Reformation process of the neuronal template for nestmate-recognition cues in the carpenter ant <Emphasis Type="Italic">Camponotus floridanus</Emphasis>

Ants use cuticular hydrocarbons (CHC-profiles) as multicomponent recognition cues to identify colony members (nestmates). Recognition cues (label) are thought to be perceived during ant–ant encounters and compared to a neuronal template that represents the colony label. Over time, the CHC-profile may change, and the template is adjusted accordingly. A phenotype mismatch between label and template, as happens with CHC-profiles of foreign workers (non-nestmates), frequently leads to aggressive behavior. We investigated the template reformation in workers of the carpenter ant Camponotus floridanus by masking their antennae with postpharyngeal gland (PPG) extracts from nestmates or non-nestmates. The behavioral response of manipulated workers encountering unmanipulated workers was measured independently after 2 and after 15 h. After 2 h of incubation, workers treated with either of the two PPG-extracts showed low aggression towards nestmates and high aggression towards non-nestmates. In contrast, after 15 h of incubation, workers treated with non-nestmate PPG-extract showed low aggression towards both nestmates and non-nestmates. The slow (>2 h) adjustment of the template indicates a reformation localized in the central nervous system rather than in chemosensory neurons. In addition, our data show that template adjustment to a new CHC-profile does not impair the assessment of the old CHC-profile as nestmate label.     

Species and Colony Components in the Recognition Odor of Young Social Wasps: Their Expression and Learning (Polistes biglumis and P. atrimandibularis; Hymenoptera: Vespidae)

In Polistes, nestmate recognition relies on the learning of recognition cues from the nest. When wasps recognize nestmates, they match the template learned with the odor of the encountered wasp. The social wasp Polistes biglumis use the homogeneous odor of their colony to recognize nestmates. When these colonies become host colonies of the social parasite P. atrimandibularis, colony odor is no longer homogeneous, as the parasite offspring have an odor that differs from that of their hosts. In trying to understand how the mechanism of nestmate recognition works in parasitized colonies and why parasite offspring are accepted by hosts, we tested the responses of resident Polistes biglumis wasps from parasitized and unparasitized colonies to newly emerged parasites and to nestmate and non-nestmate conspecifics. The experiments indicate that immediately upon eclosion both young parasites and young hosts lack a colony odor and that colony odor can be soon acquired from the accepting colony. In addition, while residents of nonparasitized colonies recognize only the odor of their species, resident hosts of parasitized colonies have learned a template that fits the odors of two species.     

Nestmate discrimination in the harvester termite <Emphasis Type="Italic">Hodotermes mossambicus</Emphasis>

Summary Nestmate discrimination was studied in the African harvester termite Hodotermes mossambicus in trophallactic feeding experiments. The results show that Hodotermes mossambicus is able to discriminate nestmates from non-nestmates, feeding nestmates significantly more than non-nestmates. Experiments in which termites were treated with the antibiotic tetracycline for 48 hours and then used either as donors or as recipients in the trophallactic assay indicate that the degree of similarity in the intestinal flora strongly affects nestmate discrimination. Antibiotic-treated termites were fed more frequently by treated non-nestmates than by untreated nestmates. Pairs of non-nestmates, which were both treated with the antibiotic, exchange food as frequently as untreated nestmates. Pairs of nestmates, on the other hand, out of which only one animal was treated, exhibit nearly as little trophallactic contacts as the non-nestmate controls. The results parallel similar recent results in Reticulitermes speratus and indicate that termites can make use of recognition cues which are quite different from those of social hymenopterans, cues that are produced by intestinal symbionts.Received 5 November 2002; revised 17 and 27 February 2003; accepted 9 March 2003.     

Nestmate recognition by guards of the Asian hive bee <Emphasis Type="Italic">Apis cerana</Emphasis>

When a honey bee colony becomes queenless and broodless its only reproductive option is for some of its workers to produce sons before the colony perishes. However, for this to be possible the policing of worker-laid eggs must be curtailed and this provides the opportunity for queenless colonies to be reproductively parasitized by workers from other nests. Such reproductive parasitism is known to occur in Apis florea and A. cerana. Microsatellite analyses of worker samples have demonstrated that the proportion of non-natal workers present in an A. cerana colony declines after a colony is made queenless. This observation suggests that queenless A. cerana colonies may be more vigilant in repelling potentially parasitic non-natal workers than queenright colonies. We compared rates of nestmate and non-nestmate acceptance in both queenright and queenless A. cerana colonies using standard assays and showed that there is no statistical difference between the proportion of non-nestmate workers that are rejected in queenless and queenright colonies. We also show that, contrary to earlier reports, A. cerana guards are able to discriminate nestmate workers from non-nestmates, and that they reject significantly more non-nestmate workers than nestmate workers. Received 25 February 2008; revised 21 May 2008; accepted 25 June 2008.     

