>' :

HARVARD UNIVERSITY
%

Library of the

Museum of

Comparative Zoology

S-NA-P^

BRIGHAM YOUNG UNIVERSITY
SCIENCE BULLETIN

mUS. COMP. ZOOL
LIBRARY

JAN 2 8 1966

HARVARD
UNIVERSITY

A COMPARATIVE POPULATION STUDY OF SMALL

VERTEBRATES IN THE URANIUM AREAS OF THE

UPPER COLORADO RIVER BASIN OF UTAH

by

Wilmer W. Tanner

Biological Series — Volume VII, Number 1
MAY, 1965

BRIGHAM YOUNG UNIVERSITY
SCIENCE BULLETIN

A COMPARATIVE POPULATION STUDY OF SMALL

VERTEBRATES IN THE URANIUM AREAS OF THE

UPPER COLORADO RIVER BASIN OF UTAH

by

Wilmer W. Tanner

Biological Series — Volume VII, Number 1
MAY, 1965

tviuS. COMP. ZOOL
LIBRARY

JAW

HARVARD
UNIVERSITY

PREFACE

This paper constitutes a terminal report to the United States Atomic Energy
Commission for contract AT(11-1)819. It provides not only a faunistic and floristic
Study of areas having a high and low surface radiation, but also includes a con-
sideration uf their general ecology. Major emphasis is placed on the variation
nl the external characteristics of the small vertebrates of these areas with a
statistical analysis of those characters examined in the lizards.

The study was conducted in southeastern Utah (Tanner 1963, see map)
from October 1959 through 1964, and was conducted as a test of the variation
effects of natural surface radiation on the external characteristics of small
vertebrates.

TABLE OF CONTENTS

INTRODUCTION 1

PROCEDURE AND ACKNOWLEDGEMENT 3

GENERAL ECOLOGICAL BACKGROUND 6

Climate 6

GENERAL DESCRIPTION OF THE AREAS 8

Temple Mountain Mesa 8

Mamie Stover Area 10

Yellow Cat Area 10

Indian Creek Area 10

REVIEW OF THE OBJECTIVES 10

Objective 1 10

Cold-blooded Vertebrates 11

Warm-blooded Vertebrates 11

Plant Species and Vegetation Analysis 11

Objective 2 13

Home Range 15

Longevity 15

Growth 16

Objective 3 17

Objective 4 17

LIZARD SPECIES STUDIES 18

Uta stansburiana 18

Sceloporus undulatus elongatus 20

Cnemidophorus tigri? septentrionalis 23

MAMMALS 25

Ammospermophilus leucurus notium 26

Etttamias quadrivittatns hopiensis 26

Neotoma lepida sanrafaeli 26

Feromijscus maniculatus sonariensis 26

Peromyscus maniculatus nebrascensis 26

Peromyscus crinitus 27

STATISTICAL ANALYSES OF THE DATA 27

CONCLUSIONS 30

LITERATUDE CITED 31

LIST OF ILLUSTRATIONS
Figure Page

1. General view of the Temple Mountain study area, as viewed from north to south 2

2. General view of the Mamie Stover area, showing a small ravine which cuts through the
overlying cap rock 4

3. Dorsal view of hands and feet of a Sceloporus undulatus elongatus showing the num-
bers assigned to the digits 4

4. Six views of the toe marks as recorded on the rocks. All taken at Temple Mountain .... 5

\ diagramatlc sketch oi the Temple Mountain area showing the location of the home
ranges ol sum,- Uta stansburiana. The legend provides the basic information. See first foldout
6 \ diagramatic sketch oi the Temple Mountain area, showing the location of the home
ranges oi Sceloporus undulatus ehngatus. Tlic legend provides the basic information

Sec second foldout

7. Number ol scale rows at midbody in six populations oi Uta stansburiana. The range of
variation is indicated by the horizontal line, the mean l>\ the vertical line, the rectangle
encloses one standard error on each side of the mean and the shorter horizontal line
above the vertical line I mean) represents one standard deviation of the mean. The num-
ber of scales (rows) is indicated both at the bottom and the top of the graph, with the
populations (areas V(>1 | to the left of the range ( long horizontal line) and size of the
population sample on the right margin of the range 19

8. Dorsal scales in Uta stansburiana counting from parietal to base of tail. See Fig. 7 for
explanation of graph 19

9. Ventral scales in Uta stansburiana counting from the enlarged scales just posterior to
the gular fold (this is approximately at the interior square of the front legs) to the
granule scales at the vent. See Fig. 7 for explanation of graph. 19

10. Total number of femoral pores on both femurs in Uta stansburiana. See Fig. 7 for ex-
planation of graph. 20

11. Total number of supralabial scales on both lower lips in Uta stansburiana. See Fig. 7

for explanation of graph 20

12. Total number oi infralabia] scales on both lower lips in Uta Stansburiana. See Fig. 7

for explanation of graph. 20

13. Number of scale rows at midbody in Sceloporus undulatus ehngatus. See Fig. 7 for
explanation of graph 21

14. Dorsal scale rows in Sceloporus undulatus ehngatus, counting from parietal to base of tail.
See Fig. 7 for explanation of graph. 21

15. Ventral scales in Sceloporus undulatus ehngatus counting from the enlarged scales just
posterior to the gular fold (this is approximately at the anterior square of the front
legs) to the granule scales at the vent. See Fig. 7 for explanation of the graph. 21

16. Total number of femoral pores on both femurs in Sceloporus undulatus ehngatus. See
Fig. 7 for explanation of the graph 22

17. Total number of supralabial scales on both upper lips in Sceloporus undulatus elon-
gatus. See Fig. 7 for explanation of graph 22

18. Total number of infralabial scales on both lower lips in Sceloporus undulatus ehnga-
tus. See Fig. 7 for explanation of graph 22

19. Number of scale rows at midbody in six populations of Cnemidophorus tigris septen-
trionalis. See Fig. 7 for explanation of graph 23

20. Dorsal scales in Cnemidophorus tigris septentrionalis, counting from parietal to base of
tail. See Fig. 7 for explanation of graph. 23

21. Ventral scales in Cnemidophorus tigris septentrionalis, counting from the enlarged scales
just posterior to the gular fold (this is approximately at the anterior square of the front
legs) to the granule scales at the vent. See Fig. 7 for explanation of the graph 24

22. Total number of femoral pores on both femurs in Cnemidophorus tigris septentrionalis.
See Fig. 7 lor explanation of graph 24

2-3. Total number of supralabial scales on both upper lips in Cnemidophorus tigris septen-
trionalis. See Fig. 7 for explanation of graph. 25

24. Total number of infralabials scales on both lower lips in Cnemidophorus tigris septen-
trionalis. See Fig. 7 for explanation of graph 25

A COMPARATIVE POPULATION STUDY OF SMALL

VERTEBRATES IN THE URANIUM AREAS OF THE

UPPER COLORADO RIVER BASIN OF UTAH

INTRODUCTION

The discovery of uranium and vanadium
ores in the Colorado Plateau area of Utah and
Colorado near the turn of the century has fin-
ally led to a mining operation far beyond the
dreams of those early prospectors who made
their first ore shipments in 1902 (Sorensen:
282). In this early period of development of
the area the ores were important because of the
radium that could be extracted. Although the
amounts obtained were small and the process
of extraction was slow, tedious, and expensive,
radium was extracted from Utah ores until 1923,
with the greatest production occurring during
the last ten years. During these few years the
Colorado Plateau area of Utah and Colorado
was the leading radium-producing area of the
world. However, the processing of these ores
was marginal and the discovery of high-grade
ores in the Belgian Congo in 1923 made further
operations too expensive.

During the first fifty years of development
the area actually passed through three rather
distinct mineralogical stages which may be rep-
resented as follows: (a) Radium to 1923. A
sidelight of these early explorations is the ship-
ping of carnotite ores from the San Rafael Swell
fossiliferous deposits to Pierre and Marie Curie,
in Paris, France, in 1903 (Boutwell 1905:207).
At this same time the new yellow mineral was
named Carnotite in honor of the French physi-
cist Adolph Carnot. During this period the re-
covery of radium from the uranium ores placed
this area in world prominence and produced
more than 32 grams of radium in 1921. (b) Van-
adium, 1923 to 1944, and (c) uranium, 1944 to
present (Sorensen, 1963). In contrast to the pre-
ceding period, during the present period van-
adium has been reclaimed as well as molyb-
denum even though uranium is now the primary
product. A history of the development of this
mining industry cannot be discussed further;
however, for those interested the following stu-
dies provide some of the historical facts and
cite additional references on this subject: Bout-
well 1905, Sorensen, 1963, Hilpert and Dasch
1964. The latter is particularly valuable because
of the large bibliography included in that study.

The explorations since 1945, however, have
been much more extensive and intensive, thus
providing a far greater knowledge of the size,
quality, depth and extent of ore bodies through-
out the entire plateau producing area of Utah
and the adjoining states of Colorado, New Mex-
ico, Arizona and Wyoming.

By the enormous numbers of claims through-
out the area, the prospecting boom of the early
1950's demonstrated that uranium is widespread
throughout the area. Although most of the es-
tablished claims did not become producing
mines because the ore was mostly of a low
grade, the great extent of the natural ground
level radiation particularly in "hot" areas was
demonstrated.

Before leaving this aspect of the study it
should be pointed out that three of the areas stu-
died have been mined at various times for the
past sixty years and are still producing, although
not to any extent, from the same shallow de-
posits used in the early days. A general descrip-
tion of each area will serve as an orientation
and as an introduction for necessary additional
details. (Also see Hayward et. al. 1958.)

The first field work on this contract was
begun on June 1, 1960, with two graduate as-
sistants and me; thus the present investigations
include four years of field work. It should be
indicated at the outset, however, that most of
the field work has been done and must be done
during the summer months, primarily because
we are dealing mostly with the cold-blooded
vertebrates which are not available in these
areas for more than about six months of the
year.

The role of isolation in this inland desert
basin must have played its part in providing an
opportunity for variation to occur. The fact that
so much variation does occur in most species of
this area suggests that perhaps other factors
may be operating. Few species occurring in the
Upper Colorado Basin are of the same sub-
species as those inhabiting the basins to the
west (Great Basin), to the southeast (Upper
Rio Grande) or to the south (Little Colorado).
In fact, some variation has been observed be-

1

BfUGHAM YOUNC I NIVERSm Si l) \. I Hi I I I i l\

populations occupying opposite sides ol
iorado Hiver in Utah.

Speciation, but more particularly subspecia-
tion, has occurred regularly in the several val-
leys extending northward from the southern
deserts. Such areas as Death Valley and thos<
basins extending north, Lowei Colorado Basin
and the Virgin Valley in southwestern i tah,
adjoining Arizona and Nevada and S. California
and the l pper Colorado Basin are examples.
One example of subspeciation found in this
area is Arizona elegans with Candida in the
Death Vallej .u<a. eburnata in lower Colorado
Basin and the Virgin Valley and philipi in the
Upper Colorado Basin. The latter Basin is par-
ticular!) impressive in regard to the many sub-
present. Since several have been describ-
ed and others discovered or recognized only re-
cently, the Upper Colorado Basin lias been of
special interest to us, leading directly to the
request for research funds to further investigate
certain possible causes (Fig. 1).

It was hypothesized, therefore, that perhaps
the numerous small areas of high surface radia-
tion may be playing their parts by inducing var-
iation into the populations of small vertebrates

inhabiting these areas Thus it was proposed
thai several areas in the • ppei ( olorado
with a higher-than-average surface radiation be
studied and thai local populations of small ver-
tebrates I lizards, and it possible the small mam-
mals) be compared for externa] morphological
variations. It was further deemed necessary to
compare populations occurring in the .ocas ot

high surface radiation (25 to 100 mr./hr.)

with populations in the nearby or adjacent areas
with a lower surface radiation | less than 20).

These comparisons consist of a series ot char-
acter comparisons including external morphol-
ogy, color and color pattern variations. In the
lizards, scale count variations are most impor-
tant, whereas in the mammals color and color
pattern are used. An examination of the internal
gross anatomy or histological comparisons is not
included. Although physiological (including ser-
ological ) and genetical ( chromosomal cytology )
studies may have been fruitful, the project was
not designed to include internal cell or tissue
variations.

The question of what is an unusual or note-
worthy variation and when are such variations
significant in small populations has been a ma-

l-'n. l. General view ol the Temple Mountain studs area as viewed From north to south.

Biological Series, Vol. 7, No. 1. May, 1965

jor challenge throughout the course of this in-
vestigation. At the outset we postulated that
there would be some variation and that these
might take the form of, (a) an increase in the
range of variation of a character within a single
population or, (b) a skewing of the mean to
the right or left of the means of other popula-
tions.

We realized at the outset that there was no
certain way of determining the amount of radi-
ation received by any one animal during a sea-
son or its life, and were therefore obligated to
determine the general radiation for the area.
From this we hoped to determine the average
dosage for the animals included in the areas
studied. Therefore, measured radiation listed in
Table 12 represents the approximate dosage re-
ceived per hour. Because there is uneven radi-
ation, with some areas serving as pockets of high
radiation and others of low, only the approxi-
mate or general area average is considered.

Although there have been many and varied
studies carried forward to determine the effects
of induced radiation on the inherited morpho-
logical characteristics of vertebrates, little is
actually known about the effects of natural rad-
iation on populations. A study by Daniel and
Blair (1959) has indicated an increase of soil

radiation in some areas of the San Luis Valley
of southern Colorado. Ecological studies now in
progress at Arco, Idaho, and Dugwav, Utah, are
primarily concerned with controlled induced
radiation on native animals. Those ecological
studies in the Atoll areas of the Pacific, at the
Nevada test site, and other natural and labora-
tory areas represent environments where nuc-
lear reactions have introduced radiation. In
neither the ecological nor the many laboratory
and field-controlled studies have the experiment-
ers dealt with the major problems considered
in this study.

During the first year four areas were select-
ed, each of which is associated with an ore
body and with an increase in surface radiation
over adjoining areas. These were designated by
the following names and will appear in this
report as named here: Temple Mountain Mesa,
Mamie Stover Incline, Yellow Cat area and In-
dian Creek area. The first two are located 45
and 15 miles respectively southwest and west of
Green River, Emery Countv, Utah. The Yellow
Cat area is approximately 15 miles southeast of
Thompson's, Grand Countv; and Indian Creek is
about 35 miles southwest of La Sal Junction,
San Juan County, Utah. In the latter area only
the reptiles are included in the variation studies
(Fig. 2).

PROCEDURE AND ACKNOWLEDGEMENT

Several techniques were used in gathering
the field data and to collect the animal samples.
The mammals were caught in live traps, frozen
in a field dry ice cabinet and returned to the
University for mounting. Standard weights and
measurements were taken, and the skins and
skulls were prepared as is usual in museum col-
lections. Previous studies (Brant 1962, Burt 1940,
Calhoun and Casby 1958, Mohr and Stumpf
1964, and Tonaka and Teramura 1958), seem-
ingly provide adequate home-range data to in-
dicate that the species inhabiting these areas
did not need additional home-range study. How-
ever, in the lizards I could not be sure; and
since the areas under consideration are relativ-
ely small, it became mandatory that the extent
of annual movement, at least, be determined for
Sceloporus undulatus elongatus, I 'la standsbur-
iana stansburiana and Cnemidophorus tigris scp-
tentrionalis. These species were common to all
areas studied. These were intensively studied
only at Temple Mountain, the other areas not

being included in the home-range studies. In
each species the toe clip method was employed.
An illustration of how the toes were counted
is seen in Figure 3, After the first twenty lizards
(Ufa) were marked, two toes were clipped,
such as 1 and 12 and the next series 2 and 15.
Each lizard caught was marked, measured and
the sex determined. Care was taken to release
each lizard in the exact area where it was caught
so that established home ranges would be less
likely upset.

Lizards were caught by hand rather than by
traps. This proved successful particularly with
Uta and Sceloporus but was difficult for Cnem-
idophorus. In order to determine lizards and the
exact spot taken the toe marks were recorded
in the field book and at the same time painted
on a rock. Figure 4. Thus a toe clip of five on the
front foot and of 17 on the hind foot was record-
ed as 5-17. The painted marks on the rocks
proved relatively permanent, thus making the
measurement of distances between points of re-

Huh. ham ^01 s(. UmvEBSm St IKSCE Blixetin

a, showing a small ravine which cuts through the overlying

capture easy. Second captures were recorded
in the same way but with a line underscoring
the figures. Vta were recorded as indicated;
Sceloporus were recorded with an X following

I tc I Dorsal view <>! hands and feet of a Scelo-
porta unduiattu elongattu showing the numbers assigned
to the digits.

the figure (4-11X): and for Cnemidophorus the
X was circled, 4-1 1®

The methods of home-range study used in
this project were in part dictated by the rough,
rocky terrain but also from the fact that the pit-
fall trap method was being used for similar
studies by Jorgensen and Tanner (1963) at the
Nevada Test Site. Although the surface funnel
traps have been used successfullv bv Fitch (1951
and I960) in Kansas, the near absence of shade
to protect captive lizards from the intense heat
argued against their use.

In all other areas lizards were collected by
using dust shot in .22 caliber pistols. This meth-
od proved highly successful in providing a rel-
atively large series of specimens from the various
.mas However, in most areas almost the entire
population was taken in order to provide the
needed series. At Temple Mountain because of
the home-range study only a few lizards were
taken until the summer of 1963. By shooting, a
lew specimens were damaged for one or two of
the standard counts made, as reflected in the
variable numbers used in the variation graphs,
but the increase in specimens made available by

Biological Series, Vol. 7, No. 1, May, 1965

Fig. 4. Six views of the toe marks as recorded on rocks. All taken at Temple Mountain.

BHIGHAM Vol m. I m\ I Msll 1 Si II N. I Hi I I h I IN

the method more than made up foi the few

IllSSfS.

Iii ordei tu better evaluate the total en\ iron
incut a careful analysis ul the kinds ol pi. mis
percentage ol species in three areas, percentage
of plant covei and soil depths were recorded.
Hen .is with the lizards the plant species
were Inst determined al remple Mountain and
tin i parisons were made to determine vari-
ations between the areas. The plant canopy
cover was determined liv following a transecl
and placing a painted pint frame over the plants
along even meter ol the transect. While meas-
uring the plant canopy, the soil depths were de-
termined by using a steel penetrometer at regu
lar intervals along the transect.

Various instruments have been used to de-
tect an increase in ground level radioactivity. In

tins Study, measurements ol radioactivity were

made in the field with a Tracerlab gamma scin-
tillation detector, employing a 2" sodium iodide
crystal, The equipment was powered by a port-
able 110-volt generator with a voltage regulator.
The minimum counting time for each station was
ten minutes. The equipment was standardized
against known background rates at Provo, Utah,
Inline and alter use in the field. Supplemental
measurements were made with a portable beta-
gamma survey meter (geiger).

\s previously indicated, some mining opera-
tions, either active or abandoned, occur in the

areas. In our radiation measurements, these were
avoided. Thus our data | Table 1- i in< hides only

those leadings taken m the natuial undisturbed
habitat ol ea< h area studied.

I am indebted to the following foi then as-
sistance and for man) liclplul suggestions dur-
ing the ionise ol the project. These Graduate

students have been responsible tor much ol the

field woik: \V Gerald Robinson, Robert E. Bul-
lock, Lynn Findlay, Elden Willis, fames Hop-
kin, Clyde Pritchett, and John Childs. Their
devotion to the project, often under adverse des
ert conditions, is greatly appreciated.

The plants were collected by Veil Mlman
and Clyde Pritchett. Mr. Allman did the clas-
sification, with final designations of a few
species being made by Dr. Stanley Welsh of the
Brigham Young University, Botan) Department.
The mammals were examined b) Dr. Stephen D.
Durrant. To him I am grateful tor helpful sug-
gestions concerning the variations occurring be-
tween populations in these species.

I am grateful also to Dr. A. Lester Allen of
the Brigham Young University, Department ol
Zoology and Entomology for the use ol equip-
ment and for the radiation measurements.
Thanks is also extended to the field officials ol
the Union Carbide Company, who permitted
us complete freedom ol movement while at
Temple Mountain.

GENERAL ECOLOGICAL BACKGROUND

Climate

The general climatic conditions for those
parts ol the Upper Colorado Basin included in
this studs are similar to those ot the Great Basin
of western Utah, which are at about 4,500 to

5,500 feet in elevation.

There are four weather stations in the gen-
eral area i ll.mksville. Green River, Thompson's,

and Moab). These, except lor Thompson's, are
at lower elevations than the Study areas and are
therefore noi completely representative ol the

climate. Although we have1 not been able to

carry forward a year-round check of climatic
conditions at the stmK areas, some summer
maximums and minimums were taken at Temple
Mountain, Both temperatures and humidity were

obtained from an Abheon indicator placed in a
juniper tree about three and one hall leel above

the ground surface and approximately at the

middle ol the ana

\s is suggested b\ the low relative humidity
of desert areas, there is only sporadic precipita-
tion. Actually the climate is relatively moderate
but with seasonal extremes. Temperatures range
to above LOOT, in summer to 0 F. and below
in winter; summer rains may be of the cloud-
burst variety, providing lor heavy flooding, and
heavy wind storms occasionally produce intense
dust storms. Based on 30 or more years ol Wea-
ther Bureau records lor stations at Lmcrv and
Green River, Emery County; Hanksville, Wayne
County; Moab and Thompson's; Grand County,
Utah, the averages listed in Table 1 have been
summarized,

Mr. E. Arlo Richardsen, State Climatologisl
at the Weather Bureau \irport Station. Salt Lake
City, Utah, was kind enough to provide similar
averages lor the studv areas. These he calculated
bv using the data available troin the nearest
weather station and used the differences in tie
vation, slope and sun exposure to provide an es

Biological Series, Vol. 7, No. 1, May, 1965

Table 1. Temperature and precipitation averages for five weather stations in Emery and Grand Counties,
Utah for 30 or more years.

Jan. Feb. Mar. April May June July Aug. Sept. Oct. Nov. Dec. Annual

Emery

Dailv Maximum

36.0

40.5

49.2

59.5

67.9

76.8

83.3

Daily Minimum

11.7

16.4

23.8

31.5

39.2

46.3

53.5

Monthly Mean

23.9

28.5

36.5

45.5

53.6

61.5

68.4

Mean Precipitation

0.51

0.43

0.52

0.36

0.61

0.56

0.77

Green River

Daily Maximum

38.0

48.1

59.6

71.1

81.5

91.7

98.9

Daily Minimum

9.3

18.1

26.1

36.0

44.6

52.1

60.1

Monthly Mean

23.7

33.1

42.9

53.6

63.1

71.9

79.5

Mean Precipitation

0.36

0.39

0.40

0.43

0.38

0.44

0.51

Hanks ville

Daily Maximum

38.8

48.5

60.4

71.1

81.0

91.5

96.7

Daily Minimum

8.2

17.9

27.1

34.7

42.8

51.4

58.9

Monthly Mean

23.5

33.3

43.7

53.1

61.9

71.5

77.9

Mean Precipitation

0.37

0.30

0.32

0.29

0.35

0.39

0.71

Moab

Daily Maximum

42.3

50.2

61.2

72.8

81.9

91.6

98.4

Daily Minimum

17.3

23.4

31.2

40.5

57.6

55.4

62.3

Monthly Mean

29.8

36.8

46.2

56.7

64.8

73.5

80.4

Mean Precipitation

Thompson's

Daily Maximum

36.6

44.7

55.8

66.9

77.5

87.8

94.4

Daily Minimum

12.9

21.5

29.2

36.9

46.2

54.4

62.2

Monthly Mean

24.8

33.0

42.6

51.9

62.0

71.1

78.5

Mean Precipitation

0.59

0.56

0.73

0.67

0.51

0.51

0.85

80.6 73.9 62.7 48.4 37.9 59.7

51.5 43.8 34.2 22.6 15.1 32.5

66.1 58.9 48.4 35.5 26.5 46.1

1.27 0.78 0.73 0.30 0.53 7.41

95.9 87.5 73.0 53.8 42.2 70.1

58.0 47.7 35.2 20.8 14.5 35.2

77.0 67.6 54.1 37.3 28.4 52.7

0.82 0.52 0.70 0.38 0.45 5.78

93.6 85.0 72.4 57.5 44.0 70.0

55.5 46.5 33.8 20.7 12.7 34.2

74.6 65.7 53.1 39.1 28.4 52.2
0.77 0.54 0.68 0.27 0.36 5.35

95.2 86.9 74.3 56.2 44.8 71.3

59.8 51.1 39.1 26.5 19.9 39.5

77.5 69.0 56.7 41.4 32.4 55.4

9.00

91.6 82.4 70.0 54.7 40.9 66.9

60.4 50.4 38.6 25.8 17.2 38.0

76.1 66.4 54.2 40.3 29.2 52.5

0.90 0.97 1.10 0.49 0.72 8.60

timation. These calculated averages are listed
in Table 2.

An examination of Tables 1 and 2 indicates
that perhaps the most important factor contrib-
uting to the differences in temperature, not only
at the weather stations but also at the study
areas, is the difference in elevation. However,
these elevation differences in temperature do
not appear when comparing all stations; for ex-

ample, Hanksville weather station at 4450 ft.
contrasted with Thompson's station at 5150 ft.
varies less than one degree for the annual temp-
erature. Other factors, air drainage, seasonal pre-
cipitation, slope, etc. must influence temperature
as well as elevation.

Precipitation is sporadic and basically rep-
resents a continental and dry climate. The area
as a whole has from 6-10 inches of precipitation

Utah.

Table 2. Temperature averages and precipitation totals for three study areas in Emery and Grand Counties,

Jan. Feb. Mar. April May June July Aug. Sept. Oct. Nov. Dec. Annual

San Rafael (Mamie Stover)

Daily Maximum

40.0

47.0

Daily Minimum

9.4

16.6

Monthly Mean

24.7

31.8

Mean Precipitation

Temple Mountain

Daily Maximum

37.2

41.5

Daily Minimum

12.8

17.3

Monthly Mean

25.0

29.4

Mean Precipitation

Yellow Jacket Wash

Daily Maximum

38.7

44.7

Daily Minimum

15.1

21.3

Monthly Mean

26.9

33.0

Mean Precipitation

57.8 68.7 78.5 88.6 93.9 91.9
24.4 32.7 40.3 48.6 56.3 53.9
41.1 50.7 59.4 68.6 75.1 72.9

50.5 60.9 69.8 79.0 85.6 83.0
25.1 32.9 41.0 48.6 55.8 53.8
37.8 46.9 55.4 63.8 70.7 68.4

56.5 67.4 77.3 88.2 94.7 91.7
30.1 37.4 46.3 54.8 62.9 60.7
42.0 52.4 61.8 71.5 78.8 76.2

83.7

71.3

55.3

43.9

68.3

45.1

32.7

18.5

12.7

32.7

64.4

52.0

36.9

28.3

50.5
6.00

76.3

64.3

49.4

39.7

61.4

46.1

35.9

23.6

16.9

34.1

61.2

50.1

36.5

28.3

47.8
8.50

84.2

71.3

54.0

42.1

67.4

52.2

39.7

25.2

18.7

38.6

68.2

55.5

39.6

30.4

53.0
7.00

Bricham V(/i si. JmvERsm St iknce Bulletin

during tin- year. Thunder storms during July
.mil August provide more rainfall than Foi an)
othei period. Both Moab and Thompson's re
port more precipitation than the station al < Ireen
River and those stations to the west. This may
reflect the Colorado River air drainage or the

influence oi die Rock) Mountains to die east
rado.
In summary, the climatic conditions in the
general area where this stud) was made are
surprisingl) similar, particularly if the annual
averages an- considered as the basic criteria.

GENERAL DESCRIPTION OF THE AREAS

After having carefully examined the selected
areas, I became apprehensive of the role played
by plant cover, soil types and soil depths. Each
of these may be a determining ecologieal factor
in any natural system where conditions might
produce an increase of variation in a gene pool.
It is not seemingly possible to rule out the prob-
able effects which may be produced by such
environmental factors in disjunct but similar
areas. Therefore, in order to establish the extent
of physical and biological factors within the
areas, we determined in a general way the plant
species in each area and their approximate per-
centage of surface cover. We also determined
the soil depth and general composition. Although
there are some slight variations, both in kinds of
plant species found and the depth of the soil
in antic, these may represent fluctuations which
would be expected to occur as a result of random
sampling.

Temple Mountain Mesa. The Temple Moun-
tain area consists of a small ridge extending for
approximately one thousand yards north and
south and varying in width from approximate!)
one hundred yards to slightly over two hundred
yards near its middle. This formation is adjacent
to and just west of the eastern uplift of the sand-
stone of the San Rafael Reef. The western edge
of the Temple Mountain area consists of a ledge
of white limestone-conglomerate cap rock clip-
ping to the east and forming a natural barrier
on the western edge. The narrow northern
boundary lies along a fault, with the area to
the north uplifted and sloping up to the base
of Temple Mountain.

This study site lies at an elevation of about
5,500 feet. It, as well as the Mamie Stover area,
lies along the eastern edge of the San Rafael
Swell. Both fall within the influence of a desert
climate similar to that of the Great Basin Desert

■that is. they are both characterized by low
rainfall and atmospheric humidity, hot sunny
days, a great daily range of temperature, inter
mittent streams, high salt content of the soil
and other desert environmental features. Im-

mediately to the east i one mile from study area)
is the San Rafael Desert. Leaching of desert soils
proceeds very slowly because of the low rainfall:
hence, surviving species must be able to tolerate
the highly alkaline soils usually present in such
anas The two plant communities most wide-
spread in the Great Basin Desert are those in
which sagebrush Artemisia tridentata Nutt. and
shadscale Atriplex confertifolia (Torr. and
Frem.) S. Watts, are the dominant plants Atri-
plex is found in the more alkaline soils.

The area slopes from north to south, with a
low hill on the western edge and the high reef
to the east, giving the hillsides of the study
area essentially a southeastern exposure. A fairly
large wash drains the area from north to south,
with smaller washes draining into it from the
slopes to tlu- west and north. No measurements
of the precipitation were taken within the study
area, but records at the city of Green River
nearby show an average precipitation of 5.78
inches falling throughout the year but with the
greatest amounts during the months of July,
August, and September. Precipitation, however,
is very erratic in such environments and signifi-
cant variations occur from year to year.

Besides the dominant Atriplex confertifolia
the hillsides are covered with a great variety ol
herbaceous and shrubby plants. Many annual
and perennial grasses are found in scattered
clumps with Oryzojjsis hymenoides (Roem. and
Schult. ) being more abundant but with ephem-
erals abundant on the slopes in the springtime.
The latter germinate, grow rapidly and mature
in the span of a few short weeks. These include
various species such as I.cpidimii densiflorum
Schred., l.cpidium montanum Nutt . Streptan-
thus longirostiris S. Watts., Euphorbia fendleri
Torr. and Gray, Oenothera eaespitosa Nutt.,
Oenothera mutiifuga Watson. Oenothera walk-
eri (A. Nelson). Cilia gracilis (Dougl. ) Hook,
Descurainia sophis (L.) Webb, Lappula redow-
ska (Hornem.) Greene, Plantago purshU Roam,
and Shult, Lygodesmia exigua (Nutt.) Greene,
and others. (See plant list.)

Biolocical Series, Vol. 7, No. 1, May, 1965

In addition to the dominant Atriplex there
are a number of other shrubs. Some of these
are Ephedra viridis Coville, Cercocarpus intri-
catus S. Wats., Cowania stansbariana Torr.,
Amelanchier utahensis Gatt. Pomac, Rhus tri-
lobata Nutt. Torr. and Gray, Rhus utahensis
Godding, Artemisia tridentata Nutt., Chryso-
thamnus nauseosus (Pall.), C. viscidiflorus
(Hook) Nutt., Tetradijmia glahrata A. Gray, and
an occasional Opuntia rhodantha Schumann,
Gesamtb.

Many species of plants typical of areas of
higher rainfall are found adjacent to desert
washes. Found chiefly near the wash in the
study area are occasional small trees of the Pinon-
Juniper Association which is so typical of the in-
termountain region, Pinus edulis Engelm, and
Juniperus utahensis (Engelm). Confined mainly
to the wash are other species such as Amelan-
chier utahensis Koehne, Gatt. Pomac, Fraxinus
anomala Torr. S. Wats., Rhus trilobata Nutt.
Torr. and Gray, Rhus utahensis Godding, Astra-
galus spp., Cowania stansburiana Torr., Salsola
kali L., Cercocarpus intricatus S. Wats, and Sym-
phoricarpos oreophilus A. Gray.

Significant among the plants of the area are
those that serve as indicators. R. F. Nelson
(1957) reports that "in every case investigated,
Beath, Eppson and Gilbert found Stanleya pin-
nata growing on seleniferous soils in 11 western
states." The use of Stanleya pinnata as an indi-

cator of selenium is confirmed by many writers
since it is found to occur almost exclusively on
selenium bearing soils. This plant is common
throughout the area.

Even more significant to this study is the
work of Helen Cannon (1952). She found that
uranium ores of the Colorado Plateau contained
not only uranium but also considerable quanti-
ties of vanadium and selenium, and that these
elements were absorbed in unusual amounts by
the roots of plants growing in these areas. She
states: "A much higher content of uranium has
been reported by the Canada Department of
Mines ... in plants growing in the vicinity of
pitchblend deposits. Plants growing on the as-
phaltic sandstone deposits at Temple Mountain,
Utah, and on limestone deposits at Grants, New
Mexico, also absorb uranium readily. The max-
imum values are considerably higher than those
found in plants associated with carnotite de-
posits of the Colorado Plateau." She gives (Table
3) the following analysis of the Asphaltic sand-
stone, Shinarump conglomerate, and organic
materials of the Temple Mountain area and for
the carnotite ore of Yellow Cat study site.

Thus we have a variable in the amounts of
minerals taken up by plants. When one considers
the availability of uranium in plants which are
used as food either directly or indirectly (eaten
directly as in the case of rodents or indirectlv
as in the case of lizards eating plant-feeding in-

Table 3.1 Analyses of Plants Growing on Uranium-bearing Roeks and Soils of the Colorado Plateau 1949-50°

U

PPM

Se

Locality and Sample

Ash Soil

(per or

cent ) rock

Plant
ash

Soil
or
rock

Soil
Plant or Plant
ash rock ash

Temple Mountain, Utah

Asphaltic ss. ore, Shinarump conglomerate

Wood in ore

Juniperus monosperma tops 4.0

Juniperus monosperma tops 4.0

Oryzopsis hymenoides tops 5.5

Stanleya pinnata tops 10.0

Sta. 1 Yellow Cat area, Thompson's District, Utah

S. 2 Sandstone near carnotite ore

"°P. 8 Oryzopsis hymenoides — tops 7.3

Oryzopsis hymenoides — roots 35.2

S. 3 Surface soil 5 ft. above carnotite ore

S. 4 Shale layer in ore

P. 10 Ephedra viridis — tops 14.7

P. 12 Atriplex confertifolia — tops 31.3

P. 56 Artemisia spinescens — tops 18.4

Artemisia spinescens — roots 13.2

760
4,900

1,500
7,800

66

100

20

37

350
470
260
120

290

30
40

70
1,600

180

23,000

20

50

70

100

0.8
3.0

n.d.

n.d.

10

190

'Adapted from Cannon 1952, Tables 6 and 8.

"F.S. Grimaldi, Ruth Kreher, Claude Huffman, and F. N. Ward, chief analysts.
°° Selenium-indicator plants

BlUCKAM YODNC UNIVEBSm SCKNCl Hi i i mis

the problem ol determining radiation
source becomes more complicated and its effects
more difficult to determine,

riw s\/r 11! tins area (approximately thirty-
five ac res I has piw ided .1 sufficiently large pop-
ulation dI small vertebrates to make a variation
study possible. This has been the area in which
we nave conducted .ill of our studies dealing
with home range in the lizards.

Approximately two miles west of Temple
Mountain are two areas represented bv small
drainage ravines, both showing a reduced radia-
tion count, which have been used for the secur-
ing of comparative specimens. Each of these
anas, designated as upper and lower ranges,
represents approximately one half of the area
at Temple Mountain, and has been intensely
collected. In both of these areas the ore-bearing
strata have been eroded away. They are, how-
ever, at approximately the same elevation as the
Temple Mountain Mesa because of the San Ra-
fael Swell uplift and the accompanying erosion.
Even though there are some slight differences
in the soils, it appears to support the same gen-
eral type of plant cover and to represent approx-
imately the same percentage of plant canopy-
cover when compared with the Temple Moun-
tain Mesa Area.

Mamie Stover Area. This area is located in
the Northeastern edge of the San Rafael Swell.
but lies just east of the San Rafael Reef. Overly-
ing this area is a cap rock varying from five to
seven feet in thickness. Embodied in the cap rock
are a series of fossil trees. Much of the radioac-
tive ore is associated with these fossils, and in
some instances rather high grade pockets have
been mined. Erosion has cut a series of channels
through the cap rock. Other areas are partly
eroded. This, plus the fact that small cracks and
crevices have developed in the cap rock, has pro-
vided for a high surface gamma radiation in this
area (Table 12). In some areas where erosion
has exposed the fossilferous material the ore is

at the surface and actually provides for tin- high-
est ground-level radiation found in any of the
areas under investigation. The basic nature of
the cap rock provides only for small pockets for
vegetation development other than what is
found in the larger ravines. Thus, there is an
ul iv inns bareness in this area which is not ap-
parent in the other areas. Transects were taken
in the more vegetated areas and therefore the
figures in Table 5 do not reflect accurately the
area as a whole but rather the vegetated pocket
areas where the animals live.

Yellow Cat Area. This area is approximately
fifty-five miles east of the San Rafael Reef and
fifteen miles Southeast of Thompsons, Grand
County, Utah. We have examined the area lying
south of the first (east) junction and west of
the road along the base of a ledge area in
which is found pockets of uranium and vana-
dium ores. This area is approximately one half
mile long, but not more than one hundred yards
wide at any place. Because of its basic geological
nature and based on our data gathered at Tem-
ple Mountain, it is considered that the small ver-
tebrates are rather restricted to this narrow belt.
The general area is less variable in altitude, with
fewer high ridges and deep vallevs. The eleva-
tion is perhaps a few hundred feet below that
of the other two areas (according to Cannon,
4,900 ft. ) , but does support approximately the
same plant and animal species. Much of the
data gathered by Cannon (loc. cit.) indicates
that approximately the same plant species occur
here as in the San Rafael Swell areas.

Indian Creek Area. This is a small area at
the bottom of a rather large canyon. Erosion
has exposed the ore-bearing strata so that they
are exposed on each side of the small creek for
a short distance. The greatest contrast here is
in the greater abundance of vegetation, particu-
larly along the stream. The apparent increase in
plant cover is seemingly reflected bv the more
dense lizard populations, particularly in Cnemi-
dophorus tigris.

REVIEW OF THE OBJECTIVES

In order to be more explicit, each of the
tour original objectives will be summarized with
as much detail as the data will permit. The ob-
jectives as originally listed have required some
modifications. Perhaps the most significant oi

these is the inclusion of the hasie plant ecology
and of soil depths for the major areas listed

above. Most of this will be included as a part
of the first objective.

Objective 1. Determine kinds of populations
of vertebrates occurring in these natural radia-
tion areas as compared with nearby nonirradi-
ated areas. (The word iwnirradiatcd should be

jioLociOAL Series, Vol. 7, No. 1, May, 1965

changed to read low surface radiation, since all
areas show some surface radiation. )

A. Cold-Blooded Vertebrates. Three species
of lizards are relatively common throughout
the area ( Uta stansburiana subsp., Sceloporus
undulatus elongatus, and Cnemidophorus tigris
septentrionalis). Other reptiles taken in limited
numbers are these: a few specimens of Crota-
phtftus colhnis, Crotaphytus wislizeni, Urosaurus
ornata, Masticophus taeniatus and Crotalus vir-
idis concolor. Amphibians have been found at
Mamie Stover and at Yellow Cat. In both in-
stances only a few individuals of Bufo punctatus
and Bufo woodhousi were taken, and they are
not therefore being included in the project
study.

B. Warni-Blooded Vertebrates. Seven species
of mammals were found to be most abundant
in the study areas. Phylogenetically they are as
follows: Ammospermophilus leucurus, Eutamias
quadrivittatus, Peromyscus crinitus, Peromyscus
maniculatus, Peromyscus truii and Neotoma le-
pida. Other species taken but not in sufficient
numbers to justify variation studies are Dipo-
domijs ordii, Neotoma cineria, Lepus sp. and
Sylvilagus sp. Signs of large carnivores of both
canids and felids were observed in all areas.

A more detailed comparative studv of the
populations of each species will be made later;
however, the skin and skull material available
for study is listed in Table 4 for each area stu-
died. This material is now available at the
Brigham Young University Museum.

Those listed from Temple Mountain Mesa
were taken during the late summer and fall of
1960. They appeared to be abundant, and we
were therefore selective, taking only a small
sample of adults. For reasons not fully under-
stood by us, all small mammals disappeared dur-
ing the winter of 1960-61. In June of 1961 we
did not trap a single mammal. An investigation
of the wood rat (Neotoma) nests revealed no
live inhabitants, but did provide several skele-
tons. It is our guess that we experienced a local
crash cycle which eliminated the Temple Moun-
tain populations. We did not notice a reduction
in populations at other areas, although there
was not a large population of any mammal spe-
cies at Mamie Stover.

Also of importance was the apparently com-
plete elimination of all small mammals in the
nearby upper and lower ranges. There are there-
fore no comparative mammals from these areas.

The geographical extent of this "crash" was
,not determined. It did not affect those popula-
tions at Mamie Stover or at Yellow Cat, but was

so complete at Temple Mountain that no small
mammals were seen or trapped during the sum-
mer of 1961 and onlv a few chipmunks were
seen in 1962 and 1963.

C. Plant Species and Vegetation Analysis.
The vegetation was analyzed in all areas being
studied in order to determine the dominant plant
species within each area.

The canopy-cover was determined by follow-
ing a transect and placing a pointed plot frame
over the plants along the transect every meter.
The purpose of using this method was to obtain
a two-dimensional evaluation of the influence
each plant taxon exerts over the other compon-
ents of its ecosystem. All of the individuals of
one taxon that fall within the plot frame are con-
sidered as a unit. An imaginary line is drawn
around the leaf tips and projected onto the
ground. The taxon is then placed into a cover
class according to the percentage of the plot
frame the imaginary polygon covers. The re-
maining taxa within the plot frame are then

Table 4. Lists of mammal species taken at three of
the study areas.

Species

No. of Specimens

TEMPLE MOUNTAIN MESA, EMERY COUNTY

Ammospermophilus leucurus
Eutamias quadrivittatus
Di)>odomys ordii
Peromyscus crinitus
Peromyscus maniculatus
P. truii
Neotoma lepida

18
13
1
4
20
4
8

MAMIE STOVER. INCLINE, EMERY COUNTY

Ammospermophilus leucurus 2

Eutamias quadrivittatus 5

Peromyscus crinitus 6

Peromyscus truii 1

Neotoma lepida 15

29

YELLOW CAT MININC DISTRICT
GRAND COUNTY

Ammospermophilus leucurus

23

Eutamias quadrivittatus

15

Dipodomys ordii
Peromyscus crinitus

2
9

Peromyscus maniculatus

3

Peromyscus truii

Onychomys leucogastcr melanophrys

Neotoma lepida

4
6
13

Neotoma cineria

3

81

12

Hltll. HAM ><>! M. I.MIHIMII S( l| M | BcLLETIN

iMni 5. Five dominant plants In the Temple Mountain, Mamie Stovet and Yellow Cat Areas.

Temple Mountain

Mamie Stover

Yellow C:.,t

Taxon

Per ■■lit

Taxon

Percent

Taxon

Percent

i Composition

Composition

Composition

Ililuria famesii

28.00%

Hilaria jamesii

26.48*

tularin jamesii

27.50*

< )ryzoptis hymenoides

20.50*

Con hum itansburiana

13.20*

Gutierrezia tarothrae

8.10*

Ephedra viridis

14.00*

Rhus trilobata

11.05*

Oryzopsis hymenoides

7.35*

Atriplex confertifolia

11.00*

Sttpa tpeciota

8.90*

Atriph I tp

7.16*

EHogonum inflatum

79.90*

Artemisia tridentata

8.75*
88. 183

Sphaeralcea pan (folia

7.02*
57.13*

added together, which results in an approxi-
mation of the total plant canopy-cover for the
area being studied.

