Polyembryony, referring here to situations where a nucellar embryo is formed along with the zygotic embryo, has different consequences for the fitness of the maternal parent and offspring. We have developed genetic and inclusive fitness models to derive the conditions that permit the evolution of polyembryony under maternal and offspring control. We have also derived expressions for the optimal allocation (evolutionarily stable strategy, ESS) of resources between zygotic and nucellar embryos.

It is seen that (i) Polyembryony can evolve more easily under maternal control than under that of either the offspring or the ‘selfish’ endosperm. Under maternal regulation, evolution of polyembryony can occur for any clutch size. Under offspring control polyembryony is more likely to evolve for high clutch sizes, and is unlikely for low clutch sizes (<3). This conflict between mother and offspring decreases with increase in clutch size and favours the evolution of polyembryony at high clutch sizes, (ii) Polyembryony can evolve for values of “x” (the power of the function relating fitness to seed resource) greater than 0.5758; the possibility of its occurrence increases with “x”, indicating that a more efficient conversion of resource into fitness favours polyembryony. (iii) Under both maternal parent and offspring control, the evolution of polyembryony becomes increasingly unlikely as the level of inbreeding increases, (iv) The proportion of resources allocated to the nucellar embryo at ESS is always higher than that which maximizes the rate of spread of the allele against a non-polyembryonic allele.

Finally we argue that polyembryony is a maternal counter strategy to compensate for the loss in her fitness due to brood reduction caused by sibling rivalry. We support this assertion by two empirical evidences: (a) the extent of polyembryony is positively correlated with brood reduction inCitrus, and (b) species exhibiting polyembryony are more often those that frequently exhibit brood reduction.

Monoecious figs reward their pollinators—agaonid wasps—by allocating a proportion of the flowers for egg laying, and retain the rest for seed production. It has been suggested that these proportions could be regulated by producing short-styled and long-styled flowers such that pollinator wasps could only use the former as their ovipositor does not reach the ovules of the latter. Thus the wasps can lay eggs only in the short-styled flowers and raise their offspring, and the ovules of uninfested, long-styled flowers can develop into seeds. This implied that figs bear dimorphic female flowers, with a bimodal distribution of style length. However, recent studies have shown that style length is distributed normally, with no evidence of bimodality. Therefore the regulation of allocation of flowers to the wasps does not seem to be through the production of two distinct kinds of female flowers. In this article we suggest that two factors govern the proportion of flowers rewarded to the wasps: (i) passive regulation, which is a consequence of the optimization of wasp ovipositor length, and (ii) active regulation, where figs are selected to enhance the variance of style length. We show that these arguments lead to certain predictions about the optimum ovipositor length, the proportion of the flowers available to the wasps, and the coefficient of variation of style length. We also show that data for 18 fig-wasp associations conform well with these predictions. We finally suggest that the regulatory process outlined here can be extended to evolution of style length in dioecious fig species also.

Given the increasing anthropogenic pressures on forests, the various protected areas—national parks, sanctuaries, and biosphere reserves—serve as the last footholds for conserving biological diversity. However, because protected areas are often targeted for the conservation of selected species, particularly charismatic animals, concerns have been raised about their effectiveness in conserving nontarget taxa and their genetic resources. In this paper, we evaluate whether protected areas can serve as refugia for genetic resources of economically important plants that are threatened due to extraction pressures. We examine the population structure and genetic diversity of an economically important rattan, Calamus thwaitesii, in the core, buffer and peripheral regions of three protected areas in the central Western Ghats, southern India. Our results indicate that in all the three protected areas, the core and buffer regions maintain a better population structure, as well as higher genetic diversity, than the peripheral regions of the protected area. Thus, despite the escalating pressures of extraction, the protected areas are effective in conserving the genetic resources of rattan. These results underscore the importance of protected areas in conservation of nontarget species and emphasize the need to further strengthen the protected-area network to offer refugia for economically important plant species.