Flight loss in insects is most often associated with environmental stability (Southwood, 1962; Harrison, 1980; Roff, 1990; Denno et al., 1991, 2001b; Wagner and Liebherr, 1992; Zera and Denno, 1997, among others). The logic is that flightless morphotypes are inclined to persist in spatially homogeneous, temporally stable habitats where food, shelter, and mates are continuously accessible to pedestrians. Conversely, flight is retained in insects living in temporary habitats, so that fluctuating levels of resource quality and abundance may be tracked. Although a number of studies support this hypothesis (e.g., Roff, 1990; Denno et al., 1991), the association of cockroaches with their habitat is not as clear as it is in insects such as stenophagous herbivores on annual plants, or waterstriders that live in temporary versus permanent ponds. Few cockroaches are exclusively associated with ephemeral or periodically disturbed habitats, although they may utilize them if available. Some species exhibit seasonal habitat shifts, but there are no known cockroaches with seasonal variation in wing morphology.
Several hurdles to understanding the role of habitat in structuring cockroach wing morphology must be added to those noted earlier. First, there can be a great deal of in-traspecific variation in habitat choice. A good example is Chorisoneura carpenteri from the Galapagos, a species with both brachypterous and macropterous forms. The fully winged morphs have been collected at elevations of 30-1000 m in agricultural areas, arid zones, pampa, humid forest, and Scalesia forest; the brachypterous form has been collected at 120-700 m in all of the listed habitats but one—the agricultural zone (Peck and Roth, 1992). Second, many cockroaches defy being described by just one aspect of their habitat, and it is difficult to tease apart the relative importance of a hierarchy of overlapping ecological levels. Is a canopy cockroach more likely to be wingless if the forest is on a mountain? Is it valid to compare a list of wingless cockroaches found in caves to a list of wingless cockroaches found in Texas (Roff, 1990, p. 395)? Finally, the fact that so many cockroaches in different habitats utilize the same microhabitats confounds analysis. Whether they are found in a desert, grassland, forest, or elsewhere, many cockroaches are associated with a continuum of dark, humid, enclosed spaces that they find or make.
The strength of the association of a given cockroach species with these subterranean and other spaces appears influential in wing development. Cockroaches that live their entire lives in burrows, galleries, or crevices, except for a brief dispersal period at the subadult or young adult stage or when the habitat becomes unsuitable, seem most prone to winglessness. It is apparent from an examination of the Panesthiinae (Fig. 2.13) that the habit of burrowing in wood or soil may be connected to the prevalence of reduced, absent, or deciduous wings in this subfamily. Cockroach species that spend their lives in the loose spaces beneath bark also fall into this category. Shaw (1918) noted that flightless cockroaches are generally cryptic in their habits, and that there was a "definite correlation" between a flattened morphology and the absence of wings. In deserts, cockroach microhabitats include the base of grass tufts and the spaces beneath debris and boulders. The majority of desert cockroaches, however, live a partially or entirely subterranean existence. Half of the 28 desert cockroaches listed by Roth and Willis (1960) live in the burrows of small vertebrates, and additional species burrow into loose sand. It should be noted that obligate cavernicoles are an extreme case of this same continuum. The ecological influences that promote wing loss in all these cockroaches, then, may differ more in degree than in type.
Several characteristics of crevices and burrows may influence wing loss in the cockroaches that permanently or periodically inhabit them. First, these are temporally stable habitats. Logs, leaf litter, and other rotting vegetable matter are continuously or periodically replenished from source plants, and migration to fresh resources, if required, is often a local trip. Second, these are homogeneous microhabitats, in that they are interchangeable dark, moist, protected quarters. If leaf litter on the forest floor loses moisture during the tropical dry season, for example, cockroaches normally found in ground-level litter are known to move into moist, arboreal accumulations of leaves (Young, 1983). Third, these are chiefly two-dimensional microhabitats, particularly for cockroach species that either rarely venture from shelters or have a modest ambit around them. Schal and Bell (1986) found that many of the flightless cockroach species in Costa Rican rainforest ground litter did not move very far in vertical space during their active period. Recent evidence suggests that it is the interaction of habitat dimensionality and habitat persistence that may have the most significant effect on insect wing morphology (Waloff, 1983; Denno et al., 2001a, 2001b). Finally, these cockroaches are able to feed within their shelter (in logs, under bark, in leaf litter, in vertebrate burrows, in social insect nests, in caves), or the shelters are situated in the immediate vicinity of potential food (soil burrowers, under rocks, under logs). The proximity of widespread, persistent, often abundant but low-quality food has two potential implications for the evolution of cockroach wing morphology. First, the insects are less tied to the seasonality of their food source. Flightlessness in insects tends to be positively correlated with their ability to remain throughout the year in their developmental habitat (Anderson, 1997; Denno et al., 2001a). Second, wing reduction and loss is often associated with nutrient limitation (Jarvinen and Vepsalainen, 1976; Kaitala and Hulden, 1990), and cockroaches that rely on rotting vegetable matter as a primary food source may be living close to their nutritional threshold. In caves, wing loss and associated morphological changes occur more frequently in organisms that rely on plant debris than those that rely on bat or bird guano (Culver et al., 1995).
Was this article helpful?