246 194 
var. joinvillensis Ros. Hedwigia 43: 225. 1904. 
Syn. Dryopleris joinvillensis Ros. Hedwigia 46: 120. 1907. 
Sta. Catharina: Joinville, ScaMarz nr. 100 (R). 
A very large form: pinne 25 x 4 em; veins 18—20 to a side. Rachis gemmi- 
ferous. Pubescence about as in the former variety, but stellate hairs fewer and 
scales more numerous. 
var devolvens (Bak.) — Fig. 39. 
Syn. Nephrodium devolvens Bak. Journ. of Bot. 1885: 217. 
Dryopteris devolvens C. Chr. Ind. 261. 1905. 
Differs from the type by 1) size: lamina 3—4 dcm long; pinne 12—15 cm 
long, 2'/2 em broad, 2) the basal pinne being distinctly reflexed, 3) lamina up- 
wards suddenly narrowed into a broad, pinnatifid apex, 4) segments close, even 
contiguous, acute, subfalcate; veins 10—12 to a side. In pubescence it does not 
differ materially; still the scales of rachis and coste beneath are very few, and 
the rachis and costae beneath are clothed with fine, simple hairs besides the stel- 
late hairs. The upperside is hairy towards the margins, the underside finely 
downy by simple hairs. Veins not prominent. 
A distinctly looking variety, well-marked by its close, acute segments and the 
upwards suddenly narrowed lamina, but I find it impossible to distinguish it spe- 
cifically from D. lugubris. Several of the specimens referred to the type show now 
one now another of the characters, which mark the variety. 
The following specimens are rather uniform: 
Rio: Graziou nr. 15766 (B, H, type-number) — Minas Geraes: Caldas, Mosén nr. 2156 (Rg, Hb. 
Brux.) — Sao Paulo: Piritabo (?), F. W. BauER nr. 49 (R) — Capivary, GEnpER nr. 88 (R) — 
Campinas, HEINER nr. 568 (Rg) — Toledo, UrnsnicuTr nr. 51 (Rg; proliferous) — Matto 
Grosso: Santa Anna da Chapada, MALME nr. 2131 (Rg). 
Synonyms of D. lugubris are probably: 
Aspidium coadunatum Klf. Enum. 239. 1824 and 
Nephrodium inaequale Schrad. Gott. gel. Anz. 1824: 869, both from the vicinity 
of Rio. A description of the former by MerTTEntus is to be found in MS in (B) 
and from this description it seems to be identical with var. devolvens. The name 
of KaAurruss has priority, if his species is a form of our D. lugubris, and I have 
no doubt that this is the case. We have thus here a common Brazilian species, 
which is described as new at least 6 times, viz 1824 by Kaurruss (coadunatum) 
and SCHRADER /inaequale), 1858 by METrTENIUs (lugubre/, 1869 by FEE /quadrangu- 
lare, 1885 by Bakrn /devolvens and 1907 by Rosenstock /joinvillensis/. This 
shows how necessary it is to compare specimens with descriptions and authentical 
specimens of described species before describing them as new. 
Aspidium nitidulum Kze.; Ettingsh. Farnkr. 188 tab. 123 f. 3. 1865 is perhaps 
the same. 