The plant canopy-cover at Temple Mountain
Mesa was 21 percent, leaving about 78 percent
of the ground surface bare. Thirty-seven differ-
ent plant species formed this cover percentage.
However, four plant species, Ililuria jamesii,
Oryzopsis hymenoides, Ephedra viridis, and
Atriplex confertifolia, respectively were the
dominant plants. Together they formed 72 per-
mit dI the total plant cover.

At the Mamie Stover area the canopy-cover
was 23.71 percent, leaving about 76 percent of
the ground bare, mostly exposed standstone and
limestone. Here Hilaria jamesii was the domi-
nant plant species, followed closely by Cowania
Stansburiana, Rlius trilobata, and Stipa speciosa
respectively.

The plant cover at the Yellow Cat Mine ap-
pears to be greater than the two previous areas,
33.58 percent. This was not exactly a random
sampling, because the transects were taken at
the base of the ledges and through the valley.
Hilaria jamesii was still the dominant plant,
with C'.utierrezia sarothrae, Oryzopsis hymen-
oides, and two species of Atriplex making up
about 50 percent of the plant cover.

The five most dominant plant species for
each area and the percentage of composition
of each are listed in Table 5.

In order to determine the percentage of plant
Cover at (lie study sites a complete list of the
plant species for each area, excepting Indian
('nek. was made. There arc a tew different
species in each list, hut since there is consider-
able duplication only the complete list for the
Temple Mountain ana is presented.

In 1964 Harrison, Welsh and Moore pub-
lished a complete list ol the plant species occur-
ring in the Arches National Monument. Since
the Yellow Cat area is onlv a lew miles north
of the Arches Monument, I suggest that those

desiring more information mi the Yellow- Cat

flora use their list of plants.

In all the areas the plant cover forms an in-
tegral part of the cover for the small saurians
and provides a habitat for the insects which in
turn serve as food for most lizards.

The composition of the soil was either clay,
sand, or a mixture of the two, depending on the
surrounding geological formation. The rocks on
the surface of the soil provide shelter for most
of the small vertebrates. A penetrometer reading
was taken at each point where the flora was
sampled to determine the relative stoniness and
hardness of the soil.

Table 6. List of the plant species taken at the
Temple Mountain study area and arranged alphabetically
by families.

PLANT SPECIES COLLECTED

Anacardiaceae.

Rhus trilobata (Nutt.) T. and G.

Rhus utahensis Clodding
Boraginaceae,

CrypUintha sp.

Cnjptantha crussisepala ( T. and G.) Greene

Cryptantha fluid ( A. Nels. ) Payson

Descurainui spohia ( L. ) Webb

Lappula redowshii ( Hornem. ) Greene
Cactaceae.

Opuntiu rhodantha Schumann. Gesamtb.
Caprifoliaceae.

Symphoricarpos oreophiku A. Gray
Chcnopodiaceae.

Atriplex confertifolia (Torr. and Fram. ) S. Wats
Compositae

Artemisia tridentata Nun"

Aster sp.

Aster arenosui Blake

Aplopappus m aulii | Nutt. > A. Gray

BrickeUla longtfolia S Wats.

Chaenactis steoioides Hook and Am.

Chrysopsis oillosa (Pursh) Nutt

Chrysothamnus nauseosus (Pall.) Britton

Chrysothamnus i iscidtflortu ( I look) Nutt.

Laphamia stansburii A Gray

Lygodesmia exigua (Nutt.) Greene

Lygodesmia funcea (Pursh) D. Don

/■ ■triuli/miu glabrata A. Graj

Thelesperma subnudum A Gra)

Townsendia mcana A. Gray
i i ui iferae.

Lepidium denstftorum Schrad.

Biological Series, Vol. 7, No. 1, May, 1965

Lepidum montanum Nutt.
Malcolmia africana ( L. ) R. Br.
Physaria australis (Payson) Rollins
Stanleya pinnata ( Pursh ) Britton
Streptanthella longirostris S. Wats.
Euphorbiaceae.

Euphorbia fendleri T. and G.
Gnetaceae.

Ephedra viridis Coville
Gramineae.

Agropijron inerme (Schribn. and Smith) Rydb.
Agropyron smithii Rydb.
Aristida longiseta Steud. Syn.
Bouteloua gracilis (H.B.K. ) Lag.
Bromus tectorum L.
Hilaria jamesii ( Ton. ) Benth
Oryzopsis hymenoides (Roem. and Schult.)
Sitankm hystrix ( Nutt) J. G. Smith
Sporobolus crypUindrus ( Torr. ) A. Gray
Stipa columbiana Macoun.
Stipa comata Trin and Rupr.
Stipa speciosa Trin. and Rupr.
Tridens pulchellus (H.B.K.) Hitch.
Hydrophyllaceae.

Phacelia crenulata Torr.
Leguminosae.

Astragalus flavus ( Nutt. ) T. and G.
Astragalus lentiginosus palans ( Dougl., Jones. Hook )
Astragalus praelongus Sheld.
Loasaceae.

Mentzelia albicaulis Doug.
Malaceae.

Amelanchier utahensis Koehne, Gatt. Pomac
Malvaceae.

Sphaeralcea coccinea ( Pursh ) Britton
Nyctaginacese.

Albronia elliptica A. Nels.
Allionia linearis (Pursh)
Onagraceae.

Oenothera caespitosa Nutt.
Oenothera multijugas Wats
Oenothera walkeri (A. Nels.) Raven
Oleaceae.

Fraxinus anomala (Torr.) S. Wats.
Pinaceae.

Pinus edulis Engelm.
Plantaginaceae.

Plantago purshii Roem. and Schult.
Polemoniaceae.

Gilia gracilis ( Dougl. ) Hook.
Polygalaceae.

Poly gala subspinosa S. Wats.
Polygonaceae.

Eriogonum bicolor Jones
Eriogonum fasciculatum Benth.
Eriogonum hookeri S. Wats.
Eriogonum inflatum Torr.
Eriogonum pusillum T. and G.
Rosaceae.

Cercocarpus intricatus S. Wats.
Cowania stansburiana Torr.
Scrophulariaceae.

Penstemon utahensis /Eastw.

The penetrometer is a pointed steeel rod,
about one-half inch in diameter and ten deci-
meters long. A short pipe welded on one end
makes it possible to push the penetrometer into

the soil. The depth is measured to the nearest
decimeter.

The relative stoniness was greatest at the
Mamie Stover area; the average soil depth is 1.1
decimeters. At Temple Mountain the average
soil depth, 1.2 decimeters, was slightly greater
than at Mamie Stover. The soil depth at Yellow
Cat was much greater than that in the two pre-
vious areas. Here the soil depth was 2.9 deci-
meters. The minimum, maximum, mode and
average soil depths are listed in Table 7.

Tarle 7. Soil Depth Data for Temple Mountain,
Mamie Stover and Yellow Cat Areas. Measurements are
in Decimeters.

Temple Mt.

Mamie Stover

Yellow Cat

Min. 0. dm. Min. 0. dm. Min. 0. dm.

Max. 5. dm. Max 5. dm. Max. 9. dm.

Mode 1. dm. Mode 0. dm. Mode 3. dm.

Ave. 1.2 dm. Ave. 1.1 dm. Ave. 2.9 dm.

Objective 2. Determine the approximate
home range, seasonal distribution and migration
of individuals in these areas.

In the area here designated as Temple
Mountain, we have toe marked over four-hund-
red lizards. During the summer of 1960 (June 1
to September 1) marked ones were also recap-
tured in the course of capturing for marking.
However, most recaptures were made during
the summer of 1961, with some being made dur-
ing 1962 and 1963. Figures 5 and 6 provide the
approximate shape and size and indicates home
ranges of a few Ufa and Sceloporus of the Tem-
ple Mountain area.

Tables 8, 9, and 10 provide data from which
the home ranges are determined. Other useful
data (to be discussed below) such as size, rate
of growth and longevity are also indicated in
these data.

Many Uta and Sceloporus were captured two
or more times during the same summer and
found to have remained within a few feet or
yards of the original point of capture. The data
gathered during the summers of 1960 and 1961
were generally confirmed by the report of Tinkle
et. al. published in early 1962. Because of the
similarity between our data and that of Tinkle
only the larger seasonal ranges were plotted on
the map. However, by 1962 it became evident
that Uta of the Temple Mountain area were
longer lived than those reported by Tinkle; thus
the home range studies were continued. What
seemed pertinent and obviously our primary
concern was to determine if these lizards actu-
ally did remain in a limited area during their
lifetime.

Bricham Yoi m. i \i\ i. us 11 ■, s< u st h Bulletin

Table 8. Via ttatuburiana mbsp Two I

Growth in mm.

Days between

Distance (in ft.) Traveled

Lizard No

Sex

Snout-vent Length

' ill hes 1 and 2

Between Catches

13-20

M

2.0

56

0

10

M

5.7

820

20

18-19

F

1.5

85

10

19

F

0.6

90

57

2-20

F

0.0

95

10

2-5

M

3.0

93

42

2-10

M

1.0

93

45

2-17

F

0.5

93

20

1-2

M

3.3

730

225

3-4

F

0.0

730

375

2-7

F

0.0

90

0

9

M

2.9

60

254

10-13-14

M

0.0

20

45

9-10

M

0.0

19

30

3-15

M

2.0

90

90

3-6

M

1.0

90

45

3-8

M

0.5

90

60

3-17-18-19

M

1.0

90

102

4-19

M

2.5

90

0

5-20

F

0.0

90

30

1-2-6

F

0.0

45

24

8-10-15

M

2.0

630

300

Data gathered in 1962 and 1963 did not alter
our conclusions based on seasonal distribution
but did indicate that those lizards caught in
two or more seasons ( summers) had, with few
exceptions, increased the size of their home
range. This increase in range during two sea-
sons is explainable on the basis of a new home
range each season. Thus the lizard apparently
establishes a new home range each spring as
hibernation ends. Since the lizard is in the same
general area at emergence in the spring, the sec-
ond or third ranges tend to overlap in part at
least or to be adjacent to the one of the previous
year. Unfortunately this point did not become
obvious until the toe-clipped specimens were

collected as the field work was being termin-
ated. This is seemingly the only logical explana-
tion which will account for the increase in Uta
home range.

Our data dealing with longevity in Uta does
not agree with that recently published by Tinkle,
McGregar and Dana (1962). We have a number
of recapture records of this species marked in
1960 and recaptured during the summers of 1962
and 1963 (See Table 8). This is confirmed by
data gathered by Tanner and Jorgensen (1963)
at the Nevada Test Site in which the longevity
records are approximately parallel to those re-
ported in this study. The three areas of study in
Nevada, Utah and Texas are different in eleva-

Table 9. Vta stansburiana subsp., Three Catches

Crowth in

mm.

Days between

Distance ( in ft. ) Traveled

Lizard No.

Sex

Snout-vent Length

Catches 1

and 3

Between Catches

3-6

F

1.0

380

90

16-18

M

6.0

371

138

5-11

M

0.0

456

54

6-9

M

0.0

710

111

8-15

F

0.0

319

90

4-14

M

0.0

405

275

7-13

M

0.0

45

210

20

F

1.5

90

18

19

F

1.6

738

75

1-7

F

1.0

100

190

17

M

0.0

720

171

2-7

F

0.0

99

54

2 I

M

0.0

35

150

Ki-ll

M

13.0

628

125

4-5

M

1.0

724

30

3-17-18-19

M

2.0

354

102

Biological Series, Vol. 7, No. 1, May, 1965

15

Table 10. Sceloporus undulatus elongatus, Two and
Three Catches

Growth

Distance (ft.)

in mm. S-V

Days Between

Between

Toe No.

Sex

Length

Catches

Catches

4X

M

8.0

78

174

IX

F

2.5

360

165

19X

M

2.0

270

105

20X

M

0.0

330

105

7-15X

F

8.0

303

135

7-1 IX

M

5.0

332

141

8X

F

105

60

1-13X

F

33

109

4-5X

F

384

150

4-12X

F

22

20

6-8X

M

1

0

7-16X

F

25

10

tion as well as latitude. It must certainly be
true that Uta in Texas have at least three, per-
haps more, months a year in which to be active
over what is commonly the case in Utah. At the
Nevada Test Site the yearly activity of Uta ap-
pears to be intermediate between the study sites
in Texas and Utah. This range of activity also
may have contributed to such variations in lon-
gevity as are now appearing in our data.

According to Tanner and Jorgensen (1963)
the Uta found at the Nevada Test Site belong to
the subspecies stansburiana, the same as those
populations from central Utah, and are not stej-
negeri as reported by Stebbins (1954). If the
Nevada and Utah populations belong to the
subspecies stansburiana and the Texas popula-
tions to stejnegeri, a genetical difference as well
as climatic variables may be partly responsible
for the reported longevity differences.

Following is a summary of our findings for
this objective:

1. Home Range. — A. Both Uta and Sceloporus
have limited home ranges. In both species stu-
died the home ranges seemingly consist of a
general area, including areas for hiding, shelter
and perhaps basking (this is usually a ledge or
exposed rock), surrounded by a feeding area.
Home ranges vary in size from those having a
surface less than 500 square feet to those of
several thousand square feet, with Sceloporus
having the larger seasonal territories. Our data
include home ranges established for one season
and for those over a period of three summers.
This added time appears to have a decided ef-
fect on the size of ranges, for it has been noted
that these lizards do not move far during the
course of one summer, but may alter their range
as they emerge from hibernation. The extent of

range alteration must yet be thoroughly inves-
tigated.

The population of C nemidophorus tigris was
not large on the Temple Mountain Mesa. Be-
cause of their size a few may be deceiving as
to numbers. Twenty five were marked and five
were recaptured. In all cases those recaptured
were not far from the original point of capture.
The few data available from this study are
similar to studies presented on this subject by
Fitch (1958); Jorgensen and Tanner (1963)";
and Bellis (1964). An attempt to find marked
lizards 100 to 200 feet beyond the boundary of
the plot area was negative for all three species
marked.

The size of home ranges indicates that the
three species of lizards studied at Temple Moun-
tain remained in the area and would therefore
be subjected to a rather uniform radiation.

B. Female Uta do not have as large a home
range as do the males. It is also common to find
several females in close association; however,
males do not seem to show this association even
after the mating season. During April and May
our observations indicate that Uta are nearly
always found in pairs. Whether or not this spe-
cies is strictly monogamous is not yet estab-
lished. However, we have not found the home
ranges of males overlapping, but males and
female, and female ranges do overlap.

C. There is little evidence thus far that Uta
or Sceloporus change locations during a summer
season. Once they have established a territory,
it is not readily abandoned. If, however, the
lizard is removed from its range or, as in a few
cases, has its shelter destroyed, males will move
for rather long distances. While we were catch-
ing gravid females in 1963, 40 males were mark-
ed by placing a white paint mark on their heads.
They were then released at a central location
over one hundred yards or more from then-
home ranges. These males very rapidly dispersed
and a few days later several were seen between
two and three hundred yards from the point of
release. In all cases they had moved downhill
from the point of release.

2. Longevity. We have several records which
indicate that some Uta live for at least three
years. These records were first made in 1960,
at which time over a hundred Uta were marked
by toe clipping. Six of these were recaptured in
June of 1962. A continuation of these records
through the summer of 1963 repeatedly demon-
strated that in this species longevity is not an-
nual as indicated by Tinkle et. al. (1962), but

16

Bhicham Voi si. UmvEBsm Science Hi i 1 r 1 is

may in some individuals include .it least two
and perhaps three breeding seasons and include
.it least three to four years <>f lift-.

In l'J'il the opportunity came to make simi-
lar test-, at the Nevada Test Site, near Mercury,
lunty, Nevada. Under the supervision of
D. M. Allred and the direction of Clive D. Jor-
gensen pints were established for the purpose
of testing home range, growth and longevity.
The latter was demonstrated to be at least three
years for Uta in the populations of southern Nev-
ada (Tanner and Jorgensen, 1963).

The three locations where these studies have
been made (western Texas, southern Nevada
and southeastern Utah) all vary as to elevation,
yearly temperature ranges, yearly and perhaps
also seasonal precipitation and other basic eco-
logical factors. However, they are all in desert
areas with sparse vegetation and much exposed
surface of either sand or rocky surface areas.

A general consideration of the elevation may
have the greatest effects on the longevity, since
a decrease to less than 2,000 ft. elevation in-
creases the general range of temperatures; al-
though this may, and undoubtedly does, make
for higher temperature during the summers in
Texas, it also extends the activity range for Uta
to ten or more months and for mild winters
some activity throughout the year. The Temple
Mountain area is approximately 5.500 feet in ele-
vation and several hundred miles north of either
the Texas or southern Nevada study areas. At
this elevation and north latitude Uta are forced
to hibernate for at least three and usually for
four or five months. Some activity may occur
in March and October; however, in general only
six months (April through September) provide
climatic conditions suited to full lizard activity.

The effect of such climatic conditions in Utah
is to extend the reproductive cycle into the sum-
mer rather than the spring months. Eggs are
laid in late June and July. Most hatchlings ap-
pear in August. Thus only two, or possibly '2lA
months, remain for growth before hibernation.

Tins requires additional growth after hiberna-
tion and provides for a mixed population of
adults and juveniles during the spring breeding
season. It is not until June that the population

has reached a Dearly uniform adult status.

Our observations indicate that mating oc-
< ins in April and May, much too early for juv-
eniles to have reached sexual maturity. Our lim-
ited data in this area indicate that Uta do not
mate until the second spring and may then live
for at least one additional reproductive cycle;
perhaps some individuals live for a total of
three.

An example is found in Table 1, Specimen
No. 10, a male toe-clipped on June 9, 1960, and
measuring 43.3 mm., recaptured on September

2, 1962, nearly 27 months later, at which time
it measured 49 mm., a growth of 5.7 mm. This
lizard was a young adult, hatched in late July
or August of 1959 and having been extant for at
least 37 months when last seen in September of
1962. According to our field notes this lizard
was active and in apparent good health when
last seen.

3. Growth. A summary of the field data dealing
with growth and age in Uta is presented in Table
6. A comparison of these general data to those
of Tanner and Jorgensen (1963) for the Nevada
test site are surprisingly similar and reveal that
there is little ecological difference between the
population in east central Utah and southern
Nevada.

Uta are the most abundant lizard in the areas
studied. Although no attempt was made to de-
termine density of any species population, there
appeared to be little difference in general abun-
dance between the Temple Mountain, Utah and
Nevada Test Site populations of Uta.

There is undoubtedly some competition be-
tween the three species here considered. Scel-
oporus and Cneniidophorus are both larger than
Uta and may feed on larger insects, thus re-
ducing their competition with Uta. Scelopoms

Table 11. Size and Approximate Growth Hates for Uta stansburiana Summarized from Data Compiled over
a Period of Four Years at Temple Mountain, Emery County, Utah.

Date

Age Group

Snout-Vent
Length

I it. |nl\ and August

August
Middle to Late September
March and April
Ma)
[une
July
2nd Year

Hatchlings is they emerge from eggs 21-25 mm.

Hatchlings approximate!) one month old 25-30 mm.

[uveniles six. weeks to two months old 31-35 mm.

[uveniles emerging from hibernation 36-39 mm.

SubadultS reaching adult size !>nt not showing the nuptial color 38-41 mm.

Subadults or adults 42-45 mm.

Adults 44-47 mm.

Adults 45-51 mm.

Biological Series, Vol. 7, No. 1, May, 1965

17

inhabits the rocky sidehills and ledges, whereas
Cnemidapharus is in the depressions and along
gullies where there are fewer rocks and open
spaces of sand and eroded soil. Thus the size of
food items taken and the ecological niche in-
habited by these species reduces the competi-
tion for food and space.

Because these three species live together —
at least Uta is found throughout the ranges of
the larger species — it was suspected that the
larger species may prey upon the smaller Uta.
To determine this a thin metal sheet fence of
about 2Ji feet high was set up in the Temple
Mountain area, and a known number of individ-
uals of each species was placed inside. The
fence surrounded several small brush and an
open sandy area. Rocks were placed in a loose
pile for added shelter. For nearly a month the
lizards were fed insects gathered from an alfalfa
field some miles away. The insects varied in
size and kind and were readily taken by the
Uta and Sceloporus. Cnemid&phorus was shy
at first and one refused to eat. By controlled
feeding we could be assured that the larger liz-
ards were hungry, yet neither of the larger liz-
ards preyed on the Uta.

Objective No. 3 — The third objective was to
establish the amount of radiation in the various
areas under consideration.

The ground surface radiation count of gam-
ma and beta rays was made at various times
since June 1, 1960. Tests were made at Temple
Mountain, three mLles west of Temple Mountain,
and at Mamie Stover during August 1960. In
September 1961 tests were made at Yellow Cat
and in May 1962 Indian Creek was tested and
rechecks were made at the other areas. In order
that all radiation measurements would be figur-
ed the same, all tests were made by Dr. A. Les-
ter Allen and me. Table 12 provides the avail-
able data.

Objective No. 4. — The fourth objective was
to determine the degree of external morpholog-
ical variation existing in populations inhabiting
areas of natural radiation and to compare these
with each other and with populations occurring
in nearby areas having a lower surface radiation.

The data presented in the following graphs
have been taken from those external character-
istics for which a more exact statistical analysis
could be made. Therefore, scale counts in lizards
are used for the graphs and a discussion of color
and color pattern is provided for the lizards and
small mammals.

Measurements were taken for all species ex-
amined. These data are, however, dependent on
such variables as age and available food and
thus were not included as graphed data. An
analysis of growth and size is provided for Uta
and Cnemidophorus in the Nevada Test Site
study by Tanner and Jorgensen (1963). These
data correspond closely to that observed in this
study. Although growth and size were not ma-
jor considerations, some information has been
included in previous discussions or will be in-
cluded in subsequent discussions under each
species.

Before considering the variations in lizards,
it should be emphasized that in the areas stu-
died only a few snakes (four Crotalus and two
Masticaphis) were seen during the four years
of field work. Thus adequate data for such a
study are available only for the lizards.

An examination of reptiles in general and
Sauria in particular shows that these as a group
have scale patterns which are used as standard
characteristics in determining their classifica-
tion. I have used six of these scale counts in
gathering the variation data: dorsal scales, from
parietal to base of tail; ventral scales, beginning
with enlarged scales at or near anterior margins
of front legs to the last enlarged scales anterior

Table 12. Radiological Survey (mr/hrs). Units: X background (11 mr/hr, at Provo, Utah)

Test Area Location

Generalized Area
Gamma only Gamma and Beta

Specific
Gamma only

Pockets

Gamma and Beta

Temple Mountain

1.5-3.5x

Not Measured

Up to 3.5x

Not measured

3 miles West of Temple

Mountain
Mamie Stover

1.5-1.8x
1.6-5.5x

Not measured
1.5s

No increase
general area
Up to lOOx

over Not measured
Up to lOOOx

Yellow Cat

1.5-2x

l-5x

Up to 25x

Up to 500x

Indian Creek

1.5-2x

l-10x

Up to 12x

Up to 500x

Note: Utah Valley background radiation increased 2.5x (gamma) to 8x (gamma and beta) during the
late spring of 1962 because of the nuclear tests just prior to that time. This increase was found elsewhere in the
west perimeter, proportional to the rainfall. None of the test areas had any noticeable increase, which may be
attributed to the relatively small amount of precipitation in those areas since the Russian and United States tests
in 1961 and 1962.

Bun. I i \M YOUNC I'm. I km II SCIENCI Hi I I Ills

tn Miit, scale rows around midbody; total fe-
moral pores; supralabials, counting each side
separately; and infralabials, counting each side
separately. These characteristics are graphed

tin . .M li spri irs and tin each population.

The Uppei Colorado River Basin "I I tah
.mil the adjoining states to the south and easl is
a vast area ol distribution lor these lizards, ami
there an- undoubtedly clinal variations which
provide For sanation among the populations. In
order to remove as much clinal variation as
possible the anas ol stud) were selected within
a radius ol tiltv miles and in areas ol similar
ecology In Fact, the lizards taken in the anas
of high surface radiation inhabit the same rock
Formations in the Stud) aieas. thus providing
as nearl) as possible a similar physical and bio-
logical environment. Except lor the Mamie Sto-
ver ana there is also a close similarity in the
radiation Factor.

Each area and each Species lias been given
a code number. For Uta the code number is A
Sceloporus, B; and Cnemidophorus, C. The areas
an- coded as ol. 02, 03, etc. Table 13 will provide
the code information for all graphs, species and

areas.

Each graph provides the Following informa-
tion: the area, siieh as A.-01; the range ol vari-
ation, indicated In- the long horizontal line, and
the mean by the vertical line above the range.

TABLE 13. Ana Code Numbers fur Uta stans-
buriana*

A-01 Temple Mountain Mesa

\-o_: Lowei remple Mountain Range

A-03 Upper Temple Mountain H.inge

\ oi Mamie SI

\-<r. Yej]

\-nd Indian ( !n i V.

loporus substitute B and for Cnemidophorus C.

I he rectangle encloses one standard error of the
mean and the horizontal line above the mean
represents one standard deviation on each side
nl tin mean. The figures on the left margin of
the horizontal lines are the number of speci-
mens included in the calculations for each area
studied.

For each of the six areas three species of liz-
ard were studied, and for each species six char-
acteristics were considered in the analysis. Each
characteristic is graphed lor each population,
thus pro\'iding 18 graphs.

The six graphs prepared for each species | I ta
stansburiana, Sceloporus undulatus and Cnemi-
dophorus tigris) will be briefly discussed as
each species is considered. A general considera-
tion of variation, as it pertains to the lizards and
small mammals, will then summarize the results
of the project.

1 T/AKD SPECIES STUDIED

Uta stansburiana stansburiana Baird and Cirard

The Northern side-blotched lizard inhabits
most biotic communities in the Upper Colorado

River basin From the low river valleys up to the

funiper-Pinyon pine Forests. In this general area
Uta is the most abundant of the lizard species
and is more commonly seen in the brushy habi-
tat than on the ledges or rocky talus slopes.

Our Field records indicate that of the three
lizards studied Via is the most active. That is. in-
dividuals leave hibernation earlier in the spring
and seek it later in the tall. During the summer

hi months ol activity, Uta is the Inst tn le.ive

shelter ill the moinuig and the last to seek it at

evening. The availability of Uta lor study, not

only its abundance in the study aieas but also

its wide range ol activity, has permitted me to

use larger series in all characteristics analyzed.

Figures 7, 8, 9, 10, 11 and 12 provide the data

lor the six external morphological characteristics
used in the analyses of ("/<;. Each one will be
briefly reviewed. As indicated in Table 12. the
area coded as A-01, A-Of. A.-05 and V.-06 were
the stuck areas with a higher surface radiation.
The other two were adjoining A-01 and had a
lower surface radiation. Hereafter only the code
numbers will be referred to rather than a re-
peating ol localities.

Results: 1. The scale rows as analyzed in
Figure 7 do not provide significant differences

in the variations between the populations stu
died. Are, is \ 01 and \ 0~> are the most diver-
gent, yet in both of these populations the stan-
dard deviation ol their means is overlapping

one another.

Equally significant is the lack ol variation
between area \-0l and the adjoining areas ol
\ 02 and V-03.

Biological Series, Vol. 7, No. 1, May, 1965

19

n

I i i i i I i i i ' | i i i i I i i i i | i i i i I i i i i |

II I I I I I I | ' I ' I I I I I ' | ' Ml I

I I I I ■ I I I I ■ I I I I I | I I I I I I I I I I I M I I I l I I | I '■ I I I I I I

A-OI
A-02
A -03 —

A-04 .

A-05
A-06
' » ■ ' ■ I ■ l I I

jtL

315

251

261

-T±L

JZtL

80

90

100

,1m

110

Fig. 7. Number of scale rows at midbody in six
populations of Uta stansburiana . The range of variation
is indicated by the horizontal line, the mean by the
vertical line, the rectangle encloses one standard error
on each side of the mean and the shorter horizontal line
above the vertical line ( mean ) represents one standard
deviation of the mean. The number of scales ( rows ) is
indicated both at the bottom and the top of the graph,
with the populations (areas A-01 ) to the left of the
range (long horizontal line) and size of the population
sample on the right margin of the range.

2. The number of scales in the dorsal series
(Fig. 8) is noticeably more variable than the
number of scale rows (Fig. 7). A skewing to the
right of the populations A-04, -05 and -06 in
this character is obvious. Again the areas A-01,
-02 and -03 are similar. The greatest variation
between these populations occurs between A-02
and A-05. In these there is a difference of 6
scales between the means; and although the
standard deviations overlap, this is less than 50
percent.

3. The numbers of ventral scales as graphed
in Figure 9 show little variation.

4. Figure 10, representing the total number of
femoral pores, provides a different pattern of
variation. In the first four populations there is a
decrease in the mean from A-01 to A-04. The
means of A-05 and A-06 are noticeably skewed
to the left, yet as graphed the means vary only
approximately three pores and the standard dev-
iations of all populations overlap.

5. Variation of the supralabials in these pop-
ulations is nil. The variation between the means
is less than one scale ( Fig. 11 ).

A-03

A-04
130 A-05

280 A"06

rh

rh

rh

rh

_rb_

40 80

110

120

311
253
272

122
265

38
J

Fig. 8. Dorsal scale in Uta stansburiana counting
from parietal to base of tail. See Fig. 7 for explanation
of graph.

6. The infralabials (Fig. 12) in contrast to
the supralabials are more variable. Extreme var-
iation between the means is approximately 1.2
scales. This is in contrast to a mean variation
in the supralabials of approximately 0.3 scales.
A total range of variation in the supralabials is
6 and for the infralabials 5 scales. A variation of
more than 1 scale in a character with so small
a range of variation seems noteworthy, yet the
standard deviations of all populations overlap.

| I i i i | i i i i I i I i i | i i I i I i i i i | ' i i i I i i i i |

■ i i i | i i i i i i

A-01

A-02

A-03

A-04

A-05

A-06

jh.

J±L

JtL

I i i i i I i i i i 1 1 i i i I i i i i I i i i i I

314
258
264

133

277

38

50

60

70

Fig. 9. Ventral scales in Uta stansburiana counting
from the enlarged scales just posterior to the gular fold
(this is approximately at the anterior square of the front
legs) to the granule scales at the vent. See Fig. 7 for
explanation of the graph.

20

r

A-OI-
A-02-
A-03
A-04
A-05
A-06

"i r

"i 1 1 r

_r±i_

BlUCHAM Yi'l NC IMsikmii Si itNCE BULLETIN

71 — ' ' ' — ' — i

290

_L±L

I— I— I

177

192

98

.255

-33

21 22 23 24 25 26 27 28 29 30 31 32 33 34 35

Fie. 10. Total number of femoral pores on both femurs in Uta stansburianu . See Fig. 7 for explanation of graph.

"I 1 1 1 1 1 1 1 1

A-OI , ttl ,

i r

296

A-02

rh

_m.

_r±L

A-OI

A-02

A-03

A-04
271

37 A-05

183

98

Fie. 11. Total number of supralabial scales on both
lower lips in Uta statwburiana. See Fig. 7 for explana-
tion of graph.

Sceloporus undulatus elongatus Stejneger

The Northern Plateau Lizard inhabits the
rocky boulder-strewn sidehills and the ledgy out-
croppings within its range. It is found in north-
western New Mexico, northeastern Arizona,
western Colorado and all ol the Upper Colorado
Basin oi Utah extending as far north in the
Green River Valley as southern Wyoming.

Foi this species the same general procedures
will be followed .is foi ' to. The graphed data
foi Sceloporus are found in Figures 13 to 18,

A-06

10

rh

rh

rh

rti

r-h

II

12

13

14

298
177
190
94
263
35

15

Fig. 12. Total number of infralabia] scales on both
lower lips in Uta stansburiana. See Fig. 7 for explana-
tion of graph.

and the scale characteristics will follow in the

same sequence as was used in considering 1 7a.

rhe reduced numbers oi individuals reported

in the SceloporUS populations Samples represent

all the individuals we could secure, and are not
small samples In choice. The population at B-04
was Original!} Small, and alter several collecting

trips the lust year (1960), the population was
depleted.

Biological Sebies, Vol. 7, No. 1, May, 1965

21

I I I l | l l I ' I ' l I ' | I I l I I I I I I

I l 1 I I I I | I l l I | I ' l | |

B-OI —

rh

8-02

TT1

R-O1^ —

rh

B-04 —

r-h

R_rt*S .-

rti

R.Ofi ^

rh

42

B-OI

70

B-02

58

B-03

17

B-04

75

B-05

56

8-06

_L±L

A.

_lH

. m

_LiL

jtL

41
74
60

17
75
56

i i i i l i i >

u_i_i_

' > ■ » i I » *-

-Li.

50

60

40

50

Fig. 13. Number of scale rows at midbody in
Sceloporus undulatus elongatus. See Fig. 7 for explana-
tion.

Results: 1. The scale rows around the body
vary little in these populations. An extreme var-
iation of only 2 scales exists between the popu-
lation means. Perhaps the most interesting fac-
tor in this character is the narrow range of vari-
ation (8 scales) in the B-01 population.

2. The dorsal scale series is very uniform,
with the variation between any two population
means being less than 1 scale. In contrast to the
scale rows around the body the population B-01
has the greatest range of variation (12 scales),
and population B-04 the least with only 7.

3. There is more variation in the ventral
scale rows than in either of the preceding char-
acters. In both the total range (9 to 16 scales)
and in the population means ( 2 scales ) there are
greater variations. In this character, in spite of
the larger number of individuals the population
B-05 has the least variation, and B-01 is larger
than any other by 3 scales.

In this character as in the two preceding
characters there is considerable overlapping of
the standard deviations.

4. The total number of femoral pores ranges
from 30 to 47 pores; however, the means for the
populations vary only approximately 2 scales.
This is not a great variation as is indicated by

Fig. 14. Dorsal scale in Sceloporus undulatus elon-
gatus, counting from parietal to base of tail. See Fig. 7
for explanation of graph.

| I » i » I I i I I | I I I I I i I i I | i '' I I I I I I I

B-OI

B-02

B-03

B-04

B-05

B-06

rh

rh

rr-i

i — 1 — i

rh

rh

42

71

59

17

76

53

I ■ » ■ ' I i . » i < i > » ■ I . . i ■ i . .■ i I i i i i i
40 50 60

Fig. 15. Ventral scales in Sceloporus undulatus
elongatus counting from the enlarged scales just posterior
to the gular fold (this is approximately at the anterior
square of the front legs ) to the granule scales at the
vent. See. Fig. 7 for explanation of the graph.

BRICHAM Vm MC I Milium S< it s< i Hi i i i iin

B-OI

B-02

B-03

B-04

B-05

B-06

I — 1 — I

r-t-i

i — r

c=t

cb

40
54
47
13
75
53

30

35

40

45

FlC. 16. Total number of femoral pores on both femurs in Sceloporus undulatus elongatus. See Fig. 7 for
explanation of graph.

the almost complete overlapping of the stan- 5. There is little variation in the supralab-

dard deviations. ials. The population means are all between

scales 8 and 9, and the standard deviations are
also basically in this same range. Noticeably dis-
tinct from ah other graphed data for this species

I 1 1 1 1 1 1 1

1 1 1 1

1 1

B-oi sh

b 02 n~i

41
51

B-03 J=b ■

36

1

B-04 J — 1 — ' ■

13
75

B-06 — nti .

till

56

1 1

B-OI

B-02

B-03

B-04

B-05

B-06

JZtL

JZtL

r— hn

_t±l

J±L

40
51

36
13

78

52

8

10

II

8

10

II

12

13

Fie 17 Total number of supralabial scales on both Fig in. Total number of infralabials scales on both

upper lips in Sceloporus undulatus elongatus. See Fig. 7 lower lips in Sceloporus undulatus elongatus. See Fig. 7
for explanation ol graph !,,r explanation ol graph.

Biolocical Series, Vol. 7, No. 1, May, 1965

II i i i i i i i i

C-OI

I I I I I I I I I I I I I I I I I I

C-02

C-03

C-04

C-05

C-06

ri— i

_T±L

r-M

24
25
48
53
48
18

I,

80

90

100

Fig. 19. Number of scale rows at midbody in six
populations of Cnemidophorus tigris septentrionalis. See
Fig. 7 for explanation of graph.

is the extreme skewness to the right of the means
and standard deviations for all populations. In
those populations with the greatest range of
variation (4 scales, 8 to 11) the skewness is

greatest. This emphasizes the regular occurrence
of 8 scales in all populations.

6. In the infralabials there is a greater range
of variation, 6 scales, but much less skewness.
Populations B-01, B-02 and B-06 are skewed to
the left, whereas the others are slightly skewed
to the right. The variation between the popula-
tion means is less than 1 scale, and the standard
deviations of all populations overlap.

Cnemidophorus tigris septentrionalis Burger

The northern whiptail is the largest and
fastest of the common lizards found in the gen-
eral study area. It is more commonly found
along the washes but forages along ledges and
among the larger shrubs and juniper trees. This
species does not climb on the ledges as is com-
monly done by Sceloporus and occasionally by
Uta, but remains on the ground. It may find
shelter under ledges or large rocks, but is seem-
ingly more at home in rodent holes under the
branches of a shrub or a hole under or along
the side of a rock.

As in the case of Sceloporus, the whiptails
are not as abundant as Uta and were practically
eliminated from the study areas as we secured
the series of specimens.

Those graphs with the data for CnemidopJio-
rus are found in Figures 19 to 24. The same

i l I I l ' i | i i i I I i ' i i I ' i i i I i i i i | '

I i i i | '

' ' ' ' | I i

C-OI

23

C-02

J=t

26

C-03

I I I

48

C-04

C-05

■ — I — I

I I I

52
48

C-06

18

U-L

i I I i I i I I I

I i I I I

I ■ I I

I I I I I I I

150

160

170

180

190

Fig. 20. Dorsal scales in Cnemidophorus tigris septentrionalis, counting from parietal to base of tail. See
Fig. 7 for explanation of graph.

J I

Hun. ham Vounc Untvebsiti S( i knce Bulletin

| I I T I I I I | I | I I I I I I I I I | I I I I |

c-oi — cb

c-02 — cb-

C-03 —

c-04 ih-

c-05 — cb

C-06

lfa_

J±L

24
26
50
54
49
18

' ■ ■ ■

■ ■ ■ i i ■ . i ■ l ■ ■ i i i ■ i i i

30

40

I i.. 21. Ventral scales in Cnemidophorus tigris
septentrUmaUs, counting from the enlarged scales just
posterioi to the gular fold (this is approximate!) .it the

. interior square of the front legs) to the granule scales
.it the sent See Fig. 7 for explanation of the graph.

characteristics presented in the same sequence as
in the two preceding species will be given fur

this spi

1. There is little variation in the range of

the stale rows. The greatest range of 22 rows is
found in population C-04 and least of 14 rows
in < 0] and C-06. An extreme of 4 rows (be-
i !-03 and C-04) occurs in the popula-
tion means; however, the standard deviations for
all populations are not significantly variable.

2. There is little variation in number of
stales in the dorsal rows. In this characteristic
the i rtremes oi the range of variation is 41
scales, and although this extreme is not rea< hed

bv all populations, the minimums is 27 scales.
It is noteworthy that the means for these popula-
tions vary less than 3 scales.

3. The number of scales in the ventral rows
of these populations are remarkably uniform and
show little variation. An extreme variation in
the range of only 12 scales and of only approx-
imately 2 scales in the population means is evi-
dence of the sameness of these populations in
this character.

4. The range of variation in the total num-
ber of pores is from 34 to 49, or 15 pores. The
two population extremes are between C-03, with
34 to 48, and C-05, with 40 to 49. Population

C-01

C-02

C-03

C-04

C-05

C-06

i— i— i

i— t — i

r-1-1

m

■ i ■

24

23

42

54

49

18

35

40

45

50

in. iz Total number ■>! Femoral pores on both femurs in Cnemidophorus tigrlt septenMonaUs. See Fig. 7
lor explanation "t graph

Biological Series, Vol. 7, No. 1, May, 1965

25

I 1

C-OI ■ —

r±3

C-02

c-03 -dn.

C-04 • 1— L

C.05 55

C-06

r-H .

— I

23

26
49
52
48
18

10

II

12

13

Fig. 23. Total number of supralabial scales on both
upper lips in Cnemidophorus tigris septentrionalis . See
Fig. 7 for explanation of graph.

means vary only 2.5 pores, and the standard
deviations overlap widely.

5. Perhaps the most significant variations
observed in this species occur in the supralabials.
Although one might expect the extremes of var-
iation to occur between those populations in-
habiting the areas with high surface radiation, in
this instance at least those two areas with lowest
radiation (C-02 and C-03) have not only the
greatest variation in range but also the greatest
mean variation.

In a character such as this, with a total range
of variation equaling only three scales, a popu-
lation means which varies 1 scale is seemingly

significant, not perhaps in this case to establish
an important and significant population differ-
ence, but to indicate the separateness of popula-
tions in a given small area. Populations C-02
and C-03 occur in adjoining small tributary
washes which are separated by a small ridge.
The washes join about half a mile below the
designated areas. Yet these populations in spite
of their close approximations do not apparently
interbreed. This is an added evidence of close
territorialism in this species.

6. The infralabials show no more variation
in the population means than occurred in the
supralabials even though the total range is twice
as great (6 scales). Figures 23 and 24 show a
very similar pattern of variation but with more
overlapping of the standard deviations in the
infralabials.

1 T

C-OI

C-02

C-03

rm

i— t-i

i 1

26

C-04

C-05

C-06

50

47

Fig. 24. Total number of infralabials scales on both
lower lips in Cnemidophorus tigris septentrionalis. See
Fig. 7 for explanation of graph.

MAMMALS

Five species of small mammals were found
in each of the study areas. Several other species
were taken in limited numbers in some of the
areas, thus not providing enough specimens for
adequate comparative purposes. Table 4 lists
the species taken at each area and the speci-

mens now available at the Brigham Young Uni-
versity Museum. Species with at least four adult
specimens from an area have been carefully
compared as to their pelage and color variations.
In making a preliminary comparison of the
allopatric species, several variations are notice-

Biucham Voi mc University s< ^se i Bulletin

according to our present understanding, they
belong in .1 single subspei ies However, it is very
(ili\i(nis that tin- Temple Mountain-Mamie Sto-
vei serii s n present a single population, whereas
those from Yellow < at represent .1 second dis-
tinct population. The latter r\liil>its .1 distinct
pelage pattern in which the four stripes stand
out sharpl) in contrast to the reddish ground
color. In contrast those taken at Temple Moun-
tain and Mamie Stover show less distinctness,
with both the stripes and the reddish ground
color appearing washed out or laded. This is re-
flected also in the tail and particularly in the
ventral-lateral region, which in some of the Em-
ery County population lias laded to a yellowish
linll rather than retaining the reddish ground
color. In the chipmunks a distinct pelage vari-
ation occurs between those populations sampled
in the San Rafael populations and the Yellow
Cat population. There is seemingly a different
gene pool than that occurring in the Yellow Cat
population.

Neotoma lepida sanrafaeli

The common packrats provide a similar pre-
sumably anomalous phenomenon in which the
pelage color and the basic variations in the
Mamie Stover population are duplicated by the
series from Yellow Cat, whereas the population
from Temple Mountain is relatively uniform in
color ,md distinct when compared with the
other populations. The pattern of the Temple
Mountain series has a greater amount of yellow,
particularly laterally, but blending into the
brownish-grey dorsal pelage and giving a more
yellowish, almost buff, color. In A', lepida. as in
A. leucurus discussed above, a similar popula-
tion variation relation exists; that is, the Yellow
( at Mamie Stover series seem to represent or ap-
proximate the same variations within each popu-
lation,

Peromyscus maniculatus sonariensis

The deer mouse from the Temple Mountain
area represents a population with considerable
pelage variation, ranging from slate to gray to
yellowish-buff, with several gradations in be-
tween these extremes. The range of variation is
striking, particularly when contrasted with such
uniform color patterns as are found in Scotoma
lepida and Eutamias. The remarkable factor in

this species is its absence from the trapping areas

at Mamie Stover.

Eutamias quadrivittatus hopiensis Peromyscus maniculatus nebrascensis

The chipmunks ot the area oiler a better The maniculatus population at Yellow Cat

example for comparative variation studies since represents the subspecies nebrascensis. This sub-

able The squirrels provide a variation in the
color ot the pelage, Those- taken at Temple
Mi tain show greatei variation within the sei

us and arc- not like the series from eithei Yellow
Cat or those taken thirty miles to (he north along

the s.m Rafael Reel in the Mamie Stover

',, is not as we had expected, In lac t

other series. Neotoma lepida sanrafaeli and
PeromySCUS CrinitUS doutti, show similar popula-
tion variations. Based on the nearness ol the
Temple Mountain and the Mamie StOVl 1
and the complete lack ol barriers, one would
expect greater similarities between these popu-
lations than between those occurring across the
Green River in the Yellow Cat district. How-
ever, only in the chipmunk and white footed
deer mouse do we Find a close similarity between
the two San Rafael populations. In the mammals,
as is evident in the reptiles, relatively minor
barriers seemingly have served to provide for
population and subspecies variations, or other
environmental factors operating in conjunction
with the barriers are responsible for the pelage
variations.

