Sympatric species that initially overlap in resource use are expected to partition the environment in ways that will minimize interspecific competition. This shift in resource use can in turn prompt evolutionary changes in morphology. A classic example of habitat partitioning and morphological differentiation are the Caribbean Anolis lizards. Less well studied, but nevertheless striking analogues to the Anolis are the Southeast Asian Draco lizards. Draco and Anolis have evolved independently of each other for at least 80 million years. Their comparison subsequently offers a special opportunity to examine mechanisms of phenotypic differentiation between two ecologically diverse, but phylogenetically distinct groups. We tested whether Draco shared ecological axes of differentiation with Anolis (e.g., habitat use), whether this differentiation reflected interspecific competition, and to what extent adaptive change in morphology has occurred along these ecological axes. Using existing data on Anolis, we compared the habitat use and morphology of Draco in a field study of allopatric and sympatric species on the Malay Peninsula, Borneo and in the Philippines. Sympatric Draco lizards partitioned the environment along common resource axes to the Anolis lizards, especially in perch use. Furthermore, the morphology of Draco was correlated with perch use in the same way as it was in Anolis: species that used wider perches exhibited longer limb lengths. These results provide an important illustration of how interspecific competition can occur along common ecological axes in different animal groups, and how natural selection along these axes can generate the same type of adaptive change in morphology.
Many carnivorous mammals consume fruits and disperse the intact seeds to specific sites. Few studies have attempted to quantify this seed dispersal or evaluate its effectiveness, despite its potential importance and functional uniqueness. In the study reported here, we found that a frugivorous carnivore, the common palm civet (Paradoxurus hermaphroditus), generated seed shadows that are distinct from those of the sympatric frugivore, the pig-tailed macaque (Macaca nemestrina), and played a unique and important role in the regeneration of Leea aculeata (Leeaceae). We found that macaques dispersed the seeds randomly, while civets dispersed them non-randomly to sites such as the banks of small rivers, rain-flow paths, abandoned trails, and treefall gaps, which are characterized by low stem density and canopy cover. Seeds of L. aculeata that were dispersed by civets to the banks of rivers and gaps had significantly higher survival and growth rates than those dispersed to rain-flow paths or abandoned trails. Seeds dispersed by macaques or to random locations also had low survival. Although the effects of the civets on seed fate were not straightforward, compared with macaques and random dispersal, civets significantly enhanced the survival and growth of L. aculeata seeds after 1 year. These results indicate that non-random dispersal by civets is important for the persistence of L. aculeata. Civets may disperse other plant species and thus could have profound effects on forest dynamics.
Differences in the density of conspecific tree individuals in response to environmental gradients are well documented for many tree species, but how such density differences are generated and maintained is poorly understood. We examined the segregation of six dipterocarp species among three soil types in the Pasoh tropical forest, Malaysia. We examined how individual performance and population dynamics changed across the soil types using 10-year demographic data to compare tree performance across soil types, and constructed population matrix models to analyze the population dynamics. Species showed only minor changes in mortality and juvenile growth across soil types, although recruitment differed greatly. Clear, interspecific demographic trade-offs between growth and mortality were found in all soil types. The relative trade-offs by a species did not differ substantially among the soil types. Population sizes were projected to remain stable in all soil types for all species with one exception. Our life-table response experiment demonstrated that the population dynamics of a species differed only subtly among soil types. Therefore, species with strong density differences across soil types do not necessarily differ greatly in their population dynamics across the soil types. In contrast, interspecific differences in population dynamics were large. The trade-off between mortality and growth led to a negative correlation between the contributions of mortality and growth to variations in the population growth rate (λ) and thus reduced their net contributions. Recruitment had little impact on the variation in λ. The combination of these factors resulted in little variation in λ among species.
