Genetica 88: 29-36, 1993. 9 1993 Kluwer Academic Publishers. Printed in the Netherlands.

Inbreeding depression and male-mating behavior in Drosophila

melanogaster R S. Miller, J. Glasner "k & R W. Hedrick *~ Department of Zoology, Arizona State University, Tempe, AZ 85287-1501, USA * Present address: Genetics Program, University of New Hampshire, Durham, NH 03824, USA ** Author for correspondence Received 5 October 1992

Accepted 3 December 1992

Key words: conservation genetics, Drosophila melanogaster, fitness components, inbreeding depression, male-mating behavior Abstract

There have been relatively few studies designed to investigate the effects of inbreeding on behavioral traits. To study this phenomenon, five experimental lines of Drosophila melanogaster made isogenic for chromosome 2 were evaluated for their male-mating ability and, subsequently, male courtship behavior. All lines showed significant reductions in overall mating ability, and males from all of these lines displayed impaired mating behavior, with two lines displaying particularly aberrant courtship patterns. Line 16 displayed an inability to successfully initiate copulation following successful courtship, while line 17 displayed significant reduction in locomotor activity, resulting in virtually no successful courtship or copulatory activity. The implications of these findings for competitive mating ability in wild Drosophila populations are presented. Further, the importance of mating success as a fitness component in the management of potentially highly inbred populations of endangered species is discussed.

Introduction

Inbreeding depression and its consequences for reproductive fitness in animals have been the topics of an extensive and diverse literature, with study subjects including laboratory populations of Drosophila (e.g., Latter & Robertson, 1962; Dobzhansky et al., 1963) as well as natural (Ralls et al., 1986) and captive (Ralls& Ballou, 1983) populations of vertebrates. It is quite likely that behavioral traits are influenced by inbreeding in much the same way as other reproductive traits, although this view is somewhat speculatory because the same level of research effort has not yet been directed toward investigating the effects of inbreeding on behavior. Work that does exist in this area, such as that of Lynch (1977) on the nesting behavior of inbred Mus musculus, suggests that some behavioral traits can indeed be impaired by inbreeding.

Nevertheless, it is clear that much more work needs to be done to investigate this phenomenon more systematically. The importance of male-mating ability as a component of reproductive fitness in Drosophila (e.g., Parsons, 1974), combined with the recognition of this ability as a complex series of behaviors involving both male and female elements (Bastock & Manning, 1955), makes a study of the behavioral basis of reduced mating ability of inbred Drosophila males particularly relevant to our understanding of this phenomenon. In a general study of inbreeding depression in relationship to conservation genetics (Miller & Hedrick, 1993), we conducted a detailed investigation of male-mating success, female fecundity, and egg-to-adult survival in a group of isogenic second chromosome lines of D. melanogaster and found a dramatic reduction in male-mating success in all inbred lines studied.

30 Consequently, we conducted further experiments to examine the behavioral basis of the lowered mating ability in these lines, the results of which we report here.

Materials and methods

Lines and culture conditions To generate isogenic chromosome 2 lines, we used Drosophila melanogaster collected in June, 1987 at the University of California, Davis (Clark, 1989). This population (hereafter referred to as UCD) has been maintained in the laboratory in twelve halfpint milk bottles through a discrete generation transfer protocol, where each generation (14 days) adult flies from each of the bottles were mixed and distributed into twelve new bottles containing fresh medium. All stock and experimental populations were reared at 25 ~ with a 12-h photoperiod using Carolina 4-24 Instant Medium. The experiments described here were carried out between June and October, 1991.