{mmospermophUus leucurus notum

The antelope ground squirrel occurring in
the three areas in which mammals were studied
represents two subspecies, notutu in Emery
County and pennipes in Grand County. The pop-
ulations indicate a very definite pelage varia-
tion, particularly between the Yellow Cat and
the Temple Mountain populations. In the Yel-
low Cat population and dorsal-lateral line stripes
are usually straight or with only a slight curve,
whereas in the Temple Mountain series the
stripes are noticeablv curved, often providing
for two undulations. The basic pelage color in
tin Temple Mountain population is lighter, with
a more buff-reddish color. Furthermore, there is
considerably more variation in the series, ex-
tending from dark brownish-gray to a light red-
dish buff. The lack ol variation between indi-
viduals of the Yellow ( !at population is most
sinking when compared with the Temple
Mountain series. The Mamie Stover area is rep-
resented by only two specimens, both of which
are very similar to those taken at Yellow Cat.
'They do not exhibit the basic pelage or pattern
characteristic ol the 'Temple Mountain series in
spite ol the fad that they inhabit a veiv simi

lar habitat.

Biological Sehies, Vol. 7, No. 1, May, 1965

27

species provides a number of interesting varia-
tions ranging from a dull slate gray to a rich
buffv brown. Specimens representing the ex-
tremes in pelage color are adults, and both
were taken within a day of each other in 1962.
Even though we cannot compare direetlv the
maiiicuhitus because of the different subspecies,
the variation within the population is obvious
and marked.

Peromyscus crinitus

The crinitus deer mouse occurring in the
areas trapped consists of two subspecies, doutti
in the Emery County areas and auripectus at
Yellow Cat. The latter is a dark brownish-gray
pelage with some buffv colors, principally on
the dorsal and lateral parts of the body and
snout regions, whereas the subspecies doutti is
much lighter, with yellowish-buff pelage nearly
uniform over the body and head. The series
from Temple Mountain is noticeably lighter than
is the one from Mamie Stover, although it is ob-
vious that these two populations are actually
not at as great a variance with each other as
are the individual members of Peromyscus
maniculatus sonariensis from Temple Mountain.

The Kangaroo rats, Dipodomys ordii, occur in
the Temple Mountain and Yellow Cat areas, but
represent two subspecies. Only a few specimens
are available for comparison, and these show a
distinctly lighter pelage pattern for the Temple
Mountain material corresponding very closely
to a few specimens taken at Green Water Spring
in San Juan County. A few other species are
present in the areas, such as Neotoma cinerea

and Peromyscus truei, but not in sufficient num-
bers to provide for reliable variation data.

A study in variation based on pelage pattern
and color is neither satisfactory nor satisfying.
The lack of small mammals at Mamie Stover
and the crash of the population at Temple
Mountain limited the numbers and this greatly
affected the securing of adequate data for the
mammals. Also the apparent similarities between
some species in the Yellow Cat and Mamie Sto-
ver populations would require much more samp-
ling of intermediate populations to permit any
conclusions.

Although several obvious variations have been
found between these mammal populations,
indicating a definite difference in the gene pool
of several species, the cause of these variations
is not at all obvious. One is not, on the basis of
the data available, able to suggest that the var-
iations occurring in these populations is not due
to normal individual variation induced by those
(actors in nature responsible for induction of
variation in most populations and maintained in
the population by limited geographical or eco-
logical isolation. Data now available from the
study areas, particularly the limited amount from
the San Rafael populations, indicate that most
species are variable and that these variations
represent, as in the squirrel A. leucurus and the
packrat N. lepida, a noticeable and pronounced
variation. Unfortunately these variables cannot
be subjected to statistical analyses, and there
is therefore no satisfactory test for these exter-
nal characteristics.

STATISTICAL ANALYSES OF THE DATA

In attempting to test the effects of surface
radiation on the external characteristics of small
vertebrates, I have investigated the general eco-
logical factors in the immediate environment of
the lizards and mammals and have measured
the approximate soil mantle which covers the
rocks bearing the radiation minerals. Insofar as
possible the study areas selected were as simi-
lar in elevation, general climatic conditions,
type and percent of plant cover, animal types
and geological formations as one could obtain
and at the same time include the areas of high
surface radiation. An attempt was also made to
keep the study areas on approximately the same
southeast slope.

In spite of all such attempts and the relative
nearness of the study areas, some physical and
perhaps biological differences, even though
small, still exist. The statistical tests indicate
that these differences did exist in the areas tested
(see Table 14). The extent to which these physi-
cal and biological differences created or influ-
enced the variations cannot be separated. The
predominance of the similarities, however, is
considered to be of great importance and to pro-
vide as much ecological uniformity as can be
expected in such a study program. It is there-
fore assumed that if the data indicate substan-
tial variations in some of the characteristics stu-
died, radiation may be, in part at least, respon-

Brich m v<» m. i mvBtum Senates Bulletin

I mi i II The analysis of variance (tests humus I Ix-twirii high and low exposure groups and areas within
exposure.

A

B

C

Exposure

Vreas

Error

Exposure

Areas

Error

Exposure

Areas

Error

Scale Rows

I) I
\l s
F

1
(i 112

4

235.8125
10.533

1271
10.533

1
78.5703

3.541

4

22.1875

3.417

312
6.4937

1
198 5468

8.813

4

45.2227
2.108

210

21.4545

D.F

\1 s
F

1
3206.7500
2.618

4

1225.0000

60.560

1255
20.2280

1
9.8047
4.832

4
2.0293

0.510

317
3.9776

1

38.87.50

0.706

4

55.0781

0.672

209
81.9931

Ventrals

D.F.

M.S.

F

1

191.3750

1.203

4

159.1406

10.855

1278
14.6601

1

30.0391

0.472

4

63.6367

5.026

312
12.6603

1

40.6406
2.694

4

15.0879
3.254

215
4.6366

Pores
D.F.
M.S.
F

1

77.1094
1.638

4
47.0703
13.317

1335
3.5345

1
56.6289
13.167

4
4.3008
0.625

323
6.8780

1

61.9180

4.350

4

14.2344

1.947

213
7.3091

S. Labials
D.F.

M S
F

1

6.2822
5.160

4

1.2175

1.607

1360

0.7577

1
2.4045
1.536

4

1.5651
4.418

324
0.3543

1

7.6895
1.226

4
6.2736
7.911

210
0.7931

I. Labials
D.F.
M.S.
F

1

19.0898
1.084

4

17.6113
27.116

1330
0.6495

1
5.6943
1.213

4

4.6930
5.050

324
0.9293

1

6.0088
1.298

4
4.6301
3.469

211
1.3349

sible. Therefore large samples of lizards from
each study area were collected and the data
provided in Figures 4 to and including Figure
21 were plotted. In order that the data are pro-
perly tested and all variations compared statis-
tically for their variation significance, all data
were submitted to Dr. Melvin W. Carter, Sta-
tistics Department, Brigham Young University,
for perusal and testing.

Under Dr. Carter's direction the data were
subjected to several tests, all intended to pro-
vide further evidence of the degree of signifi-
cances inherent in the variations. In developing
the following tests in which the variations in the
areas of high and low exposure were tested, the
variations associated with the lack of homogen-
eity of the test areas were taken into account.

1 . The first F test is an analysis of variance
with the Study areas divided into high- and low-
exposure groups and the areas within exposure
are used as the error in testing the effect oi
exposure. This test, the ordinary F tesi used in
the analysis of variance, determines whether or
not tlic average (mean) oi high exposure can
be s.inl to differ from the average (mean) of
low exposure. Table 1 1 provides the test data.

2. The second F test partitions the sum of
the square of the areas in the above analysis
so that a measure of the variation in the means
of the groups (areas) in high and low exposure
can be compared. Thus an F test of the varia-
tion of means of high exposure over that of low
exposure (divided by respective df) was used
to test whether or not the variation in the means
of the high exposure could be said to be greater
than that of the low exposure.

The statistical formula used to compute the
above F test is as follows:

F:„ =

where,

HE(SS)/3
LE(SS)/1

LE(SS) = n, (x,-x')'- + n2(x2-x')-'
HE(SS) = SS Areas-LE(SS)

n,x,4 ii \

x = i

n,+n;

The computed F values for the second test
are listed for the three lizard species in Table
15.

It should be noted that there are no signifi-
cant differences in the F values, as listed in

Biological Series, Vol. 7, No. 1, May, 1965

29

Table 15. An analysis of the variation in the means of the area group (high and low exposure), with a
listing of the F values for each characteristic.

Scale Rows
Source D.F. M.S. F

Dorsals Ventrals

M.S. F M.S. F

Pores S. Labials

M.S. F M.S. F.

I. Labials
M.S. F

H.E. 3 776.673 1.55 1577.245 9.37 209.887 30.41 62.428 62.18 1.361 1.73 20.133 2.01
L.E. 1 166.577 168.264 6.903 1.004 .787 10.035

H.E.
L.E.

3
1

11.688
53.687

0.218

0.635
6.212

0.102

C

H.E.

L.E.

3
1

49.076
33.662

1.46

71.241
6.588

10.814

0.635 0.102 67.851 1.33 5.553 10.23 1.994 7.15 5.209 1.657
50.995 0.543 0.279 3.143

7.595 0.202 15.634 1.56 3.940 0.297 4.034 0.629
37.568 10.025 13.274 6.407

Table 15, for any of the characteristics tested.

3. This test (solid variance of high exposure
different from pooled variance of low exposure)
was devised to test whether or not the average
variance of the groups with high exposure was
different from that of those with low exposure.
Table 16 provides the test data. The test used
was the F test as follows:

This F test was a one-tail test since under the
alternative hypotheses the high exposures were
expected to have a larger variance on the aver-
age. Table 16 provides the test data. Note that
the species are listed as A, B and C and the
characteristics as 1 to 6 and in the same order
as occurs in Table 15.

Table 16. Tests of the variance of high exposure
compared with that of low exposure.

F

dfd

df

A,

1.11

510

761

Bt

0.728

126

186

ci

0.892

71

139

A2

1.14

523

732

B2

1.29

132

185

c2

0.741

72

137

A3

0.838

520

758

B3

0.562

128

184

c3

1.35

74

141

A4

0.978

542

793

B4

0.976

134

189

C4

0.856

72

141

A5

0.845

558

802

B5

1.39

137

187

c5

1.25

73

137

A6

1.08

549

781

B6

1.12

136

188

C6

1.24

74

137

4. The following and final F test was de-
vised to test whether or not the heterogeneity
(differences) of variances was different in low-
exposure areas compared with high-exposure

F(2(ni-1), 2(11,-1))=

2 (n,-l) S,V2(n,-l)

2(11,-1) 8,72(111-1)

areas. The null hypothesis was that this hetero-
geneity, if present, would be the same for both
types of areas. The alternative hypothesis was
that the differences in variances would be larger
in the higher exposure than in the low expo-
sures. The test devised is approximate. The pro-
cedure used here was to calculate a Bartlett's
chi-square for each of the two type areas, high
and low exposure; then an approximate F ratio
was formed as follows:

Fa.u —

•X2L/1

None of these F ratios is significant at the
.05 percent level. There was insufficient evi-
dence to say that the variances of the high ex-
posure differed more than the variances of the
low exposure. Table 17 provides the data analy-
ses for this test. It should be noted here that the
test would not be sensitive to differences caused
by exposure to radiation because of the small
number of degrees of freedom in the F ratio.
In this test it would have been better to have
more groups in the study.

Id

BrICHAM Vh m. UNIVERSm SciENCl Bill. KM IN

I A in h IT The results <>l il»- approximate test, testing whether <>r not the lack of b i variance wa

eatet In high exposure >» low exposure

greate

Si ale Rows

x,,3

V

F».i

N,

N2

N3

N,

N5

N.

A,

1

3.0237

7.5191

0.1340

315

251

130

280

40

B,

2

5.7209

0.2836

6.7248

42

70

.58

17

75

56

C,

3

1 2448

0.3087

1.3440

21

25

IS

53

48

18

Dorsals

\

4

'i I iiil

0.1281

23.7512

311

253

272

122

265

(S

H.

5

i.2261

41

74

60

17

75

56

c"2

6

1.8337

0.5177

1.1806

23

26

48

52

48

18

Ventrals

\

7

10.7015

L.708]

314

258

264

133

277

38

B

8

25.6118

0.0466

183.3694

1-1

71

17

76

53

c3

9

2.2078

1 .3300

24

26

50

54

49

18

Pores

V

10

0.1780

9.4717

322

263

281

132

302

41

».

11

11.4748

0.5617

6.8093

42

75

61

17

78

56

c,

12

3.0224

1.5009

0.6712

24

26

48

54

49

18

S I„il)i.i]s

\

13

10.5368

0.191]

18.3768

322

270

290

134

310

40

B„

14

28.51(12

0.1852

51.3156

42

76

63

17

78

54

cs

15

1.7658

5.4965

0.1071

23

26

49

52

48

18

1. Labials

A„

16

29.6565

32.7505

0.3018

321

268

283

130

297

37

B„

17

4.6422

3.0460

0.5080

42

75

63

17

78

55

(',.

18

0.7661

0.0362

7.0532

23

26

50

53

47

18

CONCLUSIONS

Although the graphed data and the statis-
tical analysis of these data show some significant
variations, the great majority of the character-
istics tested do not show, at the 95 percent
level, significant variations between the popu-
lations in the areas of high and low exposure.
We did find thai the lizard populations in the
different areas varied in some characteristics.
These variations were significant. The differ-
ences between the population means of the
same species are indicated by the larger figures
in Table 1 1 in which the F values for the areas

(A=Uta) are noticeably higher than the F
values for exposure. Only in the I Labials is the
F value for exposure larger than the area value.
In B (Sceloporus) and C (Cnemidophorus)
characteristics there is closei F value between
exposure and area in most characteristics; how-
ever in the case of dorsals and pores in Scelo-
porta, .md scale rows and pores in Ctiemido-
phorus, the F values for the exposure groups are
noticeably higher. However, it should be pointed
oul that an analysis "I the data provided 1>\ the

F tests shows that the variations occur in areas
with similar exposure and do not represent sig-
nificant variation.

As indicated in Tables 14, 15 and 16. the
differences in the high- and low-exposure areas
did not vary more than in areas with similar
exposure; therefore we had insufficient evidence
(or data) to establish thai the variations were
caused bv or resulted from exposure to radiation.

It is possible that other tests and additional
data based on more areas or other areas and on
additional characteristics (perhaps internal) may
yet demonstrate that those- areas with a higher

surface radiation do produce in the small verte-
brates inhabiting them a greater range of varia-
tion or other evidences of genetieal change
than occur in areas ol lower surface radiation.
However, in this stud) no significant variation
is seemingly present when the areas ol higher

surface radiation are compared with areas cil

low surface radiation — that would not be ex-
pected to occur in nature bj chance alone.

Biological Series, Vol. 7, No. 1, May, 1965

31

Literature Cited

Bellis, Edward D. 1964. A Summer Six-lined Race-
runner ( Cnemidophorus sexlineatus ) Population in
South Carolina, Herpetologica, 20(1):9-16.

Boutwell, J. M. 1905. Vanadium and Uranium in

Southeastern Utah, United States Geological Survey

Bulletin, 260:202-210.
Brant, Daniel H. 1962. Measures of the movements

and population densities of small rodents. Univ.

Calif. Publ. in Zool., 62:105-184.

Burt, W. H. 1940. Territorial behavior and popula-
tions of some small mammals in southern Michigan.
Misc. Publ. Mus. Zool., Univ. Michigan, 45:1-58.

Calhoun, John B. and James N. Casby. 1958. Cal-
culations of home range and density of small mam-
mals. U. S. Public Health Monograph, 55:1-24.

Cannon, Helen. 1952. The effect of uranium-van-
adium deposits on the vegetation of the Colorado
Plateau, American Journal of Science, 250:735-770.

Daniel and Blair. 1959. Background radiation and
endemic faunal range in the San Luis valley of
Southern Colorado. Great Basin Naturalist, 19(1):
37-42.

Durrant, Stephen D. 1952. Mammals of Utah. Uni-
versity of Kansas Publ., Mus. Nat. Hist., 6:1-549.

Fitch, H. S. 1951. A simplified type of funnel trap
for reptiles. Herpetologica, 7:77-80.

. 1960. Autecology of the Copperhead. Uni-
versity of Kansas Publ.. 13(4):85-288.

Harrison, Bertrand F., Stanley L. Welsh, Glen
Moore. 1964. Plants of Arches National Monu-
ment. Brigham Young University Sci. Bull., Biol
Series, 5(l):5-23.

Hayward, C, Lynn, D Elden Beck, and Wii.meh W.
Tanner. 1958. Zoology of the Upper Colorado
River Basin. The biotic communities. Brigham
Young University Sci. Bull. Biol. Series, 1(3): 1-74.

Hilpert, L. S. and M. D. Dasch. 1964. Mineral and
water resources of Utah, Report of the U. S. Geo-

logical and Mineralogical Survey and the Utah
Water and Power Board. Prepared at the request
of Senator Frank E. Moss. U. S. Government
Printing Office, Washington, 1964:1-275.

Jorcensen, C. D. and W. W. Tanner. 1963. The
application of the density probability function to
determine the home ranges of Uta s. stanshuriana
and Cnemidophorus t. tigris. Herpetologica, 19(2):
105-115.

Mohr, C. O. and W. A. Stumpf. 1964. Relation of
tick and chigger infestations to home areas of Cali-
fornia meadow mice. Jour. Med. Ent., 1:73-77.

Nelson, R. F. 1957. A preliminary investigation of
the role of selenium in the growth of S. pinnata.
B.Y.U. Master's Thesis — unpublished.

Sorensen, Don. 1963. Wonder mineral: Utah's Ura-
nium. Utah History Quarterly, 31(3) :280-290.

Stebbins. Robert C. 1954. Amphibians and reptiles
of western North America. McGraw Hill Book Co.,
Inc., New York. 537 pp.

Tananka, R. H., Suciyama and S. Teramura. 1958.
A clumped distribution of trombiculid mites viewed
from the true home ranges of host roles. Japanese
Jour. Sanitary Zoology, 9:28-32.

Tanner, Wilmer W. 1963. Natural Radiation Ef-
fects on Vertebrate Animals inhabiting the Uran-
ium Areas of Southeastern Utah, pp. 325-326. In:
Radioecology, Proceedings of the First National
Symposium on Radioecology held at Colorado State
University, Fort Collins, Colorado, September 10-
15, 1961. Vincent Schultz and Alfred W. Klement,
editors. Reinhold Publ. Corp., New York and The
American Institute of Biol. Sci., Washington, D. C,
746 pp.

Tanner. Wilmer W. and Clive D. Jorcensen. 1963.
Reptiles of the Nevada Test Site. Brigham Young
University Sci. Bull. 3(3): 1-31.

Tinkle, Donald W., Don McGregor, and Dana Sum-
ner. 1962. Home Range Ecology of Uta stans-
huriana stejnegeri. Ecology, 43(2) :223-29.

Temple Mountain.

UTA STANSBURIANA
Z^ ROADS
//fl„ EDGE OF WASHES

7^" LEDGES

^§&»~ UTA HOME RANGES

,--— DISTANCE BETWEEN TWO CATCHES

»«ne tL5„ A,diagralni,;c *<** of the Temple Mountain area, showing the location of the home ranges of
™ Mo Xamlmrmna. The legend provides the basic information. 8

SCELOPORUS UNDULATUS ELONGATUS

" ROADS

„„„ EDGE OF WASHES

T^jf LEDGES

pggg» SCELOPORUS HOME RANGES

— DISTANCE BETWEEN TWO CATCHES

Sr„; K A diagramatic sketch of the Temple Mountain area, showing the location of the home
™porw unduktm ebngatus. The legend provides the basic information.

•ivn I s^T; LIBRARY

JUL 28 1966

HARVwr>J
UNIVERSITY

Brigham Young University
Science Bulletin

SIPHONAPTERA (FLEAS) OF THE
NEVADA TEST SITE

by

D ELDEN BECK

and

DORALD M. ALLRED

BIOLOGICAL SERIES — VOLUME VII, NUMBER 2
JUNE 1966

BRIGHAM YOUNG UNIVERSITY SCIENCE BULLETIN
BIOLOGICAL SERIES

Editor: Dorald M. Allred, Department of Zoology and Entomology,
Brigham Young University, Provo, Utah

Associate Editor: Eabl M. Christensen, Department of Botany,
Brigham Young University, Provo, Utah

Members of the Editorial Board:

J. V. Beck, Bacteriology

C. Lynn IIayward, Zoology

W. Derby Laws, Agronomy

Howard C. Stutz, Botany

Wilmer W. Tanner, Zoology, Chairman of the Board

Stanley Welsh, Botany

Ex officio Members

RudgerH.
Scienci

Ernest L. Olson, Chairman, University Publications

Rudger H. Walker, Dean, College of Biological and Agricultural
Sciences

The Brigham Young University Science Bulletin, Biological Series, publishes
acceptable papers, particularly large manuscripts, on all phases of biology.

Separate numbers and back volumes can be purchased from University
Publications, Brigham Young University, Provo, Utah. All remittances should be
made payable to Brigham Young University.

Orders and materials for library exchange should be directed to the Division
of Gifts and Exchange, Brigham Young University Library, Provo, Utah.

Brigham Young University
Science Bulletin

SIPHONAPTERA (FLEAS) OF THE
NEVADA TEST SITE

by

D ELDEN BECK

and

DORALD M. ALLRED

BIOLOGICAL SERIES — VOLUME VII, NUMBER 2
MAY 1966

COMP. ZOOL
LIBRARY

JUL 28 1966

UNIVERSITY

TABLE OK CONTENTS

Page

INTRODUCTION 1

LITERATURE REVIEW 1

METHODS AND PROCEDURES 1

ACKNOWLEDGMENTS 3

SCHEME OE CLASSIFICATION 3

FAMILIES AND SPECIES 3

SPECIES PRESENTATION 4

Echidnophaga gallinacea 4

Ptdcx irritans 4

Cediopsylla inaequalis 4

Hoplopsyllus anomalu.i 6

Atyphloceras echti ........ 6

Epitedia tvenmanni ............ 6

Catallagia decipiens ..9

Meringis dipodomys 9

Meringis parkeri .... 11

Meringis hubbardi 11

Jordanopsylla allredi II

Stenistomera alpina 13

CaUistopsyllus deuterux .... 13

Megarthroglossus procus .... 13

Anomiopsyllus amphiholus ....................... 13

RlmdiiuipM/lIu heiseri 15

Rhadinopsylla sectilk 15

Carteretta carteri 15

Thrassis hacchi .....18

Thrassis aridus 18

Dactyhpsylla bluei ... IS

Foxclla ignot/i 20

Diamanus montanus 20

Orchopeas sexdentatus ... 20

Monopsyllm wagneri ............. .. . 22

Monopsyllus eumolpi 22

Malaraeus telchinum .... 22

Malaraeus sinomus 22

Malaraeus euphorbi 24

Amphipsylla neotomae 24

OdontopsyUus dentatus 24

Peromyscopsylla hesperomys ................ 24

Nyctcridopsylla oarwouverensU ... 24

CONCLUSIONS 26

LITERATURE CITED 20

MSI (il III IMIi VI IONS

Figure Page

1 Map of the .southeastern corner ol Nevada showing the regional location of thi Nevada rest Sit< 2

2 Extent "i thi majoi plant communities ol the Nevada ["esl Site 2

igraphii distribution nl Echidnophaga gullinacca, Pulex irritant, and Cediopsylla inequalix 5

i Geographii distribution ol Hoplopsyllus tmomalits 7

Geographic distribution ol Uyphlocera echis Epiteiliu wenmanni, and Catallagia tlecipieru H

i' Geographic distribution ol \feringw diftodoniys and Weringh ptirkeri 10

7. Si-. ivun.il occurrenci ol WeringAi dipadamys I]

s Geographic distribution ol Weringw huhbardi, lordanopsyUa allredi, and Stenixtomera alpina 12

I Geographic distribution ol Cullistopsyllm deuterus, Anomio)>xyllus amphiholws, Wegarthroglossus

procus, .iiul Rhadiuopsylla lu iseri 14

10. Geographic distribution nl Rbadinopsylla sect ilia and Carteretta carteri 16

11 Geographic distribution ol Thrtuma hucchi IT

12 Seasonal occurrence "I Tbrassis bacchi .... is

Geographic distribution ol Thrassu aridi-i, Foxella ignota, Dactylopsylltt bluet, and Dianianin mon-

tanus 19

It Seasonal occurrence ol Thrassis aridh 20

IS Geographic distribution ol Orchopeat texdentatti* 21

16. Geographic distribution ol Monopsyllua wagneri, Malaraetm tclchinum, Monopsyllui eumolpi, and

Malaraeus rinomus 2 ;

!". Geographic distribution ol Malaraeus euphorbi, Odontapvyllm dent at us, Nycteridopsylla oancou-
oerensis, Amphipsylla neotomae, and Peromyscopsylla hesperomys

SIPHONAPTERA (FLEAS) OF THE NEVADA TEST SITE1

by

D Elden Beck and Dorald M. Allied

INTRODUCTION

Ecological studies at the Nevada Test Site
near Mercury, Nevada, were begun in 1959 as
a cooperative project between the United States
Atomic Energy Commission and the Brigham
Young University Department of Zoology and
Entomology. Initial research was directed to
mammals, birds and reptiles (Jorgensen & Hay-
ward, 1965; Hayward, Killpack & Richards, 1963;
Tanner & Jorgensen, 1963). As the investiga-
tions continued, however, parasites and other
eonsortes were collected. Reports on some of
these collections have been published ( Goates,

1963; Beck, Allred & Brinton, 1963; Allied, 1963;
Allred & Beck, 1963. 1964; and others— see list
of references ) .

This report deals with data on the fleas
collected at the Nevada Test Site. Geographi-
cal and ecological distribution, host relation-
ships, and seasonal occurrence are emphasized.
The biotic communities and areas of study were
designated by Allred, et al. (1963 a & b), and
our references generally follow their classifica-
tion (Figs. 1 and 2).

LITERATURE REVIEW

Previous records of fleas from the Nevada
Test Site are unknown, although there are re-
ports for the state of Nevada. Every record
from Nevada is not made here, but some gen-
eral ones are worthy of mention.

Most reports of fleas for Nevada are the
result of the extensive work by C. A. Hubbard
(1947). Practically all references made by him
are tor collections made in the northern half of
the state near or above the 39th parallel al-
though there are some collections from southern
Nevada. For example, Hoplopsyllus anomalus
was taken from the White-tailed Antelope Squir-
rel (Ammospermophilus leucurus), and Malara-

eus sinomus from the Deer Mouse (Peromyscus
maniculatus) at Searchlight. Hystrichopsylla
gigas dippiei (most likely //. dippiei dippiei)

was collected from the Sonoran Deer Mouse
(P. maniculatus sonoriensis) at Charleston Park
on Charleston Mountain near Las Vegas. Orch-
opeas sexdentatus nevadensis was taken from
the Desert Wood Rat (Neotoma lepida) from
Clark County. Hopkins and Rothschild (1962)
reported a contribution to the British Museum
of Anomiopsyllus amphibolus taken from the
White-throated Wood Rat (Neotoma albigula —
most likely .V. lepida) in Nye County. All lo-
calities are in the southeastern corner of Ne-
vada, relatively near the test site.

METHODS AND PROCEDURES

Most of the fleas were taken from the bodies
of host animals which were trapped or shot
specifically for the puqjose of collecting ecto-
parasites or selected from mammals and birds
which were collected for other purposes. In
each case the host was placed immediately into
a paper sack which was then sealed, data writ-
ten on it, and returned to the laboratory.

In some instances nests were removed from

'1! Y U -A E ( publication No. C00-1355-1+. This work was support
department of Zoology and Ento logy, Brigham Young Uni

the houses or other recesses of the Desert Wood
Rat, and the eonsortes extracted by the use of
a modified Berlese funnel. Data contained with
each collection included date, host or source,
and the biotic community where found. These
were coded for computer analysis.

Fleas were retrieved by brushing the fur of
the host which was held under a 75-watt lamp

ed (in part) by funds from the United States Atomic Energy Commission,
ty, Provo, Utab

Bhirham Young University v m s. i Bulletin

SOUTHERN NEVADA

Map of the southeastern comer of Nevada showing the regional location ol the Nevada Test Site.

Fleas of the Nevada Test Site

in a deep, white enamelware pan. All con-
sortes were preserved in 707c ethyl alcohol un-
til thev were mounted individually on micro-
slides to which were applied the collection data.

Specimens were identified from the slide pre-
parations, all data then placed on collection rec-
ord forms, and IBM punch cards prepared to
enable computer analysis.

ACKNOWLEDGMENTS

Many technicians were involved in the field
collection and laboratory preparation of hosts
and parasites. Without their careful work this
study would not have been as readily accom-
plished. We are most grateful for their unself-
ish, enthusiastic participation.

Dr. William L. Jellison, Rocky Mountain
Laboratory, Hamilton, Montana, corroborated
and assisted in most of the flea determinations.
Dr. Robert Traub, School of Medicine, Univer-
sity of Maryland, also helped to untangle some
specific problems in flea determinations. This
was likewise the case with Frank M. Prince,
Communicable Disease Center. U. S. Public

Health Service, San Francisco, California. Dr.
C. Lvnn Havward, Department of Zoology and
Entomology, Brigham Young University, Provo,
Utah, helped with the specific identification of
mammal and bird hosts.

We are especially grateful for the excellent
laboratory services and transportation facilities
which were provided by the Civil Effects Test
Operations (CETO), Division of the Atomic
Energy Commission, at Mercury, Nevada. We
are likewise grateful for similar conveniences
provided by Brigham Young University, Provo.
Utah.

SCHEME OF CLASSIFICATION

In general we have followed the taxonomic
and phvlogenetic arrangement described by Hop-
kins and Rothchild (1953, 1956, 1962). Where
special taxonomic studies have recently been
made on specific groups, such as the genus
Thrassis discussed bv Stark (1958) and unpub-
lished information on Malaraeus by Frank M.
Prince, information as deemed appropriate has

been judiciously inserted in the Hopkins-Roths-
child arrangement. The species of fleas from
the Nevada Test Site are listed below.

For the hosts we used the "Vernacular Names
for North American Mammals North of Mexico"
as approved by the American Society of Mam-
malogists (Hall, 1957).

FAMILIES AND SPECIES

Pulicidae

Echidnophaga gallinacea (Westwood)
Ptdex irritans Linne
Cediopsi/lla inaequalis (Baker)
Hoplopsi/llus anomalus (Baker)

Hystrichopsyllidae

Atyphloceras echis Jordan & Rothschild
Epitedia wenmanni (Rodischild)
Catallagia decipiens Rothschild
Meringis dipodomys Kohls
Meringis parkeri Jordan
Meringis hubbardi Kohls
Jordanopsylla allredi Traub & Tipton
Stcnistomera alpina (Baker)
Callistopsyllus deuterus Jordan

Megarthroglossus procus Jordan & Rothschild
Anomiopsyllus amphibolus Wagner
Rhadinop.ii/lla heiseri (McCoy)
Rhadinopsylla sectilis Jordan & Rothschild
Carteretta carteri Fox

Ceratophyllidae

Thrassis bacchi (Rothschild)
Thrassis aridis Prince
Dactylopsijlla bluet ( Fox )
Foxella ignota (Baker)
Diamanus montanus (Baker)
Orchopeas sexdentatus ( Rothschild )
Monopsi/Uus wagneri (Baker)
Monopsijllus ettmolpi ( Rothschild )
Malaraeus telchinum (Rothschild)

BlUGHAM YOUNC 1 'Sl\ I llsIM Si II M 1 I.

Malaraeus tinomus \ [ordan
Malaraeus euphorbi i Rothschild

Vmphips> llidae

AiiijihijisiiUii neotomae Fox
Odontopsyllus dentatus I Baker)

1 <eptops) llidae
Peromyscopsylla hesperomys (Bala

[schnopsyllidae
Sycteridopsylla vancouvernensis Wagner

SPECIES PRESENTATION

For each species listed, the following se-
quence in presentation ol information was fol-
lowed: i a I specific .ind subspecific identity and
other pertinent taxonomic data; (1)) ecological
ind geographical distribution with maps, (c)
host association; and i <l | seasonal occurrence
accompanied b) graphs where sufficient popu-
lations made such presentation worthwhile.
Seasonal occurrence was interpreted on the
basis thai a collection of fleas from a single
host constituted an encounter, regardless ol

the number ol fleas taken. For those species
for which data are minimal, the presentations
are given as summary statements without head-
ings.

Some hosts at the test site were collected in
greater numbers than others which were taken
only occasionally during a particular season ol
the year. Although it is most unfortunate that

all hosts at the test site were not collected on a

daily schedule, it was considered not econom-
ically or conservationallv feasible to do so.

Echidnophaga gallinacea
(Westwood ). 1875

Distribution. A total of 56 specimens was

collected at widely separate parts of the site

(Fig. 3). Too few specimens were collected to
indicate a community predominance.

Host Associations. The Black-tailed Jack
Rabbit (Lepus californicus) and the Kit Fox

( Vtilpes maCWtii I were the only animals on
which fleas ol this species were tound. The

fleas wen- about equalh distributed among

those hosts collected.

Seasonal Occurrence. \\\ specimens were
found m Vugust and December.

Comments. Sufficient data are not available
to ai curatel) delimit geographic boundaries Ex
tensive rabbit collections were made over sev-
eral years at various seasons, yet only lour rab-
bits were found infested, and then onlj with

a single Ilea ol this species On each rabbit.

Vccording to Stark i L958) and Wheeler and
Douglas i L945), E. gallinacea has been found
naturall) infected with plague organisms and
demonstrated a high vector efficiency.

Pulex irritans Linne, 1758

Distribution. The greatest numbers of fleas
collected were in the Cravia-Lvcium commun-
ity. This is one ol several communities in the
valleys where the most common host, the Kit

Fox was collected. Geographically, the hosts
and their fleas were widely distributed, espec-
ially along the foothills and in the valleys Fig.
3).

Most Associations. Most of the 73 specimens
obtained were collected from the Kit Fox. Other
hosts were the Black-tailed Jack Babbit, the
Coyote (Cants latrans), and the Western Pipis-
trelle Bat (Pipistrellus hesperus) which repre-
sents an unusual collection. Fleas oi this species
have been taken in most abundance from bur-
rowing animals, which may explain the greatest
number found on the Kit Fox.

Seasonal Occurrence. Apparently fleas ol this
species are not restricted seasonally, but may be

found at any month of the year when their

host is active. Fleas were collected in |.muarv.

February, May, August, and December.

Comments. To better understand the dis-
tribution and seasonal occurrence of /'. irritant
at the lest site, a larger series of predators such
as the Kit Fox, Coyote, and Bob Cat ( Li/nx
rufttS | need to be collected. Their dens also
should be carefully examined. The records
from the bat and rabbit most likely were acci-
dental encounters.

CediopsyUa inaequalis (Baker). 1S95

The male He. is collected indicate that the
subspecies is C. inaequalis interrupta [ordan
( L925).

Distribution. Most of the fleas (45 of 59)

were taken bom hosts from the Pinv on-|uniper

n

Fleas of the Nevada Test Site

N 800,000

GRAPHIC SCALE. FT.

Tonopah
1 50 miles from Mercury

N 650,000

Irig. 3. Geographic distribution of Echidnophaga gallinacea -At Pulex irritans 0 and Cediopsylla inequalis £^

Bingham \m si. University Scienci Hi i i i i in

communit) on Rainiei Mesa Thirteen were
t.tki-M hum the vicinit) "I Cane and Tippipah
Springs which are in mixed vegetational com-
munities. One collection was mad* in
og) in- community i Fig

Host Associations. I In- Rainiei Mesa col
In linns were From Nuttall's Cottontail Rabbil
(Sylvilagus nuttallii), Thirteen specimens were
collected from the Desert < .' < > 1 1 < > 1 1 1 . i i 1 Rabbit
i Sylvilagus audubonii I and one from the \\ hite
tailed Antelopi Squirrel,

Seasonal Occurrence. The greatesl number
ol specimens was From a Nuttall's Cottontail
Rabbit collected in April on Rainier Mesa. All
others were collected in November From the
Desert Cottontail Rabbit at lower elevations.

( omments Extensive collections of mam-
mals at the test site have not been made at
the higher elevations, and collections at lower

i li val s have been limited in some instances.

Vlthough studies ol the abundant mammalian
Fauna in the valleys and Foothills have been
carefully made, collections of cottontail rabbits
living in the vicinity of springs on the desert
have been limited to avoid their elimination
lis trapping and shooting. The vast areas ot
desert highland clothed with sage brush (Arte-
misia tridentata) in the western and northwest-
ern part of the test site, and the extensive f*in-
von juniper woodland to the north and north-
west could well afford a close natural history
scrutin) These undoubtedly provide For an
extended range For rabbits as well as other mam-
mals and their ectoparasites,

Hoplopsy litis anomalus (Baker), 1904

Distribution. Fleas were taken from hosts
in the valleys and foothills, and to a limited ex-
tent on Rainier Mesa. They were most Fre-
quently encountered in Grayia-Lycium, Larrea-
Franseria, and Coleogyne biotic communities.
Nevertheless, they were taken in all ol the plant
communities except Atriplex-Kochia. This is

unusual, lor the most common host, A. leUCtirUS,

was Frequently collected from this latter com-
munit) i Fig 1 I.

Host Associations. ( >iil\ a Few Rock Squirrels
(Spermophilus oariegatus) were taken on Rain-
ier Mesa in the Pin) on- |nniper community, but
.ill possessed lleas. A Roi 11 id tailed Squirrel

Spermophilus tereticaudus) ssas collected in
ssest Frenchman Flat, and two (leas were ob
tamed One Flea was removed From the Little
Pocket Mouse (Perognathus longimembris) , and
eight specimens From several Chisel-toothed

liais Dipodomys microps) Most specimens
260 ol 286) ol // anomalus were obtained
From the White-tailed Antelope Squirrel.

Seasonal ( kcurrcncc. Most Ilea collections

were made in June, xsith lesser encounters in
April, Max, and August. Two collections were
made m September and one in Dea mbei

Comments. Collections ol //. anomalus

show them to OCCU] in |une. ssith a total ab-
sence in fuly, and reoccurrence in August. This

ssas true regardless ol elesation Or host. These
data should not be interpreted .is conclusive,

for there is a difference in the seasonal activity

of the hosts from which Fleas ol this species
have been taken. For example, one should not

Compare the seasonal occurrence for a species

ol Ilia on the Rock Squirrel xsith the- same
species on the White tailed Antelope Squirrel.
While the latter host is active during the- win-
ter in the valleys, the Former max be hiberna-
ting in talus covered In snow at a much higher

elesation. All specimens of S. VariegOtUS xx ere
taken during April. Max', and |unc, xsith the
exception of one collection in August at the
west side ol Frenchman Flat along the loot-
hills.

According to Stark (1958), lleas of this
species in New Mexico have been tound natural-
Is infected xsith plague organisms.

Atyphloceras echis

Jordan and Rothschild, 1915

The subspecies ol our collections is \. echis
echis.

Distribution. Hosts and their parasites were
confined primarily to the Coleogyne commun-
ity. One specimen was taken in a mixed vege-
tational situation and one in a Pinyon-Juniper
community ( Fig. 5).

Host Association. Eleven lleas of this species
were taken from the Desert Wood Rat.

Seasonal Occurrence. Collections wore made
in [anuary and March, ssith most encounters in
December.

Comments. Fleas ot several species are

known to live on A. lepida and in its nests.
The natural history of the Desert Wood Rat
and its parasites at the test site should he cue-
lullx studied, for the' rats have a wide range ol

distribution at the site.

Epitedia wenmanni (Rothschild), L904

Only tsso specimens were collected, a male
and a Female. On the- basis of the male, this

Fleas of the Nevada Test Site

Gold Flat
N 950,000 L !

Kawich Valley

GRAPHIC SCALE, FT.

N 800,000

N 750,000

Tonopah
1 50 miles from Mercury

N 700,000-^

N 650,000

Fig. 4. Geographic distribution of Hoplopsyllus anomalus

BRICHAM VOUNC I'mmiimm Si li. si I Hi ill lis

Geographic distribution ol Atyphloceras <Wn\0 Epitedia loenmanni A and Catallogia decipiens yk

1'i.i \s of the Nevada Test Site

is most likely of the subspecies E. wenmanni
wenmanni.

Distribution. The two fleas were taken at
Tippipah Spring, a mixed type of biotic com-
munity ( Fig. 5 ).

Host Association. Both specimens were col-
letted from a Western Pipistrelle Bat.

Seasonal Occurrence. The fleas were collect-
ed in November.

Comments. Main specimens of Peromyscus
spp. and N. lepida were collected during this
survey. It is unusual that we did not collect
fleas of this species from these hosts. More
unusual is that the only specimens taken were
from a bat. In the original descriptions made
in 1904, the male was taken from a White-
footed Mouse (Peromyscus leucopus) and the
female from a Bushy-tailed Wood Hat ( Neotoma
cinerea). Most published records show fleas of
this species to occur on species of Peromyscus
and Neotoma.

Catallagia decip

Both

ild, 1915

Distribution. Hosts of two different species
were collected, and one flea was collected from
each. One was in a mountainous vicinity, the
other in the Pinyon-Juniper community on Rain-
ier Mesa (Fig. 5).

Host Associations. One host animal, a Pin-
yon Mouse (Peromyscus truei) was from Rain-
ier Mesa; the other, a Desert Wood Bat, was in
the vicinity of Tippipah Spring.

Seasonal Occurrence. The collection from

/'. truei was made in March, and from .V.
lepida in December.

Comments. The few collections of this
species perhaps can be explained on the basis
that fewer hosts have been collected at higher
elevations at the test site compared to the num-
ber made in the valleys and foothills. In other
surveys during the past 20 years and in liter-
ature references, the Deer Mouse has been the
main host. Others mentioned in the literature
are mainly mammals whose habitats art1 at high
elevations or in cool environs. Peromyscus man-
iculatus at the test site is most abundant in the
Pinvon-Juniper community which has not been
extensively surveyed by us.

Meringis dipodomys Kohls, 1938

Distribution. Fleas of this species were the
ones most commonly encountered at the test

site. This is due to the wide distribution and
abundance of their common hosts, kangaroo rats.
These mammals were under continuous study
for several years, primarily in the valleys and
foothills where they are most usually found.
From the standpoint of biotic community dis-
tribution, many collections were from Larrea-
Franseria with about equal numbers from Cole-
ogyne and Grayia-Lycium. The next ranking
community was Salsola, with few collections
from mixed vegetative communities. Two col-
lections were in Pinyon- Juniper, and one was in
Atriplex-Kochia (Fig. 6).

Host Associations. As the specific name of
the flea indicates, the most common hosts are
species of Dipodomys. The Chisel-toothed
Kangaroo Bat is the species on which these
fleas were most often encountered. The next
was Merriam's Kangaroo Bat (Dipodomys mcr-
riami). Only two collections were taken from
the large Desert Kangaroo Bat ( Dipodomijs
deserti ) which is so restricted in its distribution
that not main collections were expected. A
single collection was made from Ord's Kangaroo
Bat (Dipodomys ordii). Other animals from
which fleas were taken, in order of abundance,
are the White-tailed Antelope Squirrel, the
Southern Grasshopper Mouse (Onychomys tor-
ridus), the Desert Wood Bat, the Great Basin
Pocket Mouse (Perognathus parvus), the Deer
Mouse and the Kit Fox.

Seasonal Occurrence. Fleas of this species
were not taken in May, July, or August, and
only one collection was made in June when
three males were taken from D. microps. Most
collections were made in October and Novem-
ber. The common hosts. Dipodomys spp., were
present during these months, and many were
trapped and examined during May, fune, fuly,
and August as well as in other months. Two
collections were made in September, 32 in Oc-
tober, and 86 in November. A relatively high
incidence was maintained throughout the fall,
winter, and early spring until May (Fig. 7).

Comments. Kangaroo rats of several species
and the White-tailed Antelope Squirrel were
studied extensively at the test site. This pro-
vided an opportunity to examine their ectopara-
site fauna on a year-round basis.

