The Genus Riccardia in Chile. 205 



ing to current views dorsi ventral ity is not a primitive feature but 

 is derived from a condition of radial symmetry. Among the 

 bryophytes many of the mosses show radial symmetry clearly, not 

 only in their sporophytes but also in their gametophytes, but this 

 condition is exceedingly rare in the gametophytes of the Hepaticae. 

 It is found in a typical form only in the two closely allied genera 

 Calohryuni and Scalia, the relationships of which to the other 

 Hepaticae are still in doubt. There are certain other leafy genera, 

 however, which approximate a radial condition very closely. In 

 Herbcrfa, for example, the large and deeply bifid underleaves are 

 essentially like the leaves, and the same thing is true of other genera 

 of the Ptilidiaceae. An approach to radial symmetry is to be 

 observed also in the reproductive branches of such genera as 

 Cephalozia, Bassania and Lcpidosia, while the flagelliform 

 branches of the last two genera with their minute scale-like leaves 

 are often strikingly radial in appearance. 



It is possible that the types which have just been considered 

 may give some idea of the primitive leafy hepatics. At any rate 

 they show a closer approach to the radial mosses than do the more 

 definitely dorsiventral types. In the Riccardiaceae, however, 

 approximations to a radial condition seem to be derived second- 

 arily from dorsiventral types. We find such approximations in 

 the wingless and subterete rhizomes of Hy mono phy turn and the 

 dendroid species of Pallavicinia and Synipliyogyna; in the stalks 

 of the photosynthetic branch-systems in the same forms ; and in 

 the axis and stolons of certain species of Riccardia, such as R. pre- 

 hensilis and R. TJiaxteri. In all these cases the subterete axes 

 form portions of highly dififerentiated thalli and play little or no 

 part in photosynthesis, but the photosynthetic branches are still flat 

 and dorsiventral and thus retain the features of the undifferentiated 

 type of thallus from which the others have probably been derived. 

 There are apparently no subterete axes in any of the thalloid 

 Jungermanniales that are clearly the phylogenetic predecessors of 

 the flattened axes. 



The phylogeny of the sporophyte in Riccardia is, if possible, 

 even more of a mystery than that of the gametophyte. Goebel 

 inclines to the idea (11, p. 544) that the sporophyte of Aiifhoccros 

 approaches the primitive condition more closely than that of any 

 other hepatic, basing his opinion largely on the fact that the sporo- 



