Ml

1.3-

^nh Caro'^ State Library

Raleigh

' X !QQ/

*f£>

gcaSS w

number 11 October 1985

EDITORIAL STAFF

John E. Cooper, Editor

Eloise F. Potter, 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

Carolina, 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 0193-4406

The Mammal Fauna of Carolina Bays, Pocosins,

and Associated Communities in North Carolina:

An Overview

Mary K. Clark, David S. Lee, and John B. Funderburg, Jr.

North Carolina State Museum of Natural History,

P.O. Box 27647, Raleigh, North Carolina 27611

ABSTRACT. — This study represents the first attempt to inventory

and evaluate the mammals associated with pocosins and Carolina

bays. During a 4-year period, approximately 17,000 trap-nights and

200 field-days in 12 North Carolina habitat types produced specimens

or signs of 40 species of mammals. Early, intermediate, and advanced

serai stages of pocosin-associated plant communities varied considera-

bly in faunal composition. Species regularly trapped or observed

included Blarina sp., Pipistrellus subflavus, Sylvilagus palustris, Sciu-

rus carolinensis, Peromyscus gossypinus, Ochrotomys nuttalli, Uro-

cyon cinereoargenteus, Procyon lotor, and Odocoileus virginianus.

Additional uncommon or geographically restricted, but apparently

regular, associates were Condylura cristata, Plecotus rafinesquii,

Oryzomys palustris, Microtus pennsylvanicus, Synaptomys cooperi,

and Ursus americanus. Most mammal associates are ubiquitous spe-

cies. Although total documented diversity is high, a large percentage of

the fauna is either associated with edges of communities or of irregular

occurrence. At least eight species and several additional subspecies

reach the northern or southern limits of their ranges in pocosin-rich

areas.

Fires, storms, and certain man-related disturbances, by creating a

patchy mosaic of habitats, seem to exert positive influences on mam-

mal density and diversity in pocosin communities. Since uninterrupted

or unaltered successional development eventually leads to minimal

habitat diversity, management of extensive pocosin areas is desirable if

mammal diversity is to be maintained.

INTRODUCTION

Recent authors have commented on the almost complete lack of

information on the vertebrates associated with Carolina bays, pocosins

and successionally related southeastern Coastal Plain habitats (e.g. Wil-

bur 1981; Sharitz and Gibbons 1982). Although some mammal surveys

have been conducted within or adjacent to pocosin habitats, published

reports have addressed taxonomic status or geographic, not ecological,

distribution, making it difficult to relate most existing information to

specific plant communities. The most notable of these studies have been

Brimleyana No. 1 1:1-38, October 1985

2 Mary K. Clark, David S. Lee, John B. Funderburg, Jr.

various investigations of the Dismal Swamp fauna of northeastern

North Carolina and southeastern Virginia (e.g. Merriam 1895a,b, 1896;

Handley 1979). Rose (1981a) investigated mammals associated with

"openings" in the Dismal Swamp. Sharitz and Gibbons (1982) presented

some preliminary information on studies they are conducting in South

Carolina bays, but only three mammal species — Blarina brevicauda,

Reithrodontomys humulis and Sigmodon hispidus — were mentioned.

Lee et al. (1982) provided preliminary mammal species lists for the var-

ious plant communities in North Carolina, including information on

pocosins and Carolina bays. With these exceptions, information on the

mammal fauna of these specific systems was previously unavailable.

From 1980 to 1985 we surveyed the mammals of pocosins, Carolina

bays, and their associated plant communities in the North Carolina

Coastal Plain in order to obtain a better understanding of diversity and

relative density of the mammal fauna.

Considering the widespread geographic distribution of pocosins

and Carolina bays it is quite surprising that their vertebrate fauna is so

poorly known. There are estimates of over 55,000 Carolina bays between

southern Maryland and Florida (Sharitz and Gibbons 1982). Wells

(1946) estimated pocosin habitats to have originally occupied over 20

percent of the Coastal Plain of North Carolina alone, and noted that

there were over 300 square miles of pocosin in just three southeastern

North Carolina counties. Since that time a good percentage of these

areas have been drained, partially drained, and cleared for agricultural or

silvicultural purposes, and some areas have been dammed to create mill

ponds. Other areas have been protected from fire for so long that the

plant communities have progressed beyond pocosin serai stages.

There is some uncertainty about the extent of loss of such habitats

and the need for concern. Heath (1975) and Richardson (1981) provided

general summaries of the decline of these wetlands, and most subse-

quent studies relied on these sources as the basis for major concern for

pocosin habitats. Originally, Richardson (1981) stated that only 31 per-

cent of North Carolina pocosins remained in a natural state, but

Richardson (1983) acknowledged that his data sources were in error.

McMullan (1984) suggested that the reasons for concern may be less

serious than previously stated, owing to faulty data sources and incom-

plete or nonexistent inventories. McMullan (1984) also demonstrated,

through an analysis of a 300-year historical land use study of the

Albemarle-Pamlico peninsula of North Carolina, that pocosin commun-

ities have persisted in spite of a long history of clearing and draining,

and many present-day pocosins have developed (or redeveloped) on

abandoned farm lands. Assuming that the more recent reports are cor-

rect, it appears that the original estimates of habitat loss were too high.

Mammals of Carolina Bays 3

Furthermore, because little information on the vertebrate fauna asso-

ciated with pocosins is available, definitive statements made by previous

authors concerning wildlife values of pocosins were premature. Cur-

rently, discussions about the unique biological value of pocosins and

Carolina bays on the one hand, and consideration for their use in agri-

business, silviculture, peat mining, and waste disposal on the other, are

commonplace, but in most cases detailed information on which to base

management decisions is lacking.

Although the information presented here pertains only to North

Carolina, we suspect that our findings could apply generally to other

pocosins and Carolina bays in the southeast. However, we have little

experience with these communities outside North Carolina. Our efforts

to date have been focused on making species inventories of a large

number of different communities throughout the North Carolina

Coastal Plain. While we consider our results more than preliminary, prob-

lems associated with sampling the wide array of Carolina bays and

pocosin communities make it impractical at this time to compare rela-

tive abundance and density of species in specific habitat types based on

cumulative trap-night success. Additional studies are planned to develop

more elaborate population profiles for specific pocosin plant

communities.

HABITATS STUDIED

Pocosin habitats are defined with difficulty, since considerable con-

fusion persists in the use of the term. It originated from the Algonquin

Indian word "poquosin" and is one of the few Algonquin words

adopted by European settlers. Tooker (1899) provided a detailed discus-

sion of the origin, meaning and use of the term. In tracing its early use,

by both Indians and early settlers, Tooker found that "pocosin" referred

to a wide variety of low, wet areas extending from New England through

the Carolinas. Among European settlers, the term was locally inter-

changeable with "dismals" and "galls" for describing swampy thickets.

Botanists and ecologists have likewise used the word to describe a

variety of low, wooded, wetland habitats, and in many instances the

terms bay, bayhead, shrub bog, or evergreen shrub bog have been used

to describe pocosin vegetation types. The term "bay" is particularly con-

fusing because it refers to a number of successional stages of Southeast-

ern wetlands that support several species of bay trees (Sweet Bay, Mag-

nolia virginiana; Red flay, Persea borbonia; and Loblolly Bay, Gordonia

lasianthus), while the term Carolina bay, partly named for the presence

of bay trees, refers to elliptical depressions that often support pocosin

vegetation. Carolina bays are permanent geological features and often

are specifically named sites (e.g. Wolf Bay, Bladen County), while the

4 Mary K. Clark, David S. Lee, John B. Funderburg, Jr.

bay forests are successional stages of wetlands. Strict definition and

delimitation are further hampered by the fact that many pocosins are

situated within extensive palustrine systems and /or border estaurine

systems. Such mixed areas often provide a rich mosaic of wetland habi-

tats involving broad zones of transition and complex successional pat-

terns. Extensive areas called pocosins are, in fact, often composed of

swamp forest, hardwood forest, and marshes. There seems to be no pre-

cise botanical definition of pocosin, but the tongue-in-cheek description,

"any low, wet area so thick you can't walk through it", captures well the

nature of a pocosin.

Pocosins

Wells (1946) provided a general botanical analysis of pocosins in

Holly Shelter, Pender County, North Carolina, and Kologiski (1977)

investigated the vegetative communities of the Green Swamp, including

several types of pocosin, savanna and related successional communities.

Buell and Cain (1943) described the successional role and ecological

requirements of Atlantic White Cedar, Chamaecyparis thyoides, forests

in southeastern North Carolina. White cedar forests and savannas are

both closely allied with pocosins. Additionally, Wells (1932), Woodwell

(1956), and Sharitz and Gibbons (1983) provided overviews of pocosin

vegetation, and Wells and Whitford (1976) presented a good outline of

the successional development and fate of stream-head swamp forests,

pocosins, and savanna communities.

Carolina Bays

Carolina bays vary in size from only a few to many hundreds of

hectares, and an exposed sand rim of varying width normally occurs

around a bay's perimeter. These depressions are naturally wetter at all

seasons than are most surrounding areas, contrasting markedly with the

dry sand rims, which support xeric plant communities. Most Carolina

bays house pocosin communities in various serai stages, but some also

contain sizable lakes, ponds, marshes, bogs, and swamps. In many bays,

natural fire has been suppressed so long that the plant communities in

them are now mature deciduous bay forests. The elliptical shape and the

tendency for the deepest portion of the depression to be southeast of

center often causes concentric vegetative zonation rings in the interior of

the bays as well as an ecotonal ring around the perimeter. This type of

vegetative zonation occasionally allows for considerable faunal diver-

sity, even in small areas.

A vegetative profile of one Carolina bay near Jerome, Bladen

County, was provided by Buell (1946a,b). We found, however, that by

1983 the area had been drained and lumbered so extensively that this

bay no longer resembles Buell's description. This is unfortunate since it

Mammals of Carolina Bays 5

was the only North Carolina bay where animal communities could have

been related to a published vegetational analysis.

Unlike most other pocosin sites, Carolina bays are often located

within xeric and mesophytic systems. Their islandlike nature often

makes them more visually delineated and ecologically discrete. The wide

spectrum of successional stages, combined with their close proximity to

each other, makes them excellent study sites. In one of our principal

study areas in Bladen County, bays are highly concentrated and succes-

sional stages probably achieve their greatest diversity.

Community Development and Structure

Various environmental factors dictate the type of pocosin commun-

ity that develops on a particular site. The most conspicuous factors are

surface and subsurface soil types, hydroperiod, and fire. The importance

of the regularity and intensity of fire as it relates to season, hydroperiod,

wind, and the accumulation of combustible vegetation cannot be over-

stated. Natural fires, and those started by Indians for game exploitation

and later by Europeans for livestock range management, were all impor-

tant for long-term maintenance of various serai stages of pocosins. Fire

exclusion policies of the middle part of this century were detrimental to

certain communities (particularly savannas), but recent understandings

of the importance of regular controlled burning in certain Southeastern

vegetation types for game and habitat enhancement and for wildfire

control has, in part, alleviated this problem.

The characteristic and conspicuous plants of pocosins and Carolina

bays are comparatively few. In most instances each species occurs in a

majority of the vegetative community types and only its relative abun-

dance or growth form changes. These variations in relative composition,

however, may be dramatic, both visually and ecologically. The major

plant associates (alphabetically by genus) are Red Maple, Acer rubrum;

Wire Grass, Aristida strict a\ Atlantic White Cedar, Chamaecyparis

thyoides; Titi, Cyrilla racemiflora; Loblolly Bay, Gordonia lasianthus;

Sweet Gallberry, Ilex coriacea; Inkberry, Ilex glabra', Fetterbush, Lyonia

lucida; Sweet Bay, Magnolia virginiana; Black Gum, Nyssa sylvatica\

Red Bay, Persea borbonia; Pond Pine, Pinus serotina; Bamboo, Smilax

laurifolia; Pond Cypress, Taxodium ascendens; and Honey-cup, Zeno-

bia pulverulent a. Species less uniformly distributed include Lamb-kill,

Kalmia Carolina; gooseberries, Vaccinium sp.; rushes, Juncus sp.; sedges,

Carex sp.; Loblolly Pine, Pinus taeda; Longleaf Pine, Pinus palustris;

and Cane, Arundinaria gigantea. These latter species are, however,

often the dominant vegetation on certain sites.

Major pocosin community types include shrub bogs with scattered

Pond Pine overstory, mixed conifer-hardwood shrub bogs, and pine-

6 Mary K. Clark, David S. Lee, John B. Funderburg, Jr.

shrub savannas. Early successional stages of all these types appear to be

grass-sedge communities and later ones mature conifer-hardwood and

evergreen bay forests. With long-term absence of fire, all eventually

become deciduous bay forests. In these later stages cypress and Black

Gum emerge on the sites with protracted hydroperiods, and Sweet Gum,

Liquidambar styraciflua, and pines grow on drier ones. Thus, pocosins

in general can be viewed as intermediate successional communities,

often maintained in a subclimax stage by fire and hydroperiod, with the

mature vegetational stages being suppressed for long periods on the wet-

test sites but developing relatively quickly on drier ones. Figure 1

depicts our perception of a general successional model of pocosin com-

munities. Figure 2 shows various examples of the communities discussed.

Development of white cedar forests is unusual in that this species

needs fire or other disturbance to remove vegetation so seedlings can

develop. However, extremely hot fires destroy the peat soil and dormant

seeds, and white cedar forests do not appear. Conversely, low intensity,

fast moving fires do not destroy enough of the root stocks of competi-

tive shrubs for cedar to become well established. When established,

white cedar is extremely fire susceptible and persists only in the absence

of fire. Young white cedar forests are usually pure, nearly even-age

stands, and the density of such forests often inhibits the establishment

of other tree species for about 40 years. After that time the trees begin

to thin out and the nature of their crowns changes, which permits light

to penetrate to the forest floor. At this stage bay forests develop rapidly,

although individual white cedar trees may persist for long periods. The

open savanna community requires a periodic disturbance by fire. If fire

is suppressed for several consecutive years, many characteristic savanna

plants vanish. (The above analysis is summarized from Wells 1946;

Buell and Cain 1943; Kologiski 1977; Wells and Whitford 1976; and our

personal observations.)

We have included the sand rims associated with Carolina bays and

Coastal Plain stream-head forests in our discussion of animal distribu-

tions. In both sand rim and stream-head forest communities, fire plays

an important role in maintaining community structure. The stream-head

forest communities were considered by Wells and Whitford (1976) to be

vegetatively similar to certain pocosin communities and we have also

found them to be similar faunistically. Plant communities of the sand

rims, while in direct contrast to pocosin vegetation types, are a charac-

teristic vegetational feature of Carolina bays and are included in our

discussion. These sand rims are dominated by Longleaf Pine; Turkey Oak,

Quercus laevis; and Wire Grass.

Mammals of Carolina Bays

□

POCOSIN COMMUNITIES

Comparatively Short

recent

fire or

disturbance

infrequent

fire

suppressed

fire

Fig. 1. Suggested pattern of vegetation development of pocosins and associated

communities as related to disturbance, time, and hydroperiod (from Lee, in

manuscript. Breeding bird communities in pocosins.)

METHODS

Almost 17,000 trap-nights were involved in this study. (Trap-nights

equal the number of baited traps multiplied by the number of nights the

traps are in place). Trap-night success is the percentage of catch per

trap-night effort. Most trapping was done with Museum Specials baited

with peanut butter, but Sherman live traps (2 sizes), Conibear, leg hold,

Have-a-Hart type live traps, and mole traps were also used. Traps were

8 Mary K. Clark, David S. Lee, John B. Funderburg, Jr.

Mammals of Carolina Bays

Fig. 2. Habitats discussed in text. A. Aerial view of Carolina bays in Bladen

County; note development of sand rim. B. Clay-based Carolina bay in Hoke

County, vehicle parked on sand rim. C. White cedar forest in Green Swamp,

Brunswick County; shrub bog in foreground. Note natural opening in mature

forest, caused by storm damage. D. Pine-shrub bog, Bladen County. E. Juncus

marsh developing in roadside ditch next to deciduous bay forest, Dare County.

F. Pine-Wire Grass savanna, Carteret County. G. Longleaf Pine-Turkey Oak sand

rim, Hoke County. H. Stream-head forest, Hoke County.

10 Mary K. Clark, David S. Lee, John B. Funderburg, Jr.

selectively placed within particular plant communities rather than dis-

tributed in grid patterns. Pitfall traps at two study sites, with and with-

out drift fences, were left in place for extended periods and, although

not included in our trap tallies, represent thousands of additional trap-

nights. In addition, museum records, interviews with local residents,

trappers and others familiar with the areas studied, personal sight

records, and examinations of tracks and other sign, were used in compil-

ing the faunal list presented here. Additional random information

obtained from over 200 field-days and 25 evenings of mist netting and

shooting bats is also included.

All North Carolina pocosins and Carolina bays that had been

extensively studied and described in publications were visited to ensure

that the community terminology used here was in general agreement

with that of past studies. The difficulty of trapping small mammals in

the Coastal Plain in general, of trapping in pocosins in particular, and

unequal field effort in various community stages or areas, make it

unlikely that our mammal lists for each plant community are definitive

(bat information is particularly scarce). Because many of the areas

studied are in transition from one community type to another, and many

of our records are from ecotonal areas and disturbed or altered sites,

assessing the species composition of specific communities is difficult.

The mammal fauna of Carolina bays, pocosins, and associated

communities was studied in parts of Bladen, Brunswick, Carteret, Curri-

tuck, Dare, Hoke, Moore, Pasquotank, and Pender counties, North

Carolina, between October 1979 and April 1984 (Fig. 3). Sites were not

inventoried with equal field effort. The following sites were studied

(total trap-nights per county are in parentheses): Bladen Co. (4,009) —

Bay Tree Lake (Black Lake), Jones Lake, Little Singletary Lake, Salters

Lake, Singletary Lake, Suggs Mill Pond (Horseshoe Lake), White Lake,

one unnamed bay 3.2 km east of Kelly on NC 53, and another 17.8 km

east of Kelly on NC 53. Brunswick Co. (266) — Green Swamp, 17.8 km

north of Supply on NC 211; Sunny Point area. Carteret Co. (335)—

Croatan National Forest, 4.8 km east of Newport. Columbus Co.

(219) — Lake Waccamaw (town), Lake Waccamaw State Park. Curri-

tuck Co. (2,123) — Coinjock area and northward. Dare Co. (3,675) —

mainland between US 64 and US 264. Hoke Co. (4,400+)— North Caro-

lina Biological Survey Study Site at McCain. Moore Co. (200) — Wey-

mouth Woods State Park. Pasquotank Co. (1,261)— Dismal Swamp,

"Big Ditch." Pender Co. (0) — Holly Shelter Game Management area.

The Currituck and Pasquotank counties data are from the eastern and

southern edge of the Dismal Swamp, but include no true pocosin habi-

tats. Nevertheless, the comparative geographic and abundance informa-

tion obtained from these sites is informative. Specific specimen records

Mammals of Carolina Bays

2

11

Fig. 3. Pocosin communities and Carolina bays in North Carolina (darkened

areas; modified from Richardson 1981). Numbers correspond with study sites.

Dismal Swamp area: (1) Currituck Co., Coinjock; (2) Pasquotank Co., "Big

Ditch." Pocosins and savannas: (3) Dare Co. mainland, near East Lake; (4)

Carteret Co., Croatan National Forest, 4.8 km east of Newport; (5) Pender Co.,

Holly Shelter Game Management Area; (6) Brunswick Co., Green Swamp, 17.8

km north of Supply. Carolina bays and sand rims: (7) Columbus Co., Lake

Waccamaw; (8) Bladen Co., Bladen Lakes; (9) Hoke Co., N.C. Biological Sur-

vey Study Site, McCain. Stream-head forest: (9) Hoke Co., Biol. Survey Site,

McCain; (10) Moore Co., Weymouth Woods State Park.

of only the more unusual species are cited in the following accounts.

Plant names are from Radford et al. (1964).

RESULTS

Forty-one species of mammals were found in or adjacent to poco-

sins and Carolina bays. Fauna was composed of 1 species of marsupial, 5

insectivores, 8 bats, 2 rabbits, 15 rodents, 8 carnivores, and 2 hoofed

mammals. Only 2 of these 41 species are exotics, and interestingly

neither Myocaster nor Rattus was encountered. Two additional species,

recently extirpated, are known to have been inhabitats of pocosin com-

munities. Documented occurrence of extant species and trap success is

12

Mary K. Clark, David S. Lee, John B. Funderburg, Jr.

s 3

x xxx XXX x x x x x x XXX

XX XX

X X X X X X

X X X X X

XX XX X

X X

X X

<3

5 £

•a k

^

III

05

K

1 1

g S I 4 S3 £

1 * 5 ! 1*

" 2 .3 8 -S

53 ft, *? ^

3 5

5"^

S-3 §

QcoQqCcoC^S;'^^

Mammals of Carolina Bays

13

xxxxxxxxxxx XXX x *£^p?

m © —

en on

x xx x «o o

X X X X

X XX

x x m r» o

m m oo

2 <*

X CO «A> »/">

— CO rfr

xxxxx **r:£j:5;

rs © r-

xxxxx * £; 2£ ;=:

r) vi o

£^4

X «n v> *o

X X © © — <

CN O Tf

xxx **£;:;£

(S (S >D

© — ;

X o o

xxx

x a ^ oo

m o

X <N <N m

u

00

W

.§ I S »

co O W Q

o i-5 cs rn

CM Ui

£ ffl

U 8

to

pa T3

O T

II

at u

oo S

5

S

-5!

s

o

s

■8

3

3

3.1

&a

I 3

b -Si

c 5 ^

8, 8*8

I s «

1«« M ^-

rv •» *»

II

O 3

8 3

CO

3 «

O »*. Q

-s. O *> ^

s ^

I*

P 3

^

^^^^5o

,3 3 **" on <u

3> tS •« C "

^ w Q co

\0 t> 00 On

00

O

pq

X>

Wh

J3

CO

t3

O

60 ©

O £

pa -o

CO

on (U <o "F

> a c §

g O co £ £ U

cd

K - ri w Tt wi

.«

C3

14

Mary K. Clark, David S. Lee, John B. Funderburg, Jr.

° £

i: <3

^ c

U 3

o

1 *

o §

O X

X X X X X

XXX

XXX

X X X X X X

XXXXXX XXXX X xxxxxx

XXX

X X X X X

XX XXX X

X X

X X

XX X XX

XXX XX XXX xxxxxxxxxx

3 5

I

II

•a 5

Si 3

O 3

i s

*5 «0

^S k' C O Q •=" ft, 5j . .J^-3">s.'S . S .« Jfe

sal I

* ^ ^ 1 5

*1 k 15 C B

^ ^> "^ ^ :q

J 3 §

5i

Mammals of Carolina Bays

15

XXX

X X XX XXX X X

X X vo o

*3S

X X X X X X

X X X X X X

X rn o

O

X X o ©

«0 ©

XXXXXXXXXXO«r>

vO

m

XXX

X X X X

X X

X oo co

SO

X r^ Os

X x <n «n

m

CO

X X X X X

so SO

vO

xxxxxxxxx«^o

§

3 .5

3

3 "EL £

c .5? 5

? Q S9

2 •§ 5

o o ©

O co ^

si

■B.4I

3

R

log

S-&

pa

>3*

* 3 Ifl -a

.3 3

K -P

3 ^5 ^

5. &

S^Si^45o

.X

C

B

<u O w-

u 3 «

o o"2

O 55 G

2 Oh Oh

. . O

oo on —

.x

u

3 eG O

U Q K

o 3

s S i

d «J o

- (N m ^

16 Mary K. Clark, David S. Lee, John B. Funderburg, Jr.

presented for 12 habitats in Table 1 and for the 10 Coastal Plain coun-

ties surveyed in Table 2. Typically, only 10 mammal species were

encountered on a regular basis, and only 8 species were found in 50

percent or more of the habitats studied. However, 20 species were

recorded in 50 percent or more of the counties surveyed, suggesting that

habitat was far more locally restrictive to distribution than was geo-

graphy. Disturbed habitats with early successional communities yielded

the greatest diversity and density.

During this study only 366 mammal specimens were trapped in

9,472 trap-nights, but many additional animals were obtained by other

means, and approximately 450 specimens were taken from borders of

pocosin communities (7,000+ trap-nights). A composite trap yield suc-

cess for all pocosin habitats sampled was 5.13 percent; trap success was

generally higher for surrounding communities. Voucher specimens and

series of the common species collected at each study site are deposited in

the mammal collection of the North Carolina State Museum (NCSM).

SPECIES ACCOUNTS

Marsupialia: Opossums

Didelphis virginiana virginiana Kerr, Virginia Opossum. Although

common throughout a wide range of Coastal Plain habitats and abun-

dant in certain parts of Bladen County, the opossum is not usually asso-

ciated with pocosins or Carolina bays and individuals were rarely

observed in or around these habitats. Only one specimen was taken

from the Dare County mainland and over a four-year period few road-

killed individuals were seen there. This species is only slightly more

common on the sand rims of bays.

Insectivora: Shrews and Moles

Sorex longirostris ssp., Southeastern Shrew. This shrew is uncommon

but is found in a wide variety of habitats, including sand ridges adjacent

to stream-head forests with dense ground cover of Aristida, shrubby

ecotones of stream-head forests, /wwci/s-dominated clearings, and ever-

green and deciduous bay forests. In Bladen, Dare and Hoke counties

the subspecies represented is Sorex I. longirostris Bachman. Pagels et al.

(1982) noted that their records for this race were evenly divided between

open fields and young forests where ground cover is heavy. One speci-

men of Sorex I. fisher i Merriam (NCSM 2723) was collected by us in a

swamp forest in Currituck County. Rose (1981a) collected individuals of

fisheri in openings dominated by herbaceous vegetation in the Dismal

Swamp. Both races are apparently absent from typical pocosin (shrub

bog) communities.

Blarina sp., Short-tailed shrews. The systematics of Blarina in the

Mammals of Carolina Bays 17

southeastern Coastal Plain of North Carolina is in need of study. Even

though the area is within the range of B. carolinensis, specimens we

have obtained from Carolina bays, principally in Bladen County, are of

a large form closely approaching B. brevicauda in size and appearance. In

upland areas of Bladen County, however, we found only B. carolinensis.

French (1981) reported large specimens of Blarina from Sampson

and Columbus counties and also remarked on the need for additional

work on this genus in the North Carolina Coastal Plain. The large

forms we collected were found in mature evergreen bay forests. Blarina

brevicauda telmalestes occurs in pocosins and associated communities in

northeastern North Carolina, and for many years it was known only

from the Dismal Swamp region. Paul (1965), however, reported the

subspecies in Hyde County, and we found it at several sites in Dare,

Currituck, and Pasquotank counties. We have also found B. b. telma-

lestes as well as a large Blarina (presumably also B. b. telmalestes) in a

wide range of successional communities in southeastern North Carolina,

but most of our records are from wet forest floors.

Cryptotis parva parva (Say), Least Shrew. The Least Shrew is a

common and characteristic species of Longleaf Pine-Turkey Oak-Wire

Grass associations, sand rims of Carolina bays, Wire Grass savannas, and

early successional communities with open canopies and dense ground

cover. This shrew is not expected to occur in typical pocosin vegetative

stages, although it frequently was collected in pitfall traps at several of

the sand rim study areas.

Scalopus aquaticus howelli (Jackson), Eastern Mole. The Eastern

Mole is characteristic of sand rims of Carolina bays, but is uncommon

or absent from lower and wetter portions of Carolina bays, pocosins,

and savanna habitats. Eastern Moles commonly invade partly drained,

disturbed areas and may range into damper soils for short distances, but

in wetter systems they are probably replaced by Condylura. The Eastern

Mole was not found in the extensive palustrine system of Dare County.

Condylura cristata parva (Paradiso), Star-nosed Mole. The few

Coastal Plain records for the Star-nosed Mole are probably an artifact

of the difficulty encountered in trapping this species. Normally it is

limited to damp areas around springs, creek bottoms, and bogs. Mole

runs encountered in a stream-head forest in Hoke County and around

the wet margins of Carolina bays in Bladen County were almost cer-

tainly made by this species. Hall (1981) listed Condylura from the Dis-

mal Swamp in Virginia and from Garland near White Lake, a Carolina

bay in Bladen County. We have additional records from the following

Coastal Plain localities: New Hanover Co., Carolina Beach (NCSM

3243); Washington Co., Wenona (NCSM records); Pitt Co., precise

18 Mary K. Clark, David S. Lee, John B. Funderburg, Jr.

locality not identified (NCSM records); Robeson Co., Lumberton

(NCSM records); and Scotland Co., near Laurinburg (NCSM 3037).

Most of these records, if not all, are associated with Carolina bays and

mature bay forest communities.

Chiroptera: Bats

Lasionycteris noctivagans (LeConte), Silver-haired Bat. On the even-

ing of 7 April 1984 we collected 2 (NCSM 4179-80) of 20 or 25 of these

bats while they were foraging over a borrow pit pond in Brunswick

County. The pond was in a savanna and the direction from which the

bats emerged indicated that they were roosting either in the savanna, in

a shrub-savanna-pocosin, or in both. The bats were foraging with Lasiu-

rus cinereus and Nycticeius humeralis, and Lasionycteris became more

abund.ant as darkness approached. Searching above the pond after dark

with high intensity spotlights, however, revealed few bats. Silver-haired

Bats are uncommon spring and fall migrants and winter residents on the

North Carolina Coastal Plain (Lee et al. 1982). The normal period of

occurrences for the species in eastern North Carolina is documented

from 17 November to 3 May (100 years of records from NCSM files).

Pipistrellus subflavus (F. Cuvier), Eastern Pipistrelle. This small bat

is common in low pine-shrub bogs but is expected in most of the other

vegetation types discussed. Approximately 25 pipistrelles were seen feed-

ing at treetop level in Holly Shelter, Pender County, on 25 July 1983; 2

of them (NCSM 4064-5) were collected. A single specimen (NCSM

4181) was collected of three found in 1983 in an abandoned house near

Lake Waccamaw, Columbus County. Three specimens (NCSM 3535-6,

3725) were taken on various dates while they foraged over a pond in

Hoke County. This site, dominated by Turkey Oak-Longleaf Pine habitat,

is adjacent to a small Carolina bay and extensive stream-head forest.

Eptesicus fuscus fuscus (Palisot de Beauvois), Big Brown Bat. This

bat is not particularly common on the outer Coastal Plain of North

Carolina. A single specimen (NCSM 3888) was taken and one other

observed, in an opening in a deciduous bay forest at Buffalo City, Dare

County, on 27 April 1983. Another specimen (NCSM 3846) was col-

lected on 6 July 1982 at McCain. Additionally, we saw a bat that

appeared to be this species foraging adjacent to and over a pine-shrub

bog in Pender County. In Bladen County, the Big Brown Bat was occa-

sionally found associated with Plecotus rafinesquii in abandoned

buildings.

Lasiurus borealis borealis (Muller), Red Bat. Red Bats were regularly

seen feeding at subcanopy height above roads and other openings in all

Mammals of Carolina Bays 19

mature forest types, as well as over water, in all study areas. Single

specimens were collected from the sand rim at Singletary Lake and Sal-

ters Lake, Bladen County; the specimen from Salters Lake was col-

lected while it was foraging on 20 February 1984. Records from Dare

and Hoke counties indicate that this bat is also common above both

xeric and palustrine communities, and we saw individuals feeding over

study sites in Pender County. We have several times observed Red Bats

migrating by day (April) through Pond Pine-shrub bogs.

Lasiurus seminolus (Rhoads), Seminole Bat. We have a specimen of

this species (NCSM 3701) collected over the sand rim adjacent to a

Carolina bay in Hoke County. Bill Adams, U.S. Army Corps of Engi-

neers, has collected many of these bats over large bodies of water in

southeastern North Carolina (NCSM). We therefore find it likely that

Seminole Bats occur regularly over many bay lakes.

Lasiurus cinereus cinereus (Palisot de Beauvois), Hoary Bat. Migrants

and winter residents of this species were seen and collected adjacent to

savannas, pine-shrub bogs, bay forests, and similar areas. We think they

were seeking cleared, open areas for foraging and were not found in

pocosins per se. We have also observed the Hoary Bat over large rivers

in the southeast and suspect that it regularly forages over bay lakes. The

thick vegetation of white cedar and evergreen bay forests provides

potential roost sites, but roosting in these habitats has not been con-

firmed. We saw a Hoary Bat flushed from a hollow stump during a

controlled winter burn of a stream-head forest and sand ridge at Wey-

mouth Woods State Park, Moore County, and several were seen in the

spring of 1984 foraging over a borrow pit pond in a Brunswick County

savanna. Our documentation indicates that dates of occurrence range

from 28 September to 17 April.

Nycticeius humeralis humeralis (Rafinesque), Evening Bat. The Even-

ing Bat was commonly seen flying at canopy height in savannas, and

along the edges of mature bay forests and swamps adjacent to fields. We

were told of a local pest exterminator who gassed and removed "bucket-

fulls" of these bats from a boathouse on White Lake, Bladen County. In

July 1983 we collected one specimen of many seen flying over a savan-

na in Brunswick County. The stomach of this bat, collected during late

dusk, was already filled with fragments of recently consumed insects.

Plecotus rafinesquii macrotis (LeConte), Rafinesque's Big-eared Bat.

We have numerous records of this rare bat from the Bladen Lakes area,

Bladen County, although only four of them were directly associated

with bays. One individual was reported by a local property owner in a

hollow Black Gum cut from the edge of White Lake, another was seen

20 Mary K. Clark, David S. Lee, John B. Funderburg, Jr.

by us in an abandoned hotel on this lake, one (NCSM 4018) was from

an abandoned building at Lake Waccamaw, and another was found on

a building at Singletary Lake State Park (NCSM records). The Dismal

Swamp, where Handley (1959) reported Plecotus collected from hollow

cypress trees in Lake Drummond, is the northernmost locality for this

species on the Atlantic Coastal Plain. We have a specimen (NCSM

3938) from the southeastern edge of the Dismal Swamp, Gates County,

and recent records from Dare County. Unpublished studies by Lee and

Clark show that on the Coastal Plain this bat is restricted to river

swamps and bay lakes bordered by mature swamp forests.

Lagomorpha: Rabbits

Sylvilagus palustris palustris (Bachman), Marsh Rabbit. The Marsh

Rabbit is common in nearly all stages of pocosins, although it is most

abundant in ecotonal areas adjacent to clearings, roads, canals, and lake

edges. At Bay Tree Lake, Bladen County, we found it to be sympatric

with Sylvilagus floridanus in a low Pond Pine-shrub pocosin adjacent to

the lake. Bill Adams (pers. comm.) reported this same situation in a

pocosin in Brunswick County.

Sylvilagus floridanus (Allen), Eastern Cottontail. This rabbit is char-

acteristic of but not common on sand rims, and is rare to absent in most

pocosin and Carolina bay areas. We did not encounter a single rabbit of

this species on the Dare County mainland except around residential

areas. So many other races of this rabbit have been stocked in eastern

North Carolina that subspecific recognition of the original native form,

Sylvilagus/. mallurus, probably has little meaning.

Rodentia: Rodents

Sciurus carolinensis carolinensis (Gmelin), Gray Squirrel. The Gray

Squirrel is. common in many stream-head forests, pocosins with mature

trees, and bay and swamp forests. It is particularly common in areas

dominated by mature Pond Pine where it forages extensively on cones.

It is recorded from most mature habitats, both natural and disturbed,

and is occasionally seen crossing sand rims, but is apparently absent

from savannas. In Dare County we found extensive Gray Squirrel use

of habitats containing mature Pond Pine. These trees retain their seeds

for long periods, and fire is a major triggering mechanism for seed

release. Thus, mature cones are available throughout the year and

represent a major, and perhaps in some areas exclusive, food source.

Sciurus niger niger (Linnaeus), Fox Squirrel. This species is not usu-

ally regarded as a pocosin associate. The sand rims of Carolina bays,

however, provide the open, fire-maintained pine forests that Fox Squir-

rels prefer and here they are often common. They do forage in stream-

Mammals of Carolina Bays 2 1

head forests, and we have seen them regularly taking cover and foraging

along the edges of pocosins. Although savannas would appear to pro-

vide ideal habitat for this squirrel, it occurs in them only when the

savannas are adjacent to upland areas with oaks that provide the mast

on which the squirrel depends. Most savanna records are from the fall.

Glaucomys volans volans (Linnaeus), Southern Flying Squirrel. The

Southern Flying Squirrel is probably present at most sites inhabited by

Gray or Fox squirrels, but we have no records from pocosins. A nest

with young was discovered in the 100-foot fire tower at Jones Lake,

Bladen County, in 1983, and the species is abundant on sand rims at the

Hoke County study site where we regularly found Glaucomys in hollow trees

and bird nest-boxes in ecotonal areas of stream-head bay forests and a

Carolina bay. It is present but apparently not common in savannas, and

often is found in cavities made by the Red-cockaded Woodpecker,

Picoides borealis.

Castor canadensis Kuhl, Beaver. The Beaver was extirpated from

North Carolina in the early 1900s, but was later restocked and is mak-

ing a successful comeback. Castor is not a conspicuous or important

part of the mammal fauna of pocosins and Carolina bays at this time.

Active colonies exist on the Hoke County study area at McCain, in

close proximity to Jerome Bog and Suggs Mill Pond, Bladen County,

and along the southwestern edge of the Dismal Swamp. Flooding and

removal of many larger trees by beavers maintain boggy areas in which

many of the characteristic pocosin shrubs thrive. Both the southeastern

Castor c. carolinensis (from Alabama), the subspecies native to North

Carolina, and the northern form, canadensis, have been stocked on

North Carolina's Coastal Plain.

Oryzomys palustris palustris (Harlan), Rice Rat. The Rice Rat prefers

marshes and other open, wet areas abundant with grasses, rushes, and

sedges, but such habitats do not usually occur in those pocosins or

Carolina bays with intermediate or advanced successional development.

One specimen of Oryzomys was trapped in a clearing at Little Singletary

Lake, a Carolina bay in Bladen County, and another was taken in an

evergreen bay forest in Dare County. Fire and man-made disturbances

create or maintain early successional stages, and in such habitats the

Rice Rat is often abundant.

Reithrodontomys humulis humulis (Audubon and Bachman), Eastern

Harvest Mouse. Typically associated with early successional stages of

pocosin communities, this mouse appeared to be equally common on

both wet and dry soils, and was also present on higher ground adjacent

to estuarine systems. Harvest mice were most common in unnatural dis-

22 Mary K. Clark, David S. Lee, John B. Funderburg, Jr.

turbed areas, trash piles, mowed road shoulders, and agricultural areas.

We have no records from wooded habitats, although the species is

found on the sand rims of Carolina bays and probably occurs in open

savannas.

Peromyscus leucopus leucopus (Rafinesque), White-footed Mouse.

On sand rim areas in Hoke and Bladen counties this mouse replaces

Peromyscus gossypinus, but it is present in extremely low densities and

trapped regularly only around trash piles and other places with ample

cover. The species was more widespread in Dare County. In recently

disturbed areas, it occurred sympatrically with P. gossypinus but was

seldom as common. We found P. leucopus in drained mature deciduous

bay forests with good ground cover off US 64 in Dare County, but

failed to find P. gossypinus there. This was the only "natural" site where

we collected leucopus in any vegetation stage remotely related to

pocosin habitats.

Peromyscus gossypinus gossypinus (LeConte), Cotton Mouse. The

Cotton Mouse was the most common species collected during our

study. All pocosins, Carolina bays, stream-head forests, and swamps

with woody vegetation, and most disturbed sites, were inhabited by

these mice. We also collected specimens from clear cut pine-shrub

pocosins.

Ochrotomys nuttalli aureolus (Audubon and Bachman), Golden

Mouse. Ochrotomys was common in mature forests in all areas studied,

and occurred sympatrically with Peromyscus gossypinus. We found the

Golden Mouse to be common in flooded Pond Pine-cane pocosins in

Dare County, and in evergreen bay and stream-head forests. It was

most common in ecotonal areas where light permitted vines (particu-

larly Smilax) to flourish, and absent from savannas, sand rims, and

unforested habitats.

Sigmodon hispidus komareki (Gardner), Cotton Rat. The Cotton Rat

was most commonly associated with dry, early successional stages, Wire

Grass savannas, and various disturbed communities, and was uncom-

mon on sand rims dominated by Wire Grass. At Bay Tree Lake we found

a few Sigmodon in the detritus line along the eastern shore of the lake.

Interestingly, these rats were not at all common on the Dare County

mainland, and those that we did find were not in natural communities.

At one Dare County site, where the plant community in a roadside

swale matured from Juncus to grasses and shrubs over a 2-year period,

Sigmodon replaced Microtus. The Cotton Rat apparently is absent

from the Dismal Swamp (Handley 1979), although we have records

from the southern and western edges of the swamp.

Mammals of Carolina Bays 23

Microtus pennsylvanicus nigrans (Rhoads), Meadow Vole. The Mea-

dow Vole was abundant in wet, early successional communities on the

Albemarle-Pamlico peninsula. We found it wherever Juncus was domi-

nant, and in Spartina marshes and road shoulders bordering pocosins.

Dark individuals, which appeared to be Microtus p. nigrans, were col-

lected as far west as Gates County, along the western edge of the Dismal

Swamp, and on the Dare County mainland. Except for a few barrier

island and salt marsh populations of Microtus p. pennsylvanicus, the

Meadow Vole does not occur on the Coastal Plain south of Pamlico

River. Consequently, it was not caught in any of our study sites in south-

eastern North Carolina.

Microtus pinetorum pinetorum (LeConte), Pine Vole. We found M.

pinetorum to be rare, but collected it in the ecotonal area between

Carolina bays and their adjacent sand rims in Hoke (NCSM 3830) and

Bladen (NCSM 4182) counties, and in dry areas adjacent to stream-

head forest. Most of the study sites we visited were too damp to support

this vole, but it is common in some drained agricultural areas that

apparently had once been Carolina bays.

Ondatra zibethicus macrodon (Merriam), Muskrat. Although found

throughout most of the Coastal Plain in marshes, ponds, and shallow

areas of lakes and impoundments, Muskrats are not generally associated

with pocosins or Carolina bays. Only in the drainage canals within

pocosin areas in Dare and Pasquotank counties were they common. We

have reports of muskrats in Sugg's Mill Pond, Bladen County, but in

this particular Carolina bay the water levels are artificially maintained

by earthen dikes and dams.

Synaptomys cooperi helaletes (Merriam), Southern Bog Lemming.

This disjunct Dismal Swamp race of the Bog Lemming had not been

found between 1896 and 1979 and there was some concern that it was

extinct. Subsequently it was collected by Rose (1981b) in both the Vir-

ginia and North Carolina portions of the Swamp. In addition, we have

a single specimen (NCSM 4019; Lee et al. 1982) from near Elizabeth

City, and David Webster, UNC-Wilmington, informed us that he has

collected this lemming in a young upland pine plantation near Mer-

chant's Mill Pond, Gates County, North Carolina. Thus, the race is

considerably more widely distributed than previously known. Rose

(1981a) documented its ability to invade clearings with heavy herba-

ceous ground cover. Intensive trapping on the Albemarle-Pamlico penin-

sula by NCSM personnel, in what we regarded as optimum habitat for

the species, indicated that this rodent does not occur south of Albemarle

Sound. In fact, it appears to occupy a pocket in northeastern North

Carolina and southeastern Virginia in which Sigmodon is absent. Based

24 Mary K. Clark, David S. Lee, John B. Funderburg, Jr.

on recent habitat information it seems that these two rodents are ecolog-

ically similar in the Southeast.

Mus musculus (Linnaeus), House Mouse. This is the only exotic spe-

cies that we encountered regularly in our study. Although common

along road shoulders and windrows in Dare County, and in suburban

sites in drained or altered bays in Bladen County, it was not found in

any natural sites.

Carnivora: Carnivores

Vulpes fulva fulva (Desmarest), Red Fox. The Red Fox apparently

avoids pocosins and is uncommon on sand rims. Agricultural workers

in Dare County informed us that Red Foxes did not appear locally until

extensive areas had been cleared for agriculture.

Urocyon cinereoargenteus cinereoargenteus (Schreber), Gray Fox.

Unlike the Red Fox, Urocyon is common in most densely wooded habi-

tats, including pocosins. Individuals are often seen running on sand rims

and ridges between Carolina bays and stream-head forests.

Ursus americanus americanus (Pallas), Black Bear. With the excep-

tion of the Sandhills sites in Hoke and Moore counties, bear popula-

tions still persist in all of our study areas. In 200 field-days we saw five

individuals, and fresh tracks were found on about a dozen occasions.

Pocosin communities contribute to the diets of Black Bears. Buell and

Cain (1943), for example, reported bears feeding on Smilax fruits, and

on young Smilax vines growing from seeds in bear scats.

Radio telemetry tracking of bears has been conducted in Dare (Hardy

1974) and Bladen counties (Hamilton 1978; Landers et al. 1979). In

Dare County bears appeared to use all cover types, including pocosins

(Hardy 1974). The preferred habitats were characterized by the presence

of diverse and generally dense vegetation and close proximity to rela-

tively extensive roadless areas. Hardy (1974) listed the order of habitat

preference as forested areas, older burns, more recent burns, and clear-

cuts. Landers et al. (1979) related seasonal habitat use in Bladen County

to foraging, denning and escape behavior. Carolina bays, which com-

prise about 44 percent of the county, received the most use by foraging

bears and contributed the greatest volume of natural foods to their diet.

Corn was a major component of the diet in every month, and the prin-

cipal food item during seven months of the year. All radio-monitored

bears that denned were found to bed on nests on the ground in very

dense thickets of Fetterbush and greenbrier. Large swamps provided the

best escape cover, which is probably the most critical component of

Black Bear habitat (Landers et al. 1979). Of 45 known bear mortalities

Mammals of Carolina Bays 25

in Bladen County from 1974 to 1976 none occurred in swamp forests

(Hamilton 1978). Most bays containing dense vegetation, though, were

too small to provide adequate cover. Access roads on sand rims and

between bays also increased the vulnerability of bears to hunters.

Procyon lotor lotor (Linnaeus), Raccoon. Raccoon tracks or foraging

animals were seen or collected in all communities except white cedar

forest, and at all sites studied. At no site were Raccoons particularly

common.

Mustela frenata noveboracensis (Emmons), Long-tailed Weasel. Based

on tracks, road-killed specimens, and interviews with trappers, weasels

are regular but uncommon inhabitats of pocosin communities. We have

reports from savannas (Brunswick County), bay and white cedar forest-

ed Carolina bays (Bladen County), and pine-shrub bogs (Dare and

Bladen counties). This weasel is probably found in most woodland

habitats.

Mustela vison mink (Peale and Palisot de Beauvois), Mink. We have

only one personal record of this animal from the study areas, but Dare

County fur trappers informed us that minks are rather common along

local drainage canals. Specific sites described to us were in mature ever-

green bay and swamp forests. We saw tracks in a dirt road bisecting a

wet section of Jerome Bog on the Bladen-Cumberland County line.

Lutra canadensis lataxina (F. Cuvier), River Otter. This mustelid is

relatively common in freshwater canals and estuarine systems bordering

pocosins in Dare County, and we have accumulated enough records to

assume it occurs in all the Bladen Lakes of Bladen County. Otters were

not recorded elsewhere.

Lynx rufus floridanus (Rafinesque), Bobcat. Although we found Bob-

cats or their tracks in only a few habitats, they probably occur occa-

sionally in most Coastal Plain habitats. Tracks were seen at nearly all

our study areas, and their frequency indicated that Bobcats must be

relatively common. Using radio telemetry, Lancia et al. (in press) fol-

lowed eight Bobcats for one to five months in the Croatan National

Forest, Carteret County. Home range was larger than reported in other

studies in the Southeast and varied from 12.37 to 50.35 km2, with males

having larger ranges. Females avoided pocosins and preferred agricultur-

al lands, but otherwise habitat use was in proportion to availability.

Lancia (pers. comm.) noted that the animals they studied were asso-

ciated with edges of pocosins when these habitats were used, and gener-

ally avoided interiors of extensive pocosins. A little-known book by the

Hon. Wm. Elliott (1918) contains a chapter on hunting Bobcats in

26 Mary K. Clark, David S. Lee, John B. Funderburg, Jr.

South Carolina. Excerpts from part of this chapter (see Appendix to

this paper) provide some anecdotal insights into the nature of pocosins

as a refuge for game animals, the behavior of pursued Bobcats, and the

perpetuation of eyewitness accounts of "black" panthers in the Southeast.

Artiodactyla: Hoofed Mammals

Odocoileus virginianus virginianus (Zimmerman), White-tailed Deer.

White-tailed Deer were observed in all habitats and at all study sites.

Deer are now widely distributed throughout the Coastal Plain, although

from the turn of the century through the 1930s they were reduced to a

few remnant herds confined to the pocosin areas of the Albemarle-

Pamlico Peninsula and to the Green Swamp area. Restocking and pro-

tection have been successful and the adaptability of deer is seen in their

presence in 12 of the 13 habitats presented in Table 1.

Capra hircus (Linnaeus), Domestic Goat. Feral goats are quite com-

mon in the low shrub pocosins in southern Dare County. Land owners

queried by us were not aware of their origin or how long they have been

present. Hill (1973) reported on some goats "gone wild" in the Dismal

Swamp in Virginia. We have not included them in Tables 1 or 2.

Recently Extirpated Mammals

Canis sp., Wolf. Based on place names and bounty records of the

early 1700's through the 1800's, there is no doubt that wolves ranged

throughout eastern North Carolina. There is some question, in our

minds, however, whether the wolves were Canis lupus, the Gray Wolf,

or Canis niger, the Red Wolf, or both. The Red Wolf was found at least

as far north as Charleston, South Carolina (records at the Charleston

Museum), and there is no reason to suspect that it did not range into

coastal North Carolina as well. Pocosins and Carolina bays would

appear to make ideal haunts for Red Wolves. Elliot (1918) noted that

wolves were almost extinct in the maritime sections of the Carolinas and

Georgia in 1867. We are not aware of any bounties paid on wolves in

the Coastal Plain of North Carolina after the mid- 1700s. As indicated

by records in the North Carolina State Archives, in 1721 Chowan

County paid bounties on "bobcats, panthers and wolves this year."

Places named for wolves in eastern North Carolina include Wolf

Bay, Bladen County; Wolf House Point, Currituck County; Wolf Pit

Creek, Hoke County; Wolf Pit Township, Richmond County; Wolf-

scape Township, Duplin County; and Wolf Swamp, Onslow County.

Felis concolor ssp., Panther. We know of no Atlantic Coastal Plain

specimen records that would indicate the subspecific status of Felis con-

Mammals of Carolina Bays 27

color in North Carolina. It is reasonable to assume that Felis c. coryi

(Bangs), like various other "Florida" races of mammals, ranged into the

southeastern part of the state. Specific records for Felis concolor in the

North Carolina Coastal Plain are few. We are aware of the following:

1721. Chowan Co. bounty records. (North Carolina State Archives).

1776. "Wicker Davis was paid by court 10 shillings for killing a

panther." Carteret Co. court minutes, vol. X, page 327.

Some years before the Civil War. One killed in Rose Bay, Hyde

County (NCSM records).

1900. Trapped in a pocosin in Craven County (NCSM records).

1930. Washington County, Lake Phelps, "skin seen by biologists"

(NCSM records).

