I

N.C. DOCUMENTS
CLEARINGHOUSE

DEC 2 0 REC'D
STATE UBRAP.^Oe-

I

number 9

June 1983

EDITORIAL STAFF

John E. Cooper, Editor
Alexa C. Williams, Managing Editor
John B. Funderburg, Editor-in-Chief

Board

Alvin L. Braswell, Curator of David S. Lee, Chief Curator

Lower Vertebrates, N.C of Birds, N.C

State Museum State Museum

John C. Clamp, Associate Curator William M. Palmer, Chief Curator

(Invertebrates), N.C of Lower Vertebrates, N.C

State Museum State Museum

James W, Hardin, Department Rowland M. Shelley, Chief

of Botany, N.C State Curator of Invertebrates, N.C

University State Museum

Brimleyana, the Journal of the North Carolina State Museum of Natural His-
tory, will appear at irregular intervals in consecutively numbered issues. Con-
tents will emphasize zoology of the southeastern United States, especially North
Carolina and adjacent areas. Geographic coverage will be limited to Alabama,
Delaware, Florida, Georgia, Kentucky, Louisiana, Maryland, Mississippi, North
Carohna, South Carolina, Tennessee, Virginia, and West Virginia.

Subject matter will focus on taxonomy and systematics, ecology, zoo-
geography, evolution, and behavior. Subdiscipline areas will include general
invertebrate zoology, ichthyology, herpetology, ornithology, mammalogy, and
paleontology. Papers will stress the results of original empirical field studies, but
synthesizing reviews and papers of significant historical interest to southeastern
zoology will be included.

Suitability of manuscripts will be determined by the Editor, and where neces-
sary, the Editorial Board. Appropriate specialists will review each manuscript
judged suitable, and final acceptability will be determined by the Editor.
Address manuscripts and all correspondence (except that relating to subscrip-
tions and exchange) to Editor, Brimleyana, N. C. State Museum of Natural
History, P. O. Box 27647, Raleigh, NC 27611.

In citations please use the full name — Brimleyana.

North Carolina State Museum of Natural History

North Carolina Department of Agriculture

James A. Graham, Commissioner

CODN BRIMD 7
ISSN D193-4406

A New Species of Woodland Salamander

of the Plethodon glutinosus Group
from the Southern Appalachian Mountains

Richard Highton

Department of Zoology,
University of Maryland, College Park, Maryland 20742

ABSTRACT. — A new species of woodland salamander, Plethodon
aureolus, occurs between the Little Tennessee and Hiwassee rivers on
the western slopes of the Unicoi Mountains and nearby lowlands in
southeastern Tennessee and adjacent North Carolina. It is a small-
sized member of the P. glutinosus group and was discovered by a study
of electrophoretic variation in 22 genetic loci. It is sympatric with the
white-spotted form of P. glutinosus (here recognized as a distinct spe-
cies, P. teyahalee) at 28 localities, and at one of these it is also sympat-
ric with typical brassy-spotted P. glutinosus. Plethodon aureolus
hybridizes with Unicoi Mountain P. Jordani on Sassafras Ridge, but
there is no evidence of significant hybridization between P. aureolus
and P. teyahalee or P. glutinosus.

Two unpublished electrophoretic studies of geographic genetic
variation in eastern woodland salamanders of the Plethodon glutinosus
group, one by Peabody (1978) and the other in preparation by Highton,
have revealed the existence of an undescribed species of the group. Its
range appears to be largely restricted to the western slopes of the Unicoi
Mountains and adjacent lowlands, between the Little Tennessee and
Hiwassee rivers, in Monroe and northern Polk counties, Tennessee, and
adjacent Graham and Cherokee counties, North Carolina.

Highton (1970) called attention to the presence of three distinct
geographically parapatric color pattern variants of P. glutinosus in the
southern Appalachian Mountain region: (1) populations in the moun-
tains of western North Carolina are characterized by having small dor-
sal white spots; (2) in populations from northeastern Georgia many
individuals lack dorsal spotting; and (3) populations to the west and
south of the above areas are characterized by having brassy-colored
dorsal spots. Highton (1972) mapped the distribution of three parapat-
ric dorsal pattern variants of P. glutinosus in Pennsylvania, Maryland,
Virginia and West Virginia. Two of these resemble the first and the
third southern Appalachian types in the color of their dorsal spots,
while a third, smaller. Coastal Plain variant is characterized by its very
small dorsal brassy-colored spots. I suggested that there may be limited
gene exchange between some of these parapatric forms and some pairs
may be at or close to the species level of evolutionary divergence. Our
unpublished genetic studies have shown that hybridization often occurs

Brimleyana No.9:l-20. June 1983.

2 Richard Highton

in the narrow overlap zones where the ranges of some of the above
parapatric forms are in contact. The species described here differs genet-
ically from all of the above types and throughout its range it is sympat-
ric with the white-spotted southern Appalachian variant of P. glutino-
sus. The two forms appear to be both morphologically and genetically
distinct at all 28 localities where they have been taken sympatrically.

The new species is characterized by possessing abundant brassy-
colored dorsal spots and by its small size. It is very distinct from the
sympatric, white-spotted, large-sized populations of P. glutinosus, but
some other nearby brassy-spotted populations of P. glutinosus are very
similar to it in appearance. Using the same electrophoretic methods and
genetic loci described in Highton and MacGregor (1983), the new spe-
cies was compared genetically to 11 samples of P. kentucki, 128 samples
of P. glutinosus and 41 samples oi P. jordani taken from localities scat-
tered throughout their ranges. It was also compared with a single sam-
ple of each of the other four species of the group (yonahlossee, cad-
doensis, ouachitae and fourchensis). Geographic genetic variation in the
latter three species was studied by Duncan and Highton (1979) and the
remaining results are being prepared for publication. The new species is
genetically distinguishable from samples of all of these species, just as it
is from sympatric P. glutinosus. However, I have failed to find any
morphological characters that may be used to distinguish it from some
allopatric types of brassy-spotted P. glutinosus. The diagnosis presented
here is therefore valid only for comparisons with the three forms of
eastern large Plethodon with which it is sympatric. This is the second
cryptic species of Plethodon discovered by electrophoretic studies of
genetic variation in proteins, the first being P. websteri Highton (1979).

The new species is named for its brightly-colored brassy dorsal
spots. The name is from the Latin word meaning gilded, ornamented or
very beautiful.

Plethodon aureolus, nevj species

Diagnosis. — An eastern Plethodon of the P. glutinosus group
(Highton and Larson 1979). It differs from sympatric white-spotted P. glutin-
osus by its smaller size, its relatively larger dorsal spots, the presence of
abundant brassy flecking in the dorsal iridophore spots, and more
abundant lateral white or yellow spotting. It differs from most nearby
populations of brassy-spotted P. glutinosus by its smaller size and light-
er chin. It differs from Unicoi Mountain P. jordani by the presence of
dorsal spots and by its more abundant white iridophore spotting on the
sides and legs.

Holotype.— USNM 238341, an aduh male collected at Farr Gap
(locality 1, Table 1), Unicoi Mountains, Monroe County, Tennessee, on
30 June 1979, by Richard Highton and Jeffrey K. Streicher.

New Species of Plethodon 3

Paratypes. — USNM 238342-51, topotypes, same collecting data as
the holotype.

Other material. — Plethodon aureolus from 31 localities (Table 1)
have been identified electrophoretically, and preserved specimens from
all of these sites will be deposited in the National Museum of Natural
History (USNM).

Description of holotype. — Before preservation, the length from the
tip of the snout to the anterior angle of the vent was 54 mm, to the
posterior angle of the vent 58 mm, and the total length 122 mm. There
are 16 costal grooves (equivalent to 17 trunk vertebrae) and the vomer-
ine teeth number 8 on the right side and 9 on the left. In life there were
abundant dorsal white iridophore spots with much associated brassy
flecking scattered on a black ground color. Similar spots were also pres-
ent on the dorsal surfaces of the legs and the top of the head. There
were abundant yellow iridophore spots on the sides of the head and
body and a few yellow iridiphore spots were also present on the chin
and belly. The chin is lighter than the belly.

Distribution. — Plethodon aureolus is known from southern and east-
ern Monroe and northeastern Polk counties, Tennessee, and also occurs
in northwestern Cherokee and western Graham counties, North Caro-
lina (Fig. 1).

Variation in P. aureolus. — There is little morphological variation in
P. aureolus throughout its small range, except that at higher elevations
in the northeastern part of its range the dorsal spotting may be very
reduced or absent in some individuals.

Remarks. — Although I had collected several P. aureolus during my
earlier study of variation in southern Appalachian large Plethodon
(Highton 1970), it was not recognized as distinct from other brassy-
spotted P. glutinosus until recently when we obtained new material for
our electrophoretic studies. The results have indicated that the large-
sized, brassy-spotted, dark-chinned populations from eastern Tennessee
are closely related to P. glutinosus from the northern part of its range
(New York west to Illinois and south through Kentucky, West Virginia,
western Virginia and eastern Tennessee). Since the type locality (Prince-
ton, New Jersey) of P. glutinosus is within this area, this form will
retain the name P. glutinosus regardless of the eventual taxonomic sta-
tus of the other geographic variants. These northern populations of P.
glutinosus are characterized by possessing much darker chins than those
of the white-spotted populations (Highton 1962), P. kentucki (Highton
and MacGregor 1983) and P. aureolus. However, in the immediate vi-
cinity of the range of P. aureolus, many individuals of otherwise geneti-
cally typical P. glutinosus possess unusually light chins, making it very
difficult to distinguish the two species without an analysis of their pro-
teins. The two have been found sympatrically only at locality 4, along

Richard Highton

ed

U

.0

-o

S

u

■»->

0

ex

•5b

0

^

Im

CA

*■(

D

0

^

"S

•4^

etf

^

'rt

^

a>

T3

03

2

_«

'0

_rt

^

X

Vm

X

0

0

;z:

> w

-30

ON <N — r-

rvjcs-^ rsj-^-^-* — <«r>m— H

vO O "T* 00 fNJ O O

\D G m *r^
c^ CO ro

oooooocxooooc»oooor--oooooooooooooooooooooooo 000000000000

<NooavrnONTf<NTt'sO

00 — <N«NW^<N\0^»n

so -H
so —

TttNOOOOs<N<N«r^r*^«r>

w-^— <voa\osr--«n-HfNr~-
ooTrrN<N(Nm«OTtr<im

00 — < Tf O -^ 00

Tt O V^ O »0 OS
r^ rN m m Tt «n

a. as

<L> 0 ^ <L) p

^ =3 1 -I c I

i S a>

050

u :5 2 u
;z: :^ H ;z

^ . z ^ . .

g s §

u

— '<Nf<)Ttinsor--ooa\0 — <Nrn'^>o^r-ooON

New Species of Plethodon

— — « tN m

-^ m «ri — ^ «r^ —

^ O — — —

— ' — <N — ' OO Tf

m rN rs«

m o

— H <N — ro —

X5 X) X)
OO O On

Tt Tt Tf

0<N — — <Nr4 — w-iw-iiou-i — — — — — OO OO — O— f*^ — TtTt«r^r<-) — Tt

■^■^TtTt TtTtTtmroror^TtTtTtTtTtTt-^ "^TtTtTtTt TtTtrnrnr«-ir«^Ttm

OOOOOOOO OOOOOOOOOOOOOOOOOOOOOOOOOOOO OOOOOOOOOO OOOOOOOOOOOOOOCX)

— "Tt-^o Tt«/-^o«/^u^»n<NOOr^<N«ntN«o TtTt— ^u-^cs rNicoTt«OTtmTt«n

Osr-vdsO Ttm<NvOTt<NrsJCS<NasONOOTtO OOOOvr^-i r-ior^r-i^tNONso

^ ^ ^ ^ ««— ,CM<N<NfN— <m<Nr4«NrsJ<N <N<N«N — — — ,— ,04 — — — OO

CO ro ro f*^ rn f*^ f*^ (^ o r*^ r*^ ro c^ ro r^^ ro r*^ c^ r^^ c^ r*^ f*% f*% r^^ f^ c^ c^ rn r*^ f^ fO

r^ o <N OO

m <N — OO
«r> (^ O On

00 Tt O Tt «/^ vO

Tt r~ — W-) Tt J^

CO tN \o m OS Tt

<N O I^ O ON Tt

<N Tt OO OO w-i r-

so vo <N <N rN fN

00 <N

m NO t^

ON t--

— rs rsi

•r> m

Csj ro Tt

00 O Tt OO 00
— — Tt Tt OV

O tN r- rsi ov

00 fN O — 00

CN \o r^ U-) —

. u. .

a

^ 2

OJ o

U

;z: . .

Monroe-
Graham

Monroe
tl

It

- c i; o t

C/3

u

e c s

C o p e«

o

ON o —
rvi m m

r*^Tt^nvor~-oooNO —

00 ON

Tt Tt

o —

<NfnTt»n^r>-oooN

Richard Highton

soo — •/^«n«or--ro«r>m'^rf«r>

m r- in w-^

m fN m «r> r--

OOTfOfNO'^«n — «NTl-<Nr«^»n<NrnTtinr'^Tt»n-^0<NTtTt

•nTf»n»n-^ — O— ''-^rN|<N<NfNm<N<N<NromTtro>n— "— "O"^

OO0OOO0OOOOOOOCX)0O0O0O0OCX)0O0OOO0O0O0O0OOO0O0OOO0O0O

«nmmrsimfomcs<NrNrv|-^-^<N<N-H-^— .000<NOOO«n

r*^ ro f^ r*% c^ ro r*^ r*% ro fO r^^ co r*^ rn r*% f^ r*^ r*^ (^ f^ ro r*^ c*^ r*% (^ ro

S5 S

pq

O »n fNJ <N oo r'^
rN oo r- so <N r-
ro Tt (^ <N so "n f*^

— r-r-sOTtav-HtNON>ri

<NfNfNsO«n«n«n'^r*^

O 3

c o
^5

o

u

rt o

o — <N m Tj-

\0 *0 so ^ ^

»r> NO r^ 00 On o —

so so so so so l^ t~-

t^oooNO-^fNmTt»n

r--r-r>-oooooocx)c»oo

"5*. O

bo

■| 'I

New Species of Plethodon 7

Ellis Branch of Spring Creek, near Springtown, in northern Polk
County, Tennessee, with no genetic evidence of current hybridization
between them (see below).

At 28 of the 3 1 P. aureolus localities, white-spotted P. glutinosus
has been taken in sympatry. Highton (1972) pointed out that the name
Plethodon jordani teyahalee Hairston (1950) is available for the latter
form. The population at the type locality, Teyahalee Bald, Graham-
Cherokee County Une, North Carolina, is probably of hybrid origin, but
is much more like white-spotted P. glutinosus than P. jordani (Highton
1970). As shown below, this form occurs sympatrically with both P.
aureolus and P. glutinosus at locality 4, where apparent reproductive
isolation exists between all three forms. On the basis of this evidence, I
suggest that the white-spotted form should also be recognized as a dis-
tinct species, P. teyahalee. Its distribution and genetic relationships will
be discussed in a later paper.

^'^^

.^^^

^^\r'-^

'-1

\ MONROE CO.
\

\

\

•15

20. •18

19

'36

25

• -''•37

2^*38 GRAHAM CO. /

• 27, t /^'ll I /

4^^^ /*r2 .- \ ^-'-v .W

3 . "-• ^' .40

i.^^ n

POLK CO.

TN.

I CHEROKEE CO

N.C.

GA

Fig. 1. Distribution of P. aureolus in southwestern North Carolina and south-
eastern Tennessee between the Hiwassee and Little Tennessee rivers. Crosses
represent localities (46-48) where P. aureolus and P. jordani hybrids occur.

8 Richard Highton

Plethodon jordani, another Ught-chinned species, occurs at higher
elevations in the Unicoi Mountains (Highton 1962, 1970). An electro-
phoretic analysis of geographic genetic variation in P. jordani and P.
teyahalee in the southern Appalachian Mountains by Peabody (1978)
showed that the latter species is more closely related to some popula-
tions of P. jordani than it is to most other P. glutinosus. Plethodon
jordani and P. teyahalee hybridize extensively in a number of contact
zones (Highton 1970; Highton and Henry 1970; Peabody 1978), includ-
ing the entire periphery of the range of P. jordani in the Unicoi Moun-
tains, but in many areas the two species overlap extensively without
evidence of hybridization. Although we have no evidence of current
hybridization between P. aureolus and P. teyahalee, a transect through
the only known contact zone between P. jordani and P. aureolus on
Sassafras Ridge provided evidence of hybridization between these two
species (unpublished data).

The small size of P. aureolus is one of its most distinctive features.
As discussed in Duncan and Highton (1979), size is a difficult character
to use in salamander taxonomy. In Plethodon, however, mean adult size
and maximum length are sometimes quite consistent among genetically
closely related populations. At the type locality of P. aureolus, 356 indi-
viduals were collected for a study of the life history of the species. The
largest specimen is 67 mm from snout to anterior angle of the vent.
Only three other individuals are over 61 mm in snout-vent length. All of
the other species of southern Appalachian large Plethodon attain much
larger sizes (Highton 1970). The mean adult size of both P. glutinosus
and P. teyahalee is usually at least 70 mm and large adults are often
over 80 mm (record size, a P. teyahalee from Davis Ridge in the Great
Smoky Mountains, Sevier Co., Tennessee, 94 mm). In a sample of 78 P.
teyahalee from the type locality of P. aureolus the 10 largest females
range from 75-90 mm (mean 81.6) and the 10 largest males range from
74-90 mm (mean 78.5).

I suggest that an appropriate common name for P. aureolus is the
Tellico salamander. Tellico Plains is located centrally in its range and
much of this region is drained by the Tellico River and its tributaries.

Electrophoretic Genetic Analysis of Proteins.
Once collections from the type locality of P. aureolus were found to
differ genetically from all other Plethodon species, extensive field work
was done in the Unicoi Mountains and adjacent areas to determine the
geographic distribution of each of the four species and to study their
geographic and genetic interactions. Salamanders were obtained from 84
localities, and individuals from each of these were compared using the
same 22 genetic loci and methods described in Highton and MacGregor
(1983). When compared with a sample of P. glutinosus from near the type

New Species of Plethodon

Fig. 2. Sites of samples of P. glutinosus (hollow triangles), P. teyahalee (solid
circles) and P. jordani (solid squares) identified electrophoretically in southwestern
North Carolina, southeastern Tennessee and northern Georgia. Dotted lines outline
known range of P. aureolus.

locality in New Jersey (locality 10), 32 samples are genetically very similar
and are referred to this species. All are from areas west, north or south of
the range of P. aureolus (Fig. 2) and represent large, brassy-spotted
animals similar in appearance to those from New Jersey, although some
have chins that are much lighter than those in northern populations of P.
glutinosus. Plethodon teyahalee was found at 44 localities, and at 28 of
these P. aureolus was taken in sympatry. Plethodon teyahalee occurs at
low and intermediate elevations east of the range of P. glutinosus (Fig. 2).
Since P. teyahalee may be distinguished from the other two species by its
color pattern in life, its distribution west of the French Broad River is
probably accurately indicated by the map in Highton (1970, Fig. 5).

Two samples from populations of Unicoi Mountain P. jordani are
included in order to compare this species with the three other forms. These

10 Richard Highton

are the same salamanders studied by Peabody (1978) and are from near
Junction, on Snowbird Creek (locality 13) and Johns Knob (locality 14).
The distribution of P. jordani in the Unicoi Mountains was mapped by
Highton (1970) and its range supposedly includes all of the higher areas of
this mountain range. It was therefore surprising to discover that at four
localities in the northernmost part of the Unicoi Mountains, populations
resembling P. jordani in coloration (no dorsal spotting and reduced lateral
spotting), are assigned genetically to P. aureolus. These are from near
Cherry Log Gap (localities 22 and 23), Naked Ground (locality 38), and
Stratton Bald (locality 39). Thus the northernmost known sites for P.
jordani in the Unicoi Mountains are now in the vicinity of Johns Knob.
Highton (1970) called attention to the apparent hybridization between P.
jordani and P. teyahalee all around the periphery of the range oi P. jordani
in the Unicoi Mountains. Two transects reported by Peabody ( 1 978) (from
Johns Knob west along the North River, and from Junction east along
Snowbird Creek) have confirmed this hybridization genetically.

At some of the 84 localities few animals were collected. Voucher
specimens were preserved from all of the sites, and tissue samples (usually
viscera and tail muscle) from some were used for electrophoresis instead of
the material used in my previous work on Plethodon (whole animal
homogenates). The three general protein loci usually cannot be scored
from homogenates of viscera and tail muscle. Rather than use many with
small sample sizes and/ or incomplete genetic data, I present here the
results of a complete genetic analysis of only 1 7 populations in addition to
the sample of P. glutinosus from New Jersey (locality 10). These are from
scattered sites throughout the local ranges of the four species: 4 P. aureolus
(localities 1-4), 6 P. glutinosus (localities 4-9), 5 P. teyahalee (localities 1 , 2,
4,11,12) and 2 P. jordani (localities 1 3 and 1 4). Three species are sy mpatric
at locality 4, and P. aureolus and P. teyahalee are sympatric at localities 1
and 2. Material from all other localities shows no unusual genetic variation
beyond that observed in the 18 samples for which complete genetic analysis
is presented.

Table 2 provides the frequency data of genie variation of 18 popula-
tions from 14 localities. Of the 22 presumed genetic loci evaluated, 3
(Mdh-1, Pep, and Pt-3) show no variation. Three loci are monomorphic
except for a single population: a-Gpd has a rare slower allelomorph in P.
glutinosus (.02) at locality 8, Gdh has a slower allelomorph (.29) in P.
teyahalee at locality 1, and Mdh-2 has a rare faster allelomorph (.02) in P.
aureolus at locality 2. Table 3 gives Nei standard genetic distances {D) and
normalized identity of genes (/) (Nei 1972) for all comparisons and the
mean heterozygosity (//) estimated from allelomorph frequencies. The /
values are clustered by the UPGM A method (Sneath and Sokal 1973) in a
phenogram in Figure 3.

New Species of Plethodon

11

— O — Ors|(N— -O
' d1 o- ^ '•'^

^ 'i

X O 0)

VO >0 O^
Ov O (N

JO -o C

o o

ov o r^.

"cs "C c

o o — o

in (N —

^ r 'i

^ -51 'i

o o o o so o

^ X ^ ir

o o

— OS

CT; O rn

Id ^ c

O CX) o o o

O "/^ w^ O

o

[^

S

^

S

;^

^

S

"rt

:^

~CJ

"w

2"

'i

:^

"o

^ r

o o

rsl oo

2" ^

O o m rsi <N

— CO.

O OS

00 00 <N

o o

— OS

O sO — fN

O On

rr^ sO

cT ^ 'i

• II • • 1 1 -tJ 1 1

2- 'i

<5 •>*

00 — •

12

Richard Highton

c 2" "o ^ ~«' £■

^ ^ ^ 'i

i> m i^
<N o o

tr^ c "rt'S'trs'-^ c

q q

«■ s-

o t-- -^

oo — q

<?N O O

^ s- tr

so — o o — <^

'± « ^

o<> o^q ovq w-i^t — <z> a <z> <:s <z> -^

oo [^ oo

C9|L "O CO CO CO "On "^1!

r^oo r^r^»Nr~>/^i/-i(N

CO oa^ OfNqqoTT— ■

rsoo OOO ^rT— <N — — ^^

a P^ <o ni CO "^'^ ~P'^'=!'^'^P'^

oor? r^r^ S^fNrT^— rs^i-^

OvO r-r-1 fN — — — — OfNr«-)

O O fN O 00 5^

.^T3 COCO CO jO OOOC

^ £, o "u ■- ^

_ _ 00<N_ -<^ _ <NOfN — O — O — OfN

CO T3 ^ ^ CO CO II ^ ^ ^ ^ ^ O O ^ O O II

o o o o o o

CO ~^co^ — ov>nu-) ^

~co ^ Id ^ £" •o

ONiooor-'^ roi^r--

OO^ro — OO 00^""^

CO "O CO XI ^OIIO^^ ^^^11

O O^ r^ir-000<NOOOv</^'/^00 vnu-iOOOrvlvD

_rsi ^<^ oo — ^-tj-fN — oofNOOrM m — •^OOCN

""ll ^11 ^OOOIIOOIIOOII OOOOOII

_c _t cax)eox_cjs-o_cxi-o3 utso--^^a-^

— ON r<-i m (N\Oror<-i — rn'0<r>OOOONOvO'^rO — — — NO

OON _ /^f^ ^c^ OI^INmoOOO — O — — OCslOOOOOl^

"^ >J,0 * ■'^ll ^1! OOOIIOOOO^^OO— >OOOOOOIi

XiT3 C C c0X!T31=c0J3CJT3iu'— aoj=----^J»!CCO,aOC

New Species of Plethodon

13

c
o

.5 aj

-^ £

^ o

a> -^

tso *o

ffj a"

o

00
at

^

rn

T3

<1)

H

>
O

o

X)

•s

1^'

c«

E

;/5

C/5

C

a>

(i>

ao

^-c

Vi_

=1;

O

^w

o

N D
O (D

oo

?

S

OO

oo

oo

ON

OO

oo

NO

oo

On

OO

00

oo

oo
oq

oo
oo

r^

oo

OO

OO

oo

oo

oo

OO

oo

so
OO

oo

oo

oo

NO Tf rsi Tt

NO r^ NO NO

ON «ri w-^
NO r- r^

ON NO O

NO NO r-

OO On ON On
On On On ON

«r»m— <<N Nomr»^NOrfooTf onOvoo — h oom tj-
Nor~-NONO Nor-~r-NONONor-- ononon o — — o

t^<^o— NO«N<NNOTtr-ro ooon fs— < r--m o\
Nor~-NONO Nor~-r--NONONor-- onon oo — — o

«r>ro-^rs| r^rorooo»OoOTf on -m — -^ nOCsI On

Nor~-NONO Nor^r~-NONONor~ on ooo — — o

ro— ^ON-^ NOc'^r'^r-rtoOTt

Nor-u->NO Nor-r-~NONONor-

— fN — rs o NO r-

o o o o <N — o

or-Ttoo rsjTfoo — oorN ono— 'Or- onon no

r-r^r--r- ovOnononooon rsimmro<N — .-^ -^

-^«or-m iTi-HfTir^— ON onooononno — o ^n
r-r-r-r-- OnOnOnOnOn o mroror*^ro (Nrsl o

r-r^oor-- onooonon o— ' TtTfTf^j-Tt tN<N o

NO

o
oo

;5

?

ON

?

q

q

o

?

ON

"^^

^,

oo

On

(N

r>i

rNj

ON

in

ON

o

^

oo
q

q

r«-i

^

OO

ON

m

°

NO

On

q

o

2

o

q

rn

r^

oo

On

^

q

On "^ NO r^

NO r^ r^ r-

ON«nNOOv-)oo -^or^i— "OO <N<N m
ooo— 'OO -^-^TfTtm r^fNj —

«r> NO w-)

oo oo On

moNooooN(N»n or-ONr-m «nfN o
m<Nr^rn<NrnfN «r>TfrtTtTt mrn —

oo
oo

00

•n

q

^

^

m

(N

o
m

m
«n

?

u->

o
»n

oo

m

o

OO

q

ON

2;

m

°

^

*

^

^

m

rn

rn

m

o
m

^

»n

m

o

m

r--

rn

en

^

^

NO

rn

NO

^

m

^

m

2

— '<NmTt «r)Nor-oooNO"«!t — «(N — <NTt mTj- >

14

Richard Highton

o

r

"T

ro
"T

T

o

r

a:

a>

o
a>

CO

3

O

c

a:

CO

o

a:

ro_

_r

.; " ,.

~1

"1

-

-

evil

,

^-J

— I •

CD '

■ . k

00 J

^ —

to

.. ^ „„

CM L

[

_l

o
U

< ON

2; OO

0^

CX)

New Species of Plethodon 15

The 18 samples cluster into 4 groups, each representing a taxonomic
species. There is very little geographic genetic variation within P. teyaha-
lee, but there is a considerable amount in both P. glutinosus and P.
aureolus. Indeed, some samples within both of the latter species are as
different genetically as are some comparisons of P. teyahalee and P.
jordani. The southwestern populations oi P. jordani from the Unicoi and
Nantahala Mountains are genetically very similar to P. teyahalee (Pea-
body 1978); in fact, these two species are more closely related than any
other two species of Plethodon yet examined (see Highton and Larson
1979). It is therefore not surprising that they hybridize so extensively
(Highton 1970).

Plethodon jordani is the most variable species (mean //=.21), P.
aureolus (mean H-.\2) and P. glutinosus (mean H-. 1 1) are intermediate,
while P. teyahalee (mean H=.01) is the least variable. Compared to more
northern populations of P. glutinosus and P. jordani (Highton and Mac-
Gregor 1983), these southern samples are much more variable. The P.
glutinosus, however, have slightly lower average H values than Arkansas
and Oklahoma P. glutinosus (Duncan and Highton 1979).

In light of my unpublished evidence that P. aureolus and P. jordani
hybridize at localities 46-48 on Sassafras Ridge (the only known area of
contact between the two species), the proper taxonomic relationship
between the two forms is difficult to decide. The average D of the 8
comparisons between the two forms (.3 1 ) is not very different from that of
the 28 comparisons between P. aureolus and P. glutinosus (.29) or the 20
comparisons between P. aureolus and P. teyahalee (.43), two species with
which P. aureolus is sympatric and is not known to hybridize. In the
southwestern isolates of P. jordani (in the Great Smoky Mountains,
Cowee Bald, the Nantahala Mountains, Cheoah Bald, and in the Unicoi
Mountains), P. jordani is always a high altitude species, whereas P. aureo-
lus is mostly a lower altitude form. The color pattern is very different
(except in the northern Unicoi Mountains where hybridization between
the two has occurred). None of the above mentioned populations of nearby
P. jordani has as abundant lateral and dorsal yellow, white or brassy
spotting. Plethodon aureolus is not significantly more similar genetically
to adjacent samples of Unicoi Mountain P. jordani than it is to other
populations of P. jordani throughout its range (Peabody 1978). The only
similarity between P. jordani and P. aureolus is that they are both smaller
than Appalachian populations of P. glutinosus and P. teyahalee. I there-
fore regard the interbreeding between the two on Sassafras Ridge as a case
of hybridization between species rather than intergradation between con-
specific populations. Considering the very extensive hybridization between
P. jordani and P. teyahalee throughout their contact zone in the Unicoi
Mountains, it is curious that P. teyahalee does not appear to hybridize with
the hybrid populations of P. aureolus and P. jordani on Sassafras Ridge.

16 Richard Highton

The isolating mechanisms that keep P. aureolus from interbreeding with P.
teyahalee apparently are present in the aureolus-jordani hybrids in suffi-
cient degree to prevent the usual interbreeding of P. jordani and P.
teyahalee at all 3 Sassafras Ridge sites (localities 46-48).

Since without genetic data it is extremely difficult to correctly identify
many individuals of this complex of southern Appalachian large Pleth-
odon, particularly some P. aureolus and P. glutinosus, there is a problem
in assigning individuals to species before examining the genetic data. After
a long search for a site where the two widely sympatric species, P. teyahalee
and P. aureolus, contact the parapatric species, P. glutinosus, a localilty
was discovered along Ellis Branch, near Springtown, Polk County, Ten-
nessee (locality 4), where the three forms are sympatric. Nei genetic identi-
ties between all 24 individuals from this localilty are clustered in a
UPGM A phenogram (Fig. 4). The results clearly show that the 24 animals
are separable into three groups consisting of 1 3 P. teyahalee, 5 P. aureolus
and 6 P. glutinosus. Four additional very small animals from locality 4
were also examined at some of the diagnostic loci and were identified as
P. teyahalee, but are not included in the genetic analysis because of
incomplete genetic data for several loci. Each of these samples clusters
with others of its own species (Fig. 3), and only the P. glutinosus sample
has a higher than average H value (Table 3). The alleUc data in Table 4
indicate that there is only one locus (Ldh-muscle) in which there are
fixed differences between all three species. At the other differential loci,
sometimes two of the species have identical electromorphs and some-
times there are rare electromorphs of the same kind found in one or
both of the other species. This latter pattern of variation is also present
in sympatric populations of P. glutinosus and P. kentucki (Highton and
MacGregor 1983) and could result from inheritance of the same elec-
tromorphs from their common ancestor, or occasional hybridization
between the species after complete differentiation had occurred. The
relationships in Figures 3 and 4 and the data in Tables 2, 3 and 4 are
considered strong evidence for the recognition of all three forms as dis-
tinct species. The Ldh-muscle data clearly show that there is not a single
F| hybrid between any of the three species at the Ellis Branch locality,
as does the pattern of variation at the other differential loci.

I have no explanation as to why in three cases an electromorph from
another species appears as a rare homozygote instead of in the expected
heterozygous condition [P. teyahalee #9, Alb; P. aureolus #21, Idh-2; and
P. glutinosus #16, Ldh (heart)] (see Table 4).

The Pep electromorphs of P. aureolus are faster than those of the
other two species at locality 4 and are indicated as different in Table 4. This
difference could not be consistently detected on comparison gels of sam-
ples of the three species at other localities and is therefore not regarded as a

New Species of Plethodon

17

o

-:

CVi

ro

^

in

^.

N:

• 1

1 1 «

I *

I

o

e

c«
u>
too
O
C
(U

ex
<

a

Oh
D

a>

00

CO

18

Richard Highton

in

^ .:^ 6 ^ s> ^ ^ ^ T3

73 T3 T3 -O

X) Xi X) X) X> .O X> X) X> X5 Xi X) JD X) X "O X X X) 73 X T3 X X

XX X> 73

cdX>xxiXc«X}XXX)X)XX> ei a a a a a TS-aTa-ox)

cOcdc^cdcQcde^cdcdcdcdcdcO cdcscdcacded ^XXXX

T3 73 "O T3 73 73 73 73 73 73 73 73

XXXXXX)XX>XXiX>X^ XX)XXX>X ai a (a a ei

a>(L)D(U(U(UflJ<U<U<U<U(U(U •0737373'073 OOOOO

X X X5 X X X X X X X) X) X XJ X X> nJ X X X> X X> X X X

X> X X) X) X) Xi X) X) X X) X> X X) X) X) X X) X) X) u X) U U u

cdc^cd^^^^cd^cdc^^c^ cOcOc^cdcO^ c^c^cd

U U X X

XXXXX)XX)XX)X>X)XXi cdcdcdcdcdX X^cdcdcd

XjDXU:2XU3XXXXXX43 u^ouuX uucjuu

cdcdcdcdnScdctiXcdcdcdcd XXiXXXX XJ^XXX)

*- bo

New Species of Plethodon 19

polymorphic locus in Table 2. Thus the D values between P. aureolus and
the other two species indicated in Figure 4 are slightly higher than they
would be if this Pep difference had not been detected (see Fig. 3.)

Although there is little or no evidence for hybridization between P.
teyahalee and P. glutinosus in the samples from locality 4, at locality 44
there are 4 individuals in a sample of 17 that have dorsal spots of interme-
diate coloration. In addition, there is a low frequency of P. glutinosus
electromorphs at all of the loci that differentiate the two species (Alb, Est,
Ldh (muscle), Pt-2, and Trf). This is interpreted as evidence for hybridiza-
tion between the two species at this locality. It is not surprising that P.
teyahalee and P. glutinosus hybridize in some areas and not in others, since
this same pattern occurs between P. teyahalee and P. jordani in the
southern Appalachian Mountains.

ACKNOWLEDGMENTS.— I wish to thank the many persons who
aided in the field and laboratory work, especially Ivette Aguirre, Debra
Glover, Karol Jackson, Scrap Oktay, Robert Peabody, Joshua Schwartz,
Jeffrey Streicher and Susan Fogle. Richard L. Hoffman suggested the
scientific name. I am also grateful for the financial support of the work by
the National Science Foundation (grants DEB-79-03858 and DEB-81-
17973) and the Computer Science Center of the University of Maryland.
David B. Wake made helpful comments on the manuscript.

LITERATURE CITED

Duncan, Robert, and R. Highton. 1979. Genetic relationships of the eastern large
Plethodon of the Ouachita Mountains. Copeia 1979(1): 95-1 10.

Hairston, Nelson G. 1950. Intergradation in Appalachian salamanders of the genus
Plethodon. Copeia, 1950(4): 262-273.

Highton, Richard. 1962. Revision of North American salamanders of the genus
Plethodon. Bull. Fla. State Mus. Biol. Sci. 6;235-367.

1970. Evolutionary interactions between species of North American

salamanders of the genus Plethodon. Part 1. Genetic and ecological rela-
tionships of Plethodon jordani and P. glutinosus in the southern Appa-
lachian Mountains. Evol. Biol. 4:2\\-24\.

1972. Distributional interactions among eastern North American

salamanders of the genus Plethodon. Pp. 139-188 in P.C. Holt (ed.). The
distributional history of the biota of the southern Appalachians, Part III:
Vertebrates. Res. Div. Monogr. 4, Va. Polytech. Inst. State Univ., Blacks-
burg. 306 pp.
1979. A new cryptic species of salamander of the genus Plethodon

from the southeastern United States (Amphibia: Plethodontidae). Brim-

leyana 1:31-36.

, and S. A. Henry. 1970. Evolutionary interactions between species of

North American salamanders of the genus Plethodon. Part 2. Variation in the
electrophoretic migration of plasma proteins of Plethodon Jordani, P. gluti-
nosus, and their natural hybrids. Evol. Biol. -^.-241-256.

