Oecologia (Berl.) 15, 287--304 (1974) by Springer-Verlag 1974

The Population Dynamics and Host Utilization of Geomydoecus oregonus, a Parasite of Thomomys bottae

R. W. Rust Department of Entomology and Applied Ecology, University of Delaware, Newark, Delaware 19711

Received January 28, 1974

Summary. This study treats the population dynamics and host utilization of Geomydoecus oregonus Price and Emerson, a mallophagan parasite of the pocket , Thomomys bottae (Eydous and Gervais). Over 135000 lice were collected from 393 over a period of 20 months. Average infestation on all gophers was 357 lice. Bimonthly mean densities showed an increase in June-July of both years, and these data were statistically different from the rest. Population age structure remained relatively constant in time with 9.3% females, 7.2% males, 47.4% nymphs and 36.1% eggs. Sex-age class separation of the gophers showed juveniles of both sexes to average 86 lice; subadult males averaged 210 lice; adult males averaged 544 lice; subadult and adult females averaged 255 and 296 lice, respectively. Lice were not randomly distributed on the gopher, but were most numerous on the head and anterior dorsal body. Lice eggs were restricted to hairs around the ears and eyes of the host. Over 80% of the sampled had eggs restricted to that region. Embryonic development and eelosion of G. oregonus proceeded over a wide range of environmental parameters. Over 80% of the ova tested survived and hatched in conditions between 33 ~ and 37~ and 22 % and 84 % relative humidity. The greatest survival was 98% at 35~ 75% R. H. and in a 3% COs atmosphere. The generation time of G. oregonus was 40• days. Duration of embryogenesis and nymphal stadia approximated 10• days for each. Adult lice lived 30-t- days on gophers. Age frequency mortalities were calculated as 0.02 for eggs, 0.18, 0.24, and 0.06 for nymphal instars and 0.50 for adult lice. This indicates a Type II survivorship curve. There was a direct linear relationship between the number of female lice on a gopher and the number of lice eggs. The average number of eggs per female was four. Using the pivotal frequency for reproduetives, it was possible to calculate R o for the at 1.272. Thus, r was equal to 0.24 per generation or 0.006 per day, and the population doubles in 2.8 generations. Speculations regarding population regulation are also included.

Introduction The relationshi p between an ectoparasite and its host is governed by a complex set of interacting biotic and abiotie parameters. For the ecto- 1913 Oecologia(Bet1.), Vol. 10 288 R.W. Rust parasite, the biotic parameters are in general related to or originate from the host. The degree of adaptation of the parasite to the biotic parameters corresponds to the host specificity of the parasite. The degree of host specificity also indicates the adaptations of the parasite to the abiotie parameters influencing the host. Mallophaga, biting lice, are host specific (Hopkins, 1949; Werneck, 1950), and the restriction of all life stages to the body of the host refines the specificity and adaptations of a louse to its host and host's environment (see Clay, 1949; Ash, 1960; Baum, 1968; Foster, 1969, for avian lice; Ewing, 1924; Hopkins, 1949; Waterhouse, 1953; Craufurd-Benson, 1941; Scott, 1952; Murray, 1957b-d, 1965; Hopkins and Chamberlain, 1969, for mammalian lice). For lice, the combination of interacting mortality factors and a species innate capacity for increase will determine the population level on a particular host. Evans and Smith (1952), in their classic demographic analysis of Pedi. culus humanus Linnaeus, were the first to study the reproductive poten- tials of an ectoparasite. Murray and Gordon (1969) modeled the popu- lation dynamics of Bovicola ovis (Linnaeus), a biting louse of sheep. They showed mathematically the development of the population from a low summer density to a winter high density. The genus Geomydoecus, parasites of the fossorial rodent family Geomyidae, possesses morphological adaptations not found in other trichodectid lice (Ewing, 1936). Degenerate adaptations as the absence of eyes and lack of abdominal spiracles and tergites might be a response to the altered conditions of the host's burrow environment (Kennerly, 1964; McNab, 1966; Darden, 1970), and specializations of the antennae and well-developed hair-groove could relate to life on the gopher. Re- cently, Price and Emerson (1971) studied Geomydoecus and found 45 taxa on 29 of the 36 species of gophers. Most gophers have only one louse species; exceptions were hosts with extensive geographical distributions. The relative accessibility, small size, and peculiar habitat of the host provide a system for analyzing a parasite's population dynamics. Thus, Geomydoecus oregonus Price and Emerson, a parasite of Thomomys bottae (Eydoux and Gervais), was selected for analysis and interpretation.

