74 REVUE BRYOLOGIQUE 



If 



* 



two genera in different^subfamilies. Climaciiim and Iso- 

 thecium have almost no gametophy te characters in common 

 with tlie other genera, except, of course, such as are 

 common to all the Hypnaceae. In all gametophy te cha- 

 racters Pylaisia and Climachim are as far apart as any 

 two genera of the family (Uypnaceae). The gametophyte 

 cliaracters of Pylaisia show it to be much more closely 

 related to certain species of Uypnum than to any genera 



of the Isotheciae. 



If the number of cilia were constant in each species and 



were there no vestiges of cilia in species in wich they are 



usually lacking, the absurdity would be great enough, but 



as a matter of fact the number of cilia in the endostome 



of the Hypnaceac is known to vary even on the same 



capsule, in Brachytheciinn Fendlen^ which is discribed 



by SuUivant as having cilia solitary and short or none, 



other capsules from the type specimen possess two well 



developed cilia. It is well known that in Pylaisia^ parti- 

 cularly in P. poh/ffniha pseiidO'p/alyf/yria (Klndh.) yes- 



tigial cilia are quite conspicuous. ' In Brachylhecimn 

 Utafiense^ a close relative of />. co I lijiiim, the cs.ps,u\e is 

 erect and sytnmetric and the cilia ve^^tigial as in Pylaisia. 

 No argument is neccessary to show the worthlessness of 

 the erect symmetric capsule as a distinction between fa- 

 milies and subfamilies; the occurrence of capsules of 

 both shapes within the same genus is of too common 

 occurrciice. Another illustration of the anadequacy of 

 sporophyte classification is ihe inclusion of such diverse 

 forms as Etirhynchiitm Stokesii and E. piliferiim in the 

 same genus because both have rostrate opercula- The 

 illustrations given above have been taken from Limpricht's 

 work as he is, perhaps, the latest and best exponent of 

 Schimper's classification. Mr. E. N. Dixon in his admi- 

 rable Handbook of the British Mosses has avoided most 

 of the conspicuous errors of this system. 



The placing of Porotrichum in the Neckeraceae because 

 of the habit and structure of the leaf cells, all the other 

 characters indicating the closest kind of a relationship 

 with the Hypnaceae^ illustrates Lindberg's tendency to 

 give single characters, especially gametophyte characters, 

 undue weight. Another illustration is his p\3ic'mg flyp/i2im 

 Itulhrniciim VVeinm. with Climacium. 



The question of what characters shall be chiefly consi- 

 dered in the classification of the pleurocarpous mosses is 

 not easy to answer. Unlike the zoologists we have little 

 or no aid from geology and must rely wholly on the struc- 

 tural details of existing forms. In our present systems the 

 general habit, the structure and arrangement of the leaf 