20 Richard Highton

and A, Larson. 1979. The genetic relationships of the salamanders of

the genus Plethodon. Syst. Zool. 25;579-599.

, and J. R. MacGregor. 1983. Plethodon kentucki Mittleman: a

valid species of Cumberland Plateau woodland salamander. Herpetologica

39.189-200.
Nei, Masatoshi. 1972. Genetic distance between populations. Am. Nat. 706;283-292.
Peabody, Robert B. 1978. Electrophoretic analysis of geographic variation of

two Appalachian salamanders, Plethodon jordani and Plethodon glutinosus.

Ph.D. Dissert., Univ. of Maryland, College Park. Ill pp.
Sneath, Peter H. A., and R. R. Sokal. 1973. Numerical Taxonomy. W. H.

Freeman and Co., San Francisco. 573 pp.

Accepted 31 December 1983

Lower Wilson Creek, Caldwell County, North Carolina:
A Thermal Refugium for Reptiles?

Robert Wayne Van Devender

Department of Biology, Appalachian State University,

Boone, North Carolina 28608

AND

Paul F. Nicoletto

Department of Biology, Virginia Polytechnic Institute and

State University, Blacksburg, Virginia 24061

ABSTRACT. — Fourteen species of reptiles were found in a year-long
study of lower Wilson Creek, Caldwell County, North Carolina. Of
these, four — Anolis carolinensis, Scincella laterale, Eumeces inexpecta-
tus, and Tantilla coronata — represented significant range extensions
into the northwestern mountains of the state. Climatological studies in
the area revealed a relatively equable temperature regime, with lower
daytime highs and higher daytime and winter lows than nearby Lenoir.
A combination of exposed rock for basking, deep fissures, and warmer
night and winter temperatures is probably responsible for the con-
tinued presence of these species well outside their heretofore known
ranges. These populations may represent relicts dating from late in the
Hypsithermal Interval, approximately 7000 years ago.

INTRODUCTION

Wilson Creek and its tributaries drain most of the southeastern
slope of Grandfather Mountain in Avery, Burke and Caldwell counties,
North Carolina. These streams descend steeply from over 1220 m
(4000') to about 330 m (1080') where Wilson Creek flows into the Johns
River on its way to the Catawba River. Wilson Creek is one of the more
rugged and scenic streams of the eastern Blue Ridge front, with water-
falls and rapids in the main gorge. The discovery of several green
anoles, Anolis carolinensis Voigt, in the lower part of the Wilson Creek
gorge in August 1980 indicated that this area might also harbor other
species normally found only in lower, warmer parts of North Carolina.
We therefore set out to determine which species of reptiles occur in the
area, whether they represent natural populations, and what aspects of
the area are responsible for maintaining the reptile community.

MATERIALS AND METHODS

The study was carried out in the gorge of lower Wilson Creek
approximately 18 km (1 1 mi) west of Lenoir. The site was in the Pisgah
National Forest and included the riparian zone along Wilson Creek and
the steep, northeastern side of the gorge up to about 765 m (2510') on

Brimleyana No.9:21-32. June 1983. 21

22 Robert Wayne Van Devender and Paul F. Nicoletto

the southern and western slopes of Adams Mountain and the western
slope of Loose Mountain (USGS Collettsville 7.5' topographic map).
The site was centered near 81°44'W 35°36'N along county road 1328
three to five km north of county road 1337 and about 2.5 km north of
Brown Mountain Beach resort.

In this area, Wilson Creek cuts a gorge that drops off from about
700 m (2300') to the streambed at 365 m (1200'). The northeastern side
of the gorge, which includes the study area, consists of steep, rugged
slopes with exposed granite bedrock interdigitating with woods domi-
nated by stunted oaks {Quercus ssp.), maples {Acer ssp.), sweet gum
{Liquidambar), and Virginia pines (Pinus virginiana). Occasional small
patches of vegetation are isolated in extensive areas of bare rock. Deep
crevices and fissures are common throughout the area and abundant in
the streambed.

The site was searched for reptiles several times in August 1980 and
weekly from September 1, 1980 until March 15, 1981. Sporadic visits
were then made through August 1981. Thirty-six site visits were made,
and over ninety hours were spent searching for reptiles. During warmer
periods visual searches for active animals were supplemented by turning
rocks and logs. Winter searches were mostly visual searches of the fis-
sured granite along the creek. Species and activity were recorded when-
ever a reptile was encountered. Voucher specimens of most species were
deposited in the vertebrate collection at Appalachian State University.

In order to learn more about the local environment and to deter-
mine whether it was unique in any way, temperature and rainfall data
were obtained for the Wilson Creek area and compared with similar
data for comparable elevations at nearby Lenoir (U. S. Weather Ser-
vice). A thermograph was placed at Brown Mountain Beach resort on
September 1, 1980 and checked weekly until March 15, 1981. Data were
recorded for 186 days when the thermograph functioned properly. Rainfall
data for Wilson Creek consisted of fifteen years of data from Mortimer,
a short distance upstream from the study area (U. S. Forest Service
environmental statement for the Wilson Creek area, pers. comm., Mr.
Pat Cook, District Ranger). Lenoir was chosen for comparison because
of its proximity and because Nicoletto's experience in the area indicated
that none of the unusual species occurred there.

Since temperature, in its various guises, potentially limits reptile
distributions, daily maximum and minimum temperatures at Wilson
Creek were used separately as independent variables in linear regres-
sions against paired Lenoir data (MINITAB computer program pack-
age; Ryan et al. 1976). These regressions and long term data for Lenoir
(49 years, U. S. Weather Service) were used to estimate mean monthly
minimum temperatures at Wilson Creek. Monthly and annual rainfall

Lower Wilson Creek NC Reptiles 23

data at the two sites were compared with paired /-tests. Statistical sour-
ces were Sokal and Rohlf (1969) and Rohlf and Sokal (1969). Signifi-
cance levels of P < 0.05 and P < 0.01 are indicated throughout this
paper by one (*) and two (**) asterisks, respectively.

RESULTS

Fourteen species of reptiles were found at the study site (Table 1).
Local residents also reported that rattlesnakes, Crotalus horridus Lin-
naeus, were occasionally seen in the area. A medium-sized black snake
seen but not captured was probably a black rat snake, Elaphe obsoleta
(Say). Most species found along Wilson Creek are those to be expected
in the mountains, but four species represent significant range extensions
into this part of North Carolina of forms that are usually in lower,
warmer areas. These species are the anole; the ground skink, Scincella
laterale (Say); the southeastern five-lined skink, Eumeces inexpectatus
Taylor; and the southeastern crowned snake, Tantilla coronata Baird
and Girard.

Lower Wilson Creek is apparently an unusual microclimate in the
mountains. It is wetter and warmer than the Lenoir area, where the
unusual species do not occur. Annual precipitation averaged 136.9 cm
(53.9 in) at Wilson Creek and 125.3 cm (49.3 in) at Lenoir. In matched
data for 15 years prior to 1980, annual and monthly averages were
higher at Wilson Creek (**), with differences of 0.23 to 1.7 cm (0.09-
0.67 in).

Figure 1 shows the relationship between daily minimum tempera-
tures at Wilson Creek and Lenoir. The regression is highly significant
(**) and indicates not only that Wilson Creek is warmer but also that
the magnitude of the difference is greatest when the temperature is low-
est. At 0°C in Lenoir, Wilson Creek should be about 3.8° C. Wilson
Creek was cooler than Lenoir on only six of the 186 days (**, sign test).
Daily maximum temperatures at the sites are presented in figure 2. The
regression is highly significant (**), and the two sites are rather differ-
ent. Wilson Creek had lower maximum temperatures 169 of 178 days
(**, sign test). There is about 2°C difference between the sites at 0°C in
Lenoir, but this difference gets larger as do maximum temperatures.
Daily temperature ranges are narrower at Wilson Creek than at Lenoir.
Monthly means of minimum daily temperature at Lenoir (49 years) and
estimates for Wilson Creek are presented in figure 3. Wilson Creek is
generally warmer and does not have a month with a mean low tempera-
ture below freezing; even in winter there are numerous warmer days.
Coldest temperatures recorded during the study were -12°C (10° F) at
Wilson Creek and -15° C (5° F) at Lenoir.

24

Robert Wayne Van Devender and Paul F. Nicoletto

>. -o

Xi u

-o-S

5 6

c3 in

b != b

«-: >. lJ .J >> ij "

C 00 t-<

il

Q

< 1 < <

£

CX5 oo m

en iS 0° "^ iS oo oo

OO 00 OS

'£

^ <N -^ CM <N -^ tS

Si

o >

a> u

cg^W

u

-H ^_

(d

> >

■^ > >

> db

^i

Q

o o ,

5 jlj jl i

.0,3

1 iz 1 <

if

Tt oo

ea -o

o >»

oo X>

2-0

u

e« "eS

C O

?

8 o o^

r- m — 1 Tt -^

«o

11

\r\ so -^ OS 00

?

.

1 1 1

5 Ji

?:

oo so VO

so m o vo — so so

«n "^ o -^

u

r- vo vo

so so Os so <N so so

■^ 00 — < 00

s

m en ro

m rr> rn m Tt m r^i

Tf fn Tf f«^

o -s;

;^ S3

1

S ^

(U

3 t.

c

ir» O o

«n <N •^ r^ ro rvi m

^-t •— 4 ViH 1^

O «

1

— O tN

:«:

-O •—

-5 J3

CO (J

U rt

.s a

-. Cl,

5 <

t« ^

^ 2

V c

t .2

U t5

r. "

^_^

•und at the Wilsoi
ecimens in the col
r.

l¥3

1 i

o s g 1?
S g J ^

III

<^ o S

2 ^ S

«;3 fe o

1. Reptiles fo
voucher sp
by altimete

s

i' ^ 2 S S ^ -S
1 ^ 5 :§ § •? -^

fill

C 05 Vj C

S -a c S
-2 1 ^ 1

3

1

Si o s
X c^ t^

(2

C/5

H-H-4-

■^•^•^■^ -^

1

Lower Wilson Creek NC Reptiles

25

c

%

-

c*

x.«

^

^

.*••

X

-

••

^

II

•yv

>-

_

•\

'♦,

•

•^V'. .

-

^ X

'•*c-..

«

'

"ri'l

-

CN

• ' • fi '*•

a

1 * •^^'•.

b

**'^S%--.

-

+

•*•%

00

00

• •• h\

-

CO

•• \l . *.

II

II

* >;;

>-

c

' ^<' •

1

J 1

1 i 1 L 1

• -

o

CO

m

CM

o

CM

O

IT)

i

o

I

T3

O
^ ) C

o ^

G

m U

o !

Z 5

oe sj 03 SI 01 s

S 01- SI

Dq d W3i >I33 y 3 NOS 1 IM

I'

. o

^ oo

26

Robert Wayne Van Devender and Paul F. Nicoletto

>

v

•

_

*♦,

a;;

•

*J

•7^

•••
*. •

^

w

-

*;^aS.

•

•

X

-

•

s*

<.

•

II

*»•

•Xi

:

>-

• '

f.

•X

^V

*♦ '^.^.

*• .*>

i»

•? *

m.* *

•V «•

• *•.

X

•*. ^^»

CO

%: .

_

a

*•• jw •

6

,-■'"

♦

•**»^v

-

^

*♦. ^v

a

o

♦•X.

00

'♦.^

\

-

II

li

\\

>-

1

c

L

1 1 ! L-

-

u

G
ei
ei

CO

O

CO

m

CM

O

CM
lO

o

c

u

U i

O M

u
u

^ U

2 ^ ^

o

I

o

I

m

I

se oe S3 03 si

Dr. d W3 i

01 S 0 S- 01- SI

>l33aD NO SUM

Z ii

00

Lower Wilson Creek NC Reptiles

27

U

Z o^

-I Z

— UJ

\

\

\

\

o

I

I

,o

CO

O

CN

»o

o »n o

dwgi wnwiNiw nv3W

cd

6

c«

T

D

^_

-TD

L^

C

r^

c«

Z

"o

O

2

2

D

d

a

•o

3

k-i

g

'o

c

*c

u

S

J

T3

>.

C

;75

cd

cd

-o

(U

v^

u

O

3

«+H

bO

C/5

IS

C

o

g

c

^
A

1)

L4

C

60

o

D

s

1-1

C
0

rn

tab

c«

ti. x>

28 Robert Wayne Van Devender and Paul F. Nicoletto

DISCUSSION

Four of Wilson Creek's reptiles represent significant range exten-
sions into the northwestern mountains of North Carolina (Conant 1975;
Martof et al. 1980; pers. comm., W. M. Palmer and A. L. Braswell) as
discussed below.

Anolis carolinensis, — Anoles are common in the Coastal Plain of
North Carolina almost to Virginia and are known from a couple of
Piedmont Plateau and Mountain areas (Martof et al. 1980; pers.
comm., Palmer and Braswell). There are only a few published records
for this species in the mountains. Weller (1930) found them not
uncommon at about 457 m (1500') near Chimney Rock, Rutherford
County. Bruce (1965) found several individuals in the gorges of the
southern Blue Ridge escarpment, Transylvania County, where they
reached an altitude of 670 m (2200'). Anoles have also been reported at
elevations below 550 m (1800') in the Tennessee part of the Great
Smoky Mountains National Park (King 1939; Huheey and Stupka
1967). Another population near Townsend, Tennessee, just north of the
Park, has been studied recently (Wade and Echternacht 1980, and pers.
comm.). All montane sites for this species are associated with exposed,
fractured rock with deep fissures, usually near the head of rather steep
gorges. These conditions apparently ameHorate the cool conditions that
exclude anoles from most of the mountains. The exposed rocks allow
behavioral thermoregulation in the face of the cold, while the crevices
provide a means to escape severe cold weather, for which they seem
otherwise ill adapted (Mount 1975).

The Wilson Creek population represents a range extension of about
66 km (41 mi) north and 50 km (31 mi) east of the Rutherford County
site, the closest published record. Wilson Creek is also about 161 km
(100 mi) west of a Randolph County population in the northern Pied-
mont. The northernmost Tennessee population is near Pigeon Forge,
which is about 48 km (30 mi) south and 170 km (105 mi) west of the
Wilson Creek site. The anoles in Caldwell County are one of the most
northern populations known, and are probably the population that
experiences the most severe winter conditions encountered by the species.

Eumeces inexpectatus. — Southeastern five-lined skinks are not
found in the higher mountains of North Carolina, even though they do
extend to the north on either side of the mountains (Conant 1965; Mar-
tof et al. 1980; pers. comm.. Palmer and Braswell). As with anoles, there
are few records for this species in the mountains. Bruce (1965) reported
them at elevations up to 550 m (1800') in the gorges along the south-
eastern escarpment of the Blue Ridge Mountains, Transylvania
County. The species has been found infrequently in the Great Smoky
Mountains National Park up to about 610 m (2000') (King 1939;

Lower Wilson Creek NC Reptiles 29

Huheey and Stupka 1967). Western Piedmont records from North
Carolina are from Gaston, Randolph and Stokes counties (pers. comm.,
Palmer and Braswell). The Wilson Creek population apparently fills in
a rather large hiatus between known populations. The site is about 225
km (140 mi) east-northeast of Swain County sites, 141 km (90 mi) north-
east of Transylvania County records, 105 km (65 mi) north-northwest
of Gaston County sites, 160 km (100 mi) west of Randolph County
populations, and 140 km (85 mi) west-southwest of the Stokes County
site. The closest Tennessee population is at least 160 km (100 mi) to the
west. These lizards seemed to prefer rather xeric situations compared to
other skinks at the site, as reported by Mount (1975) in Alabama and
Minton (1972) in Indiana. Otherwise, there was little difference in the
sites where this species and Eumeces fasciatus were found.

Scincella laterale. — Ground skinks are widely distributed in south-
eastern United States except at higher elevations (Conant 1965; Mar-
tof et al. 1980; pers. comm.. Palmer and Braswell). Bruce (1965) found
this species up to 945 m (3100') in southwestern North Carolina. It has
been reported from the Great Smoky Mountains National Park up to
about 793 m (2600') (King 1939; Huheey and Stupka 1967). Weller
(1930) also found the species near Chimney Rock, Rutherford County,
North Carolina. Several general works suggest that this species is rela-
tively tolerant of dry, hot places (Mount 1975; Minton 1972), but it can
apparently also live in some cooler, higher places.

The Wilson Creek site is about 40 km (25 mi) north-northeast of
the closest record in McDowell County (pers. comm.. Palmer and
Braswell) and represents the most northern montane population yet
reported.

Tantilla coronata. — The distribution of the southeastern crowned
snake includes many areas in the Coastal Plain and Piedmont Plateau
and several areas in the mountains (Conant 1975; Martof et al. 1980;
pers. comm.. Palmer and Braswell). The highest record for North Carol-
ina is 428 m (1400') in Transylvania County (Bruce 1965). In the Great
Smoky Mountains National Park this species has been recorded only on
the Tennessee side at elevations up to 610 m (2000') (King 1939; Huheey
and Stupka 1967). The Wilson Creek specimen extends the known
range of this species in North Carolina 40 km (32 mi) north-northeast of
a Burke County site and about 80 km (50 mi) west of an Alexander
County locality. It also extends slightly the known elevational range in
North Carolina.

Co-occurrence of four species in one area outside their previously
known distributions argues fairly strongly that the populations are
native to the site and brings up questions about why they persist at the
site and when and how they arrived there. Their survival seems to be tied
to the microenvironment of Wilson Creek gorge. Exposed rocks provide

30 Robert Wayne Van Devender and Paul F. Nicoletto

abundant basking sites to overcome locally cooler daytime tempera-
tures. Warmer evening and winter temperatures provide a relatively sta-
ble environment, and deep fissures permit escape from the occasional
severe surface conditions. This combination of conditions probably
occurs in other protected montane areas that could also support popula-
tions of these and/ or other species that prefer warmer conditions.

Arrival of these species in the Wilson Creek area may be a rela-
tively recent event or a more ancient one. Recent arrival would proba-
bly be the result of immigration by populations expanding up the
Catawba and Johns rivers into the area. Since there is no other evidence
for recent range expansions of native reptiles in the eastern United
States, it is unlikely that this hypothesis is accurate. We consider these
populations to be relicts and offer the following hypothesis about their
origins.

During the last glacial maximum, about 18,000 years ago (yr BP),
essentially all of the Coastal Plain of the Carolinas was occupied by a
relatively Boreal forest of spruce and jack pines, indicative of a much
cooler climate that excluded most southern plants and many southern
animals (e.g. Watt 1979, 1980; Watt and Stuiver 1980; Flint 1971; Del-
court 1980; Wright 1976). The reptile species now characteristic of the
Coastal Plain, including the four relicts at Wilson Creek, were found
only much farther south. The warming trend that led to the retreat of
the glaciers also resulted in the immigration of southern plants and
animals into the Carolina Coastal Plain (Delcourt 1980; Webb 1981;
Watts 1979, 1980; Watts and Stuiver 1980). The warming trend reached
its climax in the Hypsithermal Interval when conditions were warmer
and drier than they are today (Webb 1981; Wright 1976; Deevey and
Flint 1957; Watts 1980; Watts and Stuiver 1980). In a few areas the
Hypsithermal lasted until about 5000 yr BP (Delcourt 1980; Watts 1979)
when the cooler, wetter climate of today was established.

By the end of the Hypsithermal, animals and plants should have
spread throughout climatically suitable areas. Reptiles are somewhat
limited by climate, so their distributions should have been more exten-
sive at the end of the Hypsithermal than they are today. The '^ooling
trend of the last 5000 to 7000 years should have resulted in local extinc-
tion of reptile populations as conditions became too severe, and in a
general contraction of species' ranges to those seen today. Special local
conditions, such as those of lower Wilson Creek or the gorges of the
Southern Blue Ridge escarpment, could forestall local extinctions of
reptiles by limiting the impact of the cooling trend (Billings and Ander-
son 1966; Anderson and Zander 1973; Bruce 1965). The result of this
protection would be isolated, relict populations in refugia well separated
from the main range of the species, as seen in the reptiles of lower Wil-
son Creek.

Lower Wilson Creek NC Reptiles 31

ACKNOWLEDGMENTS.— We are pleased to thank those whose
help was invaluable to the successful completion of this study: Messrs.
William Palmer and Alvin Braswell, North Carolina State Museum of
Natural History, gave us access to their unpublished notes on species
distributions. Dr. J. Frank Randall was supportive in the early stages of
the project. Mr. Bob Green allowed us to place the thermograph in a
protected area at Brown Mountain Beach. Mr. Pat Cook, District
Forest Ranger, supplied us with weather data for Wilson Creek. Deborah
Nicoletto, assistant city planner, City of Lenoir, provided weather data
for Lenoir. Computer time was donated by the Appalachian State Uni-
versity Computing Center. Janice Ashley typed the final draft of the
manuscript. An anonmyous reviewer provided helpful criticisms of the
manuscript.

LITERATURE CITED
Anderson, Lewis E., and R. H. Zander. 1973. The mosses of the southern

Blue Ridge Province and their phytogeographic relationship. J. Elisha

Mitchell Sci. Soc. 59(1-2): 15-60.
Billings, W. D., and L. E. Anderson. 1966. Some microclimatic characteristics

of habitats of endemic and disjunct bryophytes in the Southern Blue Ridge.

The Bryologist 69:76-95,
Bruce, Richard C. 1965. The distribution of amphibians and reptiles on the

southeastern escarpment of the Blue Ridge Mountains and adjacent pied-
mont. J. Elisha Mitchell Sci. Soc. 57(1): 19-24.
Conant, Roger. 1975. A Field Guide to Reptiles and Amphibians of Eastern and

Central North America. 2nd edition. Houghton Mifflin Co., Boston. 429

pp.
Delcourt, Paul A. 1980. Goshen Springs: Late Quaternary vegetation record for

southern Alabama. Ecology <57(2):37 1-386.
Deevey, E. S., Jr., and R. F. Flint. 1957. Postglacial hypsithermal interval.

Science 725:182-184.
Flint, Richard Foster. 1971. Glacial and Quaternary geology. John Wiley and

Sons, Inc., New York, xii + 892 pp.
Huheey, James E., and Arthur Stupka. 1967. Amphibians and Reptiles of Great

Smoky Mountains National Park. U. Tenn. Press, Knoxville. xii + 98 pp.
King, Willis. 1939. A Survey of the Herpetofauna of Great Smoky Mountains

National Park. Am. Midi. Nat. 27(3):53 1-582.
Martof, Bernard S., William M. Palmer, Joseph R. Bailey, Julian R. Harrison,

III and Jack Dermid. 1980. Amplibians and Reptiles of the Carolinas and

Virginia. U. North Carolina Press, Chapel Hill. 264 pp.
Minton, Sherman A., Jr. 1972. Amphibians and Reptiles of Indiana. Indiana

Acad. Sci. Monogr. No. 3. 346 pp.
Mount, Robert H. 1975. The Reptiles and Amphibians of Alabama. Auburn

Univ. Agric. Exp. Stn., Auburn. 347 pp.
Rohlf, F. James, and Robert R. Sokal. 1969. Statistical Tables. W. H. Freeman

and Co., San Francisco, xi + 253 pp.

32 Robert Wayne Van Devender and Paul F. Nicoletto

Ryan, Thomas A., Brian L. Joiner and Barbara F. Ryan. 1976. MINITAB Stu-
dent Handbook. Duxbury Press, North Scituate, MA. ix + 341 pp.

Sokal, Robert R., and F. James Rohlf. 1969. Biometry. W. H. Freeman and
Co., San Francisco, xxi + 776 pp.

Wade, Julia, and A. C. Echternacht. 1981. A comparative study of reproductive
cycles in two populations of Anolis carolinensis. Abstract. ASB (Assoc.
Southeast. Biol.) Bull. 25(2):96-97.

Watts, W. A. 1979. Late Quaternary vegetation of central Appalachia and the
New Jersey Coastal Plains. Ecol. Monogr. -^9(4): 427-469

1980. Late Quaternary vegetation history at White Pond on the inner

Coastal Plain of South Carolina. Quat. Res. 75:187-199.

, and M. Stuiver. 1980. Late Wisconsin climate of northern Florida

and the origin of species-rich deciduous forests. Science 270:325-327.
Webb, Thompson, IIL 1981. The past 11,000 years of vegetational change in

eastern North America. BioScience i7(7):501-506.
Weller, W. H. 1930. Records of some reptiles and amphibians from Chimney

Rock Camp, Chimney Rock, N. C. and vicinity. Proc. Junior Soc. Nat. Sci

7(8-9):(5 l-54)(pages unnumbered).
Wright, H. E., Jr. 1976. The dynamic nature of Holocene vegetation, a problem

in paleoclimatology, biogeography, and stratigraphic nomenclature. Quat.

Res. 5:581-596.

Accepted 15 April 1983

Comparative Food Studies of Yellowfin Tuna, Thunnus

albacares, and Blackfin Tuna, Thunnus atlanticus

(Pisces: Scombridae) from the Southeastern

and Gulf Coasts of the United States

Charles S. Manooch, III

AND

Diane L. Mason

National Marine Fisheries Service,

Southeast Fisheries Center, Beaufort Laboratory,

Beaufort, North Carolina 28516-9722

ABSTRACT. — Two hundred and six yellowfin tuna, Thunnus alba-
cares, and 98 blackfin tuna, T atlanticus, were sampled from sport
fisheries in the South Atlantic and Gulf of Mexico, April 1980 to July
1982. Stomach contents were analyzed by frequency of occurrence,
number of food items, and volume. Invertebrates (85%) and fish (77%)
occurred in the diet of yellowfin relatively equally. Major invertebrates
by frequency of occurrence were cephalopods (62%) and crustaceans
(52%). Fishes were represented primarily by the families Scombridae
(12.2%), Balistidae (11.2%), and Syngnathidae (8.2%). Yellowfin also
ingested floating materials such as plastic, feathers, seagrasses, and
balls of tar. Invertebrates occurred in 82% of the blackfin stomachs
with food, and represented 75% and 31% of the foods by number and
volume, respectively. Fish were found in 67% of the stomachs and con-
stituted 26% and 68% of the food number and volume, respectively.
The most frequently occurring invertebrates were crustaceans (67.4%)
and cephalopods (36.0%). Fishes were represented primarily by the
families Balistidae (10.1%), Trichiuridae (5.6%), and Carangidae (4.5%).
Blackfin also consumed floating materials, such as plastic and sea-
grasses. Statistical comparisons of the diets of the two species indicated
no significant correlation. Overall, their diets appear to reflect those of
fast, aggressive predators, and also of fish that use their gill apparatus
to strain small, near-surface items from the water.

INTRODUCTION

The family Scombridae includes many species of pelagic fish that
are very important to the world's fisheries. Some, such as the mackerels
Scomberomorus spp. and Scomber spp., are primarily coastal, migrat-
ing north in spring and summer and south in fall and winter. Others,
including members of the genus Thunnus, are usually much larger than
the mackerels and are reputed for their more complex, often transoce-
anic migrations.

Two species of Thunnus, the yellowfin tuna, T. albacares, and the
blackfin tuna, T. atlanticus, are highly esteemed food and sport fishes
whose distributions include the southeastern and Gulf coasts of the Uni-

Brimleyana No.9:33-52. June 1983. 33

34 Charles S. Manooch, III and Diane L. Mason

ted States. The yellowfin is the largest and more prized of the two,
attaining a weight of at least 176 kg (compared with 19 kg for blackfin).

On the whole, tuna landings in the western Atlantic are sporadic
and are much smaller than those made by the large-scale, international
hook and line and seine tuna fisheries that operate in the eastern Atlan-
tic and Pacific. The total United States commercial landings of all tunas
was 341,149,000 pounds in 1981, 326,860,000 pounds from the Pacific
and 14,289,000 from the Atlantic. Only 131,000 pounds were landed in
the South Atlantic Region — North Carolina, South Carolina, Georgia,
and the east coast of Florida (D. S. Fitzsgibbon, pers. comm.). Of the
South Atlantic total, only 5,000 pounds were identified as yellowfin
tuna, and none as blackfin, although the 55,000 pounds of unclassified
tunas undoubtedly included blackfin. Recreational catches of yellowfin
and blackfin tunas tend to be greater than the commercial catches for
the southeastern United States. In North Carolina, for instance, anglers
fishing from charter boats in 1978 caught approximately 151,000 pounds
of yellowfin tuna and 38,000 pounds of blackfin tuna (Manooch et al.,
1981). No information is available for 1981.

Considering the disproportionately large commercial catch of tunas
in the Pacific, it is not surprising that many publications pertaining to
life histories, population dynamics and exploitation have resulted from
research on species in that region. Relatively few studies have been con-
ducted on Atlantic stocks. Dragovich (1969) in his review of food stud-
ies on Atlantic tunas mentioned that the papers he read emphasized
the need for additional research on the foods and feeding habits of
Atlantic stocks. The limited information available from the western
Atlantic usually resulted from fish collected aboard scientific vessels
that did not operate along the southeastern or Gulf coasts of the United
States, or that operated well offshore of the normal sport fishing
grounds (Dragovich 1969, 1970).

To obtain more data pertinent to the management of pelagic
stocks, studies were initiated on oceanic species important to fisheries
along the southeastern and Gulf coasts of the United States. Our study
is the result of a cooperative effort that included the Oceanic Pelagic
Program, SEFC, Miami Laboratory, and the Bioprofiles Task, SEFC,
Panama City Laboratory. The objectives were to 1) identify the food
habits of yellowfin and blackfin tunas; 2) compare the diets of the spe-
cies collected from the same geographic area during the same period of
time; and 3) determine if changes in the diets occur for different sizes of
fish.

METHODS

Of the 206 yellowfin and 98 blackfin stomachs examined, 169 and
55, respectively, were from fish landed at Oregon Inlet or Hatteras,

Foods of Yellowfin and Blackfin Tuna 35

North Carolina during the spring, summer and fall of 1980, 1981 and
1982. A few additional samples, indicated in parentheses as yellowfin
and then blackfin, were obtained from locations along the southeast
Atlantic and Gulf of Mexico coasts: South Carolina (31,8), Georgia
(3,1), east coast of Florida (0,2), northwest Florida (3,1), Mississippi-
Louisiana (0,6), and south Texas (0,25).

Samplers at all locations apportioned their efforts to coincide with
local charter boat activities, primarily April through October. Port
samplers met boats at the docks as a day's catch was being unloaded.
Most fishermen either wanted to save their fish whole for mounting, or
to have them filleted and packed on ice or frozen upon returning to the
dock. Data were obtained only from the latter group, either in exchange
for cleaning the fish, or from fish cleaners who worked at local markets.
Fish were measured to the nearest millimeter (FL) and weighed to the
nearest tenth of a kilogram. Stomachs and gonads were placed in
labeled cloth bags or cheese cloth and preserved in 10% formaUn.

In the laboratory, stomach contents were identified to the lowest
possible taxon and were enumerated, thus providing the relative number
of each food type in the stomachs. Frequency of occurrence of materials
was determined by counting every stomach that contained at least one
specimen or part of a specific item (taxon). Empty stomachs were
excluded. The volume of each taxon was obtained by water displace-
ment and was later converted to weight by a linear regression equation.
Larval and juvenile fish in the stomachs were identified after they had
been cleared and stained following the methods discussed by Dingerkus
and Uhler (1977) and Taylor and Van Dyke (1978). Crustaceans were
identified by Steven G. Morgan and Joseph W. Goy, Duke University
Marine Laboratory, Beaufort, North Carolina. Parasites, encountered
only occasionally, were separated from food items, counted, identified
and preserved. A stomach containing only parasites was considered
empty.

All data were analyzed as percent frequency of occurrence, percent
of total number, and percent of food volume. Once frequencies, volumes
and numbers of the various foods were obtained, an index of relative
importance (IRI) was used to estimate the contribution of major food
groups to the diet (Pinkas et al., 1971). The index was calculated as: IRI
= (N + V) F, where N = numerical percentage of a food, V = its volumet-
ric percentage, and F = its percentage frequency of occurrence.

The Spearman rank correlation (r^) was used to evaluate differences
in diets of the two species based on IRI values of foods from fish col-
lected in the same geographic area and over approximately the same
period of time. Two different equations may be used. One, where there
are no ties (rankings are equal for two or more food categories), and the

36 Charles S. Manooch, III and Diane L. Mason

other where ties do occur. The equation for tied food categories (Fritz
1974) was used:

"^ - where Sx^ = — STx; Sy2 = zTy;

2 Sx2 Xy^ N -^ N

J - ^^ ' ^ ; N = numbers of ranks; d = difference between ranks; T =

N
correlation factor for ties and t = number of observations tied at a given
rank. Pearson and Kendall's Tau B Correlation Coefficients, in addition
to the Spearman rank, were also derived to evaluate differences in the
diets.

RESULTS AND DISCUSSION

Composition of Stomach Contents

Stomach contents of both species could be grouped into four prin-
cipal categories: fish, cephalopods, crustaceans and miscellaneous non-
food items (Tables 1, 2; Fig. 1). Major representatives of each group will
be discussed below under separate headings and will also be analyzed
later to identify differences in diets related to the species of predator and
its size. A graphic presentation of the overall contribution of selected
foods to the diet (IRI plots) is presented in Figure 2.

Fish. — Fishes occurred in 77% of yellowfin and 67% of blackfin stom-
achs that contained food (Tables 1, 2; Fig. 2) and consisted primarily of
older larvae and juveniles often associated with floating Sargassum. In
all, 23 famiHes were identified. Adult exocoetids, scombrids and syngna-
thids were found occasionally in yellowfin, as were syngnathids, serran-
ids, sciaenids and stromateids in blackfin. For all life stages, fish that
occurred most frequently in yellowfin tuna were Scombridae (12.2%),
Balistidae (11.2%), Syngnathidae (8.2%), Diodontidae (5.1%) and Exo-
coetidae (4.6%). Fifty-three percent of stomachs with food contained
unidentifiable fish remains. Fish that occurred most often in blackfin
tuna stomachs were Balistidae (10.1%), Trichiuridae (5.6%), Carangidae
(4.5%) and Syngnathidae (4.5%). Unidentifiable fishes were found in
44.9% of the stomachs containing food.

Cephalopods. — Cephalopods constituted almost all the molluscan
food of both species. One exception was unidentifiable mollusk tissue,
possibly cephalopod, from a yellowfin captured in the Gulf of Mexico.
Two groups were represented: Teuthidida and Octopodida. Teuthoids
(squids) were the most important by frequency of occurrence and by
volume: 50.5% and 41.0% for yellowfin, 31.5% and 21.5% for blackfin.
By comparison, octopodids, represented by the paper nautilus, Argo-
nauta argo, appeared in only 7.7% of the yellowfin tuna and 3.4% of the

Foods of Yellowfin and Blackfin Tuna

37

y^

CO

^

3

o

^ ^

CO

^z

z

<

<

-J

LU

-J

O^

LU

<^.

O

H 00

CO

CO^

S

CC

O

CO

<

3

z

O

3

>

H

Z

^

u.

)^

T3

o

t/3
C/5

<

Si

rv

-J

X

CD

C
C

o

ON

oo

-o

c

c3
C

o

so

ON

o
o

<

z

o

C2

H

y^

T3

^1^

\^

Z

3

o>

CO

U-

O

i2

CO

z

3

o

LU

Z

5

O

-1

-J

V5

c

c
o

<

<

LU

o

LU

-J

>-

a,

3

o

<

-J
LU

o

O
u

o

CO

CO

■J?

3

2

OC

—

o

38

Charles S. Manooch, III and Diane L. Mason

o

II

On

z

^

_C

. ».

C/3

"^

to

00

C4

O

w-T

(J

1 1

o

Z

o

^-^

IS

(/3 II

C/5 >»
• - 3

c o

a> >

S3 o

II g

u ^

C) (U

fc: ^ ■-►

<U (U *^

^ y <»

13 c 3

ON (U II

^

U fN

g

e oo

3

o

>.