Male-mating success The method used for evaluating male-mating success is based on that of Sharp (1974). For each of the five experimental lines, ten replicate 8-dram vials were set up containing five virgin chromosome-2 homozygote males, five virgin chromosome-2 heterozygote males, and five virgin UCD stock females. The males were allowed to compete for females for two hours, after which the females were removed and placed in individual vials. This time period has been shown to minimize the occurrence of double-insemination events while maximizing the proportion of successful matings (Bundgaard & Christiansen, 1972). The females were removed on day 3, and the vials were scored on day 14 for the presence of chromosome-2 heterozygote progeny, thereby indicating the paternal genotype responsible for the mating. This procedure was repeated for a total of 100 experimental vials per line. The relative male-mating ability (1 m) for each line was calculated from the estimator

I - m = _ _

Chromosome extractionprocedure The chromosome 2 extraction procedure utilized in this experiment was very similar to that employed by Sved & Ayala (1970). The laboratory marker stock used was Bl/SM5, with the Bl (bristle) chromosome balanced by the SM5 complex containing the dominant phenotypic marker Cy (curly wing) (Lindsley & Zimm, 1992). For 18 different lines, chromosome homozygote (%/%) and heterozygote (+i/SM5) genotypes were combined and allowed to compete according to the protocol outlined in Miller & Hedrick (1993). Six of these lines exhibited low chromosome homozygote fitness (relative to the corresponding chromosome heterozygotes) as determined from this chromosome equilibration experiment: we designate these lines herein as 4, 7, 8, 16, 17, and 21. Because of their inherent interest as low-fitness inbred lines, five of these six lines were then tested further as described below (line 7 homozygotes had such low viability that a sufficient number of adults could not be collected for mating estimation).

(1)

~M+I where N++ and N+Mindicate the number of females fertilized by ++ (homozygote) and +M (heterozygote) males, respectively (after Haldane, 1956).

Mating behavior Virgin male and female flies were collected from uncrowded bottles over two days using light CO 2 anesthesia. They consisted of males isogenic for the five different second chromosomes, males heterozygous for the isogenic chromosome and the wild-type balancer chromosome in the equilibrium lines, and females from the UCD stock. The flies were kept separately by sex and genotype in 8-dram vials in uncrowded conditions (5-7 per vial) to avoid stimulation for 4 to 6 days. Mating behavior was then observed in 8-dram vials that had been cut in half to create a more confined mating chamber. First, single males either isogenic or heterozygous for a given line were aspirated without anesthesia into individual chambers. Single UCD females were then aspirated into these chambers, and a foam plug was inserted to a con-

31 stant depth to close the chamber. Mating behavior was observed for a thirty-minnte period during which both courtship latency and duration periods were recorded. Initiation of courtship was judged as onset of one of the following behaviors: male wing vibration, male following female, or male orienting towards the female (for a description of these behaviors, see Bastock & Manning, 1955). Copulation initiation was defined as the first attempt on the part of the male to mount the female. Males from lines 16 and 17 displayed particularly aberrant mating behavior (see Results); to further investigate mating behavior in line 16, we repeated the mating behavior experiment with homozygous line 16 females to test the effect of female genotype on malemating performance.

Locomotor activity Locomotor activity (Ewing, 1963) was measured on single flies from line 17 introduced into a 30-cm long loop of clear Nalgene tubing (series 63013062) connected to itself by a larger diameter short section of clear tubing. This loop was then placed over a circular grid of 48 lines regularly spaced over the 30 cm length of tubing. Flies were introduced into the loop without anesthesia, allowed to acclimate for 100 seconds, and then tested for the number of lines crossed during the following 100 seconds (Burnet & Connolly, 1974). Measurements on the same set of flies were made on days 2 and 3 after eclosion.

