were present that divided the holes vertically and made
them appear more slit like. These structures were increased in size in the remaining
sets of fetuses.

In the fetus Set 1, the tips of the pinnae were folded inward to close the openings
in the ears; the bases of the pinnae were not adhered to the head. However, in fetus
Set 2, the bases of the pinnae were adhered to the head, but the tips were not folded
inward. These positions of the pinnae were present in all of the other sets of fetuses.
In fetus Set 4, development of scutes was noted at the bases of the pinnae.
Approximately half of the pinnae were covered with scutes in fetus Set 5. In fetus
Set 6, the pinnae were equipped fully with scutes. The distances between the right
and left pinnae were similar in all of the fetuses examined. It seemed that the head
increased in width and the pinnae enlarged, thus creating the perception that the
ears became closer together and moved from the sides to the tops of the heads of the
developing fetuses.

The tongues were protruding conspicuously from the mouths of all fetuses of
Setl and Set 2. In fetus Set 3, the tongues were enclosed completely within the
mouths of the animals. Perhaps the rates of growth of the jaws exceed those of the
tongues. In the smallest fetuses (Sets 1-3), the tongues were round in cross section;
in Sets 4-6, the tongues were flattened in cross section.

In the smallest fetuses, the jaws were flattened and without well-defined edges.
However, the edges of jaws were outlined in the fetuses of Set 2, and well developed
in Set 5 and Set 6. All fetuses examined lacked teeth. In fetus Sets 1 and 2, the lower

16

Tulane Studies in Zoology and Botany

[Vol. 31, No. 1

o

o

D U U jC

Tail
Tru n I

Head

“O

c

b:

D

UJ

Jdysi

• ^

□

★

O

Joyyoi

^ •

□

★ O

Joy^si

• □

★ O

-IDVVi

qadu

• CH ^ CD

qejBz

^ ★O

<rC] O

o oe ^ N

:3

o

K-i

c

c

O

c6

Symbols represent the mean value of the respective embryos.

1999]

Observations on Nine-Banded Armadillo

17

Figure 4. Photographs of representatives from fetus Set 1, Set 2 and Set 3 of
armadillos from southern Louisiana.

18

Tulane Studies in Zoology and Botany

[Vol. 31, No. 1

Figure 5. Photographs of representatives from fetus Set 4 and Set 5 of armadillos
from southern Louisiana.

and upper jaws were of the same length. In subsequent sets of fetuses, the upper
jaws were more and more protruding; most of the increases in the lengths of the
heads were due to the growth of the upper jaws. In fetus Set 6, the upper jaws were
8 mm longer than the lower jaws. Incidentally, this was when fleshy lips were first
detected on the fetuses studied.

In fetus Set 1, sutures of the underlying bones of the cranium were visible. In
fetuses of Set 2, primordials of scutes were noticeable on the posterior ends and the

1999]

Observations on Nine-Banded Armadillo

19

Figure 6. Photograph of representative from fetus Set 6 of armadillos from southern
Louisiana.

central parts of the tops of the heads. In the remaining fetuses, the individual scutes
were distinguishable; the head shields were well defined on the posterior parts of
the head (by a skin fold) and also on the snouts of the fetuses. On the snout of each
fetus, the shields were interrupted by a single line extending from the edges of the
eyes anteriorly to the tip of the snout.

There was noticeable progression of the development of the trunks in each of the
sets of fetuses studied. The trunks became changed from arched shapes to extended
ones; this trend was conspicuous as demonstrated by examination of the six sets of
fetuses (Figures 4-6). The areas that would be eventually covered by the carapaces
were already well defined in the smallest fetuses. All nine bands already were
present in the dorsal regions, and the anlages of the first five rows of scutes could be
discerned. The edges of the carapaces already slightly overlapped the venters and
upper extremities of the limbs. In fetus Set 2, the anlages of all scutes were
distinguishable from each other. The anlages of the scutes started as small, white
spots, but soon became enlarged until the different scutes almost were touching each
other. In subsequent sets of fetuses, the individual bands were increasingly longer.
The different shapes of the scutes (elongate-triangular in the bands, octagonal on the
pelvic shields) were established first in fetus Set 4.

In fetus Set I, genital papillae were present. The umbilical cords were situated
short distances anterior to the papillae. In fetus Set 2, whitish blotches were
arranged in rows on the venters of the animals. However, by fetus Set 5, the whitish
blotches were in irregular patterns. These white spots may be the anlages for the
dermal nuclei that become the sites for hair tufts in adult armadillos. The spots

20

Tulane Studies in Zoology and Botany

[Vol. 31, No. 1

initially were located on the bellies of the animals, but later (fetus Set 4) were
noticeable on the chins, throats, and the upper parts of the legs. No hairs were
noticed on the fetuses examined. The anal aperatures were apparent first in fetus Set

5. In fetuses of Set 6, the anal aperatures were slit-like at the crests of slight
elevations. In fetus Set 6, nipples were visible.

In the smallest fetuses (Set 1), the tails were curled anteriorly between the hind
legs. In fetus Set 2, the tails were uncurled partially; development of bony rings on
the tails was detected. The anlages of seven different rings were distinguished; the
tails became separated from the carapaces by folds of skin. In fetus Sets 3 and 4,
anlages for eight rings were present, and different scutes became distinguishable
in the 4th set of fetuses. In fetuses of Set 5, 13 different rows could be counted; in
the anterior ten rows, each of the rings consisted of two rows of scales which is the
pattern present in adult armadillos.

Initially, the development of the forefeet seemed more rapid than that of the hind
feet. In fetuses of Set 1, the four digits of the forefeet were weU separated, with some
webbing at the bases of the digits. The middle digits were the largest, and the
anlages for the claws were present. In fetus Set 2, the webbing was absent and the
anlages for the claws had ventral grooves. In fetuses of Set 3, toe-pad like
enlargements of the claws were apparent, and the anlages for the scutes (in rows)
became obvious. The two middle toes were considerably larger than the first digits.
In fetuses of Set 5, the individual scutes were distinguishable; the scutes were
partially overlapping. The pad-hke structures at the tips of the digits were dimin-
ished in relative size; the bases of the claws showed some pigmentation. In fetus Set

6, the claws almost were completely pigmented. The development of the hind feet
followed about the same patterns as that of the forefeet. However, in the first four
sets of fetuses, the development of the hind feet was less rapid than that of the
forefeet. In addition, five digits were present on the hind feet; the third toe was
decidedly the largest. In fetus Set 4, anlages of scutes were present on the undersides
of the hind feet; fetuses of Set 5 had fewer scute anlages.

Acknowledgments

We extend our sincere gratitude to Rudolph Meier, who did a preliminary
analysis of part of the home range data and for his excellent descriptions of six of
the seven sets of fetuses. Also, we wish to thank the following individuals for their
help with the field work associated with the study of home range: Frances Miller
Cashner, Robert C. Cashner, Anthony Laska, Kaneta Lott, Jayson S. Suttkus,
Ramona L. Suttkus, Frank I. Thomas, and Bruce A. Thompson. Lastly, we thank
Craig Hood and an anonymous reviewer for their critical review of the manuscript.

Literature Cited

Breece, G. a. and J. L. Dusi. 1985. Food habits and home ranges of the common long-nosed
armadillo Dasypus novemcinctus in Alabama, pp. 419^27. In: G. G. Montgomery (ed.). The
evolution and ecology of armadillos, sloths, and vermilinguas. Smithsonian Institution
Press, Washington, D.C., 451 pp.

1999]

Observations on Nine-Banded Armadillo

21

Case, R. M. 1978. Interstate highway road-killed animals: a data source for biologists. Wildl.
Soc. Bull. 6: 8-13.

Choate, J. R., J. K. Jones, Jr., and C. Jones. 1994. Handbook of mammals of the south-central
states. Louisiana State University Press, Baton Rouge, 304 pp.

Clark, W. K. 1951. Ecological life history of the armadillo in the eastern plateau region. Amer.
Midi. Nat. 46: 337-358.

Denson, R. D. 1979. Aggression and tumbling among armadillos. Southwest. Nat. 24: 697-698.
Enders, a. C. 1966. The reproductive cycle of the nine-banded armadillo {Dasypus novemcinctus).
pp. 295-320. In: I. W. Rowland (ed.). Comparative biology of reproduction in mammals.
Academic Press, London.

Fitch, H. S., P. Goodrum, and C. Newman. 1952. The armadillo in the southeastern U.S. J.
Mammal. 33: 21-37.

Galbreath, G. j. 1982. Armadillo, pp. 71-79. In: J. A. Chapman and G. A. Feldhamer (eds.).
Wild mammals of North America. Johns Hopkins University Press, Baltimore, Mary-
land, 1147 pp.

Hamlett, G. W. D. 1933. Polyembryony in the armadillo; genetic or physiological? Quart. Rev.
Biol. 8: 348-358.

FLamlett, G. W. D. 1935. Delayed implantation and discontinuous development in mammals.
Quart. Rev. Biol. 10: 432-447.

Jones, C. and J. Pagels. 1968. Notes on a population of Pipistrellus subflavus in southern
Louisiana. J. Mammal. 49: 134-139.

Kalmbach, E. R. 1943. The armadillo: its relation to agriculture and game. Texas Game, Fish
and Oyster Commission, Austin, 60 pp.

Layne, j. N. and D. Glover. 1977. Home range of armadillo in Florida. J. Mammal. 58: 411-413.
Loughry, W. j. and C. M. McDonough. 1996. Are road kills valid indicators of armadillo
population structure? Amer. Midi. Nat. 135: 53-59.

Lowery, G. H., Jr. 1936. A preliminary report on the distribution of the mammals of Louisiana.
Proc. Louisiana Acad. Sci. 3: 11-39.

Lowery, G. H., Jr. 1943. Checklist of the mammals of Louisiana and adjacent waters. Occ.

Papers Mus. Zool. Louisiana State Univ. 13: 213-257.

Lowery, G. H., Jr. 1974. The mammals of Louisiana and its adjacent waters. Louisiana State
University Press, Baton Rouge, 565 pp.

McBee, K. and R. j. Baker. 1982. Dasypus novemcinctus. Mammalian Species 162: 1-9.
McDonough, C. M. 1994. Determinants of aggression in nine-banded armadillos. J. Mammal.
75: 189-198.

McDonough, C. M. 1997. Pairingbehavior of the nine-banded armadillo {Dasypus novemcinctus).
Amer. Midi. Nat. 138: 290-298.

McDonough, C. M. and W. J. Loughry. 1997a. Influences on activity patterns in a population
of nine-banded armadillos. J. Mammal. 78: 932-941.

McDonough, C. M. and W. J. Loughry. 1997b. Patterns of mortality in a population of nine-
banded armadillos, Dasypus novemcinctus. Amer. Midi. Nat. 138: 299-305.

Newfang, D. M. 1947. Sex differentiation in the nine-banded armadillo, Dasypus novemcinctus.
J. Morphol. 81: 283-316.

Newman, H. H. 1913. The natural history of the nine-banded armadillo in Texas. Amer. Nat.
47: 513-539.

Newman, H. H. and J. T. Patterson. 1910. The development of the nine-banded armadillo from
primitive streak to birth, with special references to the question of polyembryony. J.
Morphol. 21: 359^23.

Patterson, J. T. 1913. Polembryonic development in Tatusia novemcinctus. J. Morphol. 24: 559-
684.

Storrs, E. E., H. P. Burchfield, and R. J. W. Rees. 1989. Reproduction delay in the common long-

22

Tulane Studies in Zoology and Botany

[Vol. 31, No. 1

nosed armadillo, Dasypus novemcinctus L. pp. 535-548. In: K. H. Redford and J. F. Eisenberg
(eds.). Advances in Neotropical mammalogy. Sandhill Crane Press, Inc., Gainesville,
Florida, 614 pp.

SuTTKUS, R. D. AND C. JoNES. 1991. Observations on winter and spring reproduction in
Peromyscus leucopus (Rodentia: Muridae) in southern Louisiana. Texas J. Sci. 43:179-189.

Taber, F. W. 1945. Contributions on the life history and ecology of the nine-banded armadillo.
J. Mammal. 26: 211-226.

Talmage, R. V. AND G. D. Buchanan. 1954. The armadillo (Dasypus novemcinctus). A review of
its natural history, ecology, anatomy and reproductive physiology. Rice Institute Pam-
phlet, Monographs in Biology 41: 1-135.

Taylor, W. P. 1946. Armadillos abundant in Kerr Co., Texas. J. Mammal. 27: 273.

Thomas, W. D. 1980. Summer movement of a male armadillo in central Texas. Texas J. Sci. 32:
363-365.

Warkentin, M. j. 1968. Observations on the behavior and ecology of the nutria in Louisiana.
Tulane Stud. Zool. Bot.l5: 10-17.

SYSTEMATIC REVIEW OF SUBGENUS FUSCATELUM OF
ETHEOSTOMA WITH DESCRIPTION OF A NEW SPECIES FROM
THE UPPER BLACK WARRIOR RIVER SYSTEM, ALABAMA

Henry L. Bart, Jr.

AND

Michael S. Taylor

Tulane University Museum of Natural History
Belle Chasse, Louisiana 70037

Abstract

The subgenus Fuscatelum was thought to contain a single, widespread, lowland
species, the goldstripe darter, Etheostoma parvipinne. However, comparison of popu-
lations from throughout the goldstripe darter's reported range has revealed that
populations in the upper Black Warrior River system represent a new species. The
new species has lower scales counts than the goldstripe darter, lacks the distinctive
gold stripe, and has a less intensive pattern of brown pigmentation along its sides.
There are concerns about the conservation status of the new species, because only
three rather widely separated populations are known, and the species cannot be
reliably taken in any of these areas. In addition to describing the new species, we
redescribe the goldstripe darter and review morphological attributes of subgenus
Fuscatelum.

Introduction

The goldstripe darter, Etheostoma parvipinne Gilbert and Swain, inhabits small,
vegetated Coastal Plain streams across a broad area of the Gulf and extreme lower
Atlantic slopes. Maps of the distribution of this species show the species occurring
well above the Fall Line only in the Black Warrior River system in Alabama (Rohde,
1980; Page, 1983, Figure 1). Careful study of these populations has revealed that they
represent a distinct species.

In this paper we describe the new species, redescribe E. parvipinne based on
material from throughout its geographic range, and review morphological attributes
and systematics of subgenus Fuscatelum.

Materials and Methods

Material used in this study was obtained from fish collections at Auburn Univer-
sity (AUM), Geological Survey of Alabama (GSA), Southern Illinois University (SIU),
Tulane University Museum of Natural History (TU), University of Alabama (UAIC),
and the University of Texas Memorial Museum (TNHC). Paratypes of the new
species were deposited at Academy of Natural Sciences, Philadelphia (ANSP), Illi-
nois Natural History Survey (INHS), University of Alabama, University of Florida

Tulane Studies in Zoology and Botany 31: 23-50. 1999.

23

24

Tulane Studies in Zoology and Botany

[VoL 31, No. 1

Figure 1. Map of the distribution of darters in subgenus Fuscatelum in relation to the
Fall Line (heavy line). Open circles within dashed line represents known popula-
tions of E. phytophilum in the upper Black Warrior River system of Alabama. Solid
circles mostly below the Fall Line show the distribution of £. parvipinne. (Map mod-
ified from Richards, 1963).

(UF), University of Michigan (UMMZ), and the United States National Museum of
Natural History (USNM).

Counts and measurements follow Hubbs and Lagler (1958) except that trans-
verse body scales were counted from the origin of the anal fin diagonally upward to
the base of the spinous dorsal fin, and only branched rays in the caudal fin were
counted. The great range in variation of lateral-line and pored lateral-line scale
counts made it difficult to format the tables to fit on a page. Data for these counts are
reported in two-scale intervals simply to reduce the width of Tables 1 and 2. Thus,
counts of 38 and 39 in Table 1 are tabulated under "'38-39", and so on. Vertebral
counts were made from radiographs of 20 specimens from a Tombigbee River
system population of £. parvipinne (TU 76977), and 27 specimens from the Turkey
Creek (AUM 27017) and Clear Creek (TU 162757) populations of the new species.

Data for 15 body measurements, including standard length (SL), were gathered
for samples of 17 males and 13 females from the lower Black Warrior population of
Etheostoma parvipinne (GSA 6848.06, GSA 6855.06, GSA 7204.09, GSA 7249.08, GSA
7384.11, GSA 7993.11 and TU 168007, all Tuscaloosa Co., AL) and 15 males and 15
females from Turkey and Clear Creek populations of £. phytophilum (AUM 18747,

1999]

New Species of Fuscatelum

25

AUM 21148, AUM 21155, AUM 27017, TU 163757, TU 176230 and Holotype, TU
187503). Measurements were made with needle point calipers to the nearest 0.1 mm.
Data for 14 of the body measurements were converted to ratios of standard length.
The ratios were arcsine transformed to normalize ratio distributions. The trans-
formed ratios were subjected to ANOVA to test for differences between sexes within
species, and between species within each sex.

The arcsine-transformed mensural data were also subjected to principal compo-
nents analysis (PCA) to characterize shape differences between the two species.
PCA was performed on the covariance matrix because the data were not standard-
ized. All statistical analyses were performed on a personal computer with Microsoft
Excel 97 and Statistical Analysis System (SAS) Inc., software. Names used for species
associates of the new species follow Robins et al. (1991).

Subgenus Fuscatelum, Page, 1981

Type Species: Etheostoma parvipinne Gilbert and Swain (in Gilbert, 1887).

Diagnosis: Subgenus Fuscatelum, as diagnosed by Page (1981), is distinguished by
the absence of bright breeding colors (red, blue, green or yellow) on body or fins,
presence of a frenum, and presence of tubercles on the anal fin of breeding males
(Page, 1981). As pointed out by Shaw (1996), the combination of these characters
distinguishes E. parvipinne from all other Etheostoma except E.fusiforme. The addition
of caudal peduncle depth/SL ratio greater than 0.11 (vs. less than 0.11 in E.fusiforme)
distinguishes Fuscatelum from all Etheostoma inclusive of E.fusiforme.

Description: The following is a summary of the morphological characteristics of
subgenus Fuscatelum, based on the new species and all populations of E. parvipinne
examined in this study. For completeness, some of this description is repeated in the
species accounts below.

Lateral line is straight and nearly (91-98%) complete. The nape, cheek, opercle,
prepectoral area, breast and anterior belly are fully scaled (cheek scales sometimes
partially embedded). A broad premaxillary frenum is always present. Branchioste-
gal membrane is moderately to broadly connected across the isthmus.

The head is brown dorsally with an even darker brown horizontal stripe running
on either side from top of opercle through eye, across snout and onto upper jaw. A
thick dark brown subocular bar extends down onto underside of the head. Side and
underside of head below lateral stripe light yellow-brown to whitish-yellow with
only a scattering of melanophores. Lower jaw and chin also light yellow-brown to
whitish-yellow with darker brown edging or flecking. Opercle brown; humeral spot
weakly developed. Brown pigment on body diffuse, giving the sides a more or less
continuous brown cast, or producing a dense mottling of brown that dissipates
dorsally and ventrally. When most intense, the brown pigment forms a series of 8-
10 irregular dashes or blotches along the side of body, which may be extended into
a series of vertical bars. Dorsum darker than ventrum but lighter than sides.

Medial fins have dark brown pigment forming a varying number of wavy bands.
The bands are most numerous on the caudal fin and least developed on the anal fin.

26

Tulane Studies in Zoology and Botany

[VoL 31, No. 1

Most of the pigment forming the bands is along the edges of fin rays. Fin membranes
adjacent to the bands have only a scattering of melanophores, making the bands
appear interrupted. The base of the caudal fin has two to four dark brown spots of
varying intensity. Paired fins lack pigment in non-breeding individuals.

Breeding males differ from the above pattern mainly in having black melano-
phores scattered over all of the body and fins (least so on the pectoral), giving the
body a dusky cast. Black pigment is concentrated on the anterior interradial mem-
brane of the spinous dorsal fin in mature males, producing a distinct blotch. The rest
of the first dorsal fin is dusky except for a white (pigmentless) strip along the distal
margin. The soft dorsal, anal, pelvic and (less so) pectoral fins also develop black
melanophores on the interradial membranes, making the fins appear dusky. Breed-
ing tubercles develop only on anal soft rays of nuptial males. Female genital papillae
are bulbous at the base and extended into a tube distally . Papillae of males very short
and tubular, though wider at base than at tip.

As mentioned above, the caudal peduncle in subgenus Fuscatelum is relatively
deep (more than 0.118 when expressed as a proportion of standard length), most like
the situation in subgenera Catonotus and Nothonotus (Page 1981, Brooks 1993). The
snout is very short (0.054-0.58 of standard length for both E. parvipinne and E. phy-
tophilum). In this regard, subgenus Fuscatelum is most like subgenus Boleichthys
{sensu Page, 1981). Pectoral fin length (0.205-0.224 of standard length) and soft dor-
sal fin base length (0.14 of standard length) are short relative to other Etheostoma
(Page, 1981; Brooks, 1993). In this regard subgenus Fuscatelum is most like £, exile
(subgenus Boleichthys, sensu Page, 1981; subgenus Oligocephalus, sensu Bailey and
Etnier, 1988). Finally, the caudal fin is relatively long (0.212-0.229 of standard length),
similar to the condition in subgenera Boleichthys, Boleosoma, Catonotus, Psychromas-
ter, Villora, and Ozarka.

Systematics: Etheostoma parvipinne has been traditionally allied to darters in sub-
genus Oligocephalus. Richards (1963) concluded that E. parvipinne formed a natural
group with E.fricksium and E. mariae, then also classified in subgenus Oligocephalus.
Richards (1963) speculated that E. parvipinne was derived from an E. fricksium-like
prototype through advancements such as reduction in dorsal fin color and interrup-
tion of the supratemporal canal. Page (1981) removed E. parvipinne from subgenus
Oligocephalus, and established for it the subgenus Fuscatelum to emphasize its lack
of bright breeding colors. Page (1981) considered Fuscatelum to be part of a highly
evolved group of darters, which included members of the subgenera Nothonotus,
Belophlox (E.fricksium and E. mariae), Ozarka (E. punctulatum and allies), Oligoceph-
alus, Catonotus, and perhaps Villora (E. edwini). Bailey and Etnier (1988) commented
that Fuscatelum shares a number of traits with subgenus Ozarka, including tubercu-
late anal fin, a distinctively depigmented nape, and late winter habits which suggest-
ed reproduction in "Temporarily flooded fields" as seen in subgenus Ozarka.

No consistent hypothesis of relationships has emerged from three phylogenetic
analyses involving subgenus Fuscatelum (Shaw, 1996; Wood and Mayden, 1997;
Turner, 1997). Shaw (1996) analyzed 141 osteological, external morphological, and
behavioral characters for 82 species of darters (genera Crystallaria, Ammocrypta, Per-
cina, Etheostoma) and three Eurasian percids. In this analysis, Fuscatelum (E. parvipinne)

1999]

New Species of Fuscatelum

27

was sister to all other Etheostoma (76 species representing all traditionally allied
subgenera, except loa), suggesting that Fuscatelum is the most ancestral subgenus in
genus Etheostoma.

Wood and Mayden (1997) analyzed allozyme variation at 32 presumptive gene
loci for 26 species of darters (genera Crystallaria, Ammocrypta, Percina, and Etheosto-
ma) and three other percids. The most parsimonious solution was a tree based on
allozyme frequencies with a topological constraint among outgroups consistent
with previously published work. Subgenus Fuscatelum (E. parvipinne) occupied a
more derived position in this phytogeny than in the analysis by Shaw (1996), and
was sister to a group consisting of Ozarka (E. cragini )+ Catonotus (E. flabellare + E.
crossopterum). This group was sister in turn to subgenus Boleichthys (E. gracile + E.
microperca). Subgenus Belophlox was not represented in the analysis.

Turner (1997) analyzed sequence data from the mitochondrial DNA control re-
gion. Again, subgenus Fuscatelum (E. parvipinne) was more derived than in the anal-
ysis by Shaw (1996). It was most closely related to subgenus Villora in the most
parsimonious topology, and was in a group of darters — together with representa-
tives of the subgenera Belophlox (E. mariae), Boleichthys (E. exile, E. gracile), Ozarka (E.
pallididorsum), Villora (E. edwini), and Oligocephalus (in part) — which lacked a tan-
dem repeat at the proline tRNA end of the control region.

Fuscatelum shares a depigmented lateral line and a similar pattern of blotches or
dashes along the side (most like E. mariae in this regard) with darters in subgenus
Belophlox (Richards, 1963). The spinous dorsal fin of breeding males of E, parvipinne
and E. phytophilum has a clear to milky-white marginal band with black pigment
concentrated in the anterior interradial membranes, a situation most like that seen
in E. mariae of subgenus Belophlox (Richards, 1963) and in some Nothonotus, but
without the other band(s) of color seen in the spinous dorsal fin in these groups.
Shaw (1996) mistakenly coded the spinous dorsal fin marginal band color as black
in Fuscatelum (state 3), rather than opaque /tan /milky-yellow (state 4), as interpreted
for E. mariae, members oi Nothonotus and E. blennioides. She attributed character state
0 (marginal band clear) to E.fricksium, when in fact the margin of the spinous dorsal
in breeding males of this species is clear to opaque- white with a stippling of black
melanophores.

Etheostoma phytophilum, new species
Rush Darter
Figure 2a, b

Etheostoma parvipinne. Mettee 1978:144 (account. Fishes of the Birmingham-
Jefferson County region of Alabama with ecological and taxonomic notes);

Rohde 1980:680 (distribution, in part); Page 1983:245 (distribution, in part);

Kuehne and Barbour 1983:141 (distribution, in part); Mettee et al. 1989a:181
(distribution, in part); Mettee et al. 1989b:144 (distribution, in part — Atlas);

Page and Burr 1991:413 (distribution, in part — Map 359); Mettee et al.

1996:654-655 (distribution, in part, reference to distinct form in Turkey Creek
system, Jefferson Co.).

FiOLOTYPE: TU 187503: Spring run trib. to trib. to Turkey Creek along Alabama

28

Tulane Studies in Zoology and Botany

[Vol. 31, No. 1

Highway 79, Pinson, Jefferson Co, AL, 1 April 1994, H. Bart, J. Buckley and C.
Gradney, collectors.

ParatoPOTYPES: TU 176230 (3, collected with holotype); AUM 18747 (1), 11
March 1979; AUM 21155 (2), 8 July 1979; AUM 27017 (17, distributed as follows:
AUM 27017 (5), ANSP 177501 (2), INHS 48900 (2), UAIC 1295.01 (2), UP 110739 (2),
UMMZ 234787 (2), USNM 351328 (2)) 6 November 1990; UAIC 11064.01(2), 17 April
1994.

Other Paratypes: AUM 21148 (1): AL, Jefferson Co., Penny's Spring, 8 July
1979; TU 60078 (1): AL, Jefferson Co., Turkey Creek , 20 October 1969.

Additional material examined, not designated as types: Wildcat Branch, trib. to
Clear Creek, Winston Co., AL: AUM 27031 (11), 27 November 1990; TU 162757 (10), 11 April
1992; TU 167887 (2), 26 February 1993; TU 167895 (3), 27 February 1993; UAIC 10645.01 (in
part 16), 25 February 1993; Cove Spring, Etowah Co., AL: UAIC 4139.04 (1), 4 May 1975.

Diagnosis: Etheostoma phytophilum is a member of subgenus Fuscatelum as diag-
nosed herein and by Page (1981). It differs from the only other species in the subge-
nus, £. parvipinne, in usually having 47 or fewer lateral line scales (lateral line scales
usually 48 or higher in £. parvipinne), usually 13 or fewer transverse scales (tranverse
scales usually 14 or higher in £. parvipinne), and 22 or fewer caudal peduncle circum-
ferential scales (caudal penduncle scales usually 23 or higher in £. parvipinne). Lat-
eral line not distinctly depigmented at any stage (lateral line distinctly depigmented,
especially anteriorly, in female and non-breeding £. parvipinne).

Description: Etheostoma phytophilum is a moderate-sized species of Etheostoma: the
average size of the 68 specimens examined in this study (none less than 30 mm SL)
is 40.22 mm SL; the largest known specimen is a 58 mm SL female. Lateral line with
39-51 scales (usually 47 or fewer) of which 32-50 are pored (Tables 1 and 2). Trans-
verse scales 10-14, (usually 13 or fewer. Table 3). Caudal peduncle circumferential
scales 18-22, usually 21 (Table 4). Only 10 of 68 £. phytophilum examined (15%) had
more than 47 lateral line scales, only 2 of 68 (3%) had more thanl3 transverse scales,
and none had more than 22 caudal peduncle scales. The combination of lateral line
scales 47 or fewer, transverse scales 13 or fewer, and caudal peduncle scales fewer
than 22, was seen in only 25 of 376 specimens (6.7%) of £. parvipinne examined from
through the species' geographic range.

First dorsal fin has 8 to 11 (usually 10) spines and is entirely separate from the
second dorsal fin, which has 9-13 (usually 10-11) soft rays (Tables 5 and 6). The right
side pectoral fin has 14-16 (usually 15) rays (Table 7). The pelvic fin always has one
spine and six rays. The anal fin almost always has two spines (although one may be
reduced) and 7 to 10 (usually 9) soft rays (Table 8). The caudal fin has 11-15 (usually
13) branched rays (13-15 principal rays. Table 9). Vertebrae number 35-37 (x = 36.4).

Cephalic lateralis system is as follows: lateral canal complete with 5 pores, supra-
temporal canal incomplete with two pores; supraorbital canal complete with four
pores; a coronal pore is usually present; infraorbital canal complete with 5-10 (usu-

1999]

New Species of Fuscatelum

29

Figure 2. a) Etheostoma phytophilum, holotype (TU 187503), breeding male, 44 mm SL,
spring run in the Turkey Creek system; b) E. phytophilum, breeding male, 43 mm SL,
Wildcat Branch trib to Clear Creek (TU 167895).

ally 8) pores (Table 10); preoperculomandibular canal complete with 9-12 (usually
10) pores (Table 11).

Color in life khaki (light yellowish brown) to golden brown, with a pattern of
darker gray-brown pigment overlaying a yellowish to golden ground color. Head
gray-brown dorsally with a dark brown stripe running horizontally from top of
opercle, through eye, across snout, and onto upper jaw. A dark brown vertical bar
runs below eye to underside of the head. Brown pigment on body diffuse, giving the
sides a brown cast or producing a dense mottling of brown that dissipates dorsally
and ventrally . At its most intense, the brown pigment forms a series of 8-10 irregular
dashes along the side of body. Dorsum darker than ventrum but lighter than sides.

Even at its most intense, the brown pigment on the sides of the rush darter is
seldom as intense as in the goldstripe darter, and mid lateral dashes are never drawn
into vertical bars, as frequently seen in breeding males of E. parvipinne. The distinct
gold stripe typical of E. parvipinne is never well developed in E. phytophilum. Only
a very narrow strip may be depigmented along the anterior portion of the lateral
line, above the midlateral dashes.

Brown pigment occurs at intervals along the edges of fin rays in median fins,
producing a varying number of wavy bands. Fin membranes adjacent to the bands
have a scattering of melanophores that make the bands appear interrupted. The base

30

Tulane Studies in Zoology and Botany

[Vol. 31, No. 1

of the caudal fin has two to four dark brown spots of varying intensity. Paired fins
lack pigment in non-breeding individuals.

Breeding males of E. phytophilum have black melanophores scattered over all of
the body and fins (least so on the pectoral fin), making the body appear dusky. A
concentration of black melanophores forms on anterior interradial membranes of
the spinous dorsal fin in mature males, producing a distinct blotch. The rest of the
first dorsal fin is dusky except for a white (pigmentless) strip along the distal margin.
The soft dorsal, anal, pelvic and (less so) pectoral fins also develop a dusky appear-
ance due to the presence of black melanophores on interradial membranes. Breeding
tubercles develop only on anal soft rays of nuptial males, female genital papillae are
bulbous at the base and extended into a tube distally, and genital papillae of males
are short and tubular, though wider at base than at tip.

Variation: The population in the Clear Creek system has slightly higher average
counts of lateral line and pored lateral line scales than the Turkey Creek population.
The Clear Creek population also has higher modal counts of right pectoral (16) and
branched caudal fin rays (14) than the Turkey Creek population (modes of 14 or 15,
and 13, respectively). Dorsal soft rays are modally lower in the Clear Creek popu-
lation (10) than in the Turkey Creek population (11 or 12). None of the differences
is statistically significant. In the Clear Creek system population, breeding males tend
to have more mottling on the lower sides than in males from the Turkey Creek
population. Clear Creek breeding males also develop a yellowish cast in the spinous
and soft dorsal, pectoral and caudal fins (Figure 2b). We collected only two males in
high reproductive stage in the Turkey Creek system (the holotype and a paratype
form the same collection), and neither of them had yellow pigment in the fins. An
older March-collected specimen was too faded to discern fin coloration. The Cove
Creek population cannot be adequately compared because it is represented by a
single specimen.