-c "2

JO

O C

,

KJ c«

c

e -

J^

o 28

S^

*- ON

a>

C/5 —

D,

«

^

3

c«

H

Ov* O O wS «0 Tf O U^ '^' [_ t_

^ -: -: Dc: D!i oi '^ ^. oi -: »: a: »:

<NOvOOOOOOOOOO«N«NvO<sO— <

vd sd -^ «o uS «o — " — ^ (^ -H* -J r>; r^' o o — '
TtooooTtTtr4«ri r^ (^(^«__

'*.'". ^ ^ ^. <^ oi Oi '^^ ffJ ^ '^. ^

N0»OTtTtOO«O«OU->

o o rsi so so" o o o <N

vdHwSoO-^-^OrnoOodcxJ — --' — OO-^'r^cSoOOO

o m — ' —
U-) o — '

OsrN — SO— <— <vOvCi<N<N<N-^— .OOVO — -^-^ — -^

c

T3

3 IS

o S

X ha

^-^

Pu

C/5

G .S
^ 3 T3

JD X) JJ

cd cd ^
G G o

^^.^

D D ,«J

5-1

cd
. T3

;« -5
05 C
3 cd
Cl ^

i£
I-

Ph

ed "^ J3

"o 'S
•2 'c ^

(U

Cd
T3

C X)
cd cd

5d hs
U ^

3 D

>3 a 2

5 ? c

lU V. o ^ t^

H ^ ^ ^ ^

c g. >. t: >.

w ^ E o E

^ ^ fc ^ PLh

3 o

Foods of Yellowfin and Blackfin Tuna 39

•nooooo«o«r^oovnroooooooooas>o'rj-<N<No ■^•/^»ooorsivooo

r4 rn rn O wS wS -^* — sd od O rn ---* (XJ sd O O C> O O •^* --" ro rn — -^' ro O ro r-'

•i^<Nrviooo«ou->tNO«/^'>o«/-i»r>u->oo^no«o — vo«n
o H --*—■--" -h' o o -^ --^ o '<i- c> --" <N -^ — —■—«■ c> «o Tt o

<N

q

q

CsJ

u-j

<N

-H

00

^*

'-'

r4

d

uS

Ill gil.S5||| .SJim-g i SS3|5S

^ I « o -C t3 X = .2 .2 .2 g--S S S -2 2 .S -S ^ d. 2i S c 0.0 .2 | 9
!:«SS-<«|Qt«SSS32|OSHs2Qg| s'^.^tSuS

c« .ii

^ • s 3 >< -"^ J:; • - t: -"^ 3 3 3 ^ ^ -^ • - .^ -"^ ."2 J • - ^ :^ ii 3 ^ 3 "H ^ c

40

Charles S. Manooch, III and Diane L. Mason

>

P O — On ON p^ -- 0^ — - Q^ Tt 0^ P^ ^ P^ P^ rn — . <N (N p^ P^ j^ —• Q^ p^

Tf -^t Tt ^ vq "^ r- 00 ov

"oooo«n>r!H»r5o'^'Pvdooo'--'oo>odood— ^

ir^ m en QO oo
Tt — — r^ r^

^t^fNrs»r)0«o«r)sc>(^iom\0»n«r>p>oo

^c5oo— 'oonOCJOOO-^OOOnUhC)
Tt rN m m <N r- '^

r^ so

rsj Tt -^ <X3 c» — »r» p^ ir^ Q/ -. j^ p^ -.--;-< I-. -. r- w-> -. -;

r- r- ^ ^ ^ ^ t-

Oi '^. qC ^.

O O [— I w-i [_ O

SO (N fN ^ SO 0> 00

m vo vo

o

O Ov OO On fN

»nr-r-OOrsir-»r)NO»nir>u-imoN«nmmor4<Nsow->inm«nvO

or-r--<N<N<Nr-o^o

Ov u-» »o rN <N Tj-

ON — ' — o o

— «o — — <Nm — ooTtTto«orn— ' — — u->

ON <N <N fN OO

cd

cx

"i

a>

•2

g.

cd

(U

£

g

^

Sk

o

e

cd
1^

■•3
"3

■■f

O

cd

T3
O

Ou
O

a
B
o

.2

(-1

i

ex
o

>
>-<
cd

i

1

Cd

o

cd

o

E

cd
O

cd
-a
O

cd

cd
>

XJ

cd

.5
c
B

"cd

1/3

cd

cd

'C

cd

c

>>

i

1

cd
>

d

1

1
1

.2
c

B
ex
<u
oC

X

5

HO

<3

<

3

c

Cn

"cd

-2

t«

<

"H

Q

c

"2

<U

S

5

g

1

T3

Ui

t_

o

u

u

Vi

"O

£

T**

Ui

ui

o

Ul

T3

"O

o

,^

.^

.^

hM

3

•o

4>

T3

^

B

3

c«

.-a

«u

'S

O

2 <L> (U
OT3 TJ

X)

3

T3

'S

"c

3

O U VJ CO

o

X)

'c

"c

OO

X

Oh

PO

D 6n CO

OO

C/5

6

D 3

C/5

3^3

on

Foods of Yellowfin and Blackfin Tuna

41

H H H H

-. -: f^^. ttiocja:-: a:c^a:a:p^D!:-

(N — rN-^roO\Ov«ri»nOO

OOOUi-^vOOO<NU^rnvOfSO

<NrooNf<iO<NOOT}-(Nrsl'-HrslO

— «■ — od — ' O — * O O O <N

— — (N 00

^oo^rsi— 'cJooooo^oo

m-^oor^mi^oo'^Ttr-<N»r>m
»r> u-i in ro ro m ro

•no«n»r»in— ' — >no«r>«nom»o>r)
O <N — ■

or>-sor--— '<N<N — — ooo — ^o

vo»nvoo«'^»^«noovn»om«n
<N — ." — ■ rsi — — ■ — o o o o

C C
-2 c

^ 6

C/3 C/5

O o

2 ^

o JS

cd ci-

.2 ^

Q Q

1 =

O u.

2 • (i ^

Cv D, o OJ c« c«
r*i cr> O o ei JZ

6 S c

O Q.

.§2

•S I E o

"" '^ r5

(U

c« CO

ex
JO -2 ^ O

I £

g c^ N f-^

a, cd

^ IS

ex

W)

w

^

o

J=

cd

^

S

GO

Cd

X)

_o

o

Cj

vj

lyj

on

^

cd

cd

.-•

^

ex

ex

cx

u>

u>

(U

cd

cd

3

^

JJ

5 u u

42

Charles S. Manooch, III and Diane L. Mason

oo

c

On

u

^

C

^

Vi

II

*•>

cd

Z.

O

^

o

(J

o

fS

s

u

•-

<4— 1

II

o

(yj

a

-§

(4-1
0

cd

s

3

CAl

C

-o

II

W

Z

'S

. .<

D

^

cr

W5

cd

Si

li

(U

<4-l

g

^

^

c

_3

3
O

8

"o

V3

u

>

g.

0

II

4_>

Vt-I

UU

O

^

Si

^

II

>

Z

^«-^

^

cd

(J

AC

c

e

r-;

c

3

^M

j;:

0

-^

^

u

•n

CJ

0

ri

iS

^M

II

2 g 6
033

i2 ST <=>

■•-> *> •-».

4> ^ II

c " >

ed 0^ C

C/5 cd xi

cd
H

so — , -^ r- r^

SO Ov On — -^

oo(»vooooooooa>r---rvjoo«noo«o 000000
or4rn»r^w-^or^ir4wSwSrKrsi-— "ror^TtNd(»orsir4wSv-ir^r-^

u-j fs f*^ OS «/^ Tt rvj <N — ^

>/Sodrniric5«nOOO

— • r- \o — — — — ON « — Tt Tt r-. r- — —

0—.*—;oOC><NO — 0000000

sO<Nr--sOsOO<N<N — — OsOO — — -*rOO<N — TfTfooOO— *-^
OOOsr^sosO -H— . (s| — —

rf OS OS Tt fN — <N <N — -- 1/-^ Tt — -^

r>-" Tt r-^ ro <N -^' rsi (N -^* -^ Tf" ro — ' -^

^ "^ *^. ^ ^. "^. "^. *~i ^.
-^ Tt <N -^ — ' ro ro <N rj

so Tt

(^^^^Tt(N'-^'-^r^m(N(N^^

Cd

ex cd

3 IS

cd

s I
^ 0 d

OQ t: cAi

ass

r- iS «-

s ^s 1) ** eg

2 Sc/5 ^U

^ Cd

. cd
D,T3

1-2 --^g-^
;g'C S S 6 . ^

^ "^ [£ ^ £ ^

•2

K cd

*• cd

^ D. _

I e § 6

§ ^ "^ "a

Foods of Yellowfin and Blackfin Tuna 43

m <N ^ <N r- r- [—

ooocJooo— *oooo Tfsdo^ooodo — -^ooo«oorsio<No

1^ vo rn ro rn

^ ^ rsj —< — —< — -^ <N Tt Tt —^ 'p O '"t ''^ "^. "^. '^. '^. o*^ f^. ^ ^ ^. ^ "^ ^ '^. "^
uS >r^ <N — —*—' — o <N ro ro -- r^ sO — ^ — ro r<i r^ rn Tt r-j —* (N — «' w-i — Tt fN ■^'

-^ oomrn somro «^ — -^

T3
O

a,
IS

"^ .. != ^ c« «J C

■■25-5 s S -2& 1 .-S-l^^ &^ = g 1151^1 .

44

Charles S. Manooch, III and Diane L. Mason

'~'. Pi PC ^. '^. a!, ^. ^. "-'. ^. a^ ^. *^. ^. ^. '^. oC '^Pp^^^pd

Ttmo-^w-^tNvoooooo — <Noor^ow->rsi o^w->ovo

r^i r4 «— * w-i <N rsi Tt «/S \d od -^" wS rsi Tt ---' <N — *

Tt — — <N — —

r-- -^ ?N Tf <N

w-i (N <N ro rN

Tt — — rsj — —

Tt — r4 r«-i <N

bo C

§■1
= I

U U a^ Co Co

ll

■ (J

ex

.5 ^
c B

60

.1:: <u
c 2 o

1) is N

o

.- J= Xi

""^ S c g

«<-i -rr 'rj

X) tj ^

c/5 c/:

C/5

It

^ ?^ ^

-§ .-2 -I .. ..

c o t: ^ =^ ^

DdQCU ^ 3 3 g

I £

CO N

o

c -^

Foods of Yellowfin and Blackfin Tuna

45

<D

CO

■o

<

z

0) C

z>

CO £-

3 CO

J

z

i

^

E

^

o

^

o

o

CO

<

1 1 f

-"'"'^ (

D (]> M 0)
0 CO "D CO

-J
m

c

■O T3 O -O

taceans

\

m ^ ^ ^

1 si'^

03

C

C

(0

CO CO w

o

c^

3

^

s

O to

-

o

00

>-
o

z

C
03

o,^^

o

LU

C

(O

3

; "^

To
a £

_

o

o

UJ

^

^

0) eo----'

cc

"u

o iZ

1 1 1

1 1 1

1

o

CM

O

ooooooooo

OOCD-^CVJ CSi^COOO

a39i^nN % 3i^moA %

"§

<2

-o

0)

<

z

aj

(0 r

T3

3

1

iE^-"""''^

h-

C '■ TT

a

CO

Z

oc

U-

C/5

O

-J

1

^

\

0) (D 0) (0 0)

"H-

"^^^ CO 08 CO "D CO

^^^^■D -o T3 O-O

-J

^

(0

c

LU

>

D^

(0

CO

8

♦-

>,co i: ^

(0

3

0) CO CO

ti

»-

o

o

D,

(0

£

"O /

>

o X
a X

~

o

CO

o

>

CO ^

Q) <0-^

O iZ

o

CO

o
-^

o

liJ

O
LU
OC

J3
o

lilt

1 1

C\J

^

X
T3

ooooooooo

C

OOCD^CVJ CVJ'^CDCO

CN

aaaiAinN % aiAiniOA %

tz

46 Charles S. Manooch, III and Diane L. Mason

blackfin tuna. Percent volumes of these mollusks were less than 2% for
both predators, and whereas over 430 squid were consumed by the
tunas, less than 30 paper nautilus were eaten. At least three genera of
squids were ingested: Loligo, Sepioteuthis, and Illex. Generic identifica-
tions were obtained by comparing saved, pooled samples with reference
collection specimens and therefore do not appear in the tables.

Crustaceans. — Crustaceans, important foods of both species and
second only to fish in overall frequency of occurrence, were identified in
52% of the yellowfin and in 67.4% of the blackfin. The majority were
immature stages (larvae, megalopa and glaucothoe). Due to the small
sizes of the animals, the relative percentages of the total food volume —
5.9% for yellowfin and 8.4% for blackfin — were comparatively small.
Major taxa in the diet of yellowfin by frequency of occurrence were
Raninidae (27.5%), Penaeidae (12.2%), Stomatopoda (7.7%), Portuni-
dae (7.1%), and Dromiidae (6.1%). For blackfin tuna, the most fre-
quently encountered were Stomatopoda (34.8%), Diogeninae (16.9%),
Raninidae (15.7%), Penaeidae (14.6%), and Dromiidae (12.4%). In all,
over 5,000 individuals were enumerated, and on one occasion a single
predator contained hundreds of these small, mesopelagic invertebrates.

Our findings of the overall food habits seem to agree closely with
those of Dragovich (1970), who described fish, cephalopods and crusta-
ceans as the major foods of yellowfin and skipjack, Katsuwonus pela-
mis, tunas in the Atlantic. He also mentioned that larval and juvenile
stages were prevalent for ingested fishes and macrozooplanktonic crus-
taceans.

Miscellaneous. — The very nature of tuna feeding, near-surface strain-
ing as well as actively pursuing and capturing larger animals, results in a
variety of items being consumed that are probably ingested by accident
along with natural foods. Yellowfin tuna had the most diverse assemb-
lage of non-food items (31.6% frequency): plants {Sargassum, Zostera,
Thalassia and Spartina), feathers, globs of tar, and plastic. Miscellane-
ous items occurred in only 15.7% of the blackfin, represented by Sargas-
sum, Zostera and plastic. Sargassum was found in 26.5% of the stom-
achs with food, and usually occurred in tunas captured off North
Carolina. This percentage is similar to the 37.8% reported for Sargas-
sum removed from the digestive tracts of skipjack tuna captured earlier
from approximately the same geographical area (Batts 1972).

Other studies also revealed a dominance of fish, squid and crusta-
ceans in tuna diets for the Atlantic and Pacific. Reintjes and King
(1953) investigated the food habits of 1,097 yellowfin from the Central
Pacific and found that fish occurred in 70.4% of the stomachs; squid in
55.4%; and crustaceans (mostly immature, pelagic stages) in 66.9%.
Alverson (1963) found fish, squid and crustaceans occurring in 53.8%,

Foods of Yellowfin and Blackfin Tuna 47

23.9% and 76.1%, respectively, of the yellowfin he examined from the
Pacific. Similar occurrences were reported for yellowfin from the Atlan-
tic (Dragovich 1970), and for skipjack tuna (Alverson 1963; Nakamura
1965; Batts 1972); bluefin tuna, T. thynnus (Pinkus et al., 1971); and
albacore, T. alalunga (Pinkas et al., 1971) from the Pacific.
Comparative Diets

Since temporal and spatial variations in the diets were so great
(data collected over a period of three years, and from several widely
different geographical locations), we believed that only by analyzing
small, discrete samples could we detect important differences in them.
To achieve this, we used only stomach contents of the two species col-
lected together off Oregon Inlet on 10 different days from May through
September 1981 (Table 3).

Index of Relative Importance. — Indexes of Relative Importance
(IRI), which present the combined contributions of volume, frequency
of occurrence, and numbers of each food item to the diet (Table 3),
showed that, surprisingly, invertebrates were very important foods for
both species. The first five categories (ranks) for yellowfin were Teuthid-
ida (squids), unidentifiable fish, Raninidae, Scombridae, and unidentifi-
able crustaceans. For blackfin they were unidentifiable fish, Teuthidida,
Raninidae, Stomatopoda, and unidentifiable crustaceans. Obvious dif-
ferences were more clupeids and unidentifiable diogenid crabs in black-
fin, and more scombrids and squids in yellowfin. Other items were also
different, but their respective IRI values were relatively small (i.e., exo-
coetids for yellowfin = 9.7, for blackfin = 0.0).

Correlation Coefficients. — Data from Table 3, ranked by IRI values,
were used to obtain quantitative comparisons of local food habits of the
two species. Three different measures were used: Spearman Rank Corre-
lation Coefficient (Fritz 1974); Kendall Rank Correlation Coefficient
(Bray and Ebeling 1975); and Pearson Product-moment Correlation
Coefficient (Goodall 1973). The first two require no assumption of
normality with regard to the distribution of the two predator species,
whereas the latter does. Cailliet and Barry (1978), who compared the
three methods of analyzing diets that have different distributions of prey
items, found that Spearman and Kendall correlation coefficients are
somewhat unpredictable when there are 1) a large number of ties, 2) a
considerable nonoverlap of prey items, and 3) high prey richness and
evenness (i.e., diversity). They felt that the Pearson method was best.
Although our data have a fairly low richness and evenness, there are
relatively few ties (2 for yellowfin, 3 for blackfin) and there is a fairly
good overlap in the diets. For these reasons all three methods of meas-
uring diet similarity are probably appropriate. Qualitatively, both spe-
cies feed extensively on epipelagic and mesopelagic fishes and inverte-

48

Charles S. Manooch, III and Diane L. Mason

^

o^

u

^

"w

a

>,

3

*ri

C

^

>.

B

x>

o

o

1/3

Vi

S.

•<-•

c

1'

o
o

;^

J3

^

o

c«

. .N

B

^

o

c

c«

*©

g'

3)

>^

c

4-<

C

'^

u

2

2

"§

II

a

b

^

^

13

oo

>

On

tfj

60 C
O 60

o

O

yn p

"^ p

>n

On'

u-i

a:

so

^2

pin>no«/^pp>nin<n>nomu-iioo
0>nw-iod>/S«nsd<NfNi«rirviod<N>nw-ir«^
(N rs — rN(N — — — _____ fN)

po\pooppvop«nr~-sc>»opso^o^oo
oosD m m fN o — ov

o «n >n o >n
CT^ v-j »n r-' <ri

U-) O CM

ON

p

Ov o

o

ro O

;2;

«n

<N d

d

<N d

p ^

od d

<N «n (N (N

vS d — — <

o o
d d

— o o
d d d

d — '

m (N

OS 0^ <N rs)

rsi «^ <N r^

m — Tt

<n d

(> Tt d d

d> d> d> d>

^ rn —

£ >^ =

> 5

"^ S

c iS

I -

" 3

2> " §

^ .S o

s :2 >

fc IS ^

m

1^

r~-

r-

<^

—

ON

so

OS

ON

ON

ON

r^

ON

ON

ON

so

so

ON

in

>n

u-i

iri

rvi

(N

od

(N

<N

fN

<n

(N

(N

<N

od

od

(N

p

>n

p

p

«n

O

o

«n

p

P

p

p

p

p

«n

p

p

m

«n

>n

m

«n

p

P

o

o

p

«n

od

■^'

o^

r-4

<>

d

so

•^

od

r-^

o^

<n

Tt

""

f^

Tt

Tt

■^'

Tt"

ON

(N

<N

■^"

sd

Tt

„_

o

r^

rn

^

't

rs|

o

^.

oo

ON

r-

SO

p

o

_

SO

p

p

p

p

■Tf;

«n

OO

o

'*.

''t

o

(N

d

o^

od
«n

d

"^

d

ON

od

so
1^

«n

1

d

"—

<N

d

d

d

d

d

""

csi
«n

^

od

d

6

b.O <N

^

o

On on

r- >n •rf oo Ov «^

d

— > r^

O rn -^ d> -^ 'Zi

«n r<-) (N t^ so

d r--' d d —

Tt sd

r- (N Tt m

r-- >n r*^ >n On — •

d d <N d rn On

t^ (N On rf ON r<-i

«5 i2

T3 :o

"S O C cJ
3 o c .2
U W OO Du

^ o

ii i« w -2 -5 ^

2 -^ O 3 w •-

Uc/5c/5paQHOP

CO i:

O p ro p -St ON

d d d d d —

r^^ r^ ON On Tt (N

d d rn d rn r^

(N rsl OO -t <N

(N <N r-' Tt" rvj

•a 2

O

^ :^

ex -o
B •£
< O

a 'c:

= 5 Ci,

■z ."H

cd flj C
>^ ° S

C/5 Q oi

q5

Foods of Yellowfin and Blackfin Tuna 49

brates. Eleven of the 28 food categories occurred in the stomachs of
both species, and 6 of the 10 most important categories to blackfin also
ranked in the top 10 for yellowfin. The obvious conclusion is that both
species have similar diets when they occur together off the coast of
North Carolina. Statistically, however, the correlation coefficients were
all nonsignificant at the 0.05 (0.344; 29 dO level. The correlation coeffi-
cients were: Spearman, 0.2273; Kendall, 0.1451; and Pearson, 0.2273.

Comparative Diets by Predator Size

Differences in stomach contents by fish size may of course be
attributable merely to the availability of food in the environment, but
they may also be attributable either to a change in food preference, or
to the ability of the predator to capture and swallow certain organisms
as it increases in size. Our objectives of comparing diets by tuna size
were to determine if near-surface feeding was related to tuna size and to
ascertain if basic changes in the diets occurred as the fish grew larger.

Different studies throughout the world's oceans generally suggest
that as tunas grow larger, their diets change. Reintjes and King (1953)
reported that the overall high occurrence of crab larvae, stomatopod
larvae, squid, and juvenile fishes indicates a preference by Pacific yel-
lowfin tuna for small food items. These authors further explained that
small tuna feed predominantly on crustacean larvae; medium-size fish
feed on fish, crustacean larvae, and squid; and large yellowfin mainly
consume fish and squid. These findings were substantiated by Nakam-
ura (1965) and Batts (1972) for skipjack tuna whose diets reflected a
decline in crustaceans and a subsequent higher percentage of fish, as
tuna size increased.

To accomplish our evaluations we first grouped the fish into size
classes (mm FL) (Tables 4, 5). Next, selected food groups — fish, adult
fish, juvenile fish, invertebrates, squid, larval crustaceans and plants —
were established to demonstrate food size (i.e., adult fish vs. larval crus-

Table 4. Selected food items consumed by different sized yellowfin tuna,
expressed as percent frequencies of occurrence.

Fish size (mm FL)
Contents 501-700 701-900 901-1100 >1100

Fish

Adult fish

Juvenile fish

Invertebrates

Squid

Larval crustaceans

Plants

77.8

81.8

75.0

73.8

5.5

10.9

15.0

9.5

16.7

12.7

36.4

11.9

77.8

89.1

76.3

85.7

44.4

34.5

56.3

64.3

38.9

70.9

35.0

35.7

55.5

32.7

30.0

14.3

50 Charles S. Manooch, III and Diane L. Mason

Table 5. Selected food items consumed by different sized blackfin tuna,
expressed as percent frequency of occurrence.

Fish size

(mm FL)

Contents

<500

501-700

701-900

901-1100

Fish

50.0

57.4

87.1

100.0

Adult fish

0.0

7.4

25.8

0.0

Juvenile fish

50.0

16.7

9.7

0.0

Invertebrates

100.0

90.7

64.5

100.0

Squid

50.0

31.5

25.8

100.0

Larval crustaceans

100.0

66.7

38.7

0.0

Plants

50.0

14.8

9.7

0.0

taceans) and materials that we believed to be consumed on or near the
surface (i.e., floating plants). Contents are presented as percent fre-
quency of occurrence (Tables 4, 5).

Yellowfin Tuna. — Size of food items showed little change as fish size
increased or decreased (Table 4). The three key food categories — adult
fish, juvenile fish, and larval crustaceans — neither steadily increased
nor decreased in occurrence as tuna size increased. This finding is con-
trary to that of Dragovich (1970), who found that the frequency occur-
rence offish in stomachs of yellowfin increased with fish size. However,
he discovered no relationship between squid in the diet and tuna size. In
our study, the occurrence of floating plants decreased for the larger size
classes, indicating that perhaps smaller individuals fed more extensively
near the surface.

Blackfin Tuna. — The size of prey items and feeding proximity to the
surface appeared to change with fish size. As fish size increased, large
food items (i.e., adult fish) generally occurred more frequently, and
small food items (i.e., larvel crustaceans and juvenile fish) occurred less
frequently (Table 5). Surface feeding, as suggested by the incidental
ingestion of floating plants, decreased as fish attained larger sizes.
Volumes of Contents Related to Species and Fish Body Weight

Since the quantity and types of foods ingested by fishes are often
converted into caloric equivalents for energetics studies, we present fre-
quencies of the range of food volumes for the two species (Table 6). The
displacement volume for yellowfin averaged 67.9 ml (72.2 g), compared
with 28.6 ml (29.6 g) for blackfin tuna. Volumes of stomach contents of
yellowfin and blackfin varied from 0.1 to 745.0 ml and from 0.1 to 257.5
ml, respectively. The largest volumes were found in a 40 kg yellowfin
and an 8.8 kg blackfin. The volume range for yellowfin from the Pacific
was similar, 0.1 to 1,000 ml (Reintjes and King, 1953). The extremes in
our data were much greater than those described by Dragovich and
Potthoff (1972): 0.1 to 20.0 for skipjack, and 0.1 to 60.0 ml for yellow-
fin tunas collected off the west coast of Africa. In our study, approxi-

Foods of Yellowfin and Blackfin Tuna 51

Table 6. Frequencies of food volumes by species of tuna.

Yellowfi

n tuna

Blackfin tuna

Volume range (ml)

Number

Percent

Number

Percent

O.I- 10.0

64

32.6

46

51.7

10.1- 50.0

67

34.2

26

29.2

50.1-100.0

24

12.2

10

11.2

100.1-150.0

15

7.7

3

3.4

150.1-200.0

9

4.6

3

3.4

200.1-250.0

6

3.1

-

250.1-300.0

3

1.5

1

1.1

300.1-350.0

3

1.5

-

-

350.1-400.0

1

0.5

-

-

400.1-450.0

1

0.5

-

-

450.1-500.0

1

0.5

500.1-550.0

-

-

-

-

550.1-600.0

-

-

-

-

600.1-650.0

-

-

-

-

650.1-700.0

1

0.5

-

-

700.1-750.0

1

0.5
99.9

89

-

Totals 196

100.0

mately 33% of the yellowfin had food volumes exceeding 50 ml, a pro-
portion similar to that of the 29% found by Reintjes and King (1953).
By comparison, Dragovich (1970) noted volumes of less than 20 ml for
85% of the yellowfin from the Atlantic. We found that only 19% of the
blackfin, a much smaller species, had contents over 50 ml.

To determine the relationship of volume to fish body weight, we
first derived the following equation for converting volume in ml to
volume in grams:

Volg = -1.4009 + 1.0846 (Volj^i), N=25, r=0.999.
Comparisons were then made between estimates of stomach contents
and the body weights of some of the tunas selected at random. Percent-
ages of food weight to fish weight varied from trace (<0.002) to 2.02
for yellowfin, and from 0.02 to 3.20 for blackfin tuna. Qnly 10% of the
yellowfin had contents exceeding 1% offish body weight, whereas 20%
of the blackfin tuna had contents exceeding this percentage. Usually our
observations were well below 1%, as were those of Dragovich (1970).

In summary, yellowfin and blackfin tuna appear to be fast, aggres-
sive predators capable of capturing swift, relatively large prey. On the
other hand, they use their gill apparatus to strain small, near-surface
items from the water. During feeding, non-food materials (inorganic as
well as organic) are ingested, probably incidental to normal prey. The
variability of specific food organisms within the major categories (fish,
cephalopods, and crustaceans) in the diets suggests that tunas are non-
selective feeders. This is undoubtedly a factor in their wide geographic

52 Charles S. Manooch, III and Diane L. Mason

distribution, and one would expect, therefore, for the diets of such well-
traveled fish to be rather cosmopolitan.

LITERATURE CITED

Alverson, Franklin G. 1963. The food of yellowfin and skipjack tunas in the

eastern tropical Pacific Ocean. Inter-Am. Trop. Tuna Comm. Bull.

7:295-396.
Batts, Billy S. 1972. Food habits of the skipjack tuna, Katsuwonus pelamis, in

North Carolina waters. Chesapeake Sci. 7i: 193-200.
Bray, Richard N., and A. W. Ebeling. 1975. Food, activity and habitat of three

"picker-type" microcarnivorous fishes in the kelp forests off Santa Barbara,

California. Fish. Bull. U.S. 75:815-829.
Cailliet, Gregor, and J. P. Barry. 1978. Comparison of food array overlap

measures useful in fish feeding habit analysis. Pp. 67-79 in S. Lopovsky and

C. Simenstad (eds.). Fish food habits studies. Proc. Second Pacific

Northwest Tech. Workshop, Univ. Washington, Seattle.
Dingerkus, G., and L. D. Uhler. 1977. Enzyme clearing of alcian blue stained

whole small vertebrates for demonstration of cartilage. Stain Technol.

52:229-232.
Dragovich, Alexander. 1969. Review of studies of tuna food in the Atlantic

Ocean. U.S. Fish Wildl. Serv. Spec. Sci. Rep. Fish. 593. 21 pp.
1970. The food of skipjack and yellowfin tunas in the Atlantic

Ocean. Fish. Bull. U.S. 65:445-460.
, and T. Potthoff. 1972. Comparative study of food of skipjack and

yellowfin tunas off the coast of West Africa. Fish. Bull. U.S. 70:1087-1101.
Fritz, Eugene S. 1974. Total diet comparison in fishes by Spearman rank

correlation coefficients. Copeia 1974(1):210-214.
Goodall, David W. 1973. Sample similarity and species correlation. Pp. 106-156

in R. H. Whittaker (ed.). Ordination and Classification of Communities.

Junk, The Hague.
Manooch, Charles S., Ill, L. E. Abbas and J. L. Ross. 1981. A biological and

economic analysis of the North Carolina charter boat fishery. U.S. Natl.

Mar. Fish. Serv. Mar. Fish. Rev. ^i(8):l-ll.
Nakamura, Eugene L. 1965. Food and feeding habits of skipjack tuna {Katsu-
wonus pelamis) from the Marquesas and Tuamota Islands. Trans. Am.

Fish. Soc. 9^:236-242.
Pinkas, Leo, M. S. OHphant and I. L. K. Iverson. 1971. Food habits of alba-
core, bluefin tuna, and bonito in California water. Calif. Dep. Fish Game

Fish Bull. 752:1-105.
Reintjes, John W., and J. E. King. 1953. Food of yellowfin tuna in the Central

Pacific. U.S. Fish Wildl. Serv. Fish. Bull. 5¥:91-110.
Taylor, William R., and G. C. Van Dyke. 1978. Unpublished manuscript.

Staining and clearing small vertebrates for bone and cartilage study.

Smithsonian Institution, Washington, DC. 19 pp.
Waldron, Kenneth D., and J. E. King. 1963. Food of skipjack in the Central

Pacific. FAO Fish. Rep. 6(3): 143 1-1457.

Accepted 17 September 1983

Benthic Macroinvertebrates of Cane Creek, North Carolina,
and Comparisons with Other Southeastern Streams

David R. Lenat

Biological Monitoring Group,

N.C. Division of Environmental Management,

Archdale Building, Raleigh, North Carolina 27611

ABSTRACT.— The macrobenthos of Cane Creek, in the Piedmont
Plateau of North Carolina, have been sampled by several investigators.
This information was combined to generate a list of 272 invertebrate
taxa. Cane Creek is compared to other unstressed Piedmont streams to
define characteristics of a "normal" stream in this geographic area. If
used cautiously, this data set can provide control information for bio-
logical monitoring. Average taxa richness appears to be the best tool
for environmental assessment work. It shows little variability across a
wide range of North Carolina streams, even outside the Piedmont.
Such a pattern suggests a constant number of niches in stream ecosys-
tems.

INTRODUCTION

The study of pollution in freshwater ecosystems is a complex prob-
lem. Water quality degradation may be caused by an immense number
of pollutants, many of which have an alarming degree of temporal
and/ or spatial variability. To deal with this variability, water quality
monitoring often includes some biological sampling.

The North Carolina Division of Environmental Management
(DEM) has used biological monitoring to analyze a wide variety of
water quality problems (Penrose et al. 1980). Specifically, the Division's
biologists use the structure of the benthic macroinvertebrate community
to detect stress in aquatic systems. There are many ways to examine
such data (Lenat et al. 1980), but all are based on comparisons of actual
data with some expected pattern. The expected pattern is often derived
from a control area, but in many situations it may be difficult to locate
good control stations. This difficulty can often be overcome by using
control data sets. The Division's Biological Monitoring Group has
attempted to generate control data sets by compiling information from
many unpolluted North Carolina streams and rivers. An earlier contri-
bution examined the benthos of a Mountain river system (Penrose et al.
1982). This paper describes the benthic macroinvertebrates of a typical
Piedmont stream.

Brimleyana No.9:53-68. June 1983. 53

54

David R. Lenat

STUDY SITE

Cane Creek is a third-order stream located in Orange County,
North CaroHna (Fig. 1). The total watershed is about 90 km^ and aver-
age discharge is roughly 0.7 m^s (N.C. Division Environmental Man-
agement 1975). Cane Creek is classified as A-II water, i.e. suitable for
drinking (after treatment), body contact, recreation, and "fish and wild-
life propagation".

ci:

c^

Bear Creek

Saxapahaw

White Cross

N

Orange Co.

Chatham c7. -

Fig. 1. Sampling Stations, Cane Creek, North CaroHna.

Benthic Macroinvertebrates Cane Creek NC 55

Fish collections from Cane Creek (unpublished data on file with
N.C. Wildlife Resources Commission) include 25 species. The list appears
typical of Piedmont streams in North Carolina.

The watershed contains both forest and agriculture, principally
dairy farming and row crops. According to the water basin plan (N.C.
Division Environmental Management 1975), there are no point source
discharges in this area, but runoff often causes high turbidity in Cane
Creek. In 1978 the Soil Conservation Service (unpubUshed) cited Cane
Creek as a high priority area for land treatment to reduce erosion rates.
However, little accumulation of coarse bedload sediment was noted in
biological surveys, which may be due to local geology. Cane Creek is
located in the "slate belt" of North Carolina, a zone of metamorphosed
volcanic rock (Simmons and Heath 1979). DEM surveys within this
land type (unpublished data) suggests that little "sandy" stream sedi-
ment is produced through erosion.

METHODS

Several investigators have collected benthic macroinvertebrates from
Cane Creek (Smock and Hughes 1975; Mozley 1978; Penrose et al.
1980). Extensive collections have also been made by the author and by a
limnology class at North Carolina State University. Collection methods
included Hester-Dendy muhiple plate samplers (FuUner 1971), "kicks"
(Frost et al. 1971), and various qualitative techniques. The most inten-
sive collections have been at Station 1 (Lower Cane Creek), but all areas
of the Cane Creek watershed have been sampled (see Fig. 1). Areas
sampled included temporary streams and stream orders 1 through 3. All
records were vertified by the author.

RESULTS AND DISCUSSION

Taxa List

Table 1 lists 272 taxa collected from Cane Creek. A list of taxo-
nomic references used for identification of these organisms is available
from the author. This table also contains data on frequency (rare, com-
mon, or abundant), a classification that is somewhat subjective since
most collections were qualitative. The list contains few species not col-
lected in other Piedmont streams (DEM, unpubl. surveys). The most
unusual record was Mystacides alafimbriata Hill-Griffin, a common
edge species found at several stations in July 1979. This caddisfly had
not been collected east of the Mississippi River. Identification was based
on young larvae and should be confirmed by collection of adults.
Another unusual caddisfly record was Dibusa angata Ross, a species
strongly associated with red algae (Wiggins 1977).

The turbellarian Hydrolimax grisea Haldeman was collected sev-
eral times in Cane Creek. Pennak (1978) listed this species as rare and

56 David R. Lenat

Table 1. List of benthic macroinvertebrates from Cane Creek. Under frequency,
A = abundant, C = common, R = rare.

Taxon Frequency

EPHEMEROPTERA

Baetis amplus A

B. flavistriga A

B. intercalaris A

B. pluto C

B. propinquus R

Callibaetis sp. R

Centroptilum sp. R

Cloeon alamance R

Pseudocloeon spp. C

Caenis cf. diminuta C

Ameletus lineatus C

Isonychia spp.' C

Siphloplectron basale R

Leptophlebia sp. C

Paraleptophlebia sp. A

Hexagenia munda C

Ephemerella (E. ) catawba^ R

E. {Attenella) attenuata R

E. (Dane I la) simplex R

E. {Seratella) deficiens C

E. (Eurylophella) bicolor C

E. (E.) temporalis C

E. (E.) funeralis \ R

Heptagenia aphrodite C

Stenonema modestum^ A

S. smithae C

S. vicarium C

S. (femoratum) R

Stenacron interpunctatum A

S. pallidum C

PLECOPTERA

Allocapnia spp. C

Leuctra sp. R

Acroneuria abnormis C

/I. evoluta R

Eccoptura xanthenes R

Perlesta placida C

Taeniopteryx metaqui A

T. burksi A

Strophoteryx fasciata A

Amphinemura sp. R

Isoperla clio C

Isoperla namata R

Hastaperla brevis R

Benthic Macroinvertebrates Cane Creek NC 57

HEMIPTERA

Belastoma fluminea R

Sigara spp. A

Gerris remigis C

Limnogonus sp. R

Trepobates sp. R

Metrobates hesperius R

Rheumatobates palosi R

Mesovelia mu Isanti R

Rhagovelia obesa C

Microvelia americana C

NEUROPTERA

Climacia sp. R

MEGALOPTERA

Corydalus cornutus C

Nigronia serricornis C

Chauliodes pectinicornis R

5/fl/w A

ODONATA

/4r^/fl spp. C
A. sedula
A. trans lata
A. moesta
A. tibialis

Enallagma spp."* C

Ischnura spp. C

Calopteryx sp. C

Baesiaeschna Janata R

Boyeria vinosa C

Cordulegaster sayi R

Helocordulia selysii R

Neurocordulia obsoleta R

Epitheca cynosura R

Libellula sp. R

Perithemis tenera R

Macromia allegheniensis C

Didymops transversa R

Gomphus spp C

Lanthus parvulus R

Stylogomphus albistylus R
Hagenius brevistylus . R

Dromogomphus spinosus C

TRICHOPTERA

Diplectrona modesta C

Cheumatopsyche spp. A

Hydropsyche betteni A

Macronema Carolina C

Nectopsyche sp. R

58 David R. Lenat

Ceraclea ancylus C

C. tarsipunctata R

Oecetis spp.^ C

Triaenodes injustus R

Triaenodes tardus R

r. cf. sp. b C

Mystacides alajimbriata C

Dibusa angata R

Stactobiella sp. R

Pycnopsyche guttifer R

P. gentilis R

Hydatophylax argus R

Neophylax cf. oligius R

Ptilostomis sp. R

Brachycentrus sp. R

Polycentropus spp.^ C

Phylocentropus sp. C

L>'/7e diversa R

Molanna blenda R

Chimarra cf. aterrima C

Wormaldia sp. R

Psilotreta sp. R

Lepidostoma sp. R

Rhyacophila Carolina R

/?. acutiloba R

/?. /^^ro R
COLEOPTERA

Helichus fastigiatus C

Ancyronyx variegata C

Macronychus glabratus C

Stenelmis spp. C

Oulimnius latiusculus R

Optioservus ovalis R

Dubiraphia quadrinotata C

Ectopria nervosa R

Psephenus herricki C

Anchy tarsus bicolor R

Hydroporus spp. C

Hydrovatus sp. II R

Rhantus sp. R

Tropisternus sp. R

Helophorus sp. R

Laccophilus sp. R

Copelatus glyphicus R

Dineutes sp. C

Gyrinus sp. R

Benthic Macroinvertebrates Cane Creek NC 59

DIPTERA (Miscellaneous)

Palpomyia (complex) C

Anopheles punctipennis C

Culex restuans R

Chaoborus punctipennis R

Dolichopodidae R

Empididae R

Simulium vittatum A

Prosimulium mixtum A

P. rhizophorum C

Chrysops sp. C

Tabanus sp. R

Antocha sp. - C

Dicranota sp. R

Hexatoma sp. R

Limonia sp. R

Pseudolimnophila sp. R

Tipula sp. R

r. abdominalis C

£)ixa sp. R

DIPTERA: CHIRONOMIDAE

Chironomus sp. R

Cryptochironomus blarina R

C fulvus gr. R

Demicryptochironomus sp. R

Dicrotendipes nervosus R

D. neomodestus A

Glyptotendipes sp. R

Kiefferulus dux R

Microtendipes pedellus A

A/, nr. rydalensis R

Paratendipes albimanus C

Phaenopsectra sp. R

P. flavipes C

Polypedilum aviceps R

P. convictum C

P.fallax R

P. illinoense C

P. scalaenum R

Stenochironomus sp. C

Stictochironomus sp. R

Thbelos jucundus C

Xenochironomus xenolabis R

Cladot any tarsus spp. R

Const empellina sp. R

Micropsectra sp. R

60 David R. Lenat

Paratanytarsus sp. ' R

Rheot any tarsus spp. A

Tanytarsus spp. A

T. guerlus gr. C

T. nr.glabrescens C

T. glabrescens R

Zavrelia sp. R

Ablabesmyia mallochi R

A. ornata C

A. parajanta R

Clinotanypus pinguis C

Conchapelopia group C

Labrundinia neopilosella R

L. nr. virescens , C

Larsia sp. R

Natarsia sp. R

Nilotanypus sp. R

Procladius bellus R

P. sublettei R

Psectrotanypus dyari R

Zavrelimyia sp. R

Sympotthastia sp. R

Briilia spp. '' R

Xylotopus par C

Corynoneura spp. C

Cardiocladius sp. R
Cricotopusj Orthocladius %r.

