FLEAS 0F FOX SQUIRRELS
3N SOUTHERN WCHIGAN

Thesis for the Degree of M. S.
Mics-{EGAN STATE UNWERSITY
JOHN C.. ARNSMAN
1870

* LIBRAR 3—" 5

Michigan Sty .55

University

  

    

 

 

 

 

ABSTRACT

FLEAS 0F FOX SQUIRRELS IN SOUTHERN MICHIGAN

By

John C. Arnsman

During 1967, 1968 and 1969 samples of flea populations were
removed from 197 fox squirrels in Southern Michigan. Orchopeas howardii
(Baker) was found to be the major flea of the fox squirrel. It made
up 99 per cent or 556 of the 559 fleas examined.

The annual cycle of this flea on the fox squirrel was delineated.
The population peaks every year in March, and may peak again in August
and September. The population cycle of this flea closely follows the
squirrel breeding cycle and may possibly be associated with it.

The age and sex of a host squirrel are determining factors in
its susceptibility to parasitism by fleas. Adult females have signifi-
cantly fewer fleas per animal (1.6) than the immature females (2.7), the
immature males (3.2) or the adult males (3.0). The low number of fleas
parasitizing the adult females was not a factor of the number of
squirrels in this host class that were parasitized.

A "mangy condition" frequently observed in squirrels is apparently
not related to flee parasitism. Although, it does occur at the same
time as the flea population peak in March, it is perhaps more.re1ated

to population densities and climatological factors.

FLEAS OF FOX SQUIRRELS IN SOUTHERN MICHIGAN

By

John C. Arnsman

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE

Department of Entomology

1970

Cé/Mé’
3”/u '70

ACKNOWLEDGEMENTS

I would like to thank Dr. R. L. Fischer for his encouragement,
and his material and editorial help in conducting and writing of this
study. Dr. G. E. Guyer and Dr. R. H. Baker also contributed by giving
encouragement and in helping to develop ideas. Dr. L. W. Gysel very
kindly made a number of his traps available to me and helped in
selecting the study site for this study. The Michigan Department of
Natural Resources made certain unpublished records available as well
as offering the help of their personnel with technical problems.

Last, I would like to thank my wife for complaining long enough about
my graduate student status that I had to finish this degree to get a

little peace and quiet.

11

LIST OF TABLES .

LIST OF FIGURES

INTRODUCTION . .

METHODS .

RESULTS .

DISCUSSION

SUMMARY .

LITERATURE

TABLE

OF CONTENTS

iii

Page

iv

13
18

19

Table

II.

III.

IV.

LIST OF TABLES

Number of Fleas per Host in Four Host Classes;
lst Year Female, lst Year Male, Adult Female,
and Adu1t Male 0 O O O O O O O O O O O O O 0

Number of Fleas per Host for Adult Versus
lat Year 0 O O O O O O O O O O O O O O O O 0

Number of Fleas per Host for Male Versus
Fmale quirrels O O O O O O O O O O O O O I

Number of Fleas per Host in Four Host Classes
Using only Hosts That Had at Least One Flea .

Summation of Data on Squirrels Captured More
man once 0 O O O O l O O O O O O O O O O I 0

IV

Page

12

LIST OF FIGURES

Figure Page
1. The Average Number of Fleas per Squirrel on a

Time B3818 O I O O O O O O O O O O O O O O O O O O O O O 7

2. The Per Cent of Fleas Examined Which Were Females . . . . . 11

INTRODUCT ION

As a major game animal the fox squirrel, Sciurus nigg£_Linn.,
has been the subject of much biological investigation. Allen (1943),
for example, published a monograph on the biology and management of the
fox squirrel in Michigan. Brown and Yeager (1945) and Baumgartner
(1940) published on the fox squirrel in Illinois and Ohio respectively.
All three authors mention finding fleas on the squirrels; the latter

two record the fleas, Orchopeas howardii (Baker) and Orchopeas nepos

 

(Baker). Both authors give numbers in terms of "common", "very common",
"occurs regularly" and "in very large numbers in over-wintering nests",
but no other numerical data.

