The multitude of floral architectures within angiosperms are complex adaptations (or adaptive strategies) insuring the transfer of pollen from a plant to another of the same species, the subsequent growth of the pollen tube in the style, and the fertilization of the ovule. During the past decade, ecologists and evolutionists have learned much about floral structure in relation to biotic and abiotic dispersal agents of pollen, specifically how the shape and sizes of flower parts optimize pollen transport and deposition by various pollinators (Kugler, 1970; Frankel and Galun, 1977; Faegri and van der Pijl, 1979; Cruden and Miller-Ward, 1981). Much more attention has been given to floral syndromes in relation to pollinators than to the integrated evolution of floral components, even though it is apparent that if the system is to remain functional, most evolutionary alterations in one component must be compensated for by changes elsewhere. Studies on that aspect have been rather few (Galinat, 1961; Pandey, 1971; Lee, 1978; Cruden and Miller-Ward, 1981). Many plant families have undergone extensive radiation in pollination biology, and correspondingly in floral form and size. Some of the changes are in characteristics involved in pollinator attraction; others are in characteristics immediately involved in pollen presentation, transfer and receipt. The position of stigmas is a notable example in the latter category. In species which are bird-pollinated stigmas may be exserted, whereas in butterfly-pollinated relatives the stigmas may be deep within the corolla tube; in bee-pollinated relatives stigmas may be near the orifice of the corolla tube. The position of the stigma is determined by the length of the style. Although this position is pivotal in maximizing pollen receipt, the great diversity in style length (10-fold differences in many genera) receives relatively little attention. Similarly, pollen characteristics and diversity have been described at length with an emphasis either on their taxonomic value (Erdtman, 1969, 1970) or on their biochemistry and physiological function (Barbier, 1970; Heslop-Harrison, 1971, 1979; Stanley and Linskens, 1974; Muller, 1979). Given that pollen grains must send tubes through the style on their way to the ovule, we may ask whether differences in style length are accompanied by differences in some attribute of pollen, and whether changes in style length of necessity are accompanied by changes in pollen. In general, the germination of pollen and growth of the tube are conditioned by the components of the pollen grain, as by physiochemical processes involving stigma and style (Rosen, 1971; Stanley, 1971; de Nettancourt, 1977; Heslop-Harrison, 1978, 1979; Clarke et al., 1979). The pollen may contain carbohydrates and lipids in several forms, enzymes, membranes, and amino acids, all of which are essential for germination and growth; however, many of them are not synthesized during tube growth. Some nutrients, such as polysaccharides and amino acids, can be absorbed from the style, whereas some other materials (especially m-RNA) cannot be obtained this way and thus function as limiting factors (Barbier, 1970; Mascarenhas, 1971, 1975; Stanley, 1971; Vasil, 1974). Since many similar provisions for growth must be present in the pollen grain
Read full abstract