Specific recognition of reproductive parasite workers by nest-entrance guards in the bumble bee <Emphasis Type="Italic">Bombus terrestris</Emphasis>

Background

The impact of social parasites on their hosts’ fitness is a strong selective pressure that can lead to the evolution of adapted defence strategies. Guarding the nest to prevent the intrusion of parasites is a widespread response of host species. If absolute rejection of strangers provides the best protection against parasites, more fine-tuned strategies can prove more adaptive. Guarding is indeed costly and not all strangers constitute a real threat. That is particularly true for worker reproductive parasitism in social insects since only a fraction of non-nestmate visitors, the fertile ones, can readily engage in parasitic reproduction. Guards should thus be more restrictive towards fertile than sterile non-nestmate workers. We here tested this hypothesis by examining the reaction of nest-entrance guards towards nestmate and non-nestmate workers with varying fertility levels in the bumble bee Bombus terrestris. Because social recognition in social insects mainly relies on cuticular lipids (CLs), chemical analysis was also conducted to examine whether workers’ CLs could convey the relevant information upon which guards could base their decision. We thus aimed to determine whether an adapted defensive strategy to worker reproductive parasitism has evolved in B. terrestris colonies.

Results

Chemical analysis revealed that the cuticular chemical profiles of workers encode information about both their colony membership and their current fertility, therefore providing potential recognition cues for a suitable adjustment of the guards’ defensive decisions. We found that guards were similarly tolerant towards sterile non-nestmate workers than towards nestmate workers. However, as predicted, guards responded more aggressively towards fertile non-nestmates.

Conclusion

Our results show that B. terrestris guards discriminate non-nestmates that differ in their reproductive potential and respond more strongly to the individuals that are a greatest threat for the colony. Cuticular hydrocarbons are the probable cues underlying the specific recognition of reproductive parasites, with the specific profile of highly fertile bees eliciting the agonistic response when combined with non-colony membership information. Our study therefore provides a first piece of empirical evidence supporting the hypothesis that an adapted defensive strategy against worker reproductive parasitism exists in B. terrestris colonies.

     

Colony structure and reproductive sharing among queens in a tropical paper wasp, <Emphasis Type="Italic">Polistes olivaceus</Emphasis>

Reproductive partitioning among group members is a key feature in social Hymenoptera. We investigated the genetic colony structure of a tropical paper wasp Polistes olivaceus, with an emphasis on variation in the number of queens and reproductive sharing among queens. Among 22 P. olivaceus colonies, 6 were monogynous, 9 polygynous, and 7 were queenless. Adults and brood (eggs and larvae) were genotyped based on six polymorphic microsatellite loci. In each of the polygynous colonies, progenies were assigned to their mothers using maximum-likelihood methods. Nestmate queens were full sisters. The vast majority of reproduction appeared to be monopolized by the dominant queen (α), and the overall reproductive skews were 0.63 ± 0.04 (B index) and 0.97 ± 0.02 (S _c index). Although all nestmate queens had equal reproductive potential, the high magnitude of reproductive skew was enigmatic in this species. Although 9.55 ± 2.07 workers contained developed ovaries in 11 of 15 queen-right colonies, they were unrelated to the nestmate queens but related to each other as full sisters, suggesting that they were the remaining offspring of superseded queens. In 2 of the 11 colonies, we detected male eggs produced by reproductive workers. On average, 7.27 % of the total genotyped male eggs were derived from reproductive workers among the colonies. These results suggest three possibilities regarding the presence of reproductive workers in the P. olivaceus colonies: drifting between colonies, putative remaining offspring from superseded queens, and the offspring of unrelated females who joined the colonies and reproduced there. We found no worker-derived larvae or adult males, suggesting that male eggs were removed by nestmates at some point between oviposition and hatching.     