Methods and Materials Field A minimum of 20 pocket gophers per month as live-trapped from two irrigated alfaHa fields located on the University of California, Davis Farm, and 2 consecutive months of collection data were combined for analysis. The field sizes, agricultural practices, measurement of environmental parameters, gopher collecting and labora- tory handling techniques and parasite recovery methods are treated in Rust (1973a, b). Three age classes were established for both sexes of gophers: juvenile, subadult and adult. Separation was based on reproductive condition, skull features, and body Population Dynamics of the Gopher Louse 289 measurements (Miller, 1946, 1962; Hansen, 1960; Thaeler, 1968). Juveniles had the following sutures unfused: parietal-squamosal, maxilla-alisphenoid, and alisphenoid- squamosal; the dorsal surface of the maxilla (in the zygomatic arch) and palatine showed distinct porosity. Juveniles had gray pelage and were still in association with their mother, as determined from trapping. Subadults had partial fusion of the cranial sutures, reduced porosity and brownish pelage. Subadult females had a closed pubic symphysis and limited follicular development; subadult male testes were only 3-6 mm long. Adults had skull sutures and nonporous bones. Adult females had the pubic symphysis open, placental scars, embryos, or were lactating. Adult males had testes from 7-15 mm long. Body length and maximum skull length and width showed overlap due to a continuous breeding population (Miller, 1946) but were also used in the separation. Time for the age categories would be approximately one and one-half to two months for juveniles, 2-5 months for subadults and the mean length of life for adult gophers has been estimated by Howard and Childs (1959) as 7.6 months for males and 12.3 months for females. Laboratory Eelosion and first instar survival in vitro were examined under various environ- mental conditions. 8 different temperatures (27-41~ were maintained in Percival environmental chambers. The chambers were set for total darkness with the photophase only when eggs or nymphs were checked. Eight different relative humidities (10-96%)were maintained in liter desiccator jars containing various saturated salt solutions (Winston and Bates, 1960). Eclosion was tested in normal air and a 3% COz atmosphere. The high CO 2 atmosphere was created by removing 3 % of the gas volume from the jars and replacing it with CO 2. The CO 2 concentration was measured with a Unieo Gas Analyzer. Four replicates of 25 eggs each were tested at 64 different temperature-relative humidity combinations. Translucent eggs were plucked individually from gophers for the temperature-relative humidity tests. Eclosion and nymphal survival in 3% CO 2 atmosphere were recorded at 33 ~ and 35~ both at 62%, 75% and 84% R. It. Nymphal survival was also checked in normal CO 2 atmosphere. The duration of embryonic development was obtained by allowing female lice to oviposit in small petri dishes. This gave eggs of a known age (-4-12 h), and these eggs were placed in separate containers in various temperature-relative humidity conditions. Louse survival on the host was checked in a normal CO 2 atmosphere at various humidities. Humidities tested were 35 % • 5, 50 % =J:5, 75 % • 5 and 100 % -- 2 and were maintained in environmental chambers at 21~1771. Louse demography was studied (in rive) in an artificial burrow system simulat- ing natural conditions (Darden, 1970). Four artificial burrows were housed in adjoining chambers in a soil box (60.9 X 91.4 X 121.9 cm). Burrows were constructed of 7 • 7 mm hardware cloth welded together. A system consisted of an above ground holding cage (14.2 • 20.3 X 30.4 ore), an initial diagonal tunnel (30 cm), a horizontal tunnel (24 cm) and a nest chamber (19x24 era). The tunnels were 7.5 cm in dia- meter. Before an was placed in a burrow system, carrot pieces and fresh alfalfa were placed in the next chamber. The system was packed with soil and buried in the soil box, with 25 to 30 em of soil over the nest chamber. The surface was moistened with approximately 2 1 of water and allowed to stand for 24 h before a gopher was introduced. Additional fresh alfalfa was provided daily. Two Plexiglas tubes (6 mm O. D.) were attached to the roof of the nest chamber and horizontal tunnel. Both gas samples and temperatures were measured via the tubes. The entire system was housed in a greenhouse with a regime of controlled temperature and 290 R.W. Rus~ natural light. Every seventh day of an experiment an additional 21 of water were added to the surface. Two different treatments were designed for gophers placed in the artificial system. Newly eclosed first instar nymphs were introduced on a clean gopher, and the animal was released in a system for a set period--6, 10, 15, 20, 25, and 30 days. Freshly caught gophers were cleaned of parasites, empty lice eggs were plucked from the animal, and the remaining lice eggs were counted and mapped. These animals were released in a system for 20, 30 and 40 days. At the end of an experi- ment, the animal was sacrificed and cleaned. Statistical analyses used in this study were Student's t-test, linear regression and correlation, single class analysis of variance and curvilinear regression.

Results Macro- and Microenvironmental Parameters During this study, average monthly temperatures varied from a low of 3.3~ in winter to a high of 25.4~ in summer. The Davis weather station mean maximum temperatures were 16.2 ~ and 30.2~ and the mean minima were 5.1 ~ and 11.8~ in winter and summer, respectively. Average monthly precipitation was from a high of 61.5 mm in winter months to 1 mm in summer months. Average monthly soil temperatures at 101 mm depth ranged from a low of 7.2~ in January to a high of 32.3~ in July. The soil moisture averaged 6.6% in winter and 5.3% in summer. The high summer soil moisture level was maintained by irri- gation. The alfalfa fields were flood-irrigated at approximately 30-day intervals from April to September, with water covering the fields, except irrigation levees, for nearly 24 h. The relative humidity of the burrow measured 97.4• (N-=96). Kennerly (1964), McSTab (1966), and Darden (1970) also recorded the near saturated atmosphere of the gopher burrow. Darden (1970) deter- mined that the COS concentration of the gopher burrow averaged 2.3 %, and 03 averaged 18%. MeNab (1966) found an average COS concen- tration of 1.4 % in the burrow of Geomys pinetis Rafinesque, and Kennerly (1964) measured only 0.8 % COS in Geomys bursarius (Shaw) burrows. The burrow temperatures showed less fluctuation than ambient temperatures and were out of phase with them. Rust (1973b) showed the relationship between ambient and burrow temperatures at two extreme times, August and January. In the artificial burrow system, the nest chamber temperature averaged 20.5-V0.6~ C (N-=30) and the COS concentration was 2.4-4-0.1% (N=24). The soil moisture averaged 6.2% and ranged from 7.2% to 4.1%. The greenhouse temperature was maintained between 14.4 ~ and 25.0 ~C. Thus, the conditions for experimentation are considered similar to those that would be found in a natural burrow system, except during the cooler winter months. Population Dynamics of the Gopher Louse 291