Fox (1940) lists Orchopeas wickhami (Baker) as the flea of the
fox squirrel. Ellis (1955), in a small sample, found an average of

8.3 chicken fleas, Echidnophaggflgallinaceae (Westwood), per fox squirrel

in Oklahoma. Mathewson and Hyland (1964) found Cediopsylla simplex

 

(Baker), Ctenocephalides felis (Bouche), Orchopeas howardii (Baker)
and Mgggbothris asio (Baker) on S, nigg£_in Rhode Island. Mathewson
and Hyland do not mention how many fleas per squirrel were present.
Orchopeas howardii is a species closely associated with tree
squirrels in North America. Its range extends from the Atlantic sea—
board (Fuller 1943, Tindall and Dorsie 1961, and Mathewson and Hyland

1964) to the Rocky Mountains (Dunn and Parker 1923), from the Gulf

2
coast (Baker 1944, and Fox 1940) north at least to Vermont (Osgood
1964) in the east and Montana (Dunn and Parker 1923) in the west.
The present study was designed to determine the composition
and numbers of the flea species on fox squirrels occurring in

Southern Michigan.

METHODS

The areas in which this study was carried out were three small
woodlots on the property of Michigan State University. One is an un-
named woodlot of approximately 40 acres located in section 6 of T. 3 N.,
R. 1 W. of Ingham Co. Collections of fleas were made from squirrels
found in this woodlot from August to October of 1967. Hudson woodlot,
approximately 15 acres in size, was trapped from October 1967 to
August 1968; and Minnis woodlot was studied from August 1968 to
September 1969. Both of the latter are located in section 30 of
T. 4 N., R. l W. of Ingham Co.

All woodlots botanically consisted primarily of sugar maple-

 

beech forests, Acer saccharum Marsh and Fagus grandifolia Ehrh.

 

respectively. Other trees in common, but nondominant numbers, in—
cluded: basswood, Tilia americana Linn.; elms, ylmgg spp.; white oak,
Quercus alba Linn.; black oak, Quercus velutina Lam. and black cherry,
Prunus serotina Ehrh. The woodlots were surrounded by agricultural
areas which varied from pasture and hay to row crops.

The traps used were all.box traps, of various designs, but with
a minimum size of 5 X 7 X 20 inches inside dimensions and a maximum
size of 8 X 10 X 30 inches. The traps included both enclosed, wood-
sided traps, and traps made of hardware cloth over a wire frame. Data
on the efficiency of the different kinds of traps were not kept, but

3

4
no difference in either efficiency of capture or of mortality of
trapped squirrels were noted. Ear corn was used for bait.

Trapping began in August 1967 and ended in September 1969.

The traps were set in every month of the year for a minimum of 10
days except from May through August of 1968.

When a squirrel was trapped it was released into a nylon mesh
cone and forced into the small end of the cone. Either chloroform or
ether were applied to a clean rag and held over the head of the
squirrel while it was immobilized in the cone. When the squirrel began
to relax it was withdrawn from the cone in a gloved hand by holding
the back of its neck between the thumb and first finger and supporting
the thorax with the other three fingers of the gloved hand. Addi-
tional anesthesia was applied until the animal was completely relaxed
whereupon the animal was examined for fleas.

To remove the fleas, the squirrel was held over a white piece
of new flannel cloth (15 X 24 inches) and the breath blown through the
fur against the natural lay of the hair. The fleas, easily seen moving
through the fur, were picked out by hand and placed on the white
flannel cloth. After examining the squirrel once in this manner the
animal was reexamined by rubbing the fur against the grain and
searching for additional fleas.

When all visible fleas were removed to the white flannel cloth
the squirrel usually had become quite active and was released. The
fleas were removed from the cloth and placed in 95 per cent alcohol
where they were stored until they could be cleared and permanently

mounted.

5

Twelve squirrels which were accidently killed during handling
were examined for fleas in the same manner as for the live squirrels,
then placed in polyethylene bags and brought to the laboratory for
thorough inspection by washing in a_white enamel pan. This provided
a check of the efficiency of the field techniques for flea removal.
These 12 squirrels had 93 fleas when examined in the field, and no
additional fleas were recovered when they were examined in.the
laboratory.

The fleas were cleared in 10 per cent potassium hydroxide and
mounted on slides in diaphane. The host specimen of each flea, the

age, sex and general health of each host were recorded.

