Overall, insects that exhibit parental care may be expected to show low early mortality when compared to non-parental species (Ito, 1980). This pattern, however, does not seem to apply to the few species of subsocial cockroaches for which survivorship data are available. In Macropanesthia, mortality is about 35-40% by the time the nymphs disperse from the nest at the fifth to sixth instar (Rugg and Rose, 1991; Matsumoto, 1992). Both Sal-
ganea esakii and Sal. taiwanensis incubate an average of 15 eggs in the brood sac, but average only six nymphs (third instar) in young, field-collected families (T. Matsumoto and Y. Obata, pers. comm. to CAN). Family size of Cryptocercus punctulatus declines by about half during the initial stages; a mean of 73 eggs is laid, but families average only 36 nymphs prior to their first winter (Nalepa, 1988b, 1990). These data suggest that neonates may be subject to mortality factors such as disease or starvation despite the attendance of adults.
An alternative explanation for high neonate mortality in these species is that it represents an evolved strategy for adjusting parental investment after hatch (Nalepa and Bell, 1997). Unlike other oviparous cockroaches, in Cryp-tocercus the hatching of nymphs from the egg case is not simultaneous, but extended in time. Hatching asyn-chrony results in variation in competitive ability within a brood, a condition particularly conducive to the consumption of young offspring by older siblings (Polis, 1984). Nymphs of C. punctulatus 12 days old have been observed feeding on dead siblings, and attacks by nymphs on moribund siblings have also been noted. Age differentials within broods may allow older nymphs to monopolize available food, leading to the selective mortality of younger, weaker, or genetically inferior siblings. Ne-crophagy or cannibalism by adults or older juveniles may then recycle the somatic nitrogen of the lower-quality offspring back into the family (Nalepa and Bell, 1997). The production of expendable offspring to be eaten by siblings can be viewed as an alternative to producing fewer eggs, each containing more nutrients (Eickwort, 1981; Polis, 1981; Elgar and Crespi, 1992).
The behavioral mechanisms balancing supply (provisioning by parents) and demand (begging or solicitation by nymphs) are unstudied in subsocial cockroaches. In Cryptocercus, adults appear to offer hindgut fluids periodically, with juveniles competing for access to them. It is probable that, like piglets, nymphs that struggle the hardest to reach parental fluids will gain the biggest share. Competition for food may be a proximate mechanism for adjusting brood size and eliminating runts in other subsocial cockroaches as well. Perisphaerus sp. females possess just four intercoxal openings, but nine nymphs were associated with one of the museum specimens studied by Roth (1981b). Sibling rivalry for maternally produced food is also observable in G. portentosa and Sal. taiwan-ensis (Fig. 8.3). In Cryptocercus, there is some evidence of parent-offspring conflict in the amount of trophallactic food that an individual nymph receives. Adults can deny access to hindgut fluids by closing the terminal abdominal segments, like a clamshell. In the process of doing so the head of a feeding nymph is sometimes trapped, and the adult attempts to either fling it off with abdominal
wagging, or to scrape it off by dragging it along the side of the gallery (CAN, unpubl. obs.).
It should be noted that in Cryptocercus there are cooperative as well as competitive behaviors among nymphs when procuring food. Wood is not only nutritionally poor and difficult to digest, but physically unyielding. Like young nymphs in aggregations, early developmental stages of Cryptocercus may need the presence of con-specifics to help acquire meals when they begin including wood in their diet. Nymphs have been observed feeding cooperatively on wood slivers pulled free by both siblings (Fig. 8.7) and adults (Nalepa, 1994; Park and Choe, 2003a).
Was this article helpful?