The disappearance of adult flea populations
during Ma\-, June, July, and August is an enig-
ma. The host is active during the summer
months, but fleas were not found on those
examined during that period. One might sur-
mise that fleas of this species are sensitive to
the high temperatures during the hot summer

Ill

BnicHAM Younc University Scienci Hi i i i i in

N 950,000

Tonopoh
1 50 miles from Mercury

nzoo.ooo-^^^

N 650,000

GRAPHIC SCAL£, FT

LEGEND
Ar»o Boundary.

Paved Rood

Dirt Rood

Fig, 6 Geographic distribution i>l Meringis dipodomysQ and Meringis parkeri Jf

Fleas of the Nevada Test Site

\

\

c

^

\

\

.

Jan. Feb Mor. Apr May Jun. Jul. Aug. Sep Oct Nov Dec

Fig. 7. Seasonal occurrence of Meringis dipodomys.
The figures on incidence do not represent total
numbers of specimens taken during the month, but
indicate the total number of collections (encoun-
ters) wherein fleas were found.

period and perhaps exist in the nests as an
immature stage — egg, larva, or pupa — until cool-
er temperatures occur.

Meringis parkeri Jordan, 1937

Distribution. All collections were made from
hosts taken on the high mesas or along the
foothills. Collections came from the Coleogvne.
Pinvon-Juniper, and mixed biotic communities,
as well as Artemisia associations. Most encount-
ers were from mixed vegetative situations ( Fig.
6).

Host Associations. With the exception of
one collection from the Deer Mouse, the re-
mainder were about equally encountered from
the Great Basin Pocket Mouse, Merriam's Kanga-
roo Rat, the Chisel-toothed Kangaroo Rat, and
Southern Grasshopper Mouse.

Seasonal Occurrence. The 16 specimens
taken occurred in the same pattern as M. dip-
odomys. Collections were taken in September,
December, March, and April with no encounters
in other months, although the usual hosts were
collected during all months of the year.

Comments. This species was originally de-
scribed from specimens from Dipodomys sp.
taken at Powderville, Montana. From literature
references and twenty years of field studies of
fleas, one may assume that the most likelv host
is Ord's Kangaroo Rat. This is especially true
for the Great Basin Province. Dipodomys ordii
has a scattered distribution at the test site —
northward it is predominant, but southward

it is replaced by D. merriami. Nevertheless, D.
ordii that were collected did not produce a speci-
men of M . parkeri.

Meringis hubbardi Kohls, 1938

Distribution. All specimens were taken at
higher elevations, principally on Rainier Mesa
in the northwestern portion of the test site. All
were in the Pinvon-Juniper biotic community,
except one from the Gold Meadow area north
of Rainier Mesa, a region predominantly veg-
etated by the sage brush, Artemisia tridentata
(Fig. 8).

Host Associations. Three collections were
from the Great Basin Pocket Mouse, two from
the Chisel-toothed Kangaroo Rat. and one from
the Little Pocket Mouse.

Seasonal Occurrence. All collections were
made in September except one in April.

Comments. The lack in numbers and breadth
of distribution of this species may be related to
the higher elevations where its hosts more eom-
monly occur, areas not so extensively surveyed
by us.

Jordanopsylla allredi
Traub and Tipton, 1951

Fleas of this genus and species were de-
scribed by Traub and Tipton ( 1951 ) from fe-
males taken from the Cactus Mouse ( Peromys-
cus fremitus) at Grafton, Washington County,
Utah, the western gateway to Zion National
Park.

In our investigations three males and two
females of ]. allredi were collected.

Distribution. All specimens of fleas were col-
lected from .V. lepida at the southern boundary
of the test site. Hosts were taken from rocky
ledges in a mixed vegetative biotic community
in each instance (Fig. 8).

Host Associations. Although the first col-
lection was from the Cactus Mouse in Southern
Utah, it seems probable that the principal host
will eventually be shown to be the Desert Wood
Rat. We collected the type specimens of this
genus and species and are well acquainted with
the habitats where both the initial and later
collections were made. It is on these observa-
tions that we make the tentative host-preference
designation.

Seasonal Occurrence. Fleas were taken in
October and November.

Comments. Many specimens of N. lepida and
their nests at the test site were examined during

linn .11 \ m l"i\i. I NivEBsm Science Hi i i i i is

Fig. 8. Geographic distribution ol Meringia liuhlumlig^ JordanopsijUa ullndij^ and Stenistomera o/pi

Fleas of the Nevada Test Site

13

the summer of 1965 without obtaining a single
flea of this species. Collections of nests were
made in the same locality where fleas of this
species previously had been collected. The studv
of Howell (1955) revealed a very low popula-
tion of all species of fleas in the nests of N.
lepida in Utah during the summer periods.
The specimens from which the original descrip-
tions were made were collected in December
and November.

Stenistomera alpina (Baker), 1895

Distribution. Fleas of this species have been
considered by some workers as rare in occur-
rence. In our studies at the Nevada Test Site,
they occurred commonly on the Desert Wood
Rat at some seasons of the year. This host is
not restricted to any one biotic community at
the test site, but is found in the Pinyon-Juniper,
Coleogyne, Salsola, Larrea-Franseria, and Gray-
ia-Lvcium communities. About the only restric-
tive influence affecting its distribution is tin'
absence of house-building materials, rocky led-
ges, or large shrubs in which a house may be
constructed with appropriate situations for nest-
ing either in the house or in close association
with it. Most of the fleas were collected from
hosts trapped in the mixed biotic communities
(Fig. 8).

Host Associations. The Desert Wood Hat
is the principal host. Six fleas were taken from
a Canyon Mouse (Peromyscus crinitis), and
three from a White-tailed Antelope Squirrel.

Seasonal Occurrence. Extensive collections
of the Desert Wood Rat were not made con-
tinually throughout any one year. A collecting
schedule set up to include the months of Oc-
tober, November, December, and January re-
sulted in thirty collections in December com-
pared to only four encounters each for January,
October, and November. During the summer
months of 1965 (June, July, August), collection
of N. lepida and its nests was emphasized. Sev-
en fleas of this species were taken from one
nest in June. None was found in 58 other
nests collected during the summer.

Comments. Although the principal host, the
Desert Wood Rat, is widely distributed, there
seems to be some relationship between the pres-
ence of fleas of this species and the seasonal
occurrence of the rat.

Callistopsyllus deuterus Jordan, 1937

The only specimen of this species taken is a
male from a Canyon Mouse. Hubbard (1947:

281) stated: "The two species (of Callistop-
syllus ) come consistently off deer mice, occasion-
ally off other rodents." This has been our ex-
perience in many years of collecting in Utah.
The host, /'. crinitis, has been collected in most
months of the year and at widely-separated
parts of the test site, mainly along the foothills
in mixed communities; yet this is the only flea
of this species we have taken (Fig. 9).

Megarthroglossus procus

Jordan and Rothschild, 1915

Insofar as we can determine, our single spec-
imen belongs to the subspecies M. procus procus.
The male was collected from a Desert Wood
Rat in the vicinity of Tippipah Spring in a
Coleogyne biotic community in November (Fig.
9).

Anomiopsyllus amphibolus

Wagner. 1936

Distribution. This species was taken at wide-
ly-separated points at the test site, but most
frequently at the northwestern part along the
foothills or at higher elevations. The biotic
community association was mainly with Pinyon-
Juniper, Grayia-Lycium, and in mixed vegetative
types ( Fig. 9).

Host Associations. One collection each was
made from a pocket mouse. Deer Mouse, Pinyon
Mouse, and Cactus Mouse. Five separate en-
counters were with the Desert Wood Rat.

Seasonal Occurrence. Two collections were
made in March, one in April, six in October,
and one in December.

Comments. In Utah studies, hundreds of
specimens ot A. amphibolus were found in the
nests of N. lepida, whereas few were taken
from the host's body (Beck et ah, 1953; Howell,
LS55). During June, July, August, and Septem-
ber collections in Utah, nests were relatively
free of fleas of this species compared to other
months of the year. Nests were not examined
during fall, winter, and spring months at the
test site.

The collection of this flea from the host's
body indicates a similarity of occurrence at the
test site and in collections made in Utah. Addi-
tional support to this view are our studies dur-
ing the summer of 1965. More than fifty nests
of N. lepida were collected, but not a single
flea of any species was found. We did take
seven specimens of Stenistomera alpina from
one host, but A. amphibolus was not encoun-
tered.

BniCHAM V.i si. I SIMI1SII-, Si || s. f |,

GotdFlot
N 950,000 . S

N 650,000

Fig. '■>. Geographic distribution ol Callistopsytttu deuterus^^ inomiopsylhis nmphibolusjf Vfegarthroglossus

/inninjfc .mil Rhadinopsylla heiseri 0

Fleas of the Nevada Test Site

15

Rhadinopsylla heiseri (McCoy), 1911

This species has been listed as Actenopthal-
mus heiseri in some publications. We follow the
taxonomic placement by Hopkins and Roths-
child ( 1962) in which Acte'nopthalmus is listed
as a subgenus and the generic status is Rhadi-
nopsylla. According to Jellison ( personal cor-
respondence), there is a close relationship be-
tween heiseri and multidenticulata of Morlan
and Prince (1954). Nevertheless, we feel these
specimens more closely fit the description for
heiseri.

Distribution. Most of the 127 fleas collected
of this species were taken from hosts found in
mixed vegetation communities in Jackass Flats
at the southwestern part of the test site. The
next most often encountered collections were in
Larrea-Franseria communities in the southeast-
ern section. Another community with about
equal encounters is the Coleogyne community
in the northeastern part. All other communities
except the Pinyon-Juniper produced occasional
collections (Fig. 9).

Host Associations. The host on which most
fleas of this species were taken is the White-
tailed Antelope Squirrel. Occasional collections
were also made from the Desert Kangaroo Rat,
Chisel-toothed Kangaroo Rat, Southern Crass-
hopper Mouse, and Southern Pocket Gopher
(Thomomys umbrinus). Onychomys torridus is
a predator and thus may be expected to have
fleas from various rodent victims.

Seasonal Occurrence. Most collections were
in January, with December, November, and
February ranking in relative sequence for other
collections. Only one collection was made in
March, one in April, and one in October. Fleas
were not found in other months of the year.

Comments. The original description of this
species made by McCoy (1911) was of a female
taken at Mojave, California. The host is un-
known for that collection, but in later collections
by several workers, the most common host en-
countered was the White-tailed Antelope Squir-
rel. Hubbard contributed a male and female of
this species which he had collected in Decem-
ber of 1949 at Carson City, Nevada, to the
Rritish Museum in 1950. These were from A.
leucurus. The great majority of our collections
were likewise from this same host. No doubt
the instances of occurrence on hosts other than
A. leucurus are accidental, for several species
of animals live in close association with this
squirrel.

Rhadinopsylla sectilis

Jordan and Rothschild, 192.)

Rhadinopsylla sectilis was listed by Hubbard
(1947) and Holland (1949) as Micropsylla sec-
tilis. Hopkins and Rothschild (1962) listed
Micropsylla as a subgenus of Rhadinospsylla. As

far as we can determine, our specimens are of
the subspecies R. sectilis sectilis.

Distribution. Most hosts were taken in the
Grayia-Lycium community. A few encounters
were from Larrea-Franseria and Coleogyne. The

geographical distribution was generally in Yuc-
ca and Frenchman Flats ( Fig. 10).

Host Associations. The Chisel-toothed Kan-
garoo Rat is the host on which fleas of this
species were most commonly encountered. The
next most common host is the White-tailed An-
telope Squirrel. One specimen was removed
from O. torridus, and three were from Merriam's
Kangaroo Rats.

Seasonal Occurrence. About equal numbers
were collected during January, February, March,
and December. One specimen each was taken
in October and November.

Comments. Apparently fleas of this species
are not abundant <>n a variety of hosts. Collec-
tions ot such hosts as /'. maniculatUS at higher
elevations may produce a greater number of
specimens.

Carteretta carteri Fox, 1927

Dr. William L. Jellison (personal corres-
pondence ) is of the opinion that these fleas are
of the subspecies C. carteri clavata.

Distribution. The geographical and ecolog-
ical distribution is throughout Frenchman and
Yucca Flats. Most encounters were in the Lar-
rea-Franseria community, followed by Coleogyne
and mixed vegetative communities (Fig. 10).

Host Associations. Three collections (total of
seven fleas) were taken from the Long-tailed
Pocket Mouse (Perognathus formosus). One
specimen each was from a Merriam's Kangaroo
Rat, a Canyon Mouse, and a White-tailed Ante-
lope Squirrel.

Seasonal Occurrence. Specimens were taken
in March, April, October and November.

Comments. The occasional appearance of
this flea on Merriam's Kangaroo Rat and the
White-tailed Antelope Squirrel may be accident-
al, for only a few of several hundred of these
hosts that were examined were infested with

BnlCHAM YOUNC L'NIVEBSITl Si ii \. I III I I I MS

GRAPHIC SCALE, FT

LEGEND
Area Boundary

Pov«j Rood

Dirt Rood

Fig 10 I ctpliii distribution "I Rhadinopsijlla sectllisQ find Carteretta carteri^

Fleas of the Nevada Test Site

17

Gold Flat
N 950,000 l

Fig. 11. Geographic distribution of Thrassis hacchi

liim.iiwi YOL'NC I \nnwn Si n \i i Bmiiiis

fleas ul tins spei ies, < >l tin- man) mice examined
during tin' five-veai period, onl) six specimens
a/in clavata wire taken. It is possible
ih.il Hi. is <il tins species occur in low density.

Thrassis hacchi i Rothschild I, 1905

\ total nl 1,491 specimens of tins species
uas collected during the period <•! lliis study.
making it the must abundant flea .it the test
sit.- I sing Stark s L958 publication on i tah
Siphonaptera .is .1 guidi we conclude that the
subspecies is /' hacchi gladiolus

Distribution. The hosts from which the fleas
were taken are principall) valley and foothill
inhabitants. The) were found in all commu-
nities relegated (<> these elevations ( Fig. 11 ).

Most Associations. I In- White-tailed Ante-
lope Squirrel was most often encountered as the
dost. The Southern Grasshopper Mouse was
the next most prominent host, with the Chisel-
toothed Kangaroo Ral about equal in host pref-
erence. Other hosts on which specimens were
infrequentl) taken are Merriam's Kangaroo Rat,
the Desert Kangaroo Rat, Little Pocket Mouse.
Canyon Mouse. Pinyon Mouse, and Coyote.

Seasonal Occurrence. There were about
equal numbers of collections made during most
ol the months ol the \ear except the summer
months ol 1 1 il\ and AugUSl (Fig. 12). There
was an absence ol fleas in July, and only one
specimen was taken in August. White-tailed

Jan F.b Ma

Aug S«p. Oct Nov D«<.

Fig. I- Seasonal occurrence of Thrassis hacchi. The
111411ns on incidence do not represent the total
numbers ol specimens taken during .1 month, hill

11,, li, .a. ili.- total numbci "l 1 ollei '
■1, „!, . hi 1, 11, Fleas were found.

Antelope Squirrels wen- collected during the

simimer hut the) did not possess lleas.

Comments Fleas oJ this species reflect the
general pattern ol seasonal distribution ol other

fleas at the lower elevations ot the test site

where the incidence ol occurrence is greatl)
reduced during the summer months. Burrow
examination and nesting site collections would

be a worthwhile endeavor, lor thev might help
determine what happens to the adult Ilea popu-
lations during the summer months.

Thrassis <ui<li\ I'i nice I1) 1 I

According to Stark's keys and descriptions
I 1j~>s ' . our specimens are T. aridis hoffmani.
This species ranks next to '/'. bacchi gladiolus
in abundance at the (est site.

Distribution. There was widespread distri-
bution at the test site in the biotic communities

of the valleys and foothills. Most of the Heas
came from the Grayia-Lycium, Coleogyne, Lar-
rea-Franseria, and Salsola communities j Fig. 13).

Host Associations. Thrassis aridis hoffmani

was most often encountered on the ('hisel-

toothed Kangaroo Hat. Merriam's Kangaroo Hat

was tin next most frequently infested, with
the White-tailed Antelope Squirrel producing

occasional specimens. One lo live encounters

were made with the Southern Grasshopper
Mouse. Kit Fox, Canyon Mouse. Southern Pocket
Gopher, Long-tailed Pocket Mouse, and Hock
Squirrel.

Seasonal Occurrence. The greatest numbers

were collected in November, followed by Oc-
tober. December, and January. Incidental col-
lections made at other times ol the year were
one each in June and August (Fig. 14).

Comments. The same picture for seasonal
distribution as seen in others is also reflected
m this species. Summer incidence is low or
lacking, with the highest being late fall to mid-

w inter,

DactulojMijlla blut 1 I Fox 1, I'M*)

Foxella and Dactylopsylla are variously used
bv workers in generic designation lor pocket
gopher lleas There is sufficienl difference in
genital anatom) alone to separate them as two
distinct genera, and we follow Prince (1945)

and Holland I' 19 m lliis arrangement. Speci-
mens o| /). IiIiki wen submitted to Dr. W, I..

|< IlisiMi who recommended the subspecific des-
ignation ol n bluri ftsilm

Fleas <>e the Nevada Test She

19

Gold Fiat
N 950,000 I §

N 900,000

Kawich Valley

GRAPHIC SCALE, FT.

N 800,000

Tonopah
1 50 miles from Mercury

N 700,000-^

N 650,000

Fig. 13. Geographic distribution of Thrassis aridis £ FoxeUa ignota aft Dactylopsylla hluei^^ and Diamanus
mdntanus -^.

Hlili.ll \M Yoi'NC I'NIVKIlsm Si II \i I Hi I I I I IN

F«b Mar Apf May Ju

Jul Aug. Sop Oct

Fie. 1 I Seasonal occurrence "I Thrassis aridis. The
figures on incidence <l<> not represent total num-
bers ot specimens taken during tl»- month, but
indicate the total number of collections (encoun-
ters I wherein fleas were found.

Distribution. Most of the specimens were
From gophers trapped in Grayia-Lycium com-
munities. A few were from Coleogyne and some
from mixed Mutational types. One (lea was
collected from an area ol much gopher activity
in the west-central area of the test site which is
at a higher elevation in an Artemisia association
I ig 13 i.

Host Associations. The Southern Pocket
Gopher was the principal host. Collections were
also made trom the Southern Grasshopper
Mouse and liom an unknown species of pocket
mouse.

Seasonal Occurrence. Collections were made

during fanuary, February, March. May. June.
|uh. and September. Most of them were en-
countered in March.

Comments. The principal host lor this flea
is T. umbrinus, although it is occasional!) found

on other mammals living in the same vicinity.

Before a true picture on host relationships, sea-
sonal occurrence, and distribution ma) he seen,
more concentrated surveys will have to he made
in all of the hiotic communities, especiall) in
the western hall and at higher elevations ol the

test site.

Foxella ignota | baker |, isi)~>

()nl\ two hosts were' found infested with
fleas oi tins species \ Deseit Cottontail col-
lected in a 1 .atrea I'Yanseii.i community vicldcd

sixteen fleas. Vn unusual record was thi ex-
traction "I 121 lie. is from a Southern I'ocket

Gophei which was collected from a Pinyon
|uniper communit) I Fig

Collections were made in |anuar\ and Jul\.

/ oxella ignota should he further studied, es-
peciall) in the Pinyon-Juniper communit) and
tin Vrtemisia associations found at higher ele-
vations.

Diamanus montanus > Bakei . 1895

In our surveys in Utah we found the Rock
Squirrel to lie a common host lor this flea.
Stark (1958) likewise reports this host prefer-
ence. Literature records lor other hosts indi-
cate their habitats as foothills and median mon-
tane- elevations. At the test sit. s variegatus
was collected in the Pirn on- Juniper communit)
on Rainier Mis,, i Fig. 13 . In our studies over
the years, especiall) in Utah, the usual cast- was
to find the host heavil) infested, but at the test

site onl) twelve fleas were collected. These

were taken during April. May, and June.

Obviously a more extensive surve) is needed.
especially at higher elevations, to determine the-

extent ol distribution and seasonal occurrence
Of this Ilea.

The Ilea is ot public health importance in
that it has been considered moderately effective
m plague transmission I Stark. L958

Orchopeas sexdentatus (Rothschild), llH>~>

We consider this Ilea to he ol the subspecies

0. sexdentatui agilis. The taxonomic characters
which have been used to distinguish the sub-
species of 0. sexdentatus are variously inter-
preted by different workers. A rather careful
Stud) needs to be made- ol this species over a
wide geographical range. On the basis ol lit-
erature descriptions and accompanying illustra-
tions, our specimens .uc nearer agilii than a
closely related form, nevadensis. Hubbard
i 1957) stated; "During 1938 this flea, }i<t<ulcn-

s/v. was found to be a constant parasite on
Neotoma desertorum (Wood Rat) in Clark
( i ii 1 1 it x Nevada, and Kane County, Utah.

Distribution. Hosts and then lleas were

about equall) distributed in Coleogyne and Lar-
rea-Franseria communities. Other communities
include Pinyon-Juniper, Salsola, Grayia-Lycium,

and several mixed vegetative types i Fig L5).

Host Associations. Four ol 85 separate col-
lections were from the Canyon Mouse. The re-

Fleas of the Nevada Test Site

2!

N 650,000

Fig. 15. Geographic distribution of Orchopeas sexdentatus

I VOUNC I M\ 1 |ISI II Sc II S( I. hi I I I I IN

mainder were taken from the bodies ol Desert
Wood Rats.

Seasonal Occurrence. Twenty-three collec-
tions were made in [anuary, .2* » in November
I l m I Vi embei , 9 in < >< tober and 9 in Man h.
During February, Vpril, fune, and Septembei
one to three em ountei were made in eat li "I
the months.

Comments. Fleas ol O. iexdentatus (no
subspecific designation | have been found to
harbor plague in nature and are considered ef-
ficient vectors ol the disease (Eske) <s Haas,
& L940)

Mm mil; the L965 s ner surveys of wood

rat habitats .it the test site, we were impressed
l>v the similarity "l conditions encountered in
surveys iii Othei parts ol the western United
States. We found situations in which whole
colonies ol wood rats apparently had died.
\ests and houses showed varying degrees ol

n .. lit us.' and dead hudles weir found ill till'

houses. \s stated b) Beck I L955), ". . . the
disappearance ol once abundant rodent popu-
lations has been too consistent to he happen-
stance. These extreme fluctuations in rodent
populations may he due in part to man's inter-
ference with environmental conditions, or per-
haps it is a reflection "I inherent population
rhythm. ... It is possible that the almost, and.
m sonic cases, complete disappearance is due
to a disease agent and the vectors involved."

Monopsy litis wagneri (Baker), 1904

Our specimens arc of the subspecies M. ir</<,'-
iii i i wagneri.

Distribution. All specimens were taken from
hosts located along the foothills or at higher
elevations. The majority were from the Pinyon-
[uniper biotic community i Fig. L6).

Host Associations. In our extensive surveys
in Utah and in sin\c\s In other workers, speci-
mens ol M. wagneri wagneri are considered to

lie the most lrei|ueutK collected fleas. They
have been collected at elevations much higher
than those "I tin desert valleys ami lowlands

at the test site. Most published data list the

Deei Mouse as the preferred host. \t the test
site /'. maniculatus lias been found mainl) at
highei elevations in the 1'invon Juniper com-
munity, and has provided some specimens of
\/ u agneri u >agm ri. additional collet ii«>ns wen
made from the Pinyon Mouse, Southern Grass-
hoppei Mouse and Chisel-toothed Kangaroo
Rat.

Seasonal Occurrence. Specimens were col-
lected about equalh during \pril and Novem-
ber. Two encounters wen m |uly, and on< ea< h

in VugUSl and < >< toller

Comments. Extensive year-round surveys

need to be in. nil m tin Pinyon-Juniper and
\rtcmisia associations in the western and north
Western parts ol the test site.

Monopsyllus eumolpi (Rothschild), L905

Distribution. All specimens were collected
in the north-central part ol tin test site in a
Pirn on-Juniper communit) i Fig. 1 6

Host Associations. This Ilea is commonly
termed the "chipmunk" Ilea. Most collections

at the test site were from the C.'litt Chipmunk
(Eutamias clorsalis). Additional fleas were tak-
en from the Deer Mouse and Great Basin Pocket
Mouse.

Seasonal Occurrence. Most collections were
in April with minor occurrences in |ulv and
\d\ ember.

( lomments. Eighty-three specimens w ere col-
lected. Although this ma) be considered com-
paratively lew. all collections were in the I'm

yon-Juniper comn ity, a biotic situation where

we have done relativel) little surve) work with
the vertebrate fauna

Malaraeus telchinum (Rothschild), 1905

Only two specimens of this species were col-
lected at the test site. One was taken from the
Chisel-toothed kangaroo Hat, and the other from
the Deer Mouse. Both were in a Pinyon-Juni-
per community during October and November

, Fig. 16).

Malaraeus sinomus (Jordan), L925

One ol the principal characteristics used tu

separate M. sinomus from M eremicus is the

comparative length of the first metatarsal seg-
ment. In eremicus it is longer than the com
bined 2nd. 3rd. and 1th tarsal segments, where
as in sinomus it is shorter. The majority of our
specimens lit sinomus, but there are borderline
cases. Nevertheless, the structure ol the ninth
Stemite of the male is unniistakahlv that of M.
Sinomus It ma) be that specimens at the test

sit. arc at the meeting place ol eremicus popu-
lations from the southeast and the sinomus pop-
ulations from the north ami southwest,

Distribution. Must collections at the test site
were made alone the foothills or in basins where

Fleas of the Nevada Test Site

23

Kawich Valley

GRAPHIC SCALE, FT.

LEGEND

Area Boundary

Paved Road

Dirt Road

Fig. 16. Geographic distribution of Monopsyllus wagneri^ff Malaraeus telchinum'f^ Monopsyllus eumolpi^^
and Malaraeus sinomus Q

Illiir.MAM ^ i it ni. I Nivciisrn SriKNin Hi i i i i is

(In elevation is little highei tlian Fren< hman and
Yucca Flats. Vboul e(|tial encounters were madi
in the Larrea-Franseria and Coleogyne com-
munities, | tins occasional collections from Gra)
i.i Lycitim, Pinyon |uniper and mixed vegetative
areas Fig, Ifi

Host Vssociations. Members ol the genus
Peromyscus are considered to he the preferred
hosts. Most n| our collections were from the
Canyon Mouse, Deer Mouse, and Desert Wood
Rat with occasional collections from the White-
tailed Vntelope Squirrel, Southern Grasshopper
Mousi and Pin) i >n Mouse.

Sc.iMni.il Occurrence. ( ollections were made
even month "I the year except August and
September, with most encounters in November,
December, [anuary, and February, successively.
( >i 1 1 \ one nr two encounters wen made during
other months ni the yeai

Comments. Tins flea is associated with hosts
living along the loiitliills.

Malaraem < uphorbi i Rothschild I, 1905

Only one collection ol fleas ol this species
was made. Two males and si\ females were re-
moved from a Deer Mouse in November from
a Pinyon-Junipei communit) on Rainier Mesa
Fig. 17).

Amphipsylla neotomae Fox, 1940

\ male and female specimen were taken
from a Deserl Wood Rat in the eastern end ol
|ackass Flats in a Larrea-Franseria community
I ig 17).
Frank VI. Prince (personal correspondence)
indicated thai \ neotomae should be listed as
Malaraem neotomae. Nothing has been pub-
lished to this elteet to date, so we are obligated
to leave neotomat in its present generic alloca
tion — \mphipsylla.

Odontopsyllm dentatus i Baker 1904

Three fleas ol tins species were collected in
\|in| from Nuttall's Cottontail Rabbit from a
Pinyon-Juniper communit) on Rainier Mesa
17).

Fleas ol tins species, although not found in
great numbers on the preferred hosts of several
species ol / .i fins and Si/h ilagus | jackrabbits and
cottontails, respectivel) I, are considered com-
mon. Man) specimens ol the Black-tailed [ack-
rabbit wen- collected during the period ol tins
study, and large numbers ol cottontails were
also taken and checked for ectoparasites, OnK
one collection ol ( >. tlcutiitus was encountered.

Fleas ol tins species likel) are to l» found on
rabbits at higher elevations and northward into
the (.real Basin region, (.'ollections ol rabbits
al higher elevations at the test site need to be
in. ii li in eorroborati tins point ol \ iew . So far,
inosi collections ol rabbits have been made only

III the vallcN s at the lesl site

Peromt/scopsylla hesperomys

Baku .. 1901

Vccording to ke\s and descriptions given 1>\
|ohnson and Traiib 1954 specimens be-
long to the subspecies P hesperomys adelpha.

Distribution. Most hosts and their fleas were
taken along the foothills or at highei elevations.
The biotic community most commonly repre-
sented was Coleogyne, followed by the mixed
vegetative types, and to a lesser extent the
Larrea-Franseria and Pinyon-Juniper communi-
ties i Fig. 17).

Most Associations. The Southern Grasshop-
per Mouse and C.muiii Mouse were equally en-
countered with flea consortes. The next most
often encountered was the Deer Mouse. Single
encounters were made with the Pinyon Mouse.
Great Basin Pocket Mouse. White-tailed Ante-
lope Squirrel, and Southern Pocket Gopher.

Seasonal Occurrence. January, November,

December, and April were the mouths o| the
year when most encounters were made. Single
encounters were made in February, Max. fury,
September, and October.

Comments. Records ol this species empha-
size the need for more intensive surveys along
the foothills and at higher elevations ol the

test site.

According to Holdenried and Morlan I I1'""'
fleas ol this species have been found naturally
infected with plague organisms.

Nycterhlopsylla t>ancouoerensk
Wagner, 1936

Both Hubbard I L947) and Holland | 1949)

listed this ilea under the generic name. Eptes-

COpsylUl. Following Hopkins and Rothschild

1953), we arc using the generic name. Nycter-

idopsylla.

Five specimens were collected Irom a West-
ern Pipistrelle Bat shot while flying in the vicin-
i(\ ol Tippipah Spring in November. This area
is characterized as a mixed vegetative type "I
biotic communit) ( Fig, 17 ).

The flea fauna ol hats at the test site is
relatively unknown, for few hats have been
collected.

Fleas of the Nevada Te.si Site

25

Gold Flat
N 950,000 L

N 900.000

N 800,000

Tonopah
1 SO miles from Mercury

N 700,000--^

GRAPHIC SCALE, FT.

N 650,000

Fig. 17. Geographic distribution of Malaraeus euphorbi^L Odontopsyllus dcntatus ■ Nt/cteridopsylla (Epte
copsylla) vancouvercnsisJk Amphipsylla neotomae J^. and Peromyscopst/Ila hesperomys A

2<>

Riiii.iivM Young University Science Bulletin

CONCLUSIONS

Naturally the flea fauna is best known from
those host animals which have been most fre-
quently collected. Kangaroo rats and \\ bite
tailed Antelope Squirrels were the animals most
often collected. In those anas oi the test site
where studies were made over several years
a conspicuous reduction in numbers ol fleas
d ig the summer months was observed. Geo-
graphically, the fauna ol Frenchman, Yucca,

and [ackaSS Mats is liest known, and animals
ol the foothills and mesas are least known.

Uthough thirty-three species are listed in
this paper, this does not represent all that arc
expected to occur at the test site. The extensive
stands ol Artemisia tridentata in the western
hall ol the test site arc relatively unsurveyed
biologically, and comparatively little has been
done with the Pinyon-Juniper community on the
mesas and elsewhere. To a certain extent this

applies to the foothill environs. The test site does

not contain high mountain ranges, and one would
not expect to find hosts and their ectoparasites
at the test site characteristic of high mountain
elevations such as the Sheep Range to the east
and the Charleston Mountains to the southeast.

About One-fourth ol the species of fleas re-
ported from the test site ale 1 1 on i the Desert
Wood Hat. Nevertheless, tor the most part these
records are not a result ol vear-roimd collecting
in the various blOtic communities where the rat

is Found. Such a studs would assist in resolving
some of the problems in the taxonomy ol fleas
ol the genus McHaraeus and related groups.

There is a junction of the Great Basin and the

Mojave biota at the test site, and the fleas

characteristic of these two provinces may dem-
onstrate an unusual distribution pattern once it
is known.

LITERATURE CITED

Allied, D. M., D E. Beck and C. D. Jorgensen. 1963b.
Biotic communities of the Nevada Test Site. Brig-
ham Young Univ. Sei. Bull., Biol. Ser„ 2(2): 1-5.1.

Allied, D. M., D E. Beck and C. D. Jorgensen. 1963b.
Nevada list Site stuck aieas and specimen depos-
itories. Brigham Young Univ. Sci. Bull., Biol. Ser.,
2| t):l-15.

Beck I) E. 1955. Distributional studies of parasitic
arthropods in Utah, determined as actual and po-
tential vectors ot Rocky Mountain spotted fevei
and plague, with notes on vector-host relationships.
Brigham Young Univ. Sci. Bull., Biol. Ser., 1(1):
I i.l

Beck D E., A. H. Barnum and L. Moore. 1953.
Arthropod consortes found in the nests of Neotoma
limn a iicrcnu ( Ord ) and Neotoma lepida lepida
(Thomas). Proc. Ut;ih Acad. Sci., Arts, and Letters
30:43-52.

Beck, I) E., 1) M. Allied and E. I'. Brinton. 1963.

Ticks ol til,- Nevada Test Site. Brigham Vomit;

Univ. Sci. Bull., Biol. Ser., 4(1):1-11. '
Eskej C H aui\ V. 11. Haas 1939. Plague in the

western part of tin- United States. Infection in

rodents, experimental transmission of fleas and
inoculation tests for infection. Public Health Re-
ports, 54:1 167-1481.
Eskey, C. R. ami V. H. Haas. 19KI Plague in the

western part ot the United Stales Public Health
Bull.. 2,54:1-83.

Goates, M. A. 1963. Mites on kangaroo rats at the

Nevada Test Site. Brigham Young Univ. Sci. Bull..
Biol Ser., 3(41:1-11. '

Hall. E. It. (Ed 1957 Vernacular nanus „f North

American mammals Inn ol Kansas Museum ol

Natural History. Miscellaneous Publication No. 14:

1-16.

Hayward C. I.. M. L. Killpack and G. I.. Richards,

L963. Birds of die Nevada Test Site. Hre4ha.11

Young Univ. Sci. Bull., Biol. Ser.. 3(11:1-27.

Holdenried, R. and II. B. Morlan. 1955. Plague-
infected fleas from northern New Mexico wild

rodents. Jour. Infect. Dis., 96:1.33-137.

Holland. G. P. 1949. The siphonaptera of Canada.
Science Service. Division of Entomology, Livestock
Insects Laboratory, Pub. SIT. Tech. Bull.. 70:1-306.

Hopkins, G. II. E. and M. Rothschild. 19.53. An ill-
ustrated catalogue of tin- Rothschild collection of
fleas ( Siphonaptera ) in the British Museum ( Na-
tural History). Univ. Press, Cambridge, Vol I.

Hopkins. G. H. E. and M. Rothschild. 1956. An illus-
trated catalogue ol the Rothschild collection of fleas
(Siphonaptera) in the Britisb Museum (Natural

History). Univ. Press, Cambridge, Vol. II.

Hopkins C. H. E. and M, Rothschild. 1962. An
illustrated catalogue ol the Rothschild collection of

fleas (Siphonaptera) in the British Museum Na

tural History). Univ. Press, Cambridge, Vol. III.

Howell, F I 1955. A stuiK ol the- aspcctional var-
iations of Siphonaptera associated with the nests
of the Thomas Wood Rat. Neotoma lepida Upida
Thomas. The Great Basin Nat.. 15: ( 1-4 ) :35-49.

Hubbard. C. A. 1947. Fleas of western North Amer-
ica. Iowa State College Picss. Ames.

Johnson, P. T. and R. Tiaub. 19.54. Revision of the
Ilea genus PewmysCOpsylla. Smithsonian Misc. Col-
lections. 123(4 1 1-68.

[orgensen, C. D. and C. L. Hayward. 1965. Mam-
mals ol tin- Nevada lest Site. Brigham Young
Univ. Sci. Bull., Biol. Ser., 8(3):1-81.

McCo) G. W. 1911 A new flea Ctenophthalmus

heiseri spec. QOV. (Siphonaptera). Entomol. Neves.
22:445-448.

Fleas of the Nevada Test Site

27

Morlan, H. B. and F. M. Prince. 1954. Notes on the
subfamily Rhadinopsyllinae Wagner, 1930 ( Siph-
onaptera, Hystrichops yllidae ) . and description of a
new species. Rhadinopsylla multidenticulatus. Tex-
as Reports on Biol, and Med.. 12:1037-1046.

Prince, F. M. 1945. Descriptions of three new species
of Daetijlopsylhi Jordan and one new subspecies
of Foxella Wagner, with records of other species
in the genera ( Siphonaptera ). Canadian Ento-
mol., 77(1): 15-20.

Stark, H. E. 1958. The Siphonaptera of Utah: Their
taxonomy, distribution, host relations and medical

importance. U. S. Dep. Health, Educ. and Welfare,
Public Health Service, Commun. Dis. Center, At-
lanta, Georgia.

Tanner, W. W. and C. D. Jorgensen. 1963. Reptiles
of the Nevada Test Site. Brigham Young Univ.
Sci, Bull.. Biol. Ser., 3(3):1-31. '

Traub, R. and V. J. Tipton. 1951. Jordanopsylhi allredi
a new genus and species of flea from Utah ( Siph-
onaptera). J. Washington Acad. Sci., 41(8) :264-270.

Wheeler, C. M. and J. R. Douglas. 1945. Sylvatic
plague studies. V. The determination of vector
efficiency. J. Infect. Dis.. 77:1-12.

Brigham Young University
Science Bulletin

COMP, _QOL.
LIBRARY

JUL 28 i966

hARVAKU

UNIVERSITY

ANTS OF THE NEVADA
TEST SITE

by
ARTHUR C. COLE, JR.

BIOLOGICAL SERIES — VOLUME VII, NUMBER 3
JUNE 1966

BRIGHAM YOUNG UNIVERSITY SCIENCE BULLETIN
BIOLOGICAL SERIES

Editor: Dorald M. Allred, Department of Zoology and Entomology,
Brigham Young University, Provo, Utah

Associate Editor: Earl M. Christensen, Department of Botany,
Brigham Young University, Provo, Utah

Members of the Editorial Board:

J. V. Beck, Bacteriology

C. Lynn Hayward, Zoology

W. Derby Laws, Agronomy

Howard C. Stutz, Botany

Wdlmer W. Tanner, Zoology, Chairman of the Board

Stanley Welsh, Botany

Ex officio Members:

Rudcer H. Walker, Dean, College of Biological and Agricultural
Sciences

Ernest L. Olson, Chairman, University Publications

The Brigham Young University Science Bulletin, Biological Series, publishes
acceptable papers, particularly large manuscripts, on all phases of biology.

Separate numbers and back volumes can be purchased from University
Publications, Brigham Young University, Provo, Utah. All remittances should be
made payable to Brigham Young University.

Orders and materials for library exchange should be directed to the Division
of Gifts and Exchange, Brigham Young University Library, Provo, Utah.

Brigham Young University
Science Bulletin

ANTS OF THE NEVADA
TEST SITE

by
ARTHUR C. COLE, JR.

BIOLOGICAL SERIES — VOLUME VII, NUMBER 3
JUNE 1966

5 COMP. 200L..
LIBRARY

JUL 28 1966

UNIVERSITY

TABLE OF CONTENTS

Page

INTRODUCTION 1

Glossary of Special Terms 1

FAMILY FORMICIDAE 1

Myrmica emeryana tahoensis 3

Pogonomyrmex californicus 4

Pogononujrmcx occidenUdis 5

Pogonomyrmex rugosus 5

Pogonomytmex sulinus 6

Pogonomyrmex imherbiculus - - 6

Stenamma smithi 7

Aphaenogaster boulderensis 9

Aphaenogaster megommatus 9

Veromessor luriiersi 12

Veromessor lohognaihus 12

Veromessor pergandei 13

Veromessor smithi 14

Pheidole bicarinata paiute 15

Pheidolc desertorum 15

Pheidole inquilina 15

Pheidole pilifera coloradensis .... 15

Crematogaster coarctata vermiculata 16

Crematogaster depilis 16

Monomorium minimum 16

Solenopsis aurea 17

Solenopsis molesta validiuscula 17

Solenopsis saline 17

Solenopsis xyloni 17

Leptothorax andrei 17

Leptolhorax nevadensis rudis 17

Dorymyrmex bicolor 18

Dorymyrmex pyramicus 18

Liometopum occidentale luctuosum 18

Iridomyrmcx pruinosum emails 18

Camponotus hyatti 19

Camponotus maccooki 19

Camponotus ocreatus 20

Camponotus vicinus 20

Lasius crypticus 20

Lasius sitiens 20

Acanthomyops latipes 20

Myrmecocystus comatus 21

Myrmecocystus lugubris 21

Myrmecocystus mexicanus 21

Myrmecocystus mimicus 22

Myrmecocystus mojave 22

Formica fusca 23

Formica integroides planipilis 23

Formica lasioides 23

Formica limata -4

Formica microgyria -4

Formica moki "4

Formica neogagates 26

Formica neorufibarbis 26

Formica obtusopilosa -"

Formica luhptAila cumponolicvpi y,

\. n inmjrmi v minor j(,

I I I II. \ I I 1:1 ( I I I I) m

LIST OF IK a 1(1 s

Page

I Disturbed Pinyon-Junipci communit) on Rainier Mesa 2

in -.1 i.l Pogonomyre* valifornicm in mixed communit) S

' mi -I gravel mi tl nl Pogonomijrmex occidentals w :in Vrtemisia (.ommunit) 5

I I I ii (»ravel ilisi nl Pagonomyrmex rugosus in the Grayia-Lyeiiim communih 0

"> Lo' (ravel mound ol Pogonomijrmex salinus in a Pinyon-Juniper communih 7
fi Contour, in profile, ol tlioraeii dorsum and nl petioli ind pnstpetinle of Steiuimma smithi,

ii. s|i , I loli itypi 7

ntours, viewed from above, ol petiolar and postpetiolai nodes ol Stcmimmt, smithi.

n. sp. Iiolotypi 8

H Contoiirv m profile view, ol petiole and postpetiole o( a male \phaenogaster megommatui 'i

■ni. mi nl paramerc ol .i male Aplwenogastei megommatui 9

in Volsella ol a mali \pliacnogastei megommatui 9

II. Acdeagus nl .i male Kphaenogaster megommatui 9

12 Abdominal si. nut. I\ i.l .i mali \phaenogaster megommatui 10
' iiiitiiin in profile, ol thoracic dorsum and of petiole and postpetiole "I a female iphaenogastet

mi gommatus 10

II Contours, viewed from above, of petiole and postpetiole ol ,i femali Kphaenogaster megommatus in

II. Mound nl Vvromessoi Uiricen-i in a mixed plant community 12
16. Contotu nl thoracic dorsum and of petiole and postpetiole \ Veromessor smithi;

H \ lariversi, C V. hbognathus. Worker caste I ;

17 Mound nl Veromes&o oergandei in a Larrea community 13

Is Mound nl Veromcssoi smithi in .i Coleogyne community II

19. Mound ol \lyrmecocy.itiii mexicunm in a mixed community 22

20. Thatched mound ol Formica integroides planipilii in a Pinyon-Juniper communit) 21

21 Paramen ol a mali Formica moki 25

22 Volsella nl .i male Fonnica moki 2".
23. Aedeagus nl a male Formica moki 25
2-4. abdominal sternite IX nl a male Formica moki 25

I is i in rAXONOMIC KEYS

Page

Subfamilies ol Formicidac 2

'i ol \l\ nun in. if .!

Spi . i. - ol Pogonomyrnu i I

S| s ill Apliacnogaili i 8

Spei ies nl \ ' i rami vtoi 1 I

nl Pheidole 14

fi igasti i H>

Species nl Solcnopsii l(i

Si ii cies i .1 Li pti 'ili' 'rax 17

nl I >i ilii In idi i in. i' 17

Specii nl / Utrymyrmi \ IS

Genera nl I mum inai 19

Spei ies ol Camponotin 19

S] I / usim 2(1

ii i/tiin 21

irmica 22

ANTS OF THE NEVADA TEST SITE'

bv

Arthur C. Cole, Jr.;

INTRODUCTION

During the summer of 1962 I had the priv-
ilege of being stationed at the Nevada Test
Site, as a part of Brigham Young University's
research team, to make a study of the ants of
the site. Previously I had identified extensive
samples of ant populations that had been taken
from can traps imbedded in the soil. The can
traps served admirably in supplying me with
nearly a complete list of the taxa occupying the
site, and my field collections provided only a
very few forms that had not appeared in the
can samples. Thus, before I went to the site
I had learned not only what taxa were there,
but also which geographic areas and plant com-
munities were involved.