Additionally, there are at least five Coastal Plain localities named after

Panthers, presumably each representing encounters of early settlers with

this cat. There are three different Panther Creeks, one each in Duplin,

Pitt, and Sampson counties; Panther Swamp, Northhampton County,

and Panther Swamp Creek, Greene County.

Sight records of Panthers still continue to be reported. Lee (1977)

surveyed the numerous reports, and the information in our files has led

to the following conclusions: 1) no recent reports of panthers in North

Carolina are authenticated by specimens, photographs, identifiable

tracks, hair samples, or scats; 2) seemingly reliable reports accumulated

over the last 80 years have clustered in a few specific areas; 3) the fre-

quencies and localities of Panther reports are directly related to past

and present distributions and numbers of White-tailed Deer; and 4)

based on a four-point scale of reliability (Lee 1977), nearly all reliable

Coastal Plain sight records are from pocosin-rich areas. Forty-four

records of Panther sightings in 20 eastern North Carolina counties were

reported to the State Museum in the 1970's, but none was substantiated

by photographs, footprints, or by other means. Unless evidence to the

contrary appears, we consider the Panther extirpated.

DISCUSSION

Of the 40 mammals found, probably only Blarina sp., Pipistrellus

subflavus, Sylvilagus palustris, Sciurus carolinensis, Peromyscus gossy-

pinus, Ochrotomys nuttalli, Urocyon cinereoargenteus, Ursus america-

nus, Procyon lotor, and Odocoileus virginianus occur with enough den-

sity or regularity to be considered typical (although not characteristic)

inhabitants of pocosin/ Carolina bay communities. When open savannas

are included in this system, Cryptotis parva and Sigmodon hispidus

should be included. All species, except possibly the Black Bear, can be

found in equal or greater abundance in many other Coastal Plain habi-

tats and therefore are not to be regarded as index species for pocosins.

28 Mary K. Clark, David S. Lee, John B. Funderburg, Jr.

In North Carolina, only a few species are so geographically re-

stricted that they would not be expected wherever pocosin habitats exist.

Two North Carolina mammals, Microtus pennsylvanicus and Synap-

tomys cooperi, are limited to the northeast section of the state. All oth-

ers, except Sciurus niger, Castor canadensis and to a lesser extent Ursus

americanus, appear to be more or less uniformly distributed throughout

the Coastal Plain. The Fox Squirrel is absent from pocosin areas in the

northeastern counties, probably because of a lack of adjacent sand rims,

open canopy pine forests, and mast-producing oaks. Through restock-

ing, the beaver is now widely, but not uniformly, established in the

Coastal Plain. Bears have been locally extirpated from the Sandhills

area.

Most of the mammals discussed here are opportunistic species that

exploit early and intermediate successional stages of many types of plant

communities. Several (Sciurus niger, Glaucomys volans, Microtus

pinetorum and the aquatic mammals) are associated only with pe-

ripheral communities of bays (sand rims) or aquatic systems and not with

pocosin vegetation per se. Most species appear to exist normally in low

densities within true pocosins and become common only in disturbed

areas or ones with temporary vegetative shifts caused by fire or storms.

Sequence of Succession

We interpret the sequence of successional changes of the plant/ -

mammal communities in the stages shown below (mammals listed in

approximate order of abundance). It should be emphasized that these

lists of characteristic mammals do not represent total faunal lists as

presented in Table 1, but indicate only species regularly found in each

major community type. The sand rim associates of Carolina bays, as

well as those of other community types, are listed in Table 1.

Early Stages. — Sedge/ grass /rush communities (canopy and shrubs

removed by fire or man) and savannas.

Characteristic: Oryzomys palustris, Sigmodon hispidus, Microtus

pennsylvanicus, Cryptotis parva, Reithrodontomys humulis, Syl-

vilagus palustris.

Occasional: Peromyscus leucopus, Mus musculus, Blarina sp.

Intermediate Stages. — Pine-shrub bogs.

Characteristic: Peromyscus gossypinus, Pipistrellus subflavus, Syl-

vilagus palustris, Odocoileus virginianus, Ursus americanus.

Occasional: Sylvilagus floridanus.

— White cedar forests (mature dense forest).

Characteristic: Blarina brevicauda.

Occasional: Sciurus carolinensis.

Mammals of Carolina Bays 29

Advanced Stages. — Evergreen and deciduous bay forests.

Characteristic: Sorex longirostris, Blarina sp., Sciurus carolinensis,

Peromyscus gossypinus, Ochrotomys nuttalli, Urocyon cinereoar-

genteus, Ursus americanus, Procyon lotor.

Occasional: Sylvilagus palustris, Lynx rufus, Mustela vison, Pleco-

tus rafinesquii, Didelphis virginiana.

We have not adequately surveyed white cedar forests and savannas.

Our limited information (Table 1), however, suggests low diversity and

density in white cedar forests and variable or highly fluctuating densities

in savannas. In the cedar forests, most mammal activity seems to be

restricted to edges (we also found this to be true of breeding birds). In

savannas, periodic flooding and fire limit populations, but the quick

response of grasses and other herbaceous plants after burning provides

excellent cover and food, and denuded areas are probably repopulated

quickly.

Factors Affecting Density and Diversity

Natural ecotones, openings, and edges caused by land-use practic-

es, were far more productive for mammal trapping and observing

mammal signs than were the interiors of pocosins. This can be attri-

buted to generally good cover in these areas, a richer diversity of food

plants, and a slight relief in topography that provides temporary refuge

from seasonal flooding. In grass and sedge stages we found high mam-

mal densities, and in several instances had over 50 percent trapping suc-

cess. Semi-flooded Juncus areas produced some interesting results.

Early in our studies of one such site we found Microtus (60% of total

catch), Reithrodontomys (23%), and Oryzomys (8%) to be the dominant

mammals. In following years, however, as the community matured and

grasses dominated the vegetation, Oryzomys (50%) and Sigmodon

(33%) became the more abundant species. Artificially maintained sys-

tems (mowed, grazed and drained, etc.) created habitats in which species

not typical of pocosins, such as Scalopus, Sylvilagus floridanus, Mus,

and Vulpes, appeared and often became numerous.

Although small mammals of many types quickly colonize early

(open canopy) successional stages (Rose 1981a, and this study), the

limited plant diversity probably does not provide a year-round food

base in the intermediate (i.e., pocosin) successional stages adequate to

attract or support small mammals. Thus, except for ecotones and natu-

ral or man-made openings, typical pocosin communities support very

low densities of mammals, and at any single site usually a low diversity

of species as well (see above lists of early and intermediate successional

species). In contrast, efforts were made to sample the interiors of poco-

sins and bays. The interior of these dense communities where the canopy

30 Mary K. Clark, David S. Lee, John B. Funderburg, Jr.

or subcanopy is closed do not seem to be frequented by any mammal

species, although they are certainly used for refuge by several large

species.

Overall, the influence of fire is more positive than negative for

mammals. Plants associated with pocosins respond quickly to burning,

and the new growth of herbaceous species and the temporary openings

in shrub layers generally result in an increase in small mammals. Areas

we trapped in mainland Dare County one and two years after a major

fire in extensive Pond Pine-shrub and Pond Pine-cane pocosins pro-

duced higher trap yields than did adjacent unburned areas. Burned or

clear-cut sand rims and burned stream-head forests also had higher trap

yields than those that were unburned for long periods. The fire resistant

pines, most shrub root stocks, and moist soils would protect arboreal

and terrestrial species of mammals from burning.

In many areas the plant communities are perched on hard subsoils

that form natural basins; these basins retain surface water in the organic

topsoils. Root systems of many of the bay forest trees usually do not

penetrate the subsoils, and the limited support offered by these shallow

peat soils makes larger trees extremely vulnerable to strong winds and

ice storms. In addition, frequency of blow downs is high because many

such trees are "crown heavy," a result of early competition with the

normally dense understory vegetation. During the spring of 1983 we

found extensive uprooting and limb breakage of trees (particularly Red

Maple, Red Bay and Atlantic White Cedar) in Carolina bays in Bladen

County and white cedar forests in Brunswick County. This damage was

caused by late March snow and ice storms. Hurricanes and tornados

would certainly cause even greater damage. Buell and Cain (1943)

observed areas where the weight of Smilax climbing into the canopies of

white cedar forests caused trees to uproot. Thus, natural openings in

advanced successional stages are commonplace. They provide numerous

sites for shade intolerant plants and for early successional and ecotonal

faunas to maintain populations during periods when pocosins are in

intermediate and advanced stages of development. Modest mammal

diversity and density is apparent in such openings.

Degree and duration of flooding of pocosin communities is

extremely variable. In general, areas with organic soils have protracted

hydroperiods, whereas those with mineral soils have comparatively

short hydroperiods. Local topography, the nature of soil types of adja-

cent communities, and land drainage also affect the amount of standing

water. As expected, the soils in most communities remained saturated

for extended periods. However, presence or absence of small mammal

populations is dictated by the retention of standing surface water. The

following comments provide examples of the effects that flooding has

Mammals of Carolina Bays 3 1

on the composition of local mammal diversity and density (as gauged by

trap-night success, shown in parentheses).

Dare County. — North side of Milltail Lake; 525 trap-nights in a 40-

70 year pure stand of white cedar yielded three Blarina brevicauda

(0.56%). A dense ground cover of sphagnum and other mosses, and

numerous stumps and fallen logs, provided ideal microhabitats for small

mammals. However, a later visit to the site revealed that strong winds

regularly pushed lake water deep into the cedar forest, leaving only

small, isolated hummocks unflooded.

Bladen County. — Salters Lake. Regular winter flooding, and period-

ic partial flooding at other seasons, of a mature Carolina bay forest

apparently limits the local fauna in the bay forest. In 1,120 trap-nights

in winter we collected only 6 Peromyscus gossypinus and 1 Ochrotomys

nuttalli (total 0.62%), but in April, after standing water subsided, trap

success (1.39%) was somewhat higher. Blarina sp. were also present but

collected only in pitfall traps.

Hoke County. — McCain study site. In trapping the interior of a 2

hectare Carolina bay dominated by high shrub, in which areas with

good cover for small mammals were numerous, we collected one adult

Peromyscus gossypinus in 460 trap-nights (0.2%). Although the interior

of the bay remained free of long-term standing water for at least two

years (1980-82), heavy rains for extended periods in the spring of 1983

flooded over 80% of the interior of this bay and the clay-based subsoils

retained the water for at least four months. Deer and rabbit (presuma-

bly S. palustris) sign were the only indications that the interior was used

by mammals other than Peromyscus. Density and diversity were greater

around the edges of the bay.

Pender, Dare, and Bladen Counties (shrub bogs). — On all visits,

standing water was so prevalent that we did not normally attempt to set

traps. (One attempt at trapping in Dare County yielded no mammals in

250+ trap-nights.) Even in the long-term absence of flooding, mammal

populations in the interiors of pocosins probably are depauperate, an

effect of seasonally limited food supplies. The characteristic plants are

not true mast producers, and the majority of plant species hold their

seeds for extended periods, rendering them inaccessible to ground-

foraging mammals. In the interior of dense, intermediate-to-advanced

pocosins, the exclusion of light prohibits flowering and fruiting of most

understory plants (pers. observ.). From midwinter through early spring

(normally the typical flood period) food resources in pocosin interiors

are minimal. In all seasons very low raptor densities correlate with low

densities of small mammal prey, a further indication of minimal availa-

bility of food plants for rodents. Breidling et al. (1983) addressed the

problems of interpreting low density and diversity in four Dismal Swamp

32 Mary K. Clark, David S. Lee, John B. Funderburg, Jr.

plant communities. Interestingly, they suggested that no single factor

was responsible, but discussed the impacts of low food availability and

flooding.

The paucity of small mammals in pocosins and related communi-

ties can be illustrated by comparing trap-night success in those habitats

with success in other habitats. For example, 3,000 trap-nights in pocosin

communities (data pooled from Bladen, Hoke and Dare counties) had

only a 2.06 percent capture success, and 1,000 trap-nights in mature

evergreen bay forest in Bladen County had 1.23 percent success. On the

other hand, 2,000 trap-nights in non-pocosin habitats in Currituck

County averaged 6.1 percent trapping success, and 365 trap-nights in

upland habitats of various types adjacent to Carolina bays in Bladen

County yielded 7.12 percent success. The point here is that plant com-

munities, not geography, are responsible for the relative densities.

Pocosin Systems as Refugia

Lee et al. (1982) discussed the occurrence in pocosin-rich areas of

relicts such as Synaptomys cooperi, semi-relicts such as Condylura cris-

tata and Marina brevicauda, and species at the limits of their ranges.

About 25 percent of the mammals associated with pocosins fall into one

or another of these categories. Species that reach their northern distri-

butional limits on the Atlantic slope within the Dismal Swamp area

include Plecotus rafinesquii, Sylvilagus aquaticus, Peromyscus gossypi-

nus, Ochrotomys nuttalli, and to a lesser extent Sigmodon hispidus.

Except for fragmented populations of Condylura cristata and saltmarsh

populations of Microtus pennsylvanicus that occur farther south, the

Dismal Swamp area is also the southern limit for northern species that

invade the southeastern Coastal Plain on the Atlantic slope. These

include C. cristata, M. pennsylvanicus, and S. cooperi, and possibly M.

lucifugus and B. brevicauda. Furthermore, Condylura cristata and

Sorex longristris each reach the southern extremes of their ranges in

pocosin-like communities, the former in the Okefenokee Swamp, Geor-

gia (Paradiso 1959) and the latter in the Green Swamp, Polk County,

Florida (Hill 1945). Their southernmost populations appear to be

somewhat disjunct from populations to the north.

Pocosin habitats may have provided refugia for both northern and

southern species since Pleistocene times, allowing populations to persist

beyond otherwise normal distribution limits. The protective evergreen

vegetation and the heat retention of high water tables should buffer the

extremes of severe winter weather, while shade and evaporative cooling

could be expected to ameliorate extreme high temperatures of Coastal

Plain summers. Hibbard (1960) postulated that in the late Pleistocene

(ca. 16,000 years ago) a period of climatic equality existed, in which

Mammals of Carolina Bays 33

milder winters and cooler summers prevailed. These conditions allowed

northern species to extend farther south and subtropical species farther

north than they do today. This hypothesis is supported by evidence

from various Southeastern fossil deposits (Holman 1976, 1982;

Slaughter 1975). It is well documented that boreal elements were estab-

lished in the Southern Applachians during the Pleistocene, but the pres-

ence of northern species in the Coastal Plain is not generally recognized.

Whitehead (1963) discussed northern elements of Pleistocene flora in

the Southeast and included information on two North Carolina Coastal

Plain bays. Based on fossil pollen from Singletary Lake and Rockhound

Bay, Whitehead showed that northern plants once were present in east-

ern North Carolina.

Whether floral and faunal elements reached their distributional lim-

its as a result of the effects of glacial displacements and interglacial

warming periods as is widely accepted, or did so simultaneously during

a period of climatic equality, is not critical to this discussion. Either

situation could produce the current assemblage of northern and south-

ern elements that persists on the outer Coastal Plain of North Carolina.

However, it is interesting that in some areas, well beyond their typical

distributional limits, both northern and southern elements now coexist.

As the climate shifted to the present regime, some relict and semi-relict

populations were stranded along the outer Coastal Plain as well as in

the southern Appalachians. Just as frost pocket bogs, areas of high ele-

vation, and cove forests provided local refugia in the mountains, poco-

sins and Carolina bays apparently have done so on the Coastal Plain.

Pocosins as Key Coastal Plain Habitats

In addition to their roles as geographic refugia and climatic buffers,

pocosins and Carolina bays were important as natural stands of early

successional habitat. Their subclimax communities and complex zona-

tion provided habitats for early to intermediate successional mammals

that in precolonial times would not otherwise occur regularly on the

Coastal Plain. Today, pocosins and related communities are not criti-

cally important for the geographical maintenance of most early succes-

sional mammal species because grazing, mowing, lumbering and similar

activities produce a wide array of early successional stages over exten-

sive areas. For example, Robinson and Lee (1980) pointed out that

Marmota monax was unable to invade the Piedmont Plateau and Coast-

al Plain in the Southeast prior to extensive artificial maintenance of

early upland communities and corriders for dispersal. The same is prob-

ably true for Vulpes fulva (Lee et al. 1982). Thus, for animal species

already associated with pocosins and therefore widely distributed across

the Coastal Plain, local expansion of their populations into disturbed

34 Mary K. Clark, David S. Lee, John B. Funderburg, Jr.

upland habitats is not surprising. In other words, species that may have

been confined to pocosins in the past because of ecological restrictions

are now able to exploit a wide array of disturbed community types as a

result of agricultural and other activities.

Since pocosins and Carolina bays are extensive on the outer Coast-

al Plain, dispersal of small lowland mammals from one to another is

feasible. River floodplain swamps can also facilitate dispersals of pocosin

colonizers. Although these swamps would provide climatic buffers sim-

ilar to those found in pocosins and Carolina bays, their frequent and

rapid flooding and periods of protracted submergence make them un-

suited to support permanent small mammal populations. Compared to

adjacent upland communities, our trapping experience indicates that

Coastal Plain swamps have extremely depauperate mammal faunas.

Bears, deer, bobcats and, until recent times, probably panthers and

wolves, frequented pocosins and Carolina bays in the Coastal Plain.

Within the last century most of those species had become confined to

these areas, but this may have been an artifact of uncontrolled harvest

before modern game management programs were developed and not a

reflection of the specific habitat needs of these animals. Populations

simply persisted in the inaccessible reaches of extensive pocosins longer

than they were able to in other parts of the Coastal Plain. In eastern

North Carolina, White-tailed Deer became restricted to a few pocosin

areas by the turn of the century, and only in the last 30 years or so have

they again become common in other areas and habitats.

Preservation and Management

We are concerned about pocosin preservation, but find the argu-

ment that these areas harbor many rare and unique faunal elements to

be overstated, at least for mammals and birds (personal observations).

As we have shown, pocosin mammal faunas generally consist of species

with wide ecological tolerances and abilities to exploit early successional

stages and areas disrupted by human activities. Pocosin areas should be

preserved for a variety of reasons, but our present knowledge of the

vertebrate fauna leads us to suggest that wildlife values for mammals

(except the Black Bear), may not be a primary concern. Our results indi-

cate that some types of alterations, when followed by normal succes-

sional patterns, actually increase species diversity and density of mam-

mals and birds in Carolina bays and pocosins. Unpublished results of

research by others indicate that this may be true of a variety of verte-

brates. This is not surprising, because many such alterations simply

change pocosin habitats in ways similar to those in which they are mod-

ified by fire and fluctuating water tables.

Lee and Funderburg (1977) discussed the conservation status of

Mammals of Carolina Bays 35

North Carolina mammals, and listed as "status undetermined" Blarina

brevicauda telmalestes, Sorex longirostris fisheri, Lasiurus seminolus,

Plecotus rafinesquii macrotis, Microtus pennsylvanicus nigrans, and

Synaptomys cooperi helaletes, all associated with pocosins and Carolina

bays. Ursus americanus was considered of special concern, and Felis

concolor as endangered and possibly extirpated. Later information on

the "undetermined" species and races has shown them to be more com-

mon or widespread than previously suspected (see Lee et al. 1982; Rose

1981a; and this study). The number of known localities of extant popu-

lations has at least doubled for all these mammals, and some are now

known to be widespread and even common. In the case of the Black

Bear, though, pocosins and related habitats play a more than minor role

where survival is concerned. A significant percentage of the surviving

Coastal Plain bears is closely associated with pocosins and Carolina

bays. Modern hunting methods, however, may make even extensive

pocosins unsafe sanctuaries for bears.

The classification of pocosins and Carolina bays as types of wet-

lands, although in our opinion correct, has led others to a general

assumption of high wildlife values despite the lack of systematic inven-

tories. Misconceptions, absence of standard definitions, and lack of

comparative information from other Southeastern wetland habitats,

have also contributed to the problems of inventory and projection of

habitat loss. Accurate assessments of the wildlife values of pocosins are

further hampered by the lack of comparable historical comparisons of

the mammalian fauna in the Southeast in general, and in the North

Carolina Coastal Plain in particular.

ACKNOWLEDGMENTS.— We thank the U.S. Fish and Wildlife

Service for partial support of our studies in Dare County (contract

number 14-16-0004-81-056). Bryan Taylor, N.C. State Parks, provided

permission and encouragement to study certain areas in the state parks

system. Bill Adams, Eloise Potter, Gilbert Grant, Steve Platania, Marty

Williams, Warren Parker, Paul Kumhyr and Danny Smith all assisted

in portions of our inventory effort. The Explorers Club provided,

through Clark, a grant for D. Smith to determine the mammals of Salt-

ers Lake Bay. The U.S. Army Corps of Engineers provided Lee and

Platania funds (contract number DACW 54-78-F-0069) for a study of

the Coinjock area. David K. Clark, Elizabethtown, gave assistance in

aerial surveys of Carolina bays and lodging to field party members in

Bladen County.

36 Mary K. Clark, David S. Lee, John B. Funderburg, Jr.

LITERATURE CITED

Breidling, F. Elizabeth, F. P. Day, Jr. and R. K. Rose. 1983. An evaluation of

small rodents in four Dismal Swamp plant communities. Va. J. Sci.

34(1): 14-28.

Buell, Murray F. 1946a. Jerome Bog, a peat-filled "Carolina Bay." Bull. Torrey

Bot. Club 73:24-33.

1946b. The age of Jerome Bog, "A Carolina Bay." Science 703:14-15.

, and R. L. Cain. 1943. The successional role of southern white cedar,

Chamaecyparis thyoides, in southeastern North Carolina. Ecology24: 85-93.

Elliott, William. 1918. Carolina Sports by Land and Water. State Co. Printers,

Columbia, SC. 260 pp.

French, Thomas W. 1981. Notes on the distribution and taxonomy of short-

tailed shrews (Genus Blarina) in the Southeast. Brimleyana 6:101-1 10.

Hall, E. Raymond. 1981. The Mammals of North America. John Wiley and

Sons, New York. 1 181 + 90 pp., 2 volumes.

Hamilton, Richard J. 1978. Ecology of the black bear in southeastern North

Carolina. Masters thesis, Univ. Georgia, Athens. 214 pp.

Handley, Charles O., Jr. 1959. A revision of American bats of the genera

Euderma and Plecotus. Proc. U.S. Natl. Mus. 110:95-264.

1979. Mammals of the Dismal Swamp: A historical account. Pp.

297-357 in P. W. Kirk (Ed.). The Great Dismal Swamp. Univ. Press Vir-

ginia, Charlottesville. 427 pp.

Hardy, David M. 1974. Habitat requirements of the black bear in Dare County,

North Carolina. Masters thesis, Virginia Polytech. Inst. State Univ.,

Blacksburg. 121 pp.

Heath, R. C. 1975. Hydrology of the Albemarle-Pamlico Region, North Caro-

lina. Water-Resource Investigation 9-75, U.S. Geological Survey, Raleigh.

98 pp.

Hibbard, C. W. 1960. Pliocene and Pleistocene climates in North America. 62nd

Annu. Rep. Mich. Acad. Sci. Arts Letters (1959-60): 5-30.

Hill, D. 1973. The great swamp's brightened future: anything but dismal. Com-

monwealth 40( April): 12-24, 48-52.

Hill, J. E. 1945. Bachman's shrew from Polk County, Florida. J. Mammal.

2(5:88.

Holman, J. Alan. 1976. Paleoclimatic implications of "ecologically incompati-

ble" herpetological species (late Pleistocene: southeastern United States).

Herpetologica 32(3): 290-295.

1982. Late Pleistocene fossils in Georgia. Explor. J. 60(4): 160-163.

Kologiski, R. L. 1977. The phytosociology of the Green Swamp, North Caro-

lina. N. C. Agric. Exp. Stn. Tech. Bull. 250. 100 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.

Landers, J. Larry, R. J. Hamilton, A. S. Johnson and R. L. Marchinton. 1979.

Foods and habitat of black bears in southeastern North Carolina. J. Wildl.

Manage. 43(1): 143-153.

Lee, David S. 1977. Unscrambling rumors, the status of the panther in North

Carolina. Wildl. N.C. 47(3):6-9.

Mammals of Carolina Bays 37

In manuscript. Breeding bird communities in pocosins.

, and J. B. Funderburg. 1977. Mammals. Introduction, and species

accounts. Pp. 385-389, 392-407 in J. E. Cooper, S. S. Robinson and J. B.

Funderburg (Eds.). Endangered and Threatened Plants and Animals of

North Carolina. N.C. State Mus. Nat. Hist., Raleigh. 444 pp.

, and M. K. Clark. 1982. A Distributional Survey of

North Carolina Mammals. Occas. Pap. N.C. Biol. Surv. 1982-10. 70 pp.

Merriam, C. Hart. 1895a. Revision of the shrews of the American genera Bla-

rina and Notiosorex. North Am. Fauna 10:5-34.

1895b. Synopsis of the American shrews of the genus Sorex. North

Am. Fauna 10:57-100.

1896. Revision of the lemmings of the genus Synaptomys with de-

scriptions of new species. Proc. Biol. Soc. Wash. 70:55-64.

McMullan, Philip S., Jr. 1984. Land-clearing trends on the Albemarle-Pamlico

Peninsula: 300 years of development in a wetlands region. Privately pub-

lished, McMullan Consulting, Durham. 131 pp.

Pagels, John F., C. S. Jones and C. O. Handley, Jr. 1982. Northern limits of the

southeastern shrew, Sorex longirostris Bachman (Insectivora: Soricidae),

on the Atlantic Coast of the United States. Brimleyana 8:51-59.

Paradiso, John L. 1959. A new star-nosed mole (Condylura) from the south-

eastern United States. Proc. Biol. Soc. Wash. 72:103-108.

Paul, John R. 1965. Blarina telmalestes in North Carolina. J. Mammal.

46(3):496.

Radford, Albert E., H. E. Ahles and C R. Bell. 1964. Manual of the Vascular

Flora of the Carolinas. Univ. North Carolina Press, Chapel Hill. 1183 pp.

Robinson, Sarah S., and D. S. Lee. 1980. Recent range expansion of the

Groundhog, Marmota monax, in the Southeast (Mammalia: Rodentia).

Brimleyana 3:43-48.

Richardson, Curtis J. (Ed.). 1981. Pocosin Wetlands: An Integrated Analysis of

Coastal Plain Freshwater Bogs in North Carolina. Hutchinson Ross Publ.

Co., Stroudsburg, PA. 364 pp.

1983. Pocosins: Vanishing wastelands or valuable wetlands? Bio-

Science 53(1 0):626-633.

Rose, Robert K. 1981a. Small mammals in openings in Virginia's Dismal

Swamp. Brimleyana 6:45-50.

1981b. Synaptomys not extinct in the Dismal Swamp. J. Mammal.

60(4):844-845.

Sharitz, Rebecca R., and J. W. Gibbons. 1982. The ecology of Southeastern

shrub bogs (pocosins) and Carolina bays: a community profile. U.S. Fish

Wildl. Serv., Div. Biol. Serv., Washington. FWS/OBS-82/04. 93 pp.

Slaughter, B. H. 1975. Ecological interpretation of the Brown Sand Wede local

fauna. Pp. 179-192 in F. Wendorf and J. J. Hester (Eds.). Late Pleistocene

Environments of the Southern High Plains. Publ. Fort Burgwin Res. Cen-

ter 9:179-192.

Tooker, W. W. 1899. The adapted Algonquin term "Poquosin." Am. Anthropol.

(Jan): 162-170.

Wells, B. W. 1932. The Natural Gardens of North Carolina. Univ. North Caro-

lina Press, Chapel Hill. 458 pp.

1946. Vegetation of Holly Shelter Wildlife Management Area. N.C.

38 Mary K. Clark, David S. Lee, John B. Funderburg, Jr.

Dep. Conserv. Develop. Bull. 2. 40 pp.

, and L. A. Whitford. 1976. History of stream head swamp forests,

pocosins, and savannas in the Southeast. J. Elisha Mitchell Sci. Soc.

92:148-150.

Whitehead, Donald R. 1963. "Northern" elements in the Pleistocene flora of the

Southeast. Ecology 44:403-406.

Wilbur, Henry M. 1981. Pocosin fauna. Pp. 62-68 in Richardson, C. J. (Ed.).

Pocosin Wetlands: An Integrated Analysis of Coastal Plain Freshwater

Bogs in North Carolina. Hutchinson Ross Publ. Co., Stroudsburg, PA.

364 pp.

Woodwell, George M. 1956. Phytosociology of Coastal Plain wetlands of the

Carolinas. Masters thesis, Duke Univ., Durham. 50 pp.

Accepted 28 February 1985

Appendix

The Hon. Wm. Elliott and his friends regularly employed dogs to hunt

Bobcats in Carolina bays and bay heads near Beaufort, South Carolina. The

following is Elliott's (1918:142-145) description of the habitat and the hunt.

"...quagmire at the surface, briers above (wherever their places were not

preoccupied by bay-trees, that, for want of elbow room, had grown up as

straight as canes, and almost as close)... and where the cat... had ensconced

himself behind an entrenchment of briers, which hounds, unless their blood was

heated by pursuit, would not willingly enter — so that he remained undetected.

"The hounds had not long entered the thicket, in which (from finding at its

edge the remains of a half-devoured rabbit) we concluded that the cat still

lurked,... and soon, a burst from the pack assured us that the cat was roused...

But he keeps the cover, which is so thick as to defy the keenest sight; and circles

it securely, leaving the dogs to tear their way through the briers. 'Ha! what is

that? a shot! — another!'... Another shot! ah, now they pause — one savage

growl — one stifled cry — and all is hushed..." [Three hunters surrounded the bay

and each shot one or more times at the cat.]

"'How now?' says the judge, 'what hocus pocus is here? This is a tawny,

leopard-like animal, while I pronounce the cat I fired at to be bigger and

blacker; I saw it clearly as it rolled over in the swamp at the flash of my gun.'

"'My opinion, in this case, is precisely the same,' said the doctor. 'I fired at

a black cat; the dogs must have changed cats during the chase!'

"'So much the better, gentlemen,' said I; 'we shall then have two cats,

instead of one. Put on the hounds, boys!' They were taken to the point from

which the doctor fired; but the stupid animals could find no trail, but that which

led them again to the spot on which the tawny cat lay dead!

"... the cat, being duly subjected to a post-mortem examination, was found

to have been struck by four out of the six shots fired at him — and the doctor's

shot, of peculiar size, being lodged in his body, left no doubt of the fact, that the

black cat of the doctor and judge was no other than the tawny cat of the rest of

the field. Whether the change of color was in the skin of the cat, or the eye of

the sportsman, or the distribution of light, we leave philosophers to determine."

Sympotthastia Pagast (Diptera: Chironomidae), an

Update Based on Larvae from North Carolina,

S. diastena (Sublette) comb, n.,

and Other Nearctic Species

Jan S. Doughman1

U. S. Geological Survey,

National Water Quality Laboratory,

6481-H Peachtree Ind., Doraville, Georgia 30340

ABSTRACT. — Chironomids from the Nearctic believed to belong to

Sympotthastia Pagast have been verified. They have been described

and keyed against the well-known Palaearctic species. Generic descrip-

tions of the male imago and pupa are expanded to include these spe-

cies. Species found in the Nearctic in the larval stage, S. fulva

Johannsen, a species near fulva, and S. zavreli Pagast, are contrasted

with a generic larval description. The nearctic S. diastena (Sublette)

comb, n., known only in the adult stage, is redescribed using two

imagos collected in Oregon.

INTRODUCTION

This paper offers a taxonomic and ecological summary of all

known species of Sympotthastia, which should be identifiable using the

generic descriptions and keys provided. Species are separable only by

minor differences, but two species groups are discernable.

The Palaearctic S. zavreli-group consists of S. zavreli Pagast, S.

spinifera Serra-Tosio, and S. macrocera Serra-Tosio. Most of these spe-

cies are well described in all stages (Serra-Tosio 1971, Ferrarese and

Rossaro 1981). An unassociated larva found in North Carolina is keyed

and described herein as zavreli (cf. Thienemann 1952), and this is the

only member of the zavreli-group to be found in the Nearctic.

The Nearctic S. fulva-group consists of S. fulva (Johannsen), S.

diastena (Sublette), a species near fulva, known in the larval stage, and

a pupa described by Saether (1969).

Sympotthastia fulva was described at all stages by Johannsen

(1921, 1937) with an emended description of the adults by Sublette

(1967). I believe that this species is properly placed because the imagoes

of both sexes were described from associations with the immatures

reviewed for this paper. Dr. Dean Hansen identified two slides of male

imagoes as S. diastena. This is a new combination with which I concur;

•Present address: U.S. Geological Survey— WRD, FB^4, 301 W. Congress,

Tucson, Arizona 85701

Brimleyana No. 1 1 :39-53, October 1985 39

40 Jan S. Doughman

see its new description below. The holotype was poorly mounted and

much smaller and less setiferous than these finds of Hansen, but the

phallapodeme unmistakably belongs to Sympotthastia. Following the

criteria of Serra-Tosio (1971), an attempt was made to diagram the male

imagoes of Sympotthastia in their phylogenetic order (Fig. 14).

Terminology used in my keys and descriptions follows Saether

(1980), but Saether did not illustrate a Diamesini. Several terms used

for parts of the male hypopygium were seemingly interchangeable,

namely superior volsella and aedeagus. I decided to use his term, supe-

rior volsella (SVo), to describe the anteromedial portion of the phallapo-

deme that is apodemal in nature, that is, this structure is heavily sclerot-

ized and articulates with the sternapodeme. This is illustrated for the

Protanypini (Saether 1980: Fig. 13), but I think this structure should be

consistently present in some form or another in all Diamesinae. The

putative aedeagus is only a flaring of the lower phallapodeme in the

Orthocladiinae (Saether 1980:Fig. 16). Saether, however, changed his

terminology to aedeagus sensu Hansen and Cook (1976) for most of the

Diamesini. Aedeagus, or a penislike, intromittent structure, is present as

a secondary, microsetigerous lobe of the phallapodeme in most of the

species of the sister genera Potthastia and Sympotthastia. It seems

proper to refer to this lobe of the phallapodeme as the median volsella

(MVo). The Diamesini have an anteromedial, hirsute lobe, which may

not be very well expressed, on the gonocoxite resembling that of

Saether (1980: Fig. 16). This report uses Saether's term, inferior volsella

(IVo) for this lobe such as it exists in Sympotthastia (see Fig. 12).

The following ratios are expressed in key as percentages:

Leg ratio = first tarsomere/ tibia;

Palp to face = lengths of palpal segments 2-5/ facial width at eyes;

Venarum ratio = length of cubital vein (from the arculus to its fork)/

length of medial vein.

Sympotthastia Pagast

Sympotthastia Pagast, 1947:457, type species, S. zavreli Pagast, orig.

design.; Serra-Tosio, 1968:129-130 and 1971:267-268 (descriptions of

adults and pupae).

Diamesa heterodentata Botn. et C.-Cure(?), Pankratova, 1970 (incom-

plete larval description but spoonlike premandible possessing lateral

teeth).

Nee Symp. sp. Simpson and Bode, 1980:30 (= P. gaedii Mg.).

Male. — Color uniform and dark. Wing 3-4 mm; body 3-6 mm. Eye

naked between lenses (X400) and reniform. Antenna with 13 plumose

flagellomeres; AR usually 2-3; A relatively long and slightly swollen

apically below the subterminal spine. Postorbital setae biserial at mid-

Sympotthastia update

41

3s/" -

Figs. 1-3. Sympotthastia diastena from Oregon: 1, wing; 2, head showing

endoskeleton; 3, hypopygium showing endoskeleton.

eye; outer verticals few, extending only to top of eye; inner verticals and

orbital setae absent. Clypeal setae 0-15. Palpal length less than width of

face; second and third segment partially fused (Fig. 2). Antepronotum

with medial commissure closed under scutal process (i.e., antepronotal

halves not gaping) and only lateral setae present. Acrostichals absent.

Dorsocentrals usually uniserial at mid-thorax but otherwise multi-serial.

Prealars with some setae forward and free of main cluster. Scutellar

setae numerous. Wing macrotrichia on cells absent but punctations

gross and visible at X400; setae on R4+5 usually absent; anal lobe

rounded; squama setae dark; alula without setae; venarum ratio 0.87-

0.90. Spiniform setae of legs present only on 1st and 2nd tarsomere (ta)

of P2and P3; other leg setae finer. Ta4 cylindrical and shorter or longer

42 Jan S. Doughman

than ta5. Hypopygium with a rectangular sternapodeme. Sternite IX

with a small tubercle protruding over base of each gonocoxite; anal

point (AnP) absent or a small, awl-like protrusion of tergite IX which is

not supported by oblique tergal apodemes. Pars ventralis small. Median

volsella (MVo) clavate and setigerous; superior volsella (SVo) a strongly

sclerotized and curving band (Figs. 3, 12). Gonocoxite tapering, nearly

parallel sided, with longish setae scattered on the lower two-thirds;

proximal half of inferior volsella (IVo) with a pile of microtrichia and

distal half with short, perpendicular setae and ending slightly free of

gonocoxite; and a second medial 'lobe' covered with short setae extend-

ing the length of the middle third of the gonocoxite. Gonostylus tubu-

lar, pubescent and with a few setae distally, terminally with thick chitin

surrounding its face below the perpendicular macroseta, and two strong,

perpendicular setae stand just proximal to macroseta (Figs. 3, 7).

Female. — Antenna with six flagellomeres (first two somewhat

fused). Alula of wing with setae and anal lobe not rounded. Tergite IX

apparently large, segment X without setae, lateral parts darkly sclerot-

ized, postgenital plate apparently well developed, cerci hexagonal with

subequal sides, and three oval seminal capsules. (Serra-Tosio 1971:Fig.

124; Saether 1977, based on S. zavreli).

Pupa. — Length 5-7 mm. Frontal apotome without cephalic tuber-

cles, but long frontal seta arising from each rugose spot. Thoracic horn

absent. Middle precorneal seta longer and thicker than the two, sub-

equal, flanking setae. Thoracic, antennal and wing sheaths smooth. Leg

sheath of Pending in a Z-curve. Tergite I (Tl) with only a polygonal

pattern and only L2 and L4 present. Tergites II-VII with an anterior

band of color only (apophyse) followed by some group shagreen border-

ing polygons; then spinulae dispersing evenly and ending in rows of

small, triangular spines on the posterior margin; finally rows of strong,

anteriorly pointing spines on conjunctiva II /III - VII / VIII (Fig. 4). L2

dorsal. L|3 evenly spaced but moving toward lateroposterior on each

successive segment. L3and L4in close proximity. L-setae usually api-

cally forked, up to eleven branches on L2 on VII and VIII. D-setae

small, dispersed, and usually simple. MD1 somewhat transverse and

MD2 longitudinal. Shagreen on sternites similar to that of tergites.

Segment IX with a fringe of very short setae on lateral border; each lobe

having a terminal, tooth-like ventral tubercle near the last of the three

straight macrosetae. Gonopodal sacs straight and not extending past

caudal margin.

Larva (4th stadium). — Head quadrate, rather thinly sclerotized,

and color uniform luteous or darker. Mentum with a yellow or smoky,

subequally trisected ventromentum (median area) that is 6-7X the width

of the much darker first dorsomental (lateral) tooth; six to seven

Sympotthastia update 43

obliquely arrayed, dark lateral teeth. Ventromental plate (VmP) present,

teardrop-shaped, or absent. Antennal AR 1.5 or 2.0-2.5; A2 appearing

very bifid with the blade base fused to it laterally, or appearing normal

with the blade base fused to A nearer to the apex of Af; ring organ in

basal fourth; blade and style elongate, reaching at least to A4; A annu-

late. Labrum with simple S-setae, mostly simple chaetae, and simple

spinulae over premandible. SIII hairlike and SIV lanceolate. Labral

lamellae (LL) two broad, simple or slightly denticulate plates. Chaeta

media broad at base and sometimes frayed apically. Epipharyngeal area

with a three scaled pecten flanked on each side by a pair of larger blades

and several thinner spines. Premandible mitten-shaped with 1-4 incon-

spicuous lateral teeth on its medial, curved margin; brush spike-like lat-

eral spine only. Mandible normal or sickle-shaped as in Potthastia spp.

Prementum with three groups of long, flat bristles. Maxilla with a low

palp and sensilla; galea without a row of lanceolate pegs. Body moder-

ate in length (6-12 mm) and without conspicuous lateral setae even on

the 10th segment (base of parapod). Procercus button-like, heavily

sclerotized anally, height to width about one, and supporting seven long

(about 400 um) anal setae (AS) and two small, unequal lateral setae.

Supraanals not reduced, but shorter than anal setae. Anal tubules (TA)

fingerlike, rounded or pointed apically, and shorter than parapod. Pos-

terior parapod moderately elongate with 16 dark claws.

Remarks. — The adult males of Sympotthastia, according to Serra-

Tosio (1971), is relatively plesiomorphic to Potthastia based on the

trends of unreduced chaetotaxy and the cylindrical ta4. Using the reduc-

tion in chaetotaxy within the genus, a phylogenetic scheme is proposed

that includes the Nearctic species and that leaves S. zavreli the most

apomorphic (Fig. 14).

Sympotthastia is a very uniform genus in all stages, but two species

groups may be separated. The zavreli-group is mostly Palaearctic, and

the /w/va-group is Nearctic. Adults have differences in the palpal 2nd

segment, color of capitellum of haltere, and anal point. Pupae show

only slight variations in chaetotaxy, and larvae have or do not have

ventromental plates and have differently shaped mandibular armature.

Specifics are used in the following keys. Because of this homogeneity no

subgeneral are proposed.

Ecology. — The temperate Sympotthastia species can be found in

the peneplaned or filled valleys of the foothills-piedmont within the alti-

tude of 60 to 220 m. The small streams of these lowland valleys origi-

nate in forested hills of about 350 m altitude, and they are secondarily

cutting into the floodplain, exposing rock and sand substrate. They are

moderately mineralized and relatively free of silt (and pollution). Sur-

rounding land is used for moderate agriculture, or remains forested.

44 Jan S. Doughman

Larvae can be found in crude cases in pools or in reaches with laminar

flow. Their guts contain predominantly diatoms. Flight time is from

March to June when water reaches 10-15°C. This ecological summary

was taken from Serra-Tosio (1971), Johannsen (1937), Pagast (1947),

Ferrarese and Rosaro (1981) and conversations with U.S. Geological

Survey personnel and with David Lenat of the North Carolina Depart-

ment of Natural Resources and Community Development.

KEYS TO MALES, PUPAE AND LARVAE OF SYMPOTTHASTIA

MALES

1. AnP awl-like, bare, 60-90 um long. Capitellum dusky. Palpal 3d segment

without a keel of dark setae {fulva-gp.) 2

AnP absent or conical, hairy and one or two spines apically (Figs. 9-1 1).

Capitellum clear. Palpal 3d segment expanded bearing a keel of dark

setae (zavreli-gp.) 3

2. AR 2.4-2.6; A)3800 um long. LR 73, 49, 53; P3 with ta4 110 um and

shorter than ta Palpal length 500 um; palp to face about 85. Clypeal

setae about 15. SVo straight distally (Fig. 13). Body and wing about 3 mm

S.fulva (Jon.)

AR 2.7-2.9; A]3995 um long. LR 88, 54, 52; P3 with ta4 170 um and

longer than tay Papal length 850 um; palp to face 87-98. Clypeals 12-15.

SVo sigmoid (Fig. 3). Body 5-6 mm; wing 3.6 mm . . . . S. diastena (Subl.)

3. AnP absent or minuscule and hairy. Clypeal setae absent. LR 81, 49, 59.

AR 2.4-3.0; A|3 1150 um long. Palp to face 67-72. VR 85-92. SVo near

Fig. 12, but lacking spinulae. Body 4-6 mm; wing 3-4 mm . . .S. zavreli Pag.

AnP conical, hairy, and stong spine(s) apically. Clypeals present. LR 75,

52, 58 or 64. VR 87-90 4

4. AnP about 35 um long with one spine apically (Figs. 7, 9, 10). Clypeals 4

or 5. Palp to face 79-85. AR 1.6-1.8; A)3 750 um long. SVo sigmoid,

narrow, and smooth (Fig. 8). P3with ta4 125 um long and LR 64. Body

4.3-4.8 mm; wing 3.5 mm S. spinifera Tosio

AnP longer with two unequal spines apically (Fig. 11). Clypeals 1 1. Palp

to face 89. AR 3.0; A 1095 um long. SVo band-like with fine spinulae

on surface (Fig. 12). P3with ta4 195 um long and LR 58. Body 5.5 mm;

wing 4.3 mm S. macrocera Tosio

PUPAE

1. L-setae of VII-VIII subequal and not darker than those on preceding

segments and L2 with 8 branches or less. D4 of III-VII larger than D5

and sometimes bifid on VII or VIII. Shagreen near apophyse of Til

weakly grouped. Thoracic spiracle (bulb) without spinulae. Coloration

yellow-brown with transparent-brown setae and spines, and muscle scars

clear or colored (fulva-gp.) 2

L-setae of VII-VIII shorter (by 1/3) and darker than those of VI and L2

with more than 8 branches. D4 of III-VII simple and subequal to D5.

Thoracic spiracle (bulb) with spinulae. Muscle scars dark (zavreli-gp) ... 3

Sympotthastia update 45

Fig. 4. 5. spinifera from Italy, pupa. Typical of genus.

2. L simple or bifid on III-VI. Muscle scars dark. Few spines on conjunc-

tiva II/III S. sp. (Saether 1969)

L3on II-VI simple. Muscle scars clear. Many spines on conjunctiva II/III

S.fulva (Joh.)

3. L3 on II-VI usually bifid or trifid and not much finer than the other

L-setae. Macrosetae of IX subequally spaced. Shagreen near apophyse of

II strongly grouped. Color pale yellow with darker areas . . S. zavreli Pag.

L on II-VI fine and usually simple, at most bifid. Anterior macroseta of

IX noticeably separate from terminal pair (Fig. 4). Shagreen near the apo-

physe of Til weakly grouped. Color of abdomen II-VIII clear-brown with

darker spines S. spinifera Tosio

S. macrocera Tosio

LARVAE

1. Ventromentum yellow and VmP absent. Mandible with subequal lateral

teeth set in as a group. Premandible with 1-2 small lateral teeth. Blade

fused to mid-A2. AS about 1.5X longer than supraanals (fulva-gp.) 2

Ventromentum smoky colored; VmP present, teardrop shaped. Mandible

normal; lateral teeth thorn-like, individual outgrowths. Premandible with

about 3 inconspicuous lateral teeth. A appearing less bifid as blade base

is nearer to apex of Ar AS 2X longer than supraanals (zavreli-gp.) 3

2. AR 2.2. Head capsule with flecks of chitinous thickening. Alaskan arctic

S. cf. fulva

AR 1.5. Head thinly sclerotized. New York S.fulva (Joh.)

3. AR 2.5. Blade reaching A . Premandible dark and with 2-3 lateral teeth.

Ventromentum standing above laterals. LL two nondenticulate plates

S. zavreli Pag.

AR 2.2. Blade reaching A4. Premandible dark and with 3-4 lateral teeth.

First pair of lateral teeth projecting above ventromentum. Two LL, each

with about 3 denticulations S. spinifera Tosio

46 Jan S. Doughman

Sympotthastia fulva (Johannsen)

Diamesa fulva Johannsen, 1921:229, orig. design, (holotype female

description).

D. {Psilodiamesa) fulva Jon., Johannsen, 1937:33-34 (description of

pupa, Fig. 91; description of larva, Figs. 92-96; keys; also noted that

male and female imago descriptions were associated with the imma-

ture stages); Pagast, 1947:51 1-512, 569 (remarked that this species lies

within Sympotthastia; Thienemann, 1952:248 (keyed it near Potthas-

tia gaedii).

Psilodiamesa fulva Jon., Johannsen and Townes, 1952:13.

D. fulva Jon., Roback, 1957:51, 53 (keys); Sublette, 1964:129-130

(color description of a female that is perhaps this species); Sublette

and Sublette, 1965:276 (distribution); Sublette, 1967:480-483 (sup-

plemented description of female and male with accurate figures of

male hypopygium).

Sympotthastia fulva (Jon.), Saether, 1969:34 (description of a probable

new species near this one); Serra-Tosio, 1971:281 (placement within

genus near S. spinifera); and Hansen and Cook, 1976:142 (generic

placement).

Larva. — Head yellowish and thinly sclerotized without flecks of

reinforcing chitin. Mentum as described for genus; specifically, ventro-

mentum yellow and VmP absent. AR 1.5; A | length/ width (ALAW)=

3; antenna to mandible 6:10. Contrasted to S. zavreli below, the SI is

shorter, wider and more blade-like; SII subequal to SHI; LL two non-

denticulate plates. Premandible with 1-2 lateral teeth. Mandible as in

Potthastia spp., i.e., somewhat sickle-shaped, toothed area dark, and

four, subequal laterals set in as a group. Si with 20 finely serrated

branches, proximal ones longest. Maxilla as in Johannsen (1937: Fig.

94). Body 7 mm. Procercal height/ width (H/W)= 0.8. Seven AS, 1.5X

longer than supraanal setae. TA three-fourths the length of parapod.

Posterior parapod with the nominal 16 claws {versa Johannsen 1937:33).

Material. — USA: New York, Tompkins Co., Cascadilla Cr. at

Cornell Univ., Ithaca. Paratype female, pinned; allotype male hypopy-

gium and antenna and pupal and larval skin, Cornell Univ. no. 2326,

slide mounted (specimens squashed), and the remainder of the pinned

male was slide mounted in Euparal.

Sympotthastia cf. fulva (Joh.)

Larva (4th stadium, N=l). — Fig 6. Head rather robust, brown with

flecks of heavier sclerotization. AR 2.2; blade fused nearer to base of A2

than in fulva, resembling zavreli. Body length and color indeterminate.

Remaining characters identical to that of fulva.

Material. — USA: Alaska, Happy Valley Cr. (Sagavanirktok R.

Sympotthastia update

47

il IS V J

Fig. 5. S. zavreli from North Carolina, larval head.

Fig. 6. S. cf.fulva from Alaska, larval head.

basin) nr. Sagwon, 69°09'N 148°50'W, 7 July 1976. 4th stadium larva.

Ecology. — This specimen was drift-netted in an arctic creek with

abundant orthoclads, simuliids and baetids. The following parameters

were noted for this creek a few years earlier: June - Sept. 1972, temp, to

11°C, conductance very low, pH circumneutral, alkalinity 12-18mg/l,

discharge l-1.5m3/s (Nauman and Kernodle 1973).

Sympotthastia zavreli Pagast

Syndiamesa sp. Thienemann, 1934, in the keys of Johannsen (1937) and

Roback(1957).

Sympotthastia zavreli Pagast, 1947: 458-459 and 510-512 (description of

48 Jan S. Doughman

male with Figs. 12-15 and description of pupa); Serra-Tosio, 1968:

130-134 (redescription of male, Pl.III hypopygium); Serra-Tosio,

1971:268-274 (description of male, P1.122 hypopygium and Pl.123.1

SVo; description of female, PI. 124 caudal section, wing, and antenna;

description of pupa, Pis. 125-126 abdominal segments); Pinder, 1978:44

(key and Fig. 95B of hypopygium).

Larva (4th stadium, N=2). — Fig. 5. Head brownish-yellow. Ven-

tromentum smoky in color. VmP clear, teardrop shaped. Antennal AR

2.2-2.5; ALAW 4-5. Antenna to mandible 6:10. Labral S-setae all strong

spines. SI subequal to SII; SIII hairlike. LL two nondenticulate plates.