Cricotopus (C) bicinctus C

C (C) tremulus gr. sp. 1 R
(=C infuse at us)

C. (C.) tremulus gr. sp. 2 R

C. (C.) cf. cylindraeeus C

Orthoeladius (O.) robaeki R

O. (O.) nr. dorenus C

O. (O.) cf. obumbratus €

O. (O.) cf. nigritus R

O. (O.) nr. c/arA:e/ R

O. (Euorthoeladius) sp. 1 R

a (£.) sp. 2 » C

Diploeladius eultriger C

Eukiefferiella elaripenis gr. R

Tretenia bavariea gr. R

r. diseoloripes gr. R

Heterotrissoeladius marcidus R

Hydrobaenus spp. R

Nanoeladius spp. C

Genus nr. Nanoeladius R

Benthic Macroinvertebrates Cane Creek NC

61

Parakiefferiella sp. 1

P. sp. 3

P. nr. triquetra

Paraphaenocladius sp. T

Paracricotopus sp.

Parachaetocladius sp.*

Pseudosmittia sp.

Psectrocladius sp.

Rheocricotopus cf. robacki

Synorthocladius sp.

Thienemaniella sp.
MOLLUSCA

Somatogyrus sp.

Ferrissia rivularis

Phy sella sp.

Stagnicola sp.

Gyraulus sp.

Heliosoma anceps

Elimia sp.

Campeloma decisum

Eupera cubensis

Pisidium spp.

Sphaerium simile

Elliptio camplanata

E. icterina

Strophitus undulatus
CRUSTACEA

Cambarus acuminatus

Procambarus acutus

Palaemonetes paludosus

Lirceus sp.

Crangonyx spp.

Hyallela azteca
OLIGOCHAETA

Aulodrilus pigueti

A. pluriseta

Limnodrilus hoffmeisteri

Ilyodrilus templetoni

Peloscolex variegatus

Branchiura sowerbyi

Nais bretscheri

N. variabilis

Slavinia appendiculata

Stylaria lacustris
HIRUDINEA

Helobdella elongata

Placobdella multilineata

62 David R. Lenat

P. papillifera I R

Mooreobdella tetragon R

TURBELLARI

Cura foremanii C

Dugesia tigrina C

Hydrolimax grisea R

BRYOZOA

Plumatella repens C

Fredericella sultana R

MISCELLANEOUS

Porifera: Eunapius sp. C

Nemertea: Prostoma graecens R

Hydracarina C

Nematoda C

' Includes Isonychia bicolor.

^ Subgenera listed for Ephemerella are considered as genera by some authors.

^ Old identifications of Lewis (1974) have been revised in accordance with Bed-

narik and McCafferty (1979).
■* Includes Enallagma divergens.
^ Includes Oecetis cf. cinerascens.
^ Probably includes Cernotina sp.
^ This species listed as Parametriocnemus in many other surveys.

Near Pseudorthocladius; identified by Len Ferrington.

"reported from New Jersey and east Pennsylvania". This interpretation
of range probably results from taxonomic difficulties rather than rarity.
I have found this species to be widespread in Piedmont and Coastal
Plain streams, especially in the latter.

Several common Piedmont macroinvertebrates were rare or absent
in Cane Creek. Ephemerella catawba Traver, which I have found to be a
highly abundant organism in most Piedmont streams, was collected
only once. This mayfly prefers sand-gravel areas, a rare habitat in Cane
Creek. The lack of sand substrates probably also accounts for the
absence of Baetisca Carolina Traver, Progomphus obscurus Rambur,
and Robackia demeijerei (Kruseman).

Taxa collected in first-order tributaries were markedly different
from the fauna at lower stream stations. Many of these first-order
stream taxa were more typical of small mountain streams: Diplectrona
modesta Banks, Ephemerella funeralis McDonnough, Amphinemura
sp., Micropsectra sp., and Heterotrissocladius sp. This may reflect the
colder water temperature normally found in headwater areas (Vannote
and Sweeney 1980). Other taxa collected only in first-order tributaries
included Paraleptophlebia sp., Eccoptura xanthenes (Newman), Pyc-
nopsyche gentilis (McLachlan), and Molanna blenda Sibley.

Benthic Macroinvertebrates Cane Creek NC

63

Total Taxa Richness

Table 2 presents Cane Creek taxa richness by group, and includes
data from several other southeastern Piedmont stream investigations.
These data show a remarkable degree of consistency if allowances are
made for geographic area, collection methods, and special interest of the
investigators. Some differences, as seen in South Carolina and Georgia
samples, are attributable to collections of nonaquatic adult insects. This
usually increases estimates of species richness, as greater taxonomic pre-
cision may be attained with adult specimens.

Collections in semiaquatic areas also may increase the number of
species collected, especially for Coleoptera. Holeski and Graves (1978)
found that 30+ species of "shore beetles" were usually found at most
stations, but that this type of data was not useful in assessing environ-
mental stress. Eighty chironomid taxa were identified from Cane Creek,
about 30% of the total fauna. Use of adults or pupal exuviae might have
doubled this figure. Coffman (1973), using pupal exuviae, identified 143
chironomid taxa from Linesville Creek, while larval sampHng from the
same area produced only 77 taxa.

Table 2. Total taxa richness at Cane Creek, with selected data from other intensive stream
investigations. Numbers in parentheses omit data based only on nonaquatic
adults. NI = Not Identified.

Stream:
State:

Cane'

NC

Wildcat'
SC

Little

Garvin'

SC

20(19)

Upper

Three Runs*

GA

Linesville'
PA

EPHEMEROPTERA

30

54

32(27)

13

ODONATA

22

29

30(16)

24

7

PLECOPTERA

13

44

6

28(17)

5

HEMIPTERA

10

4

8

27

5

MEGALOPTERA

4

2

5

6(4)

2

LEPIDOPTERA

1

0

0

9(1)

0

NEUROPTERA

1

0

0

2

0

COLEOPTERA

16

20

15

86^

15

TRICHOPTERA

31

48

10

108(46)

25

DIPTERA/(MISCELLANEOUS)

19

17

12

NI

14

DIPTERA: CHIRONOMIDAE

80

50

61

NI

77

OLIGOCHAETA

13

NI

NI

NI

?

HIRUDINEA

4

NI

NI

NI

?

MOLLUSCA

14

NI

NI

NI

?

OTHER

• ^y.: -

8

272

NI
266

NI

167

NI

321

?

TOTAL

192

This study.

White and Fox (1980), based largely on the collections of Paul Carlson.

White and Fox (1980).

Morse etal. (1980).

Coffman etal. (1971).

Includes many semiaquatic species.

64 David R. Lenat

Table 3. Average taxa richness by group for North Carolina Piedmont streams. Ranges
are rounded to integer values.

Stream:

# Collections:

County:

Cane Bolin Gates

3 3 3

Orange Orange Orange

UT Lanes
Creek

3
Union

Long
Branch

4
Gaston

4-Mile

4

Davidson

Group

Range

EPHEMEROPTERA

8.3

8.7

8.7

7.7

8.0

9.4

8-9

PLECOPTERA

1.7

3.7

3.7

2.0

3.0

5.0

2-5

ODONATA

0.7

0.7

3.7

0.7

1.0

1.6

1-4

TRICHOPTERA

5.0

4.0

5.6

2.0

7.3

3.4

2-7

COLEOPTERA

5.0

4.7

5.7

5.3

3.7

3.0

3-6

MEGALOPTERA

2.0

2.0

1.3

2.7

1.0

1.8

1-3

DIPTERA

21.0

13.7

19.7

16.0

17.3

14.6

14-21

MOLLUSCA

3.3

2.3

1.7

2.3

1.3

1.4

1-3

OLIGOCHAETA

3.7

1.0

2.7

2.3

1.7

1.8

1-4

CRUSTACEA

1.7

3.0

2.7

1.7

2.0

1.7

2-3

OTHER

2.7

2.0

0.3

0.6

-

-

0-3

TOTAL

55.1

45.8

55.8

43.3

46.3

44.0

43-56

Average Taxa Richness/ Percent Density

Environmental assessment is often based on quantitative data from
a single collection or on data averaged over several collections. It may
be difficult to relate these data to water quality if good control informa-
tion is lacking. Tables 3 and 5 present average taxa richness and density
values from Cane Creek and make comparisons with five other rela-
tively unstressed Piedmont streams. All samples were taken during
DEM investigations. Collection methods (kicks) and identification
techniques were identical for all streams. From this data set, one may
attempt to define the normal characteristics of Piedmont stream
macroinvertebrate communities.

Table 3 gives average taxa richness values in the range of 43 to 56
(x = 48.5). There was often a remarkable constancy at the group level.
For example, Ephemeroptera varied only from 8.3 to 9.4. Plecoptera
values were generally in the range of 2 to 4, except at 4-Mile Creek. The
greater number of Plecoptera in this stream, which is at a higher eleva-
tion, may reflect colder water temperatures. The expected number of
Odonata is close to the average (1.4), except at Cates Creek. This is a
very narrow, slow stream and might be expected to show a high propor-
tion of bank associated (edge) species in quantitative samples. Odonata
are most frequently collected in bank areas. Variations in Coleoptera
and Trichoptera, especially the latter, are not easily explained. Various
regional differences in water chemistry, temperature, and gradient may
be responsible. The number of Diptera, especially chironomids, had a

Benthic Macroinvertebrates Cane Creek NC 65

wide range (14-21), but was always high. Even stressed areas may have a
diverse chironomid fauna (Penrose et al. 1980) although tolerant species
will become dominant. The "other" category includes miscellaneous
Insecta (Neuroptera, Lepidoptera), Turbellaria, Hirudinea, Nemertea,
Porifera, and Nematoda. Variation in this category is very unpredictable.

Table 4 presents average taxa richness values for streams in three
broad geographic areas in North Carolin: Mountains (data expanded
from Penrose et al. 1982); Piedmont (as in Table 5); and Coastal Plain.
These three physiographic regions have differing physical characteris-
tics. Going from the mountains to the coast one would expect increasing
water temperature, lower gradient, and increasing amonts of sand and
silt. The Coastal Plain data set is based on collections in Craven, Hert-
ford, and Northampton counties. These coastal streams are not entirely
unstressed, but the data are adequate to illustrate geographic trends.

The overall trend suggests a relatively constant number of species
in stream environments. Average taxa richness for Coastal Plain and
Mountain streams is well within the range expected for Piedmont
streams (Table 3). From this pattern one might advance the hypothesis
that unpolluted streams have a relatively constant number of niches.
Furthermore, these data imply that taxa richness is an excellent moni-
toring tool across a wide range of stream types. A similar hypothesis
was advanced by Patrick (1975:448). She examined species richness in
nine different rivers and concluded that "similar-sized areas of different
streams support similar numbers of species." This constancy of "a -
diversity" prevailed even when the number of species shared between
streams was low.

Table 4. Average taxa richness by group for Coastal Plain, Piedmont and
Mountain streams in North Carolina. See text for data sources.
Number of streams shown in parentheses.

Group

Coastal Plain (7)

Piedmont (6)

Mountains (9)

EPHEMEROPTERA

2.9

8.5

9.7

PLECOPTERA

0.8

3.2

6.5

ODONATA

2.5

1.4

0.8

TRICHOPTERA

3.9

4.6

7.8

COLEOPTERA

2.8

4.6

2.5

MEGALOPTERA

0.7

1.8

0.2

DIPTERA

23.0

17.1

14.5

MOLLUSCA

3.7

2,1

0.5

OLIGOCHAETA

3.6

2.2

0.9

CRUSTACEA

2.5

2.1

0.9

OTHER

2.0

0.9

0.2

TOTALS

48.4

48.9

44.5

66 David R. Lenat

At the group level, taxa richness varies considerably across the
three types of streams. Each stream type has a different assemblage of
invertebrate predators. Plecoptera and Trichoptera are most diverse in
Mountain streams, Megaloptera in Piedmont streams, and Odonata in
Coastal Plain streams. The "other" category, also most important in
Coastal Plain streams, includes many other predators: Hirudinea, Tur-
bellaria, and Prostoma graecens (Bohmig). Shifts may also be observed
in the collector-gatherer groups. Going from the mountains toward the
coast, Ephemeroptera-Trichoptera decline and are replaced by Coleop-
tera (Piedmont only), Oligochaeta, Crustacea, and Mollusca.

Table 5 shows density (expressed as a percentage of total density)
for Cane Creek and five other North Carolina Piedmont streams. The
density values show much greater between-stream variability than does
average taxa richness. These data may serve as controls only if used
with extreme caution. Variability is imposed by such factors as stream
size, substrate, and geographic locality. The data are further biased by
the selection of riffle areas and by the mesh size used in sample process-
ing. A smaller mesh size will increase the importance of Diptera and
Oligochaeta.

Table 5. Density by group (as percent of total density) for Cane Creek and
other unstressed North Carolina Piedmont streams. Data are the aver-
age of 3-4 collections, rounded to integer value.

Stream:

Cane

Bolin

Gates

UT Lanes Long Br.

4-Mile

Group

EPHEMEROPTERA

13

12

16

6

27

38

PLECOPTERA

3

5

9

3

4

14

ODONATA

-

-

1

-

1

1

TRICHOPTERA

50

70

3

14

25

10

COLEOPTERA

5

3

4

29

2

9

MEGALOPTERA

2

1

1

-

-

-

DIPTERA

20

8

61

45

39

27

MOLLUSCA

3

-

-

2

-

-

OLIGOCHAETA

2

-

2

1

1

-

CRUSTACEA

-

1

2

-

-

1

OTHER

2

-

1

-

-

-

ACKNOWLEDGMENTS.— I would particularly like to acknowl-
edge the help and support of two colleagues: Samuel Mozley, N.C.
State University, and David Penrose, N.C. Division of Environmental
Management. Taxonomic assistance was received from many sources:
Chironomidae — Samuel Mozley; Patrick Hudson, U.S. Fish and Wild-
life Service; Leonard Ferrington, University of Pittsburgh; Robert Bode,

Benthic Macroinvertebrates Cane Creek NC 67

N.Y. Department of Health; Broughton Caldwell, Georgia Department
of Natural Resources; N.W. Boesel, Miami University; Don Oliver, Bio-
systematic Research Institute, Canada; and David Smith, U.S. Envir-
onmental Protection Agency, Athens, GA. Ephemeroptera — Paul
Carlson, S.C. Department of Health and Environmental Control. Tri-
choptera — John Morse, Clemson University. Crustacea — John E.
Cooper, N.C. State Museum of Natural History. People assisting in col-
lections included Barbara Burchard, Ross Green, and Ken Eagleson.
Two anonymous reviewers and John E. Cooper provided valuable
comments on the manuscript.

LITERATURE CITED

Bednarik, A. F., and W. P. McCafferty. 1979. Biosystematic revision of the
genus Stenonema (Ephemeroptera: Heptageniidae). Can. Bull. Fish. Aquat.
Sci. 207:1-73.

Coffman, William P. 1973. Energy flow in a woodland stream ecosystem: II.
The taxonomic composition and phenology of the Chironomidae as deter-
mined by the collection of pupal exuviae. Arch. Hydrobiol. 77:281-322.

, K. W. Cummins and J. C. Wuycheck. 1971. Energy flow in a

woodland stream ecosystem: I. Tissue support trophic structure of the
autumnal community. Arch. Hydrobiol. 65:232-276.

Frost, S., A. Huni and W. E. Kershaw. 1971. Evaluation of a kicking technique
for sampling stream bottom fauna. Can. J. Zool. 49:167-173.

Fullner, Richard W. 1971. A comparison of macroinvertebrates collected by
baskets and modified multiple-plate samplers. J. Water Pollut. Control
Fed. 45:494-499.

Holeski, Paul M., and R. C. Graves. 1978. An analysis of shore beetle com-
munities of some channelized streams in northeast Ohio (Coleoptera).
Great Lakes Entomol. 77:23-26.

Lenat, David R., L. A. Smock and D. L. Penrose. 1980. Use of benthic
macroinvertebrates as indicators of environmental quality. Pp. 97-1 12 in D.
L. Worf (ed.). Biological Monitoring for Environmental Effects. Lexington
Books, D. C. Heath Co., Lexington. 227 pp.

Lewis, P. A. 1974. Taxonomy and ecology of Stenonema mayflies (Heptage-
niidae: Ephemeroptera). EPA Monitoring Series EPA-670/ 4-74-006.

Morse, John C, J. W. Chapin, D. D. Herlong and R. S. Harvey. 1980. Aquatic
insects of Upper Three Runs Creek, Savannah River Plant, South Caro-
lina. Part I. Orders other than Diptera. J. Ga. Entomol. Soc. 75:74-101.

Mozley, Samuel C. 1978. Preimpoundment survey of species of benthic
macroinvertebrates in Cane Creek. Unpubl. report Environmental Services
Corp., Chapel Hill. 40 pp.

N. C. Division Environmental Management. 1975. River Basin Plans.

Patrick, Ruth. 1975. Stream communities. Pp. 445-459 in M. S. Cody and J. M.
Diamond (eds.). Ecology and Evolution of Communities. Belknap Press,
Harvard Univ., Cambridge. 545 pp.

Pennak, Robert W. 1978. Freshwater Invertebrates of the United States. 2nd
edition. Ronald Press, New York. 780 pp.

68 David R. Lenat

Penrose, David L., D. R. Lenat and K. W. Eagleson. 1980. Biological evalua-
tion of water quality in North Carolina streams and rivers. N. C. Div.

Environ. Manage., Biological Series #103. 181 pp.
, and . 1982. Aquatic macroinvertebrates of the

Upper French Broad River basin. Brimleyana 8:27-50.
Simmons, Clyde E., and R. C. Heath. 1979. Water quality characteristics of

streams in forested and rural areas of North Carolina. U. S. Geol. Survey

Water Resour. Invest. 79-108. 49 pp.
Smock, L. A., and H. L. Hughes. 1975. Benthic collections. Haw and New

Hope rivers. May 1974-April 1975. Rep. to U.S. Army Corps. Engineers,

Wilmington District, NC. 52 pp.
Vannote, Robert L., and B. W. Sweeney. 1980. Geographic analysis of thermal

equilibria: a conceptual model for evaluating the effect of natural and

modified temperature regimes on aquatic insect communities. Am. Nat.

7/5:667-695.
White, Tina R., and R. C. Fox. 1980. Recolonization of streams by aquatic

insects following channelization. Clemson Univ. Water Resour. Inst. Tech.

Rep. 87, Vol. 1. 119 pp.
Wiggins, G. B. 1977. Larvae of North American caddisfly genera (Trichoptera).

Univ. Toronto Press, Toronto, Canada. 401 pp.

Accepted 30 September 1982

Pleistocene Mammals from the Rock Springs Local Fauna,

Central Florida

Kenneth T. Wilkins^

Florida State Museum,
University of Florida, Gainesville. Florida 3261 1

ABSTRACT. — Although several of the interesting fossil mammals
from Rock Springs, Orange County, Florida, have been mentioned in
scientific papers, this is the first comprehensive report of its mammal-
ian fauna. The fauna, which evidently accumulated during several
intervals of late Pleistocene deposition, consists of at least 27 species,
including both marine and terrestrial forms. The faunal composition,
coupled with modes of preservation (e.g., a barnacle-encrusted speci-
men of a terrestrial species), provides direct evidence of at least one
marine transgression (probably corresponding to the Pamlico shore-
line) into central peninsular Florida. Two extralimital taxa character-
ize the Rock Springs fauna: (1) current affinities of Thomomys sp.
(pocket gophers) are with western North America, and (2) Mormoops
sp. (leaf-chinned bats) now occur in western North America and the
Neotropics.

INTRODUCTION

The Rock Springs site in central Florida has yielded an interesting
sample of late Pleistocene vetebrate fossils. Although the first collec-
tions were made in the 1920s, the first mention of fossil vertebrates from
Rock Springs was a list of seven taxa (Gut 1939). Of the five vertebrate
classes represented in this fauna, only the avifauna has been comprehen-
sively reported (Woolfenden 1959). Auffenberg (1963) included two ser-
pent species — Drymarchon corais, an indigo snake, and Crotalus gigan-
teus, a large rattlesnake — from Rock Springs in his review of the fossil
snakes of Florida. Ray et al. (1963) discussed the presence of Mor-
moops megalophylla, a leaf-chinned bat, and Ray (1964) and Gillette
(1976) studied a species of small cat, Felis amnicola, from this site.
Webb (1974) listed some 17 mammalian species from the Rock Springs
site.

This paper presents the first comprehensive compilation of the
Rock Springs mammalian fauna. Re-examination of all available fossil
material, including a new collection made in 1982, reveals 27 mammal-
ian taxa, some 14 of which are extinct (Table 1). Ten species in the
Rock Springs local fauna presently occur in Florida. Because one spe-
cies was not identifiable to genus (i.e., the large felid), its current status

' Present Address: Department of Biology, Baylor University, Waco, Texas 76798.

Brimleyana No.9:69-82. June 1983. 69

70 Kenneth T. Wilkins

Table 1. Mammalian fauna of the late Pleistocene Rock Springs site, Orange
County, Florida.

Extant Extant, not

Species

Extinct

in Florida in Florida

Blarina cf. carolinensis

X

Mormoops megalophylla

X

Myotis austroriparius

X

Holmesina septentrionalis

X

Dasypus bellus

X

cf. Glossotherium

X

Sylvilagus sp.

X

Geomys pinetis

X

Thomomys cf. orient alls

X

Castor canadensis

X

cf. Sigmodon

X

cf. Tursiops

X

Urocyon cinereoargenteus

X

cf. Canis dims

X

Tremarctos floridanus

X

Ursus americanus

X

Felis amnicola

X

cf. Felidae

gen. et sp. indet.

Mammut americanum

X

Mammuthus sp.

X

Trichechus manatus

X

Tapirus veroensis

X

Equus sp.

X

Mylohyus nasutus

X

Paleolama mirifica

X

Odocoileus virginianus

X

Bison sp.

X

is uncertain. The one extant species in the Rock Springs fauna not now
occurring in Florida is Mormoops megalophylla. The other extralimital
genus in the Rock Springs fauna is Thomomys, the smooth-toothed
pocket gophers. General affinities of Mormoops and Thomomys lie in
western North America and the Neotropics.

Rock Springs is located in Kelly Park about 10 km (6 mi.) north of
Apopka, Orange County, Florida (Sorrento quad. NEl/4, NEl/4,
NWl/4, Sec. 15, T20S, R28E). Its elevation is about 7 m above present
sealevel. The spring is one of many flowing into the Wekiva River, a
tributary of the St. Johns River. The spring discharges at a mean rate of
1.83 cubic m/ second (64.6 cubic ft. /second) from a partially-submerged

Pleistocene Mammals Central Florida 71

cavern in a 5 m high limestone bluff of the early Miocene Hawthorne
Formation (Rosenau et al. 1977). The mouth of the cavern is about 1.5
m in diameter. The Rock Springs fauna has been collected along the
initial 200 m of the spring run, and the underlying Crystal River Forma-
tion of the Ocala Group (late Eocene age) is exposed in places along the
run. Present habitats in the surrounding region are the longleaf pine-
turkey oak sandhill ecosystem and the live oak xeric hammock associa-
tion, as is common in the upland karst regions of peninsular Florida.
Mesic hardwoods and other floodplain and freshwater marsh species
occupy the adjacent riparian corridor.

The material comprising the Rock Springs mammal fauna repre-
sents the efforts of many amateurs and professionals, including J.
Bauer, R. and J. Franz, L. F. Lovell, Mrs. C. A. Meyer, G. S. Morgan,
J. W. Pierce, G. M. Ponton, A. E. Pratt, R. Savage, C. Simpson, J. R.
Todd, K. T. Wilkins and G. E. Woolfenden. Additional Rock Springs
fossils are contained in the private collection of H. James Gut, which is
now in the custody of his son, Robert M. Gut of Jacksonville, Florida.
Specimen numbers cited in the following annotated species list refer to
the catalogue of the Florida State Museum Vertebrate Paleontology
Collection at the University of Florida (UF). Catalogue numbers pre-
ceded by "V" are part of the Florida Geological Survey collection which
is also housed with the UF collection.

ANNOTATED SPECIES LIST

Order Insectivora

Family Soricidae

Blarina cf. carolinensis (Bachman) 1837

Material— Partial right mandible with iT, P4-M2, (UF 48997);
partial left mandible with Ml, (UF 48998).

Remarks. — Short-tailed shrews in Pleistocene deposits in Florida
have previously been referred to B. brevicauda, the name formerly app-
lied to most living populations of Blarina in eastern North America.
Recent morphometric and karyotypic study of modern populations
indicates that northern, southeastern and southwestern populations are
separate species, with southeastern populations referred to Blarina caro-
linensis (Genoways and Choate 1972; George et al. 1982). I apply caro-
linensis to the Rock Springs Blarina solely on geographic grounds. Sta-
tistical comparisons of later Pleistocene material from Florida with
modern B. brevicauda, B. hylophaga and B. carolinensis are necessary
to establish species identification.

72 Kenneth T. Wilkins

Order Chiroptera

Family Mormoopidae

Mormoops megalophylla (Peters) 1864

Material. — Cranial and post-cranial material (UF 3860-3866) as
described by Ray et al. (1963).

Remarks. — Identification was established by Ray et al. (1963),
who compared Rock Springs Mormoops with samples of the two living
species of the genus. Kurten and Anderson (1980) noted Rock Springs as
the only Pleistocene occurrence of the species in continental North
America.

Family Vespertilionidae

Myotis austroriparius (Rhoads) 1897

Material. — Skull with partial maxillary dentition (UF 8925); ros-
trum with partial maxillary dentition (UF 8926); 6 dentaries with partial
dentitions and 1 edentulous dentary (UF 8927); 20 humeri (UF 8928); 25
radii (UF 8929); 30 metacarpals (UF 8930); 4 femora (UF 8931).

Remarks. — This material is assigned to M. austroriparius rather
than to M. grisescens Howell 1909 because of its relatively small size.
M. austroriparius is a very common cave-dwelling bat of the karst
regions of present-day Florida (Rice 1957).

Order Edentata

Family Dasypodidae

Holmesina septentrionalis (Leidy) 1889

Material— 2 movable scutes (UF 24845); 4 fixed scutes (UF 24846).

Dasypus bellus (Simpson) 1930

Material. — 1 fixed scute (V-4455).

Family Mylodontidae

cf. Glossotherium (Owen) 1840

Material— Distal phalanx (V-4394).

Remarks. — Poorly ossified epiphyses indicate that this claw is that
of a juvenile. The small size (greatest length = 51 mm) and roundness of
the claw suggest Glossotherium. Measurements taken midway between
proximal and distal ends are: mediolateral width 15 mm; dorsoventral
depth 22 mm. Claws of other ground sloth genera are more mediolater-
ally compressed than in Glossotherium.

Pleistocene Mammals Central Florida 73

Order Lagomorpha
' Family Leporidae

Sylvilagus sp. Gray 1867

Material— Left dentary with P3-P4, M2-M3 (V-4467); proximal
femur (UF 49201); distal femur (UF 49202).

Remarks., — Two species of similar-sized cottontails, Sylvilagus
palustris and S.floridanus, presently range throughout most of Florida.
The distribution of a third, much larger species, S. aquaticus, closely
approaches the Florida panhandle in Alabama; quite possibly the spe-
cies has recently extended its range into extreme western Florida. Left
mandibular toothrow lengths were measured in five specimens each
(mixed sex) of modern S. palustris and S. floridanus from Alachua
County, Florida, in efforts to assign specific identity to the Rock
Springs rabbit material. The sample mean for toothrow length for S.
floridanus is 15.20 mm (s = 0.29 mm) and for S. palustris is 15.24 mm (s
= 0.45 mm). Toothrow length in the Rock Springs specimen is 14.9 mm.
Lack of mensural differences between these three samples dictates refer-
ral of the Rock Springs rabbit to Sylvilagus sp.

Order Rodentia

Family Geomyidae

Geomys pinetis Rafinesque 1817

Material— Isolated upper P4, (UF 49205).

Remarks. — Geomys pinetis is the only extant pocket gopher in the
eastern United States. The southeastern pocket gopher is presently
abundant in the vicinity of Rock Springs, and in other parts of the
southeastern United States characterized by the longleaf pine-turkey
oak sandhill ecosystem. Despite the limited referred material, its generic
identity is certain. Extant Geomys species possess an enamel band on
the posterior surface of the fourth upper premolar. This posterior
enamel investment on upper P4 is absent in Thomomys species and in
the early-to-middle Irvingtonian species of Geomys in Florida (Wilkins
1984).

Thomomys cf. orientalis Simpson 1928

Material — Six mandibles with partial to complete dentitions (UF
46571-46576).

Remarks. — Generic assignment of these specimens is indicated by
the presence of both anterior and posterior enamel bands on the lower
molars; anterior bands are absent in Geomys species. The genus Thom-
omys presently occurs in western North America (Hall 1981). The only

74 Kenneth T. Wilkins

other Thomomys material reported from the southeastern United States
is from Sabertooth Cave (Rancholabrean, Citrus Co., Florida). Simp-
son's (1928) description of the species referred only to skull material,
none of which is available in the Rock Springs deposit. Pending further
study, the Rock Springs Thomomys is tentatively referred to orientalis.

The Rock Springs site has yielded additional geomyid material for
which generic assignment cannot be made on qualitative grounds. Sim-
ilarity in size and preservation to the Thomomys material Hsted above
suggests that four mandibles lacking cheekteeth (UF 46577-46580) are
referable to Thomomys. No effort has been made to identify isolated
lower P4's (UF 46581) and molars (UF 46582) beyond the family level.

Family Castoridae
Castor canadensis ¥.\x\i\ \%20

Material. — Lower molar (V-4399).

Remarks. — The presence of beaver in the Rock Springs fauna was
previously noted by Johns (1958), who argued that its present distribu-
tion (restricted in Florida to the panhandle and northern peninsula) is
less extensive than it was in historic and Pleistocene times. The Rock
Springs specimen, coupled with two nearly complete mandibles with P4-
M3 (V-5403) from the shores of nearby Lake Monroe, Volusia County,
comprises the species' southernmost Pleistocene record in Florida. Post-
Wisconsinan records include specimens from middens in Seminole,
Volusia and Brevard counties (Furgeson 1951) and along the Indian
River (Allen 1942).

Family Cricetidae
cf. 5/gmo<io« Say and Ord 1825

Material. — Two isolated upper incisors (UF 48999).

Remarks. — Comparison of these incisors with modern Sigmodon
hispidus and other rodents from Florida suggests this generic assign-
ment. Martin (1974) recognized the occurrence of two species of cotton
rats, S. bakeri and S. hispidus, in the United States and Mexico during
the late Irvingtonian and Rancholabrean. Identification to species
requires molar teeth.

Pleistocene Mammals Central Florida 75

Order Cetacea

Family Delphinidae

cf. Tursiops

Material— Vertebra (UF 48976).

Remarks. — The specimen is a complete centrum (diameter = ca. 38
mm) with one complete transverse process (length = ca. 75 mm). The
paired portion of the neural arch ventral to its fusion into the spine is
present. Comparison with modern small odontocetes from Florida sug-
gests Tursiops. The relatively intact nature of this specimen, and the
lack of water-wear, suggests in situ deposition and comprises further
evidence of marine transgression above the Rock Springs elevation.

Order Carnivora

Family Canidae

Urocyon cinereoargenteus (Schreber) 1775

Material.— Distal humerus (UF 49000).

Remarks. — This specimen resembles humeri of modern individuals
of the gray fox from central Florida.

cf. Canis dirus Leidy 1858

Material. — Canine fragment (V-4397).

Remarks. — This identification follows Webb (1974).

Family Ursidae

Tremarctos floridanus (Gidley) 1928

Material— Molar fragment (UF 8946); 2 upper M3's (UF 8947-
8948).

Remarks. — The genus is extinct in North America, with the sole
living species restricted to the South American Andes. Tremarctos flori-
danus, the Florida spectacled bear, was a common member of many of
Florida's Pleistocene faunas; Ursus and Tremarctos remains are often
found in the same deposits. Ecological differences of the two species
apparently allowed coexistence (Kurten 1966). Adaptation of T. florida-
nus to a highly herbivorous lifestyle is evident from its relatively broad
molars, which allowed for increased occlusal surface areas. The rela-
tively narrow and elongate molars of U. americanus indicate more
omnivorous adaptation.

76 Kenneth T. Wilkins

Ursus americanus

Material— Two upper M3's (UF 8951-8952); femur (UF 8953).

Remarks. — The black bear presently occurs throughout much of
Florida. An unfossilized femur attests to its relatively recent occurrence
in the Rock Springs vicinity. Generic identity of other ursid material is
uncertain: distal humerus (UF 49203); canine fragment (UF 49204).

Family Felidae

Felis amnicola Gillette 1976

Material— Right dentary with P3-MT (UF 4522)

Remarks. — This specimen was originally recognized and illustrated
by Ray (1964) as a jaguarundi, Felis yagouaroundi. Subsequently,
Gillette (1976) described a new species of small river cat, F. amnicola, to
which he referred this Rock Springs dentary.

cf. Felidae

gen. et sp. indet.

Material— Phalanx (UF 8954)

Remarks. — Webb's (1974) tabulation of "Felis sp." at Rock Springs
probably refers to this specimen, which compares favorably in size with
the modern Florida panther, Felis concolor coryi Bangs 1899.

Order Proboscidea

Family Mammutidae

Mammut americanum (Kerr) 1791

Material. — Right dentary fragment (V-4378); molar fragments (UF
48986, V-4385, V-4464); molar enamel fragment (V-4465).

Family Elephantidae
Mammuthus sp. (Falconer) 1857

Material— Molar (V-4473); molar plates (UF 48987, V-4383); tusk
fragments (V-4384).

Remarks. — In accordance with Kurte'n and Anderson (1980),
mammoth material is provisionally referred to Mammuthus sp. pending
comprehensive study of the genus.

Order Sirenia

Trichechus manatus Linnaeus 1758

Material— Cheektooth (UF 48981); cheektooth fragment (UF
48982); edentulous maxillary fragment (V-4451); ear ossicle (UF 48983);

Pleistocene Mammals Central Florida 77

tympanic (UF 48984); vertebrae material (UF 48979-48980, V-4469);
ribs (UF 48977-48978, V-4380).

Remarks. — Various states of preservation are evident in this mana-
tee material. Some is well permineralized, whereas other specimens are
only slightly fossilized. The material was probably deposited during sev-
eral different intervals, including times when the site was marine or
estuarine, or when it comprised a freshwater system occupied by mana-
tees much as are many other river systems in Florida today.