Results

Male-mating success Table 1 presents a summary of the fitness component results from Miller & Hedrick (1993). Only two of the six experimental lines showed significantly lower homozygote viability relative to +[ SM5 heterozygotes (lines 7 and 16); similarly, only one of the five lines tested showed reduced female fecundity relative to +[SM5 heterozygotes (line 17). In contrast, all five lines tested for male-mating success showed a greatly reduced ability for chromosome 2 homozygotes to successfully sire offspring. The average viabilities and female fecundities in these lines were close to unity (0.894 and

Table 1. Fitness estimates for chromosome homozygotes in six experimental lines (from Miller & Hedrick, 1993). Line

Viability

Male-mating ability

Female fecundity

4 7 8 16 17 21 Average

1.260 0.198 1.063 0.800 0.999 1.041 0.894

0.607 -0.297 0.212 0.023 0.237 0.275

1.116 -1.202 1.325 0.554 0.878 1.015

(0.053) *" (0.012)* (0.043) (0.084)* (0.051) (0.040) (0.371)

(0.133)* (0.103)* (0.062)* (0.017)* (0.071)* (0.212)*

(0.202) (0.202) (0.225)* (0.110)* (0.179) (0.305)

* P < 0.05 a Values indicate the estimate relative to +]SM5 heterozygotes, with standard deviations given in parentheses.

1.015, respectively) while the average male-mating ability was much lower than unity (0.275). In line 17, the worst mating line, only 2 out of 87 stock females were inseminated by homozygous males, and in line 16, the next worst mating line, 14 out of 79 females were inseminated by homozygous males. Even in line 4, despite being the most successful mating line, homozygous males successfully inseminated only 34 out of 89 stock females.

Mating behavior Results of the mating behavior observations are given in Table 2 for homozygous and heterozygous males. All inbred lines show some trend towards a decrement in male courtship latency and/or courtship duration. With the exception of line 16 using UCD females and line 4, all inbred lines showed an increase in both courtship latency and duration, with increased latency being statistically significant in lines 4 and 17 and duration showing significance in line 16 when using isogenic females. Males from two inbred lines showed particularly noteworthy behavioral patterns that further impaired their mating ability dramatically. The homozygous males from line 16 initiated courtship rapidly and their mating behavior did not appear different from that of other flies observed. However, the males experienced extreme difficulty initiating copulation (only 35% and 30% successfully initiated copulation in trials with heterozygous and homozygous females, respectively), often due to the female kicking off the male before successful mounting could take place. Because of this anoma-

32 Table 2. The time until courtship and copulation initiation. Sample size is 40 males unless otherwise indicated. Line

Male genotype

Male courtship latency a

Proportion of males courting

Courtship duration a

Proportion of courting males copulating

4

+/+

6:59 (l:09)t *** 1:31 (0:09)

0.975 (0.025) 0.975 (0.025)

7:17 (1:04) 8:07 (1:11)

0.82 (0,062) 0.795 (0.065)

3:06 (0:32) 2:30 (0:25)

0.975 (0.025) 1.00

9:33 (1:16) 7:03 (1:00)

0.795 (0.065) 0,775 (0.066)

2:18 (0:21) 2:40 (0:44)

1.00 0.975 (0.025)

6:29 (1:05) 5:29 (0.54)

0.35 (0.075) *** 0.846 (0.058)

3:57 (1:04) 2:11 (0:46)

1.00 1.00

10:38 (1:38) ** 4:59 (0:46)

0.30 (0.084) *** 0.73 (0.081)

13:12 (1:42) *** 2:33 (0:16)

0.70 (0.072) *** 1.00

0:135 5:47 (0:52)

0.036 (0.035)*** 0,80 (0.063)

2:27 (0:24) 2:14 (0:16)

1.00 1.00

9:00 (1:15) 7:08 (0.57)

0.90 (0.047) 0.775 (0.066)

+/SM5 8

+/+

+/SM5 16

+/+

+/SM5 16b

+/+

+/SM5 17

+/+

+/SM5 21

+/+

+/SM5

a Times are in minutes and seconds, with standard deviations in parentheses. b Homozygous wild-type females from line 16 were used. Also, the number of males used in this trial was 30. t All statistical comparisons are between +]+ and +[SM5within each experimental line. $ Mean based on only one observation. ** P < 0.01 *** P < 0.001

