22 



ARKIV FÖR BOTANIK. BAND 1. 



Geraes (Regnell, n. III, 1490), and my own specimens from 

 Paraguay. In Hook, and Bak. Syn. Fil., ed. 2, p. 77, the species 

 (»pusillum», non Sw.!) has this area: from Mexico and the West 

 Indies southwards to Brazil and Peru, and Natal in S. Africa. 

 The illustrations above quoted and the specimens I have 

 obtained, prove that Tr. cpicrcifolmm is of irregular and va- 

 riable shape, but also 

 that certain marks are 

 invariable, at least in 

 most of the fronds in 

 a tuft, viz.: 



the size is modera- 

 tely large (frond 3 — 5 

 cm long, or a little 

 more); fig. 7, A and D; 

 the fertile frond is 

 in typical cases gra- 

 dually narrowed to- 

 wards the base and 

 then widened to the 

 top, the greatest width 

 being above the middle ; 

 the young frond is 

 also very gradually nar- 

 rowed at the lower 

 part, the young blade 

 most often linear-spa- 

 thulate; 



the fullgrown frond 

 is deeply pinnatifid, 

 the lower segments ge- 

 nerally short, broad, 

 and somewhat narrow- 

 ed towards the point, 

 and the rachis with a 

 broad wing (1—2 mm) on each side; the upper segments often 

 more extented, linear or lingulate, fig. 10, A, bearing the 

 involucre in the emarginate end; 



the segments, or primary lobes, are separated by a rather 

 narrow and acute sinus, and their margin is slightly sinu- 

 ated or toothed, but not deeply' pinnatifid; 



Fig. 10. Trichomanes quercifolium Hook, ct 

 Grev. (X 5), A the middle of a trond from Para- 

 guay, Exped. 1. Regnell. n. A, ::i207 (see also 

 fig. 7, I)), showing broad segments in the lower 

 part and produced narrow segments in fhe toj): 

 B specimen from Mirador in Mexico, collected by 

 Liebmann (Herb. Stockholm), two fronds of dift'e- 

 rent shape as to the base, both from the same 

 rhizome. 