Premandible short and darkened distally with 2-3 inconspicuous lateral

teeth. Mandible normal with each of the four lateral teeth appearing as

individual outgrowths. Si with 12-17 nearly smooth branches. Body

about 11 mm. Color brownish. Procercal H/W= 1.0 and AS 2X longer

than supraanals. TA rounded or pointed distally and nearly half the

parapod length.

Material. — USA: North Carolina, Durham Co., simipermanent

tributary to Little R.; Wake Co., trib. to Swift Creek. Several 4th sta-

dium larvae. Both stations were sampled 5 Feb. 1980. Moore Co., Deep

Creek (Lumber R. basin) 9 Feb. 1982. One larva. All leg. D. Lenat;

collection N.C. Department of Natural Resources and Community

Development.

Remarks. — These Carolina specimens obviously do not belong to

the fulva-gp. by diagnosis since this species has ventromental plates.

Although unassociated, they fit no known species description but that

of S. zavreli (cf. Thienemann 1952).

Ecology. — The Carolina specimens were found with Paraphaeno-

cladius and Eukiefferiella (b.s.) in small Piedmont streams with sand and

gravel substrate and slow current. Thienemann (1952) stated that zavreli

was found in shallow, unshaded trout streams having slow current,

spring runs, and meadow ditches, and that diatoms were prevalent in

guts. These new Nearctic finds had consumed mostly Synedra and

Gomphonema.

Sympotthastia diastena (Sublette) comb. n.

Pseudodiamesa (P.) diastena Sublette, 1964:128, orig. design. (Fig. 7b,c

of male hypopygium and description of allotype).

P. diastena Sublette, Serra-Tosio, 1976:135 (stated placement ques-

tionable.

Sympotthastia diastena (Sublette), Dr. D. Hansen in 1973 determined

two slide mounted males.

Male (N=2).— Figs. 1-3. The holotype was decidedly smaller than

the following described specimens (measurements of holotype in paren-

Sympotthastia update

49

Figs. 7-10. S. spinifera (cf. Serra-Tosio, 1971): 7, hypopygium; 8, phallapodeme;

9-10, anal point.

Figs. 11-12. S. macrocera (cf. Serra-Tosio, 1971): 11, anal point; 12, phallapo-

deme. Labels added: MVo, median volsella, Pha, phallapodeme, SVo, supe-

rior volsella, t, tubercle of SIX, and TSa, transverse sternapodeme.

theses). Unspecified lengths are in micrometers. Body length 5.7-6.0 mm

(holotype 2.9 mm); wing 3.6 mm (2.9 mm). Coloration of head, body,

wing veins and coxae dark; haltere dark, pubescent with capitellum

somewhat lighter. Extremities of the legs and the sternites lighter.

Antennal flagellomere lengths: A, 73-80 (60): A2 12 24-27: A|3 995

(880). AR 2.7-2.9. A3 width greater than length, excepting the last

few which become squarish. Width of head at eyes 750-820. Clypeus

with 12-15 setae proximally in a staggered double row. Palpal segment

2-5 lengths: 105 (70): 220-280 (195): 250-280 (205): 295 (230). Ante-

pronotum with 14 lateral setae: dorsocentrals variable — staggered sin-

gle to triple row of 25-30 setae (holotype with a single to double row of

22 setae); prealars with a main cluster of 15 and anteriorly 7-8 isolated

setae staggering forward (holotype with only 4 or 5 isolated setae). Scu-

tellum with numerous setae. Wing VR 0.88. One specimen with a full

row of setae on R (Fig. 1).

50 Jan S. Doughman

Tibial spur lengths: P^O, P271 and 72, P362 and 96.

Spiniform setae of ta)5 often paired: P^one, P2 12-14,2-3,0,0,0, and

P3 13-16,2-5,0,0,0.

Tergite IX with each half with two groups of 6-8 moderately long setae;

anal point naked, awl-like, 70-90 long. Transverse sternapodeme thickly

rectangular. Straight coxite portion of phallapodeme, 70-170 long,

articulating with a strongly sclerotized, narrow, sigmoid SVo and a

membranous, ovate, setigerous MVo (Fig. 3). Gonocoxite, gonostylus

and other details typical for the genus, and found in the diagnostic de-

scription above.

Material. — USA: California, Marin Co., Mill Valley. 12 Apr.

1957. Leg. H. L. Mathis, light trap. Holotype, U.S. National Museum

no. 65522 (poor mount). Oregon, Benton Co., Berry Creek (Willamette

R. basin), 9 mi N of Corvallis, 60 m alt. 17-24 Mar. 1960, leg. D.

Hansen; det. Hansen, 1973. 2 males. Univ. Minn. Collection nos.

DH69-280 and -281 (in balsam?).

Remarks. — The holotype of diastena is very near fulva, but these

new specimens show diastena can be much larger and darker than fulva.

I doubt fulva will be found in the western Cordilleran. The numerous

dorsocentral setal rows and forward running prealars, and the unique

presence of setae on R4+5(Fig. 1), demonstrate that this species is the

most plesiomorphic species known in Sympotthastia. Its SVo is identi-

cal to spinifera and differs slightly from fulva.

The presence of a species different fr om fulva in the western Cordil-

leran is also demonstrated by the pupa found in Waterton National

Park, Alberta, by Saether (1969) and keyed here as Sympotthastia sp.

Also, the larva from the Alaskan arctic, S. cf. fulva, is separable. It may

be that these unconnected metamorphic stages represent a single Cor-

dilleran species.

Sympotthastia spinifera Serra-Tosio

S. spinifera Serra-Tosio, 1968:134-140, orig. design. (Figs. 1-4 of hy-

popygium; ecology); Serra-Tosio, 1971:224-227 (PL 123.2 phallapo-

deme; PL 127 hypopygium; Pis. 128-129 pupa; p. 277 ecology); Ferra-

rese and Rossaro, 1981:77-80 (larva description, Fig. 36 with mentum,

labrum, antenna, premandible, procercus; pupal description, Fig. 37).

Sympotthastia macrocera Serra-Tosio

S. spinifera forma macrocera Serra-Tosio, 1968:137-138.

Sympotthastia update

51

Fig. 13. S.fulva, phallapodeme of allotype no. 2326.

Cd

:>

Fig. 14. Proposed cladogram of Sympotthastia.

52 Jan S. Doughman

S. macrocera Serra-Tosio, Serra-Tosio, 1971:277-280 (PI. 130 phallapo-

deme, gonostylus, and ta3 ; p. 279 ecology).

Sympotthastia sp. Saether

The two pupal exuviae (Saether 1969:34) were examined from Dr.

Saether's collection. This species differs from fulva by having only a few

anteriorly pointing spines on conjunctiva 11/ III. The lengths of these

specimens were about 5 mm instead of the 8 mm reported by Saether.

Sympotthastia virendri (Singh)

This oriental species was placed in Sympotthastia by Sublette and

Sublette (1973). I am inclined to think this is not suitable, since Singh

(1958) described the species with hairy wings, pictured wing without

strong anal lobe, with acrostichals, and, in a diminutive drawing, the

SVo appears to be that possessed by the Diamesae. The number of

spermathecae was not mentioned for the female described.

ACKNOWLEDGMENTS.— I thank the curators, Drs. Q. E.

Wheeler, Paul Clausen and W. W. Wirth, for their efforts that allowed

me to see important types, North Carolina's environmental biologists,

and Drs. Bruno Rossaro and L. C. V. Pinder who loaned me Palaearc-

tic examples. Special appreciation is extended to Dr. Bernard Serra-

Tosio for allowing me to reproduce his excellent figures (Figs. 7-12).

LITERATURE CITED

Ferrarese, Uberto, and B. Rossaro. 1981. Guide per il riconoscimento delle spe-

cie animali delle acque interne italiane 12. Chironomidi, 1 (Diptera, Chiro-

nomidae: Generalita, Diamesinae, Prodiamesinae). Consiglio Nazionale

Delle Ricerche AQ/ 1/ 129. 97 pp.

Hansen, Dean C, and E. F. Cook. 1976. The systematics and morphology of

the Nearctic species of Diamesa Meigen, 1835 (Diptera: Chironomidae).

Mem. Am. Entomol. Soc. 50:1-203.

Johannsen, O. A. 1921. The genus Diamesa Meigen (Diptera: Chironomidae).

Entomol. News 32:229-232.

1937. Aquatic Diptera. Part III. Chironomidae subfamilies Tanypo-

dinae, Diamesinae, and Orthocladiinae. N.Y. Agric. Exp. Stn. Ithaca Mem.

205:1-84.

, and H. K. Townes. 1952. Tendipedidae (Chironomidae). Pp. 3-47 in

Guide to the Insects of Connecticut, Part VI. The Diptera or true flies of

Connecticut. Conn. Geol. Nat. Hist. Surv. Bull. 80.

Nauman, Jon W., and D. R. Kernodle. 1973. Field water-quality information

along the proposed trans-Alaskan pipeline corridor. U.S. Geol. Surv. Basic

Data Rep. 22 pp.

Sympotthastia update 53

Pagast, Felix. 1947. Systematik and Verbreitung der um die Gattung Diamesa

gruppierten Chironomidae. Arch. Hydrobiol. 47:435-596.

Pankratova, V. Ya. 1970. Larvae and pupae of midges of the sub-family Ortho-

cladiinae of the fauna of the USSR. Nauka (Leningrad) 702:1-345.

Pinder, L. C. V. 1978. A key to the Adult Males of the British Chironomidae

(Diptera) the non-biting midges. Freshw. Biol. Assoc. Sci. Publ. 37, 2 vols.

169 pp.

Roback, Selwyn S. 1957. The immature tendipedids of the Philadelphia area

(Diptera: Tendipedidae). Monogr. Acad. Nat. Sci. Phila. 9. 152 pp.

Saether, Ole A. 1969. Some nearctic Podonominae, Diamesinae, and Orthocla-

diinae (Diptera: Chironomidae). Bull. Fish. Res. Board Can. 770:1-154.

1977. Female genitalia in Chironomidae and other Nematocera:

Morphology, phylogenies, keys. Bull. Fish. Res. Board Can. 797:1-209.

! 1980. Glossary of chironomid morphology terminology (Diptera:

Chironomidae). Entomol. Scand. Suppl. 74:1-51.

Serra-Tosio, Bernard. 1968. Taxonomie phylogenetique des Diamesini: les

genres Pott hast ia Kieffer, Sympotthastia Pagast, Parapotthastia n.g. et

Lappodiamesa n.g. (Diptera: Chironomidae). Trav. Lab. Hydrobiol. Pisci-

cult. Univ. Grenoble 59-60: 117-164:

1971. Contribution a Tetude taxonomique, phylogenetique biogeo-

graphique et ecologique des Diamesini (Diptera: Chironomidae) d' Europe.

2 vols. These Universite Scientifique Medicale de Grenoble. 462 pp.

1976. Chironomides des Alpes: The genre Pseudodiamesa (Diptera:

Chironomidae). Trav. Sci. Pare Natl. Vanoise 7:117-138.

Simpson, Karl W., and R. W. Bode. 1980. Common larvae of Chironomidae

(Diptera) from New York State streams and rivers. N.Y. State Mus. Bull.

439:1-105.

Singh, Santokh. 1958. Entomological Survey of the Himalaya. Part 29. Proc.

Natl. Acad. Sci. India (B) 25:308-314.

Sublette, James E. 1964. Chironomid midges of California. II. Tanypodinae,

Podonominae, and Diamesinae. Proc. U.S. Natl. Mus. 775:85-135.

1967. Type specimens of Chironomidae (Diptera) in the Cornell

University Collection. J. Kans. Entomol. Soc. 40:477-564.

, and M. S. Sublette. 1965. Family Chironomidae (Tendipedidae). Pp.

142-181 in A. Stone, et al. A Catalogue of the Diptera of America North of

Mexico. Agric. Res. Serv. Agric. Handb. 176. Washington, D.C. 1,696 pp.

1973. Family Chironomidae. Pp. 389-422 in M. D. Delfinado and

D. E. Hardy (Eds.). A Catalog of the Diptera of the Oriental Region, Vol. 1.

Univ. Hawaii Press.

Thienemann, August. 1952. Bestimmungstabelle fuer die Larven der mit Diamesa

naechst verwandten Chironomiden. Beitr. Entomol. 2:244-256.

Accepted 10 May 1984

54

ATLAS OF NORTH AMERICAN FRESHWATER FISHES

by

D. S. Lee, C. R. Gilbert, C. H. Hocutt, R. E. Jenkins,

D. E. McAllister, J. R. Stauffer, Jr., and many collaborators

This timely book provides accounts for all 777 species of fish

known to occur in fresh waters in the United States and Canada. Each

account gives a distribution map and illustration of the species, along

with information on systematics, distribution, habitat, abundance, size,

and general biology.

". . . represents the most important contribution to freshwater

fishes of this continent since Jordan and Evermann's 'Fishes of North

and Middle America' over 80 years ago." — Southeastern Fishes Coun-

cil Proceedings.

1980 825 pages Indexed Softbound ISBN 0-917134-03-6

Price: $25, postpaid. North Carolina residents add 4l/2% sales tax. Please make

checks payable in U. S. currency to NCDA Museum Extension Fund.

Send to FISH ATLAS, N. C. State Museum of Natural History,

P. O. Box 27647, Raleigh, NC 27611.

ATLAS OF NORTH AMERICAN FRESHWATER FISHES

1983 SUPPLEMENT

by

D. S. Lee, S. P. Platania, and G. H. Burgess

The 1983 supplement to the 1980 Atlas of North American Fresh-

water Fishes treats the freshwater ichthyofauna of the Greater Antilles.

In addition to this bound supplement, there are 19 accounts, mostly

species not described in 1980, in looseleaf form to be added to the 1980

volume. Illustrated by Renaldo Kuhler.

1983 67 pages Indexed Softbound

Price: $5, postpaid. North Carolina residents add 4'/2% sales tax. Please make

checks payable in U. S. currency to NCDA Museum Extension Fund.

Send to FISH ATLAS, N. C. State Museum of Natural History,

P. O. Box 27647, Raleigh, NC 27611.

Genetic Variation in the Eastern Cottonmouth, Agkistrodon

piscivorus piscivorus (Lacepede) (Reptilia: Crotalidae) at

the Northern Edge of its Range

Donald A. Merkle

Department of Natural Sciences, Longwood College,

Farmville, Virginia 23901

ABSTRACT. — Genetic variation was examined by electrophoresis in

six populations of Agkistrodon p. piscivorus from the northern edge of

the species' range in southeastern Virginia. Twenty-three presumptive

loci were found to be monomorphic, while three loci were polymorphic

in some populations. Average observed heterozygosity values ranged

from 0.9% to 2.9%, with a mean for all populations of 1.6%. Nei's

index of genetic identity reveals that all Virginia populations sampled

have a very high degree of genetic similarity, with a minimal value of

.964.

INTRODUCTION

Advances in biochemical techniques have enabled researchers to

collect a great deal of information on the amount of genetic variation

that occurs in natural populations (Selander and Johnson 1973). Most

of the studies dealing with snakes have involved the family Colubridae

(Sattler and Guttman 1976; Gartside et al. 1977). Those studies that

included species of the family Crotalidae analyzed only components of

the venom.

The Eastern Cottonmouth, Agkistrodon piscivorus piscivorus

(Lacepede), occurs primarily in the Coastal Plain of the eastern United

States, where it ranges from Alabama to its northern distributional lim-

its in the southeastern corner of Virginia. For years, the James River

was thought to mark the northernmost limit for this species, just as it

does for a number of other reptiles and amphibians (Wood 1954;

Conant 1975). Cottonmouths, however, occur north of the James River

in the vicinity of the Newport News-Hampton area (Engeling 1969; Lin-

zey and Clifford 1981).

The distribution of the cottonmouth in Virginia is discontinuous.

There is an isolated population along Swift Creek and the Appomattox

River just west of Hopewell in Chesterfield County. "This population is

at least 60 km from the nearest known locality in the main part of the

range. . ." (Blem 1981:117). Populations of this species on the barrier

beaches along Back Bay are separated from the mainland by several

miles of water. Other populations, such as that at Sea Shore State Park,

Brimleyana No. 1 1:55-61, October 1985 55

56 Donald A. Merkle

are essentially isolated by "urban sprawl". Thus, there appear to be a

number of potential barriers to gene flow in populations of this species

in Virginia.

Several other species of snakes whose northern limits are essentially

restricted to the Coastal Plain of southeastern Virginia also occur on the

peninsula between the James and York rivers. They include the Timber

Rattlesnake, Crotalus horridus; the Brown Water Snake, Nerodia taxis-

pilota; the Redbelly Water Snake, Nerodia erythrogaster erythrogaster,

and the Glossy Crayfish Snake, Regina rigida rigida (Linzey and Clif-

ford 1981).

The distribution of such populations is open to two interpretations.

The first is to assume that populations occurring north of the James

River were established by animals that crossed the river in the same area

despite its width (approximately 6 km). The second hypothesis is to

assume that all of these species once occurred throughout the Coastal

Plain of southeastern Virginia. This would enable colonizers to invade

areas north of the James by crossing in the general vicinity of the Fall

Line, where the river is a less formidable barrier. Subsequent elimina-

tion of intervening populations would then produce the distributional

patterns currently observed. The potential isolation of populations

could result in genetic differentiation due to lack of gene flow.

In this study, I used electrophoresis to examine genetic variation in

several populations of A. p. piscivorus to determine: 1) levels of genetic

variation that might accompany potential isolation; and 2) if patterns of

genetic variation reveal any insights into the origin of populations of A.

piscivorus north of the James River.

MATERIALS AND METHODS

Specimens collected in the field from May 1980 until September

1982 were returned to the laboratory and sacrificed by freezing. Attempts

were made to collect specimens from every drainage system in Virginia

where the species has been recorded. Localities where snakes used for

this study were collected (Fig. 1) are: 1) Newport News Reservoir, New-

port News; 2) Appomattox River near Hopewell; 3) Sea Shore State

Park, Virginia Beach; 4) Gum Swamp near North Landing River, Vir-

ginia Beach; 5) Northwest River, Chesapeake; and 6) False Cape State

Park, Virginia Beach.

Prior to electrophoresis, specimens were thawed and extracts of

soluble proteins prepared by homogenizing samples of heart, liver, and

skeletal muscle with equal volumes of 2% 2-phenoxyethanol. Homoge-

nates were centrifuged at 4,000 G for 30 minutes. Supernatants were

then decanted and used for electrophoresis.

Genetic Variation in Cottonmouth 57

Fig. 1. Populations of A. piscivorus sampled for electrophoretic study. 1 = New-

port News, 2 - Hopewell, 3 r Sea Shore State Park, 4 = Gum Swamp, 5 =

Northwest River, 6 = False Cape State Park.

Starch gel electrophoretic techniques using the basic procedures of

Selander et al. (1971) were used with the following modifications:

albumin, hemoglobin, general proteins, and esterases were best demon-

strated on the Poulik buffer, while all other enzymes were isolated on

gels using the Tris-Maleic acid buffer. All gels were 12.5% starch (Elec-

trostarch Lot 307, Otto Hiller, Madison, Wisconsin).

Isozymes of various proteins were designated in order of decreasing

anodal mobility. Alleles present at polymorphic loci were designated

alphabetically by a superscript following the locus designation. Nei's

(1972) index of genetic identity was used to compare genetic similarities

between populations, and mean heterozygosity per individual (H) was

calculated for each population.

RESULTS

Of the 29 protein loci examined in this study, 23 were found to be

monomorphic in all individuals examined. These loci included: albumin,

hemoglobin, three general proteins, three esterases, two malate dehy-

drogenases, two lactate dehydrogenases, two phosphoglucomutases, one

phosphoglucose isomerase, a-glycerophosphate dehydrogenase, gluta-

mate dehydrogenase, two superoxide dismutases, isocitrate dehydroge-

nase, 6-phosphoglucose dehydrogenase, sorbital dehydrogenase, and

glutamate oxaloacetate transaminase (Got-1). Only three loci were poly-

morphic; their gene frequencies are listed in Table 1.

58

Donald A. Merkle

Table 1

Allelic frequencies at polymorphic loci in_Virginia populations of

Agkistrodon piscivorus. N = sample size; H = mean heterozygosity

per individual.

Variability at Polymorphic Loci

Got-2:

Xdh-1

Lap-1

This locus was polymorphic in all populations, with the fre-

quency of Got-2a ranging from .111 in the Hopewell popula-

tion to .400 in the False Cape State Park population.

Two alleles were present in all populations, with the excep-

tion of Sea Shore State Park where only Xdh-lb was present.

Xdh-lb was the predominant allele in the False Cape State

Park and Northwest River populations, while Xdh-la pre-

dominated in the populations from Hopewell, Newport News,

and Gum Swamp.

Most individuals were monomorphic for Lap-lb, but one

individual from both Newport News and Hopewell popula-

tions possessed Lap-la in the heterozygous condition.

Observed mean individual heterozygosity values (H) ranged from a

low of 0.9% in the Sea Shore State Park population to a high of 2.9% in

the population at False Cape State Park. The mean heterozygosity per

locus per individual for all populations was 1.6%. Nei's (1972) index of

genetic identity values for all pairings is presented in Table 2.

Table 2. Nei's index of genetic identity values between populations of Agkis-

trodon piscivorus from Virginia.

Genetic Variation in Cottonmouth 59

DISCUSSION

The average heterozygosity value of 1.6 reported here for Virginia

populations of A. piscivorus is less than that of 4.1% - 8.3% reported for

Thamnophis sirtalis by Sattler and Guttman (1976), and the 9.2% and

7.7% reported for Thamnophis proximus and Thamnophis sauritus,

respectively, by Gartside et al. (1977). It is interesting to note the

extremely low heterozygosity in the Sea Shore State Park population

(0.9%). Virginia populations of A. piscivorus appear to exhibit less vari-

ation than other species of snakes. Two specimens of the Florida Cot-

tonmouth, Agkistrodon piscivorus conanti, were found to be almost

identical to Virginia specimens based on electrophoretic analysis of the

same loci. Only the presence of a unique allele at the Got-1 locus distin-

guished it from Virginia A. piscivorus. Since there is so little variation

between these two subspecies, the low amount of variation and hetero-

zygosity reported in this study may be typical for the species. Thus, A.

piscivorus may have had a rather conservative biochemical evolution, if

we can draw inferences from data at hand.

The highest level of heterozygosity was observed in the population

at False Cape State Park (2.9%), while the lowest was seen in the Sea

Shore State Park sample (0.9%). This population is unique in that hous-

ing developments separate it from other populations. The surrounding

area is one of the most rapidly developing areas in the United States.

Although the park once supported high-density populations of cotton-

mouths, the species has become extremely rare there in recent years.

Despite extensive field efforts over a two-year period, only two speci-

mens were collected. This apparent decline may be an effect of the

extensive droughts this area has suffered during the past few years.

Nei's (1972) index of genetic similarity reveals all populations are

extremely close genetically. Even the lowest pairing value (.964), obtained

in comparing Sea Shore State Park with the Newport News population,

indicates a very high degree of genetic similarity among all Virginia

populations sampled. Although all Virginia populations of A. piscivorus

sampled are very similar genetically, the Hopewell, Gum Swamp, and

Newport News populations show the highest values for genetic similar-

ity. All three displayed a very high frequency for Xdh-la, which occurred

in much lower frequencies in the other populations and was absent from

the Sea Shore State Park samples. While the Newport News population

shared a slightly higher identity with the Gum Swamp population than

with the Hopewell population (.99880 vs .99826), this extremely small

difference can be explained by the very low frequency of Got-2a (.111)

and higher frequency of Xdh-la (.944) in the Hopewell population. Both

of these loci appear to be heading towards complete fixation of alleles,

either by selection or by drift, in this isolated population. Such patterns

60 Donald A. Merkle

have been observed in isolated populations of other species (Avise

1976). Fixation of Xdh-lb in the Sea Shore State Park population also

appears to be in progress. Additionally, the presence of Lap-la in only

the Hopewell and Newport News populations indicates the close genetic

relationship between them. This suggests that the populations of A. pis-

civorus north of the James River are derived from populations in the

vicinity of Hopewell rather than farther down the river. This hypothesis

is supported by other species of snakes with distributions similar to that

of the cottonmouth. Nerodia taxispilota occurs on the Peninsula, but

does not extend to its lower end. The single Virginia record for Regina

r. rigida is also from the uppermost reaches of the Peninsula. Whether

this represents a relict population or a short-lived introduction is prob-

lematical. While these species would be able to cross the James River at

its widest expanse, they are absent along the river in its lower reaches.

The distribution of the species least likely to cross large expanses of

water, C. horridus, lends further credence to the hypothesis. The timber

rattlesnake has a more extensive distribution on the Peninsula than does

the cottonmouth, but it also is not recorded from the counties directly

below the James River. Yet there are records for this species from

Prince George County, very near the Hopewell population of A. pisciv-

orus. Both N. taxispilota and N. e. erythrogaster also persist in this

same general region.

The elimination of intervening populations for these species may be

a result of climatic factors, at least in the case of A. piscivorus. It

appears that the distributional range of this species in Virginia is con-

tracting. Richard Hoffman (in Russ 1973) reported that cottonmouths

were once common east of the Fall Line in Virginia, but are now only

rarely found in many areas. Today, even in the Dismal Swamp, the

species is so rare that the Park Service will not issue permits to collect

this species. Many areas that once supported large populations no

longer do so, and it appears that the range of this species is being

pushed to the southeast. Blem (1981) felt that the record cold winters

during the last decade may have had a decimating effect on the survival

of this species at the northern edge of its range.

In summary, it appears that the levels of genetic variation observed

in six Virginia populations of A. piscivorus are lower than those

reported for other snakes. This species and several others of the Coastal

Plain probably had more extensive distributions in the past, but appear

now to be undergoing range contractions. Virginia cottonmouth popu-

lations that still seem to be thriving also display the highest genetic

variation.

Genetic Variation in Cottonmouth 61

ACKNOWLEDGMENTS.— I would like to thank the following

individuals for assistance in the field: Chris Pague, Charles Blem, Gary

Williamson, John Foster, Joe Mitchell, Bob Bader, Charles Hooper,

David Breil, and Costello Craig. The assistance of authorities at New-

port News Reservoir Park, Sea Shore State Park, False Cape State

Park, and Northwest River Park is greatly appreciated. This research

was partly funded by a grant from the Longwood College Faculty

Research Committee.

LITERATURE CITED

Avise, John C. 1976. Genetic differentiation during speciation. Pp. 106-122 in

Molecular Evolution. F.J. Ayala (Ed.). Sinauer Assoc, Sunderland, MA.

Blem, Charles R. 1981. Reproduction of the Eastern Cottonmouth Agkistro don

piscivorus piscivorus (Serpentes: Viperidae) at the northern edge of its

range. Brimleyana 5:117-128.

Conant, Roger. 1975. A Field Guide to Reptiles and Amphibians of Eastern and

Central North America. 2nd ed. Houghton Mifflin Co., Boston. 329 pp.

Engeling, Glen A. 1969. Reptiles and amphibians of York Co., and the

Newport-Hampton area. Va. Herpetol. Soc. Bull. (52:1-3.

Gartside, Donald F., J. S. Rogers and H. C. Dessauer. 1977. Speciation with

little genie and morphological differentiation in the ribbon snakes Tham-

nophis proximus and T. sauritus (Colubridae). Copeia 1977(4):697-705.

Linzey, Donald W., and M. J. Clifford. 1981. Snakes of Virginia. Univ. Virginia

Press, Charlottesville. 159 pp.

Nei, Masatoshi. 1972. Genetic distance between populations. Am. Nat.

706:283-292.

Russ, William P. 1973. The rare and endangered terrestrial vertebrates of Vir-

ginia. Unpubl. M.S. thesis, Va. Polytech. Inst. State Univ. 339 pp.

Sattler, Paul W., and S. I. Guttman. 1976. An electrophoretic analysis of

Thamnophis sirtalis from western Ohio. Copeia 1976(2):352-355.

Selander, Robert K., and W. E. Johnson, 1973. Genetic variation in vertebrate

species. Annu. Rev. Ecol. Syst. 4:75-91.

Selander, Robert K., M. H. Smith, S. Y. Yang, W. E. Johnson and J. B. Gen-

try. 1971. Biochemical polymorphisms and systematics in the genus Pero-

myscus. I. Variation in the old-field mouse Peromyscus polionotus. Stud.

Genet. VI. Texas Univ. Publ. 7103:49-90.

Wood, John T. 1954. The distribution of the poisonous snakes in Virginia. Va.

J. Sci. 5:152-167.

Accepted 4 June 1984

62

A DISTRIBUTIONAL SURVEY

OF NORTH CAROLINA MAMMALS

by

David S. Lee, John B. Funderburg, Jr., and Mary K. Clark

This book lists all the mammals of North Carolina and offers spe-

cies accounts and range maps for all of the non-marine species. Intro-

ductory chapters describe the plant communities of the state as they

relate to mammal distribution and discuss local zoogeographic patterns.

1982 72 pages Soft bound

Price: $5, postpaid. North Carolina residents add 4!/2% sales tax. Please make

checks payable in U. S. currency to NCDA Museum Extension Fund.

Send to MAMMAL BOOK, N. C. State Museum of Natural History,

P. O. Box 27647, Raleigh, NC 27611.

Seasonal Weight Changes in

Raccoons (Carnivora: Procyonidae)

of North Carolina

Samuel I. Zeveloff1 and Phillip D. Doerr

Department of Zoology,

North Carolina State University,

Box 7617, Raleigh, North Carolina 27695-7617

ABSTRACT. — Raccoons, Procyon lotor L., were studied in North

Carolina to determine if seasonal changes in body weight occur in this

species in a mid-latitude region. The sample consisted of live-trapped

animals and intact carcasses from a fur buyer. Juvenile male body

weights increased from the end of July to mid-December 1975. Body

weights of juvenile males and females tended to decline between mid-

December 1975 and February 1976. During the midwinter of both

1975 and 1976, adult male and female body weights decreased; the

decline was less extreme and occurred about two weeks later than

declines reported for the northcentral United States. This pattern of

weight loss at higher latitudes may reflect the greater energetic cost of

raccoon winter foraging at northern locations. Such sites experience

lower temperatures than more southerly sites, and their ground vegeta-

tion is less accessible because of deeper snows.

INTRODUCTION

Systematic seasonal changes in body weight occur in a variety of

nonhibernating mammals (for examples see Keller and Krebs 1970;

Markham and Whicker 1973; Iverson and Turner 1974; Mautz 1978).

This is not surprising since food quality and/ or quantity normally vary

within a year. Except for those of Iverson and Turner (1974) and Mautz

(1978), most reports of such seasonal weight changes are presented

without ecological explanations, or as events associated with population

cycles (e.g., Keller and Krebs 1970). In this study we examined raccoons,

Procyon lotor L., in North Carolina to determine whether such changes

occur in this species in a mid-latitude region. Seasonal body weight

changes have been documented in raccoons from the northern (Stuewer

1943a; Mech et al. 1968) and southern (Johnson 1970) United States.

Thus, we could also evaluate geographic variation in this phenomenon.

METHODS

This study was conducted in the North Carolina piedmont from

January 1975 through February 1976. Study areas included the Ecologi-

1 Present address: Department of Zoology, Weber State College, Ogden, Utah 84408.

Brimleyana No. 1 1:63-67, October 1985 63

64 Samuel I. Zeveloff and Phillip D. Doerr

cal Research Area of North Carolina State University (NCSU), 3.2 km

southwest of Raleigh, and woods in two nearby locations: Schenck

Forest and the Faculty Club, both of NCSU.

Live traps baited with sardines and corn were placed along stream

bottomlands and checked daily. Twenty-two captured raccoons were re-

strained in a wire cone (Stuewer 1943b), weighed (±1.0 g), measured

(±0.5 mm), sexed (Stuewer 1943b), ear tagged, and released. Additional

data were obtained from 123 intact carcasses from a fur buyer in Smith-

field, 48.3 km south of Raleigh. All of these raccoons were trapped

within a 150 km radius of Smithfield less than two weeks prior to exam-

ination (L. Barbour, pers. comm.). Exact capture dates were not known

for about 75% of the specimens. As approximations used to compute

regressions of body weight over time, dates for these animals were esti-

mated to be seven days prior to the date of necropsy.

The dead raccoons were aged by degree of epiphyseal closure seen

in X-rayed radii and ulnae (Sanderson 1961). We analyzed the data

from the noncastrated raccoons in Sanderson's (1961:9) Illinois sample

and found that the mean ages of females with broad, thin, and closed

epiphyses were all significantly different (P<0.01) with little overlap

between the means' 95% confidence limits. Since these means corre-

sponded nicely with year classes, we categorized our females with broad

epiphyses as juveniles (— zero-one yr), with thin epiphyses as subadults

(^ one-two yr), and with closed epiphyses as adults (^ two yr). In

Sanderson's study, the only significant difference (P<0.01) was between

the mean ages of males with broad and thin epiphyses. Again, matching

epiphyseal closure with mean age, our males were classified as juveniles

(=* zero-one yr) or adults (^ one yr). Fifty raccoons, including those

live-trapped, were not X-rayed. Instead, they were assigned to age

classes by comparing their total lengths and body weights with the

means and 95% confidence intervals of these variables displayed by the

animals we did X-ray.

Weights of males and females were plotted separately against dates

by age. We noted that weight typically increases during the fall or early

winter through December, and is followed by a midwinter decrease.

Weights of juvenile males increased by midsummer. Stepwise regression

analyses were used to determine if these visually observed changes were

statistically significant.

RESULTS AND DISCUSSION

Although adult weights decreased during January and February of

1975 and 1976, the only statistically significant change was for males in

1975 (r = -0.45; P<0.05; N = 20). However, since the pattern was con-

sistent in both sexes each year, it appeared real and justified data pool-

Raccoon Seasonal Weight Changes 65

ing. Slopes and intercepts of the descending lines of adult male body

weight in January and February were similar in 1975 and 1976 (P>0.05,

F-test); the same was true of adult female body weights. The negative

slope of the pooled male sample was significant (r = -0.45; P = 0.01; N =

30) and the pooled female data also provided a clearly decreasing body

weight trend (r = -0.44; P = 0.08; N = 18). These adult weight decreases

in January and February were compared between sexes (F-test). Slopes

were similar (P>0.05) but intercepts were not (P<0.01), with males

being typically heavier.

In Minnesota, Mech et al. (1968) reported a 50% weight decline in

adult raccoons from late November through mid-March followed by a

weight gain beginning in mid-April. Adult raccoon weights were also

found to be minimal in the spring in Michigan (Stuewer 1943a) and

Alabama (Johnson 1970). In Alabama, however, spring weights were

only about 20% less than those in the fall. We also found this midwinter

weight decline to be less extreme in North Carolina. In the pooled sam-

ple regressions, expected values of adult weight decreased by 22% for

males and 27% for females through January and February. Further, this

winter decline in adult raccoon weights begins about two weeks later in

North Carolina than it does in the northcentral states (see Stuewer

1943a; Mech et al. 1968).

Body weights of juvenile male raccoons increased from the end of

July to mid-December 1975 (r = 0.71; P<0.05; N = 12), followed by an

insignificant decreasing trend through February 1976 (r = -0.43; P>0.05;

N = 5). Juvenile female weights also declined from mid-December 1975

through February 1976 (r = -0.65; P = 0.06; N = 9). Similarly, both

Stuewer (1943a) and Mech et al. (1968) found that body weights of

juvenile raccoons in Michigan and Minnesota increased until November

of their first year and then declined. We also observed yearling female

weights to increase from November 1975 through January 1976 (r =

0.85; P = 0.08; N = 5), another pattern consistent with those found

farther north by these authors.

Iverson and Turner (1974) suggested that mammals lose weight

when it is adaptive to lessen energy requirements in certain seasons.

However, a weight decrease normally occurs by fat loss, which need not

imply lower energy requirements. Energy demands could even be rela-

tively high in fat depleted individuals. For example, raccoon weights are

lowest by winter's end, a time when energy requirements for upcoming

breeding events should be high.

We suggest a less complicated explanation, which is simply that

animals put on weight while food is readily available to prepare them

for the harsher winter and early spring. Mautz (1978) argued that white-

tailed deer, Odocoileus virginianus, add fat in summer and fall to offset

66 Samuel I. Zeveloff and Phillip D. Doerr

the lower nutritive value of winter browse. Both situations are, of

course, analogous to that occurring in many hibernators (e.g., Davis

1976). In raccoons, fat stores would become depleted in winter because

of (1) higher metabolic costs of staying warm, (2) more energetically

expensive foraging associated with greater reliance on predation (John-

son 1970), which is a more active type of foraging, and (3) reduced

overall food intake because of sparser food sources. Northern latitudes,

with colder temperatures and vegetation made less accessible by deeper

snows, should be the most energetically demanding places for raccoons

in the winter. This would explain the latitudinal differences in adult

weight loss. Other comparisons support this explanation. The periods of

yearling weight increase and adult weight decrease occur later in North

Carolina than in the northcentral states and are less extreme. Where

winter arrives later and is not as harsh, weight gains need not occur as

early to ensure survival. Instead, energy stores that last longer into the

winter should delay weight loss.

Winter foods of raccoons might be of low nutritive value. But,

since raccoons are extremely omnivorous, this should be less a factor in

their weight changes than for those in obligatory herbivores like deer

(see Mautz 1978). Furthermore, since raccoons exhibit denning behav-

ior and often forage at night, when winds are typically less severe, they

might have lower winter energy demands per unit weight than deer.

Certainly, other factors such as the insulative quality of the fur and

physiological adaptations also need to be considered for a valid compar-

ison to be made.

Finally, breeding activity in the later winter and early spring might

influence the pattern of adult raccoon weight decrease by causing a

further drain on fat stores. In a separate study (Zeveloff and Doerr

1981) we found a high negative correlation between mean body weights

and mean testis weights of 16 adult male raccoons from mid-January

through February 1976 (r = -0.99; P<0.01). This indicates that male

body weight decreases at a time of increased reproductive activity. To

fully understand raccoon body weight dynamics, one might consider

their reproductive events in addition to geographic variation in climatic

seasonality and food availability.

ACKNOWLEDGMENTS.— R. A. Lancia, R. J. Monroe, L. C.

Ulberg, H. Underwood and J. R. Walters of NCSU and J. E. Cooper of

the North Carolina State Museum of Natural History provided useful

criticisms of this work. V. C. Schmidt, NCSU, helped prepare materials

for X-raying. L. Barbour of Smithfield, NC, permitted us to inspect

raccoon specimens at his fur business. This is paper number 857 1 of the

Raccoon Seasonal Weight Changes 67

Journal Series of the North Carolina Agricultural Research Service,

Raleigh, NC 27695. Use of trade names does not imply endorsement.

LITERATURE CITED

Davis, David E. 1976. Hibernation and circannual rhythms of food consump-

tion in marmots and ground squirrels. Q. Rev. Biol. 51:477-514.

Iverson, Stuart L., and B. N. Turner. 1974. Winter weight dynamics in Microtus

pennsylvanicus. Ecology 55:1030-1041.

Johnson, A. Sydney. 1970. Biology of the raccoon (Procyon lotor varius Nelson

and Goldman) in Alabama. Auburn Univ. Agric. Expt. Sta. Bull. 402:1-148.

Keller, Barry L., and C. L. Krebs. 1970. Microtus population biology; III.

Reproductive changes in fluctuating populations of Microtus ochrogaster

and Microtus pennsylvanicus in southern Indiana, 1965-1967. Ecol. Monogr.

40:263-294.

Markham, O. D., and F. W. Whicker. 1973. Seasonal data on reproduction and

body weights of pikas (Ochotona princeps). J. Mammal. 54:496-498.

Mautz, William W. 1978. Sledding on a bushy hillside: the fat cycle in deer.

Wildl. Soc. Bull. 5:88-90.

Mech, L. David, D. M. Barnes and J. R. Tester. 1968. Seasonal weight changes,

mortality, and population structure of raccoons in Minnesota. J. Mammal.

49:63-73.

Sanderson, Glen C. 1961. Techniques for determining age of raccoons. 111. Nat.

Hist. Surv. Div. Biol. Notes 45:1-16.

Stuewer, Frederick W. 1943a. Raccoons: their habits and management in Mich-

igan. Ecol. Monogr. 13:203-257.

1943b. Reproduction of raccoons in Michigan. J. Wildl. Manage.

7:60-73.

Zeveloff, Samuel I., and P. D. Doerr. 1981. Reproduction of raccoons in North

Carolina. J. Elisha Mitchell Sci. Soc. 97:194-199.

Accepted 10 June 1984

68

FISHERMAN'S GUIDE

FISHES OF THE SOUTHEASTERN UNITED STATES

by

Charles S. Manooch, III

Illustrated by Duane Raver, Jr.

Remarkable for its breadth of coverage, this book details the hab-

its, range, and appearance of more than 250 species offish, 150 of which

are illustrated in color. Each account also includes tips on catching the

fish and preparing it for the table.

Manooch is experienced in the field of Fisheries Management and

Technology, and Raver is nationally known for his paintings of wildlife.

"An excellent general reference book for scientific or non-scientific

audiences. ... It contains information not easily found in any other

source." — Carter R. Gilbert, Curator of Fishes, Florida State Museum.

1984 376 pages Index Bibliography ISBN 0-917134-07-9

Price: $24.95, plus $1.25 for shipping. North Carolina residents add 4]/2% sales

tax. Please make checks payable in U. S. currency to NCDA Museum

Extension Fund.

Send to FISHERMAN'S GUIDE, N. C. State Museum of Natural

History, P. O. Box 27647, Raleigh, NC 27611.

Age, Growth, Food Habits, and Reproduction of the

Redline Darter Etheostoma rufilineatum (Cope)

(Perciformes: Percidae)

in Virginia

James C. Widlak and Richard J. Neves

Virginia Cooperative Fishery Research Unit1,

Department of Fisheries and Wildlife Sciences,

Virginia Polytechnic Institute and State University

Blacksburg, Virginia 24061

ABSTRACT. — Life history aspects of Etheostoma rufilineatum, the

redline darter, were investigated from May 1981 to May 1982 in the

North Fork Holston River, Virginia. Analysis of scale samples indi-

cated that males and females grew at approximately the same rate, but

males reached a greater maximum length. Estimated annual survival

rate for age II and older males was 0.28 and for females 0.03. Aquatic

insect larvae were the major food items, and dipterans predominated

numerically (68-87%) year-round. Feeding over a 24-hour period

peaked from early to late afternoon (1600-2000 hr). The sex ratio

favored males throughout the year (2.5:1) and was attributed to differ-

ential survival and distribution. Age I fish more than 42 mm long, of

both sexes, were sexually mature. Ripe males were first collected in

March, although spawning coloration was evident in December. Female

ovaries began maturing in late February, and spawning occurred from

mid-May to mid-August.

INTRODUCTION

The darters are small members of the family Percidae and consti-

tute a diverse group of North American fishes, with 145 species in 3

genera and 28 subgenera (Collette 1967; Page 1983). They reach maxi-

mum lengths of 35-200 mm and differ widely in morphological charac-

teristics, habitat preference, and behaviors. A comprehensive review of

the biology and ecology of darters was published by Page (1983).

The redline darter, Etheostoma rufilineatum (Cope), is one of 13

species in the subgenus Nothonotus, one of the more gaudy groups of

darters. Nothonotus species exhibit strong sexual dimorphism. Males of

most species display brilliant coloration during most of the year. The

species occur in riffle habitats of clear upland streams. Most occupy the

'The Unit is jointly supported by the U. S. Fish and Wildlife Service, the

Virginia Commission of Game and Inland Fisheries, -and Virginia Polytechnic Insti-

tute and State University.

Brimleyana No. 11:69-80. October 1985 69

70 James C. Widlak and Richard J. Neves

Ohio River basin, but one species occurs in direct tributaries of the

lower Mississippi River, one in the Mobile Bay drainage, and two in the

Ozarks. Of the 13 species, life history studies have been conducted only

on Etheostoma acuticeps Bailey, the sharphead darter (Jenkins and

Burkhead 1975; Bryant 1979) and Etheostoma maculatum Kirtland, the

spotted darter (Raney and Lachner 1939), and comparative ecological

studies were done on 3 species (Stiles 1972).

The redline darter occurs in tributaries of the Tennessee and Cum-

berland rivers in Virginia, North Carolina, Tennessee, Kentucky, Geor-

gia, Mississippi, and Alabama (Etnier 1980). It is found in swift, shallow

riffles of clear streams and may exist in riffles shallower than those pre-

ferred by other Nothonotus species (Zorach 1970). Except for studies on

systematics (Zorach 1970) and breeding and food habits of three

Nothonotus species (Stiles 1972; Bryant 1979), no biological informa-

tion is available on the redline darter. The present study was conducted

to describe age, growth, food habits, and spawning of a population in

Virginia.

MATERIALS AND METHODS

Study Area

Field sampling was conducted on the North Fork Holston River, a

fourth-order stream in the Ridge and Valley Province of southwestern

Virginia. The study site was a 350 m section at River Mile 86.9

(36°55'N, 81°40'W), about 8 km upstream from Saltville, Smyth

County. The river at this site averages 29 m wide and consists primarily

of riffle habitat with cobble and boulder substrate. Water temperatures,

recorded daily with a Ryan 30-day thermograph, ranged from 1°C in

February to 29° C in July. Water quality characteristics, collected

monthly and analyzed with a Hach DR-EL/ 1 field kit, are summarized

in Table 1. Detailed water quality data for the North Fork Holston

River were compiled by Poppe (1982). A total of 41 fish species occurs

at the study site (Widlak 1982).

Fish Collections

Redline darters were collected twice monthly from May to August

1981, and monthly from September 1981 to May 1982. Sampling was

done exclusively with a Coffelt BP-1 backpack electroshocking unit

with direct current output, and dip nets. Waterscopes were used to facil-

itate sighting of darters during high water levels. All habitats were elec-

trofished to obtain representative samples of redline darters, and an

attempt was made to collect at least 10 darters on each sampling date.

Seventeen to one-hundred twenty-six specimens were collected per sam-

ple during summer and fall sampling, but only three to seventeen were

collected during the winter months because of high water levels or ice

conditions. Specimens were placed on ice to reduce regurgitation of

Biology of Etheostoma rufilineatum 1 1

Table 1. Water chemistry characteristics collected monthly at North Fork Hol-

ston River Mile 86.9, January 1981-March 1982.

stomach contents before preservation in 10% buffered formalin. Preser-

vation produced a mean shrinkage of 2 mm in length and gain of 0.2 g

in weight. In the laboratory, the preserved fish (499 males, 183 females)

were measured (total lengths to nearest 1.0 mm) and weighed (to nearest

0.1 g).

Age and Growth

Scale samples from 126 males and 65 females were taken from the

left side above the lateral line and below the spiny dorsal fin at the tip of

the depressed pectoral fin, mounted on optical plastic slides, and exam-

ined under a compound microscope at 10X magnification. Measure-

ments were made with an ocular micrometer from the center of the

focus to each annulus and to the scale margin in the anterolateral field.

Length-frequency distributions were plotted as a check on scale read-

ings. Age and growth data for males and females were analyzed using a

computer program developed by Marques et al. (1982). Regressions for

body length-scale radius relation were fitted by linear regression and

length-weight relations were computed. Growth curves were fitted to the

von Bertalanffy growth equation (Ricker 1975):

where Lt = total length (mm) at time t, L^ = asymptotic length (mm), K

= growth coefficient, t = time (age), and tQ = hypothetical age at zero

length. Annual survival of males and females was computed by the

unbiased minimum variance estimator of Chapman and Robson, which

is based on coded ages and the frequency of individuals in each age class

(Everhart et al. 1975). Since age 0 and I fish were less vulnerable to the

sampling gear, they were not included in these survival estimates.

Food Habits

Ten darters per month, sampled randomly from collections made

during each sampling period, were dissected to determine seasonal food

72 James C. Widlak and Richard J. Neves

habits. If fewer than 10 fish were collected in a month, all specimens

were examined. Stomach contents were removed, sorted, identified to

order or family (Hilsenhoff 1975; Merritt and Cummins 1978; Barnes

1980), and counted. Sampling to determine feeding chronology was

conducted on 2 July, 13 August, and 9 September 1981. Twelve to

twenty-four darters were collected during each of six 4-hour intervals

(1200, 1600, 2000, 2400, 0400, and 0800). Stomach contents were pooled

for each sampling time, blotted on a paper towel, and volumetric dis-

placement was measured with a 1 cc plastic syringe. Mean stomach

volume for each time interval was computed. These 24-hour samples

(329 fish) were also included in seasonal food habits analyses.

Reproduction

The reproductive cycle of the male was studied by recording the

size and appearance of testes of 427 males, and by external body colora-

tion. Ovaries and digestive tracts of females were removed, and the

ovaries weighed to the nearest 0.01 g. Adjusted body weights (body

weight after removal of stomach, ovaries, intestinal tract, and liver) of

females were also obtained. Ovaries were examined under a dissecting

microscope and classified as (1) gravid - containing maturing eggs; (2)

ripe - containing ripe eggs; (3) spent - containing some ripe eggs and

showing apparent resorption; and (4) resting - containing no mature or

maturing eggs. All eggs in both ovaries were counted and categorized

as (1) mature (ripe) - largest in size, translucent, indented, and contain-

ing a single large oil globule; (2) maturing - intermediate in size, opaque,

and yellow; and (3) immature - smallest in size, round, and white. Sam-

ples of 10 mature and maturing eggs from each pair of ovaries were

measured to the nearest 0. 1 mm with an ocular micrometer. A gonoso-

matic index (GSI) was calculated for females by multiplying the ovary

weight by 1000 and dividing by the adjusted body weight (Burr and

Page 1978, 1979); values were then plotted over time. No female darters

were collected in January and only two in December; consequently,

these months were not represented in the GSI computations. Analysis of

covariance tested for homogeneity between ovary weight and body

weight, total length, login body weight, and cube of total length (de

Vlaming et al. 1982). Relations between ova diameter, number of

mature and maturing ova, and total length of female, and ova number

and adjusted body weight in pre-spawning fish were computed by sim-

ple linear regressions. Attempts to observe spawning at the study site

and in the laboratory, and to rear darters for larval descriptions, were

unsuccessful.

RESULTS AND DISCUSSION

The redline darter was the predominant darter and one of the most

abundant fish species at the study site. A total of 682 was collected

Biology of Etheostoma rufilineatum 73

during the study. Large males (longer than 50 mm) occurred consis-

tently in the swiftest sections of riffles; smaller males and females were

collected in swift riffles during summer, but were found frequently in

areas with moderate current or along the margins of riffles in other

seasons. Young-of-the-year darters were not well represented in sam-

ples, and were generally found in protected areas with low current veloc-

ity, adjacent to emergent vegetation or streamside brush. Larval darters

apparently drift passively into pool areas (Scalet 1973) and have been

found at depths of 3 meters (Stiles 1972). The habitat of redline darters

in the North Fork Holston River concurs with that reported for other

Nothonotus species (Raney and Lachner 1939; Raney and Suttkus 1964;

Zorach 1969, 1970).

Age and Growth

Scale radius and total length of fish were strongly correlated (r =

0.85, females; r = 0.93, males). Regenerated scales were common on fish

of all sizes, but at least three readable scales were available from each

fish collected. Annuli were recognized by crowding of circuli in the

anterior field and cutting over in the lateral field. Annulus formation

was in early to mid-March at water temperatures of 5° to 20° C, as is

typical of other darter species (Fahy 1954; O'Neil 1981; Shute et al.

