The changes in the dominance hierarchy over time of a complete field-captured colony of Cryptomys hottentotus hottentotus

The common mole-rat, Cryptomys h. hottentotus , is a social subterranean rodent occurring in colonies in which one female and one to three males are involved in reproduction and the remaining colony members are non-reproductive. Within each sex the reproductive animals are usually the largest and most dominant animals.
The dominance hierarchy amongst a field-captured colony was linear ( h = 0.95, calculated from Landau's linearity index) soon after capture. The non-reproductive females were ranked low in the dominance hierarchy; many were subordinate to non-reproductive males. The order of capture of mole-rats was not related to the position in the dominance hierarchy. The hierarchy became non-linear ( h = 0.56) after six months in captivity during which two juvenile animals became adult. The breakdown in the hierarchy may result from the lack of opportunity in captivity for animals to disperse and establish satellite colonies, or from colony members becoming co-dominant in the hierarchy as a result of a rise in rank by young animals.
Dominant mole-rats are involved in a greater proportion of interactive behaviours than subordinates. Popularity studies show that females tend to be more popular animals than males. The largest reproductive male was the least popular animal in the first study, whereas a beta male was the least popular animal in the second study period. The reproductive female was the most popular in both periods.     

The social structure and dominance hierarchy of the Mashona mole-rat, Cryptomys darlingi (Rodentia: Bathyergidae) from Zimbabwe

Cryptomys darlingi is a social subterranean rodent mole which inhabits the mesic regions of south-eastern and central Africa. Mashona mole-rats live in small colonies (5-9 animals) in which reproduction is normally restricted to the largest male and female in the colony. The non-reproductive members in a mature colony cannot be placed into clearly defined work-related groups based on body mass.
The dominance hierarchy of a young colony was found to be linear, with a value of 1.00 calculated from Landau's linearity index, while that of a mature colony of nine mole-rats was almost linear (0.77). Dominance was found to be related to gender in the mature colony, with males more dominant than females, and to age in the young colony. The reproductive mole-rats are the dominant animals within their respective genders. Dominance appears to correlate positively with body mass (rs = 0.77 in the mature colony and rs = 0.93 in the young colony).
Popularity studies show that smaller animals and females tend to be more popular than the larger massed individuals or males. In the mature colony which contained predominantly adult animals, the reproductive pair was among the least popular. While in the young colony, composed predominantly of sub-adult and juvenile animals, the reproductive pair was the most popular.
Social organization within Mashona mole-rat colonies is compared with other southern African Cryptomys species.     

The social structure and reproductive biology of the common mole-rat, Cryptomys h. hottentotus and remarks on the trends in reproduction and sociality in the family Bathyergidae

Cryptomys h. hottentotus occurs in mesic and semi-arid regions of South Africa. It lives in colonies ofup to 14 individuals, in which reproduction is limited to one ofthe largest males and the largest female in the colony.
Reproduction and details of colony size and the number of breeding animals in a colony are described for one complete and two incomplete field-captured colonies. One incomplete colony was kept in captivity for more than 3 years, whereas the complete colony was autopsied.
The social hierarchy of the common mole-rat is almost linear and has a value of 0.7 1 calculated from Landau's linearity index. Subordinate mole-rats weighed less than dominant mole-rats. The reproductive mole-rats are the dominant animals within the colony. The non-reproductive females rank lowest in the hierarchy.
Mating was not confined to a particular season, but occurred more frequently during the summer months. The alpha male initiated pre-copulatory behaviour. The gestation length is 55 to 66 days (n = 2). Two litters each of three pups were born to the colony during captivity. The pups were relatively altricial. They wandered out of the nest 5 days after birth, began to eat solids when 10 days old, and were fully weaned at 4 weeks. When 2 weeks old the pups began to spar with each other and with some of the adult colony members, but levels of aggression were never high. The pups did not disperse but were incorporated into the colony, this suggesting that C. h. hottentotus colonies are largely composed of related individuals.     