RESULTS

In the 25 months of the study 197 fox squirrels were examined.
A total of 559 fleas were removed of which 556 or 99 per cent were

Orchopeas howardii (Baker). A single female Nosophilis fasciatus

 

(Bosc), one male Opisodasys pseudarctomys (Baker), and a female
Ctengphthalmusgpseudoagyrtes Baker were also recovered. Since
9, howardii was the preponderant species, all the population calcula-
tions were based on this species.

There were 2.8 fleas per squirrel over the entire study period.
Figure 1 shows the incidence of fleas per squirrel on a monthly basis.
A major cycle in numbers of fleas per squirrel occurred, beginning
with a low in November or December and reaching a peak in March of
both years. Numbers of fleas were also high in September of 1967
but data are insufficient to determine whether this is a peek or the
downgrade of a prior peak. Since the flea population did not peak
again in September 1968, special effort was made to check this period
in 1969. However, the flea population was again low. Fifteen
squirrels killed and examined from areas outside of the study area
during September and October of 1969 harbored a total of only five
fleas.

The host squirrels were broken into four classes: first year
male, first year female, adult male, and adult female, as shown in

6

_ das
- Snv
P Inf
.. unf
- New
+- JdV
_ 19w
b 99.1
_ nap

 

.. oaa
_ AON
.. zoo
.. dag
_.8nv
. '[nf
..unf
..Aew
_.1dv
..zan
..qag
.— nap
r oaa
-.AON

- 330

 

 

I I F fl T F l I I l
C o C!) N \O In <1' «1 N H
H

Iamue/ seat 5r ° aAv

dag

1969

1968

1967

Figure 1.-The average number of fleas per squirrel on a time basis.

8
Table I. The average number of fleas per host, within these classes
was compared (Table I) using a Chi2 test. The summation of the Chi2
values for the four segments of the population equalled 22.8794.
Adult females were found to have significantly fewer fleas per animal
(1.6) than any other class. First year females had 2.7 per animal,
first year males 3.2, and adult males 3.0.

TABLE I.--Number of fleas per host in four host classes; 1st year
female, lst year male, Adult female, and Adult male

 

 

 

 

Host No. of No. of Chi2

Class Squirrels fleas values

1st Year Female 75 204 0.2470

lst Year Male 34 108 1.5661
Adult Female 36 56 20. 3179*

Adult Male 51 154 0.7485
2 *

Summation Chi 22.8794

 

*Significant at the l per.cent level.

No significant differences were found in the fleas per squirrel
for the adults versus first year animals or for the males versus
females (Tables II & III).

To determine whether the low number of fleas per adult female
was a reflection of fewer animals with fleas or fewer actual fleas per.
animal, the data were compared in a different manner. The number of
hosts with fleas was compared to the total number of squirrels examined.
The four host classes were not significantly different in the number of

hosts with fleas as shown in Table IV.

9

TABLE II.--Number of fleas per host for adult versus lst year

 

 

 

 

Host No. of No. of Chi2
Class Squirrels fleas values
Adults 87 210 4.994
lst Year 110 312 0.016
Summation of Chi2 5.010

 

TABLE III.--Number of fleas per host for male versus female

 

 

 

 

squirrels
Host No. of No. of Chi2
Class Squirrels fleas values
Males 86 262 1.615
Females 111 260 7.254
Summation of Chi2 8.869

 

TABLE IV.--Number of fleas per host in four host classes using
only hosts that had at least one flea

 

 

 

Host No. of No. with Chi2

Class Squirrels fleas values
lst Year Female 34 13 0.026
lst Year Male. 51 21 0.076
Adult Female 75 32 0.292
Adult Male 36 10 1v124

 

Summation of Chi2 1.517

 

10

The sex distribution of fleas was also checked. Female fleas
significantly outnumber males in every collection, as shown in Figure 2.
The females never made up less than 57 per cent of the papulation and
averaged 72 per cent.

Records on the reinfestation of individual host animals are
shown in Table V. Five squirrels were captured twice, and four were
captured 3 times. Two squirrels which were caught twice and one
squirrel which was caught 3 times were never found to have fleas. The
other three squirrels which were caught twice had fleas the second time.
they were captured. The time interval between any of these recaptures
did not exceed three days. The other three squirrels which were caught
3 times had fleas at least two times out of the three. The interval
between captures with fleas for these three animals was between one and
23 days. Therefore, six of the nine recaptured squirrels became in-
fested with fleas between captures. The time interval between captures
when an individual squirrel became reinfested with fleas ranged from
one to 23 days. On three occasions squirrels were captured one day
after a previous capture and were reinfested. A squirrel, without fleas
at its first capture, appeared to remain free of fleas for 46 days,
during which time it was caught, examined, and released two additional
times.