Population Density and Structure Over 135000 Geomydoecus oregonus were removed from the gophers; 5 animals were clean of lice. The average infestation for all gophers was 357.5q-13.9 lice, and 1941 was the maximum number of lice recovered from an adult. Five other gophers (2 females and 3 males) had more than 1000 lice on them. The bimonthly mean densities (Fig. 1) showed an increase in June-July of both years, and these sample means were found to be statistically different from the rest (F=2.7417, P~0.005). The 15 "dry land" gophers from August-September 1971 had a mean density of 376.1-~66.4 lice. This was not found to be significantly different from the mean of the "irrigated land" sample for the same period or the overall mean (t=0.4253, 0.9~P ~0.5 and t=0.2709, 0.9~P~0.5, respectively). The population age structure remained relatively constant in time with 9.3% females, 7.2% males, 47.4% nymphs and 36.1% eggs (Fig. 1). Juveniles had an increased proportion of eggs (44.8) and reduced nymphs (36.2), but adults were nearly equal. Geomydoecus oregonus has three nymphal instars as determined from head capsule measurements and increased number of abdominal setae. The proportion of the nymphs represented by each was 19.9% first, 13.3% second and 14.2% third. 1Vlean louse populations on male and female gophers were strikingly different with males averaging over 430 and females under 290. Further analysis of the population based on gopher age classes of juvenile, sub- adult, and adult revealed a greater difference. Juvenile gophers (N~--18), regardless of sex, averaged only 86.3~ 13.7 lice. Subadult males (N=46) had 210.8~19.8 lice, and adult males (N----120) averaged 544.2-~25.6. However, subadult (2V=71) and adult females (N----123) averaged nearly the same number of lice 255.0q-26.3 and 296.0• respectively. The bimonthly sex-age class averages are plotted in Fig. 2. When the sex-age class averages are considered, the seasonal changes are at most slight although when combined, they show a statistically significant increase in June-July. This early summer increase in 1970 was a result of large louse populations on adult male gophers with 8 of 10 having over 600 lice. The 1971 louse increase resulted from an increase in both adult female and subadult populations. Three of the adult females had over 800 lice, which if removed from the sample, reduces the mean to 299. Lice were not randomly distributed on the gophers, but were most numerous on the head and anterior dorsal body. Selective cleaning produced the following distribution of lice: 19.7% dorsal head, 14.0% ventral head, 49.5% anterior dorsal body, 7.5% posterior dorsal body and 9.3 % ventral body. Lice on the ventral body were restricted to the area between the forelegs and axillae. If the above were based on the 292 R. W. Rust

600 5OO

g~~ 400 ~300 I 2OO 6O Nymphs 5O 40 .... ~ ~ "'"~"~ 30 20 Females 10

-- Males A-IM J'-J A:S O'-N D% F'-M A:M J:J A'-S O'-N Months Fig. 1. Population structure of Geomydoecus oregonus from 378 Thomomys bottae collected from Davis, California, from April 1970 through November 1971. Left and above: mean population density for all gophers, vertical lines equal standard error. Right and below: population age structure

800

700

600

5OO

~ 400 Adutt Femc~le~/"/'x"-, g /, x, 300 X ,,... ~ ~~?" "',",,~"- ...... 7~ ""...... 200 ..~" " Subadu[[ Mole 100 Mean Juvenile

A"M J"J A~S' O"N D~'J F~'M AIM J.J A"S O~'N Months :Fig. 2. Mean population density of Geomydoecus oregonus based on sex-age class separation of Thomomys bottae. Juvenile gophers represented by overall mean only Population Dynamics of the Gopher Louse 293

Fig. 3. Distribution of Geomydoecus oregonus eggs on 89 Thomomys bottae, above. Mean regional body temperatures in degrees centigrade, below

proportion of the gopher's body inhabited, then 92.5 % of the lice were found on 55 % of the animal's surface.