Differential growth patterns between successive litters of the eusocial Damaraland mole-rat, Cryptomys damarensis, from Namibia

Differential mean rates of growth were calculated for successive litters of mole-rats born to genetically unrelated pairs of C. damarensis. The Logistic equation provided the closest fit to the growth data.
Both intra- and inter-colonial variation in mean maximum growth rate between successive litters incorporated into the natal colonies were determined.
The mean asymptote (A), growth rate constant (K) and inflection time (I) were modelled for the first five litters of pups born to pairs of mole-rats. Inter-colonial analysis of recruitment to colonies revealed litters 1, 2, and 5 to grow faster than litters 3 and 4. Litters 1 and 2 had significantly higher asymptotes and inflection times than litters 3, 4, and 5 ( P < 0.001). All litters were assimilated into their natal colonies. There were no significant differences between males and females for the asymptote, growth rate constant, or inflection time ( P < 0.05).
Intra-colonial variation of litters 1 to 4 born to a single colony revealed comparable results to that obtained for inter-colonial comparisons. Multiple comparisons showed litters 1 and 2 to have a significantly higher absolute growth rate than litters 3 and 4. The lowest rate of growth being for litter 4.
The patterns of mean differential growth are discussed in the light of the social organization of the colony. It is postulated that growth rates in litters 1, 2, and 3 are more rigid since they constitute the main functional unit of colony organization. Litters 4 and 5 show a greater plasticity in growth; it is speculated that these colony members become incorporated into the various worker and defender groupings which are characteristic of C. damarensis colonies.     

Colony size and structure, activity patterns and foraging behaviour of a colony of the social mole-rat Cryptomys damarensis (Bathyergidae)

A complete colony of 20 Cryptomys damarensis was trapped in the Kalahari Gemsbok National Park, South Africa. The colony consisted of 15 males and five females and formed three distinct size classes, two of which were presumed to represent worker (smallest mole-rats) and soldier (largest males) castes. Body masses ranged from 86–197 g, with the dominant male being the largest mole-rat. The activity patterns of five animals in the colony were monitored by radiotelemetry. Activity patterns were not influenced by diel fluctuations in burrow temperature and photoperiod. It was suggested that the chance of hyperthemia influenced activity, such that in order to maximize the daily distance burrowed (foraging efficiency), the mole-rats had to engage in frequent (5.58 times. day^-1), but short ( c. 60 min) foraging bouts.     

Anovulation in non-reproductive female Damaraland mole-rats (Cryptomys damarensis)

Within colonies of Damaraland mole-rats (Cryptomys damarensis), anovulation in non-reproductive females is thought to play an important role in maintaining reproductive skew. Pituitary sensitivity and ovarian structure were examined in three groups of females that differed with respect to their social environment and breeding status to determine whether anovulation is due to inhibitory social cues or is merely the result of a lack of copulatory stimulation. The contribution of gonadal steroid negative feedback to neuroendocrine differences in the reproductive systems of the respective groups was also investigated. LH secretion after a 0.5 micrograms GnRH challenge in females that had been removed from the presence of the breeding individuals for at least 6 months (removed non-reproductive females) was significantly higher than in non-reproductive females in the colony, but significantly lower than in reproductive females. In both removed non-reproductive females and reproductive females, corpora lutea were observed in ovaries of seven of eight females, indicating that ovulation occurs spontaneously in subordinate females on removal from the breeding pair. Circulating progesterone concentrations in removed non-reproductive females were significantly higher than in non-reproductive females, indicating that circulating progesterone is not responsible for infertility in non-reproductive females. Indeed, after hystero-ovariectomy, reproductive females continued to show significantly greater GnRH-stimulated LH secretion than non-reproductive females. Thus, differential inhibition of gonadotrophin secretion in breeding and non-breeding females occurs independently of gonadal steroids. It is concluded that female Damaraland mole-rats are spontaneous ovulators and that anovulation results from inhibitory social cues within the colony, not a lack of copulatory stimulation. Since non-reproductive females are infertile, inhibition of the hypothalamo-pituitary-gonadal axis has the potential to play a causal role in maintaining reproductive skew in colonies of C. damarensis.     

Recrudescence of sexual activity in a reproductively quiescent colony of the Damaraland mole-rat (Cryptomys damarensis), by the introduction of an unfamiliar and genetically unrelated male—a case of incest avoidance in'queenless'colonies

