﻿FERN ANCESTRY AND ANGIOSPERM ANALOGIES. 245 



Cycadeoidea; belonged, as in the case of every highly organized plant type, 

 presented members of infinite variety. Also, in the ease of the sole remote type of 

 which we have now gained a fortuitous knowledge, striking analogies to living 

 angiosperms are suggested, no difference whether, laying histological structure 

 somewhat aside, we fasten our attention upon one set of characters and Liriodendron 

 be called to mind, or upon another with the result that the male and female 

 catkins of Amentacea.' first suggest themselves, or upon a third set that call to 

 mind some other hint of characters that must have been present in the countless 

 members of a great proangiosperm complex, just as the monocotyl Panda/ins 

 thus suggested itself to Saporta. We should not ask too much of isolated 

 evidence, nor yet be content with a scant interpretation of highly suggestive facts. 

 And taking up the general question not from the position of organs so much as 

 their general morphology, what right would we have, in surmising as to the homol- 

 ogies between living plants and hypothetical primitive angiosperms, to look for 

 instance upon the bifurcate stamens of Ginkgo as separated by an unbridgeable 

 hiatus from those of the multibrauched stamens of Ricinus ? Assuredly none. For 

 the purposes of broader generalization, fern-like fronds upon which were doubtless 

 borne the pollen of Lyginodendron, the staminate fronds of Cycadeoidea of true 

 Marattiacean type, the mega- and microsporophylls of Cycas, the stamens of Cor- 

 dages and Ginkgo, and finally of Ricinus and Liriodendron, all belong to a series. 

 Nor is there from a plain point of view an unbridgeable gap between the stami- 

 nate disk of Cycadeoidea and that of Uehcitsc/iia, for the latter could arise similarly 

 to one of the hypothetical one-seeded and bisporangiate forms of the cycadeoidean 

 alliance by one of the simplest of all evolutionary processes, namely, increased number 

 of flowers to the plant and decrease in bulk until there was left of each original frond 

 but a single filament bearing a pair of pollen sacs (as in Ginkgo biloba), and finally 

 but a single pollen sac, the filaments retaining the original cyclic arrangement of 

 the fronds from which they were derived. 



In discussing this question we are forced to the conclusion that it is clearer than 

 ever that there is a universal homology betweeu the mega- and microsporophylls 

 of the spermaphyta and the asexual spore-bearing fronds of the ferns, from which 

 it becomes more and more probable that all the spermaphytes of the existing flora 

 have sprung. And for the sake of a clearer conception of the views proposed it 

 has seemed preferable to speak of the stamens or earpophylls of particular plants 

 rather than to attempt some completer form of generalization. The series of plants 

 named is not asserted to be an evolutionary one, except in the most general sense. 

 Other forms in the direct relationship yet remain to be discovered. 



It is, then, in Cycadeoidea, and especially in the persistence in such highly 

 organized plants of the Marattiacean synangium, that we get the first unmistakable 

 hint of the nature of angiosperin evolution. And the further view would seem to 

 be abundantly justified— that while the staminate disk surrounding the ovulate axis 

 of Cycadeoidea indicates primarily an evolution terminating, so far as now possible 

 to trace, in the gymnosperms, the juxtaposition of parts is exceedingly suggestive 

 of the possibility, if not the manner as well, of angiosperin development directly 

 from filicinean forms. 