In the present study eight squirrels were caught (March and
April of 1968) which had a "mangy condition". Three, or 37.5 per cent
were infested with fleas and 38.6 per cent of the total squirrel popu-
lation were infested during this same time interval. However, the
eight squirrels with a "mangy condition" had fewer fleas (3.3 per animal)

than did the total trapped population (5.3 fleas per animal).

11

[deg
.an
~Inf
-unf
.Xen
radv
.19“
tqag
-uef
~aaq
~AoN
r330
-das
rfinv
*Inf
.unr
-5914
-1dv
~19n
pqag
.uaf

-oaq

 

100-

95.

90«

In
w

c5636
comm

SGIBWSJ 31190 .19 (I

I
In
\0

b
\o

55~

 

deg

50

1969

1968

-AON'\
\O

4003}

Figure 2.-The per cent of fleas examined which were females.

12

TABLE V.-—Summation of data on squirrels captured more than once

 

 

 

Number Days Total days
Host Recapture of fleas between between lst &
present captures last captures
0‘1 1 31 31
2
C-19 l 0 2 2
C-23 l 2 2
2
C-26 1 O 3 3
B-122 1 1 l
B-28
2
3
B-47 3
43 46
3
C-21 l 21 17
O 6 23
3
D-23 l
2 1 8

 

DISCUSSION

The annual adult population of Q, howardii has at least one well
defined peak in fox squirrels in southern Michigan. This occurs in

March. A single annual peak population of fleas has also been noted

 

for Spilgpsyllus cuniculi (Dale) on the European rabbit, Ogyctolagus
cuniculus (Linn.) by Allan (1956). Parker (1958) had two fleas,

nglopsyllus anomalus (Baker) and Rhadinopsylla heiseri (McCoy), on the

 

antelOpe ground squirrel, Citellus leucurus leucurus (Merriam), with

 

one population cycle per year; and Ryckman g£_§l, (1954) had three

fleas, Echidnophaga gallinacea (Westwood), Damanus montanus (Baker) and

 

H, anomalus on the California ground squirrel, Citellus beecheyi

 

beecheyi (Richardson), with one population cycle per year.

During 1967, however, a high population level was also found in
September (Figure 1). While the March population peak appears to
initiate from a low level in November and December and to increase con-
tinously until March when numbers begin to decrease. This peak popula-
tion of fleas follows the mating season of the fox squirrel in southern
Michigan (Allen, 1940) by approximately a month.

Mead-Briggs and Rudge (1960) have described the necessity for a
"factor" from pregnant rabbits for ovarian maturation in the rabbit

flea, Spilopsyllus cuniculi (Dale). While data are scanty, circum-

 

stantial evidence would lend support to Mead-Briggs and Rudge's work

13

14
in the present study with Q, howardii. The primary mating season of
the fox squirrel begins in mid-winter and lasts until early spring
(Allen, 1940). A second mating season may occur in Michigan but only
when the habitat can support more squirrels, and its extent appears to
be dependent on the condition of the habitat (Allen, 1940). Therefore,
if Q. howardii does require a "factor" from a pregnant female host to
breed successfully it would have a yearly population peak in the late
winter or early spring, following the squirrel mating season which
peaks in January and February. Another population peak could occur
during August and September in occasional years. This second peak
would occur only during the years that the squirrels had an extensive
summer mating season and might account for the high numbers of fleas
found in late summer of 1967.

Adult female squirrels were_found to have fewer fleas per animal
than first year females, adult males or first year males. This dif-
ference in numbers of fleas per host was not a factor of age or sex
alone but was a factor of both combined, adults not having more fleas
than first year animals and males not having more than females. It
was also shown that this was not a factor of the number of animals
parasitized. So it.can be deduced that it is an actual difference in
the number of fleas per host. The most obvious reason for this dif—
ference is in differing behavior patterns of the host classes. Allen
has shown that adult males have a significantly larger range than
females and that immature squirrels move frequently during their first
year. By moving frequently from one nest to another these three

classes would be more susceptible to infestation. Once infested they

15
would spread the fleas to other nests within their present and future
home ranges. The adult female, on the other hand, has both a close
association with her home range as well as a smaller home range. This
could be responsible for the smaller number of fleas per animal.