Colonies of the eusocial Damaraland mole-rat, Cryptomys damarensis , are founded from a single reproductive pair of animals genetically unrelated by common descent. All non-reproductive members of the colony are progeny of this reproductive pair. In colonies where the reproductive female has been experimentally removed or has died a natural death, there is a strict incest avoidance and the colony remains reproductively quiescent. Reinstatement of sexual activity in a queenless colony may be brought about in the laboratory by the introduction of an unfamiliar and unrelated adult male.
In the queenless colony under study, there was a marked change in social structure with an increase in Landau's index of linearity from 0.8 to 0.9 on introduction of the new male. The unrelated male became a high ranking dominant reproductive male. The youngest, but most dominant n on-reproductive female became sexually active and subsequently became pregnant and hence acquired the position of reproductive female. The new reproductive female exhibited heightened progesterone (9nmols/mmol creatinine) and oestradiol (3000pmols/mmol creatinine) concentrations in the urine relative to the other non-reproductive females. These hormone concentrations were indicative of a reproductively active female.
Behavioural and hormonal data are presented to show that sexual activity can be re-instated in queenless colonies of laboratory maintained mole-rats by the introduction of unrelated male mole-rats.     

Colony structure and parentage in wild colonies of co-operatively breeding Damaraland mole-rats suggest incest avoidance alone may not maintain reproductive skew

Colonies of co-operatively breeding African mole-rats have traditionally been thought to be composed of a single breeding female, one or two breeding males, and their offspring. In the naked mole-rat (Heterocephalus glaber), the occurrence of facultative inbreeding means incest avoidance cannot prevent reproduction in subordinate group members, and physiological suppression of reproductive function by the breeding female occurs in both sexes. In contrast, previous studies of captive colonies of the Damaraland mole-rat (Cryptomys damarensis) suggest that breeding within a colony is restricted to a single breeding pair, simply because all other colony members are highly related (first- or second-order relatives) and this species is an obligate outbreeder. Using microsatellite markers, we investigated parentage and colony composition in 18 wild Damaraland mole-rat colonies to determine whether inbreeding avoidance alone can explain the high levels of reproductive skew in this species. Multiple and unidentified paternity was widespread within colonies and immigrants of both sexes were regularly identified. Unrelated, opposite-sex nonbreeders were found coexisting in two colonies. These results suggest that, in the wild, conditions exist where nonreproductive females can come into contact with unrelated males, even when they do not disperse from their natal colony. Inbreeding avoidance alone is therefore insufficient to maintain the high levels of reproductive skew identified in this species suggesting that the breeding female somehow suppresses the reproductive function in nonbreeding females.     

Reproductive suppression in eusocial Cryptomys darnavensis colonies: socially-induced infertility in females

Inhibition of reproduction occurs in colonies of the Damaraland mole-rat ( Cryptomys damarensis ), where one female and usually one male are reproductively active. They remain the sole reproductive animals thoughout their stay in the colony, which in the field may exceed five years.
An 18-month study on a captive colony of C. damarensis shows that non-reproductive females remain anovulatory and always have concentrations of urinary progesterone lower than that of the reproductive female 10.7 8.8 nmols/mmol creatinine (n = 85), although their progesterone concentrations are slightly elevated when the reproductive female is early in pregnancy. In contrast the reproductively active female has elevated concentrations of progesterone 63.3 70.1 nmols/mmol creatinine (n= 14).
Standard histological together with immunohistochemical examination of the ovarian structure shows that follicular development in the non-reproductive females is halted at varying stages prior to ovulation and that the unruptured follicles luteinize. These unruptured luteinized follicles stain positive for 3β hydroxysteroid dehydrogenase—the enzyme catalysing the synthesis of progesterone from pregnenolone. It is possible that the low levels of circulating progesterone produced by the luteinized follicles in the non-reproductive females in the colony are sufficient to feedback on the hypothalamo-pituitary axis and prevent the surge of LH necessary for ovulation.
Behavioural studies provide corroborative evidence for sexual suppression. Thus the non-reproductive females in the colony are never involved in courtship or copulatory behaviour. nor do they actively solicit males in the colony. The reproductive female, however, plays an active role in mate selection and is the initiator in courtship behaviour.
The suppression of ovulation in non-reproductivc females is maintained for as long as there is a reproductive female in the colony.     

Behaviour and social organization in a colony of the Damaraland mole-rat Cryptomys damarensis

This paper reports qualitative and quantitative behavioural studies of the social mole-rat C. damarensis . The behaviours were subdivided into four major categories: nest, maintenance, interactive and auto-behaviours. Social interactions and behavioural types were analysed by sex, age category and caste. The nest area is a focal point in the burrow system with individuals spending more than 70% of the day there. Many of the interactive behaviours occurred there and the mole-rats showed definite preferences with regard to which individuals they rested next to or avoided.
The Damaraland mole-rat colonies show some division of labour, into frequent and infrequent worker groups; these castes differed significantly in the amount of both digging and soil movement undertaken.     

