Arboreal and Canopy Habitats

Rainforest canopies are structurally complex habitats with many niches favorable for maintaining cockroach populations: living and dead leaves, branches, bark crevices, sub-bark spaces, vines, epiphytes, suspended soils, hollow branches, vine-tree interfaces, treeholes, and bird and insect nests, among others. Canopies also contain an exceptionally rich array of organic resources (Novotny et al., 2003) known to be incorporated into cockroach diets. These include nonvascular plants, sap, bird excrement, plant litter, leaves, flowers, and fruit. In most studies of canopy invertebrates cockroaches are a consistent but minor component of the fauna. At times they are relegated to the "other" category (e.g., Nadkarni and Longino, 1990) because of the low number collected. Species-level identification is rarely attempted. In a recent eye-opening review of canopy arthropods worldwide, however, Basset (2001) concluded that while cockroaches represented only 5.3% of the individuals collected, they dominated in the amount of invertebrate biomass present. Blattaria represented 24.3% of the biomass, with Hymenoptera (primarily ants) coming in second at 19.8%, and Coleoptera ranking third at 18.8%. The revelation that nearly a quarter of the arthropod biomass in tree canopies may consist of cockroaches is particularly significant because the most commonly used canopy techniques almost certainly under-sample Blattaria. These are fogging, light traps, suspended soil cores, beating foliage, bromeliad bagging, and branch bagging (Table 3.5). Fogging is most effective on insects out in the open and is typically conducted early in the morning when the air is still. At that time, however, nocturnal and crepuscular cockroach species have likely entered harborage for the day. While the insecticide fog might kill them, they may not drop from their shelters. The same is true for cockroaches that live in tree hollows, epiphytes, insect nests, and other enclosed canopy habitats. Light traps, on the other hand, capture only volant cockroaches (Basset et al., 2003b) like Gyna gloriosa, taken at a height of 37 m in Uganda (Corbet, 1961). Branch bagging under-represents highly mobile taxa, and must be well timed. More cockroaches were collected at night than during the day using this method (Schowalter and Ganio, 2003), possibly because cockroaches perching on leaves during their active period were included in the night samples. A combination of the above methods may give a clearer picture of cockroach diversity and abundance in the canopy, with the additional use of baited traps and hand collecting from vines, suspended dead wood, treeholes, and other cryptic habitats (Basset et al., 1997). There is evidence that canopy cockroaches are a taxonomically rich group. In a fogging experiment in Borneo cockroaches were about 2% of the catch, but 40 presumed species were represented (Stork, 1991). A difficulty in documenting cockroach diversity, however, is that it is rarely possible to identify cockroach juveniles, and these can make up the bulk of Blattaria collected; 90% of the cockroaches collected by Fisk (1983) in Central American canopies were nymphs. In Venezuela, Paoletti et. al (1991) categorized cockroaches collected in their study as "microinvertebrates" because all were less than 3 mm in size. It is unclear, however, if these were small species or immatures.

Despite the high amounts of precipitation in rainforests, the canopy is a comparatively harsh environment characterized by high mid-day temperatures and low relative humidities, wind turbulence, and intense solar radiation (Parker, 1995; Rundel and Gibson, 1996).Cock-roach canopy specialists no doubt have physiological and behavioral mechanisms that allow them to function in these conditions, but we currently have little information on their biology. These taxa are distinct from species commonly collected near the forest floor by light traps and other means (Fisk, 1983), and have been characterized as "smaller, aerial varieties endowed with unexpected beauty" (Perry, 1986). Conspicuously colored beetle mimics like Paratropes bilunata live in the canopy; this species imitates both the appearance and behavior of a lycid beetle (Perry, 1986). Fisk (1983) considered the following blattellid genera as canopy indicators in Panama and Costa Rica: Imblattella, Nahublattella, Chorisoneura, Ria-tia, and Macrophyllodromia. In Costa Rican lowland rainforest, Schal and Bell (1986) collected Car. imitans and two species of Imblattella from attached, folded, dead leaves in successional stands, and noted Nyctibora noctivaga and Megaloblatta blaberoides on trees in mature forest.