lous behavior observed in line 16, we looked at the mating behavior of line 16 males when paired with isogenic line 16 virgin females to test for the effect of female genotype on male-mating behavior. The aforementioned mating difficulty was apparently enhanced when homozygous males were presented with their homozygous female counterparts, as evidenced by a highly significant increase in courtship duration relative to the heterozygous males (Table 2). Even when copulation was successfully initiated, the duration of copulation was always less than 10 rain, considerably less than that reported to be characteristic of D. melanogaster (17-20 rain: Spieth, 1952) and less than that generally required for the initiation of sperm transfer (10 rain: Gilbert et al., 1981). Additionally, line 17 homozygous males showed extreme difficulty with both courtship and copulation. The males appeared clumsy and awkward, often unable to climb the sides of the mating chambers. A limited number of males exhibited some courtship behaviors, i.e., orientation and wing vibration, but were not able to actively follow fe-

males and had to wait for the females to move to the bottom of the vial, resulting in a low proportion of courting males (70%) as well as a highly delayed courtship initiation with a mean courtship latency of over 13 rain. Copulation was extremely difficult for males of this line, with only 1 in 28 courting males successfully copulating.

Locomotor activity Because of the poor ability of line 17 flies to move about, we attempted to quantify their locomotor activity. Table 3 gives the locomotor behavior of males and females from line 17 (both chromosome homozygotes and heterozygotes). Data from the two days were found to be homogeneous using a heterogeneity G-test (Sokal & Rohlf, 1981) and were subsequently pooled for analysis. There was a highly significant difference in activity between homozygotes and heterozygotes, with homozygotes having only 0.488 and 0.609 the activity of heterozygotes in males and females, respectively. In other words, line 17 homozygous flies in this

33 Table 3. The locomotor activity of chromosome 2 homozygotes and heterozygotes from line 17; N is the sample size.

Males

+/+

+/SM5 Females

+/+

+/SM5

Age in days 2 (N = 15)

3 (N = 35)

Combined

21.00 (8.97) a ** 45,23 (22.13)

12.57 (8.35) *** 25.63 (14.09)

15.10 (9.30) *** 30.94 (18.61)

15.27 (5.70) *** 30.40 (13.20)

17.69 (10.19) ** 26.74 (13.10)

16,96 (9.10) *** 27,84 (13.10)

a Values indicate the average number of gridlines crossed in the test chamber, with standard deviations in parentheses. ** P < 0.01 *** P < 0.001

simple movement test had roughly only half the activity of heterozygotes.

Discusssion Our research has demonstrated that inbreeding in laboratory populations of D. melanogaster can dramatically impair male-mating ability, and that specific behavioral components which contribute to this impairment can be identified. The data presented in this paper strongly suggest (hat this reduced mating ability is due to an increase in either male courtship latency or courtship duration, or both, in all inbred lines studied. There appears to be considerable genetic variation present for a number of specific male courtship behaviors in Drosophila in the form of either additive or directional dominance variance, or both (Collins & Hewitt, 1984 and refs. therein); furthermore, selection appears to favor males exhibiting fast mating speed (Collins & Hewitt, 1984). Consequently, it follows that inbreeding depression following the construction of isogenic chromosomes is responsible for the aberrant courtship behavior and subsequently reduced mating success observed in our inbred lines. The results from our study cannot provide definitive evidence as to the mechanistic basis of this fitness reduction, i.e., whether specific visual (Connolly et aL, 1969), auditory (Bennet-Clark & Ewing, 1969), or olfactory (Averhoff & Richardson, 1974) responses have been impaired by inbreeding. It is likely that the specific mechanistic element responsible is variable across inbred lines (Eastwood & Burnet, 1979), or that all sensory inputs may be affected to varying degrees (Markow, 1981). In the mating success assay presented in Table 1 and the mating behavior assay presented in Table 2,