While the importance of local-scale habitat niches in shaping tree species turnover along environmental gradients in tropical forests is well appreciated, relatively little is known about the influence of phylogenetic signal in species' habitat niches in shaping local community structure. We used detailed maps of the soil resource and topographic variation within eight 24-50 ha tropical forest plots combined with species phylogenies created from the APG III phylogeny to examine how phylogenetic beta diversity (indicating the degree of phylogenetic similarity of two communities) was related to environmental gradients within tropical tree communities. Using distance-based redundancy analysis we found that phylogenetic beta diversity, expressed as either nearest neighbor distance or mean pairwise distance, was significantly related to both soil and topographic variation in all study sites. In general, more phylogenetic beta diversity within a forest plot was explained by environmental variables this was expressed as nearest neighbor distance versus mean pairwise distance (3.0-10.3 % and 0.4-8.8 % of variation explained among plots, respectively), and more variation was explained by soil resource variables than topographic variables using either phylogenetic beta diversity metric. We also found that patterns of phylogenetic beta diversity expressed as nearest neighbor distance were consistent with previously observed patterns of niche similarity among congeneric species pairs in these plots. These results indicate the importance of phylogenetic signal in local habitat niches in shaping the phylogenetic structure of tropical tree communities, especially at the level of close phylogenetic neighbors, where similarity in habitat niches is most strongly preserved.
Dispersal-assembly theories of species coexistence posit that environmental factors play no role in explaining community diversity and structure. Dispersal-assembly theories shed light on some aspects of community structure such as species-area and species-abundance relationships. However, species' environmental associations also affect these measures of community structure. Measurements of species' niche breadth and overlap address this influence. Using a new continuous measure of niche and a dispersal-assembly null model that maintains species' niche breadth and aggregation, we tested two hypotheses assessing the effects of habitat heterogeneity on the ability of dispersal-assembly theories to explain community niche structure. We found that in both homogenous and heterogeneous environments dispersal-assembly theories cannot fully explain observed niche structure. The performance of the dispersal-assembly null models was particularly poor in heterogeneous environments. These results indicate that non-dispersal based mechanisms are in part responsible for observed community structure and measures of community structure which include species' environmental associations should be used to test theories of species diversity.
Leaf litter and its breakdown products represent an important input of organic matter and nutrients to mangrove sediments and adjacent coastal ecosystems. It is commonly assumed that old-grown stands with mature trees contribute more to the permanent sediment organic matter pool than younger stands. However, neither are interspecific differences in leaf decay rates taken into account in this assumption nor is our understanding of the underlying mechanisms or drivers of differences in leaf chemistry sufficient. This study examines the influence of different plant species and ontogenetic stage on the microbial decay of mangrove leaf litter. A litterbag experiment was conducted in the Matang Mangrove Forest Reserve, Malaysia, to monitor leaf litter mass loss, and changes in leaf litter chemistry and microbial enzyme activity. Four mangrove species of different morphologies were selected, namely the trees Rhizophora apiculata and Bruguiera parviflora, the fern Acrostichum aureum and the shrub Acanthus ilicifolius. Decay rates of mangrove leaf litter decreased from A. ilicifolius to R. apiculata to B. parviflora to A. aureum. Leaf litter mass, total phenolic content, protein precipitation capacity and phenol oxidase activity were found to decline rapidly during the early stage of decay. Leaf litter from immature plants differed from that of mature plants in total phenolic content, phenolic signature, protein precipitating capacity and protease activity. For R. apiculata, but not of the other species, leaf litter from immature plants decayed faster than the litter of mature plants. The findings of this study advance our understanding of the organic matter dynamics in mangrove stands of different compositions and ages and will, thus, prove useful in mangrove forest management.
Mangroves, woody halophytes restricted to protected tropical coasts, form some of the most productive ecosystems in the world, but their capacity to act as a carbon source or sink under climate change is unknown. Their ability to adjust growth or to function as potential carbon sinks under conditions of rising atmospheric CO2 during global change may affect global carbon cycling, but as yet has not been investigated experimentally. Halophyte responses to CO2 doubling may be constrained by the need to use carbon conservatively under water-limited conditions, but data are lacking to issue general predictions. We describe the growth, architecture, biomass allocation, anatomy, and photosynthetic physiology of the predominant neotropical mangrove tree, Rhizophora mangle L., grown solitarily in ambient (350 μll(-1)) and double-ambient (700 μll(-1)) CO2 concentrations for over 1 year. Mangrove seedlings exhibited significantly increased biomass, total stem length, branching activity, and total leaf area in elevated CO2. Enhanced total plant biomass under high CO2 was associated with higher root:shoot ratios, relative growth rates, and net assimilation rates, but few allometric shifts were attributable to CO2 treatment independent of plant size. Maximal photosynthetic rates were enhanced among high-CO2 plants while stomatal conductances were lower, but the magnitude of the treatment difference declined over time, and high-CO2 seedlings showed a lower Pmax at 700 μll(-1) CO2 than low-CO2 plants transferred to 700 μll(-1) CO2: possible evidence of downregulation. The relative thicknesses of leaf cell layers were not affected by treatment. Stomatal density decreased as epidermal cells enlarged in elevated CO2. Foliar chlorophyll, nitrogen, and sodium concentrations were lower in high CO2. Mangroves grown in high CO2 were reproductive after only 1 year of growth (fully 2 years before they typically reproduce in the field), produced aerial roots, and showed extensive lignification of the main stem; hence, elevated CO2 appeared to accelerate maturation as well as growth. Data from this long-term study suggest that certain mangrove growth characters will change flexibly as atmospheric CO2 increases, and accord with responses previously shown in Rhizophora apiculata. Such results must be integrated with data from sea-level rise studies to yield predictions of mangrove performance under changing climate.