Egg Distribution, Numbers and Environmental Tolerance Distribution of the lice eggs was more restricted than thedistribution of nymphs and adults. The proportional distribution of eggs on 89 gophers is shown in Fig. 3. Over 80 % of the gophers had lice eggs located around the ears and eyes. If juvenile animals and adults with high infestations (800q-lice) were removed from the sample, the lice eggs on the mid-dorsal back were excluded from the distribution. An egg of G. oregonus is glued to the base of the hairs from one follicle and only one egg]hair group was observed. Although oviposition was not observed, the placement of the egg on the hairs suggests an oviposition behavior similar to that de- scribed for other lice (Murray, 1957a). The chorion remains attached after eclosion and until the hair group is shed with the subsequent molt. There was a direct linear relationship (Fig. 4) between the number of female lice on a gopher and the number of lice eggs (bz3.253=k0.208), with a correlation of 0.868. Thus 74% of the egg variation was explained by the presence of the female lice (r~=0.7396). The average number of eggs per female was 4 (3.93) and ranged from 0.6 to 11.2. If juveniles, females with litters, and old gophers were removed, the range was narrowed to 2.9 to 8.2 eggs/female. In terms of eggs per day, G. oregonus lays 0.2 eggs/day in an average life span. 294 R.W. Rust

~6 400 .', ' '"

350

3oo

~" 250 0 ~200 ....

-'0 100 , ,,,-/:," , , r =0.868 n 89

50 ~.~,," "~.X, ,: , : ,

, _ , , , , , , 20 40 60 80 100 120 140 Female Lice per Gopher Fig. 4. Regression of number of female Geomydoecus oregonus per Thomomys bo#ae to number of G. oregonus eggs per gropher

Embryonic development and eclosion of G. oregonus proceeded over a wide range of environmental parameters. An 80 % or greater plateau of ova survival and hatching (Fig. 5) was found between 33-37~ and 22-84% relative humidity. The greatest survival was 96% at 35~ and 75% R. It. The mean rectal temperature of T. bottae was 36.5~0.4~ (N=26). Darden (1970) recorded an average of 36.7~ for the same species. This temperature was near the upper end of the survival range. However, surface temperatures (Fig. 3) of the cheek and head region were appro- ximately 1~ lower and corresponded to the highest eclosion rate. Development and hatching occurred in 3 % CO 2 atmosphere and with a slightly higher survival rate than eggs reared in a normal CO S atmos- phere. Ninety-eight percent of over 500 eggs hatched in vitro and 97.7% of the eggs hatched on gophers in the artificial burrow system. Darden (1970) determined the CO s concentration of the gopher burrow to average 2.3 %, and O 3 averaged 18 %. I found an average CO 2 concen- tration of 2.4% in the nest chamber of the artificial burrow system. Normal atmospheric CO 2 was determined to slightly reduce the survival of the louse. Eggs on 3 gophers maintained in normal atmosphere and exposed to a relative humidity greater than 75 % for 30 days hatched and developed to third instar and adult lice with an average mortality of 56.3%. Population Dynamics of the Gopher Louse 295

100 100

m "U

zs zs

D -I

50 50 C

U| ~..7

D- e~

25 25

~'/ qb

Fig. 5. Survival and eclosion of Geomydoecus oregonus eggs reared in vitro under constant temperature and relative humidity

Population Development The embryonic development of G. oregonus averaged 10.5days (N=18) and ranged from 9 to 12 days. This time plus the duration of nymphal development (Table 1) indicates a generation time of 40=k6 days. Duration of nymphal stadia approximated 10• days for each. The longevity of adult lice was not ascertained. However, 6 females and 3 males out of a total of 80 lice lived 30~-days on gophers in the artificial burrow system. The mortality rate (Table 1) from cohorts beginning with first instar lice was extremely high as indicated by a 50 % mortality of first instar lice after 6 days. This high death rate was probably induced by handling 296 R.W. Rust

Table 1. Recovery of Geomydoecus oregonus from gophers, each artificially infested with newly hatched first instar nymphs. Gophers maintained in artificial burrows

Days Inoc- I~umber and stage of development Total % after ulum Mortality infesta- Immatures Adults tion 1st 2nd 3rd ~

6 36 18 .... 18 50.0 10 58 1 16 ------17 70.3 15 49 -- 8 1 -- -- 9 81.7 20 32 -- 2 2 -- -- 4 87.5 25 31 -- -- 2 -- -- 2 93.5 30 40 ------2 1 3 90.0

the lice and a delay (12 h) before they could feed or assume a correct posture on the gopher. The results (Table 2) obtained from cohorts starting with the egg stage, indicate a Type II survivorship curve for the louse population. Thus, age frequency mortalities can be calculated as 0.02 for eggs, 0.18, 0.24, and 0.06 for the nymphal instars, and 0.50 for adult lice. Due to the 10 day age difference of eggs used to initiate these tests, females and new eggs were already present on the gophers in the 30 day samples, and at 40 days, new first instar lice appeared in the samples. Using the pivotal frequency for reproductives, it is possible to estimate the net reproductive rate, Ro=Zlxm x. Thus, R o for G. oregonus is 1.272 (0.53 X 2.4= 1.272). This assumes a 1:1.2 sex ratio, equal survival of sexes and an average of 4 eggs per female louse. The intrinsic rate of increase (r) can be calculated from Ro=e% Thus, r is equal to 0.24 per generation or 0.006 per day, and the population doubles in 2.8 genera- tions.