line 17 inbred males displayed the worst performance with line 16 inbred males exhibiting a similar degree of disability. Lines 4, 8, and 21, displaying equally poor mating success (Table 1), did not show the same level of inbreeding depression for mating behavior, though a reduction in overall mating speed was clearly apparent (Table 2). These results suggest that a mating scenario involving competition among different male genotypes for access to a limited number of females, as simulated in the male-mating success assay, can magnify differences in male-mating ability that are not as distinct in single-pair mating behavior assays. Such highly competitive situations are common in natural populations of Drosophila (Gromko & Markow, 1992;Markow & Sawka, 1992); consequently, we feel that even a small decrease in mating speed among inbred males in the laboratory would translate into a significant disadvantage to inbred males competing for mates in the wild. When inbred males from line 16 were presented with inbred females from the same line, there was a large increase in courtship latency and a highly significant increase in courtship duration compared to these males paired with outbred stock females. It appears from these data that line 16 inbred females have a higher threshold for male acceptance, or equivalently, a higher level of courtship summation (Manning, 1967). Such alteration of female receptivity has been described in D. melanogaster (Connolly et aL, 1974), and can serve to further reduce the viabiUty of small populations experiencing inbreeding depression for reproductive and behavioral traits. I t has been argued (Robertson, 1982; Cobb et al., 1987) that the common approach of initiating observation of mating behavior immediately after in-

34 troducing a male-female pair into a simple mating chamber may lead to spurious results. This is due to the female being in a highly agitated state following aspiration into the chamber, a state in which she is unresponsive to any male courtship cues. In this situation, male courtship can become stimulatory only after the female has had time to adapt to the new environment. Short male latencies will therefore lead to prolonged and, for the most part, ineffective courtship while long male latencies will result in short and highly stimulatory courtships. Consequently, if male latency and courtship duration are negatively correlated, any conclusions regarding mating behavior in the population under study may only be an artifact of the experimental design. While techniques such as introducing males and females into opposite sides of a mating chamber containing a temporary divider for pre-mating acclimitization (Cobb et al., 1987) may help to eliminate any experimental shortcomings, we think our data represent actual behavioral differences between inbred and outbred males. Females from the same outbred base population were used as a standard throughout the mating assays, and experimental protocol was kept constant in all mating tests; if any degree of female agitation did exist in our mating chambers, it would very likely be the same across all such chambers and any observed differences in mating parameters would be a consequence of male genotype. Nevertheless, this hypothesis was investigated further by calculating Pearson product-moment correlations between log-transformed male latency and courtship duration data for inbred and outbred groups for each line as well as the grouped inbred and outbred data (Table 4). Because only one male from line 17 successfully copulated, this line could not be included in the analysis. Most correlations for the outbred lines are near zero and all are highly non-significant, but each inbred line has a negative correlation and a lower P-value with that for line 8 being significant. More striking, however, is the comparison of correlations for the pooled data, the inbred group showing a highly significant negative correlation while the outbred group shows no such relationship. Because this strong effect is present only in inbred males, we interpret these results not as support for Robertson's hypothesis of laboratory-induced alteration of female summation proc-

Table 4. Pearson product-moment correlation analysis between male courtship latency and courtship duration. Line 4

+[+

+]SM5 8

+/+

+/SM5 16

+/+

+]SM5 16 a

+/+

+[SM5 21

+/+

+]SM5 Total

+]+

+]SM5

r

P

-0.292 -0.132

0.111 0.478

-0.367 0.029

0.042 e 0.877

-0.433 -0.124

0.122 0.506

-0.329 0.054

0.387 0.811

-0.173 0.163

0.313 0.382

-0.297 -0.055

0.001 ~'~ 0.462

a Using line 16 isogenic females.