Daily collections in the field provided many
important data about the ants of the site and
disclosed the presence of some taxa that were
either new records for the state or species new
to science. Three new species were represented;
one of these (Aphaenogaster megommatus M. R.
Smith) was described recently (Smith, 1963),
one (Veromessor smithi Cole) was described
subsequent to my studies on location at the site
(Cole, 1963), and the other (Stenamma n. sp. ) is
described in this paper. Males of Formica moki
Wheeler, which were found in a nest with as-
sociated workers, are described herein for the
first time, as are also both sexual castes of
Aphaenogaster megommatus M. R. Smith.

Although ants were represented to some ex-
tent in all plant communities of the test site,
the disturbed Pinyon-Juniper community on
Rainier Mesa supported the greatest number of
taxa. Of the total of 52 taxa discovered at the
test site, 28 were inhabitants of that communitv
and 19 were found only there. Two conditions
probably account for this distribution: (1) the
natural condition of the habitat and plant com-
munity, and (2) the altered situation produced
by the atomic detonations. The undisturbed
Pinyon-Juniper community provides some degree
of shade, a different type of soil texture, more

■B.Y.U.-A.E.C. Publication No. COO-1355-15. This work was suppoiled under reseanli ronlr
tween the U.S. Atomic Energy Commission and Brigham Young University.

-Department of /oology and Entomology. University of Tennessee. Kno.wille.

:lKor glossaries of common ant terminology the reader is referred to publications by Smith (An
Wheelei (Ants ,,\ North Dakota. lnl>3JU5-Rh and Cole (J. Tennessee Acad. Sci.. 24:86-8).

soil moisture, and cooler diel temperatures than
do the other communities of the test site. As a
result of atomic detonations, the natural environ-
ment was considerably altered, apparently fav-
orably so for many ant species. Most of the
trees were killed by the atomic detonations, and
surface rocks were disturbed considerably (Fig.
1). Unless otherwise stated, a disturbed com-
munity is one which has been affected by one
or more atomic detonations.

For excellent accounts of the plant commun-
ities of the test site, the reader is referred to
the publications in this series by Allred, et al.
(1963) and Barnum (1964).

I am indebted especially to Dr. Dorald M.
Allred and Dr. D Elden Beck, project super-
visors, who invited me to participate in the
project and were very helpful in providing ex-
cellent facilities for my use. I should like to
express my appreciation to Merlin Killpack for
his interest in my studies at the site and for his
patience during our joint trips to collecting
stations.

Glossary of Special Terms'

Cephalic index (CI). Head width X 100/head

length.
Epinotum. The first abdominal segment fused

with the thorax; the propodeum.
Eye length (EL). Maximum length of compound

eye.
Eye width (EVV). Maximum width of compound

eye.
Funiculus. Portion of antenna distal to the scape.
Head length (HL). Length of head, in full face

view, from anterior border of lateral clypeal

lobes to the extreme posterior limits.
Head width (HW). Maximum width of head,

excluding the eyes, in full face view.
Ocular index (OI). Eye length X 100/head

length.

,1. contract. ATl I I -1 I -7Nli and AT(11-1)-H55 lie

Midland Nat., 37:626-0), Whe

BfUCHAM YOUNC UNIVERSm Si ii\i i Mi i i i I in

Petiolar node length (PNL Length of onlj the Scape index SI Scape length X l<>0 head

1 11 h li- cil the petiole as measured in profile

or dorsal \ iew
Petiolar node width PNW). Maximum width

of the petiolar nodi m dorsal \ ievt .
Postpetiolar length PPL lot., I length of the

postpi tiole 111 profile or dorsal view.
Scape. Basalmost segment of an antenna,

■ idtli

Scape length si. Length of the scape exclud-
ing the hull) at the base.

Thorax length II. Length of thorax, in pr<>-
File view, from anterior margin of pronotum
(excluding the collar) to the tip of the me-

tastern.il lobe.

!
In;. 1. Disturbed Pinyon-Juniper community on Rainier Mesa.

FAMILY FORMICIDAE
\ Ki \ to the Subfamilies of Formictoae for Identification of the Workers

1. Abdominal pedicel consisting of two segments

Vbdominal pedicel consisting of one segment

2. Cloaca! orifice distinctly circular and surrounded by a fringe of hairs

Myrmicinae

Formicinae

Cloaca! orifice slit-shaped, the hairs, when present, not forming an encircling fringe

Dolichoderinae

hlnn Bull Mm i p /■-■! . mh J". Ill I

Ants of the Nevada Test Site 3

Subfamily Myrmicinae
A Key to the Genera of the Subfamily Myrmicinae for the Identification of the Workers4

1. Postpetiole attached to dorsal surface of first gastric segment, the gaster flattened dorsally
but much more convex vertically, acutely pointed behind Crematogaster

Postpetiole attached to anterior end of first gastric segment, the gaster about equally con-
vex above and below and not notably pointed behind 2

2. Antennae with 10 segments, the last two forming a distinct club Soletwpsis

Antennae with more than 10 segments, the club, if present, with more than two segments ... 3

3. Spurs on middle and hind tibiae distinctly pectinate 4

Spurs on middle and hind tibiae not pectinate 5

4. Thoracic dorsum not impressed between the mesonotum and epinotum; psammophore pres-
ent, but sometimes weak Pogonomyrmex

Thoracic dorsum impressed between the mesonotum and epinotum; psammophore absent
Mijrmwa

5. Epinotum unarmed, the basal face at the same level as the dorsum of the mesonotum

Monomorium

Epinotum armed with spines or teeth 6

6. Workers dimorphic, with the head of the major disproportionally large Pheidole

Workers monomoqihic 7

7. Thoracic dorsum with the mesoepinotal suture absent or very faint, epinotum not depressed
well below level of mesonotum Leptothorax

Thoracic dorsum with the mesoepinotal suture prominent, epinotum depressed well below
level of mesonotum 8

8. Clypeus with a pair of prominent, parallel, longitudinal carinae; eyes minute, set very low
on sides of head Stenamma

Clypeus without a pair of prominent, parallel, longitudinal carinae; eyes notably larger, set
higher on sides of head 9

9. Head quadrate, not notably narrower behind the eyes than in front of them; psammophore
present Veromessor

Head longer than broad, much narrower behind the eyes than in front of them, the occip-
ital corners more broadly rounded; psammophore absent Aphaenogaster

Genus Mi/rmica Latreille

Myrmica emert/ana tahoensis Wheeler and small, and were found under stones in the

more shaded areas. The workers, docile and

Colonies of tahoensis, the only member of sluggish, can be distinguished from their close

its genus known from the test site, were re- relatives elsewhere by the epinotal spines, which

stricted to the Pinyon-Juniper community and are distinctly bent downward, and by the red-

chiefly to that of Rainier Mesa. They were few dish yellow thorax and blackish head and gaster.

4 Bricham Younc University Science Bulletin

Genus Pogonomyrmex Mayr

\ Key to Species dI the Genus Pogonomyrmex lor Identification oi the Workers

1 Mandible with six teeth; eyes placed decidedly below center <>t sides of head; clypeus w it Ji
a prominent tooth-like projection in front Of each antenna] fossa, head, thorax, and petiolar
node strongly rugo-reticulose; inetastern.il flanges strongl) developed and prominently
acute, psammophore weakly developed; small ants, length of largest workers less than 5.0
nun. imberbiculus Wheeler

Mandible with seven teeth; eyes placed .it approximately the center of the sides of the head.

clypeus without prominent tooth-like projections; head, thorax, and petiolar node not
Strongly rugO-reticnlose; mctasternal Manxes less well developed and more rounded; |>sam-
mophore strongly developed; larger ants, length of largest workers notably greater than
5.0 mm. 2

2. Ventral surface of petiolar peduncle with a tew, long, erect, downward-directed hairs, eyes
small, weakly convex, not extending beyond sides of head with head in full-face view, the
head length between the occipital corner and the mandibular insertion more than three
times the greatest eye diameter rugosus Emery

Ventral surface of petiolar peduncle without hairs; exes notably larger, more Strongly com ex.
extending well bevond sides of head with head in hill-lace view, the head length between
the occipital corner and the mandibular insertion distinctly less than three times the great-
est eye diameter 3

3. [nterrugal spaces on head and thorax shining, free of sculpture or with tine punctures that
do not obscure the shining surface; base of antenna! scape weakly dexcloped. epinotum
without armature californicus ( Buckle)

[nterrugal spaces on head and thorax opaque, densely and strongly punctate, the punctures
(especially on the head) giving a headed appearance, base of antenna] scape strongly de-
xcloped; epinotum with a pair of short to long spines I

4. Basal-most mandibular tooth offset, meeting the basal mandibular margin at a pronounced
angle; superior lobe of base of antenna] scape truncate, the extreme base with a distinct
carina extending to the apex of the superior lobe; interrugal punctures of sides of the pro-
notum not tending to obscure the rugae; dorsum of petiolar node irregularly rugose; base
of dorsum of first segment of gaster strongly shining, at most only shagreened

occidentals t Cresson I

Basal-most mandibular tooth not offset, meeting the basal mandibular margin evenly at a
Straight angle; superior lobe of base of antenna] scape broadly rounded, the extreme base
without a carina; interrugal punctures of sides of pronotum tending to obscure the rugae;
dorsum of petiolar node with a number of rugae that are generally transverse and subpar-
allel; base of dorsum of first segment of gaster usually densely and finely punctate, the

surface subopaque or opaque valinus (<)lscn>

Pogonomyrmex (Pogonomyrmex) petiole) black, wholh or in part. Both forms

californicus | Buckley | occur at the test site, often at the same stations

I include, as a portion of tins overall species and "' the same '"'stv
population, the ant which has been known as Colonies were found abundantly in Grayia-

caiifornicus estebanius, lor reasons that have Lycium and Salsola communities. They were less

been stated in my generic revision (Cole, numerous in Larrea-Franseria, and rather sparse

1965b). The latter has been distinguished from in Atriplcx-kochia. mixed, and Coleogyne com-

the typical, concolorous, ferrugineous red cat- munities. The nest of californicus is surmounted

ifornicus by its bicolored body — the head and by a circular or semicircular crater of loose sand

thorax being ferrugineous red and the gaster bearing a single, centra] entrance t fig. 2). The

(as well as generall) the petiole and post worker max be distinguished Irom tli.it of all
■Adapted Irom Cnl« M'Hrthi Validation ol ihp nan I ihi i nod lulh in the revision

Ants of the Nevada Test Site

Fig. 2. Crater nest of Pogonomi/rmcx californicus in
mixed community.

other species of Pogonomyrmex at the test site
by a lack of epinotal armature. The sexual
castes were found in nests in early and middle
July.

Pogonomyrmex ( Pogonomyi mix )

occidentals ( Cresson )

Colonies of this species are characteristic of
sagebrush (Artemisia) areas below the zone of

Pinyon-Juniper. Thev contact the range of P. sali-
nus in Pinyon-Juniper, but do not enter it. The
characteristic gravel cones, or domes, surrounded
by cleared areas and with basal entrances, pre-
dominate the landscape (Fig. 3).

The worker, which bears a close superficial
resemblance to that of salinus, has the basal-
most mandibular tooth offset from the margin,
the superior lobe of the antenna! scape base
truncate, the petiolar and postpetiolar nodes
without transverse rugae, and the dorsum of the
base of the first gastric segment shining and
without punctures. Far from being docile, the
workers attack the intruder vigorously when a
nest is disturbed and can inflict very painful
stings.

Pogonomyrmex ( Pogonomyrmex )
rugosus Emery

This is the taxon that, together with certain
of its variants, has been known as barbatus
rugosus and which I have elevated to full species
status in my unpublished revision of the genus
Pogonomt/imcx in North America (Cole, 1965b).
Although colonies were most numerous in the
Grayia-Lycium community, thev occurred also

£*fv

Fig. 3. Conical gravel mound of Vogonomyrmex occidentalis in an Artemisia community. Ruler is one foot in
length.

Bricham Younc UKivERsm Science Bulletin

in Vtriplex Kochia, < oleogyne, Larrea Franseria,
SaJsoIa, .mil mixed conn lities. The nest su-
perstructure consists oi aven low gravel m id

or disc w i 1 1 1 a large irregular central entrance
Fig. I I. The workers, which defend their nests
with pugnacity, are the largest in the genus .is
represented at the test site. They are black,
reddish black, or reddish brown in color, and
can be distinguished from all other Pogonotnyr-
met oi the site 1>\ tin- presence <it a few erect
hairs on the venter oi the petiolar pedum le

Pogonomyrmex [Pogonomyrmex
talinus Olsen

This species ol liai\ ester is one of the domi-
nant ants ol the Pinyon-Juniper community. In
fact, it is prettv much ot an "indicator species,"
for it is restricted to that communiU and there-
in replaces P. OCCidentalis at higher elevations.

The nest is constructed in an open area between
shrubs and is surrounded by a rather flat gravel
bed winch hears one or more saucer-like de-
pressions with a central entrance (Fig. 5). It is
surrounded bv an area which has been cleared

The worker, which is rather closely allied to
/' occidentals is characterized by its mandible.
the basal-most tooth of which is not offset from
tin mandibular margin, the convex superior lobe
ot the base o| the .uiteiiii.il si ape, the generally
transversel) rugose dorsum ot the petiole and
postpetiole. and the usually strongly punctate

anil subopaque dorsal base o| tin- first i^astric

segment. Unlike those oi occidentals, the work-
ers are very docile ants and retreat to cover
rather than attacking.

Pogonomyrmex \ Ephebomyrmex)

itnberbiculm W heeler

The diminutive, sluggish workers of this
species form small, obscure colonic^ beneath
stones. At the test site they were found in the
Grayia-Lycium community. The only member of
its subgenus known from the site, imberbiculus
is distinctive in its small size (length 4-5 mm.),
the pair of projections in front of the antenna!
fossae, the rugo-rcticiilose sculpture of head and
thorax, and the very thick postpetiole. It is to-
tally inoffensive in its behavior and retreats
readily when disturbed.

>-*

"•*->•

*«4

I l.ii gravel disc ol Pogonomyrmex rugosus in the Grayia-Lycium community. Rulei is one loot in length.

Ants of the Nevada Test Site

-

Fig. 5. Low gravel mound of Pogonomi/rmex salinus in a Pinyon-Juniper community. Note the two
depressions and their central entrances. Ruler is one foot in length.

Genus Stenamma Westwood

Stenamma smithi, new species

Holotype, worker. HL 1.02 mm, HVV 0.85
mm, CI 83.3, SL 0.08 mm, SI 94.1, EL 0.22
mm, EW 0.15 mm, OI 21.5, WL 1.22 mm, PNL
0.17 mm, PNW 0.22 mm, PPL 0.29 mm, PPW
0.31 mm.

Funicular segments two through six distinct-
ly broader than long, last segment only slightly
longer than combined lengths of the two pre-
ceding segments. Eve large, oval, with 12 om-
matidia across its greatest diameter which is
nearly equivalent to distance from lower eye
margin to the mandibular insertion.

Contour of thoracic dorsum and of petiole
and postpetiole as shown in Fig. 6; mesoepi-
notal impression pronounced, its greatest length
0.22 mm, its greatest depth 0.05 mm, bearing
a distinct, transverse, median welt; declivity of
epinotum long and steep; epinotal spines short
but pronounced, acute apically and broad basal-

ly, directed rather strongly upward; petiolar
peduncle rather short, its venter with a small,
distinct, blunt process. Contour of petiolar and
postpetiolar nodes, viewed from above, as shown
in Fig. 7; petiolar node with a broad, blunt
apex; postpetiolar node subspherical.

Sculpturing of head pronounced, the rugae
rather coarse and wavy, especially laterally
where they tend to form weak reticulations;
front with subparallel, longitudinal striae; inter-

Fig. 6. Contour, in profile, of thoracic dorsum and of
petiole and postpetiole of Stenamma smithi, n. sp.
holotype.

Bnn.li wt Young Universiti Scienci Bulletin

Fig, 7 Contours, viewed from above, of petiolar and
postpt'tiol.ir nodes "1 Stenatnma smithi, n. sp., holo-
type

nr4.1l spans densely and finely punctate and
subopaque. Pronotal rugae longitudinal, chiefly
moderately coarse, somewhat wavy, widely and
unevenly spaced, those of anterior border finer
and transverse; interspaces shining, very finch'
and sparsely punctate. Base and sides of epino-
tum moderately rugose, the rugae longitudinal
and unevenly spaced, the interspaces subopaque.
Base of gaster with prominent, short, longitu-
dinal rugulae with a maximum length of 0.086
mm. Gaster smooth, strongly shining.

Body with abundant, suberect to erect, gold-
en hairs, those on the head notably shorter and
denser than elsewhere. Hairs on legs mostly
subappressed and reclinate, except on femora
where they are distinctly suberect and erect.

Antennae, clypeus, genae, mandibles, thorax.
legs, petiole, and postpetiole uniformly, medium
ferrugineous red; frons, vertex, and occiput
deeply and uniformly infuscated; dorsum of
gaster a very deep, uniform, blackish brown.

Type locality. The holotvpe and 11 paratype

workers were extracted from can samples taken
on January 3, 5, 8, and 10, 1962, in B.Y.U. Study
Area ECH of die test site. The nest was not

found. I am pleased to name tins interesting

hew species tor Dr. \1 H Smith, Ills gOOG

friend and colleague, who recently revised the
genus Stenamma in North America north of
Mexico (Smith, 1957). I had sent the specimens

to Dr. Smith for Ins inspection, and whin he
informed me that they appeared to represent

a new species, I invited him to describe and

name the population. However, inasmuch as he
retired from his position at the ('. S. National
Museum before he could accomplish this. Dr.
Smith kindlv arranged tor the return of the
specimens ti > me.

Variation in paratype series. Variation among
the spec uncus is remarkably slight. In a few
of the workers the epinotal spines deviate some-
what in shape and size from those of the liolo-
type. In some they are a little longer and less
robust; in others they are a little shorter and
more angular. The pronotal rugae in three
specimens are rather annate and somewhat
coarser. The number of ommatidia across the
maximum diameter of the eye varies from 10 to
13. The pronotal width yaries from 0.53 to
0.59 mm. Variations in other parts are as fol-
lows: HL 0.88 to 0.97 mm, H\V 0.75 to 0.S0 mm,
CI 82.5 to 87.5, SL 0.73 to 0.77 mm. SI 93.7
to 96.3, EL 0.19 to 0.22 mm, EW 0.12 to 0.14
mm, OI 20.0 to 21.6, TL 1.07 to 1.09 mm,
PXL 0.0S to 0.12 mm. PNW 0.17 to 0.21 mm.
PPL 0.22 to 0.26 mm, PPW 0.27 to 0.31 mm.

Location of types. The holotvpe and one
paratype will be deposited in the U. S. National
Museum; other paratvpes will be deposited in
the Museum of Comparative Zoology and in the
author's collection.

Affinities. In Smith's (1957, p. 112) revision
of the genus the species keys to brevicorne
(Mayr), an eastern species which ranges west-
ward to Nebraska and Minnesota. It differs
markedly from that species, however, especially
in its greater body size and larger eves, the
differing dorsal contour of the thorax, the long
declivity ol the epinotum, ami the conformation
of the petiolar and postpctiolar nodes.

Genus Aphaenogaster Mayr

A Key to Species of the Genus Aphaenogaster for Identification of the Workers

1 I've very large, strongly protuberant, with IS to 20 facets in its greatest diameter, its
lower border approximately the greatest ocular diameter from the mandibular articulation.
body a sordid, pale, yellowish tan megommatus M. R. Smith

Eye notably smaller, with about 15 I. nets in its greatest diameter, its lower border about
l!j times the greatest ocular diameter from the mandibular articulation; body terrugineous

red boulderensis M. R. Smith

Ants of the Nevada Test Site

Aphaenogaster boulderensis
M. R. Smith

This species was first described from workers
collected on Horseshoe Island, Lake Mead,
Boulder Dam (now Hoover Dam), Nevada. The
distinctive worker is rather small ( length 4.5 to
5.5 mm ) , very slender, and with stronglv round-
ed posterior corners of the head and a rounded
occipital border. The epinotal armature con-
sists of faint tubercles. The bodv is a rather
uniform, pale, fernigineous red, with the gaster
sometimes slightly darker.

No nest was found at the test site. Two
workers were taken from soil cans in mixed
vegetation in October. The ants probablv nest
beneath stones and form small colonies. Ap-
parently boulderensis is poorly represented at
the test site. It mav indeed be a rare ant any-
where.

Aphaenogaster megommatus
M. R. Smith

This species was described recently (Smith,
1963) from workers collected in Nevada, Ari-
zona, California, and Oregon. Although no nest
was found at the test site, the ants rather
commonly appeared in can traps from areas of
Larrea, Larrea-Franseria, and Grayia-Lycium.

A. megommatus is a crepuscular and noc-
turnal forager. The worker is readily recogniz-
able by its pale color, extraordinarily large,
black eyes, and long, slender antennal scapes.
Bodv length varies from 4.0 to 5.5 mm.

The previously undescribed sexual castes
are described as follows:

Male. HL 0.71 to 0.88 mm, HW 0.63 to 0.70
mm, CI 112.7 to 125.7, SL 0.31 to 0.34 mm, SI
48.5 to 48.9, EL 0.48 to 0.68 mm, EW 0.37 to
0.43 mm, OI 60.1 to 77.3, TL 1.99 to 2.14 mm,
PNL 0.14 to 0.19 mm, PNW 0.22 to 0.24 mm,
PPL 0.26 to 0.31 mm, PPW 0.34 to 0.41 mm.

Head distinctly longer than broad; eyes very
large; maximum width of head across the eyes
0.82 to 0.88 mm; antennal scape short, its
length less than the combined lengths of first
two funicular segments; anterior border of me-
dian lobe of clypeus with a shallow but distinct
notch. Epinotal armature consisting of a pair
of short, blunt, tubercles; petiolar and post-
petiolar contours as illustrated in Fig. 8; petiolar
node not pronounced, its apex broadly convex.
Contour of paramere of genitalia as shown in
Fig. 9, of volsella as in Fig. 10, and of aedeagus
as in Fig. 11; abdominal sternite IX as shown
in Fig. 12.

Fig. 8. Contours, in profile view, of petiole and post-
petiole of a male Aphaenogaster megommatus.

Fig. 9. Contour of paramere of a male Aphaenogaster
megommatus.

Fig.

10. Volsella
matus.

of a male Aphaenogaster megom-

Fig.

11. Aedeagus of a male Aphaenogaster megom-
matus.

Ill

Uhk.ii km Young University Sc n \< i Hi 1 1 1 1 1\

Fig. 1^. Abdominal sternite IX oi a male Apluicnogus-
tet megommatus.

Head very densely and finely punctate, sub-
opaque; thorax faintly and finelv punctate, the
punctures extremely delicate, the integument
translucent and somewhat shining; gaster prom-
inently, densely, and finely shagreened, the sur-
face highly polished.

Hairs on head, thorax, petiole, and postpe-
tiole sparse, rather long, erect and suberect,
ul unequal length, pointed, slender, and golden.
especially delicate on scutum; a few suberect
hairs on occipital corners of head: numerous on
gular region; on the thorax most abundant on
tuberculate areas ol epinotum; unusually long
and sparse on petiolar and postpetiolar nodes;
rather long, scattered, delicate, pointed, erect,
and suberect on gaster.

Thorax, appendages, petiole, |x>stpetiolc.
gaster, and anterior half of head a rather uni-
form, dusky yellow; posterior half of head
strongly infuscated, especially deeply within
the ocellar triangle: mandibular teeth deep red-
dish brown; eves black.

Female. III. 1 .37 to L.52 mm, HW 1.25 to
1 29 mm, CI 82.2 to 91.5, SL 1.56 to 1.60 mm,
SI 124.0 to 124.8, EL 0.49 to 0.53 mm, EW
0.38 to 0.46 mm, ()l 34.9 to 37.6, TL 2.67 to
2.81 mm, PXL 0.27 to 0.34 mm, PNW 0.38 to
0.42 mm, PPL 0.38 to 0.46 mm, PPW 0.53 to
0.65 mm.

I had notably longer than wide, occipital
corners rather broadly and evenly rounded; ex-
treme occipital margin with a low, narrow,
transverse carina; eyes subovute. very large,
their greatest diameter approximately one-third
more than distance from lower eye margin to
mandibular articulation, with 20 to 23 facets
in their greatest diameter; maximum head width

across the exes 111 to 1.52 mm. antenna! scapes

Fig. 13. Contour, in profile, of thoracic dorsum and of
petiole mil postpetiole ol .i female Aphaenogaster
megommatus.

long, in repose extending well beyond posterior

corners ol head.

Contours ol thorax, petiole, and postpetiole
as shown in Fig. 13; epinotal declivity short and
very steep; epinotal spines robust, broad, finger-
like, bluntly rounded apicallw flattened lateral-
ly, directed strongly backxvard; petiolar peduncle
with a rather well-developed, longitudinal keel.
petiolar node broadly and evenly rounded
apicallv; postpetiole with a small, prominent,
sharp, anterovcntral. downwardly directed spine,
the node with a long anterior declivity and a
short posterior declivity. Contours of petiolar
and postpetiolar nodes, viewed from above, as
shown in Fig. 14; apex of petiolar node broadly
truncate, notably narrower than the base; post-
petiolar node considerably broader than long,
the sides rather strongly convex.

Head prominently, unevenly, longitudinally
rugulose, the interspaces somewhat shining and
with dense and xerx delicate punctures; mandi-
bles finely, longitudinally striate, densely and
finch punctulate, and subopaque; frontal area

Fig. 14. Contours, viewed from above, of petiole and
postpetiole1 of a f( male Aphaenogaster megom-

Ants of the Nevada Test Site 11

with three or four longitudinal carinae. Pro- Body and appendages mostly a rather uni-

notum, scutum, and scutellum mostly smooth form, sordid light brown; scutum sometimes not-

and strongly shining; epinotum transversely, un- ably infuscated; eves jet black; mandibular teeth

evenly rugulose, the interspaces finely and deep reddish brown.

densely punctulate and subopaque; petiolar and These descriptions are basea Upon the fol-

postpetiolar nodes subopaque, finely and densely ,t)wing specimens: Last Chance Canvon, El Paso

punctulate, and with sparse, irregular rugulae. Mts Kem Co CaHf Juj 10 1964 R R Snel.

Gaster moderately shining, densely and finely lmg leg (6rf> 4$ ). 7 miies S. of Benton, Mono

shagreened. C() Cahf August 9 1962, R. R. Snelling leg

Hairs yellow, moderately abundant, and scat- (2o* , 4 9 ); Las Vegas, Nev., July 22, 1963, col-

tered; shorter, less delicate, and more blunt on lector unknown (lcf); Nevada Test Site, can

scutum than on head and scutellum; sparse, traps, July 28, 1961 (1 nest queen), July 1, 1962

rather long, and robust on epinotum, petiole, (lcf), August 12, 1962 (1 nest queen), August 19,

and postpetiole. 1963 (1 2 ), August 25, 1963 (1 nest queen).

Genus Veromessor Forel
A Key to Species of the Genus Veromessor for Identification of the Workers"

1. Middle of anterior border of clypeus with a prominent projection; eye with a distinct an-
teroventral angle; color piceous brown to black; strongly polymorphic; forage in files

pergandei Emery

Middle of anterior border of clypeus without a projection; eye without a distinct antero-
ventral angle; color pale brownish yellow to deep ferrugineous brown; not strongly poly-
morphic; forage singly 2

2. Head strongly, longitudinally, and rather evenly rugose, the rugae extending into the oc-
cipital region; eye small (OI 22), the distance from its posterior margin to the occipital
margin greater than L2 times its maximum diameter; epinotal spines very long, about I'A
times the distance between their bases; nests generally beneath stones in compact gravel-
ly soil; diurnal; length 6.29 to 6.87 mm; thoracic profile as shown in Fig. 16C; base of an-
tennal scape flattened, broader than apex; color a rather uniform, deep, ferrugineous brown

lobognathns (Andrews )

Head not strongly rugose, the rugae not extending into the occipital region; eye large (OI
30 to 31), the distance from its posterior margin to the occipital margin no greater than or
notably less than I'A times its maximum diameter; epinotal spines shorter, no longer than the
distance between their bases; nests in open and surmounted by craters; crepuscular or noc-
turnal; length 3.61 to 5.86 mm 3

3. Occipital region of head shining and faintly shagreened; rugae around antennal fossae con-
centric; maximum diameter of eye subequal to distance from posterior margin of eye to
occipital margin; SI 87; base of scape narrower than apex and not flattened, the dilation
reduced, involving only the extreme base, and meeting the shaft at strong angles; thorac-
ic contour as shown in Fig. 16B; epinotal spines notably shorter than distance between then-
bases; apex of petiolar node acute; head, thorax, petiole, and postpetiole brownish yellow;
gaster darker; length 3.61 to 5.86 mm; nocturnal lariversi M. R. Smith

Occipital region of head subopaque, densely and finely punctate; rugae around antennal fos-
sae not concentric; distance from posterior margin of eye to occipital margin of head about
1 1/6 times maximum diameter of eye; SI 96; base of scape flattened, broader than apex,
the dilation involving more of the shaft and meeting it at weaker angles; thoracic contour
as shown in Fig. 16A; length of epinotal spines and their interbasal distance subequal; apex
of petiolar node broadly convex; head, thorax, petiole, and postpetiole medium reddish
brown, gaster slightly darker; length 5.70 to 5.86 mm; crepuscular smithi Cole

"Adapted from Cole (10(,f, pp. (.SI-

12

Bricham Young University Si umi Hi uhin

\ . romessor /<"" ersi \l R. Smitli

Although generall) sympatric with V. smithi
and occupying the same stations, lariversi was
relatively uncommon. Most colonies were in
sand) areas with mixed vegetation. Nests oc-
curred verj sparsel) . however, in Grayia-Lycium
Salsola, and Coleogyne communities. Each nesl
\\ as surmounted 1>\ one or two circular craters
d| sand, three inches or Irss in diameter, and
marked bj a rather large, irregular entrance
(Fig. 15). Maximum depth ol a nesl was two
feet, well above the hardpan layer. The work-
ers are nocturnal foragers, as their pale color
and large cms would seem to indicate. \\ inged
males and females, few in number, were in
some <>t the nests between |ul\ 1 and 27.

WOrkcrs ol lariversi can be distinguished
casiK from other species ol Veromessor at the

test site by tlicir brownish yellow head, thorax.

petiole, postpetiole, and tlicir rather contrasting-
ly darker gaster; and by the shining occipital
region ol the head, the large eyes, the very short
epinotal spines, and the acute apex ol the petio-

lar node. The dorsal contours of the thorax.

petiole, and postpetiole arc shown in Fig L6B
I'm furthei information on this species the read-
er is referred to papers bj Smith | L951 I and

Col.- ( L955, 1963).

Veromessor lobognathus (Andrews)

Probably more colonics o| this species have
been found .it the test site than at all other lo-
calities combined, \lthough lobognathus had
been considered to he a rare ant. it has become

evident that it can he a common occupant oi
certain stations such as that on Rainier \hs.i
in disturbed Pinyon-Juniper. Firs! described
trom Colorado (Andrews. H)lf>i. it was redis-
covered there by Gregg (1955) and near Ely,
Nevada, !>v Cole (Gregg, lor. cit.), and subse-
quently in North Dakota and South Dakota hv
Wheeler and Wheeler ( L956, 1959, 1965). At the

test site the nests were largclv confined to
Rainier Mesa and were restricted to I'invon
Juniper areas where they were sympatric with
Pogonomyrmex saliwis Olsen — a species whose
workers rcmarkahlv resemble superficially those
ol lobognathus. On the mesa I excavated and

^i — s-vv

mm** ..*

■ .. ■

Fig L5 Mi

Ants of the Nevada Test Site

13

studied 47 nests of lobognathus and located 11
others which were not disturbed. The colonies
were very populous and lived beneath large
rocks, some of which were deeplv and strongly
embedded in the soil and banked peripherally
with a light coating of gravel. One nest opened
into a gravel crater beside the covering rock.

The workers moved rapidlv and agilely when
a nest was opened. They did not attack the in-
truder. Distinct and perpetual stridulation was
heard. During late June and early July numer-
ous males and females were in the nests. V.
lobognathus seems to be one of the most abun-
dant and successful components of the ant fauna
on Rainier Mesa.

The worker can be identified easily by its
strong, superficial resemblance to Pogonomyr-
mex occidentalis and salinus; its deep, ferrugine-
ous brown color; its small eyes; and its very long
epinotal spines. The contours of the thorax,
petiole, and postpetiole, in profile, are illustrated
in Fig. 16C.

Veromessor pergandei Emery

Probably by far the glossiest black ant of
the low desert, pergandei is the dominant species

of the Larrea community where it is able to
flourish under extremelv xeric conditions. It is
a diurnal forager which can remain active dur-
ing periods of intense heat. The long trails of
streaming, black workers represent a character-
istic pattern of extranidal activity. The nests,
which are in exposed soil, are surmounted bv
low, semicircular or circular craters (sometimes
multiple ) of sand ( Fig. 17 ) and are often cov-

Fig. 16. Contour of thoracic dorsum and of petiole and
postpetiole. A. Veromessor smithi; B. V. lariversi;

"V>3

fc!

. ^>

•vf^v>

Fig. 17. Mound of Veromessor pergandei in a Larrea community. Ruler is one foot in length.

1 1

Bum. 1 1 \\i ^01 \<. I Mm iiMii Sc 11 \i i Hi ii i 1 1\

ircd with a dense layer of (halt from harvested
seeds. The workers .ue strongl) polymorphic,
This characteristic, tin- concolorous black body,
.Hid the presence ol a prominent projection on
tin- middle oi (Ik- anterior border ol the clypeus
will ser\e to separate the species from all others

in tin- minis .it the test site .mil elsewhere.

Veromessor smithi Cole
Tins species was described from numerous

Specimens Oi .ill castes collected at the test site
i Cole. L963). Most nests were in the ColeOgt/m
area at the loot ol Rainier Mesa, where they
were numerous, hut one nest was observed ill

open, sandy desert with mixed vegetation. They
mi lined m fully exposed areas between plants,

.mil each was marked l>\ a small, circular (rater

of soil about five inches in diameter Fig. 18).

The timid sluggish workers appear to he cre-
puscular foragers. They can he recognized by
the medium, reddish brown body color, the large
eyes, the punctate and subopaque occiput, the
epinotal spines ol medium length, and the dis-
tinctive thoracic contour and broadly subtrun-

cate petiolar node i Fig. 16A .

\\ inged castes appeared abundantly in the
nests between Julv I and 27 Mating flights
were not observed.

Fig. 18. Mound of Veromessor smithi in a Coleogyne community. Ruler is sL\ inches in length.

Genus Pheidole Westwood

A Key to Species of the Genus Pheidole for Identification of the Workers

1. Antenna] scape ol major suqwssing the occipital corner; large ants, length of major 4.5 to
5.0 mm, of minor 3.0 to 3.5 mm desertorum Wheeler

Vntennal scape ol major not reaching the occipital corner; smaller ants, length of major less

than -1.5 mm. ol minor less than 2.5 mm 2

2. Major with transverse occipital striae which are confined chiefly to the top of the occiput

pilifcrn coloradensis Emery

Ants of the Nevada Test Site

15

Occiput of major without transverse striae or rugae 3

3. Epinotum of the minor bearing a pair of thick, bluntly rounded spines of approximately
the same diameter throughout; promesonotal suture of the major and minor distinct; in-
quilines in nests of P. pilifera coloradensis inquilina (Wheeler)

Epinotal spines of the minor sharp, tapered, and more slender; promesonotal suture of the
major and minor faint or absent; not inquilines 4

4. Anterior clypeal margin of the major sinuate; pronotum smooth and shining; basal face of
epinotum somewhat shining bicarinata paiute Gregg

Anterior clypeal margin of the major bluntly bidentate; pronotum punctate and usually with
transverse striae or rugae; basal face of epinotum opaque bicarinata oinelandica Forel7

Pheidole bicarinata paiute Gregg

The types of this recently described sub-
species (Gregg, 1958) came from Goldfield,
Nevada, at an elevation of 5,800 feet. Colonies
were numerous at the test site in areas of
Grayia-Lycium, Larrea-Franseria, Coleogyne,
and Atriplex-Kochia. The nests were beneath
stones. Majors were numerous in a colony, and
the ratio between them and the minors seems to
be unusually great for a Pheidole. The docile,
small, brown majors, with a body length of 4.0
to 4.5 mm, are readily distinguishable by the
broadly sinuate anterior margin of the clypeus,
the apunctate and strongly shining pronotum,
and the strong rugae enclosing laterally the
rather shining but distinctly punctate basal sur-
face of the epinotum.

Pheidole desertorum Wheeler

A common occupant of our southwestern
deserts, deseiiorum was most prevalent at the
test site in Grayia-Lycium and mixed plant
communities. It occurred to a decidely more
limited extent in Coleogyne, Larrea-Franseria,
and Salsola areas. Nests were beneath the larg-
er stones. Colonies were generally large and
the occupants very active and pugnacious. The
majors, which attain a body length of about
4.5 to 5.0 mm, can be identified easily by their
long antennal scapes and by their reddish yel-
low color.

Pheidole inquilina (Wheeler)

At the test site this especially interesting
species is an inquiline in nests of Pheidole
pilifera coloradensis Emery. Described as the
type species of the monotypic genus Epipheidole
by Wheeler (1903, p. 664), this species was re-
cently transferred to the genus Pheidole by Cole

'Althnugh this ant has not been
will probably be found eventually

aken at the test site, I ha

(1965a) who, at the test site, discovered the
unknown soldier caste and rediscovered the
worker, and subsequently showed that they are
representative of Pheidole, not a genus of their
own.

Nests of the host ants, which were confined
to disturbed Pinyon-Juniper on Rainier Mesa,
were under loose stones. A total of only one
soldier and two workers of inquilina was found,
indicating the rarity ot these castes, but several
alate males and both alate and dealate females
of the inquiline were in the nests. The sexes,
as well as the small workers, can be recognized
by their finger-shaped epinotal spines. The
soldier (major worker) shows close affinities
to that of coloradensis, the host ant, but it is
of notably smaller stature, the occiput is smooth
and shining and lacks the transverse rugae or
striae that are characteristic of the host soldier,
the dorsal contours of the thorax, petiole, and
postpetiole are distinctively different, and the
body color is much lighter. Cole (ibid., p. 174,
figs. 1, 2) has compared the contours, in profile,
of the thorax, petiole, and postpetiole of the
soldiers in the two taxa.

The discovery of inquilina at the test site has
extended considerably westward its range which
included previously only Nebraska and Colorado.

Pheidole pilifera coloradensis Emery

Nests of this taxon were found only be-
neath rather large stones in disturbed Pinyon-
Juniper on Rainier Mesa. Colonies were rather
numerous. Winged forms were in some of the
nests between June 22 and July 4, 1962.

The timid, dark reddish brown soldiers can
be recognized easily by their transversely rugu-
lose or striate occiput and the presence of well-
developed lateral connules on the postpetiole.

ide.l it in the key because it lias been collected from adjoining areas

Hi B rich am Young University Science Bulletin

( .1 mis < 'rematogaster I .mid
A Ki\ iu Species ol the Genus Crematogaster for Identification <>t the Workers

1 Dorsum oi thorax without erect hairs; promesonotum densely punctate, the punctures larg<
K replacing the rugae depilis Wheeler

Dorsum of thorax with our, long, erect hair al each humeral angle; rugae on dorsum ol
promesonotum coarse and vermiculate, the interrugal punctures prominent, the surface
opaque coarctata vermiculata Emer)

( 'rematogaster coarctata
i ermiculata Emery

Although 1 have assigned all pertinent col-
lections to vermiculata, the test Site appears to
he in a region ol transition between vermiculata
and court lata, inasmuch as some of the workers
seem to represent Lntergrades of the two popu-
lations. Problems of this kind will have to be
resolved by a generic revision following an in-
spection of large series of ants from all parts ol
their range.

\ests at the test site were chiefly in open
soil, hut in Pinvon-Juniper on Rainier Mesa they
were found sparsely under stones. Although
being most numerous in the Larrea-Franseria
community, they occurred also in considerable
numbers in Atriplex-Kochia and less abundantly

m both mixed and Grayia-Lycium communities.

'The characters used in the key will serve ade-
quately to separate the two species oi Cremato-
gaster known from the test site. Males and fe-
males were found in nests on Rainier Mesa on
July is.

Crematogaster lUjnhs Wheeler

Strangely and apparently absent from the
Pinvon-Juniper community, colonies ol depilis
were most numerous in the Larrea-Franseria ind
mixed communities. The nests were in open
are. is between shrubs or at the bases of the
plants. Colonies appeared not to be so populous
as those of vermiculata. The workers can be
recognized by the lack ol hairs on the pronotum
and by their generally brown color.

Genus Monomorium Mavr
Monomorium minimum t Bucklcv )

This minute, black species nests beneath

stones chiefly in the Coleogyne and mixed com-
munities, although a tew scattered colonies were
found in the Pinvon-Juniper community of Rain-

ier Mesa. The worker is readily recognizable In

its small si/e. black color, smooth and highly
polished body, and the unarmed epinotum. The
populous colonies contained very active workers
and multiple queens. The sexual castes were not
found.

Genus Solenopsis Westwood
A Key to Species ol the Genus Solenopsis lor Identification ol the Workers

1. Second and usually the third funicular segment of antennae at least 1 '_• tunes as long as
broad: workers polymorphic -

Second and third funicular segment ol antennae at most only slightly longer than broad.
usually broader than long, workers verj small, not polymorphic, their nests frequently in

those of other ants 3

2. Eyes of major with no more than 50 facets, those ol minor with about 20 facets; front of
head of major with onl) a lew widel) scattered punctures; dorsum of epinotum of minor,

in lateral view, rather strongl) convex, bod\ a coneolorous golden yellow uurcu Wheeler

Eyes ol major with 70 to SO facets, those ot minor with about .50 I. nets, trout of head
ol major with numerous punctures; dorsum ol epinotum ol minor, in lateral view, weakly
onl broadly convex; head and thorax yellow to red faster iisualK notablv darker.

xi/loni McCook

Cdil !. ■■ , ■ \ ■ ■ ■■ nil Pop* i Nn

Ants of the Nevada Test Site yj

3. Punctures on head dense, small, distinct, clearly greater in diameter than the hairs which
arise from them saUm Wheeler

Punctures on head sparser, smaller, often visible onlv under high magnification, only a
little larger than the hairs which arise from them ......... molesta validiuscula (Emery)

Solenopsis aurea Wheeler

Nests of this species were in open soil of the
Coleogyne and Grayia-Lveium communities,
and were more representative of the former.
The colonies were small and few. The kev char-
acters will suffice for recognition of this ant.
The uniformlv golden vellow color of the body
in all workers is a characteristic feature.

Solenopsis molesta validiuscula ( Emerv i

This minute (length 1.8 to 2.0 mm), brown-
ish vellow species nests independently or in colo-
nies of other ants. Most colonies were found in
Pinvon-Juniper, and especially in nests of Phei-
dole pilifera coloradensis, but the species oc-
cupied also, to a very limited degree, the Gravia-
Lycium and Larrea-Franseria communities.

Solenopsis salina Wheeler

Three colonies of what probablv represents
this species were found beneath stones in the
disturbed Pinvon-Juniper community of Rainier
Mesa. A species difficult to separate from mo-
lesta validiuscula, salina has numerous cephalic
punctures that are of greater diameter than the
hairs which rise from them, and a distinct ven-
tral tooth on the venter of the petiolar peduncle.

Solenopsis xyloni McCook
The stronglv polymorphic workers of xyloni
were found infrequently in the Larrea-Franseria
communitv where they nest at the base of shrubs.
The minor workers are considerably darker than
the larger workers.

Genus Lcptotliorax Mavr
A Key to Species of the Genus Leptothorax for Identification of the Workers

1. Antennal scape in repose failing to reach the occipital margin by an amount twice as great
as its greatest thickness; epinotal spines reduced to short, broad angles; thoracic dorsum
densely and finely punctate andrei Emery

Antennal scape in repose failing to reach the occipital margin by an amount no greater than
its greatest thickness; epinotal spines well-developed; thoracic dorsum coarsely, longitudi-
nally rugose nevudensis rudis Wheeler

Leptothorax andrei Emery

A few workers, assignable to this species,
were taken in disturbed Pinvon-Juniper on Rain-
ier Mesa. The nest was not found. This small
species can be recognized by its yellow color,
its feebly shining head, its very short, angulate
epinotal armature, and the broad petiolar node
which, viewed in profile, is nearly as wide api-
cally as basally. The ants generally construct
small nests beneath stones. For additional in-
formation on andrei and its allies, the reader is
referred to a paper by Cole ( 1958 ) .