Unfortunately, if Q, howardii does need a "factor" from a
pregnant female host, the fleas would have to feed on a female. The
distribution of fleas within the host population found in this study
does not indicate that many fleas have this opportunity. The distri—
bution of fleas among the host age classes might change at different
times of the year. If additional collections could be made it might
be possible to show whether or not the fleas changed their preference
for host-age classes at a time of the year that allowed them to feed
on females immediately after breeding.

The data from this study indicate that the flea population on
the host is composed of 72 per cent females and 28 per cent males.
Many other studies, including Evans and Freeman (1950) and Stannard
and Pietsch (1958), have observed similar ratios of male to female
fleas on the host. They have postulated that the females might live.
longer than the males or that the females might spend more time on the
host and less in the nest compared to the males. This study cannot_
elucidate further on this phenomenon, but-widens the range of species
in which it is known to occur.

Fleas are known to vary widely in their requirements for the
presence of a host. Some species spend nearly all their adult lives
attached to a single host while others spend the majority of their time

in the nest with only short periods on the host for feeding. However,

' Iill llll'.l||ll|.i

 

16
most species fall between the two extremes in the amount of time spent
on the host. The data given in Table V were collected to help deter-
mine where Q, howardii fits on this host requirement gradation. How-
ever, the low number of squirrels examined, plus the variability in
time taken for repopulation of different individual hosts by fleas,
made this impossible. One squirrel (B-28) was repopulated by fleas
within one day, while another (C-Zl) was not repopulated in 17 days.
Squirrel C-l apparently spent 31 days without becoming infested while
B-47 apparently spent 46 days without any fleas. Some squirrels, such
as B-28 and D-23, became repopulated readily, while others, such as
C-1 and B-47, seemed never to acquire fleas. This phenomenon seems
best explained by saying that the nest is the primary focus of the
infestation. Those animals whose nests are not infested do not have
fleas while those animals whose nests are infested become repopulated
rapidly.

Allen (1940) and Brown and Yeager (1947) suggest a possible
relationship between fleas and a "mangy condition".of the squirrel's
pelage. The time when squirrels are most susceptible to mange in late
winter. The only "mangy" squirrels observed were seen during March
and April of 1968. This coincides with the peak of the flea population
cycle. Of the eight squirrels handled with a "mangy condition" three
(37.5 per cent) had fleas, which was no different from that found in‘
the total population at this time (38.6 per cent). However, the eight-
squirrels with a "mangy condition" had 3.3 fleas per animal while the
total population at this time had 5.3 fleas per animal. The sample

size does not allow statistical comparison, but the difference in the

17

numbers of fleas per animal in "mangy" and "non-mangy" squirrels indi-
cates that the "mangy" condition may not be attractive to fleas.

Sarcoptic mange is usually associated with period of stress due
to either weather or food conditions. If the flea population cycle is
dependent on the breeding cycle of the squirrels, then the occurrence
of large numbers of fleas and mange would appear at the same time in
late winter but for different reasons. Since a "mangy condition" was
not noted during the fall of 1967, or during late winter of 1967 and
1969 when large numbers of fleas were present, it appears that the two

phenomena are caused by different factors.

SUMMARY

During 1967, 1968 and 1969 samples of flea populations were

removed from 197 fox squirrels in Southern Michigan. Orchopeas howardii

 

(Baker) was found to be the major flea of the fox squirrel. It made up
99 per cent or 556 of the 559 fleas examined.

The annual cycle of this flea on the fox squirrel was delineated.
The population peaks every year in March, and may peak again in August
and September. The population cycle of this flea closely follows the
squirrel breeding cycle and may possibly be associated with it.

The age and sex of a host squirrel are determining factors in
its susceptibility to parasitism by fleas. Adult females have signifi—
cantly fewer fleas per animal (1.6) than the immature females (2.7),
the immature males (3.2) or the adult males (3.0). The low number of
fleas parasitizing the adult females was not a factor of the number of
squirrels in this host.class that were parasitized.

