8 Development of the Fern Leaf 



in sand mixed with leaf -mold, on a bank near the sea, on Shelter 

 Island, New York; in the series borne by plants growing on 

 rocks, on an exposed hillside, in central Vermont; and in the 

 series borne by plants raised in mixed soil in flower-pots kept 

 under glass in a greenhouse. 



There is, however, one hypothesis which offers a rational 

 explanation of such changes; namely, that in the species' leaf 

 during its development are indicated characters that the leaves 

 of the plant's ancestry possessed. That is what we should expect 

 to be true if we accept as true the alleged fact that in other living 

 things traces of the individual's ancestry are to be seen in 

 changes which take place in the individual during its develop- 

 ment from the first to the adult stage. 



The fact that, in the case of ferns, the changes in the leaf, 

 which is a part of the fern individual, are exemplified in a series 

 of successive ephemeral leaves rather than in one persistent leaf, 

 need not militate against this hypothesis; which finds support in 

 such facts as the following, difficult to account for on any 

 other. 



(i) In some species peculiarities are present in the early 

 stages of the leaf which disappear in the later stages: in some 

 peculiarities are absent in the first stages which appear and are 

 gradually intensified in the later. For example, in the early 

 stages of the leaf of Nephrokpis exaltata, according to Mr. A. A. 

 Eaton,* the leaf's pinnae are crisped and bristly at margin with 

 excurrent nerves, but lose these characters as the leaf becomes 

 mature. In the first stages of the leaf of Polystichum acrostichoides 

 spinulose points on the leaf's margin are absent, but in later 



* See Fern Bull, n: 47, 1903. 