In Peninsular Malaysia ten species of lycaenid butterflies use leaf flushes or inflorescences of the legume tree Saraca thaipingensis as larval hostplant. Resource partitioning among these species is regulated by a complex mixture of patterns of interaction with ants. Females of obligately myrmecophilous species lay their eggs exclusively on trees colonized by their specific host ants. On trees colonized by weaver ants, only specialist mutualists adapted to these territorial ants are able to survive, while larvae of other species are killed. The formicine ant Cladomyrma petalae, which inhabits hollow twigs of the myrmecophytic hostplant, likewise precludes oviposition by female butterflies. Lycaenid larvae confronted with this ant species never survive, but one concealed feeding species (Jamides caeruleus) escapes removal due to the cryptic life-habits of the larvae. Two facultative myrmecophiles associate in a mutualistic way with a wide and largely overlapping range of ant genera which forage at the extrafloral nectaries of leaf flushes. One species (Cheritra freja) is not myrmecophilous, but is tolerated by all but the most territorial ants. Ant-dependent hostplant selection and egg-clustering characterize the obligate mutualists, whereas facultative myrmecophiles and the non-myrmecophile distribute their eggs singly over appropriate hostplants. Signals mediating caterpillar-ant communication are highly specialized in one obligate myrmecophile (Drupadia theda), but rather unspecific in four other species tested. Altogether our observations indicate that colonization and establishment of lycaenid butterflies on S. thaipingensis trees are governed by specializations as well as opportunistic use of resources (ants and hostplant parts). Therefore, the diversity of this species assemblage is maintained by deterministic as well as stochastic factors.
The macrofauna of the soils on the west ridge of Gunung Mulu in 4th Division, Sarawak, were sampled during February and March 1978. The eleven sampling sites ranged from near the base of the mountain at 130 m a.s.1. to the summit at 2.376 m.Altitudinal changes from lowland rain forest (mixed dipterocarp forest) to lower montane and upper montane rain forests are concomitant with changes in soil from red yellow podzolics and regosols to peaty gley podzolics and organic peats.The abundance of the total macrofauna declined from 2,579 individuals m(-2) at 130 m to 145 m(-2) at 2,376 m. Declining population densities of ants and termites correlated significantly with increasing altitude but the effect on other groups was variable. Changes in total biomass were erratic and varied from 4.1-6.2 g m(-2) (alc. w.w.) in the dipterocarp forest soils to 5.8 g m(-2) in the lower montane, 9.3-20.2 g m(-2) in the upper montane (tall facies) and 1.9-9.5 g m(-2) in the upper montane (short and summit facies). Only the decline in the biomass of termites and ants correlated significantly with altitude. Other groups remained fairly constant, varied erratically or increased in the middle altitudes.The dipterocarp forest soil macrofaunal biomass was dominated by termites, beetles and earthworms (Megascolecidae and Moniligastridae), with ants the dominant predators. The lower montane forest was a transitional and ill-defined zone on the mountain and the soil macrofauna was also transitional to some extent. Termite biomass fell substantially and earthworms replaced them as the dominant detritivores, with beetles in a secondary role. Formicidae remained as the major predators. With the inception of peats in the upper montane forest (tall facies), the macrofauna was dominated by Coleoptera with earthworms, Diptera larvae and Blattodea in lesser roles. With increasing exposure in the upper montane forests (short and summit facies), several major groups disappeared altogether. The soils were dominated by Blattodea with Coleoptera and Megascolecidae of lesser importance. Chilopoda and Arachnida replaced Formicidae as the dominant predators.