Leptothorax nevadensis
rudis Wheeler

Sparse and small colonies were found in the
disturbed Pinvon-Juniper community on Rainier
Mesa. The workers were sluggish and docile.

The worker of this subspecies is a small,
brownish black ant. The thoracic dorsum is
completely covered with coarse, longitudinal
rugae except for a punctate area on the meso-
notum. The epinotal spines are rather well de-
veloped, and the postpetiole is a little less than
twice the width of the petiole.

Subfamily Dolichoderinae

A Kev to the Genera of the Subfamily Dolichoderinae for Identification of the Workers

1. Epinotum with a prominent tooth-like protuberance directed upward from the junction of
the basal and declivious faces; third segment of the maxillary palp very long, as long as or
longer than the three succeeding segments combined Dorymyrmex

IS

Biiii.ium Yin \i. l\i\i nsin Si ii m i Hi i i i 1 1\

I i > 1 1 1 • >t 1 1 1 1 1 without such ,i protuberance unarmed the junction of the two faces rounded or
angular; third segmenl ol the maxillar) palp nol usual 1) long shorter than 1 1 »« - three succeed-
ing segments combined -

2 Dorsum ol the thorax without a conspicuous impression a| the mesoepinotal suture, hairs
on the thorax abundant; gastric pubescense dense; workers somewhat polymorphic

Liometopum

Dorsum ol the thorax with a conspici s impression at the mesoepinotal suture, hairs on

the thorax verj sparse; gastric pubescence dilute Iridomyrmex

Genus Dorymyrmex Forel
\ ke\ to Species of the Genus Dorymyrmex for Identification ol the Workers

1. Clypeus broadly and evenly rounded, without a trace of a median angle or carina; head
and thorax deep red or reddish yellow, gaster brownish black or black bicolor Wheeler

Clypeus distinctly angular or subcarinate medially; body not contrastingly bicolorcd

pyramicus \ Rogei

Dorymyrmex bicolor Wheeler

Rapidly moving, foraging workers of this
species were encountered repeatedly in the
Larrea-Franseria community to which the nests
were restricted. This common, typically desert
species constructed, in unshaded areas, nests
marked 1>\ exquisite, semicircular or circular
craters of fine loose sand. The worker can be
distinguished from that of its nearest relative,
pyramicws, bj its bicolored body, a feature that

appears to be genetically fixed.

Dorymyrmex pyramicus \ Rog< i

Verj closely allied to bicolor, pyramicus was
also found nesting, and in its greatest numbers.
in the Larrea-Franseria community, where it was
often sympatric with bicolor at the same stations.
It occurred also, but to a much more limited
extent, in the Coleogyne and mixed communi-
ties.

Genus Liometopum Mavr

Liometopum occidentale
luctuosum Wheeler

\ single small Colony Of this species was

found in disturbed Pinyon-Juniper on Rainier

Mesa. The nest was in soil beneath light detri-
tus at the base o| a juniper.

The worker caste shows a considerable vari-
ation in size (2.5 to 5 mm in length I, has \ er\
sparse pilositv. and is rather uniformly deep
brown in color and rather strongly shining. As

is true ol other members o| its genus, luctuosum
possesses the characteristic and unpleasant "Lio-
metopum odor.

Genus Iridomyrmex Mavr

Iridomyrmex pruinosum

analis I E. Andre )

Present in all plant communities studied at the
test site, this common .nit constructs its nests
beneath stones, at the base of plants, ami in
fully exposed areas where it makes a small, ir-
regular or Circular mound ol soil. The small
winkers run about verj rapidb and apparently

erraticalb The} can withstand very high soil

surface temperatures, and were observed forag-
ing on trails during hot summer days.

The pale \ellow' and vcllow and brown work-
ers arc characterized by their long, sparse, erect
body hairs and the dilute pubescence on the
head and thorax which does not obscure the
shining surface. The worker is \irtualK identi-
cal to that ol bkolor, except lor its concol-

orous brown body. Nesl location and construc-
tion ol the tW" species alothc same.

Ants of the Nevada Test Sitk 19

Subfamily Formicinae
A Key to the Genera of the Subfamily Formicinae for the Identification of the Workers

1. Thoracic dorsum, in lateral view, evenly convex, the epinotum not depressed below the
level of the promesonotum, the mesoepinotal suture slightly or not at all impressed; antennal
scapes inserted well behind the posterior edge of the clypeus Camponotus

Thoracic dorsum, in lateral view, with the epinotum distinctly depressed below the level of
the promesonotum, the mesoepinotal suture well impressed; antennal scapes inserted at or
near the posterior border of the clypeus 2

2. Maxillary palps very short, 3-segmented Acanthomyops

Maxillary palps notably longer, 6-segmented 3

3. Maxillary palps longer than the head, the third and fourth segments very long and as long

as or longer than the two terminal segments combined; psammophore present Myrmecocystus

Maxillary palps shorter or no longer than the head, the third and fourth segments shorter than
the two terminal segments combined; psammophore absent 4

4. Frontal carinae prominent, their lateral margins slightly reflected upward; ocelli very dis-
tinct Formica

Frontal carinae less well marked, their lateral margins flattened; ocelli indistinct or absent
Lasius

Genus Camponotus Mayr
A Key to Species of the Genus Camponotus for Identification of the Workers

1. Middle of anterior clvpeal border depressed, bearing a narrow, median notch; length of major
less than 8 mm hyatti Emery

Middle of anterior clvpeal border not bearing a narrow, median notch; length of major
notably greater than 8 mm 2

2. Antennal scape of the major distinctly flattened at the base, the flattened portion forming

a small lateral lobule maccooki Forel

Antennal scape of the major not flattened at the base or, if flattened, without a lateral lob-
ule 3

3. Cheeks strongly shining, with minute inconspicuous punctures; antennal scape not flattened
at the base; head black; thorax and gaster yellow, often suffused with brown; tibial- and
tarsi strongly infuscated ocreatus Emery

Cheeks feebly shining, the punctures coarser and conspicuous; antennal scape flattened at
the base; body medium to very deep red uicinus Mayr

Camponotus hi/atti Emery known from the test site. The highly polished,

. . , 11 l r ; ,.• r, a i<'t black "aster contrasts sharply with the shin-

A single small colony of luiatti was found ' , . , & , , , , , ' -

.°.i .,i ., i • j ■ i. i ing, bright red head and thorax,

nesting in the soil beneath a dead juniper limb &' °

in a mixed community. The species is a member

of the subgenus Myrmentoma, which is char- Camponotus maccooki forel

acterized by having a narrow, median notch in This species was limited largely to the Gray-

the depressed middle of the anterior clvpeal ia-Lycium community, but it occurred spottedly

border, and is the onlv member of that subgenus in Pinyon-Juniper also. Nests were beneath

■m

Huk.ii wi Yoi'.vc: I viversiti Science Hi i i i i is

stiiins. I In- major workers are rather large ants yellow
ill .1 rathei uniform yellowish hrown or reddish strikini
brown color. The) are readih identifiable l>\
their antenna! scapes which are distinct!) flat-
tened hasalh and hear a lateral loliule.

Cainponotux ocreatus Emen

Except tin two workers found in the I. .una
l- i.nisii ia community, all members ol tins spe-
i ■■ -s occurred in the mixed community. \ singli

thorax and faster give
appearance.

In- winker a

( 'amponotus i U inus \\.i\ i

\rsts ut cicinus were found frequently in

tin- Pinyon-Junipei comi it\. ol which the)

arc a characteristic element, but were seen no
where else at the test site The workers are
very large ants with a deepl) ferrugineous red
thorax and a black head and gaster. Colonies

nest was located beneath a small stone. The were beneath the largei rucks in parti) shaded
black head and appendages contrasting with the .mas.

Genus Lasius Fabricius

\ kc\ to Species ut the Genus Lasius for Identification of the Workers

I \er\ small ants (width ot pronotum usually less than 0.54 mm.); eyes minute, typically
with 11 ommatidia in a line along the lout; axis ol the eye; color yellowish brown

titiens Wilson

Larger ants (width ol pronotum usually more than 0.54 mm.); eyes larger, typically with
1 1 or 15 ommatidia in a line along the long axis ol the eye; color consistentl) darker, us-
ually dark hrown crypticus Wilson

Lasius crypticus Wilson

A few nests ol crypticus were found ill (lis
turbed Pinyon-Juniper on Rainier Mesa. All
were in the soil beneath stones in open areas.

The dark brown workers have a sparse pilos-
ity, and erect li.urs ,ne absent Irom the antenna!

scapes and tibiae.

Lasius sitiens Wilson

('ninnies ol this species were found Only in

disturbed Pinyon-Juniper, where the) were con-

siderably more common than those ot crypticus.
The nests were beneath stones at the same
stations as those occupied by crypticus. The
sexual castes were iii nests between June i'3
and July It).

The workers are of a pale hrown color and
arc UOtabl) smaller than those ol crypticus. The
exes are small, having only 10 to 12 ommatidia
across the maximum diameter ot the eye rather
than the usual 1 t or 15 ol crypticus.

This is a new Nevada record for tliis species.

Genus Acanthi
Acanthomyops latipes i Walsh i

Colonies ol latipes were found only in dis-
turbed Pinyon-Juniper on Rainier Mesa. The

nests were Incited rather deeply ill the soil lic-

neath stones in partially shaded areas. Queens

nl both alpha and beta types were taken. The

latter is an especially interesting ant. The body
is extremely hairy, the femora and tibiae are
greatl) compressed and enlarged, so that the)
contrast strongly with the slender tarsi, and

""'/"/'

Mavr

gradually much incras-
including the gula, is

the antenna] scapes arc
sated from base to apex.

The workers body,

densel) hair) and ol a rather concolorous brown-
ish yellow. The antenna! scape, in repose, docs
not surpass the occipital holder. Viewed in
profile, the scale ol the petiole has a blunt apex:
seen from behind, the scale is somewhat flat-
tened and entire. The worker possesses the

"citronella odor" that is characteristic ot all
members ol its genus.

Ants of the Nevada Test Site

Genus Myrmecocystus Wesmael

A key to Species of the Genus Myrmecocystus for Identification of the Workers

1. Eves large, their greatest diameter notably longer than the length of the first funicular seg-
ment; ocelli small and obscure or absent; mandibles with eight teeth

Eyes small, their greatest diameter about equal to the length of the first funicular segment;
ocelli large and prominent; mandibles with seven teeth

2. Petiolar scale thick from front to back, erect hairs on crest and sides of petiolar scale num-
erous and conspicuous; length range 4.0 to 9.5 mm mexicanus Wesmael

Petiolar scale thin from front to back; erect hairs on crest and sides of petiolar scale very
sparse or absent; length range 2.5 to 5.5 mm mojave Wheeler

3. Erect hairs delicate, sparse or absent on the cheeks; pubescense short and sparse, not not-
ably obscuring the ground surface; body rather shining 4

Erect hairs coarse and numerous, abundant on cheeks; pubescence longer and dense, most-
ly obscuring the ground surface; both often silvery but not shining contains Wheeler

4. Erect hairs sparse or absent on the femora and tibiae, confined chiefly to the flexor sur-
faces; small ants, length 2.0 to 3.5 mm; body a concolorous dark brown lugubris Wheeler

Erect hairs abundant on all surfaces of the femora and tibiae; larger ants, length 3.0 to 6.5
mm; body bicolored, the head and thorax notably lighter than the gastcr, and generally
red or reddish brown mimicus Wheeler

Myrmecocystus comatus Wheeler

M. comatus was well represented in the
Gravia-Lycium, Larrea-Franseria, Atriplex-Koch-
ia, and mixed communities; scarce in the Cole-
ogvne and Salsola communities; and rare in
Pinvon-Juniper areas. It nests in open areas be-
tween shrubs. Each nest is marked either by
onl\- a hole in the soil or by a crude soil crater.
Most colonies were very populous.

By far the hairiest member of its genus and
accordingly well named, comatus is best, but
not easily, distinguished from its close relatives,
by that trait. The dull red head, the darker
infuscated thorax, and the silvery, densely pu-
beseeri, black gaster are unique features. The
workers vary in length from 3.5 to 6.5 mm.

Myrmecocystus lugubris Wheeler

This species was found only in the Atriplex-
Kochia community where it was a minor com-
ponent of the ant fauna. The small colonies
occupied exposed areas, and their nests were
marked by an entrance around which soil was
loosely and irregularly dispersed. The workers
were active during the intense heat of the sum-
mer day. They appeared to be most numerous
on the shrubs over which they moved swiftly.

The very dark brown or black, polymorphic
workers are of rather small stature, with a body
length of 2.5 to 3.5 mm. Erect hairs on the
body, and especially on the appendages, are

rather sparse. The entire body is somewhat
shining.

Myrmecocystus mexicanus Wesmael
Nests of this common desert species were
found most abundantly in the Gravia-Lycium
community, but they were rather numerous in
the mixed, Salsola, and Coleogyne communities,
and were scarce in the Larrea-Franseria com-
munity. All colonies were in unshaded soil be-
tween plants, and were marked with a broad,
circular entrance of coarse, tightly packed sand
(Fig. 19). The ants are nocturnal foragers.

This species has strongly polymorphic work-
ers ( 4.5 to 9.5 mm in length ) which can move
with great rapidity and agility. Although most
specimens were referable to the typical mexi-
canus, some (even in the same nests) were
characteristic of the subspecies hortideorum Mc-
Cook, which supposedly differs from mexicanus
by its nearly concolorous yellow body. I pre-
dict that future revisionary studies of Myrme-
cocystus will show that the two forms are syn-
onymous.

Hhk.m \m Viim. I \1\11wn Sc ii m 1 Hi l.i i i in

L9 Mound ..I

mixed rommimitv

Myrmecocystus mimicus Wheeler

\ common occupant of the Larrea-Franseria
community, mimicus was well represented in
the Grayia-Lycium, but only moderately so in
the Vtriplex-Koehia, Salsola, and mixed com-

munities, ami poorly so in Coleogyne. The nests
and nesting sites were like those of comattu.

I In workers possess erect hairs t hut are
notahl) more delicate and less abundant than
those dI comatus. The two species are not sep-
arable l>\ color differences, but the bodv of
mimu i« is more shining, and the pubescence
on the gaster is not sufficiently dense to obscure
the shining surface or produce a silvery luster.

Myrmecocystus mojan e Wheeler

At the test site mojave is restricted to Pinvon-
[uniper areas. Strong colonies were round nest-
ing beneath stones or in soil without cover and
marked by a circular crater or small pebbles
with a large, irregular, central entrance. The
workers are nocturnal and crepuscular foragers.
The sexual castes, as well as semirepletes, were
in nests on Rainier Mesa from middle to late
[une,

The workers of mojaoe, like those of mexi-
canus, are polymorphic, but less strongly so.
They vary in length from 3 to 5 mm. The scale
ol the petiole is notably thinner, viewed lateral-
ly, than is that of mexuxmus.

Genus Formica Linne

\ key to Species ol the Genus Formica tor Identification of the Workers

1 The combined second and third funicular segments ol the antenna not more than 115
limes as long as the first segment and often not that long 2

The combined second and third funicular segments at least 1J« times as long as the first
segment and usually longer 4

2. Extensor surface of antenna] scape with a number of short, delicate, erect, silvery hairs

lasioiaes Emery

Extensor surface <>l antenna! scape without erect hairs, except for a few at the extreme tip 3

3. Thorax with numerous erect hairs; body surface moderately shining neogagates Emery

Thorax without erect hairs or with only one or two erect hairs; bodv surface strongly shin-
ing ............ limata Wheeler

1 Erect hairs on the pronotum distinctly clavate or spatulatc microgyria Wheeler

Erect hairs on the pronotum simple, not clavate or spatulatc 5

5. \iitcnnal scape considerably longer than distance from middle of clypeal border to mid-
occipital border, base of epinotum notably eonvex. the angle between it and the decisions

surface poorh defined; posterior surface or petiole convex moki Wheeler

Vntennal scape not longer than distance from middle ol clypeal bonier to midoccipital

border, or if longer, the epinotum distinctly angular, posterior surface ot petiole not convex 6

6. Anterior border ol eK pens distinctly excised, gaster evenly covered with stout, rather
long, blunt, erect, silvery hairs obtusopilosa Emery

Ants of the Nevada Test Site 23

Anterior border of clypeus not excised; gaster without such pilosity 7

7. Head and thorax of the larger workers ferrugineous red and generally notably lighter
than the gaster or, if infuscated, the infuscation not obscuring the ferrugineous ground
color; frontal area shining; frontal lobes strongly divergent behind 8

Bodv concolorous or, if bicolored, the thorax lighter than the head and gaster; frontal area
opaque, frontal lobes not stronglv divergent behind, often parallel 9

8. Head of the major worker, excluding the mandibles, as broad as long or broader than
long; erect hairs on the thorax unequal in length; cephalic hairs nearly as numerous as,
and only a little longer than, those on the thorax obscuripcs ForelT

Head of the major worker, excluding the mandibles, longer than broad; erect hairs on the
thorax short, subequal in length; cephalic hairs considerably longer and less dense than
those on the thorax integroides planipilis Creighton

9. Gula with erect hairs; worker caste stronglv polymorphic; head of largest workers (man-
dibles excluded) quadrate, as broad as or broader than long
subpolita camponoticeps Wheeler

Gula without erect hairs; worker caste not polymorphic; head longer than broad

10

10. Body black; when in full-face view, eyes of largest workers reaching or projecting beyond
the head margin fusca L.

Thorax more or less red; when in full-face view, eyes of largest workers not reaching the
head margin neorufibarbis Emery

Formica fusca Linne

This species appeared to be restricted to Pin-
von-Juniper areas, where the nests were be-
neath individual or clusters of stones. Colonies
were sparse and also unusually weak for this
species. The ants are active scavengers.

The workers are entirely black and of me-
dium size (4 to 7 mm in length), and they have
a moderately shining and very sparsely hairy
bodv. The gaster is moderately pubescent and
often has a silky luster.

Formica integroides
planipilis Creighton

This ant was limited to Pinyon-Juniper areas
and was more representative of the undisturbed
than the disturbed ones. Each nest was sur-
mounted by a dome-shaped, thatched mound
fashioned of assorted, rather tightly packed de-
tritus, and constructed against or around a
shrub that was ultimately killed by the ants
(Fig. 20). Brood was found in chambers with-
in the mound as well as in those underground.
The colonies were very populous. When the
mounds were disturbed, workers in large num-
bers attacked the intruder and ejected their
"formic acid" spray.

The workers vary considerably in size, and
may be designated as minors, media, and ma-
jors. The majors are distinctly bicolored, with
dull red head and thorax and brownish black
gaster; the red portions of the media are more
or less marked with brown; and the minors are
stronglv and extensively infuscated. The legs
are brownish black in all workers. The body
is very hairy, the erect hairs of the thorax being
short and subequal in length, whereas those of
the head are longer and sparser. Erect hairs on
the middle and hind tibiae are numerous and
generally cover all surfaces. The gaster is
opaque and densely pubescent.

Formica lasioides Emery

This species was sparsely represented in the
ant fauna of disturbed Pinyon-Juniper on Bain-
ier Mesa. The nests were beneath small stones
in open areas. The workers were very timid.

The worker of this comparatively small ant
(3.5 to 4.5 mm in length) has a smooth and
shining body which is generally of a medium
brown" color. There are a number of erect, short,
delicate hairs along the extensor surface of the
antennal scape.

Briciiam Young University Scienci Mi i i.i i i\

Fig. -'»• Thatchi
length.

j>Uinif>ilis in a Pinyon-Junipei communit}

Formica Umata Wheeler

Two small colonies ol tin's ant were Found
under small stones in disturbed Pinyon-Juniper
on Rainier Mesa. The species appeared no-
where else at the test site.

The worker resembles that ol lasioides and
neogagates. The characters in the key will seg-
gregate it readily. Its most stinking character-
istic is the ven strongh shining body.

I in mica microgyria Wheeler

\ single worker assignable to tins species
was taken from a can trap in Pinyon-Juniper.
No nest was found.

The worker ol microgijna lias a rather light
ferrilgineOUS head and thorax, and a black,
strongh opaque gaster. It is characterized chiel
K In the presence ol a few clavate hairs on the
pronotum and b\ the pilosih ol the femora and
tibiae which includes erect hairs in addition to
those on the I lexoi surfaces.

\csts ol microgyria arc generall) constructed

in open areas under stones which are ultimately
hanked peripherally with detritus.

Formica moki Wheeler
\ single colon) ol this species was found
in a Pinyon-Juniper area near Tippipah Spring.

The nest was in open. slon\ soil beneath a small
rock. Numerous workers and a lew males were
collected, but no females were encountered. The
workers arc 1.0 to 5.8 mm in length and ol a
dull ferrugineous red color, with the postero-
dorsal part ol the head, and the petiole, gaster,
and le'_;s dark hrown. The hairs are sparse and
erect I h> l>od\ surface is suhopaque and fine-
l\ and denseh grannlose. In certain lights the
posterodorsal portion ol the head and tin- faster
1 1 . i \ i a rather l>ron/\ luster.

I'h, male caste is described herewith for
the In si i inn

Ants of the Nevada Test Site

25

Male. ML 1.41 to 1.41 mm, M\V 1.60 to 1.74
iniii. CI 113 to 123, SL 1.67 to 1.71 mm, SI 'is
to 104, EL 0.87 to 0.91 mm, EW 0.49 to 0.53
mm, 01 62 to 65, TL 3.50 to 3.S0 mm, P.M.
0.49 to 0.53 mm, PNW 0.95 to 1.03 mm.

Antennal scape long, its length approximate-
ly equal to the combined lengths of funicular
segments 2 to 5, inclusive. Median lobe of
clvpeus strongly and evenly convex, its anterior
border entire.

Petiolar node, in profile, notably broader
basally than apically, the apex acute hut rather
thick. Petiolar node. Viewed from above and
behind, with its apex prominently, broadly, and
evenly concave; the corners rather sharply
rounded.

Paramere of genitalia as shown in Fig. 21,
vulsella as in Fig. 22, and aedeagus as in Fig.
23. Stemite IX of abdomen as illustrated in
Fig. 24.

Mead, thorax, and petiole densely and finely
grannlosc; subopaque. Gaster densely and more
finch' grannlosc; the surface somewhat shining.

Cephalic pilosity mostly very sparse, the
hairs limited chiefly to the mandibles (where
they are long and dense) and the anterior mar-
gin of the clypeus; absent from occipital corners
of the head. Thorax (except the epinotuin )

Fig. 22 Volsella of a male Formica moki.

Fig. 23. Aedaegus of a male Formica moki.

:>f a male Formica moki.

Fig 24, Abdominal stemite IX of a male Formica

moki.

mostly densely pubescent, with a few scattered,
short, slender, pointed, golden hairs on scutum
and scutellum; epinotuin, petiole, and dorsum
of base of first segment of gaster covered with
a dense pile of very short, erect and subercct,
slender hairs forming a rather even, plush-like
investiture; each corner of apex of petiolar node
with two or three comparatively long, erect
hairs. Gaster chiefly without hairs, except for
those described on the first segment and for a
verv lew. short, siiheiect, delicate ones along
the posterior border of the segments; pubes-
cence on gaster appressed, widely spaced, not
obscuring the surface.

Mead deep brown; thorax, and especially the
dorsum of the scutum, a little lighter; petiole,
legs, and gaster dusky yellow; antennae light
brown.

Mkii. 1 1 wi Young I mversiti Si n \< i M< i i i i i\

ribed from six males collected bj the
writei on |ul\ 25 L962 / moki, heretofore im-
ported ( i < •■ 1 1 Nevada, is known also from south-
ern Utah .iiitt northern Vrizona where it ap-
pears to be a verj uncommon species

Formica neogagates Emerj

Colonies ol neogagates were found at the
test site only in Pinyon-Juniper areas, but they
were considerabl) better represented than were
those "I Utsioides and limata. Nests and nesting
siii-s were similar iii those ol the latter two
species.

The worker ol neogagates resembles closely
thai nl both lasioides ami limata in color and
averagi size. Unlike thai ol lasioides, however,
its antenna] scapes have no erecl hairs. Its body
is iiiiK i I. ratel) shining.

/ minim neorufibarbis Emery

\ few colonies "I tins ant were found under

st s in partial shade in the Pinyon-Juniper

community .

'I'lu' worker "I tins medium-sized ant I length
3 to 6 mm I lias a brown to black head and gas-
ter, and a light to deep red thorax which is
often strongl) infuscated, especially on the pro-
notum. The head and thorax arc subopaque; the
gaster is rather smooth and shining.

Formica obtusopilosa Emery
A few colonies oi this ant wen found in a

mixed plan! COmmunit) with scattered sparse

junipei mil Vrtemisia. The nests wen m rather

gravelh soil ol unshaded arc is. The)
were marked b\ a rather largi irregular en-
trance surrounded b) a narrow, uneven cir-
cle! ol small pebbles. Workers were n TOUS

III the colonies. \|\ attempts to find males.
which have never lieen descrilied. were unsuc-
cessful

The large ' up to 7 mm m length I, hand-
some workers ol obtUSOpUosa have the head

and thorax colored a rich, uniform ferrugineous
red. and the gaster an opaque black. The con
spicuous, long, robust, blunt, silver) hairs which
evenh cover the gaster are definitive ol this
taxon. As in other closely related taxa, the an-
terior clypeal margin is distinctly, mediall)
emarginate.

Formica mbpolita camponoticeps

Wheeler

Restricted to the Pinyon-Juniper community,

colonies ol this taxon were rather common un-
der stones in sunm areas. The workers are
strongly polymorphic and \ar\ in length from
2.5 to (>.() mm.

The workers have a shining body with cas-
taneous brown head, brownish or reddish yel-
low thorax, and piceous brown gaster. The head
ol the larger workers is at least as broad as
long, excluding the mandibles, and its gula

hears erect hairs. There are a few erect hairs
also on the front and occiput as well as on the
dorsum of the pronotiim.

Subfamil) Hi no i in u
(anus Neivami/rmex Borgmeier
Neivamyrmex miner (Cresson) to black light on Jul) 21 in Pinyon-Juniper on

\ few males ol this species were attracted Manner Mesa. The worker caste is unknown.

Ants of the Nevada Test Site

27

LITERATURE CITED

Allred, I). M., D E. Beck, and C. D. Jorgensen. 1963.
Biotic communities of the Nevada Test Site. Brig-
ham Young Univ. Sci. Bull.. Biol. Ser., 2(2):l-52.

Andrews, H. 1916. A new ant of the genus Messor
from Colorado. Psyche. 23:81-3.

Barnuin, A. H. 1964. Orthoptera of the Nevada Test
Site. Brigh.im Young Univ. Sci. Bull.. Biol. Ser..
4(3): 1-134.

Cole, A. C. 1955. Studies of Nevada ants. I. Notes
on Veromessor lariversi M. R. Smith and a descrip-
tion of the queen ( Hymenoptera : Formicidae). f.
Tennessee Acad. Sci., 30:51-2.

Cole. A. C. 1958. North American Leptothorax of the
nitens-carinutus complex (Hymenoptera: Formici-
dae). Ann. Entomol. Soc. Amer., 51:535-8.

Cole. A. C. 1963. A new species of Veromessor from
the Nevada Test Site and notes on related species
(Hymenoptera: Formicidae). Ann. Entomol. Soc.
Amer., .56:678-682.

Cole, A. C. 1965a. Discovery of the worker caste of
Pheidole (P.) inquilina, new combination (Hy-
menoptera: Formicidae). Ann. Entomol. Soc. Amer.,
58:173-5.

Cole, A. C. 1965b. A monographic revision of the
genus Pogonomyrmex Mavr in North America I Hy-
menoptera: Formicidae). (Unpub. ms.).

Gregg H. E. 1955. The rediscovery ol Veromessor

lobognathus (Andrews) (Hymenoptera: Formici-
dae). Psyche, 62:45-52.

Gregg, R. E. 19.58. Key to the species of Pheidole
( Hsmenoptera: Formicidae) in the United States.
J. New York Entomol. Soc, 66:7-48.

Smith. M. R. 1941. Two new species of Aphaeno-
gaster (Hymenoptera: Formicidae). Great Basin
Nat., 2:118-21.

Smith, M. R. 19.51. Two new ants from western Ne-
vada (Hymenoptera: Formicidae). Great Basin
Nat., 11:91-6.

Smith. M. R. 1957. Revision of the genus Stcnamma
Westwood in America north of Mexico ( Hymenop-
tera: Formicidae). Amer. Midland Nat., 57: 133-74.

Smith, M. R. 1963. A new species of Aphaenogaster
(Attomyrma) from the western United States (Hy-
menoptera: Formicidae). |. New York Entomol.
Soc, 71:244-6.

Wheeler, G. C. and J. Wheeler. 1956. Veromessor
lobognathus in North Dakota (Hymenoptera: For-
micidae). Psyche. 63:140-5.

Wheeler. G. C. and ]. Wheeler. 1959. Veromessor
lobognathus: second note (Hymenoptera: Formici-
dae). Ann. Entomol. Soc. Amer.. 52:176-9.

Wheeler. G. C. and |. Wheeler. 1965. Veromessor
lobognathus: third note (Hymenoptera: Formici-
dae). J. Kansas Entomol. Soc, 38:55-61.

v u _l

iy65

Brigham Young University
Science Bulletin

THE SPHAERODACTYLUS DECORATUS

COMPLEX IN THE

WEST INDIES

by

RICHARD THOMAS

and
ALBERT SCHWARTZ

BIOLOGICAL SERIES — VOLUME VII, NUMBER 4
AUGUST, 1966

BRIGHAM YOUNG UNIVERSITY SCIENCE BULLETIN
BIOLOGICAL SERIES

Editor: Dorald M. Allred, Department of Zoology and Entomology,
Brigham Young University, Provo, Utah

Associate Editor: Earl M. Christensen, Department of Botany,
Brigham Young University, Provo, Utah

Members of the Editorial Board:

J. V. Beck, Bacteriology

C. Lynn Hayward, Zoology

W. Derby Laws, Agronomy

Howard C. Stutz, Botany

Wllmer W. Tanner, Zoology, Chairman of the Board

Stanley Welsh, Botany

Ex officio Members:

Rudger H. Walker, Dean, College of Biological and Agricultural
Sciences

Ernest L. Olson, Chairman, University Publications

The Brigham Young University Science Bulletin, Biological Series, publishes
acceptable papers, particularly large manuscripts, on all phases of biology.

Separate numbers and back volumes can be purchased from University
Publications, Brigham Young University, Provo, Utah. All remittances should be
made payable to Brigham Young University.

Orders and materials for library exchange should be directed to the Division
of Gifts and Exchange, Brigham Young University Library, Provo, Utah.

Brigham Young University
Science Bulletin

THE SPHAERODACTYLUS DECORATUS

COMPLEX IN THE

WEST INDIES

by

RICHARD THOMAS

and
ALBERT SCHWARTZ

BIOLOGICAL SERIES — VOLUME VII, NUMBER 4
AUGUST, 1966

OCT 1 3 1966

UNIVERSITY

TABLE OF CONTENTS

Page

INTRODUCTION 1

HISTORICAL SUMMARY 2

PROCEDURES AND TERMINOLOGY 3

THE BAHAMAN SITUATION 3

S. decoratus — , 4

S. </. decoratus 5

S. (/. fUnicaudus 7

S. d. atessares, new subspecies 8

THE CUBAN SITUATION 9

S. </. granti, new subspecies 10

S. d. lissodesmus, new subspecies 12

S. </. strategus, new subspecies 13

Specimens Not Allocated to Subspecies 14

S. alayoi 14

S. torrei - - 15

S. t. torrei 16

S. /. OCUJal, new subspecies 16

S. t. spielmani 18

S. stejnegeri 19

S. intermedins 20

S. ruibali 22

HABITAT 22

CONCLUSIONS 23

LITERATURE CITED 25

LIST OF FIGURES

1. Map of the Bahama Islands 5

2. S. </. decoratus female, dorsal view 6

3. S. d. decoratus male, dorsal view 6

4. S. decoratus ssp. female, dorsal view 6

5. S. d. atessares female, dorsal view 6

6. Map of Cuba - H

7. S. d. granti female, dorsal view 12

8. S. d. lissodesmus female, dorsal view 12

9. S. </. strategus female, dorsal view - 12

10. S. d. alayoi female, dorsal view 12

11. Map of eastern Cuba 14

12. S. t. torrei female, dorsal view 1'

13. S. (. ocujal female, dorsal view 1'

14. S. t. spielmani female, dorsal view 1'

15. Dorsal and ventral views of heads of three subspecies of S. torrei 18

16. S. stejnegeri female, dorsal view 1"

17. Map of Haiti 20

18. S. intermedins, dorsal views of two females -1

19. S. ruibali, dorsal views ^1

20. Lateral views of heads of four geckos to show hypothetical evolution of pattern from generalized
decoratus-type head pattern through intermedins to ruibali 24

THE SPHAERODACTYLUS DECORATUS COMPLEX
IN THE WEST INDIES

by
Richard Thomas and Albert Schwartz1

INTRODUCTION

On the Greater Antillean islands of Cuba
and Hispaniola, and the Bahama Islands occur
geckos of the genus Sphacrodactijlu.s in which
the juveniles are typically boldly marked with
from two to five dark crossbands. Adults are
sexually dichromatic — the females retain a
modified and usually less intense juvenile
pattern, and the males are either unicolor
and patternless, or have large dark dots on a
paler ground on the dorsum and head. The group
was recently reviewed in part bv Schwartz
(1958). Inasmuch as additional material has
been collected since that time and several new
names applied to members of the group, it
seemed appropriate to revise the complex com-
pletely. A large amount of the new material was
collected in Cuba through the efforts of Major
Chapman Grant, whose interest in these geckos
resulted in his description of three new species
of this (the decoratus) complex. Although there
are still grossly inadequate numbers of these
lizards from many areas of eastern Cuba, the
material available indicates that a complete re-
assessment of the named forms and their inter-
relationships is in order. As will be seen, this
strikingly marked group of geckos is considerablv
more complex than originally thought.

We collected decoratus group members in the
Bahamas and Cuba. These specimens are in die
Albert Schwartz Field Series (ASFS) and die
collection of the senior author (RT). Other
Cuban material collected by the junior author
and John R. Feick, William H. Gehrmann, Jr.,
Ronald F. Klinikowski, David C. Leber, James D.
Smallwood, and Barton L. Smith between the
years 1957 and 1960 under the sponsorship of
National Science Foundation grants G-3865 and
G-6252, has been deposited in the American
Museum of Natural History (AMNH). Messrs.
Klinikowski and Leber were also students in the
National Science Foundation Undergraduate
Participation program during 1959 and 1960. We
are indebted to Dennis R. Paulson and C. Rhea
Warren for visiting Cat and Long islands in the

'10,000 S.W. 84th St., Miami, Florida.

Bahamas on our behalf, and to Miss Patricia A.
Heinlein for assistance in collecting these geckos
on South Bimini. The senior author visited the
United States Naval Base at Guantanamo Bay,
Cuba, through the cooperation of the U.S. Coast
Guard and the officers and men of the U.S. Coast
Guard cutter Holk/hock. We extend our sincere
thanks to all of these people for their assistance
and companionship in the field; without their
efforts far less material would be available for
our use.

In addition to our own material we borrowed
material from the following collections whose
respective curators and their assistants were most
cooperative in lending us pertinent specimens;
American Museum of Natural History (AMNH),
Charles M. Bogert and George W. Foley; Acad-
emy of Natural Sciences of Philadelphia
(ANSP), James N. Bohlke and Edmond V. Mal-
nate; Brigham Young University (BYU), Wil-
mer W. Tanner; Carnegie Museum (CM), Neil
D. Richmond and Clarence J. McCoy; Chicago
Natural History Museum (CNHM); Robert W.
Inger and Hvmen Marx; Museum of Compara-
tive Zoology (MCZ), Ernest E. Williams; Uni-
versity of Illinois Museum of Natural History
(UIMNH), Hobart M. Smith; Museum of Zool-
ogy, University of Michigan (UMMZ), Charles
F. Walker and Dale L. Hoyt; United States Na-
tional Museum (USNM), Doris M. Cochran and
James A. Peters; Yale Peabody Museum (YPM),
Charles A. Reed. Attention is also directed to the
collections made by George B. Rabb while on
the Van Voast-American Museum of Natural
History Bahama Islands Expedition between De-
cember, 1952 and May, 1953. These lizards, pres-
ently divided between the American Museum
and the University of Michigan, are of special
interest and add significantly to the quantity and
quality of the available Bahaman material.

Most of the illustrations are the work of
Wayne King. His attention to detail aided great-
ly in visualization of the various forms involved,
and we thank him for his excellent delineations.

Bbicham Younc UNivERsm Si ii \( i Bulletin
HISTORICAL SUMMARY

The names applied to members <>l tin- decora-

tUS group .mil their type localities arc tin lol
lowing:

i 1 ) decoratus Carman, 1888 - Rum Cay, Ba-
hama Ulands

J flavicaudus Harbour. HXM - Mangrove
( lay, Andros Island, Bahama Islands

(3) torrei Barbour, 1914 - Santiago de Cuba,
Oriente Province, Cuba

(4) gibbus Barbour, 1921 - Stocky Island,
Exuma Cays, Bahama Islands

5 Hejnegeri Cochran, 1931 - St. Michel de
I'Atalaye, Dept. de I'Artibonite, Haiti

(6) drapetiscus Schwartz, 19.58 - 2 miles east
of Playa de Guanabo, Cueva de Rincon
de Guanabo, Habana Province, Cuba

(7) spielmani Grant, 19.58 - Guantanamo,
Oriente Province, Cuba

(8) ahu/oi Grant, 1959 - United States Nasal
Base, Guantanamo, Oriente Province,
Cuba

(9) niibali Grant, 1959 - United States Naval
Base, Guantanamo, Oriente Province.
Cuba

We examined the holotypes and syntypes ol
all the above forms. A total of 762 specimens of
this complex was at our disposal.

All of the names listed above were proposed
as full species with the exception of drapetiscus.
Barbour (1921:227-228) considered flavicaudus
a synonym ol decoratus: the former name was
based upon a series of fourteen svntvpes, all
males and without pattern. Barbour indicated
that in decoratus, males were patternlcss and
females were transversely crossbarred, a condi-
tion which had heretofore gone unnoted. In the
same publication, Barbour (op. < ir.:22.S-229 )
described S. gibbtiS, based on two specimens
from Stocks Island and one from a cay off
Rosevillc. Great Exuma. This species was diag-
nosed as being "coarsely spotted with large dots";
the holotvpc is a male. One1 of the paratypes
| NIC/. 13437) is barely spotted dorsally.

The name torrei was based upon two females
and one juvenile (holotype and two paratypes);
all arc prominently crossbanded. Later, Barbour
( 1921:230-231 ) noted that male torrei are "uni-

form gray-brown or very faintly barred." How-
ever, Bail r commented under the discussion

ot gibbus that two Cuban specimens (one from
Santiago de Cuba, the other from Guantanamo)
were spotted like gibbus, but in habit and scjua-
m.ition were like turret. Even at this early date
in the history ot the decoratus group, confusion
was existent.

Hecht ( 19.54:133) first suggested that decora-
tus and gibbtiS were conspecilic. This course of
action was followed by Schwartz. ( 1958), who
also considered torrei conspecific with decoratus
and de-scribed drapetiscus as a subspecies ot
decoratus. Grant (1956:247-248) had previously
discussed the status of gibbus ( which had been
reported from Cuba on the basis of spotted
males), and stated that Cuban specimens of
"gibbus" were in actuality sported male torrei.
Thus, the distribution of torrei was extended to
those Bahaman islands where spotted males
(= gibbus) occur.

To complicate the situation further, Grant
( 19.58, 1959a and b) named three species (spiel-
mani, alaijoi, ruibali) from Guantanamo and the
United States Naval Base at the Bahia de Guan-
tanamo. Since torrei had already been recorded
from this area (at Boqueron by Cochran, 1934:
12), the interrelationships of these four species
was at once questioned. Such an abundance of
similar species in one restricted area was not
only unusual, but somewhat suspicious. Since
Grant had not examined previously reported
Oriente material, it was imperative that all these
specimens be restudied to ascertain their status.

From this brief historical summary, it is ob-
vious that the status of at least the Cuban mem-
bers of the complex — and in part their Bahaman
relatives — has been confused. This is clue to tliree
major factors: (1) the relative paucity of speci-
mens from many regions ( a condition which still
persists >; ( 2) lack of centralized source for ma-
terials (no one has ever assembled all available
material together at once); and (3) the inherent
complexity of dealing with a number of syin-
patric (or even svntopic) forms, all of which are
sexually dichromatic and some of which appar-
ently have trenchant pattern changes in males.
It speaks highly for the perspicacity of Major
Grant that lie was able to discern differences
between lizards where others had failed. Had he
examined in detail specimens other than his own,
the decoratus complex would likely be better
understood.

Geckos of the West Indies

PROCEDURES AND TERMINOLOGY

We used the method of counting body scales
first proposed by Grant (1937:507) but first
used consistently and further elaborated upon
by King (1962). Dorsal scales are counted dor-
solaterally on a line from axilla to groin; ventrals
are counted in the midline from the level of the
axilla to that of the groin; and a count is taken
around the midbody. Internasal scales are the
small scales between the enlarged nasals which
border the rostral posteriorly. Otherwise the
counts taken are standard or self-explanatory.

The group of geckos dealt with here generally
has been considered to have granular scales in
contrast to some other forms which have flat-
tened and imbricate scales. Although the two
extremes are quite different, there is no true
dichotomy, and transitions from one to the other
can be found. Thus none of the forms considered
can be said to have simply granular dorsal body
scales ( such as are found on the head ) which
are juxtaposed and not imbricate. All forms show
a tendency toward imbrication, even where no
actual overlap of adjacent scales occurs.

We used the following terms (in part after
King, 1962) to categorize the types of dorsal
scales. Scales may be imbricate or granular
(simple granular scales are not typical of any
of the forms in this assemblage, but those scales
which show a distinct separation from adjacent
ones are so called). Granular scales, as here used,
have the posterior edge somewhat raised ( true
nondirectional granules, when present, are usual-
ly the result of injury ) . They may be swollen or
flattened, or they may be erected or flat-lying.
Erected scales must in some degree be swollen;
but if scales are strongly erected, somewhat
swollen but still flattened, imbricate, and close
packed (serried), the term papillate is used.
Extremely erected granular scales are termed
conical. In outline (i.e., viewed normal to the
plane of the scale), scales are categorized as
rounded, acute, or obtuse. Scales may be smooth
or keeled, although keeling in the group is usual-
ly very weak and when present may not be evi-
dent on all of the dorsal scales.

Various categories of scales may be found on
a single individual. For instance, the more granu-
lar scales are usually found in the middorsal

zone, while on the sides they tend to be more
imbricate and erected. Nevertheless, there are
distinct trends in some forms on which certain
categories of scales occur over a greater surface
of the body and with greater frequency than on
other forms.

The decoratus group of Sphacrodactijlus is
characterized by

( 1 ) Large size

(2) Sexual dichromatism

(3) Ontogenetic pattern changes in both
sexes ( greater in males )

(4) Banded patterns (basically), i.e., females
and juveniles

(5) Basically granular dorsal scales (with
nonabsolute trends in the various species
toward flattening, imbrication, keeling,
or swelling and erection )

(6) Large compact escutcheons with little or
no extension onto thighs

(7) Few large hair-bearing scale organs
around the apex of dorsal bodv scales

Sphaerodacti/lus ruibali differs in some of
these requisites, but we include it as a periph-
eral member of the decoratus group because
of its apparent relationship to intermedins.
Sphacrodactijlus cincrcus, a large, granular-
scaled form, is not included in the decoratus
group because of strong differences in the pat-
tern ( although the basic juvenile pattern is
banded), scalation, and ungual sheath. Sphacro-
dactijlus cinereus may be a peripheral member
of this group, but its relationships are not ob-
viously with the forms here included, and it
does not demand discussion for that reason.

The sequence in which we discussed the
forms in this study is not intended to be phylo-
genetic. Rather, we started with decoratus,
which, as the oldest named and most widespread
geographically, makes an ideal starting point.
Thereafter, the forms are associated roughly ac-
cording to relationships as we see them — inso-
far as can be done in a linear sequence.