Distribution: Etheostoma phytophilum is endemic to upland portions (Appalachian
Plateau and Valley and Ridge provinces) of the Black Warrior River system in Ala-
bama. Only three rather widely separated populations are known (Figures 1 and 3):
one in streams in the Clear Creek system, a tributary of the Sipsey Fork of the Black
Warrior River in Bankhead National Forest; one in springs tributary to Turkey Creek
north of Birmingham, AL; and one in a spring around Little Cove Creek near Gads-
den, AL (a single specimen known from this location). The latter two streams are
tributary to the Locust Fork of the Black Warrior River. Considering the area of the
upper Black Warrior River system over which E. phytophilum occurs (Figure 3), and
the large number of springs in this area, it would seem likely that other populations
exist. The sampling locations depicted in Figure 3 give the impression that the upper
Black Warrior Basin has been extensively sampled. However, the vast majority of
the sampling locations are on streams larger than those that support E. phytophilum.
Small streams and spring runs in the upper Black Warrior system need to be sur-
veyed to determine the full extent of the rush darter's distribution.

Habitat and Biology: Etheostoma phytophilum is an uncommon species that is

1999]

New Species of Fuscatelum

31

Figure 3. Map of the Black Warrior River system in Alabama, showing collection
sites (open circles) and the distributions of Etheostoma parvipinne (solid circles), E.
phytophilum (solid triangles, with circled star representing the type locality) in rela-
tion to the Fall Line (heavy broken line). Map modified from Mette et al. (1989a).

32

Tulane Studies in Zoology and Botany

[VoL 31, No. 1

Figure 4. Type locality of Etheostoma phytophilum, showing the proximity to Ala-
bama Highway 79. Note the abundant stream-side vegetation (prefered habitat of
the rush darter) and the evidence of spring influence (watercress in mid channel).

usually taken in low numbers (fewer than 10 specimens). We have had the most
success collecting the species from the root masses of emergent vegetation in very
shallow water along the margin of small spring-fed streams (Figure 4). Vegetation
types include rushes (Juncus spp.), dock {Rumex spp.), and a variety of unidentified
grasses (Gramineae). Although water along the margins of these streams is shallovv^,
it is typically clear, cool and flowing. The rush darter is never taken in the thick
growths of water cress found both along the sides and at mid-channel in spring runs
in the Turkey Creek system. The two streams in the Clear Creek system (Sipsey Fork
of the Black Warrior) where rush darters have been collected appear to be less
spring-influenced than streams in the Locust Fork, and the streamside vegetation is
thick, tall and dominated by rushes and grasses. Cove Spring in Etowah Co., AL,
where a single rush darter was collected, flows into a reach of Cove Creek which is
impounded by beaver. Lower reaches of Cove Spring Run appear to offer good
habitat, but are difficult to access and sample effectively.

Males taken in March and April are tuberculate and exhibit dusky breeding
pigmentation. Females collected in March and early April have enlarged ovaries

1999]

New Species of Fuscatelum

33

with large ripening oocytes. Spawning and other aspects of life history are presently
unknown.

In the Turkey Creek system, species associates of the rush darter include Campos-
toma oligolepis, Cyprinella callistia, Luxilus chrysocephalus, Semotilus atromaculatus,
Lepomis cyanellus, Etheostoma whipplei artesiae and Coitus carolinae. In the Clear Creek
system, species associates include Ichthyomyzon gagei, Campostoma oligolepis, Luxilus
chrysocephalus, Nocomis leptocephalus, Notropis baileyi, Semotilus thoreauianus, Erimyzon
oblongus, Lepomis cyanellus, and Percina nigrofasciata.

Concerns about the conservation status of E. phytophilum are increasing because
of its sporadic occurrence in streams in the Clear Creek and Turkey Creek systems,
increasing threats to habitat quality in the latter, and our inability to collect the
species in Cove Spring. The species is known from three locations in the Turkey
Creek system, but cannot be reliably collected at any of these locations. The type
locality is a spring run that forms part of the road side drainage system along Ala-
bama Highway 79, southwest of Pinson, AL. The stream has been modified exten-
sively and receives runoff from several business. Mainly because of the spring influ-
ence, upper reaches of the spring run continue to offer what appears to be favorable
habitat. Specimens of E. phytophilum have also been taken in a spring north of Pinson
on two occasions (a total of two specimens, only one of which is archived). The
watercress darter E. nuchale was introduced into this spring by W. Mike Howell in
1988 to reduce a then looming threat of extinction of this species due to destruction
of its native habitat (Mettee et al., 1989a). The watercress darter appears to be doing
well in the spring; the rush darter is rare.

Etymology: The name phytophilum is derived from the Greek words phytum,
meaning plant or vegetation, and philo meaning to love or loving, and is in refer-
ence to the preference of this species for emergent vegetation along stream margins.
The vernacular name, rush darter, is suggested because rushes (Juncus spp.) are an
obvious component of the stream side vegetation in streams inhabited by E. phyto-
philum.

Etheostoma parvipinne Gilbert and Swain
Goldstripe Darter

Etheostoma (Etheostoma) parvipinne Gilbert and Swain in Gilbert, 1887:59-60
(original description).

Etheostoma squamiceps Bollman, 1886:464-465; Escambia River at Flomaton.
Gilbert, 1891:156,159 (Alabama records: Little Escambia River; 1 mile south of
Pollard; Sandy Creek, 3 miles east of Evergreen), and Escambia River system.
Jordan and Evermann, 1896: 1096 (in part, incorrectly synonymized E.
parvipinne in E. squamiceps).

Claricola squamiceps Jordan et al., 1930: 292 (in part, incorrectly synonymized E.
parvipinne in C. squamiceps).

Poecilichthys parvipinnis. Hubbs and Black, 1941: 9 (validation, not a synonym of P.
squamiceps). Moore and Cross, 1950: 145-146 (confirmation of Hubbs and
Black's, 1941 validation; comparison with P. squamiceps; characters; distribu-
tion). Moore and Rigny, 1952: 7.

34

Tulane Studies in Zoology and Botany

[VoL 31, No. 1

HoloTYPE: USNM 36716, male 43.0 mm SL, small spring branch tributary to the
Black Warrior at Tuscaloosa, AL, summer 1884, C.H. Gilbert and J. Swain, collectors.

Material ExAMiNED:Colorado River Dr.: TNHC 23901 (6): TX, Bastrop Co., Alum Creek;
TNHC uncataloged (8): TX, Bastrop Co., Alum Creek. Brazos River Dr.: TNHC 11436 (1): TX,
Leon Co., Long Hollow Creek. San Jacinto River Dr.: TNHC 1205 (13): TX, Montgomery Co.,
Trib. to Peach Creek; TNHC 1346 (1): TX, Polk Co., Trib. to Trinity River; TNHC 2614 (1): TX,
Polk Co., Morgan Creek; TNHC 2722 (1): TX, Polk Co., Bluff Creek; TNHC 2749 (13): TX,
Walker Co., Gum Branch; TU 74010 (1): TX, Montgomery Co., Stewart Creek. Sabine River
Dr.: TNHC 1523 (1): TX, Smith Co., Trib. to Harris Creek; TNHC 2455 (1): TX, Anderson Co.,
Copperas Creek; TU 60256 (3): LA, Beauregard Par., Hoosier Creek; TU 86888 (1): TX, New-
ton Co., Big Cow Creek; TU 86935 (2): LA, Beauregard Par., Anacoco River; TU 103251 (5):
TX, Tyler Co., Drakes Branch. Red River Dr.: TNHC 2454 (3): TX, Morris Co., Boggy Creek;
TNHC 3654 (8): TX, Upshur Co., Trib. to Big Cypress Bayou; TNHC 3928 (11): TX, Bowie Co.,
Caney Creek; TNHC 12381 (1): TX, Upshur Co., Jones Creek. Obion River Dr.: SIU 3335 (16):
TN, Weakley Co., Old Knob Creek; SIU 4126 (14): TN, Weakley Co., Old Knob Creek. Ten-
nessee River Dr.: TU 87968 (1): MS, Alcorn Co., Sevenmile Creek; TU 88014 (5): MS, Tisho-
mingo Co., Little Yellow Creek; TU 88580 (7): MS, Tishomingo Co., Little Yellow Creek; TU
88656 (1): MS, Tishomingo Co., Caney Creek; TU 88666 (4): TN, McNairy Co., Trib. to
Tennessee River; TU 88689 (1): TN, McNairy Co., Owl Creek; TU 89326 (1): TN, Hardeman
Co., Hatchie River; TU 153375 (2): MS, Tippah Co., Hatchie River. Lower Mississippi River:
TU 33049 (2): MS, Lafayette Co., Trib. to Yacona River; TU 55713 (1): MS, Jefferson Co., North
Fork Coles Creek; TU 56552 (1): MS, Wilkinson Co., Little Buffalo Creek; TU 57952 (1): MS,
Wilkinson Co., Little Buffalo River; TU 61654 (4): MS, Wilkinson Co., Buffalo Bayou; TU
64850 (2): MS, Copiah Co., Bayou Pierre; TU 67983 (8): MS, Lincoln Co., Homochitto River;
TU 68014 (2): MS, Lincoln Co., Homochitto River; TU 68060 (1): MS, Lincoln Co., Homochit-
to River; TU 69003 (1): LA, West Feliciana Par., Thompson Creek; TU 75460 (1): MS, Hinds
Co., Big Black River; TU 78730 (1): MS, Lincoln Co., Homochitto River; TU 84066 (1): MS,
Copiah Co., Homochitto River; TU 91107 (1): MS, Hinds Co., Tallahala Creek; TU 91201 (4):
MS, Copiah Co., Foster Creek; TU 124195 (1): MS, Lincoln Co., Homochitto River. Pearl
River Dr.: TU 17652 (3): MS, Marion Co., Pearl River; TU 26537 (1): MS, Marion Co., Harvey
Creek; TU 26601 (1): LA, St. Tammany Par., Talley's Creek; TU 28920 (1): MS, Marion Co.,
Trib. to Pearl River; TU 36048 (2): LA, Washington Par., Talisheek Creek; TU 50186 (1): LA,
Washington Par., Pearl River; TU 53615 (1): MS, Simpson Co., Strong River; TU 54389 (2):
MS, Simpson Co., Strong River; TU 55521 (2): MS, Simpson Co., Strong River; TU 56069 (4):
MS, Simpson Co., Strong River; TU 56489 (1): MS, Simpson Co., Strong River; TU 56611 (2):
MS, Simpson Co., Strong River; TU 57437 (1): LA, Washington Par., Pearl River; TU 58215
(3): MS, Simpson Co., Strong River; TU 59264 (2): MS, Simpson Co., Strong River; TU 62465
(1): MS, Simpson Co., Strong River; TU 64472 (1): MS, Simpson Co., Strong River; TU 170514
(1): MS, Lincoln Co., Fair River. Pascagoula River Dr.: TU 4284 (19): MS, Jasper Co., Minnows
Inc. Hatchery; TU 55480 (1): MS, Covington Co., Okatoma Creek; TU 56653 (1): MS, Jones
Co., Leaf River; TU 66040 (2): MS, Jones Co., Leaf River. Tombigbee River/lower Black
Warrior: AUM 19506 (1): AL, Greene Co., Gainesville Lake; GSA 6607.05 (3): AL, Tuscaloosa
Co., Little Tyro Creek; GSA 6619.03 (2): AL, Tuscaloosa Co., Little Tyro Creek; GSA 6848.06
(10): AL, Tuscaloosa Co., Little Tyro Creek; GSA 6855.06 (8): AL, Tuscaloosa Co., Little Tyro
Creek; GSA 7000.07 (2): AL, Tuscaloosa Co., Little Tyro Creek; GSA 7204.09 (2): AL, Tusca-
loosa Co., Little Tyro Creek; GSA 7249.08 (3): AL, Tuscaloosa Co., Kepple Creek; GSA
7384.11 (7): AL, Tuscaloosa Co., Kepple Creek; GSA 7393.11 (7): AL, Tuscaloosa Co., Kepple
Creek; GSA 7496.14 (2): AL, Hale Co., German Creek; GSA 7560.11 (2): AL, Greene Co.,

1999]

New Species of Fuscatelum

35

Minter Creek; TU 76977 (36): AL, Greene Co., Needham Creek; UAIC 1541.01 (7): AL, Tus-
caloosa Co., seepage pool. Alabama River: AUM 6791 (5): AL, Lee Co., Trib. to Choctafaula
Creek; AUM 10885 (8): AL, Macon Co., Sandy Creek; AUM 19968 (1): AL, Macon Co.,
Opintloco Creek; AUM 23151 (1): AL, Lowndes Co., temporary pond; AUM 27682 (1): AL,
Lowndes Co., Dry Cedar Creek; TU 35057 (11): AL, Dallas Co., Oakmulgee Creek; TU 35142
(1): AL, Dallas Co., Pine Flat Creek; TU 70675 (1): AL, Wilcox Co., Alabama River; TU 163142
(1): AL, Wilcox Co., Alabama River. Perdido River Dr.: TU 23980 (1): AL, Escambia Co.,
Brushy Creek. Escambia River Dr.: AUM 20999 (4): AL, Covington Co., Blue Spring. Choc-
tawhatchee River Dr.: AUM 4693 (30): AL, Barbour Co., Easterling Fish Hatchery. Apalach-
icola River Dr.: AUM 1285 (1): AL, Russell Co., Snake Creek; AUM 2850 (19): AL, Lee Co.,
Trib. to Flake Creek; AUM 9266 (1): AL, Lee Co., Flake Creek; AUM 10749 (3): AL, Lee Co.,
Trib. to Little Uchee Creek; AUM 10815 (3): AL, Lee Co., White's Creek; AUM 13373 (1): AL,
Russell Co., Padgett Branch; AUM 15153 (1): AL, Lee Co., Trib. to Little Uchee Creek; AUM
24067 (1): AL, Lee Co., Trib. to Uchee Creek. Altamaha River Dr.: Uncataloged (1): GA, Butts
Co., Plymale Creek; AUM 10370 (5): GA, Ben Hill Co., Merritts Pond; AUM 14212 (1): GA,
Dekalb Co., Swift Creek.

Diagnosis: A species of subgenus Fuscatelum, as diagnosed herein and by Page
(1981), differing from the only other recognized species in the subgenus, E. phytophi-
lum, in usually having 48 or more lateral line scales (lateral line scales usually 47 or
lower in E. phytophilum), 14 or more transverse scales (tranverse scales usually 13 or
lower in E. phytophilum), and 23 or more caudal peduncle circumferential scales
(caudal penduncle scales usually 22 or lower in E. phytophilum). The lateral line in
females and non-breeding males is distinctly depigmented, especially anteriorly
(not distinctly depigmented in E. phytophilum.).

Description: E. parvipinne is a moderate-sized species of genus Etheostoma, similar
in size to E. phytophilum: the average size of 376 specimens examined in this study
(none less than 30 mm SL) is 38.78 mm SL; maximum size, 55 mm SL. Lateral line
scales 42-66 (usually 48 or more. Table 1), of which 33 to 61 are pored (Table 2);
transverse scale rows 11 to 18 (usually 14 or more. Table 3); caudal peduncle circum-
ferential scales 20-30 (usually 23 or more. Table 4). The combination of 48 or more
lateral line scales, 14 or more transverse scales, and 23 or more caudal peduncle
scales characterizes 93% of the specimens of E. parvipinne examined. Nape, cheeks,
opercle, prepectoral area, breast and anterior belly fully scaled (scales sometimes
partly embedded). First dorsal fin with 7 to 14 (usually 9 or 10) spines (Table 5);
second dorsal fin with 9 to 13 (usually 10 or 11) soft rays (Table 6); right pectoral rays
13 to 17, (usually 15 or 16, Table 7); pelvic fin invariant with one spine and 5 soft rays;
anal fin with two spines (one sometimes reduced) and 7 to 10 (usually 8 or 9) soft rays
(Table 8); caudal fin with 10-16 branched rays (12-16 principal rays. Table 9). Ver-
tebrae 34-38.

Etheostoma parvipinne is nearly identical to E. phytophilum in characters of the
lateral canal (complete with 5 pores), supratemporal canal (incomplete with two
pores); supraorbital canal (complete with four pores); coronal pore (present); in-
fraorbital canal (complete with 3-10 pores. Table 10); and preoperculomandibular
canal (complete with 6-12 pores. Table 11).

Color in life taupe (brownish gray), with darker brown pigment forming a dis-
tinctive pattern over a lighter yellow to whitish body color. Dark brown pigment

36

Tulane Studies in Zoology and Botany

[VoL 31, No. 1

also covers most of upper sides of the head and body. A somewhat darker brown
stripe runs from top of opercle, through eye, along snout and onto upper jaw. A
thick, brown bar runs vertically from eye to underside of the head. Side of head
below the lateral eye stripe light yellow to white, although the entire area may be
flecked with melanophores in densely pigmented specimens. Lower jaw and chin
also light with brown flecking. Brown pigment on body either diffuse or producing
a dense brown mottling that dissipates ventrally. Brown pigment forms a series of
8-10 irregular dashes along the mid sides, which may be drawn into vertical bars in
breeding males. The bars are sometimes confluent, forming a brown mid-lateral
stripe. The lateral line is distinctly depigmented along most of its length just above
the mid-lateral dashes, producing the ''gold stripe" typical of the species. The stripe
is sometimes obliterated by brown pigment on the sides of the body. As many as 10
narrow saddles may cross the top of the back generally in line with the vertical bars.

Medial fins have a varying number of wavy bands of brown pigment. The bands
are most numerous on the caudal fin and least developed on the anal fin. Most of the
pigment forming the bands is along the edges of individual fin rays; fin membranes
adjacent to the bands have only a scattering of melanophores, making the bands
appear interrupted . Two to as many as four dark brown spots may be seen at the base
of the caudal fin. Paired fins without pigment, except in breeding males.

As described for the rush darter and subgenus Fuscatelum, the body and fins of
breeding male goldstripe darters become dusky during the breeding season. A con-
centrated spot of black pigment forms in the anterior portion of the spinous dorsal
fin. The rest of the fin is dusky except for a pigmentless strip along the distal margin.
Breeding males of E. parvipinne differ from E. phytophilum in sometimes having the
midlateral dashes extended into a series of dark vertical bars that are generally in
line with the saddles crossing the back. This pigmentation pattern has been associ-
ated with aggressive encounters (Johnston 1994). When present, the bars obliterate
the gold stripe running along the side. Male breeding tubercles, and male and female
genital papillae are as described for E. phytophilum. No tubercular ridges were ob-
served on the pelvic fin as noted by Richards (1963).

Variation: Scale counts exhibit the greatest variability east to west across the
range of E. parvipinne. Populations from the Colorado River system in Texas, east-
ward to the Alabama River, AL vary more or less as a group, and tend to have higher
means for all scale counts. Counts tend to be slightly lower for populations in upper
portions of the Mississippi River Valley (Obion and Tennessee river systems), and
highly variable for populations in the Tombigbee and lower Black Warrior river
systems. A more pronounced shift toward lower scale counts occurs east of Mobile
Bay, especially Choctawhatchee River to Altamaha River. The counts approach the
condition seen in E. phytophilum. Sample sizes are too small to judge how well the
Perdido and Escambia river populations fit this pattern. A pattern of vicariance
involving divergence of populations east of Mobile Bay from a more widespread,
western group of populations extending to or including the Mississippi River, has
been observed for a number of other groups of fishes (Wiley and Mayden, 1985).

Distribution: Etheostoma parvipinne is primarily restricted to lowland streams on

1999]

New Species of Fuscatelum

37

the Coastal Plain, below the Fall Line. The species occurs from the Colorado River
drainage in Texas, east to the Altamaha River drainage in Georgia (the only Atlantic
Slope population), and in lowland tributaries of the Mississippi River northward as
far as extreme southern Missouri and Kentucky (northern extent of the Mississippi
Embayment, Figure 1). E. parvipinne occurs above the Fall Line only in the Ocmulgee
River system (Altamaha River drainage) the easternmost extent of the distribution.
The Brazos River was believed to be the westernmost limit of the distribution (Ro-
hde, 1980). However, Doyle T. Mosier recently discovered a population in Alum
Creek in the Colorado River drainage. Counts from these specimens are included in
comparisons.

Habitat and Biology: The goldstripe darter generally inhabits very small streams,
spring runs, and seeps, where the water is clear, current is slow to moderate, and the
substrate is silt, sand or clay. It is usually associated with aquatic vegetation. The
goldstripe darter frequently colonizes small ponds, apparently entering these water
bodies from small, vegetated tail waters.

The peak breeding season of the goldstripe darter is March and April but may
extend into May. Johnston (1994) reported on the breeding behavior of specimens
of E. parvipinne captured in April, amid Vallisneria sp. along the margin of a spring
in the Tallahatchie River system. The specimens were transferred to aquaria where
they spawned a short time later. The following account on spawning and early life
history is summarized from Johnston (1994).

Males are nonterritorial but aggressive. During spawning and aggressive inter-
actions with other males, body coloration changes from olive brown (sometimes
with dark blotches) to brown with a series of black vertical bars. The suborbital bar
darkens, the eyes become intensely red, the pelvic and anal fins change from dusky
to black, and the dorsal fin become intensely black.

During aggressive encounters, males perform stationary, lateral displays with
erect dorsal fins. Males chase females, mounting them with the caudal peduncle
along the females' side when the female moves onto a spawning surface. The pair
vibrates as eggs are released. Spawning is promiscuous, with both males and fe-
males spawning with more than one partner. The eggs are strongly adhesive and
stick to a variety of spawning substrates in the aquarium, including plant stems,
leaves, roots, rock and the sides of the aquarium. The species has the spawning
attributes of an egg-attacher; that is, its eggs are relatively small and adhesive, and
the genital papillae of females are long and [apically] tubular.

Fertilized, water-hardened eggs ranged from 0.85 to 1.10 mm in diameter and
hatched in 8 days when incubated at 20°C. Multiple stages of egg development were
observed in the ovary, indicating that more than one clutch of eggs is spawned
during the course of the spawning season. The average number of ripening ova per
ovary was 33. Young-of-the-year specimens (~20 mm TL) were collected in May
with two different size groups of adults (45 to 50 and 55 to 65 mm SL), suggesting
the presence of three year classes.

Etymology: parvipinne is derived from the Latin words parvus, meaning "small"

38

Tulane Studies in Zoology and Botany

[Vol. 31, No. 1

and pinna meaning ''fin", and apparently is in allusion to the very short pectoral and
pelvic fins (Gilbert 1887).

Morphometrics

Univariate comparisons. — Table 12 lists means for 14 body proportions, expressed
as hundredths of standard length, for males and females of E. phytophilum, and
topotypical E. parvipinne from streams tributary the Black Warrior River system in
the vicinity of Tuscaloosa, AL. All of the differences mentioned below are significant
at or below the 0.10 level.

Sexual dimorphism was evident in a number of body proportions, but expression
differed between species. Males of E. phytophilum have higher means for caudal
peduncle length, longest dorsal spine (height of the spinous dorsal fin), and caudal
fin length than females. Males of E. parvipinne have higher means for snout length,
longest dorsal spine and ray length, and pelvic fin length than females. The sexual
dimorphism in body depth noted for E. parvipinne (females deeper bodied than
males) is undoubtedly related to the peak breeding condition of the specimens; i.e.,
the abdomens of females were distended with eggs. Most of the specimens of E.
phytophilum used in this comparison were not in peak spawning condition.

The head in E. phytophilum tends to be proportionately longer and deeper than
that in typical E. parvipinne (males and females). E. phytophilum also has a shorter
caudal peduncle, and longer caudal and pectoral fins than E. parvipinne. Other dif-
'ferences are sex specific. Male E. phytophilum have larger orbits that male E. parvipinne;
however, orbit size does not differ between species in females. Similarly, females of
E. phytophilum have shorter spinous dorsal fins and longer pelvic fins than female E.
parvipinne, but these characters do not differ between males of the two species.

Principal Components Analysis. — Clusters representing E. phytophilum and E.
parvipinne from the lower Black Warrior showed nearly complete separation in the
space define by the first two components derived in the Principal Components
Analysis (PC A) of body proportion data (Figure 5). Characters loading heavily on
PC 1 included head length, and length of the pectoral, pelvic and caudal fins, all of
which tended to be longer in E. phytophilum than in E. parvipinne. Characters loading
heavily on PC 2 included caudal peduncle length, and first and second dorsal fin
height, which tended to be longer and higher in E. parvipinne than in E. phytophilum.

Ackowledgments

For assistance in the field, we thank Royal Suttkus, Joseph Buckley, Clark Grad-
ney, and Mark Bailey. We thank the following individuals for loans of specimens
under their care: Rick May den and Bernie Kuhajda, University of Alabama Ichthy-
ological Collection; Brooks Burr, Southern Illinois University; Maurice F. Mettee Jr.,

PC 2

1999]

New Species of Fuscatelum

39

Figure 5. PCA Polygons representing projections of morphometric data for E.
parvipinne and E. phytophilum on the first two principal components. Body propor-
tions loading heavily on these components and the direction of variation are shown
along the axes.

40

Tulane Studies in Zoology and Botany

[VoL 31, No. 1

Geological Survey of Alabama; and Dean Hendrickson and Allison Anderson Texas
Memorial Natural History Museum. We also thank Jon Armbruster for allowing us
to distribute specimens from a type lot cataloged in the Auburn University Museum
Fish Collection, and Stuart Poss for use of radiographic equipment at Gulf Coast
Research Laboratory. Lastly, we thank Bob Cashner for serving as referee on re-
views, and two anonymous reviewers for editorial suggestions that improved the
quality of the paper.

Literature Cited

Bailey, R. M. and D. A. Etnier. 1988. Comments on the subgenera of darters (Percidae) with
descriptions of two new species of Etheostoma (Ulocentra) from southeastern United
States. Misc. Publ. Mus. Zool. Univ. Michigan 175: 1-48.

Bollman, C. H. 1886. Notes on a collection of fishes from the Escambia River, with
description of a new species oiZygonectes {Zygonectes escambiae). Proc. U.S. Natl. Mus. 9:
462-465.

Brooks, A. P. 1993. A phylogenetic analysis of darter subgeneric relationships (Percidae:
Etheostoma) using morphometries. Unpubl. M.S. Thesis, Auburn Univ., Auburn, AL.
79pp.

Etnier, D. A. and W. C. Starnes. 1993. The fishes of Tennessee. University of Tennessee Press,
Knoxville. 681pp.

Gilbert, C. H. 1887. Descriptions of new and little known etheostomids: Proc. U.S. Natl. Mus.
10: 47-64.

Gilbert, C. H. 1891. Report of explorations made in Alabama during 1889, with notes on the
fishes of the Tennessee, Alabama and Escambia rivers. Bull. U.S. Fish. Comm. 9: 143-159.

Hubbs, C. L. and J. D. Black. 1941. The subspecies of the American percid fish, Poecilichthys
whipplii. Occ. Pap. Mus. Zool. Univ. Michigan 429: 1-27.

Hubbs, C. L. and K. F. Lagler. 1958. Fishes of the Great Lakes region. University of Michigan
Press, Ann Arbor. 213pp.

Johnston, C. E. 1994. Spawning behavior of the goldstripe darter {Etheostoma parvipinne
Gilbert and Swain) (Percidae). Copeia 1994: 823-825.

Jordan, D. S. and B. W. Evermann. 1896. The fishes of North and Middle America. Bull. U.S.
Natl. Mus. 47: 1-3313.

Jordan, D. S., B. W. Evermann, and H. W. Clark. 1930. Check list of the fishes and fish-like
vertebrates of North and Middle America, north of the northern boundary of Venezuela
and Colombia. Rept. U.S. Comm. Fish (1928) pt. II (1055): 1-670.

Kuehne, R. a. and R. W. Barbour. 1983. The American darters. University Press of Kentucky,
Lexington. 177pp.

Mette, M. F. 1978. The fishes of the Birmingham-Jefferson County region of Alabama with
ecologic and taxonomic notes. Geol. Surv. Alabama Bull. 115. 183pp.

Mettee, M. F., P. E. O'Neil, J. M. Pierson, and R.D. Suttkus. 1989a. Fishes of the Black Warrior
River system in Alabama. Geol. Surv. Alabama Bull. 133. 201pp.

Mettee, M. F., P. E. O'Neil, J. M. Pierson and R. D. Suttkus. 1989b. Fishes of the western
Mobile River basin in Alabama and Mississippi. Geol. Surv. Alabama Atlas 24. 170pp.

Mettee, M. F., P. E. O'Neil and J. M. Pierson. 1996. Fishes of Alabama and the Mobile Basin.
Oxmoor House, Birmingham, Alabama. 820pp.

Moore, G. A. and F. B. Cross. 1950. Additional Oklahoma fishes with validation of
Poecilichthys parvipinnis (Gilbert and Swain). Copeia 1950: 139-148.

Moore, G. A. and C. C. Rigney. 1952. Taxonomic statusof the percid fish Poecilichthys radiosus

1999]

New Species of Fuscatelum

41

in Oklahoma and Arkansas, with the descriptions of two new subspecies. Copeia 1952:
7-15.

Page, L. M. 1981. The genera and subgenera of darters (Percidae, Etheostomatini). Occ. Pap.
Mus. Nat. Hist. Univ. Kansas 90: 1-69.

Page, L. M. 1983. Handbook of darters. T.F.H. Publications, Neptune City, New Jersey.
271pp.

Page, L. M. and B. M. Burr. 1991. A field guide to freshwater fishes. Houghton Mifflin
Company, Boston. 432pp.

Rhode, F. C. 1980 et. seq. Etheostoma parvipinne. In: D. S. Lee, C. R. Gilbert, C. H. Hocutt, R.
E. Jenkins, D. E. McAllister, and J. R. Stauffer, Jr. (eds.). Atlas of North American
freshwater fishes. N.C. State Mus. Nat. Hist., Raleigh, North Carolina. 867pp.
Richards, W. J. 1963. Systematic studies of darters from southeastern United States (Pisces,
Percidae). Unpubl. Ph.D. Diss., Cornell Univ. 189pp.

Robins, C. R., R. M. Bailey, C. E. Bond, J. R. Brooker, E. A. Lachner, R. N. Lea, and W. B. Scott.
1991. Common and scientific names of fishes from the United States and Canada. Am.
Fish. Soc. Spec. Publ. 20: 1-183.

Robison, H. W. 1977. Distribution, habitat, variation and status of the goldstripe darter,
Etheostoma parvipinne Gilbert and Swain, in Arkansas. Southwestern Nat. 22: 435-442.
Smith-Vaniz, W. F. 1968. Freshwater fishes of Alabama. Auburn University Agricultural
Experimental Station, Auburn, Alabama. 211pp.

Shaw, K. A. 1996. Phylogenetic analysis and taxonomy of the Oligocephalus group of darters
(Teleostei: Percidae). Unpubl. Ph.D. Diss., Univ. Kansas. 338pp.

Turner, T. F. 1997. Mitochondrial control region sequences and phylogenetic systematics of
darters (Teleostei: Percidae). Copeia 1997: 319-338.

Wiley, E. O. and R. L. Mayden. 1985. Species and speciation in phylogenetic systematics,
with examples from the North American fish fauna. Ann. Missouri Bot: Gard. 72: 596-
635.

Wood, R. M. and R. L. Mayden. 1997. Phylogenetic relationships among selected darter
subgenera (Teleostei: Percidae) as inferred from analysis of allozymes. Copeia 1997: 265-
274.

Table 1. Frequency distribution of lateral-line scale counts in Etheostoma phytophilum and across the range of E. parvipinne.

[Vol. 31, No. 1

42

Tulane Studies in Zoology and Botany

in vOrO^Dr-tONOfNCOONtN

T-HOfOCxONOmoNCorsit^t^oiricocrji^

aN'^oocsiinodtNcocNfNodisirHodvovd

iOrOrOOrs|CNOfOv£)Oi-i^OO'£>
cdt— ifsicorsicocDfNiioco mcD

(N T-H

CNCOONCOC30

i-^rHCO ONvOtS.OCN t-hcS

'vO'^LO lOLOOOOtN-^

CO COvOCOCOCOt^CO'^vOvO T-HO-^tN

vO COr-H-^GOCNCNlT-l OOOi-H T-llO<

< O (N t-h ID

C» IN t-h CNI T-l

Si 2 ^

. 1 alls

•S -§ 2 c 2 ^2 ^ S -P. 2 ^

2 S ^ g 2 c '

O N ^ -S ^ O

.►r w ns .;h ^

2 2 'o ^ s

JO E

£ -2
5 u

^ g

2 (U — , R3

b > ’I u c Xs t3

i-i (S (Q ^ o o) fc op <L» '^•>n^'35

S^U PDcOOOCSiOHHJCl.ClHH<&HWU<<

, o ”3

Turkey Creek 1 2 8 10 4 25 43.68 2.79

Little Cove Creek 1 1 40.00

Table 2. Frequency distribution of pored lateral line scale counts in Etheostoma phytophilum and across the range of E.