Order Perissodactyla

Family Tapiridae

Tapirus veroensis Sellards 1918

Material— Palate with left U, 13, PJ_, M3 and right PJ_-M3 (UF
18702); left maxilla with MNM3 (V-4389);_upper molar fragment (UF
48970); left dentary and_symphysis with P4-M3 (V-4390/4391); partial
right dentary with_P4-M3 (UF 12485); left 12 (UF 48971); deciduous P2
(UF 8945); left P2 (V-4396); lower molar (UF 48969); 5 isolated teeth
(V-4454).

Remarks. — All referred material resembles T. veroensis rather
than the larger T. copei Simpson 1945.

Family Equidae

Equus sp.

Material— Upper cheektooth (UF 48975); lower molar (UF 48974);
incisors (UF 48972-48973); radius (V-4471).

Remarks. — The material represents several states of preservation.
The unfossilized radius, that of an immature individual, probably
represents the domestic Equus caballus. The dental material is fossilized
and is provisionally referred to Equus sp.

Order Artiodactyla

Family Tayassuidae

Mylohyus nasutus (Leidy) 1869

Material— Partial right mandible with P3-M3(UF 17720).

Remarks. — This specimen is particularly interesting because of its
preservation. Overall coloration of the specimen is black, in contrast to
buff or chalky colors of most other Rock Springs material. The ventral
surface of the dentary is broken away to expose the mandibular canal.
External surfaces of the jaw, as well as the internal surfaces of the man-
dibular canal, are encrusted with growths of various marine organisms,
including barnacles. Assuming in situ deposition, this specimen further

78 Kenneth T. Wilkins

documents marine transgression of the Rock Springs site. The most
recent occurrence of a transgression sufficient to cover the Rock Springs
site probably corresponds to the Pamlico marine terrace (Healy 1975).
Hence, the minimum age estimate for this specimen is about 125,000
years (see Discussion).

Family Camelidae

Paleolama mirifica (Simpson) 1929

Material— Deciduous P4 (V-4398); P4 (V-4453); M3 (UF 12491).

Remarks. — Webb (1974:183) recorded this stout-legged llama as a
member of the Rock Springs fauna in his study of Florida Pleistocene
Lamini.

Family Cervidae

Odocoileus virginianus (Zimmermann) 1780

Material — Right dentary with P3-M3 (V-4395); isolated cheek-
teeth (UF 48988-48989, V-4462); distal tibia (UF 48990); distal metatar-
sal (UF 48991).

Remarks. — The white-tailed deer presently occurs throughout
Florida. Because of differential preservation, deer material in the Rock
Springs fauna appears to have been deposited during several intervals.

Family Bovidae

Bison sp.

Material— Upper cheektooth (UF 10037); M3 (UF 52184); lower
cheektooth (UF 15065).

Remarks. — Robertson (1974) stated that unequivocal species
determination for the three species of Bison in Florida can be made only
from horn core material. He was unable to identify diagnostic dental
characters, although he noted that, by its smaller size. Bison bison (Lin-
naeus) 1758 could sometimes be distinguished from the two larger spe-
cies, Bison latifrons (Harlan) 1825 and Bison antiquus Leidy 1852.
Mesurements of the nearly unworn lower M3(UF 52184) are: greatest
alveolar length = 47.7 mm, greatest alveolar width =21.2 mm, and
greatest crown height = 69. 1 mm. The large size of the Rock Springs M3
suggests the presence of one of the larger Bison species rather than B.
bison.

DISCUSSION

Chronology

As with many late Pleistocene fluvial sites in Florida, the Rock
Springs local fauna accumulated during several intervals. Some material

Pleistocene Mammals Central Florida 79

is Recent or even unfossilized (e.g., some of the Equus and Odocoileus).
In addition, some Miocene fossils of marine invertebrates and sharks
have eroded from the limestone bedrock through which the stream
flows. The late Pleistocene fossilized materials represent accumulation
in two types of situations: one of higher sealevels and another of much
lower sealevels. The cetacean vertebra, the barnacle-covered Mylohyus
dentary, and the Trichechus material comprise evidence of marine intru-
sion. The last time seas could have covered the Rock Springs locale
(elevation 7 m above present sealevel) was some 125,000 years before
present (ybp). According to eustatic studies, 125,000 ybp was the most
recent occasion that seas have been higher than at present (Shackleton
and Opdyke 1973; Bloom et al. 1974; Chappell 1974). In Florida, this
late Sangamonian interglacial transgression corresponds to the Pamlico
shoreline at about 9 m (Healy 1975). It is possible, yet unHkely, that
some earlier transgressions of similar or greater extent introduced the
marine influence. Thus, the cetacean, the barnacles encrusting the
previously-deposited peccary jaw, and at least some of the sirenian
material are probably of Sangamonian age.

Most of the terrestrial vertebrates presumably were deposited dur-
ing one or more phases of reduced sealevel. During at least one such
interval, sealevel and the piezometric surface dropped sufficiently to
produce habitats suitable to vertebrates that do not now occur in east-
ern North America. Woolfenden (1959) noted a western contingent in
the Rock Springs avifauna. A portion of the Rock Springs mammal
fauna also exhibits noneastern United States affinities. Smooth-toothed
pocket gophers (genus Thomomys) are known from the Rock Springs
deposit, but the genus now occurs only in Mexico and western North
America (Hall 1981). The range of Mormoops megalophylla is now res-
tricted to the southwestern United States, southward through much of
Central America and into northern South America (Smith 1972). Mor-
moops blainvilli, the only other extant species of the genus, occurs in
the West Indies (Hall 1981).

Paleoenvironmental Interpretations

The genus Mormoops (as well as the species M. megalophylla
itself) is neotropical and temperate in distribution and occupies humid
to semiarid to arid situations at elevations generally less than 3000 m
(Smith 1972). Similarly, Thomomys pocket gophers occupy habitats
including deserts, prairies, montane meadows and forests ranging over
some 35 degrees of latitude and 3000 m of elevation. It is apparent that
no single temperature or rainfall regime characterizes the entire range of
either genus. Therefore, because of their broad habitat and environmen-
tal tolerances, the presence of M. megalophylla or Thomomys at Rock
Springs need not be directly indicative of any particular habitat types.

80 Kenneth T. Wilkins

Yet, their presence does offer clues in reconstructing certain aspects of
the paleoenvironment.

Bats. — The two species of bats recorded as fossils at Rock Springs
are usually considered obligate cave-dwellers, although alternate roosts
are also used (Barbour and Davis 1969). Their presence suggests a
reduced piezometric surface. No modern records of M. austroriparius
are known from the Rock Springs cavern, although the species is
reported from various northwestern Orange County caverns in the
immediate vicinity (Rice 1957). The present water level in the Rock
Springs cavern is probably too high to accomodate sizable colonies of
bats. Additionally, bat bones are quite delicate; that most of the fossil-
ized bat material is intact suggests the absence of flowing water in the
cavern during the depositional intervals. Yet, some water was probably
on the cave floor during occupation by Myotis austroriparius. This is
perhaps the most abundant extant bat species in Florida and it resides
in caves with either still or flowing water beneath roosting sites. Avian
remains from Rock Springs also showed Httle evidence of water-wear
(Woolfenden 1959). As with many other stream deposits, the context
and associations of fossil specimens is unknown and cannot be recon-
structed. However, the similarity of preservation of both Mormoops
and Myotis fossils indirectly suggests that occupation of the cave could
have been contemporaneous.

Pocket gophers. — The contemporaneous occurrence of two con-
trageneric pocket gopher species in a fossil deposit was previously
reported by Dalquest and Kilpatrick (1973). They interpreted their cave
deposit (Shulze Cave, early Holocene, Edwards County, Texas) as the
roost of barn owls that foraged over a broad area, including the differ-
ent habitats occupied by Thomomys bottae and Geomys bursarius,
which had mutually exclusive microgeographic distributions. It is quite
possible that the Rock Springs cavern served as an owl roost during
times of reduced water table levels. Fossil barred owls, Strix varia, are
known from Rock Springs (Woolfenden 1959). Lack of stratigraphic
context disallows determination of whether Geomys and Thomomys
occurred contemporaneously in the Rock Springs vicinity rather than
being members of faunas of different time intervals. Nevertheless, the
mere presence in the fauna of the two species of pocket gophers allows
inferences regarding soil characteristics. Miller (1964) examined soil
preferences and competitive interactions of three genera of pocket
gophers in Colorado. He found that Geomys species required deeper,
sandier and more friable soils than either Pappogeomys species (includ-
ing Cratogeomys species) or Thomomys species. Thomomys, in con-
trast, was well suited to shallow, gravelly, less friable soils, although
Thomomys could and would inhabit deeper soils where available.
Requirements of Pappogeomys were intermediate to the other genera.

Pleistocene Mammals Central Florida 81

Miller found in competition experiments that, in its preferred deeper
soils, Geomys excluded species of other genera, whereas Thomomys was
the superior competitor in shallow soils. Hence, presence of Geomys
and Thomomys in the Rock Springs deposit suggests local occurrence
of two markedly different substrates. The deep, sandy soils predominat-
ing the vicinity today favor Geomys and generally represent sediments
of marine terraces of interglacial periods. Shallower, gravelly soils often
occupied by Thomomys in western North America are uncommon in
Florida today. However, such soils could be developed via erosion of
deep sands overlying Umestone. Shallow sands mixed with gravel formed
of eroding limestone could form a substrate inhabitable by Thomomys
but not by Geomys.

ACKNOWLEDGMENTS.— I thank Richard Franz, Jeff Franz,
Gary Morgan and Ann Pratt for field assistance. Discussions with
Richard Franz, Richard Hulbert, Gary Morgan and David Webb aided
me in identifying specimens and in making paleoecological interpreta-
tions. The efforts of Mary Ellen Ahearn, Jon Becker, Richard Hulbert,
Bruce MacFadden, Gary Morgan, Ann Pratt and David Webb in
reviewing drafts of this manuscript are greatly appreciated. This paper is
University of Florida Contribution to Vertebrate Paleontology No. 223.

LITERATURE CITED

Allen, Glover M. 1942. Extinct and vanishing mammals of the western hemis-
phere with the marine species of all the oceans. Intelligence Printing Co.,
Lancaster. 620 pp.

Auffenberg, Walter. 1963. The fossil snakes of Florida. Tulane Stud. Zool.
70(3):131-216.

Barbour, Roger W., and W. H. Davis. 1969. Bats of America. Univ. Press
Kentucky, Lexington. 286 pp.

Bloom, Arthur L., W. S. Broecker, J. M. A. Chappell, R. K. Matthews and K.
J. Mesolella. 1974. Quaternary sea level fluctuations on a tectonic coast:
230 Th/234 U dates from the Huon Peninsula, New Guinea. Quat. Res.
(N.Y.) ¥.185-205.

Chappell, John. 1974. Geology of coral terraces, Huon Peninsula, New Guinea:
a study of Quaternary tectonic movement and sea-level changes. Bull. Geol.
Soc. Am. ,^5.-553-570.

Dalquest, Walter W., and C. W. Ki;patrick. 1973. Dynamics of pocket gopher
distribution on the Edwards Plateau of Texas. Southwest. Nat. 75(1): 1-9.

Furgeson, Vera M. 1951. Chronology at South Indian River Field. Anthropol-
ogy No. 45, Yale Univ. Press, New Haven.

Genoways, Hugh H., and J. R. Choate. 1972. A multivariate analysis of syste-
matic relationships among populations of the short-tailed shrew (genus
Blarina) in Nebraska. Syst. Zool. 27.- 106-1 16.

George, Sarah B., H. H., Genoways, J. R. Choate and R. J. Baker. 1982. Kary-
otypic relationships within the short-tailed shrews, genus Blarina. J. Mam-
mal. 65(4):639-645.

82 Kenneth T. Wilkins

Gillette, David D. 1976. A new species of small cat from the late Quaternary of
southeastern United States. J. Mammal. 57(4):664-676.

Gut, H. James. 1939. Hitherto unrecorded vertebrate fossil localities in south-
central Florida. Q. J. Fla. Acad. Sci. 5:50-53.

Hall, E. Raymond. 1981. The Mammals of North America. John Wiley and
Sons, New York, 1 181 + 90 pp., 2 volumes.

Healy, Henry G. 1975. Terraces and shorelines of Florida. Fla. Dep. Nat. Re-
sour., Bur. Geol. Map Series No. 71.

Johns, Bette A. S. 1958. The distribution of the beaver in Florida and a study of
its ecololgy in the Flint-Chattahoochee-Apalachicola region. Unpubl. M.S.
thesis, Univ. Florida, Gainesville. 48 pp.

Kurt^n, Bjorn. 1966. Pleistocene bears of North America. I: Genus Tremarctos,
spectacled bears. Acta Zool. Fenn. 775:1-120.

, and E. Anderson. 1980. Pleistocene Mammals of North America.

Columbia Univ. Press, New York. 442 pp.

Martin, Robert A. 1974. Fossil mammals from the Coleman HA fauna, Sumter
County. Pp. 35-99 in S. D. Webb (ed.). Pleistocene Mammals of Florida.
Univ. Presses Fla., Gainesville. 270 pp.

Miller, R. S. 1964. Ecology and distribution of pocket gophers (Geomyidae) in
Colorado. Ecology 45:156-212.

Ray, Clayton E. 1964. The jaguarundi in the Quaternary of Florida. J. Mam-
mal. ¥5(2):330-332.

, S. J. Olsen and H. J. Gut. 1963. Three mammals new to the Pleis-
tocene fauna of Florida, and a reconsideration of five earlier records. J.
Mammal. ¥¥(3):373-395.

Rice, Dale W. 1957. Life history and ecology of Myotis austroriparius in Flor-
ida. J. Mammal. i5(l): 15-32.

Robertson, Jesse S., Jr. 1974. Fossil Bison of Florida. Pp. 214-246 in S. D.
Webb (ed.). Pleistocene Mammals of Florida. Univ. Presses Fla., Gaines-
ville. 270 pp.

Rosenau, Jack C, G. L. Faulkner, C. W. Hendry, Jr. and R. W. Hull. 1977.
Springs of Florida. Bull. Fla. Bur. Geol. 57.1-461.

Shackleton, Nicholas J., and N. D. Opdyke. 1973. Oxygen isotope and palaeo-
magnetic stratigraphy of equatorial Pacific core V28-238: Oxygen isotope
temperatures and ice volumes on a 10^ year and 10^ year scale. Quat. Res.
(N.Y.)5(l):39-55.

Simpson, George G. 1928. Pleistocene mammals from a cave in Citrus County,
Florida. Am. Mus. Novit. 525:1-16.

Smith, James D. 1972. Systematics of the chiropteran family Mormoopidae.
Univ. Kans. Mus. Nat. Misc. Publ. 56.1-132.

Webb, S. David (ed.). 1974. Pleistocene Mammals of Florida. Univ. Presses
Fla., Gainesville. 270 pp.

Wilkins, Kenneth T. 1984. Evolutionary trends in Florida Pleistocene pocket
gophers (genus Geomys), with description of a new species. J. Vertebr.
Paleontol. 5(3):166-181.

Woolfenden, Glen E. 1959. A Pleistocene avifauna from Rock Springs Florida.
Wilson Bull. 77(2): 183- 187.

Accepted 8 August 1983

Bird Density and Habitat Use in Forest Openings Created

by Herbicides and Clearcutting in

The Central Appalachians

William C. McComb and Robert L. Rumsey ^

Department of Forestry, University of Kentucky

Lexington, Kentucky 40546-0073

ABSTRACT. — Winter and breeding bird communities were sampled
on clearcuts, on plots receiving four rates of picloram herbicide appli-
cation, and on untreated plots. Four-year-old picloram treated plots
did not increase bird species diversity, but did increase bird density
over untreated plots. Positive response of some bird species to treat-
ment was due to creation of an edge. Bird density was affected by
physiographic site. Hairy woodpeckers, Picoides villosus; wintering
red-bellied woodpeckers, Melanerpes carolinus; and Carolina chicka-
dees, Parus carolinensis, used south-facing slopes more than north-
facing slopes, while red-eyed vireos, Vireo olivaceus, and ovenbirds,
Seiurus aurocapillus, used north-facing slopes more than ridgetop
sites. White-breasted nuthatches, Sitta carolinensis', red-bellied wood-
peckers; tufted titmice, Parus bicolor; and Carolina chickadees exhi-
bited changes in habitat use from winter to spring. A variety of piclo-
ram herbicide treatments (27-68 kg/ ha) and clearcutting in small (0.5-
1.0 ha) blocks, on both south-facing and north-facing slopes, is
recommended to increase bird density and provide more diverse habi-
tat than exists in undisturbed forests.

INTRODUCTION

Herbicides are used by foresters for timber stand improvement and
by wildlife managers to create forest openings (McCaffery et al. 1974;
Loftis 1978; Dewey 1980). Any herbicide use or timber harvesting
affects the structure and composition of forests as wildlife habitat. The
effects of forest cutting on bird communities have been reported by
McComb and Noble (1980), Strelke and Dickson (1980), and Crawford
et al. (1981), who found that tree thinning tended to increase bird spe-
cies diversity (BSD) and bird abundance, probably due to an increase in
foliage height diversity of the stand. In contrast, little information is
available on the effects of herbicide application on forest bird communi-
ties. Beaver (1976), Savidge (1978), and Best (1972) reported the effects
of 2,4,5-T and 2,4-D applications to shrubs and trees on birds and
found either no effects or reductions in bird density and /or diversity.
Shipman (1972) reported greater wildlife diversity for up to 10 years
after application of fenuron herbicide on 0.04-ha plots in Pennsylvania.

Present address: Department of Agriculture, McNeese State University,
Lake Charles, Louisiana 70609

Brimleyana No.9:83-95. June 1983. 83

84 William C. McComb and Robert L. Rumsey

Picloram-based herbicides are desirable for creating forest openings
because of their low toxicity to many vertebrates (Kenaga 1969), but we
could find no studies that compared bird use of picloram-created clear-
ings with use of clearcut or uncut areas. Our objectives were to: 1) com-
pare the relative abundance and diversity of winter birds and breeding
birds among 4-year-old herbicide-created forest clearings, 4-year-old
clearcuts, and mature untreated areas, 2) evaluate the effect of edge
creation on bird use of habitats, and 3) identify the characteristics of
habitats used by common species.

STUDY AREA AND METHODS

Snag Ridge Fork watershed, in the University of Kentucky's
Robinson Forest, Knott and Breathitt counties, Kentucky, contains a
second-growth mixed-mesophytic forest typical of much of the central
Appalachians (Carpenter and Rumsey 1976). Ridges are dominated by
shortleaf pine, Pinus echinata\ pitch pine, P. rigida; chestnut oak, Quer-
cus prinus; and scarlet oak, Q. coccinea. South-facing slopes are domi-
nated by hickories, Carya spp.; white oak, Q. alba; black oak, Q. velu-
tina; and sourwood, Oxydendrum arboreum. North-facing slopes are
dominated by northern red oak, Q. rubra; cucumbertree. Magnolia
acuminata; and yellow-poplar, Liriodendron tulipifera.

Fifteen of 18 0.4-ha square plots were treated in the watershed.
This included four plots on each of a north-facing slope, south-facing
slope, and ridge-top, which were randomly assigned one of the follow-
ing hand-broadcast herbicide treatments applied in May 1976: 23 kg/ ha
TORDON 10k; 45 kg/ha TORDON lOK; 68 kg/ha TORDON lOK; or
90 kg/ ha M-3864. TORDON lOK is a pelletized picloram-based (10%
4-amino-3,5,6-trichloropicolinic acid) herbicide and M-3864 is a 5% pic-
loram pellet. (Mention of trade names is for identification and does not
imply endorsement by the Kentucky Agricultural Experiment Station,
Lexington.) A fifth plot on each aspect was clearcut (all stems cut, no
residuals); fallen trees were not removed. A sixth plot on each aspect
was established as a control in the untreated forest at least 75 m from
any treated plot. Plots were located along the contour, and treated plots
were 15 m to 50 m apart.

Fifteen stations were established perpendicular to the contour
through the center of each plot. Thirty environmental characteristics
(Table 1) chosen based on Anderson and Shugart (1974), Stauffer and
Best (1980), and Crawford et al. (1981) were measured at each station.
Estimates of cover of rocks, logs, leaves, canopy, and midstory followed
methods described by James and Shugart (1971). Environmental char-
acteristics other than understory vegetation were measured in May
1980. Understory vegetation was quantified on one 4-m^ circular plot
(radius = 1.2 m), 2 m away from each station along the contour in

Bird Response to Clearings 85

Table 1. Habitat variables used in correlation analyses, bird habitat preference
on clearcut, herbicide-treated and uncut plots, January to June 1980
and 1981, Knott County, Kentucky.

Description

Number of trees>10 cm dbh, within 2 m of a station.

Diameter of nearest tree (cm).

Distance to nearest tree (m).

Basal area of living stems (m^/ha), wedge prism.

Number of snags> 10 cm dbh, > 1 .8 m tall, within 2 m of a station.

Diameter of nearest snag (cm).

Distance to the nearest snag (m).

Number of stumps>10 cm in diameter and<1.8 m tall, within 2 m of a station.

Distance to the nearest stump (m).

Number of logs>10 cm diameter,>1.8 m long, within 2 m of a station.

Maximum diameter of the nearest log (cm).

Length of the nearest log (m).

Distance to the nearest log (m).

Percent of ground covered by logs within 2 m of a station.

Distance of the nearest rock>5 cm above ground.

Percent of ground covered by rocks within 2 m of a station.

Percent crown cover above 6.1 m at a station.

Percent vegetation cover between 1.8 and 6.1 m at a station.

Percent of the ground covered by fallen leaves within 2 m of a station.

Percent understory cover<1.8 m tall per 4-m^ in Jan. and in Apr.

Number of understory stems per 4-m^ in Jan. and in Apr.

Number of understory taxa per 4-m^ in Jan. and in Apr.

Understory species diversity per 4-m^ in Jan. and in Apr.

Distance to water (m).

Slope (%).

Foliage height diversity.

January and April of 1980. A modified Aldous method, similar to that
described by Murphy and Noble (1972), was used to quantify vegeta-
tion. Percent cover and stem density were determined for each plant
taxon (at least to genus) on each 4-m^ plot. Total cover, total stem
density, plant species richness, and a Shannon-Weaver plant species
diversity index were calculated for each station for both sampling peri-
ods. April understory values were used in conjunction with May mid-
story and overstory values to calculate a foliage height diversity index
based on estimated cover of the three layers. Mean habitat characteris-
tics were calculated for each 0.4-ha plot and used to characterize the
plot. A more detailed discussion of habitat characteristics was given by
McComb and Rumsey (in press).

86 William C. McComb and Robert L. Rumsey

Birds were counted 15 times on each 0.4-ha plot from 15 January
to 5 March 1980 and 1981 (winter birds), and 13 times on each plot
from 20 March to 15 June 1980 and 1981 (breeding birds). Approxi-
mately 10 minutes were spent on the center of each plot during each
visit counting birds seen or heard on the plot. Locations of birds in the
plot were judged to be within 9 m of the plot edge or within the plot
center. Only winter birds observed foraging were assumed using a plot
while breeding birds which were either singing or foraging were tallied.
Birds were counted within three hours of sunrise or sunset. Eight morn-
ing and seven evening visits were made to each plot during winter, and
seven morning and six evening visits were made to each plot during the
breeding season. Shannon-Weaver BSD, equitability, richness, and
average density per plot per visit of each species were compared among
treatments blocking on aspects with analysis of variance and Duncan's
New Multiple Range Test. Tests among treatments are conservative due
to variability among aspects. Linear correlation was used to identify
plot characteristics potentially important in determining use by common
bird species. A t-test was used to compare weighted mean environmental
characteristics for plots used by a species vs. overall means and to com-
pare mean BSD, equitabiHty, richness, and density between seasons.
Chi-square analysis was used to compare occurrences of birds at the
edge vs. center of plots.

RESULTS AND DISCUSSION

Seasonal Effects. — Not unexpectedly, the breeding bird communi-
ties had higher diversity (x = 0.91 1), richness (x = 2.2), equitability (x =
0.238), and density (x = 2.4 birds/ 0.4 ha) than the winter bird commun-
ities (x = 0.265, 1.0, 0.069, and 1.0, respectively) (Table 2). Winter bird
communities were dominated by hairy woodpeckers, Picoides villosus,
and Carolina chickadees, Parus carolinensis (Table 2), and the breeding
bird communities were dominated by red-eyed vireos, Vireo olivaceus,
and hooded warblers, Wilsonia citrina (Table 3). Several resident spe-
cies changed their habitat usage from winter to spring. White-breasted
nuthatches, Sitta carolinensis, used areas in spring with a higher den-
sity (x = 1/43 m^) of larger diameter (x = 1.23 cm) snags and fewer trees
(x = 1/32 m^) than in winter (x = 1/171 m^, 7.9 cm and 1/27 m^ respec-
tively) (P<0.05). Red-bellied woodpeckers, Melanerpes carolinus, used
areas in spring with more snags (x = 1/50 m^), higher understory density
(x = 45.2 stems/ 4 m^), and lower foliage height diversity (x = 1.23) than
in winter (x = 1/171 m^ 34.7 stems/4 m^, and 1.40, respectively)
(P<0.05). Both white-breasted nuthatches and red-bellied woodpeckers
probably used areas with higher snag density in spring for nesting pur-
poses, since both will forage on dead or living trees in the winter
(Conner 1979).

Bird Response to Clearings

87

■a >

o ^

o S

■5 £:

"2 s:

3 -5

1.2

i 2

^ "S ^

o o

.z. o .:2

>/^

ON

sO

sO

rsi

0

d

d

00

Tt

vo

-^

0

6

d

-*

u-i

-^

—

0

d

d

m

r--

ON

m

0

d

d

ir^

00

1^

ON

r»-)

0

d

d

>/-i

0

0

00

r^

0

d

d

r<-,

0

0^

10

0

d

d

>^

a>

>

■a

C/5

>,

.ii

0

1

03

■0

3

a-

m

UJ

U g V,

C <3 O
> - C

2 J^

3 .

u-' X)

3 :| '^

^ ^ •-

.Si 5 ^

X) o *i

oJ O 4J

i If

■n 1- ^

O o

0.1 a,

c ^

<U 73

^ 3
C Q

>. o

1 i

' S S S
t '^ ^

■II -■

■~^ , s:

0 w 3

£^■50
T3 5 lU

'E 2 o
^ .2 ..

^'^ I

2 2 5

? •? a

1 S "
« s: 5

to <3 ^
^' U (U

■^ -^ .5
c .^'-^

o ^ V

•2 c

3

William C. McComb and Robert L. Rumsey

o

o

ON

'^

oo

00

so

flj

-o

o

X)

m
-^

(U

K

m

(N

CQ

03
-a o

X)

> if

> Co

!> .2

x; a

? o

X

U

-o >
o ^
o ^

X

Bird Response to Clearings

89

2 S

— , oo

— f^

ON rsi

o o

-_; oo

-:; ON

oo

O

d

q

OO

-"

d

ON

o

o :: -

O —
O <N

^ O

OO —

so O

IZ o

<L>

'^

M

3

O

•S

(U

o

T3

v^

O

a

O

^

^

2

-o

Ci,

a>

c

<i>

M

c

u

?3

X5

^

"S

^

= 1

O <L>
o -f.

Si

(U t>c
o ^

2 ^ '^

.2 -

o

= <

Oil

^ ^ .

"S ^ -5

^ u -^

C (U 5^

en ap O

o 03 is

^ B ?\

< 1

03

(u <u •!::

^ >

C GO

1^

O (U
00 ^

-o o
<u o

I i

^ ^'

Iff

-^ '^ "^5
t- 5 3

03 •> I-

- cd

*i -2 2 X)

03 >;

° I

o ^

^1

O ^
^ 00

uT ^ X)

-So
X) Ci, OX)

i 1 1

CT3 V^ • -

^ C "<

2-Q ^

90 William C. McComb and Robert L. Rumsey

Carolina chickadees used areas with greater (P<0.05) understory
species richness in the spring (x = 7.3 species/ 4 m^) than in the winter (x
= 6.6), and tufted titmice used areas with lower (P<0.05) leaf cover (a
more open canopy) (x = 46.4%) in the spring than in the winter (x =
63.5%).

Winter Birds. — Bird species diversity, equitability, and richness did
not vary significantly among treatments nor among aspects (P>0.05).
Winter bird density was higher on clearcuts, 23 kg/ ha, and 45 kg/ ha
plots than on the control plots (P<0.05) (Table 2), and bird density on
ridge-tops (x = 1.1 bird/ plot/ visit) (P<0.05) was higher than on north
facing slopes (x = 0.53 birds/ plot/ visit) (P<0.05). South-facing and rid-
getop areas are warmer than north-facing slopes throughout the day-
light hours and probably influenced bird activity (Shields and Grubb
1974).

Hairy woodpecker abundance was similar among treatments, but
they used south-facing slopes (17 individuals) more than north-facing
slopes (7 individuals), and they were observed using plot edges (20 indi-
viduals) more than plot centers (4 individuals) (P<0.05). Carolina
chickadee abundance was similar among treatments, but they used
ridge-tops (25 individuals) more than north-facing slopes (5 individuals)
(P<0.05). Strelke and Dickson (1980) found breeding Carolina chicka-
dees associated with pine-hardwood clearcut edges in Texas. Carolina
chickadee numbers were correlated with distance to the nearest stump
(r =+ 50.6), slope (r =- 50.5), and understory density (r =+ 49.4) (P< 0.05).
Distance to stumps on plots used by chickadees (23.2 m; SD = 7.1)
differed significantly (P<0.05) from overall stump distance (17.4 m; SD
= 7.9). Large, soft stumps may afford roosting sites for chickadees.

White-breasted nuthatch abundance was similar among aspects
(P>0.05), but more individuals were observed on 45 kg/ ha plots than
on any other treatment except 23 kg/ ha plots (P<0.05) (Table 2).
White-breasted nuthatch numbers were correlated with log length (r =
+ 51.1). Log presence indicates prior snag presence so nuthatches may be
using an area that recently had snags present. Plots used by wintering
white-breasted nuthatches had higher crown cover (x = 57.5%), lower
understory diversity (x = 4.73), and lower understory density (x = 26.3
stems/ 4 m^) than overall plot means (x = 31.6, 5.15, and 36.2, respec-
tively). Wintering white-breasted nuthatches occurred on plots with
open crowns, sparse understory, and many long logs on the ground.

Red-bellied woodpeckers used ridge-tops (13 individuals) more
than south- or north-facing slopes (2 individuals each) and they were
found more frequently at plot edges (12 individuals) than centers (5
individuals) (P<0.05). Red-bellied woodpecker numbers were negatively
correlated with slope (r = 45.7), a result of their use of ridge sites.

Bird Response to Clearings 91

Wintering northern cardinals, Cardinalis cardinalis, used clearcuts
more than other treatments except 23 kg/ ha plots (P<0.05) (Table 2).
Both treatments resulted in dense understory vegetation (>100%) and
low basal area (<17.5 m^ha). Cardinals showed no association with
edge. Northern cardinal abundance was correlated with snag and tree
distance (r =+ 62.4, + 57.7), and basal area (r =- 54.1). Snags provide
singing perches for cardinals.

Rufous-sided towhees, Pipilo erythrophthalmus, used clearcuts
more than any other treatment (P<0.05) (Table 2). Clearcuts had the
lowest basal area (9.3 m^/ha 4-year-old regrowth) and the highest
ground cover (logs, 3.5/ 12 m^; rocks, 6.7% cover; stumps, 0.8/ 12 m^) of
any of the plots. Towhee numbers were correlated with distance to the
nearest tree (r =+ 80.0) and basal area (r =- 70.3) (P<0.01). Trees near
young clearcuts probably provide singing perches for towhees. These
results concur with those of Crawford et al. (1981) and Conner and
Adkisson(1975).

Sample sizes for 13 other winter resident species were too small to
detect meaningful differences among treatments or aspects (Table 2).

Breeding Birds. — We found no differences in BSD or equitability
among treatments or aspects (P>0.05). Average breeding bird density
was higher on 23 kg/ ha plots than on other treatments except 45 and 68
kg/ ha plots (P<0.05) (Table 3). Density was higher on south- and
north-facing slopes (x = 1.6 birds/ plot/ visit) than on ridge-tops (x = 1.2
birds/ plot/ visit) (P<0.05). Smith (1977) reported that moist forests
were of more importance to most species of breeding birds in Arkansas
than extremely dry sites.

The red-eyed vireo was the most abundant species and it used
south- and north-facing slopes (31 and 26 individuals) more than ridge-
tops (13 individuals), and edges (54 individuals) more than centers (16
individuals) (P<0.05). Because edges of most plots had more complete
canopy development than plot centers (except control plots), it is not
surprising that red-eyed vireos used edges. Anderson and Shugart
(1974), Stauffer and Best (1980), and Crawford et al. (1981) reported
red-eyed vireos associated with closed canopies and sparse understory.
We also found that red-eyed vireo numbers were correlated with tree
density (r=+ 47.8).

Hooded warblers, Wilsonia citrina, were ubiquitous over treat-
ments and aspects. Plots used by hooded warblers had lower basal area
(x = 7.1 m^/ha) than plots not used (18.6 m^ha) (P<0.05). Occurrence
of hooded warblers was associated with understory diversity (r =+ 59.3)
and understory density (+ 50.1), similar to Anderson and Shugart 's
(1974) findings.

Black-throated green warblers, Dendroica virens, were ubiquitous
over treatments and aspects. Plots used by black-throated green

92 William C. McComb and Robert L. Rumsey

warblers had less leaf cover (x = 26.4%) (hence lower crown cover),
greater slope (x = 49.0%), and higher understory diversity (x = 5.49)
than overall plot averages (x - 56.0, 36.2, and 5.15, respectively).
Freedman et al. (1981) reported adverse effects of forest cutting on
black-throated green warblers in Nova Scotia, but we found this species
tolerant of forest disturbance in Kentucky, though densities were low on
clearcut, 45 kg/ ha and 90 kg/ ha plots.

Ovenbirds, Seiurus aurocapillus, used control and 90 kg/ ha plots
more than clearcut and 68 kg/ ha plots (P<0.05) (Table 3). The 90
kg/ ha of M-3864 resulted in dense understory (x = 105%), partial mid-
story (x = 13%), and a discontinuous overstory (x = 31.9%) (McComb
and Rumsey, in press). Further, ovenbirds used edges more frequently
(13 individuals) than plot centers (2 individuals) (P<0.05), and they
were found more often on north-facing slopes (10 individuals) than on
ridges (1 individual) (P<0.05). Allaire (1978), Robbins (1979), Stauffer
and Best (1980) and Crawford et al. (1981) indicated that ovenbirds are
a forest interior species and that they are adversely affected by forest
disturbance or edge presence, but our study and that of Freedman et al.
(1981) indicated that ovenbirds will occur on edges of some disturbed
areas surrounded by undisturbed forest. Plots where ovenbirds occurred
in our study had higher crown cover (x = 49.7%), leaf cover (x = 73.2%),
basal area (x = 41.7 m^/ha), foliage height diversity (x = 1.44), lower log
density (x = 0.8/ 12 m^), and a shorter distance to water (x - 189.8 m)
than overall plot means (x = 31.5, 56.0, 33.3, 1.30, 1.5, and 335.7,
respectively). Ovenbird numbers were correlated with basal area (r =
+ 54.6), crown cover (r -+ 48.0), and distance to water (r =- 47. 1).

Kentucky warblers, Oporornis formosus, used 68 kg/ ha plots more
than control and clearcut plots (P<0.05) (Table 3). Abundant (x = 1/31
m^) and large-diameter (x = 34.8 cm) snags, low basal area (17.2 m^/ha),
and low midstory cover (17.1%) characterized 68 kg/ ha plots. Kentucky
warblers were found on plots with higher snag density (x = 1/13 m^),
rock cover (x - 8.8%), and lower midstory cover (x = 15.5%), snag dis-
tance (x = 16.2 m), rock distance (x = 14.9 m), and log length (x = 19.7
m), than overall averages ( x = 1/20, 3.3, 28.0, 3.4, 32.8, and 28.4, respec-
tively). Kentucky warbler numbers were correlated with midstory cover
(r =- 52.3) and snag diameter (r =+ 48.6). Kentucky warbler habitat on
our plots was characterized by many logs, rocks and large snags and
little midstory on sites away from permanent water. Log and rock pres-
ence may be important as a nest location for this ground-nesting
species.

Tufted titmice had no association with aspect or edge, but they
were observed most frequently on 68 kg/ ha plots (P<0.05) (Table 3).
The abundance of large-diameter snags (x = 34.8 cm dbh) that occurred
on 68 kg/ ha plots may have influenced use of these plots by this cavity-

Bird Response to Clearings 93

nesting species. Stauffer and Best (1980) reported this species tolerant of
habitat disturbance and Crawford et al. (1981) found titmice using a
wide range of habitats.

Red-bellied woodpeckers used control and 45 kg/ ha plots more
frequently than clearcut or 23 kg/ ha plots (P<0.05) (Table 3). Control
and 45 kg/ ha plots were characterized by having higher basal area (x =
22.5 m^ha) than either clearcut (9.3 m^ha) or 23 kg/ ha plots (17.5
m^/ha). Red-bellied woodpeckers exhibited no association with aspect
or edge, but hairy woodpeckers used ridge-tops (9 individuals) more
(P<0.05) than north- or south-facing slopes (3 and 2 individuals, respec-
tively). Red-bellied woodpeckers used habitats with an open understory
(high crown cover) and high basal area.

