The genus Ceratopteris; a preliminary revision 



Ralph Curtiss Benedict 



The genus Ceratopteris comprises a complicated group of hy- 

 drophytic leptosporangiate ferns which vary considerably not only 

 in leaf-form, habit, etc., but also in the more fundamental charac- 

 ters of the sporangium, but which for nearly a century have com- 

 monly been regarded as constituting but a single species. Re- 

 cently, however, Hieronymus,* and Underwood t have dissented 

 from this view and have recognized a second species, but without 

 attempting any revision of .the genus as a whole. The present 

 paper has as an object the presentation of such a revision. Owing 

 to insufficient material the revision is necessarily incomplete, but 

 it is believed that it represents a step in advance, and will aid in 

 the further study of these plants. 



In reaching conclusions as to the taxonomic status of the various 

 forms, it has been necessary to examine microscopically the spo- 

 rangia of a large amount of material from many localities. This 

 has resulted in the discovery, of some general morphological inter- 

 est, that the number of spores per sporangium varies with the 

 different species. Also of interest is the establishment of the fact 

 that this variation, and the variation in the development of the an- 

 nulus, can be correlated with the variation in leaf-form and that the 

 genus consists not of a single species, anomalous both in variation 

 and distribution, but oi several species whose variations do not 

 exceed reasonable h'mits, and whose distribution, although possibly 

 anomalous in one case, is for the most part comparable with that 

 of the great majority of species. This correlation is furthermore 

 of interest as additional evidence, if additional evidence is needed, 

 as to the value of the study of minute characters m connection 

 with taxonomic work with the ferns. The synopsis finally 

 obtained seems clear enough now, even with respect to the grosser 

 differences in leaf-form, but these differences became clear only 



*Bot. Jahrb. 34 : 561. 1905. 

 t Torreya 7 : 195. 1 907, 



463 