43

1999]

New Species of Fuscatelum

000^'£)T^lNO00^-H0^l^) OSt^tNCO

oqLncoT^T};cocNvqLOfS tN-'ijpp

r-ifOCOlO

ONOOCOOOCOONOOrOOrJ^txOOKtNO
“jPpcNN;pppplN;ppppppp

i'^odo^odo^iNiooN'^coio

In fO IT) In 00 T
■Ttl LO '

iOCOCOOCNCNOCOvOOt-I'^OOIN
COT-ifNCOCNfOCOtNlOCO COCO

(fN ^ T-irO'=r(N'=:t^«N

(Ni-ifSI iNOOCOfMlO

COcOLOr-irO'OOOvOCOfN (N-^T-HfN

CM rJHOOfOlNvOaN T-HlOfN(N

m ir>rorsivo(Ni'^co(N^DcOT-H ont

fN lNr-irOir)(N(NT-lr-lcnOO t-HIOt

(fNiNvO t-i COt— I INvOt

IN CO T

^ c2
(0
I

o

5^u

> 2>

2 .g 2 ^ s

.2
2 cn

> I VH
2 ^ ^

s .2 <2 a:

Of .52

(U >5
cn 1*0 •

^ -3

In ^ y2-^^ 2'o*

^2 a,.s>|.§.g|-s|

Pc?5CDDi:OHHJPPH<Cl.WU<<

Clear Creek 1 4 6 7 13 6 4 1 42 43.57 3.20

Turkey Creek 1 2 7 8 6 1 2 2 25 37.92 4.23

Little Cove Creek 1 1 37.00

Table 3. Frequency distribution of transverse scale counts in Etheostoma phytophilum and across the range of E. parvipinne.

44

Tulane Studies in Zoology and Botany

[Vol. 31, No. 1

t-hOOOOOOOOO

o o o o

I ^ ^ (N CN ^

CD l\0'sOOCMO^CO'^T—

IT) CO LOCOvOCNCOCxO T— (CNt— ILO IDIO,— i

CN VO rH On

^ O ^

0» > kI” J-l

• O

(2^2
aS2 g

c :s

^ VH cc o7 S K

• l-H <D OJ ^ rV ^

$, > S ^ -2

S ^ 2 I

U ^ ^

ns 'o ^

5 -tC vh u

iS y ^
u S B

~ n

- V O)

5 J! s;

2^0

sr'-'

. i: X!

s=^U CQcotOC!iOHKJP^pL,H<P^WU<< H H-

Table 4. Frequency distribution of caudal peduncle circumferential scale counts in Etheostoma phytophilum and across the
range of E. parvipinne.

1999]

New Species of Fuscatelum

45

VO m ! CO

CM LOCOVO T— l^sCOCN^v t-h

cMT-HLO^cOT-H(N^tvT-Hirim

T— l^fNIVOrHCMT-HVOavaN 00

'=*1 t^rHCO^(N'OiiOOOCV4vOT-HCMvOUOCO

ONT-HVOCO'^f'lO(N LOOvt— I

^ CN r-< t-h to

CO CN CO

..2 s; o ^

g ■§ 2 .S S

CsS

'S.

vh Cl

o) .i=r

> eft

S 22

Cft

> >

O N

- ^ ^ W

^ ^ IS

« 22 ^ p-

53 . S^2 s
”1:2^

O) jQ S o

bO C o o

™ s I ;2 g § s g

. » go rag

O) OS O'— }.0> 5^3)

s^u CQc/5cOP!^OH.-IPhPhH<PhWU<<:

Clear Creek 2 2 12 17 9 ■ 42 20.69 1.27

Turkey Creek 2 4 12 7 ,^25 20.96 1.05

Little Cove Creek 1 1 20.00

Table 5. Frequency distribution of dorsal spine counts in Etheostoma phytophilum and across the range of E. parvipinne.

46

Tulane Studies in Zoology and Botany

[Vol. 31, No. 1

O OOOt-hOOOOOO OOOt-h

iop^vOT-H(Nco^qvqpioioppaNis.T-H
On ^ ON 00 On O O On On X— I ^ ON On O On On O

CN tN NO

(N T-HOOOOtN^ T—

t^s^-lNOI^^OTtLooo^Nl'^coT-HT— i^oocnI

m r-H CN| CO T-I

>

^ 2
0» ^

> 2 >

S 22

C T3 '
2

O N

fC C

E -

cs O „

CQ cn
laj

■Jc o S

-H

iS e ^
.£iyr\C>fC(/)5(C5-(U^n-^
O^DjJOHh-1C1^C1hH<C1hWU<<

.> > a;

i c5 ^ 2 ^

' ni '■'N

^ ns a; 2 ^

.22 o) o) 1— I ^

5! > 5 ^ -2

O bO C O Xi

jh CiJi bo-^ S -o c

« -2 5 lo c

Clear Creek 3 16 18 5 42 9.60 0.82

Turkey Creek 3 5 17 25 9.56 0.69

Little Cove Creek 1 1 10.00

1999]

New Species of Fuscatelum

47

Table 6. Frequency distribution of dorsal ray counts in Etheostoma phytophilum
and across the range of E. parvipinne.

9

10

11

12

13

N

X

SD

E. parvipinne

Colorado River

8

6

14

10.43

0.51

Brazos River

1

1

10.00

San Jacinto River

1

16

13

30

10.40

0.56

Sabine River

5

7

1

13

10.70

0.63

Red River

1

11

10

1

23

10.48

0.67

Obion River

12

16

1

1

30

10.70

0.70

Tennessee River

7

11

4

22

10.86

0.71

Lower Mississippi R.

3

20

7

2

32

10.25

0.72

Pearl River

1

12

16

1

30

10.57

0.63

Pascagoula River

1

4

14

4

23

10.91

0.73

T ombigbee / W arrior

1

14

26

13

1

56

10.98

0.80

Alabama River

3

13

12

1

1

30

10.47

0.86

Perdido River

1

1

10.00

Escambia River

3

1

4

10.25

0.50

Choctawhatchee R.

7

17

5

30

9.93

0.65

Apalachicola River

3

22

5

30

10.07

0.52

Altamaha River

4

3

7

10.43

0.53

E. phytophilum

Clear Creek

24

16

2

42

10.48

0.65

Turkey Creed

1

10

13

1

25

11.52

0.78

Little Cove Creek

1

1

11.00

Table 7. Frequency distribution of right pectoral ray counts in Etheostoma
phytophilum and across the range of E. parvipinne.

12 13 14 15 16 17 N jc SD

E. parvipinne
Colorado River
Brazos River
San Jacinto River
Sabine River
Red River
Obion River
Tennessee River
Lower Mississippi R.
Pearl River
Pascagoula River
T ombigbee / W ar rior
Alabama River
Perdido River
Escambia River
Choctawhatchee R.
Apalachicola River
Altamaha River

E. phytophilum
Clear Creek
Turkey Creek
Little Cove Creek

5 9

1

3 14 9 4

116 5

2 15 5

1 4 16 9

14 8

20 10

1 15 14

3 11 9

12 37 6

1 3 13 10

1

2 2

2 17 10

1 2 14 13

1 2 4

3 16 23

14 11

1

14

15.64

0.50

1

14.00

30

14.47

0.86

13

15.15

0.90

1

23

15.22

0.67

30

15.10

0.76

22

15.36

0.49

2

32

15.44

0.62

30

15.43

0.57

23

15.26

0.69

1

56

14.93

0.63

3

30

15.33

1.03

1

16.00

4

15.00

1.15

30

15.28

0.59

30

15.30

0.75

7

15.43

0.79

42

15.48

0.68

25

14.44

0.67

1

16.00

48

Tulane Studies in Zoology and Botany

[VoL 31, No. 1

Table 8. Frequency distribution of anal ray counts in Etheostoma phytophilum
and across the range of E, parvipinne.

7

8

9

10

N

X

SD

E. parvipinne

Colorado River

7

7

14

7.50

0.52

Brazos River

1

1

8.00

San Jacinto River

12

18

30

8.60

0.50

Sabine River

1

8

3

1

13

8.31

0.75

Red River

2

15

6

23

8.17

0.58

Obion River

4

22

4

30

9.00

0.53

Tennessee River

7

14

1

22

8.73

0.55

Lower Mississippi R.

3

18

11

32

8.25

0.62

Pearl P^ver

8

18

4

30

7.87

0.63

Pascagoula River

1

8

12

2

23

8.65

0.71

T ombigbee / W arrior

3

23

26

3

56

8.53

0.69

Alabama River

5

22

3

30

7.93

0.52

Perdido P^ver

1

1

7.00

Escambia P^ver

2

2

4

8.50

0.58

Choctawhatchee R.

3

24

2

30

7.97

0.42

Apalachicola River

9

20

1

30

7.73

0.52

Altamaha River

5

2

7

8.29

0.49

E. phytophilum

Clear Creek

14

24

4

42

8.76

0.64

Turkey Creek

1

6

15

3

25

8.80

0.66

' Little Cove Creek

1

1

10.00

Table 9. Frequency distribution of branched caudal ray counts in Etheostoma
phytophilum and across the range of E. parvipinne.

10

11

12

13

14

15

16

N

X

SD

E. parvipinne

Colorado River

5

6

3

14

13.86

0.77

Brazos River

1

1

14.00

S^ Jacinto River

3

8

10

7

28

13.75

0.97

Sabine Phver

1

1

9

2

13

13.92

0.76

Red River

2

8

12

1

23

13.52

0.73

Obion River

2

7

20

1

30

12.67

0.66

Tennessee P^ver

1

2

7

7

3

1

22

.13.57

1.16

Lower Miss. R.

1

8

15

6

1

31

13.93

0.85

Pearl River 2

1

2

3

14

8

30

13.67

1.40

Pascagoula Pliver

1

7

10

5

23

13.83

0.83

Tombigbee / Warrior

1

4

27

16

7

56

13.44

0.88

Alabama River

1

6

16

7

30

13.97

0.76

Perdido River

1

1

15.00

Escambia River

2

2

4

13.50

0.58

Choctawhatchee R.

15

7

7

30 .

13.72

0.84

Apalachicola Phver 1

1

3

3

14

8

30

13.73

1.26

Altamaha Pliver

1

5

1

7

13.00

0.58

E. phytophilum

Clear Creek

3

12

16

11

42

13.83

0.95

Turkey Creek

1

5

17

2

25

12.80

0.84

Little Cove Creek

1

1

12.00

1999]

New Species of Fuscatelum

49

Table 10. Frequency distribution of infraorbital pore counts in Etheostoma
phytophilum and across the range of £. parvipinne.

3

5

6

7

8

9

10

N

X

SD

E. parvipinne

Colorado River

4

9

1

14

7.79

0.58

Brazos River

1

1

8.00

San Jacinto River

2

24

4

30

8.07

0.45

Sabine River

3

8

2

13

7.92

0.64

Red River

1

3

19

23

7.74

0.69

Obion River

1

4

23

2

30

7.87

0.57

Tennessee River

2

19

1

22

7.95

0.38

Lower Miss. R.

1

25

5

1

32

8.16

0.63

Pearl River 1

3

23

3

30

7.83

1.02

Pascagoula River

2

18

3

23

8.04

0.47

T ombigbee / Warrior

9

43

4

56

7.91

0.48

Alabama River

5

23

2

30

7.90

0.48

Perdido River

1

1

7.00

Escambia River

4

4

8.00

Choctawhatchee R.

1

22

6

30

8.17

0.47

Apalachicola River

1

27

2

30

8.03

0.32

Altamaha River

1

4

2

7

8.14

0.69

£. phytophilum

Clear Creek

1

1

8

29

3

42

7.76

0.58

Turkey Creek

13

8

3

1

25

7.68

0.97

Little Cove Creek

1

1

8.00

Table 11. Frequency distribution of preoperculomandibular pore counts in
Etheostoma phytophilum and across the range of E. parvipinne.

6

7

8

9

10

11

12

N

JC

SD

£. parvipinne

Colorado River

1

1

9

3

14

9.93

1.00

Brazos River

1

1

10.00

San Jacinto River

1

26

3

30

10.07

0.37

Sabine River

1

11

1

13

10.00

0.41

Red River

3

20

23

9.87

0.34

Obion River

2

27

1

30

9.97

0.92

Tennessee River

1

1

17

2

1

22

10.05

0.72

Lower Miss. R.

2

28

2

32

10.00

0.36

Pearl River 1

1

3

21

4

30

9.83

0.95

Pascagoula River

3

18

2

23

9.96

0.47

T ombigbee / Warrior

1

8

38

7

2

56

10.00

0.76

Alabama River

1

1

6

22

30

9.63

0.72

Perdido River

1

1

8.00

Escambia River

3

1

4

10.25

0.50

Choctawhatchee R.

4

16

5

4

30

10.31

0.89

Apalachicola River

6

24

30

9.80

0.41

Altamaha River

3

4

7

9.57

0.53

E. phytophilum

Clear Creek

9

27

5

1

42

9.95

0.70

Turkey Creek

1

24

25

9.96

0.19

Little Cove Creek

1

1

10.00

50

Tulane Studies in Zoology and Botany

[Vol. 31, No. 1

CL ^

o

CL ^

3 £

CD C/i

0> 'P

> C

O

2

< «

CC

O)
c«

S o'
ten
C O
QJ V

^ 4->

T3 (t3
C u
(Z cc

c

5 t/5

C QJ

■£ o

O C

CL

C "^3

fC ~o

T3

0> rn

c/i

c/i

QJ ~

2 £X

CD

O)

CL S

g S 5

CD O CD

c c

0) -s: O

E

Si w 5

S "2 .S
S « X

o;

_ CD

>..S

"O Cc.

g s O

SL.-0
cs a>

<N

^ o c

CD •-

^ C C

(T3 O C

SX_2

Pi >

COOOCJOLOOt^COOrslOCDON<NvO

oo^ixmoasomotxjcoO'T-HtN

oo^^l^)Oc^l^x^-HO(J^cr)OOo

oooooc5oooo<dooo

CD ^ O O ^ '

I>s CD CN I>N

ON Os CN ID
O csi K ID

C30IDCNt^tNC3N _ .

lr^O^(NOOO(N^T-H^
OOONOOr^vDOTj^ND

oocdcziooooociocpcbcp

CD CD C30 ID O

■ LD O O ^

CD ON t>. O '

Ot^OOOONOfN

CDCNllDOOvO^^CiOtONlDNOOOON
t^tN'.ONCNl^CNLDNOOO'^CD^CMCX)
CNt—It— iT-HfNr-(OOO^r-lfN|(NT-H

ooocdciooooczioooci

ooooooooooo

o o o o o o o

NO O ID O CN LD 00
NO O ^ t-h o
O ^ T-H ,— I CN CN ^
O O O CD O O O

OCD^^^^^OOCOlDOOfMOCDlDOO
txNOOOCNTti— iIDnCOID'^t— iO<»
CNt— I,— (,— iCNl^OOrHrHT-HCNCNrH

CDOCDCD0Oc5oOC5OO00

QJ QJ

3 •£

3 3 X

f 8-g-:2 S-S-g

.2iQ-^ S-=T;-.£i»

C/J 5-1 4-> J

_ — 03 t)C ,

03 03 ^ C i

^ ^ CD — QJ ^

^T3 "3 TO .£ ^

;-L. ;_L, -

— CD CD CD-r
■^co(nj_,"QjQjQjfn
■u ^_^"0 ”D 3 KIN hr KrT-S

^ tiObCiiCO

;H?i"0"033o-3CCC3
^^CCR303cSrOCO«3

XXcqcqUUcdOjj— lU

FUSCONAIA APALACHICOLA, A NEW SPECIES OF FRESHWATER
MUSSEL (BIVALVIA: UNIONIDAE) FROM PRECOLUMBIAN
ARCHAEOLOGICAL SITES IN THE APALACHICOLA BASIN OF
ALABAMA, FLORIDA, AND GEORGIA

James D. Williams

U.S. Geological Survey, Biological Resources Division
7920 NW 71st Street, Gainesville, Florida 32653

AND

Arlene Fradkin
Department of Anthropology
Florida Atlantic University, Boca Raton, Florida 33431

Abstract

Fusconaia apalachicola, a new unionid mussel, is described from four precolumbi-
an archaeological sites in the Apalachicola Basin. The sites range in age from A.D.
500 to 1350. Fusconaia apalachicola is absent from recent collections but was likely
extant when Europeans arrived in North America. Its extinction was likely due to
pollution and habitat destabilization resulting from agriculture and development in
the Apalachicola Basin beginning in the early 1800s. This species is known only from
this basin and only from archaeological contexts. It is the first species of the genus
Fusconaia known to occur within the Apalachicola Basin. Based on conchological
characters it appears to be most closely related to F. rotulata which is endemic to the
Escambia River drainage in south Alabama and west Florida.

Introduction

Within the southeastern United States, the eastern drainages of the Gulf of Mex-
ico harbor a diverse unionid mollusk fauna. This fauna consists of widespread taxa
that occur from the Mississippi Basin eastward and south Atlantic drainages west-
ward into the eastern Gulf (Burch, 1975). This region is also inhabited by a relatively
large number of endemic mussel taxa. The endemic mussel fauna in the eastern Gulf
drainages, Escambia River east to Suwannee River, consists of 26 species and one
genus, Elliptoideus Frierson, 1927 (Butler, 1989; Williams and Butler, 1994).

Much of the eastern Gulf region endemism is centered in the Apalachicola Basin,
which is drained by the Apalachicola, Chattahoochee, and Flint rivers. The Apalach-
icola Basin is the largest drainage east of the Mobile Basin and drains portions of
eastern Alabama, western Georgia, and northwestern Florida. Within the region,
the Apalachicola Basin supports the greatest number of freshwater mollusk species
as well as endemics (Butler, 1989; Clench and Turner, 1956). The mussel fauna of the
Basin consists of 33 species, of which 10 are endemic (Brim Box and Williams, in
press).

The genus Fusconaia is widespread in the Mississippi Basin and Gulf of Mexico
drainages where it is represented by 13 species. In the Atlantic Coast drainages.

Tulane Studies in Zoology and Botany 31; 51-62. 1999.

51

52

Tulane Studies in Zoology and Botany

[Vol. 31, No. 1

however, the genus Fusconaia is represented by a single species, f . masoni (Conrad,
1834) (Potomac River drainage south to the Savannah River drainage). In the eastern
Gulf drainages from the Mobile Basin eastward, the genus Fusconaia is represented
by F. cerina (Conrad, 1838) and F. ebena (Lea, 1831) (Mobile Basin and westward), F.
escambia Clench & Turner, 1956 (Escambia and Yellow river drainages), F. rotulata
(Wright, 1899) (Escambia River drainage), and F. succissa (Lea, 1852) (Escambia
River drainage east to Choctawhatchee River). Fusconaia apalachicola n. sp. was en-
demic to the Apalachicola Basin, the only species of Fusconaia known from this basin,
and was the easternmost representative of the genus on the Gulf Coast.

Fusconaia apalachicola was discovered in the process of identifying mollusks in a
faunal sample from a precolumbian archaeological site located along the main chan-
nel of the upper Apalachicola River, Liberty County, northwest Florida. An addi-
tional 13 species of unionid mussels were found in archaeological samples in the
Apalachicola Basin. While all 13 species are well represented in museum collections
of recent mollusks, it appears that F. apalachicola disappeared before modern mala-
cologists or shell collectors could collect live or fresh shells.

Materials and Methods

Comparative archaeological material of Fusconaia apalachicola was obtained from
three museum research collections. We examined faunal collections housed in the
Florida Museum of Natural History, Gainesville, Florida (FLMNH), the Bureau of
Archaeological Research, Division of Historic Resources, Department of State, Tal-
lahassee, Florida (FLBAR), and the Columbus Museum, Columbus, Georgia (uncat-
aloged). Material of other species of Fusconaia was obtained from the FLMNH, the
Alabama Malacological Research Center, Mobile, Alabama (AMRC), the National
Museum of Natural History, Smithsonian Institution, Washington, DC (USNM),
and U.S. Geological Survey, Gainesville, Florida collections.

Type material was designated for specimens in the collection of the FLMNH.
Additional specimens from the FLBAR collections were designated as type material
and deposited at the North Carolina State Museum of Natural Sciences, Raleigh,
North Carolina (NCSM), the University of Tennessee McClung Museum, Knoxville,
Tennessee (UTMM), and the USNM.

A total of 56 specimens of Fusconaia apalachicola was identified in the FLMNH
precolumbian archaeological sample from Site 8LI76, Liberty County, Florida. We
also identified specimens of F. apalachicola in the FLBAR collections from the Sy-
camore Site 8GD13, Gadsden County, Florida, and from the Scholz Steam Plant Site
8JA1 04, Jackson County, Florida. Of the specimens examined, we borrowed 24 of the
most complete valves for comparative study, 14 from the Sycamore Site and 10 from
the Scholz Steam Plant Site. These three sites were located along the main channel
in the upper reach of the Apalachicola River (Figure 1).

We also examined material from the Chattahoochee and Flint rivers housed in
the Columbus Museum, Georgia. One additional specimen of Fusconaia apalachicola
was found in a faunal sample from the Omussee Creek Site 1H026, Houston Coun-
ty, southeast Alabama (Figure 1). This specimen is the only record from the Chatta-
hoochee River and the only occurrence of the species outside the state of Florida.

1999]

New Species of Precolumbian Fusconaia

53

Figure 1. Distribution of Fusconaia apalachicola (stars) in the Chattahoochee River
system, Alabama and Georgia, and in the Apalachicola River system, Florida, and
Fusconaia rotulata (triangles) in the Escambia River drainage, Alabama and Florida.

None of the valves of Fusconaia apalachicola were complete, although many were
missing only a small portion of the posterior end of the shell. Also, there was only
a trace of periostracum on a few of the archaeological specimens. The lack of com-
plete valves prevented us from taking total length measurements for comparative
purposes. Instead of using total length of the shell, we utilized a measurement of the
inter-adductor muscle scar distance taken from the anterior margin of the anterior
adductor mussel scar to the anterior margin of the posterior mussel scar. These
landmarks were chosen because they were moderately distinct and represented the
only measurement of length that could be taken with any degree of accuracy. The
width of the shell was taken by measuring the maximum extent of inflation in one
valve. The interdentum was used as a base for this measurement as it provided a
relatively flat platform to rest one arm of the calipers. This measurement was diffi-
cult to obtain on the archaeological material since a portion of the posterior and/or

54

Tulane Studies in Zoology and Botany

[Vol. 31, No. 1

ventral margin was usually missing. Measurements were taken on both left and
right valves. Dial calipers were used to take all measurements which were rounded
off to the nearest tenth of a millimeter.

Fusconaia apalachicola, new species
Apalachicola ebonyshell
Figures 2a and 2b

Fusconaia sp. or Fusconaia cf. rotulata, Fradkin (1995a).

Quincuncina infucata (in part) Percy (1976).

Fusconaia succissa, Quincuncina burkei, and Q. infucata (in part) Milanich (1974).

FiOLOTYPE: FLMNH 05260690.1 (left valve). Florida, Liberty County, Site 8LI76,
located 500 m east of the Apalachicola River (TIN; R8W; SE V4 Sec. 1) near river mile
88 (U.S. Army Corps of Engineers), about 5 miles N of Bristol (Figure 2a).

ParatypeS: Malacological Collections. — Florida, Gadsden County, Sycamore Site
8GD13, located near the east bank of the Apalachicola River, at N 30°37'39" latitude
and W 84°53’46" longitude: FLMNH 271727 (formerly cataloged as FLBAR 74.68.53.1
[right valve]); NCSM 4694 (formerly cataloged as FLBAR 74.68.156.3 [right valve];
FLBAR74.68.156.4[leftvalve]);NCSM4695(formerly cataloged as FLBAR74.68.168.1
[left valve]; FLBAR 74.68.168.2 [left valve]; FLBAR 74.68.168.3 [left valve]); USNM
880618 (formerly cataloged as FLBAR 74.68.150.1 [right valve]; FLBAR 74.68.150.2
[right valve]); USNM 880617 (formerly cataloged as FLBAR 74.68.157.1 [left valve]);
UTMM 3410 (formerly cataloged as FLBAR 74.68.157.2 [right valve]; FLBAR

74.68.158.1 [right valve]; FLBAR 74.68.158.2 [left valve]; FLBAR 74.68.158.3 [left
valve]).

Florida, Jackson County, Scholz Steam Plant Site 8JA104, located near the west
bank of the Apalachicola River, 3.25 miles southeast of Sneads (T3N ; R7W ; SW 1/4 Sec.
12): FLMNH 271728 (formerly cataloged as FLBAR 75.138.01.1 [left valve]).

Archaeological Collections. — Florida, Liberty County, Site 8LI76, located 500 m east
of the Apalachicola River (TIN; R8W; SE V4 Sec. 1) at river mile 88 (U.S. Army Corps
of Engineers), about 5 miles N of Bristol: FLMNH 05260528.1 (right valve); FLMNH
05260528.2 (right valve); FLMNH 05260528.3 (left valve); FLMNH 05260528.4 (right
valve); FLMNH 05260528.5 (left valve); FLMNH 05260528.6 (left valve); FLMNH
05260528.7 (left valve); FLMNH 05260528.8 (right valve); FLMNH 05260528.9 (right
valve); FLMNH 05260528.10 (right valve); FLMNH 05260528.1 1 (right valve); FLMNH

05260560.1 (left valve); FLMNH 05260560.2 (right valve); FLMNH 05260560.3 (right
valve); FLMNH 05260560.4 (right valve); FLMNH 05260560.5 (left valve); FLMNH
05260560.6 (right valve); FLMNH 05260690.2 (right valve); FLMNH 05260690.3 (right
valve).

Florida, Jackson County, Scholz Steam Plant Site 8JA104, located near the west
bank of the Apalachicola River, 3.25 miles southeast of Sneads (T3N; R7W; SW V
4 Sec. 12): FLBAR 75.138.01.2 (right valve); FLBAR 75.138.01.3 (right valve); FLBAR
75.138.01.4(leftvalve);FLBAR75.138.01.5(leftvalve);FLBAR75.138.01.6(leftvalve);
FLBAR 75.138.01.7 (right valve); FLBAR 75.138.01.8 (right valve).

1999]

New Species of Precolumbian Fusconaia

55

Figure 2. A) Fusconaia apalachicola. Holotype, FLMNH 05260690.1 (right valve). Flor-
ida, Liberty County, Site 8LI76, located 500 m east of the Apalachicola River (TIN;
R8W ; SE V4 Sec. 1) near river mile 88, about 5 miles N of Bristol, Florida. B) Fusconaia
apalachicola. Paratype, FLMNH 05260528.9 (right valve). Florida, Liberty County,
Site 8LI76, located 500 m east of the Apalachicola River (TIN; R8W; SE V4 Sec. 1) near
river mile 88, about 5 miles N of Bristol, Florida. Photo © by Richard Bryant.

Other material examined but not designated as types: Alabama, Houston County,
Omussee Creek Site 1H026, located at the mouth of Omussee Creek (Area I NW slope)
where it joins the Chattahoochee River, at Columbia, Alabama: uncataloged, Columbus
Museum, Georgia (left valve). Florida, Liberty County, Site 8LI76, located 500 m east of the
Apalachicola River (TIN; R8W; SE V4Sec. 1) at river mile 88 (U.S. Army Corps of Engineers),
about 5 miles N of Bristol: FLMNH 05260585.1 (right valve); FLMNH 05260585.2 (left valve);
FLMNH 05260585.3 (left valve); FLMNH 05260603.1 (right valve); FLMNH 05260603.2 (left
valve); FLMNH 05260603.3 (left valve); FLMNH 05260603.4 (right valve); FLMNH 05260603.5
(left valve); FLMNH 05260603.6 (left valve); FLMNH 05260603.7 (left valve); FLMNH
05260603.8 (left valve); FLMNH 05260603.9 (left valve); FLMNH 05260603.10 (left valve);
FLMNH 05260631.1 (right valve); FLMNH 05260631.2 (right valve); FLMNH 05260631.3
(left valve); FLMNH 05260647.1 (left valve); FLMNH 05260647.2 (right valve); FLMNH
05260647.3 (right valve); FLMNH 05260647.4 (right valve); FLMNH 05260647.5 (left valve);
FLMNH 05260660.1 (rightvalve); FLMNH 05260702.1 (leftvalve); FLMNH 05260718.1 (right
valve); FLMNH 05260718.2 (leftvalve); FLMNH 05260724.1 (right valve); FLMNH 05260724.2
(left valve). Florida, Jackson County, Scholz Steam Plant Site 8JA104, located near the west
bank of the Apalachicola River, 3.25 miles southeast of Sneads (T3N; R7W; SW 1/4 Sec. 12):
FLBAR 75.138.02.1 (left valve); FLBAR 75.138.02.2 (right valve). Florida, Gadsden County,
Sycamore Site 8GD13, located near the east bank of the Apalachicola River, at North 30°37'39"
latitude and West 84°53'46" longitude: FLBAR 74.68.164.1 (right valve).

Diagnosis: Fusconaia apalachicola can be distinguished from other species of
Fusconaia by the following characteristics. Fusconaia apalachicola has a smooth shell,
lacks any trace of a posterior ridge, moderately compressed, valve depth about 2.4
times into inter-adductor muscle scar distance, circular in outline, hinge plate broad
and angular, and the beak cavity is deep and compressed (Figures 2a and 2b). Fusconaia

56

Tulane Studies in Zoology and Botany

[VoL 31, No. 1

succissa has a poorly defined broadly rounded posterior ridge, moderately inflated,
the shell is oval in outline, rounded anteriorly, broadly convex posteriorly, hinge
plate broad and arcuate, and the beak cavities are moderately deep and open. Fusconaia
escambia has a well developed posterior ridge ending in a point posteriorly, the hinge
plate is broad and strongly arcuate, and the beak cavities are moderately deep and
open. The posterior slope is slightly concave. Fusconaia rotulata is moderately inflat-
ed and has a faint posterior ridge with one or two additional ridges on the posterior
slope. Outline of the shell circular to oval, typically oval, beak cavities are moderate-
ly deep and compressed, and the hinge plate is broad and angular.

Fusconaia apalachicola has been misidentified as Quincuncina infucata, the sculp-
tured pigtoe, by Milanich (1974) and Percy (1976). This species has a moderately
thick shell, subcircular in outline, and moderately inflated. Surface of shell varies
from almost smooth to faintly sculptured with small nodules or nodulous ridges
arranged in chevron-shaped pattern. Posterior slope flat to slightly concave; the
posterior ridge narrowly rounded, shell tapering to a blunt point on the base line
posteriorly. TTie beak cavities are shallow, compressed.

Description: Fusconaia apalachicola is characterized by a smooth moderately thick
shell, lack of a posterior ridge, circular in outline, compressed, and with beaks (erod-
ed in all individuals in this sample) anterior to the center of the shell. Internally the
beak cavities are deep and compressed, and the interdentum is broad and smooth.
There are two thick pseudocardinals in the left valve. The crest of the dorsal pseudocar-
dinal tooth ranges from parallel to the lateral teeth to a slightly anterior-dorsal/
posterior-ventral orientation. The crest of the ventral pseudocardinal tooth is tan-
gential to or passes through the posterior margin of the anterior adductor muscle
scar. There is a single pseudocardinal in the right valve. The laterals are typically
straight, but may be slightly curved, two in the left valve and one in the right valve.
While nacre color cannot be determined with certainty on archaeological material,
there has been no indication that it was colored. The periostracum remained partial-
ly intact on two individuals and appeared to be black.

The two best characters for distinguishing Fusconaia apalachicola (Figure 2) from
F. rotulata (Figure 3) are its circular outline and more compressed shell. InF. apalachicola
the ratio of valve depth to inter-adductor muscle scar distance ranges from 2.2 to 2.7
with a mean of 2.4 (N=20). The ratio of valve depth to inter-adductor muscle scar
distance in F. rotulata ranges from 1.9 to 2.4 with a mean of 2.1 (N=13). The relation-
ship of shell width and length are presented in Figure 4. Use of outline and degree
of compression of the shell for taxonomic purposes in unionids, however, is often
viewed with skepticism as these characters can be influenced by environment. In
headwater populations of some species the shell is more compressed and round in
shape when compared to downstream populations (Ball, 1922; Ortmann, 1920).
However, our use of these characters in distinguishing F. rotulata from F. apalachicola
is appropriate as both species occur in similar environments, the main channel of the
lower portions of large rivers on the lower Coastal Plain.

