8 Rhodora [January 



his superb work on the Ferns of North America, and, as until recently, 

 generally adopted in the United States, comprises all of the shield 

 ferns with orbicular indusia centrally attached to the receptacle (pel- 

 tate), and those also with cordate reniform indusia attached at the 

 sinus, without regard to the character of the venation except in so far 

 as it may serve the purposes of sectional relations. 



As originally proposed by Olaf Swartz in 1800 or 1801 (Schrad- 

 er's Journal), and more fully established by him in his Synopsis Fili- 

 cum later on (1806), the genus consisted of a heterogeneous collection 

 of species having little cohesion from natural affinities, but represent- 

 ing the best arrangement which had been made up to that time. 



Subsequent revisions, fortified by the views of such eminent pten- 

 dologists as Drs. Presl, Mettenius and Hooker, gradually reduced its 

 proportions to the limitations recognized by Prof. Eaton. 



The genus as thus constituted has consisted of three well charac- 

 terized types, which, while closely related, are readily separable into 

 distinct entities. 



The first of these types is represented by Aspidium trifoliatum 

 Swartz, which stands as the type of Aspidium proper, and which is a 

 species with orbicular, peltate indusia, and anastomose venation. 



The second type is represented by Aspidium Lonchitis Swartz, 

 which, while it has peltate indusia, as in Aspidium proper, has its 

 venation wholly free. Not only this but it possesses additional char- 

 acters of such a nature that many able authorities have treated it as 

 generically distinct in every way from Aspidium. Now it was upon 

 this species that Roth founded his genus Polystkhum somewhere 

 between 1798 and 1800 (exact date uncertain, but the same as that 

 at which he established Athyrium ; see previous note in Rhodora, 

 ii. 267), and Polystkhum Lonchiiis Roth, with its peltate indusia, 

 free venation, and special characters to be described, stands as the 

 type for Roth's genus. 



In considering the history of the early genera it appears that nearly, 

 if not all of them consisted of a heterogeneous medley of species 

 which are now seen to have represented different genera, and this 

 justifies adopting the first correctly named species under the genus 

 as the generic type ; only such other species as conform to such type 

 characters being accepted as rightfully belonging to the genus. 



Following out this principle we find our third type in those mem- 

 bers of the Aspidium genera with a cordate reniform indusium 