Table 2. Recovery of Geomydoecu8 oregonus from gophers, each naturally infested with eggs. Gophers maintained in artificial burrow systems

Dura- Num- Inoc- Number and stage of development Mortality %a tion ber ulum age days of eggs Immatures Adults Eggs % % inter- gophers eggs total val 1st 2nd 3rd ~ !~ Unhatched new

20 2 295 65 127 34 7 -- 2.3 20.0 18.0 30 4 551 -- 7 187 42 43 11 21 1.9 49.4 30.0 40 5 914 13 b -- 30 150 180 ? 157 ? 60.7 ll.0 a Determined initially from 2 % egg mortality and rounded to nearest whole number. b First instars of second generation. Population Dynamics of the Gopher Louse 297

Discussion Population Structure and Development Coffman (personal communication) studied Geomydoecus geomydis (Osborn), a parasite of Geomys bursarius. He found nearly twice the mean density of lice as I did for G. oregonus as well as a seasonal shift with a spring increase for G. geomydis in South Dakota. Geomys bursarius is one and one-half to two times as large (body weight) as T. bottae from the Davis area (Hall and Kelson, 1959), breeds once a year, is subject to different climatic regimes and possibly different microenvironmental conditions such as lower CO S concentration (Kennerly, 1964). Size of host (decreased surface area), continous breeding (Miller, 1946), the effects of a different climatic regime, and flood irrigation are elements which probably reduce the population size of G. oregonus on T. bottae in the Davis area. Other trichodectid lice are found primarily on fissiped carnivores, hyraxes and ungulates (Hopkins, 1949; Emerson, 1964), and seasonal fluctuations in other small mammal mallophagans are unknown to me. Large mammal mallophagan species show a general late fall to winter increase and a summer decline (Crauford-Benson, 1949 ; Matthysse, 1944, 1946; Scott, 1952; Enzie, 1956; Murray, 1960, 1963a, b). Vysotos- kaya (1950) and Murray (1961) have suggested that the fall to winter increase may relate to the host's responses to the severity of the winter conditions. However, the general uniformity of G. oregonus population structure through time may result from the modified summer environ- mental conditions brought about by irrigation. The homogenous condi- tions in irrigated lands are also responsible for year-round reproduction of gophers (Miller, 1946). If the louse is affected by the environmental conditions in the late summer as is the host (Miller, 1946; Howard and Childs, 1959), then a population decline may be predicted. The small "dry land" sample mean, however, equalled the mean number on hosts from irrigated lands. Howard and Childs (1959) suggested that gophers may retreat to lower burrows and seal shallower ones to escape the summer heat. This could equalize environmental conditions and allow louse reproduction to proceed. The different louse densities on the sex-age classes of gophers result from at least two factors : the size of the animal and the loss of a portion of adult female's parasites to the young of each litter. Thus, the time of maximum gopher reproduction would also be the most disruptive period for the lice on female gophers. The reduced mean density of lice on adult females in February-March is concomitant with maximum gopher reproduction (Miller, 1946), and the increased density of lice in the early winter months corresponds to minimum gopher reproduction. The solitary habits of gophers (Ingles, 1952 ; Howard and Childs, 1959) suggest that the exchange of parasites takes place primarily between females with 298 1%.W. Rust litters and between litter mates. This explains the adult sex differences and small louse population (inoculum) on juvenile gophers. Miller (1946) found an average of 5 young/litter and 2 litters/year for gophers inhab- iting irrigated fields in the Davis area. Thus, ideally 84% of the female's louse population could be lost twice a year. Actual measurements of the proportion of the population lost ranged from 56% to 66% for litter sizes of 2 and 4 (N=4). Lice are generally found on the head and foreparts of the small mammals and may spread over the dorsal and lateral aspects of the host as the population increases (Vysotoskaya, 1950; Dubinin, 1953; Murray, 1961). In general, louse distribution is controlled by the grooming characteristics of the host (Murray, 1961). Observations of gopher grooming indicate that lice can be nipped from the ventral body and possibly the hind quarters. The short neck and large pectoral girdle prevent more extensive preening with incisors. Preening with the hind and fore claws in the form of scratching probably does not remove many lice because of their tenacity. This action could destroy stationary nits, but gophers were observed to rub the head region with their fore claws rather than scratch with claw tips. Preening is an important control agent in the distribution of nits (Murray, 1961), but hair size and configuration (Murray, 1957b-d; Hopkins and Chamberlain, 1969) and host regional body temperatures (Murray, 1957e, 1963b, 1965) also affect their distribution. Another factor on the gopher is the structure of the hairs on which eggs are glued. Most hair follicles on the gopher contain 9-18 hairs (Morejohn and Howard, 1956), but the character of hairs per follicle differs in the various body regions. Hair follicles on the dorsal and lateral aspects of the head bear 4-6 large guard hairs and 7-12 fur hairs, whereas follicles on the dorsal body lack the stiff guard hairs. Follicular separation is nearly equal over the body, with the hairs follicles being separated by 0.25-0.33 ram. The response of female lice to a particular hair con- figuration for oviposition may result in the limited distribution as seen on over 80% of the gophers sampled. On juveniles, all hairs are of the fur type and thus the possible wider distribution of nits, but as the gopher loses its juvenile pelage the hair per follicle change to the fur hair-guard hair arrangement. The wider distribution of nits on gophers with large infestations (800-~) possibly indicates a spreading to less favorable sites. Molting restricts the availability of oviposition sites and may result in a loss of eggs already laid. Evidence for the latter is sparse, although Vysotoskaya (1950) considered molting a major factor for population decline. Three gophers with active molt lines in and around the ears and eyes had no unhatched nits in the area. When examined 30 days later the Population Dynamics of the Gopher Louse 299 molt line had moved posteriorly and the empty eggs were gone. Morejohn and Howard (1956) studied the molt patterns of captive and field gophers. They found two distinct seasonal pelages, winter and summer, and a constant molt period with a short non-molt period from December to February. It is therefore possible to have a reduction in the hair follicles available twice a year. The dense replacement hairs and thickened dermis during active molting renders an area inaccessible to lice and oviposition. Lice were not observed in this thick, new hair. Using an average juvenile gopher inoculum of 86 lice and no upper restriction on population growth, the average mean density of lice on subadults can be reached in about 200 days and for adult males in 300 days. The mean louse density of adult female gophers would also be reached in 200 days, and this represents an approximation of sexual maturity (Miller, 1946). On male gophers, the louse density would reach nearly 1000 in 400 days, which is the mean longevity of male gophers (I-Ioward and Childs, 1959). The small sample of gophers with 1000 or more lice suggests that factors must be operating to maintain the density at a lower level than predicted. To examine the population growth curve, an arbitrary measure of time was necessary because of the difficulty in accurate aging of gophers. The measurement selected was the increase in surface area of the gopher (cm 2) and was obtained by combining body length and circumference of the thorax cavity, legs and tail excluded from the measurement. An accuracy check was made by measuring fresh, fiat skins. It has been shown that gophers continue to grow as adults (Miller, 1952 ; Howard and Chflds, 1959). Therefore, a clumping of older animals in a younger age group is not expected. The average densities on categoric age groups showed nearly equal numbers of lice on subadult and adult female gophers. Thus, the louse population on a female gopher should reach an asymptote at some body size or point in time. The relationship between female gopher area and lice per area was curvilinear (Fig. 6) and the curve was defined by a second degree polynomial, Y=--8.9990+O.15884X--O.OOO56X 2. On female gophers, the louse population leveled at 2.2 lice per cm ~ over 140 cm 2 of body area, or 308 lice. The mean densites of lice on male gophers differed according to age groups, and the relationships were analyzed for both linear and eurvilinear regression. The test for curvi- linear regression (Snedecor, 1956) was highly significant (F--~7.1604, P>0.001), and a linear relationship was rejected. The curve (Fig. 6) was defined by a second degree polynominal, Y------5.4723-~0.09352X-- 0.0026X 2. The louse population on male gophers leveled at around 2.9 lice per cm ~ over an area of 175 cm ~, or 507 lice. The respective asymptotes are equal to the overall mean densities for adult animals. 300 R. W. Rust