Additional evidence that supports the recognition of Fusconaia apalachicola as a
distinct species is found in the zoogeography of aquatic organisms in the eastern
Gulf drainages. There are several distinct and well documented breaks in east/ west

1999]

New Species of Precolumbian Fusconaia

57

Figure 3. A) Fusconaia rotulata. Holotype, USNM 159969, length 46 mm. Escambia
River, Escambia County, Florida. B. FI. Wright. Photo by Jim Williams. B) Fusconaia
rotulata. AMRC 4997.1, length 61 mm. Conecuh River, 1 mile above Alabama High-
way 41, south of East Brewton, Escambia County, Alabama. 17 August 1996. Photo
© by Richard Bryant.

distribution of aquatic species in the eastern Gulf drainages. One of the most pro-
nounced breaks in distribution of aquatic fauna occurs between the west Florida
rivers, the Escambia, Yellow, and Choctawhatchee, and the Apalachicola, Chatta-
hoochee, and Flint rivers. This is reflected in the large number of endemic mussels
and fishes in the two areas and the reduced number of species that the two areas
share in common (Swift et al., 1986; Brim Box and Williams, in press; Clench and
Turner, 1956).

Distribution: Based on the material available, the Apalachicola ebonyshell ap-
pears to have been restricted to the Apalachicola Basin (Figure 1). We currently have
material from the Apalachicola River in Florida and Chattahoochee River on the
Alabama and Georgia border in southeastern Alabama. We also examined two

58

Tulane Studies in Zoology and Botany

[Vol. 31, No. 1

Figure 4. Relationship of valve width to length (distance from anterior margin of the
anterior muscle scar to the anterior margin of the posterior muscle scar) in Fuscomia
rotulata and Fuscomia apalachicola.

archaeological faunal collections housed at the Columbus Museum, Georgia. Of the
two samples, one was from a site on the Chattahoochee River in east Alabama, the
Abercrombie Site (1RU61), Russell County. The other site was the Cannon Site on the
Flint River (9CP108), Crisp County, Georgia. These samples had similar species
assemblages but did not contain Fusconaia apalachicola. We also examined one sam-
ple from an archaeological site (8WL81) in the Choctawhatchee River drainage,
Walton County, west Florida, but no specimens of F. apalachicola or F. rotulata were
present (Fradkin, 1995b).

The four archaeological sites, three in Florida and one in Alabama, that con-
tained specimens of Fusconaia apalachicola range in age from A.D. 500 to 1350. The
three Florida sites are located along the mam channel of the upper reach of the
Apalachicola River (Figure 1). Site 8LI76, Liberty County (Maymon et al., 1996), and
the Sycamore Site (8GD13), Gadsden County (Milanich, 1974), are situated along
the east side of the river. The Scholz Steam Plant Site (8JA104), Jackson County, is
located on the west side of the river (Percy, 1976). All three are inland sites and were
occupied during the middle or late Weeden Island culture period, ranging from
approximately A.D. 500 to 1000. The Omussee Creek Site (1H026) (Figure 1), Hous-
ton County, Alabama, is located on the lower Chattahoochee River and dates to the
Rood phase of the Mississippian period, approximately A.D. 900 to 1350 (Schnell
et al., 1981).

1999]

New Species of Precolumbian Fusconaia

59

Etymology: Derivation of the word Apalachicola is not clear. One interpretation
is that Apalachicola is derived from the Hitchiti Indian words apalahchi = on the
other side, and okli = people, referring to people who live on the other side of the
river (Read, 1937). Another interpretation is from the Choctaw word Apelichi =
ruling place, and Okla = people, meaning people of the ruling place (Boyd, 1956). We
apply this name in reference to the species' restricted distribution to the Apalachi-
cola Basin. Apalachicola ebonyshell is suggested for the common name for Fusconaia
apalachicola.

Discussion

Mussel collections from the Apalachicola Basin date from 1833, when T. A. Con-
rad traveled across the Chattahoochee and Flint rivers. He collected mussels from
the Flint River but was unable to sample the Chattahoochee River (Wheeler, 1935).
Between 1830 and 1870, Isaac Lea described more than 60 species from the Apalach-
icola Basin, of which only 10 are currently recognized as valid. In the early 1900s, H.
H. Smith of the Alabama Museum of Natural History collected large numbers of
mollusks from the Apalachicola Basin. In a comprehensive study of the Apalachi-
cola Basin mussels. Brim Box and Williams (in press) examined most of the available
museum collections and 350 samples collected between 1990 and 1997, but no
Fusconaia were found. During the past 150 years, there has been considerable effort
expended in the collection of mussels in the Apalachicola Basin. Many of these
samples were collected prior to major alterations in the watershed. It appears that
no individuals of F. apalachicola were ever encountered.

The causal factors and time of extinction of Fusconaia apalachicola are not known.
The presence and abundance of this mussel in archaeological remains dated to A.D.
1350 suggests that it was most likely extant when the Europeans arrived in North
America. If this assumption is correct, then it likely disappeared in response to
environmental change associated with settlement and development of the Apalach-
icola Basin. Significant habitat alterations in the Apalachicola Basin were well un-
derway in the early 1800s. Glenn (1911) reported that 60% of the uplands on the
Chattahoochee River, between its headwaters near Gainesville, Georgia, down-
stream to Atlanta, had been cleared. Sediment from the cleared lands had filled the
river channel to the point that many ferries on the Chattahoochee River could not
operate during most of the year (Glenn, 1911). Logs cut along the Chattahoochee
River were floated downstream to sawmills on an island at Columbus, Georgia,
where they were cut into lumber and the sawdust and bark dumped in the river.
Cotton mills in Columbus were also dumping waste, fibers, and metal based-dies
into the river (Brim Box and Williams, in press).

One of the more common mussels in archaeological faunal samples in the Apalach-
icola and Chattahoochee rivers, Elliptoideus sloatianus, is known in the Chattahoochee
River from only two individuals collected in the 1830s (Brim Box and Williams, in
press). While E. sloatianus persisted in the Apalachicola and Flint rivers, its early
extirpation from the Chattahoochee was most likely due to the pollution of the river
in the 1800s.

Based on conchological characters, Fusconaia apalachicola appears to be most closely

60

Tulane Studies in Zoology and Botany

[VoL 31, No. 1

related to f . rotulata, which is endemic to the Escambia River drainage in Escambia
County, Alabama, and in Escambia and Santa Rosa counties, Florida. Wright (1899)
described Fusconaia rotulata as a new species and placed it in the genus Unio based
on a single specimen. Simpson (1900a) reexamined the type specimen and, based on
the characters of this individual, assigned it to the genus Obovaria. Subsequent workers
(Heard, 1979; Johnson, 1967a, 1967b, 1969; Simpson, 1900b, 1914; Turgeon et al.,
1998) continued to recognize rotulata as a species of Obovaria. After examining the
type and additional specimens of rotulata and comparing conchological characters
(teeth, hinge plate, and deep umbo pocket) with other species of Obovaria and
Fusconaia, Williams and Butler (1994) concluded that, based on conchological char-
acters, rotulata was a species of Fusconaia, not Obovaria. Fusconaia rotulata most close-
ly resembles F. ebena, which is widespread in the Mississippi Basin and along the
Gulf coast eastward to the Alabama River system. The other two species of eastern
Gulf drainage Fusconaia, F. escambia and F. succissa, are easily distinguished from f .
rotulata by the presence of a well developed posterior ridge.

The habitat of Fusconaia apalachicola and F. rotulata appears to have been very
similar. Fusconaia rotulata is confined to the main channel of the Escambia River in
areas with moderate current and a mixture of sand and gravel substrates. Based on
museum records and recent collections, both species also appear to have very re-
stricted distributions, being limited to large rivers in the lower Coastal Plain ecore-
gion. The small number of F. rotulata located in museum collections along with those
found in a recent survey of the main channel of the Escambia River indicate that it
has been rare during the past 100 years. Until additional archaeological samples
from the Apalachicola Basin are analyzed, it is not possible to assess the prehistoric
rarity of F. apalachicola. In proportion to other unionids in archaeological samples
where it was found, F. apalachicola appeared to be uncommon; however, this percep-
tion may be due to the small number of samples examined. The rarity of F. rotulata
may be due in part to the difficulty of collecting mussels that occur in deeper waters
of the main channel of a river.

While the extinction of Fusconaia apalachicola is significant, it also has serious
implications for the future of F. rotulata. The extinction of F. apalachicola was most
likely due to the alteration of its main channel riverine habitat. The abusive land use
practices in the 1800s, during the development of the Chattahoochee and Flint river
watersheds, resulted in drastically altered hydrology of these systems (Glenn, 1911;
Brim Box and Williams, in press) which most likely extended downstream to the
Apalachicola River. Fusconaia rotulata, like F. apalachicola, has a limited geographic
distribution and is restricted to the main channel of the Escambia and Conecuh
rivers. These factors, in combination with the current alteration of land, riparian, and
stream habitats in the Escambia River watershed, make F. rotulata an at risk species
with a high probability of extinction.

Mussel samples from the three Florida sites where Fusconaia apalachicola was
found also contained the following 13 species: Amblema neislerii, Elliptio crassidens,
E. icterina, Elliptoideus sloatianus, Glebula rotundata, Lampsilis straminea claibornensis,
L. teres,Medionidus penicillatus, Megalonaias nervosa, Fleur obemapyr if orme, Quincuncina
infucata, Villosa lienosa, and V. vibex. Only five mussels, Elliptio crassidens, E. icterina,
Elliptoideus sloatianus, Pleurobema pyriforme, and Quincuncina infucata, were present

1999]

New Species of Precolumbian Fusconaia

61

in the sample from the mouth of Omussee Creek on the Chattahoochee River, Hous-
ton County, Alabama. In the Apalachicola Basin, these species typically occur in
large creeks and rivers in a substrate of gravel, sand, clay, or mixture of sand and
clay. Fusconaia apalachicola is the only species of unionid known to become extinct
before it was discovered and described. Based on abundant archaeological samples
from the southeastern United States, no other extinct species have been found which
were undescribed.

Acknowledgments

We would like to thank the following individuals for their assistance in provid-
ing access to specimens and collections in their care: Dr. David Dickel, Bureau of
Archaeological Research, Division of Historic Resources, Department of State, Tal-
lahassee, Florida; Mr. Kurt Auffenberg, Dr. Fred Thompson, and Dr. Elizabeth Wing,
Florida Museum of Natural History, Gainesville, Florida; Dr. Frank Schnell, Colum-
bus Museum, Columbus, Georgia; Mr. Doug Shelton, Alabama Malacological Re-
search Center Collection, Mobile, Alabama; Dr. Art Bogan, North Carolina State
Museum of Natural Sciences, Raleigh, North Carolina; Dr. Paul Parmalee, Frank H.
McClung Museum, University of Tennessee, Knoxville, Tennessee, and Kevin Cum-
mings, Illinois Natural History Survey, Champaign, Illinois. We also appreciate the
review and comments on an early draft of this manuscript by Drs. Bogan and Par-
malee.

We would also like to thank Mr. Dick Bryant for the use of his photographs, Ms.
Amy Benson for the preparation of the distribution map, and Ms. Sherry Bostick for
her assistance in manuscript preparation.

Literature Cited

Ball, G. H. 1922. Variation in fresh-water mussels. Ecology 3: 93-121.

Boyd, M. F. (ed.). 1956. A provisional gazetteer of Florida place-names of Indian derivation
either obsolescent or retained together with others of recent application. Florida Geo-
logical Survey, Special Publication No. 1. 158 pp.

Brim Box, f., and ]. D. Williams. In press. Unionid mollusks of the Apalachicola Basin in
Alabama, Florida, and Georgia. Bull. Alabama Mus. Nat. Hist. 320 pp.

Burch, ]. B. 1975. Freshwater unionacean clams (Mollusca: Pelecypoda) of North America.

Malacological Publications, Hamburg, Michigan. 204 pp.

Butler, R. S. 1989. Distributional records for freshwater mussels (Bivalvia: Unionidae) in
Florida and south Alabama, with zoogeographic and taxonomic notes. Walkerana 3:
239-261.

Clench, W. J., and R. D. Turner. 1956. Freshwater mollusks of Alabama, Georgia, and
Florida from Escambia to the Suwannee River. Bull. Florida State Mus. Biol. Sci. 1: 97-
239.

Fradkin, a. 1995a. Animal resource use at an inland Weeden Island Site: Analysis of faunal
remains from Site 8LI76, Liberty County, Florida. Report on file. Environmental Archae-
ology, Florida Museum of Natural History, Gainesville, Florida. 52 pp.

Fradkin, A. 1995b. Animal resource use at an inland Weeden Island village: Analysis of
faunal remains from Site 8WL81, Walton County, Florida. Report on file. Environmental
Archaeology, Florida Museum of Natural History, Gainesville, Florida. 43 pp.

62

Tulane Studies in Zoology and Botany

[Vol. 31, No. 1

Glenn, L. C. 1911. Denudation and erosion in the southern Appalachian region and the
Monongahela Basin. U.S. Geological Survey Professional Paper 72. 137 pp.

Heard, W. H. 1979. Identification manual of the freshwater clams of Florida. Florida Dept,
of Environmental Regulation Technical Series 4: 1-82.

Johnson, R. I. 1967a. Illustrations of all the mollusks described by Berlin and Samuel Hart
Wright. Occasional Papers on Mollusks, Museum of Comparative Zoology, Harvard
University 3: 1-35.

Johnson, R. 1. 1967b. Additions to the unionid fauna of the Gulf drainage of Alabama,
Georgia and Florida (Mollusca: Bivalvia). Breviora 270: 1-21.

Johnson, R. 1. 1969. Further additions to the Unionid fauna of the Gulf drainage of Alabama,
Georgia and Florida. The Nautilus 83: 34-35.

Maymon, J. H., T. F. Majarov, F. Vento, M. Williams, A. Fradkin, C. T. Brown, K. M. Child,
AND J. C. Clarke. 1996. Phase 111 data recovery at Site 8L176 for the proposed Florida gas
transmission phase 111 expansion project. Liberty County, Florida. Final report prepared
for Florida Gas Transmission Company, Houston, Texas. 418 pp.

Milanich, j. T. 1974. Life in a 9th century Indian household. A Weeden Island Fall-Winter
site on the upper Apalachicola River, Florida. Bureau of Historical Sites and Properties,
Bulletin 4: 1-44. Division of Archives, History, and Records Management, Department
of State, Tallahassee, Florida.

Ortmann, a. E. 1920. Correlation of shape and station in fresh water mussels (naides). Proc.
Amer. Philos. Soc. 59: 269-312.

Percy, G. W. 1976. Salvage investigations at the Scholz Steam Plant Site (8JA104), a middle
Weeden Island habitation site in Jackson County, Florida. Bureau of Historic Sites and
Properties, Miscellaneous Projects Report Series No. 35. Division of Archives, History,
and Records Management, Department of State, Tallahassee, Florida. 150 pp.

Read, W. A. 1937. Indian Place Names in Alabama. Louisiana State University Studies, No.
29. Revised edition with a Foreword, Appendix, and Index by J. B. McMillan. 1984.
University of Alabama Press, Tuscaloosa, Alabama. 107 pp.

ScHNELL, F. T., V. J. Knight, Jr., and G. S. Schnell. 1981. Cemochechobee. Archaeology of a
Mississippian Ceremonial Center on the Chattahoochee River. University Presses of
Florida, Gainesville, Florida. 290 pp.

Simpson, C. T. 1900a. New and unfigured Unionidae. Proc. Acad. Nat. Sci. Phila. 52: 74-86.

Simpson, C. T. 1900b. Synopsis of the naiades, or pearly freshwater mussels. Proc. U.S. Natl.
Mus. 22: 501-1044.

Simpson, C. T. 1914. A descriptive catalogue of the naiades or pearly freshwater mussels.
Bryant Walker (Detroit). Parts I-III. 1540 pp.

Swift, C. C., C. R. Gilbert, S. A. Bortone, G. H. Burgess, and R. W. Yerger. 1986. Zoogeogra-
phy of the freshwater fishes of the southeastern United States: Savannah River to Lake
Pontchartrain. pp. 213-266. In: C. H. Hocutt and E. O. Wiley (eds.). The Zoogeography
of North American Freshwater Fishes. Wiley, New York. 526 pp.

Turgeon, D. D., j. F. Quinn, Jr., A. E. Bogan, E. V. Coan, F. G. Hochberg, W. G. Lyons, P. M.
Mikkelsen, R. j. Neves, C. F. E. Roper, G. Rosenberg, B. Roth, A. Scheltema, F. G. Thompson,
M. Vecchione, and J. D. Williams. 1998. Common and scientific names of aquatic
invertebrates from the United States and Canada: mollusks, 2nd edition. American
Fisheries Society, Special Publication 26, Bethesda, Maryland.

Wheeler, H. E. 1935. Timothy Abbott Conrad, with particular reference to his work in
Alabama one hundred years ago. Bull. Amer. Paleo. 23: 1-157.

Williams, J. D., and R. S. Butler. 1994. Class Bivalvia, Order Unionoida, freshwater bivalves,
pp. 53-128. In: M. Deyrup and R. Franz (eds.). Rare and Endangered Biota of Florida, Vol.
II: Invertebrates. University Florida Press, Gainesville, Florida. 798 pp.

Wright, B. H. 1899. New southern unios. The Nautilus 13: 22-23.

' .-if : "K'

V

’ I *( ^*. *^'» 5^ - -.«

i ■ •' ’ ’ ■ <• **>C ic^- -

‘.1 yf . ‘ -

,-*- *>'. .^i^ryr. 4-^ V ^ 4i-“-* .•■ "•-•llic^i

•* • 71 ’.•iTwtSt'T.Ol (- ' , "

V } '

■ . ■ - a - - .

l:-- - ' J Vif..

- - •

' 5-^ « V'4 U/f jl? r>. ^

A**» 1 .f» ^ lif ^ I ^VTNrx>l4'Jb

^ '^. U : *•: "j . vjcifi4%

_ ..- -.r^ ■_ -■

■ ■* l*J ^ •■ ' -v-V^ <^**’5 >%r«''

* 4-, ^ • .■•^-^-‘>^Xr„ *L .,

■ < T <!r«t^-., M*>.

I:'-- ;, A7.;..^>'.'t#^.:,

I' r

*;'K4 (+

N.-**/ .--

»*rf ir-ti?5*

■f. jrr. IMS

r> fX\W. —

^ V‘nXi |ii|ifi||'l5*,cd^ %

; Vrl^ ^'-vW.. *i‘- ■ ■ :- 5 * ■•• :•>•'

. r myrC^T^ ,.• *-w‘ ..* ■•- -i

v*..^ :.J*. , , ^ - . : . .

*■ ‘■■*’^* "* 1* t ■* '% ~j t . ± tj. > I. ■p,.i*ij ' ■ '' - -fc . _ _ .

.* i^*»4 - ‘ i ^ i - ■*'fir I

r • :-* • >*. i^ sk/i44r«%l.f . -

* ■ A *^C! ' ri* : ^ ^ *■ . ''^* ■• W:t * ’. - . .

«Si:!

4 ^ I . . *r, y>V9^‘iiAMI

• •■■ -■ ■ •

»■. Vu-r*|gy4«^

- -■'•V \ ri*^

; i-U*:

>A .

' :'Vt-, ^.X ’t'-' ■ ^

. -iv> '

.» -K..? ’'•.-;v »*. *V*CS

\ ii : .hiJlf'iAn'

' > ■ -. _V *■?' '. ■‘■’ ' .

li^./ . • --!>,?

. 7 I-

^■■■•. V‘ i-‘ ri 7W4

•4^

• ;•?- f?- .V M|'us-*.'*j- r-riv<>;--ih — a-Ci^tiflilKtS' '■ :■

« w«-

Recent Issues of Tulane Studies of Zoology and Botany
Other back issues available; please consult our website.
http://www.museum.tulane.edu/pubs/

Volume 29, Number 1.

Suttkus, R. D. and R. M. Bailey. Etheostoma colorosum and E. bellator, two new darters, sub-
genus Ulocentra, from southeastern United States.

Bart, H. L., Jr. and M. S. Taylor. Type specimens of fishes in the Tulane University Museum
of Natural History.

Bradburn, A. S. and S. P. Darwin. Type specimens of vascular plants at Tulane University,
with a brief history of the Tulane Herbarium.

Volume 29, Number 2.

Suttkus, R. D., R. M. Bailey, and H. L. Bart, Jr. Three new species of Etheostoma from the Gulf
Coastal Plain of southeastern United States.

Warren, M., R. C. Cashner, and R. D. Suttkus. Fishes of the Buffalo River system, southwest-
ern Mississippi.

Knight, C. L. and R. W. Hastings. Fishes of the Tangipahoa River system, Mississippi and
Louisiana.

Volume 30, Number 1.

White, D. A. and S. P. Darwin. Woody vegetation of tropical lowland deciduous forests and
Maya ruins in the north-central Yucatan Peninsula, Mexico.

Garcia-Franco, J. G. and V. Rico-Gray. Population structure and clonal growth in Bromelia
pinguin L. (Bromeliaceae) in dry forests of coastal Veracruz, Mexico.

Reid, J. W. and G. G. Marten. The cyclopoid copepod (Crustacea) fauna of non-planktonic
continental habitats in Louisiana and Mississippi.

Volume 30, Number 2.

Johnston, C.E. and W. R. Haag. Life history of the Yazoo darter (Percidae: Etheostoma raneyi),
a species endemic to north-central Mississippi.

Rico-Gray, V., J.G. Garcia-Franco, A. Trigos-Landa, R. Mata, and P. Castaneda. Leaf-miner
defenses in Bromelia pinguin L. (Bromeliaceae) in Veracruz, Mexico.

Guzman, G. Observations on some fungi from Louisiana and Mississippi in comparison
with those of Mexico.

Dundee, H.A. Some reallocations of type localities of reptiles and amphibians described
from the Major Stephen H. Long expedition to the Rocky Mountains, with comments on
some of the statements made in the account written by Edwin James.

TULANE STUDIES IN
ZOOLOGY AND BOTANY

Volume 31, Number 2

25 March 2002

NEW SPECIES AND NEW RECORDS OF PSEUDOTHELPHUSID CRABS
(CRUSTACEA: BRACHYURA) FROM COLOMBIA
GILBERTO RODRIGUEZ

Centro de Ecologia, Instituto Venezolano de Investigaciones Cientificas
Apartado 21827, Caracas 1020 A, Venezuela

MARTHA R. CAMPOS

Universidad Nacional, Instituto de Ciencias Maturates
Apartado Aereo 53416, Bogota, Colombia

BEATRIZ LOPEZ

Centro de Ecologia, Instituto Venezolano de Investigaciones Cientificas
Apartado 21827, Caracas 1020 A, Venezuela

A REVIEW OF THE SPINYCHEEK SLEEPERS, GENUS ELEOTRIS
(TELEOSTEI: ELEOTRIDAE), OF THE WESTERN HEMISPHERE, WITH
COMPARISON TO THE WEST AFRICAN SPECIES

AND

FRANK PEZOLD

Department of Biology and Museum of Natural History (Zoology)
The University of Louisiana - Monroe
Monroe, LA 71209-0500

MCZ

AND

LIBRARY

BRYAN CAGE

JUN 1 8 2002

Department of Biology
University of Mississippi
University, MS 38677

HARVARD

UNIVDR3JTY

Tulane
University
New Orleans

TULANE STUDIES IN ZOOLOGY AND BOTANY
ISSN 0082-6782

Tulane University Museum of Natural History
Building A-3, Wild Boar Road
Belle Chasse, Louisiana 70037

Henry L. Bart, Jr.

Editor

editor @ museum, tulane . edu
INFORMATION FOR AUTHORS

Manuscripts dealing with all aspects of ecology, evolution, and systematic biology
are encouraged. Manuscript style should follow the latest edition of the Council of
Biology Editors Style Manual. Manuscript (with abstract) must be printed on 8.5 x
1 1 inch paper, double-spaced throughout (including Literature Cited, figure cap-
tions, footnotes, and tables). Do not hyphenate words at end of lines. Last name(s)
of author(s) should appear on each page with page number; this includes tables and
figure captions. Lettering used in figures must be easily read after reduction to a
page size of 5 x 7.75 inches. Photographs (halftones) should be made for no or lit-
tle reduction. Prospective authors should request a copy of the detailed guidelines
for manuscript preparation (including footnotes, typefaces, literature and specimen
citation, and figure dimensions) from the Editor at the above address.

The original and three copies of the manuscript, along with figures, should be sub-
mitted. All manuscripts are peer-reviewed. Following review, authors are required
to submit revised manuscripts (text, tables, and figures, if possible) in electronic
form. Consult the Editor for acceptable formats and modes of electronic submis-
sion.

Page charges are assessed at the rate of $50 per page. Charges may be reduced or
waived if authors are unable to meet these costs.

EXCHANGES AND STANDING ORDERS

Tulane Studies in Zoology and Botany is available for exchange with institutions
publishing comparable series. Otherwise, standing orders are accepted, with issues
priced individually (foreign postage extra). A list of back issues with prices is
available at http://www.museum.tulane.edu/pubs/ or on request. Remittances should
be made payable to Tulane University.

NEW SPECIES AND NEW RECORDS OF PSEUDOTHELPHUSID CRABS
(CRUSTACEA: BRACHYURA) FROM COLOMBIA

Gilberto Rodriguez

Centro de Ecologia, Institute Venezolano de Investigaciones Cientificas
Apartado 21827, Caracas 1020 A, Venezuela
Martha R. Campos

Universidad Nacional, Institute de Ciencias Naturales
Apartado Aereo 53416, Bogota, Colombia
And

Beatriz Lopez

Centro de Ecologia, Institute Venezolano de Investigaciones Cientificas
Apartado 21827, Caracas 1020 A, Venezuela

Abstract

The collections of freshwater crabs from Colombia deposited at the Tulane
Natural History Museum contain representatives of 14 species. Three of them,
Hypolobocera noanamensis, Lindacatalina sinuensis and Moritschus altaqueren-
sis, are new species. Several taxa present noteworthy disjunctions. Lindacatalina
orientalis (Pretzmann, 1968) and L. sumacensis Rodriguez and von Sternberg,
1998, display a trans-Andean distribution, with areas widely separated in Ecuador
and Colombia. Hypolobocera beieri Pretzmann, 1968, and H. lloroensis Campos,
1989, are found in the Pacific and Atlantic slopes of the Andes. The new species
Lindacatalina sinuensis comes from the Sinu River, which drains into the
Atlantic, while other members of the genus are located more than 800 km away,
in basins draining to the Amazon.

Introduction

The areas of distribution of most species of pseudothelphusid crabs usually
cover continuous tracts along the watercourses. In the case of many Andean
species, these areas are small, usually covering less than 100 km^. Long-range
specific distributions, particularly involving trans-basin disjunctions, are very
rare. Among the few examples that could be found in the literature are
Hypolobocera bouvieri angulata (Rathbun, 1915), from the Sierra Nevada de
Santa Marta, Colombia, and the Sierra de Perija, Venezuela (Rodriguez, 1982a),
and two species from Ecuador, Hypolobocera aequatorialis (Ortmann, 1897) and
Lindacatalina orientalis (Pretzmann, 1968), both with trans-Andean areas of dis-

Tulane Studies in Zoology and Botany 31:1-17, 2002.

1

2

Tulane Studies in Zoology and Botany

[Vol. 31, No. 2 2002]

tribution in the Pacific and Amazonian basins (Rodriguez and von Sternberg,
1998).

For more than 25 years. Dr. Alfred C. Smalley, from Tulane University, gath-
ered an important collection of freshwater crabs from the Neotropics (Rodriguez
and Fitzpatrick, 1996). Among the 14 species and subspecies from Colombia rep-
resented in this collection, three new species were discovered, one of them dis-
playing a conspicuous generic disjunction between the Atlantic slopes of northern
Colombia, the Pacific slopes of southern Colombia, and the Amazon basin of
Ecuador. One species, previously known in the literature from localities in the
Pacific slopes of the Andes, were found in the Atrato River that drains into the
Atlantic, and two other display a trans-Andean distribution, with areas widely
separated in Ecuador and southern Colombia.

In the present contribution we describe the three new species discovered and
discuss the size of the areas of distribution of the Andean Pseudothelphusidae, in
the light of the new data provided by the Smalley collection.

All the materials recorded are deposited in the Museum of Natural History of
Tulane University, New Orleans (TU), the Museo de Historia Natural, Instituto de
Ciencias Naturales, Universidad Nacional de Colombia, Bogota, Colombia (ICN-
MHN-CR) and the Museo de Biologia Marina, Universidad del Valle, Cali,
Colombia, (MBMUV). Other abbreviations used are cl. for carapace length and
cb. for carapace breadth.

Systematic Account

Family Pseudothelphusidae Rathbun, 1893
Subfamily Pseudothelphusinae Ortmann, 1893
Tribu Strengerianini Rodriguez, 1982a

Genus Chaceus Pretzmann, 1965
Chaceus pearsei (Rathbun, 1915)

Material Examined: 1 male, cl. 16.7 mm, cb. 25.9 mm, 1 soft-shelled female, 9
August 1965, road above Minca, 1,020 m alt., Cesar Department, Colombia, A.
E. Smalley, TU 5388.

Genus Phallangothelphusa Pretzmann, 1965
Phallangothelphusa dispar (Zimmer, 1912)

Material Examined: 1 female, cl. 20.0 mm, cb. 34.0 mm, 1 August 1981, El
Triunfo (4°33’N, 74°29’W), Cundinamarca Department, Colombia, TU 6304.

New Pseudothelphusid Crabs

3

Genus Strengeriana Pretzmann, 1971
Strengeriana taironae Rodriguez and Campos, 1989

Material Examined: 1 male, cl. 12.3 mm, cb. 21.6 mm, 1 juvenile female, cl.

6.0 mm, cb. 11.8 mm, 26 February 1964, San Lorenzo near Santa Marta, 2,700 m
alt., Magdalena Department, Colombia, C.A. Velasquez and F. Medem, TU 4917.

Tribu Hypolobocerini Pretzmann, 1971

Genus Hypolobocera Ortmann, 1897
Hypolobocera beieri Pretzmann, 1968
Figs. IJ-K

Material Examined: 2 males, cl. 16.0, 15.3 mm, cb. 25.0, 24.2 mm, 1 female,
cl. 11.0, cb. 16.5 mm, 24 June 1965, Cauca River tributary of Melendez River, 3
km upstream from Carmelo near Cali, Valle del Cauca Department, Colombia, A.
E. Smalley and Zapata, TU 5381; 1 male juvenile with broken carapace, 2
October 1982, Alto Anchicaya, Valle del Cauca Department, Colombia, von
Prahl, TU 6383; 1 female, cl. 18.5 mm, cb. 31.5 mm, 1 juvenile, cl. 10.5 mm, cb.
15.8 mm, 1 June 1965, Melendez River, Valle del Cauca Department, Colombia,
Moberly, TU 5459; 1 male, cl. 18.2 mm, cb. 28.4 mm, 6 August 1965, El Salto
de Jorge, Anchicaya river, tributary of Dagua River, highway to Buenaventura,
Valle del Cauca Department, Colombia, A. E. Smalley and Zapata, TU 5383; 1
male juvenile, cl. 11.2 mm, cb. 17.0 mm, 1 female, cl. 9.4 mm, cb. 14.6 mm, 30
June 1965, tributary of Melendez River, 3.5 km upstream from Carmelo, about
14.5 km SE Cali, Valle del Cauca Department, Colombia, A. E. Smalley and
Zapata, TU 5377; 7 males, cl. range 11.5 - 7.9 mm, cb. range 17.8 - 12.5 mm, 4
females, cl. range 8.6 - 9.8 mm, cb. range 12.6 - 14.5 mm, 3 juveniles, cl. 7.0,

6.5, 6.2 mm, cb. 10.5, 9.5, 8.0 mm, 14 July 1965, Quebrada de Nidios, forestry
station near Pico de Loro and 1.5 km S Ponce, Valle del Cauca Department,
Colombia, A. E. Smalley, Zapata and Adler, TU 5379; 3 males, cl. 13.5, 12.4, 9.2
mm, cb. 20.7, 19.1, 13.9 mm, 2 female, cl. 14.9, 11.4 mm, cb. 22.3, 16.9 mm, 1
June 1965, tributary of Anchicaya River, 9 km west of El Salado, Valle del Cauca
Department, Colombia, A. E. Smalley and Zapata, TU 5376; 1 male juvenile, cl.
12.7 mm, cb. 20.4 mm, 3 ovigerous females, cl. 21.8, 19.1, 18.9 mm, cb. 38.8,

31.8, 30.2 mm, 3 females, cl. 18.1, 16.5, 16.2 mm, cb. 30.2, 26.5, 25.9 mm, 1
juvenile with broken carapace, 15 March 1980, left margin of Mazamorra River,
tributary of Guachicono River, 1,400 m alt., Valle del Cauca Department,
Colombia, Matidnell, TU Acc N° 94-100; 3 males, cl. 18.5, 18.4, 15.4 mm, cb.

