330 NATURAL SELECTION [CH. 



undoubtedly have a case in which Natural Selection can scarcely 

 be a factor of any great importance, and yet there is a quite 

 extraordinary variety of specific forms, many of which are 

 confined to extremely limited areas. 



That specific forms may be markedly definite and distinct, 

 and that yet the differences between them may be such that it 

 is scarcely possible to imagine that they have any special sur- 

 vival value, is also indicated in the case of a number of aquatics 

 outside the Podostemaceae. Water plants in general have the 

 character of being Protean, and there is undoubtedly great 

 individual variability associated with varying conditions of life, 

 but, at the same time, the opinions of those best qualified to 

 judge, tend to the conviction that there is great fixity rather than 

 plasticity of specific characters. It is probable that the general 

 impression as to the specific variability of aquatics is partly 

 attributable to the fact that, owing to the prevalence of vegeta- 

 tive reproduction, local races readily come into being, since any 

 variation may be perpetuated by this means for a considerable 

 time. But there is no reason to suppose that such local races 

 would come true from seed. In the case of the Eu-callitriches, 

 great variation may often be observed in the form of the leaves 

 and the size of the floating rosette. Little groups of plants 

 growing together often conform to one type in these respects ; 

 but it is probable that such homogeneous groups are merely 

 the vegetative progeny of one individual. The Potamogetons 

 are proverbially variable, and their specific identification pre- 

 sents almost insuperable difficulties to the tyro, yet a great 

 authority on this group was led, by a critical study of some of 

 these puzzling forms, to write: "All I have observed during 

 the past summer induces me to believe that, at the present time, 

 each form of the lucens group is so far constant that seed of each 

 form produces its like. Their imitation of one another under 

 variation, induced by abnormal circumstances, may betray a 

 comparatively recent common origin, but at the present day 

 our fenland pondweeds certainly seem to be ' fixed quantities 1 .' " 

 1 Fryer, A. (1887). 