4 % Adult Mate 3 o. SubaduIt ..,.'J ...... _J Male ..,'": ...... 2 ,.'..;~;:'~"~ j Adult Ferna[e

Juveniles ..;'if"'" Female

, I " | , t I , i .... , 4O 80 100 120 140 160 180 200 Gopher Body Area (cm 2) Fig. 6. Population growth curve for Geomydoecus oregouus on male and female Thomomys bottae as determined from regression analysis of the relationship of the number of lice per body area to the total body area available

Population Regulation Population growth as presented above can be controlled by both density dependent and density independent factors. Food is not consid- ered to be an important factor, since it is in constant supply on the host (Waterhouse, 1953). One dependent factor appears to be the availabil- ity of a proper oviposition site. This was indicated by the narrow distri- bution of nits on gophers and the inaccessibility of a site until the hairs were replaced. Cannibalism has been reported for avian mallophagans (Waterston, 1926; Rothschild and Clay, 1952; Nelson, 1971). Hopkins and Chamberlain (1969) found destruction of nits, as high as 25% in in vitro colonies of B. crass@es. I found no evidence for cannibalism in G. oregonus. Empty chorions on the gophers appeared intact and undam- aged. Lice held in small containers for 5 days did not destroy or consume newly laid eggs or immature individuals. This does not, however, exclude the possibility of cannibalism for G. oregonus. Density independent factors such as preening, molting, and social interaction of the host can lead to population reduction or control. Environmental conditions in the Davis area and especially in irrigated fields do not appear to restrict the population densites. Irrigation may destroy a proportion of the population at the time of flooding. Wet gophers were found in the fields during the initial stages of irrigation, but more commonly, dry animals were found on field levees, murray (1963a) showed that adults and nymphs of B. ovis can survive immersion in Population Dynamics of the Gopher Louse 301 water for 1 h with less than 20 % mortality. Eggs could be immersed for up to 7 days before a substantial reduction in hatching occurred. It is therefore possible to have a reduction in the nymph and adult population with each irrigation. The distribution of the lice on the gopher and the inability of the gopher to withstand immersion for even one hour pro- bably reduces the severity of irrigation. Interaction and competition with other ectoparasites may influence the upper densities (Hopkins, 1949; Vysotoskaya, 1950). Geomydoecus oregonus co-inhabits the gopher in the Davis area with three other parasites (Rust, 1973 a, b). The small size and restricted distribution of Geomylichus n. sp. and the nidicolous habits of Haemolaelaps geomys Strandtmann remove them from interactions with the louse (Rust, 1973 b). Hirstionyssus ]emuralis Allred is found both in the nest and on the host in low numbers and are not considered to pose any serious deterrent in the louse density increase. Finally, the physiological state of the host, either environmentally or pathologically induced, may influence population size. Sick animals have been reported by several workers as having unusually heavy infestations of lice (Thompson, 1936; Spencer, 1939; Craufurd-Benson, 1941; and Eichler, 1942). The state of health of the 6 gophers with heavy infesta- tions could not be determined. However, they appeared healthy. The only "unhealthy" animals encountered were wounded males, but this is probably a common condition resulting from their pugnacious behavior (Howard and Childs, 1959). Increased louse reproduction induced by ingestion of host hormones, as in the classical case of Spilopsyllus cuniculi Dale and its host the rabbit (Mead-Briggs and Rudge, 1960; Rothschild and Ford, 1964 a, b), is not considered important. The food and feeding habits of trichodectid lice (Waterhouse, 1953) would prevent contact with host's blood. If estrogens enter the dermal tissue fluids, then they may be ingested and could then be considered an important factor.

