716 



THE AMERICAN NATURALIST [Vol. XL VII 



rest to variation and change. The law of recapitulation, 

 which declares that ontogeny repeats phylogeny, is now 

 accepted in a more or less modified form by almost all 

 zoologists, and despite differences in the interpretation 

 of embryology as a guide to a knowledge of ancient ani- 

 mals, it is generally agreed that early developmental 

 stages are much more conservative than are later ones. 



Xot as many striking examples of recapitulation are 

 known among plants as among animals, but Darwin long 

 ago noticed resemblances between the leaves of certain 

 seedlings and of their supposed ancestors, and others have 

 cited many similar instances. Attention has more recently 

 been called, particularly by Jeffrey, to the fact that the 

 internal structure of the young plant or of a first annual 

 ring of the mature plant, even more clearly than their 

 external form, is slow to change and therefore frequently 

 displays primitive characters. The woody axis of one 

 of the higher ferns begins in the sporeling as a solid rod, 

 which, after forming a medullated cylinder, gives rise to 

 the complicated vascular system of the adult, the various 

 steps of its development representing stages through 

 which its ancestors doubtless passed and which now char- 

 acterize the more primitive living families of ferns. In 

 the first annual ring of certain conifers occur resin canals, 

 "bars of Sanio," parenchyma cells and other structures 

 present throughout the wood of more primitive and pre- 

 sumably ancestral types. The first few annual rings 

 of many angiosperms, as well, show in the structure of 

 their rays and vessels characters which are undoubtedly 

 ancient. On an abundance of such evidence as this it 

 must be admitted that the validity of the law of recapitu- 

 lation has been demonstrated for plants almost as thor- 

 oughly as for animals. 



We have seen that conservative characters vary consider- 

 ably in their constancy according to the part of the body 

 or the stage of development with which they are associ- 

 ated. Still more notable cases of differences in fixity are 

 evident in similar characters occurring in different fami- 



