1 82 AQUATIC STEMS [CH. xin 



with its numerous aquatic species. Further instances of poly- 

 stely have been subsequently discovered among the Nymphaea- 

 ceae 1 . Though the anomalous structures met with in this family 

 cannot perhaps be explained on quite the same lines as those 

 of Auricula and Gunnera, their existence does not invalidate 

 Scott's view; they are of interest as furnishing another example 

 of the tendency towards the development of distinct steles or 

 vascular zones in aquatic plants in which secondary increase 

 in thickness is lacking. 



The present writer would like to suggest that there is possibly 

 some significance in the fact that nearly all the known cases of 

 polystely in Angiosperms occur in plants whose main vegetative 

 axis takes the form of a rhizome. This organ, not being sub- 

 jected to the same mechanical strains as an erect stem which has 

 to support leaves and branches, is not so irrevocably committed 

 to the ' continuous cylinder ' type of vascular system, which 

 is the best form of structure for withstanding bending forces. 

 That the polystelic type of anatomy does not make for strength, 

 is indicated by the recent observation, concerning a gigantic 

 Hawaian species of Gunner a^ that "the rhizome is very soft, 

 and can be severed by a single machete stroke 2 ." 



One special point of interest connected with the hypothesis 

 of the origin of polystely through an aquatic ancestry, lies in the 

 fact that, if it be accepted, it forms a particularly salient in- 

 stance of the working of a certain principle of evolution which 

 the present writer proposes to call "the Law of Loss 3 "; this 

 law will be discussed in Chapter xxvm. 



1 Gwynne-Vaughan, D. T. (1897); see also Chapter in, p. 37. 



2 MacCaughey, V. (1917). 3 Arber, A. 