Acknowledgements. I thank K. Emerson and 1%. D. Price for their identification of Geomydoecus oregonus. 1%. F. Denno, D. L. McLean, R.W. Thorp and R.K. Washino read various stages of the manuscript. Their criticisms and suggestions were appreciated. Charles Coffman provided unpublished data which was most helpful. Aileen K. Rust assisted throughout the study. Her comments and sugges- tions are sincerely appreciated.

References Ash, J. S. : A study of the Mallophaga of birds with particular reference to their ecology. Ibis 120, 93-102 (1960) Bantu, H. :Biologie nnd Okologie der Amselfederl~use. Angew. Parasit. 9, 129-175 (1968) Clay, T.: Some problems in the evolution of a group of eetoparasites. Evolution 3, 279-299 (1949) 20 Oeeologia(Berl.), Vol. 15 302 R.W. Rust

Craufurd-Benson, H. J. : The cattle lice of Great Britain. Parasitology 88, 331-358 (1941) Darden, T. R. : Respiratory adaptations of a fossorial mammal, the pocket gopher (Thomomys bottae). Unpublished Ph.D. Dissertation, Univ. California, Davis (1970) Dubinin, V. B.: Parasitic fauna of Murid rodents and its changes in the fauna of the Volga Delta. Parazit. Sbornik 15, 252-301 (1953) Eichler, W. : Die Entfaltungsregel und andere Gesetzm~Bigkeiten in den parasito- genetischen Beziehungen der Mallophagen und andere st~ndiger Parasiten zu ihren Wirten. Zool. Anz. 187, 77-83 (1942) Emerson, K. C.: Checklist of Mallophaga of North America (north of Mexico), p. 1-171. Dugway, Utah: Dugway Proving Grounds 1964 Enzie, F.D.: Mange and lice of horses and mules. Yearbk. Agr. USDA 1956, 555-558 (1956) Evans, :F. C., Smith, F. E.: The intrinsic rate of natural increase for human louse, Pediculus humanus L. Amer. Nat. 86, 299-310 (1952) Ewing, H. E. : On the taxonomy, biology and distribution of the biting lice of the family Gyropidae. U. S. Nat. Mus. Prec. 68, 1-42 (1924) Ewing, H. E.: The taxonomy of the mallophagan family Trichodeetidae, with special reference to the New World fauna. J. Parasit. 22, 233-246 (1936) Foster, M. S.: Synchronized life cycles in the Orange-crowned Warbler and its mallophagan parasites. Ecology 50, 315-323 (1969) Hall, E. R., Kelson, K. P.: The mammals of North America, Vol. 1. New York: Ronald Press 1959 Hansen, R. M.: Age and reproductive characteristics of mountain pocket gophers in Colorado. J. Mamm. 41, 323-335 (1960) Hopkins, G. H. E. : The host-associations of the lice of mammals. Zool. Soc. London Prec. 119, 387-604 (1949) Hopkins, D. E., Chamberlain, W. F. : In vitro colonization of the goat biting lice, Bovicola crassipes and B. timbata. Ann. ent. See. Amer. 62, 826-828 (1969) Howard, W. E., Childs, H. E. : Ecology of pocket gophers with emphasis on The- morays bottae mewa. Hilgardia 29, 277-358 (1959) Ingles, L. G. : The ecology of the mountain pocket gopher, Thomomys monticola. Ecology 88, 87-95 (1952) Kennerly, T. E.: Mieroenvironmental conditions of the pocket gopher burrow. Texas J. Sci. 16, 395~41 (1964) Matthysse, J. G. : Biology of the cattle biting louse and notes on cattle sucking lice. J. econ. Ent. 87, 436-442 (1944) Matthysse, J. G.: Cattle lice, their biology and control. New York State Agr. Exp. Sta. Ithaca Bull. 832 (1946) MeNab, B.K.: The metabolism of fossorial rodents: a study of convergence. Ecology 47, 712-733 (1966) Mead-Briggs, A. R., Rudge, A. J. B.: Breeding of the rabbit flea, Spilopsyllus cuniculi (Dale): Requirement of a "factor" from a pregnant rabbit for ovarian maturation. Nature (Lond.) 187, 1136 (1960) Miller, M. A.: Reproductive rates and cycles in the pocket gopher. J. Mamm. 27, 335-358 (1946) Miller, M. A.: Size characteristics of the Sacramento Valley pocket gopher (Thomo. mys bottae marvus Merriam). J. Mamm. 83, 442-456 (1952) Morejohn, G.V., Howard, W. E.: :Holt in the pocket gopher, Thomomys bottae. J. Mamm. 37, 201-213 (1956) Population Dynamics of the Gopher Louse 303