1982). The body-scale relations for males and females were linear. Equa-

tions for the fitted regression lines were as follows:

Males L = 14.3 + 0.619 (S) (R2 = 0.859)

Females L = 10.0 + 0.708 (S) (R2 = 0.720)

where L = total length (mm) and S = scale radius magnified (focus to

margin in the lateral field in mm). Back-calculated lengths, based on

scale measurements, approximated actual lengths of darters at capture

for all age classes (Table 2). Length-frequency distributions for males

and females did not provide an adequate indication of age class struc-

ture. Scale readings were assumed to be accurate because a random

sample of 30 scales was aged, with few discrepancies, by several fishery

biologists. Young-of-the-year darters appeared as an identifiable age

class in June, but overlap between lengths of 40 to 60 mm obscured

separation of the intermediate age classes (I and II). Scale analysis indi-

cated that maximum age for males and females was four years and three

years, respectively. Chapman-Robson estimates for annual survival were

0.28 (±0.003) for males and 0.03 (±0.001) for females.

Growth of redline darters was rapid and uniform for all age classes.

Because only one age III female (67 mm) was collected, length-at-age

data were insufficient to obtain a reliable growth equation for females.

The von Bertalanffy growth parameters for males were as follows:

L- = 88 mm, t0 = -0.815 years, K = 0.378

74

James C. Widlak and Richard J. Neves

Table 2. Actual and back-calculated lengths (linear regression) at capture for

redline darters, North Fork Holston River, May 1981-May 1982.

Table 3. Numerical abundance and percent of total number of food items for

different categories of food in 546 redline darter stomachs, 1981-1982.

Biology of Etheostoma rufilineatum 75

Males grew rapidly during the first two years of life and slowed thereaf-

ter. The largest male collected was 80 mm (age IV), approaching the

estimated maximum length of 88 mm for males. Length-weight relations

for male and female redline darters were described by the following

regression equations:

Males log W = -5.42 + 3.32 log|QL (R2 = 0.987)

Females logioW = -5.11 + 3.14 log^L (R2 = 0.956)

where W = weight (g), and L = total length (mm). Both sexes grew at

similar rates.

Food Habits

Contents of 546 stomachs were identified and tabulated (Table 3).

Empty stomachs made up only 9.9% (range, 5.6 to 11.4%) of the total

sample by season (Table 3). Dipteran larvae were the dominant food

items throughout the year. Chironomidae and Simuliidae made up from

67.1% of the diet in June- August to 87.0% in December- February, and

occurred in 10.0 to 83.3% of the stomachs examined monthly. Ephe-

meroptera and Trichoptera larvae were important items in summer and

fall; 9.8 to 29.3% of stomachs examined from June to November con-

tained these larvae. Hydracarina were eaten throughout the year, but

were most common in stomachs in March to May (5.1% of the diet).

Food of young-of-the-year was not determined, but may have been

largely zooplankton, as reported for the young of other species (Scalet

1972; Burr and Page 1978, 1979; Page 1980). The food of juveniles 25 to

40 mm was similar to that of adults. Items found infrequently in stom-

achs (less than 1% of the diet) included Plecoptera, Coleoptera, Nema-

tomorpha, and unidentifiable invertebrate eggs. Fish eggs and small

crayfish (Decapoda) were found in the stomachs of several large males

(60+ mm).

Sample sizes for the 24-hour feeding studies on 2 July, 13 August,

and 9 September were 95, 126, and 98 fish, respectively. Stomach con-

tent analyses indicated a distinct feeding chronology and similar feeding

patterns on all three sampling dates (Fig. 1). Mean food volumes in

stomachs increased from mid-morning through late afternoon and early

evening, and decreased from late evening to early morning. Feeding

peaks were at 2000 hr on 2 July (0.021 ml/ stomach) and 9 September

(0.028 ml/stomach), and at 1600 hr on 13 August (0.027 ml/stomach).

None of the fish collected at 1600 hr on the three dates had empty

stomachs, but 18 to 30% of those collected at 2400 hr and 76 to 78%

collected at 0400 hr were empty. These observations suggest that redline

darters are diurnal sight feeders. Scalet (1972) reported that Etheostoma

radiosum when feeding may rely on visual cues, particularly on move-

ments of prey, and similar visual feeding has been proposed for other

darter species (Mathur 1973; Adamson and Wissing 1977; Schenck and

Whiteside 1977).

76 James C. Widlak and Richard J. Neves

Reproduction

Darters less than 35 mm long could not be sexed by external char-

acteristics or by gonadal examination and were considered juveniles.

Adult males and females were easily distinguished by their sexual di-

chromism. Males in breeding condition were readily distinguishable from

non-breeding males by body coloration. Non-breeding males lacked

coloration on fins and pelvic region and were similar in appearance to

non-breeding females. Their testes were small and translucent. In con-

trast, the vertical fins of breeding males were edged with a band of dark

green and inner bands of white and bright red. The basal portion of

these fins was dusky gray. Paired fins were yellow or greenish-yellow

basally and edged with bright red. The pelvic region was dark green and

the abdomen creamy white to yellow. Body coloration varied from light

tan to dark brown, with bright red lateral spots. Testes were enlarged,

creamy white, and opaque. Bright coloration was apparent on males

throughout the year, but enlargement of testes was not observed until

March. Males in spawning condition were collected throughout spring

and summer, but testes were reduced in size in all fish collected in Sep-

tember. About 72% of age I males, all those longer than 40 mm, were

sexually mature.

Females were less colorful although breeding females were gener-

ally darker than others. They lacked the bright coloration on the fins

and body; vertical and paired fins were heavily spotted with black. The

body was darker than that of the male and lacked red spots, but the

caudal base of both sexes had prominent white spots. All females col-

lected between September and mid-February had resting ovaries, and

ova development first became apparent in March. By May the ovaries

contained both maturing and fully ripe ova. Females also reached sex-

ual maturity at age I; 56% of age I females were in spawning condition.

The smallest ripe female (42 mm long) was collected in July 1981.

Sex ratios of redline darters at the study site were strongly skewed

in favor of males. The overall ratio of males to females was 2.7:1, a

significant deviation from a 1:1 sex ratio (x2 = 146.4, p < 0.005). Sea-

sonal differences or sampling biases may have contributed to the appar-

ent temporal changes in distribution of the sexes. During summer, the

sex ratio averaged 2.5:1, since both sexes were present in riffles for

spawning. As females moved out of shallow riffle areas after spawning,

the sex ratio increased to 5:1 (September) and then returned to 2.5:1 as

age 0 individuals were recruited into the population. It is likely that

electrofishing was selective for males. Darters shocked in swift riffles

were swept into the water column and easily netted, while those in other

areas often remained on the bottom and were more difficult to collect.

Total egg complements of 85 female redline darters ranged from 50

Biology of Etheostoma rufilineatum

11

1200 1600 2000 2400

SAMPLING TIME (h)

0400 0800

Fig. 1. Twenty-four hour feeding chronologies for redline darters in the North

Fork Holston River, 2 July, 13 August, and 9 September 1981.

to 331; numbers of maturing (diameter 0.7-1.5 mm) and mature (1.6-2.2

mm) eggs combined ranged from 23 to 131. Diameter of mature ova

and total length of female were not significantly correlated (r = 0.08).

The number of mature ova and total length of female, and the number

of mature ova and adjusted body weight, were slightly correlated (r =

0.498 and 0.537, respectively).

Spawning occurred between May and August 1981 at water

temperatures of 14° and 26° C. Ripe females were collected as early as 3

May and as late as 18 August. The gonosomatic index (GSI) was not

used to identify the spawning season of redline darters because previous

78 James C. Widlak and Richard J. Neves

Table 4. Comparison of mean ovary weights of sexually mature redline darters

in the North Fork Holston River, 1981-1982.

statistical analyses have shown the ovary-body relationships of fishes to

be misleading (de Vlaming et al. 1982). Analysis of covariance indicated

that relationships between ovary weights and body weights and lengths

in female redline darters were not homogeneous for all stages of ova

development (p<0.001); therefore, the gonosomatic index did not give

an accurate indication of gonadal activity. Mean ovary weights for var-

ious size classes presented in Table 4 indicate that ovary weights were

highest from May to July (0.01-0.29 g) and lowest from November to

April (0.01-0.08 g). As judged by gonadal weights and egg maturity,

spawning apparently occurs from May to August in the North Fork

Holston River.

Assuming that all mature eggs produced during the spawning sea-

son by a gravid female are laid (Winn 1958), egg numbers for redline

darters in this study (23 to 131) are lower than those reported for other

darter species (Winn 1958; Bryant 1979; Burr and Page 1979; Lindquist

et al. 1981). Winn (1958) reported that in several species females spawn

with different males and lay only a few eggs at each spawning. Female

redline darters have been observed burying themselves in the gravel sev-

eral times during spawning, laying several eggs at a time (Stiles 1972).

Females may already have laid a portion of their eggs before collection

in summer, and egg numbers reported here may not represent total

numbers of eggs laid during the spawning season. The unusually low

correlation between total length of females and number of eggs tends to

support that conclusion.

ACKNOWLEDGMENTS.— We thank C. W. Fay, H. E. Kitchel, S.

Moyer, P. Pajak, and L. R. Weaver for assisting with field collections

Biology of Etheostoma rufilineatum 79

and G. B. Pardue, D. J. Orth, L. A. Helfrich, and R. E. Jenkins for

reviewing the manuscript. This study was funded in part by the U.S.

Fish and Wildlife Service.

LITERATURE CITED

Adamson, Scott W., and T. E. Wissing. 1977. Food habits and feeding periodic-

ity of the rainbow, fantail, and banded darters in Four Mile Creek. Ohio J.

Sci. 77:164-169.

Barnes, Robert D. 1980. Invertebrate Zoology. Saunders College Press, Phila-

delphia. 1089 pp.

Bryant, Richard T. 1979. The life history and comparative ecology of the sharp-

head darter, Etheostoma acuticeps. Tenn. Wildl. Resour. Agric. Tech. Rep.

No. 79-50. 60 pp.

Burr, Brooks M., and L. M. Page. 1978. The life history of the cypress darter,

Etheostoma proeliare, in Max Creek, Illinois. 111. Nat. Hist. Surv. Biol.

Notes 106. 15 pp.

, and 1979. The life history of the least darter, Etheos-

toma microperca, in the Iroquois River, Illinois. 111. Nat. Hist. Surv. Biol.

Notes 112. 15 pp.

Collette, Bruce B. 1967. The taxonomic history of the darters (Percidae: Etheos-

tomatini). Copeia 1967(4):814-819.

de Vlaming, Victor, G. Crossman and F. Chapman. 1982. On the use of the

gonosomatic index. Comp. Biochem. Physiol. 75:31-39.

Etnier, David A. 1980. Etheostoma rufilineatum (Cope). Redline darter. P. 687

in D. S. Lee, C. R. Gilbert, C. H. Hocutt, R. E. Jenkins, D. E. McAllister

and J. R. Stauffer, Jr. (Eds.). Atlas of North American Freshwater Fishes.

N.C. State Mus. Nat. Hist., Raleigh. 867 pp.

Everhart, W. Harry, A. W. Eipper and W. D. Youngs. 1975. Principles of

Fishery Science. Cornell Univ. Press, Ithaca. 288 pp.

Fahy, Wiliam E. 1954. The life history of the northern greenside darter, Ethe-

ostoma blennioides blennioides (Rafinesque). J. Elisha Mitchell Sci. Soc.

5/5:139-205.

Hilsenhoff, William L. 1975. Aquatic Insects of Wisconsin. Wis. Dep. Nat.

Resour. Tech. Bull. No. 89. 52 pp.

Jenkins, Robert E., and N. M. Burkhead. 1975. Recent capture and analysis of

the sharphead darter Etheostoma acuticeps, an endangered percid fish of

the upper Tennessee River drainage. Copeia 1975(4):73 1-740.

Lindquist, David G., J. R. Shute and P. W. Shute. 1981. Spawning and nesting

behavior of the Waccamaw darter, Etheostoma perlongum. Environ. Biol.

Fishes 6:177-191.

Marques, Kenneth, G. B. Pardue, R. J. Neves, L. Fortunato and A. Tipton.

1982. Computer program for the computation of age and growth statistics

of fish populations. Dep. Fish. Wildl. Sci., Va. Polytech. Inst. State Univ.

Manage. Series No. 1. 92 pp.

Mathur, Dilip. 1973. Food habits and feeding chronology of the blackbanded

darter, Percina nigrofasciata (Agassiz), in Halawakee Creek, Alabama.

Trans. Am. Fish. Soc. 702:48-55.

80 James C. Widlak and Richard J. Neves

Merritt, Richard W., and K. W. Cummins. 1978. An Introduction to the Aquat-

ic Insects of North America. Kendall-Hunt Co., Dubuque. 441 pp.

O'Neil, Patrick E. 1981. Life history of Etheostoma coosae (Pisces: Percidae) in

Barbaree Creek, Alabama. Tulane Stud. Zool. Bot. 23:75-84.

Page, Lawrence M. 1980. The life histories of Etheostoma olivaceum and E.

striatulum, two species of darters in central Tennessee. 111. Nat. Hist. Surv.

Biol. Notes 113. 14 pp.

1983. Handbook of Darters. TFH Publications Inc., Neptune City,

NJ. 271 pp.

Poppe, Wayne L. 1982. Cumberlandian mollusk conservation program - activity

6: analysis of water quality. Tennessee Valley Authority, Chattanooga. 94 pp.

Raney, Edward C, and E. A. Lachner. 1939. Observations on the life history of

the spotted darter, Poecilichthys maculatus (Kirtland). Copeia 1939(3):

157-165.

, and R. D. Suttkus. 1964. Etheostoma moorei, a new darter of the

subgenus Nothonotus from the White River system, Arkansas. Copeia

1964(1): 130-139.

Ricker, William E. 1975. Computation and interpretation of biological statistics

offish populations. Bull. Fish. Res. Board Can. No. 191. 382 pp.

Scalet, Charles G. 1972. Food habits of the orangebelly darter, Etheostoma

radiosum cyanorum (Osteichthyes: Percidae). Am. Midi. Nat. 57:515-522.

1973. Reproduction of the orangebelly darter, Etheostoma radiosum

cyanorum. Am. Midi. Nat. 59:156-165.

Schenck, John R., and B. G. Whiteside. 1977. Food habits and feeding behavior

of the fountain darter, Etheostoma fonticola (Osteichthyes: Percidae).

Southwest. Nat. 27:487-492.

Shute, Peggy W., J. R. Shute and D. G. Lindquist. 1982. Age, growth, and early

life history of the Waccamaw darter, Etheostoma perlongum. Copeia

1982(3):561-567.

Stiles, Robert A. 1972. The comparative ecology of three species of Nothonotus

(Percidae - Etheostoma) in Tennessee's Little River. Ph.D. Dissert., Univ.

Tennessee, Knoxville. 97 pp.

Widlak, James C. 1982. The ecology of freshwater mussel-fish host relationships

with a description of the life history of the redline darter in the North Fork

Holston River, Virginia. Unpubl. M.S. Thesis, Virginia Polytechnic Insti-

tute and State University, Blacksburg. 121 pp.

Winn, Howard E. 1958. comparative reproductive behavior and ecology of four-

teen species of darters (Pisces — Percidae). Ecol. Monogr. 25:155-191.

Zorach, Timothy. 1969. Etheostoma jordani and E. tippecanoe, species of the

subgenus Nothonotus (Pisces: Percidae). Am. Midi. Nat. 57:412-434.

1970. The systematics of the percid fish Etheostoma rufilineatum

(Cope). Am. Midi. Nat. 54:208-225.

Accepted 23 May 1984

Rete Mirabile Ophthalmicum

and Intercarotid Anastomosis in Procellariiformes

Taken off the North Carolina Coast

Gilbert S. Grant1

North Carolina State Museum of Natural History,

P.O. Box 27647, Raleigh, North Carolina 27611

ABSTRACT. — Dissections of arterial circulation patterns were made

in eleven species of procellariiform birds taken off the coast of North

Carolina. All species possessed well-developed rete mirabile ophthal-

micums (RMO) and intercarotid anastomoses, both playing a role in

selectively shunting blood flow and counter-current heat exchange to

facilitate thermoregulation and maintaining brain temperatures lower

than body temperatures during heat stress. There was no correlation

between relative size of the RMO and nesting latitude, but RMO size

was relatively greater in the smaller members of the order.

INTRODUCTION

Kilgore et al. (1979) and Bernstein et al. (1979a, 1979b) showed that

the presence of a rete mirabile ophthalmicum (RMO) was associated

with a reduction in brain temperature in heat stressed birds. The rete

facilitates counter-current heat exchange between the arterial blood

supply to the brain and the venous blood draining the evaporative sur-

faces of the mouth and the cornea. RMO's have been reported in a

number of species (Richards and Sykes 1967; Lucas 1970; Kilgore et al.

1976; Crowe and Crowe 1979; Pettit et al. 1981). Pettit et al. (1981)

examined the anatomy of the RMO of Hawaiian seabirds that may

encounter stressful thermal environments at their tropical nesting sites.

An additional site for possible counter-current heat exchange lies in the

cavernous sinus that houses the carotid vein and the intercarotid anas-

tomosis (Baumel and Gerchman 1968). These authors described three

major types of intercarotid anastomoses in birds.

My study was undertaken to determine if the RMO and the type of

intercarotid anastomosis in procellariiform birds breeding near the

poles, in the temperate zones, and in the tropics differed, perhaps in

response to thermal stresses encountered at the nesting colony. One

might predict that tropical, open, ground-nesting species are exposed to

greater thermal stress and therefore would have relatively larger retes.

METHODS

Most specimens were collected 30 to 60 km off the North Carolina

'Present address: Route 2, Box 431, Sneads Ferry, North Carolina 28460

Brimleyana No. 11:81-86, October 1985 8 1

82 Gilbert S. Grant

coast, but a few specimens of the Northern Fulmar, Fulmarus glacialis,

and the Sooty Shearwater, Puffinus griseus, were found dead on North

Carolina beaches. The lone Antarctic Petrel, Thalassoica antarctica, was

obtained from D. G. Ainley via D. W. Johnston in trade. All specimens

on which this note is based are deposited in the collection of the North

Carolina State Museum of Natural History. Dissections under a binocu-

lar microscope were in most cases performed on pickled specimens,

using the methodology of Pettit et al. (1981).

RESULTS

Two of the three types of intercarotid anastomoses were found in the

eleven procellariiform species examined in this study (Table 1). There

was no correlation of anastomosis type with latitude (as an index of

potential thermoregulatory stress).

In all cases the RMO was found in the temporal region of the head,

between the orbital ridge and the otic process of the quadrate. The

RMO is derived from the external ophthalmic branch of the internal

carotid artery and from branches of the facial, maxillary and mandibu-

lar veins. To crudely assess the relative size of the rete in 1 1 species of

Procellariiformes ranging in weight from 34 to 774 grams, I measured

the surface area of the RMO using a transparent grid. This measure

does not take into account the relative thickness or, more importantly,

the actual area of contact of the arterial and venous components of the

rete. However, it does give a first approximation of the relative size of

the rete. Surface area/ body weight ratios exhibited a slight but insignifi-

cant (P>0.05) increase with latitude (Table 2), while an inverse correla-

tion (P<0.05) of body weight and RMO surface area/ weight ratio was

evident. Thus, the smaller procellariiforms have relatively larger RMO's.

DISCUSSION

A well-developed intercarotid anastomosis unites the two carotids

caudal to the hypophysis in most birds examined by Baumel and

Gerchman (1968). They found that injection of the cervical portion of

one carotid resulted in bilateral filling of both the intra- and extracra-

nial arteries via this anastomosis. The avian intercarotid anastomosis

may effectively substitute for the mammalian circle of Willis in main-

taining brain-to-body temperature differences (Baumel and Gerchman

1968; Kilgore et al. 1976; Pettit et al. 1981).

As Kilgore et al. (1976) pointed out, the effectiveness of the RMO

heat exchange depends on the arterial-venous temperature differential,

on rete blood flow and velocity, and the area and closeness of arterial-

venous contacts within the rete.

Several species, including Sooty and Greater Shearwaters, and Wil-

son's Storm-Petrel, Oceanites oceanicus, are transequatorial migrants

Seabird Rete 83

Table 1. Pattern of intercarotid anastomosis in 46 specimens of 11 species of

Procellariiformes. N = number examined.

'X-type is defined as having cerebral carotids anastomosing side-to-side, H-

type has a pronounced transverse anastomosis joining the two carotids, and

X-H type is intermediate with a short transverse anastomosis (after Baumel

and Gerchman 1968).

2from Palmer (1962) or Watson (1975).

that may be exposed to thermal stress while flying across the doldrums

(equatorial zone with little wind). Their retes are not appreciably larger

than those of north or south temperate zones or Antarctic species. Birds

collected while they were flying at sea off North Carolina did not

exhibit elevated body temperatures (Platania et al., in press). The RMO

of the Black-capped Petrel, Pterodroma hasitata, a tropical species, is

not different from that of other species of higher latitudes. The bird is

not subjected to heat stress at its nesting grounds because it is a winter

(Northern Hemisphere) breeder, and because at high elevations it exca-

vates burrows. Altitude, and the extent of the use of burrows for nest-

ing, further cloud simple correlations of RMO ratios with latitude. In

general, the smaller birds nest exclusively in burrows or crevices while

the larger shearwaters and fulmars are open ground or cliff nesters. In

addition, most ground-level activity of burrowing species occurs at

night, further reducing heat stress.

As arterial blood may reach the brain via several routes (Richards

and Sykes 1967; Richards 1970; Crowe and Crowe 1979), involving both

direct and indirect (via extensive anastomoses) flow, the potential exists

for selectively regulating flow under varying conditions. Flow of arterial

blood may be shunted through the RMO to the brain via anterior anas-

tomoses with intracranial branches of the internal carotid. This could

84

Gilbert S. Grant

Table 2. Surface area of rete and body weight ratios of 1 1 species of Atlantic Procellarii-

formes. Data presented as mean ± 1 standard deviation (sample size).

serve to maintain brain temperatures lower than core body temperatures

during heat stress (Kilgore et al. 1979; Bernstein et al. 1979a, b). During

cold stress the arterial blood flowing to the anterior surface of the head

is cooled by returning venous blood in the RMO (Frost et al. 1975).

Undue loss of body heat is prevented in the RMO by counter-current

heat exchange. In this example, the anastomoses with the intracranial

arteries are not open; blood flow to the brain is achieved directly via the

internal carotid. Therefore, the lack of correlation between size of RMO

and latitude (as an indicator of temperature stress) may indicate that the

RMO functions during both cold and heat stress. The relatively larger

RMO in smaller birds is probably related to their relatively larger sur-

face/volume ratios, and the relatively greater stress they encounter as

environmental temperatures fluctuate.

Seabird Rete 85

In summary, all 1 1 species of procellariiform birds examined pos-

sessed a rete mirabile ophthalmicum. There was no correlation between

relative size of RMO and nesting latitude, but RMO size was relatively

greater in the smaller members of the order.

ACKNOWLEDGMENTS.— Most of the specimens examined were

collected by David S. Lee and colleagues at the N.C. State Museum of

Natural History under a contract (#92375-1 130-621-16) from the Slidell,

Louisiana, laboratory of the U.S. Fish and Wildlife Service. This is con-

tribution number 1985-4 of the North Carolina Biological Survey.

LITERATURE CITED

Baumel, Julian J., and L. Gerchman. 1968. The avian intercarotid anastomosis

and its homologue in other vertebrates. Am. J. Anat. 722:1-18.

Berstein, Marvin H., M. B. Curtis and D. H. Hudson. 1979a. Independence of

brain and body temperatures in flying American Kestrels, Falco sparverius.

Am. J. Physiol. 257:R58-R62.

, I. Sandoval, M. B. Curtis and D. H. Hudson. 1979b. Brain tempera-

ture in pigeons: effects of anterior respiratory bypass. J. Comp. Physiol.

729:115-118.

Brown, Leslie H., E. K. Urban and K. Newman. 1982. The Birds of Africa. Vol.

1. Academic Press, New York. 521 pp.

Crowe, T. M., and A. A. Crowe. 1979. Anatomy of the vascular system of the

head and neck of the helmeted guinea fowl Numida meleagris. J. Zool.

(Lond.) 755:221-233.

Frost, P. G. H., W. R. Siegfried and P. J. Greenwood. 1975. Arteriovenous heat

exchange systems in the jackass penguin, Spheniscus demersus. J. Zool.

(Lond.) 775:231-241.

Kilgore, Delbert L., M. H. Bernstein and D. H. Hudson. 1976. Brain tempera-

tures in birds. J. Comp. Physiol. 770:209-215.

, D. F. Boggs and G. F. Birchard. 1979. The role of the rete mirable

ophthalmicum in maintaining the body-to-brain temperature difference in

pigeons. J. Comp. Physiol. 729:119-122.

Lucas, Alfred M. 1970. Avian functional anatomic problems. Fed. Proc.

29:1641-1648.

Palmer, Ralph S. (Ed.). 1962. Handbook of North American Birds. Vol. 1. Yale

Univ. Press, New Haven. 567 pp.

Pettit, Ted N., G. C. Whittow and G. S. Grant. 1981. Rete mirabile opthalmi-

cum in Hawaiian seabirds. Auk 95:844-846.

Platania, Steven P., G. S. Grant and D. S. Lee. In press. Core temperatures of

non-nesting Western Atlantic seabirds. Brimleyana 12.

Richards, S. A. 1970. Brain temperature and the cerebral circulation in the

chicken. Brain Res. 25:265-268.

86 Gilbert S. Grant

Richards, S. A., and A. H. Sykes. 1967. Responses of the domestic fowl (Gallus

domesticus) to occlusion of the cervical arteries and veins. Comp. Biochem.

Physiol. 27:39-50.

Watson, George E. 1975. The Birds of the Antarctic and Sub-Antarctic.

National Geophysical Union, Washington, D.C. 350 pp.

Accepted 19 July 1984

Notes on Virginia (Reptilia: Colubridae)

in Virginia

Charles R. Blem and Leann B. Blem

Department of Biology,

Virginia Commonwealth University, Academic Division,

Richmond, Virginia 23284

ABSTRACT. — Female Virginia striatula in central Virginia produce

an average of 6.0 young/ litter and reproduce annually. Litter size, fre-

quency of reproduction, oviductal egg size and size of newborn young

are greater than those of V. striatula from the southwestern part of its

range in Texas. Absence of size classes below those of mature snakes

suggests high mortality of subadults or perhaps sampling bias caused

by behavioral differences between adults and young. Ventral and sub-

caudal counts of V. striatula from Virginia are low; comparison of

these with the few measurements from the rest of the range indicates

there is significant geographic variation, although the pattern is not

clear. Data from Virginia v. valeriae collected in the same area are also

provided.

INTRODUCTION

The genus Virginia includes two species of small, secretive, ground-

dwelling snakes: the Rough Earth Snake, Virginia striatula (Linnaeus),

and the Smooth Earth Snake, Virginia valeriae (Baird and Girard). In

Virginia, V. valeriae is represented by the nominate subspecies, V. v.

valeriae^ the Eastern Earth Snake. Both species of Virginia are found

throughout much of the southern tier of states from eastern Texas and

Oklahoma to the Atlantic Coast (Conant 1975). The Rough Earth

Snake reaches the northern edge of its known distribution along the

Atlantic Coast in central Virginia. The Eastern Earth Snake occupies

most of Virginia, and its distribution extends northward to New Jersey

and Pennsylvania.

Few papers containing quantitative data have been published

regarding V. striatula (Clark 1964; Clark and Fleet 1976), and nothing

has been reported regarding the species in Virginia. As part of a long-

term study of reptiles at the northern edge of their range, we report here

morphometric and reproductive data for a population of V. striatula in

central Virginia. The collecting location is near the apparent northern

extreme of the species' range (Conant 1975; Martof et al. 1980). Addi-

tionally, we provide information about a smaller sample of V. v. vale-

riae collected in the same area.

Brimleyana No. 1 1:87-95, October 1985 87

88 Charles R. Blem and Leann B. Blem

MATERIALS AND METHODS

Data were taken from freshly collected specimens and preserved

material in the Virginia Commonwealth University herpetological col-

lection. Twenty-six male and forty-six female V. striatula are included.

Four of the females lacked tail tips, so sample sizes of subcaudal counts

and tail lengths are reduced accordingly. Thirteen male and seventeen

female V. valeriae are also included. Most specimens were collected in

eastern Henrico County, about 16 km east of Richmond. Two V. stria-

tula and three V. valeriae were obtained from sites in Chesterfield

County, approximately 20 km south of the Henrico locality. Snakes

were collected in the period 1974-1980, but 85% of the sample was

obtained during 1978-1980. The Henrico site is covered by a secondary

growth of loblolly pine, Pinus taeda, within which are piles of roofing,

discarded furniture, and tires. Besides Virginia, 14 other species of rep-

tiles have been collected at the site, including one turtle, Terrapene

carolina\ five lizards, Sceloporus undulatus hyacinthinus, Scincella later-

alis, Cnemidophorus sexlineatus, Eumeces fasciatus and Eumeces

inexpectatus; and eight other snakes, Storeria occipitomaculata, Store-

ria dekayi, Carphophis amoenus, Diadophis punctatus, Heterodon pla-

tyrhinos, Coluber constrictor, Elaphe obsoleta, and Agkistrodon con-

tortrix mokasen. Virginia striatula is the most common reptile at this

locality.

Snout-vent and tail lengths were measured to the nearest millime-

ter. Ventral and subcaudal scales were counted in standard fashion (e.g.

see Schmidt and Davis 1941) and the style is comparable to that used by

Clark and Fleet (1976). Sex, number of ova/ embryos, and stomach con-

tents were determined by dissection. Testes and ova lengths were meas-

ured to the nearest millimeter. Embryos were removed and examined

microscopically to determine extent of development. Developmental

stages were assigned according to criteria in Zehr (1962). The 5% level

of significance (P<0.05) was used in all statistical tests.

RESULTS

MORPHOMETRICS

There is no statistical intersexual difference in snout-vent lengths of

V. striatula (t = 1.5, df = 70; Table 1), but significant differences exist

between tail lengths (t = 2.6, df = 63), tail length/ snout- vent length ratios

(t = 6.9), ventral scale counts (t = 2.6) and subcaudal counts (t = 9.3). There

also is no statistical difference between snout-vent length of sexes of V.

valeriae (t = 0.7, df = 28; Table 2). However, statistical differences do exist

between tail lengths (t = 3.0, df = 28), tail length/ snout-vent length ratios

(t = 10.1), ventral scale counts (t = 5.7) and subcaudals (t = 13.3).

Notes on Virginia in Virginia 89

Table 1. Comparison of Virginia striatula from Virginia and Brazos County,

Texas. All values are means ± one standard error; sample sizes are in

parentheses. All lengths are in mm; SVL = snout-vent length.

Table 2. Morphometric data for Virginia valeriae in central Virginia. Values are

means ± one standard error. See text for sample sizes.

Measurement Males Females

Snout-vent length (mm) 1 35. 1 ± 8.5 1 45.4 ± 12.5

Tail length (mm) 30.6 ± 2.2 22.5 ± 1.6

Tail length /snout-vent length 0.225 ± 0.004 0.160 ± 0.005

Ventrals 114.6 ±0.5 119.5 ±0.7

Subcaudals 35.4 ± 0.5 26.0 ± 0.5

Sex Ratios and Reproduction

Significantly more female V. striatula were collected than males

(X2 = 5.0, Yates continuity correction performed). We found no signifi-

cant difference in sex ratios of V. valeriae.

The smallest gravid V. striatula was 175 mm snout-vent length.

Thirty-one of the forty-six females (67%) in this study reached or

exceeded this size, and mean size of mature females was 210.0 mm

(SE = 3.9). Excluding those females below 175 mm and those mature

females captured outside the reproductive period (see below), 100%

(24/24) possessed yolked eggs and /or embryos or gave birth to young in

the laboratory. Size of ovarian eggs apparently begins to increase in late

90 Charles R. Blem and Leann B. Blem

March or early April, and ovulation occurs by mid to late May (Fig. 1,

Table 3). Birth probably occurs in late July or early August. One female

bore young in the lab on 10 August. Mean litter size, as based on the

number of fertilized or yolked eggs or young, was 6.0 (range 4-10,

N = 24).

The smallest gravid V. valeriae was 185 mm snout-vent length.

Only 6 of the 17 females in this study reached or exceeded this size. Of

these, 83% (5/6) possessed enlarged ova or embryos. Mean litter size is

6.6 ± 0.8 (N = 5). Ovarian eggs appear to increase in size at about the

same time as those of V. striatula, and one female bore four young in

the laboratory on 3 August.

Developmental stages (Table 3) of gravid female V. striatula seemed

to conform closely to Zehr's (1962) scheme. In some females there was a

small amount of variation in degree of development of embryos, but

never more than three stages were present. Although Zehr recognized 37

stages, in practice the first 5 stages (pre-blastodisc) are difficult to rec-

ognize. Later stages can be recognized with some precision, and in our

material, development of embryos confirms the timing of reproduction

described above.

Most of the male V. striatula collected in our study were sexually

mature as concluded from convolution of the vasa deferentia and en-

largement of the testes. Clark (1964) found that minimum body length of

mature males was 142 mm; 69.2% (18/26) of our sample exceeded 164

mm (none in the size interval 138-163 mm were collected), and all had

convoluted vasa deferentia. Right testes length in mature V. striatula

ranged from 5 to 12 mm. No seasonal cycle in testes size could be dem-

onstrated as the length of testes varies with snout-vent length, obscuring

temporal variations. At least 53.8% (7/ 13) of the male V. valeriae were

over 125 mm and appeared to be sexually mature. These had enlarged

testes (6-12 mm) and also had convoluted vasa deferentia.

Food

Only 19.6% (20/102; 12 V. striatula, 8 V. valeriae) of the Virginia

collected in this study contained food. All recognizable items consisted

of small pieces of red annelids that we were not able to identify further.

DISCUSSION

Clark (1964) published an analysis of an extensive series (324 spec-

imens) of V. striatula collected in Brazos County, Texas, and Clark and

Fleet (1976) provided ecological data for a population in the same area.

Both sites are near the southwestern edge of the species' range (see

Conant 1975). Snout-vent lengths and tail lengths of the males in

Clark's study differed significantly from those in our study (t = 2.3 and

Notes on Virginia in Virginia

91

E

E

E

>

o

x"

CO

18

16

14

12

10

8

6

4

2

•••

f

•••

•• •

••

M

M

Month

Fig. 1. Seasonal variation in maximum size of ova or ovarian follicles of mature

female Virginia striatula from Virginia.

Table 3. Stage of development of eggs of the rough earth snake, Virginia

striatula.

2.4; df = 193, 157), while those of females did not (t = 1.3 and 1.3; df = 199,

170). There was no significant difference in tail length/ snout-vent length

ratios (males: t = 1.3; females: t = 0.4). In both our data and those of Clark

and of Clark and Fleet, shrinkage corrections were made after the speci-

mens were preserved. Scale counts of Virginia specimens are lower than

those of Texas material. Ventral counts differ greatly (males: t = 20.8;

92 Charles R. Blem and Leann B. Blem

females: t = 22.4). and smaller but significant differences occur between

subcaudal counts (males: t = 2.3; females: t = 8.9). It appears that signifi-

cant geographic variation occurs in scale number of V. striatula,

although the pattern is not clear. Mount (1975) gave subcaudal counts

for Alabama striatula that are lower than those of Virginia or Texas

specimens (males: 38.5; females: 34.2). Mean ventral counts of Alabama

males (121.2) are intermediate between those from Virginia and Texas,

while ventral counts from females (124.8) are lower than those of the

other states.

Although our collecting efforts were not evenly divided over all

months, it is obvious that earth snakes were more available in March,

April and May (63/102 = 61.8% were collected in these months). This

may be due to increased exposure as a result of mating activity. Clark

(1964) concluded that mating of V. striatula occurred in Texas during

March and April, as judged from the presence of spermatozoa in the

lumina of the oviducts. D. Greene (pers. comm.) observed copulation in

a group of 30 or more individuals accidentally excavated from a hiber-

naculum at Richmond, Virginia, on 30 March 1982. This suggests that

first mating may occur shortly after emergence from the hibernaculae in

early spring.

Size distribution of V. striatula in central Virginia seems skewed

toward larger snakes. For example, males of striatula as small as 123

mm snout-vent length were collected in May, yet only 28.0% (7.25;

newborn young excluded) of the entire sample was less than 164 mm

(minimum size at maturity). For females, 123 mm also was a minimum

size of May specimens, yet only 33.0% did not exceed 175 mm. This bias

toward large size indicates either that subadult earth snakes are difficult

to find, or that mortality rates of young are relatively great. We believe

that the latter hypothesis is correct. Subadult snakes do not seem to

behave differently from older snakes, as they often were found with

adults or in similar sites. Also, the abundance of small V. valeriae sup-

ports this contention. About 41.7% (5/12; newborns excluded) of male

V. valeriae and 57.1% (8/14) of female valeriae were shorter than

mature individuals. Virginia valeriae is not at the northern edge of its

range as is V. striatula at this collection site. We hypothesize that mid-

winter mortality of young snakes may be important, as it is in the east-

ern cottonmouths, Agkistrodon p. piscivorus that also reach the

northern edge of their range in this area (Blem 1981).

Gravid female V. striatula near the northern edge of the range are

smaller than those reported by Clark (1964) and by Clark and Fleet

(1976) in Texas (Table 1). Since error terms are not available for the

Texas sample, statistical comparison is not possible. Clark and Fleet

demonstrated a significant regression of litter size on snout-vent length

Notes on Virginia in Virginia

93

N

"35

"O

o

o

DO

200

250

SVL Imml

Fig. 2. Brood size as a function of snout-vent length (SVL, mm) in female Vir-

ginia striatula. Solid circles represent Texas specimens (fclark and Fleet 1976);

hollow circles represent Virginia specimens. The dash line is the least-squares

best-fit line for Texas snakes (brood size - 0.052 SVL - 6.647, r = 0.77); the solid

line is the best-fit line for Virginia snakes (brood size = 0.038 SVL - 1.993, r =

0.46).

in which larger females produce larger numbers of ova or young. Statis-

tical comparison of their data with ours demonstrates a significant inter-

locality difference (F = 20.4); slopes and intercepts of equations predict-

ing brood size from female snout-vent lengths are significantly different

(see Fig. 2). This means that female V. striatula in Virginia produce

more eggs per reproductive attempt than do Texas females of similar

size. Absolute brood size, as judged from numbers of fertilized ova or

young per female, is significantly higher in Virginia females (t = 2.1;

df = 38) than the values given by both Clark and Clark and Fleet, but the

difference is not statistically significant in the latter comparison.

Further reproductive adjustment may occur through increased ova

size. Clark (1964) found that the largest right ovarian follicle was larger

than 5 mm only twice in a sample that included over 100 females (many

of them over 180 mm snout-vent length), and concluded that one of

these was abnormal. In our study, largest right ovarian follicles reached

12 mm and oviductal eggs were 12-17 mm long. Also, while 100%

(24/24) of our females were gravid during the reproductive period, not

all mature Texas females were. It may be significant that newborn

94 Charles R. Blem and Leann B. Blem

Texas V. striatula ranged from 61-69 mm in snout- vent length (N = 8)

while the range of Virginia specimens was 73-79 mm (N = 7). Clark did

not calculate the percentage of mature females that produced young

each year, but we interpret his data as indicating that about 79% (30/38)

should have produced young. Clark and Fleet (1976) assumed that

mature females produced young once per year, but pointed out that in

some years this was not achieved.

Blem (1981), in an analysis of the eastern cottonmouth at the

northern edge of its range in Virginia, found that a major limiting factor

was high overwinter mortality of young during very cold winters.

Reproductive rates were very high; 83% of mature females were gravid

during the breeding season (Blem 1982). The similarity of these findings

to those of the present study suggests that the spread of snakes north-

ward on the coastal plain of Virginia may sometimes be limited by the

balance between reproductive rate and cold-induced mortality. Further

comparison of our studies of Virginia and Agkistrodon, and a vast liter-

ature (e.g. Fitch 1970), suggest that some generalities may be recognized

regarding reproductive "strategies" of small colubrids and larger viper-

ids. In general, it appears that less than 100% of mature female viperid

snakes reproduce each year and the proportion is often nearer 50%

(Aldridge 1979; but see Kofron 1979 and Blem 1982 for exceptions).

Conversely, 90-100% of the females of many colubrid species, particu-

larly small snakes such as Carphophis amoenus, Diadophis punctatus

and Thamnophis sir talis, reproduce annually (Aldridge 1979). It there-

fore appears that one might expect a large proportion of the females of

a population of V. striatula to be involved in the production of young

each year. In many studies addressing reproductive output of snakes, a

relationship has been noted between litter size, or some other measure

of reproductive output, and female size (see Blem 1981, 1982). Both

small colubrids (e.g. Clark 1964) and viperids show this phenomenon.

However, frequency of reproduction appears to be a size-related phe-

nomenon in some viperids (Burkett 1966; Blem 1982), while that rela-

tionship has not been demonstrated for small colubrids.

ACKNOWLEDGMENTS.— We are indebted to William Gutzke

for sharing one of his favorite collecting sites and for companionship in

the field. Several Virginia Commonwealth University herpetology classes

provided most of the manpower necessary for collecting specimens.

Cheryl Roeding, Gerald Schaefer and Tom Thorp also provided assist-

ance in the field. We also thank Carolyn Conway for assistance in

staining and staging embryos.

Notes on Virginia in Virginia 95

LITERATURE CITED

Aldridge, Robert D. 1979. Female reproductive cycles of the snakes Arizona

elegans and Crotalus viridis. Herpetologica 35(3):256-261.

Blem, Charles R. 1981. Reproduction of the eastern cottonmouth Agkistrodon

piscivorus piscivorus (Serpentes: Viperidae) at the northern edge of its

range. Brimleyana 5:117-128.

1982. Biennial reproduction in snakes: an alternative hypothesis.

Copeia 1982 (4):961-963.

Burkett, Ray D. 1966. Natural history of the cottonmouth moccasin, Agkis-

trodon piscivorus (Reptilia). Univ. Kans. Publ. Mus. Nat. Hist. 17(9):

435-491.

Clark, Donald R., Jr. 1964. Reproduction and sexual dimorphism in a popula-

tion of the rough earth snake, Virginia striatula (Linnaeus). Tex. J. Sci.

/<5(3):265-295.

, and Robert R. Fleet. 1976. The rough earth snake (Virginia stria-

tula): ecology of a Texas population. Southwest. Nat. 20(4):467-478.

Conant, Roger. 1975. A Field Guide to the Reptiles and Amphibians of Eastern

and Central North America. Houghton Mifflin, Boston. 429 pp.

Fitch, Henry S. 1970. Reproductive cycles in lizards and snakes. Univ. Kans.

Mus. Nat. Hist. Misc. Publ. 52:1-247.

Kofron, Christopher P. 1979. Reproduction of aquatic snakes in south-central

Louisiana. Herpetologica 35(l):44-50.

Martof, Bernard S., W. M. Palmer, J. R. Bailey and J. R. Harrison III. 1980.

Amphibians and Reptiles of the Carolinas and Virginia. Univ. North Caro-

lina Press, Chapel Hill. 264 pp.

Mount, Robert H. 1975. The reptiles and amphibians of Alabama. Auburn

Univ. Agric. Exp. Stn., Auburn. 347 pp.

Schmidt, K. P., and D. D. Davis. 1941. Field Book of Snakes of the United

States and Canada. G. P. Putnam's Sons, N.Y. 365 pp.

Zehr, David R. 1962. Stages in the normal development of the common garter

snake, Thamnophis sirtalis sirtalis. Copeia 1962 (2): 322-329.

Accepted 22 May 1984

96

THE SEASIDE SPARROW,

ITS BIOLOGY AND MANAGEMENT

Edited by

Thomas L. Quay, John B. Funderburg, Jr., David S. Lee,

Eloise F. Potter, and Chandler S. Robbins

The proceedings of a symposium held at Raleigh, North Carolina,

in October 1981, this book presents the keynote address of F. Eugene

Hester, Deputy Director of the U. S. Fish and Wildlife Service, a bibli-

ography of publications on the Seaside Sparrow, and 16 major papers

on the species. Authors include Arthur W. Cooper, Oliver L. Austin, Jr.,

Herbert W. Kale II, William Post, Harold W. Werner, Glen E. Wool-

fenden, Mary Victoria McDonald, Jon S. Greenlaw, Michael F. Delany,

James A. Mosher, Thomas L. Merriam, James A. Kushlan, Oron L.

Bass, Jr., Dale L. Taylor, Thomas A. Webber, and George F. Gee. A

full-color frontispiece by John Henry Dick illustrates the nine races of

the Seaside Sparrow, and a recording prepared by J. W. Hardy supple-

ments two papers on vocalizations.

"The Seaside Sparrow, with its extensive but exceedingly narrow

breeding range in the coastal salt marshes, is a fascinating species. All

the authors emphasize that the salt marsh habitat is at peril. . . . The

collection is well worth reading." — George A. Hall, Wilson Bulletin.

1983 174 pages Softbound

Price: $15, postpaid. North Carolina residents add 4'/2% sales tax. Please make

checks payable in U. S. currency to NCDA Museum Extension Fund.

Send to SEASIDE SPARROW, N. C. State Museum of Natural History,

P. O. Box 27647, Raleigh, NC 2761 1.

Fossil Bats (Mammalia: Chiroptera)

from the Late Pleistocene and Holocene Vero Fauna,

Indian River County, Florida

Gary S. Morgan

Florida State Museum,

University of Florida, Gainesville, Florida 32611

ABSTRACT. — Six species of bats are reported from the late Pleisto-

cene and Holocene Vero fossil vertebrate locality on the east coast of

peninsular Florida: Eptesicus fuscus, Lasiurus intermedins, L. cf.

seminolus, Nycticeius humeralis, Tadarida brasiliensis, and Eumops

glaucinus. This is the first known fossil occurrence of Lasiurus semino-

lus, and the first record of Nycticeius humeralis from the Pleistocene

of Florida. Previous reports of Myotis austroriparius from Vero are

shown to be in error, as they were based on a misidentified humerus.

The bats from Vero represent the most diverse fossil chiropteran fauna

yet known from Florida and one of the richest in the North American

Quaternary. This site is unique among Florida fossil vertebrate locali-

ties as it samples species of bats that roost primarily in trees, rather

than cave-dwelling forms. The six species present at Vero constitute

the entire native chiropteran fauna of present-day South Florida, indi-

cating that the bat fauna of this region has remained relatively stable

over the past 10,000 years.

INTRODUCTION

Recent curation of the abundant microvertebrate fossils collected

by Robert D. Weigel in 1956 and 1957, during his re-excavation of the

classic Vero Site on the Atlantic coast of Florida (Weigel 1962),

revealed the presence of a relatively large sample of bat remains. Based

on only four elements, Weigel (1962) recognized three genera of bats

from Vero — Myotis, Eptesicus, and Lasiurus. He did not assign any of

his material to species, and his identification of Myotis was incorrect.

Detailed study of the bat fossils from Vero, especially the postcranial

elements, and a re-examination of the small sample identified by Wei-

gel, allows for more precise identification of most of the material. In the

present study, six species of bats are recognized from the Vero deposits

based on 37 elements representing 16 individuals. Comparison with data

in Webb (1974:14, Table 2.1) indicates that the six species of bats at

Vero make it the most diverse fossil bat fauna yet known from Florida.

DESCRIPTION OF LOCALITY

The Vero fauna is one of the best known late Pleistocene (Rancho-

labrean) local faunas in Florida (see Weigel 1962 for a complete list of

Brimleyana No. 11:97-117, October 1985 97

98 Gary S. Morgan

fossil vertebrates from Vero). Vero engendered considerable controversy

in the early part of this century, as it was the first fossil site in the New

World where human bones and artifacts were supposedly found in asso-

ciation with extinct Pleistocene vertebrates. The site was discovered in

November 1913 during excavation of an east-west drainage canal through

the town of Vero Beach by the Indian River Farms Company. Between

1913 and 1917, Isaac M. Weills and Frank Ayers collected the majority

of the vertebrate fossils and human remains that formed the basis for a

large number of publications on the site (see Ray 1957 and Weigel 1962

for a complete bibliography). The fossil site is located within the present

city limits of Vero Beach, Indian River County, Florida (center of

SE1/4, sec. 35, T32S, R39E, Vero Beach Quadrangle, USGS 7.5 min.

series; 27°39'N latitude, 80°24'W longitude), southeast of the Vero

Beach airport and immediately south of the Florida East Coast Rail-

road. The paleontological and historical significance of Vero, coupled

with the paucity of microvertebrate fossils in the early collections,

prompted Weigel to conduct extensive field work at the site during the

summers of 1956 and 1957.

The fossil-bearing deposits at Vero consist of three distinct units,

designated from bottom to top as Strata 1, 2, and 3 by Sellards (1917)

and all later workers except Weigel (1962). He called them Beds 1, 2,

and 3. According to Weigel, the three strata are easily recognized

throughout the site and fill a shallow sedimentary basin approximately

100 m in diameter. A typical stratigraphic section at Vero and a map of

his various excavations within the site can be found in Weigel (1962).

The total thickness of strata at Vero does not exceed 3 m, of which only

1.5 to 2 m constitute the bone-bearing Strata 2 and 3. Stratum 1 is a late

Pleistocene marine shell marl referred to the Anastasia Formation by

Sellards (1916) that has produced no terrestrial or freshwater vertebrate

fossils. Lying above the Anastasia Formation and separated from it by

an erosional unconformity is Statum 2, consisting of white beach sands

at the base, grading upward into coarse and fine brown stained sands

that become darker toward the top of the bed. The vertebrate fossils

from Stratum 2 are heavily permineralized and include 17 species of

extinct Pleistocene megafauna. The contact between Strata 2 and 3 is

horizontal, and is sharply demarcated by the contrast between the rela-

tively dark brown upper portion of Stratum 2 and the relatively light

colored sands of Stratum 3. Stratum 3 consists of loose white sands,

muck, and peat, banded with decayed plant material. Bones from this

layer are extremely abundant, stained very dark brown, and are barely

permineralized. In his excavations, Weigel found no remains of extinct

vertebrates in Stratum 3, except at his Locality 1, which corresponds

with the area where much of the early fossil material was collected by

Fossil Bats 99

Weills and Ayers. According to Weigel, the beds appeared to be dis-

turbed at Locality 1. A small creek flowed through this locality, appar-

ently cutting through Strata 2 and 3 and mixing fossils from these beds

with more recent artifacts and human bone. In six other stratigraphic

sections at Vero, Weigel found no extinct vertebrates in Stratum 3 and

no evidence of stream channel fills or other reworked deposits.

Owing to the presence of human remains at Vero, the age of the

various strata there has raised much controversy. Weigel (1962:8-9) gave

five radiocarbon dates for Stratum 2, ranging in age from 3,550 years

before present (ybp) to greater than 30,000 ybp. Based on a radiocarbon

date from the top of Stratum 2, Weigel (1962) hypothesized that now

extinct vertebrates may have persisted in Florida until 3,500 years ago.

