MAXON—STUDIES OF TROPICAL AMERICAN FERNS. 557 
4, POLYPODIUM UNDULATUM Fourn. Mex. Pl. 1: 75. 1872. 
Under this name Fournier published a description of certain plants from 
Mexico, Guiana, and Ecuador, these having rigid, linear, or lanceolate-linear 
pinnatisect fronds, 10 to 18 cm. long, 2 to 2.5 cm. broad, the segments adnate, 
subauriculate, gibbous, undulate, with simple veinlets. The sori are said to be 
4 or 5 pairs to the segment, dorsal upon the veinlets, and borne midway between 
the costule and the margin. The last are characters which clearly exclude this 
species from the group of P. trichomanoides and indicate that it belongs pre- 
sumably with P. pilosissimum and allied species, 
Fournier described also a variety parvulum, characterized by “shorter, 
monosorous pinne,” and cited P. gibbosum Fée as a synonym, with mention of 
Galeotti’s no. 6878 (in part). This association of a monosorous plant with P. 
undulatum, as a variety, is almost certainly erroneous and has led also to the 
mistaken reduction’ of P. undulatum to P. gibbosum, which is itself a critical 
species with monosorous segments. 
Type specimens of P. undulatum Fourn, have not been seen by the writer. 
If these should prove to represent a valid species it must nevertheless bear a 
new name, on account of Willdenow’s earlier description? of a plant from 
Tranquebar as Polypodium undulatum. 
POLYPODIUM FURFURACEUM AND RELATED SPECIES. 
Perhaps no members of the genus Polypodium, taken in its broad 
sense, have received more varied and.at the same time less satis- 
factory treatment than those species with lepidote fronds. For the 
accommodation of the different types more than a few genera have 
been proposed, as, for example, Pleopeltis, Marginaria, Lepicystis, 
and Lopholepis on the basis of venation, habital characters, and 
soriation. From the standpoint of venation alone several flexible 
categories might be recognized, which, with further subdivision on 
habital differences, would result in an excessive multiplication of 
genera. The faultiness of such a treatment is at once obvious from 
a consideration of the large number of intermediate species. . 
Again, it is possible to group the lepidote species under a single 
genus, as Diels* has done, apportioning them among several sections 
according to their venation. The objections to this are several: (1) 
It is scarcely possible to draw an obvious line of demarcation between 
lepidote and nonlepidote species; (2) even if such a separation were 
made the resulting arrangement would be altogether unnatural, 
necessitating the reference of very closely allied species to different 
genera; (3) it can not be admitted that the presence of scales upon 
the lamina in varying profusion is in this instance a character of 
generic importance. In Christensen’s Index we find, for example, 
Polypodium plebejum referred to the subgenus Marginaria, and one 
of its closest allies, P. leucosticton, to Eupolypodium; P. fallax to 
Marginaria, and P. typicum to Eupolypodium. These are very simi- 
lar in their scantily paleaceous surfaces, and their venation is 
20. Chr. Ind. Fil. 572. 1906. 
? Sp. Pl. 5: 155, 1810. 
* Lepicystis. In Engl. & Prantl, Pflanzenfam. 1‘: 322-324. 1899. 
