252 6 



which I here call "species", but also in the great power of variation of the latter 

 in almost all possible directions. This variation within the same species has 

 resulted in a great many forms, which may perhaps be ascribed to numerous 

 mutations quickly following after each other and which in the most different 

 manner pass into one another; I therefore (ind it of only small importance to give 

 the single forms special names. This might be justified, if one was enabled to 

 study the species and their forms in the field. By studying dried specimens alone 

 one stands on unsafe ground, especially as most collectors give too little informa- 

 tion as to the quality of the soil, the humidity, shade, etc., which might possibly 

 explain to us in which direction the evolution proceeds, and to what degree the 

 outer factors are of any significance to the variation. I have, therefore, as a rule 

 desisted from establishing forms and varieties under a species, but have confined 

 myself to indicating in what respects the species mostly vary. 



In the following pages I have tried to define 82 species altogether, specimens 

 of which I have seen; still I have no doubt that several more are to be found in 

 tropical America. Of these 82 species, two (D. ptarmica and D. aspidioides) only 

 doubtfully belong to the group, but are included here because they to a certain 

 extent fall within the above given limitation, differing from all the other species 

 however by having stalked pinnæ. The well known D. deltoidea (Sw.) O. Ktze. is 

 not included, as it may upon the whole be placed very near to the group of D. 

 patens, although its lower pinnæ are much reduced yet in a manner unknown 

 within the group of D. opposita. More closely allied to this group are such 

 species as Ü. Leprieurii (Hk.) O. Ktze and D. decussatn (L.) Urban, but because of 

 their having the lower pinnæ not at all reduced I have excluded them. With 

 some others these species might be united into a little group connecting the 

 group of D. opposita with that of D. patens. 



In North America our group is represented by three species: D. novebora- 

 censis (L.) A. Gray, D. oregana C. Chr. (Aspidiiim nevadense Eat.) and the also Eu- 

 ropean D. oreopteris (Ehrh.) Maxon. These three as well as the somewhat more 

 distant relatives D. simulata Dav. and D. thelijpteris (L.) A. Gray, which stand rather 

 outside the limits of the group, I have not included, as they are supposed to be 

 so well known that they can hardly be confounded with other species. 



In the Old World the group is represented by a long series of species, 

 which to a great extent need revision. That any of them should be identical 

 with the American species is not very probable. Still it is likely that in West 

 Africa forms are to be found, which can only with great difficulty be distinguished 

 from Brazilian ones. I have not had material, however, to undertake such a 

 comparison, but I think that no described African species can be united with 

 any known American one. 



A grouping of such closely related species as are being dealt with here, is 

 very difficult. If we have arrived at the conclusion that the characters ascribed to 