29.9, 29.7, 24.5 mm, 3 females, cl. 21.4, 19.9. 10.4 mm, cb. 36.1, 31.5, 16.2 mm,
5 January 1967, mountain stream near Pichinde, Valle del Cauca Department,
Colombia, Dale Little, TU 5588; 2 males, cl. 14.6, 11.8 mm, cb. 22.9, 17.8 mm.

4

Tulane Studies in Zoology and Botany

[Vol. 31, No. 2 2002]

17 June 1965, small tributary of Paila River, 4.3 km E Comito, Cauca River,

Valle del Cauca Department, Colombia, A. E. Smalley, TU 5380.

Hypolobocera noanamensis, new species
Figs. lA-H

Holotype: Adult male, cl. 50.9 mm, cb. 80.3 mm, 8 August 1969, Noanama, San
Juan River (4°42’N, 26”56’W), Choco Department, Colombia, Dale Little, TU
6191.

Paratype: 1 mature female, cl. 53.9 mm, cb. 80.3 mm, collected with holotype,
TU 5337.

Diagnosis: First male gonopods with caudal ridge strong, fusiform, thickened at
middle, ending in narrow ridge beyond lateral lobe, crossed by thin transverse
lines; lateral lobe small, subtriangular, with distal angle rounded, projected, prox-
imal border traverse or slightly convex, placed transversely in relation to axis of
appendage; cephalic surface with tuberculated crest. Apex funnel-form, distal
angle not projected, outline of apex oval in distal view, with meso-cephalic bor-
der rounded, expanded, prominent papilla near cephalic border, spermatic channel
with spiny ridge on meso-caudal side and flat, wide papilla on caudo-lateral side.

Description of Holotype: Carapace 1.58 times as wide as long, surface smooth,
regions strongly marked by deep depressions; cervical grooves deep and straight
becoming indistinct toward margins of carapace. Anterolateral margins with deep
V-shaped notch behind outer orbital angle, devoid of papillae; margin posterior to
cervical groove with approximately six wide obsolescent lobes, rest of margin
with eight wide subtriangular teeth. Postfrontal lobes rounded, forming two
prominent swellings; median groove very wide between postfrontal lobes, indis-
tinct in front of postfrontal lobes, making deep V-shaped incision on upper mar-
gin of front. Surface of carapace in front of postfrontal lobes inclined anteriorly
and towards middle. Upper margin of front bilobed in dorsal view, angulated,
devoid of defined papillae; lower margin strongly sinuous in frontal view and
conspicuously thickened; space between both margins very narrow.

Third maxilliped with lateral border of merus of endognath straight, ending
distally in rounded lobe and deep depression; exognath 0.3 length of ischium of
endognath. Orifice of efferent branchial channel open. Chelipeds moderately
unequal, finger of larger one (right) with small gap when closed, palm slightly
swollen; rows of small black points on external and upper surfaces of mobile fin-
ger and external surface of fixed finger, scattered small tubercles on internal
lower surface of palm. Carpus with prominent spine on internal upper margin;
internal upper margin of merus with row of teeth, row of smaller teeth on exter-

New Pseudothelphusid Crabs

5

Figure 1. Hypolohocera noanamensis, new species, male holotype from
Noanama, San Juan River, Choco Department, Colombia (TU 6191), A-D, first
left gonopod: A, caudal view; B, lateral view; C, cephalic view; D, apex, distal
view; E, aperture of efferent channel; F, larger chela, right; G, left third maxil-
liped; H, dorsal view of right side of carapace; Hypolohocera beieri Pretzmann,
1968, male cb. 24.2 mm, from Cauca River, near Cali, Colombia (TU 5381), first
left gonopod; I, caudal view; J, lateral view; K, apex, distal view; r, spiny ridge;
p, papilla. Scales A-D = 1 mm, E, G = 3 mm, F,H = 1 cm, I, J, K = 1 mm.

6

Tulane Studies in Zoology and Botany

[Vol. 31, No. 2 2002]

nal margin and prominent tubercles on lower margin.

First male gonopods slender, arched cephalically ; caudal ridge strong,
fusiform, thickened at middle, ending in narrow ridge beyond lateral lobe,
crossed by thin transverse lines; lateral lobe small, subtriangular, with distal angle
rounded, projected, proximal border traverse or slightly convex, placed trans-
versely in relation to axis of appendage; cephalic surface with tuberculated crest.
Apex funnel-form, with distal angle not projected in lateral view, outline of apex
oval in distal view, with meso-cephalic border rounded, expanded, prominent
papilla near cephalic border, spermatic channel with spiny ridge on meso-caudal
side and flat, wide papilla on caudo-lateral side.

Remarks: The first gonopods of Hypolobocera noanamensis resemble those of
H. bouvieri rotundilobata Rodriguez, 1994, and H. beieri, but it can be distin-
guished from both by the spiny ridge on the caudal side of the spermatic channel
(Fig. ID, r), the shape and relative development of the lateral lobe, and the out-
line of the apex. In H. beieri the apex forms a beak-like projection on the cephal-
ic side (Fig. IJ) and the apical outline in distal view (Fig. IK) has the meso-
cephalic border transverse, not rounded and expanded (Fig. ID). In H. bouvieri
rotundilobata the apical outline in distal view has the meso-cephalic border sub-
triangular, projected.

Size: This is a large species. The mature female paratype has a cb. of 80.3 mm

Etymology: The species is named after the Noanama locality in the San Juan
River, Colombia, where the species was collected.

Hypolobocera bouvieri bouvieri (Rathbun, 1898)

Material Examined: I male, cl. 25.6 mm, cb. 40.4 mm, 22 September 1982,
Finca Villa Christina, Municipio Santardercito, Cundinamarca Department,
Colombia, von Prahl, TU 6382; I male, cl. 35.0 mm, cb. 57.5 mm, 7 November
1981, La Mesa, 1,500 m alt., Cundinamarca Department, Colombia, TU 6355; 2
males, cl. 21.8, 16.4 mm, cb. 32.1, 24.4 mm, 21 April 1982, Sasaima, 1,500 m
alt. (4°58’N, 74°26’W), Cundinamarca Department, Colombia, von Prahl, TU
6376; I male, cl. 52.4 mm, cb. 89.6 mm, 28 November 1981, Doima River,
Municipio Doima, Tolima Department, Magdalena basin, Colombia, TU Acc N°
94-100.

Hypolobocera bouvieri angulata (Rathbun, 1915)

Material Examined: I immature male, cl. 20.0 mm, cb. 30.2 mm, 9 August
1965, highway from Santa Marta to Riohacha, 10 km from Bonda, Magdalena
Department, Colombia, A. E. Smalley, TU 5382.

New Pseudothelphusid Crabs

7

Hypolobocera gorgonensis von Prahl, 1983

Material Examined: 1 male, cl. 50.6 mm, cb. 83.3 mm, 22 August 1980,
Gorgona Island, Valle del Cauca Department, Colombia, von Prahl, TU 6303.

Hypolobocera lloroensis Campos, 1989

Material Examined: 1 male, cl. 24.4 mm, cb. 40.0 mm, 2 females, cl. 26.3,

13.8 mm, cb. 43.9, 21.8 mm, 8 August 1969, Istmina, San Juan River, Choco
Department, Colombia, Dale Little, TU 6193.

Hypolobocera merenbergiensis von Prahl and Giraldo, 1985

Material Examined: 1 male holotype, cl. 13.1 mm, cb. 20.1 mm, 3 females, cl.
14.0, 13.4, 11.5 mm, cb. 23.4, 21.8, 18.2 mm, MBMUV-82051, 1 female
paratype, cl. 14.2 mm, cb. 22.5 mm, ICN-MHN-CR 0541, 2 males, cl. 12.3,

10.5 mm, cb. 18.8, 16.5 mm, TU 6369, 8 April 1982, Finca Merenberg, 2,300 m
alt.. Cordillera Central, Huila Department, Colombia, H. von Prahl and J.
Giraldo.

Remarks: We list above the seven extant specimens from the original collection
deposited in several museums. Von Prahl & Giraldo (1984) gave a list of six
specimens with their respective measurements and, separately, they designated
the holotype and four paratypes, without indication of their sizes. They did not
recorded the sex of the specimens measured, but gave the length of the gonopod
for one of them (cb. 19.8 mm). This is the only male in the collections of the
MBMUV and consequently must be considered as the male holotype designated
by von Prahl & Giraldo (1984). They designated as male paratype a specimen
deposited at TU. Since there are two male specimens at TU, this probably refers
to the largest one.

Von Prahl & Giraldo (1984) recorded three female paratypes, one of them
deposited at the ICN-MHN-CR, and two at the MBMUV. They stated that these
last were marked as numbers 48 and 49, but we could not find the respective
labels, and furthermore, there are three females at the MBMUV instead of the
two implied in the designation of the paratypes. Consequently, we cannot deter-
mine to which of the females in the MBMUV corresponds the status of
paratypes. Our measurements of the cb. show slight discrepancies with the
measurements given by Von Prahl & Giraldo (1985). The discrepancies are even
larger for cl. measurements.

8

Tulane Studies in Zoology and Botany

[Vol. 31, No. 2 2002]

Lindacatalina Pretzmann, 1977
Lindacatalina sinuensis, new species
Figs. 2A-H

Holotype: Adult male, cl. 10.0 mm, cb. 15.6 mm, 3 August 1965, 80 km up Sinu
River, from Tierra Alta, small waterfall tributary of Sinu River, Cordoba
Department, Colombia, Kraig Adler, TU 5337.

Paratype: 1 ovigerous female, cl. 14.9 mm, cb. 23.8 mm, collected with holo-
type, TU 5337.

Diagnosis: First male gonopods with lateral margin divided longitudinally into
two distinct halves, one caudal, small, rounded, another lateral, larger, circular in
outline, covered by stout spines. Apex in distal view elongated in mesolateral
axis, oval, with flat papilla over field of spines.

Description of Holotype: Carapace 1.56 times as wide as long, surface covered
with small papillae not visible to naked eye, regions of carapace clearly defined;
cervical grooves shallow and straight, becoming indistinct toward margins of
carapace. Anterolateral margins with wide shallow notch behind outer orbital
angle, notch with papillae on margin; not meeting anteroextemal orbital angle,
continuing above orbital margin; posterior to cervical groove with approximately
15 small teeth. Postfrontal lobes small, forming two slight swellings placed trans-
versely in relation to middle axis of carapace; median groove shallow, ill defined.
Surface of carapace in front of postfrontal lobes inclined anteriorly. Upper margin
of front bilobed in dorsal view and bearing row of ill defined papillae; lower
margin strongly sinuous in frontal view and conspicuously thickened; space
between both margins moderately high.

Third maxilliped with lateral border of merus of endognath forming regular
curve, with deep depression on distal part of external margin. Exognath of third
maxilliped 0.3 length of ischium of endognath. Orifice of efferent branchial chan-
nel open. Chelipeds unequal, finger of larger one with small gap when closed,
palm slightly swollen. Carpus with prominent spine on internal upper margin;
internal upper margin of merus with rows of teeth.

First male gonopods arched cephalically, caudal ridge prominent, with few
transverse wrinkles at middle section. Lateral margin divided longitudinally into
two distinct halves, one caudal, small, rounded, formed by distal expansion of
caudal ridge, and another lateral, larger, circular in outline, covered by stout
spines. Apex in distal view elongated in mesolateral axis, oval, with flat papilla
over field of spines.

Remarks: We are tentatively placing this species in the genus Lindacatalina due
to the presence of two parallel lateral lobes, one of them densely covered with

New Pseudothelphusid Crabs

Figure 2. Lindacatalina sinuensis, new species, male holotype from Sinu River,
Cordoba Department, Colombia (TU 5337), A-D, first left gonopod: A, caudal
view; B, lateral view; C, latero-cephalic view; D, apex, distal view; E, aperture of
efferent channel; F, left third maxilliped; G, larger chela, right; H, dorsal view of
right side of carapace. Scales A-F = 1 mm, G-H = 3 mm.

10

Tulane Studies in Zoology and Botany

[Vol. 31, No. 2 2002]

spines. However, the length of exognath in Lindacatalina is usually more than
0.45 times the length of the ischium of endognath, and the spines over the lateral
lobe of the gonopods are similar to those over the field of spines (Rodriguez and
von Sternberg, 1998), whereas in the new species the proportions of the above
structure is 0.3, and the spines over the lateral lobe are larger and stronger than
those over the field of spines. Lindacatalina sinuensis resembles L. sumacensis
in the caudal ridge covered by minute transverse wrinkles, but the latero-cephalic
half of the lateral lobe is not expanded distally as is the case in L. sumacensis.

Size: This is a relatively small species for the family. The ovigerous female
paratyper has a cb. of 28.3 mm.

Etymology: The species is named after the Sinu River, in whose basin the
species was collected.

Lindacatalina orientalis (Pretzmann, 1968)

Material Examined: 1 male, cl. 15.7 mm, cb. 24.7 mm, 1 female, cl. 12.5 mm,
cb. 18.8 mm, 24 October 1969, Sibundoy, Putumayo Department, Colombia,

Dale Little, TU 6192.

Lindacatalina sumacensis Rodriguez and von Sternberg, 1998

Material Examined: 1 male, 19.6 mm, cb. 31.2 mm, 7 July 1969, Mocoa,
Putumayo Department, Colombia, Dale Little, TU 6194.

Moritschus Pretzmann, 1965
Moritschus altaquerensis, new species
Figs. 3A-H

Holotype: 1 male, cl. 16.0 mm, cb.26.7 mm, 29 July 1999, Vereda El Mirador,
Corregimiento Altaquer, 1,200 m alt. (1^13’49”N, 78^2’ 18” W), Municipio
Barbacoas, Narino Department, M. R. Campos, ICN-MHN-CR 1769.

Paratype: 9 males, cl. range 13.5 - 10.6 mm, cb. range 22.1 - 16.7 mm, 7
females, cl. range 15.0 - 10.8 mm, cb. range 24.0 - 17.1 mm, collected with
holotype, ICN-MHN-CR 1770.

Additional Material Examined: 5 males, cl. range 12.5 - 10.0 mm, cb. range
20.2 - 16.0 mm, 6 females, cl. range 15.2 - 11.8 mm, cb. range 24.4 - 18.8 mm,
30 July 1999, Vereda Ospina, 1,3(X) m alt. (1^13’ 18”N, 78^02’ 10”W), Municipio
Ricaurte, Narino Department, Colombia, M. R. Campos, ICN-MHN-CR 1771; 6

New Pseudothelphusid Crabs

11

Figure 3. Moritschus altaquerensis, new species., male holotype from Altaquer,
Municipio Barbacoas, Narino Department, Colombia (ICN-MHN-CR 1769), A-
D, first left gonopod: A, caudal view; B, lateral view; C, cephalic view; D, apex,
distal view; E, aperture of efferent channel; F, larger chela, left; G, left third max-
illiped; H, dorsal view of right side of carapace. Scales A-E = 1 mm, F-H = 3

mm.

12

Tulane Studies in Zoology and Botany

[Vol. 31, No. 2 2002]

males, cl. range 15.2 - 12.6 mm, cb. range 25.1 - 19.4 mm, 1 female, cl. 13.8
mm, cb. 22.1 mm, 30 July 1999, stream affluent of Nembi River, Vereda Ospina,
1,200 m alt. (1^13’23”N, 78^02’ 15”W), Municipio Ricaurte, Narino Department,
Colombia, M. R. Campos, ICN-MHN-CR 1772; 1 male, cl. 15.3 mm, cb. 25.4
mm, 1 female, cl. 14.6 mm, cb. 23.5 mm, 11 July 1969, stream near Altaquer,
Narino Department, Dale Little, TU 6178.

Diagnosis: First male gonopod with lateral lobe prominent, forming winged sub-
triangular expansion, extending from middle of appendage to below apex, caudal
end produced in short beak. Apex in distal view strongly elongated along meso-
lateral axis, arched, with minute spines on caudal margin; elongated process over
field of spines with two rudimentary papillae.

Description of Holotype: Carapace 1.66 times as wide as long, dorsal surface
smooth, limits between regions well marked; cervical groove shallow and
straight, wide posteriorly, narrow anteriorly, reaching margin of carapace.
Anterolateral margin with shallow notch behind orbit, not reaching outer orbital
angle, shallow notch at level of cervical groove, anterolateral border with about
12 papillae. Postfrontal lobes small, oval, indicated anteriorly by transverse
slightly depression, median groove obsolete. Surface of carapace in front of post-
frontal lobes inclined anteriorly. Superior frontal border smooth, not bilobed in
dorsal view, inferior border visible in dorsal view, strongly sinuous in frontal
view and conspicuously thickened, front narrow.

Chelipeds strongly unequal, palm of larger one (left) inflated, fingers gaping,
movable finger slightly arched. Third maxilliped with merus of endognath regu-
larly curved, with shallow depression on distal part of external margin; exognath
0.36 length of ischium of third maxilliped. Orifice of efferent branchial channel
open.

First male gonopod slender, arched laterally; caudal ridge elongated longitu-
dinally; lateral lobe prominent, forming winged subtriangular expansion, extend-
ing from middle of appendage to below apex, caudal end produced in short beak;
apex in distal view strongly elongated along mesolateral axis, arched, with
minute spines on caudal margin; elongated process over field of spines with two
rudimentary papillae.

Color: The holotype is dark brown with paler brown specks on the dorsal side of
the carapace. The walking legs are brown dorsally, and light brown ventrally. The
chelae are brown dorsally, and buffy-brown ventrally. The ventral surface of the
carapace is buffy-brown.

Habitat: The specimens were collected in shady, moist banks of springs and
small streams. They were found in soft mud, under rocks, or in borrows.

New Pseudothelphusid Crabs

13

Figure 4. Geographic distribution of Colombian species of Pseudothelphusidae.
Triangles, new species; open circles, new records; dots, previous records.

14

Tulane Studies in Zoology and Botany

[Vol. 31, No. 2 2002]

Remarks: The first gonopods of the new species resemble those of Moritschus
ecuadorensis, but its lateral margin forms a well defined subtriangular expansion,
whereas in M. ecuadorensis it stretches out progressively towards the apex.
Furthermore the elongated process over the field of spines in M. ecuadorensis
presents a notch between two rudimentary papillae, while in M. altaquerensis the
apex of this process is continuous.

Size: This is a relatively small species for the family. A mature female has a cb.
of 24.4 mm (ICN-MHN-CR 1771.

Etymology: The specific epithet is derived from the locality’s name where the
species was collected.

Biogeography

The records given above shed new light on the sizes of the areas of distribu-
tion of several Colombian species. The smallest area is that of Hypolobocera gor-
gonensis, an insular species restricted to Gorgona Island. The total surface of this

island is only 15 km^. Chaceus pearsei and Strengeriana taironae are restricted
to localities on the north and western slopes of the Sierra Nevada de Santa Marta.
Our material of these three species come from the type localities, as well as that
of Hypolobocera bouvieri angulata, but this last species posses a very large area
of distribution which extends to the Sierra de Perija and the Cordillera de Merida
in Venezuela (Fig. 4).

Hypolobocera bouvieri bouvieri has a very extensive distribution on the
slopes of the Central and Eastern Cordilleras draining to the Magdalena River.
Our specimens are nested within this area of distribution, towards the southern
portion of its range. Our specimens of Phallangothelphusa dispar are also nested
within the known area of distribution. The distribution area of this species over-
laps, in the southern portion with that of H. bouvieri bouvieri. Hypolobocera
merenbergiensis is known only from a single locality in the Cordillera Central, in
the upper reaches of the Magdalena River.

The only previous record of Hypolobocera lloroensis is from the headwaters
of the Atrato River, which flows northwards into the Gulf of Uraba, on the
Atlantic coast. Our material comes from a region that, notwithstanding its prox-
imity to the type locality, is situated on the basin on the San Juan River, which
flows southwards to the Pacific coast of Colombia. This phenomenon is possibly
due to intermittent communication of both basins during seasonal flooding in
areas below 1(X) m of altitude. A similar phenomenon has been recorded for the
Orinoco, Amazon, and the Atlantic rivers of the Guianas (Rodriguez, 1982b).

Hypolobocera beieri displays an interesting trans-Andean distribution.
Numerous specimens of H. beieri present in the Tulane Collection come from the

New Pseudothelphusid Crabs

15

main area of distribution of the species, in the upper reaches of the Cauca River,
which drains into the Atlantic, but others were collected in the basin of the
Anchicaya River, in the Valle del Cauca Department, that drains into the Pacific.
Rodriguez (1994) also gives an isolated record of this species from the upper
reaches of the San Juan River, near the type locality of another Hypolohocera, H.
bouvieri rotundilobata Rodriguez, 1994. In the same San Juan River basin,
towards its middle course, was collected Hypolobocera noanamensis.

Of special interest are the distribution areas of those Colombian species relat-
ed to taxa distributed in the Ecuadorian Andes (Rodriguez and von Sternberg,
1998). The presence of Lindacatalina orientalis in Colombia has already been
recorded by von Prahl (1988). The range of this species in Ecuador covers the
basins of rivers draining both to the Amazon basin, through the Napo and Pastaza
rivers, and to the Pacific through the Esmeraldas River. The present record fol-
lows the same pattern of trans-Andean distribution in southern Colombia, through
the Putumayo and Telembi rivers, respectively. The record of L. sumacensis also
extends the range of the species to Colombia, but in this case, within the Amazon
basin.

The new species Moritschus altaquerensis belongs to a genus present in
Ecuador and the southern Department of Narino in Colombia. The only localities
known for Moritschus altaquerensis are in the Telembi River (affluent of Patia
River), less than 30 km from the type locality of M. narinnensis Campos and
Rodriguez, 1988, and on the same river basin. The record of Lindacatalina ori-
entalis given by von Prahl (1988) also comes from this river. Notwithstanding the
close geographical proximity of Moritschus altaquerensis and M. narinnensis, the
former species is more closely related in the morphology of its gonopod to M.
ecuadorensis (Rathbun, 1897), a species restricted to the upper reaches of the
Esmeraldas River in Ecuador, which is the next large river to the south of the
Patia River, than to M. narinnensis.

Noteworthy is the disjunction of Lindacatalina sinuensis in regard to the rest
of the species in the genus. The species of Lindacatalina are restricted to a sector
of the eastern slope of the Ecuadorian Andes that drains to the Amazon basin
(Rodriguez and von Sternberg, 1998), with the exception of Lindacatalina orien-
talis discussed above. The area of L. sinuensis is restricted to the Sinu River that
drains into the Atlantic at the northernmost comer of Colombia, at a distance of
more than 800 km from the main distribution area of the genus.

Acknowledgements

We express our thanks to Drs. Henry L. Bart, Jr., and Dr. Joseph F.

Fitzpatrick, Jr. for entmsting the curation of this valuable collection to the senior
author, and to Mr. Mike Taylor for his untiring help with the handling of the
material.

16

Tulane Studies in Zoology and Botany

[Vol. 31, No. 2 2002]

Literature Cited

CAMPOS, M. 1989. Nuevas especies de cangrejos de agua dulce del genero
Hypolobocera (Crustacea: Decapoda: Pseudothelphusidae) para Colombia.
Trianea 3: 143-147.

CAMPOS, M. and G. RODRIGUEZ. 1988. Notes on the freshwater crabs of the
genus Moritschus Pretzmann, 1965 (Crustacea: Decapoda:

Pseudothelphusidae) with description of M. narinnensis from southern
Colombia. Proceedings of the Biological Society of Washington 101(3): 640-
643.

ORTMANN, A. 1893. Die Dekapoden-Krebse des Strassburger Museums, mit
besonderer Beriicksichtigung der von Herm Dr. Doderlein bei Japan und bei
den Liu-Kiu-Inseln gesammelten und zur Zeit in Strassburger Museum aufbe-
wahrten Formen. VII. Theil. Abtheilung: Brachyura (Brachyura genuina Boas)
II. Unterabtheilung: Cancroidea, 2 Section: Cancrinea, 1. Gruppe:
Cyclometopa. Zoologische Jahrbticher, Abtheilung fur Systematik, Geographie
und Biologie der Thiere 7: 411-495.

ORTMANN, A. 1897. Carcinologische Studien. Zoologische Jahrbiicher,
Abtheilung fiir Systematik,Geographie und Biologie der Thiere 10:258-372.

PRAHL, H. VON. 1983. Hypolobocera gorgonensis sp. nov. (Crustacea:

Brachyura: Pseudothelphusidae) un nuevo cangrejo de agua dulce de la isla de
Gorgona, Colombia. Cespedesia 12(45-46): 105-110.

PRAHL, H. VON. 1988. Fresh- water crabs (Crustacea: Brachyura:
Pseudothelphusidae) of the Pacific drainage of Colombia. Zoologische
Jahrbiicher, Abtheilung fiir Systematik, Geographie und Biologie der Thiere
115: 171-186.

PRAHL, H. VON, and J. GIRALDO. 1985. Un nuevo cangrejo de agua dulce de
la Cordillera Central de Colombia. Lozania 49: 1-5.

PRETZMANN, G. 1965. Vorlaufiger Bericht uber die Familie

Pseudothelphusidae. Anzeiger der Mathematisch Naturwissenschaftliche
Klasse der Osterreichischen Akademie der Wissenschaften (1) 1: 1-10.

PRETZMANN, G. 1968. Neue Sudamerikanische Siisswasserkrabben der
Gattung Pseudothelphusa. Entomologisches Nachrichtenblatt, Wien 15(1): 1-
15.

PRETZMANN, G. 1971. Fortschritte in der Klassifizierung der

Pseudothelphusidae. Sitzungsberichten der Osterreichischen Akademie der
Wissenschaften, Mathematisch Naturwissenschaftliche Klasse (1) 179 (1-4):
15-24.

PRETZMANN, G. 1977. Zur Taxonomie, Chorologie und Systematik der mitte-
landischen Hypolobocerini. Sitzungsberichten der Osterreichischen Akademie
der Wissenschaften, Mathematisch Naturwissenschaftliche Klasse (1) 186:
429-439.

New Pseudothelphusid Crabs

17

RATHE UN, M. J. 1893. Descriptions of new species of American freshwater
crabs. Proceedings of the United States National Museum 16 (959): 649-661.

RATHE UN, M. J. 1897. Descriptions de nouvelles especes de Crabes d’eau
douce appartenant aux collections du Museum d’Histoire naturelle de Paris.
Eulletin du Museum nationale d’Histoire naturelle (Paris) 3 (2): 58-61.

RATHE UN, M. J. 1898. A contribution to a knowledge of the freshwater crabs of
America. The Pseudothelphusinae. Proceedings of the United States National
Museum 21(1158): 507-537.

RATHE UN, M. J. 1905. Les crabes d’eau douce (Potamonidae). Nouvelles
Archives du Museum d’Histoire Naturelle, Paris (4) 7: 159-321.

RATHE UN, M. J. 1915. New fresh-water crabs (Pseudothelphusa) from
Colombia. Proceedings of the Eiological Society of Washington 28: 95-100.

RODRIGUEZ, G. 1982a. Les crabes d’eau douce d’Amerique. Famille des
Pseudothelphusidae. Faune Tropicale 22: 1-223.

RODRIGUEZ, G. 1982b. The freshwater shrimps (Crustacea, Decapoda,

Natantia) of the Orinoco basin and the Venezuelan Guayana. Journal of
Crustacean Eiology 2: 378-391.

RODRIGUEZ, G. 1994. A revision of the type material of some species of
Hypolobocera and Ptychophallus (Crustacea: Decapoda: Pseudothelphusidae)
in the National Museum of Natural History, Washington, D. C. with descrip-
tions of a new species and a new subspecies. Proceedings of the Eiological
Society of Washington 107: 296-307.

RODRIGUEZ, G., and M. CAMPOS. 1989. Cladistic relationships of freshwater
crabs of the tribe Strengerianini (Decapoda: Pseudothelphusidae) from the
northern Andes, with comments on their biogeography and descriptions of new
species. Journal of Crustacean Eiology 9: 141-156.

RODRIGUEZ, G. and J.F. FITZPATRICK. 1996. Alfred Evans Smalley. Journal
of Crustacean Eiology 16: 214-215.

RODRIGUEZ, G., and R. VON STERNEERG. 1998. A revision of the freshwa-
ter crabs of the family Pseudothelphusidae (Decapoda: Erachyura) from
Ecuador. Proceedings of the Eiological Society of Washington 111: 110-139.

ZIMMER, C., 1912. Eeitrag zur Kentniss der SUsswasser dekapoden

Kolumbiens. In: O. Fuhrmann and E. Mayor, eds.. Voyage d’exploration scien-
tifique en Colombie. Memoires de la Societe neuchateloise des Sciences
naturelles 5: 1-8.

•4^*^ .lAifil Hi I? •'

t.*^, f '''.

A REVIEW OF THE SPINYCHEEK SLEEPERS, GENUS ELEOTRIS
(TELEOSTEI: ELEOTRIDAE), OF THE WESTERN HEMISPHERE, WITH
COMPARISON TO THE WEST AFRICAN SPECIES

Frank Pezold

Department of Biology and Museum of Natural History (Zoology)
The University of Louisiana - Monroe
Monroe, LA 71209-0500
And

Bryan Cage

Department of Biology
University of Mississippi
University, MS 38677

Abstract

Species of the genus Eleotris from the eastern Pacific and western
Atlantic are reviewed. Three species are recognized from the eastern Pacific
region. The wide-ranging Eleotris picta Kner, Eleotris tubularis Heller and
Snodgrass (endemic to Cocos Island), and Eleotris tecta Bussing (limited to
Costa Rica, Panama and Colombia) are distinguishable by scale counts (size),
cephalic neuromast features and morphology of the urogenital papilla. Three
western Atlantic species are recognized. Eleotris pisonis (Gmelin) is a continental
South American species ranging from southern Brazil to the Orinoco River delta
in eastern Venezuela. A second primarily continental species, E. amblyopsis
(Cope), is distributed from Brazil through the Caribbean basin and Gulf of
Mexico to North Carolina. Eleotris pemiger (Cope), largely Caribbean in distri-
bution, is the prevalent species in the Antilles and Quintana Roo, but is also sym-
patric with E. amblyopsis in Central America. The three western Atlantic species
differ in scale counts and cephalic neuromast patterns. Eleotris daganensis
Steindachner of West Africa is morphologically indistinguishable from E, ambly-
opsis. Lacking evidence of connectivity between eastern Atlantic populations of
E. daganensis and western Atlantic populations of E. amblyopsis, the species are
not synonymized. Eleotris annobonensis Blanc, Cadenat and Stauch is similar to
E. pemiger and North American populations of Eleotris amblyopsis. The remain-
ing three West African species of Eleotris are most similar to E. picta of the east-
ern Pacific. Eleotris and Erotelis are recognized as distinct genera.

Tulane Studies in Zoology and Botany 31:19-63, 2002.

19

20

Tulane Studies in Zoology and Botany

[VoL 31, No. 2 2002]

Introduction

Spinycheek sleepers in the genus Eleotris are sit-and-wait predators char-
acterized by a distinctive eleotrid morphology - moderately blunt large head, tor-
pedo-like body form, broad, rounded caudal fm and prominent lower jaw (Fig.

1). In the tropics and subtropics, they are a common element in estuaries, insular
freshwater streams and small continental streams with poorly developed freshwa-
ter ichthyofaunas. Although they are not important as food fishes in most places,
spinycheek sleepers are likely an important component of these subtropical and
tropical ecosystems, both as predators and, in the larval stage, as a food source
(e.g. Nordlie, 1979, 1981; Perrone and Vieira, 1990, 1991). Despite the growing
interest in their ecological roles in freshwater and estuarine communities, little
attention has been given to the systematics of Eleotris species beyond the
description of new species. The lack of significant interest is likely the synergis-
tic result of the cosmopolitan distribution of the genus, a plethora of nominal
species, few diagnostic characters with which to work and a confusing amount of
state variation for those characters.

When it was extricated from Gobius 200 years ago, species included in
the genus Eleotris were distinguished in that they lacked a cup-like base formed
by joined pelvic fins, although the pelvic fins were described as connected by
membranes (Bloch and Schneider, 1801). Through time the genus came to be
identified with gobioid species with separate pelvic fins; species characterized by
that feature were routinely tossed into the Eleotris bin throughout the 1800’s and
early 1900’s (more than 200 species [Eschmeyer, 1998]). Although most of these
species have been removed in turn to other genera, and in some cases families,
the genus has never been revised or reviewed in its entirety and a number of
nominal taxa remain for which validity has never been tested. Only two regional
reviews have been accomplished. Akihito (1967) examined species from Japan
and compared them to several other Indo-Pacific species. His study was particu-
larly important in demonstrating the significance of the free neuromast patterns
on the head for diagnosing species. More recently. Miller (1998) reviewed
species from the eastern Atlantic. As did Akihito, he found cephalic free neuro-
mast patterns and differences in squamation the most useful characters in separat-
ing species. Miller also offered what he termed a phenetic diagnosis of the genus
based on the antrorse spine on the preoperculum, axial osteology and features of
the cephalic lateralis system. He proposed that the genus Erotelis was a junior
synonym of Eleotris as the former’s two included species shared these diagnostic
features.