Pruning of host plant neighbours as defence against enemy ant invasions: <Emphasis Type="Italic">Crematogaster </Emphasis>ant partners of <Emphasis Type="Italic">Macaranga</Emphasis> protected by "wax barriers" prune less than their congeners

Many ant partners of tropical ant-plants prune the leaves and shoot tips of other plants growing around their hosts. According to the hypothesis proposed by Davidson et al. (Ecology 69:801-808), this specialized behaviour not only protects the host plants against overgrowth, but it also conveys a direct benefit to the ant colony as it removes contact points to the neighbouring vegetation where invasions of enemy ants could occur. Here we test this hypothesis by comparing pruning intensity in five closely related Crematogaster (subgenus Decacrema) plant-ant species (and one species of Technomyrmex) that differ in their exposure to competition by other ants. Pruning intensity was quantified by measuring the area loss of paper tape pieces wrapped around the stems of Macaranga host plants. All Crematogaster (Decacrema) ants tested but not Technomyrmex sp. pruned, but the intensity of the behaviour varied strongly between and within species. Pruning was significantly weaker in the three tested Crematogaster species inhabiting Macaranga host plants with a slippery, waxy stem surface, which functions as a mechanical barrier protecting the specific ant partners against generalist competitors. Pruning was generally stronger on more densely ant-populated trees. Even though the number of ants per twig length was lower in associations of ants with glaucous Macaranga hosts, only part of the variation of pruning activity could be explained by "ant density". When corrected for ant density, "wax-running" Crematogaster (Decacrema) ants still pruned more weakly than their congeners inhabiting non-glaucous Macaranga hosts. Pruning is obviously most important when an ant-plant is potentially accessible to intruders, but less necessary when the ant colony is isolated by a protective wax barrier. Our results support the hypothesis that "selfish" defence against invasions is the major selective pressure that has led to the development and maintenance of pruning behaviour in weakly competitive plant-ants.     

Pupal experience and nestmate recognition in <Emphasis Type="Italic">Polistes dominula</Emphasis> wasps

Social insects can discriminate between nestmates and aliens by comparing the chemical phenotype of an individual with the neural representation of their own colony odor (template). For social paper wasps of the genus Polistes, a general recognition model has been proposed and tested on few North American species: wasps learn colonial recognition cues from the nest paper during the first hours after emergence as adults. However, a recent study revealed that workers of Polistes dominula do not necessarily use the nest paper for early post-emergence cue-learning, suggesting that cues used for the formation of the referent template in this species could be learned at different life stages. Pre-natal learning is a widespread phenomenon in animals and it can shape various behaviors in adults. Here, we investigated whether pre-imaginal learning affects later nestmate recognition in P. dominula wasps. We reared worker pupae in artificial conditions to test whether the absence of nest material, or the exposure to nest material taken from a foreign conspecific colony, during pupal development would alter the nestmate recognition ability in adult life. Our results show that wasps maintain their correct recognition ability regardless of the treatment, suggesting that wasps do not form their referent template during the pupal stage from the nest paper. Alternative hypotheses for template formation timing and source of recognition cues are discussed. Moreover, we investigated whether young wasps already possess, on their own body, reliable chemical cues to form a recognition template by self-referent phenotype matching.     

Acanthocephalan (Oligacanthorhynchidae) parasitism of the Caribbean termite <Emphasis Type="Italic">Nasutitermes acajutlae</Emphasis>: implications for reproductive success

Summary Few studies have examined the impact of parasitism on free-living social insects. We documented the association between an acanthocephalan parasite and survival, reproduction and growth of a Caribbean termite (Nasutitermes acajutlae). We measured these parameters yearly in 100 termite colonies for 4 years. In 2001 and 2002 we also compared the rate at which parasitized and unparasitized colonies rebuilt damaged foraging trails because trails protect foragers from predation as well as fluctuations in temperature and humidity. Although there were no significant differences in growth or survival between parasitized and unparasitized colonies, parasitized colonies reproduced significantly less often and rebuilt foraging trails significantly more slowly than unparasitized colonies. The average parasitized colony may sustain a loss of alate production =30% in the year in which it is parasitized, and a loss of up to 19% if it is parasitized once in a 5-year period. We conclude that the acanthocephalan parasite is associated with significant alteration of fitness parameters of individual N. acajutlae colonies. However, prevalence of parasitism is low (6.8%) and alate production by non-parasitized colonies is likely to compensate for losses due to parasitism. Therefore, it seems unlikely that this parasite affects N. acajutlae at the population level.Received 5 October 2003; revised 19 November 2003; accepted 3 December 2003.     