Sample sizes were insufficient to identify treatment, aspect, or habi-
tat preferences of 22 other species of breeding birds encountered during
the course of our study (Table 3).

MANAGEMENT IMPLICATIONS

Long-term effects of picloram herbicide application in field situa-
tions on soil invertebrates, invertebrates inhabiting herbicide-created
snags, plant succession, and reproductive physiology of terrestrial verte-
brates have not been thoroughly investigated. If future studies indicate
minimal effects of picloram herbicides on these ecological processes,
then pelletized picloram application may be a more desirable and less
expensive non-game bird management tool in remote or rugged terrain
than manually cutting trees. Four years after application of picloram
herbicide or clearcutting, increases in BSD through changes in species
composition and increased density of some species in the forest may be
expected, but no single treatment will accomplish this goal. Although
we found no differences in BSD among treatments, some species pre-
ferred or occurred exclusively on one or several treatments.

Total bird density was higher on treated plots than control plots in
both winter and spring. This may have been due to the presence of edge
for some species (wintering hairy and red-bellied woodpeckers, red-eyed
vireos, and ovenbirds) and/ or to the changes in the biotic and abiotic
habitat characteristics produced by the treatment. For instance, changes
in habitat characteristics due to clearcutting produced predictable
occurrences of rufous-sided towhees and cardinals. Changes in habitat
structure brought about by herbicide application allowed predictable
occurences of some species (e.g., Kentucky warblers and ovenbirds) but
not others (black-throated green warblers and tufted titmice).

We recommend that a range of pelletized picloram rates from 27 to
68 kg/ ha be used on small plots (0.5 - 1.0 ha) in conjunction with small
clearcuts to provide desirable habitat requirements for as many species
of birds as possible, and to increase the density of many of these species

94 William C. McComb and Robert L. Rumsey

over what would occur in an undisturbed forest. Ninety kg/ ha of
M-3864 was not a desirable treatment since none of the species used
plots with this rate over one of the other treatments. If ten plots per
1,000 ha were treated each year, habitat would be provided on a sus-
tained basis while allowing 100 years for recovery on each site.
Herbicide-treated plots and clearcuts placed on both north-facing and
south-facing slopes would increase the advantages to breeding migrants
and wintering residents or breeding Picids, respectively.

ACKNOWLEDGMENTS.— We wish to thank R. L. Anderson,
D. M. Allen, M. J. Immel, and C. Rowell for advice on statistical ana-
lyses; Dow Chemical Corp. for supplying herbicides; J. B. Davis and the
Robinson Forest staff for assistance with field work; and R. N. Conner,
P. D. Doerr, B. A. Thielges, P. N. Allaire, R. E. Noble, R. B. Hamilton,
R. D. Shipman, R. N. Muller, and C. J. Liu for reviewing an early draft
of the manuscript. The investigation reported in this manuscript (No.
82-8-40) is in connection with Kentucky Agricultural Experiment Sta-
tion Project No. 620 and is published with the approval of the Director.

LITERATURE CITED

Allaire, Pierre N. 1978. Effects on avian populations adjacent to an active
stripmine site. Pp. 232-240 in D. E. Samuel, J. R. Stauffer and W. T.
Mason (eds.). Surface mining and fish/ wildlife needs in the eastern United
States. U. S. Dep. Inter., Fish Wildl. Serv. Program FWS - OBS 78/81.

Anderson, Stanley H., and H. H. Shugart, Jr. 1974. Habitat selection of breed-
ing birds in an east Tennessee deciduous forest. Ecology 55:828-837.

Beaver, Donald L. 1976. Avian populations in herbicide treated brush fields.
Auk 95:543-553.

Best, Louis B. 1972. First-year effects of sagebrush control on two sparrows. J.
Wildl. Manage. 36:534-544.

Carpenter, Stanley B., and R. L. Rumsey. 1976. Trees and shrubs of Robinson
Forest, Breathitt County, Kentucky. Castanea ^^7:277-282.

Conner, Richard N. 1979. Snag management for cavity-nesting birds. Pp. 120-
128 in R. M. DeGraaf (Tech. Coord.). Proceedings of the Workshop on
Management of Southern Forests for Nongame Birds. U. S. Dep. Agric.
For. Serv. Gen. Tech. Rep. SE-14.

, and C. S. Adkisson. 1975. Effects of clearcutting on the diversity of

breeding birds. J. For. 7i:78 1-785.

Crawford, Hewlette S., R. G. Hooper and R. W. Titterington. 1981. Songbird
response to silvicultural practices in central Appalachian hardwoods. J.
Wildl. Manage. 45:680-692.

Bird Response to Clearings 95

Dewey, John B. 1980. "Gridball Pellets" — a new tool for brush control in
pines. For. Farmer ^0(2): 14-15, 34.

Freedman, B., C. Beauchamp, I. A. McLaren and S. I. Tingley. 1981. Forestry
management practices and populations of breeding birds in a hardwood
forest in Nova Scotia. Can. Field-Nat. 95:307-31 1.

James, Francis C, and H. H. Shugart, Jr. 1971. A quantitative method of habi-
tat description. Audubon Field Notes 2'/:727-736.

Kenaga, Eugene E. 1969. Tordon herbicide — evaluation of safety to fish and
birds. Down-to-Earth 25:5-9.

Loftis, David L. 1978. Preharvest herbicide control of undesirable vegetation in
southern Appalachian hardwoods. South. J. Appl. For. 2:51-54.

McCaffery, Kenneth R., F. L. Johnson and L. D. Martoglio. 1974. Maintaining
wildlife openings with pellets containing picloram. Wildl. Soc. Bull. 2:40-45.

McComb, William C, and R. E. Noble. 1980. Small mammal and bird use of
some managed and unmanaged forest stands in the mid-south. Proc. Annu.
Conf. Southeast. Assoc. Fish Wildl. Agencies 5^:482-491.

, and R. L. Rumsey. In press. Habitat characteristics of forest clear-
ings created by picloram herbicides and clearcutting. Proc. Annu. Conf.
Southeast. Assoc. Fish Wildl. Agencies 55.

Murphy, Patrick K., and R. E. Noble. 1972. The monthly availability and use of
browse plants by deer on a bottomland hardwood area in Tensas Parish,
Louisiana. Proc. Annu. Conf. Southeast. Assoc. Fish Wildl. Agencies
2(5:39-57.

Robbins, Chandler S. 1979. Effect of forest fragmentation on bird populations.
Pp. 198-212 in R. M. DeGraaf (Tech. Coord.). Proceedings of the Work-
shop on Management of North Central and Northeastern Forests for Non-
game Birds. U. S. Dep. Agric. For. Serv. Gen. Tech. Rep. NC-51.

Savidge, Julie A. 1978. Wildlife in a herbicide-treated Jeffrey pine plantation in
eastern California. J. For. 7(5:476-478.

Shields, William M., and T. C. Grubb, Jr. 1974. Winter bird densities on north
and south slopes. Wilson Bull. 5(5:125-130.

Shipman, Robert D. 1972. Converting low-grade hardwood forests to Japanese
larch with fenuron herbicides. Tree-Planters' Notes 2^:1-3.

Smith, Kimberly G. 1977. Distribution of summer birds along a forest moisture
gradient in an Ozark watershed. Ecology 55:810-819.

Stauffer, Dean F., and L. B. Best. 1980. Habitat selection by birds of riparian
communities: evaluating effects of habitat alterations. J. Wildl. Manage.
^^:1-15.

Strelke, William K., and J. G. Dickson. 1980. Effect of forest clearcut edge on
breeding birds in east Texas. J. Wildl. Manage. 44:559-561.

Accepted 29 April 1983

Drainage Records and Conservation Status Evaluations
for Thirteen Kentucky Fishes

Melvin L. Warren, Jr. ' and Ronald R. Cicerello

Kentucky Nature Preserves Commission,

Frankfort, Kentucky 40601

ABSTRACT.— Recent ichthyofaunal surveys in Kentucky have
resulted in drainage records, refinement of distributional patterns, and
evaluation of conservation status for 13 fish species. The following
species are recorded for the first time from the Kentucky drainages
indicated in parentheses: Lampetra appendix (Cumberland River),
Umbra limi (West Fork Clarks River), Hybognathus hayi (Tennessee
River), Hybopsis insignis (Red River), Notropis ariommus (Kinnicon-
ick Creek), Notropis telescopus (Green River), Lepomis marginatus
(Tennessee River), Etheostoma camurum (Licking River), Percina
phoxocephala (Red River), and Percina shumardi (Little Sandy River).
In addition, the continued existence of Percina macrocephala in the
Barren River system and Kinniconick Creek, and of Fundulus chryso-
tus in extreme southwestern Kentucky, are confirmed. Evaluation of
recently acquired distributional data necessitates recommended changes
in the conservation status (established by the Kentucky Academy of
Science) of four species: (1) Lepisosteus oculatus should be removed
from the threatened category and reassigned to special concern status;

(2) Hybopsis insignis and Fundulus chrysotus should be elevated from
special concern to threatened and endangered status, respectively; and

(3) Percina phoxocephala does not warrant conservation status recog-
nition.

INTRODUCTION

The freshwater ichthyofauna of Kentucky is one of the most speci-
ose in North America, ranking third behind Tennessee and Alabama
(Burr 1980). Nevertheless, attempts to thoroughly document the distri-
bution and conservation status of this fauna have only recently been
reaHzed through publication of Clay*s (1975) book on Kentucky fishes,
an updated distributional checkHst (Burr 1980), and a Hst of endangered
and threatened Kentucky fishes endorsed by the Kentucky Academy of
Science (Branson et al. 1981b). These efforts stimulated renewed interest
in the Kentucky fish fauna, with emphasis on taxonomic status, refine-
ment of distributions, and re-evaluation of conservation status as exem-
plified by Starnes and Starnes (1978, 1979), Bauer and Branson (1979),
Burr and Mayden (1979), Burr et al. (1980), Starnes (1981), Warren
(1981), Page and Burr (1982), Warren and Cicerello (1982), and others.

' Present address: Department of Zoology, Southern Illinois University at Car-
bondale, Carbondale, Illinois 62901.

Brimleyana No.9:97-109. June 1983. 97

98 Melvin L. Warren, Jr. and Ronald R. Cicerello

Recent ichthyofaunal surveys conducted under the auspices of the
Kentucky Nature Preserves Commission have further contributed to the
refinement of distributional patterns and evaluation of the Kentucky
Academy of Science conservation status (Branson et al. 1981b) of sev-
eral fish species. It is the primary purpose of this report to elucidate and
summarize these findings for a better understanding of Kentucky's speci-
ose ichthyofauna.

SPECIES ACCOUNTS

Several new drainage records, significant distributional informa-
tion, and conservation status evaluations for 13 Kentucky fishes are
presented in the following species accounts. Each account includes col-
lection numbers followed in parentheses by the number of specimens,
standard or total length (TL) range in millimeters, stream and major
drainage, locality, county, and date of collection. All scientific and
common names follow Robins et al. (1980). Collecting materials and
methods were the same as those presented by Harker et al. (1980). All
collections are housed at the Kentucky Nature Preserves Commission
(KNP), pending deposition at Southern Illinois University at Carbon-
dale (SIUC). A number of specimens, as noted, are deposited at the
Kentucky Department of Fish and WildHfe Resources (KFW), the Ken-
tucky Department of Transportation (KDOT), or SIUC.

Lampetra appendix (DeKay). American brook lamprey. KNP uncat.
(1, 182 TL), Cumberland R. (Ohio R. dr.), 30 m below the mouth of
Sulphur Cr., Monroe Co., 22 October 1982.

Burr (1980) considered Lampetra appendix (as L. lamottei) to be
occasional in the upper Barren, Green, Kentucky, and Big Sandy rivers
of Kentucky. The specimen reported here is the first published record
for the Cumberland River of Kentucky, and one of three from the entire
drainage (Rohde 1980). Based on development of dentition, eyes, and
fins, the specimen was judged a sub-adult in the latter stages of trans-
formation; myomeres numbered approximately 69-70. Seagle and Nagel
(1982) noted that metamorphosis of this species in streams of eastern
Tennessee occurred from mid-August through October; this is sup-
ported by capture of a metamorphosing specimen in Kentucky in late
October. The specimen was collected from shallow (15 cm), clear water
with moderate current over a gravel riffle near the shore. When col-
lected the specimen was sluggish, relatively easy to capture, and hemor-
rhaging at the base of the dorsal fins. It should be noted that the Cum-
berland River was at unusually low flow and high water temperatures
due to temporary curtailment of hypolimnetic discharge from upstream
Wolf Creek Dam. Extensive seining in the area produced only a few
specimens of Coitus carolinae (Gill) and a single juvenile Etheostoma

Kentucky Fishes 99

rufilineatum (Cope). We speculate that the lamprey was a waif from
nearby Sulphur Creek, which more closely approximates the preferred
small river and stream habitat (Rohde 1980). Lampetra appendix is
regarded as a species of special concern in Kentucky (Branson et al.
1981b).

Lepisosteus oculatus (Winchell). Spotted gar. KNP Tn04GRV (1,
464 TL; 1 not retained), unnamed wetland (W. Fk. Clarks R. dr.), 0.9
km ENE Clear Springs, Graves Co., 4 May 1982; KNP OOl MCC (1,380
TL; 1 not retained). Metropolis Lake (Ohio R. dr.), 2.3 km N KY 1420
and KY 996 jet, McCracken Co., 6 June 1982; KNP M02CAE (1, 603
TL), Fish Lake (Mississippi R. dr.), at Burkley, Carlisle Co., 20 June
1982; KNP M05HIC (1, 557 TL; 1 not retained), Obion Cr. (Mississippi
R. dr.), 0.1 km downstream from KY 307 crossing, Hickman Co., 22
June 1982.

In Kentucky, the spotted gar was thought to be restricted to the
mainstem and tributaries of the Mississippi, lower Ohio (exclusive of
the lower Tennessee and Clarks rivers), and lower Cumberland rivers
(Burr 1980), until specimens from the Tradewater and Green river
drainages were reported by Warren and Cicerello (1982). More recently.
Rice et al. (1983) pubUshed the first substantiated Kentucky record for
L. oculatus from the lower Tennessee River drainage and, in additon,
six records for locaHties in CarUsle, Fuhon, and Hickman counties. In
light of these records and those presented herein, we believe a reapprai-
sal of the threatened status assigned to L. oculatus by Branson et al.
(1981b) is warranted. The spotted gar is more widespread in Kentucky
than previously thought despite the continuing loss of wetland habitat
favored by the species (Trautman 1981; Warren and Cicerello 1982).
The apparent rarity of the species is probably a result of the difficulty of
sampling the preferred vegetated, wetland habitat which has only recently
begun to be surveyed in Kentucky. We recommend that the status of L.
oculatus be changed from threatened to special concern, so that the
impact of habitat modification can be monitored.

Umbra limi (Kirtland). Central mudminnow. KNP M02GRV (1,
74), unnamed wetland (Terrapin Cr. dr.), 1.45 km S KY 97 and KY
1485 jet at Bell City, Graves Co., 27 April 1982; KNP Tn03GRV (1, 64),
old channel (W. Fk. Clarks R. dr.), at KY 131 crossing. Graves Co., 4
May 1982; KNP Tn04GRV (2, 46-53), unnamed wetland (W. Fk. Clarks
R. dr.), 0.9 km ENE Clear Springs, Graves Co., 4 May 1982; KNP
Tn08GRV (4, 18-21), unnamed wetland (W. Fk. Clarks R. dr.), 1.1 km
SSW mouth Spring Cr., Graves Co., 5 May 1982; KNP M04FUL (4,
40-66), unnamed wetland (Reelfoot Lake dr.), 0.4 km NE Tyler on E
side KY 94, Fulton Co., 24 June 1982; KDOT Q18 (2, -), slough along
KY 94 (Reelfoot Lake dr.), NW of KY 94 and 1500 m NE of Tennessee-

100 Melvin L. Warren, Jr. and Ronald R. Cicerello

Kentucky line, Fulton Co., 12 June 1979; KDOT Q20 (2, -), tributary to
Blue Pond (Reelfoot Lake dr.), KY 311 bridge, Fulton Co., 17 June
1979; KDOT Q27 (-, -), Rittenhouse Slough (Reelfoot Lake dr.), 1700 m
S Bondurant, Fulton Co., 24 August 1978; KDOT Q32 (2, -), unnamed
trib. Running Slough (Reelfoot Lake dr.), N Illinois Central Railroad
and 170 m SE Ledford, Fulton Co., 22 August 1978.

This species was previously known from only three Kentucky local-
ities (Sisk 1978; Burr 1980). Burr (1980) regarded it as rare and Branson
et al. (1981b) listed it as threatened. Umbra limi is apparently firmly
established in Terrapin Creek (Brooks M. Burr, pers. comm.) and the
Reelfoot Lake drainage of extreme southwestern Fulton County, Ken-
tucky, and is sporadically distributed in Clarks River, being most pre-
valent in the West Fork. Although the populations in Clarks River
represent the only published localities of the species in the Tennessee
River drainage, U. limi has also been collected in the Big Sandy River
of Tennessee (David A. Etnier, pers. comm.). These populations are
near the southern periphery of the range (Gilbert 1980a). The apparent
absence of the fish in streams of western Kentucky draining directly into
the Mississippi River (e.g.. Bayou du Chien, Obion and Mayfield
creeks) is zoogeographically puzzling; however, further intensive sam-
pling of wetland habitats in these drainages will probably reveal its
presence. The future existence of U. limi in Terrapin Creek (Obion R.
dr.) and West Fork Clarks River may be jeopardized by drainage of
remaining wetlands as witnessed at two of our collection sites (i.e., KNP
M02GRV, KNP Tn04GRV). Likewise, rapid erosion of the Mississippi
loess bluffs and expansion of agriculture in the floodplain threaten this
and other species inhabiting the Reelfoot Lake drainage in both Tennes-
see (Starnes and Etnier 1980) and Kentucky.

Hybognathus hayi Jordan. Cypress minnow. KNP Tn04GRV (1,
74), unnamed wetland (W. Fk. Clarks R. dr.), 0.9 km ENE Clear
Springs, Graves Co., 4 May 1982.

Recent works addressing the distribution of H. hayi in Kentucky
have revealed records for direct Mississippi and Ohio river tributaries
and floodplain lakes in the extreme western part of the state (Burr et al.
1980) and a relictual population in lower Green River (Warren and
Cicerello 1982). The discovery of the species in the West Fork Clarks
River system represents the first record for the Tennessee River drain-
age in Kentucky, although records are available for the drainage in
Tennessee (David A. Etnier, pers. comm.) and Alabama (Gilbert 1980b).
The specimen was secured from a shallow (<0.6 m) pothole in a recently
cleared and drained wetland. Hybognathus hayi is considered threa-
tened in Kentucky (Branson et al. 1981b) and in consideration of the
elimination and destruction of the preferred wetland habitat by oil

Kentucky Fishes 101

exploration and coal mining (Warren and Cicerello 1982), channeliza-
tion (Burr et al. 1980), and drainage as witnessed herein, the future of
the species in Kentucky is increasingly tenuous.

Hybopsis insignis Hubbs and Crowe. Blotched chub. KNP C04LOG
(1, 89), S. Fk. (Red R. dr.), 1.7 km NE Smith Grove Church, Logan
Co., 13 July 1982; KNP C05LOG (15, 47-79), Red R. (Cumberland R.
dr.), at Dot, Logan Co., 13 July 1982.

In Kentucky, H. insignis was known to persist only in the Little
South Fork Cumberland River of southeastern Kentucky (Marker et al.

1979) and was formerly known to occur in the mainstem of Cumberland
River (Harris 1980) and the lower Tennessee River (Hubbs and Crowe
1956) before impoundment. In the Cumberland River drainage of Ten-
nessee, the species is known from four localities (Harris 1980). The dis-
covery of a substantial population in the Red River represents a new
record for that drainage in Kentucky and adds hope for the continued
existence of the species in the state. Branson et al. (1981b) considered
the species of special concern in Kentucky. In light of strip-mine and oil
field related water pollution in Little South Fork (Harker et al. 1979,

1980) and heavy siltation and pesticide pollution in Red River (pers.
observ.) the species should be considered at least threatened within
Kentucky.

Notropis ariommus (Cope). Popeye shiner. KFW uncat. (2, 45),
Kinniconick Cr. (Ohio R. dr.), near mouth Pipe Lick Cr., Lewis Co., 7
May 1981; KFW uncat. (3, 45-48), Kinniconick Cr. (Ohio R. dr.),
downstream Laurel Fk. mouth, Lewis Co., 7 May 1981.

The distribution of the popeye shiner in Kentucky was previously
defined as the upper Cumberland, Green, Barren, RoUing Fork, and
Kentucky river drainages (Gilbert 1969, 1980c; Burr 1980). Despite this
rather wide distribution, the popeye shiner is sporadic in occurrence and
seldom common and was thus listed as of undetermined status in Ken-
tucky (Branson et al. 1981b). In middle and upper Ohio River tributar-
ies other than those aforementioned, A^. ariommus is known from six
widely separated populations in Indiana, Pennsylvania, and West Virgi-
nia (Gilbert 1969, 1980c). Several of these represent old records for
populations apparently extirpated (Gilbert 1969). The collections
reported herein are a significant eastward range extension in the Ohio
River valley of Kentucky and also close the hiatus between the widely
separated middle and upper Ohio River populations. It is increasingly
apparent from data presented by Gilbert (1969, 1980c) and that of this
report that A^. ariommus once occupied much of the Ohio River valley,
but is now extirpated or reduced in the northern and upper regions of
the valley to widely disjunct, sporadically distributed localities. Kinnic-
onick Creek is a high quality stream with a predominantly forested

102 Melvin L. Warren, Jr. and Ronald R. Cicerello

watershed that has fortuitously escaped degradation from development.
However, proposed plans to develop extensive Devonian age oil shale
deposits in the watershed (Harker et al. 1980) may alter these conditions
and threaten the future existence of A^. ariommus in the drainage.

Notropis telescopus (Cope). Telescope shiner. KNP uncat. (1, 47),
E. Fk. (Barren R. dr.), at mouth of Isenburg Cr., Monroe Co., 12 June
1979; SIUC 3931 (17, 39-57), E. Fk. (Barren R. dr.), at KY 63 crossing,
Monroe Co., 24 September 1981.

Notropis telescopus was known in Kentucky only from the Cum-
berland River (below the Falls) in the southeastern part of the state
(Burr 1980) where it is common to abundant in small to medium-sized,
high quality streams and rivers. Collection of the telescope shiner in
East Fork Barren River represents the first substantiated report of the
species from the Green River system. Woolman (1892) noted the species
as rare in the Little Barren River; however, Gilbert (1969) regarded the
record as erroneous; the specimen(s) was apparently not available for
examination. These records are of particular interest because several
other taxa typical of the Cumberland River of Kentucky and Tennessee,
such as Nocomis effusus Lachner and Jenkins, Notropis leuciodus
(Cope), Fundulus catenatus (Storer), and closely related members of the
subgenera Catonotus, Nanostoma, and Nothonotus of the genus
Etheostoma, are also represented in the ichthyofauna of the Barren and
Green rivers. Furthermore, Lachner and Jenkins (1971) and Zorach
(1972) cited evidence that stream capture has been responsible for ich-
thyofaunal exchange between the Cumberland and Green rivers. The
ichthyofauna of the upper Barren and Green rivers is relatively well
known (Burr 1980), and the failure of workers subsequent to Woolman
to capture A^. telescopus in the drainage presents an enigma in interpret-
ing its native or non-native status. Since Woolman was apparently quite
familiar with A^. telescopus and differentiated it from the similar A^. ari-
ommus (Cope), the likeUhood of misidentification of the Little Barren
River specimen(s) is reduced. Also of interest is the fact that both East
Fork Barren River and Little Barren River headwaters lie in close prox-
imity to Cumberland River streams (Meshack and Marrowbone creeks,
respectively) which harbor large populations of A^. telescopus (pers.
observ.). The possibility of headwater capture is heightened, especially
along the Meshack Creek-East Fork Barren River divide, by the pres-
ence of karst development including numerous sinkholes and subterra-
nean drainage. This is best illustrated on the southeast corner of the U.
S. Geological Survey 7.5 minute Sulphur Lick quadrangle map (Harris
1964). The current extent of dispersal of A^. telescopus in the Barren
River is unknown, but the capture of the species only in East Fork
implies a localized distribution. Based on the available evidence, the
origin of A^. telescopus in the Green River is unclear; however, as noted

Kentucky Fishes 103

by Jenkins et al. (1971) even the most suggestive evidence concerning
stream capture may lead to invalid conclusions, and limited distribu-
tions may result from introduction rather than natural factors. Further
collecting aimed at probable theatres of stream capture should further
elucidate the native or non-native status of A^. telescopus as well as rela-
tionships among other shared taxa.

Fundulus chrysotus (Giinther). Golden topminnow. KNP M03FUL
(9, 42-51), Running Slough (Reelfoot Lake dr.), at Ledford, Fulton Co.,
24 June 1982.

According to Burr (1980), the golden topminnow was known in
Kentucky only from Open Pond, Fulton County, where two collections
were made by Sisk (1973). Field observations made during the summer
of 1982 revealed that Open Pond and surrounding wetlands had been
drained, cleared, and converted to agricultural land. Recent collecting
efforts in appropriate habitat throughout the general drainage area in
Kentucky revealed that the golden topminnow is now known from a
single extant population in Running Slough. Although F. chrysotus is
listed as of special concern in Kentucky by the Kentucky Academy of
Science (Branson et al. 1981b), the species should be placed in the
endangered category in light of its extremely limited distribution in
Tennessee (Starnes and Etnier 1980) and the potential for rapid loss of
habitat as witnessed for Open Pond.

Lepomis marginatus (Holbrook). Dollar sunfish. KNP Tn04GRV
(1, 49), unnamed wetland (W. Fk. Clarks R. dr.), 0.9 km ENE Clear
Springs, Graves Co., 4 May 1982; KNP Tn08GRV (1, 72), unnamed
wetland (W. Fk. Clarks R. dr.), 1.1 km SSW mouth Spring Cr., Graves
Co., 5 May 1982.

Lepomis marginatus, previously known from only two Kentucky
localities, neither of them in the Tennessee River drainage (Burr 1980),
was discovered by Rice et al. (1983) at a third site in West Fork Clarks
River. These records represent an addition to the ichthyofauna of the
Tennessee River in Kentucky. The species is apparently widely distrib-
uted in West Fork Clarks River and at some localities occurs in good
numbers (Rice et al. 1983). The populations in West Fork, although
threatened by wetland drainage as witnessed by the authors (e.g., at
KNP Tn04GRV), may prove to be critical in preserving the species as a
viable member of the native Kentucky ichthyofauna. Branson et al.
(1981b) listed the species as threatened.

Etheostoma camurum (Cope). Bluebreast darter. KNP LOIBAT
(26, 34-54), Licking River (Ohio R. dr.), at mouth Slate Cr., Bath Co.,
16 September 1982.

Etheostoma camurum was previously known in Kentucky from the
upper Cumberland (below the Falls) and upper Kentucky river drain-
ages (Burr 1980; Zorach 1972). The collection reported herein represents

104 Melvin L. Warren, Jr. and Ronald R. Cicerello

an addition to the ichthyofauna of the speciose Licking River system
and reduces the distributional hiatus among known populations in the
upper and middle Ohio River drainage (Zorach 1972). Ahhough pre-
vious surveys of the Licking River have not revealed E. camurum
(Woolman 1892; Welter 1938; Clark 1941a, b; Jones 1970), other
members of the subgenus Nothonotus (i.e., E. maculatum Kirtland and
E. tippecanoe Jordan and Evermann) have been reported from the
drainage (Woolman 1892; Clark 1941a, b; Burr 1980). However, E.
maculatum, often confused with E. camurum (Zorach and Raney 1967),
is known in the drainage only from a collection made by J. A. Henshall
in South Fork (Woolman 1892). Henshall apparently recognized both
species, according to details presented in Trautman (1981), and the E.
maculatum record is considered valid by some authors (Zorach and
Raney 1967; Burr 1980), although Etnier (1980) did not include the
Licking River in its distribution. We speculate that E. camurum was
missed in the Licking River by previous investigators because popula-
tions are often localized (Stauffer 1980). Moreover, large stream or river
habitat often occupied by members of the subgenus Nothonotus is diffi-
cult to collect and has elsewhere recently yielded species missed during
many years of collecting (Williams and Etnier 1978). Our specimens
were collected from a 0.3-0.6 m deep riffle habitat with moderate to
swift current. Substrate consisted of bedrock overlain with slab boulder
where the current was swiftest, and cobble/ gravel in areas of moderate
flow. Additional collecting in the poorly sampled mainstem of the Lick-
ing and South Fork Licking rivers (Burr 1980) is necessary to determine
the extent of distribution of E. camurum and to verify the existence of
E. maculatum in the drainage.

Percina macrocephala (Cope). Longhead darter. KNP uncat. (1,
72), Kinniconick Cr. (Ohio R. dr.), 2.1 km upstream from mouth Pine
Br., Lewis Co., 5 May 1981; KNP uncat. (4, 45-52), Kinniconick Cr.
(Ohio R. dr.), near mouth Pipe Lick Cr., Lewis Co., 7 May 1981; KNP
uncat. (2, 51-75), Kinniconick Cr. (Ohio R. dr.), between Mill and Lea-
therbelly branches, Lewis Co., 14 May 1981; KNP GOl ALL (15, 67-90),
Trammel Fk. (Barren R. dr.), at old state rd. ford 1.55 km N of Red
Hill, Allen Co., 14 July 1982; KNP G05WAR (6, 36-72), Trammel Fk.
(Barren R. dr.), at ford 0.2 km upstream from mouth Drakes Cr., 16
July 1982.

According to Page (1978) and Burr (1980), the longhead darter
occurs sporadically in Kentucky in the upper Barren, upper Green, Ken-
tucky, Licking, and Big Sandy river systems and Kinniconick Creek and
has apparently been extirpated from the Cumberland River. Burr (1980)
noted that the species was once common in the Barren River prior to
impoundment of Barren River Reservoir as indicated in pre-

Kentucky Fishes 105

impoundment surveys conducted by the Kentucky Department of Fish
and Wildlife Resources. Our collections in Trammel Fork (Barren R.
dr.) indicate the species continues to persist in good numbers in tributar-
ies unaffected by impoundment. The only previous collection of P.
macrocephala in Kinniconick Creek is based on a single specimen col-
lected in the 1930s and housed at the University of Louisville. Our
observations indicate that the species is moderately common in approp-
riate habitat along the Kinniconick Creek mainstem. In spite of these
relatively healthy populations in the Barren River drainage and Kinnic-
onick Creek, the conservation status of the species in Kentucky should
remain threatened (Branson et al. 1981b) because of pollution and habi-
tat destruction associated with coal mining in the upper Kentucky, Lick-
ing, and Big Sandy rivers (Harker et al. 1979), the extirpation of the
species from the Cumberland River of Kentucky (Page 1978), and the
threat of oil shale development in the Kinniconick Creek drainage
(Harker etal. 1981).

Percina phoxocephala (Nelson). Slenderhead darter. KNP CO 1 LOG
(1, 49), Whippoorwill Cr. (Red R. dr.), 0.7 km W Millertown Church,
Logan Co., 8 July 1982; KNP C03LOG (3, 58-63), S. Fk. (Red R. dr.),
Kentucky-Tennessee line, Logan Co., 13 July 1982; KNP C04LOG (5,
65-69), S. Fk. (Red R. dr.), 1.7 km NE Smith Grove Church, Logan
Co., 13 July 1982; KNP C05LOG (3, 61-84), Red R. (Cumberland R.
dr.), at Dot, Logan Co., 13 July 1982; KNP C02TRI (5, 55-67), Little R.
(Cumberland R. dr.), 1.3 km downstream from KY 1253 crossing, Trigg
Co., 6 July 1982.

Percina phoxocephala was previously unknown from the Red
River of Kentucky and Tennessee (Starnes and Etnier 1980; Thompson
1980), and Burr (1980) noted its former occurrence in the lower Cum-
berland River. The collections noted herein are new records for the Red
River and substantiate the persistence of the species in a lower Cumber-
land River tributary. In Kentucky, the species was regarded as of special
concern (Branson et al. 1981b) and was noted by Burr (1980) as occa-
sional within several drainages. Recent collections made by the authors
and fresh material examined by us indicate good populations of P.
phoxocephala in Tygarts Creek of eastern Kentucky (Warren 1981, and
unpublished), and upper and lower Green, Barren, and Rough rivers
(Retzer et al. 1983; Warren and Cicerello 1982, and unpubhshed). The
slenderhead darter has been taken by others in recent years from several
stations each in Eagle Creek (lower Kentucky River) (Horseman and
Branson 1973) and Salt River and tributaries (Hoyt et al. 1979). It
therefore appears that retention of the species on the Kentucky Academy
of Science list of rare fishes is unwarranted.

106 Melvin L. Warren, Jr. and Ronald R. Cicerello

Percina shumardi (Girard). River darter. KNP SOI GUP (2, 40-44),
Little Sandy R. (Ohio R. dr.), 0.5 km W Argillite, Greenup Co., 14
September 1982.

According to Burr (1980), P. shumardi is sporadic and uncommon
in every major river of the state except the Salt and Big Sandy rivers
and the direct Ohio River tributaries in extreme northeastern Kentucky.
Collection of the species in the Little Sandy River is a new record for
this system and suggests the stream is in need of further study despite
recent collections (Marker et al. 1979; Branson et al. 1981a). It further
suggests that the species will eventually be taken in adjacent rivers and
streams (e.g.. Big Sandy River, Kinniconick and Tygarts creeks). The
rarity of the species is partially attributable to difficulty in collecting the
preferred habitat (Trautman 1981). The specimens reported herein were
taken from a deep (1.3 m), swift chute over a substrate of large cobble.
Apparently this is the last shoal present on the Little Sandy River
before its waters are embayed by the Greenup Lock and Dam on the
Ohio River. The river darter is considered threatened in Kentucky
(Branson et al. 1981b); however, we believe that further collecting in
appropriate habitat will reveal new populations and result in its removal
from the state threatened category.

ACKNOWLEDGMENTS.— Our appreciation is extended to Richard
R. Hannan, Director of the Kentucky Nature Preserves Commission,
for support throughout this effort. Special thanks go to Brian D. And-
erson, Keith E. Camburn, and Dan E. VanNorman for assistance with
field work. The following individuals and their agencies or institutions
kindly confirmed identifications, shared collecting information, or pro-
vided specimens and other courtesies: Brooks M. Burr, Southern Illinois
University at Carbondale; David A. Etnier, University of Tennessee;
John L. Harris, Arkansas Department of Transportation; Lewis E.
Kornman, Kentucky Department of Fish and Wildlife Resources; and
Stephen P. Rice, Kentucky Department of Transportation. Our grati-
tude is also expressed to Branley A. Branson, Eastern Kentucky Univer-
sity; Brooks M. Burr; Keith E. Camburn, Kentucky Nature Preserves
Commission; and David A. Etnier for critical review of the manuscript.

LITERATURE CITED

Bauer, Bruce H., and B. A. Branson. 1979. Distributional records for and

additions to the ichthyofauna of Kentucky. Trans. Ky. Acad. Sci.

-^0(l-2):53-55.
Branson, Branley A., D. L. Batch and S. Rice. 1981a. Collections of fishes from

the Little Sandy River and Tygarts Creek drainages, Kentucky. Trans. Ky.

Acad. Sci. '/2(3-4):98-100.

Kentucky Fishes 107

, D. F. Harker, Jr., J. M. Baskin, M. E. Medley, D. L. Batch, M. L.

Warren, Jr., W. H. Davis, W. C. Houtcooper, B. Monroe, Jr., L. R. Phil-
Hppe and P. Cupp. 1981b. Endangered, threatened, and rare animals and
plants of Kentucky. Trans. Ky. Acad. Sci. ¥2(3— 4):77-89.

Burr, Brooks M. 1980. A distributional checkHst of the fishes of Kentucky.
Brimleyana No. 3:53-84.

, andR. L. Mayden. 1979. Records of fishes in western Kentucky with

additions to the known fauna. Trans. Ky. Acad. Sci. ¥0(l-2):58-67.

, M. E. Retzer and R. L. Mayden. 1980. A reassessment of the dis-

tributional status of five Kentucky cyprinids. Trans. Ky. Acad. Sci

¥/(l-2):48-54.
Clark, Minor E. 1941a. A list of the fishes in northeastern Kentucky. Ky. Dep.

Fish Wildl. Resour., Fish Bull. 7:1-11.
1941b. Biological survey of the Little Sandy and upper Licking River

watersheds. Ky. Dep. Fish Wildl. Resour., Fish. BulLJ: 1-146.
Clay, William M. 1975. The Fishes of Kentucky. Ky. Dep. Fish Wildl. Resour.,

Frankfort. 416 pp.
Etnier, David A. 1980. Etheostoma maculatum Kirtland, Spotted darter, p. 664

in D. S. Lee, et al. Atlas of North American Freshwater Fishes. N. C. State

Mus. Nat. Hist., Raleigh, x + 867 pp.
Gilbert, Carter R. 1969. Systematics and distribution of the American cyprinid

fishes Notropis ariommus and Notropis telescopus. Copeia 1969(3):474-492.
1980a. Umbra limi (Kirtland), Central mudminnow. p. 129 in D. S.

Lee, et al. Atlas of North American Freshwater Fishes. N. C. State Mus.

Nat. His., Raleigh, x + 867 pp.
1980b. Hybognathus hayi Jordan, Cypress minnow, p. 176 in D. S.

