Influence of seed size on feeding preferences and diet composition of three sympatric harvester ants in the central Monte Desert, Argentina Abstract Selective seed consumption by harvester ants may affect seed abundance and composition and, ultimately, plant communities. We evaluated the influence of seed size on preferences and diet of Pogonomyrmex mendozanus, P. rastratus, and P. inermis in the central Monte Desert, Argentina. In choice experiments with Pappophorum spp. seeds of different sizes, P. mendozanus and P. rastratus preferred large seeds, maximizing energy reward. P. inermis showed a less-marked preference for large seeds, which was probably due to morphological constraints imposed by its small body size. Under natural conditions, none of the three species selected larger Pappophorum spp. seeds probably because of high travel and handling costs. Seeds of intermediate size predominated in the diet of the three species but a slight size match was detected as P. mendozanus carried larger seeds than P. rastratus and this than P. inermis, matching body-size differences. Thus, ants probably maximize energy reward but face morphological restrictions and higher costs when carrying and holding large seeds. While seeds of intermediate size are the most vulnerable ones to ant predation, small seeds are favored, as they are abundant in the soil seed bank and lowly predated.
Habitat degradation caused by cattle grazing may be a serious threat for seed-eating birds because the availability of beneficial seeds usually diminishes in grazed areas. Ecologically plastic species might, however, circumvent food deprivation via changes in foraging behaviour. We studied the limits of feeding flexibility and factors affecting seed preferences in Zonotrichia capensis, Diuca diuca, and Saltatricula multicolor. We experimentally assessed preferences for seeds of eight grass and eight forb species by using a protocol that combines choice and non-choice trials, and employed a different batch of experiments to evaluate some plausible causes of different feeding flexibility. On average, birds consumed 45-140% more grass than forb seeds, confirming previous results. Z. capensis preferred several grass and forb seeds, and showed maximum feeding flexibility. S. multicolor and, to a lesser extent, D. diuca, were grass specialists that preferred large and medium-sized grass seeds. The size of forb seeds did not affect preferences. Coat thickness of grass seeds did not seriously reduce consumption levels. Birds showed low ability to feed on resources characteristic of degraded environments (i.e. annual grass seeds). Species-specific differences in behavioural flexibility could be used to predict dietary and numerical responses of seed-eating birds to habitat degradation.
1. The abundance and composition of soil seed banks is a key determinant of plant community structure. Harvester ants can remove huge quantities of preferred seeds close to the nest affecting composition and spatial distribution of plants.2. In the central Monte desert (Argentina) ants of the genus Pogonomyrmex have high seed removal rates, especially of the five main grasses. The aim of this study was to establish if their foraging activity affects spatial patterns of the soil seed bank around their nests. Our hypotheses were: (1) removal by ants decreases seed abundance of preferred species in the soil; and (2) the effect varies in time.3. Soil seed abundance was assessed at different distances from Pogonomyrmex nests in the litter and in bare soil at the beginning, the middle, and the end of the season (late spring-early autumn). 4. A lower seed abundance of preferred species was observed close to the nest in the litter at the end of the season. Non-preferred species showed no distance gradient.5. The lower foraging activity and seed consumption at the beginning of the season could explain the temporal variation of the spatial effect. This was only observed in the litter, probably because of the higher removal frequency in this substrate.6. Colonies of Pogonomyrmex spp. could enhance the heterogeneity of soil seed banks in the central Monte desert from the summer to the beginning of the autumn. Implications for vegetation dynamics depend on the degree to which seed density limits perennial grasses recruitment after ant activity season.
Background
Consumers should show strong spatial preferences when foraging in environments where food availability is highly heterogeneous and predictable. Postdispersal granivores face this scenario in most arid areas, where soil seed bank abundance and composition associates persistently with vegetation structure at small scales (decimetres to metres). Those environmental features should be exploited as useful pre-harvest information, at least to avoid patches predicted to be poor. However, we did not find the expected spatial association in the algarrobal of the central Monte desert by observing foraging seed-eating birds, a field technique influenced by how much they exploit visited patches. In this work we tested if the first stage of foraging by granivorous birds (patch visit, encounter or exploration) is positively associated with environmental indicators of patch quality by recording the removal of single seeds from 300 scattered experimental devices during seasonal trials. Spatial selectivity was analysed by comparing the structural characteristics of used vs. available microhabitats, and evaluated against bottom-up and top-down hypotheses based on our previous knowledge on local seed bank abundance, composition and dynamics. Their foraging activity was also explored for spatial autocorrelation and environmental correlates at bigger scales.
Results
Postdispersal granivorous birds were less selective in their use of foraging space than expected if microhabitat appearance were providing them relevant information to guide their search for profitable foraging patches. No microhabitat type, as defined by their vegetation structure and soil cover, remained safe from bird exploration. Analyses at bigger temporal and spatial scales proved more important to describe heterogeneity in seed removal.
Conclusions
Closeness to tall trees, probably related to bird territoriality and reproduction or to their perception of predation risk, seemed to determine a first level of habitat selection, constraining explorable space. Then, microhabitat openness (rather than seed abundance) exerted some positive influence on which patches were more frequently visited among those accessible. Selective patterns by birds at small scales were closer to our predictions of a top-down spatial effect, with seed consumption creating or strengthening (and not responding to) the spatial pattern and dynamics of the seed bank.
Electronic supplementary material
The online version of this article (10.1186/s12898-019-0242-z) contains supplementary material, which is available to authorized users.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.