Bricham Vdi \(. t MVERsrn Scienci Bulletin
THE BAH WI.W sill \TI<>\

Tin- Bahama Islands are for the most pari
clustered in loose associations mi shallow banks.
Slight changes in sea level would cause profound
changes in tin- land exposure oi this archipelago.
The Mali. im. is are presumed in have been nun-
pletely <>r nearh completely submerged during
tin- Pliocene and Pleistocene i Rabband Hayden,
1957 8). The relatively slight changes in sea
level necessary to produce alterations in the land
area pins the probable effect of current erosion
in enlarging channels (Clench, 1938:485) may
well mean that proximal parts ol adjacent islands
have been only recently separated. These parts
may show closer faunal affinities than parts ot a
presently continuous land mass which may have
been united recently after having been separated
tor a long period. For instance, the geckos of the
south tip ol Eleuthera are more similar to those
of Cat Island than to those of the rest of Eleu-
thera. Both of these islands are now separated by
a substantial water gap lint connected liv a shal-
low- strip (Clench, 1938) which may indicate
that their association has been closer in the past.
\lso, it appears thai islands may hear inter-
mediate populations which have only recentK
become isolated, and so while not true intcr-
grades, they still have the characteristics of the
original intergradient populations.

Sphaerodactylus decoratus Carman

Sphaerodactylus decoratus Garman, 188S, Bull.
Essex. Inst., 20:111. Type locality. Ruin Cay,
Bahama Islands; type specimen MCZ 6221).

Definition. Dorsal scales granular to swollen,
rounded to acute and slightly imbricate; some-
times with conical or papillate scales, especially
on posterior flanks; taint keeling present (in
some specimens only in a small area, usually the
lumbar region | in majorit) of specimens; large
hair-bearing organs | three hairs) present on
posterior face ol scales; dorsals, axilla to groin,
|fi-(il). ventral scales | including those of throat
and chest I smooth, flattened, acute to rounded
and imbricate (axilla to groin. 29-46); midbocK
scales 61-89; dorsal scales ol tail smooth. Hat-
King, imbricate and rounded to obtuse; scales
beneath tail smooth, rounded, imbricate and en-
larged midventrally, Internasals 0 to 1. mode 1
or 2. upper labials to mid-eye 3 to 5, mode );
escutcheon large, compact, roughly triangular

and with virtually no extensions onto thighs (.">

Hi \ (> 2T Habitus stout; snout moderate in

length and width, size moderate to large I" K)

mm snout-vent >.

female coloration; Dorsal ground color light

tan. gray, gray-brown); dark crossbands (3 to

5, axilla to groin. 5 to 7, snout to groin) some
shade of darker brown with black edges, ocelli in
bands white, with or without darker edges. Dark
facial coloration extends to level just posterior to
eyes; light areas on snout delimit loreal dark
Stripe and median snout stripe which lorks pos-
teriorly; narrow light band just behind facial
markings followed by wide dark head band with
ear situated about midway ol its width. D.irk
neck band with two prominent paramedian light
ocelli. Venters oft-white to gray; dorsal mark-
ings extend onto venter ill chin and throat region,
usually becoming scrambled medially, forming
a marbled pattern or fading out centrally. Juve-
nile coloration is like that ot female but markings
are more solid and contrasting.

Male coloration: ( f ) Unicolor dark to yellow-
ish brown to bluish gray, tail and head more
yellow, or (2) as just described but with pattern
of scattered black or dark brown spots.

Range. Eastern Cuba from central Camagiiey

east, and the Bahamas north of the Crooked
Island Passage (except San Salvador and unre-
ported from many smaller islets) and excluding
the Little Bahama Bank ( Fig. 1 ) .

Remarks. The dichromatism and particularly
the variation in the males has caused consider-
able confusion. On the basis of spotted males
from the Bahamas, Barbour ( 1921 ) named S.
gibbus and later recorded it from Cuba. Grant
(19.56, 1959a) showed that S. decoratus was
sexually dichromatic and interpreted the spotted
pattern on males to be a transitory ontogenetic
character. The juvenile pattern was thought to
give way to the spotted which in turn was lost
in older unicolor males. This sequence was pre-
sumed to have been demonstrated in a photo-
graph i Grant, 1956: Fig. 1 ) of 12 males from the
Bancs area showing various degrees of spotting
or its lack. We examined oxer 100 specimens from
northern Oriente. most of them taken by (•rant,
and all males .ire spotted or are becoming so
with the exception of a single series ol 15 speci-
mens that contains five unspotted males. We
have reason to doubt the data for this scries
i see below). To us the evidence indicates that
the degree ol spotting is constant in the indivi-
dual male after maturity, and its incidence in a
population is constant and often geographically

Geckos of the West Indies

circumscribed. In some populations males vary
from completely spotted to completely unspotted,
but in others either spotting or its lack is the pre-
dominant or exclusive condition. Bahaman popu-
lations exist in which the preponderance of
spotted or unspotted males has remained con-
stant in repeated collections over the past half-
century.

In commenting on the specific distinctiveness
of drapetiscus (=intcrmcdius, vide infra) Grant
( 1959a ) stated that the number of bodv bands
is "fixed in each group" ( presumably he meant
species). On the basis of our broader survey of
the group, we find this not to be so; band num-
ber, although constant in some species, is vari-
able within single populations of some forms.

Sphaerodacti/lus decoratus decoratus
Garman
Sphaerodactylus gibbus Barbour, 1921, Mem.
Mus. Comp. Zool., 47:228.

Definition. A race of S. decoratus charac-
terized by moderate size; 3 or 4 bands ( modallv
4) between axilla and groin in females; bands
relatively light, not heavy and dark, bordered
posteriorly bv light punetulations; ocelli lacking
on body bands or present but small on the first
band; throats faintlv marked or not at all, and
collar not prominentlv cleft middorsallv; males
spotted.

Range. The Exuina chain from Warderick
Wells Cay south, Long Island, the Ragged Is-
lands (known only from Great Ragged), and
Rum Cav (Fig. 1). Records are not continuous

Fig. 1. Map of the Bahama Islands. Solid symbols indicate localities whence specimens of Sphaerodactylus decora-
tus have been examined. Ranges of subspecies are as follows: S. d. decoratus, vertical lines; S. d. flavicaudus,
diagonal lines; S. d. aiessares, horizontal lines. Overlap of symbols in the northern Exuma Cays indicates area
of atessarcs influence on S. d. decoratus. Cat Island and the southernmost locality symbol on Eleuthera Island
remain unshaded (see text for discussion).

Bricham loi \i. University Scienci Bulletin

| see specimens examined I and much ol tins
range is presumptive. \lso the range is some
what arbitrarily defined in the Exumas i see
remarks below l.

Variation. In <>| specimens, dorsal scales
axilla to groin, are 48-67, mean 59.0; ventrals,
axilla to groin, 31-46, mean 37.1; midbody scales
67-86, mean 76.1. Internasals 1-4, mode 1 (Ex-
umas) or 2 i Long); upper labials to mid-eye 3.
4 ( mode) or 5. The largest male is 34 mm snout-
vent, tin- largest female 37 mm snout-vent.
Escutcheons compact with little extension onto
thighs, range 6-8 X 7-13.

Female coloration | Fig. J : Dark hands be
tween axilla and groin 3(11 specimens t, 3/4 (4
specimens ) . 4 (37 specimens I. Bands are typical-
ly light (brown to grav brown ), minutely flecked
with light, with a narrow black edge, and a
border of buffv spots along the posterior edge;
ocelli are white and minute when present (sel-
dom); ground color light grav to tan. Collars
typically darker than body hands, a tew speci-
mens show some middorsal splitting of collar;
ocelli white and small, not always present. Tails
banded with light grav to white and black.
Throats faintlv marbled in few specimens but
usually lack markings. Males tan to brown or
light grav. with a well-developed pattern of black

spots i Fig. 3); a few specimens at tin- extremes
iion have only sparse spotting one speci-
men from Big Farmers Cay lacks spots, ground

color ot head dull yellow, tail usually brighter.

Ins color rich brown dolors given are from
notes taken on Great Exuma specimen

Aberrant head pattern variants, in which the

facial hand is joined to the head hand by two
dorsolateral bridges of dark pigment or in which
the light interband between these two elements
has a pair of dorsolateral diverticula into the head
hand, occur in a high percentage of Long Island
specimens. These variants are found more rarelv

elsew here, .mil one occurs in the single specimen

from Great Ragged.

Remarks. The exact limits of the range of
this subspecies cannot be presently given. We
have rather arbitrarily defined the northern limit
in the Exuma chain. Specimens from the- upper
Exumas, particularly Feat Cay, we believe to
show the influence of the Flenthera race in that
they have only three body bands, modally two
internasals, and the only male is unspotted. How-
ever, since the shift in internasal modality actual-
ly occurs much further down the chain, it is pos-
sible that the transition is more gradual or that
various populations on the Exuma Cays have di-
verged independently to some extent. The Rag-

Figs. 2-.1 Fig 2. Sphaerodactulus decoratus decoratus, female dors.il ci,-«, ASKS V7UU8, 0.8 mi NW George
Town. Cn-.it Exuma Island Bahamas. Fig, 3. S. d. decoratus, male dorsal view; ASKS V7070, 9.8 mi NW
George Town, Great Exuma Island, Bahamas. Kij». 4. S. decoratus subsp., female dorsal view; MCZ 39554.
Arthur's Town. Cat Island, Bahamas, Fig. 5. S. (/. utvwi/rev. female, dorsal view; MCZ 811(H). holotype, 4 mi
\ 2 I mi E Rock Sound, Kindlier,! Island, Bahamas.

Geckos of the West Indies

ged Islands are included on the basis of only one
specimen from Great Ragged at the southern end
of the chain; it agrees with S. (/. decoratus as
presently defined, but obviously nothing can be
said about variation in the Raggeds. Long shows
some divergence from the Exuma populations in
internasal modality and coloration (splitting of
the collar); it shows some influence of the rather
divergent Cat Island population, which is not
subspecifically allocated herein. We have called
this cluster of populations decoratus because the
single juvenile type from Rum Cay, which is
geographically close, falls within the variation of
these populations as far as can be determined.
Should the Rum Cay population be found to be
distinct, gibbus Barbour will be the next avail-
able name for those populations we are calling
S. d. decoratus.

Specimens examined: Exuma Cays: Warder-
ick Wells Cay, AMNH 76237; Great Guana Cay,
CM 41046, 41058; Big Farmers Cay, AMNH
76232-36 (12 specimens), CM 41019-20; Cave
Cay, CM 41022; Darby Island, UMMZ 117017-
18 (8 specimens); Great Exuma: 9.8 mi NW
George Town, ASFS V6978-81, V7070-75; approx.
2 mi SE Rolle Town, ASFS V6987-92; 0.8 mi NW
George Town, ASFS V7001-09; 3.2 mi NW
George Town, ASFS V7012-14, RT 1387; approx.
4 mi NW George Town, ASFS V7020; 1.0 mi SE
Moss Town, ASFS V7095; Stocky (Stocking?)
Island, MCZ 13436 (type of gibbus), 13437
( paratype ) ; cay off Roseville ( Rolleville? ) , MCZ
13435 (paratype); Little Exuma: 5.7 mi SE The
Ferry, ASFS V7045-50; Hog Cay, MCZ 55620;
Long Island: MCZ 37957; Simms.CNHM 22744-
47, MCZ 42278-79, 84396; Grays Settlement,
ASFS V6812; 2 mi E Grays Settlement, ASFS
V8613-28; Deadman's Settlement, UMMZ
117020; Clarence Town, MCZ 42280-81, UMMZ
117019 (2 specimens); Rum Cay: MCZ 6220 (type
of decoratus); Great Ragged Island: UMMZ
118029.

The population of S. decoratus on Cat Island
shows certain peculiarities of its own. Most not-
able is the greater degree of middorsal splitting
of the collar (Fig. 4); this appears to be onto-
genetic, and the majority of adult females show
nearly complete division of the collar, which is a
greater incidence than found in other popula-
tions of the species (the Bahaman races are the
only ones in which the trend is present). Some
specimens from Long Island and the Exumas
show a trend in this direction, with that of Long
being strongest. More complete collections may
show a definable population to exist on Cat and

the proximal northwestern end of Long (in the
present sample greater splitting of the neck band
appears to occur in the north). Of the six Cat
Island males four are unspotted and two have
some indication of spotting. This also suggests
some divergence from the bulk of the material
assigned to S. d. decoratus, but it is not conclu-
sive. Body bands are 3/4 in 3 specimens, 4 in 20
specimens, and 4/5 in 2 specimens. This is a
trend toward higher numbers of bands than is
found in the other races. For the Cat Island
specimens dorsals are 49-64, mean 54.8; ventrals
35-46, mean 40.1; midbodv scales 70-85, mean
76.5; fourth toe lamellae 7-13, mode 10; inter-
nasals 1-4, mode 1; escutcheons 5-9 X 7-11.

Sphaerodactylus decoratus flavicaudus
Barbour

Sphaerodactylus flavicaudus Barbour, 1904, Bull.
Mus. Comp. Zool., 46:56. Type locality, Man-
grove Cay, Andros Island; cotypes, MCZ 84385-
95, 6953, 13564, UMMZ 107614.

Definition. A moderately sized race of
decoratus characterized by 3 or 4 body bands,
relatively light, not heavy and dark, bordered
posteriorly bv light punctulations; throats of fe-
males moderately to slightly marked; collar not
prominently cleft midtlorsallv; males unspotted.

Range. Known from the islands of Andros
and South Bimini ( Fig. 1 ) ; may be reasonably
expected on the rest of the Bimini chain with
the possible exception of North Bimini.

Variation. In 60 specimens, dorsals range
from 46-66, mean 56.1; ventrals 31-42, mean 36.9;
midbody scales 61-86, mean 73.6; fourth toe
lamellae 8-13, mode 10; internasals 1-2, mode 1;
upper labials 3-5, mode 4; escutcheon 6-8 X 7-13.
Largest male and female 35 mm snout-vent.

Female coloration: Ground color gray to tan-
nish-grav; bands wood brown with narrow, dark,
not sharply distinct edges and bordered poster-
iorly by light flecks. Body bands 3 ( 8 specimens ) ,
3/4 (13 specimens), or 4 (34 specimens). Collar
black in juveniles, dark brown in adults with
white ocelli, middorsal break lacking or not
prominent. Ocelli of moderate size and promi-
nence, sometimes present on first body band. A
distinct but not boldly contrasting throat pattern
is present in a few females, but in most the
markings are faint or virtually lacking.

Male coloration: Dull brown; head, throat
and tail pale yellow. None of the extensive series

Bhk.h wi Young University Sciencj Bulletin

ol males "I this form is spotted, Ins coloi in
Biminj specimens was noted .is "greenish-graj to
silver) I \ll above colors are from field notes
on South Biminj specimens. I

( Comparisons. Sphaerodactylua </ flat it audus
differs from decoratus primarily in having un
sported males. In scale characters there are no
significant differences, except in the number "I
fourth toe lamellae which is modally K) in flavi-
caudus .uid 12 in decoratus. The frequencies for
the different categories oi body handing are
about the same in both. Female flavicaudus tend
to have more prominent throat markings, lint
there is considerable overlap. On the basis oi the

Biminj material there appears to he an iris color

difference, that of flavicaudus greenish grav to
silvery and that of decoratus ( Great Exuma I
rich brown.

Specimens examined. "Bimini," AMXH 73489,
73493-96; South Bimini: western side, ASFS
X4663-69, AMMI 76S73, 68806 (2 specimens),
68807 (2 specimens), 68808 (2 specimens); western
end, ASFS X 4733-39. X4753-60, X47S0-S1. CM
34133; Andros: NIC'/ 6221. 84369; approx. 0.5 mi
N of Nicholl's Town, ASFS V6966-71; west side
Morgans Bluff, ASFS V6974-75; Mangrove Cav:
MCZ 6952, S4375-84, 84385-95 (cotypes), 695.3
(cotype), 1.3.564 (cotype), I'MMZ 107614 (co-
type), AMMI 24715; Bastian Pt. on South Bight,
UMMZ 117022; finder Village (not mapped),
UMMZ 118028 (2 specimens); South Bight,
Driggs Hill shore, I'MMZ 117023 (2 specimens).

Sphaerodactylus decoratus atessares*,
new subspecies

Holotype. MCZ 81100, an adult female from
4 mi N, 2.3 mi E Bock Sound, Eleuthera, Bahama
Islands, collected 5 October llKv> by Richard
Thomas.

Paratypes. Bahama Islands, Eleuthera: ASFS
17557, Alicetown, native, 6 November 1961;
AMXH 69245-46, 692-49, Hatchet Bay, G. Camp-
bell, 5 April - 17 July 1948; ASFS V6805-06, 6
mi N Tarpum Bay, R. Thomas, 3 October 1965;
\M''S V6833-40, V 6842-54, same data as type;
ISWI 15789.5-902, MCZ 81101-09, same locality
and collector as type, 7 October 1965; II' 21.504-
09, IIMXII 61651-58, CM 40586-90, ASKS

V6946-53. same locality and collector as type, 9

October 1985; KU 93349-51, 1.5 mi E, 2.9 mi S
Hock Sound, R. Thomas, 7 October 1965; KU

93,348, 5 mi XXW Southeast Point, R. Thomas,
4 October 196.5.

Associated specimens. New Providence: MCZ
6973. I MMZ L0074J Nassau, Ft Charlotte.

Rose Maud: (near New Providence) UMMZ

117021.

Diagnosis. [Tie largest subspecies oi decora-
tus, characterized by possessing only three body
bands, which are nearly solidly pigmented, have
Only a narrow light central /one. and are edged
posteriorly by continuous narrow light areas in-
stead of discrete punctulations; bands in juve-
niles solid, not lightened by taint light stippling.

Range. The island oi Eleuthera, excluding

the southern tip ( Fig. 1 ); there arc no specimens
from north of Hatchet Bay; also New Providence
and Rose Island, at least tentatively.

Description of type (Fig. 5). An adult female.
39 mm snout-vent length, total length 73 mm;
dorsal scales, axilla to groin, 63; ventrals 37, mid-
body scales 77; 10 fourth toe lamellae; four
labials to mid-eve on either side; two intern. isals.
Median dark snout stripe present, slightly forked
posteriorly, joins dark interocular area at level
of posterior part of eyelid; head band black,
slightly lightened centrally; collar nearly solid
black with no median interuption, ocelli small,
white; three body bands with wide dark mar-
gins and narrow, indistinct central light area;
light outlines to dark bands give appearance of
accessory indistinct dark bands in middle of
interband areas. Faint small ocelli present on
all body bands. Tail alternately banded with
black and white; throat markings indistinct; ven-
ter light grayish,

Variation. In 60 paratypes dorsal scales
range from .50-69, mean 58.6; ventrals 34-44.
mean 37.9; midbodv scales 64-88. mean 76.0;
fourth toe lamellae 8-14, mode 10; intcrnasals
1-3, mode 2, escutcheons compact as in the rest
of the species, 5-9 scales long and 6-13 wide.

Female coloration: Most female paratypes
arc similar in color and pattern to the type. The
most notable variant is a tendency toward frag-
mentation of the body pattern in some large
specimens in which the dark areas are sharply
set off, and the ground color lighter. Fight flecks
form a posterior border to the dark stripes in-
stead of a continuous light edge in a tew speci-
mens, but the flecks arc not so prominent and
discrete as in other races. Small paired ocelli are
found in all three body bands in about one third
oi the specimens, others have ocelli in the first
or second body band, and others in none. In life,
females were noted as having a light grav to

Geckos of the West Indies

gray-brown ground color, bands dark brown,
variably hollowed with lighter brown; venters
pinkish gray; tails banded black and white. Juv-
eniles have the female coloration but the bands
are solid except for the occasional presence of
the minute paired ocelli, and the interbands are
lighter, almost white, sometimes with a pinkish
tinge, and with no central darkening; transition
to adult coloration is gradual.

Male coloration: Males vary from completely
unspotted to completely spotted; the ground col-
or is brown to gray brown or light gray, with
the heads dull yellow and tails usually somewhat
brighter yellow. Iris color for specimens of ates-
sares was noted as brown.

Comparisons. In possessing only three body
bands atessares differs in a strongly modal man-
ner from all other Bahaman races. The heavily
pigmented body bands differ from those of the
vast majority' of specimens of the other races;
the small, irregular and relatively sharply set off
central light areas of the dark bands also differ
from the more extensive light areas and narrower
dark margins of the other forms. The collars of
atessares show no tendency toward middorsal
splitting as is found in at least some specimens
of the other Bahaman forms, especially in the
Cat Island population.

The discrete light flecks that form the pos-
terior margins of the dark bands in the other
Bahaman populations do not characterize atess-
ares. Even where there is some indication of
punctulations, the light border is more continu-
ous and the flecks not so prominent as in the
other subspecies. The solid banding and con-
trasting ground color of the juvenile atessares
apparently is a constant difference, as the dark
bands of the juveniles of the other Bahaman
populations are stippled with light areas, and
result in a hazier, less substantial appearance to
the pattern. Although male atessares may be
thoroughly covered with spots, they never seem
to attain the neatness and regularity found in
some spotted males of decoratus. Sphaerodac-
tylus d. atessares exceeds all other members of

the species in size, the maximum snout-vent
length being 40 mm (six specimens). The maxi-
mum size for the other races varies but does not
exceed 37 mm. The impression received when col-
lecting the paratypes was that they were distinct-
ly larger than other members of the species which
had been collected. In scalation the differences
are slight; the modal internasal condition of 2
differs from other races discussed except for the
Long Island population of decoratus; the modal
fourth toe lamellar count of 10 differs from that
of decoratus (12) but not from flavicadus.

Eleven freshly collected geckos from the
southern tip of Eleuthera (Southeast Point) are
not assignable to atessares. The two males are
unspotted and very pallid. Of the remaining
specimens, three have four body bands, two have
four bands on one side, and three have three
bands. All (except the males) show the lighter
body bands bordered posteriorly by light flecks
which are not characteristic of atessares. The one
adult female has a nearly completely divided
collar. It is our belief that these specimens show
the influence (apparently strong) of the Cat
Island geckos immediately to the south.

Two specimens from New Providence and
one from Bose Island near New Providence are
provisionally referred to atessares. One of the
New Providence specimens has three heavy dark
body bands, the other is a juvenile in poor con-
dition which has four bands. Little else can be
determined. The Bose Island specimen has three
bands but is not typical of atessares in all ways.
Only additional material from these islands will
show their true affinities.

Six specimens from Leaf Cay and one from
Little Norman's Cay in the Upper Exuma chain
( which closely approaches Eleuthera ) may show
the influence of atessares. They have three body
bands and two intemasals, and the one male is
unspotted. Otherwise they bear no particularly
close resemblance to atessares. They likely are
the result of past genetic continuity with the
Eleuthera race, and the resultant influence may
extend some distance down the Exuma chain
(see remarks under S. d. decoratus).

THE CUBAN SITUATION

Unfortunately for systematists, the city of
Guantanamo in Oriente Province has the same
name as the extensive bay to the south, which
has in turn bestowed its name upon the United
States Naval Base located at the mouth of the

bay. The city is situated well inland (about 18
miles), nearly in the foothills of the Sierra del
Guaso which here forms the southern margin of
the mountainous interior of the province. To the
south and slightly to the east of the city lies

1(1

linK.ii wi Voi \(. I mm Usui Sc ii \( i Bulletin

the bay, extending about 13 miles inland in a
north-northeasterly direction, and .it its entrance
occupying both tides is the I S. naval base (al
though the main pari ol the Base is on the easl
side oi the ba) I The naval base, then, is at besl
(straight line distance) about 13 miles From the
city, and such a path would necessitate cutting
across the ba) itself, Thus the naval base is a
substantial distance from the city, and its princi-
pal part is on the opposite side oi a very signifj
cant physical harrier.

To mU\ to the general uncertainty ol most
herpetological collections from the region, in
which people have in the past tailed to distin-
guish a city from its surrounding countryside,

we find that some (and tear that others! have
not distinguished between the city, the haw and
the naval base bearing the name Guantanamo.
For example, the type locality of Sphaerodacty-
lus alayoi, as stated in the original description, is
the naval base hut the locality of the holotype as
catalogued is "Guantanamo." There appears to
he no simple way of determining the truth of
the matter, for collections apparently were made
by Grant at both the naval base- and the city
1 e.g., the type locality of spielmani, fide the or-
iginal description is "Guantanamo"). To confuse
matters further, virtually no collectors have made
any effort to specify where on the naval base
specimens were collected — an elemental consid-
eration for anyone hoping to further scientific
knowledge. It is probable that most collections
labelled "U. S. Naval Base, Guantanamo Bay,"
or some such, came from the eastern side of the
hay, hut then one can never be sure. In situa-
tions such as we face in the present stuck, it
would he most helpful to he certain. In conclu-
sion, knowledge of the complex and we do in-
deed mean complex | of sphaerodactvls occur-
ring in the vicinity ol Guantanamo must remain
to a certain degree in contusion. Our results in
some instances are tentative, and in such cases
we will make ready reference to the' state ol af-
fairs.

Barbour ( 1914) applied the Inst name to this
complex of handed Cuban tonus when he named
torrei. His description of the type said that it
had five' dark transverse hands from snout to
sacrum. Additionally, and perhaps most perti-
nent, he never figured the type hut only other
specimens that he presumed t<> pertain to this
form. Thus, when Grant 1 1958) described spiel-
mani he assumed the name torrei to applv to a
three-handed 1 between axilla and groin, six from
snout to i^rom | form as illustrated hv Barhour

I 1921). Unfortunately the assumption w .ts in

the type and paratypes ol torrei have
two hands between the axilla and groin and per-
tain to the form to which (.rant assigned the

name tpielmani. Thus tpielmani is a junior syn-
onym oi torrei. With tin- more widespread three-
handed form ol Cuba we associate the name S.

decoratus.

Sphaerodactylus decoratus granti,
new subspecies

Holotype. BYU 17233, an adult female, col-
lected at Bancs, ( )riente Province-, Cuba, bv
Chapman Grant, January, 195.S.

Paratvpes. Cuba: Camaguev Province: MCZ

57339-43; 7 km S Playa Santa Lucia. R. Molina
and R. Ruibal, 9 July 1957; MCZ 59316. 15 km
S Playa Santa Lucia, R. Molina, E. K. Williams,
R. Ruibal, 24 August 1959; Oriente Province:
MCZ 36941-42, Las Calabazas, Holguin, S.
\guavo, 4 June 1923; BYU 17192-2.32. 17234-
17268, 22889-22920, and two untagged juveniles,
same data as type; AMXII 61404-10, Marcane.
C. M. Jervis. 30 January 1930.

Definition. A race of S. decoratus charac-
terized bv moderate size, three uniformly col-
ored, brown body hands with sharp chirk edges
and with prominent, dark-edged paired ocelli
usually on each hand; hands not prominently
edged with light or with posterior light flecks;
collar not black hut colored like body bands;
throats prominently marbled in the majority of
specimens; males spotted.

Range. Presently known from extreme north-
eastern Camaguev Province, the northern half
ot Oriente Province whence it is presumed to
follow areas of low or intermediate elevation
around the western edge of the central Oriente
mountains (Sierra de N'ipe and its southern
affiliates) to the vicinity of the city of Guan-
tanamo ( Fig. 6).

Description of type (Fig. 7). An adult female,
27 mm snout vent, 51 mm total length; dorsals,
axilla to groin, 51; ventrals 37; midbodv scales
81; fourth toe lamellae 12, upper labials 4/4,
one intcmasal. Coloration: Head, neck and three
body hands hrovv n with distinct dark edges.
prominent dark-edged, paired, cream-colored
ocelli present on neck hand, and present hut less
prominent on 3 bodv hands; snout stripe com-
plete, forked posteriorly, lading into dark inter-
ocular area, head coloration extending onto
underside ol head with hand edges retaining
their alignment hut mixing centrally, resulting

Geckos of the West Indies

11

Fig. 6. Map of Cuba. Solid symbols indicate locality records for Sphaerodactylus decoratus, hollow symbols for
S. intermedins . Ranges of subspecies of S. decoratus are as follows: S. d. grant i, horizontal lines; S. d. lisso-
desmus, diagonal lines; S. d. strategus, vertical lines. Overlap of symbols of granti and strategus in the vicinity
of Guantanamo indicates area of intergradation.

in prominent brown on cream marbling; ground
color of body tan to pale, nearly cream; body
bands narrowly outlined in light; tail banded
with alternating dark-edged brown bands and
tan interbands; venter cream, faintlv stippled
with dark pigment.

Variation. Of the 51 specimens counted dor-
sal scales are 47-66, mean 56.3; ventrals 29-41,
mean 35.0; midbody scales 69-89, mean 78.1;
fourth toe lamellae 7-13, mode 10 or 11; inter-
nasals 0-2, mode 1; escutcheons 5-10 X 8-17. The
largest specimens of both sexes are 32 mm snout-
vent.

Female coloration: The other female speci-
mens are very similar to the type in coloration.
There is some variation in intensity of the dark
edges to the bands, and some have accessory
dark spotting in the bands. Ocelli occur on the
neck bands and all three body bands in over
half of the specimens; in the rest of the ocelli
usually occur as far posteriorly as the second
body band. No adult specimen and few juveniles
are without some indication of ocelli on at least
the first two body bands. Throat patterns are
prominent in all but a very few specimens, and
in those they are indicated. As is typical of the
species, juvenile patterns are more solid, and
hatchlings appear banded black and white.

Male coloration: Adult males are spotted;
spots vary in size, some large or distinctly small-
er and more numerous; most are fully spotted
dorsally, but a common variant is the lack of
spotting on the head. Variation in degree of
spotting includes a few which are sparsely
spotted and five which have no spotting at all.

These five are all part of a series ( BYU 22927-
41 ) which was catalogued at a later time than
the balance of the specimens, and all of this
series are badlv dessicated. It is evident that
their history has in some way been different
from the other specimens. The incidence of un-
spotted individuals (5 out of 15) compared to
no unspotted specimens among the other north-
ern Oriente males (37 specimens) throws fur-
ther doubt on these specimens. Additionally, the
series in question immediately follows in the
catalogue a series of six (BYU 22921-26) also
much dessicated S. notatus, whose locality is
given as Guantanamo. Thus, we regard this
series of males supposedly from Banes with high
suspicion and believe that their locality is prob-
ably Guantanamo.

Comparisons. From decoratus and flavicau-
dus, granti differs in having only three body
bands ( versus a mode of four ) , and in having the
bands more distinctly dark edged and more
regular in outline. In gianti there is a lack of the
hazy aspect of the pattern of these two Bahaman
races and the punctulate posterior border to the
body bands. The ocelli are more frequent and
more prominent in granti and the collars are
never solid black as they often are in all of the
Bahaman races. The boldness of the throat pat-
tern is also greater in granti. Sphaerodactylus d.
granti is distinctly smaller than the Bahaman
races but the contrast with atessares is most
striking. In atessares there are heavily darkened
margins to the body bands with smaller and
irregular light centers as opposed to the wide,
lighter bands with distinct narrow dark edges
of granti.

12

HlUc.ll \\l Vol NC I \» I Hsi 1 1 Si ii \i I Hi I l.l II \

Sphaerodactylus decora t us lissodcsmus .
new subspecies

Holotype. M< Z 57344 Sierra de Cubitas near
H. in. in. Camagiie) Province Cuba, collected 22
August, 1957 by R. Molina and R. Ruibal.

Paratype. M< / 57345, same data .is type.

Definition. \ race ol S decoratus charac
terized by having three light brown body bands
with dark brown edges fading into the color ol
(In- band, not abruptl) set <>tl from it; ocelli in
distinct or lacking; head band joined or approxi-
mated to dark facial area by two or three
"bridges" ol pigment near the dorsal midline;
posterior light interband ol the head distinctly
wider than the centra] head band; snout region
not heavil) pigmented and extending onto under-
side of head along margin ol lower jaw instead
ot to midline. Males unspotted.

Range, known only from the Sierra de
Cubitas in Camagiiey Province, Cuba (Fig. 6).

Description of typo (Fig. 8). An adult female
29 mm snout-vent; tail missing; dorsal stales.
axilla to groin, 53; ventrals 33; midbody scales
S4; fourth toe lamellae 11; upper labials 1 4; one

intiin.is.il Coloration: Ground color pale tan,
three bod) bands brown, smooth edged and

with narrow dark edges that t.ule gradually
into wide centra] area ol bands; collar with
heavy dark edges that lade into a light center,
and with one small, not dark-edged ocellus on
the right side. Facial pattern much simplified:
loreal and median snout stripes join in the mid-
line at a level posterioi to the eyes awA form an
elongate trident like figure with a short pos-
terior spue (hat nearly meets the dark head hand.
which in turn has two small anterior projections
On either side ol the spire; light interband on
head and neck wider than dark head hand, snout
pattern not extending toward midline ol under-
side ol head hut ending along the sides of the

lower jaw.

Variation. The single paratype, a male, is
attaining the unicolor phase, a faint pattern,
much like that ot the female, is yet evident; the
trident-like cephalic figure is present, but the
spire and the projections ol the head hand con-
nect both elements instead ol just approximating
them. Dorsal scales, axilla to groin, I1*, ventrals
36; midbodv scales 79; fourth toe lamellae 12;
labials 4/4; internasals 2. escutcheon evident but
not fullv developed; snout-vent length 30 mm.

Figs 7-10. Fig. 7. Sphaerodactylus decoratus granti, female, clors.il view; BYU 17233, holotype, Banes, Orients
Province Cuba Fig. 8. S */ lissodesmus, female ilnrs.il view; MCZ 54344, holotype, Sierra de Cubitas near
Banao, Camagiie) Province Cuba Fig. 9 S. d. strategus, female dorsal view; MCZ 81110 holotype, east side
of the Bahia de Guantanamo, United States Naval Base, Oriente Province, Cuba. Fig. It'. S. atayoi, female,
dorsal view i i\i\ll 44219, paratype, U. S Naval Base, Guantanamo, Oriente Province Cuba.

Geckos of the West Indies

13

Comparisons. Although only two specimens
assignable to lissodesmus are available, they are
so different from the other races of decoratus
that we have little doubt that they represent a
distinct form endemic to the limestone massif
of the Sierra de Cubitas. The complex head pat-
tern joining the facial markings and head band
serve to distinguish lissodesmus from all other
subspecies. The smooth-sided and distinctly
colored body bands are not found, although
rarely approached, in the other races. The slight
extent of the facial pattern onto the throat is
also distinctive and in contrast to the Bahaman
specimens, which frequently have a verv faded,
virtually absent throat pattern. The single male
is unspotted, and if this is true for the popula-
tion, lissodesmus is additionallv distinct from its
nearest geographical relative granti, which has
spotted males.

Sphaerodacttjlus decoratus strategus*,
new subspecies

Holotype. MCZ 81110, an adult female from
the east side of the Bahia de Guantanamo,
United States Naval Base, Oriente Province,
Cuba, collected 24 April 1965, by Bichard Thom-
as.

Paratypes. MCZ 81111, USNM 15903-04,
ASFS V6244, V6246-47, same locality as type, 23
April 1965, B. Thomas, native; ASFS V6256-60,
V6262, same data as type; ASFS Y6275, same lo-
cality as type, 25 April 1965, B. Thomas; MCZ
68732, 69441, same locality as type, B. V. Lando,
June 1962 and August 1958;' UIMM1 44232.
same locality as type, A. Spielman. January 1958.

Associated specimens. USNM 59221-31, Cai-
manera, Oriente Province, Cuba.

Definition. A race of decoratus closely allied
to granti in possessing three wide bodv bands
with narrow black margins, but differing from
that form in that die males are unspotted.

Range. The south Oriente coast, possibly
only that part to the east of the Bahia de Guan-
tanamo (Fig. 6); presumed to intergrade with
S. d. granti in the vicinity of the cirv of Guan-
tanamo.

Description of type (Fig. 9). An adult female,
32 mm snout-vent, total length 60 mm. Dorsal
scales, axilla to groin, 64; ventrals 42; midbody
scales 84; fourth toe lamellae 11; upper labials
3/4; one internasal. Coloration: Three wide body
bands bordered with a sharply distinct narrow

*strategus, from Greek: strategos, admiral

dark margin; ocelli present on collar and in-
distinct on first two body bands; head band
wide, only narrowly separated from collar and
facial markings; snout stripe entire, forked pos-
teriorly and connected with interocular dark
area; head band continues across throat as do
the facial markings; collar interrupted in mid-
line of throat; tail banded with broad dark and
narrow light bands. ( For color in life see be-
low. )

Variation. Dorsal scales, axilla to groin, 50-
67, mean 59.1; ventrals 33-42, mean 36.8; mid-
body scales 73-84, mean 78.3; fourth toe lamellae
8-13, mode 11; escutcheons 7-10 X 8-23. Female
coloration is much the same as that of the type;
one specimen has a split bodv band on one side
giving a 3/4 formula. Specimens were noted in
life as having die dark bands dark brown or
gray-brown, light interbands tan or light gray or
gray-brown; ocelli white; venter gray, light bands
on throat cream; light bands near tip of tail
white. Iris color dark brown. Males were noted
as having the dorsal coloration grav or blue-gray
with faint lighter flecking evident, sometimes
outlining the banded "female" pattern. Heads
were dark dull yellow, tails yellow to dark yel-
low-green; venters light gray; throats and tails
dull yellow beneath. Iris color gray to gray-
brown.

Comparisons. The female coloration of this
form is closest to that of granti on the north
coast. There appears to be an at least average
greater width of the body bands in strategus.
Grant ( 1959a ) commented that the south coast
"torrci" (=zdecoratus as used herein) was light-
er in color than those on the north coast. This is
probably correct but cannot be verified in the
preserved specimens. The specimens from the
city of Guantanamo that are assignable to deco-
ratus (see below, specimens not subspecifically
allocated) are considered to be intergrades be-
tween granti and strategus, as they include both
spotted and unspotted males. The associated
specimens from Caimanera on the west side of
the bay are only tentatively referred to strategus
because of geographical proximity; there are no
males in the series to allocate them definitely.
Otherwise, the comparisons of strategus are for
all practical purposes the same as those of
granti.

The paucity of specimens of S. decoratus
from the southern coast of Oriente is strange,
and it is possible that the primary distribution
of the species in this region is to the east of the

1 1

Hiiii.ii \\i Young I \miwn Si u \< , Hi i i i i in

Bahia de < luantanamo. Uthough there are sped
mm us from Caimanera on the \\<st side ((| the
bay, there is no evidence ol the occurrence "I
S. decoratus any farther to the west. The onh
Grant specimen ol rtrategus thai we know of is
one that was included in the type scries ol S.
alayoi. The fact that Grant collected other forms
Inii apparent!) not man) decoratus on the Naval
Base may indicate that he collected in a differ-
ent ecological situation than that preferred by
decoratus.

Specimens Not Allocated to Subspecies

Specimens examined hut not subspecifically
allocated. S. decoratus: Bahamas: Exuma Cavs,
Leal Cay, WINN 76238-43; Little Norman's
Cay, MCZ 13478; Eleuthera, Southeast Point,
ASKS Y6S12-I7. V6818-22; Cat Island: Orange
Creek, Arthur's low,, MCZ 39556-62; Orange
Creek. 1 .5 mi NW Arthurs low,,, I'MMZ 79444
(5), 79445 ( 1); Arthurs Town, MCZ 39546-55; The
Bight, \SFS V2138-39. S. decoratus granti X
strategus: Cuba. Oriente Province: Cuantanamo,
BYU 17183-91, ,J22927-41. USNM 59<X)4. 63218,
MCZ 8508, 13482-83, UMMZ 90632; Cuantana-
mo, Monte Libano, MCZ 14665.

Sphaerodactylus alayoi Grant

Sphaerodactylus alayoi Orant, 1959, Heqicto-
logica, 15(1):49. Type locality, Cuantanamo
\.,\al H.lsc, U.S.N., Oriente. Cuba; type speci-
men, II MX 1 1 14215.

Definition (males have not been included in
the computations tor reasons that will become
apparent). Dorsal stales swollen, granular to
imbricate above, becoming erected and papillate
mi tin- posterior Hanks, faint keeling present in
adult specimens; dorsals, axilla to groin, 53-60,
mean 55. S (23 specimens ); ventral scales smooth,
flattened, rounded and imbricate, 32-39 axilla
to groin, mean 35.9 (20 specimens); midbody
scales 73-82, mean 76.4 (21 specimens); scales
of tail acute to rounded, smooth, flattened, im-
bricate, enlarged in midventral line; fourth toe
lamellae 6-11, mode 10; intemasals 1-2. mode 1;
upper labials to mid-eye 3-5, modi' 4; escutcheon
large, compact, roughly triangular and with no
extension onto thighs i 5-8 X 9-12); habitus stout.
snout moderate m length and width; size mod-
erate, to 32 mm snout-vent.

Female coloration (Fig. 10): Pattern handed
with three hands between axilla and groin; dark
body bands twice or only slightlv more than
twice the width ot light (cream to tan) inter-

hands, dark hands sharply demarcated with wide
dark borders and relativel) narrow light cen-
tral portions, no d.uk spotting or dark-edged
ocelli present; ocelli usually present ,,, neck
band, not prominently dark-edged, transversely
elongate; ocelli seldom present ,,, bodv bands,
not i onspicuous and not prominently dark-edged
and tending to be transversely elongate; neck
and head hands generally darker and more solid
than body bands, loreal and median snout stripes

present, snout stripe joining interocular dark
area; ventrallv, head stripes ( including dark
area anterior to c\csi usually continue across
throat; neck stripe interrupted, or with only an-
terior dark edge continuing across throat. Young
have solid hands as in other members of this
group. Males which are assigned to this form
are becoming unicolor but have some banded
pattern remnants; others show some degree of
spotting.

Range. Reported only from the type locality

(Fig. 11).

Remarks. The status of S. alayoi is more
questionable to us than that of any of the other
species. Then- is no doubt whatsoever that the
female coloration of this lizard is eminently dis-
tinct from that of S. decoratus which it resembles
most closely. Included in the type series was one

Fig. 1 1 Map of eastern Cuba with detailed inset of the

region between the Bahia do Santiago and the
Bahia de Guantanamo. Solid symbols indicate
localih records for Sphaerodactylus t<>n<i. hollow
symbol for S. ultiyni, vertically semisolid s\ml>ol fur
S. niiluili. horizontally semisolid symbol for S.
ruihnli + S. alayoi, Subspecies of S. tomrt are as

follows: S. ' tOirei, vertical lines; S. t. ocujal, hori-
zontal lines; S. '. Spielmani, diagonal lines.

Geckos of the West Inthes

15

female (UIMNH 44232) which we consider to
be a specimen of decoratus and not an almjoi.
Males are less easily identifiable but remnant
patterns do seem to ally them with the females.
We have assigned two additional male specimens
(MCZ 69438, 69440) to almjoi on the basis of
their resemblance to the male paratypes.

It is possible that alayoi is a local pattern
variant (it has been found nowhere else) or a
very circumscribed race of S. decoratus in the
region of the Bahia de Guantanamo ( see Thomas
and Schwartz, 1966, on S. beaityi). Obviously
we are at a distinct disadvantage in dealing with
alm/oi, for we have not seen it in the field or in
life. Unfortunately neither ecological informa-
tion regarding the type series nor a statement of
other species collected with the type series was
given. We do not know if specimens of decoratus
were found syntopically with alayoi, whether all
specimens of almjoi were found together at one
precise locality (the U.S. naval base is too big
to be so considered), or if found in a varietv of
places with decoratus, ecologically separate, or
whatever. Information on any of these points
could influence our decision. Only carefully doc-
umented collecting and field observation will
determine the truth.

Specimens examined. Cuba, Oriente Province:
UIMNH 44215 (type), U.S. Naval Base, Guan-
tanamo, November, 1957, C. Grant; UIMNH
44216-31, 44233-38, MCZ 61230-31 (female para-
types), U.S. Naval Base, Guantanamo, January,
1958, Andrew Spielman; UIMNH 44240-44
(male paratypes), data same as preceding;
USNM 81822-23, Boqueron. Questionably refer-
red: MCZ 69438, 69440, U.S. Naval Base, Guan-
tanamo.

Sphaerodaetijlm torrei Barbour

Sphacrodactijlus torrei Barbour, 1914, Mem.
Mus. Comp. Zool., Harvard, 44(2) :260. Type
locality, Santiago de Cuba, Oriente Province,
Cuba; type specimen, MCZ 6916.