In retrospect, it appears clear that this date is erroneous, as recent stud-

ies based on extensive series of radiocarbon dates (Meltzer and Mead

1983) suggest that no members of the extinct Pleistocene megafauna

survived in North America beyond 10,000 ybp. Although no radiocar-

bon dates are available from Stratum 3, the absence of extinct Pleisto-

cene megafauna and the predominance of species found in the imme-

diate vicinity at the present time, indicate that this part of the fauna is

Holocene in age. Holocene faunas are uncommon in Florida, or at least

they have rarely been recognized and studied. The late Pleistocene and

early Holocene Devil's Den fauna (Martin and Webb 1974), and the

Nichol's Hammock fauna (Hirschfeld 1968) of unknown but probably

late Holocene age, are the best known. In this paper, vertebrate fossils

from Stratum 2 are regarded as late Wisconsinan (late Pleistocene,

Rancholabrean), while fossils from Stratum 3, in particular the ex-

tremely rich microvertebrate sample from Weigel's Site 3a, are consi-

dered Holocene. Only four bat fossils were recovered from Stratum 2 in

Weigel's excavations, the mandible he referred to Eptesicus sp. and

three specimens of Nycticeius humeralis. All six species identified from

Vero are present in Stratum 3, where the great majority of the bat

remains occur.

METHODS AND MATERIALS

Skulls and postcranial skeletons of all 1 1 species of Recent bats

native to Florida were available for comparison with the fossil material

from Vero. Where possible, specimens from localities in southern Flor-

ida were used for comparisons. Only one maxillary fragment is present

among the Vero chiropteran fossils, while mandibles are slightly more

common. The most important mandibular characters used in differen-

tiating the various species were overall size, number and form of the

premolars, morphology of the molars, length of the ramus, shape of the

coronoid process, and development of the masseteric fossa. Very few

previous studies of bats from Florida Pleistocene localities have included

100 Gary S. Morgan

postcranial material, even though limb elements are often quite abund-

ant in sites where bat fossils occur. In fact, two of the bat species identi-

fied from Vero are based only on postcranial material. The classic study

of Miller (1907) used characters of the humerus, in addition to more

conventional cranial and dental characters, to diagnose many of the

higher taxonomic groups of bats. In her work on the fossil bats from

the Miocene Thomas Farm Site in northern Florida, Lawrence (1943)

discussed the taxonomic importance of the humerus in bats, demon-

strating that almost all Recent genera of North American vespertilionids

could be distinguished using characters of the proximal and distal ends

of the humerus. The radius, especially the proximal end, is useful in

distinguishing between certain groups of bats, although it lacks the large

number of diagnostic characters found in the humerus. Terminology for

various structures on the humerus and proximal end of the radius fol-

lows Vaughan (1959) and Smith (1972). Miller (1907) and Lawrence

(1943) used the terms trochiter and trochin for the greater and lesser

tuberosity of the humerus, respectively. However, since these structures

are homologous with the greater and lesser tuberosity of other mam-

mals, the latter terms will be used in this paper. Dental terminology is

standard for mammals (Szalay 1969). Site names followed by Roman

numerals refer to fossil sites listed in the Florida State Museum verte-

brate paleontology locality files. Cranial and dental measurements were

taken with a Gaertner measuring microscope accurate to 0.01 mm.

Postcranial measurements were taken with dial calipers accurate to 0.10

mm.

All recent comparative material is from the Mammal Collection of

the Florida State Museum, University of Florida (UF). The Vero fossils

are from the Florida Geological Survey Collection, formerly housed in

Tallahassee and now merged with the Florida State Museum Fossil Ver-

tebrate Collection in Gainesville (UF/FGS, catalogue number preceded

byV).

SYSTEMATIC PALEONTOLOGY

Order Chiroptera Blumenbach

Family Vespertilionidae Gray

Eptesicus fuscus (Palisot de Beauvois, 1796)

Referred material— Stratum 2-V7200, partial left mandible with

m3; Stratum unknown-V7201, complete edentulous left mandible.

Recent distribution.— Eptesicus fuscus is one of the most wide-

spread bats in the New World. It occurs throughout the United States,

southern Canada, Greater Antilles, Bahamas, Middle America, and

northern South America. In Florida, the big brown bat has been

Fossil Bats 101

recorded as far south as Englewood in southern Sarasota County on

the west coast and from southern Highlands County in the central por-

tion of the peninsula.

Fossil record. — Vero is the only Pleistocene fauna in Florida from

which Eptesicus fuscus has been reported (listed as Eptesicus sp. by

Webb 1974 and Weigel 1962). I have recently identified E. fuscus in two

additional late Pleistocene (Rancholabrean) faunas from Florida: Arre-

dondo IIA, Alachua County, and Monkey Jungle Hammock, Dade

County. Eptesicus fuscus is the most widespread Pleistocene bat in

North America, having been reported from more than 25 Ranchola-

brean faunas, ranging from Pennsylvania and Florida in the east to

Wyoming and New Mexico in the west, and as far south as Nuevo

Leon, Mexico (Martin 1972). This species has also been reported from a

number of late Pleistocene and Holocene cave deposits in the West

Indies.

Description and comparisons. — Based on their large size, the two

mandibles here referred to Eptesicus fuscus can be distinguished from

all other Florida bats except Eumops and Lasiurus intermedius. The

mandible of Eumops differs in its larger size, reduced coronoid process,

and shallow masseteric fossa. Lasiurus intermedius can be separated

from the fossils by its shorter, more robust mandibular ramus, vertical

mandibular symphysis, smaller triangular coronoid process, shallower

masseteric fossa lacking a strong anterior ridge, and the more reduced

m3. The two mandibles are readily identified as E. fuscus by the long

and relatively slender mandibular ramus, high rounded coronoid pro-

cess, and deep masseteric fossa with a strong anterior ridge. Measure-

ments of the two fossil mandibles compare closely with measurements

of recent E. fuscus from Florida (Table 1).

Discussion. — Although single specimens of Eptesicus fuscus have

been collected from a number of localities in the northern two-thirds of

Florida, it is considered rare in the state. Likewise, E. fuscus is uncom-

mon as a fossil in Florida, having been recorded from only three late

Pleistocene sites based on a small handful of specimens. Most recent

individuals of E. fuscus from Florida have been found in buildings, in

association with colonies of Tadarida brasiliensis. According to Jen-

nings (1958), the absence of E. fuscus from Florida caves is due to the

high humidity and damp walls characteristic of these caves. Eptesicus

fuscus also roosts in hollow trees and rock crevices, the former probably

serving as the preferred roosting site in Florida before the appearance of

man-made structures. A minimum of two individuals of E. fuscus is

represented in the Vero deposit based on the presence of two left man-

dibles. The mandible from Stratum 2 represents one of the few late Pleis-

tocene bat fossils from Vero and was the basis for Weigel's (1962:32)

identification of Eptesicus sp. from the site.

102

Gary S. Morgan

2

in jo mpiM

ui jo qixhiq

ui jo qjpi/ft

ui jo qjgu3[

SS3DOJCI piOUOJOD

jo jqgiaq

ui MOjaq

snuiBJ jo qjdsp

moj qjooj JBinqipuBUi

jo q;§u3f jb|03A(b

3{qipUBUI jo

qjSuai pnoj

o

+|m

o

C^

+1

Tl — CS

+12.

o '

cm

+1— '

'OO —

6 °; <i -•

© w o — w

+ -A

+12.

+'- s

cm cm'

o

T1— 'CM

m

+12,

^' — «/->

r- so •

oo

<* *> wS *? ^ vo ^

r- «r> w <n so ^ so

©' ^

■H2T

</-> r- fi so

+i2t

o 2

SJ

a-

6

3

-ft*-

SJ

-ft c

-S w

r

55 CM

OC-

c

<u

o

4)

o

c

>

'5b

Fossil Bats 103

Lasiurus intermedins H. Allen, 1862

Referred material— Stratum 3-V7202, proximal two-thirds of left

humerus; V7203, proximal end of right radius.

Recent distribution. — Lasiurus intermedius occurs primarily in the

southeastern United States, from South Carolina to Texas, and into

lowland tropical Middle America as far south as Honduras. The yellow

bat is found throughout Florida, with records from as far south as Lee

County on the west coast and Palm Beach, Broward, and Dade counties

on the Atlantic coast (Layne 1974).

Fossil record. — Lasiurus intermedius is known as a fossil only from

Florida. Webb (1974) recorded this species from three late Pleistocene

sites in the state: Haile XIB, Alachua County; Devil's Den, Levy

County; and Reddick I A, Marion County. Martin (1972) also identified

L. intermedius from Arredondo IIA. I have recently identified a mandi-

ble of this species from the Glyptodont Site in Pinellas County.

Description and comparisons. — Based on its large size, the humer-

us referred to Lasiurus intermedius can readily be distinguished from

all other Florida bats except Eumops and Eptesicus. The proximal end

of the humerus differs from that of Eptesicus by the elliptical humeral

head oriented at a 45° angle to the shaft, the more prominent greater

tuberosity, and the smaller lesser turberosity. It is also readily separable

from the humerus of Eumops by its smaller size, almost perfectly ellipti-

cal head, relatively longer and less expanded pectoral ridge, and lack of

a deep groove on the lateral surface of the greater tuberosity. The fossil

is identical in size and morphology to humeri of recent L. intermedius

from Florida (see measurements in Table 2). Although the other large

North American species of Lasiurus, L. cinereus, has been recorded

from Florida on several occasions, it occurs there only as a rare

migrant. The humerus of L. cinereus is larger than the fossil from Vero,

with a broader proximal end and relatively thicker shaft.

As with the humerus, the proximal radius referred to Lasiurus

intermedius needs comparison only with Eumops and Eptesicus. It is

completely unlike the radius of Eptesicus, differing from that genus in

the more robust shaft, considerably shorter ridge extending distally

from the flexor fossa, lack of a deep groove in the articular surface for

the capitulum of the humerus, and the acutely triangular shape of the

proximalmost extension. The fossil radius can be distinguished from

Eumops by its smaller size, more laterally placed flexor fossa, more

slender shaft, relatively smaller articular surface that is rounded in out-

line rather than distinctly triangular, and lack of a deep central groove

on the articular surface. Although essentially identical to the radius of

Lasiurus cinereus, the fossil is somewhat smaller, as is the radius of L.

intermedius.

104 Gary S. Morgan

Discussion. — Lasiurus intermedins roosts almost exclusively in trees

and appears to be closely associated with Spanish moss (Barbour and

Davis 1969). Although the yellow bat is known from more fossil sites

(six) in Florida than any other bat species except Myotis austroriparius,

it is uncommon in the sites where it occurs, generally being represented

by only one or two specimens. The rarity of L. intermedins remains in

fossil sites is not difficult to explain, because the majority of fossil chi-

ropteran faunas in Florida are derived from deposits formed in caves,

fissures, or sinkholes. Yellow bats are not known to enter caves, so

apparently their presence in cave foss*il deposits results from being

brought into caves by predators, most likely the Barn Owl, Tyto alba.

According to Jennings (1958), L. intermedins commonly feeds over

water, thus providing a possible explanation for the presence of the spe-

cies at Vero, based on WeigeFs (1962) interpretation of the site as a

pond or marsh. The two elements of L. intermedins identified from Stra-

tum 3 represent one individual.

Lasiurus cf. seminolus (Rhoads, 1895)

Referred material. — Stratum 3-V7204, nearly complete right humer-

us; V7205-7206, proximal ends of right humeri; V7207, distal end of

left humerus.

Recent distribution. — Lasiurus seminolus occurs primarily in the

southeastern United States from North Carolina to Texas. The Seminole

bat is found throughout most of Florida, as far south as Lee County on

the Gulf Coast and Broward and Dade counties on the east coast.

Fossil record. — This is the first fossil record of Lasiurus seminolus,

assuming the identification is correct. In general, the small species of

Lasiurus have a poor fossil record. Lasiurus borealis has been reported

from only five fossil sites: Reddick IA, Florida (although this could just

as easily represent L. seminolus); Bat Cave, Missouri; Natural Chimneys

and Clark's Cave, Virginia; and Organ-Hedricks Cave, West Virginia

(Kurten and Anderson 1980).

Description and comparisons. — The proximal humeri referred to

Lasiurus cf. seminolus are readily distinguished from all Florida vesper-

tilionids, except Lasiurus, by the elliptical humeral head oriented at a

45° angle to the shaft. They can be separated from the proximal humer-

us of Tadarida brasiliensis, the only similar-sized molossid in Florida,

by the relatively smaller humeral head, reduced greater and lesser tu-

berosities, and less expanded pectoral and medial ridges. The single distal

humerus agrees with Lasiurus and differs from all other Florida bats in

the presence of a deeply excavated olecranon fossa. In addition, the

fossil and Lasiurus can be separated from other Florida vespertilionids

by the prominent distal spinous process. Unlike Tadarida and most

other molossids in which the spinous process is free, the spinous process

Fossil Bats 105

Table 2. Comparison of measurements (in mm) of the humerus of fossil bats

from Vero with Recent Florida bats.1

Mean, standard deviation, sample size, and observed range (in parentheses),

respectively, are given for Recent specimens and fossils of Nycticeius humeralis.

in Lasiurus is attached to the distal articular surface for most of its

length.

The humeri referred to L. cf. seminolus are much smaller than the

corresponding element in L. inter me dius and L. cinereus. There are two

smaller species of Lasiurus known from Florida, L. seminolus and L.

borealis, that have humeri within the size range of the fossils. The

106 Gary S. Morgan

humeri of these two species are broadly overlapping in size (see mea-

surements, Table 2). Examination of a series of humeri of Recent L.

borealis and L. seminolus from Alachua County, Florida revealed no

reliable morphological characters that would distinguish them. These

species are also very similar in external and cranial morphology. They

can, however, be separated by pelage color and the presence of a lach-

rymal ridge in L. borealis. Unfortunately, fossils cannot be definitely

assigned to one species or the other without a skull.

Lasiurus borealis occurs primarily in the northern half of the Flor-

ida peninsula, having been recorded as far south as Hardee County,

although it is not common south of Pasco County. Lasiurus seminolus

occurs sympatrically with L. borealis throughout most of north Florida;

however, the Seminole bat is more widely distributed in the southern

half of the peninsula and is the only small Lasiurus presently found on

the east coast of Florida as far south as Indian River County. Because

the mammalian fauna from Stratum 3 at Vero closely approximates the

Recent fauna of that vicinity, these fossils are tentatively referred to L.

seminolus.

Discussion. — Like L. intermedius, L. seminolus roosts primarily in

clumps of Spanish moss hanging from trees. Although L. borealis is not

necessarily associated with Spanish moss, it too roosts almost exclu-

sively in trees. The tree-roosting habits of these two small species

undoubtedly account for their rarity in the fossil record. The occurrence

of L. borealis or L. seminolus in the Reddick site is probably a result of

Barn Owl predation, while the presence of L. seminolus at Vero is most

likely related to their preference for feeding near water. Although only

four fossils referable to L. seminolus were identified from Vero, three of

these were proximal ends of right humeri representing a minimum of

three individuals.

The proximal half of a right humerus (V7205), identified as Myotis

sp. by Weigel (1962:32) and later referred to M. austroriparius by Webb

(1974:14), is actually referable to Lasiurus cf. seminolus. The fossil

differs from Myotis and agrees with the smaller species of Lasiurus in its

larger size, elliptical humeral head oriented at a 45° angle to the shaft,

and the reduced lesser tuberosity. The left humerus (V7204) referred to

Lasiurus sp. by Weigel (1962:32) and later to L. borealis by Webb

(1974:14) is a right humerus instead. I could not locate the left femur

from Stratum 3 identified as Lasiurus sp. by Weigel.

Nycticeius humeralis (Rafinesque, 1818)

Referred material— Stratum 2-V7228, nearly complete right

mandible with m2-m3; V721 1, nearly complete right humerus; Stratum 3-

V7229, partial edentulous right mandible; V7209, complete right humer-

Fossil Bats 107

us; V7212, distal end of right humerus; VI 603, complete left humerus;

V72 13-72 16, V7231, proximal portions of left humeri; V7230, distal end

of left humerus, V7217, one right and one left femur; Stratum

unknown-V7208, nearly complete edentulous right mandible; V7210,

complete right humerus; V7232, nearly complete left humerus; V7233,

distal end of left humerus.

Recent distribution. — Nycticeius humeralis occurs throughout the

eastern United States and along the Gulf Coast as far south as the state

of Veracruz, Mexico. It is found throughout Florida and is one of the

most common bats of south Florida (Jennings 1958), specimens having

been taken as far south as Collier and Dade counties.

Fossil record.— The specimens of N. humeralis from Vero represent

the first fossil record of the evening bat from Florida. The two speci-

mens from Stratum 2 constitute the second record of N. humeralis from

the Pleistocene of North America, the other occurrence being in Baker

Bluff Cave in northeastern Tennessee (Guilday et al. 1978).

Description and comparisons, — The three mandibles referred to N.

humeralis can be readily distinguished from Myotis and Plecotus by the

presence of only two premolars, from Pipistrellus by their larger size,

and from Eptesicus, Eumops, and Lasiurus intermedius by their consid-

erably smaller size. The mandibles are generally similar in size to the

two smaller species of Lasiurus, but they differ from them in possessing

a longer, more slender mandibular ramus and a larger coronoid process,

and in lacking a deep cleft between the paraconid and metaconid on all

molars. The fossils can be differentiated from Tadarida brasiliensis by

smaller size, presence of a single-rooted rather than a double-rooted P3,

strong rounded coronoid process, deep masseteric fossa, small entoco-

nids on molars, and relatively large incisors (the incisors are small and

compressed in Tadarida). The characters of these three mandibles,

including size, length and depth of ramus, shape of coronoid process

and masseteric fossa, and morphology of the dentition, agree closely

with specimens of N. humeralis (see measurements in Table 1).

Twelve humeri from Vero are referable to Nycticeius humeralis.

They can be separated readily from Eptesicus and Eumops by their

smaller size, and from Tadarida and all species of Lasiurus by the

hemispherical humeral head and reduced distal spinous process. Based

on a number of characters, the humeri can easily be narrowed down to

Myotis, Pipistrellus, and Nycticeius. The most reliable character on the

proximal end of the humerus for separating the fossils from Myotis and

Pipistrellus is the more prominent medial ridge extending ventrally from

the lesser tuberosity and producing a larger fossa or concavity for the

origin of the lateral head of the triceps muscle. In a posterior view of the

proximal end, that portion of the humerus medial to the pectoral ridge

108 Gary S. Morgan

is wider in Nycticeius as a result of the better developed medial ridge.

The distal half of the posterior surface of the humeral shaft is distinctly

flattened in Nycticeius and the fossils, but is round in cross-section in

the other two species. On the distal end of the humerus, the lateral edge

of the articular surface (lateral epicondyle of the capitulum) extends

lateral to the edge of the shaft in Myotis and Pipistrellus, but is in line

with the shaft in the fossils and Nycticeius. Nycticeius and the fossils

possess a prominent notch immediately proximal to the lateral edge of

the capitulum that extends around the lateral edge almost to the ante-

rior surface of the humeral shaft. This notch is not as well developed in

the other two species. Finally, in Nycticeius there is a well developed,

rounded tubercule on the lateral edge of the shaft just proximal to the

notch, which is absent in P. subflavus and M. austroriparius. The region

medial to the medial epicondyle (trochlea) is relatively large in Myotis,

somewhat smaller in Pipistrellus, and very reduced in the fossils and

Nycticeius. Therefore, although the humeri in these three species are

superficially very similar, a number of characters can be used to separ-

ate them, and the fossils are clearly referable to N. humeralis (see mea-

surements on humeri in Table 2).

Discussion. — Nycticeius humeralis is the most abundant fossil bat

in the Vero site, with 17 identifiable elements representing a minimum

of seven individuals. Evening bats roost primarily in buildings, hollow

trees, and under the loose bark of trees. They seem to show a preference

for cypress trees and are the common bat in Florida near cypress stands

(Jennings 1958). Like the species of Lasiurus, N. humeralis is not known

to enter caves, thus explaining the absence of this species from other

Pleistocene sites in Florida that have produced bat fossils. Apparently,

TV. humeralis is not as subject to raptor predation as is Lasiurus, since

species of the latter genus do on occasion appear in cave fossil deposits.

Identification of fossil cypress, Taxodium distichum, from Stratum 3

(Berry 1917) supports Weigel's statement (1962:42) that there was a

cypress pond in the vicinity of the Vero site. The presence of cypress

trees and the preference of Nycticeius humeralis for roosting in cypress

offer an explanation for the abundance of evening bat fossils at Vero.

Family Molossidae Gill

Tadarida brasiliensis (I. Geoffroy St. -Hilaire, 1824)

Referred material— Stratum 3-V7219, proximal end of left humer-

us; Stratum unknown-V7218, nearly complete left mandible with m2-m3.

Recent distribution. — Tadarida brasiliensis is found primarily in

the southern and western United States and then southward through

Middle America, the West Indies, and much of South America. Brazil-

Fossil Bats 109

ian free-tailed bats occur throughout Florida, and according to Layne

(1974) the species is the most successful bat in southern Florida, where

it has been recorded as far south as Dade and Collier counties.

Fossil record. — Tadarida brasiliensis is known from three other

fossil sites in eastern North America, two in Florida (Reddick IA and

Nichol's Hammock) and the other in Mammoth Cave, Kentucky (out-

side the present range of the species). There are numerous Pleistocene

records of T. brasiliensis from the southwestern United States and the

West Indies.

Description and comparisons. — The mandible referred to T. brasil-

iensis is distinguishable from Eumops by its considerably smaller size

and from Pipistrellus by its considerably larger size. The fossil differs

from Lasiurus, Eptesicus, Nycticeius, Myotis, and Plecotus in the

reduced coronoid process, shallow masseteric fossa, small compressed

incisors, double-rooted p3, and larger ni3 relative to m2. Myotis and

Plecotus both have the same number of premolar alveoli as Tadarida,

but they possess single-rooted P2 and p3, while Tadarida lacks p2 and

has a double-rooted p3. Based on the above combination of characters,

the fossil mandible is readily identified as T. brasiliensis (see measure-

ments, Table 1).

Although poorly preserved and lacking the lesser tuberosity, the

proximal humerus here referred to T brasiliensis is identifiable. Based

on its small size and elliptical humeral head, the humerus can be distin-

guished from that of all Florida bats except the two small species of

Lasiurus and Tadarida. The humerus is identified as T brasiliensis by

its broader and shorter pectoral ridge and greater distal extension of the

medial ridge.

Discussion. — Only two fossils of T. brasiliensis, probably represent-

ing a single individual, have been identified from Vero. The Brazilian

free-tailed bat is rare as a fossil in Florida, having been recorded from

only three sites based on less than ten specimens. At the present time, T.

brasiliensis in Florida roosts almost exclusively in man-made structures,

such as in houses and under bridges (Jennings 1958). Although it has

been observed in small numbers in several caves in Marion County,

Florida (R. Franz, pers. comm.), these probably do not represent roost-

ing colonies. In marked contrast to the southwest, where T. brasiliensis

inhabits caves in colonies sometimes numbering into the millions, it is

not known to roost in caves in the southeastern United States. Appar-

ently, the warm humid atmosphere of Florida caves offers an unsuitable

environment for roosting colonies (Jennings 1958). Tadarida brasiliensis

in Florida has also been observed roosting under the dead fronds of

palm trees in Lee and Charlotte counties in southwestern Florida, and

in hollow mangrove trees in the Tampa Bay area (Jennings 1958). Palm

1 10 Gary S. Morgan

trees provide the natural roosting site for many species of Neotropical

molossids, and it seems reasonable to hypothesize a similar roosting

ecology for T. brasiliensis in Florida prior to the extensive construction

of buildings. The probable tree-roosting habits of T. brasiliensis, coupled

with their extremely rapid flight, would limit predation and help to

explain their absence from most Florida fossil sites.

Eumops glaucinus floridanus (G. M. Allen, 1932)

Referred material— Stratum 3-V7222, partial edentulous left man-

dible; V7224, proximal end of right radius; V7226, proximal end of left

radius; V7227, one proximal and one distal end of femur; Stratum

unknown- V7220, right maxilla with P4-M3; V7221, left mandible with

p3-ni3; V7223, proximal end of right humerus; V7225, proximal half of

left radius.

Recent distribution. — Eumops glaucinus has the most restricted

distribution of any bat in the United States, being known only from

Charlotte and Dade counties in southernmost Florida. Wagner's mastiff

bat also occurs in tropical America from southern Mexico south

through Middle America, much of South America, and Cuba and

Jamaica in the Greater Antilles. The species has a disjunct distribution,

as it is not known to occur between southern Florida and southern

Mexico.

Fossil record. — The fossil record of E. glaucinus is restricted to

Florida, where it is known from Vero, Monkey Jungle Hammock (Mar-

tin 1977), and the late Pleistocene Melbourne fauna, Brevard County

(Allen 1932; Ray et al. 1963). The fossil records from Brevard County

and Indian River County (this paper) extend the known range of the

species in Florida some 200 km north.

Descriptions and comparisons. — The cranial and postcranial ele-

ments here referred to E. glaucinus are from a very large bat, and thus

need only be compared with the three largest species found in Florida —

E. glaucinus, Eptesicusfuscus, and Lasiurus intermedius. A maxilla and

partial rostrum agree with E. glaucinus and differ from E. fuscus and

L. intermedius as follows: presence of a tiny peg-like P3, stronger hypo-

cone on M1 and M2, lack of a deeply incised nasal notch, vertical slit-

like infraorbital foramen, and vertical orientation of rostrum dorsal to

orbit, reflecting deep, laterally compressed snout (rostrum is dorsoven-

trally flattened in the two large vespertilionids). The mandible with p3-

m3, can be differentiated from E. fuscus and L. intermedius as follows:

presence of only two tiny incisors that are crowded between the canine

and mandibular symphysis, double-rooted p3, P3 and p4 subequal in size,

and the posterior margins of trigonid and talonid on molars not at right

angles to long axis of tooth row.

Fossil Bats 1 1 1

The proximal humerus referred to E. glaucinus can be readily dis-

tinguished from all other Florida bats by its very large size, teardrop-

shaped humeral head oriented at a 45° angle to the shaft, short

expanded pectoral ridge, and proximal extension of the greater tuberos-

ity. The three radii are identified as E. glaucinus by the large, deep

flexor fossa on the anterior surface just distal to the proximal articula-

tion, the acutely triangular proximal end, and the strongly concave

articular surface with a deep central groove for reception of the medial

portion of the capitulum on the distal end of the humerus. The proximal

and distal femur can be separated from all Florida vespertilionids by the

small femoral head relative to the greater and lesser trochanters, rela-

tively broader distal end, and more widely separated articular condyles.

Among Florida bats, only Tadarida brasiliensis has femora with a sim-

ilar morphology, but their small size eliminates them immediately.

Discussion.— Even though Eumops glaucinus is the second most

abundant bat at Vero based on the total number of elements present

(nine), a minimum of only two individuals is represented. The presence

of Eumops glaucinus at Vero is of particular interest since this site is

over 100 km north of the northernmost locality from which recent indi-

viduals of this species have been collected. A single fossil mandible of E.

glaucinus is known from the Melbourne Site, located approximately 50

km north of Vero (Allen 1932; Ray et al. 1963). Until recently, living

specimens of E. glaucinus floridanus had been collected only from man-

made structures in the Miami area of Dade County in extreme

southeastern Florida. Belwood (1981) discovered a small colony of E. glau-

cinus roosting in a hollow long-leaf pine, Pinus palustris, near Punta

Gorda in Charlotte County on the southwest coast of Florida. Hollow

trees appear to be the preferred natural roosting site of this species

(Belwood 1981). The discovery of E. glaucinus in Charlotte County

extends the modern range of the species in Florida 200 km westward

and 100 km northward of Miami. With the addition of the three fossil

records from Florida discussed above, the species is now known from

three different localities in south Florida and two localities from the

central portion of the state (Fig. 1).

Martin (1977) suggested that the presence of Eumops glaucinus in

central Florida during the late Pleistocene represented a northward shift

in winter isotherms indicative of tropical or subtropical conditions.

Belwood 's recent discovery of E. glaucinus in a part of Florida and in

an ecological situation from which the species was previously unknown

suggests that our knowledge of this bat is far from adequate. If Eumops

did extend its range northward in response to warmer climates, why is it

known in central Florida only from a late Wisconsinan site (Melbourne)

in which climatic conditions were presumably drier and cooler than at

1 12 Gary S. Morgan

present, and a Holocene site (Vero) in which the climatic conditions

were essentially similar to those at present? It would seem more likely

that Eumops would have been found in one of the Sangamonian inter-

glacial sites (Reddick, Haile, Arredondo, etc.), at a time during which

climates were probably somewhat more tropical than they presently are.

As noted by Eger (1977) and Koopman (1971), the endemic Florida

subspecies, Eumops glaucinus floridanus, is the most distinct form of

the species. The Florida animal is characterized by its larger size, a fea-

ture also seen in the fossil representative of the species from Florida (see

measurements in Table 3). According to Eger (1977), all Neotropical

representatives of E. glaucinus, including those from the West Indies,

are referable to the nominal subspecies, while only the Florida popula-

tion is recognizable as a distinct subspecies. Baker and Genoways (1978)

suggested the possibility that E. glaucinus invaded Florida from Cuba, a

distance of only 200 km. However, the strong mainland Neotropical

component of Florida's Pleistocene fauna, and the total lack of any

other West Indian bats in the state, suggest strongly that the present

distribution of E. glaucinus resulted from a warmer interglacial period

when the Neotropical fauna was continuous around the Gulf Coast.

Two other bats found in Florida during the late Pleistocene, Desmodus

stocki and Mormoops megalophylla, also reflect this mainland Neotrop-

ical influence.

DISCUSSION

The fossil bat fauna from Vero is significant for several reasons.

First, more species of bats (six) are represented at Vero than in any

other fossil vertebrate fauna yet described from Florida. The two most

diverse fossil chiropteran faunas from Florida listed by Webb (1974:14)

were Reddick 1 A, Marion County, with five species — Desmodus stocki,

Myotis austroriparius, Lasiurus borealis, L. intermedius, and Tadarida

brasiliensis — and Devil's Den, Levy County, with four species — M. aus-

troriparius, M. grisescens, Pipistrellus subflavus, and L. intermedius.

Second, among the ten or so Pleistocene and Holocene vertebrate fau-

nas in Florida that contain abundant bat fossils, only the Vero deposit

represents a depositional environment other than a cave, fissure, or

sinkhole. The fossil deposits at Reddick consist of unconsolidated sedi-

ments filling caverns and solution pipes in the surrounding Eocene

limestones. A cave-dwelling species, Myotis austroriparius, accounts for

the great majority of bat remains at Reddick. The Devil's Den site is a

water-filled sinkhole and cave system, presumably inhabited by the bats

during a period of lower sea level and water tables in the late Wisconsi-

nan and early Holocene. Cavernicolous bats also predominate at Devil's

Den. In fact, all of the major North American Pleistocene sites listed by

Kurten and Anderson (1980) that contain large bat faunas were depos-

ited in caves and are dominated by cave-inhabiting species.

Fossil Bats

113

AELBOURNE

VERO

Fig. 1. Pleistocene and Recent occurrences of Eumops glaucinus floridanus in Florida.

Asterisks (*) indicate Recent records, daggers Of) and name of fauna indicate Pleistocene

records.

Based on studies of the sediments, fossil plants, and fossil verte-

brates, Weigel (1962) concluded that the fossil deposits at Vero repre-

sented a pond or marsh habitat. Berry (1917) studied the fossil plants

from Stratum 3 at Vero. Among the more informative plants he identi-

fied were cypress, Taxodium distichum, and three species of obligate

pond inhabitants: water lettuce, Pistia; pond apple, Anona glabra', and

water shield, Brasenia purpurea. A number of the other fossil plants

from Vero also have aquatic tendencies. According to Weigel (1962),

almost 50% of the vertebrates from Strata 2 and 3 are forms associated

114

Gary S. Morgan

o

«-

>

1

3

I

5

ui jo qipiM

ui jo qjSusi

ui jo qipiM

ui jo qjSusi

ui Mopq

snuiBJ jo qjdap

:N jo qjpiM

j/\l jo qj§u3|

,p\ joq;piM

W jo qjguai

M-M JoqjSuai

moj qiooj Xjbiiixbui

JO qjSu3l lU\OQ\\U

o

+|r-

oo

o

©

00

o

o

o

© s

O

°" 2

1 2

00

c

OS

>

Fossil Bats 1 15

with aquatic habitats, including such obligate freshwater species as gar,

Lepisosteus; bowfin, Amia\ Siren and Amphiuma; two species of ranid

frogs; Alligator; two species of water snake, Nerodia; four species of

kinosternid mud turtles; a number of species of ducks, rails, herons, and

egrets; and the round-tailed muskrat, Neofiber alleni. The large compo-

nent of aquatic vertebrates supports the sedimentological and paleobo-

tanical evidence that the deposits were formed in a shallow freshwater

pond or marsh. Based on the presence of a number of strictly terrestrial

forms in the fauna, several other habitats were certainly present in the

immediate vicinity, including mesic hammock and pine flatwoods.

The presence of a diverse bat fauna at Vero is somewhat difficult to

explain in the context of the freshwater pond or marsh habitat sug-

gested by Weigel (1962). In Weigel's scenario of Vero (1962:49), "Bats

flew over the pond and marsh in search of insects. . . ." It is true that bats

commonly fly over open water, both in search of insects and to drink,

but bats are usually absent or extremely rare in fossil deposits sampling

such habitats. A large number of Pleistocene sites in peninsular Florida

were deposited in marshes, swamps, or fluvial environments, several of

which have abundant microvertebrate samples. Yet, except for Vero,

only two specimens of fossil bats are known from Florida sites sam-

pling such habitats — the type specimen of Molossides floridanus

{-Eumops glaucinus floridanus) from Melbourne, and a mandible of

Lasiurus intermedins from the Glyptodont Site, Catalina Gardens,

Pinellas County.

The roosting ecology of the bats recorded from Vero provides some

insight into the problem, as all six species are known to roost in trees.

None of the bats from Vero normally roost in caves in the southeastern

United States. In contrast, the two most abundant and widespread bats

found as fossils in northern Florida cave and fissure deposits are Myotis

austroriparius and Pipistrellus subflavus, both of which roost in caves at

certain times of the year. The absence of these two species from the

modern fauna of south Florida, except for accidental occurrences, is

almost certainly related to the absence of dry caves south of Citrus and

Marion counties.

Unlike any other Quaternary bat fauna known from Florida, Vero

offers a unique view of the late Pleistocene and Holocene bat fauna

associated with riparian habitats. Generally, tree-roosting bats and bats

associated with freshwater habitats are rare or totally absent from fossil

sites deposited in caves, fissures, or sinkholes in north Florida, as in the

case of Nycticeius humeralis and Eumops glaucinus. The exact mode of

deposition of the bat fossils at Vero is still a matter of speculation.

Perhaps the bat carcasses accumulated in hollow trees alongside the

pond or marsh and were eventually washed in when the trees fell. The

great abundance of other small mammals in the Vero deposit, especially

shrews and small rodents, suggests the possibility of a raptor roost in

the vicinity of the pond, most likely that of Tyto alba.

1 16 Gary S. Morgan

Layne (1974) recorded seven bat species from Florida south of

Lake Okeechobee. Although Vero is slightly north of this area, it is

located in the southern half of the Florida peninsula. Among these

seven species, Myotis austroriparius and Pipistrellus subflavus almost

certainly do not breed in south Florida, and Layne considered their

occurrence in the region to be accidental. The remaining five species

comprise the native chiropteran fauna of south Florida: Lasiurus semi-

nolus, L. intermedins, Nycticeius humeralis, Tadarida brasiliensis, and

Eumops glaucinus — all of which are known from Vero. All six species

of bats from Vero, including Eptesicus fuscus, might be expected to

occur in that vicinity at the present time, with the possible exception of

Eumops glaucinus. Apparently, the chiropteran fauna of south Florida

had become established by the early Holocene and has remained essen-

tially unchanged to the present time.

ACKNOWLEDGMENTS.— I thank R. Franz, A. E. Pratt, and

K. T. Wilkins for their helpful comments on an earlier draft of this

manuscript. L. Wilkins and C. A. Woods allowed me access to the Flor-

ida State Museum Recent mammal collections. Shaen Williams typed

the final draft of the paper. This is Univeristy of Florida Contribution

to Vertebrate Paleontology No. 225.

LITERATURE CITED

Allen, Glover M. 1932. A Pleistocene bat from Florida. J. Mammal. 73:256-259.

Baker, Robert J., and H. H. Genoways. 1978. Zoogeography of Antillean bats.

Pp. 53-97 in F. B. Gill (Ed.). Zoogeography in the Caribbean. Spec. Publ.

Acad. Nat Sci. Phila. 73:1-128.

Barbour, Roger W., and W. H. Davis. 1969. Bats of America. Univ. Presses

Kentucky, Lexington. 286 pp.

Belwood, J. J. 1981. Wagner's mastiff bat, Eumops glaucinus floridanus,

(Molossidae) in southwestern Florida. J. Mammal. 62:411-413.

Berry, Edward W. 1917. The fossil plants from Vero, Florida. J. Geol. 25:661-666.

Eger, Judith L. 1977. Systematics of the genus Eumops (Chiroptera: Molossi-

dae). Royal Ont. Mus. Life Sci. Contrib. 770:1-69.

Guilday, John E., H. W. Hamilton, E. Anderson and P. W. Parmalee. 1978.

The Baker Bluff cave deposit, Tennessee, and the late Pleistocene faunal

gradient. Bull. Carnegie Mus. Nat. Hist. 77:1-67.

Hirschfeld, Sue E. 1968. Vertebrate fauna of Nichol's Hammock, a natural trap.

Q.J. Fla. Acad. Sci. 37:177-189.

Jennings, William L. 1958. The ecological distribution of bats in Florida.

Unpubl. Ph.D. dissert., Univ. Florida, Gainesville. 126 pp.

Fossil Bats 117

Koopman, Karl F. 1971. The systematic and historical status of the Florida

Eumops (Chiroptera, Molossidae). Am. Mus. Novit. 2478:\-6.

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

Columbia Univ. Press, New York. 442 pp.

Lawrence, Barbara. 1943. Miocene bat remains from Florida, with notes on the

generic characters of the humerus of bats. J. Mammal. 24:356-369.

Layne, James N. 1974. The land mammals of South Florida. Pp. 386-413 in P.

J. Gleason (Ed.). Environments of South Florida: Present and Past. Miami

Geol. Soc. Mem. 2:1-452.

Martin, Robert A. 1972. Synopsis of late Pliocene and Pleistocene bats of North

America and the Antilles. Am. Midi. Nat. 57:326-335.

. 1977. Late Pleistocene Eumops from Florida. Bull. New Jersey

Acad. Sci. 22:18-19.

, and S. D. Webb. 1974. Late Pleistocene mammals from the Devil's

Den fauna, Levy County. Pp 114-145 in S. D. Webb (Ed.). Pleistocene

Mammals of Florida. Univ. Presses Florida, Gainesville. 270 pp.

Meltzer, David J., and J. I. Mead. 1983. The timing of late Pleistocene extinc-

tions in North America. Quat. Res. (NY) 79:130-135.

Miller, Gerrit S., Jr. 1907. The families and genera of bats. U.S. Nat. Mus. Bull.

57:1-282.

Ray, Clayton E. 1957. A list, bibliography, and index of the fossil vertebrates of

Florida. Fla. Geol. Surv. Spec. Publ. 5:1-175.

, 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. 44:373-395.

Sellards, E. H. 1916. Human remains and associated fossils from the Pleistocene

of Florida. Fla. Geol. Surv. 8th Annu. Rep. pp. 122-160.

. 1917. On the association of human remains and extinct vertebrates

at Vero, Florida. J. Geol. 25:4-24.

Smith, James D. 1972. Systematics of the chiropteran family Mormoopidae.

Univ. Kans. Mus. Nat. Hist. Misc. Publ. 56:1-132.

Szalay, Frederick S. 1969. Mixodectidae, Microsyopidae, and the insectivore-

primate transition. Bull. Am. Mus. Nat. Hist. 740:193-330.

Vaughan, Terry A. 1959. Functional morphology of three bats: Eumops,

Myotis, and Macrotus. Univ. Kans. Publ. Mus. Nat. Hist. 72(1): 1-1 53.

Webb, S. David. 1974. Chronology of Florida Pleistocene mammals. Pp. 5-31 in

S. D. Webb (Ed.). Pleistocene Mammals of Florida. Univ. Presses Florida,

Gainesville. 270 pp.

Weigel, Robert D. 1962. Fossil vertebrates of Vero, Florida. Fla. Geol. Surv.

Spec. Publ. 70:1-59.

Accepted 27 May 1984

New Trechine Beetles (Coleoptera: Carabidae)

from the Appalachian Region

Thomas C. Barr, Jr.

School of Biological Sciences,

University of Kentucky, Lexington, Kentucky 40506

ABSTRACT. — New taxa of Pseudanophthalmus from caves in south-

central Kentucky are described and illustrated: P. menetriesi campes-

tris, P. simulans, pilosus, glo biceps, transfluvialis , cerberus cerberus,

cerberus completus, darlingtoni darlingtoni, darlingtoni persimilis, and

pubescens intrepidus. Two new species of Trechus — T (Trechus) cali-

ginis and T (Microtrechus) inexpectatus — are described and illustrated

from Camp Creek Bald, North Carolina/ Tennessee.

The trechines are a group of small carabid beetles that includes

many species restricted to cool, moist microhabitats. In the Appalachian

region they are abundant in the Unaka mountain province of western

North Carolina and adjacent Tennessee as well as in caves of the

Appalachian Valley and Interior Low Plateaus (Barr 1979a, 1980, 1981).

The following previously undescribed taxa are from both Unaka and

Interior Low Plateau regions.

Pseudanophthalmus menetriesi campestris, new subspecies

Fig. 1

Etymology. — Latin campestris, "of the plain."

Description. — Differs from nominate Pseudanophthalmus menetriesi

(Motschulsky) in narrower elytra, L/W for Mammoth Cave 1.56 ± .04

(N = 45) and for Walnut Hill Cave (type locality) campestris 1.60 ± .04

(N = 45, P = .01); humeri less angular, striae deeper, intervals subcon-

vex, pubescence of pronotum and elytral discs relatively dense. Length

4.6 - 5.7, mean 5.0 ± 0.1 mm (N = 65). Aedeagus about as in P. m.

menetriesi.

Type series. — Holotype male (American Museum of Natural His-

tory) and 41 paratypes, Walnut Hill Cave, 3.3 km S Park City, Barren

Co., Kentucky (Park City 7%' Quadrangle), 18 March 1966, T. C. Barr,

R. M. Norton, T. G. Marsh. Measurements of holotype (mm): total

length 5.20, head 0.90 long X 0.82 wide, pronotum 0.98 long X 1.05

wide, elytra 2.93 long X 1.83 wide, antenna 3.24 long.

Distribution. — This is the geographic race from the Sinkhole Plain

south of Mammoth Cave National Park described but not named by

Barr and Crowley (1981). It ranges from the vicinity of Hardyville, Hart

Brimleyana No. 1 1:1 19-132, October 1985 1 19

120 Thomas C. Barr, Jr.

County, through Barren County to Smiths Grove and Three Forks in

Warren County, Kentucky, hybridizing with nominate menetriesi in

caves at Park City and Cave City.

Pseudanophthalmus simulans, new species

Fig. 2

Etymology. — Latin simulans, "simulating."

Diagnosis. — Closely similar to menetriesi but larger, pronotum wider,

elytra pubescent over entire disc, elytral chaetotaxy +++, humeri more

pronounced, with slight posthumeral sinuation in margin; aedeagus

larger than that of menetriesi, its apex much wider in dorsal view.

Description.— Length 5.2-6.4, mean 5.8 ± 3.0 (N = 19), larger than

menetriesi (P = .01). Form robust, convex, pubescent, elytral micro-

sculpture not pruinose. Head as wide as long, labrum doubly emargin-

ate. Pronotum about 0.9 as long as wide, disc pubescent, sides curved

in apical 2/3, barely sinuate in basal 1/7, hind angles as in menetriesi.

Elytra with humeri more prominent, humeral serrations larger and

deeper, margin with shallow posthumeral sinuation; disc densely pubes-

cent, posterior discal seta present. Aedeagus as in Figure 2, similar to

that of menetriesi but significantly larger.

Type series. — Holotype male (American Museum of Natural His-

tory) and 17 paratypes, Cub Run Cave, at Cub Run, Hart Co., Ken-

tucky (Cub Run iyi Quadrangle), 18 November 1964, T. C. Barr, W.

M. Andrews; one paratype, same cave, 29 December 1956, L. Hubricht.

Measurements of holotype (mm): total length 6.18, head 0.93 long X

0.93 wide, pronotum 1.08 long X 1.18 wide, elytra 3.50 long X 2.17

wide, antenna 3.75 long.

Distribution. — The species is an isolate known only from the type

locality. Extrinsic isolation of Cub Run Cave is a reasonable hypothesis,

supported by absence of any vicar taxa related to Neaphaenops tell-

kampfi (Erichson), P. striatus (Motschulsky), or P. pubescens (Horn),

which coexist with P. menetriesi. By similar reasoning, this species and

the next three (described below) are judged to have arisen along with P.

menetriesi through multiple cave colonization by a common ancestor.

Pseudanophthalmus pilosus , new species

Fig. 3

Etymology. — Latin pilosus, "hairy."

Diagnosis. — Closest to menetriesi, differing in more convex, vaulted

elytra, flat near middle of disc and with abruptly declivous sides; elytral

disc uniformly pubescent and not pruinose; 6 discal striae usually deeply

impressed and seriate-punctulate, intervals subconvex, discal setae +0+

New Trechine Beetles

121

Fig. 1-5. Aedeagi of Pseudanophthalmus spp., left lateral view. 1: menetriesi

campestris, n. subsp. 2: simulans, n. sp. 3: pilosus, n. sp. 4: globiceps, n. sp. 5:

transfluvialis, n. sp.

in about half the individuals in most populations (except the northern-

most, which are +++); pronotum wider (L/W 0.80-0.85); aedeagus

straighter in middle portion, apex finely reflexed.

Description. — With the characters of P. menetriesi except as noted

above. Length 4.4-5.7, mean 5.5 ± 0.3 mm (N = 118). Pronotum disc

evidently pubescent, hind angles more consistently sharp, sometimes

1 22 Thomas C. Barr, Jr.

with small secondary angles on base. Elytra with one or both posterior

discal setae absent in about half of specimens examined (+0+). Aedea-

gus 0.71-0.80, mean 0.75 ± .04 mm long (N = 18).

Type series.— Holotype male (American Museum of Natural His-

tory) and 18 paratypes, Bland Cave, 1.8 km NW Spurrier on N side

Akers Valley, Hardin Co., Kentucky (Millerstown 7%' Quadrangle), 28

December 1962, T. C. Barr, R. A. Kuehne. Measurements of holotype

(mm): total length 5.52, head 0.93 long X 0.74 wide, pronotum 0.99 long

X 1.09 wide, elytra 3.10 long X 2.02 wide, antenna 3.97 long.

Distribution.— Limited to northwestern Hart County, Kentucky,

north of the Hart County Ridge, an extrinsic barrier (Barr 1979b), and

southwestern Hardin County (to Star Mills).

Pseudanophthalmus globiceps , new species

Fig. 4

Etymology. — Latin globus, "round," + -ceps, "head."

Diagnosis.— Resembles pilosus in very convex, vaulted elytra with

dense discal pubescence, but head more rounded and pronotum propor-

tionately wider, base more nearly rectilinear than in other species close

to menetriesi, barely emarginate behind hind angles; aedeagus with

median lobe more strongly arcuate than other species of the menetriesi

series.

Description.— Length 4.8-5.8, mean 5.3 ± 0.3 mm (N = 10). Head

and pronotum as described above. Elytra broad and quite convex but a

little less so than in pilosus; 6-7 finely impressed and strongly punctulate

striae, sutural stria deepest, intervals flat or nearly so, discal chaetotaxy

normal (+++); subhumeral margin slightly sinuate, greatest width behind

middle. Aedeagus of a paratype 0.73 mm, about as in pilosus but a little

narrower and more strongly arcuate, apex not briefly reflexed.

Type series. — Holotype male (American Museum of Natural His-

tory) and 3 paratypes, Barnes Smith Cave, 5.7 km N Hinesdale, Hart

Co., Kentucky (Canmer ll/2' Quadrangle), 30 December 1956, T. C.

Barr; 6 additional paratypes, same cave, 30 August 1963, T. C. Barr, J.

R. Holsinger; 2 July 1980, T. C. Barr, Jr., T. C. Barr, III. Measure-

ments of holotype (mm): total length 5.84, head 1.05 long X 0.98 wide,

pronotum 1.13 long X 1.23 wide, elytra 3.39 long X 2.26 wide, antenna

3.63 long.

Distribution. — Known only from the type locality, a cave at the

south base of the Hart County Ridge, where P. globiceps coexists with

Neaphaenops tellkampji from the Mammoth Cave region and P. orien-

talis Krekeler, from the Greensburg area. There are few other caves in

the vicinity; an interaction with P. menetriesi is perhaps feasible,

although no suitable intervening caves have been sampled. The unusu-

ally convex elytra suggest common ancestry with P. pilosus, however,

not menetriesi.

New Trechine Beetles 1 23

Pseudanophthalmus transfluvialis, new species

Fig. 5

Etymology. — Latin trans-, "across," + fluvialis, "pertaining to a

river."

Diagnosis. — Closest to P. menetriesi but heavily pubescent, sides of

head a little less rounded, pronotum narrower, prehumeral borders less

oblique and humeri more prominent, elytral intervals subconvex instead

of flat, striae distinctly impressed, disc not pruinose.

Description.— Length 4.6-5.8, mean 5.2 ± 0.3 mm (N = 32). Head

about 0.15-0.18 longer than wide. Pronotum about as wide as long,

mean L/W 0.96 ± .03 (N = 32); disc pubescent, hind angles as in mene-

triesi. Elytra moderately convex (not vaulted as in pilosus or glo biceps,

nor flattened as in nominate menetriesi), about 1.6 times longer than

wide, pubescent; humeri stronger than in menetriesi (either subspecies),

intervals subconvex, striae deeper, strongly punctured; chaetotaxy nor-

mal (+++) and microsculpture not pruinose. Aedeagus 0.67-0.76, mean

0.69 ± .03 mm (N = 8), closely similar to that of menetriesi .

Type series. — Holotype male (American Museum of Natural His-

tory) and 5 paratypes, McGinnis Cave, 4.2 km SW Bowling Green,

Warren Co., Kentucky (Bowling Green South 1%' Quadrangle), 26 Sep-

tember 1949, J. M. Valentine, W. B. Jones, I. C. Royer. Measurements

of holotype (mm): total length 5.52, head 1.01 long X 0.86 wide, prono-

tum 1.09 long X 1.09 wide, elytra 3.16 long X 1.99 wide, antenna 3.43

long, aedeagus 0.76 long.

Distribution. — Described on 43 specimens from the type cave and

other caves in Bowling Green (Bypass, Horseshoe, State Trooper); the

westernmost limit is Wheeler Cave, 3.3 km northeast of South Union, in

eastern Logan County. The species appears to have a range that runs

along the base of the Dripping Spring escarpment from Bowling Green

to Wheeler Cave; it does not extend south into the range of P. princeps

(see Barr 1979b), but coexists with Neaphaenops meridionalis Barr and

P. loganensis Barr in all of the caves where it has been collected (Barr

1979b). The trivial name refers to the barrier status of Barren River at

Bowling Green, where the river separates P. menetriesi, P. striatus, P.

pubescens, and N. tellkampfi from P. transfluvialis , P. loganensis, and

N. meridionalis.

Pseudanophthalmus cerberus cerberus, new species and subspecies

Fig. 6

Etymology. — Named for Cerberus, the mythical dog guarding the

gates of Hades, usually depicted with three or more heads; P. cerberus is

the most widely distributed of three closely similar species in south-

central Kentucky.

