﻿246 RELATIONSHIPS. 



In more fully describing the micros porophylls of Cycadeoidea, in 1901, I was 

 led, in speaking of the new light shed by the investigation of the American fossil 

 cycads on the nature of the evolutionary course that had culminated in seed repro- 

 duction, to call attention to the hypothetical position occupied by Lyginodendron 

 and its allies as follows (192) : 



* The most important possibility suggested by the staminate fronds 

 above described is as to the character of fructification in the cycadofilices. If we are 

 permitted to imagine, as we surely are, a plant, vegetatively like Lyginodendron, and 

 either monoecious or dioecious, with microsporophylls like those of the staminate fronds 

 of Cycadeoidea, and megasporophylls like those of Cycas, do we not picture an ancestral 

 form which almost beyond doubt existed and may any day be found? * * * It 

 certainly becomes more than ever probable that the dimorphism of various Paleozoic 

 plants usually referred to the ferns is intimately connected with forms of heterospory 

 and the acquiring of the seed habit by types immediately ancestral to the cycads, if not 

 also to the angiosperms." 



This remark has been fully justified. Nearly two years later Oliver and Scott 

 (106) reinvestigated the cycad-like seeds described by Williamson under his genus 

 Lagcnostoma and proved on the basis of isolated similar parts that seeds of L. 

 Loniaxi are really the megaspores of " the Paleozoic quasi-fern " Lyginodendron 

 Oldhamium. And since then evidence has been rapidly accumulating that the 

 majority of the Paleozoic plants with filicinean foliage were either heterosporous, or 

 actually bore pollen and well-developed seeds of cordaitean or cycadeau type. In 

 this country the seeds of Ancimitcs, in England those of Nenroplcris, in France 

 those of Pecopteris, have been found in quick succession; while the identity of 

 Trigonocarpon with Meditllosa is all but established. Moreover, as these pages 

 go to press, Kidston has determined the pollen-bearing fronds of Lyginodendron 

 Oldhamium to be of the Crossotheca type. It is thus clear that whatever the 

 connections of the advanced condition of floral development of the Cycadeoideae 

 with angiosperm evolution, among the most fundamental organs involved in the 

 origin of seed-bearing plants yet discovered is the staminate frond as first demon- 

 strated in the silicified trunks from the Black Hills. Previously the bearing of 

 pollen or of microspores, incipiently pollinial in nature, by fronds yet retaining a 

 more or less filicinean structure, was only hypothetically involved in various attempts 

 to unravel the manner in which the origin of heterospory had taken place ; although 

 the ferns of Marattiaceous type were always first looked to, because of their great 

 abundance in ancient and primitive florae taken together, more especially with the 

 presence of certain vegetative and fruit characters in the existing cycads, testifying 

 more or less directly to an ancient relationship with such tree-fern types as Angi- 

 opteris. But undeniably the evidence morphologically and chronologically of most 

 weight in the raising of this hypothesis to the dignity of a definite probability was 

 afforded by the fact that pollen-bearing structures so nearly identical to those of the 

 asexual spore-bearing syuangia of Marattia as to be in themselves at most no more 

 than generically separable, and preclude all reasonable possibility of homoplastic 

 development, should actually persist in highly organized seed-bearing plants of the 

 late Jurassic and Lower Cretaceous. 



