infinitesimal fraction of the plants that he sees around him. Of that fraction many are simply freaks or sports ; and as there is a tendency for such freaks to he collected and to find their way into herbaria it follows that herbaria may often give an exaggerated impression of the range of variability of species generally. Into the genetic phenomenon of freaks this paper cannot enter.
Over and above freaks we have to recognize a number of other types of variability.
First there is the phenomenon of geographical change, which may be either gradual or sudden. As an example of the first, the colour of the spikelets of many Cyperacece and GraminecB is paler in North India and gradually darkens as one goes south, until on the Nilgiris it is almost black. As an example of the second, Ageratum conysoides Linn, is on the Nilgiris a delicate-stemmed plant with pure mauve flowers, while in the Dharwar District of the Bombay Presidency it is a coarse-stemmed and coarse-leaved plant with dirty white flowers. The causes of geographical change are very obscure. Complex climatic factors have to be reckoned with, as well as geological formations. There is also the phenomenon of isolation of "lines," which are discussed below.
Secondly there is a type of variability of a purely edaphic character within the same geographical region. Thus Flueggea hucopyros Willd. would seem to be merely an edaphic (xerophytic) form of F. microcarpa Bl.; Leucas Montana Sr. would seem to be the xerophytic form of L. mollissima Wall. Ihe various edaphic forms of many species not hitherto split off by the Floras are well known.
Thirdly there are distinct cases of variability in life-period. Thus Fimbristylis diphylla Vahl., a perennial, has an annual form var. annua (sp.) R. and S. : and similarly Cypents Iria L. (see p. 693 of Journ. Bom. Nat. Hist. Soc. Vol. XXV No. 4). Cypcrus fiavidus Retz. would seem to be only an annual form of C. Haspan L.
Fourthly there would seem to be plants which exhibit a sort of seasonal dimorphism, not of course homologous with the same phenomenon in insects. Thus some species of Smithia have small flowers when they mature in the rains and large ones when they mature in the dry weather. S- flava Dalz. (var. in Cooke. F. Bom. Pres.), if not a valid species is apparently a seasonal form of S. sensltiva Ait. ; while last winter the writer found at Yellapur in North Kanara a plant which corresponded exactly with S. bigemina Dalz. except that it was much larger in all its parts especially the flowers, and had matured in the winter instead of the rainy season.
Fifthly there is the much more difficult phenomenon of "lines". That these are due to the interplay of Mendelian characters is now