Growth rate and body weight of foundress-reared offspring in a paper wasp, <Emphasis Type="Italic">Polistes chinensis</Emphasis> (Hymenoptera,Vespidae): no influence of food quantity on the first offspring

Summary. Laboratory experiments with Polistes chinensis (Pérez) showed that growth rates and weights of the first-hatched offspring were not different between food-available and food-deficient conditions. In food-deficient colonies, some first-hatched offspring grew quickly, but their dry weights were not different from that of later offspring. The present results suggest that foundresses of P. chinensis may intensively feed some first-hatched larvae to produce the first workers quickly under food-deficient conditions. Since food availability of field foundresses of this species has been estimated as very low, it seems likely that they may produce the first offspring as early as possible by such feeding. On the other hand, heavier offspring were produced when extra food is available, especially in the latter half of the order of emergence. Foundresses of this species may change the amount of prey fed to each larva in response to food availability.     

Natural patterns of caste and sex allocation in the stingless bees <Emphasis Type="Italic">Melipona favosa</Emphasis> and <Emphasis Type="Italic">M. trinitatis</Emphasis> related to worker behaviour

Summary. This paper is the report of a long-duration study of the production of gynes and males in Melipona stingless bees in Trinidad and Tobago, including a behavioural study of laying workers. In 167 brood samples (16342 pupae) of M. favosa, taken from 78 colonies in the natural environment over the years 1993-2001, workers represented 78.4% and males represented 17.3%. Gynes represented 5.1% of all female brood. Gyne proportion never exceeded 22.5%. Gynes were found to be common throughout the year. Like the gynes, males were present all year, but their occurrence varied over the months. Males were most common in the months of July and August. At the colony level male occurrence in the brood was characterized by very high values (maximum 74.2%), next to a frequent complete absence (in 26.5% of all samples). The finding of bout-like production of males at the colony level is supported by the observations on reproductive worker behaviour. The colonies appeared to be in different phases of male production. When male producing laying workers were present, they competed heavily for egg-laying opportunities. Various behaviours of laying workers and the queen are described and interpreted on the basis of reproductive competition and kin conflict theories. From these results, we conclude that in M. favosa, workers have a distinct influence on reproduction, including the production of gynes and males.     

Colony founding and social parasitism in <Emphasis Type="Italic">Lasius</Emphasis> (<Emphasis Type="Italic">Acanthomyops)</Emphasis>

We investigated colony foundation behavior of three species of the Lasius claviger group, L. latipes, L. interjectus and L. claviger, using field observations and laboratory experiments. Laboratory studies included observing gyne behavior prior to interaction with host colonies (e.g. grouping, overwintering, feeding) and experiments in which gynes were introduced to putative host colonies. Observations of gyne parasitic behavior immediately after mating flights in the field and the discovery of mixed colonies both confirm the parasitic behavior of these species and point toward a wide host species range including two cases of hyperparasitism. L. interjectus and L. latipes enter the host colony immediately after the nuptial flight, while L. claviger may hibernate and enter in the spring. Laboratory experiments showed that L. interjectus gynes prefer to group and suffer lower mortality when in a group, which is consistent with field observations of L. interjectus gynes entering host colonies in large numbers. Details on behavior of L. latipes alpha and beta form gynes are also given.     

Pushing Wasps to Work: Decentralized Aggression Induces Increased Activity in the Paper Wasp <Emphasis Type="Italic">Polistes versicolor</Emphasis>

When helpers from cooperative breeding animals have some expectation of direct reproduction, there is potential for conflict over how much aid they should provide to the colony. For example, if food is shared among all colony members, then higher levels of foraging by a helper would be desirable for the colony as a whole. However, because foraging is risky and physiologically costly, hopeful reproductive helpers could avoid foraging. Evidence suggests that this work-conflict could be resolved if helpers are aggressively coerced by their nestmates to provide aid. Here, we showed that in the primitively eusocial paper wasp Polistes versicolor, colony food starvation leads to an increasing in aggression that results in an increasing activity level (including foraging). We propose that aggression affects forage levels because (i) attacks from nestmates are directed toward known foragers rather than non-foragers; and (ii) resting wasps generally respond to aggression by becoming more active while already active wasps generally respond by switching the task they were doing. In P. versicolor, direct reproduction is an option for helpers. It means that they can be considered as hopeful reproductive individuals seeking to avoid performing risky behaviours, like foraging. In this sense, decentralized aggression from nestmates could be a coercive mechanism to force wasps performing undesirable tasks, while simultaneously enhancing the performance of a variety of other tasks.     