A "mangy condition" frequently observed in squirrels is apparently
not related to flea parasitism. Although, it does occur at the same
time as the flea population peak in March, it is perhaps more related

to population densities and climatological factors.

18

LITERATURE CITED

Allen, R. M. 1956. A Study of the Populations of the Rabbit flea
Spilgpsyllus cuniculi (Dale) on the Wild Rabbit Ogyctolagus
cuniculus in Northern Scotland. Proc. R. Ent. Soc., London.
31: 145-152.

 

Allen, D. L. 1943. Michigan Fox Squirrel Management. Game Division
Pub. 100. Michigan Dept. of Cons. 404 pp.

Baker, R. H. 1944. An Ecological Study of Tree Squirrels in Eastern
Texas. Jour. Mamm. 25: 8-24.

Brown, L. G. and L. E. Yeager. 1945. Fox Squirrels and Gray Squirrels
in Illinois. Bull. Ill. Nat. Hist. Survey. 23: 5: 1-95.

Dunn, L. H. and R. R. Parker. 1923. Fleas Found on Wild Animals in
the Bitterroot Valley, Montana. U. S. Pub. Health Rpts. 38:
2763-2775.

Ellis, L. L. 1955. A Survey of the Ectoparasites of Certain Mammals
in Oklahoma. Ecology 36: 12-18.

Evans, F. C. and R. B. Freeman. 1950. On the Relationship of Some
Mammal Fleas to Their Hosts. Ann. Ent. Soc. Amer. 43: 320-333.

Fox, I. 1940. Fleas of the Eastern United States. The Iowa State
College Press. Ames, Iowa. 191 pp.

Fuller, H. S. 1943. Fleas of New England. Jour. N. Y. Ent. Soc.

Holland, G. P. and A. H. Benton. 1968. Siphonaptera from Pennsylvania

Hopkins, G. H. E. and M. Rothschild. 1953-1956. An Illustrated
Catalogue of the Rothschild Collection of Fleas in the British
Museum of Natural History. Brit. Mus. (Nat. Hist.) London.
V010 1-40

Hopkins, G. H. E. 1954. Nomenclatural Notes on the Flea of the Gray
Squirrel. Entomologist. 87: 197-198.

19

20

Knipping, P. A., Morgan, B. B. and R. J. Dicke. 1950. Preliminary
List of Some Fleas From Wisconsin. Wis. Acad. Sci., Arts and
Letters. 40: 199-206.

Layne, J. N. 1954. The Biology of the Red Squirrel, Tamiasciurus

hudsonicus lgquax (Bangs), in Central New York. Ecol. Monog.

 

Mathewson, J. A. and K. E. Hyland. 1964. The Ectoparasites of
Rhode Island Mammals. Jour. Kans. Ent. Soc. 37: 157-163.

Mead-Briggs, A. R. and A. J. B. Rudge. 1960. Breeding of the Rabbit
Flea, Spilopsyllus cuniculi (Dale). Nature, London. 187:
1136-1137.

 

Osgood, F. L. 1964. Fleas of Vermont. Jour. N. Y. Ent. Soc. 72:
29-33 0

Parker, D. D. 1958. Seasonal Occurrence of Fleas on the Antelope
Ground Squirrel in the Great Salt Lake Desert. Jour. Econ. Ent.
51: 32-36.

Poorbaugh, J. H. and H. T. Gier. 1961. Fleas (Siphonaptera) of Small
Mammals in Kansas. Jour. Kansas Ent. Soc. 34: 198-204.

Rapp, W. F. and D. B. Gates. 1957. A Distributional Check-List of
the Fleas of Nebraska. Jour. Kansas Ent. Soc. 30: 50-53.

Ryckman, R. E., Lindt, C. C., Ames, C. T. and R. D. Lee. 1954.
Seasonal Incidence of Fleas on the California Ground Squirrel
in Orange County, California. Jour. Econ. Ent. 47: 1070-1079.

Tindall, E. E. and R. F. Dorsie. 1961. New Delaware Records for
Mammalian Ectoparasites, Including Siphonaptera Host List.
Bull. Brooklyn Ent. Soc. 56: 89-99.

HICHIGQN STRTE UNIV. LIBRQRIES

| 11 Ill

3129301 1072869