Mating frequency, genetic structure, and sex ratio in the intermorphic female producing ant species Myrmecina nipponica

1. Myrmecina nipponica has two types of colonies: a queen colony type, in which the reproductive females are queens and new colonies are made by independent founding, and an intermorphic female colony type, in which reproductive females belong to a wingless intermediate morphology between queen and worker, and where colonies multiply through colonial budding. 2. The mating frequencies of reproductive females in both types indicate monoandry. The relatedness among nestmates in both types was almost 0.75, however relatedness between mother and daughter in intermorphic female colonies was slightly higher than that of queen colonies. 3. The sex ratio (corrected investment female ratio) was 0.70 at the population level, suggesting that the sex ratio is controlled by workers in this species, however the ratio differed greatly between the two types of colonies. Queen colonies (n = 37) had a female‐biased sex ratio of 0.77 while intermorphic female colonies (n = 33) had a ratio of 0.56. 4. Each reproductive intermorphic female was accompanied by an average of 2.9 workers (including virgin intermorphic females) in the colonial budding, and when the investment to those workers was added to the female investment, the sex ratio reached 0.81. 5. The frequency distribution of sex ratio was bimodal, with many colonies producing exclusively males or females, however mean estimated relatedness within colonies was almost 0.75. These data are inconsistent with the genetic variation hypothesis, which is one of the predominant hypotheses accounting for the between‐colony variation in sex ratio.     

Inbreeding avoidance and reproductive skew in a cooperative mammal

In animal social groups, socially subordinate individuals frequently show low reproductive success or completely fail to breed. This suppression of subordinate reproduction is currently typically attributed to control by dominant individuals. However, subordinates in cooperative groups often lack access to unrelated mates, and an alternative possibility is that their reproduction is limited by inbreeding avoidance. Using the eusocial Damaraland mole-rat Cryptomys damarensis, this paper provides the first experimental evidence, to our knowledge, for this explanation. Subordinate, non-breeding female mole-rats were given access to unrelated mates while remaining in the presence of dominant females, and many became reproductively active soon after unrelated males were introduced. Inbreeding avoidance and the availability of unrelated mates provides a plausible and untested explanation for variation in reproductive skew across animal societies.     

Dominance and queen succession in captive colonies of the eusocial naked mole-rat,Heterocephalus glaber.

Naked mole-rat colonies exhibit a high reproductive skew, breeding being typically restricted to one female (the ''queen'') and one to three males. Other colony members are reproductively suppressed, although this suppression can be reversed following the removal or death of the queen. We examined dominance and queen succession within captive colonies to investigate the relationship between urinary testosterone and cortisol, dominance rank and reproductive status; and to determine if behavioural and/or physiological parameters can be used as predictors of queen succession. Social structure was characterized by a linear dominance hierarchy before and after queen removal. Prior to queen removal, dominance rank was negatively correlated with body weight and urinary testosterone and cortisol titres in males and females. Queen removal results in social instability and aggression between high ranking individuals. Dominance rank appears to be a good predictor of reproductive status: queens are the highest ranking colony females and are succeeded by the next highest ranking females. The intense dominance-related aggression that accompanies reproductive succession in naked mole-rats provides empirical support for optimal skew theory.     

Influence of gonadal sex hormones on behavioral components of the reproductive hierarchy in naked mole-rats

Naked mole-rats (Heterocephalus glaber) are fossorial, eusocial rodents that live in colonies which typically include 60-80 individuals. Generally, only one of the females and 1-3 of the males in a colony are reproductives. The reproductives engage in mutual genital nuzzling behavior that is rarely exhibited by subordinates (non-reproductives). Thus, genital nuzzling may represent a mechanism of bonding and/or specific recognition between reproductive individuals. We investigated whether gonadal hormones are involved in the maintenance of genital nuzzling behavior and mating behaviors in isolated pairs of mole-rats and also in established breeding pairs of mole-rats within colonies. We also explored whether sex hormone deprivation would alter the strict partner preference for performance of nuzzling within colonies. Our results indicate (a) considerable variation between pairs in the frequency of nuzzling, (b) a reduction in the frequency of nuzzling following castration of the male and restoration of the 'baseline' frequency after replacement of testosterone in castrated males, (c) the failure of either castration or combined castration and ovariectomy to eliminate genital nuzzling in established pairs, and (d) the exhibition of nuzzling behavior by some of the subordinates in all three experimental colonies beginning several weeks after gonadectomy of both of the reproductives. No cases of lordosis behavior were seen during the approximately 109 h of behavioral observations. This is not surprising, since female mole-rats have an approximately 30-day ovulatory cycle, and lordosis only occurs during a peri-ovulatory period of a few hours. A total of 44 cases of mounting behavior were recorded; all these involved breeding males in colonies or males from isolated pairs, and all occurred when males were either gonad-intact or castrated with testosterone replacement. Thus, in contrast to nuzzling behavior, male sex behavior appeared to be eliminated during androgen deprivation.     