124 Thomas C. Barr, Jr.

Diagnosis. — Similar to P. menetriesi in having hind angles of the

pronotum tipped forward instead of produced backward (as in striatus

and darlingtoni), differing in impunctate or vaguely punctulate elytral

striae, strongly pruinose elytral disc, and more oblique prehumeral

borders; elytral chaetotaxy +0+, the posterior discal puncture absent.

Description. — Length 4.6-5.8, mean 5.2 ± 0.3 mm (N = 77). Robust,

subconvex, pubescent, elytral apical groove vestigial (diagnostic for

menetriesi group). Head and mandibles less slender than in menetriesi,

dorsum subglabrous; labrum with low median lobe. Pronotum 0.83-0.86

as long as wide, widest in apical third, sides rounded apical half then

convergent to small, approximately right, sharp, and reflexed hind

angles; anterior angles prominent, base 0.75 maximum width, apex 0.87

as wide as base; small secondary angles of base internal to very deep,

oblique, basolateral impressions; median antebasal impression quite

deep and linear; disc convex, with long, rather sparse pubescence; ante-

rior marginal setae placed before greatest width, posterior marginals

before hind angles. Elytra elongate-oval, 0.55-0.64 longer than wide;

humeri prominent, setose with moderately coarse serrations, prehumeral

borders oblique to mid-line; disc densely pubescent and strongly prui-

nose overall; striae rather shallow but well-defined and regular, impunc-

tate, intervals weakly subconvex; posterior discal setae constantly absent

(+0+). Aedeagus 0.97-1.13, mean 1.05 ± .05 mm long, much larger than

that of menetriesi (or the species described above in this paper); basal

bulb larger and deflexed, median lobe moderately arcuate, briefly pro-

duced, apex spout-shaped and not reflexed; transfer apparatus typical

for menetriesi group.

Type series. — Holotype male (American Museum of Natural His-

tory) and 76 paratypes, Rhoton Cave, 3.3 km SW Hestand on N side

valley of Sweetwater Creek, Monroe Co., Kentucky (Tompkinsville 7%'

Quadrangle), 7 August 1963, T. C. Barr, R. A. Kuehne. Measurements

of holotype (mm): total length 5.28, head 1.02 long X 0.93 wide, prono-

tum 1.02 long X 1.24 wide, elytra 3. 19 long X 2.05 wide, antenna 3.97.

Distribution. — Described on 279 specimens from 17 caves. The range

of this taxon is roughly Y-shaped, centered in Monroe County, Ken-

tucky, but extending northwest into southeastern Barren and south-

western Metcalfe counties, northeast to southern Adair and northwestern

Cumberland counties, and south to Clay and northern Jackson coun-

ties, Tennessee, where it inhabits caves in Ordovician limestone at Cen-

tral Basin level.

Pseudanophthalmus cerberus completes, new subspecies

Fig. 7

Etymology. — Latin completes, "complete."

Description.— Similar in size (4.6-5.4 mm) and habitus to cerberus

New Trechine Beetles 1 25

cerberus, differing in presence of both pairs of elytral discal setae (+++)

and longer and straighter aedeagus (length 1.15-1.43, mean 1.30 ± 0.90

mm) with slightly reflexed tip (as in Fig. 7).

Type series. — Holotype male (American Museum of Natural His-

tory) and 11 paratypes, Cole Cave, 1.8 km N Austin, Barren Co., Ken-

tucky (Austin iy2' Quadrangle), 12 February 1966, T. C. Barr; 11 addi-

tional paratypes, same cave, 13 April 1973, R. Pape. Measurements of

holotype (mm): total length 5.52, head 0.87 long X 0.81 wide, pronotum

0.87 long X 1.02 wide, elytra 2.76 long X 1.67 wide, antenna 3.66.

Distribution. — Described on 33 specimens, all from central Barren

County, Kentucky: Cole, Beckton, Bryant Edmonds, Geralds, Mitchell,

Hansons, and Slick Rock caves. Hybridization with nominate cerberus

occurs in Bowles Branch Cave, 8.4 km southeast of Glasgow. Ranges of

P. m. campestris and P. c. completus are almost parapatric, rare exam-

ples of the latter having been collected in Beckton and Hansons caves,

where P. m. campestris is more abundant. This taxon coexists with P.

striatus (hind angles produced; elytra not pruinose), P. pubescens (api-

cal groove well developed), and Neaphaenops tellkampfi.

Pseudanophthalmus darlingtoni darlingtoni, new species and subspecies

Fig. 8

Etymology. — Patronymic honoring the late Philip J. Darlington, Jr.

Diagnosis. — A large species of the menetriesi group with shallow,

impunctate striae, convex elytra with normal chaetotaxy (+++), disc

weakly pruinose overall; pronotum with 1-2 long setae each side of disc

and hind angles produced backward as in P. striatus.

Description.— Length 4.9-6.3, mean 5.7 ±0.3 mm (N = 207). Robust,

convex, pubescent. Head about 0.14 (mean index) longer than wide,

dorsum subglabrous; labrum with small median lobe. Pronotum about

0.85 (mean) as long as wide, transverse-cordiform, widest in apical

fourth; anterior angles scarcely produced; sides rounded apical 0.4 then

convergent with very shallow or no sinuation to hind angles, which are

small, sharp, and slightly obtuse; base with small, rounded, secondary

angles, not obliquely inclined forward at sides (thus differing from

menetriesi and cerberus); disc with scattered, rather long pubescence

and 1-2 long setae each side; apex as wide as base and two-thirds great-

est width. Elytra oblong-oval, very convex with deplanate circumscutel-

lar area, 1.56 ± .04 times longer than wide (range 1.40-1.60), widest

behind middle and with posthumeral marginal sinuation (as in striatus

but deeper), humeri very prominent, prehumeral borders not quite per-

pendicular to mid-line, humeral serrations very coarse; disc microsculp-

ture finely transverse, forming tight meshworks, disc with weak prui-

nose microsculpture over entire surface; longitudinal striae moderately

and evenly impressed, rather shallow, intervals subconvex to nearly flat,

1 26 Thomas C. Barr, Jr.

with 2-3 rows of rather long pubescence each, inner 5 striae deeper but

outer striae still usually all discernible, striae impunctate, rarely evanes-

cently punctulate; apical recurrent groove vestigial but usually faintly

traceable to 5th stria, rarely to 3rd; discal chaetotaxy normal (+++).

Aedeagus 0.82-1.03, mean 0.90 ± .06 mm long (N = 56); basal bulb

large, forming less than right angle with main axis of median lobe,

which is moderately arcuate, gradually and briefly attenuate, slightly

reflexed at apex; copulatory pieces subequal in length, typical for mene-

triesi group; parameres with 4 apical setae.

Type series. — Holotype male (American Museum of Natural His-

tory) and 39 paratypes, Jones Cave, 4.3 km NNE Columbia on E side

valley of Butler Branch, Adair Co., Kentucky (Cane Valley 7%' Quad-

rangle), 27 July 1963, T. C. Barr. Measurements of holotype (mm): total

length 5.60, head 1.07 long X 0.89 wide, pronotum 1.04 long X 1.19

wide, elytra 3.24 long X 2.02 wide, antenna 3.79 long.

Distribution. — Described on 306 specimens from 20 caves in north-

eastern Metcalfe, northern Adair, and southern Green counties, Ken-

tucky. Distribution and variation in P. darlingtoni is discussed in detail

elsewhere (Barr, in press). Hybridization with the following subspecies

takes place in populations near the mouth of Little Barren River.

Pseudanophthalmus darlingtoni persimilis, new subspecies

Fig. 9

Etymology. — Latin persimilis, "resembling."

Description. — Length 4.8-6.4, mean 5.5 ± 0.3 mm (N = 40). Head,

pronotum, and elytra more slender than in nominate darlingtoni; pro-

notum widest in apical third, usually with more pronounced antebasal

sinuation, hind angles correspondingly less prominent, prehumeral

borders a little more oblique to mid-line, circumscutellar depression

shallower, posthumeral marginal sinuation very feeble. Aedeagus sim-

ilar to that of P. d. darlingtoni; mean length smaller, 0.77-1.00, mean

0.87 ± .06 mm (N = 16).

Type series. — Holotype male (American Museum of Natural His-

tory) and 25 paratypes, Woodard Cave, 5.0 km N W Donansburg near

Little Barren River, Green Co., Kentucky (Hudgins 7%' Quadrangle), 22

September 1963, T. C. Barr, J. R. Holsinger, R. M. Norton. Measure-

ments of holotype (mm): total length 5.72, head 1.04 long X 0.95 wide,

pronotum 1.07 long X 1.19 wide, elytra 3.34 long X 2.11 wide, antenna

3.87 long.

Distribution. — Described on 90 specimens from 11 caves in Green

County, north of Green River (a dispersal barrier upstream from the

mouth of Little Barren River, where it flows in a bed of lower, cherty

Fort Payne formation), and in eastern Hart County, Kentucky.

New Trechine Beetles

27

Fig. 6-10. Aedeagi of Pseudanophthalmus spp., left lateral view. 6: cerberus

cerberus, n. sp. and subsp. 7: cerberus completus, n. subsp. 8: darlingtoni dar-

lingtoni, n. sp. and subsp. 9: darlingtoni persimilis, n. subsp. 10: pubescens

intrepidus, n. subsp.

Pseudanophthalmus pubescens intrepidus, new subspecies

Fig. 10

Etymology. — Latin intrepidus, "undaunted, intrepid."

Description. — Length 5.3-6.1, mean 5.7 mm, more robust and

depressed than nominate pubescens. Pronotum more transverse, sides

less strongly convergent and scarcely sinuate in basal half, width at base

0.75 maximum width. Elytral discal chaetotaxy consistently normal

(+++), disc slightly less convex than in nominate pubescens, intervals

quite flat. Aedeagus larger, length 1.06-1.19, mean 1.11 mm.

Type series. — Holotype male (American Museum of Natural His-

tory) and 2 paratypes, Buchanan Cave, 1.3 km W Gainesville and 30 m

E KY Rt. 101, at head of hollow tributary to Difficult Creek, Allen Co.,

Kentucky (Scottsville 1%' Quadrangle), 18 August 1963, T. C. Barr.

Measurements of holotype (mm): total length 6.11, head 1.05 long X

128 Thomas C. Barr, Jr.

1.05 wide, pronotum 1.25 long X 1.39 wide, elytra 3.60 long X 2.17

wide, antenna 4.43 long, aedeagus 1.12 long.

Distribution. — Described on a total of nine specimens from the type

locality and Bryant Edmonds Cave, 1.7 km southwest of Beckton at the

head of Greens Creek, Barren County, Kentucky. Although Ban and

Crowley (1981) demonstrated what appears to be substantial clinal vari-

ation in P. pubescens, this relatively uncommon southwestern geogra-

phic race hybridizes with nominate pubescens in Beckton Cave, 1.0 km

northwest of Beckton, Barren County.

Trechus (Trechus) caliginis, new species

Fig. 11

Etymology. — Latin caliginis, "of mist or fog."

Diagnosis. — A large species of the hydropicus group in which the

aedeagal apex is simply attenuate, not knobbed; pronotum margins

broadly reflexed, hind angles blunt and obtuse; closest to T. roanicus

Barr but differing in pronotal characters and more strongly impressed

elytral striae.

Description. — Length 3.9-4.2, mean 4.0 ± 0.1 mm (N = 5). Dark

castaneous to reddish castaneous; legs, mouthparts, and basal antennal

segments slightly paler. Eye diameter subequal to scape length. Prono-

tum 0.7 as long as wide, base 0.8 maximum width; reflexed margin

unusually wide, sides convergent without trace of sinuosity to blunt,

obtuse hind angles. Elytra very convex, 1.33 longer than wide, 5 striae

present, moderately impressed but intervals either flat or very feebly

convex. Aedeagus of holotype 0.52 mm long, basal bulb small and

deflexed at right angle to slender, very straight median lobe, which is

gradually attenuate and feebly reflexed at apex; left copulatory piece

slender, rod-like, about 0.6 as long as right piece, which is hemicylindri-

cal and bears small, irregular scallops on dorsal margin near bluntly

rounded apex; parameres slender, weakly arcuate, with 3 long apical setae.

Type series. — Holotype male (American Museum of Natural His-

tory) and one male paratype, Camp Creek Bald, just below summit,

elevation about 1460 m, Greene Co., Tennessee/ Madison Co., North

Carolina (Greystone 7)4' Quadrangle), 9 August 1983, T. C. Barr, Jr., T.

C. Barr III; three female paratypes, same locality, 21 August 1960, T. C.

Barr, Jr., M. C. Bowling. Measurements of holotype (mm): total length

3.96, head 0.73 long X 0.70 wide, pronotum 0.70 long X 1.01 wide,

elytra 2.30 long X 1.71 wide, antenna 2.11 long, aedeagus 0.52 long.

Distribution. — At present known only from the type locality in the

Bald Mountains between Greeneville, Tennessee, and Asheville, North

Carolina, this species will key out near T. roanicus in my key to Appa-

lachian Trechus (Barr 1979a); all other species of the hydropicus group

possess a more or less knobbed aedeagal apex.

New Trechine Beetles

129

_0.3

MM

Fig. 1 1-12. Aedeagi of Trechus spp., left lateral view. 11: T. (T.) caliginis, n. sp.

12: T. (Microtrechus) inexpectatus, n. sp.

Trechus (Microtrechus) inexpectatus, new species

Fig. 12

Etymology. — Latin inexpectatus, "unexpected."

Diagnosis. — A species of the uncifer group closest to T. uncifer Barr,

differing in larger size, more convex elytral disc with very shallow striae,

and aedeagal apex less produced.

Description. — Length 3.46-3.59 mm (N = 2). Dark castaneous; legs,

mouthparts, and basal antennal segments paler but not flavous as in

uncifer, consequently less contrasting. Eye diameter less than scape

length. Pronotum closely similar to that of uncifer: sides convergent,

then almost subparallel but very briefly so, just opposite posterior mar-

130 Thomas C. Barr, Jr.

ginal punctures; hind angles right to slightly obtuse, sharp. Elytral disc

moderately convex, 3 inner striae very feebly impressed, only sutural

complete, intervals completely flat (disc flattened near middle, and

striae deep with convex intervals in uncifer). Aedeagus of holotype 0.42

mm long, very strongly arcuate, apex more briefly attenuate and less

sharply hooked, in comparison with uncifer, internal sac with heavy

scales obscuring copulatory pieces; parameres rather short, with 4 stout

apical setae.

Type series. — Holotype male (American Museum of Natural His-

tory) and one female paratype, Camp Creek Bald, just below summit,

elevation 1460 m, Greene Co., Tennessee/ Madison Co., North Carolina

(Greystone 7%' Quadrangle), 9 August 1983, T. C. Barr, Jr., T. C. Barr,

III. Measurements of holotype (mm): total length 3.46, head 0.64 long

X 0.70 wide, pronotum 0.67 long X 1.01 wide, elytra 2.08 long X 1.47

wide, antenna 1.71 long, aedeagus 0.42 long.

Distribution. — Known only from the type locality, where it coexists

with T. caliginis, T. hydropicus beutenmuelleri Jeannel, and a small iso-

late near T. (M.) vandykei Jeannel. Compared to the last two species it

is distinctly larger; compared to caliginis it is less convex, has smaller

eyes, very shallow elytral striation, and — if males are available — an api-

cally hooked aedeagus and a single enlarged protarsomere.

DISCUSSION

All of the Pseudanophthalmus taxa described in this paper occur in

south-central Kentucky, east or south of the Mammoth Cave region.

Nine belong to the menetriesi group, which is defined by the forward

position of the anterior discal puncture of the elytron opposite the 2nd

umbilicate puncture, a vestigial apical elytral groove, and the structure

of the transfer apparatus. The copulatory pieces resemble those of the

pubescens group but are simpler and lack spines or apical knobs (Barr

1979b). The pronotum hind angles are tilted forward in menetriesi and

its close allies {simulans, pilosus, globiceps, trans/luvialis; weakest in

glo biceps) and also in cerberus. The hind angles are produced backward

in darlingtoni and striatus. Both cerberus and darlingtoni have pruinose

elytral microsculpture, which is absent in other species of the menetriesi

group (although present to a greater or lesser extent in all species of the

pubescens group). Pseudanophthalmus pubescens intrepidus, which

occurs on the southern periphery of the range of polytypic pubescens,

manages to cross the upper Barren River in the vicinity of Barren River

Dam, where Neaphaenops tellkampfi crosses (Barr 1979b). The Barren

is a barrier at Bowling Green, but not in part of its upstream course

(Barr, in press).

The two larger Trechus species described from Camp Creek Bald

New Trechine Beetles 1 3 1

were obtained during a study of electrophoretic variation in the isolates

of the vandykei group {yandykei, pisgahensis Barr, tusquitee Barr, haoe

Barr, bowlingi Barr and seven undescribed species). They represent only

2 of 18 previously undescribed Appalachian Trechus species obtained by

a sifting technique in 1982-1984 (Barr, in preparation). With four spe-

cies of Trechus, three of them endemics, Camp Creek Bald now joins

the list of eight other massifs in the Unaka region with endemic species

(see Barr 1979a:41). Trechus inexpectatus is the only species of the

uncifer group currently known east of the Asheville basin; morphologi-

cally it is closer to uncifer (Great Smokies) and satanicus Barr (Great

Balsams) than to the series of species morphologically clustered around

aduncus Barr (including talequah Barr, howellae Barr, toxawayi Barr

and coweensis Barr; see Barr 1979a for descriptions, illustrations, and

geographic ranges). Camp Creek Bald is shown on Figure 46 in Barr

(1979a) but is incorrectly labeled "Camptown Bald." The site lies at the

west end of a long ridge 1300-1500 m in elevation and about 10 km

long.

ACKNOWLEDGMENTS.— This study was supported in part by

grants from the National Science Foundation (DEB 82-02339) and the

Highlands Biological Foundation. I thank W. M. Andrews, T. C. Barr,

III, M. C. Bowling, J. R. Holsinger, R. A. Kuehne, T. G. Marsh, R. M.

Norton, R. Pape, and J. M. Valentine for field assistance or contribu-

tion of specimens. This paper is dedicated to the memory of the late

Robert A. Kuehne, friend and colleague.

LITERATURE CITED

Barr, Thomas C, Jr. 1979a. Revision of Appalachian Trechus (Coleoptera:

Carabidae). Brimleyana 2:29-75.

. 1979b. The taxonomy, distribution, and affinities of Neaphaenops,

with notes on associated species of Pseudanophthalmus (Coleoptera,

Carabidae). Am. Mus. Nov. 2682. 20 pp.

. 1980. New species groups of Pseudanophthalmus from the Central

Basin of Tennessee (Coleoptera: Carabidae: Trechinae). Brimleyana 3:85-96.

1981. Pseudanophthalmus from Appalachian caves (Coleoptera:

Carabidae): The engelhardti complex. Brimleyana 5:37-94.

. In press. Pattern and process in speciation of trechine beetles in

eastern North America (Coleoptera: Carabidae: Trechinae). Pp. 350-407 in

George E. Ball (Ed.). Taxonomy, phylogeny, and zoogeography of beetles

and ants. A volume dedicated to the memory of Philip Jackson Darlington,

Jr., 1904-1983 (Series Entomologia 33). W. Junk: Dordrecht, The

Netherlands.

132 Thomas C. Barr, Jr.

and P. H. Crowley. 1981. Do cave carabid beetles really show

character displacement in body size? Am. Nat. 777:363-371.

Accepted 21 February 1985

Fishes of Buck Creek,

Cumberland River Drainage, Kentucky

Ronald R. Cicerello

Kentucky Nature Preserves Commission,

Frankfort, Kentucky 40601

AND

Robert S. Butler

Department of Biology,

Eastern Kentucky University, Richmond, Kentucky 40475

ABSTRACT. — Fifty-nine personal fish collections and museum

records from thirty-nine sites in the drainage of Buck Creek, a major

tributary to the upper Cumberland River below Cumberland Falls in

Kentucky, revealed a total of seventy-three species and one hybrid,

representing thirteen families. New records for the upper Cumberland

River drainage in Kentucky included Ictiobus bubalus, I ctalurus f mea-

tus, and Lepomis microlophus. Notropis ariommus and Etheostoma

cinereum, two species assigned protection status in Kentucky, are

known from Buck Creek, but E. cinereum has not been collected since

1955. Analysis of faunal resemblance of species collected at twenty-one

sites along the Buck Creek mainstem revealed three faunal units. The

pattern of longitudinal distribution along the mainstem involved addi-

tion of species in the middle stream section to those present in the

upper section and replacement in the lower section by forms typical of

low-gradient, big-river habitats.

INTRODUCTION

The upper Cumberland River basin upstream from the Tennessee

border drains 13,416 sq km of eastern Kentucky (Mayes et al. 1975) and

contains many of the highest quality streams remaining in Kentucky

(Harker et al. 1980; Hannan et al. 1982). Although the fishes of the

upper Cumberland River basin have been the subject of numerous pub-

lished collections, distributional lists, and descriptions (e.g., Jordan and

Swain 1883; Kirsch 1892, 1893; Woolman 1892; Evermann 1918; Jen-

kins et al. 1972; Starnes and Starnes 1978; Harker et al. 1979, 1980;

Burr 1980; Stauffer et al. 1982), thorough ichthyofaunal surveys of trib-

utaries within the drainage have been conducted on only the Big South

Fork Cumberland River (Comiskey 1970; Comiskey and Etnier 1972)

and Rockcastle River (Small 1970). Because of this paucity of informa-

tion, we initiated our study of the fishes of the Buck Creek drainage.

The study, based on personal collections and museum records, aug-

ments the limited published faunal information available for the drain-

age (Carter and Jones 1969; Harker et al. 1979, 1980).

Brimleyana No. 1 1:133-159, October 1985 133

134 Ronald R. Cicerello and Robert S. Butler

STUDY AREA

Buck Creek, a fifth-order tributary to the Cumberland River in

southeastern Kentucky (Fig. 1), drains approximately 767 sq km of Lin-

coln, Pulaski, and Rockcastle counties and flows south 107.2 km before

discharging into the Cumberland River near river km 859. Impound-

ment of the river in 1951 to form Cumberland Reservoir permanently

ponded several kilometers of the lower portion of Buck Creek, and this

influence may extend upstream in excess of 21 km following heavy rain-

fall. The stream is generally less than 20 m wide and 2 m deep, but

achieves a maximum width of approximately 150 m and a maximum

depth exceeding 25 m near the mouth. Buck Creek is a high quality

stream with clear, well-oxygenated and buffered water (Harker et al.

1979, 1980). The average stream gradient along the mainstem of the

creek is 1.25 m/km and is also influenced by Cumberland Reservoir

backwaters. According to the United States Army Corps of Engineers

(USACE 1976), mean annual flow is 1 1.7 cu m/second.

Buck Creek lies almost entirely within the Eastern Highland Rim

Subsection of the Interior Low Plateaus Physiographic Province (Quar-

terman and Powell 1978). Surface geology is composed primarily of

Mississippian Age limestone deposits with limited exposures of shale

bedrock in the northeastern portion of the basin. Karst topography and

sinking creeks associated with limestone deposits are common in the

watershed, especially south of latitude 37°17'00". South of Kentucky

route (KY) 80 the stream is deeply entrenched within the western limit

of the Cumberland Plateau Section of the Appalachian Plateaus Physi-

ographic Province. This area, associated outlying hills to the west, and

much of the extreme eastern boundary of the watershed are overlain

with erosion resistant Pennsylvanian Age sandstone. Quaternary Age

alluvium is limited to isolated stream channel and floodplain deposits.

Watershed land use is primarily agricultural and secondarily forest.

Forested areas are small and scattered except along stream channels and

in the part of the watershed south of KY 80, much of which lies within

the proclamation boundary of the Daniel Boone National Forest. Coal

stripmines and limestone quarries also occur in the watershed south of

KY 80 and each comprises less than 1% of the watershed area. Two small

(15 and 1 1 ha) flood control reservoirs were constructed within the Lin-

coln County part of the watershed by the Soil Conservation Service (T.

A. Heard, pers. comm.).

MATERIALS AND METHODS

Fifty-nine collections are reported from thirty-nine collection sites

in the Buck Creek drainage (Table 1, Fig. 1). Each collection site

includes the stream name, locality, county, and collection date(s). Col-

lections were made by the authors, except as noted, using seines, gill

Fishes of Buck Creek, Kentucky

135

Cumberland River

Fig. 1. Map of Buck Creek drainage, Kentucky, showing collection sites.

36 Ronald R. Cicerello and Robert S. Butler

Table 1. Buck Creek collection sites, dates, and collectors.

Site 1. Buck Creek, 2.7 km SSW of Ottenheim and W of Kocher Ridge

Road and Maple Swamp Road jet, Lincoln Co., 14 March 1977, L.

M. Page and C. W. Ronto; 29 April 1979.

2. Buckeye Branch, confluence with Buck Creek, Pulaski Co., 29

April 1979.

3. Buck Creek, KY 1781 bridge, Lincoln Co., 29 April 1979.

4. Buck Creek, confluence with Gilmore Creek, Lincoln Co., 27 Sep-

tember 1980.

5. Gilmore Creek, 0.5 stream km upstream from confluence with Crab

Orchard Creek, Lincoln Co., 20 May 1979.

6. Crab Orchard Creek, 1.7 km WSW of Broughtentown at Brad Petery

Road bridge, Lincoln Co., 21 May 1979.

7. Crab Orchard Creek, 1.5 stream km upstream from confluence with

Gilmore Creek, Lincoln Co., 21 May 1979.

8. Gilmore Creek, KY 1781 bridge, Lincoln Co., 29 April 1979.

9. Glade Fork Creek, 1.5 stream km upstream from confluence with

Buck Creek, Pulaski Co., 20 May 1979.

10. Buck Creek, 0.33 stream km upstream from confluence with Bear

Den Hollow tributary, Pulaski Co., 20 May 1979.

11. Caney Creek, KY 1012 bridge, Pulaski Co., 9 November 1980.

12. Buck Creek, 6.4 km W of Bandy at KY 70 bridge, Pulaski Co., 13

June 1970, L. M. Page and N. D. Penny.

13. Indian Creek, 2.4 km W of Bobtown, Pulaski Co., 17 March 1976, B.

M. Burr, L. M. Page, and M. A. Morris; 19 March 1978, L. M. Page

and R. L. Mayden.

14. Buck Creek, KY 39 bridge, Pulaski Co., 14 March 1981.

15. Brushy Creek, KY 70 bridge, Rockcastle Co., 17 June 1979.

16. Bee Lick Creek, 0.6 stream km upstream from confluence with

Brushy Creek, Pulaski Co., 19 April 1981.

17. Brushy Creek, 0.2 stream km downstream from KY 934 bridge,

Pulaski Co., 19 April 1981.

18. Brushy Creek, 5.3 stream km upstream from confluence with Clifty

Creek, Pulaski Co., 17 June 1979.

19. Brushy Creek, 0.1 stream km upstream from confluence with Clifty

Creek, Pulaski Co., 17 June 1979.

20. Clifty Creek, 0.33 stream km upstream from confluence with Brushy

Creek at Elrod Road, Pulaski Co., 17 June 1979.

21. Brushy Creek, confluence with Buck Creek, Pulaski Co., 14 Sep-

tember 1955, C. R. Gilbert and B. C. Franklin.

22. Buck Creek, confluence with Brushy Creek, Pulaski Co., 14 Sep-

tember 1955, C. R. Gilbert and B. C. Franklin.

23. Buck Creek, KY 461 bridge, Pulaski Co., 14 September 1967, C. R.

Gilbert, W. Seaman, and C. M. Colson; 30 August 1978, 28 Sep-

tember 1980, 14 March 1981; 26 September 1981, B. M. Burr, S. J.

Walsh, and T. E. Shepard.

Fishes of Buck Creek, Kentucky 137

24. Buck Creek, 0.61 km downstream from KY 461 bridge, Pulaski Co.,

11 July 1978, S. P. Rice, E. G. Amburgey, R. C. Wilson, and J.

R. MacGregor.

25. Buck Creek, KY 1677 bridge, Pulaski Co., 11 July 1978, S. P. Rice,

E. G. Amburgey, R. C. Wilson, and J. R. MacGregor; 9 June 1980.

A. W. Berry, M. J. Linville, J. R. MacGregor, and S. P. Rice.

26. Unnamed stream in Sinking Valley, 1.0 km E of Plato School and

1.4 km NNE of Plato, Pulaski Co., 22 November 1980.

27. Buck Creek, old KY 80 bridge at Stab, Pulaski Co., 27 July 1954, J.

R. Charles; 14 September 1955, C. R. Gilbert and B. C. Franklin; 14

September 1967, C. R. Gilbert, W. Seaman, and C. M. Colson; 28

July 1973, B. A. Branson and D. L. Batch; 22 October 1976, B. M.

Burr and L. M. Page; 24 June 1978, 28 October 1980.

28. Short Creek, opening to downstream cave 0.6 km ESE of old KY 80

bridge at Buck Creek, Pulaski Co., 14 September 1955, C. R. Gilbert

and B. C. Franklin; 28 October 1980.

29. Flat Lick Creek, KY 461 bridge, Pulaski Co., 8 July 1979.

30. Flat Lick Creek, 1.5 km SSE of Shopville on Heron Cemetery Road

and 3.2 km W of Stab, Pulaski Co., 9 November 1980.

31. Buck Creek, KY 1003 bridge, Pulaski Co., 28 September 1980.

32. Buck Creek, KY 192 bridge, Pulaski Co., 9 June 1965, R. E. Jenkins,

C. Tsai, C. R. Robins, and T. Zorach; 8 September 1966, T. Zorach

and R. F. Denoncourt; 19 July 1968, B. A. Branson and D. L. Batch;

28 September 1980, 9 November 1980.

33. Buck Creek, 2.5 stream km downstream from KY 192 bridge, Pulaski

Co., 3 May 1981.

34. Unnamed stream in Silvers Hollow, 1.7 km NW on KY 192 from jet

with KY 1003, Pulaski Co., 7 November 1981.

35. Buck Creek, 10.6 stream km downstream from KY 192 bridge and

1.6 km SSW of Poplarvilie, Pulaski Co., 3 May 1981.

36. Buck Creek, boat ramp off KY 1097, 8.33 stream km upstream from

Cumberland River confluence, Pulaski Co., 9 November 1980.

37. Buck Creek, at Hound Hollow Creek, Pulaski Co., 26 October 1980.

38. Buck Creek, 4.6 stream km upstream from confluence with Cumber-

land River, Pulaski Co., 6 April 1981, 23 August 1981, 19

September 1981.

39. Buck Creek, confluence with Cumberland River, Pulaski Co., 19

September 1981.

138 Ronald R. Cicerello and Robert S. Butler

nets, an electroshocker, and an ichthyocide. An effort was made to

intensively sample all habitats at each site. Representative specimens of

all except two species collected during the current survey were fixed in

10% formalin and stored in 35-40% isopropanol. Except where other-

wise indicated, they are housed at the Kentucky Nature Preserves

Commission pending museum deposition.

Species composition of 21 mainstem collecting sites were analyzed

to determine faunal resemblance using Long's (1963) average resemb-

lance formula, in which:

average faunal resemblance =C(Ni+N2)(100)/2NiN2.

C is the number of species shared by sites 1 and 2, and Ni and N2 are

the number of species found at sites 1 and 2, respectively. Resemblance

values range from 0 to 100, where 0 indicates that sites 1 and 2 have no

species in common, and 100 indicates that sites 1 and 2 have identical

faunas.

RESULTS

Based on our collections, museum records, and acceptable litera-

ture records, the following fishes are known from the Buck Creek

drainage. Scientific and common names and the order of presentation

follow Robins et al. (1980). Distribution within the drainage is indicated

by the terms "generally distributed", "occasional", and "sporadic" as

defined by Smith (1965). Collection site numbers are presented for each

species, followed in parentheses by the number of specimens collected (if

available). Institutions where specimens are deposited and their abbre-

viations are as follows: Cornell University (CU), Eastern Kentucky Uni-

versity (EKU), Illinois Natural History Survey (INHS), University of

Louisville (UL), and University of Michigan Museum of Zoology

(UMMZ).

SPECIES ACCOUNTS

Petromyzontidae — lampreys

Ichthyomyzon bdellium (Jordan). Ohio lamprey. Occasional in rif-

fles in late winter and spring. Sites: 23(1), 24(1), 27(1), 32(-).

Lepisosteidae — gars

Lepisosteus osseus (Linnaeus). Longnose gar. Carter and Jones

(1969) reported one specimen from Buck Creek below the KY 80 bridge.

Our specimens were collected by gill net over a sloping mud bottom

from the lower mainstem where the species occurs sporadically. Site:

38(3).

Fishes of Buck Creek, Kentucky 139

Clupeidae — herrings

Dorosoma cepedianum (Lesueur). Gizzard shad. The gizzard shad

was reported by Carter and Jones (1969) from two mainstem sites and

was generally distributed and abundant in the lower mainstem. Sites:

32(1), 35(-), 36(3), 38(3), 39(-).

Dorosoma petenense (Giinther). Threadfin shad. This species is

generally distributed in downstream parts of Buck Creek influenced by

Cumberland Reservoir, where it was originally introduced in 1957 as

forage for game fish (Henley 1967). Sites: 38(2), 39(-).

Cyprinidae — carps and minnows

Campostoma oligolepis Hubbs and Greene. Largescale stoneroller.

This species was reported from Buck Creek as Campostoma anomalum

(Rafinesque), the central stoneroller, by Carter and Jones (1969) and

Harker et al. (1979, 1980). However, Burr (1980) and Stauffer et al.

(1982) indicated that the species found in the Cumberland River drain-

age is C. oligolepis. This was the most common and generally distrib-

uted species in the drainage. Sites: 1(13), 2(1), 3(5), 4(3), 5(2), 6(1), 7(4),

8(3), 9(1), 10(4), 11(2), 12(34), 13(12), 14(2), 15(1), 18(2), 19(2), 20(2),

21(5), 22(1), 23(16), 24(49), 25(3), 27(27), 28(3), 29(5), 30(1), 31(1),

32(26), 33(1), 38(1).

Cyprinus carpio Linnaeus. Common carp. Several large specimens

were collected from the lower mainstem, where the species was generally

distributed. Sites: 35(2), 38(1), 39(-).

Ericymba buccata Cope. Silverjaw minnow. This minnow was spo-

radic in the upper half of the drainage where it was also collected by

Harker et al. (1979). Sites: 7(4), 11(2), 14(3), 20(3), 23(1).

Hybopsis amblops (Rafinesque). Bigeye chub. Harker et al. (1979)

reported this species from the mainstem. The bigeye chub was generally

distributed in the mainstem in or just downstream of riffles flowing over

a variety of substrates. Sites: 22(1), 23(4), 24(-), 25(1), 27(33), 31(7),

32(5), 33(5).

Hybopsis dissimilis (Kirtland). Streamline chub. This chub was

sporadically distributed in 0.3-0.6 m deep mainstem riffles with moder-

ate current and cobble or boulder substrate. Sites: 27(1), 32(4), 33(2).

Notropis ardens (Cope). Rosefin shiner. This shiner was reported

from Buck Creek by Harker et al. (1979, 1980) and was generally dis-

tributed throughout the upper half of the drainage. Sites: 1(16), 2(18),

3(13), 4(7), 5(17), 6(13), 7(7), 8(8), 9(21), 10(24), 11(9), 12(43), 14(8),

15(6), 17(8), 18(27), 21(40), 23(47), 24(14), 25(24), 27(38), 29(4), 30(4).

Notropis ariommus (Cope). Popeye shiner. This silt intolerant spe-

cies (Trautman 1981) was sporadically distributed in the drainage. Site:

21(1), 32(1), 33(3).

140 Ronald R. Cicerello and Robert S. Butler

Notropis atherinoides Rafinesque. Emerald shiner. Although this

species was present in only three collections, further sampling of deep-

water, riverine habitat along the mainstem would undoubtedly yield

more specimens. Sites: 32(1), 33(1), 36(29).

Notropis boops Gilbert. Bigeye shiner. This shiner was sporadic in

the mainstem and tributaries in the upper half of the drainage, where

the species was collected from flowing pools and riffles over a generally

bedrock bottom. Sites: 3(1), 8(5), 17(2), 18(6), 23(7), 24(3).

Notropis buchanani Meek. Ghost shiner. This large-river species

(Smith 1979) was occasional in the lower main channel. Sites: 36(86),

37(6).

Notropis chrysocephalus (Cope). Striped shiner. Carter and Jones

(1969) and Harker et al. (1979) reported the striped shiner from Buck

Creek, and Harker et al. (1980) collected it in Brushy Creek. The species

was generally distributed in clear pools throughout all but the lower

part of Buck Creek influenced by Cumberland Reservoir. Site: 1(7),

2(5), 3(2), 4(5), 5(8), 6(2), 7(2), 8(14), 9(2), 10(10), 11(5), 12(35), 14(4),

15(6), 18(1), 21(8), 23(59), 27(4), 32(4).

Notropis galacturus (Cope). Whitetail shiner. Harker et al. (1980)

reported specimens of this fish from mainstem Buck Creek. The white-

tail shiner was generally distributed in the lower half of the drainage

and was often collected with, but was more common than, two other

species of the subgenus Cyprinella — Notropis spilopterus (Cope) and

Notropis whipplei (Girard). Notropis galacturus was commonly col-

lected from riffles flowing over gravel, cobble, or bedrock substrate and

from adjacent eddy habitat. Sites: 14(7), 18(2), 19(3), 21(14), 22(5),

23(13), 24(25), 25(3), 27(44), 31(1), 32(12), 33(6).

Notropis phot ogenis (Cope). Silver shiner. This inhabitant of mod-

erate to large streams (Trautman 1981) was apparently restricted to and

sporadically distributed in the lower mainstem of Buck Creek upstream

from the influence of Cumberland Reservoir. Sites: 27(1), 31(3), 32(19).

Notropis rubellus (Agassiz). Rosyface shiner. This shiner is gener-

ally distributed in the mainstem, where it was also collected by Harker

et al. (1979). Sites: 10(2), 14(6), 23(18), 25(1), 27(19), 32(6), 33(12).

Notropis spilopterus (Cope). Spotfin shiner. The spotfin shiner was

reported from the Buck Creek mainstem by Harker et al. (1979). We

found it generally distributed in the lower half of the drainage exclusive

of the area influenced by Cumberland Reservoir. Sites: 21(11), 23(8),

24(12), 27(1), 31(1), 32(-), 33(8).

Notropis telescopus (Cope). Telescope shiner. Harker et al. (1979,

1980) reported the telescope shiner from Buck and Brushy creeks. We

found it generally distributed and readily collected from flowing pools

and riffles over bedrock or gravel and cobble substrates. Sites: 2(36),

Fishes of Buck Creek, Kentucky 141

3(2), 8(13), 9(15), 10(2), 11(2), 12(3), 14(9), 15(8), 17(6), 18(10), 19(4),

20(2), 23(93), 24(-), 25(1), 27(37), 32(3).

Notropis whipplei (Girard). Steelcolor shiner. The steelcolor shiner

was sporadic in the lower half of the drainage. Sites: 21(2), 32(-), 33(1).

Phoxinus erythrogaster (Rafinesque). Southern redbelly dace. This

species was restricted to small headwater streams and was occasional in

distribution. Sites: 4(6), 5(6), 7(1), 26(7).

Pimephales notatus (Rafinesque). Bluntnose minnow. Harker et al.

(1979, 1980) reported this minnow from Buck and Brushy creeks, where

it was the most abundant species. We found it to be generally distrib-

uted and one of the most common fishes in the drainage. Sites: 1(18),

2(4), 3(2), 5(1), 6(4), 7(5), 8(8), 9(4), 10(14), 11(4), 12(12), 13(4), 14(4),

15(8), 16(2), 17(3), 18(7), 19(1), 20(1), 21(13), 22(4), 23(17), 24(20), 25(6),

27(18), 29(6), 30(5), 32(-), 33(1).

Pimephales promelas Rafinesque. Fathead minnow. A first-order

tributary to a sinking creek in the headwaters of Buck Creek drainage

supported a small population of P. promelas, but probably represented

a bait-bucket introduction. Site: 26(2).

Pimephales vigilax (Baird and Girard). Bullhead minnow. This

inhabitant of medium and large streams (Smith 1979) was sporadically

distributed in the lower mainstem. Site: 37(2).

Rhinichthys atratulus (Hermann). Blacknose dace. This species was

occasional in small headwater streams over gravel substrates. Sites: 1(6),

4(1), 6(1), 9(1), 13(3), 29(9).

Semotilus atromaculatus (Mitchill). Creek chub. This chub was

generally distributed and common throughout the upper half of the

drainage and present in tributaries in the remainder. Harker et al. (1980)

reported specimens from Brushy Creek. Sites: 1(5), 2(2), 4(1), 5(5), 6(4),

7(2), 8(2), 9(3), 10(3), 18(2), 20(3), 21(2), 23(1), 24(2), 26(2), 28(1), 29(1),

30(1), 34(3).

Catostomidae — suckers

Carpiodes cyprinus (Lesueur). Quillback. The quillback occurs spo-

radically in the lower mainstem of Buck Creek influenced by Cumber-

land Reservoir. The specimen retained (EKU 1190) measured 35 cm

standard length (SL), and was taken by gill net over a mud and debris

bottom in water less than 9.1 m deep. Sites: 35(1), 38(2).

Carpiodes velifer (Rafinesque). Highfin carpsucker. Two adult

specimens, one (EKU 1 190) of which measured 34 cm SL, were taken by

gill net from the lower mainstem, where the species occurs sporadically.

Site: 38(2).

Catostomus commersoni (Lacepede). White sucker. The white sucker

was sporadic in tributaries of Buck Creek. Sites: 2(1), 6(1), 30(1).

142 Ronald R. Cicerello and Robert S. Butler

Hypentelium nigricans (Lesueur). Northern hog sucker. Carter and

Jones (1969) reported this species from all three sites they sampled on

Buck Creek, and Harker et al. (1979, 1980) reported specimens from the

mainstem and Brushy Creek. The northern hog sucker was generally

distributed throughout all but the lower mainstem of Buck Creek and

was the most common catostomid in the drainage. Sites: 1(2), 5(2), 6(1),

7(1), 11(1), 12(9), 14(3), 15(1), 18(1), 21(1), 23(5), 24(11), 27(2), 29(2),

30(1), 32(5), 33(1).

Ictiobus bubalus (Rafinesque). Smallmouth buffalo. Three speci-

mens of this typically large-river buffalo (Smith 1979) were collected by

gill net from the lower mainstem, where the species occurred sporadi-

cally. One specimen was retained (EKU 1190) and measured 39 cm SL.

Site: 38(3).

Moxostoma anisurum (Rafinesque). Silver redhorse. One specimen

measuring 17.6 cm SL was collected from Buck Creek in 1954 by J. R.

Charles (UL 6865) (R. E. Jenkins, pers. comm.); however, exact locality

information was not recorded. Site: unknown.

Moxostoma carinatum (Cope). River redhorse. Although the river

redhorse has been reported to inhabit medium-size rivers with gravel

and rock bottoms and swift or strong flow (Pflieger 1975; Smith 1979),

our specimens were collected from the sluggish lower mainstem, where

the species occurs sporadically. The specimen collected at Site 35 was

taken by gill net in 3.0 m of slowly flowing water over a bottom that

graded from mud and debris to rock. Sites: 35(1), 38(3).

Moxostoma duquesnei (Lesueur). Black redhorse. Harker et al.

(1979, 1980) reported specimens of this sucker from the mainstem and

Brushy Creek. The black redhorse was occasional throughout the Buck

Creek drainage. Sites: 3(1), 11(1), 16(1), 24(-), 27(1), 35(2).

Moxostoma erythrurum (Rafinesque). Golden redhorse. Carter and

Jones (1969) collected 112 specimens from three sites on the mainstem,

some of which were probably misidentified in light of the diverse sucker

fauna present in Buck Creek. This was the most generally distributed

and common Moxostoma in the drainage. Sites: 7(2), 18(2), 19(2), 21(3),

22(8), 23(8), 24(-), 27(4), 33(1).

Moxostoma macrolepidotum (Lesueur). Shorthead redhorse. Only

one specimen of the subspecies M. macrolepidotum breviceps (CU

52283) has been collected from the drainage (R. E. Jenkins, pers.

comm.). Site: 32(1).

Ictaluridae — bullhead catfishes

let alums fur catus (Lesueur). Blue catfish. Several specimens of the

blue catfish were collected from the lower mainstem, where the species

occurs occasionally. Site: 38(-).

Fishes of Buck Creek, Kentucky 143

Ictalurus natalis (Lesueur). Yellow bullhead. Carter and Jones

(1969) reported a specimen from the mainstem near the KY 39 bridge.

Three juvenile specimens of this sporadically distributed species were

collected from a drought isolated pool. Site: 4(3).

Ictalurus punctatus (Rafinesque). Channel catfish. One specimen

was reported from the mainstem at the KY 80 bridge by Carter and

Jones (1969). Although the channel catfish was collected only from the

lower mainstem, it is probably widely distributed in the drainage. Sites:

35(1), 38(5).

Noturus flavus Rafinesque. Stonecat. Carter and Jones (1969)

reported a specimen of N. flavus from the mainstem at the KY 70

bridge. The stonecat was occasional in the lower mainstem exclusive of

the area influenced by Cumberland Reservoir. Recent monthly collec-

tions at Sites 14, 19, and 32 yielded a dozen additional specimens from

under slab boulders and cobble in areas of moderate to swift current.

Sites: 21(3), 23(3), 24(-), 25(-), 32(-).

Pylodictis olivaris (Rafinesque). Flathead catfish. Only one speci-

men was collected from the mainstem; however, the flathead catfish is

probably more common and widely distributed along the mainstem.

Site: 38(1).

Cyprinodontidae — killifishes

Fundulus catenatus (Storer). Northern studfish. Harker et al. (1979,

1980) reported the northern studfish from Buck and Brushy creeks. We

found it common and generally distributed in the upper part of the

drainage. Sites: 1(14), 2(4), 3(1), 4(4), 5(2), 6(2), 7(4), 8(5), 9(3), 10(1),

11(2), 12(24), 14(1), 15(2), 16(1), 17(1), 19(2), 20(1), 21(2), 23(12), 24(4),

27(2), 32(1).

Atherinidae — silversides

Labidesthes sicculus (Cope). Brook silverside. The brook silverside

was generally distributed in the lower mainstem. Sites: 23(1), 31(2),

32(1), 36(2), 38(2).

Percichthyidae — temperate basses

Morone chrysops (Rafinesque). White bass. Although only one

specimen was collected during the survey, Carter and Jones (1969)

stated that white bass are harvested by fishermen from Buck Creek near

Cumberland Reservoir. Site: 38(1).

Morone saxatilis (Walbaum). Striped bass. This introduced game

fish has been stocked in Cumberland Reservoir essentially every year

since 1957 (Axon et al. 1982) and, according to a local boat dock opera-

tor, is sporadically harvested by anglers from the lower, impounded sec-

tion. Site: none.

144 Ronald R. Cicerello and Robert S. Butler

Centrarchidae — sunfishes

Ambloplites rupestris (Rafinesque). Rock bass. The rock bass was

reported from the mainstem by Carter and Jones (1969) and Harker et

al. (1979) and from Brushy Creek by Harker et al. (1980). It was gener-

ally distributed in all but the lower impounded mainstem. Sites: 11(1),

14(1), 15(1), 21(2), 23(4), 24(6), 25(1), 27(3), 29(1), 32(2).

Lepomis cyanellus Rafinesque. Green sunfish. This sunfish was

reported by Carter and Jones (1969) and Harker et al. (1979, 1980), and

was generally distributed and common in pools throughout the drain-

age. Sites: 1(1), 5(1), 6(1), 7(2), 8(1), 9(1), 11(1), 15(2), 16(1), 21(1),

24(2), 27(1), 34(3).

Lepomis gulosus (Cuvier). Warmouth. This species was generally

distributed in the lower half of the drainage. Sites: 18(1), 32(-), 36(1),

37(2), 38(-).

Lepomis humilis (Girard). Orangespotted sunfish. In Kentucky,

this sunfish is sporadic in all but the extreme western part of the state

(Burr 1980). Seven specimens (INHS 76015) were collected from a Buck

Creek headwater site (L. M. Page, pers. comm.). Site: 1(7).

Lepomis macrochirus Rafinesque. Bluegill. The bluegill was reported

from Buck Creek by Carter and Jones (1969) and Harker et al. (1979,

1980). This game fish was generally distributed and common through-

out the drainage. Sites: 1(13), 2(3), 4(1), 6(4), 7(1), 8(2), 9(1), 10(2),

12(3), 15(1), 18(1), 21(2), 23(5), 24(1), 25(1), 27(2), 29(2), 31(1), 33(1),

35(1), 36(11), 37(1), 38(3), 39(-).

Lepomis megalotis (Rafinesque). Longear sunfish. Generally dis-

tributed and common throughout the drainage, the longear sunfish was

previously reported by Carter and Jones (1969) and Harker et al. (1979,

1980). Sites: 1(3), 2(3), 3(4), 7(1), 10(2), 11(1), 12(4), 14(1), 17(1), 19(1),

21(4), 22(3), 23(4), 24(4), 25(2), 27(7), 32(-), 37(2), 38(10).

Lepomis macrochirus x Lepomis megalotis. This is a relatively

common natural hybrid (Trautman 1981). Site: 23(1).

Lepomis microlophus (Giinther). Redear sunfish. One adult speci-

men of L. microlophus was collected from the lower mainstem. As sug-

gested by Burr (1980) in regard to other eastern Kentucky records, this

may have resulted from an introduction. Site: 38(1).

Micropterus dolomieui Lacepede. Smallmouth bass. Previously

reported from two mainstem Buck Creek sites by Carter and Jones

(1969), the smallmouth bass was generally distributed and common

throughout the drainage. Sites: 1(1), 2(1), 9(1), 18(1), 23(3), 24(5), 27(2),

32(3), 38(1).

Micropterus punctulatus (Rafinesque). Spotted bass. Carter and

Jones (1969) reported the spotted bass from all three mainstem sites

Fishes of Buck Creek, Kentucky 145

they surveyed. This species was common and generally distributed in the

drainage. Sites: 15(1), 19(1), 21(1), 22(1), 23(4), 24(5), 27(3), 31(2), 32(2),

36(7), 37(3), 38(1).

Micropterus salmoides (Lacepede). Largemouth bass. Although

reported by Carter and Jones (1969) from a mainstem site and by

Harker et al. (1980) from Brushy Creek, the largemouth bass was spo-

radic in Buck Creek and was collected only from the lower mainstem

during our survey. Sites: 32(-), 38(1).

Pomoxis annularis Rafinesque. White crappie. This species was

generally distributed and abundant in the lower mainstem influenced by

Cumberland Reservoir. Sites: 32(-), 35(2), 38(2), 39(-).

Percidae — perches

Etheostoma blennioides Rafinesque. Greenside darter. Harker et al.

