314 68 



tries of tropical America. I refer hereto only following few specimens, as I have 

 not had the specimens in HB and HS. 



Mexico: Baranca de Jovo et Baranca de Mirador, Liebmann (HH). 



Costa Rica, H. Pittier n. 323 (HC). 



Columbia, Moritz n. 241 (HH = Lept. rupestris Kl.). — Santa Marta, H. H. 

 Smith n. 1005 (pt.) (HC). 



San Domingo, Thouin (HH). 



Guadeloupe, L'Herminier (HC = Ph. Diichassaigniana Fée). 



The species may be distinguished from the allied species with linear sori by 

 (1) the leaf being gradually and shortly narrowed towards its base with only 2 — 3 

 pair of distant, reduced pinnæ, (2) more pairs of pinnæ below the apex of the 

 lamina being entire or nearly so, (3) medial pinnæ incised only half-way down to 

 the midrib, rarely a little more, (4) veins only 6 — 7 to a side, the basal pair run- 

 ning out in sinus. 



Sori placed above the middle of the vein, as a rule linear. The whole plant 

 more or less shortly pubescent by hamate hairs. Segments obtusely rounded, 

 c. 4 mm. broad. 



Lept. rupestris Kl. is for me the typical form. Some of the specimens from 

 Columbia leg. H. H. Smith are intermediate between this species and D. aspidioides, 

 which generally is smaller and has the larger pinnæ distinctly short-stalked. 



var. brevisora Rosenstock, Hedwigia 46 : 134. 1906. 



Syn. Nephrodiam Kaulfussii Lindman, Arkiv for Bot. 1: 225 tab. 10 fig. 10. 1903. 



1 Gymnogramme oppositans Fée, Cr. vase. Br. 1: 58 tab. 14 flg. 1. 1869. 



This variety is evidently a very common fern in southern Brazil. Besides the 

 specimens mentioned by Rosenstock 1. c. I have examined the following. 



Brazil: Minas Geraes, Lagoa Santa, E. Warming n. 796 (HH). — Serra de 

 Ouropreto, Schwacke n. 10233 (HC).— Caldas, Mosen n. 2162, 2164 (HS), 2163 (HS. 

 HH). — Rio de Janeiro, Corcovado, Mosen n. 2698 (HS). — Matto Grosso, H. Smith 

 n. 99 (HC). — San Paulo, Campinas, A.E.Severin n. 23 (HS). — Rio Grande do Sul: 

 Porto Alegre, Lindman n. A. 379 (HS). — Hamburger Berg, Lindman n. A. 539 (HS). 



Lindman's flgure cited above illustrates very well this well-marked variety, 

 which most probably is a distinct species and the same as G. oppositans Fée, 

 which name if so has priority. Besides the differences from true D. diplazioides 

 mentioned by Rosenstock: the nearly round sori — mostly only the basal ones 

 a little elongated — and the leaf nearly always proliferous, which is also some- 

 times the case in the typical form, I may add, that stem and rachis are as a rule 

 more robust and the texture of the lamina more firm, sometimes nearly coriaceous. 



59. Dryopteris consimilis (Fée) C. Chr. — [Fig. 37]. 



Syn. Gymnogramme gracilis Hew. ß G. consimilis Fée; Bak. Syn. Fil. 377. 

 1868; Gymnogramme consimilis Fée; Jenman, Bull. Dept. Jamaica 4: 203. 1897. 



Type from Guadeloupe (non vidi). After Jenman's description I identify 



