— 29 — 



the former, for it appears again^ as Astomum ahhreviatum (Mitt.) Fleisch. from 

 Java. Fleischer^ did not however report it from Java, but included it as to be 

 expected in Java on the basis of its occurrence in Ceylon, outside of which island 

 it has not, so far as I know, been found. I note also a species, Astomum chilense, 

 recently described by Williams^ which is so close to Phascum recurvirostrum 

 C. M.^*^ from Paraguay placed by Paris in Hymenostomum that even its specific 

 validity is questionable, as Mr. Williams was inclined to admit when I called his 

 attention to the other species. Roth had transferred this species to Astomum 

 and figured it.^^ It should be added that Brotherus has not included it at all 

 and that Paris^^ apparently incorrectly represents it as identical with Hymeno- 

 stomum Balansaeanum Besch. The confusion is doubtless due to the fact that 

 both were collected by Balansa. H. Balansaeanum is understood to be one of 

 the more highly developed gymnostomous so-called Hymenostomums and has 

 recently been figured as such by Felippone from new material collected by him- 

 self and evidently determined by Brotherus.^^ It may be noted in passing that 

 A. recurvirostrum and A. chilense together with some other South American 

 Astomums differ considerably in gametophyte characters from the uniform type 

 of the Astomums of the northern hemisphere and are perhaps best regarded as 

 cleistocarpous forms more immediately related to so-called Hymenostomums 

 and Trichostomums of their general region. The case as between Hymenostomum 

 and Weisia is perhaps not quite so bad, but European bryologists speak of the 

 difficulty of separating Weisia crispata (N. & H.) C. M. and Hymenostomum 

 tortile (Schwaegr.) B. & S.^* Certainly there is no sharp generic division even 

 with these last two species excluded from consideration. They are themselves 

 in something of a "twilight zone" toward Trichostomum. 



On the other hand there are some mosses, especially from the tropics, com- 

 monly included with Hymenostomum (so also by Brotherus) which are not directly 

 related to Weisia-species, but form the gymnostomous relatives of species ordi- 

 narily placed in Trichostomum Hedw., as is being more and more recognized.^^ 

 In a natural grouping these should then either be excluded from Hymenostomum 

 or the genus should be extended to cover at least some of the species now assigned 

 to Trichostomum. The hybridization of Trichostomum fiavovirens Bruch with 

 Astomum crispum (Hedw.) Hampe detected by Nicholson^'^ shows a close relation- 

 ship which might more or less justify the drawing of the broader generic lines 

 last suggested, which would further involve the inclusion of Tortella (C. M.) 

 Limpr. It is not however my intention to carry the discussion further in this 



'Op. cit. (Nachtrage), 1189. 1909. 

 ^Flore de Buitenzorg, V, I, 315. 1902. 

 sBull. Torrey Hot. Club, XLII, 393. 1915. 

 "Flora, LXXI, 5. 1888. 



"Auszereuropaisch;. Laubmoose, f, 195. 191 1. 

 i2Index Bryologicus, ed. 2, II, 356. 1904. 



i^Contribution a la Flore Bryologique de I'Uruguay, I, 17. 1909. 



iiCf. Dixon, Handbook, ed. 2, 230. 1904; L'mpncht in Rabenhorst, Kryptogamenfl 3ra, 

 IV, I, 230. 254. 1886. 



i^Cf. Fleischer, Flore de Bu'tenzorg, V, I, 313!'. 1902. 

 isRev. Bryol., XXXVII, 23f. i9io. 