<Emphasis Type="Italic">Aphidius ervi</Emphasis> Preferentially Attacks the Green Morph of the Pea Aphid, <Emphasis Type="Italic">Acyrthosiphon pisum</Emphasis>

The pea aphid, Acyrthosiphon pisum Harris (Hemiptera: Aphididae) is found in red and green color morphs. Previous work has suggested that the aphidiine parasitoid Aphidius ervi Haliday preferentially attacks green pea aphids in the field. It is not clear whether these results reflect a real preference, or some unknown clonal difference, such as in immunity, between the aphids used in the previous studies. We used three susceptibility-matched pairs of red and green morph pea aphid clones to test for preferences. In a no-choice situation, the parasitoids attacked equal proportions of each color morph. When provided with a choice, A. ervi was significantly more likely to oviposit into colonies formed from green morphs when the neighboring colony was formed from red morph aphids. In contrast, red morphs were less likely to be attacked when their neighboring colony was of the green morph. By preferentially attacking green colonies, A. ervi may reduce the likelihood of intraguild predation, as it is suggested that visually foraging predators preferentially attack red aphid colonies. Furthermore, if this host choice behavior is replicated in the field, we speculate that color morphs of the pea aphid may interact indirectly through their shared natural enemies, leading to intraspecific apparent competition.     

Prevalence of the parasite Strepsiptera in <Emphasis Type="Italic">Polistes</Emphasis> as detected by dissection of immatures

Summary. hough the paper wasp genus, Polistes, is well studied, we know little of the incidence of parasitism in this group. Here we present details of 45 nest dissections for 4 species: P. dominulus (Christ), P. gallicus (L.), P. stabilinus Richards and P. carnifex (F.) to detail levels of parasitism of colony members by the obligate parasitic group of insects, the Strepsiptera. All 4 species showed evidence of parasitism among immature members. For 3 species, more than 50% of inspected nests were parasitized and the levels of parasitism among brood (larvae and pupae) was very high and did not differ significantly between parasitized nests. One species, P. stabilinus, suffered very low levels of parasitism, which may be related to its habitat choice. The number of parasites per host was positively related to the proportion of infected brood (parasite prevalence) and in some cases reached phenomenally high levels, which casts doubt on previ ously assumed mechanisms of infection for nest-making Hymenoptera, i.e. phoresy. We also document cases of egg parasitism and encapsulation in Polistes nests. Our data show that parasitism levels greatly varied among areas. Finally, the recent debate on the competitive advantage of P. dominulus in its introduced range, USA, has credited an absence of strepsipteran parasites of this species in facilitating its spread. For the first time, we document levels of parasitism for this species in its nature P range and this would appear to corroborate previous claims. We place our work in the context of other studies of parasitism of social insects and posit that the genus Polistes may have much to offer to this field.     

Near-infrared spectroscopy identifies the colony and nest of origin of weaver ants, <Emphasis Type="Italic">Oecophylla smaragdina</Emphasis>

The ability of social insects to differentiate between colony members and others is essential for the survival of the colony. It enables individuals to direct altruistic behavior towards colony mates, while protecting the colony from intruders. Colonies have a distinct chemical signature that facilitates colony-mate recognition. However, in large polydomous colonies, this signal is likely to be modified by factors unique to each nest. We demonstrate, using near-infrared spectroscopy (NIRS), that individual weaver ants, Oecophylla smaragdina, can be differentiated with respect to their colony and nest of origin. 76.5% of individuals from four colonies could be correctly assigned to their colony of origin; and 79.6% of individuals could be assigned to the correct nest (of two) within their colony. Despite the differences between nests within colonies, in most cases individuals from one nest were more similar to individuals from the other nest within the colony than they were to individuals from any nest outside the colony. Therefore, a distinctive colony identity is maintained despite differences between nests within colonies. We discuss the advantages of using NIRS as a faster and less expensive alternative to the analysis of cuticular hydrocarbons following extraction and identification with gas chromatography/mass spectroscopy. Received 26 November 2007; revised 22 January 2008; accepted 25 January 2008.     

Social interactions between an inquiline ant, <Emphasis Type="Italic">Ectatomma parasiticum</Emphasis>, and its host, <Emphasis Type="Italic">Ectatomma tuberculatum</Emphasis> (Formicidae,Ectatomminae)

Inquilines, workerless social parasites, frequently show advanced adaptations to their parasitic life style that indicate a long co-evolutionary history with their host. Ectatomma parasiticum, the first inquiline described in the poneromorph group, usurps established colonies of E. tuberculatum and produces only sexuals. In laboratory colonies, parasites were specifically attacked by the host workers, showing a failure in their social integration. Social interactions were frequent between parasites and their hosts, especially antennation, interpreted as attempts to promote colonial odor transfer. Inquilines destroyed eggs laid by the other queens (67 out of 209 eggs laid), including conspecific parasites, which is unusual. Such partial integration into the host colony and potential parasite virulence argue for a recent evolution of social parasitism in E. tuberculatum.     

