446 



THE AMERICAN NATURALIST [Vol. XL V 



It is, however, by no means impossible that some of the earliest 

 known ferns, the Botryopteridea?, may have been related to the 

 Ophioglossacea\ Both the form of the leaves, and the sporangia 

 which were borne on special leaf segments, are suggestive of the 

 Ophioglossaoeii*, and there are also certain anatomical resem- 

 blances. 



One of the earliest fern-like fossils is the Devonian genus 

 ArcJueopteris. This fossil in the venation of the leaves suggests 

 the simpler types of linlnjrh i 11 m . and the sporangia are borne on 

 special leaf segments, which, however, it must be said more 

 nearly resemble Osmunda than they do Botrychium. Professor 

 Seward is inclined to believe that the sporangia of Arch<t opl< ns 

 are really pollen-sacs of a Pteridosperm, stating that they are 

 much larger than the sporangia of any known fern, being two or 

 three mm. in length. It is evident that Professor Seward over- 

 looked the Ophioglossaceffi in making this comparison, and it is 

 with these that the comparison really should be made. The 

 sporangia of Archaopteris are described as pear-shaped sacs, 

 two to three mm. in length. These are nearly equalled in size by 

 some species of Botrychium, such as B. Lunaria and B. silai- 

 folhnn, in which the globular >p..raneia may be 1.5 mm. in diam- 

 eter, while the sporangia of the large species of Ophioglossum 

 very much exceed in size these figures. In 0. pendulum the 

 sporangia are probably larger than those of any other living 

 Pteridophyte, and may reach a diameter of four millimeters. 

 It is clear then that the mere question of size is not a valid argu- 

 ment for considering .1 rcha opl< ris a Pteridosperm rather than 

 a homosporous fern. 



The evidence of the fossil record entirely bears out the conclu- 

 sions based upon a study of the living ferns that the condition 

 in which the sporophyll, or parts of it, are entirely devoted to 

 spore-production, as in Ophioglossum and Osmunda, is a more 

 primitive condition than that in which the sporangia are pro- 

 duced upon the backs of unmodified leaves. 



There is abundant evidence from a study of existing Arche- 

 goniates that the sporophyte of the fern is the result of the 

 elaboration of the sporogonium of some bryophytic ancestor. 

 This being the case, it necessarily follows that the sporophylls 

 are older phylogenetically than the sterile leaves, and are not 

 secondary modifications of the latter. It is to be hoped that stu- 

 dents of the Botryopteridea? and other archaic fern types will 

 make a thorough comparison of these with the existing Ophio- 

 glossaceae, in the light of the most recent developmental studies 



