Applications of allocation and kinship models to the interpretation of vascular plant life cycles

Among vascular plants/ different life cycles are associated with characteristic ranges of propagule size. In the modern flora, isospores of homosporous pteridophytes are almost all smaller than 150 urn diameter, megaspores of heterosporous pteridophytes fall in the range 100-1000 urn diameter, gymnosperm seeds are possibly all larger than the largest megaspores, but the smallest angiosperm seeds are of comparable size to large isospores. -- Propagule size is one of the most important features of a sporophyte's reproductive strategy. Roughly speaking, larger propagules have larger food reserves, and a greater probability of successful establishment, than smaller propagules, but a sporophyte can produce more smaller propagules from the same quantity of resources. Different species have adopted very different size-versus-number compromises. The characteristic ranges of propagule size, in each of the major groups of vascular plants, suggest that some life cycles are incompatible with particular size-versus-number compromises. -- Sex expression in homosporous plants is a property of gametophytes (homosporous sporophytes are essentially asexual). Gametophytes should produce either eggs or sperm depending on which course of action gives the greatest chance of reproductive success. A maternal gametophyte must contribute much greater resources to a young sporophyte than the paternal gametophyte. Therefore, smaller gametophytes should tend to reproduce as males, and gametophytes with abundant resources should tend to reproduce as females. Consistent with these predictions, large female gametophytes release substances (antheridiogens) which induce smaller neighbouring ametophytes to produce sperm. -- The mechanism of sex determination in heterosporous species appears to be fundamentally different. Large megaspores develop into female gametophytes, and small icrospores develop into male gametophytes. Sex expression appears to be determined by the sporophyte generation. This is misleading. As argued above, the optimal sex expression of a homosporous gametophyte is influenced by its access to resources. This is determined by (1) the quantity of food reserves in its spore and (2) the quantity of resources accumulated by the gametophyte's own activities. If a sporophyte produced spores of two sizes, gametophytes developing from the larger spores' would be more likely to reproduce as females than gametophytes developing from the smaller spores, because the pre-existing mechanisms of sex determination would favor production of archegonia by larger gametophytes. Thus, the predicted mechanisms of sex determination in homosporous species could also explain the differences in sex expression of gametophytes developing from large and small spores in heterosporous species. Megaspores of living heterosporous pteridophytes contain sufficient resources for female reproduction without photosynthesis by the gametophyte (Platyzoma excepted), whereas microspores only contain sufficient resources for male reproduction. Furthermore, many more microspores are produced than megaspores. A gametophyte's optimal sex expression is overwhelmingly determined by the amount of resources supplied in its spore by the sporophyte, and is little influenced by the particular environmental conditions where the spore lands. Gametophytes determine sex expression in heterosporous species, as well as homosporous species. A satisfactory model for the evolution of heterospory needs to explain under what circumstances sporophytes will benefit from producing spores of two distinct sizes. -- In Chapter 4, I present a model for the origin of heterospory that predicts the existence of a "heterospory threshold". For propagule sizes below the threshold, homosporous reproduction is evolutionarily stable because gametophytes must rely on their own activities to accumulate sufficient resources for successful female reproduction. Whether a gametophyte can accumulate sufficient resources before its competitors is strongly influenced by environmental conditions. Gametophytes benefit from being able to adjust their sex expression in response to these conditions. For propagule sizes above the threshold, homosporous reproduction is evolutionarily unstable, because the propagule's food reserves are more than sufficient for a "male" gametophyte to fertilize all eggs within its neighbourhood. A population of homosporous sporophytes can be invaded by sporophytes that produce a greater number of smaller spores which could land in additional locations and fertilize additional eggs. Such'spores would be male-specialists on account of their size. Therefore, both spore types would be maintained in the population because of frequency-dependent selection. -- The earliest vascular plants were homosporous. Several homosporous groups gave rise to heterosporous lineages, at least one of which was the progeniture of the seed plants. The first heterosporous species appear in the Devonian. During the Devonian, there was a gradual increase in maximum spore size, possibly associated with the evolution of trees and the appearance of the first forests. As the heterospory threshold was approached, the optimal spore size for female reproduction diverged from the optimal spore size for male reproduction. Below the threshold, a compromise spore size gave the highest fitness returns to sporophytes, but above the threshold, sporophytes could attain higher fitness by producing two types of spores. -- The evolution of heterospory had profound consequences. Once a sporophyte produced two types of spores, microspores and megaspores could become specialized for male and female function respectively. The most successful heterosporous lineage (or lineages) is that of the seed plants. The feature that distinguishes seed plants from other heterosporous lineages is pollination, the capture of microspores before, rather than after, propagule dispersal. Traditionally, pollination has been considered to be a major adaptive advance because it frees sexual reproduction from dependence on external fertilization by freeswimming sperm, but pollination has a more important advantage. In heterosporous pteridophytes, a megaspore is provisioned whether or not it will be fertilized whereas seeds are only provisioned if they are pollinated. The total cost per seed cannot be assessed solely from the seed's energy and nutrient content. Rather, each seed also has an associated supplementary cost of adaptations for pollen capture and of resources committed to ovules that remain unpollinated. The supplementary cost per seed has important consequences for understanding reproductive strategies. First, supplementary costs are expected to be proportionally greater for smaller seeds. Thus, the benefits of decreasing seed size (in order to produce more seeds) are reduced for species with small seeds. This effect may explain minimum seed sizes. Second, supplementary costs are greater for populations at lower density. Thus, there is a minimum density below which a species cannot maintain its numbers. -- By far the most successful group of seed plants in the modern flora are the angiosperms. Two types of evidence suggest that early angiosperms had a lower supplementary cost per seed than contemporary gymnosperms. First, the minimum size of angiosperm seeds was much smaller than the minimum size of gymnosperm seeds. This suggests that angiosperms could produce small seeds more cheaply than could gymnosperms. Second, angiosperm-dominated floras were more speciose than the gymnosperm-dominated floras they replaced. This suggests that the supplementary cost per seed of angiosperms does not increase as rapidly as that of gymnosperms, as population density decreases. In consequence, angiosperms were able to displace gymnosperms from many habitats, because the angiosperms had a lower cost of rarity. -- Angiosperm embryology has a number of distinctive features that may be related to the group's success. In gymnosperms, the nutrient storage tissue of the seed is the female gametophyte. In most angiosperms, this role is taken by the endosperm. Endosperm is initiated by the fertilization of two female gametophyte nuclei by a second sperm that is genetically identical to the sperm which fertilizes the egg. Endosperm has identical genes to its associated embryo, except that there are two copies of maternal genes for every copy of a paternal gene. -- Chapter 9 presents a hypothesis to explain the unusual genetic constitution of endosperm. Paternal genes benefit from their endosperm receiving more resources than the amount which maximizes the fitness of maternal genes, and this conflict is expressed as parent-specific gene expression in endosperm. The effect of the second maternal genome is to increase maternal control of nutrient acquisition. -- Female gametophytes of angiosperms are traditionally classified as monosporic, bisporic or tetrasporic. Bisporic and tetrasporic embryo sacs contain the derivatives of more than one megaspore nucleus. Therefore, there is potential for conflict between the different nuclear types within an embryo sac, but this possibility has not been recognized by plant embryologists. In Chapter 10, I show that many previously inexplicable observations can be understood in terms of genetic conflicts within the embryo sac.