Most studies of canopy invertebrates have been con-

Table 3.5. Studies in which cockroaches were collected during canopy sampling.

Method

Location

Habitat

Reference

Beating foliage Branch bagging

Bromeliad bagging Bromeliad bagging

Fogging Fogging Fogging Fogging Fogging

Fogging Fogging

Light traps

Suspended soil cores

Gabon Puerto Rico, Panama Venezuela Mexico

Sabah

Australia

Japan

Brunei

Thailand

Hawaii

Costa Rica,

Panama

Sarawak

Gabon

Lowland rainforest Evergreen wet forest

Cloud forest

Low, inundated forest, semi-evergreen forest Lowland rainforest Rainforest Mixed pine stand Lowland rainforest Dry evergreen forest

Varied; altitudinal transect Lowland forest

Lowland mixed dipterocarp forest Lowland forest

Basset et al.(2003a) Schowalter and Ganio (2003)

Paoletti et al.(1991) Dejean and Olmstead (1997)

Floren and Linsenmair (1997) Kitching et al. (1997) Watanabe (1983) Stork (1991) Watanabe and Ruaysoongnern (1989) Gagné (1979) Fisk (1983)

Itioka et al. (2003)

Winchester and Behan-Pelletier (2003)

ducted in the tropics. The canopies of temperate forests have proportionately fewer niches available because of the lower occurrence of lianas and epiphytes (Basset et al., 2003b; Novotny et al., 2003). In Japan, no cockroaches were listed in the results of a fogging study on a cypress plantation (Hijii, 1983) but they were recovered from a mixed pine stand (Watanabe, 1983). Miriamrothschildia (= Onychostylus) pallidiolus is an arboreal cockroach in Japan, the Ryuku islands, and Taiwan. The nymphs are very flat and semitransparent, and are found on live or dead tree leaves (Asahina, 1965). In the United States (South Carolina) Parcoblatta sp. were present in dead limbs and in and on the outer bark of longleaf pines sampled in winter. All trees had cockroaches on the upper bole, with a mean biomass of 36.2 mg/m—2. Cockroaches were present but variable on other parts of the tree (Hooper, 1996).Additional Blattaria that forage and shelter on live and dead tree boles at various heights include Aglaopteryxgemma (Horn and Hanula, 2002) and several species of Platyzosteria on tea tree (Leptospermum) in Australia (Rentz, 1996).

A number of species that shelter on or near the forest floor spend their active period on trunks or low branches (Schal and Bell, 1986). However, Basset et al. (2003a) reported no difference in the number of cockroaches collected between day and night beat samples in lowland tropical rainforest in Gabon. Seasonal movement into the canopy may occur, coincident with rainfall and its effects on tree phenology. In Central America, Fisk (1983) collected 16 arboreal cockroach species (n = 220) during the dry season, but 24 species (n = 986) during the wet season. Maximum cockroach numbers coincided with peak new leaf production of the early wet season. In a light trapping study in Sarawak, Itioka et al. (2003) monitored cockroach abundance in relation to flowering periods in the canopy. Blattaria were most numerous during the post-flowering stage, and lowest during the non-flowering stage (Fig. 3.12). This seasonal abundance was attributed to the increased amount of humus in the canopy during the post-flowering period, derived from spent flowers, fruits, and seeds. Barrios (2003) found that the number of cockroaches collected by beat sampling comparable leaf areas in Panama was higher in mature trees (n = 237) than in saplings (n = 60). Long-term fluctuations were evident in a study by Schowalter and Ganio (2003). Canopy cockroaches were more abundant in drought years, and least abundant during post-hurricane years in Puerto Rico and Panama.