Parental distance and plant density dependence of seedling leaf turnover and survival was examined to investigate predictions of the Janzen-Connell hypothesis. The focal study species, Shorea macroptera is a canopy tree species in a lowland rain forest in peninsular Malaysia. We found that the peak of the distribution of plants shifted from 3-6 m to 6-9 m during the course of the change from seedling to sapling stage. The leaf demography of the seedlings was influenced by their distance from the adult tree and also by the seedling density. Although significant density- and distance dependence in leaf production was not detected, seedling leaf loss decreased with distance from the parent tree and with seedling density. Similarly, leaf damage was not found to be distance- or density-dependent, but net leaf gain of seedlings increased with distance from the parent tree. Although no significant distance- or density-dependence was evident in terms of leaf damage, significant distance dependence of the net leaf gain was found. Thus, we concluded that positive distance dependence in the leaf turnover of seedlings may gradually contribute to a shift in the distribution pattern of the progeny through reductions in growth and survivorship.
The structure and organization of aquatic arthropod communities in Nepenthes ampullaria pitchers were studied at two sites (M in Malacca and K in Kuching) in Malaysia. The communities consisted mainly of aquatic dipteran larvae. Community M was dominated by a filter feeder, Tripteroides tenax, which reached a high density despite a strongly aggregated distribution. Community K had five trophic groups: carrion feeders, filter feeders, detritus feeders, nipping predators and hooking predators, each including multiple species. The summed density of filter feeders in Community K remained much below the level attained by filter feeders in Community M. Niche differentiation within each trophic group with regard to pitcher age and feeding behaviour was not sufficient to allow species coexistence through niche separation alone. Aggregated distributions directly reduced interspecific encounters. Nevertheless, species belonging to the same trophic group commonly shared the same pitcher, because of high occurrence probabilities of dominant species and positive associations between some taxa (due mainly to similar occupancies by pitcher age). Predator coexistence in Community K may have been facilitated by self-limitation of the large predators through intraspecific cannibalism strengthened by aggregation. Prey coexistence, on the other hand, may have relied more on population suppression by predation, especially the selective removal of old instar Tripteroides.
The seeds in fruits consumed by primates may be chewed and digested, swallowed and defecated intact, or separated from the flesh and spat out. We show by a combination of close field observations and experiments with caged animals, that long-tailed macaques (Macaca fascicularis) have a remarkably low threshold of 3-4 mm for swallowing seeds and also that wild macaques rarely break them. The seeds of 69% of the ripe fruit species eaten are spat out intact or cleaned outside the mouth and dropped. Seed-spitting significantly reduces the swallowed food bulk and may lessen the risk of releasing seed toxins during mastication. However, it requires that even small fruits are processed in the mouth one or a few at a time. We suggest that fruit storage in the cheek pouches of cercopithecine monkeys allows them to spit seeds individually without excessively slowing fruit intake while feeding on patchily distributed fruit. In contrast, Apes and New World monkeys apparently swallow and defecate most ripe seeds in their diet and colobine monkeys break and digest them, detoxifying seed defenses by bacterial fermentation.
In the Constant Frequency portions of the orientation calls of various Rhinolophus and Hipposideros species, the frequency with the strongest amplitude was studied comparatively. (1) In the five European species of the genus Rhinolophus call frequencies are either species-specific (R. ferrumequinum, R. blasii and R. euryale) or they overlap (R. hipposideros and R. mehelyi). The call frequency distributions are approximately 5-9 kHz wide, thus their ranges spead less than ±5% from the mean (Fig. 1). Frequency distributions are considerably narrower within smaller geographic areas. (2) As in other bat groups, call frequencies of the Rhinolophoidea are negatively correlated with body size (Fig. 3). Regression lines for the genera Rhinolophus and Rhinolophus, species from dryer climates have on the average higher call frequencies than species from tropical rain forests. (4) The Krau Game Reserve, a still largely intact rain forest area in Malaysia, harbours at least 12 syntopic Rhinolophus and Hipposiderso species. Their call frequencies lie between 40 and 200 kHz (Fig. 2). Distribution over the available frequency range is significantly more even than could be expected from chance alone. Two different null hypotheses to test for random character distribution were derived from frequency-size-relations and by sampling species assemblages from a species pool (Monte Carlo method); both were rejected. In particular, call frequencies lying close together are avoided (Figs. 4, 5). Conversely, the distribution of size ratios complied with a corresponding null hypothesis. This even distribution may be a consequence of resource partitioning with respect to prey type. Alternatively, the importance of these calls as social signals (e.g. recognition of conspecifics) might have necessitated a communication channel partitioning.