Lee, et al. Atlas of North American Freshwater Fishes. N. C. State Mus.

Nat. Hist., Raleigh, x -♦■ 867 pp.

1980c. Notropis ariommus (Cope), Popeye shiner, p. 229 in D. S.

Lee, et al. Atlas of North American Freshwater Fishes. N. C. State Mus.
Nat. Hist., Raleigh, x + 867 pp.

Harker, Donald F., Jr., S. M. Call, M. L. Warren, Jr., K. E. Camburn and P.
Wigley. 1979. Aquatic biota and water quality survey of the Appalachian
Province, eastern Kentucky. Tech. Rep. Ky. Nature Preserves Comm.,
Frankfort. 1152 pp.

, M. L. Warren, Jr., K. E. Camburn, S. M. Call, G. J. Fallo and P.

Wigley. 1980. Aquatic biota and water quality survey of the upper Cum-
berland River basin. Tech. Rep. Ky. Nature Preserves Comm., Frankfort.
683 pp.

R. R. Hannan, R. R. Cicerello, W. C. Houtcooper, L. R. Phillippe

and D. VanNorman. 1981. Preliminary assessment of the ecology and

ecological features of the Kentucky "Knobs" oil shale region. Tech. Rep.

Ky. Nature Preserves Comm., Frankfort. 137 pp.
Harris, John L. 1980. Hybopsis insignis Hubbs and Crowe, Blotched chub. p.

188 in D. S. Lee, et al. Atlas of North American Freshwater Fishes. N. C.

State Mus. Nat. Hist., Raleigh, x + 867 pp.
Harris, Leonard D. 1964. Geology of the Sulphur Lick quadrangle, Kentucky.

GQ-325. U. S. Geol. Survey, Washington, D.C.

108 Melvin L. Warren, Jr. and Ronald R. Cicerello

Horseman, Nelson D., and B. A. Branson. 1973. Fishes of Eagle Creek, north-
ern Kentucky. Trans. Ky. Acad. Sci. i¥(l-2):5-12.

Hoyt, Robert D., S. E. Neff and V. H. Resh. 1979. Distribution, abundance,
and species diversity of fishes of the upper Salt River drainage, Kentucky.
Trans. Ky. Acad. Sci. ^0(1 -2): 1-20.

Hubbs, Carl L., and W. R. Crowe. 1956. Preliminary analysis of the American
cyprinid fishes, seven new, referred to the genus Hybopsis, subgenus Eri-
mystax. Occas. Pap. Mus. Zool. Univ. Mich. 575:1-8.

Jenkins, Robert E., E. A. Lachner and F. J. Schwartz. 1971. Fishes of the cen-
tral Appalachian drainages: Their distribution and dispersal, pp. 43-117 in
P. C. Holt (ed.). The Distributional History of the Biota of the Southern
Appalachians, Part HI: Vertebrates. Res. Div. Monogr. 4, Va. Polytech.
Inst. State Univ., Blacksburg. 306 pp.

Jones, Albert R. 1970. Inventory and classification of streams in the Licking
River drainage, Ky. Dep. Fish Wildl. Resour., Fish. Bull. "^5:1-34.

Lachner, Ernest A., and R. E. Jenkins. 1971. Systematics, distribution, and
evolution of the Nocomis biguttatus species group (Family Cyprinidae:
Pisces) with a description of a new species from the Ozark upland. Smith-
son. Contrib. Zool. 9/: 1-28.

Page, Lawrence M. 1978. Redescription, distribution, variation and life history
notes on Percina macrocephala (Percidae). Copeia 1978(4):655-664.

, and B. M. Burr. 1982. Three new species of darters (Percidae,

Etheostoma) of the subgenus Nanostoma from Kentucky and Tennessee.
Occas. Pap. Mus. Nat. Hist. Univ. Kans. 707:1-20.

Retzer, Michael E., B. M. Burr and M. L. Warren, Jr. 1983. Fishes of the lower
Green River drainage, Kentucky. Ky. Nature Preserves Comm. Sci. Tech.
Ser. i:l-48

Rice, Stephen P., J. R. MacGregor and W. L. Davis. 1983. Distributional
records for fourteen fishes in Kentucky. Trans. Ky. Acad. Sci. -^^(3-4).

Robins, C. Richard, R. M. Bailey, C. E. Bond, J. R. Brooker, E. A. Lachner,
R. N. Lea and W. B. Scott. 1980. A List of Common and Scientific Names
of Fishes from the United States and Canada. 4th ed. Am. Fish. Soc. Spec.
Publ. No. 12. 174 pp.

Rohde, Fred C. 1980 Lampetra appendix (DeKay), American brook lamprey, p.
23 in D. S. Lee, et al. Atlas of North American Freshwater Fishes. N. C.
State Mus. Nat. Hist., Raleigh x + 867 pp.

Seagle, Henry H., Jr., and J. W. Nagel. 1982. Life cycle and fecundity of the
American brook lamprey, Lampetra appendix, in Tennessee. Copeia
1982(2):362-366.

Sisk, Morgan E. 1973. Six additions to the known piscine fauna of Kentucky.
Trans. Ky. Acad. Sci. 5-^(3-4):49-50.

Starnes, Wayne C. 1981. Listing package for the blackside dace, Phoxinus
cumberlandensis. Submitted 23 September 1981 to U. S. Fish Wildl. Ser-
vice, Area Office, Asheville, NC.

, and D. A. Etnier. 1980. Tennessee's rare fishes, pp. B1-B133 in D. C.

Eager and R. M. Hatcher (eds.). Tennessee's Rare Wildlife. Tenn. Wildl.
Resour. Agency and Dep. Conserv., Nashville, viii + 337 pp.

Kentucky Fishes 109

and L. B. Starnes. 1978. A new cyprinid of the genus Phoxinus

endemic to the upper Cumberland River drainage. Copeia 1978(3):508-516.
, and 1979. Taxonomic status of the percid fish Etheos-

toma nigrum susanae. Copeia 1979(3):426-430.
Stauffer, Jay R. 1980. Etheostoma camurum (Cope), Bluebreast darter, p. 632

in D. S. Lee, et al. Atlas of North American Freshwater Fishes. N. C. State

Mus. Nat. Hist., Raleigh, x + 867 pp.
Thompson, Bruce A. 1980. Percina phoxocephala (Nelson), Slenderhead darter,

p. 737 in D. S. Lee, et al. Atlas of North American Freshwater Fishes. N.

C. State Mus. Nat. Hist., Raleigh, x + 867 pp.
Trautman, Milton B. 1981. The Fishes of Ohio. Ohio State Univ. Press,

Columbus. 782 pp.
Warren, Melvin L., Jr. 1981. New distributional records of eastern Kentucky

fishes. Brimleyana No. 6:129-140.
, and R. R. Cicerello. 1982. New records, distribution, and status of

ten rare fishes in the Tradewater and lower Green rivers, Kentucky. South-
east. Fish. Council Proc. 5(4): 1-7.
Welter, Wilfred A. 1938. A list of the fishes of the Licking River drainage in

eastern Kentucky. Copeia 1938(2):64-68.
Williams, James D., and D. A. Etnier. 1978. Etheostoma aquali, a new percid

fish (subgenus Nothonotus) from the Duck and Buffalo rivers, Tennessee.

Proc. Biol. Soc. Wash. 9/(2):463-471.
Woolman, Albert J. 1892. Report of an examination of the rivers of Kentucky,

with lists of the fishes obtained. Bull. U. S. Fish. Comm. 70:249-288.
Zorach, Timothy. 1972. Systematics of the percid fishes, Etheostoma camurum

and E. chlorobranchium new species, with a discussion of the subgenus

Nothonotus. Copeia 1972(3):427-447.
, and E. C. Raney. 1967. Systematics of the percid fish Etheostoma

maculatum Kirtland, and related species of the subgenus Nothonotus. Am.

Midi. Nat. 77(2):296-322.

Accepted 16 April 1983

Winter Food Habits of Bobcats in North Carolina

Anne M. King
Department of Zoology,

Richard A. Lancia
Department of Forestry,

AND

S. Douglas Miller, iDavid K. Woodward and Jay D. Hair '

Department of Zoology
North Carolina State University, Raleigh, North Carolina 27650

ABSTRACT.— Carcasses of 505 bobcats, Felis rufus (229 9, 276 5),
were collected from October through March in 1978-79 and 1979-80.
Stomach contents were analyzed and the results summarized by physi-
ographic regions, sex, and age. The top eight prey groups, ranked by
frequency of occurrence, were rabbits, Sylvilagus spp.; birds; cotton
rats, Sigmodon hispidus; white-tailed deer, Odocoileus virginianus;
rodents; gray squirrels, Sciurus carolinensis; raccoons, Procyon lotor,
and opossums, Didelphis virginianus. Ranked by frequency of occur-
rence, rabbits were first in the Coastal Plain and cotton rats first in the
Piedmont Plateau. Adult male bobcats consumed larger prey than did
adult females or kittens. Kittens tended to exploit smaller and a wider
variety of prey items than did adults.

INTRODUCTION

Because of restrictions placed on harvesting many species of spot-
ted cats, the value of bobcat, Felis rufus, pelts increased greatly on the
international fur market. In 1977 concern about the national status of
bobcats prompted the Council on International Trade of Endangered
Species of Flora and Fauna (CITES) to list them as a species that could
become threatened with extinction unless trade was subject to regula-
tion. As part of a program to learn more about the status and ecology
of bobcat populations in North Carolina, we initiated a study of bobcat
food habits based on analyses of trapper-harvested carcasses. Other
aspects of our research program are reported elsewhere (Lancia et al., in
press).

Previous food habits studies in the southeastern United States
reported rabbits, Sylvilagus spp., as the major food item (Progulske
1952; Davis 1955; Fritts 1973; Buttrey 1974; Fox and Fox, in press). In
addition, cotton rats, Sigmodon hispidus; white-tailed deer, Odocoileus

' Present address: National Wildlife Federation, 1412 Sixteenth Street, N.W.,
Washington, D.C. 20036

Paper number 7035 of the Journal Series of the North Carolina Agricultural
Research Service, North Carolina.

BrimleyanaNo.9:lll-122. June 1983. Ill

112 Anne M. King, et al.

virginianus; squirrels, Sciurus carolinensis; voles, Microtus spp.; and
birds were frequently identified food items (Progulske 1952; Davis 1955;
Kight 1962; Fritts 1973; Buttrey 1974; Miller and Speake 1978).

MATERIALS AND METHODS

Skinned bobcat carcasses were provided by fur dealers and trappers
during the 1978-79 and 1979-80 trapping seasons. The carcasses were
weighed, sexed, and necropsied. Adult and kitten (< 8 months) age
classes were assigned using lacteal tooth replacement criteria (Crowe
1975).

Stomach contents, prepared following Korschgen (1980), were
identified to order and, when possible, to species. Data were recorded
by percent frequency of occurrence because not all material in each
stomach could be identified, hence percent volume could not be meas-
ured precisely. Food items were identified macroscopically or, when
necessary, mammals were identified by microscopic examination (at
lOOx and 400x) of sample hairs using hair keys (Spiers 1973; Moore et
al. 1974).
Statistical Analyses and Interpretation

The state was divided into three physiographic regions (Stuckey
1965), and the data were examined at statewide and regional levels.
Comparisons of carcass weights were made using least-squares fit to a
fixed-effect linear model using the Scientific Analysis System (SAS;
Barr et al. 1979). Weights of kittens and adults were analyzed sep-
arately. Food items were ranked by frequency of occurrence, and differ-
ences in ranks were evaluated using Wilcoxon's Rank Test (Wilcoxon
1945). Chi-square tests were used to evaluate differences in prey selec-
tion. Calculation of expected frequencies of prey items for chi-square
tests were based on an assumed equal probability of occurrence in a
particular sex and age group.

Caution must be observed when making inferences about the pro-
portion of food items present in the diet based solely on stomach anal-
yses because: 1) smaller prey species are frequently under represented in
stomach analyses due to differences in digestibility (Weaver and Hof-
fman 1979; Merriwether and Johnson 1980); 2) some food items may
have been eaten incidentally as trap bait, garbage or carrion; and 3)
frequency of occurrence data cannot be used to evaluate food preferen-
ces without estimates of both prey abundance and availability.

RESULTS AND DISCUSSION

From October through March 1978-79 and 1979-80, 505 bobcats

(229 $, 276 5) were collected. Weights between samples collected each

year were not different (p>0.05); therefore, the samples were pooled for

subsequent analyses. Among the three physiographic regions, weights

Bobcat Winter Food Habits

113

were different (p<0.01) for both age groups, but no clear biological sig-
nificance in the trends was apparent (Tables 1, 2). Weights of females
were less than males (p<0.01) for both adults and kittens. Sex ratios of
adults and kittens were not different among regions or between yearly
samples (p>0.05).

Table 1. Analysis of variance for weights of skinned carcasses of 505 bobcats
collected in 1978-79 and 1979-80 in North Carolina.

Adults (N=355)
Source DF F value

Kittens (N= 150)
Source DF F value

Sample

Region (Sample)

Sex (Region)

0.02
4.13**
134.74**

Sample

Region (Sample)

Sex (Region)

1.75

5.24**

3.84**

** Significant at the P<0.01 level.

Table 2. Skinned weights (kg) of 505 bobcats collected in the three physiograph-
ic regions of North CaroHna.

Region

Age Class
Kittens

Sex

N

Mean

SE

Coastal Plain

9

47

3.9

0.1

S

58

4.3

0.1

Adults

9

131

5.8

0.1

S

153

8.0

0.1

Piedmont

Kittens

9

10

3.4

0.2

S

23

4.0

0.2

Adults

9

26

5.9

0.2

S

33

8.2

0.2

Mountains

Kittens

9

6

3.0

0.3

S

6

2.9

0.4

Adults

9

9

4.9

0.1

S

3

9.0

1.1

Of 473 stomachs suitable for examination, 398 (179 $, 219 $) con-
tained food. This sample comprised 257 adults and 141 kittens. Stom-
achs containing food were collected from 50 of the 100 counties in the
state, with the largest samples in the Coastal Plain (307) and Piedmont
Plateau (73) regions and the smallest sample (18) from the Mountains
(Fig. 1).

Some disproportionate sampling was evident in these regions, with
one or two counties contributing more than a third of the total number

114

Anne M. King, et al.

■^

."2

c
■g
■3

o
u

I

B
o

bO

Bobcat Winter Food Habits

115

of stomachs for the region. To determine if these county samples biased
the total regional sample, we ranked the top eight food items in these
counties and compared them to the top eight food items in the remain-
ing counties of the region. The differences in ranks were not statistically
significant (p>0.05).

Mammals were the most frequently represented food group, with
20 species identified statewide (Table 3). The top eight food items
ranked by frequency of occurrence were rabbits; birds; cotton rats;
white-tailed deer; small rodents; gray squirrels; raccoons, Procyon lotor;
and opossums, Didelphis virginianus. Small numbers of reptiles, am-
phibians, fish, and insects also were identified.

Rabbits were the most frequently identified species in bobcat stom-
achs (Table 3, Fig. 2). The percent occurrence of rabbits decreased from
the Coastal Plain to the Mountain region (p<0.05), probably in response
to differences in abundance of rabbits in the regions rather than to shifts
in prey preference. The low frequency (5.6%) of rabbits in stomachs
from the Mountain region contrasted the findings of Progulske (1952)

ADULT MALES (156 STOMACHS)

MISC. PREY (1,8%)

OTHER MAMMALS

(4 4%)
SMALL RODENTS

( 13%)
SQUIRREL (4.8%)

<DS
(16 7%)

RACCOON (7 5%)

DEER (172%)

COTTON RAT (10 6%)

ADULT FEMALES (134 STOMACHS)

-MISC PREY (18%)

UNKNOWN MAMMALS
(77%)

SQUIRREL (4 7%)
RACCOON ( I % )
DEER (5.7%)

COTTON RAT (12%)

KITTENS (108 STOMACHS)

MISC PREY (3.7%)

SQUIRREL (4.3%)
DEER (1.8%)

COTTONRAT (15.8%)

Fig. 2. Relative percent frequency of occurrence of food items identified in adult
male, adult female, and kitten bobcat stomachs. Small rodents include voles,
marsh rice rat, and mice; other mammals include opossum, muskrat, pig, rats,
chipmunk, flying squirrel, and beaver; miscellaneous prey include reptiles,
amphibians, fish, and arthropods.

116

Anne M. King, et al.

o
So
2

3 oo

« 7?

*-. II

1

5 5. cr

C/3 w

Cu

k;

2

Uh

60

O

Is

2; ^

C T3

•-^ C

CA cd

6 O

••^ I

oo

«2 -
3 2

o o

Cfj

^ 00

O ro

" 5,

■3 ^

■^ J-I.
o D

OoovoOoo»or^oow-im«n«n«/^w^

Tj- ^ —

l^ »r^ •^ 00 On
oo r- u^ <N -H -H

TtO>t^vO«r><N<N<N<N

CM <N <N w-^

d d o d

^ ^ ^ ^

vq so

-HOm<NO<^tNOOOOO — — I

O O O Tt

ON t^ r«; "^

r4 On r^ -^
rri rn ^^

Tt -H — I Tt
-J Tt Tt -^

-H rn -H

^ O O O O — I

O O — ' o

fN »n m

»o -H ^

-HfSromr<^NOvovOvOr^

od«orn-^-^-^'^ddd

VO rt Tf <N -^ -H

tn -2

C/3 ^ 03

< ^ -s

2 ^^ ^

S "^ o

<: Z5 ti

S oi u

Si) p

52 ^

13

>> ft*"
a On

c o

S 8

ed ««

Si) =*

2 ^ >

■Si ^ <u

- 5 e

c* ■*- 2
5 <^> 3

o .2

s i

3

o o

-H -H O 00

>

I

o

3
3 <^

c .S
-2 >^

> O [Jh

S 6

a>

D, i:

«j :3 o

S S K

Oh W U W

C/3

B

M

Bobcat Winter Food Habits

117

^'-^OOOCJ O O OCJOO oooo

r-fN— ^-^W-i -H ^H ^^ .^ ^^ ^^ ^H<N'-^ro

I -H I I 00

I I I

orNoo"/^ o o oooo oooo

-^ I I Tt OO Tt

I I I I I I I r^

f^oo— 'c*^ -^ o oooo ooo-^

^ o -^
m

ro m m m m m \o rn \o
o oocJo oooo

ON

§

s

§

o

"S

3

>

1

^
s

1

2

^

S

-o

O

t-l

.4

IS

<§

o

^

"S

o

3

73

x>

o

«J

c

W)

c

C/3

Q

o

1

u

^

o
"^

o

n

;zi

a5

U

PQ

X) C/!)

c w
D a:

cx o

3

c c
o o

D U- D

13 "^

CO t->

g .a

J O

C/5

O H

1/3

"J «J i^ ^
< < Q I 3

118 Anne M. King, et al.

who examined 25 stomachs from the Pisgah and Nantahala National
Forests and reported 24% contained rabbits.

Birds were found in a greater percentage of the stomachs in 1978-79
than in 1979-80 (p<0.05). In both years the percent occurrence of birds
was greater in stomachs from the Mountains than from the Piedmont
region (p<0.05). Birds were found in proportionately more adult female
and kitten stomachs than in adult male stomachs (Fig. 2). Although this
relationship was not statistically significant in our study, Fritts and Sea-
lander (1978) also observed a similar trend in Arkansas.

Statewide, cotton rats were the second ranking mammal by fre-
quency of occurrence. A high occurrence of cotton rats in kitten stom-
achs, although not significantly greater than in adults (p>0.05), reflects
a trend in prey size similar to that observed for birds, i.e., smaller prey
occurred more frequently in adult females and kittens than in adult
males.

The percent occurrence of cotton rats was highest in the Piedmont
region where this species was the major food item (Table 3). Beasom
and Moore (1977) and Miller and Speake (1978) have shown that in
habitats where cotton rats were the most abundant small mammal, they
were the major prey item. Cotton rats ranked second in frequency in the
Coastal Plain region, but were found less than one-third as frequently as
rabbits, the major food item (Table 3). Cotton rats were not found in
stomachs examined from the Mountain region.

Deer were identified in stomachs from all regions and ranked third
statewide among mammals in frequency of occurrence (Table 3). Al-
though a major food item in the Mountain region, the percent occur-
rence of deer was not significantly different (p>0.05) from that found in
other regions.

Progulske (1952) reported deer as a winter food of bobcats in west-
ern North Carolina and Virginia and suggested they were eaten as car-
rion. Based on 124 scats from Virginia, he concluded deer were con-
sumed primarily during the fall and winter rather than in spring or
summer, and suggested hunting season casualties on deer might be par-
tially responsible. Managers of North Carolina wildlife areas reported
16 deer killed by bobcats in 1968 (Barrick 1969).

Deer were found more frequently in stomachs of adult males than
in adult females or kittens (Fig. 2). In the Coastal Plain, observed fre-
quencies of deer were greater for adult males and less for adult females
and kittens than expected frequencies (p<0.05), possibly indicating a
selective preference for deer by adult males. Observed versus expected
frequencies of deer were not different (p>0.05) between the Piedmont or
Mountain regions, probably due to the small number of stomachs
examined from these regions.

Bobcat Winter Food Habits 119

Greater use of deer by adult male bobcats may be related to differ-
ences in average weight and/ or size of home ranges of males and
females. Weights of males were larger than those of females, and a tele-
metry study in the Coastal Plain region indicated the average size of
adult male home ranges (N=5) was about 41% larger than home ranges
of adult females (N=3) (Lancia et al., in press). Similar ratios of male to
female ranges were observed by Bailey (1974) in Idaho, Hall and New-
som (1976) in Louisiana, and Miller (1980) in Alabama.

Gray squirrels ranked fourth in overall frequency of occurrence
among mammals (Table 3). The relative percent frequencies of squirrels
in this study were similar for all bobcat sex and age groups (Fig. 2).
Squirrels were found in two stomachs from the Mountain region and
one stomach from the Piedmont region. Similarly, Progulske (1952)
found squirrels in two of 25 stomachs he examined from western North
Carolina. In the Coastal Plain region, squirrels ranked fourth in fre-
quency of occurrence.

Raccoons ranked fifth among mammals in the statewide sample
and occurred only in stomachs from the Coastal Plain and Piedmont
regions (Table 3). Raccoons occurred primarily in adult male stomachs
and were not found in any kitten stomachs (Fig. 2). The greater occur-
rence of raccoons in adult male stomachs again may reflect the greater
weight and larger home range of male bobcats.

Small rodents were infrequent food items. Voles, Microtus penn-
sylvanicus and M. pinetorum, were the most frequently identified small
rodents and occurred in stomachs from all regions. Voles ranked first
(with deer) in percent occurrence in the Mountain region (Table 3) and
were found primarily in stomachs of kittens and adult females. In the
Coastal Plain, adult females and kittens had greater observed than
expected frequencies of voles (p<0.05). Other small rodents were prey
of adult females and kittens but were seldom taken by adult males (Fig.
2). Fritts and Sealander (1978) also reported that occurrences of rats
and mice were greater in the stomachs of adult females and kittens than
in adult male bobcats.

We examined 141 kitten (59% males) stomachs containing food.
Like adults, kittens relied primarily on rabbits and cotton rats for food
(Fig. 2); however, the occurrence of some items in kitten stomachs was
notably different from adults. Deer occurred in only three (2.1%) kitten
stomachs, compared to 51 (19.8%) adult stomachs. Two intermediate-
sized prey species, raccoons and opossums, occasionally were eaten by
adults but were absent or rarely found in the kitten sample (Fig. 2). The
low occurrence of large and intermediate-sized prey species in kitten
stomachs probably reflected an inability of kittens to capture larger-sized
prey. The infrequent occurrences of these prey species in kitten stom-

120

Anne M. King, et al.

achs may have been the result of food provided by the aduh female, or
evidence of feeding on carrion.

A greater variety of food items was eaten by kittens than by adults.
For example, kitten stomachs represented 28% of the total sample, but
accounted for 71% of the occurrences of Peromyscus spp. and 57% of
the voles. Fritts and Sealander (1978) also reported a greater use of
cricetid rodents by kittens than by adults.

Although prey species in the cotton rat-rabbit size range (0.1-2.0
kg) were major components in the diet of all bobcat sex and age classes,
some differential selection was evident. Kittens and adult females were
the principal consumers of the smallest prey (<0. 1 kg), while the largest
prey species (>2.0 kg) were consumed most frequently by adult males
(Fig 3). The greatest occurrence of prey in the 0. 1 - 2.0 kg range may
represent a specialization of bobcats for a selected prey size (Miller and
Speake 1978).

o

3
O
LU

q:

o
or

LU

40 r

30

20

10

I ADULT 9
I KITTEN
ADULT cT

oL

0.0 - 0. 1

0.1-0.4

0.4-2.0 2.0-16.0 16.0-120.0

PREY SPECIES WEIGHT CLASSES (kg)

Fig. 3. Relative percent occurrences of different weight classes of mammalian
prey items identified in adult male, adult female, and kitten bobcat stomachs.
Species included in the weight (kg) classes are: 0.0-0.1 (mice); 0.1-0.4 (chipmunk,
cotton rat, woodrat, and Norway rat); 0.4-2.0 (squirrel, rabbit, muskrat); 2.0-
16.0 (opossum, raccoon); and 16.0-120.0 (deer).

Bobcat Winter Food Habits 121

CONCLUSIONS

The results of our study are similar to previous winter food habits
studies in the southeastern United States and indicate rabbits, birds,
cotton rats, and deer were the top four prey items by percent occurrence
in stomachs of bobcats collected statewide in North Carolina during
winter. Furthermore, adult males appeared to select the larger prey
items (>2.0 kg), and adult females and kittens the smaller prey items
(<0. 1 kg). All sex and age classes apparently preferred intermediate-
sized prey (0.1 - 2.0 kg).

The identification of food habits has management imphcations.
The prey species most used by bobcats as food items were species
adapted to early successional stages (e.g., rabbits, cotton rat, and deer).
An interpretation based on prey selection suggests bobcats are not a
climax species, but rather a subclimax species, benefitting from land
management practices that maintain early successional habitats.

ACKNOWLEDGMENTS— The U. S. Fish and Wildlife Service, the
N. C. Wildlife Resources Commission (NCWRC), the N. C. Agricul-
tural Research Service, and North Carolina State University provided
funding. The NCWRC collected carcasses from fur dealers and made
them available to us. Numerous trappers cooperated by providing car-
casses. D. Hazel, L. Petrovick, D. Stewart, J. Wooding, J. Henegar, S.
Habel, and K. Lewis assisted in the necropsy of the carcasses. A. Bras-
well of the N. C. State Museum of Natural History identified the herpe-
tological material.

LITERATURE CITED

Bailey, Theodore N. 1974. Social organization in a bobcat population. J. Wildl.
Manage. 55:435-446.

Barr, Anthony J., J. H. Goodnight, J. P. Sail and J. T. Helwig. 1979. User's
guide to SAS 1979. SAS Institute, Inc., Raleigh. 474 pp.

Barrick, Frank B. 1969. Deer predation in North Carolina and other southeast-
ern states, pp. 25-31 in L. K. Halls (ed.). White-tailed deer in the southern
forest habitat. South. For. Exp. Sta., USDA, Nacogdoches, TX. 130 pp.

Beasom, Samuel L., and R. A. Moore. 1977. Bobcat food habit response to a
change in prey abundance. Southwest. Nat. 27:451-457.

Buttrey, George W. 1974. Food habits and distribution of the bobcat, Lynx
rufus rufus (Schreber), on the Catoosa Wildlife Management Area. Tenn.
Wildl. Res. Agency Tech. Rep. No. 75-12. Nashville. 64 pp.

Crowe, Douglas M. 1975. Aspects of aging, growth and reproduction of bobcats
from Wyoming. J. Mammal. 55:177-198.

Davis, James R. 1955. Food habits of the bobcat in Alabama. Unpubl. M. S.
Thesis, Alabama Polytech. Inst., Auburn. 79 pp.

122 Anne M. King, et al.

Fox, Lloyd B., and J. Fox. In press. Population characteristics and food habits
of bobcats in West Virginia. Proc. Annu. Conf. Southeast. Assoc. Fish
Wildl. Agencies.

Fritts, Steven H. 1973. Age, food habits and reproduction of the bobcat {Lynx
rufus) in Arkansas. Unpubl. M. S. Thesis, Univ. Arkansas, Fayetteville. 80 pp.

, and J. A. Sealander. 1978. Diets of bobcats in Arkansas with special

reference to age and sex differences. J. Wildl. Manage. 2^:226-228.

Hall, Harlan T., and J. D. Newsom. 1976. Summer home ranges and move-
ments of bobcats in bottomland hardwoods of southern Louisiana. Proc.
Annu. Conf. Southeast. Assoc. Fish Wildl. Agencies i0:427-436.

Kight, James. 1962. An ecological study of the bobcat {Lynx rufus Schreber), in
west-central South Carolina. Unpubl. M. S. Thesis, Univ. Georgia, Athens.
52 pp.

Korschgen, Leroy J. 1980. Procedures for food habits analysis, pp. 1 13-127 in S.
D. Schemnitz (ed.). Wildlife Management Techniques, 4th ed. The Wildlife
Society, Washington. 686 pp.

Lancia, Richard A., D. K. Woodward and S. D. Miller. In press. Summer
movement patterns and habitat use by bobcats on Croatan National Forest,
North Carolina. International Cat Symposium, Texas A&I Univ., Kings-
ville, Oct. 4-6, 1982.

Merriwether, David, and M. K. Johnson. 1980. Mammalian prey digestibility
by coyotes. J. Mammal. 67:774-775.

Miller, Stuart D., and D. W. Speake. 1978. Prey utilization by bobcats on quail
plantations in southern Alabama. Proc. Annu. Conf. Southeast. Assoc.
Fish Wildl. Agencies 52:100-1 1 1.

1980. Ecology of the bobcat in south Alabama. Unpubl. Ph.D. Dis-
sert., Auburn Univ., Auburn. 168 pp.

Moore, Tommy D., L. W. Spence and C. E. Dugonolle. 1974. Identification of
the dorsal guard hairs of some mammals of Wyoming. Wyoming Game
FishDep. Bull. 14. 177 pp.

Progulske, Donald R. 1952. The bobcat and its relation to prey species in Virgin-
ia. Unpubl. M. S. Thesis, Virginia Polytech. Inst., Blacksburg. 135 pp.

Spiers, James E. 1973. A microscopic key to the hairs of Virginia land mam-
mals. Unpubl. M. S. Thesis, Virginia Polytech. Inst., Blacksburg. 106 pp.

Stuckey, Jasper L. 1965. North Carolina: Its Geology and Mineral Resources.
N. C. Dep. Conserv. Develop., Raleigh. 550 pp.

Weaver, John L., and S. W. Hoffman. 1979. Differential detectability of rodents
in coyote scats. J. Wildl. Manage. -^3:783-786.

Wilcoxon, Frank. 1945. Individual comparisons by ranking methods. Biomet-
rics 1:80-83.

Accepted 17 September 1983

Foods and Feeding Behavior of Sauger,

Stizostedion canadense (Smith) (Pisces: Percidae),

from GaUipoUs Locks and Dam, Ohio River

Steven I. McBride and Donald Tarter

Department of Biological Sciences,
Marshall University, Huntington, West Virginia 25701

ABSTRACT. — The food habits of 151 sauger, Stizostedion canadense
(Smith), from the Ohio River at Gallipolis Locks and Dam, were stu-
died from March 1981 through April 1982. Seventy percent of the
stomachs contained food, which was analyzed for identification. Fishes
were the primary food, both by weight (99.7%) and frequency of
occurrence (100%). Emerald shiners, Notropis atherinoides Rafinesque,
and gizzard shad, Dorosoma cepedianum (Lesueur), comprised 96% of
all identifiable items. Emerald shiners were numerically and by weight
the most important item; they occurred in 49% of stomachs that con-
^' tained food, and made up 32% of food biomass. Terrestrial and aquat-
ic invertebrates were of minor dietary importance. Seasonal variation
in sauger foods was minimal. Size had no influence on the kind of
food contained in the stomachs of adult sauger. Sauger appeared to
feed most intensively in the fall. Regression analysis showed a signifi-
cant negative correlation (r = -0.87) between sauger caught by anglers
and water temperature.

INTRODUCTION

The sauger, Stizostedion canadense (Smith), is an important sport
and commercial fish (Scott and Grossman 1973). Information on its
foods and feeding behavior are essential to our understanding of percid
communities and to the effective management of these resources. Sev-
eral investigators, including Forbes (1878, 1888) and Richardson (1920),
Pearse (1921), Dendy (1946), and Priegel (1969), provided brief notes on
sauger food habits, but only one detailed study was concerned with a
population in a riverine habitat (Vanicek 1964). Our study was initiated
in March 1981 and continued through April 1982. Its objectives were to
determine: (1) what foods are most important to the sauger population,
(2) if food habits varied with season, (3) if food habits changed as
sauger increased in size, and (4) how water temperature correlated with
sauger catch.

METHODS AND MATERIALS
Study Area

Gallipolis Locks and Dam is one in a series of flood and naviga-
tional control dams situated on the Ohio River. It is located at river
mile (RM) point 279.2 (38°40'53'', 82°ir22"), between Galia County,

BrimleyanaNo.9:123-134. June 1983. 123

124 Steven I. McBride and Donald Tarter

Ohio, and Mason County, West Virginia (U. S. Army Corps of Engi-
neers 1980a). The Gallipolis pool extends 67.6 km (42 mi.) upriver to
Racine Locks and Dam and 49.4 km (31 mi.) up the Kanawha River to
Winfield Locks and Dam. The Gallipolis Dam marks the northern
boundary of the Greenup navigational pool, which extends 75.8 km (62
mi.) downriver to Greenup Locks and Dam. The floodplain of the river
at the Gallipolis Dam is slightly more than 1.61 km (1 mi.) wide. The
river channel has an average width of 366 m, and its depth varies from
approximately 7 to 10 m, with a navigable depth of 2.7 m (U. S. Army
Corps of Engineers 1980b).

Specimens were collected from several locations in the Greenup
pool and once from inside Belleville Locks and Dam. However, the
main collecting site was situated just below Gallipolis Locks and Dam.
This area is steeply sloped and becomes very rocky, then levels toward
the shoreline. Approximately 2 m from shore a submerged concrete wall
runs parallel with the shoreline for nearly 100 m. This wall provides an
excellent habitat for many types of aquatic life and fish are attracted to
the area.

Methods

Three methods were used to collect sauger. A weekly creel survey
of Gallipolis Locks and Dam conducted by the authors from 15 March
1981 through April 1982. Surveys resulted in 99 sauger stomachs,
donated by cooperative anglers. At various times during the study, gill
netting was conducted in the Gallipolis pool as a supplement to the creel
survey. Twelve days of netting yielded 18 sauger. Rotenone sampling
was done at Belleville Locks on 30 September 1981 in conjunction with
the Ohio River Valley Water Sanitation Commission (ORSANCO) and
the West Virginia Department of Natural Resources (WVDNR). Thirty-
four sauger were obtained,

Sauger and/ or sauger stomachs were transported on ice to the
laboratory, then frozen. Specimens were later thawed, and total length
of each fish measured to the nearest mm. Stomachs were excised and
their contents washed into a watch glass for sorting into various taxo-
nomic categories with the aid of a dissecting microscope. The number of
individuals in each taxon was recorded for each stomach. Standard tax-
onomic keys were used for identifications (list of references is available
from authors).

The following calculations were made for each monthly collection:
(1) percentage of stomachs in which a particular taxon occurred (per-
centage frequency of occurrence), (2) percentage of the total number of
food items that one taxon comprises, and (3) percentage of the total
volume (by weight) of all taxa made up by the combined assigned
weight of each particular taxon.

Foods and Feeding of Sauger 125

In order to overcome bias due to differences in stages of digestion
of food items, the original weight of individuals can be estimated
(Lagler 1956). This is done by averaging weight of 30 undigested indi-
viduals of the species. The figure obtained is an approximate "weight
when alive" value.

To determine if seasonal variations occur in the diet of the sauger,
151 specimens were arbitrarily grouped according to season: spring
(March, April, May), summer (June, July, August), and fall (Sep-
tember). The number of empty stomachs from each season also was
recorded, to determine if there was any seasonality in the percentage
frequency of their occurrence.

Specimens were divided into two total-length classes — 150 mm
and less, and 151 mm-400 mm — to determine if food varied with
length.

Our study was concurrent with a WVDNR creel survey, conducted
from 1 April through 31 November 1981.

The relationship between water temperature and sauger catch (from
the WVDNR creel survey) was determined by regression analysis and
tested with a coefficient of correlation. Temperatures used for the
regression analysis consisted of means calculated from monthly average
values from the Gallipolis Locks and Dam, and single day values from
other dams included in the investigation. Temperature data were
obtained from an electronic water quality monitoring system and a
manual sampling system, both maintained by ORSANCO.

RESULTS AND DISCUSSION

Of the 151 sauger stomachs examined, 105 (69.5%) contained food.
The number with food was highest in September (89.7%) and lowest in
July (50%) (Fig. 1). Thus, sauger appeared to feed most intensively in
the fall.

Food Items

Fishes were the primary food of sauger, both by weight (99.7%)
and by frequency of occurrence (100%) (Table 1). Emerald shiners,
Notropis atherinoides Rafinesque, and gizzard shad, Dorosoma cepedi-
anum (Lesueur), accounted for 56.6% of the total volume of food con-
sumed (Table 1). Together these two forage fish comprised 96% of all
identifiable food items (Table 1).