Colony kin structure and breeding patterns in the social wasp, <Emphasis Type="Italic">Polistes biglumis</Emphasis>

We used DNA microsatellites to study colony kin structure and breeding patterns in the primitively eusocial wasp Polistes biglumis. P. biglumis inhabits cool areas at high altitudes and, as a consequence, has a reduced colony cycle compared to more temperate Polistes. P. biglumis colonies are always founded and controlled by a single foundress, but nest failure is common and foundresses losing their nests do not have time to start new ones due to the short season. Instead, nests are characterized by frequent female turnover, in the form of females taking over (usurpation) other con-specific nests. Our results showed that most nests had offspring from multiple unrelated females, including some where multiple females were not observed in monitoring. Reconstruction of behavioural events from the genetic data revealed three types of multiple matriline nests: (a) nests that were usurped by another female, where the original nest owner disappeared following the usurpation event, (b) nests that were joined by another female, where the original nest owner stayed following the joining event, (c) nests that were both usurped and joined by other females. We also found, for the first time, a clear indication of multiple mating by Polistes females. Moreover, males mating with the same female were related, which may be explained by the lek mating system of P. biglumis. Finally, we analysed the nest sex ratios and how it changed during the season and found that sexes were produced sequentially, males before females.     

Food supplementation affects colony-level life history traits in the annual social wasp <Emphasis Type="Italic">Polistes metricus</Emphasis> (Hymenoptera,Vespidae)

Summary A field population of Polistes metricus Say near St. Louis, Missouri was supplemented with dilute Apis mellifera honey and Trichoplusia ni caterpillars during the entire colony development period. Offspring were collected at two times to coincide with emergence of worker and reproductive broods. Food supplementation had no effect on nest size, the number of worker offspring, or the size of workers. Supplemented colonies did produce more females in August but female size was unaffected by supplementation. Supplementation had no effect on the number of males produced, but males were slightly larger on supplemented nests. We interpret these findings in the context of social insect life history theory.Received 30 June 2003; revised 9 December 2003; accepted 18 December 2003.     

Nutrient partitioning and storage in arid-zone forests of the mangroves Rhizophora <Emphasis Type="Italic">stylosa</Emphasis> and <Emphasis Type="Italic">Avicennia</Emphasis><Emphasis Type="Italic">marina</Emphasis>

Mangrove partitioning and storage of macronutrients and trace metals were examined in different arid coastal settings of Western Australia. Total living biomass in three Rhizophora stylosa forests, which ranged from 233 to 289 t DW ha^-1, was significantly greater than biomass in three Avicennia marina forests (range: 79-155 t DW ha^-1). Although prop roots and stems were the largest single tree components for R. stylosa and A. marina, respectively, most nutrients were stored in leaves and living roots of both species. However, only a small fraction of the total nutrient pool was vested in tree biomass; the vast bulk was in soils. A large below-ground pool of dead fine roots was identified at all stands, equivalent to 36-88% DW of total living tree biomass. The amount of Ca, S, Cl, Na, Si, Fe, Mn, Zn, B, Mo and Cu vested in dead roots of both species was greater than in the total living tree biomass. The proportion of Fe and S vested in live and dead roots was exceptionally large, consistent with previous evidence of metal plaques on mangrove roots. Sulphur, iron and zinc in dead roots of both species constituted the bulk of these metals. R. stylosa trees preferentially accumulated more Mg, S, Cl, Na, Si, Fe, Mn, B and Mo than A. marina trees. Proportionally greater storage of P, N, Ca, K, Cu and Zn occurred in two of the three A. marina forests. Foliar concentrations of Mg, S, Mn, B and Mo in mangrove leaves were at the high end of the range reported for other tropical trees, but other elemental concentrations were at the low or mid-range. Nitrogen limitation in these forests is implied by a positive correlation between total tree N and net canopy production and by a lower percentage of ecosystem N in tree biomass as compared with other forests. Unlike terrestrial forests where a large proportion of nutrient capital is vested in floor litter, most elements in these mangrove forests are stored in dead roots. A large reservoir of dead roots below the forest floor may serve as a conservation mechanism, particularly in such arid oligotrophic environments.