SO2-fumigation experiments were conducted with plants of pea (Pisum sativum) and corn (Zea mays) recognized to be, respectively, sensitive and tolerant to the gas. Accumulation of inorganic sulfur in the tissues revealed that the pea takes up considerably greater amounts of SO2 than corn. Porometer measurements and transpiration coefficients show that this greater SO2 uptake is partly related to the lower leaf diffusion resistance of peas. Additional internal sinks for SO2 uptake are discussed in order to give a complete explanation of the differential SO2 uptake of the two species.
1. Density of nests, population number, biomass, nitrogen and carbon content, calorific content and respiration rate of termites were studied at Pasoh Forest, West Malaysia mainly with four dominant species of epigeous nest builders, Macrotermes carbonarius, Dicuspiditermes nemorosus type-a, type-b and Homallotermes foraminifer, to reveal their role in the ecosystem. 2. The density of nests or mounds was 15-41/ha in M. carbonarius, 60-110/ha in the two types of D. nemorosus and 85-165/ha in H. foraminifer. 3. The population number per nest or mound was about 88,000 in M. carbonarius, 45,000 in D. nemorosus type-a, 47,000 in D. nemorosus type-b and 13,000 in H. foraminifer. The population number per hectare was about 1.8x10(6) for M. carbonarius, 4.3x10(6) for D. nemorosus type-a, 5.2x10(6) for D. nemorosus type-b and 2.1x10(6) for H. foraminifer. 4. The ratio in number of workers to soldiers was 6.4 for M. carbonarius, 19.0 for D. nemorosus type-a, 23.1 for D. nemorosus type-b and 8.9 for H. foraminifer. The ratio in number of adults (workers plus soldiers) to larvae was 1.4 for M. carbonarius, 0.59 for D. nemorosus type-a, 1.76 for D. nemorosus type-b and 3.84 for H. foraminifer. 5. The number of adults per nest of D. nemorosus type-a, type-b and H. foraminifer was linearly correlated with the weight of nest on the log-log coordinates. 6. The ratio of dry body weight to live weight was 0.24-0.30 for workers, 0.19-0.29 for soldiers and 0.15-0.22 for larvae. 7. The ash content of termite body was 26-66% for workers, 2-24% for soldiers and 1-4% for larvae. 8. The carbon content ranged from 45% to 66% of ash-free dry weight and the nitrogen content from 5.6% to 12.6%. 9. The mean calorific value of termite body was 5.3 gcal/mg in workers and soldiers on an ash-free weight basis, but was greater in nymphs and winged reproductives in the nest (6.7-6.9 gcal/mg) owing probably to their large fat storage. 10. The relation of CO2 evolution rate to temperature in these termites was similar to what has been found in other insects. 11. The total biomass of the four termite species was estimated at 6.01 kg ash-free dry weight/ha, equivalent to 0.55 kg nitrogen/ha and 3.09 kg carbon/ha. 12. The role of the fungi cultivated by M. carbonarius on their fungus combs was discussed in relation to the nutrition of termites and the decomposition of leaf-litter with special reference to their nitrogen metabolism. The high nitrogen content of fungus spherules growing on fungus combs seemed to have an important bearing on the nutrition of termites. 13. It was concluded that the termites played a very important role in the organic matter decomposition cycle of Pasoh Forest.