This paper reports the results of a review of Eleotris species from the
Western Hemisphere, the eastern Pacific and western Atlantic basins. Species are
compared to West African Eleotris and a key is provided to the species of the

Review of the Spinycheek Sleepers

21

western Atlantic and eastern Pacific oceans. Eleotris and Erotelis are distin-
guished and their separate recognition is recommended.

Methods

Meristic and mensural characters and procedures follow Hubbs and
Lagler (1958), except for the following: preanal length, postanal length, head
width at the preopercle, length of the urogenital papilla, body width at pectoral
fm base, and body width at second dorsal fm origin. Preanal length is the least
distance from the vent to the tip of the snout, and postanal length is the least dis-
tance from the vent to the hypural of the caudal peduncle. Head width is the
greatest lateral distance through the fish at the preopercle. Urogenital papilla
length was measured from the vent to the distal tip of the papilla. Body widths
are the greatest distance through the body at the bases of the left and right pec-
toral fins, and the greatest distance through the body at the origin of the second
dorsal fm. Head length, head width at preopercle, pre-dorsal length, nape height,
preanal length, postanal length, caudal peduncle length, body width at pectoral
fm base, body width at second dorsal fm origin, urogenital papilla length, pec-
toral fm length, pelvic fm length, and caudal fm length are reported as propor-
tions of standard length (SL). Interorbital width, upper jaw length, orbit length,
and snout length are reported as proportions of head length (HL). First dorsal fm
pterygiophore insertion pattern formulas are as given in Birdsong et al. (1988).
Meristic information is reported as mode (range).

Principal component analyses were used to separately investigate mor-
phological and meristic variation among species. Principal component scores
derived from untransformed morphological variables were regressed against stan-
dard length. To remove size as a factor in the analyses, residual values for each
specimen were then used in the scatter plots. Preliminary analysis disclosed no
sexual dimorphism for morphological features other than urogenital papilla form,
which was not included in the analysis. Males and females were pooled in subse-

22

Tulane Studies in Zoology and Botany

[VoL 31, No. 2 2002]

quent analyses. Samples of the wide-ranging species Eleotris amblyopsis and E.
picta were studied for geographic variation. For these analyses, Eleotris picta
samples were subdivided into roughly equal latitudinal regions because there are
no obvious geographic barriers or restrictions to movement throughout its range.
Eleotris amblyopsis samples were subdivided into natural geographic units based
upon regional current patterns, land mass distributions and distribution patterns
observed for other gobioid fishes in the western Atlantic (e.g. Pezold and Grady,
1989).

Cephalic neuromast distribution patterns and urogenital papilla anatomy
were examined and illustrated using a dissecting microscope and camera lucida.
Cephalic neuromast patterns are described using terminology developed by Sanzo
(1911) with modifications employed by Miller and Wongrat (1991). Transverse
suborbital rows are designated with Arabic numbers and major horizontal rows
on the cheek are indicated with the letters b and d. To simplify references to the
particular transverse suborbital rows crossing row d, a formula of row numbers
separated by periods is used. For example, if rows 2, 3 and 4 cross row d, this
condition is represented by the formula “2.3.4.” Similarly, the notation “2.4” indi-
cates rows 2 and 4 cross d. Some species had incompletely formed rows, termed
“segments.” To be considered a “row” instead of a “segment,” a line of papillae
would have to reach at least half the distance between the eye and horizontal row
d. Transverse opercular rows are labeled ot and of . Upper and lower longitudinal
rows on the operculum are labeled os and oi, respectively (Fig.2).

Decisions of species limits were rooted in the Evolutionary Species
Concept (Wiley and May den, 2(X)0). Operational applications differed for sym-
patric and allopatric populations. Reproductive isolation was inferred from con-
cordant patterns of variation of at least two morphological characters in sym-
patric population samples (Brown and Wilson, 1952; Grady and Quattro, 2000).
For species determinations of allopatric population samples, the degree of mor-
phological differentiation between samples was compared to that observed among
sympatric samples. In the latter case, the possibility for gene flow through disper-
sion or dispersal, and probable length of separation were also considered.

Museum acronyms follow Leviton et al., 1985 and Leviton and Gibbs,
1988, except where noted in the list of materials examined.

Results

Eastern Pacihc. Three species of Eleotris are easily distinguished in the eastern
Pacific basin using meristic features, cephalic neuromast patterns, urogenital
papillae and pigmentation. Morphological variation was largely uninformative
within or among species with the exception of the urogenital papilla (Table 1).
Although some differences among species appeared in the simple examination of
relative proportions presented in Tablet, no significant interspecific distinctions

Review of the Spinycheek Sleepers

23

Figure 2. Cephalic free neuromast patterns of eastern Pacific Eleotris species,
a) Eleotris picta b) Eleotris tubularis c) Eleotris tecta.

24

Tulane Studies in Zoology and Botany

[Vol. 31, No. 2 2002]

were retained in the multivariate analysis after principal component scores were
corrected for size correlations. The urogenital papilla of female Eleotris picta is
different from that of both E. tubularis and E. tecta. The posterior margin is not
as rounded and has a fringe of fmger-like projections that flare in the largest
females (Fig. 3). Females of E. tecta and E. tubularis have robust, round urogeni-
tal papillae with fewer fringes. The urogenital papillae of female specimens of
Eleotris picta develop more slowly then their smaller relatives. Among female
specimens of E. tecta and E. tubularis, urogenital papillae vary little other than
the development of longer pre-vent furrows in E. tubularis (Fig. 3). Urogenital
papillae development in males occurs at a smaller size in Eleotris tubularis and
E. tecta compared to E. picta (Fig. 4).

Principal components analysis of meristic features clearly separated the
three species (Fig. 5). Eleotris picta has more predorsal scales than either eastern
Pacific congener (Table 2). It also has more lateral scale rows than the others, but
does overlap with E. tecta. Eleotris tubularis has a lower lateral scale count than
either eastern Pacific congener (Table 2). When comparing transverse scale
counts, E. picta has a much higher count than E. tubularis, whereas E. tecta has
an intermediate range. Eleotris tecta is distinct from E. picta in having a lower
range for caudal peduncle scale count, but is not distinguished by that feature
from E. tubularis (Table 2). Although overlap occurs, the means and modes for
pectoral-fm elements are distinct for each species. Modes observed were 18 for
Eleotris picta, 17 for E. tecta and 16 for E. tubularis (Table 2). Principal compo-
nents analysis of meristic characters among the regional samples of Eleotris picta
suggested clinal variation from high values in the region of Guatemala-Costa
Rica to lower values both southward and northward. The pattern reflected varia-
tion primarily of transverse scale row counts and caudal peduncle scale counts
(Table 3). Specimens from Mexico (regions 1 and 2 in Table 3) also had higher
numbers of lateral and predorsal scales.

Eleotris from the eastern Pacific differ consistently in cephalic free neu-
romast patterns (Fig. 2). Though intraspecific variation can be observed, species
are distinguished by the number and specificity of transverse suborbital rows
extending ventrally beyond horizontal row d, the presence of row ot\ and the
union or separation of opercular rows os and oi. The transverse suborbital rows
of Eleotris picta and E. tubularis have the same configuration, rows 2 and 4
extend well ventral to horizontal row d in both species (2.4 pattern), but addition-
al segments, incompletely formed superficial rows occur between the transverse
rows in E. picta. In contrast to its two congeners, transverse suborbital row 3
crosses horizontal row d in E. tecta (Fig. 2). The cephalic neuromast pattern of
E. picta is also distinguished from those of E. tecta and E. tubularis by the addi-
tion of an opercular posterior vertical row of . In Eleotris picta, of connects
with opercular row os to form a fork on the upper posterior portion of the oper-
cle. Row ot continues obliquely to the margin of the opercle where it connects

Review of the Spinycheek Sleepers

25

Figure 3. Comparison of the urogenital papillae among females of eastern
Pacific Eleotris. a) Eleotris picta 63.7 mm SL (TNHC 14755) b) Eleotris picta,
63 mm SL (USNM 293478) c) Eleotric tecta, 55.1 mm SL (CAS 137539) d)
Eleotris tubularis, 58.7 mm SL (LACM 200047).

26

Tulane Studies in Zoology and Botany

[Vol. 31, No. 2 2002]

Figure 4. Comparison of urogenital papillae among males of eastern Pacific
Eleotris. a) Eleotris picta, 51.4 mm SL (ANSP 144132) b) Eleotris tecta, 42.0
mm SL (CAS 66639) c) Eleotris tubularis, 50.5 mm SL (LACM 25806).

Review of the Spinycheek Sleepers

27

PC1

Figure 5. Principal components analysis of eastern Pacific Eleotris populations
using meristic features (transverse scale rows, lateral scale rows, caudal peduncle
scale rows, pectoral fin rays and predorsal scales). Dots = Eleotris picta, triangles
= E. tecta, and squares = E. tubularis.

with lower longitudinal row oi (Fig. 2). Eleotris tubularis and E. tecta lack of
altogether. The relationship between rows os and oi differs between Eleotris
tubularis and E. tecta. The upper row os in E. tecta shifts obliquely downward to
meet the posterior end of lower longitudinal row oi on the lower right posterior
margin of the opercle. Intersection of upper row os and lower longitudinal row
oi was not observed in specimens of E. tubularis (Fig. 2).

Although all three species have a basic brown body color, pigmentation
varies among the species. Eleotris tecta has horizontal rows of spots along the
sides and a prominent spot on the upper pectoral fm base. Eleotris tubularis lacks
both well-defined rows of spots along the sides and the spot on the upper pectoral
fm base. Eleotris picta has a dark band or row of blotches along the upper flanks,
lacks the dark spot on the upper pectoral fm base and often has a mottled body.

Western Atlantic. The three western Atlantic species are not so easily delin-
eated, but a combination of meristic, neuromast patterns and subtle pigmentation
differences does allow their distinction. As with the eastern Pacific species, mor-
phometry was of little help in distinguishing taxa, even with principal compo-
nents analysis (Table 4).

28

Tulane Studies in Zoology and Botany

[Vol. 31, No. 2 2002]

4

2-

-4 -

-4 -2 0 2 4

PC1

Figure 6. Principal components analysis of western Atlantic Eleotris populations
using meristic features (transverse scale rows, lateral scale rows, caudal peduncle
scale rows, pectoral fin rays and predorsal scales) and cephalic suborbital neuro-
mast patterns. Diamonds = Eleotris pemiger, stars = E. pisonis, dots = E. ambly-
opsis from the Gulf of Mexico, triangles = E. amblyopsis from SE Florida, and
squares = E. amblyopsis from Central and South America.

Meristic and cephalic neuromast features clearly separate Eleotris ambly-
opsis and E. pemiger (Fig. 6). The meristic differences are best illustrated by a
comparison of lateral scale row counts (Table 5). Eleotris amblyopsis ranged
from 40-58 with a mean of 49. Eleotris pemiger had 54-68 with a mean of 60.
The distinction between these two species is pronounced however, where they are
sympatric in the Caribbean Basin; E. amblyopsis counts from that region ranged
from 40-52 (Fig. 7 and 8). Lateral scale row number ranged from 47-63 with a
mean of 54 for Eleotris pisonis. Although there was broad overlap with E. ambly-
opsis, E. pisonis counts were usually 50-54 in Guianas and eastern Venezuela
where the two species are sympatric (Figs. 7 and 8). Eleotris amblyopsis counts
ranged from 41- 46 in that region.

Cephalic free neuromast patterns distinguish Eleotris pemiger from its
two western Atlantic congeners (Fig. 9). Whereas E. pemiger typically has a
2.4.6 transverse suborbital row pattern, the other two species have a 2.3.4 pattern.

Review of the Spinycheek Sleepers

29

25

20

15

10

5

0

Antilles, Bahamas

Venezuela

40 44 48 52 56 60 64 68

Brazil

Figure 7. Frequency distributions of lateral scale counts (x axis) for western
Atlantic Eleotris from South America, the insular Caribbean Basin and the
Bahamas. Solid bars = Eleotris pemiger, diagonal bars = Eleotris pisonis and
wavy horizontal bars = Eleotris amblyopsis.

Eleotris pemiger may also develop additional short segments between the pri-
mary rows. Eleotris amblyopsis from North America, particularly specimens seen
from North Carolina, may have an additional row developed on one side of the
head or the other. Many specimens from the Gulf of Mexico and the Carolinas
also had occasional short segments developing between the primary rows.
Intraspecific variation is detailed in the species accounts below. None of the
western Atlantic species exhibit opercular row of and os and oi are generally
connected although this may be variable within species.

Pigmentation is remarkably similar among the three western Atlantic
species, and actually throughout the genus. Preserved specimens of all three
species are basically tawny in body color with a light colored abdomen. Eleotris
pemiger tends to have well-developed horizontal rows of spots on the flanks that
in some specimens appear as continuous lines. Of the other two species, E.

30 Tulane Studies in Zoology and Botany [Vol. 31, No. 2 2002]

Central America SE Florida

Gulf of Mexico Carolines

Figure 8. Frequency distributions of lateral scale counts (x axis) for western
Atlantic Eleotris from North and Central America. Solid bars = Eleotris pemiger
and wavy horizontal bars = Eleotris amblyopsis.

amblyopsis may have horizontal rows of spots, but less regularly arranged or con-
trasted as in E. pemiger. Most often, E. amblyopsis shows scattered spots along
the upper flanks which may not be arranged in rows. Eleotris pisonis specimens
were seen with the scattered spots as in E. amblyopsis, but sometimes spots on
the sides were entirely lacking. Eleotris pemiger and E. amblyopsis both have a
dark spot on the upper pectoral-fm base. This spot was generally not observed in
E. pisonis, but when it was expressed it was poorly contrasted and not as pro-
nounced as dark pigment above the pectoral-fm base on the side of the nape.
Individuals of any of these species or the eastern Pacific species may shift to a
two-tone appearance of the trunk in which a light dorsum is sharply and evenly
contrasted with dark flanks.

Review of the Spinycheek Sleepers

31

Fig. 9 Cephalic free neuromast patterns of western Atlantic Eleotris species, E.
pemiger (above) 57.5 mm SL, NLU 69725 and E. amblyopsis (below) 60.7 mm
SL, NLU 69723. The suborbital free neuromast configuration of Eleotris pisonis
(not illustrated) is the same as that shown for E. amblyopsis.

32

Tulane Studies in Zoology and Botany

[Vol. 31, No. 2 2002]

Discussion

Our findings support the recognition of six Eleotris species in the
Western Hemisphere. Bussing (1996) and Miller (1998) have previously noted
the conservative morphology of the genus. The only significant morphometric
variation we observed was that of the form of the urogenital papilla, which is
useful in determining the sex of individuals. Eye size was a major distinguishing
feature in Jordan and Gilbert’s (1896) original description of E. abacurus and a
relatively larger eye size has been used since then to separate E. amblyopsis from
E. pemiger (referred to in most works as E. pisonis) in keys to eleotrids of the
western Atlantic (e.g. Villa, 1982). We found no concordance between eye size
variation and the other features determined diagnostic for the western Atlantic
species.

Despite the importance of meristic characters in delimiting the species,
some meristic features regarded as diagnostic in the literature were uninforma-
tive. Jordan and Evermann (1898) used the extent of cheek squamation to distin-
guish E. amblyopsis (including E. abacurus, a synonym herein) and E. pisonis
from E. pemiger. Eleotris amblyopsis and E. pisonis were described as having
fully scaled cheeks, while those of E. pemiger were only Vi scaled. We found
cheek squamation highly variable within species. The extent of squamation on
the snout and interorbital region, characters examined by Akihito (1967), were
also of no utility.

Comparisons with West African species. Eleotris daganensis Steindachner,
1869 of the eastern Atlantic is indistinguishable from E. amblyopsis from the
Caribbean Basin. The free neuromast patterns, number of lateral scale rows and
body and fm pigmentation are similar in type and variation. The number of scales
in a lateral series ranged from 44-53 with a mean of 48 in 48 specimens of E.
daganensis examined. Specimens of that species typically had a 2.3.4 suborbital
neuromast pattern, but of 38 examined one had a 2.3.5 pattern and another
2.3.4.5. Undoubtedly, E. amblyopsis and E. daganensis are closely related and
possibly sister species. Their relationship to E. annobonensis, E. pemiger and E.
pisonis is close and undecipherable from comparative morphology.

Eleotris annobonensis from the eastern Atlantic islands of Equatorial
Guinea and Pagalu (Annobon) is similar to E. pemiger. Pigmentation is compara-
ble between these two species and lateral scale row numbers are the same.
Suborbital free neuromast patterns differ slightly in the extent of supernumerary
row development. Whereas 2.4.6 is the predominant pattern in E. pemiger, 2.3.4
with intervening segments was the most common pattern observed in the 22
specimens of E. annobonensis we examined, including types. Only 3 specimens
had a 2.4.6 pattern; one specimen had a 2.4 pattern and another a 2.5.8 pattern.

Review of the Spinycheek Sleepers

33

Of all the Eleotris species in the eastern Pacific and Atlantic basins,
Eleotris picta is most similar to E. senegalensis, E. feai and E. vittata from West
Africa. All four species have the forked neuromast pattern on the upper opercu-
lum. All are large species for the genus, with E. picta possibly reaching the
largest size. Eleotris picta is similar in lateral scale row number to E. vittata and
E. feai. Twenty specimens of E. vittata had a range of 57-65 and a mean of 61.
Two paratypes of E. feai had counts of 62 scales. Although E. senegalensis has
larger scales (and therefore a lower count), most specimens examined exhibited a
2.4 suborbital neuromast row pattern as in E. picta. There was variation observed
for this feature in both E. vittata and E. senegalensis, however. The 19 specimens
of E. vittata examined for the character showed five patterns - 2 (15 specimens),

3 (1), 2.3 (1), 2/2.5 (1) and 2.4/2.5 (1). The E.feai paratypes had 2 and 2/2.5 pat-
terns. Twenty-one specimens of E. senegalensis showed 2 character states - 2 (3)
and 2.4 (18). A low degree of variation for this feature was also observed in E.
picta', out of 255 specimens examined, four had 2.3.4 and one a 2.3.4.5 pattern.

Of the four with the 2.3.4 pattern, only one had the aberrant pattern on both
cheeks. Clearly, the phylogenetic lines of Eleotris hypothesized by Miller (1998)
based upon differences in suborbital neuromast patterns may be confused when
variation is considered. The potential for homoplasy is great.

Eleotris amblyopsis is not synonymized with E. daganensis and E.
pemiger is not synonymized with E. annobonensis. The allopatric distributions of
these taxa make it difficult to evaluate the relationships of the amphi-Atlantic
species pairs considering the scant, subtle and overlapping morphological varia-
tion of diagnostic characters observed among the western Atlantic taxa. For
example, although North American populations of E. amblyopsis approach a
morphology that could be confused with that observed in E. pemiger, sympatric
populations of the two species in the Caribbean Basin are clearly distinguished.
Additionally, the distinctive compositions of the west African and western
Atlantic gobioid faunas suggest that gene flow between gobioid fishes of these
two long-isolated regions is unlikely, even considering the pelagic larval stage of
these species. It is most parsimonious and consistent with the Evolutionary
Species Concept not to synonymize the species pairs. There is considerable cir-
cumstantial evidence to suspect that these taxa have had independent evolution-
ary histories for millions of years and continue on distinct evolutionary trajecto-
ries. We also believe that application of the Morphological Species Concept to
this genus of cryptic species is inappropriate because of the concept’s insensitivi-
ty to the diversity the genus comprises.

The gobioid fauna of tropical West Africa comprises about 28 genera,
including 8 that are endemic to that region and two others endemic to Africa
(Birdsong et al., 1988). Three genera are shared with the Indo-Pacific (discount-
ing the introduced Prionobutis) and four genera are shared only with the north-
eastern Atlantic/Mediterranean region (Table 6). Five genera are shared only with

34

Tulane Studies in Zoology and Botany

[Vol. 31, No. 2 2002]

the Western Hemisphere. Five other genera are more cosmopolitan, occurring in
the Western Hemisphere, Indo-Pacific and eastern Atlantic. Thus, of the 28 gob-
ioid genera in West Africa, 10 are shared with the Western Hemisphere. One
goby species possibly common to the eastern and western Atlantic, Bathygobius
soporator (Miller and Smith, 1989), doesn’t increase the likelihood of ongoing
gene flow among other gobioids. It merely suggests a lack of differentiation
despite a long period of isolation and offers no evidence of even an intermittent
mechanism for larval transport east to west across the Atlantic. It should also be
noted that there are five species of Eleotris in the eastern Atlantic, three of which
{E. feai, E. senegalensis and E. vittata) have no counterpart in the western
Atlantic. It seems unlikely that there should be any link between populations of
E. daganensis or E. annobonensis and their New World counterparts, and not the
other three species if east to west surface currents are the potential vehicle.

Miller (1998) suggests that a subsurface equatorial countercurrent could
allow dispersal from west to east across the Atlantic. Studies of larval gobiids
have shown a typical duration of 20 - 40 days (Breitburg, 1989; Brothers et al.,
1983; McFarland et al., 1985). This is a much shorter period than the 96 days
estimated as necessary for cross Atlantic subsurface transport at the equator
(Scheltema and Hall, 1975). The western Atlantic has more than 120 gobioid
species (Birdsong et al., 1988) and an entire tribe, the Gobiosomini, that is shared
only with the more recently conjoined eastern Pacific. This degree of taxonomic
distinction is consistent with the limitations on larval transport that may be
inferred from the few studies mentioned above. However, a much longer larval
period was observed by Radtke et al. (1988) for the Hawaiian freshwater gobiids
Stenogobius genivittatus and Awaous stamineus that would span the 96 days esti-
mated above. Studies of shore fishes of the St. Paul’s Rocks (Lubbock and
Edwards, 1981), St. Helena (Edwards and Glass, 1987) and Ascension Islands
(Lubbock, 1980) are informative. No gobioid fishes were taken at St. Pauls’
Rocks. Two gobies are known from Ascension and St. Helena, the western
Atlantic Gnatholepis thompsoni and the endemic Priolepis ascensionis. One
specimen of an Eleotris species was captured at St. Helena and referred to E.
pisonis. This species’ identity has not been checked by us. These mid-Atlantic
tropical islands are more interesting for the gobioid species they lack, than for
those that are present. The lack of more of the rocky reef-associated goby fauna
suggests a passage too long or stressful for most gobioid fishes. Where is
Bathygobius soporatorl As for the single sleeper at St. Helena, was it a ballast
introduction as we presume for the Eleotris picta specimen we have observed
from Venezuela?

There is limited information available about the status of tropical Atlantic
species within the ten gobioid genera common to both sides of the ocean, but it
generally suggests isolation of the shore fauna. The only other shared eleotrid
genus is Dormitator; the common west African species, D. lebretonis, is distinct

Review of the Spinycheek Sleepers

35

from the common western Atlantic species, D. maculatus (unpubl. data). The
species limits and relationships within Sicydium, the only sicydiine genus com-
mon to streams in both the Caribbean Basin and Gulf of Guinea, are unclear, but
Harrison (1993) observed that S. bustamantei from West Africa and specimens
identified as S. plumieri from the Antilles were very similar and suggested that
they could be amphi- Atlantic conspecifics as has been suggested for Bathygobius
soporator. Watson (2000) redescribed S. plumieri, but did not include S. busta-
mantei in the synonymy. Although he indicated some familiarity with West
African species, he did not list any comparative material from West Africa, so the
question of species distinction remains unanswered. Among the shared gob-
ionelline genera, Awaous, Ctenogobius, Gnatholepis, Gobioides and Gobionellus,
information on species status is available for three genera. In a recent review of
the genus Gobioides, Murdy (1998) recognized two African species distinguished
from western Atlantic species. Gobionellus and Ctenogobius are each represented
by a single species in West Africa not found in the western Atlantic (Pezold,
1984). Reports of Gobionellus oceanicus from (e.g. Miller and Harrison, 1992)
the Gulf of Guinea are presumed to be based on misidentifications of G. occiden-
talis as no specimens of this species were found from West Africa during an
exhaustive review of the genus (Pezold, 1984). The gobiine genus Priolepis has
not been reviewed for its Atlantic species.

The phylogenetic relationships of species within the 10 shared genera are
for the most part unknown. Murdy (1998) offers evidence that Gobioides
africanus is the sister group to the other species of the genus. Preliminary data
also indicate that Dormitator lebretonis is the primitive sister group to the
Western Hemisphere species (unpubl. data). These observations further support a
long break between gobioid populations on the two sides of the Atlantic.

Instead of any active genetic exchange across the Atlantic we suggest
that Eleotris is simply morphologically conservative. Similar morphotypes to
those seen in the Western Hemisphere and the eastern Atlantic pop up in other
tropical estuaries and insular streams, albeit ever so slightly different. Present cir-
cumglobal distributions suggest that several basic lines were separated long ago.
This perspective is reinforced by the phylogenetic study of gobioid fishes using
mtDNA by Akihito et al. (2000). The several species of Eleotris included in their
analysis, E. acanthopoma, E. fusca, E. melanosoma and E. oxycephala were sep-
arated by greater genetic distances than were specimens in the morphologically
distinct pairs of genera Dormitator and Hypseleotris, and Calumia and
Gobiomorphus. The four species of Eleotris are distinguishable by cephalic free
neuromast patterns and various meristic combinations similar to those seen in the
Western Hemisphere.

Phylogenetic Relationships. Miller (1998) subdivided Eleotris into clusters of
nominal species based upon cephalic neuromast patterns. Although there is

36

Tulane Studies in Zoology and Botany

[Vol. 31, No. 2 2002]

heuristic value in his approach, cephalic free neuromast patterns must be used
with caution in any phylogenetic reconstruction. The suborbital row patterns vary
additively and ontogenetically, and that variation is significant, as can be seen in
the species descriptions that follow. Variation among species also exists for the
presence or absence of the ot' row that Miller regarded as a synapomorphy for
Leptophilypnus and Eleotris (including Erotelis). In addition to the variation
among Eleotris species as noted by Miller, it appears in only one of the two
Leptophilypnus species; Leptophilypnus fluviatilis has it, but L. panamensis does
not. The appearance of both states of the single proposed synapomorphy in
Leptophilypnus increases the homoplasy inherent in any phylogenetic recon-
struction and weakens the argument for recognizing Leptophilypnus as the sister
group to an Eleotris clade on the basis of that character alone. Before an unequiv-
ocal phylogeny of intrageneric relationships can be developed, more polarized
characters are needed, and this demands a better understanding of eleotrid rela-
tionships. Another major problem confronting any attempt at phylogenetic recon-
structions of intrageneric Eleotris relationships at this time is that morphological
variation within and among Indo-Pacific species requires attention. The seminal
work by Akihito (1967) was only a beginning.

Systematics

Eleotris Bloch and Schneider, 1801

Eleotris, Bloch and Schneider (1801): 65. Type species: Gobius pisonis Gmelin
1789, South America (Brazil). Type established by ICZN by use of ple-
nary powers (Opinion 93, Direction 56) and Eleotris Gronow 1763 listed
as a name published in a rejected work (Opinion 417).

Eleotrides, Bory de Saint- Vincent, 1825: 417. Type species: Gobius pisonis
Gmelin 1789.

Culius, Bleeker, 1856: 385, 411. Type species: Cheilodipterus culius Hamilton
1822, Bengal, India, by absolute tautonymy, not Culius fuscus Bleeker
1856 (= Poecilia fusca Bloch and Schneider, 1801 = Eleotris nigra Quoy
and Gaimard, 1824, 259, pi 60 fig 2, Waigeo, Indonesia) as subsequently
designated in Bleeker 1874: 303.

Kieneria, Mauge, 1984: 98 (subgenus of Eleotris). Type species: Kieneria

vomerodentata Mauge, 1984, Madagascar, by original designation and
monotypy.

Diagnosis. Eleotrid fishes with transverse suborbital free neuromast rows, no
cephalic lateralis canals, a spine on the angle of the preopercle, and equal num-
bers of fm elements in the second dorsal and anal fins (1,8).

Review of the Spinycheek Sleepers

37

Description. Head broad and flattened, body low, and torpedo-like. Mouth large
and oblique, posterior margin of upper jaw to vertical through middle of eye or
rear margin of orbit, lower jaw projecting. Upper and lower jaws with multiple
rows of small teeth, a few caniniform teeth in some species. Tubular anterior
nares overhanging upper lip, posterior nares open pits. Stout antrorse spine at
angle of preopercle. Small eyes high on head. Interorbital broad, frequently three
times eye width. Gill opening moderately broad, extending to below preopercu-
lum. Almost completely scaled, cycloid scales on nape, cheek, opercle, interor-
bital, pectoral-fm base, pre-pelvic region, abdomen, and in one or two rows bor-
dering median fins, ctenoid scales covering sides of trunk. Urogenital papilla in
females rounded, bulb-shaped with orifice equal to anus in size, elongate and
tapered in males. Pectoral fins longer than pelvic fins, variably reaching to verti-
cal through urogenital base to vertical through anal-fin origin. Pelvic fins sepa-
rate, I, 5, generally not reaching anus (variable within E. amblyopsis, may reach
anus in some populations). Two separate dorsal fins, first dorsal fm with six flexi-
ble spines, second dorsal fm with single leading flexible spine followed by eight
soft rays. Anal fm with single flexible spine followed by eight soft rays.
Appressed median fins not reaching caudal-fm base. Caudal fm rounded, pre-
ceeded by 8-10 procurrent fm rays that extend forward no further than above the
penultimate caudal vertebra. Caudal fm not extended forward along the caudal
peduncle. No lateral line canals. Rows of transverse and longitudinal free neuro-
masts (sensory papillae) on head; disconnected short vertical rows of free neuro-
masts along sides of trunk; three diverging rows of free neuromasts on caudal fm,
central row horizontal, upper and lower rows angled posterodorsally and pos-
teroventrally respectively. Dorsal-fm pterygiophore formula 3-1221 or 3-12210,
10-1- 15 = 25 vertebrae, two epurals, 2 or 3 preanal hemal pterygiophores
(Birdsong et al., 1988).

Remarks. Eleotris was placed in the Official List of Generic Names and Gobius
pisonis Gmelin,1789 was established as the type species by the International
Commission on Zoological Nomenclature by use of plenary powers for the sus-
pension of rules in Opinion 93, Direction 56, 1926. The summary lists E. gyrinus
Cuvier and Valenciennes as type species in error. Opinion 93 also notes that
Gobiomoroides, LacepMe, 1800: 592, is not a synonym of Eleotris as the dried
type specimen has a “single dorsal of 45 rays and canine teeth.” Epiphthalmus,
Rafmesque- Schmaltz, 1815: 86, an unnecessary substitute name for
Gobiomoroides, is also removed from synonymy.

Miller (1998) synonymized Erotelis with Eleotris because they share an
antrorse spine on the preoperculum, have the same number of precaudal and cau-
dal vertebrae, agree in first dorsal-fm pterygiophore pattern and have similar
cephalic sensory systems. He stated that they only differ in squamation. Erotelis
has small cycloid scales covering the body, while Eleotris has larger scales and
those on the lateral trunk are ctenoid. In actuality, Erotelis also differs in having a

38

Tulane Studies in Zoology and Botany

[Vol. 31, No. 2 2002]

derived condition of the caudal fin. The caudal fin is tapered and has 12-14
unsegmented procurrent rays before the hypural both dorsally and ventrally that
extend forward above and below the third caudal vertebra from the terminus. The
procurrent rays are elongate and support well-developed fm margins along the
rear portion of the caudal peduncle. In Eleotris, there are about eight to ten rays
in advance of the hypural extending no farther forward than above and below the
penultimate caudal vertebra and the caudal fm is rounded. The procurrent rays of
Eleotris also quickly taper to rudiments and are not associated with extensive fin
membranes as seen in Erotelis. Erotelis also has a much more oblique jaw, a
more elongate body, more segmented rays in the second dorsal and anal fins, one
more ray in the second dorsal fm than the anal fm (as opposed to equal numbers)
and a highlighted myomeric pattern of pigmentation on the sides (Dawson, 1969;
Ginsburg, 1953; Jordan and Evermann, 1898). We assume that Miller’s purpose
in synonymizing the two genera was to establish a monophyletic Eleotris based
upon the shared possession of a preopercular spine. As none of the other charac-
ters of the axial skeleton and cephalic lateralis are unique to Eleotris, the genus
remains undefined by a synapomorphy. The number of elements in the second
dorsal fm and anal fm, as a combination, may be synapomorphic, but this charac-
ter cannot be polarized until an outgroup is unequivocally identified. Accepting
the preopercular spine as a synapomorphy of the two genera, the three Erotelis
taxa (two species and a subspecies) are characterized by several derived features:
many small cycloid scales covering the body, the lanceolate caudal fm, the for-
ward extensions of the caudal fm and their numerous associated procurrent rays
and probably the myomeric pigmentation on the sides. Polarity of the other char-
acters described for Erotelis is less certain. Considering the conservative nature
of morphological variation within Eleotris and other genera of eleotrids, the dis-
tinction of Erotelis from Eleotris is outstanding. Synonymization will only
obscure the distinctions. If we were to apply the same rationale to gobioid fishes
as a whole, that is ignoring identified monophyletic groups if it results in a sister
group diagnosed by plesiomorphic traits, we would only have two families,
Rhyacichthyidae and Gobiidae with all but a couple of the roughly 2000 gobioids
in the latter family (Miller, 1973). Obviously much information is less readily
available in a classification developed with that approach and many of our
advances in understanding of gobioid relationships would be less apparent (e.g.
Hoese and Gill, 1993; Pezold, 1993). We recognize the two genera as distinct and
valid.