Combined Olfactory Contact with the Parent Colony and Direct Contact with Nonbreeding Animals Does Not Maintain Suppression of Ovulation in Female Naked Mole-Rats (Heterocephalus glaber)

The study investigated the role of odor cues from two naked mole-rat colonies, in conjunction with behavioral cues from nonbreeding colony members, in maintaining suppression of ovulation in subordinate female naked mole-rats isolated from the two parent colonies. Four high ranking nonbreeding female naked mole-rats were removed from their respective parent colonies and singly housed in separate burrow systems. For a 64-day period, the removed females were maintained in daily odor contact with their parent colony by daily rotating soiled bedding material between the parent colony and the burrow systems of removed females. In addition, subsets of nonbreeding animals from the respective parent colony were regularly moved into the burrow systems of removed females for 2-day periods during this 64-day period. Removed females were therefore in continual social contact with subsets of parent colony animals except for the breeding pair. All four removed females exhibited raised levels of urinary progesterone (< 2 ng/mg Cr) indicative of the onset of ovarian function within 3 days of being separated from the parent colony. Removed females exhibited a normal ovulatory cycle with levels of progesterone remaining elevated for 25–35 days (mean concentration of progesterone ± SEM; 16.2 ± 2 ng/mg Cr). Initiation of aggression and sexual behavior by removed females increased significantly when they were isolated from the parent colony. The results demonstrated that odor cues from the complete colony in conjunction with behavioral / tactile / vocal cues from the nonbreeding colony members were not the major cues maintaining reproductive suppression in nonbreeding female naked mole-rats. Instead, our results suggest that female reproductive suppression in naked mole-rats is caused by a dominance-related behavioral mechanism requiring direct contact with the breeding female.     

Stochastic effects in LMC models.

LMC (local mate competition) was first introduced by W. D. Hamilton to explain extraordinary female-biased sex ratios observed in a variety of insects and mites. In the original model, the population is subdivided into an infinite number of colonies founded by a fixed number of inseminated females producing the same very large number of offspring. The male offspring compete within the colonies to inseminate the female offspring and then these disperse at random to found new colonies. An unbeatable sex ratio strategy is found to be female-biased. In this paper, the effects of having colonies of random size and foundresses producing a random finite number of offspring are considered. The exact evolutionarily stable strategy (ESS) sex ratio is deduced and comparisons with previous approximate or numerical results are made. As the mean or the variance of brood size increases, the ESS sex ratio becomes more female-biased. An increase in the variance of colony size increases the ESS proportion of males when the mean brood size and colony size are both small, but decreases this proportion when the mean brood size or the mean colony size is large.     

Eusociality in African mole-rats: new insights from patterns of genetic relatedness in the Damaraland mole-rat (Cryptomys damarensis)

After the discovery of eusociality in the naked mole-rat, it was proposed that inbreeding and high colony relatedness in this species were the major underlying factors driving cooperative breeding in African molerats. By contrast, field and laboratory studies of the eusocial Damaraland mole-rat (Cryptomys damarensis) have raised the possibility that this species is an obligate outbreeder, although the build-up of inbreeding over several generations could still occur. Using microsatellite markers, we show that most breeding pairs in wild colonies of the Damaraland mole-rat are indeed unrelated (R = 0.02 +/- 0.04) and that mean colony relatedness (R = 0.46 +/- 0.01), determined across 15 colonies from three separate populations, is little more than half that previously identified in naked mole-rats. This finding demonstrates that normal familial levels of relatedness are sufficient for the occurrence of eusociality in mammals. Variation in the mean colony relatedness among populations provides support both for the central role played by ecological constraints in cooperative breeding and for the suggestion that inbreeding in naked mole-rats is a response to extreme constraints on dispersal. Approaches that determine the relative importance of an array of extrinsic factors in driving social evolution in African mole-rats are now required.     