esses but as providing additional insight into the mechanism of reduced male-mating speed in these inbred lines. Perhaps, in the parlance of Robertson (1982), the inbred males are somehow acting to further increase female agitation or, perhaps more appropriately, decrease female receptivity beyond that induced by the novel environment or by early, aggressive courtship (short latency). From this study, it appears that inbreeding may have dramatic effects on male-mating success in wild populations of Drosophila. The cause of this reduction in our laboratory populations was directly demonstrated as resulting from abnormal mating behavior. Observations such as this have direct relevance to conservation genetics, a discipline involving the application of population genetics principles to endangered species conservations (Schonewald-Cox et al., 1983; Soul6, 1987; Hedrick & Miller, 1992). The primary goal of a genetic and demographic management program for an endangered species is to determine the minimum population size necessary to reasonably ensure long-term population viability. In the case of many endangered organisms, populations can be small and isolated with potentially high levels of inbreeding, and mate availability may be restricted both spatially and temporally. As a result, competition for mates can be severe. A reduced mating ability in

35 inbred males may further decrease the size of the breeding population (also known as the effective population size: Wright, 1931) and may therefore be a critical factor influencing population viability (Lande & Barrowclough, 1987). Consequently, in our opinion, because male-mating behavior appears to be a fitness component that is very sensitive to inbreeding, careful monitoring of small populations of endangered species for reduced male-mating ability should be an integral part of a conservation program, and should accompany more traditional studies of the effects of inbreeding on viability and fecundity (e.g., Ralls et al., 1988). Perhaps more importantly, an increased awareness of the effects of inbreeding on behavioral traits may be essential in our efforts to increase the survival potential of small populations.

Acknowledgements We gratefully acknowledge Barb Webb, Jennifer Alexander, and Sara Good for their technical assistance. Ted Markow gave us essential advice throughout the research. We appreciate the comments of T. Markow and an anonymous reviewer on the manuscript. This research was supported by NSF BSR-8923007.

References Ashburner, M. & J. N. Thompson, 1977. The laboratory culture of Drosophila, pp. 1-109 in The Genetics and Biology of Drosophila Vol 3c, edited by M. Ashburner and T. R. E Wright. Academic Press, New York, NY. Averhoff, W. W. & R. H. Richardson, 1974. Pheromonal control of mating patterns in Drosophila melanogaster. Behav. Genet. 4: 207-225. Bastock, M. & A. Manning, 1955. The courtship of Drosophila melanogaster. Behaviour 8: 85-111. Bennet-Clark, H. C. & A. W. Ewing, 1969. Pulse interval as a critical parameter in the courtship song of Drosophila melanogaster. Anim. Behav. 17: 755-759. Bundgaard, J. & E B. Christiansen, 1972. Dynamics of polymorphisms: I. Selection components in an experimental population of Drosophila melanogaster. Genetics 71: 439-460. Bumet, B. & K. Connolly, 1974. Activity and sexual behavior in Drosophila melanogaster, pp. 201-257 in The Genetics of Behavior, edited by J. A. E Van Abeelen. North Holland Publ. Co., Amsterdam. Clark, A. G., 1989. Causes and consequences of variation in lipid and carbohydrate storage in Drosophila. Genetics 123: 131-144.