Key to Eleotris species of the Western Hemisphere

1. a. Teeth heteromorphic, several rows of fine teeth in both jaws with
larger canine teeth present laterally towards rear in outer row

Review of the Spinycheek Sleepers

39

and/or medially near symphysis in inner rows; opercular free

neuromast row of absent 2

b. All teeth fine, no canine teeth present; opercular vertical row of

present E. picta

(eastern Pacific; Mexico to Peru and the Galapagos)

2. a. Dark spot present on upper pectoral-fm base (may be covered by

opercular membrane), darker than markings on nape at opercular

margin 3

b. No dark spot present on upper pectoral-fm base or if present not
strongly-contrasted and not as dark as pigment on nape 4

3. a. Second and some combination of two of third, fourth or fifth

transverse suborbital neuromast rows extend below longitudinal

row d (usually third and fourth) 5

b. Second, fourth and sixth transverse suborbital neuromast rows

extend below longitudinal row d E. pemiger

(western Atlantic; Caribbean Basin, Mexico, Bermuda)

4. a. Only second and fourth transverse suborbital neuromast rows on

cheek extend below longitudinal row d (2.4 pattern)... E". tubularis
(eastern Pacific; Cocos Island)

b. Second, third and fourth transverse suborbital neuromast rows on
cheek extend below longitudinal row d (2.3.4 pattern)... pisonis
(western Atlantic; Brazil to Orinoco delta)

5. a. Scales in lateral series 40-58 over range, 43-52 in Caribbean

Basin E. amblyopsis

(western Atlantic; primarily continental, Guianas through North
Carolina, Cuba, occasional other Antilles)

b. Scales in lateral series 54-60 E. tecta

(eastern Pacific; Costa Rica through Colombia)

Eleotris amblyopsis (Cope, 1871)

Eleotris gyrinus, Valenciennes, in Cuvier and Valenciennes, 1837: 220, pi. 356,
Martinique and Saint-Domingue, (in part).

Culius amblyopsis Cope, 1871: 473, Surinam.

Eleotris (Culius) belizianus Sauvage, 1880: 56, Belize, (in part).

Eleotris abacurus Jordan and Gilbert, 1896: 228, Charleston, South Carolina.
Eleotris isthmensis Meek and Hildebrand, 1916: 356, 359, Mindi, Panama Canal
Zone.

40

Tulane Studies in Zoology and Botany

[VoL 31, No. 2 2002]

Diagnosis. Scales in lateral series 40-58, generally 41- 46 in NE South
America, 45 - 49 in Caribbean Basin, 50 - 56 in North America. Second, third
and fourth suborbital free neuromast rows on cheek extending ventrally past hori-
zontal row d (2.3.4 pattern); lacking of neuromast row on upper opercle. Dark
blotch present on upper pectoral-fm base; sides of trunk usually with scattered
spots along dorsum and no regular rows of spots of uniform intensity forming
horizontal pinstripes.

Description. Body morphology as described for genus. Both jaws with multiple
rows of fine teeth, larger canine teeth present laterally towards rear in outer row
and/or medially near symphysis in inner rows. Proportional measurements are
given in Table 4.

Meristics. First dorsal fin VI; second dorsal fm 1,8 (I,7-I,9); pectoral fins 16 (15-
18); pelvic fins I, 5; anal fm 1,8 (I,7-l,8). Scales in lateral series 49 (40-58), gen-
erally 43-46 in NE South America, 45-49 in Caribbean Basin, 50-56 in North
America; predorsal scales 36 (29 - 46); transverse scale rows 14 (11 - 18); caudal
peduncle scale rows 15 (11 - 24).

Cephalic lateralis. Fig. 9. Five transverse suborbital free neuromast rows of
which 2nd, 3rd and 4th extend ventrally beyond horizontal row d (2.3.4 pattern).
Short supernumerary segments may occur between two or more transverse rows
in specimens from North America, occasionally with an extra row on one or both
sides such that 2nd, 3rd and 5th (2.3.5) or 2nd 4th and 5th rows (2.4.5) extend
below row d, extra rows especially common in specimens from Atlantic coast of
US. No ot' row on upper opercle, row os connects with row oi at ventroposterior
margin of opercle.

Coloration in alcohol. Based on specimens from Costa Rica (NLU 69723),
Florida (UF 87886), and Mexico (UMMZ 209724). Body dark brown laterally,
lighter along the dorsum or tan with rows of spots on sides of varying intensity;
abdomen and gular region lighter. Cheek with two dark streaks radiating posteri-
orly from eye, one along sulcus between nape and cheek and operculum. Large
spot on upper pectoral-fm base which may be covered by the opercular mem-
brane; two elongate spots extending from pectoral-fm base onto pectoral-fm rays
seen in some specimens. First dorsal fm with dark band along base and another
reticulate band through middle of fin, dark spots generally present on spines
above mid-level band, but not at tips, bands generally converging posteriorly;
second dorsal fm with 5 to 8 wavy diagonal bands, sometimes over dusky mem-
brane; anal fm dusky or with as many as seven wavy rows of spots; caudal fin
with 10 to 14 vertical bars formed by small spots on rays against a dusky mem-

Review of the Spinycheek Sleepers

41

brane; pectoral fins peppered with small spots on rays, membrane may be clear or
dusky, pelvic fins similar but with fewer rows of spots.

Distribution. Common in continental estuaries from French Guiana to North
Carolina, but also known from Brazil and the Antilles. Enters freshwater streams
in the Caribbean Basin.

Comments. This species reaches a smaller maximum size than E. pemiger and
matures at a size of 50-60 mm (Hildebrand, 1938). The largest specimen collect-
ed by Greenfield and Thomerson (1997) in Belize was just over 73 mm SL and
Bussing (1996) reported it to reach 80 mm SL in Costa Rica. The largest speci-
mens observed in our study were 83 mm SL from the Caribbean Basin and 113
mm SL from North America.

Dawson (1969) noted the ability of this species to change color pattern
from uniformly dark to dark laterally with a light dorsum, a trait that seems to be
common to a number of Eleotris species. This is most likely the species from
Mexico karyotyped by Uribe et al. (1994), as E. pemiger, though known from
that region, is not the common species. Karyotypes revealed heteromorphic sex
chromosomes; 44 acrocentric and two metacentric chromosomes.

This species is known from both estuaries and streams in Central
America (Gilbert and Kelso, 1971; Greenfield and Thomerson, 1997; Nordlie,
1979) where it often co-occurs with E. pemiger and estuaries in North America
(Dawson, 1969; Schwartz, 1999). In Belize, it was the primary species on the
mainland (only one specimen of E. pemiger was taken), but both species were
regularly found together in Saint George’s Cay on the barrier reef (Greenfield and
Thomerson, 1997). Greenfield collected both species in Honduras, where E.
amblyopsis was most common in estuaries and E. pemiger (reported as E.
pisonis) was found more often in freshwater. Gilbert and Kelso (1971) and
Bussing (1998) found E. amblyopsis the more common species in Costa Rica.
Although Gilbert and Kelso (1971) took both species together from estuaries and
streams, Nordlie (1979) reported E. pemiger (as E. pisonis) more common in
freshwater than E. amblyopsis, but E. amblyopsis the more abundant by far in the
Tortuguero estuary. Bussing (1998) also noted it in estuaries and streams to 15m
elevation. In Panama E. amblyopsis was reported by Hildebrand (1938) (as E.
isthmensis) to be less common than E. pemiger (identified as E. pisonis).
Although usually noted from low salinity estuaries in North America (e.g.
Dawson, 1969), Gilmore and Herrema (1981) found them most common in fresh-
water in east-central Florida. Microhabitat preferences that have been noted
included mangroves in Mexico (Britton and Morton, 1989), mangrove channels
in Belize (Greenfield and Thomerson, 1997) and hyacinth roots in Costa Rica
(Gilbert and Kelso, 1971). In Louisiana, the first author has collected this species
associated with hyacinth roots and Phragmites marshes in oligohaline estuaries at

42

Tulane Studies in Zoology and Botany

[VoL 31, No. 2 2002]

the mouth of the Mississippi River. Others have been captured in a channel in a
Spartina marsh using traps composed of oyster shells in plastic crates. Eleotris
amblyopsis is a carnivorous species that feeds on arthropods and fishes (Bussing,
1998; Nordlie, 1981). Nordlie observed no change in diet associated with posi-
tion in the estuary or size.

Eleotris perniger (Cope, 1871)

Eleotris gyrinus, Valenciennes, in Cuvier and Valenciennes, 1837: 220, pi. 356,
Martinique and Saint-Domingue, (in part).

Culius perniger Cope, 1871: 473, St. Martin, West Indies.

Eleotris maltzani Hilgendorf, 1889: 53, Haiti.

Eleotris hilgendorfi Pietschman, 1913: 182. replacement name for E. maltzani
Hilgendorf (not Steindachner)

Diagnosis. Scales in lateral series 54-68, usually about 60. Second, fourth and
sixth suborbital free neuromast rows on cheek extending ventrally past horizontal
row d (2.4.6 pattern), with numerous short segments above row d between trans-
verse rows, sometimes an additional row formed resulting in 2.4.7 pattern or
other variants; lacking of neuromast row on upper opercle. Dark blotch present
on upper pectoral-fm base, sides of trunk usually with regular rows of spots or
stripes.

Description. Body morphology as described for genus. Both jaws with multiple
rows of fine teeth, larger canine teeth present laterally towards rear in outer row
and/or medially near symphysis in inner rows. Proportional measurements are
given in Table 4.

Meristics. First dorsal fin VI; second dorsal fin 1,8 (I,7-I,8); pectoral fins 18 (lb-
19); pelvic fins I, 5; anal fm I, 8. Scales in lateral series 60 (54-68); predorsal
scales 49 (39 - 62); transverse scale rows 20 (17-23); caudal peduncle scale rows
17(13-21).

Cephalic lateralis. Fig. 9. Adults with six transverse suborbital free neuromast
rows of which 2nd, 4th, and 6th extend ventrally beyond horizontal row d and
many short supernumerary segments often present between transverse rows (2.4.6
pattern). Often one or two complete extra rows formed on one or both sides such
that 2nd, 4th and 7th rows extend below row d (2.4.7), occasionally other pat-
terns produced including 2.5.7, 2.5.8, 2.4.5, 2.4.8, 2.5.9, 2.6.8, 3.5.7 and 3.5.10.
No of row on upper opercle, row os connects with row oi at ventroposterior mar-
gin of opercle. Early juveniles with five transverse suborbital rows forming 2.3.4
pattern and no intervening segments.

Review of the Spinycheek Sleepers

43

Coloration in alcohol. Based on specimens from Costa Rica (NLU 60725),
Honduras (UF 16325, UF 16333), Mexico (UMMZ 196529), Panama (ANSP
104096, ANSP 104158, ANSP 104400), Puerto Rico (ANSP 115625, ANSP
144626). Body dark brown laterally, lighter along the dorsum or tan with rows of
dark brown spots on sides sometimes forming thin horizontal pinstripes;
abdomen and gular region lighter. Cheek with two dark streaks radiating posteri-
orly from eye, one along sulcus between nape and cheek and operculum. Large
spot on upper pectoral-fm base which may be covered by the opercular mem-
brane. Other pigment as observed in E. amblyopsis.

Distribution. Common in the Caribbean Basin, predominant species in the
Antilles, also known from Bermuda and Rio de Janeiro.

Comments. Recent references to Eleotris pisonis from the Caribbean Basin (e.g.
Bussing, 1998; Greenfield and Thomerson, 1997) in which large-scaled species
{E. amblyopsis) and small-scaled species (E. pisonis) were distinguished refer to
this species. This is the largest species native to the western Atlantic. Bussing
reported that E. perniger reaches 120 mm SL. The largest specimen observed in
our study was 177 mm SL. Doug Smith (pers. comm.). University of
Massachusetts at Amherst, photographed a specimen captured on St. John Island,
VI, that was approximately 250 mm TL. Considering that the caudal fm in this
species averages 26% SL, the specimen was about 200 mm SL.

Greenfield and Thomerson (1997) reported a single juvenile specimen
from freshwater in continental Belize where E. amblyopsis is the more abundant
species. Hildebrand (1938) reported E. perniger as the common species in
Panama. It is the only species known from Bermuda where Smith- Vaniz et al.
(1999) reported two specimens from oligohaline waters: Walsingham Cave Pool
and Fern Sink Cave Pool. Bohlke and Chaplin (1993) noted two juveniles from a
shallow tidal creek on the south shore of New Providence Island. The specimens
were collected over a substrate of sand and rocks amid mangrove roots. They
stated that it was also known from freshwater Lake Killamey on that island.
Bussing (1998) regarded E. perniger as uncommon in Costa Rica. He also said
that it occurred in stagnant waters, or low velocity rivers and creeks. He found
the species in water of 25-28 C and although most abundant near the coast, it
occurs at least 60 m above sea level (which puts it farther inland than Eleotris
amblyopsis). Like E. amblyopsis, it feeds primarily on fishes and shrimps.

Nordlie (1979) found them mostly in freshwaters of the Rio Tortuguero system.

Eleotris picta Kner, 1863
Eleotris picta Kner, 1863:223. Rio Bayano, Panama.

Eleotris pictus Kner and Steindachner, 1864:18, Plate 3, fig. 1,

44

Tulane Studies in Zoology and Botany

[Vol. 31, No. 2 2002]

Rio Bayano, Panama, (spelled Eleotris picta on plate).

Culius aequidens Jordan and Gilbert, 1882:461, Rio Presidio near Mazatlan,
Mexico.

Diagnosis. Scales in lateral series 56-68, mode of 61. Neuromast configuration
with second and fourth suborbital vertical rows extending ventrally past horizon-
tal row d (2.4 pattern); opercular vertical rows os and of intersect on posterior
opercle forming an acute fork opening towards posterior margin. Urogenital
papillae small and square in shape in young females; distal margin fringed and
flared in adult females; pre-vent furrow not well developed.

Descriition. Body morphology is as described for the genus, although the head
is flatter and narrower compared to Eleotris tubularis. Broad bands of fine teeth
in both jaws, no canine teeth. Urogenital papillae small and square in shape in
young females; distal margin fringed and flared in adult females; pre-vent furrow
not well developed. Proportional measurements are shown in Table 1.

Meristics. First dorsal fm VI; second dorsal fm 1,8; pectoral fm 18 (11-19);
pelvic fins I, 5; anal fins 1,8 (I,8-I,9). Scales in lateral series 61 (56-68); predor-
sal scales 61 (53-65); transverse scale rows 24 (20-31); caudal peduncle scale
rows 18 (13-23).

Cephalic lateralis. Fig. 2. Five suborbital transverse rows of which the 2nd and
the 4th extend below horizontal row d (2.4 pattern). Supernumerary segments
occur between the 5 transverse rows. Opercular vertical row of is present, con-
necting with horizontal row os to form an acute fork opening toward the posterior
margin of the opercle.

Coloration in Alcohol Based on specimens from Honduras (TU 186087) and
Costa Rica (TU 84544, TU 24181, TU 84587, TU 84687 and TU uncat.). Color
dark tan or brown with darkest pigment laterally on trunk. Abdomen pale. Entire
body often mottled. Two dark postocular stripes present, more distinct in young;
uppermost postocular stripe extends horizontally from orbit to posterior opercular
margin; lower postocular stripe extends to upper preopercular margin.
Pigmentation on lateral scales may form dark band, especially in juveniles, con-
tinuous or broken into series of large blotches, band being formed from horizon-
tal rows of spots; often with dark speckles on top of head highlighting longitudi-
nal rows of free neuromasts. All fins possess wavy, dusky bands of spots. Dorsal
and ventral spots sometimes present above and below hypural.

Distribution. Common from Mexico to continental Ecuador and the Galapagos

Review of the Spinycheek Sleepers

45

Islands, also known from Cocos Island and Salton Valley of California.

Comments. This is the largest species of Eleotris in the Western Hemisphere.
Bussing (1998) stated that it reaches 320 mm SL. The largest specimen we exam-
ined was 276 mm SL. Uribe and Diaz (1996) described the karyotype for this
species as 52 acrocentric chromosomes.

Bussing (1998) described life colors of Eleotris picta as grayish black
dorsally, with pale gray sides occasionally marked with irregular yellow blotches,
and a tan abdomen with yellow or white blotches on young and subadults. He
also stated that juveniles are often pale yellowish brown above with spotted sides
and have a black tail with transparent margin. The fins in this species were noted
as dark with transparent spots producing bands, although he notes that this pat-
tern is less apparent in larger specimens. Grove and Lavenberg (1997) described
the species as being dark greenish brown with a light abdomen with dusky fins
“with dark speckles, [and] undulating stripes on the dorsal and anal fins”.
Juveniles were described as “mottled with blue and dark brown speckles, without
stripes or bars.”

Grove and Lavenberg (1997) reported one specimen 94 mm SL from a
freshwater stream near Playa Negra, on Floreana in the Galapagos Islands.
Bussing (1998) found them in waters ranging from stagnant to high velocity
rivers. He reported them most abundant near the coast, but inhabiting streams up
to 100 m above sea level, with larger individuals upstream. Temperatures at cap-
ture were from 25-33 C.

According to Bussing (1998), this is a lie-and-wait benthic predator, like
others in the genus, that feeds on fishes and shrimps. It lurks under stones and
overhanging shorelines. Winemiller and Morales (1989) found the diet comprised
about 32% aquatic insect larvae, 39% shrimps and 9% fishes in Corcovado
National Park, Costa Rica. The condition of seven specimens examined by
Hildebrand (1938) suggested that spawning occurs during the dry season.

A single specimen of Eleotris picta was discovered during this study
from San Juan de los Cayos, Falcon State, Venezuela. This is the first report of
this species in the western Atlantic. It most likely resulted from introduction via
ship’s ballast water.

Eleotris pisonis (Gmelin, 1789)

Gobius pisonis, Gmelin, 1789:1206, South America (based on Eleotris capite

plagioplateo of Gronow; after Marcgrave and Piso, Hist. Brasil, IV. 166,
1648; Brazil).

Gobius amorea Walbaum, 1792: 205, western Atlantic.

Eleotris gyrinus, Valenciennes, in Cuvier and Valenciennes, 1837: 220, pi. 356,
Martinique and Saint-Domingue, (in part).

Eleotris (Culius) belizianus Sauvage, 1880: 56, Belize, (in part).

Eleotris carvalhonis Starks, 1913: 65, pi. 9, Brazil.

46

Tulane Studies in Zoology and Botany

[Vol. 31, No. 2 2002]

Diagnosis. 47-63 lateral scale rows, usually 50-54 in NE South America and 54-
59 in Brazil. 2.3.4 suborbital free neuromast pattern. Spot on upper pectoral fm
base lacking or poorly defined, not as dark as patch of pigment above opercle on
edge of nape; no well-defined rows of spots or stripes on flanks

Description. Body morphology as described for genus. Both jaws with multiple
rows of fine teeth, larger canines present laterally towards rear in outer row
and/or medially near symphysis in inner rows. Proportional measurements are
given in Table 4.

Meristics. First dorsal fm VI; second dorsal fm I, 8; pectoral fins 17 (15-19);
pelvic fins I, 5; anal fm 1, 8. Scales in lateral series 54 (47-63), usually 50-54 in
Guianas and eastern Venezuela, 54-59 in Brazil; predorsal scales 40 (32-49);
transverse scale rows 16 (14-20); caudal peduncle scale rows 17 (13 - 25).

Cephalic lateralis. Fig. 9. Five transverse suborbital free neuromast rows of
which 2nd, 3rd and 4th extend ventrally beyond horizontal row d (2.3.4 pattern).
Short supernumerary segments rarely observed. No of row on upper opercle, row
os connects with row oi at lower rear margin of opercle.

(Coloration in alcohol. Based on specimens from Brazil (AMNH 20743), Guyana
(AMNH 14420, AMNH 72129, AMNH 72151), and Venezuela (MHNLS 9805,
MHNLS 9945). Body tan to dark brown, often with a few scattered brown spots
on upper flanks, not forming regular rows or thin stripes; dorsum occasionally
lighter than flanks; abdomen and gular region lighter. Two streaks radiating
across cheek posteriorly from eye, another along sulcus between nape and cheek
and operculum, expanding into patch of dark brown pigment on nape along edge
of operculum, more pronounced than pigment on upper pectoral-fm base. Spot or
dark patch variably present on upper pectoral-fm base, if present not highly
defined and not as dark as pigment on side of nape; other pigment as observed in
E. amblyopsis.

Distribution. Brazil to the Orinoco Delta of eastern Venezuela.

Comments. Eleotris pisonis reaches a maximum size similar to E. amblyopsis.
The largest we examined was a 113 mm SL individual from Brazil. Teixeira
(1994) mentioned that this species was a common element of the ichthyofauna in
estuaries of northeastern Brazil. The species was taken in salinities ranging from
0.1-18.2 ppt at temperatures of 25-33 C. It was most abundant in the fall in chan-
nels in regions of fluctuating salinity. Perrone and Vieira (1990) found that mar-
ginal vegetation served as the primary shelter and feeding area for this species in
another Brazilian river system during the wet season.

Review of the Spinycheek Sleepers

47

Teixeira (1994) discovered E. pisonis primarily fed on snails (Neritina
virginea), polychaetes, shrimps and fishes. Neritina were the principal food
resource throughout the year; secondary foods changed with season and ontoge-
ny. Smaller spinycheek sleepers ate polychaetes, larger individuals fed on fishes
and shrimps. In another Brazilian estuary, this species ate mostly dipteran larvae
in the wet season and carideans and fishes in the dry season (Perrone and Vieira,
1991). Seasonal changes in diet were associated with fluctuation in the abun-
dance of marginal vegetation, a preferred habitat when it is present. As with the
study by Teixeira, there were ontogenetic differences in diet. Juveniles fed mostly
on dipterans, while adults ate more fishes and crustaceans.

Size at maturity for this species has been reported as 57 mm TL for
males and 43 mm TL for females (Teixeira, 1994). He observed no clearly
defined reproductive period for males, but noted an April surge for females.
Perrone and Vieira (1990) gave a size at maturity of 35 mm SL for females (in
agreement with that above) and suggested a reproductive period linked to the wet
season, summer in southeastern Brazil and winter in the northeast.

Eleotris tecta. Bussing 1996

Eleotris tecta Bussing, 1996: 252, Rio Banegas, Puntarenas, Costa Rica.

Diagnosis. Scales in lateral series 54-60, mode of 59. Neuromast configuration
with 2nd, 3rd, and 4th suborbital rows extending ventrally past horizontal row d.
Urogenital papillae round and blunt in females, long and acuminate in males.

Description. Body morphology as described for genus. Both jaws with multi-
ple rows of fine teeth, larger canines present laterally towards rear in outer row
and/or medially near symphysis in inner rows. Proportional measurements are
given in Table 1.

Meristics. First dorsal fin VI; second dorsal fin 1,8 (1,7- 1,8); pectoral fins 17
(16-17); pelvic fins I, 5; anal fin 1,8 (I,7-I,8). Scales in lateral series 59 (54-60),
predorsal scales 38 (38-43); transverse scale rows 18 (17-21); caudal peduncle
scale rows 14 (13-14).

Cephalic lateralis. Fig 2. Second, 3rd, and 4th suborbital vertical rows intersect-
ing horizontal row d (2.3.4 pattern). No of row on upper opercle, opercular rows
os and oi join at ventroposterior margin of preopercle.

Coloration in Alcohol. Based on specimens available. Body straw color, darker
on dorsum, lighter on belly; mottling present in a few specimens; pigmentation

48

Tulane Studies in Zoology and Botany

[Vol. 31, No. 2 2002]

on lateral scales forming series of interrupted lines six to seven scales in depth
extending from pectoral-fm base to caudal peduncle, more distinct posteriorly.
Dark spot present on upper base of pectoral fin; dorsal and anal fins with wavy,
dusky bands parallel or diagonal to body axis; bars on the caudal, pectoral, and
pelvic fins perpendicular to body axis, some specimens lacking dusky bars on
pectoral and pelvic fins. Cheek with small dark dots; interrupted interorbital bar;
two postocular stripes, top stripe extends to the posterior margin of the opercle,
bottom stripe extends to or slightly past the posterior margin of the preopercle;
bar connecting two postocular stripes along seam of opercle and preopercle in a
few specimens.

Distribution. Pacific versant of Colombia, Panama and Costa Rica.

Comments. Bussing (1998) stated that this species reaches 80 mm SL. The
largest specimen we observed was 62 mm SL. The holotype of the species is
female, not male as originally described by Bussing (1996). There was also some
confusion in Figs. 2 and 3 of the original description. Cephalic neuromast draw-
ings in Fig. 2 are from top to bottom Eleotris picta (c in legend), E. tecta (a in
legend) and E. tubularis (b in legend). In Fig. 3, the illustrations are Eleotris
picta in the upper left comer, E. tecta in the upper right and E. tubularis on the
bottom.

From Bussing (1998) we know that Eleotris tecta inhabits streams with
little to high current up to 75 m above sea level, and has been taken in waters
with temperatures of 25-29 C. It is known to be sympatric, but not syntopic, with
E. picta. He also stated that it is carnivorous.

Eleotris tubularis, Heller and Snodgrass, 1903
Eleotris tubularis Heller and Snodgrass, 1903: 210, Plate 10, Cocos Island, Costa
Rica.

Diagnosis. Scales in lateral series 48-53, mode of 50. Neuromast configuration
with 2nd and 4th suborbital vertical rows extending ventrally past horizontal row
d. Urogenital papilla round and blunt in females; urogenital papilla long and
acuminate in males, tip extending past anal fm origin in adults; long pre-vent fur-
row.

Description. Body morphology as described for genus. Head is less robust than
Eleotris tubularis. Proportional measurements given in Table 1.

Meristics. First dorsal fm VI; second dorsal fm 1,8 (I,8-I,9); pectoral fins 16 (14-
16); pelvic fins I, 5; anal fm 1,8. Scales in lateral series 50 (48-53), predorsal

Review of the Spinycheek Sleepers

49

scales 36 (34-40); transverse scale rows 16 or 17(14-17); caudal peduncle scale
rows 13 (12-16).

Cephalic lateralis. Fig. 2. Second and fourth suborbital vertical rows intersect
horizontal row d. No of row on upper opercle, opercular rows os and oi do not
meet at posterior opercular margin.

Coloration in Alcohol. Based on specimens available. Body olive brown with
head region slightly darker; ventral surface lighter than dorsal surface. No dark
spot on pectoral base; dorsal and anal fins with dusky bands parallel or diagonal
to body axis; pectoral, pelvic, and caudal fins with dusky bars perpendicular to
body axis.

Distribution. Endemic to Cocos Island, Costa Rica.

Comments. The holotype is desiccated but counts on measurements were taken
where possible. This is also a small species, the largest individual observed in
this study was 85 mm SL.

Materials Examined

Eleotris amblyopsis. Belize: FMNH 82078 (30); MNHN 27 (1), Eleotris
belizianus, syntype; MNHN 2528 (2), E. belizianus syntypes. BraziliAMNH
211135 (1); AMNH 20752 (1); Costa Rica: ANSP 140682 (1); ANSP 140683 (1);
ANSP 174839 (2); ANSP 174841 (1); ANSP 174842 (1); NLU 69723 (8); Cuba:
CAS 66647 (4); CAS 89139 (1); MCZ 32926 (2); MCZ 159203 (1). Florida:
ANSP 71070 (5); ANSP 72915 (1); ANSP 72947 (1); ANSP 73105 (1); ANSP
144266 (1); MCZ 13435 (1); UF 59144 (5); UF 91947 (3); UF 87754 (1); UF
18133 (1); UF 47739 (1); UF 33977 (1); UF 87886 (8); UF 47003 (1).

Guatemala: AMNH 32076 (3); AMNH 35114 (1). French Guiana: NLU 76503
(3); NLU 76504 (1); NLU 76505 (2); NLU 76506 (1). Guyana: FMNH 53923
(1); FMNH 53924 (1). Haiti: ANSP 83082 (1); AMNH 19315 (1); AMNH 19421
(1). Jamaica: CAS-SU 69780 (1). Louisiana: NLU 33488 (3); NLU 33614 (1);
NLU 33958 (1); NLU 53287 (1); NLU 69722 (1); NLU 69736 (1); NLU 69843
(1); NLU 69844 (1); NLU 69845 (2); NLU 69846 (1); NLU 69851 (1); NLU
69852 (1); NLU 69853 (2); NLU 69854 (1); NLU 69900 (1); NLU 69901 (2);
NLU 71488 (1). Mexico: CAS-SU 21122 (1); GCRL 2878 (6); UMMZ 167492
(1); UMMZ 178565 (3); UMMZ 184433 (4); UMMZ 184444 (3); UMMZ
184454 (1); UMMZ 238760 (1); UMMZ 184471 (1); UMMZ 184480 (8);

UMMZ 184503 (2); UMMZ 184504 (1); UMMZ 186657 (1); UMMZ 194837
(23); UMMZ 194838 (4); UMMZ 194884 (2); UMMZ 196411 (14); UMMZ

50

Tulane Studies in Zoology and Botany

[Vol. 31, No. 2 2002]

194861 (4); UMMZ 209505 (5); UMMZ 209664 (1); UMMZ 209724 (2);

UMMZ 209772 (2); UMMZ 213610 (1); UMMZ 213611 (3); UMMZ 213612
(9). Mississippi: GCRL 497 (2); GCRL 1456 (3): GCRL 2501 (20). North
Carolina: UNC 8273 (1); UNC 10355 (5); UNC 11668 (1); UNC 14951 (6);

UNC 15379 (1). Panama: ANSP 103393 (12); ANSP 104227 (1); ANSP 104291
(4); CAS 46149 (1); CAS-SU 50905 (1); CAS-SU 50317 (1); CAS 50318 (10);
CAS-SU 50320 (5); CAS 214223 (1); (1); CAS 66651 (1); CAS 66652 (1). South
Carolina: CAS-SU 2009 (!),£’. abacurus holotype, specimen and radiograph.
Surinam: AMNH 211136 (10); ANSP 10577-10579 (3), E. amblyopsis syntypes;
CAS-SU 53292 (2); CAS-SU 53331 (5); MCZ 13429 (20); MNHN A.1672 (1),

E. gyrinus syntype; MNHN A. 1673 (3) E. gyrinus syntypes. Texas: TNHC 8115
(2). Trinidad: ANSP 144625 (1). Venezuela: CAS 50794 (2); ANSP 76244 (1);
MCNG 19176 (1); MCNG 19245 (1); MCNG 13982 (1); SCN 1428 (11); SCN
1439 (1); SCN 1443 (4); SCN 1990 (1); SCN 9945 (1); SCN 14603 (3); SCN
14604 (2); SCN 14605 (1).

Eleotris annobonensis. Pagalu (Annobon): MNHN 1965.611 (4), paratypes;
MNHN 1965.612 (4), paratypes; MNHN 1967.446 (1), holotype. Fernando Po:
MRAC 7846p229 (1); MRAC 92089p0001-04 (4); MRAC 142100-01 (2);
MRAC 7846p222-8 (7); MRAC 145418-20 (3).

Eleotris daganensis. Cameroun: MRAC 7302p2205-7 (3); MRAC 93083p0033-
34 (2). Congo- Brazzaville: MRAC 8027p257-58 (2); MRAC 9057p2607 (1);
MRAC 78027p256 (1). Gabon: MRAC 7660pl26 (1). Guinea: MRAC
92059p4031 (1). Liberia: MRAC 7310p7385-91 (7); MRAC 7310p7392 (1);
RMNH 24403 (6), E. buttikoferi paralectotypes. Namibia: RUSI 63257 (6).
Nigeria: MRAC 8803p58-60 (3); MRAC 9019p529-33 (3); MRAC 9110p918-27
(10). Portuguese Congo: MRAC 1760 (1); MRAC 1764 (1). Rio Muni: MRAC
7846p235 (1); MRAC 173332-4 (3). Sierra Leone: MRAC 73406-8 (3); MRAC
73410-11 (2); MRAC 7310p7382-83 (2); MRAC 7310p7378-81 (4). Togo:
MRAC 7313p429-31 (3).