Emerald shiners were the most important food item by number,
occurring in 49% of stomachs that contained food, and by weight mak-
ing up 32% of food biomass (Table 1). Gizzard shad were second in
importance appearing in 11% of stomachs that contained food and
comprising 25% of the total volume of food consumed (Table 1). Other
identifiable food fishes were striped shiners, A^. chrysocephalus (Rafi-

126

Steven I. McBride and Donald Tarter

100

CO

40

Z 30

liJ

O

S 20

Q.

10

M A M J J A S
MONTHS

Fig. 1. Monthly fluctuations in percent of sauger stomachs containing food,
Ohio River at Gallipolis Locks and Dam.

Foods and Feeding of Sauger

127

Table 1. Food organisms consumed by sauger from the Ohio River at Gallipolis
Locks and Dam, March 1982-April 1982. N = total number of
organisms, % N = percentage of all organisms combined, % FO = percent
frequency of occurrence of each organism, and % Wt = percentage of
the total weight of food items.

Food Organism

N
160

%N
96.9

%FO
100

% Wt

FISH

99.7

Unidentified

50

30.3

42.8

41

Cyprinidae

91

55.1

48.6

32.3

Notropis atherinoides

90

54.5

47.6

32

N. chrysocephalus

1

0.6

0.9

0.3

Clupeidae

17

10.3

11.4

24.6

Dorosoma cepedianum

17

10.3

11.4

24.6

Sciaenidae

1

0.6

0.9

1.8

Aplodinotus grunniens

1

0.6

0.9

1.8

AQUATIC INVERTEBRATES

3

1.8

2.9

0.2

Insecta

3

1.8

2.9

0.2

Diptera

3

1.8

2.9

0.2

Chironomidae larvae

3

1.8

2.9

0.2

TERRESTRIAL INVERTE-

BRATES

2

1.2

1.9

0.1

Insecta

2

1.2

1.9

0.1

Coleoptera

1

0.6

0.9

0.05

Orthoptera

1

0.6

0.9

0.05

nesque), and freshwater drum, Aplodinotus grunniens (Rafinesque).
Fishes digested beyond identification were found in 43% of the stom-
achs (Table 1).

Terrestrial and aquatic invertebrates occurred in 3% of stomachs
containing food and comprised 0.3% of the total volume of food con-
sumed (Table 1).

Forbes and Richardson (1920) examined the stomach contents of
14 sauger and concluded that this species feeds entirely on fishes. Pearse
(1921) listed the food of sauger as 75% fish, and 25% insects and plank-
ton. Dendy (1946) found food in 1,980 of 3,807 sauger stomachs col-
lected in 1943 and 1944 from Norris Reservoir. The stomach contents
consisted chiefly of shad, with crappie (Pomoxis) second in importance.
Priegel (1969) found Lake Winnebago sauger utilizing troutperch, Per-
copsis omiscomaycus (Walbaum), as their major food item. Young and

128 Steven I. McBride and Donald Tarter

yearling freshwater drum were also an important item in Priegel's study.
Vanicek (1964) examined 133 sauger collected from Lewis and Clark
Lake and the Missouri River. He found young gizzard shad to be the
most important food for both reservoir and river sauger. Although the
emerald shiner was abundant in Lewis and Clark Lake, it appeared in
only 5% of the stomachs. Swenson (1977) reported that troutperch
were the primary source of food for sauger in Lake of the Woods, Min-
nesota.

Seasonal Food Habits

Seasonal variation in sauger diet was minimal (Figs. 2 and 3).
Fishes were the dominant food item by number (100% FO) and weight
(99.7%) throughout the year (Table 1). Emerald shiners were most
important in both numerical frequency and frequency of occurrence at
all seasons. Gizzard shad ranked second in importance by number
throughout the year.

Volumetric proportions of food items generally coincided with per-
cent numbers in the diet. Emerald shiners were the most important iden-
tifiable food item volumetrically in both spring (38%) and summer
(29%). Gizzard shad comprised the largest proportion of food volume
(36%) during the fall (Fig. 3).

The greatest variation in seasonal feeding was an increase in the
number of stomachs containing food from spring (61%) and summer
(64%) to fall (89.7%).

There have been relatively few food habit studies documenting sea-
sonal variation in diet. Dendy (1946) reported that it was not uncom-
mon to find as many as 30 small shad in an individual stomach. The
shad, tending to "hibernate," are undoubtedly easy prey for sauger in
winter. Priegel (1969) reported that in the winter. Lake Winnebago
sauger consumed equal amounts of troutperch and emerald shiners.
When forage fishes were scarce, however, sauger consumed more chiro-
nomid larvae.

Food Habits by Length Class '

150 mm and less. — Our methods enabled us to collect no data
concerning food of sauger of this length class. Priegel (1969) found the
food of young sauger to vary with size, changing from zooplankton to
chironomid larvae, to immature and adult mayflies. He considered
sauger less than 50 mm to be plankton feeders; however, at 50 mm they
would cease consuming plankton if forage fishes were extremely abun-
dant. Of the invertebrates utilized by sauger in the 12-50 mm size class,
Daphnia sp. was the most important item consumed. The major forage
fishes consumed by young sauger were troutperch, freshwater drum,
and white bass, Morone chrysops (Rafinesque).

Foods and Feeding of Sauger

129

ES

GS

D

SS
Al
Tl

Spring (54)

ES

GS

D

SS
Al
Tl

Summer (58)

Fall (39)

ES

GS

D

SS
Al
Tl

0 10 20 30 40 50 60 70 80 90 100

PERCENT OCCURRENCE

Fig. 2. Seasonal frequency of occurrence of food items in sauger stomachs, Ohio
River at Gallipolis Locks and Dam. ES = emerald shiners, GS = gizzard shad, D
= freshwater drum, SS = striped shiners, Al = aquatic invertebrates, Tl = terres-
trial invertebrates.

130

Steven I. McBride and Donald Tarter

LO

-I

LO

ttC </) H--

oo

OJ

i^

''^

■a

i^rf

« J-

F-S

4) .C

= ;;=

a> (/>

LO

« 2

c/>

Foods and Feeding of Sauger 131

151-400 mm. — Fishes were the number one food item by weight
(99.7%) and by frequency of occurrence (100%) of sauger in this length
class. Emerald shiners were the most important forage fish, with gizzard
shad second in importance. Dendy (1946) found that for sauger whose
standard lengths ranged from 200 to 480 mm, size had no influence on
the kind of food contained in the stomach. We came to a similar con-
clusion for fish in that size range. However, examination of 21 fish in
the 150-200 mm size range showed that none had consumed gizzard
shad, but all stomachs with identifiable food items contained emerald
shiners. This supports Vanicek (1964), who found that young-of-the-
year and yearling sauger in July and August consumed mostly Notropis,
despite the fact that gizzard shad were the most abundant forage fish
available.

Catch Related to Temperature

The WVDNR creel survey of the Ohio River tallied 6,499 sauger
harvested from the tailwaters of six dams on the river: New Cumberland
(RM 54.4), Pike Island (RM 84.2), Hannible Island (RM 126.4), Willow
Island (RM 161.8), Belleville (RM 203.9), and Gallipolis (RM 279.2)
dams.

Regression analysis of the relationship between sauger catch and
water temperature showed a significant negative correlation (r = -0.87)
(Fig. 4). Sauger catch was greatest in November (2,268) with the mean
water temperature at 54° F, and smallest in August (44) with the mean
water temperature at 78.8° F.

Factors other than temperature may relate to sauger catch in the
Ohio River. Doan (1941) found a high correlation (r = =0.m9) between
catch and turbidity for Lake Erie sauger.

CONCLUSIONS

Populations of emerald shiners and gizzard shad dominated the
supply of forage fish in the Ohio River during the study period. Clay
and Carter (1962) reported that the emerald shiner was by far the most
abundant species in the Ohio River, making up 57.8% of 741,438 fish
collected in a three-year period. In the same study, gizzard shad were
second in importance by number (14.5%) and most important by
weight, comprising 44.6% of 16 tons of fish. The future of sauger popu-
lations in the Ohio River is apparently dependent on the existence of
large numbers of gizzard shad and emerald shiners. Whether or not
sauger would become more opportunistic feeders in the absence of these
two forage species remains questionable. However, in all likelihood
sauger, as well as other piscivorous fishes, would suffer to some degree
if either of these food sources becomes depleted.

132

Steven I. McBride and Donald Tarter

2200-1
2000
1800H
1600
1400
5 1200
« 1000

C/3

800-

600

400

200

y=4698.63 + 58.12
r=-0.87

~T — I — I — I — I — r~
0 20 40 60

80 100

TEMPERATURE

Fig. 4. Relationship of water temperature (F) and sauger catch from tailwaters
of New Cumberland, Pike Island, Hannible, Willow Island, Belleville, and Gal-
lipolis dams, Ohio River.

Foods and Feeding of Sauger 133

If sauger catch can be interpreted as active feeding behavior, then
the negative coefficient of correlation (r= -0.87), obtained from examining
the relation of sauger catch to water temperatures, may indicate that
sauger feed most intensively during periods of cool water temperatures.
This is supported by the highest percentage of stomachs with food
(89.7%) occurring in the fall.

ACKNOWLEDGMENTS.— The authors thank Mr. Glenn Moore,
ORSANCO, for supplying the water quality data for this study. We also
express our gratitude to all the members of ORSANCO and WVDNR
who were helpful in collecting fish. Special thanks go to Bernie Dowler,
Assistant Chief of Warmwater Fisheries, WVDNR, for supplying creel
survey data.

Several persons assisted in the field work during this study. We are
particularly grateful to Curtis Hardman, George Bakewell and Mark
Sheridan. Also, we express our appreciation to Ms. Vickie Crager for
typing the manuscript.

LITERATURE CITED

Clay, William M., and B. T. Carter. 1962. Aquatic-life resources of the Ohio
River. Ohio River Valley Water Sanitation Commission, Cincinnati. 218 pp.

Dendy, Jack S. 1946. Food of several species of fish, Norris Reservoir, Ten-
nessee. J. Tenn. Acad. Sci. 27(1):105-127.

Doan, Kenneth H. 1941. Relation of sauger catch to turbidity in Lake Erie.
Ohio J. Sci. ¥/(6):449-452.

Forbes, Stephen A. 1878. The food of Illinois fishes. Bull. 111. State Lab. Nat.
Hist. l(2):71-89.

1888. On the food relations of freshwater fishes. Bull. 111. State Lab.

Nat. Hist. 2:475-538.

, and R. E. Richardson. 1920. The fishes of Illinois. 111. Nat. Hist.

Survey, Vol. 3, Ichthyology. 278 pp.
Lagler, Karl F. 1956. Freshwater Fishery Biology. W. C. Brown Co., Dubuque,

Iowa. 421 pp.
Pearse, A. S. 1921. The distribution and food of the fishes in three Wisconsin

lakes in summer. Univ. Wise. Stud. Sci. No. 3. 61 pp.
Priegel, Gordon R. 1969. The Lake Winnebago Sauger. Wis. Dep. Nat. Resour.

Tech. Bull. 43. 63 pp.
Scott, W. B., and E. J. Crossman. 1973. Freshwater Fishes of Canada. Bull.

Fish. Res. Board Can. 184. 966 pp.
Swenson, William A. 1977. Food consumption of walleye {Stizostedion vitreum

vitreum) and sauger {Stizostedion canadense) in relation to food availability

and physical conditions in Lake of the Woods, Minnesota, Shagawa Lake, and

western Lake Superior. J. Fish. Res. Board Can. i^( 10): 1643-1654.
, and L. L. J. Smith. 1973. Gastric digestion, food consumption,

feeding periodicity, and food conversion efficiency in walleye {Stizostedion

vitreum vitreum). J. Fish. Res. Board Can. i0(9): 1327- 1336.

134 Steven I. McBride and Donald Tarter

and 1976. Influence of food competition, predation.

and cannibalism on walleye {Stizostedion vitreum vitreum) and sauger {S.

canadense) populations in Lake of the Woods, Minnesota. J. Fish. Res.

Board Can. ii(9): 1946- 1954.
S. Army Corps of Engineers. 1980a. Gallipolis Locks and Dam Replacement,

Ohio River. Phase 1, Advanced Engineering and Design Study. General

Design Memorandum. Appendix J, Volume 2. 202 pp.
1980b. GalHpolis Locks and Dam Replacement, Ohio River. Phase

1, Advanced Engineering and Design Study. General Design Memorandum.
Main Report. 89 pp.
Vanicek, C. D. 1964, Age and growth of sauger, Stizostedion canadense
(Smith), in Lewis and Clark Lake. Iowa State J. Sci. 55(4):48 1-502.

Accepted 3 February 1983

Notes on Breeding Period, Incubation Period, and Egg

Masses oi Amby stoma jeffersonianum (Green) (Amphibia:

Caudata) from the Southern Limits of Its Range

Charles K. Smith ^

Department of Biology,
University of Kentucky, Lexington, Kentucky 40502

ABSTRACT. — Observations on breeding and incubation periods of
Ambystoma jeffersonianum were made at three central Kentucky
ponds during 1980 and 1981. Breeding began in early January, 1980
and in early February, 1981. There were also substantial differences
between years in duration of the breeding period. During both years
hatching occurred over a two-week period between late March and
early April. Mean number of eggs per mass was 23.4 ± 13.5, and mean
mass dimensions (length x diameter, mm) were 43.1 ± 11.9 x 39.0 ±
8.6. There are indications that the number of eggs per mass for Ken-
tucky populations is greater than for more northern populations.

INTRODUCTION

Relatively little life history data exist for Ambystoma jeffersonia-
num (Green). Data published prior to Uzzell's (1964) definitive work on
the A. jeffersonianum complex may not, in fact, deal with A. jeffersoni-
anum. The species ranges throughout most of the northeastern United
States and reaches its southern limits in central Kentucky. Life history
data from Kentucky populations, since they are peripheral, should shed
light on factors that limit the species' distribution. Unfortunately, very
few published studies concern Kentucky populations. This note provides
observations on breeding, incubation periods, and characteristics of egg
masses of A. jeffersonianum in central Kentucky. To investigate annual
and local variation in breeding and incubation periods, observations
were made in three local ponds over a two-year period.

STUDY SITES AND METHODS
Data were collected during 1980 and 1981 from three abandoned,
semi-permanent farm ponds in Jessamine County. Each pond had a sur-
face area of less than 288 square meters, and an average depth of 21 cm.
Pond 1 is located 0.4 km north of Pond 2, and Pond 3 is approximately
1.5 km northwest of Pond 1. Ambystoma jeffersonianum is the only
ambystomatid salamander that regularly uses any of the ponds, although
over a four-year period I have occasionally observed eggs and larvae of
A. texanum in Ponds 1 and 2.

• Present Address: Department of Biology, University of North Carolina, Chapel Hill,
North Carolina 27514.

Brimleyana No.9: 135-140. June 1983. 135

136 Charles K. Smith

The ponds were visited during the day once a week, and on every
rainy night beginning with the first rains of late fall (19 December 1979
and 1 November 1980). During each visit the ponds were searched for
adults and egg masses. Developmental stages of the embryos were used
to determine lengths of the breeding, incubation, and hatching periods.
Masses in early stages of development (preblastula) were assumed to be
recently deposited. In addition, data on the number of eggs per mass,
mass size (greatest length and diameter), and mass placement were col-
lected from Pond 3 on 23 March 1980 and 8 March 1981 while transects
were waded across the pond.

RESULTS AND DISCUSSION

A summary of breeding and incubation periods oi Amby stoma jef-
fersonianum is given in Table 1 . Substantial differences occurred between
years in the initiation and duration of the breeding season. Hatching,
however, occurred over a 1 to 2 week period at approximately the same
time during both years.

Since the migration oi Amby stoma to breeding ponds is correlated
with precipitation and temperature (Baldauf 1952; Whitford and Vin-
egar 1966; Douglass and Monroe 1981), variable weather conditions

Table 1. Summary of dates and duration of breeding and incubation periods for
Ambystoma jeffersonianum in three central Kentucky Ponds. '

1980

1981

Ponds

1
19

2

3

P

2

3

Initiation of

3

22

21

21

21

breeding period

Jan.

Jan.

Jan.

Feb.

Feb.

Feb.

Egg deposition

6

11

9

-

3

3

period (wks)

Incubation

period (wks)

4-10

4-13

4-10

-

3-6

3-6

Hatching

23 Mar.

30 Mar.

30 Mar.

-

28 Mar.

28 Mar.

period

5 Apr.

5 Apr.

13 Apr.

-

5 Apr.

5 Apr.

' Dates are approximate

, based on first observation of the event. Duration of an

event is based on observations of first and last occurrence.

2 Drying destroyed all eggs in pond 1 during 1981;

see text.

Ambystoma jeffersonianum Reproduction 137

offer the most probable explanation for the differenes between years in
the date that breeding was initiated. Noticeably less precipitation oc-
curred at the study site between November 1980 and January 1981 than
for the same period during the previous year. Whereas the ponds filled
by January in 1980, they were not filled until February in 1981.

Differences between years in local variation in the duration of the
breeding periods, and, hence, incubation periods can best be explained
if migration to the ponds is both a function of favorable weather condi-
tions and an individual's physiological readiness to respond to those
conditions. During 1980, by the time weather conditions were favora-
ble for migration to the ponds, a relatively large group of individuals
may have been ready to migrate. Thus, a more synchronous mass
migration to the ponds resulted, and breeding and egg deposition
occurred over a shorter time period.

Synchrony of hatching is of particular interest, given the variation
that occurred in initiation and duration of the breeding season. Tem-
poral variation in oviposition was buffered and did not produce equal
variation in initiation and duration of the hatching period. Temperature
and developmental rates show positive correlation (Moore 1939; Has-
singer et al. 1970). Because ambient temperatures tend to progressively
increase during the incubation period, embryos of eggs deposited earlier
in the breeding season take longer to develop, and, as a result, the tem-
poral variation of the hatching period is much less in comparison to the
breeding period.

Worthington (1968, 1969), and Douglas and Monroe (1981) pro-
posed that temporal staggering of breeding is adaptive because it results
in a partitioning of resources among larvae of different sizes. Since the
hatching period was much shorter than the breeding period, my study
suggests that the efficacy of staggered breeding in partitioning pond
resources may be greatly reduced.

At a breeding pond approximately 135 km west of my study site,
Douglas and Monroe (1981) reported that adult A. jeffersonianum were
present from late November until early March. They did not note if
oviposition occurred over the entire period. They found a peak occur-
rence of adults during early January, as did Creusere (1972) for popula-
tions in western and central Kentucky. Seibert and Brandon (1960)
found eggs in southern Ohio as early as February, while Smith (1911)
and Bishop (1941) reported that breeding begins in late March or early
April in New York. Brandon (1961) and Creusere (1972) observed hatch-
lings in western Kentucky on dates similar to those reported here.
Bishop (1941) reported hatching to occur during April and May in New

138 Charles K. Smith

York populations. Geographic differences in initiation of breeding are
likely due to differences in temporal occurrence of weather conditions
that stimulate adult migration to the ponds. The later hatching time for
northern populations corresponds with a later breeding season.

During 1981 my study ponds filled slowly, and water levels fluctu-
ated greatly during the breeding and incubation periods. Ponds 2 and 3
did not dry until late summer, and pond drying was not a source of
embryo mortality. Pond 1, however, dried completely by 22 March, des-
troying all egg masses. Although pond 1 partially refilled before April,
no additional egg masses were found.

Egg mass data are presented in Table 2. The mean number of eggs
per mass (23.4) is similar to that reported by Brandon (1961) for popu-
lations in southern Ohio (x = 22), but greater than that reported for New
York populations by Bishop (1941; x = 16), Smith (1911; x = 14), and
Uzzell (1967; x = 14). Further study of geographic variation in the
number of eggs per mass is required in order to determine if this appar-
ent north-to-south cline is real.

Table 2. Selected data, egg masses oi Amby stoma jeffersonianum in a central
Kentucky pond (pond 3). R = range, N = sample size.

X

1 SD

R

N

Number of

eggs per mass

23.4

13.5

2-67

105

Length of

single mass (mm)

43.1

11.9

30-70

26

Diameter of

single mass (mm)

39.0

8.6

20-60

26

Length of masses

in series (mm)

117.0

40.4

40-200

13

Number of

masses in series

3.4

1.0

2-5

13

Depth to top

of mass (cm)

14.3

3.3

7-24

64

Total water depth

at mass (cm)

25.3

3.0

20-30

64

Ambystoma jeffersonianum Reproduction 139

Masses were found attached both singly and in series to blades of
herbaceous vegetation, petioles of leaves, and twigs. The descriptions of
attachment sites and mass dimensions agree with those of Uzzell (1967)
and Bishop (1941).

ACKNOWLEDGMENTS.— I thank E. K. Smith for her assistance
in the field, and J. W. Petranka for his critical review of the manuscript
and discussion of the problem. This note was taken from a larger study
of larval ecology of A. jeffersonianum which was completed for the M.
S. degree under the direction of R. W. Barbour, and was supported in
part by two Graduate Student Grants from the University of Kentucky.

LITERATURE CITED

Bishop, Sherman C. 1941. The Salamanders of New York. N. Y. State Mus.
Bull. 324. 365 pp.

Baldauf, Richard J. 1952. Climatic factors influencing the breeding migration of
the spotted salamander, Ambystoma maculatum (Shaw). Copeia 1952 (3):
178-181.

Brandon, Ronald A. 1961. A comparison of the larvae of five northeastern spe-
cies oi Ambystoma (Amphibia: Caudata). Copeia 1961 (4): 377-383.

Creusere, F. Michael. 1972. Behavioral and ecological aspects of five ambysto-
matid species. Unpubl. M. S. Thesis, Eastern Kentucky Univ., Richmond.
140 pp.

Douglas, Michael E., and B.L. Monroe Jr. 1981. A comparative study of topo-
graphical orientation in Ambystoma (Amphibia: Caudata). Copeia 1981
(2): 460-463.

Hassinger, Dawn D., J. Anderson and G. H. Dalrymple. 1970. The early life
history and ecology oi Ambystoma tigrinum and A. opacum in New Jersey.
Am. Midi. Nat. 84: 474-495.

Moore, John A. 1939. Temperature tolerance and rates of development in the
eggs of amphibia. Ecology 20: 459-478.

Smith, Bertram G. 1911. Notes on the natural history oi Ambystoma jeffersoni-
anum. A. opacum, and A. tigrinum. Bull. Wisconsin Nat. Hist. Soc. 9:
14-27.

Seibert, Henri C, and R.A. Brandon. 1960. The salamanders of southeastern
Ohio. Ohio J. Sci. 60\ 291-303.

Uzzell, Thomas M. 1964. Relations of the diploid and triploid species of
Ambystoma jeffersonianum complex (Amphibia: Caudata). Copeia 1964
(2): 257-300.

Whitford, Walter G., and A. Vinegar. 1966. Homing, survivorship, and over-
wintering of larvae oi Ambystoma maculatum. Copeia 1966 (3): 515-518.

140 Charles K. Smith

Worthington, Richard D. 1968. Observations on the relative size of three species
of salamanders in a Maryland pond. Herpetologica 24: lAl-lAd.

1969. Additional observations on sympatric species of salamander

larvae in a Maryland pond. Herpetologica 25: 227-229.

Accepted 10 March 1984

Occurrence and Habitat Preference of

Fundulus luciae (Baird) (Pisces: Cyprinodontidae) on

a Southeastern North Carolina Salt Marsh

Mark A. Shields and Carol H. Mayes

Department of Biological Sciences,

University of North Carolina at Wilmington,

Wilmington, North Carolina 28403

ABSTRACT.— Thirty-two Fundulus luciae were collected in pit traps
on a salt marsh in southeastern North Carolina during October and
November 1983. This species remained on the marsh at low tide,
exhibiting a strong preference for high marsh areas dominated by Jun-
cus roemerianus while tending to avoid low marsh habitats dominated
by Spartina alterniflora. The reported rarity of F. luciae probably
reflects inadequate sampling of its preferred high marsh habitat by
conventional collecting techniques.

INTRODUCTION

Fundulus luciae (Baird), the spotfin killifish, is found sporadically
in brackish coastal waters from Long Island, New York (Butner and
Brattstrom 1960) to Georgia (Jorgenson 1969), with most populations
apparently concentrated in the Chesapeake Bay area (Lee et al. 1980).
In North Carolina, Hildebrand (1941) reported F. luciae common only
in marsh pools on Shackleford Banks and along the Newport River,
Carteret County. More recently, Kneib (1978) found F. luciae to be
common on the high salt marsh at Tar Landing Bay on Bogue Sound,
also in Carteret County. Few records of F. luciae from other areas of
coastal North Carolina have been published. In this note we document
the occurrence and describe the habitat preference of F. luciae on a salt
marsh in southeastern North Carolina.

STUDY AREA AND METHODS

Our study site was a salt marsh bordering Bradley Creek, a small
tidal river in New Hanover County. Three marsh habitats were defined
within the study area: tall-form Spartina alterniflora (TS), short-form
S. alterniflora (SS), and Juncus roemerianus (JU). TS formed a band 1
to 3 m wide along the levees of the creek and smaller rivulets in the
marsh, while SS vegetated the flat areas behind the creekbank levees.
The higher areas of marsh farthest from the creek were dominated by JU.

One 25 X 25 X 25 cm pit trap was dug in each marsh habitat. Pits
in TS, SS, and JU were placed 1.5, 18.5, and 51.5 m, respectively, from
Bradley Creek. Each pit was lined with 1 mm mesh cloth to facilitate
removal of trapped fish. Traps were visited during daylight at low tide

Brimleyana No.9: 141-144. June 1983. 141

142 Mark A. Shields and Carol H. Mayes

when the marsh was completely drained and all fish were removed and
preserved in 10% formalin. Standard length (SL) of all specimens was
measured to the nearest mm. Specimens were deposited in the Univer-
sity of North Carolina at Wilmington (UNCW) Fish Collection.

Eleven collections were made between 19 October and 20 November
1983. Salinity of water in the pit traps ranged from 23 ppt to 33 ppt
during the study, with salinities among the three pits varying no more
than 3 ppt on any given day. Water temperature ranged from 7° C to
19° C.

The percentage of each of the three habitats in a 0.5 ha area cen-
tered over the pit traps was estimated visually from an aerial photo-
graph. TS comprised 14.5% of the available marsh habitat, SS
accounted for 60.9%, and JU comprised 24.6%. A Chi-square test,
based on the null hypothesis that fish moved at random over the marsh
at high tide and used each habitat in proportion to its availability, was
employed to measure habitat preference by F. luciae.

RESULTS AND DISCUSSION

A total of 32 F. luciae (mean SL 19.4 mm, range 5-33 mm) was
collected. The largest number trapped in one day was 12 on 11 November.
No F. luciae were caught on 14 November when mean water tempera-
ture in the pit traps was 7° C, the coldest temperature recorded during
our study. Since F. luciae becomes relatively inactive and therefore
more difficult to catch during cold weather (Byrne 1978; Kneib 1978), it
may have been even more abundant on Bradley Creek Marsh than our
limited autumn collections indicated.

Distribution of F. luciae among the three marsh habitats was not
random {x^ =8.70, df=2, P<0.025). Figure 1 shows the distinct prefer-
ence of this species for the high Juncus marsh. Preference for high
marsh habitats was also reported by Byrne (1978) and Kneib (1978).
Avoidance of strong tidal currents and predators of the low marsh and
tidal creek habitats may account for this small species' preference for
high marsh areas (Byrne 1978).

We concur with Byrne (1978) and Kneib (1978) that the purported
rarity of F. luciae is due mainly to inadequate sampling of its preferred
high marsh habitat. As demonstrated by Kneib and this study, the use
of pit traps appears to be an effective method of collecting F. luciae,
which tends to remain on the marsh at low tide. We beUeve the species
is more common and widespread than is indicated by the few published
reports based on the results of seining and trawling.

Fundulus luciae on N.C. Salt Marsh

143

Dhabitat availability

o
u

0.

Fig. 1 . Availability and use by Fundulus luciae of salt marsh habitats on Brad-
ley Creek Marsh, North Carolina, October-November, 1983. TS = tall-form
Spartina alterniflora, SS - short-form S. alterniflora, JU = Juncus roemerianus.

ACKNOWLEDGMENTS.— This study was prompted by discus-
sions with David G. Lindquist. Robin D. Bjork and Jennifer E. Slack
provided helpful field assistance. Courtney T. Hackney and Paul E.
Hosier constructively criticized the manuscript.

LITERATURE CITED
Butner, Alfred, and B. H. Brattstrom. 1960. Local movement in Meni-

dia and Fundulus. Copeia 1960 (2): 139-141.
Byrne, Donald M. 1978. Life history of the spotfin killifish, Fundulus

luciae (PiscesiCyprinodontidae), in Fox Creek Marsh, Virginia.

Estuaries 7(4):21 1-227.
Hildebrand, Samuel F. 1941. An annotated list of salt and brackish

water fishes, with a new name for menhaden, found in North

Carolina since the publication of "The fishes of North Carolina" by

Hugh M. Smith in 1907. Copeia 1941(4):220-232.

144 Mark A. Shields and Carol H. Mayes

Jorgenson, Sherrell C. 1969. A Georgia record for the cyprinodontid

fish, Fundulus luciae. Chesapeake Sci. 70(1): 65.
Kneib, R. T. 1978. Habitat, diet, reproduction and growth of the spotfin

killifish, Fundulus luciae, from a North Carolina salt marsh. Copeia

1978(1):164-168.
Lee, David S., C. R. Gilbert, C. H. Hocutt, R. E. Jenkins, D. E.

McAllister and J. R. Stauffer, Jr. 1980. Atlas of North American

Freshwater Fishes. N. C. State Mus. Nat. Hist., Raleigh. 854 pp.

Accepted 12 April 1984

145

SUBSCRIPTIONS AND EXCHANGES

The editors anticipate two issues of approximately 150 pages each annually.
Rates for subscriptions for all issues appearing within the calendar year:

Individual— United States $ 9.00

Individual — Foreign $ 1 3.00

Institution $15.00

All subscriptions must be paid in advance.

Issues will be available on an exchange basis to organizations and institu-
tions publishing general natural history and ecology journals or papers on a
fairly regular schedule. Publications received on exchange will be placed in the
State Museum's H. H. Brimley Memorial Library.

Address all subscriptions and requests for information on purchase and
exchange to Managing Editor, Brimleyana, N. C. State Museum of Natural
History, P. O. Box 27647, Raleigh, NC 2761 1. Back issues are available.

DATE OF MAILING

Brimleyana No. 8 was mailed on 12 April 1983.

PUBLICATION SCHEDULE

We apologize to all subscribers and contributors who have been inconven-
ienced by our publishing delay. Brimleyana Nos. 10 and 11 are being processed
simultaneously and should appear at about the same time this autumn (No. 10 is
devoted to various aspects of the biology of the salamander, Necturus lewisi).
We are trying to assure, also, that No. 12 will appear within a reasonable pro-
duction time.

FORTHCOMING . . .

* Fishes, of Buck Creek, Kentucky;

* An update of Sympotthastia (Diptera);

* Fossil bats from Florida;

* Seasonal weight changes in raccoons;

* A new species of Paracricotopus (Diptera);

* Aspects of the biology of Etheostoma rufilineatum (Pisces);

* Aquatic distributional patterns in the Interior Low Plateaus;

* Seabird core temperatures;

* Genetic variation in Agkistrodon piscivorus (Reptilia);

* Virginia in Virginia;

* The rete in seabirds;
. . . AND OTHERS.

STATE LIBRARY OF NORTH CAROLINA

3 3091 00748 4710

INFORMATION FOR CONTRIBUTORS

Submit original and two copies of manuscripts to Editor, Brimleyana, North Carolina State
Museum of Natural History, P. O. Box 27647, Raleigh, NC 2761 1. In the case of multiple
authorship, indicate correspondent. Manuscripts submitted for publication in this journal
should not also be submitted elsewhere.

Preparation of manuscript. Adhere generally to the Council of Biology Editors Style
Manual, Fourth Edition. Use medium-weight bond paper, SVi X 11", and leave at least an
inch margin on all sides. Double space all typewritten material.

The first page will be separate and contain the title and the author's name and address.
Where appropriate, the title will indicate at least two higher categories to which taxa
belong. Example: Studies of the genus Hobbseus Fitzpatrick and Payne (Decapoda: Cam-
baridae).

A brief informative abstract on a separate sheet follows the title page, preceding the text. In-
dicative abstracts are not acceptable. Footnotes will be used only where absolutely
necessary, numbered consecutively throughout the paper.

Individuality of writing style and text organization are encouraged, but for longer papers
the INTRODUCTION, MATERIALS AND METHODS, RESULTS, DISCUSSION
and LITERATURE CITED format is preferable, with those headings centered and
capitalized. Headings plus sub-headings must be kept to a total of three levels.

Scientific names in taxonomic papers will include the author in first usage. Descriptions of
new taxa must be in accordance with the requirements of established international codes.
Etymology is desirable.

Last item in the text will be acknowledgments, with the body of the section preceded
thusly: ACKNOWLEDGMENTS. — Authors should verify that persons mentioned in
acknowledgments acquiesce in the wording.

Appendixes: place after acknowledgments and before literature cited.

Form for literature cited: Author's last name, first name, middle initial. Year. Title. Jour-
nal (see BIOSIS list of Serials with Title Abbreviations) volume (number) :pages. Provide
total number of pages for books, dissertations, and theses. For second authors use initials
followed by last name. Examples:
Woodall, W. Robert, Jr., and J. B. Wallace. 1972. The benthic fauna in four small southern

Appalachian streams. Am. Midi. Nat. ^<5(2): 393-407.
Crocker, Denton W. and D. W. Barr. 1968. Handbook of the Crayfishes of Ontario. Univ.
Ontario Press, Toronto. 158 pp.

Authors, not the editor, are responsible for verifying references.

Form for citing references in text: parenthetical (Woodall and Wallace 1972:401), page
numbers optional, following a colon; for more than two authors use et al. (not italicized).

All tables go on separate sheets at the end of the manuscript. Do not use vertical lines in
tables. Indicate lightly in pencil in the margin of the original manuscript where tables and il-
lustrations would best fit.

Preparation of illustrations. Illustrations, including maps, graphs, charts, drawings,
and photographs, should be numbered consecutively as figures. They should not be larger
than 21.5 X 28 cm (8V2 X 11"). Plates must be prepared and presented as they are to ap-
pear, not as groups or large sheets of items for arrangement by the editors. Do not mount in-
dividual photographs. The author's name, title of the manuscript, figure number, and the
notation "Top," should be penciled lightly on the back of every illustration. Lettering on
original drawings and maps should be of adequate size to permit proper reduction where
needed. Do not type on illustrations. Legends should be typed, double-spaced, on separate
sheets. Avoid indicating scale as "X life size." Consult CBE Style Manual, pp. 39-45, for more
complete guidelines.

Page charges, reprints and proofs. A per page charge of $20 is expected from authors
who have funds available from institutions, grants, or other sources. Those without such
funds should so indicate in their correspondence with the Editor. This will not affect accept-
ance for normal publication. Contributors who pay full page costs will be furnished 100 free
reprints. Reprint order forms will be sent with galley proofs and are to be returned to the
Managing Editor. On papers with more than one author, it will be the responsibility of the
correspondent to assure that other authors have an opportunity to obtain reprints. Proofs
are to be corrected, signed and returned to the Managing Editor within 48 hours.
Changes in proofs other than type corrections will be charged to the author.

CONTENTS

A New Species of Woodland Salamander of the Plethodon glutinosus
Group from the Southern Appalachian Mountains. Richard
Highton 1

Lower Wilson Creek, Caldwell County, North CaroUna: A Thermal
Refugium for Reptiles? Robert Wayne Van Devender and Paul F.
Nicoletto 21

Comparative Food Studies of Yellowfin Tuna, Thunnus albacares, and
Blackfin Tuna, Thunnus atlanticus (Pisces: Scombridae) from the
Southeastern and Gulf Coasts of the United States. Charles S.
Manooch III and Diane L. Mason 33

Benthic Macroinvertebrates of Cane Creek, North Carolina, and
Comparisions with Other Southeastern Streams. David R.
Lenat 53

Pleistocene Mammals from the Rock Springs Local Fauna, Central

Florida. Kenneth T. Wilkins 69

Bird Density and Habitat Use in Forest Openings Created by Herbi-
cides and Clearcutting in the Central Appalachians. William C.
McComb and Robert L. Rumsey 83

Drainage Records and Conservation Status Evaluations for Thirteen
Kentucky Fishes. Melvin L. Warren, Jr. and Ronald R.
Cicerello 97

Winter Food Habits of Bobcats in North Carolina. Anne M. King,
Richard A. Lancia, S. Douglas Miller, David K Woodward and
Jay D. Hair Ill

Foods and Feeding Behavior of Sauger, Stizostedion canadense (Smith)
(Pisces: Percidae), from Gallipolis Locks and Dam, Ohio River.
Steven I. McBride and Donald Tarter 123

Notes on Breeding Period, Incubation Period, and Egg Masses of
Ambystoma jeffersonianum (Green) (Amphibia: Caudata) from the
Southern Limits of its Range. Charles K Smith 135

Occurrence and Habitat Preference of Fundulus luciae (Baird) (Pisces:
Cyprinodontidae) on a Southeastern North Carolina Salt Marsh.
Mark A. Shields and Carol H. Mayes 141

Miscellany 145