Cobb, M., K. Connolly & B. Burnet, 1987. The relationship between locomotor activity and courtship in the melanogaster species sub-group of Drosophila. Anim. Behav. 35: 705713. Collins, M. F. & J. K. Hewitt, 1984. The genetic architecture of the male courtship sequence in Drosophila melanogaster. Heredity 53: 321-337. Connolly, K., B. Burnet & D. Sewell, 1969. Selective mating and eye pigmentation: An analysis of the visual component in the courtship behavior of Drosophila melanogaster. Evolution 23: 548-559. Connolly, K., B. Burnet, M. Kearney & L. Eastwood, 1974. Mating speed and courtship behavior of inbred strains of Drosophila melanogaster. Behaviour 48: 61-74. Dobzhansky, Th., B. Spassky & T. Tidwell, 1963. Inbreeding and the mutational and balanced genetic loads in natural populations of Drosophila pseudoobscura. Genetics 48: 361-373. Eastwood, L. & B. Bumet, 1979. Variation in the importance of acoustic stimuli in Drosophila melanogaster courtship. Experientia 35:1159-1160. Ewing, A. W., 1963. Attempts to select for spontaneous activity in Drosophila melanogaster. Anim. Behav. 11: 269-378. Gilbert, D. G., R. C. Richmond & K. B. Sheehan, 1981. Studies of esterase 6 in Drosophila melanogaster. VII. The timing of remating in females inseminated by males having active or null alleles. Behav. Genet. 11: 195-208. Gromko, M. H. & T. A. Markow, 1992. Courtship and remating in field populations of Drosophila. Anita. Behav., in press. Haldane, J. B. S., 1956. The estimation of viabilities. J. Genetics 54: 294-296. Hedrick, E W. & E S. Miller, 1992. Conservation genetics: techniques and fundamentals. Ecol. Appl. 2: 30-46. Lande, R. & G. E Barrowclough, 1987. Effective population size, genetic variation, and their use in population management, pp. 87-123 in Viable Populations for Conservation, edited by M. E. Soul6. Cambridge University Press, Cambridge, England. Latter, B. D. H. & A. Robertson, 1962. The effects of inbreeding and artificial selection on reproductive fitness. Genet. Res. 3: 110-139. Lindsley, D. L. & G. G. Zimm, 1992. The Genome of Drosophila melanogaster. Academic Press, New York, NY. Lynch, C. B., 1977. Inbreeding effects upon animals derived from wild populations of Mus musculus. Evolution 31: 525537. Manning, A., 1967. The control of sexual receptivity in female Drosophila melanogaster. Anita. Behav. 15: 239-250. Markow, T. A., 1981. Genetic and sensory aspects of mating success of phototactic strains of Drosophila meIanogaster. Behav. Genet. 11: 273-279. Markow, T. A. & S. Sawka, 1992. Dynamics of mating success in experimental groups of Drosophila melanogaster (Diptera: Drosophilidae). J. Insect Behav. 5: 375-383. Miller, E S. & R W. Hedrick, 1993. Inbreeding and fitness in captive populations: Lessons from Drosophila. Zoo Biol., submitted. Parsons, R A., 1974. Male mating speed as a component of fitness in Drosophila. Behav. Genet. 4: 395-404. Ralls, K. & J. D. Ballou, 1983. Extinction: lessons from zoos,

36 pp. 164-184 in Genetics and Conservation, edited by C. M. Schonewald-Cox, S. M. Chambers, B. MacBryde & L. Thomas. Benjamin Cumminggs, Menlo Park, CA. Ralls, K., R H. Harvey & A. M. Lyles, 1986, Inbreeding in natural populations of brids and mammals, pp. 35-56 in Conservation Biology: The Science of Scarcity and Diversity, edited by M. E. Soul~. Sinauer Associates, Sunderland, MA. Ralls, K., J. D. Ballou & A. R. Templeton, 1988. Estimates of lethal equivalents and the cost of inbreeding in mammals. Cons. Biol. 2: !85-193. Robertson, H. M., 1982. Female courtship summation in Drosophila melanogaster. Anim. Behav. 30:1105-1117. Schonewald-Cox, C. M., S. M. Chambers, B. MacBryde & L. Thomas, editors, 1983. Genetics and Conservation. Benjamin Cummings, Menlo Park, CA.

Sharp, R M., 1984. The effect of inbreeding on competitive male-mating ability in Drosophila melanogaster. Genetics 106: 601-612. Sokal, R. R. & E J. Rohlf, 1981. Biometry, 2nd. ed., Freeman, New York, NY. Soule, M. E., editor, 1987. Viable Populations for Conservation. Cambridge University Press, Cambridge, England. Spieth, H. T., 1952. Mating behavior within the genus Drosophila (Diptera), Bull. Amer. Mus. Nat. Hist. 99: 398-474. Sved, J. A. & E J. Ayala, 1970. A population cage test for heterosis in Drosophila pseudoobscura. Genetics 66: 97113. Wright, S., 1931. Evolution in Mendelian populations. Genetics 16: 97-159.