Eleotris feai. Pagalu (Annobon): MNHN 1974.5.13.1 (1), paratype; MRAC
7445pl (1), paratype.

Eleotris pemiger. Bahamas: ANSP 98816 (1); ANSP 98817 (1); ANSP 148541

(1) . Barbados: CAS-SU 37267 (7); MCZ 13283 (1). Bermuda: ANSP 150163 (1);
ANSP 150164 (1). Brazil: ANSP 121269 (2); CAS-SU 69781 (1); MCZ 159204

(2) . Cayman Islands: MCZ 52513 (1). Costa Rica: ANSP 163142 (1); ANSP
163772 (1); ANSP 163773 (2); ANSP 163774 (1); ANSP 163775 (2); NLU
69725 (4). Cuba: ANSP 39935 (1); ANSP 69214 (3); CAS-SU 1892 (3); CAS
66648 (2); MCZ 13342 (1); MCZ 13334 (1); MCZ 13355 (1); MCZ 13363 (1);

Review of the Spinycheek Sleepers

51

MCZ 13364 (4); MCZ 13366 (1); MCZ 13382 (1); MCZ 13384 (1). Dominican
Republic: ANSP 10574 (1); MNHN A. 1698 (3), Eleotris gyrinus syntypes.
Florida: ANSP 55907 (1). Grenada: ANSP 52517 (3). Guadeloupe: MCZ 13440

(2) . Haiti: ANSP 83082 (1); ANSP 83660 (1); CAS-SU 25604 (4). Honduras:
CAS 35746 (1); GCRL 6003 (1); UF 16325 (6); UF 16333 (3). Jamaica: ANSP
112909 (1); ANSP 144716 (1); CAS-SU 4965 (8); MCZ 34030 (10); MCZ 58321
(8). Martinique: ANSP 152267 (1); MCZ 26070 (12); MNHN A.1597 (1), E,
gyrinus syntype; NLU 75227 (3). Mexico: ECOCH (El Colegio de la Frontera
Sur- Chetumal) 0243 (1); ECOCH 0505 (1); ECOCH 0573 (1); UMMZ 184467

(3) ; UMMZ 196529 (1); UMMZ 209702 (2). Panama: ANSP 99833 (3); ANSP
99915 (1); ANSP 104074 (6); ANSP 104096 (1); ANSP 104158 (2); ANSP
104302 (3); ANSP 104400 (5); CAS-SU 50277 (1); CAS 214224 (5); CAS-SU
50906 (5); CAS-SU 18578 (10); CAS-SU 18579 (11); CAS 66650 (1). Puerto
Rico: ANSP 23552 (4); ANSP 23587 (1); ANSP 91914 (3); ANSP 115625 (1);
ANSP 118559 (44); ANSP 144626 (3); CAS-SU 8243 (1); CAS-SU 8274 (2);
CAS 78668 (2); CAS 11702 (1); MCZ 28871 (1); MCZ 34669 (1); MCZ 49411
(18). St. Martin: ANSP 10575 (1), E. pemiger holotype. St. Vincent: MCZ 26103

(4) ; MCZ 26109 (1); MCZ 26111(11). Trinidad: ANSP 174843 (1). Venezuela:
ex-MCNG 13982 (1); SCN 2451 (1); SCN 3958 (3); SCN 3966 (2); SCN 3968
(1); SCN 5969 (2); SCN 9245 (1); SCN 10001 (2); Virgin Islands: CAS 66645
(1); UMA (University of Massachusetts at Amherst) uncatalogued (1).

E. pemiger X E. amblyopsis hybrids. Cuba: MCZ 13360 (2); MCZ 13365 (1);
Mexico: ECOCH 0272 (1); ECOCH 0861 (1); UMMZ 124299 (1). Panama:
ANSP 178003 (1); MCZ 45744 (1).

Eleotris picta. Colombia: NRM 28607 (1); NRM 28609 (1); NRM 10704 (1);
FMNH 94689 (15); CAS 11691 (1); LACM 24320 (1); NRM 39516 (3); NRM
10705 (2). Costa Rica: ANSP 144126 (1); ANSP 144155 (5); ANSP 144132
(15); LACM 30109-5 (2); LACM 2887 (3); LACM 4859 (1); TNHC 15362 (5);
TNHC 14789 (3); TNHC 15380 (1); TU uncat. (2); TU 84587 (1); TU 24181 (1);
TU 84544 (1); TU 84687 (1); TNHC 14849 (5); TNHC 15350 (3); TNHC 11986

(1) ; TNHC 12006 (1); TNHC 11496 (1); TNHC 11505 (1); TNHC 15360 (1);
TNHC 14755 (2); UCR 130.003 (12); UCR 936.006 (5); UCR 732.007 (5). Isla
del Cocos, Costa Rica: LACM 26462 (2); UCR 736.005 (2). Ecuador: MCZ
58605 (1); MCZ 54970 (1); USNM 288041 (2); CAS 66643 (2); FMNH 56864

(2) ; CAS 66644 (1); FMNH 56865 (1). Ecuador, Galapagos Islands: UWZM
10688 (1); CAS 54585 (1); SIO 59-358-58B (1); LACM 43964-1 (1).

Guatemala: UMMZ 194134 (3); UMMZ 197122 (1); UMMZ 188085 (5);

UMMZ 190541 (2). Honduras: TU 186087 (1). Mexico: UMMZ 164635 (1);
LACM 30369-2 (1); UMMZ 164646 (1); UMMZ 173605 (1); CAS-SU 2907 (5);
UMMZ 164627 (1); UMMZ 164685 (1); UMMZ 172012 (1); UMMZ 172096

52

Tulane Studies in Zoology and Botany

[Vol. 31, No. 2 2002]

(1); UMMZ 172060 (3); UMMZ 183927 (1); UMMZ 172082 (1); UMMZ
183935 (1); UMMZ 172145 (1); UMMZ 184827 (5); UMMZ 184747 (1);
UMMZ 178428 (1). Nicaragua: UCR 52.002 (10); UCR 272.003 (1); UCR
187.002 (7); UCR 270.006 (20). Panama: NMW 76866 (1), E. picta syntype;
FMNH 27274 (1); FMNH 27280 (7); ANSP 146763 (1); USNM 226407 (2);
FMNH 32313 (1); FMNH 32312 (1); FMNH 27283 (6); FMNH 27281 (3);
FMNH 32314 (1); ANSP 104276 (2); ANSP 104277 (1); ANSP 104282 (1);
ANSP 104189 (1); ANSP 104101 (1); ANSP 104403 (2); CAS 66637 (1); CAS
66638 (1); FMNH 27275 (1); ANSP 104165 (2); FMNH 36985 (1); USNM
270831 (10); USNM 260093 (3); USNM 293478 (4); NRM 35870 (1).
Venezuela: SCN 2270 (1).

Eleotris pisonis. Brazil: AMNH 3764 (1), Eleotris carvalhonis, paratype; AMNH
20743 (2); ANSP 69641 (1); CAS-SU 22208 (5), E. carvalhonis paratypes; CAS-
SU 22215 (1), E. carvalhonis holotype, specimen and radiograph; CAS 39109
(3); CAS-SU 52355 (15 of 62); CAS-SU 52356 (2); CAS-SU 52359 (1); CAS-
SU 52360 (1); CAS 53478 (1); MCZ 1196 (1); MCZ 4627 (1); MCZ 13385 (1);
MCZ 13386 (1); MCZ 13389 (2); MCZ 13391 (1); MCZ 13397 (5). French
Guiana: MNHN A.2216 (3), Eleotris belizianus syntypes; MNHN A.2217 (5), E.
belizianus syntypes. Guyana: AMNH 14420 (1); AMNH 72049 (1); AMNH
72129 (7); AMNH 72151 (1); AMNH 73056 (1); AMNH 14420 (1); CAS 51066
(1). Surinam: AMNH 54793 (31); MNHN A. 1693 (1), E. gyrinus syntype.
Venezuela: ANSP 149479 (1); SCN 1429 (7); SCN 1962 (1); SCN 9805 (11);
SCN 13142 (4); SCN 13485 (2); SCN 14602 (2).

Eleotris senegalensis. Cameroun: MRAC 7302p2208-09 (2); MRAC 7329pl630-
31 (2). Congo: AMNH 17020 (1). Congo-Brazzaville: MNHN 1901.8.1.90-3 (1);
MRAC 9057p2612-15. Gabon: MRAC 7302p2204 (1); MRAC 7660pl25 (1).
Guinea: MRAC 92059p4032-33 (2). Liberia: MRAC 7310p7567-68 (2); RMNH
5254 (1), E. buttikoferi lectotype. Nigeria: MRAC 9263p357-8 (2); MRAC
8608p42 (1); MRAC 9110p928 (1). Senegal: MRAC 771p640 (1). Sierra Leone:
MRAC 7310p7556(l).

Eleotris tecta. Colombia: NRM 43549 (1); CAS 66639 (1); CAS-SU 37538 (2);
CAS-SU 37539 (2). Costa Rica: LACM 45893-1 (1), E. tecta holotype. Panama:
USNM 357288 (3).

Eleotris tubularis. Isla del Cocos, Costa Rica: CAS-SU 6348 (!),£’. tubularis
holotype; LACM 25806 (23); LACM 20047 (20).

Eleotris vittata. Angola: AMNH 223211 (1); BMNH 1864.7.11.8 (1), E. monteiri
type. Benin: MRAC 8958pl5 (1). Cameroun: MRAC 7308p26 (1). Congo:

Review of the Spinycheek Sleepers

53

AMNH 43132 (2). Congo- Brazzaville: MRAC 9057p2416-17 (2). Cote
D’Ivoire: MRAC 173598 (1). Fernando Po: MRAC 142099 (1). Gabon: MNHN
a. 1548 (1). Liberia: MRAC 7310p7384 (1). Nigeria: MRAC 8843p463-5 (3);
MRAC 8803p56-57 (2); MRAC 9022p4 (1). Sierra Leone: MRAC 7310p7382-83
(2). Togo: MRAC 739pl41 (1).

Erotelis armiger. Mexico: CAS-SU 3455 (1), Alexurus armiger holotype, speci-
men and radiograph.

Erotelis smaragdus. Florida: CAS 51044 (1). Louisiana: UNOVC (University of
New Orleans Vertebrate Collections) 4295 (1). Panama: CAS-SU 19339 (2).

Acknowledgments

Specimens were provided by W. G. Saul and M. Sabaj (ANSP), M. Stiassny and
B. Brown (AMNH), A-M. Hine (BMNH), D. Catania and W. Eschmeyer (CAS),
J. J. Schmitter-Soto (ECOCH), B. Chemoff and M.A. Rogers (FMNH), S. Poss
(GCRL), J. Seigel and C. Thacker (LACM), D. Taphom (MCNG), K. E. Hartel
(MCZ), O. M. Lasso-Alcala (SCN), D. Nelson (UMMZ), J.-C. Hureau, R Keith
and R Rruvost (MNHN), G. Teugels and M. Rarrent (MRAC), S. Kullander
(NRM ), H. Wellendorf (NMW), M.J. R van Oijen (RMNH), V Mthombeni and
R. Bills (RUSI), C. Klepadlo (SIO), D. Hendrickson, M. Leslie and A. Rentz
(TNHC), H. Bart, N. Rios and M. Taylor (TU), W. Bussing (UCR), C. Gilbert, G.
Burgess and R. Robins (UF), D. Smith (UMA), D. W. Nelson (UMMZ), R.
Cashner and M. O’Connell (UNOVC), F. Swartz (UNC), S. L. Jewett and L.
Ralmer (USNM), and J. Lyons (UWZM). We thank M. Gambordella for lab assis-
tance and L. Olsan for Latin translations.

Literature Cited

Fringe Akihito. 1967. On four species of the gobiid fishes of the genus Eleotris
found in Japan. Japanese J. of Ichthyology 14:135-166.

Akihito, A. Iwata, T. Kobayashi, K. Ikeo, T. Imanishi, H. Ono, Y. Umehara, C.
Hamamatsu, K. Sugiyama, Y. Ikeda, K. Sakamoto, A. Fumihito, S.

Ohno and T. Gojobori. 2000. Evolutionary aspects of gobioid fishes
based upon a phylogenetic analysis of mitochondrial cytochrome b genes.
Gene 259:5-15.

Birdsong, R. S., E. O. Murdy, and F. L. Rezold. 1988. A study of the vertebral
column and median fm osteology in gobioid fishes with comments on
gobioid relationships. Bull of Mar Sci 42(2): 174-214.

Bleeker, R. 1856. Bijdrage tot de kennis der ichthyologische fauna van het eiland
Boeroe. Natuurkd. Tijdschr. Neder. Indie 11:383-414.

54

Tulane Studies in Zoology and Botany

[Vol. 31, No. 2 2002]

Bleeker, P. 1874. Esquisse d’un Systeme Naturel des Gobioides. Arch. Neerl.
Sci. Nat., Haarlem 9:289-331.

Bloch, M. E. and J. G. Schneider. 1801. Sy sterna Ichthyologiae iconibus cx
illustratum. Berlin.

Bohlke, J. E. and C. C. G. Chaplin. 1993. Fishes of the Bahamas and Adjacent
Tropical Waters. 2nd ed. With nomenclatural changes and additions by E.

B. Bohlke and W. F. Smith- Vaniz. University of Texas Press, Austin, TX.

Bory de Saint- Vincent, J. B. G. M. 1825. Dictionaire classique d’histoire
naturelle, VII. Breitburg, D. L. 1989. Demersal schooling prior to settle
ment by larvae of the naked goby. Environ. Biology of Fishes 26:97-103.

Britton, J. C. and B. Morton. 1989. Shore Ecology of the Gulf of Mexico.
University of Texas Press, Austin, TX.

Brothers, E. B., D. McB. Williams, and P. F. Sale. 1983. Length of larval life
in twelve families of fish at ‘One Tree Lagoon’, Great Barrier Reef,

Australia. Mar. Biol. 76:319-324.

Bussing, W. A. 1996. A new species of eleotrid, Eleotris tecta, from Pacific slope
streams of tropical America (Pisces: Eleotridae). Rev Biol Trop
44(l):251-257.

Bussing, W. A. 1998. Freshwater Fishes of Costa Rica. 2nd ed. Editorial de la
Universidad de Costa Rica, San Juan, Costa Rica.

Cope, E. D. 1871. Contribution to the ichthyology of the Lesser Antilles. Trans.
Amer. Phil. Soc. 14:445-483.

Cuvier, G. 1829. La Regne Animale. 2nd ed. Vol 2.

Cuvier, G. and A. Valenciennes. 1837. Histoire Naturelle des Poissons. Vol. 12.
Levrault. Paris, France.

Dawson, C.E. 1969. Studies on the gobies of Mississippi Sound and adjacent
waters II. An illustrated key to gobioid fishes. Publ. Gulf Coast research
Laboratory. 59 p.

Edwards, A.J. and C. W. Glass. 1987. The fishes of Saint Helena Island, South
Atlantic Ocean. I. The shore fishes. J. Nat. Hist. 21:617-686.

Eschmeyer, W. N. 1998. Catalog of Fishes. 3 vols. California Academy of
Sciences, San Francisco.

Gilbert, C. R. and D. P. Kelso. 1971. Fishes of the Tortuguero area, Caribbean
Costa Rica. Bulletin of the Florida State Museum: Biological Sciences
16(1): 1-54.

Gilmore, R. G. and D. J. Herrema. 1981. Fishes of the Indian River Lagoon and
adjacent waters, Florida. Harbor Branch Foundation, Inc. Technical
Report 41.

Ginsburg, I. 1953. Ten new American gobioid fishes in the United States
National Museum, including additions to a revision of Gobionellus. J.
Washington Academy of Sciences 43(1): 18-26.

Review of the Spinycheek Sleepers

55

Gmelin, J. F. 1789. Caroli a Linne...Systema Naturae per gegna tria naturae,
secundum classes, ordines, genera, species. 3 vols. G.E. Beer. Leipzig,
Germany.

Grady, J. M. and J. M. Quattro. 1999. Using character concordance to define
taxonomic and conservation units. Conservation Biology 13(5): 1004-1007.

Greenfield, D. W. and J. E. Thomerson. 1997. Fishes of the Continental
Waters of Belize. University Press of Florida, Gainesville, FL.

Gronow, L. T. 1763. Zoophylacii Gronoviani.

Grove, J. S. and R. J. Lavenberg. 1997. The Fishes of the Galapagos Islands.
Stanford University Press, San Francisco, CA.

Hamilton, F. 1822. An account of the fishes found in the river Ganges and its
branches. Edinburgh and London

Harrison, I.J. 1993. The West African sicydiine fishes, with notes on the genus
Lentipes (Teleostei: Gobiidae). Ichthyol. Explor. Freshwaters 4(3): 201-232.

Heller, E. and R. E. Snodgrass. 1903. Papers from the Hopkins Stanford
Galapagos expedition, 1898-1899. Proc of the Wash Acad of Sci 5:189-229.

Hildebrand, S. F. 1938. A new catalogue of the freshwater fishes of Panama.
Field Museum of Natural History, Zoological Series 22(4):217-359.

Hilgendorf, F. M. 1889. Uber eine Fischsammlung von Haiti, welche 2 neue
Arten, Poecilia (subg. n. Acropoecilia) tridens und Eleotris maltzani,
enthalt. Sitzungsber. Ges. Naturf. Freunde Berlin 1889:51-55.

Hoese, D.F. and A.C. Gill. 1993. Phylogenetic relationships of eleotridid fishes
(Perciformes: Gobioidei). Bull. Marine Sci. 52(l):415-440.

Hubbs, C. L.1953. Eleotris picta added to the fish fauna of California. California
Fish and Game 39(l):69-76.

Hubbs, C. L., and K. F. Lagler. 1958. Fishes of the Great Lakes Region.
Revised Edition. Cranbrook Institute of Science. Bloomfield Hills,

Michigan. International Commission of Zoological Nomenclature. 1926.
Opinion 93, Direction 56. Smith. Misc. Coll. 73 (4):5-ll.

Jordan, D. S. and B. W. Evermann. 1898. The Fishes of North and Middle
America: A Descriptive Catalog of the Species of Fish-like Vertebrates
Found in the Waters of North America, North of the Isthmus of Panama. Parts
I-IV. Bull of the US Nat Mus 47:1-3313, 392 pi.

Jordan, D. S. and C. H. Gilbert. 1882. Description of five new species of fishes
from Mazatlan, Mexico. Proc of U S Nat Mus 4:458-463.

Jordan, D. S. 1896. Notes on fishes little known or new to science. Proc Cal
Acad. Sci. (Ser. 2) 6:201-244, pi. 20-43.

Kner R. 1863. Eine Uebersicht der ichthyologishchen Ausbeute des Herm
Professors Dr. Mor. Wagner in Central Amerika. Konigl. Bayer. Akad.
Wissensch. Muchen. 2:220-230.

Kner, R. and F. Steindachner. 1864. Neve Gattungen und arten von fischen aus
Central-Amerika. Abh. Bayer Akad. Wiss. 10:1-61, 6 pis.

56

Tulane Studies in Zoology and Botany

[Vol. 31, No. 2 2002]

Lacepede, B. G. E. 1800. Histoire Naturelle des Poissons. Vol. 2

Leviton, A.E. and R.H. Gibbs, Jr. 1988. Standards in herpetology and

ichthyology. Standard symbolic codes for institution resource collections in
herpetology and ichthyology. Supplement no. 1: additions and corrections.
Copeia 1:280-282.

Leviton, A.E., R.H. Gibbs, Jr., E. Heal and C.E. Dawson. 1985. Standards in
herpetology and ichthyology: Part I. Standard symbolic codes for institutional
resource collections in herpetology and ichthyology. Copeia 3:802-832.

Lubbock, R. 1980. The shore fishes of Ascension Island. J. Fish Biol. 17:283-
303.

Lubbock, R. and A. Edwards. 1981. The fishes of Saint Paul’s Rocks. J. Fish
Biol. 18:135-157.

Marcgrave, G. 1648. Historiae rerum naturalium Brasiliae.

Mauge, a. L. 1984. Diagnoses preliminaires d’Eleotridae des eaux douces de
Madagascar. Cybium 8(4): 98-100.

McFarland, W. N., E. B. Brothers, J. C. Ogden, M. J. Shulman, E.
Bermingham and N.M. Kotchian-Prentiss. 1985. Recruitment patterns
in young French grunts, Haemulon flavolineatum, at St. Croix, Virgin
Islands. U. S. Fish. Bull. 83:413-426.

Meek, S. E. 1904. The Fresh- Water Fishes of Mexico North of the Isthmus of
. Tehuantepec. Field Columbian Museum, Zoological Series Vol. 5.

Chicago, IL.

Meek, S. E. and S. F. Hildebrand. 1916. The Fishes of the Fresh Waters of
Panama. Field Museum of Natural History, Zoological Series 10(15).

Chicago, IL.

Miller, P.J. 1973. The osteology and adaptive features of Rhyacichthys aspro
(Teleostei: Gobioidei) and the classification of gobioid fishes. J. Zool.,
London 171:397-434.

Miller, P. J. 1998. The west African species of Eleotris and their systematic
affinities (Teleostei: Gobioidei). J. Natural History 32:273-296.

Miller, P. J. and R. Mck. Smith. 1989. The West African species of
Bathygobius and their affinities. J. Zool., London 218:277-318.

Miller, P. J. and P Wongrat. 1991. The innervation of head neuromast rows in
eleotridine gobies (Teleostei: Gobioidei). J of Zool 225:27-42.

Murdy, E. O. 1998. A review of the gobioid fish genus Gobioides. Ichthyol. Res.
45(2):121-133.

Nordlee, F. G. 1979. Niche specificities of eleotrid fishes in a tropical estuary.
Revista de Biologia Tropical 27(1): 35-50.

Nordlie, eg. 1981. Feeding and reproductive biology of eleotrid fishes in a
tropical estuary. J. Fish Biology 18(1):97-110.

Review of the Spinycheek Sleepers

57

Perrone, E. C. and R Vieira. 1990. Ocorrencia e periodo reprodutivo de
Eleotris pisonis (Teleostei: Eleotrididae) na regiao estuarina do rio Jucu,
Espirito Santo, Brasil. Ciencia e Cultura 42(9): 707-710.

Perrone, E. C. and F. Vieira. 1991. Habito alimentar de Eleotris pisonis
(Teleostei: Eleotrididae) na regiao estuarina do Rio Jucu, Espirito Santo,
Brasil. Rev. Brasil. Biol. 51(4):867-872.

Pezold, F. 1984. A revision of the gobioid fish genus Gobionellus. Unpubl. PhD.
Dissertation. The University of Texas, Austin, Texas.

Pezold, F. 1993. Evidence for a monophyletic Gobiinae. Copeia 3:634-643.

Pezold, F. and J. M. Grady. 1989. A morphological and allozymic analysis of
species in the Gobionellus oceanicus complex (Pisces: Gobiidae). Bull.

Marine Sci. 45(3):648-663.

PiETSCHMAN, V. 1913. Fische der Wiesbadener Museums. Jahrb. Nas. Ver. Naturk.
Wiesbaden 66:170-201, 2 pis.

Quoy, J.R.C. and J. P. Gaimard. 1824. Description des Poissons. In Voyage
autour du Monde... execute sur les corvettes de L. M. “L’Uranie” et “La
Physicienne,” pendant lesannees 1817, 1818, 1819 et 1820, by L. de
Freycinet. Paris, France.

Radtke, R. L., R. a. Kinzie, III and S. D. Folsom. 1988. Age at recruitment of
Hawaiian freshwater gobies. Environ. Biol. Fishes 23(3):205-213.

Rafinesque-Schmaltz, C. S. 1815. Analyse de la nature, ou tableau de I’univers
et des corps organises. Palerme.

Sanzo, L. 1911. Distribuzione delle papille cutanee (organiciatiforme) e suo val
ore sistematico nei gobi. Mitt zool Stn Neapel 20:249-328.

Salvage, H. E. 1880. Description des Gobioides nouveaux ou peu connus de la
collection du Museum d’histoire naturelle. Bull. Soc. Philomath. Paris.

(Ser. 7) 4:40-58.

SCHELTEMA, R. S. AND J. R. Hall. 1975. The dispersal of pelagosphera larvae by
ocean currents and the geographical distribution of sipunculans, pp. 103-
115 in Proceedings of the International Symposium on the Biology of
Sipuncula and Echiura, I. Kotor, ed.

Schwartz, F.J. 1999. Gobioid fishes of North Carolina. J. Elisha Mitchell Society
15(4):281-293.

Smith- Vaniz, W. F, B. B. Collette and B. E. Luckhurst. 1999. Fishes of
Bermuda: History, Zoogeography, Annotated Checklist and Identification
Keys. American Society of Ichthyologists and Herpetologists, Special
Publ. No. 4, Allen Press, Inc. Lawrence, KS.

Starks, E.C. 1913. The fishes of the Stanford expedition to Brazil. Stanford
University Publ., Univ. Ser. 1-77, 15 pi.

Tedceira, R. L. 1994. Abundance, reproductive period, and feeding habits of
eleotrid fishes in estuarine habitats of north-east Brazil. J. Fish Biology
45:749-761.

58

Tulane Studies in Zoology and Botany

[Vol. 31, No. 2 2002]

Uribe, A.M. and J. R Diaz. 1996. Chromosome complements of Gobionellus
microdon and Eleotris picta collected in Mexico. J. of Fish Biology
48:796-798.

Uribe, A. M., P.R. Montes and J. P. Diaz. 1994. Chromosome complement of
Eleotris pisonis (Gobiidae; Perciformes) from Mexico: a new case of het
eromorphic sex chromosomes in fish. Cytobios 77:183-187.

Villa, J. 1982. Peces Nicaraguenses de agua dulce. Managua, Nicaragua.

Walbaum, j. j. 1792. Petri Artedi Sueci Genera piscium.

Watson, R.E. 2000. Sicydium from the Dominican Republic with description of
a new species (Teleostei: Gobiidae). Stuttgarter Beitr. Naturk., Serie A
608:1-31.

Wiley, E.O. and R. L. Mayden. 2000. The evolutionary species concept, pp. 70-
89, in Species Concepts and Phylogenetic Theory, A Debate, Q.D.

Wheeler and R. Meier, eds. Columbia University Press.

Wilson, E. O. and W. L. Brown, Jr. 1953. The subspecies concept and its taxo
nomic application. Systematic Zoology 2:97-111.

WiNEMiLLER, K.O. AND N. E. MORALES. 1989. Comunidades de peces del Parque
Nacional Corcovado luego del cese de las actividades mineras. Brenesia
31:75-91.

Table 1. Eastern Pacific Eleotris morphometric data. Standard length reported to nearest 0.1 mm. PI = pectoral fin, D2 = second dorsal fin. Orbit, interorbital,
snout and upper jaw measurements expressed as per cent of head length, all others expressed as per cent of standard length (all rounded to nearest per cent).

Eleotris picta Eleotris tecta Eleotris tubularis

Review of the Spinycheek Sleepers

59

^ CN

CN VO
rn CN

<N VO ^

(S ^ m

^ ^ ^

i i oo

<N CN Tf — . ^ -H

(N — • CS

fA rn

CN CN

m »n

WWW

OO 0\ Tt

CsJ -H

^ ^ ^ lO

VO m m >o

CNJ — I (N — I

m <N tN

-H — I VO r- ^

VO oo m

4 VO

(M ^ rn ^ ^

oo oo oo

Tt fN r-H

oo 00 oo

o U-) r-

Tj- ^ ^

oo oo oo

Zh ^

fN) <N

m ^ ^ ^

O 2:

»o m

u-i U-) »n

CN CM CM

m m
«o »n »n

CN (N (N

m m m

>n >n m

CN <N <N

— < ^ lO

m ro
in

Cs| CN

60

Tulane Studies in Zoology and Botany

[Vol. 31, No. 2 2002]

Review of the Spinycheek Sleepers

61

Table 3. Geographic patterns of meristic variation in Eleotris picta. Regions are: 1) Baja California - Puerto
Vallarta, Mexico, 2) Puerto Vallarta, Mexico - Guatemala Border, 3) Guatemala, Honduras, El Salvador,
Nicaragua, 4) Costa Rica, Panama 5) Colombia, Ecuador, Galapagos Islands.

REGION

MEAN

LOW 95% Cl

UP 95% Cl

MODE

LATERAL SCALES

1

63.00

61.97

64.03

65

2

63.00

61.92

64.08

64

3

61.72

61.26

62.18

61

4

61.18

60.90

61.46

60

5

61.95

61.44

62.47

61

PREDORSAL SCALES

1

60.63

59.31

61.94

62

2

60.09

58.11

62.07

62

3

60.34

59.84

60.84

60

4

60.29

59.97

60.60

61

5

61.51

60.95

62.08

61

TRANSVERSE SCALES

1

23.25

22.39

24.11

24

2

24.91

24.28

25.54

25

3

26.02

25.41

26.63

25

4

24.56

24.26

24.85

25

5

24.15

23.71

24.59

24

CAUDAL PEDUNCLE SCALES

1

18.31

17.85

18.78

18

2

19.18

18.34

20.02

19

3

19.70

19.24

20.16

19

4

18.59

18.40

18.78

19

5

18.15

17.78

18.51

17,18

PECTORAL FIN RAYS (LEFT)

1

17.56

17.29

17.84

18

2

17.82

17.55

1^09

18

3

17.74

17.59

17.89

18

4

17.66

17.53

17.80

18

5

17.36

17.18

17.55

17

Table 4. Western Atlantic Eleotris morphometric data. Standard length reported to nearest 0.1 mm. PI = pectoral fin, D2 = second dorsal fm. Orbit,
interorbital, snout and upper jaw measurements expressed as per cent of head length, all others expressed as per cent of standard length (all rounded to nearest

per cent).

E. amblyopsis E. perniger E. pisonis

62

Tulane Studies in Zoology and Botany

[VoL 31, No. 2 2002]

„ <N (N
<^2 2

OO CO

— ^ — IT)

— ' — o —

fS — —

c<-i m

<N fS

m OO

m — CO

Tt <N

O — — '

o^ ^

^ 111 'O ^

OO ^ 3

cn m

^

— — O

— m

Tf — —

X) .S, S ’5 on (U

;a ~

pectoral fin length 26(136) 25 -26 19 -31 25 (70) 24-25 20-35 23 (55) 23 -24 20-26

Table 5. Meristic characters for western Atlantic Eleotris. Means and 95% confidence interval values rounded to the nearest whole number. Cpd = caudal
peduncle scale rows, PI = pectoral fin.

E. amblyopsis E. perniger E. pisonis

character mean (N) 95% C.I. Range Mode mean (N) 95% C.I. Range Mode mean (N) 95% C.I. Range Mode

cpd scales iT(53) iJTb TT24 \2 13^21 17 TT(45) 16^18 i>25 ~

Review of the Spinycheek Sleepers

63

vn

CN «o

Wheelerigobius

Recent Issues of Tulane Studies of Zoology and Botony
Other back issues available; please consult our website.
http://www.museum.tulane.edu/pubs/

Volume 30, Number 1.

White, D.A. and S.P. Darwin. Woody vegetation of tropical lowland deciduous
forests and Maya ruins in the north-central Yucatan Peninsula, Mexico.

Garcia-Franco, J.G. and V. Rico-Gray. Population structure and clonal growth in
Bromelia pinguin L. (Bromelilaceae) in dry forests of coastal Veracruz, Mexico.

Reid, J.W and G.G. Marten. The cyclopoid copepod (Crustacea) fauna of non-
planktonic continental habitats in Louisiana and Mississippi.

Volume 30, Number 2.

Johnston, C.E. and W.R. Haag. Life history of the Yazoo darter (Percidae:
Etheostoma raneyi), a species endemic to north-central Mississippi.

Rico-Gray, V., J.G. Garcia-Franco, A. Trigos-Landa, R. Mata, and P. Castaneda.
Leaf-miner defenses in Bromelia pinguin L. (Bromeliaceae) in Veracruz,

Mexico.

Guzman, G. Observations on some fungi from Louisiana and Mississippi in
comparison with those of Mexico.

Dundee, H.A. Some reallocations of type localities of reptiles and amphibians
described from the Major Stephen H. Long expedition to the Rocky Mountains,
with conunents on some of the statements made in the account written by Edwin
James.

Volume 31, No. 1.

Suttkus, R.D., C. Jones. Observations on the nine-banded armadillo, Dasypus
novemcinctus, in southern Louisiana, [abstract]

Bart, H.L., M.S. Taylor. Systematic review of subgenus Fuscatelum of Etheostoma
with description of a new species from the upper Black Warrior River system,
Alabama, [abstract]

Williams, J.D., A. Fradin. Fusconaia apalachicola, a new species of freshwater
mussel (Bivalvia: Unionidae) from precolumbian archaelogical sites in the
Apalachicola basin of Alabama, Florida, and Georgia, [abstract]

, ^

■ 'H

■ /
/-