(1979, 1980) reported this species from Buck and Brushy creeks. This

darter was generally distributed but seldom abundant within the drain-

age, and was absent from the part influenced by Cumberland Reservoir.

Adults were usually collected from substrates that ranged from coarse

gravel to boulder riffles with moderate current. Sites: 2(1), 3(2), 5(1),

6(1), 7(3), 8(1), 10(1), 13(1), 15(1), 17(1), 18(1), 19(1), 21(1), 23(3), 24(1),

25(2), 27(45), 31(1), 32(15), 33(1).

Etheostoma caeruleum Storer. Rainbow darter. The rainbow darter

was reported by Harker et al. (1979, 1980) from Buck and Brushy

creeks. It was the most common and widely distributed darter in the

drainage. Collections were made over substrates ranging in size from

medium gravel to cobble in slow to moderate current. Sites: 1(26), 2(6),

3(3), 4(2), 5(9), 6(5), 7(3), 8(3), 9(1), 10(5), 11(2), 12(14), 13(2), 14(2),

15(1), 17(4), 18(5), 19(1), 20(2), 21(1), 22(2), 23(36), 24(26), 25(5), 27(14),

31(1), 32(17), 33(2).

Etheostoma camurum (Cope). Bluebreast darter. Harker et al.

(1979) reported the bluebreast darter from Buck Creek. It was generally

distributed in the swiftest areas of mainstem riffles where substrates var-

ied from coarse gravel to boulders, and was often collected with Etheo-

stoma maculatum Kirtland, another species of the subgenus Nothonotus.

Sites: 12(1), 19(7), 23(3), 24(4), 25(8), 27(1), 31(2), 32(24), 33(3).

Etheostoma cinereum Storer. Ashy darter. This darter is known

from Buck Creek as a result of one specimen collected in 1954 (UL

5392) and seven specimens collected in 1955 (UMMZ 171557, 171590)

(B. M. Burr, pers. comm.; C. R. Gilbert, pers. comm.). Sites: 22(2),

27(6).

Etheostoma flabellare Rafinesque. Fantail darter. This darter was

reported by Harker et al. (1979, 1980) from Buck and Brushy creeks. It

was generally distributed in moderate to swift riffles and flowing pools

146 Ronald R. Cicerello and Robert S. Butler

over fine to coarse gravel in the upper half of the drainage. Sites: 1(6),

2(4), 3(2), 5(3), 6(2), 7(1), 8(2), 9(1), 10(2), 12(24), 13(1), 14(1), 15(2),

17(1), 18(1), 19(2), 21(3), 23(38), 24(9).

Etheostoma maculatum Kirtland. Spotted darter. This species was

generally distributed in the lower half of the drainage and often

occurred with E. camurum under large slab boulders in moderate to

swift current. Harker et al (1979) previously reported the spotted darter

from Buck Creek. Sites: 14(3), 19(2), 23(6), 24(1), 25(12), 27(14), 31(1),

32(25), 33(1).

Etheostoma spectabile (Agassiz). Orangethroat darter. One some-

what aberrant specimen (INHS 87624) (L. M. Page, pers. comm.) was

collected from a small, headwater tributary. Site: 2(1).

Etheostoma stigmaeum (Jordan). Speckled darter. Harker et al.

(1979, 1980) reported the speckled darter from Buck and Brushy creeks.

It was generally distributed and was often collected in sluggish runs over

a silty-sand substrate. However, adults were occasionally taken from the

margins of coarse gravel riffles. Sites: 1(4), 3(5), 5(1), 8(1), 10(7), 12(5),

14(3), 16(1), 21(3), 22(2), 23(5), 24(8), 25(5), 27(6), 31(2), 32(5).

Etheostoma virgatum (Jordan). Striped darter. Harker et al. (1979,

1980) reported this species from Buck and Brushy creeks. The striped

darter was generally distributed and, as reported by Page and Schemske

(1978), is the only slab-pool Catonotus occupying the drainage. Sites:

1(20), 3(7), 4(2), 5(3), 7(1), 8(1), 10(2), 11(1), 12(1), 14(3), 15(1), 16(2),

17(2), 18(1), 20(1), 21(7), 22(1), 23(5), 24(7), 25(1), 27(4), 29(2).

Etheostoma zonale (Cope). Banded darter. The banded darter was

occasional in the mainstem of lower Buck Creek. We collected it from

the interstices of gravel over swift, boulder and bedrock riffles. Sites:

23(1), 24(3), 25(2), 27(6), 32(11).

Percina caprodes (Rafineque). Logperch. The logperch was reported

from mainstem Buck Creek by Carter and Jones (1969) and Harker et

al. (1979). Generally distributed along the Buck Creek mainstem, it was

collected from a variety of habitats ranging from swift, cobble and

boulder riffles to a slow flowing, silt covered bedrock pool. Sites: 10(2),

19(1), 22(1), 23(1), 24(2), 25(2), 27(4), 31(1), 32(13), 33(2), 36(2), 37(1).

Percina maculata (Girard). Blackside darter. Since the only collec-

tions of this percid were two made in 1955 (C. R. Gilbert, pers. comm.),

the species is either sporadically distributed or possibly extirpated from

Buck Creek. Sites: 21(1), 27(1).

Stizostedion canadense (Smith). Sauger. This species was appar-

ently sporadic in the lower mainstem of Buck Creek where, according to

a local boat dock operator, it is harvested irregularly by fishermen. Site:

none.

Stizostedion vitreum (Mitchill). Walleye. According to a local boat

Fishes of Buck Creek, Kentucky 147

dock operator, fishermen sporadically harvest walleye from the lower

mainstem. Site: 38(1).

Sciaenidae — drums

Aplodinotus grunniens Rafinesque. Freshwater drum. This primar-

ily large-river fish (Smith 1979) was reported by Carter and Jones (1969)

and was generally distributed in the lower mainstem. Sites: 32(1), 38(2),

39(-).

Cottidae — sculpins

Cottus carolinae (Gill). Banded sculpin. The banded sculpin was

reported by Harker et al. (1979) and was occasional in moderate to swift

riffles containing cobble and boulders. Sites: 25(1), 27(5), 28(7), 32(10),

33(1).

DISCUSSION

Seventy-three species of fishes and one hybrid, representing 13 fam-

ilies, were found to occur in the Buck Creek drainage. Approximately

80% consisted of members of the Cyprinidae (23 species), Percidae (14),

Centrarchidae (11), and Catostomidae (10). Of the 121 species reported

by Burr (1980) from the upper Cumberland River drainage in Kentucky,

70 are known to occur in Buck Creek. Of the remaining 51 species, 21

are known from adjacent streams or Cumberland Reservoir and poten-

tially occur in Buck Creek (Table 2).

New distributional records were obtained for Ictiobus bubalus,

Ictalurus furcatus, and Lepomis microlophus in the upper Cumberland

River drainage, and the continued existence of Carpiodes velifer within

the drainage was confirmed. Within the Cumberland River drainage,

Ictiobus bubalus was formerly known to occur only in the lower part of

the river in western Kentucky, which has been impounded to create

Barkley Reservoir (Burr 1980; Lee 1980). Our collection extends the

known range of /. bubalus in the Cumberland River upstream approxi-

mately 854 km from the nearest downstream collection made at river

km 4.8 (D. A. Etnier, pers. comm.). Ictalurus furcatus had not pre-

viously been collected from the upper Cumberland River drainage of

Kentucky (Burr 1980). It was not entirely unexpected, however, since

specimens have been taken from the river in adjacent Tennessee (Glodek

1980; D. A. Etnier, pers. comm.). Lepomis microlophus is sporadic and

uncommon throughout the state, except in the upper Cumberland River

drainage (Burr 1980). Although the specimen from Buck Creek repres-

ents the first record for the upper Cumberland River drainage, the

redear sunfish has been widely stocked in impoundments and is proba-

bly not native to the drainage. Carpiodes velifer is sporadically distrib-

uted in the eastern half of Kentucky (Burr 1980) and was previously

148 Ronald R. Cicerello and Robert S. Butler

known from the upper Cumberland River drainage as a result of two

1925 collections deposited at UMMZ (B. M. Burr, pers. comm.). The

highfin carpsucker persists in the upper Cumberland River drainage

despite extensive habitat alteration resulting from impoundment of 162

km of the mainstem Cumberland River and pollution from coal mining.

These records lend credence to speculation that other fishes, especially

large-river forms, may be collected from the Buck Creek drainage

(Table 2), and emphasize the need to sample such habitat during faunal

surveys.

A total of 14 specimens of Ericymba buccata was collected from

five sites in the Buck Creek drainage by Harker et al. (1979) and during

this survey. Moreover, four additional specimens (EKU 1215) were

recently collected from the adjacent Pitman Creek drainage, Pulaski

County. These new records significantly expand the range of the silver-

jaw minnow in the upper Cumberland River drainage as depicted by

Burr et al. (1980) and Gilbert (1980a). Buck Creek has been rather well

collected (B. M. Burr, pers. comm.), which suggests that the silverjaw

minnow has only recently dispersed into the Buck and Pitman creek

drainages. However, its current distribution closely approximates the

upper Cumberland River drainage on the Cumberland Plateau before

the upstream migration of Cumberland Falls, suggesting that the species

was simply overlooked by previous investigators. According to McGrain

(1966), Cumberland Falls originated on the Pottsville escarpment near

Burnside, Kentucky, and has eroded into the Cumberland Plateau

approximately 72 km to its present location. Prior to the retreat of the

falls, Buck Creek was the most downstream major tributary to the

Cumberland River above the falls, while Pitman Creek and Big South

Fork Cumberland River discharged below the falls. Thus, E. buccata is

now known to occupy all major tributaries to the Cumberland River

upstream from the apparent original location of Cumberland Falls.

Although it may have been introduced into Buck and Pitman creeks via

bait bucket transfer, Burr et al. (1980) mentioned evidence of recent

range expansion by the silverjaw minnow in other states and discussed

the implications of several newly discovered, isolated populations in the

lower Green and Tradewater rivers of Kentucky.

Several alternative mechanisms for the dispersal of E. buccata into

Buck and Pitman creeks, are available. The first involves movement

through subsurface channels, which potentially connect Buck Creek

with adjacent drainages. Karst topography including numerous sink-

holes and subterranean streams is common in the Buck Creek drainage

and extends east and northeast to the Dix and Rockcastle river drainages

and west to the Pitman Creek drainage. Ericymba buccata is present in

the Dix River and western tributaries to the Rockcastle River (Burr et

Fishes of Buck Creek, Kentucky

149

Table 2. Species that potentially occur in Buck Creek, their locality of occur-

rence and source.

Species

Locality of

occurrence

Source

Ichthyomyzon greeleyi

Lampetra aepyptera

A cipenser fulvescens

Polyodon spathula

Anguilla ro strata

Hiodon alosoides

Hiodon tergisus

Esox americanus

Carassius auratus

Hybopsis storeriana

Notropis leuciodus

Notropis volucellus

Notropis sp.

(undescribed sawfin

shiner)

Carpiodes carpio

Ictiobus cyprinellus

Little South Fork

Taylor Branch, Youngs

Creek, and Clear Creek

Cumberland River

Cumberland Reservoir

Statewide

Cumberland River

Cumberland Reservoir

Cumberland Reservoir

Cumberland Reservoir

Cumberland River

Fishing Creek

Little South Fork

Rockcastle River

Rock Creek

Little South Fork

Pitman Creek

Big South Fork

Cumberland River

Obey River

Wolf Creek

Cumberland Reservoir

Cumberland Reservoir

Cumberland Reservoir

Cumberland Reservoir

Upper Cumberland

River

Pomoxis nigromaculatus Cumberland Reservoir

Minytrema melanops

Ictalurus melas

Fundulus notatus

Gambusia af finis

Percina sciera

Pitman Creek

Big South Fork

Comiskey and Etnier (1972)

Walsh and Burr (1981)

Burr (1980)

Henley (1967), Charles et al.

(1979), Axon etal. (1980)

Burr (1980)

Gilbert (1980b)

Henley (1967), Charles et al.

(1979), Axon et al. (1980,

1982)

Henley (1967)

Henley (1967)

Gilbert (1980c)

Harkeretal. (1980)

Comiskey and Etnier (1972)

Gilbert and Burgess (1980)

Harkeretal. (1979)

Harkeretal. (1979)

Warren (1981)

Burr (1980)

Burr (1980)

Lee and Shute(1980)

Burr (pers. comm.)

Henley (1967)

Henley (1967)

Henley (1967)

Henley (1967)

Burr (1980)

Henley (1967), Charles et al.

(1979), Axon etal. (1980,

1982)

Page (1980)

Page (1980)

50 Ronald R. Cicerello and Robert S. Butler

«*> _ — M .-. — M _ , — .— .-.^.^..^^l^^,., VO"^

m^NtN — — «— . — — — — <N — — — (Nro inTtm^^

\conoo — ©Noor- — — TtoornoNoor^oo^v'vs^

(*) — — . — — -< n m n - c*ocr>c4m^<*'^<^

IT) On © © — " On © © OO © OO vO On NO © «/"> On s< s^ s> ^> w^

c

S f*Nr»->TfrTi-<Nu-N«r>ir>«r>vN.r^-NONOc--NO

u

•— ^NOOOON^©NOf^U^Tfrl^rn — NOs/,S^S/<S/lV'S/''S/'S>

JH f^rNlc^c^rvirfrcncoTt^NONONONo/si^i^^i^i^/s/N

J*

u fStriioNO^NONONONONOoooor — ^^,^^1^1^,^,^,/^.

^Tf©OOf^t^OO^V^^©s/'S>S/,S/l'S/l'S/|lS/lS>'S/'^'Sy

<s~©nono — N©r^NOs^v's^s/'s/,s/ls/iSrf's/i'v'ls>s/,v'

vo sO n© «/"> nO no vO

^ On so <N «r> m 00

»h «/-> «n r-» ^r

p^xxxxxxxxxxxxxxx

w* r- no r- «r> 00

JS no «o Tf

>

< « xxxxxxxxxxxxxxxxxxxxx

.0

f*5,^^^,««r^c<c<<M<M<sr*5<*>r^r*)C*>f*)r')m

Fishes of Buck Creek, Kentucky 151

al. 1980; Gilbert 1980a; Branson and Batch 1981) and could have

actively or passively moved via subsurface channels into Buck Creek

and from there into Pitman Creek. Dispersal through Cumberland

River Reservoir constitutes the second mechanism. Guillory (1978) dis-

cussed active and passive dispersal via the main channel of the lower

Mississippi River by Notropis longirostris (Hay), a small-stream species.

Burr et al. (1980) interpreted the presence of E. buccata in the Green

River main channel as evidence of direct dispersal to tributary streams.

The final mechanism, stream capture, is plausible, but potential sites of

piracy with adjacent drainages could not be identified. Whether E. buc-

cata is expanding its range or is limited to the Cumberland River drain-

age on the Cumberland Plateau and Pitman Creek can only be deter-

mined by periodic surveys of the Fishing Creek and Big South Fork

Cumberland River fish faunas, which apparently do not currently

include this species (Comiskey and Etnier 1972; Harker et al. 1980).

Two species assigned protection status in Kentucky by the Ken-

tucky Academy of Science (Branson et al. 1981) are known from the

Buck Creek drainage. Notropis ariommus was listed as of undetermined

status but must be considered rare in Buck Creek, from which only 5

specimens are known. Etheostoma cinereum was listed as endangered

and is known from only four drainages within the Cumberland river

system of Kentucky, including Buck Creek (Burr 1980; Warren 1981).

This large-stream and river darter prefers" cover such as boulders,

undercut banks, and rubble-gravel substrate mixed with detritus and /or

Justicia americana in sluggish current adjacent to swift shoals (Saylor

1980; Warren 1981). Although the ashy darter has not been collected

from Buck Creek since 1955, it may persist in suitable habitat along the

mainstem between KY 80 and KY 192.

Three faunal units were discerned when the fish faunas of the 21

mainstem collecting sites were analyzed to determine average faunal

resemblance (Table 3). The units were comprised by sites 1-27, 31-33,

and 35-39, respectively, (hereinafter referred to as Units 1, 2, and 3)

based on greater than 50% shared fauna. Divergence from this standard

within each unit is attributed to sampling artifact.

The fauna of Unit 3 (Table 4) was characteristic of low-gradient

habitats such as lakes, impoundments, and medium-to-large rivers

(Pflieger 1975; Smith 1979; Trautman 1981) and was markedly different

from that of the other units. Fifteen of the twenty-nine species collected

from Unit 3 were limited in distribution to this section of Buck Creek.

Faunal differences were less pronounced between the two remaining

units (Table 3). Sites 1-14 were faunistically similar to 21-27 but gener-

ally shared 40% or less of the fauna with Unit 2 (sites 31-33). Sites 21-27

152

Ronald R. Cicerello and Robert S. Butler

T3

3

"5b

c

o

T3

C

oj

c

o

X

X X X X

X X

X

X

X X X X X X

XXXXXXXXX XX

X XX XX

X

X

XXX XX

XXX XXX

XX XX

xxxxxxxxxxxxxxx xxxx xxxx

X X XX XXX XXX XXX XXX

XXXXX XXXXXXXXXXXXX XX XXX

xxxxxxxxxxxxxx xxxxxxx

X X

XX X

X xxxx

X

X X

X X

XXX

X XX XXX

XXXXXXXXXXXXX X X

XX XXXXXXXXX X

XX XXXXXXXXX XX

XXX XXXXX XXX XXX

XXXX X XXX XX X

XX XXXXX XXX xxxx

XXX XX X xxxxxxx

xxxxxxxxxxxxxxxx

o

a

C/2

0

S 5

? H

S e -

Jl

"a J

**> R:

SI *

2 2 5

if

uj ^ ^3 uj

■2 3

N

o §

3 s "5. ?

a I i 1 -a s

o R a ° a ■§

a to

E I 3

o fc i-

*> ^ 2

Is I

•2 *S .2 -a «

3

II

s. •*

uj < 5:'uj

S3"

§.a

3 ^

11

*1

^uj^'^^^kj^^^ujcs:

Fishes of Buck Creek, Kentucky 153

X X X X X

X X xxxxxxxxx xxxxxxxxx

X X XX

X X XXX X

XX X X X X

XXXX X XXX

XXXX X xxxxxxxxxxxxxx

X XX XX

X XXXXXX XXX

X XXX

X X XX XXX

X X XX XXXX

XXXX

XXXXXX

lis I -«

S ^ .. ] r ^» *S ^ ^ S-5o^ "5 § .a 5 S I 3 * * >, "Sac

I |if I I §-511 * |1 II a J II § J -E I o |l gA g 3| |ll I

» s 2 2 s s I It II § st 5 ?i 1 I iVls-lf 1 1-§ ■!■§■§ ^ I I S g

154 Ronald R. Cicerello and Robert S. Butler

generally shared 50% or more of the fauna with sites 31-33 and exhi-

bited a gradual increase in similarity to these sites with each successively

closer site. The transitional nature of the fauna of sites 21-27 relative to

sites 1-14 and 31-33 is reflected by faunal resemblance values calculated

by pooling the fauna of each group of sites. Sites 21-27 had 72% and

79% shared fauna with 1-14 and 32-33, respectively. Sites 1-14 and 31-33

had 49% shared fauna.

Analysis of the species characteristic of Units 1 and 2, based upon

the occurrence of a species in greater than 50% of the sites comprising

each unit, further illustrates the distinctions and similarities between the

two units (Table 4). Species characteristic of Unit 1 included Notropis

ardens, N. chrysocephalus, N. telescopus, Semotilus atromaculatus ,

Fundulus catenatus, Etheostoma flabellare , and E. virgatum. These are

generally creek or small-stream fishes that migrate downstream in fall to

overwinter in larger and deeper waters (Pflieger 1975; Smith 1979;

Trautman 1981) such as that present in sites 21-27. Hybopsis amblops,

H. dissimilis, Notropis ariommus, N. atherinoides, N. photogenis, N.

rubellus, N. spilopterus, N. whipplei, Labidesthes sicculus, Etheostoma

camurum, E. maculatum, and Cottus carolinae were characteristic of

Unit 2 and typically inhabit moderate-to-large streams and small rivers

(Pflieger 1975; Etnier 1976; Smith 1979; Trautman 1981). Elsewhere in

Buck Creek these fishes were collected almost exclusively from the

downstream portion of Unit 1, sites 21-27. This indicates that sites 21-27

supported a mixture of small-stream, large-stream, and small-river

fishes and may be considered an area of transition between the faunas

of Units 1 and 2. Several species were excluded from this analysis

because they: (1) were collected from sites in all three units and were

considered ubiquitous (Campostoma oligolepis, Lepomis macrochirus ,

L. megalotis, Micropterus dolomieui, and Percina caprodes), or (2)

occurred extensively throughout Units 1 and 2 and were thus not useful

in identifying differences between these units (N. galacturus, Pimephales

notatus, Hypentelium nigricans, Etheostoma blennioides, E. caeruleum,

and E. stigmaeum).

The pattern of longitudinal distribution of fishes along the Buck

Creek mainstem was similar to that reported by Guillory (1982) for a

Louisiana stream, and involved: (1) the addition of species in the middle

section of Buck Creek (sites 21-33) to those widely distributed through-

out sites 1-27, and (2) replacement of upper- and middle-river species

with those typical of low-gradient, big-river habitat in the lower

impounded section of Buck Creek (Table 4). A general downstream

increase in the number of species was also noted (Table 4) and has been

reported for other streams (Kuehne 1962; Larimore and Smith 1963;

Sheldon 1968; Lotrich 1973; Guillory 1982). We believe that this pattern

Fishes of Buck Creek, Kentucky 155

would have been more distinct if an additional collection site in the

intermittent headwaters of Buck Creek had been established, and through

further sampling of deepwater habitat in the downstream section of

Unit 2 and in Unit 3.

The faunas of each unit are a product of the diverse conditions

found among the units. Numerous physicochemical factors that deter-

mine stream habitat diversity have been correlated with the longitudinal

succession of stream fishes. These factors include, but are not limited to,

depth (Sheldon 1968), drainage area (Larimore and Smith 1963), gra-

dient (Trautman 1942; Burton and Odum 1945), pool size (Minckley

1963), stream order (Kuehne 1962; Lotrich 1973), and stream width

(Burton and Odum 1945). Physicochemical factors excluded as primary

causes of observed faunal differences between the upper and middle sec-

tions of Buck Creek included water quality, which Harker et al. (1979,

1980) reported as similar between the two areas, and substrate, which

was characteristically cobble, slab boulder, and bedrock throughout the

stream.

Observed faunal differences probably resulted from the interrela-

tionship of numerous physicochemical factors as postulated by Guillory

(1982). Since species characteristic of faunal Unit 1 of Buck Creek were

also an important component of the Unit 2 fauna, the habitat require-

ments of additional species characteristic of Unit 2 may explain faunal

differences (Table 4). These species (e.g., N. atherinoides, N. photogenis,

H. dissimilis, E. camurum, E. maculatum) typically inhabit moderate-to-

large streams and rivers and are thus adapted to relatively stable envir-

onments. We speculate that factors such as discharge and permanence

of flow are important determinants of faunal differences observed

between Units 1 and 2. These factors, which are more constant in the

large-stream or small-river habitat of Unit 2 than Unit 1, are closely

interrelated with several of the previously cited physicochemical factors,

and are important to the maintenance of a stable environment.

The fauna of the lower impounded river was also related to the

type of habitat present. Within this part of Buck Creek gradient is low

and the stream is deep (10-25 m), providing habitat suitable for large-

river and lentic species (Table 4).

ACKNOWLEDGMENTS.— We thank E. F. Crowell, C. F. Gor-

ham, and B. T. Kinman of the Kentucky Department of Fish and Wild-

life Resources, and J. M. Clayton and S. L. Steele, for field assistance.

For generously sharing collection information, confirming identifica-

tions, and other courtesies we are grateful to: B. A. Branson and J. C.

Williams, Eastern Kentucky University; B. M. Burr, Southern Illinois

156 Ronald R. Cicerello and Robert S. Butler

University at Carbondale; D. A. Etnier, University of Tennessee; C. R.

Gilbert, Florida State Museum, University of Florida; R. E. Jenkins,

Roanoke College; L. M. Page, Illinois Natural History Survey; S. P.

Rice, Kentucky Transportation Cabinet; and M. L. Warren, Jr., Ken-

tucky Nature Preserves Commission. For critical review of the manu-

script and helpful comments we acknowledge B. A. Branson, Eastern

Kentucky University; B. M. Burr, Southern Illinois University at Car-

bondale; and K. E. Camburn and M. L. Warren, Jr., Kentucky Nature

Preserves Commission.

LITERATURE CITED

Axon, James R., B. T. Kinman and C. Gorham. 1982. Annual performance

report for statewide fisheries research project. Part I, Subsection I: Sport

fishery investigation. Ky. Dep. Fish Wildl. Resour. Fish. Div., D-J Proj.

F-40-4, Frankfort. 76 pp.

, W. N. McLemore, D. B. Bell, B. D. Laflin, J. P. Henley, L. E.

Kornman, A. R. Jones and K. W. Prather. 1980. Annual performance

report for statewide fisheries management project. Part I, Subsection I;

Reservoirs research and management. Ky. Dep. Fish Wildl. Resour. Fish.

Div., D-J Proj. F-50-2, Frankfort. 150 pp.

Branson, Branley A., and D. L. Batch. 1981. Fishes of the Dix River, Kentucky.

Ky. Nature Preserves Comm. Sci. Tech. Ser. 2:1-26.

, 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-

lippe and P. Cupp. 1981. Endangered, threatened, and rare animals and

plants of Kentucky. Trans. Ky. Acad. Sci. ¥2(3-4):77-89.

Burr, Brooks M. 1980. A distributional checklist of the fishes of Kentucky.

Brimleyana 3:53-84.

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

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

4/(l-2):48-54.

Burton, G. W., and E. P. Odum. 1945. The distribution of stream fish in the

vicinity of Mountain Lake, Virginia. Ecology 25(2): 182-194.

Carter, James P., and A. R. Jones. 1969. Inventory and classification of streams

in the upper Cumberland River drainage of Kentucky. Ky. Dep. Fish

Wildl. Resour. Fish. Bull. 52:1-70.

Charles, James R., W. M. McLemore, D. E. Bell, B. D. Laflin, J. P. Henley, L.

E. Kornman, A. R. Jones and K. W. Prather. 1979. Annual performance

report for statewide fisheries management project. Part I, Subsection I;

Reservoirs research and management. Ky. Dep. Fish Wildl. Resour. Fish.

Div., D-J Proj. F-50-1, Frankfort. 116 pp.

Comiskey, Charles E. 1970. The fishes of the Big South Fork of the Cumber-

land River system. Unpubl. Master's thesis, Univ. Tenn., Knoxville. 92 pp.

, and D. A. Etnier. 1972. Fishes of the Big South Fork of the Cum-

berland River. J. Tenn. Acad. Sci. 47(4): 140- 145.

Fishes of Buck Creek, Kentucky 157

Etnier, David A. 1976. The Tennessee species of the family Percidae. Unpub-

lished keys. Univ. Tenn., Knoxville. 29 pp.

Evermann, Barton W. 1918. The fishes of Kentucky and Tennessee: a distribu-

tional catalogue of the known species. Bull. Bur. Fish. 35:295-368.

Gilbert, Carter R. 1980a. Ericymba buccata Cope, Silverjaw minnow. P. 156 in

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

Mus. Nat. Hist., Raleigh, x + 867 pp.

. 1980b. Hiodon alosoides (Rafinesque), Goldeye. P. 74 m D. S. Lee,

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

Hist., Raleigh, x + 867 pp.

. 1980c. Hybopsis storeriana (Kirtland), Silver chub. P. 194 in D. S.

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

Nat. Hist., Raleigh, x + 867 pp.

, and G. H. Burgess. 1980. Notropis volucellus (Cope), Mimic shiner.

P. 322 in D. S. Lee, et al. Atlas of North American Freshwater Fishes. N.

C. State Mus. Nat. Hist., Raleigh, x + 867 pp.

Glodek, Garrett S. 1980. Ictalurus furcatus (Lesueur), Blue catfish. P. 439 in D.

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

Nat. Hist., Raleigh, x + 867 pp.

Guillory, Vincent. 1978. New interpretations of fish dispersal via the lower

Mississippi River. Fla. Sci. ¥7:96-100.

. 1982. Longitudinal gradients of fishes in Thompson Creek, Louisi-

ana. Southwest. Nat. 27(1): 107- 11 5.

Hannan, Richard R., M. L. Warren, Jr., K. E. Camburn and R. R. Cicerello.

1982. Recommendations for Kentucky's outstanding resource water classi-

fications with water quality criteria for protection. Tech. Rep. Ky. Nature

Preserves Comm., Frankfort. 459 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.

Henley, James P. 1967. Lake Cumberland investigations. Ky. Dep. Fish Wildl.

Resour. Fish. Bull. 25:1-26.

Jenkins, Robert E., E. A. Lachner and F. J. Schwartz. 1972. Fishes of the cen-

tral Appalachian drainage: their distribution and dispersal. Pp. 43-1 17 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., Blacksburg. 306 pp.

Jordan, David S., and J. Swain. 1883. List of fishes collected in the Clear Fork

of the Cumberland, Whitley County, Ky., with descriptions of three new

species. Proc. U. S. Natl. Mus. 6:248-251.

158 Ronald R. Cicerello and Robert S. Butler

Kirsch, Philip H. 1892. Notes on the streams and fishes of Clinton County,

Kentucky, with a descsription of a new darter. Bull. U. S. Fish. Comm.

70:289-292.

. 1893. Notes on a collection of fishes from the southern tributaries of

the Cumberland River in Kentucky and Tennessee. Bull. U. S. Fish.

Comm. 77:257-268.

Kuehne, Robert A. 1962. A classification of streams, illustrated by fish distri-

bution in an eastern Kentucky creek. Ecology 43 (4): 608-6 14.

Larimore, R. W., and P. W. Smith. 1963. The fishes of Champaign County,

Illinois, as affected by 60 years of stream changes. 111. Nat. Hist. Surv. Bull.

25:299-382.

Lee, David S. 1980. Ictiobus bubalus (Rafinesque), Smallmouth buffalo. P. 404

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

Mus. Nat. Hist., Raleigh, x + 867 pp.

, and J. R. Shute. 1980. Ictiobus cyprinellus (Valenciennes), Bigmouth

buffalo. P. 405 in D. S. Lee, et al. Atlas of North American Freshwater

Fishes. N. C. State Mus. Nat. Hist., Raleigh, x + 867 pp.

Long, Charles A. 1963. Mathematical formulas expressing faunal resemblance.

Trans. Kans. Acad. Sci. (56(1): 138-140.

Lotrich, Victor. 1973. Growth, production, and community composition of

fishes inhabiting first-, second-, and third-order streams of eastern Ken-

tucky. Ecol. Monogr. ¥3(3):377-397.

Mayes, Sudderth, and Etheredge, Incorporated. 1975. The river basin water

quality management plan for Kentucky. Upper Cumberland River. Ky.

Dep. Nat. Resour. Environ. Prot., Frankfort.

McGrain, Preston. 1966. Geology of the Cumberland Falls State Park area. Ky.

Geol. Surv. Spec. Publ. 77:1-33.

Minckley, William L. 1963. The ecology of a spring stream Doe Run, Meade

County, Kentucky. Wildl. Monogr. 77:1-124.

Page, Lawrence M. 1980. Percina sciera (Swain), Dusky darter. P. 740 in D. S.

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

Nat. Hist., Raleigh, x + 867 pp.

, and W. Schemske. 1978. The effect of interspecific competition on

the distribution and size of darters of the subgenus Catonotus (Percidae;

Etheostoma). Copeia 1 978(3) :406-4 12.

Pflieger, William L. 1975. The Fishes of Missouri. Mo. Dep. Conserv., Jefferson

City. 343 pp.

Quarterman, Elsie, and R. L. Powell. 1978. Potential ecological/ geological

natural landmarks on the Interior Low Plateaus. U. S. Dep. Interior,

Washington. 738 pp.

Robins, Richard C, 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. 12. 174 pp.

Saylor, Charles F. 1980. Etheostoma cinereum Storer, Ashy darter. P. 635 in D.

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

Nat. Hist., Raleigh, x + 867 pp.

Fishes of Buck Creek, Kentucky 1 59

Sheldon, Andrew L. 1968. Species diversity and longitudinal succession in

stream fishes. Ecology 49(2): 194-198.

Small, James W. 1970. A study of the fish fauna of the Rockcastle River basin

of Kentucky with comments on the effect of strip mine pollution. Unpubl.

Master's thesis, Univ. Ky., Lexington. 53 pp.

Smith, Philip W. 1965. A preliminary annotated list of the lampreys and fishes

of Illinois. 111. Nat. Hist. Surv. Biol. Notes 54: 1-12.

. 1979. The Fishes of Illinois. Univ. 111. Press, Urbana. 314 pp.

Starnes, Wayne C, and L. B. Starnes. 1978. A new cyprinid of the genus

Phoxinus endemic to the upper Cumberland River drainage. Copeia

1978(.3):508-516.

Stauffer, Jay R., Jr., B. M. Burr, C H. Hocutt and R. E. Jenkins. 1982.

Checklist of the fishes of the central and northern Appalachian Mountains.

Proc. Biol. Soc. Wash. 95(l):27-47.

Trautman, Milton B. 1942. Fish distribution and abundance correlated with

stream gradient as a consideration in stocking programs. Trans. N. Amer.

Wildl. Conf. 7:211-224.

. 1981. The Fishes of Ohio, with Illustrated Keys. Ohio State Univ.

Press, Columbus. 782 pp.

US ACE. 1976. Water quality conditions in Lake Cumberland. U. S. Army

Corps of Engineers, Nashville District, Nashville. 106 pp.

Walsh, Stephen J., and B. M. Burr. 1981. Distribution, morphology and life

history of the least brook lamprey, Lampetra aepyptera (Pisces: Petromy-

zontidae), in Kentucky. Brimleyana 6:83-100.

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

fishes. Brimleyana 6:129-140.

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.

Accepted 13 April 1984

Paracricotopus millrockensis,

a New Species of Orthocladiinae

(Diptera: Chironomidae)

from the Southeastern United States

Broughton A. Caldwell

2382 Rockwood Way,

Stone Mountain, Georgia 30087

ABSTRACT. — The adult male and female, and immature stages, of

Paracricotopus millrockensis n. sp. are described from specimens col-

lected in Georgia and North Carolina. This species is also known from

South Carolina. The new species is clearly separable from other species

of Paracricotopus in the larval, pupal, and adult female stages. The

male is very similar to P. niger (Kieff.). The immature stages of P.

millrockensis have been found in small streams associated with stone

substrates, and the larvae feed mainly on detritus and algae.

INTRODUCTION

Saether (1980a) revised and emended the generic diagnoses for all

stages and both sexes of Paracricotopus Thienemann and Harnisch. In

North America only one species of this genus has been described, Para-

cricotopus glaber Saether, from South Carolina. I collected in an urban

Georgia stream specimens that, based on Saether's diagnoses, belong to

Paracricotopus, but the pupal stage clearly differed in several characters

from published descriptions -of known species. This paper describes the

new species and presents information concerning several aspects of its

ecology and distribution.

The various life stages were obtained from specimens reared in iso-

lation. All specimens were preserved in 70 percent ethanol prior to

mounting on microscope slides. Some specimens were dissected, cleared,

or fixed, and all were mounted in Euparal or Canada balsam.

In the following discussions, N= the number of specimens exam-

ined, and measurements are usually expressed as the total range. All

measurements were made with a calibrated ocular micrometer. Except

where otherwise noted, all measurements are given in jum. The numerals

in parentheses indicate the number of features used to derive the range

given. General terminology follows that of Saether (1980b).

Paracricotopus millrockensis n. sp.

Male. — N = 2; one specimen a mature pupa.

Head. Eyes pubescent, without dorsal elongation. Inner verticals

Brimleyana No. 1 1:161-168, October 1985 161

162 Broughton A. Caldwell

1-2 (4); outer verticals 2 (4); clypeal setae 6-10 (2). Lengths of palpal

segments 2-5: 34 (2), 56 (2), 63 (2), 104-106 (2). Antennal ratio 0.62-0.63

(3); flagellomeres brown with darker brown pedicel.

Thorax. Light brown with brown vittae, preepisternum, scutellum,

and postnotum. Antepronotum with 1-2 (4) lateral setae. About 5 short

acrostichals present on anterior portion of scutum, beginning near ante-

pronotum; 6-9 (4) dorsocentrals; 3 (4) prealars; about 4-6 (2) scutellars.

Haltere very faintly brown with basal area darkest.

Wing. Length from arculus to tip 0.97 (2) mm, maximum width

0.32 (2) mm; V.R. of 0.93 (2); 3 (2) squamals.

Legs. Sensilla chaetica not evident at 500X. Lengths and proportions:

Abdomen. Pale brown. Tergites with anterior row of about 2-4 setae,

and median row of about 5-7 setae. Hypopygium as shown in Figure 1.

Anal point 17-26 (2) long with 2 or 3 setae on each side. Laterosternite

IX apparently with 1 or 2 setae. Gonocoxite length 126 (4); gonostylus

length 52-60 (4); apical spine of gonostylus 9 (4) long. Phallapodeme

46-49 (4) long; transverse sternapodeme 63-74 (2) long. H.R. of

2.10-2.42.

Female. — N = 3; one specimen a mature pupa. Similar to male

except for the following differences.

Head. Inner verticals 2-3 (6); outer verticals 3 (6); clypeal setae 11-

12 (2). Lengths of palpal segments 2-5: 31-34 (2), 57 (2), 60-63 (2), 109

(2). Antennal flagellomere lengths: 56-49 (6), 29 (6), 26-29 (6), 27-31 (6),

54-66 (5). Coronal suture absent.

Thorax. 7-8 (3) acrostichals.

Wing. Length from arculus to tip 0.09-0.95 (4) mm; maximum

width 0.36-0.39 (4) mm; 3-5 (4) squamals; V.R. of 0.92 (4).

Legs. 9 and 11 or 12 sensilla chaetica evident on tai of p2 and p3,

respectively. Lengths and proportions:

fe ti tai ta2 ta3

P, 377-394(3) 389-435(4) 206-232(4) 145-157(4) 110-116(4)

P2 336-377(4) 313-383(4) 139-157(4) 75-87(4) 64-67(4)

P3 336-365(4) 383-406(4) 186-203(4) 99-110(4) 84-99(4)

ta4 ta5 LR SV BV

P, 64-81(4) 52-58(3) 0.52-0.57(4) 3.40-3.67(3) 2.55-2.56(2)

P2 35-41(4) 44-52(4) 0.39-0.44(4) 4.67-5.09(4) 3.57-3.88(4)

P3 44-46(3) 49-52(4) 0.49-0.50(4) 3.76-3.91(4) 3.15-3.20(3)

New Species of Paracricotopus

163

Figs. 1-4. Paracricotopus millrockensis, Male: 1, hypopygium, dorsal view to

the left, ventral view with internal structure to the right; 2-3, Female: 2, genita-

lia, ventral view, with internal structure to the right; 3, T IX, dorsal view; Pupa:

4, abdominal tergites and anal end.

164 Broughton A. Caldwell

Abdomen. Setae on tergites forming anterior and median rows,

though separation sometimes not distinct on posterior tergites. Anterior

row consisting of about 2-4 setae. Genitalia as shown in Fig. 2. Gono-

coxite IX with 3-4 larger, and about 3-5 smaller setae. Tergite IX, Fig. 3,

without indication of caudal emargination. Cerci 57-66 (6) long. Notum

with a light triangular area at junction of rami. Ducts of seminal cap-

sules much narrower in anterior length than in posterior length.

Pupa.— N = 6.

Cephalothorax. Pale brown. Thorax with much of dorsal surface

reticulate and faintly rugose. Thoracic horn (Fig. 5) elongate with

rounded apex, entire surface smooth, 74-157 (10) long, 10-20 (7) wide.

Median antepronotals 2; precorneals 3, with Pc2 longest and most

robust, and PC3 shortest and least robust. Dorsocentrals consisting of 2

pair with Dei and DC2 widely separated from each other. Frontal apo-

tome slightly to moderately wrinkled; FS on prefrons up to 57 long.

Abdomen. Pale brown. Chaetotaxy and other details as shown in Fig.

4. T I-VII without shagreen. T VIII-IX with shagreen, most pronounced

on T IX, and very sparse on VIII. T II-VI with median and caudal

bands of spines and spinules; a few spines bifid or trifid. Median band

of spines and spinules on T II not extended as far laterally as other

bands, and may appear interrupted medially. T VII-VIII with caudal

spines and spinules only. Spines on T VI and VII longest, with some up

to 23 long. Conjunctives of T II-VI with several rows of recurved spin-

ules. Sternites II-VIII with faint, uniformly distributed shagreen. S VI-

VII with a few caudal spinules. Pedes spurii A on S IV-VII, most pro-

nounced on S VI. Segment I with 2 non-filamentous L setae, segments

II-VI with 3 non-filamentous setae, segment VII with 3 filamentous L

setae, segment VIII usually with 4 filamentous L setae. Many D, V, and

non-filamentous L setae appearing bifid. Dorsal O setae on T II-VII.

Anal lobe with about 2-4 small apical spines and 3 curved macrosetae of

about equal length. Fringe of anal lobe with about 7-12(12) setae.

Larva. — N = 9; final instar.

Head. Yellowish brown with dark brown mandibles and mentum.

Antenna as shown in Fig. 6. Lengths of antennal segments 1-5; 40-46

(12), 17-20 (12), 7 (12), 6 (12), 4 (12); A.R. of 1.08-1.35 (4). Basal seg-

ment 17 (11) wide; ring organ at base; blade extended to middle of

fourth segment; accessory blade not visible; lauterborn organs about 6

(6) long. Premandible (Fig. 7) simple, with mesal spiny projection. S I

appearing simple in most specimens, apex at most with one weak serra-

tion. Maxilla as shown in Fig. 8. Mandible (Fig. 9) 94-103 (8) long; seta

interna with several branches. Mentum (Fig. 10) with slightly peaked,

but generally rounded median tooth 19-20 (4) wide; median tooth some-

times weakly notched on each side; fifth lateral tooth small. Ventromen-

tal plates about 2-3 (8) wide. Width of ventromental plate/ width of

median tooth 0.10-0.16. Postmentum 126-134 (8) long.

New Species of Paracricotopus

165

Figs. 5-10. Paracricotopus millrockensis, Pupa: 5, thoracic horn and precorneal

setae; 6-10, Larva: 6, antenna; 7, premandible; 8, maxilla, ventral view; 9, man-

dible; 10, mentum and postmental area.

166 Broughton A. Caldwell

Abdomen. Procerci dark brown, up to 26 long, 17 wide, with a basal

curved spur and 2 short median setae. Preapical procercal spur not

developed, represented at most by a slight protrusion. Three long and

two shorter anal setae; longest seta about 450-500 (3) long. Anal tubules

slender, gradually tapering, and about 2.5 times as long as posterior

parapods.

Etymology. — The species is found in Millrock Branch, a stream in

Rockdale County, Georgia, that supports a most diverse and interesting

chironomid fauna.

Holotype. — Reared §, with exuviae, Millrock Branch at Haralson

Mill Road (83°57'24" N, 30°45'41" W), Rockdale County, Georgia, 3

VII 83, leg. B. A. Caldwell. Holotype specimen deposited in the Florida

State Collection of Arthropods (Tallahassee).

Paratypes (9). — Reared 2 (allotype), with exuviae, same data as

holotype; reared §, with exuviae, same data as holotype; $ prepared

from mature pupa, with exuviae, same data as holotype except 3 VI 82;

mature 2 pupa, with exuvium, same data as holotype except 2 VI 78; 2

pupal exuvium, same data as holotype except 20 III 82; 2 larvae, final

instar, same data as holotype; larva, final instar, same data as holotype

except 19 VII 77; larva, final instar, Huffines Mill Creek, Rockingham

County, North Carolina, VIII 81, leg. D. R. Lenat. All paratype speci-

mens are deposited in the Florida State Collection of Arthropods

(Tallahassee).

Diagnosis. — Males of Paracricotopus niger (Kieff.) and P. uligino-

sus (Brund.) are very similar structurally, as has been noted by Albu

(1968) and Saether (1980a). Paracricotopus millrockensis is very similar

to these two species structurally, and consistent separation may not be

possible. I have not borrowed the type material, but differences in these

species might be discovered in the structure of the aedeagal lobe and

gonostylus. Males of P. millrockensis are separable from those of P.

glaber by differences in the hypopygium, especially gonocoxite length,

aedeagal lobe size and shape, and gonostylus shape. In the female, P.

millrockensis is most similar to P. niger, but differences are found in the

genitalia, especially the shape of the coxosternapodeme and the ventral

lobe of gonopophysis VIII. In lateral view, prior to embedding in bal-

sam, the allotype female was noted to have a thinner notum than that

illustrated by Saether (1980a, Fig. 2 A). Also, there are differences in leg

lengths and ratios. The larva and pupa of P. millrockensis are separable

from the other described species in the genus by several characters. In

the pupal stage, P. millrockensis can be separated by the smooth elon-

gate thoracic horn, different abdominal chaetotaxy, and different anal

lobe. Its anal lobe is very similar to that of P. niger, based upon the

figure of Thienemann (1950). Saether (1980a), however, reported 8-15

New Species of Paracricotopus 167

apical spines with an average of 10 in specimens that he examined.

Paracricotopus millrockensis differs from P. glaber in having apical

anal lobe spines and an anal fringe. The larva of P. millrockensis is

separable from these other species by the apparently simple S I, shorter

postmentum, lack of a developed pre-apical procercal spur, minor dif-

ferences in mentum structure, and the anal tubules, which are about 2.5

times as long as the posterior parapods. Other characters that may

further separate P. millrockensis from the structurally similar P. niger

include a lower A.R., more posterior position of the submental setae,

and apparent differences in the shape and size of the anterior lacinial

chaetae of the maxilla (cf. Saether 1980a: 138).

Range. — Paracricotopus millrockensis has been collected in Geor-

gia and North and South Carolina. The preferred habitat appears to be

low order streams in the Piedmont and Blue Ridge Provinces. In Geor-

gia, this species has been collected in Cascade Branch, Habersham

County, and in Millrock Branch, Rockdale County. In South Carolina,

it has been collected in Boone Creek, Oconee County. In North Caro-

lina, specimens of this species have been collected in Beaverdam Creek,

Wake County, Huffines Mill Creek, Rockingham County, and in an

unnamed stream, Macon County. The species is likely to occur in sim-

ilar streams in other southeastern states.

Ecology. — In North and South Carolina streams, and Georgia

streams other than Millrock Branch, larvae have been collected from

among stone substrates in qualitative samples. In the Georgia stream

locally known as Millrock Branch, the species has been collected by

hand-picking from very shallow water on granitic bedrock among moss

and detritus. In this stream, most larvae have been found in association

with Hudsonimyia parrishi Caldwell and Soponis. Some other chiro-

nomids found in the same microhabitat, but not necessarily at the same

time, are listed in Caldwell and Soponis (1982). Millrock Branch drains

a relatively undeveloped, unpolluted watershed with dissolved oxygen

values near saturation. D. Lenat (pers. comm.), however, has collected

larvae in some North Carolina streams influenced by pesticides and

other pollutants. This would suggest that the species is not a good indi-

cator of water quality.

Detritus, fungi, and algae constitute the majority of food observed

in gut contents of several larvae.

ACKNOWLEDGMENTS.— The author wishes to thank Dr. A. R.

Soponis, Florida A & M University, for helpful suggestions on the

initial draft of the manuscript. North Carolina larval specimens and col-

lection records were furnished by Mr. D. R. Lenat, North Carolina

168 Broughton A. Caldwell

Division of Environmental Management. Mr. D. R. Smith, U. S. Environ-

mental Protection Agency, furnished additional collection records. I

would also like to thank Dr. F. K. Parrish for permission to collect

specimens from Millrock Branch on his property.

LITERATURE CITED

Albu, Paula. 1968. Chironomide din Carpatii Romanesti (III). Stud. Cercet.

Biol. Ser. Zool. 20:455-465.

Caldwell, Broughton A., and A. R. Soponis. 1982. Hudsonimyia parrishi, a new

species of Tanypodinae (Diptera: Chironomidae) from Georgia. Fla.

Entomol. 55:506-513.

Saether, Ole A. 1980a. The females and immatures of Paracricotopus Thiene-

mann and Harnisch, 1932, with the description of a new species (Diptera:

Chironomidae). Aquat. Insects 2:129-145.

. 1980b. Glossary of chironomid morphology terminology (Diptera:

Chironomidae). Entomol. Scand. Suppl. 74:1-51.

Thienemann, August. 1950. Lunzer Chironomiden. Arch. Hydrobiol. Suppl.

75:1-202.

Accepted 11 May 1984

Aquatic Distributional Patterns in the Interior Low

Plateau

Branley Allan Branson

Department of Biological Sciences,

Eastern Kentucky University, Richmond, Kentucky 40475

ABSTRACT. — The aquatic gastropod and fish faunas of the Interior

Low Plateau of extreme southern Indiana, Illinois (Wabash River

drainage), Kentucky, Tennessee, and northern Alabama reflect the

interaction of tectonic changes, glaciation, eustatic stream modifica-

tions, piracy within the Low Plateau and in extralimital drainages

(particularly the Coosa-Alabama system), immigration from trans-

Mississippian systems, speciation within drainages that cross the Low

Plateau, and survivorship as relicts. Examples in the Pleuroceridae,

Unionidae, and several fish families are discussed, with emphasis on

percid darters and the catfish genus Noturus.

INTRODUCTION

The Interior Low Plateau of extreme southern Indiana, Illinois

(Wabash River drainage), Kentucky, Tennessee, and northern Alabama

(Fenneman 1938; Quarterman and Powell 1978) is a biological cross-

roads between regions that are faunistically and floristically rich.

Understanding of the biological importance of this unique region has

come very slowly, piecemeal really, mostly because of inadequate study

and failure in the synthesis of existing information. The present biota is

quite complex, consisting of mixtures of types from diverse centers of

origin, including a rather large number of relicts and endemics. To

extrapolate from knowledge of existing biotas back into the past in

order to understand the present is not a bad approach, particularly if

there are considerable supporting geologic and paleontologic data

available.

MOLLUSCA

In the discussion that follows, I have elected to retain Goodrich's

nomenclature (see literature cited) rather than the combinations recently

espoused by Burch (1982), since most readers are not familiar with the

resurrected combinations. Furthermore, there is still considerable dis-

agreement regarding some of the combinations.

In the early part of this century, well-diggers near Henderson, Ken-

tucky, cut into a deposit 25.5 m below the surface. The strata were

determined to be of Yarmouthian Interglacial age (Baker 1920). The

mollusks removed from those deposits included specimens of Campe-

loma crassula Rafinesque, Pleurocera canaliculatum (Say), Planorbula

Brimleyana No. 1 1 : 1 69- 1 89, October 1 985 1 69

170 Branley A. Branson

{Menetus) dilatatus (Gould), and Valvata species. The Yarmouthian fol-

lowed the Kansan glacial epoch and was the longest interglacial. During

that period the climate was slightly warmer than at present (Fenneman

1938), allowing many species to extend their ranges (Baker 1920). Later

investigations of Yarmouthian deposits near Evansville, Indiana (Baker

1920), disclosed the presence of unionid clams and various snails that no

longer live in streams of the Interior Low Plateau. For example, Quad-

rula quadrula asper (Lea 1831) now lives in streams that drain into the

Gulf of Mexico from Alabama to central Texas and northward to Kan-

sas. Another species with derivatives still present in many Low Plateau

streams, Amblema plicata Say, has its principal and parental stocks dis-

tributed from the Alabama River drainage and streams flowing into the

Gulf of Mexico west to central Texas and north to central Kansas.