Definition. Dorsal scales small, granular,
serried, swollen, erect (especially on sides where
the appearance is papillate), and weakly to
strongly imbricate; faint keeling present in a few
specimens; dorsals, axilla to groin, 45-59; dorsal
scales with few large hair-bearing scale organs
(3 hairs) on posterior face of scale. Ventral
scales (including those of throat and chest)
smooth, flattened, acute to rounded and imbri-
cate (axilla to groin, 29-39); midbody scales 68-
82; scales of tail acute to rounded, smooth flat-
tened, imbricate, enlarged in midventral line;

fourth toe lamellae 8-15. Internasals 0-2, mode 1;
upper labials to mid-eye, 3-6, mode 4. Escutch-
eon large, compact, roughly triangular and with
little extension onto thighs (6-12 X 9-27). Hab-
itus stout, snout relatively long and narrow;
size large, to 38 mm snout-vent.

Female coloration: Ground color light tan or
gray; dark crossbands (2-3 axilla to groin, 4-5
snout to groin) solid or lightened centrally but
not with well defined dark edges, ocelli when
present small and not dark edged; narrow light
fringe around dark bands sometimes present.
Head dark to just behind eyes, light areas out-
line loreal and snout lines, the latter being either
complete to interocular dark area or shortened
and not joining interocular area; wide head band
present with ear centrally positioned; neck band
without ocelli or with more than two very small
ocelli, not prominentlv outlined in black. Dark
bands typically end abruptly on underside of
throat or continue completly across; they do not
fade out or meet to form an extensive marbling;
dark color on snout represented beneath by a
small but distinct mental spot. Venter light, tail
banded like body.

Male coloration: Uni color gray to tan on
body; head and tail yellow; venter tan.

Range. The south coast of Oriente Province,
Cuba, from the city of Guantanamo to the vicin-
ity of Cabo Cruz (Fig. 11). Barbour and Rams-
den (1919:120) referred to a specimen of torrei
from Cotorro, which is near La Habana. We
have not seen the specimen ( it was received by
Ramsden and may be in a Cuban collection),
but it was in all probability a well-patterned
specimen of intermedins. The same statement
applies also to a "torrei" from Camoa ( Barbour
and Ramsden, op. cit. :85).

Remarks. In assigning the southwestern
Oriente three-banded geckos to torrei (instead
of decoratus) , we may be accused of being ar-
bitrary or of constructing a flimsy and artificial
taxonomic edifice. Certainly, without the three-
banded population, S. torrei would stand as a
grossly distinguishable form with but two body
bands, and such may be eventually found to be
the truth. However, we were led to our conclu-
sion by the facts that ( 1 ) there are no unques-
tionable specimens of decoratus (i.e., females)
either to the east in the fairly well collected re-
gion of Santiago, or to the northeast, either of
which might indicate a link with the known
Cuban populations of decoratus; (2) as discussed
below the three-banded specimens are similar
in several respects (coloration and modality of

16

I5kk.ii wi Voi M. l m\ i km i -, Si ii m i Bulletin

scalation) to other torrei, and (3) number ol
body l>.ui(ls is nut necessarily species constant

It is |)dssil)|c tint S. decoratua and S. torrei
(or parts oi them) are both members ol one
species, as their ranges are largely allopatric.
This is a possibility that we cannot deny with
complete certainty. However, on the basis of
presently known localities (imprecise though
they may be) there is a slight amount of east
west overlap and apparent sympatrj .it the eitv
of Guantanamo, in which ana the two are
eminentl) distinct, It is the western three-banded
race which causes the most doubt. As we have
defined it. S. torrei is distinguishable from S.
decoratltS by its solid body and neck handing,
not lightened centrally or dark edged, and ocelli,
when present, small, simple, and not dark-edged;
and hv the very simple throat patterns in which
the head handing extends onto the throat but
remains discrete and holdK contrasting, and
does not lade out or meld into an irregular
mottling. Other less precisely enumerable dif-
ferences in habitus and in trends of scalation
i i.e.. dorsal scales of torrei tend to be more
erected and closely pat Iced ) appear to exist.

Sphaerodactylus torrei torrei Barbour

Definition. A subspecies of S. torrei charac-
terized l>v having two dark body bands wider
than the light interspaces, a high frequency of
dorsal keeling in adults, a long median snout
stripe extending from tip of snout to interocular
dark area, and a high modal number of fourth
toe lamellae.

Range. Along the smith Oriente coast from
the vicinity of Santiago tie Cuba east at least to
Playa Juragua | Fig. 1 1 ).

Variation. In 22 specimens, dorsal scales,
axilla to groin, 45-49, mean 52.6; ventrals, axilla
to groin, 31-37, mean 35.3; midhodv scales 68-83,
mean 74.3; some degree of dorsal scale keeling
occurs in 25 of 27 adults. Internasals 0-2, mode
1, upper labials (in series from each side) to
mid-eye 4 (.50), 5 (23), or 6 (1) (number of
labial series in parentheses i. Escutcheons 6-10 X
9-21.

Female coloration (Fig. 12): All patterned
specimens (21) have all dark (black) hands
wider than light interspaces, and median dark
snout stripe joined with dark interocular areas.
Dark hands ol most adults hazily and irregularly

hollowed, outlined hv taint light fringe. Neck

hand continues across throat in most specimens,

posterior head hand broken vcntrallv in most.
and hand at eye level discontinuous across throat

and ends abruptly; dark chin spot extends pos-
tering to include first lower labial.

Male coloration; Gray to tan on bod) \«.u\

and tail yellow or yellowish.

Specimens examined. Oriente Province:

Wl\ll s.c><)|-<k; (6 specimens . \\I\II 94260
(2 specimens), MNH 94261, UtfNH 94263, 4
km \ Santiago de Cuba; MCZ 6916 (type),
MCZ 84370-71 (paratypes), MCZ S4372-7 1.
MCZ 6917, MCZ 1977F AMMI 12546, AMMI
42583-90, UMMZ 90728-29, UMMZ 90631, San-
tiago de Cuba; UMMZ 9072-1 (2 specimens),
Santiago de Cuba, Church of San Francisco;
AMMI 17721, Siboncv; AMMI 94262 (3 speci-
mens ), Playa Juragua, 3.7 mi F Siboney.

Sphaerodactylus torrei oeujal,
new- subspecies

Holotype. USNM 138015, an adult female,
collected at Oeujal, Oriente Province, Cuba, by
J. D. Hardy, 31 August 19.56.

Paratvpes. Cuba, Oriente Province: AMNH
32301, Belie, no collector, no date; USNM
138017, 5 mi N Cabo Cruz, J. D. Hardy, 6 Sep-
tember 1956; UMMZ 90727. 4 mi N Cabo Cruz,
A. Vanderhorst, January, 1939; USNM 81827-28,
Puerto Portillo, P. Bartseh, 29 August 1930; MCZ
42487, coast S of Pico Turquino, P. J. Darlington,
June 19.36; MCZ .50152, 84397, near Pico 'Tur-
quino, J. Acuna. 10-19 June 1936; USNM 138016,
1 mi E Oeujal. J. D. Hardy, 31 August 19.36;
USNM 29759, Guama, B. S. Bowdish, 2 January
1902.

Associated specimens. Cuba, Oriente Prov-
ince: UMMZ 90725, Santiago de Cuba, La
Socapa; UMMZ 90726. Santiago de Cuba, Mono
Castle.

Definition. A subspecies of S. torrei charac-
terized by Inning three dark hands only slightly

broader than light interspaces between axilla
and groin; numerous, very small, not dark-edged,

ocelli in dark hands of adults; median dark snout
line not extending to interocular dark area; male
coloration spotted; dorsal keeling absent in
most specimens.

Range. From the vicinity ol Cabo Cruz east

at least to (mama and probably to the coast
southeast of the city ol Santiago de Cuba (Fig.
11).

Description of type (Fig. 13). An adult female
with snout-vent length of 33 mm, tail missing.
Dorsals, axilla to groin, 52: ventrals. axilla to
groin. 35; midhodv scales 73. fourth toe lamellae

Geckos of the West Indies

17

Figs. 12-14. Fig. 12. Sphaerodactylus torrei torrei, female, dorsal view; MCZ 6916, holotype, Santiago de Cuba,
Oriente Province, Cuba. Fig. 13. S. t. ocujal, female, dorsal view; USNM 138015, holotype, Ocujal, Oriente
Province, Cuba. Fig. 14. S. t. spielmani, female, dorsal view; MCZ 61233, paratvpe, Guantanamo, Oriente
Province, Cuba.

10; internasal 1; 4 supralabials to mid-eye. Dor-
sal scales granular (more so middorsallv), only
slightly imbricate, swollen but not erected, and
keeled only in sacral region; gular, chest and
ventral scales smooth.

Coloration: Median dark snout stripe extends
to level with anterior border of orbits; Ioreal
stripes present, dark band at level of eyes un-
broken medially; posterior head band relatively
narrow, about width of light interspace, hol-
lowed centrally by light suffusion, and with ear
centrally positioned. Neck bands and three body
bands dark brown, approximately same width as
light interspaces, suffused with lighter brown
centrally, and bearing numerous small, not dark-
edged, ocelli. Dorsal ground color brown, lighter
at margins of dark bands. Ventrally a U-shaped
dark spot is confined to mental, followed on
either side by another dark spot occupying cen-
tral half of first lower labial; dark band at eye
level extends onto throat and broken medially in
abrupt (not fading) termination; posterior head
band continues across throat, neck band does
not, and terminates abruptly, as do body bands,
on ventrolateral surfaces; appearance of fading
is given by central light suffusion.

Variation. Dorsals, axilla to groin (5 speci-
mens), 48-54, mean 51.2; venfrals, axilla to groin

(4 specimens), 32-36, mean 34.5; midbody scales
(5 specimens) 73-79, mean 75.5; fourth toe
lamellae 9-14, mode 13; internasals 1-2, mode
1; supralabials 3 (2 labial series) or 4 (20 labial
series); escutcheons 7-9 X 9-12. Coloration of
four adult female paratypes similar to that of
type; median snout line of one (USNM 81828)
touches interocular dark zone by virtue of thin
spire-like extension, but is not broadly confluent
therewith; second chin spots lacking in USNM
29759, but mental spot entirely covers mental
and extends onto first labial. One specimen
( MCZ 42487 ) has somewhat scrambled but still
recognizable body pattern. Juvenile specimens
have very solid bands and lack ocelli, as is
characteristic of juveniles of other forms of this
group. Extent of ventral banding described for
type similar in other specimens, although neck
band continuous across throat in one and eye-
level band in another. Males are spotted with
relatively small, dark brown spots; two speci-
mens transitional from juvenile pattern to spotted
condition, and faint but complete basic patterns
agree in detail with female pattern described.
One specimen (MCZ 84397) has unicolor head,
and of two provisionally referred specimens one
(UMMZ 90725) has only head spotted, body
unicolor. These two specimens will be discussed
below.

IS

Bun. 1 1 v\l 'loi NC Um\ I lis I I v N( II \( I Bulletin

Sphaerodactylus torrei rjiielmani Grant

Sphaerodactylus tpielmani dr. int. IT>S. Herpe-
tologica II I 22Tt. Type locality, Guantanamo,
Oriente Province, Cuba; type specimen I l\l\ll
14105.

Definition. A subspecies <>l S. torrei char-
acterized b) having dark transverse bands nar-
rower (approximate!) hall .is wide) than the
light interspaces, and only two between axilla
and groin; median snout-stripe typicall) short,
never joining interocular dark area; eye-level
band reduced, not continuous across head in
must specimens; males unicolor; dorsal keeling
lacking in majorit) i>l specimens, and low modal
number ol lonrtli toe lamellae.

Range. Known only from the city of Guan-
tanamo but assumed to meet and intergrade
with S. t. torrei somewhere between Playa Jur-
agua and Guantanamo (Fig. LI). The species
is not known from the U.S. Naval Base, as far as
we know, and whether it ranges to the east ol
the citv ol Guantanamo is unknown.

Variation. In 50 specimens, dorsal scales,
axilla to groin. 43-59, mean 53.7; ventrals, axilla
(o groin, 29-37, mean 33.3; midbody scales 68-
7lJ, mean 73.0; fourth toe lamellae 8-15, mode
M). intemasals 0-1, mode I, upper labials to mid-
eve :} (five labial series), 4 (121 series), 5 (98
series!. (> ( (i series). Kscutchcons 6-12 X 11-24.
The largest specimen is a male that measures
38 mm snout-vent length.

Female coloration (Fig. 14 |: Females gener-
ally as diagnosed. Some have lightened centers to
bands or irregular Light spots within them; there
may be tendency toward further reduction ol
IxkIv bands, as some specimens have one or more
ol these broken, reduced, or incomplete. About
88 percent have eye-level band incomplete

across head; some specimens have ventral por-
tion ol band disjunct from dorsal portion; in
some (his leaves onlv narrowly circumoi hita!
dark area. Roughly halt ol specimens have no

first infralabial spot; some have tins spot joined
by narrow extension with ventral ends ol eye-
level band, man) ol those which have no first
labial spot have slight anterior extension ol eye-
level band. Thus the loss o| that spot appears
to involve fusion with the eve level band.

Males arc unicolor tan to <4r.1v brown in their

I inal adult coloi at ion.

Comparisons. Sphaerodactylus t. torrei is dis-
tinguished from <><\q<il by the possession o| two.
rather than three, body bands, bv having the
median snout stripe confluent with the dark in-
terocular region (Fig. 15), by lacking ocelli in
the bands, by having unicolor rather than spotted
males, and bv having the mental spot fused with
the first labial spot. Sphaerodactylus t. spielmani
differs from torrei in having the dark bands
much narrower than the light interspaces
(rather than distinctly wider) and in having the
median snout-stripe short and not connected
with a dark interocular area 1 Fig. L5); m lack-
ing keeling in most specimens spielmani differs
modallv from torrei. at least, it also differs Iroin
both torrei and OCUJal 111 having a modal number
of fourth toe lamellae of 10 rather than L3.

Sphaerodactylus t. ocujal differs from both the

other races in possession of three rather than
two both- bands. The reasons we have included
it in torrei in the absence of obviously intergrad-
ient specimens have been stated in the discussion
of the species. Additionally, we felt that our con-
viction of relationships of the forms should be
expressed trinomially rather than bv proposing
another species tor a region that has already had
described from it a plethora ol species.

1 [g r. Ders.il ,in.l ventral views ,>l heads of three subspecies »f S. torrei. A. S. t. ocujal, MCZ 42487. B, S. t.
torrei, AMNH 12586; C, S 1. tpieUnani, UIMNH 44212

Geckos of the West Indies

1!)

In some respects (shortened snout stripe, the
modal presence of a separate first lower labial
spot, and modal lack of keeling in both forms)
ocujal and spielmani are more similar to one an-
other than to torrei. The two male specimens
provisionally referred to ocujal are of interest
because of their proximity to Santiago de Cuba,
the locality for many of the torrei examined.
Both are from the mouth of the Bahia de San-
tiago. One of them lacks spotting except on the
head, and it may be that this region is an area
of intergradation between the two forms. The
bay would not necessarily be an absolute barrier
at its mouth ( which is verv narrow ) , because the
region has evidently been in a state of geological
flux and different channels have probably existed
as mouths to die bay at different times in the
relatively recent geological past. Therefore an
area of intergradation could exist in a region
encompassing both sides of the mouth of the
Bahia de Santiago. Whether it does or not, only
more specimens will demonstrate.

Specimens examined. Cuba, Oriente Province:
UIMNH 44105 (type), 44106-45, 44152-213,
MCZ 61232-33; BYU 17165-6 (paratypes), MCZ
59142-46, USNM 140270-73, Cuantanamo.

Sphaerodactylus stejncgcri Cochran

Sphaerodactylus stejnegeri Cochran, 1931,
Copeia, no. 3, p. 90. Type locality, San Michel,
Dept. du Nord, Haiti (here amended to St.
Michel de 1'Atalaye, Dept. de l'Artibonite );
type specimen, USNM 76640.

Definition. A Sphaerodactylus of the decora-

tus group characterized by having dorsal scales
flattened to slightly swollen, smooth, rounded,
and slightly imbricate; head granules relatively
large and cobble-like, not small and tending to-
ward conical; dorsals with few large hair-bearing
organs on tip of scales; dorsals, axilla to groin,
41-52, mean 47.0; ventrals smooth, flattened,
rounded, and imbricate (axilla to groin, 27-36;
mean, 31.1); scales around midbody 53-65, mean
57.7; dorsal scales of tail smooth, flat-lying,
rounded and imbricate; scales beneath tail
smooth, rounded, imbricate and enlarged mid-
ventrally. Internasals 1-2, mode 1; upper labials
to mid-eye 4 or 5 in one labial series; escutcheon
relatively large, compact and with extensions
well onto thighs (4-7 X 15-22; one specimen has
an escutcheon width of only 7 scales). Habitus
moderate to slender, snout moderate in length
and breadth but sharply pointed; size moderate,
to 31 mm snout-vent in males, 30 mm in females.

Female coloration (Fig. 16): Dorsal ground
color light gray to gray-brown, irregularly
spotted or mottled with brown; two solid black
body hands, very narrow and widely spaced,
and outlined with white margins; neck band
solid black, somewhat wider than body bands,
outlined in white, extending onto throat and
usually ending ventrolaterallv, but may extend
completely across; tail banded with dark hands
similar to those of body; head pattern simple and
distinct: one relatively wide dark band anterior
to ear joined broadly on each side bv wide post-
ocular dark stripe that continues anterior to eye
as wide loreal stripe to tip of snout; venter light,
cream to grayish.

Male coloration: Unicolor tan or yellowish
tan above, venter as in females. (Coloration
given here is from preserved specimens.)

Range (Fig. 17). As presently known, Haiti,
from St. Michel de 1 Atalaye west to St. Marc
(Mertens, 1939:42), thence south to the Cul de
Sac Plain ( it is not known from the Dominican
Valle de Neiba). The record from "southwestern
Haiti" (Cochran, 1941:112) cannot be taken
too precisely; the specimen was originally cited
as being from "Haiti" (Cochran, 1924:3).

Remarks. We have amended the type locality
because there appears to be no San Michel in
die Departement du Nord, and because the col-
lectors of the type, A. J. Poole and W. M. Perry-
go, visited St. Michel de l'Atalave in December
of 1928 (Wetmore and Swales,' 1931:36) when
the type was collected.

Fig. 16. S. stejnegeri, female,

Eaux Gaillees. Dept. de 1'Oui

iew; MCZ 6317.3.
t. Haiti.

20

Bricham Young UNivEBsrn Scienci Bulletin

Fig. 17. Map of Haiti. Solid .symbols indicate localities
whence specimens of S. stejnegeri have been ex-
amined, and the hatched area indicates the con-
jectural range (it the species. The northeastern ex-
tension of the hatching includes the type locality
whence we have not examined material.

Specimens examined. Haiti: Dent, de l'Arti-
bonite: MCZ 59478-87, bridge over Riviere de
I'Artibonite, St. Marc Road; Dept. de l'Ouest:
NIC/ L3481, Thomazeau; MCZ 52158-62, Hatte
Lathan, Cul-de-Sac Plain; MCZ 59488-94, 63172-
73, 84353, YPB 3089-90, Eaux Gaillees; MCZ
13442, Coutard.

Sphaerodactylus ititermedius
Harbour and Ramsden

Sphaerodactylus intermedins Barbour and Hams
den. 1919, Mem. Mns. Comp. Zool., 17:211. Type
locality, Sierra do Hato Nuevo between Hato
Nuevo and Sabanilla de la Palma, Matanzas
Province, Cuba: type specimen, MCZ 12305.
Sphaerodactylus decor at us drapetiscus,
Schwartz, L958, Proc. Biol. Sot. Wash., 71:29.
Type locality, 2 mi E Playa de Cuanabo, Cueva
de Rincon de Cuanabo, Habana Province, Cuba;
type specimen, \M\1I 77751).

Definition. \ Sphaerodactylus ot tin- decora-
tus group having the dorsal scales granular and
juxtaposed to swollen, rounded and slightly im-
bricate Hut not prominent!) erected and papil-
late, weak keeling present in most specimens.
Dorsal scales with lew large hair-bearing scale
organs on posterior Face ot scales, dorsals, axilla
to groin, 19 62; ventrals 31-44; midbody scales

65 -77. Dorsal scales ot tail smooth, flat-lying;

ventral scales of tail smooth enlarged, Hut ap-
parently not prominently s lidventral line.

[nternasals 0-2, mode 1; upper labials to mid-
eye I "i 5, mode I escutcheon large, compact.
roughly triangular .mil with virtually no exten-
sion onto thighs 6-8 X 8-11. Habitus moderate,
snout elongate, size large, to 35 mm snout-vent.

Female coloration (Fit,'. 18): Head Hand and
facial markings united to produce dark head
markings that end just behind ear; two dark-
edged light postocular stripes, one (inferior)
that proceeds diagonally onto ventrolateral sur-
face of throat, fading out on a level with ear,
and one (superior) proceeds diagonally upwards
and over temporal area, and at least in its "basic"
manifestation turns medially and joins its mate
on opposite side, forming a I -shaped light
cephalic figure; loreal and median snout stripes
present, latter forked posteriorly before joining
interocular area; collar present, heavily dark-
edged and light-centered. Hut not occllate; H(xlv
Hands 3 or I i a 1 5 Hand count is found in one
specimen I, sometimes broken up on posterior
part of body or faint in young (or long-pre-
served? i specimens, prominently dark-edged and
somewhat sinuous. Coloration in life: dorsal
ground color yellowish gray, brightest on neck
and shoulders; ground color of dorsum of head
pale yellowish gray; collar black-edged with
yellowish gray center which has a few scattered
vellow dots just posterior to the anterior black
edge; body bands black-edged becoming grav
posteriorly, center of bands always with a few
scattered vellow dots; dorsa of limbs yellowish
gray — flecked and mottled with tan; venter gray-
ish, with gray flecking on infralabials and throat.

Males heavily spotted with very dark brown
on a light yellowish-tan ground color; head dis-
tinctly more yellowish than body; ground color
of tail bright vellow | notes based on coloration
nt type and paratvpes of drapetiscus: Schwartz,
1958).

Range. Known from the northern portions of
Habana ami Matanzas province (probable rec-
ords from CotOrrO and CamOa have been noted
under the range of torrei), and from the south-
western coast of Oriente Province near Cabo

Geckos of the West Indies

21

Figs. 18-19. Fig. 18. Sptiaerodachjlus intermedins, dorsal views of two females: A, AMNH 77763, 2 mi E Playa
de Guanabo. Cueva de Rineon de Guanabo, Habana Province, Cuba; B, AMNH 94550, 10.5 mi NE Matanzas,
Matanzas Province, Cuba. Fig. 19. S. ruibali. dorsal views: A, ASFS V6252, male, e;Lst side. Guantanamo
Bay, U. S. Naval Base, Oriente Province, Cuba; B, female, UMMZ 110181. U. S. Naval Base, Oriente Pro-
vince, Cuba.

Cruz (Fig. 6). The records from the latter area
may well represent a distinct population.

Remarks. The type and paratvpe of inter-
medius have a very obscure body pattern but
a bold neck and head pattern. Upon discovering
that S. cinereus changes from a boldly banded
juvenile ("elegans") to the more uniformly
colored adult, Barbour ( 1923 ) concluded that
intermedins had been based only on a transi-
tional stage of this metachrosis and was there-
fore a synonym of cinereus, where the name re-
mained hidden until the present study. Sphaero-
dactylus cinereus in the banded phase does not
have the characteristic head pattern of inter-
medius but has a simpler one similar to that of
S. decor atus. There is no doubt that intermedins
is a close relative of decoratus ( and other mem-
bers of the group); but because of its radically
different head pattern along with other signifi-
cant pattern differences we have accorded it
species status, which we have little doubt that
it deserves. Of note are two specimens, MCZ
8510 from Cabo Cruz and USNM 81670 from
Rio Puerco, both of which are near the extreme
southwestern tip of Oriente Province. (One of
these specimens ( MCZ 8510) is figured in dorsal
aspect bv Barbour and Ramsden (1919:Plate 2,
Fig. 1 ), although it is not the type of S. torrei, as
stated there.) We consider both to represent S.

intermedins. They are old and faded but are
banded and possess the distinctive head pattern
of intermedins, including the double postocular
stripes which are but faintly visible. Due to their
somewhat distinctive patterns it is probable that
they represent another race of the species, but
we are reluctant to propose a new name because
of the difficulty of being sure what features are
due to fading and poor preservation, and what
are natural. An additional difficulty is the possi-
bility that intermedins and ruibali are derivative
forms as evidenced bv some similarities in their
patterns (Fig. 20). In such a case the population
represented bv the southwestern Oriente speci-
mens might be most significant and should not
be described until its variation can be sufficentlv
known.

Specimens examined. Cuba: Matanzas Pro-
vince: MCZ 12305, 13726 (type and paratype of
intermedins) , Sierra de Hato Nuevo between
Hato Nuevo ( Marti ) and Sabanilla de la Palma;
UMMZ 78486, ANSP 16359 (paratvpe of drape-
tiscus), Matanzas; AMNH 945.50,' 10.5 mi NE
Matanzas; Habana Province: AMNH 77759-63,
77765, 81.367-71 ( type and paratypes of drape-
tiscus), AMNH 94264, 2 mi E Playa de Guanabo,
Cueva de Rineon de Guanabo; Oriente Province:
MCZ 8510, Cabo Cruz; USNM 81670, Rio Puer-

Hit li. 1 1 Wl Vol \i. I \l\ I Its I I 1 S( II \i I BULLETIN

Sphaerodactylun ruibali Grant

Sphaerodactylus ruibali Grant, L959, Herpeto-
logica 15 i 1:53. Type localit) I ,S. Naval Base,
Cuantanamo Oriente, Cuba. i\ j>< ■ specimen
UIMNH 14246

Definition. Dorsal scales small, obtuse to
rounded, only slightly swollen, smooth and mod-
erately but uniformly imbricate; dorsals, axilla
to groin, 42-60, mean 19.5; few large hair-bearing
organs I 3 hairs > on posterior tip of dorsal scales.
ventral scales smooth, flattened, rounded imbri
■ ate. axilla tt> groin, 30-41, mean 34.8; midbody

scales 69-82, mean 75.2 Fourth toe lamellae (>
11, mode 10; upper labials to mid-eye, 3-5, mode
I. one internasal in all specimens. Escutcheon
in fullv adult specimens with large central area
and slender extensions length of thighs, escutch-
eon scales not infrequently somewhat pigmented
(dimensions 5-10 X 19-28). Habitus moderate.
snout relatively short and broad, size moderate,
to 32 mm snout-vent.

Female coloration (Fig. 19): Dark loreal
stripe continues as interior postocular stripe to
level of ear where it meets with dark transverse
head stripe that passes behind ear; superior
postocular stripes i not always present ) proceed
for short distance on top of head and stop before
meeting transverse head stripe. Pair of dark
transverse stripes on neck appear to be cognates
ot neck stripe in other members of the decoratus
group; no median snout stripe or interocular
tlark area present. Body mottled with dark
brown irregular markings, which sometimes tend
to form irregular ocelli, on tan or light brown
ground color. Sometimes there are indications
of a transverse dark stripe on anterior btxlv just
posterior to axilla; tail dull yellow with some
brown mottlings or reticulation. Venter liglit
gray; underside ot tail dull yellow. Iris color
light gray or gray in lite.

Male coloration (Fig. 19): Male loses head
and neck banding ol If male-juvenile pattern
witli maturity. Head becomes nearly uniform
yellow-brown; both color becomes yellow-brown
to gray with fine brown specklings or vermicula-

tions. In lite ventei is hidit gray, tail yellow

above and below with line dark mottling above.
I Colors from field notes On specimens m life. I

Range. Although not always so stipulated,
the naval base records are presumed to be from
the east side of the bav ( tills is known to be true
lor the MCZ specimens). The known range
therefore extends from the east side of the Bahia
de Cuantanamo to the Rio Yateras (Fig. 11),

the mouth ot which is roughly ten miles east of

the bay.

Remarks. The flattened and more imbricate
doisal scales ot ruibali distinguish it from other
members ol the decoratus group. This scale type,
however, is not very far removed from that
shown bv members of the group and probably
represents an early stage in the development of
the flattened, more strongly imbricate scales ot
some species in the genus The simplified pattern
remnants of ruibali are reminiscent of those
found in interim (litis: the postocular pattern,
the transverse head pattern extending to behind
the car instead ot ending just behind the eves,
and the prominence of the neck and anterior
body pattern versus that of the rest of the bodv
( very similar to some specimens of female inter-
medius in which the body pattern is much
broken up but the head, neck, and scapular pat-
terns arc very bold). It is possible that ruibali
may be an intermedins derivative in eastern
( )riente.

Several specimens ot S. ruibali were collected
by the senior author at a locality where S. deco-
ratus was also taken. However, ruibali was found
in a more exposed and drier area than were- the
decoratus which were taken in a well-watered
nursery.

Specimens examined. Cuba, Oriente Prov-
ince: UIMNH 44246 (type), 44247-49 (para-
types), MCZ 67380-81, 68733-35, 68931-35,
69436-37, 69439, 69442, UMMZ 110181, U.S.
Naval Base, Cuantanamo; ASKS V6252-55, east
side. Cuantanamo Bay, U.S. Naval Base; USNM
78921-22, Rio Yateras.

HAHI'I \ I

Sphaerodactylus decoratus: In the Bahamas palms. On Andros our experience was less ex-

some differences in habitat preferences were ob- tensive but specimens were' also found in beach

served for decoratus on the various islands. On areas or immediately behind beaches and in

South Himini flavicaudut was found almost ex- association with COCOS and Sid>al trash. On

elusivelv in a beach habitat of low dunes where Kleutlie i a specimens ot S. il. atessares were also

specimens were taken m litter from Tluinax most easily Collected in sancK beach areas, par-

Geckos of the West Indies

ticularly in piles of Cocos trash or around bases
of Thrinax. However, the species is apparently
more widespread (but rarer) in other situations;
one was taken in arid scrub on the southern por-
tion of the island and two others, although taken
in a stand of Thrinax, were in a coastal limestone
scrub, which is not a beach habitat. On Andros,
Bimini and Eleuthera S. notatus was more wide-
spread than S. decoratus and was found both in
beach and interior situations. On Great Exuma
S. (/. decoratus was apparently more widespread
and was often seen in the interior of the island,
usually in piles of rocks. It was most easily col-
lected in the beach situation where palm trash
was available. This is in contrast to the afore-
mentioned islands where, despite intensive col-
lecting in interior localities, virtually no speci-
mens were seen. On Cat Island two specimens
were taken at night near a light on a plaster-
walled house. On Long Island most specimens
were taken in a solution hole planted in bananas
but others were taken under rocks in shady
areas.

Our only experience with Cuban decoratus
was with S. d. strategus, which was collected in
a very xeric region but found most abundantly
in the artificially moist microhabitat of a well-
watered nursery on the U.S. Naval Base. Speci-
mens were taken by moving flowerpots and
many more were seen than were collected. Also
collected in the same situation were Amphis-
baena cubana and Eleutherodactylus atkinsi.
Only one specimen was taken in a more xeric
situation along the coast. As to the northern
habitat of the races, little or nothing is available.
However, we know that the country is generally
more mesic in northern Oriente, and Grant
(1960) stated that the Banes and Preston area
was "forested and humid," which we may pre-
sume forms the habitat for S. d. granti. The
types of lissodesmus were taken in a rotten log
in forest. The Sierra de Cubitas is in general
a mesic area.

Sphacrodactijlus alayoi: No habitat notes are
available for this species but if the specimens
came from the U.S. Naval Base, they were prob-
ably taken in xeric surroundings.

Sphacrodacti/lu.s torrci: Of the specimens of
the nominate race collected by the junior author
and company, two were taken in a motel room;
three were found in a stone wall in relatively
mesic but open surroundings, and those from
Playa Juragua in coastal arid forest. Male speci-
mens of torrci that we have included in the race
ocujal were reported by Cooper (1958) as com-
ing from a coastal sea-grape (Coccoloba) situa-
tion, where they were taken in "rocks, dried
leaves, and other plant debris."

Sphacrodacti/lu.s intermedins: This type and
paratype of this species were described by Bar-
bour ( 1921 ) as having been collected "under
loose stones on a dry, rather scrubby, hillside
pasture." Schwartz ( 1958, on the type series of
drapetiscus) described taking the species "under
rocks on the ground" at the base of limestone
cliffs; the situation was in "typical tropical decid-
uous woods with scattered palms." The specimen
from northeast of Vlatanzas was taken in leafy
trash at the base of a small tree on a cleared
beach.

Sphacrodacti/lu.s ruibali: The type series of
this form was reported as Inning been taken in
grassy hills at about 300 feet elevation ( Grant,
1959b). The specimens collected by the senior
author were found adjacent to the nursery de-
scribed above for S. d. Strategus. However, the
ruibali were taken in an exposed area where
they were found under scattered rocks and in a
rather thin and spotty covering of leaves; none
was seen in the moist nursery.

Sphaerodactylus slejnegeri: Although we
have had no first hand experience with this
species, it apparently is capable of enduring a
variety of habitats. The "bridge over the Biviere
de l'Artibonite," whence we examined specimens,
is a mesic situation, and the Cul-de-Sac is in
general dry and desert-like, but the species may
occur in oasis situations there. Seven specimens
from Eaux Gaillees ( MCZ 59488-94) have as
part of their collecting data that they were taken
on a large tree. Mertens (1939:42) noted that
the onlv S. steynegeri he collected was on the
wall of a house in the late afternoon.

CONCLUSIONS

The granular scaled forms have been re-
garded as more primitive members of the genus
(Barbour, 1921:218; King, 1962:42), and prob-
ably correctly so. The evidence seems to indi-

cate that forms with more specialized scale types
have arisen on different occasions from the gran-
ular scaled ones. Thus, of the Cuban members
of the decoratus complex, S. torrci shows a dis-

-1

linn .i i \ m Voi \i. I \i\ i mm i \ Science Bulletin

tmct trend toward flattening and imbrication ol
(In- scales. This, coupled with tin- striking sum
larity in patterns ol S stejnegeri to S torrei, Nads
us to believe tin- two to be derivative tonus. Tins
is borne out m tin- scalation ol itejnegeri which
is ,i degree furthei advanced beyond torrei in
Flattening and imbrication ol body scales i it may
be Further noted thai ol the Cuban Forms, torrei
possesses less keeling than the others and thai
keeling is lacking in stejnegeri) . The distribu-
tional picture supports tins conclusion. S, torrei
is apparently a wholl) eastern Cuban species
and S, stejnegeri is a wholl) western Hispaniolan
one. and perhaps primarily a northwestern one

at thai (although it reaches the Cul-dc-Sac
Plain, it has apparently not exploited this natural
channel into the Dominican Republic).

Although we are less certain about the af-
Finities of S. ruibali, on the basis of head pat-
terns (Fig. 20) its relationships may well lie
with S. intermedins. It such is the case, once
again we see a progression toward Flattening

and imbrication of the dorsal scales. At the same
tune there can be little doubt that S. intermedin*

and S. decoratus arc di rivative tonus, it is prob-
able that the Former partakes of a relict distribu-
tion, as it is found in western Cuba and .south-
western Oriente. Of note is the occurrence ot
other relict or apparently relict tonus in the same
general region of Oriente. Thus Leiocephahts
rat iceps has two races in southeastern and cast-

Hii Oriente and another apparent!) disjunct in
Matanzas Province (Schwartz I960); and (ri-
cosaura typica is a relict Cuban xantusiid known
onl) From extreme southwestern Oriente Sa>
age'. 1964

Sphaerodactylus torrei is presumed to have

arisen as an isolate on the southern coastal
slopes ol the Sierra \laestra. II we have inter-
preted the three-banded subspecies ocujal cor-
rectly, the close affinities of torrei with decoratus
arc certain. S. t. OCUJal is the most distal race
from the present area ol contact between S.
decoratus and S. torrei. and. unless contact with

northern (as opposed to eastern' decoratus

existed around the western end ot the Sierra
\laestra, is the most isolated of the races of
torrei From the presumed ancestral form. Tills
relative isolation may account for the retention
of the more primitive color pattern (greater
number of body bands). Present records indi-
cate that the primary distribution of S. decoratus
in Cuba is the north of Oriente Province. It
probably occurs to the east of the Bahia de
Guantanamo and is possibly continuous peri-
pherally with the north coast populations around
Cabo Maisi. Sphaerodactylus <1. lissodesmus is
the westernmost representative of the species
and max be endemic to the Sierra tic (aibitas.
Sphaerodactylus alayoi is certainly a close rela-
tive ot S. decoratus, but so little is known ot its

p-a^-s^.

B

D

Fig, 20. Lateral views ol Iic.hK ol lour geckos to show hypothetical evolution of pattern from generalized decora-
(u«-type head pattern through intermedius to ruibali: A S decoratus grant! (BYU L7232) It S. decoratus
subsp. (MCZ 12281) C, S intermedius (AMNH 81367); D, S. ruibali (MCZ 67380). rhe Cat Island sjpeci-
men ol s decoratus (hi shows a pattern variant which demonstrates the possible method ol origin ol the
intermedius-ruihali head pattern

Geckos of the West Indies

25

distribution that no explanation of its origin can
be hazarded (and foi that matter, its status
must also remain in slight doubt). The presump-
tive intermedius-ruibali relationship may reflect
an older subgroup of this complex, as indicated
by (1) a high degree of divergence, and (2) a
relict distributional pattern.

Sphaerodactylus <7< coratus as here defined
includes both Bahaman and Cuban races. Some
might prefer that these two geographical groups
be regarded as separate species. However, we
believe that on the basis of geographical prox-
imity and similarity ot coloration and scalation
of the two that we have made the choice that
best expresses their relationships. The diversity
of decoratus in the Bahamas themselves, less in
degree than that which has occurred in the group
in Cuba, is attributable to the recency of the
present Bahaman land configuration and prob-
ably also to its variability. The herpetofauna of
the Bahamas north of the Crooked Island Pas-
sage is primarily Cuban in origin and the degree
ot endemism is not generally great. It has been
postulated that the Bahamas were colonized
from Cuba by dispersal across water rather than
by direct connection with Cuba (Clench, 1938;
Babb and Harden, 1957), as there is presumed
to have been no continuous land connection be-
tween the two areas. Sphaerodactylus decoratus
is absent from the Little Bahama Bank and the
islands south of the Crooked Island Passage.
Both of these divisions are occasions for distinct
(but not absolute) faunal breaks in the Bahamas.
Rum Cay is separated from the Great Bahama
Bank by deep water (about 1500 fathoms) and
is the onlv island to which S. decoratus has
spread beyond the Great Bank.

Puerto Rico has no members of this group,
nor apparently does Jamaica (we have not had
first-hand experience with S. oxyrrhinus, but it
does not seem to pertain to the decoratus group ) .
The group is also not represented in the Lesser
Antilles; with the possible exception of S. cin-

ereus ( see below ) it is not represented on His-
paniola except by S. stejnegeri.

Sphaerodacti/lu.s cinereus has not been con-
sidered a member of the decoratus complex in
this study because of its differences in colora-
tion, lack of sexual dichromatism, and scalation
(of all the forms we have seen it probably comes
closest to having truly granular scales). If in the
future it should be considered to belong to the
group, its distribution fits well with the concept
of a Cuban origin of the complex. It is wide-
spread on Cuba ( but uncommon in Oriente Prov-
ince) but restricted to western Hispanola, al-
though in contrast to the similarly ranging torrei-
stejnegeri pair, differentation between the popu-
lations of the two islands appears to be slight in
this species.

Thus, the distributional picture presented bv
the decoratus group of sphaerodactyls is one
of a principally Cuban radiation in which limited
dispersal to nearby islands and island groups has
occurred. The great diyersity of forms in Cuba
indicates a long history of the group there. That
diversification has been greatest in eastern Cuba
is perhaps explained by the physiographic and
ecological complexity of this region.

Cuba is generally well represented by
Sphaerodactylus and has more (and more di-
verse) species than any other West Indian island,
including members of the seaher group charac-
terized by large dorsal scales and a middorsal
zone of granules, and S. not at us, which belongs
to a loosely defined assemblage including forms
(such as S. difficilis and S. inaguae and perhaps
the Puerto Rican forms ) having uniformly flat-
tened, keeled and comparatively large dorsal
scales. Sphaerodactyltts ramsdeni, a local adapta-
tion to the montane forest gecko niche, is prob-
ably derived from one of the other "groups" on
the island. S. argt(s of Jamaican origins shows a
restricted but apparently natural distribution in
Cuba and is probably a comparatively recent
arrival.

LITERATURE CITED

Barbour, T. 1914. A contribution to the zoogeography
of the West Indies, with especial reference to am-
phibians and reptiles. Mem. Mus. Comp. Zool.,
44(21:207-359.

. 1921. Sphaerodactijlus. Mem. Mus. Comp.

Zool., 47(3):218-278.

and C. T. Ramsden. 1919. The herpetology

of Cuba. Mem. Mus. Comp. Zool, 47(2) :71-213.'

Clench. W. J. 1938. Origin of the land and fresh-
water mollusk fauna of the Bahamas, with a list of
species occurring on Cat and Little San Salvador
Islands. Bull. Mus. Comp. -Zool. 80 ( 14 ) :481-541.

Cochran. D. M. 1924. Notes on the herpetological
collections made by Dr. W. L. Abbott on the island
of Haiti. Proc. U. S. Nat. Mus., 66(6): 1-15.

26

Bum. 1 1 \\i Yoi M. I m\ i us 1 1 -i SciENCi Hi i.i i 1 1\

. 1934, Herpetological collections From tin-

Wist Indies made l>\ l'r Paul H.irtM h undei the
Walter Rathbone Bacon Scholarship, 1928
Smithsonian Misi Coll., 92(7 I I is

1941 The herpetology ol Hixpaniola, Dull

1 s Nat. Mus., ITT i mi
Cooper, | E, 1958 Ecological notes on some Cuban
lizards. Herpetologica, 1 I I

II. rpi tologii il notes with new species
from tin- Am. in ,ui .mil British Virgin Islands, L936.
J. Dep. Agr. Puerto Kim -2 1 < 1 :50 S-522.
. 1956. The st.iins i.l Sphaerodactylus gib-
bus. Herpetologica, 12(3):247

L958 V new gekkonid lizard [Sphaero-

dactylus) from Cuba Herpetologica, 14(4):225-

. L959a \ new Sphaerodactylus from Guan-

tanamo, Cuba Herpetologica, 15(l):49-53.

1959b. Anothei new Sphaerodactylus from

Cuantanamo, Cuba Herpetologica, L5(l):53.

I'ii. ii. I (iffereni es in shade <>f some ri p-

tilrs from the north .mil south io.isK ol Oriente,
Cuba. Herpetologica, 161 3):174.
IL. lit. \l. K 1954 Tin comparison oi recent and fos-
sil amphibian, reptilian .mil mammalian faunas in
tin- Bahamas. Year Book Amer. Phil. Soc, pp. 133-
135.

King, W, 1962. Systemati. • I. ei tarjllean lizards

..I tin- n<-iiiis Sphaerodactylus. Hull Florida State

Mus.. 7( 1 cl-52.
Mertens, R. 1939 Herpetologische Ergebnisse einer

Reise nai Ii der Insel Hispaniola, VVestindien \l.h

senckenberg. naturi < •■ 149:1-84

Rabb, G. B. ..ml I It Hayden L959 The Van V.ust-
\m.rii.m Museum i.l Natural Historj Bahama Is-
lands Expedition record ol the expedition .mil gen-
eral feature: "t thi i lands toner. Mus. Novitates,
18 16:1-53.

| \l L964 Studies mi ilu- lizard I imilj

Xantusiidae, V, Cricosaura typica <■ II. h h .mil

Peters. Copeia no, 3:536-542.

Schwartz, \ 1958 \ new gecko of the Sphaero-
dactylus decoratus group from Cuba. Proc. BioL
Soc. Washington, 71:27-36.

. I960 Variation in the Cuban lizard

Leiocephalus raviceps Cope. Proc. Biol Soc. Wash-
ington, 75:67-82

Thomas, B and \ Schwartz. 1966 Sphaerodactylus
(Gekkonidae) in tl»- Grcatei Puerto Rico region.
Bull. Florid.. State Mus., 10(6):193-260

Wetmore, A, and B. II Swales. 1931. The birds of
Haiti and the Dominican Republic. Bull. I'. S Nat,

Mus., 155:1-483.

cyrcme
Bookbinding Co., Inc.
100 Cambn !

wn. MA 02129

2044 072 224 678