POLLEN DIMORPHISM IN THE RUBIACEAE

Abstract

striking dimorphism of pollen and stigmata fonnd in the majority of the species of the tribe Staticeae in the Plumbaginaceae was discovered by McCleod (1887). This dimorphism and derived monomorphic conditions have been extensively investigated with results that have been found extremely helpful in the elucidation of genetical, genecological, phytogeographical and evolutionary problems in the family, particularly in the genera Armteria and Limoniumn (Baker, 1948a, b, 1950, 1952, 1953a, b, 1954a, b, c). extreme usefulness of this pollenand stigma-dimorphism in investigations within the Plumbaginaceae suggests that similar phenomena should be looked for elsewhere. As there are at least 15 families in which heterostyly has been recognised (Bateman, 1954), there are at least this number of potential sites for pollen dimorphism, for it is in association with dipolar incompatibility systems that this kind of dimorphism is most likely to arise (Baker, 1954). As a preliminary to a comprehensive examination of the pollen of these families a search has been made through the existing literature. This has revealed very little that is hopeful. Differences in the sizes of grains from long-styled and short-styled plants are usual, of course. Darwin (1877, p. 135) noted that soaked pollen grains from short-styled flowers of Cinchona micrantha showed sharper angles than those from long-styled flowers. On the other hand he found that slight differences in shape between the pollen-grains of longand short-styled forms of some Primula species disappeared when the grains were wetted. However, when grains are prepared for examination by acetylation these differences persist (cf. Erdtmnan, 1952). More recently, Johnston (1953) has shown that there is a significant difference in shape between the grains of the corresponding forms of certain species of Lithospermum. A difference of another kind was found by Fritz Miiller who reported that the pollen of the short-styled flowers of a species of Villarsia was bluish in colour while the pollen from longstyled plants was yellow (cf. Darwin, 1877, p. 116). impressive differences between the pollen forms in the Plumbaginaceae, however, lie in the sculpturin-g of the exine of the grains. In Linum, there is sometimes a small difference between the pollen of longand short-styled plants in the verrucoid excrescences by which the grains are ornamented (see Erdtman, 1952) but the difference is slight and, it would seem, not always distinguishable. Only one report of pollen dimorphism comparable in clarity with that in the Plumbaginaceae has been encountered and this, strangely enough, appears to have been missed by those who have surveyed the literature during the compilation of recent works on pollen morphology. It is the account by Fritz Miiller (1869) of extreme pollen dimorphism in a species of Faramnea (Rubiaceae) from Itajahy in the Brazilian State of Santa Catarina. That this should have been overlooked is especially surprising for Miiller's account was re-presented (with illustrations) by Darwin (1877, p. 128131) in his well-known book The Different Forms of Flowers on Plants of the Same Species (fig. 1). In addition, it was abstracted by Fritz's brother Hermann (H. Muiller, 1883) in The Fertilization of Flowers.