Murray, M. D.: The distribution of the eggs of mammalian lice on their hosts. I. Description of the oviposition behavior. Aust. J. Zool. 5, 13-18 (1957a) Murray, M. D.: The distribution of the eggs of mammalian lice on their hosts. II. Analysis of the oviposition behavior of Damalinia ovis (L.). Aust. J. Zool. 5, 19-29 (1957b) Murray, M. D.: The distribution of the eggs of mammalian lice on their hosts. III. The distribution of the eggs of Damalinia ovis (L.) on the sheep. Aust. J. Zool. 5, 173-182 (1957c) Murray, M. D.: The distribution of the eggs of mammalian lice on their hosts. IV. The distribution of the eggs of Damalinia equi (Denny) and Haematopinus asini (L.) on the horse. Aust. J. Zool. 5, 183-187 (1957d) Murray, M. D. : The ecology of lice on sheep. II. The influence of temperature and humidity on the development and hatching on the eggs of Damalinia ovis (L.). Aust. J. Zoo1. 8, 357-362 (1960) Murray, M. D. : The ecology of the louse Polyplax serrata (]3urm.) on the mouse Mus musculus L. Aust. J. Zool. 9, 1-13 (1961) Murray, M. D.: The ecology of lice on sheep. V. Influence of heavy rain on popu- lations of Damalinia ovis (L.). Aust. J. Zool. 11, 173-182 (1963a) Murray, M. D. : Influence of temperature on the reproduction of Damalinia equi (Denny). Aust. J. Zool. 11, 183-189 (1963b) Murray, M. D.: The diversity of the ecology of mammalian lice. Proc. 12th int. Congr. Ent. (1964), p. 366-367. (1965) Murray, M. D., Gordon, G. : Ecology of lice on sheep. VII. Population dynamics of Damalinia ovis (Schrank). Aust. J. Zool. 17, 179-186 (1969) Nelson, ]3. C.: Successful rearing of Colpocephalum turbinatum (Phthiraptera). Nature (Load.) 282, 255 (1971) Price, R. D., Emerson, K. C.: A revision of the genus Geomydoecus (Malloph~ga: ) of the New World pocket gophers (Rodentia: Geomyidae). J. ivied. Ent. 8, 228-257 (1971) Rothschild, M., Clay, T.: Fleas, flukes and cuckoos: A study on bird parasites. London: Collins 1952 Rothschild, M., Ford, ]3. : Maturation and egg laying of the rabbit flea [SpilopsyUus cuniculi (Dale)] induced by the external application of hydrocortisone. Nature (Lond.) 208~ 210-211 (1964a) Rothschild, M., Ford, ]3. : Breeding of the rabbit flea [Spilopsyllus cuniculi (Dale)] controlled by the reproductive hormones of the host. Nature (Load.) 21)4, 103-104 (1964b) Rust, R. W. : Ectoparasites and nidicolous Acari of the pocket gopher, Thomomys bottae. Pan-Pacific Eat. 49, 59-60 (1973a) Rust, R. W. : The acarinium of the pocket gopher, Thomomys bottae. J. Med. Ent. 1O, 169-175 (1973b) Scott, M. T. : Observations on the bionomics of the sheep body louse (Damalinia ovis). Aust. J. Attic. Res. 3, 60-67 (1952) Snedecor, G. W. : Statistical methods. Ames, Iowa: Iowa State Univ. Press 1956 Spencer, G. J.: Ectoparasites of deer in ]3ritish Colombia. Proc. eat. Soc. ]3rit. Colombia 85, 15-19 (1939) Thaeler, C. S. : An analysis of three hybrid populations of pocket gophers (Genus Thomomys). Evolution 22,543-555 (1968) Thompson, G. ]3. : Mallophaga on sickly birds. Brit. Birds 29, 356 (1936) Vysotoskaya, S. O.: Seasonal changes in the infestation of the grey mole with lice (Russian). Parazit. Sbornik 12, 73-79 (1950) 20* 304 R.W. Rust

Waterhouse, D. F. : Studies on the digestion of wool by . IX. Some features of digestion in chewing lice (Mallophaga) from bird and mammalian hosts. Aust. J. biol. Sci. 6, 258-275 (1953) Waterston, J.: On the crop contents of certain Mallophaga. Zooh Soc. Lond. Proc. 1926, 1017-1020 (1926) Werneck, F. L.: Os Malofagos de Mamiferos. Parte II; (continuacao de Triehodeetidae) e Rhyncophthirina. Ed. do Inst. Osw. Cruz, Rio de Janeiro (1950) Winston, P. W., Bates, D. H. : Saturated solutions for the control of humidity in biological research. Ecology 41, 232-237 (1960)

Dr. R. W. Rust Department of Entomology and Applied Ecology University of Delaware Newark, Delaware 19711, USA