Three additonal operculated snails were found in the well deposits:

Pleurocera unciale (Haldeman), P. alveare (Conrad), and a species of

Lioplax. The last two still live in the Ohio River and many of its tribu-

taries, whereas the first is now restricted to the upper tributaries of the

Tennessee River in Virginia and eastern Tennessee (Burch 1982; Good-

rich 1940); all have their principal relatives in the Alabama River

system.

Many deposits of the previous Aftonian Interglacial and the Illino-

ian glacial period (Browne and Bruder 1963) indicate climatic condi-

tions that were considerably cooler and moister than at present. Ice

movement, completely overwhelming stream systems to the north,

caused a southward shift of faunas and, upon recession, a re-expansion

northward. Thus, many of the species found themselves exposed to new

environmental and competitive conditions that perhaps stimulated

extensive differentiation into species and races in extralimital areas.

Many gastropod species that are now restricted to more northerly lati-

tudes occurred in the Interior Low Plateau during Pleistocene times

(Browne and Bruder 1963) although some derivative species remain.

They include Amnicola, various species of Lymnaea, and Helisoma

anceps (Menke) (Branson 1972).

The presence of the Alabama and Tennessee river system deriva-

tives in Interior Low Plateau streams, as far north as the Wabash River

of Indiana and Illinois, raises questions regarding migratory pathways

and strengthens a theory previously in vogue. Baker (1920) postulated

existence of an Appalachian river that included parts of the old Teays

and Tennessee rivers and their tributaries above Chattanooga and the

Coosa-Alabama system. Whether it is necessary to invoke the existence

of such a river is open to debate; other mechanisms can explain the

observed distributions. However, that the Alabama River system, in

particular the Coosa basin, has been a potent generator of species is

Interior Low Plateau Distributional Patterns 171

scarcely open to argument. The Coosa River had more endemic species

of operculated gastropods than any other river in North America

(Goodrich 1944a), including members of the families Pleuroceridae and

Viviparidae. Two of eleven species of Somatogyrus that occur in the

Coosa also live in the Tennessee River, and Goniobasis carinifera

(Lamarck) lives in both drainages (Goodrich 1944a).

A very characteristic pleurocerid genus of the Alabama-Coosa sys-

tem is Anculosa (Leptoxis, according to Burch 1982), which has a large

number of endemic species not found outside that system but also has

derivatives that occur elsewhere (Goodrich 1922). Some of these species

occur in Low Plateau drainages, including the related genus Nitocris

(Branson 1972). Nitocris (relegated to Mudalia by Burch 1982) has been

able to extend its range into the Kanawha of West Virginia, the Hiwas-

see of North Carolina, the Tennessee River, and the Ohio and Little

Miami of Ohio, Indiana and Kentucky (Goodrich 1940, 1944a). Ala-

bama River derivatives of Anculosa occur in the Tennessee, Cumber-

land and Green rivers (Burch 1982; Goodrich 1934). Anculosa subglo-

bosa Say lives in the Cumberland River system of central Tennessee

(Goodrich 1921). Anculosa praerosa Say, a secondary Tennessee River

system derivation, occurs in the Tennessee system; in the Cumberland,

Holston, Duck, Clinch, Little Tennessee, and Obey rivers of Tennessee;

in tributaries of the Duck and Tennessee rivers in Alabama (Burch

1982; Goodrich 1944b; TV A 1975); and in the Blue and Wabash rivers

of Indiana. Thus, it is clear that streams of the Interior Low Plateau

have served as important migration pathways into the Ridge and Valley

Province and elsewhere.

There are many endemic pleurocerid species in the Tennessee and

Cumberland river systems (Goodrich 1940). Pleurocera prasinatum

(Conrad) of the Alabama River system is most closely related to P.

canaliculatum (Say) (Goodrich 1935), various forms of which occur in

the Tennessee, Cumberland, Clinch, Kentucky and Ohio rivers. It has

also been able to penetrate into the Wabash of Indiana (Goodrich

1929), doubtless via the Ohio since the ancestral Wabash of Indiana and

Illinois was almost completely overwhelmed during Wisconsin glacia-

tion. Lithasia obovata (Say) and L. geniculata Haldeman are both con-

sidered Tennessee River derivatives. Lithasia geniculata is relatively

widespread in the Interior Low Plateau of the Tennessee River in Ten-

nessee (Burch 1982). It was recently discovered in the southwestern sec-

tion of the Kentucky portion of the Low Plateau (Branson et al. 1983),

and is common in parts of the lower Tennesse (TV A 1975). Lithasia

obovata had spread from the Tennessee River basin into the Green,

Cumberland and Kentucky river basins and, via the Ohio River, into

the Wabash of Indiana and the Scioto of Ohio (Goodrich 1929).

172 Branley A. Branson

Goniobasis laqueata (Say), common in the Tennessee River basin

(TV A 1975) and Cumberland River (Branson and Batch 1982), is

another species that has spread into various segments of the Interior

Low Plateau drainages from the Tennessee basin. Goniobasis semicarin-

ata has its center of distribution in the Kentucky River drainage (Bran-

son and Batch 1981). It entered the Cumberland River (Branson and

Batch 1982), possibly by stream capture such as that documented by

Kuehne and Bailey (1961), and the Salt River of Kentucky and the

Wabash of Indiana (Goodrich 1935), perhaps by tributary hopping and

reinvasion.

Many additional examples in the Pleuroceridae and Unionidae

illustrate the principles involved, but the ones presented here shall suf-

fice. From the very rich centers of endemicity in the Alabama-Coosa

system and secondary speciation centers in the Tennessee and Middle

Cumberland systems, the Pleuroceridae and Viviparidae (Somatogyrus,

Viviparus, Campeloma) expanded northward and westward into the

Cumberland, Green and Lower Tennessee rivers, and via them into the

Ohio River basin. Depauperacy is one of the main features in the north-

ern part of the area. The Tennessee and Cumberland rivers served as

major refugia for unionid species that later reinvaded upstream Ohioan

streams, and the Green River may have served as a refugium as well. All

three rivers were sources for repopulation of the Wabash and Maumee

rivers in postglacial times (Johnson 1980). Pleurocerids were able to

reinvade Ohio and Indiana from similar sources via the Ohio River,

penetrating into the Wabash River and its tributaries. Or perhaps, as

discussed below, they came from some of the other old tributaries of the

Teays system.

FISHES

Fish and mollusk distributional patterns in the Interior Low Pla-

teau must be correlated with various physical and hydrologic phenom-

ena in order to account for observed faunal relationships between

drainage basins. According to Lachner and Jenkins (1971), there are

four principal ways that aquatic organisms have achieved or may

achieve new dispersal distributions: (a) stream capture, (b) eustatic

changes in coastal plains, (c) Pleistocene modifications of drainages,

and (d) movements from one drainage to another via past and existing

interconnecting main streams. These mechanisms were perhaps involved

in structuring the fauna of the Interior Low Plateau. Pleistocene glacia-

tion certainly affected the region, very obviously so in the Shawnee

Hills, the northern boundary of which (38th parallel) is demarked by

glacial till (Harker et al. 1980, and literature cited therein). In this same

general area, numerous karsts and faults crisscross the Green, Pond,

Tradewater, Rough, Barren, and Ohio rivers.

Interior Low Plateau Distributional Patterns 173

To the west and southwest the Plateau is separated from the Coast-

al Plain Province by the Tennessee and Cumberland rivers in Kentucky,

except along a narrow isthmus that abuts Missouri after crossing

extreme southern Illinois, and by the Tennessee River in Tennessee and

northwestern Alabama. The whole plateau is an ecotonal region, past

and present, that lies in the western mesophytic forests, although slender

extensions of the southeastern forests penetrate along rivers. The nar-

row connection between the Ozark Plateau and the Low Plateau

through Indiana and Illinois has exerted an important influence on the

floras and faunas, both east and west (Conant 1960).

In postglacial times, particularly during the so-called Climatic

Optimum, the mesic forest and its streams, many of which are now

buried (Wayne 1952), were apparently much more extensive than at

present. They extended well up into the glaciated parts of Ohio, Indiana

and Illinois. During the Xerothermic period that followed, the mesic

forest vanished from most of the area above the Ohio River (Conant

1960), forcing a shrinkage in ranges of many organisms and leaving

many of them in the narrow isthmus, an area thought to have been

stable along the upper margin of the Mississippian Embayment since at

least Tertiary times. Conant (1960) believed that many organisms used

this narrow isthmus to extend their ranges east and west, including var-

ious species of Eurycea, Plethodon, and Natrix, and the fishes Hybopsis

dissimilis (Kirtland), Noturus eleutherus (Jordan), and Notropis galac-

turus (Cope). Etheostoma microperca Jordan and Gilbert, certain spe-

cies of the percid subgenus Nothonotus, and Notropis telescopus (Cope)

may have been involved in such exchanges as well. However, Pflieger

(1971) noted that the hill country around the head of the Mississippi

Embayment may have played a role in various eustatic changes that

allowed fish exchanges through the Embayment rather than around it

via streams now absent.

Older geologic changes doubtless played a role in determining var-

ious aspects of the Interior Low Plateau's fauna, particularly pre-

Cretaceous and Cretaceous Appalachian peneplanation (Griswold 1895,

and others). One portion of such peneplains has been mapped westward

through the Mississippi Embayment into Arkansas and northern Ala-

bama, an important fact considering the putative sources of many Low

Plateau organisms. Not only is the ancestral Cumberland River sup-

posed to have crossed the northwestern Alabama axis, but there is a

possibility that a Lower Mississippi stream, working backward from the

embayment, captured much of the interior drainage from the old Teays

system (Griswold 1895), transferring many aquatic species with it. Later

events would have been of greater importance.

174 Branley A. Branson

The Teays and the Old Ohio rivers greatly influenced the distribu-

tion of fishes and other aquatic organisms in the Low Plateau. Accord-

ing to Hocutt (1979), the Old Ohio River headed near the present con-

fluences of the Salt River (Kentucky) and the Blue River (Indiana), a

major tributary being the Green River from Tennessee and Kentucky.

The Teays headed in North Carolina and flowed through Ohio, Indiana,

and Illinois, then southward into the Mississippi Embayment; its major

southern tributaries were the Old Kentucky and Old Licking rivers

(Hocutt et al. 1978). Pleistocene glaciation destroyed most of the Lower

Teays, creating the vast Upper Ohio River by westward diversion of the

Big Sandy, Little Sandy, Licking, Kentucky, and Kanawha rivers and,

during Kansan times (Wayne 1952), by diverting a large segment of the

Teays system into the Wabash River. Such drainage modifications

allowed eastern fishes, such as Percina macrocephala (Cope) and

Etheostoma blennioides blennioides Rafinesque, to disperse westward

into Low Plateau streams. This may account for the strong resemblan-

ces between the fish faunas of the Green and Licking rivers (Retzer et al.

1983) and the distribution of Percina cymatotaenia (Gilbert and Meek)

and its undescribed sibling species (Branson and Batch 1974).

Lachner and Jenkins (1971) considered the Teays River to have

functioned as a generation center, a reservoir of species, and a dispersal

pathway. According to them, for example, the ancestral stock of Noco-

mis micropogon (Cope) evolved in the Upper Teays (Kanawha-New)

system, and dispersed from there into the Lower Teays and into the

southwestern Ohio basin. Since the Wabash was nearly completely

overwhelmed by Pleistocene ice, the population in that stream has to be

the result of later re-invasion from refugia to the south. Lachner and

Jenkins (1971) presented evidence that N. micropogon entered the Ten-

nessee River via stream capture, probably by headwater piracies between

the Tennessee, Chattahoochee and Savannah rivers (Ross 1971). Ross

(1971) also documented transfers of the Duck and Elk rivers from the

Cumberland to the Tennessee and stream captures between the Coosa-

Hiwassee and Tennessee rivers in northern Alabama. A Cumberland-

Kentucky river drainage exchange of Etheostoma sagitta (Jordan and

Swain), and doubtless other fishes as well, including N. micropogon (the

only Nocomis in most of the Kentucky River), was documented by

Kuehne and Baily (1961). If true, these exchanges would explain many

of the mollusk and fish distributional patterns mentioned previously

and below, and would strengthen the idea that the Tennessee, Coosa-

Alabama , and Cumberland rivers have been potent differentiation and

dispersal centers for various piscine, molluscan, and other groups of

aquatic organisms.

There appears to have been considerable faunal exchange between

the Alabama River system and the Mobile basin and the Tennessee

River. The Tennessee and Alabama river systems share many species of

Interior Low Plateau Distributional Patterns 175

fishes (Table 1), and the Mobile and Escambia drainages share at least

seven species: Hybognathus hayi Jordan, Pimephales notatus (Rafin-

esque), Notropis baileyi Suttkus and Raney, Etheostoma histrio Jor-

dan and Gilbert, E. proeliare (Hay), Percina ouachitae (Jordan and Gil-

bert) {-P. uranidea (Jordan and Gilbert)), and Stizostedion vitreum

(Mitchill) (Lee et al. 1980; Smith-Vaniz 1968). Many of these species, of

course, range over much of the Low Plateau, although some of them are

more restricted in distribution and some species have been unable to

effect exchange between the river systems. Etheostoma squamiceps Jor-

dan, for example, occurs in southern Illinois, western Kentucky, and

southwestern Indiana (Wabash drainage in Posey County), and in the

Tennessee River system of west-cental Tennessee, Alabama, and Missis-

sippi (Page et al. 1976), but has not been reported from the Alabama

River system. Many Nothonotus and Catonotus show similar distribu-

tional patterns, albeit superimposed upon strong endemicity.

The Tennessee, Cumberland and Green river systems have received

varying contributions to their fish faunas from other systems, and have

made contributions to other drainages in the Interior Low Plateau. All

have served as piscine speciation centers and as reservoirs for endemic

species. One of the sources for Low Plateau fishes was obviously the

Lower Mississippi River system, but the species derived from there

mostly exhibit marginal or extralimital patterns in the Plateau. Excep-

tions are seen in the Green and Tradewater rivers, Kentucky, where they

have effected rather wide distribution. Elsewhere, these fishes retain

populations in more or less stable, protected, relict habitats, like those

reported by Gunning and Lewis (1955) in southwestern Illinois. The

species assemblage of that swampine environment includes fishes with

mostly southern affinities: Aphredoderus sayanus (Gilliams), common

in the Green River; Umbra limi (Kirtland); Fundulus notti (Agassiz);

Elassoma zonatum Jordan, historically common in both the Green and

Tradewater rivers; Lepomis symmetricus Forbes; Centrarchus macrop-

terus (Lacepede), common in the Green River; Etheostoma gracile

(Girard), common in the Green River; and Chologaster agassizi Put-

nam, not southern but Low Plateau. Umbra limi, which also occurs

sporadically along the margin of the Low Plateau (Clay 1975; Sisk

1973), is of northern origin. According to Wiley (1977), the Fundulus

notti species complex originated in the Lower Mississippi basin and,

abetted by stream captures between the Mississippi and the Mobile Bay

drainages, spread elsewhere. Fundulus notti barely penetrates south-

western Kentucky (Burr 1980) and Tennessee (Baker 1939) outside the

Low Plateau. In Tennessee, the only non-embayment record for F. notti

is from the Big Sandy River, a system with a host of embayment spe-

cies, although the species is widespread in the Obion, Forked Deer and

Hatchie rivers (D. Etnier, pers. comm.).

176 Branley A. Branson

A large percentage of the fishes in southern Indiana and adjacent

Illinois and Kentucky are of southern or lowland origin, many of them

doubtless gaining entry in post-glacial times by migration through the

Ohio River and its tributaries. Illinois, for example, has two species

complexes that are coincidental with the Mississippi and Ohio river

drainages, respectively (Forbes 1909; Smith 1979), entering the area via

the Wabash and smaller Ohio River tributaries. Included in this list are:

Ichthyomyzon bdellium (Jordan), Lampetra aepyptera (Abbott), No-

tropis atherinoides Rafinesque, N. fumeus Evermann, N. shumardi

(Girard), N. venustus (Girard), N. volucellus (Cope), Ericymba buccata

Cope, Nocomis micropogon (Cope), Hybopsis amblops (Rafinesque),

H. gracilis (Richardson), H. meeki Jordan and Evermann, Noturus fla-

vus Rafinesque, N. miurus Jordan, N. eleutherus Jordan, N. stigmosus

Taylor, Fundulus olivaceus (Storer), Aphredoderus say anus (Gilliams),

Lepomis megalotis (Rafinesque), Centrarchus macropterus (Lacepede),

Micropterus punctulatus (Rafinesque), Elassoma zonatum Jordan,

Ammocrypta pellucida (Putnam), Percina ouachitae (Jordan and Gil-

bert), Etheostoma blennioides Rafinesque, E. fusiforme (Girard), E.

histrio (Jordan and Gilbert), E. kennicotti (Putnam), E. proeliare (Hay),

and E. squamiceps Jordan. Properly speaking, Etheostoma proeliare is

a fish of the Coastal Plains and Mississippi Embayment (Burr and Page

1978), as is E. fusiforme (Sisk 1973), and the distributional center of

the subgenus Ammocrypta appears to have been in the Lower Missis-

sippi basin (Williams 1975). In the Low Plateau region, A. pellucida is

the most widespread member, but A. clara Jordan and Meek is known

from the Green River in Kentucky. Fundulus chrysotus (Giinther),

Notropis maculatus (Hay), and Menidia beryllina (Cope) are all distinc-

tive Gulf Coastal Plains fishes that barely impinge upon the Low Pla-

teau in Kentucky without actually penetrating its drainages (Sisk 1973;

Baker 1939; Burr and Mayden 1979).

As stated previously, in Kentucky the Land Between the Lakes

region (Lower Tennessee-Lower Cumberland rivers) separates the Low

Plateau from the Gulf Coastal Plains. The mix of fishes in this area

reflects the various centers of origin; some examples are presented in

Table 2 (McDonough 1974; Resh et al. 1972). A similar picture is pre-

sented by a partial list of fishes from Reelfoot Lake (Table 3), which

barely laps northward into Kentucky (Parker 1939; Baker 1939). Some

of the fishes in this area and elsewhere in the Low Plateau gained access

to the region from the north, possibly via a temporary post- Wisconsin

connection between the Erie and Wabash drainages, and Indiana's

White and Big Blue drainages (Gerking 1945). They include Umbra limi

(Kirtland), Rhinichthys atratulus (Hermann), Percopsis omiscomaycus

(Walbaum), and Fundulus catenatus (Storer). Jordan (1877) reported

Interior Low Plateau Distributional Patterns

177

Table 1. Fishes shared by the Tennessee and Alabama River systems (Lee et

al. 1980; Smith-Vaniz 1968).

Ichthyomyzon castaneus

Ichthyomyzon gagei

Lampetra aepyptera

Amia calva

A cipenser fulvescens

Scaphirhynchus platorhynchus

Polyodon spathula

Lepisosteus oculatus

Lepisosteus osseus

Anguilla rostrata

Alosa chrysochloris

Alosa alabamae

Dorosoma cepedianum

Hiodon tergisus

Esox americanus

Esox niger

Campostoma anomalum

Hybognathus hayi

Hybognathus nuchalis

Hybopsis storeriana

Nocomis leptocephalus

Nocomis micropogon

Notemigonus crysoleucas

Notropis atherinoides

Notropis baileyi

Notropis bellus

Notropis chrysocephalus

Notropis lirus

Notropis venustus

Notropis volucellus

Notropis whipplei

Opsopoeodus emiliae

Pimephales notatus

Pimephales vigilax

Rhinichthys atratulus

Semotilus atromaculatus

Carpiodes cyprinus

Cycleptus elongatus

Erimyzon oblongus

Ictiobus bubalus

Ictiobus cyprinellus

Minytrema melanops

Moxostoma carinatum

Moxostoma duquesnei

Moxostoma erythrurum

Moxostoma macrolepidotum

Ictalurus furcatus

Ictalurus melas

Ictalurus natalis

Ictalurus nebulosus

Ictalurus punctulatus

Noturus gyrinus

Pylodictis olivaris

Fundulus olivaceus

Gambusia affinis

Labidesthes sicculus

Morone chrysops

M or one mississippensis

Ambloplites rupestris

Lepomis cyanellus

Lepomis gulosus

Lepomis humilis

Lepomis macrochirus

Lepomis megalotis

Lepomis microlophus

Micropterus punctulatus

Micropterus salmoides

Pomoxis annularis

Pomoxis nigromaculatus

Stizostedion vitreum

Percina caprodes

Percina maculatum

Percina ouachitae

Percina shumardi

Etheostoma nigrum

Etheostoma stigmaeum

Cottus carolinae

Aplodinotus grunniens

1 78 Branley A. Branson

Etheostoma camurum (Cope), E. variatum Kirtland, E. spectabile

(Agassiz), Ammocrypta pellucida (Putnam), and Percina copelandi

(Jordan) from the White River in Indiana, as well as the minnows

Hybopsis dissimilis (Kirtland) and Notropis ariommus (Cope), and the

sucker Erimyzon oblongus (Mitchill), most of these doubtless re-invading

during post-glacial times via the Ohio River. The Wabash River, how-

ever, has been the principal Low Plateau pathway of piscine re-invasion

into Indiana and Illinois (Table 4).

At least one species in this area, Clinostomus funduloides Girard,

reported from the Lower Tennessee (Miller 1978), Cumberland (Burr

1980), and Little Sandy rivers and several other streams in northeast

Kentucky (Bauer and Branson 1979), the Wabash drainage in Indiana

(Lee et al. 1980; Gerking 1945; Blatchley 1938), and the Barren and

Green rivers (Retzer et al. 1983), has a strongly pre-glacial relict distri-

bution. This may be true also of Rhinichthys atratulus.

Fishes considered by Etnier (unpublished) to have strong lowland

and Lower Mississippi affinities are presented in Table 5. In addition,

he believes that several species of Low Plateau fishes are derivable from

areas west of the Mississippi, from the Ozarkian and Great Plains fau-

nal regions: Hybopsis gracilis (Richardson), H. storeriana (Kirtland),

Hybognathus hayi Jordan, H. nuchalis Agassiz, H. placitus Girard,

Notropis lutrensis (Baird and Girard), N. camurus (Jordan and Meek),

N umbratilis (Girard), N. fumeus Evermann, and N. stramineus (Cope).

Such faunal exchanges could have occurred either via the aforemen-

tioned isthmus across southern Illinois and Indiana, or via the

Mississippi-Ohio system. To this list should be added the percid subge-

nus Nothonotus (Zorach 1972; Harker et al. 1980) (see discussion

below).

Although many fish species enjoy wide distribution throughout the

Low Plateau, many others are restricted to certain portions of the area.

One of the most interesting of such patterned distributions is endemic-

ity, important in biogeographic studies and presenting several implica-

tions. Applied specifically to the Interior Low Plateau aquatic problem,

endemicity may reflect interrupted gene flow imposed by isolation

resulting from drainage modification and control (cut off from sur-

rounding drainages) by master rivers like the Green (Kuehne 1966) and

extralimital origins, dispersal into other drainages and modification and

divergence in the new system. For example, Zorach (1972) proposed

that the ancestral stock of Nothonotus arose west of the Mississippi in

the Arkansas or Red River systems and dispersed from there into the

Ohio and Tennessee systems, where evolutionary divergence occurred.

The Tennessee and Middle Cumberland, and the Green-Barren rivers,

seem to have been of great importance as speciation centers. Three dif-

Interior Low Plateau Distributional Patterns

179

Table 2. Some Land Between the Lakes fishes of Kentucky (from McDonough

1974, and Reshetal. 1972).

Ichthyomyzon bdellium

Lepisosteus oculatus*

Lepisosteus osseus

Lepisosteus platostomus

Amia calva*

Alosa chrysochloris

Hiodon tergisus

Hybopsis storeriana

Nocomis micropogon

No tr op is atherinoides

Notropis blennius

Notropis buchanani

Notropis spilopterus

Notropis whipplei

Opsopoedus emiliae*

Carpiodes carpio

Carpiodes cyprinus

Carpiodes ve lifer

Ictiobus bubalus

Ictiobus cyprinellus

Ictiobus niger

♦Widespread in the Lower Green River portions of the Interior Low Plateau

Minytrema melanops

Ictalurus furcatus

Ictalurus melas

Ictalurus natalis

Ictalurus punctulatus

Noturus gyrinus

Pylodictis olivaris

Gambusia affinis

Labidesthes sicculus

Aphredoderus say anus

Pomoxis annularis

Pomoxis nigromaculatus

Lepomis gulosus

Lepomis humilis

Stizostedion canadense

Stizostedion vitreum

Etheostoma asprigene*

Etheostoma caeruleum

Cottus carolinae

Aplodinotus grunniens

Table 3. A partial list of fishes from Reelfoot Lake, Tennessee (from Lee et al.

1980, and Parker 1939). Asterisk (*) denotes absence from Low Pla-

teau streams.

Lepisosteus platostomus

Amia calva

Alosa alabamae

Hiodon tergisus

Hybognathus nuchalis

Opsopoedus emiliae

Rhinichthys atratulus

Erimyzon oblongus

Noturus gyrinus

Umbra limi*

Fundulus chrysotus*

Fundulus notti*

Fundulus olivaceus

Menidia beryllina (audens)*

Elassoma zonatum

Centrarchus macropterus

Lepomis humilis

Lepomis symmetricus*

Etheostoma fusiformis*

Etheostoma gracilis

Etheostoma proeliare

1 80 Branley A. Branson

ferent subgenera of darters — four if we accept Nanostoma (Page and

Burr) — (Table 6), perhaps resulting from various impulses or cycles of

invasion from extra-limital stream basins, have markedly diversified

within these river systems, doubtless abetted by the notable niche and

habitat variability from stream to stream and within drainages.

Actually, the Low Plateau endemic species of the percid subgenera

Nothonotus, Catonotus and Ulocentra (including Nanostoma) and their

nearly 30 species, pose a biogeographic and evolutionary problem of

considerable importance that has been inadequately studied. These three

groups, in my estimation, are species swarms that have developed in

response to mechanisms similar to those proposed by Pflieger (1971) to

account for stepwise fish dispersal through the Mississippi and Ohio

rivers: "Aggradation subsequent to the last glacial stage produced the

environmental conditions now prevailing in the Embayment, restricting

further dispersal by upland fishes. All the glacial and interglacial peri-

ods were accompanied by alternate entrenchment and aggradation in

the Mississippi Embayment, and this would seem to provide an ade1

quate mechanism for the alternate dispersal and isolation of populations

east and west of the Embayment."

Thus, this mechanism may have been accompanied by cycles of

isolation, adaptation, and speciation, aided by extralimital stream cap-

tures that brought congeners back into contact to heighten competition

and perhaps establish new patterns of variation and divergence. What-

ever the mechanisms and processes, the species swarms are real and the

whole problem is deserving of detailed analysis.

Another group of fishes of considerable interest and importance is

the ictalurid genus Noturus. The fact that most species of Noturus avoid

cold water (Taylor 1969) indicates a southern origin for the group. The

center of greatest abundance of species encompasses Kentucky and

Tennessee to Virginia and North Carolina, with a derivative secondary

speciation center in the Ozarks of Arkansas and Missouri. Tennessee

has the largest number of species, mostly associated with the Tennessee

River basin, Kentucky is second followed by Alabama, and the number

decreases peripherally. Noturus gyrinus (Mitchill), a distinctive lowland

species that is widespread and common in the Lower Green and Trade-

water systems, gets into Low Plateau streams in western Kentucky and

adjacent Indiana and Illinois (Wabash drainage). Noturus exilis Nelson

is absent from southern Indiana and most of Kentucky, but occurs in

extreme southwestern Illinois, much of the Low Plateau of central Ten-

nessee and northern Alabama, and peripheral areas to the north (post-

glacial) and trans-Mississippian in Oklahoma, Kansas, Missouri, and

Arkansas. In the Low Plateau, Noturus nocturnus Jordan and Gilbert

occurs in the Tennessee River drainage of western Tennessee, northern

Interior Low Plateau Distributional Patterns

181

Table 4. Wabash River fishes in Indiana derived from Mississippi Embayment

and Low Plateau sources (from Gerking 1945, and Blatchley 1938).

Polyodon spathula

Lepisosteus oculatus

Lepisosteus spatula

Alosa chrysochloris

Hiodon alosoides

Clinostomus elongatus

Hybopsis aestivalis

Hybopsis storeriana

Hybognathus hayi

Hybognathus nuchalis

Notropis buchanani

Carpiodes carpio

Carpiodes velifer

Cycleptus elongatus

Moxostoma carinatum

Moxostoma macrolepidotum

Noturus nocturnus

Gambusia af finis

M or one mississippiensis

Centrarchus macropterus

Lepomis humilis

Ammocrypta clara

Etheostoma asprigene

Etheostoma chlorosomum

Etheostoma gracile

Etheostoma histrio

Etheostoma squamiceps

Etheostoma variatum

Percina copelandi

Percina evides

Percina ouachitae

Percina sciera

Percina shumardi

Table 5. Fishes of the Low Plateau derived from lowland and lower Missis-

sippi sources (from Etnier, unpublished).

Notropis buchanani

Notropis maculatusx

Notropis shumardi

Notropis venustus

Opsopoedus emiliae

Phenacobius mirabilis2

Erimyzon succetta

Moxostoma poecilurum1

Noturus phaeus1

Noturus stigmosus

Etheostoma asprigene

'Not in Low Plateau

2Plains origin

3Trans-Mississippian origin

Etheostoma chlorosomum

Etheostoma gracile

Etheostoma histrio

Etheostoma parvipinne

Etheostoma spectabile3

Etheostoma swaini

Percina evides2

Percina phoxocephala

Percina sciera

Percina shumardi

182 Branley A. Branson

Alabama, and Western Kentucky, from whence it has been able to

extend its range up the Ohio into the Wabash River drainage of Illinois

and Indiana, the Green River of Kentucky, and old Teays tributaries

(Kentucky and Big Sandy rivers) in eastern Kentucky. Noturus phaeus

Taylor and N. hildebrandi (Bailey and Taylor) have ranges that only

impinge upon the Low Plateau in Mississippi River drainages of north-

ern Alabama and western Tennessee, N. phaeus barely getting into

southwestern Kentucky (Terrapin Creek) (Taylor 1979). Noturus flavus,

being more tolerant of cold water, is distributed throughout much of the

upper two-thirds of the Mississippi drainage, including most of the Low

Plateau. Noturus elegans Taylor, autochthonous to the Barren-Green

system of Kentucky and adjacent Tennessee, has an apparently disjunct

population in the Tennessee River basin (Duck River). In the Low Pla-

teau, N. eleutherus occurs in the Tennessee (no published records from

Kentucky stretches of that stream), the Green River, and the Wabash

system of Illinois and Indiana. Noturus stigmosus, a member of the

furiosus species complex, avoids most of the Low Plateau but has pene-

trated into the Green and Salt river drainages of Kentucky and into the

Wabash River of Illinois-Indiana. Noturus miurus has the widest distri-

bution of all Low Plateau madtoms, having been reported from all

drainages.

It seems likely that the ancestral stock of Noturus arose somewhere

in the lowlands of the Mississippi River basin from a bullhead-like

ancestor (Taylor 1969), and spread from there into other parts of Amer-

ica. Using the Tennessee River basin as a speciation center, additional

species diverged and spread widely throughout the system in pre- and

postglacial times, particularly after development of the Upper Ohio sys-

tem. Such conclusions are supported by the paucity of Noturus species

in the Eastern Seaboard drainages.

The occurrence of Moxostoma {Thoburnia) atrippine Bailey — a

close relative of the torrent suckers of Virginia — in the Barren River

system of the Low Plateau either represents a relict or a case of immi-

gration, extinction in intervening areas, and survival and divergence.

The wide hiatus in ranges between population centers in the subgenus

Thoburnia suggest the latter.

CONCLUSIONS

The examples of Interior Low Plateau aquatic biota discussed here

merely represent the complexity of the mechanics and biologistics of

understanding such a fauna and flora. I have not discussed other taxa,

such as the swarms of unionid pelecypods and decapod crustaceans,

because that would have lengthened the paper considerably. However,

the distributional patterns of these interesting organisms also very

graphically reflect similar conclusions.

Interior Low Plateau Distributional Patterns 183

Table 6. Endemic fish species of various streams of the Interior Low Plateau and vicinity.

1 84 Branley A. Branson

The Interior Low Plateau is a unique province in many ways. The

region has been periodically convulsed by tectonic readjustments of the

crust that stimulated some reorganization of drainages. It was also

influenced by the formation and maintenance of the Mississippi Embay-

ment and the isthmus of stable land at the northwest corner of the Low

Plateau, and /or by eustatic adjustments of the Embayment, and by

stream evolution and modification in extralimital areas, mostly taking

the form of headwater piracies between various streams. Pleistocene

glaciation caused enormous changes by obliterating much of the old

Teays River system and other streams in the unglaciated area, and by

successively turning streams westward to form the Upper Ohio River,

connecting streams that had never before been in direct contact. This,

and the placement of the Mississippi Embayment, conspired to form

one of the most unusual hydrologic phenomena in North America. More

large rivers find confluence near the junction of Illinois, Indiana, Mis-

souri, and Kentucky than at any other area of equal size in North Amer-

ica: the Ohio, Mississippi, Tennessee, Cumberland, Green, and Wabash

rivers. Some of these streams run in nearly ancestral basins, but others

do not. The biological function of all these changes was, of course, to

open up faunal exchange pathways that had not previously been avail-

able. The influence of the Mississippi-Ohio connection upon the repopu-

lation of Indiana and Illinois segments of the Wabash drainage has been

very great. During meltback, many small streams were doubtless exter-

minated and larger ones modified by outwash and deposition, judging

from the thickness of known deposits in Indiana and adjacent Kentucky

and the fossils contained therein.

Stream-margin and tributary migration ("hopping"), both upstream

and downstream, is still a viable hypothesis to explain some observed

distributional patterns in small-stream (third order and smaller) species.

The meltback of even enormous glaciers is not a constant phenomenon,

but one that varies according to season and even time of day, creating

impulses of high- and slack-water conditions. Where warmer water

flows into glacial streams, the warm water does not mix immediately

with the cold. Instead, as I have observed at the Athabascan Ice Field in

Canada and at many smaller valley glaciers in British Columbia,

Washington, and Oregon, the warmer water flows along the margins of

the current. Thus, in the case of Interior Low Plateau drainages, various

species of fishes could have found their way into tributaries where they

had not previously occurred.

There were, of course, fishes already present in Low Plateau

streams prior to the onset of these influences; some of those remain as

relicts, either as endemics or as segments of the fauna that are identical

with or closely related to segments elsewhere with varying hiatuses

intervening. Newly arrived forms from elsewhere may have, because of

Interior Low Plateau Distributional Patterns 185

narrow habitat or niche requirements, remained close to the points of

entry (as in the case of some of the headwater species), or they may have

spread inexorably through systems, progressively coming into compe-

tition with members of the preexisting fauna. Such competition may

have caused some minor extinctions here and there by way of niche

replacement, but it probably stimulated subdivision of niches and,

apparently, considerable sequences of divergence, speciation, and other

types of faunal readjustments. Some speciation may have been stimu-

lated by the presence of unoccupied niches, by the habitat fluctuations

associated with long-term geologic influences, and by differences between

various segments of Low Plateau drainages.

All these influences have acted in concert to create a superior aquat-

ic fauna in the streams and rivers of the Interior Low Plateau, but one

which has remained relatively poorly understood until recent times, par-

ticularly in its relationships to extralimital drainages. Now a new factor

has been superimposed upon natural ones, a factor that not only threat-

ens the integrity of the fauna and flora (Branson 1977), but one that

also threatens our ability to understand this biota. The results of pre-

impoundment studies indicate that some species, present prior to dam-

ming of streams, are now either very rare or unknown from afflicted

areas. An example is Hybognathus nuchalis (pers. comm., David Etnier

and students, following study of University of Michigan holdings).

Every major stream that flows through the Low Plateau bears dams,

sometimes multiple ones. These structures have vastly modified habi-

tats, upset temperature regimens, and diminished nutrient flow through

the systems. Exotic species and transplanted ones, such as striped bass,

trout, and threadfin shad, have gotten into the system, creating stress

that was not earlier present. Previously autonomous drainages are now

interconnected. For example, a navigation canal joins the Cumberland

and Tennessee rivers at Land Between the Lakes; other connections are

planned, such as the Tennessee-Tombigbee Waterway. Coal mining and

channelization continue to inflict adverse changes in many Low Plateau

streams. Other recent influences are steadfastly afflicting this unique

fauna, bringing many species to levels of concern (Branson et al. 1981).

Most are associated with human population pressures, the changing

financial picture, and the energy crisis. Hopefully, we have learned

enough to preserve at least the principal genomes of the Low Plateau

biota.

ACKNOWLEDGMENTS.— I greatly appreciate the helpful and crit-

ical comments of Mr. Melvin L. Warren, Jr., Kentucky Nature Pre-

serves Commission, and the criticism of Dr. David Etnier, University of

Tennessee.

186 Branley A. Branson

LITERATURE CITED

Baker, Clinton L. 1939. Additional fishes of Reelfoot Lake. Rep. Reelfoot Lake

Biol. Stn. Tenn. Acad. Sci. 3:6-40.

Baker, Frank C. 1920. The life of the Pleistocene or glacial period. Univ. 111.

Press, Urbana. 466 pp.

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

tions to the ichthyofauna of Kentucky. Trans. Ky. Acad. Sci. 40:53-55.

Blatchley, William S. 1938. The Fishes of Indiana. Nature Publ. Co., Indianapo-

lis. 121 pp.

Braasch, Marvin E., and L. M. Page. 1979. Systematic studies of darters of the

subgenus Catonotus (Percidae), with the description of the new species

from Caney Fork, Tennessee. Occas. Pap. Mus. Nat. Hist. Univ. Kans. 78.

10 pp.

Branson, Branley A. 1972. Checklist and distribution of Kentucky aquatic gas-

tropods. Ky. Fish. Bull. 54: 1-20.

1977. Threatened fishes of the Daniel Boone National Forest. Trans.

Ky. Acad. Sci. 55:33-35.

, and D. L. Batch. 1974. Fishes of the Red River drainage, eastern

Kentucky. Univ. Press Ky., Lexington. 67 pp.

, and 1981. Distributional records for gastropods and

sphaeriacean clams of the Kentucky and Licking rivers and Tygarts Creek

drainages, Kentucky. Brimleyana 7:137-144.

, and 1982. Molluscan distributional records from the

Cumberland River, Kentucky. Veliger 24:351-354.

, 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-

lippe and P. Cupp. 1981. Endangered, threatened, and rare animals and

plants of Kentucky. Trans. Ky. Acad. Sci. 42:77-89.

, J. B. Sickel and B. M. Bauer. 1983. Notes on rare and endangered or

threatened pleurocerid snails from the Cumberland River, Kentucky.

Nautilus.

Browne, Ruth G., and P. M. Bruder. 1963. Pleistocene Mollusca from the

loesses of Kentucky. Sterkiana 77:53-57.

Burch, John B. 1982. Freshwater snails (Mollusca: Gastropoda) of North Amer-

ica. Environ. Prot. Agency 600/3-82-026. 294 pp.

Burr, Brooks M. 1980. A distributional checklist of the fishes of Kentucky.

Brimleyana 3:53-84.

, and R. L. Mayden. 1979. Records of fishes in western Kentucky

with additions to the known fauna. Trans. Ky. Acad. Sci. 40:58-67.

, and L. M. Page. 1978. The life history of the cypress darter, Ethe-

ostoma proeliare, in Max Creek, Illinois. Biol. Notes 111. Nat. Hist. Surv.

706:1-15.

Clay, William M. 1975. The Fishes of Kentucky. Ky. Dep. Fish. Wildl. Resour.,

Frankfort. 416 pp.

Conant, Roger. 1960. The queen snake, Natrix septemvittata, in the Interior

Highlands of Arkansas and Missouri, with comments upon similar disjunct

distributions. Proc. Acad. Nat. Sci. Phila. 772:25-40.

Interior Low Plateau Distributional Patterns 187

Fenneman, Nevin M. 1938. Physiography of Eastern United States. McGraw-

Hill Book Co., Inc., New York. 714 pp.

Forbes, Stephen A. 1909. On the general and interior distribution of Illinoian

fishes. Bull. 111. State Lab. Nat. Hist. 5:381-437.

Gerking, Shelby D. 1945. The distribution of the fishes of Indiana. Invest. Indi-

ana Lakes Streams 3:1-137.

Goodrich, Calvin. 1921. Anculosae north of the Alabama drainage. Nautilus

55:9-12.

1922. The Anculosae of the Alabama river system. Misc. Publ. Mus.

Zool. Univ. Mich. 7. 68 pp.

1929. The Pleurocerid fauna of the Falls of the Ohio. Nautilus

53:1-17.

1934. Studies of the gastropod family Pleuroceridae — I. Occas.

Pap. Mus. Zool. Univ. Mich. 286. 19 pp.

1935. Studies of the gastropod family Pleuroceridae — V. Occas.

Pap. Mus. Zool. Univ. Mich. 318. 12 pp.

1940. The Pleuroceridae of the Ohio River system. Occas. Pap. Mus.

Zool. Univ. Mich. 417. 21 pp.

1944a. Certain operculates of the Coosa River. Nautilus 55:1-10.

1944b. Pleuroceridae of the Coosa River basin. Nautilus 55:40-48.

Griswold, L. S. 1895. Origin of the Lower Mississippi. Occas. Pap. Boston Soc.

Nat. Hist. 26:474-479.

Gunning, Gerald E., and W. M. Lewis. 1955. The fish population of a spring-fed

swamp in the Mississippi bottoms of southern Illinois. Ecology 3(5:552-558.

Harker, Donald F., Jr., R. R. Hannan, M. L. Warren, Jr., L. R. Phillippe, K.

E. Camburn, R. S. Caldwell, S. M. Call, G. J. Fallo and B. VanNorman.

1980. Western Kentucky coal field: preliminary investigations of natural

features and cultural resources. Introduction and ecological features of the

Western Kentucky coal field, Vol. I (Parts I and II). Ky. Nature Preserves

Comm. Tech. Rep. 1. 296 pp.

Hocutt, Charles H. 1979. Drainage evolution and fish dispersal in the central

Appalachians. Part II. Bull. Geol. Soc. Am. 90:197-234.

, R. F. Denoncourt and J. R. Stauffer, Jr. 1978. Fishes of the Green-

brier River, West Virginia, with drainage history of the central Appala-

chians. J. Biogeogr. 5:59-80.

Howell, W. Mike, and G. Dingerkus. 1978. Etheostoma neopterum, a new per-

cid fish from the Tennessee River system in Alabama and Tennessee. Bull.

Alabama Mus. Nat. Hist. 3:13-26.

Johnson, Richard I. 1980. Zoogeography of North American Unionacea (Mol-

lusca: Bivalvia) north of the maximum Pleistocene glaciation. Bull. Mus.

Comp. Zool. 749:77-189.

Jordan, David S. 1877. A partial synopsis of the fishes of upper Georgia. Ann.

New York Lye. Nat. Hist. 6:307-377.

Kuehne, Robert A. 1966. Depauperate fish faunas of sinking creeks near Mam-

moth Cave, Kentucky. Copeia 1966(2): 306-3 10.

, and R. M. Bailey. 1961. Stream capture and the distribution of the

percid fish, Etheostoma sagitta, with geologic and taxonomic considera-

tions. Copeia 1961(1): 1-8.

1 88 Branley A. Branson

, and J. W. Small, Jr. 1971. Etheostoma barbouri, a new darter (Per-

cidae, Etheostomatini) from the Green River with notes on the subgenus

Catonotus. Copeia 1971(1): 18-26.

Lachner, Ernest A., and R. E. Jenkins. 1971. Systematics, distribution, and evo-

lution of the chub genus Nocomis Girard (Pisces, Cyprinidae) of eastern

United States, with descriptions of new species. Smithson. Contrib. Zool.

55:1-97.

, and 1967. Systematics, distribution, and evolution of

the chub genus Nocomis (Cyprinidae) in the southwestern Ohio River

basin, with the description of a new species. Copeia 1967(3):554-580.

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, xiv + 867 pp.

Miller, Lewis G. 1978. New distributional records for the rosyside dace in Ken-

tucky. Trans. Ky. Acad. Sci. 39:142-144.

McDonough, T. A. 1974. Fish inventory data for Barkley Reservoir 1974. Div.

For. Fish Wildl. TVA 1974:1-19

Page, Lawrence M., and M. E. Braasch. 1976. Systematic studies of darters of

the subgenus Catonotus (Percidae), with the description of a new species

from the Lower Cumberland and Tennessee systems. Occas. Pap. Mus.

Nat. Hist. Univ. Kans. 60. 18 pp.

, and 1977. Systematic studies of darters of the subge-

nus Catonotus (Percidae), with the description of a new species from the

Duck River system. Occas. Pap. Mus. Nat. Hist. Univ. Kans. 63, 18 pp.

, 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. 101. 20 pp.

, B. M. Burr and P. W. Smith. 1976. The spottail darter, Etheostoma

squamiceps (Osteichthys, Percidae) in Indiana. Am. Midi. Nat. 95:478-479.

Parker, Malcolm V. 1939. A note on the food of the shorthead gar at Reelfoot

Lake. Rep. Reelfoot Biol. Stn. Tenn. Acad. Sci. 5:186-187.

Pflieger, William L. 1971. A distributional study of Missouri fishes. Mus. Nat.

Hist. Univ. Kans. Publ. 20:225-570.

Quarterman, Elsie, and R. L. Powell. 1978. Potential ecological/ geological nat-

ural landmarks in the Interior Low Plateaus. U.S. Dep. Interior, Washing-

ton, D. C. 739 pp.

Raney, Edward C, and T. Zorach. 1967. Etheostoma microlepidum, a new per-

cid fish of the subgenus Nothonotus from the Cumberland and Tennessee

river systems. Am. Midi. Nat. 77:93-103.

Resh, Vincent H., C. R. Baker and W. M. Clay. 1972. A preliminary list of the

fishes of the Land Between the Lakes, Cumberland and Tennessee river

drainages, Kentucky. Trans. Ky. Acad. Sci. 53:73-80.

Retzer, Michael E., B. M. Burr and M. L. Warren, Jr. 1983. Fishes of the

Lower Green River, Kentucky. Ky. Nature Preserves Comm. Sci. Tech.

Ser.

Ross, Robert D. 1971. The drainage history of the Tennessee River. Pp. 1 1-42 in

P. C. Holt (Ed.). The distributional history of the biota of the southern

Interior Low Plateau Distributional Patterns 1 89

Appalachians, Part III: Vertebrates. Res. Div. Monogr. 4, Va. Polytech.

Inst. & State Univ., Blacksburg. 306 pp.

Sisk, Morgan E. 1973. Six additions to the known piscine fauna of Kentucky.

Trans. Ky. Acad. Sci. 54:49-56.

Smith, Philip W. 1979. The Fishes of Illinois. Univ. Illinois Press, Urbana. 314 pp.

Smith- Vaniz, William F. 1968. Freshwater Fishes of Alabama. Agric. Exp. Stn.,

Auburn Univ., Auburn. 21 1 pp.

Taylor, William R. 1969. A revision of the catfish genus Noturus Rafmesque,

with an analysis of higher groups in the Ictaluridae. U.S. Natl. Mus. Bull.

2*2:1-315.

TV A. 1975. Tennessee Valley streams, their fish, bottom fauna, and aquatic

habitat. Duck River drainage basin. Tenn. Valley Auth. 1975:1-27.

Wall, Benjamin R., and J. D. Williams. 1974. Etheostoma boschungi, a new

percid fish from the Tennessee River drainage in northern Alabama and

western Tennessee. Tulane Stud. Zool. Bot. 7^:172-182

Wayne, William J. 1952. Pleistocene evolution of the Ohio and Wabash valleys.

J. Geol. 60:575-585.

Wiley, E. O. 1977. The phylogeny and systematics of the Fundulus nottii species

group (Teleosti: Cyprinodontidae). Occas. Pap. Mus. Nat. Hist. Univ.

Kans. 66:1-31.

Williams, James D. 1975. Systematics of the percid fishes of the subgenus

Ammocrypta, genus Ammocrypta, with descriptions of two new species.

Bull. Alabama Mus. Nat. Hist. 7:1-56.

, and D. A. Etnier. 1978. Etheostoma aquali, a new percid fish (sub-

genus Nothonotus) from the Duck and Buffalo rivers, Tennessee. Proc.

Biol. Soc. Wash. 97:463-471.

Zorach, Timothy. 1968. Etheostoma bellum, a new darter of the subgenus

Nothonotus from the Green River system, Kentucky and Tennessee. Copeia

1968(3):474-482.

1972. Systematics of the percid fishes, Etheostoma camurum and E.

chlorobranchium new species, with a discussion of the subgenus Nothono-

tus. Copeia 1972(3):427-447.

Accepted 7 February 1983

190

NEW MANAGING EDITOR

Alexa C. Williams, Managing Editor of Brimleyana since its inception early

in 1979, has resigned her position as Director of Publications and Public Rela-

tions at the museum. She will be Associate Editor of a new international physi-

ology journal edited by Knut Schmidt-Nielsen at Duke University, Durham. We

wish Alexa all the best in this challenging endeavor. Alexa's replacement, and

new Managing Editor of Brimleyana, is Eloise F. Potter. As an author,

reviewer, and longtime Editor of The Chat, quarterly bulletin of the Carolina

Bird Club, Inc., Eloise is conversant with all aspects of journal production and

is a welcome addition to our staff.

SUBSCRIPTIONS AND EXCHANGES

The editors anticipate two issues of approximately 150 pages each annually,

but may not always be able to adhere to this schedule. Subscription rates are for

the two forthcoming issues, unless the subscriber indicates otherwise.

Individual— United States $ 9.00

Individual — Foreign $ 1 3.00

Institution $ 1 5.00

All subscriptions must be paid in advance; the cost includes postage and

handling. Back issues are available.

Brimleyana is available for exchange with institutions publishing natural

history and ecology serials or occasional papers. Publications received by

exchange are placed in the museum's H. H. Brimley Memorial Library.

Address all subscriptions, exchange queries, and requests for information

to: Managing Editor, Brimleyana, N. C. State Museum of Natural History,

P. O. Box 27647, Raleigh, North Carolina 2761 1.

DATE OF MAILING

Brimleyana No. 10 was mailed on 4 March 1985.

ERRATA

Brimleyana No. 10:

Page 3: line 12, insert lewisi after Necturus maculosus.

Page 32: LITERATURE CITED, between Neill 1963 and Shoop and

Gunning 1967 insert:

Nickerson, Max A., and R. E. Ashton, Jr. 1983. Lampreys in

the diet of hellbender Cryptobranchus alleganiensis (Dundin), and the

Neuse River waterdog Necturus lewisi (Brimley). Herpetol. Rev.

/4(1):10.

STATE LIBRARY OF NORTH CAROLINA

3 3091 00748 4942

INFORMATION FOR CONTRIBUTORS

Submit original and two copies of manuscripts to Editor, Bnmleyana, North Carolina State

Museum of Natural History, P. O. Box 27647, Raleigh, NC 27611. 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, 8V2 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.