There are numerous humid microhabitats in treetops, where cockroaches not specifically adapted to the arid conditions of the canopy thrive. Among these are habitats that are little or nonexistent in the understory, such as bird nests and the spaces in and around complex vegetation such as epiphytes, intertwining vines, lianas, tendrils,

Fig. 3.12 Average monthly numbers of cockroaches in light traps at 1, 17, and 35 m in height during three trapping periods; flowering status of the trees varied during these periods. The study was conducted in tropical lowland dipterocarp forest in Sarawak, Malaysia. After Itioka et al. (2003), with permission of T. Itioka.

Flowering Postflowering Nonflowering

Status of Flowering

Fig. 3.12 Average monthly numbers of cockroaches in light traps at 1, 17, and 35 m in height during three trapping periods; flowering status of the trees varied during these periods. The study was conducted in tropical lowland dipterocarp forest in Sarawak, Malaysia. After Itioka et al. (2003), with permission of T. Itioka.

and adventitious roots. These provide sheltered resting places and a substantial amount and variety of food, particularly in the form of suspended soils. Fisk (1983) found a general albeit inconsistent correlation between number of cockroaches collected during fogging and the number of lianas per tree. Floren and Linsenmair (1997) fogged trees from which all lianas and epiphytes were removed in Sabah, and found that cockroaches did not exceed 1% of the insects collected, on average. The substantial pool of suspended soil that accumulates in the various nooks and crannies of the canopy may be particularly important in understanding the vertical stratification of cockroach faunas (Young, 1983), yet it is commonly neglected in tropical canopy research (Winchester and Behan-Pelle-tier, 2003). Suspended soil has a high organic content derived from leaf, fruit and flower litter, epiphyte tissues, decomposing bark, and the feces, food, and faunal remains of canopy-dwelling animals. It also contains a mineral component derived from fine particles carried on wind, rain, and fog (Winchester and Behan-Pelletier, 2003).

This above-ground humus in rainforest is often thicker than the rapidly decomposing layer on the ground, and cockroaches that utilize the plant litter on the forest floor may also do so in the litter of the canopy. Leaf litter in plastic cups suspended in the lower branches of cacao trees in Costa Rica attracted cockroaches. Most abundant were species of Latiblattella and Eurycotis; the latter was also found in ground litter (Young, 1983). Studies of arthropods to date, however, generally indicate that the soil/litter fauna on the forest floor is in large measure distinct from that of the forest above (Basset et al., 2003b). One example among cockroaches is Tho. porcellana, which lives in aerial litter caught by the interlaced horizontal branches of plants in scrub jungle in India. The entire lifecycle of this cockroach is confined to suspended soil; they have no direct contact with the substratum (Bhoopathy, 1997).Winchester and Behan-Pelletier (2003) found that unidentified cockroaches collected from suspended soil cores from the crown of an Ongokea gore tree in Gabon were stratified; they were more abundant at 42 m than at 32 m above the ground.

Canopy litter is often considered ephemeral, as it can be removed by disturbances such as wind, rain, and arboreal animals (Coxson and Nadkarni, 1995). That is not true of the suspended soil trapped in some of the container epiphytes, such as the bird's nest Asplenium ferns and species of Platycerium with basal, clasping structures. In both, the litter mass acts as a sponge to retain water and nutrients (Rundel and Gibson, 1996). In the Neotropics epiphytes and hemiepiphytes may comprise greater than 60% of all individual plants, individual trees may support several hundred bromeliads, and a single bromeliad can contain more than 100 gm of soil (Gentry and Dodson, 1987; Paoletti et al., 1991). This is a substantial resource pool for cockroaches that feed on the accumulated debris and microorganisms contained within. Dejean and Olmsted (1997) found cockroaches in 67-88% of collected bromeliads (Aechmea bracteata) examined on the Yucatan peninsula of Mexico. Rocha e Silva Albuquerque et al. (1976) identified more than 30 cockroach species in bromeliads and list additional ones from the literature.

FOUR

Was this article helpful?

0 0

Post a comment