Breeding status affects motoneuron number and muscle size in naked mole-rats: recruitment of perineal motoneurons?

Naked mole-rats live in large colonies and exhibit a strict reproductive hierarchy. Each colony has one breeding female and one to three breeding males; all other individuals are nonreproductive subordinates. Subordinates show a remarkable lack of sex differences in behavior and anatomy, but can become reproductive if removed from the colony. We recently reported that the striated perineal muscles and their innervating motoneurons, which are sexually dimorphic in all other mammals examined to date, are not dimorphic in subordinate naked mole-rats. Here we asked whether sexual differentiation of this neuromuscular system occurs when a subordinate becomes a breeder. The size and number of cells within Onuf's nucleus (homologue of the rat spinal nucleus of the bulbocavernosus) as well as perineal muscle volume were examined in subordinate and breeding naked mole-rats of both sexes. Sex differences in perineal motoneurons were not observed, regardless of social status. To our surprise, however, counts of motoneurons in Onuf's nucleus were increased approximately 30% in breeders of both sexes. This was accompanied by a reciprocal decrease in cells in Onuf's nucleus that were characterized by small soma size, and lacked a clear nucleus or nucleolus. Although not exhibiting typical motoneuron morphology, some of these small cells were positive for the motoneuron marker, SMI-32. The neuronal changes correlate with increased perineal muscle volumes in breeders. We propose that small, relatively undifferentiated cells are recruited to the pool of large Onuf's nucleus motoneurons when subordinate naked mole-rats become breeders.     

Social cues elicit sexual behavior in subordinate Damaraland mole-rats independent of gonadal status

Damaraland mole-rats (Fukomys damarensis) are among a small number of eusocial mammals. Eusociality is a social system where only a few individuals within a colony engage in direct reproduction, while remaining subordinate members are non-breeders and support reproductive efforts of breeding individuals. Inbreeding avoidance precludes mating between subordinate siblings and between offspring and parents. Interestingly, non-breeders readily attempt to mate with unrelated opposite-sex individuals. This is unusual since the non-breeding females do not attain puberty while in their natal colony. Based on this finding, the present study investigated the role of the gonads in the regulation of mating behaviors in this species and identified the mechanism of inbreeding avoidance. Gonadal-intact and gonadectomized non-breeders from different colonies were removed from their colonies and tested for the expression of sexual behavior. Results indicated that gonadal status had only minor effects on the expression of sexual behavior in either males or females. In a second experiment, sexual behaviors were absent between opposite-sex siblings so long as they had frequent contact with each other; however, following 5 weeks of separation, sexual behavior between these siblings was robustly expressed. Thus, Damaraland mole-rats avoid establishing mating relationships with familiar individuals but will readily mate with unfamiliar individuals of the opposite sex, with genetic relatedness apparently playing little role. The initiation of sexual behavior in Damaraland mole-rats does not require the presence of the gonads, but does require that the members of the pair have not been in contact with one another for at least several weeks.     

COLONY SEX RATIOS,CONFLICT BETWEEN QUEENS AND WORKERS,AND APPARENT QUEEN CONTROL IN THE ANT PHEIDOLE DESERTORUM

Sex-ratio conflict between queens and workers was explored in a study of colony sex ratios, relatedness, and population investment in the ant Pheidole desertorum. Colony reproductive broods consist of only females, only males, or have a sex ratio that is extremely male biased. Colonies producing females (female specialists) and colonies producing males (male specialists) occur at near equal frequency in the population. Most colonies apparently specialize in producing one reproductive sex throughout their life. Allozyme analyses show that relatedness does not differ within male-specialist and female-specialist colonies and they do not appear to differ in available resources. In the population, workers are nearly three times more closely related to females than males; however, the investment sex ratio is near equal (1.01, female/male), which is consistent with queen control. Selection should be strong on workers to increase investment in reproductive females, so why do workers in male-specialist colonies produce only (or nearly only) males? One hypothesis is that queens in male-specialist colonies prevent the occurrence of reproductive females, perhaps by producing worker-biased female eggs. An earlier simulation study of genetic evolution of sex ratios in social Hymenoptera (Pamilo 1982b) predicts that such mechanisms can result in the evolution of bimodal colony sex ratios and queen control. Results on P. desertorum are generally consistent with that study; however, information is not currently available to test some of the model's predictions and assumptions.