An investigation and evaluation of characters of possible taxonomic importance in the genus Dianthus.

Living material and herbarium specimens of European and Oriental species have been used in the investigation. Characters of the vegetative and reproductive phases have been examined in the first part of the work. Species were found to share the same basis-habit, but to differ in the quantitative growth of the various parts. Among the leaf characters such as size, shape, amount of fibre tissue, density of crystal clusters and pubescence, quantitative continuous variation was found throughout the species investigated. Two aspects of vegetative growth were observed: the length of the intercalary meristem in species with non-capitate and capitate inflorescences, and the development of the pairs of axillary buds found at various nodes. Some study has been made of the chromosome numbers of various populations, and variation within a species has been found. The transition from a vegetative to a floral apex has been recorded. Among floral characters continuous variation in size of calyces, calyx teeth, petals, and in depth of dissection of the margin and in barbulation has been found. Inflorescences vary continuously through-out the species seen in length of pedicel and in number of flowers per axis. The pollination mechanism, especially with regard to efficiency of seIf-pollination,has been observed. Male sterility is of direct taxonomic importance because morphology of the flowers is affected. Hybridisation has been found to occur between quite diverse species so that it is not directly useful in estimating affinity between species. However hybrids have been used in various aspects of the work and have provided a useful taxonomic tool. In the fruiting phase quantitative continuous variation has again been found in capsule length, seed length and in number of seeds per capsule. A correlation between these three characters has been attempted. In the second part of the work some taxonomic problems among various species have been outlined, and partially answered in the light of knowledge gained from Part 1. In the separation of species the quantitative continuous variation which has been found among almost all characters has been born in mind. An attempt has been made to divide the genus into small groups of closely similar species, but no attempt has been made to relate these groups to one another.