In the humid tropics of SE Asia there are some 14 myrmecophytic species of the pioneer tree genus Macaranga (Euphorbiaceae). In Peninsular Malaysia a close association exists between the trees and the small, non-stinging myrmicine Crematogaster borneensis. These ants feed mainly on food bodies provided by the plants and have their colonies inside the hollow internodes. In a ten months field study we were able to demonstrate for four Macaranga species (M. triloba, M. hypoleuca, M. hosei, M. hulletti) that host plants also benefit considerably from ant-occupation. Ants do not contribute to the nutrient demands of their host plant, they do, however, protect it against herbivores and plant competition. Cleaning behaviour of the ants results in the removal of potential hervivores already in their earliest developmental stages. Strong aggressiveness and a mass recruiting system enable the ants to defend the host plant against many herbivorous insects. This results in a significant decrease in leaf damage due to herbivores on ant-occupied compared to ant-free myrmecophytes as well as compared to non-myrmecophytic Macaranga species. Most important is the ants' defense of the host plant against plant competitors, especially vines, which are abundant in the well-lit pioneer habitats where Macaranga grows. Ants bite off any foreign plant part coming into contact with their host plant. Both ant-free myrmecophytes and non-myrmecophytic Macaranga species had a significantly higher incidence of vine growth than specimens with active ant colonies. This may be a factor of considerable importance allowing Macaranga plants to grow at sites of strongest competition.
The ratio of stable carbon isotopes (δ(13)C) in plants and animals from Malaysian mangrove swamps, coastal inlets, and offshore waters was determined. Vascular plants of the swamps were isotopically distinct ( x±s.d.=-27.1±1.2‰) from plankton (-21.0±0.3‰) and other algae (-18.7±2.2‰). Animals from the swamps (-20.9±4.1‰) and inlets (-19.8±2.5‰) had a wide range of isotope ratios (-28.6 to-15.4‰), indicating consumption of both mangrove and algal carbon. Several commercially important species of bivalves, shrimp, crabs, and fish obtained carbon from mangrove trees. Mangrove carbon was carried offshore as detritus and was isotopically distinguishable in suspended particulate matter and sediments. Animals collected from 2 to 18 km offshore, however, showed no isotopic evidence of mangrove carbon assimilation, with ratios (-16.5±1.1‰, range-19.1 to-13.1‰) virtually identical to those reported for similar animals from other plankton-based ecosystems. Within groups of animals, isotope ratios reflected intergencric and interspecific differences in feeding and trophic position. In particular, there was a trend to less negative ratios with increasing trophic level.
Declines in density of one natural, three artificially-seeded and eight experimental populations of Anadara granosa were monitored over periods ranging from seven to twelve months. For three of four large populations the decline in density has been adequately described by a model of the form N t=N 0·e zt where N t is density at time t, N o is initial density and z is the instantaneous mortality coefficient. The mean value of the latter was found to be-1.88·yr-1 with 95% confidence limits-1.54 to-2.25. The data obtained from the experimental populations demonstrated that mortality is unaffected by initial densities upto 2,500·m-2, but increases as shore elevation decreases due at least partly to greater access by predators at the lower levels.The data on mortality rates have been combined with previously-published information on growth rates in order to arrive at estimates of production. At one of the artificially-seeded sites the 95% confidence limits of estimates of mean production are 24 and 62 g dry tissue·m-2·yr-1 with an average value of 42 g·m-2·yr-1. Examination of the effects of shore elevation on production revealed that the latter is greatest at the uppermost shore level studied because down-shore increases in growth rate are more than offset by increases in mortality. Production per individual has been shown to decrease with increasing density. This latter fact has been used to estimate the maximum possible production which, for one site 250 cm above chart datum, lies between 49 and 72 g·m-2·yr-1 (95% confidence intervals). It has been demonstrated that for one of the culture sites variation about an estimate of mean mortality rate does not contribute as much to the variation of the final estimate of production as does the variation about the estimates of the mean values of the constants in a growth equation.As far as possible production of Anadara granosa has been compared with values reported for other marine and estuarine bivalve molluscs and it is concluded that A. granosa may be considered moderately to highly productive.
To estimate the rate of consumption of leaf litter by termites on the forest floor of Pasoh Forest Reserve, Negeri Sembilan, West Malaysia, newly fallen leaves were marked and distributed on the ground. The loss of leaf area due to termites was determined either photometrically or visually. An average of 1.70% of the total surface area of the leaf litter disappeared per week in experiment 1 and 1.25% in experiment 2 in Plot 1, and 2.9% per day in other plots located near the mounds of Macrotermes carbonarius. The amount of leaf litter accumulation in the Ao layer was estimated at about 2.3 t/ha at Plot 1, so it was likely that an amount equivalent to about 32% of the daily leaf-litter fall was transported by M. carbonarius to their mounds in experiment 1 and 22% in experiment 2. It was considered that the termites had an important role in the detritus food chain of the ecosystem.