—72— 



rant forms of the three species: 5. cymhifolium, S. medium, S. papillosum 

 which is true of the last only in so far that forms of 5. papillosum with weak or 

 no papillae might very easily be confused with it. That it combines forms of 

 S. palustre {S. cymhifolium) and of 6". magellanicum {S. medium) is the con- 

 clusion which I have also reached and appears to coincide with Russow's orig- 

 inal conception (1887) of the relation between these two species.^ Apart from 

 the fact that Russow's detailed description of 5. intermedium includes no feat- 

 ures that do not belong to the one or the other of the two species in question 

 one need but consult Warnstorf's figures^ of leaf-sections of the Europea 

 species of Inophloea to be convinced of the real nature of his S. sub- 

 Licolor Hampe. In his figure 3 are given three sections (a, b, c) of branch- 

 leaves of S. subbicolor, which may be compared with those preceding of S. cym- 

 hifolium (fig. 2a, b, c, d, e) and those following of 5. papillosum (fig. 4a, b) and 

 S. medium (fig. 5). Certainly nothing but extreme " Haarspalterei " can make 

 any distinction between 3a and the various forms of 2, nor is there a 

 much greater difference between 3c and 5, while that between 3a and 3c is in- 

 finitely greater than either. My conclusion then is,, and all the facts that I 

 have observed seem to confirm it, that S. subbicolor as understood by Warn- 

 storf is an artificial pecies, attempting to embrace intergrading or apparently 

 intergrading forms of two very closely related species. The conception implied 

 in Jensen's name, 6". centrale, applies rather to the aberrant forms of 6". magel- 

 lanicum, which species alone of Inophloea Jensen found in the station from which 

 he first recorded 5. centrale. Such variation of 6". magellanicum involves only a 

 slight thickening of its cell walls at their junction. In the summer of 1904 I 

 had the pleasure of meeting Herr Warnstorf in Neu-Ruppin and expressed to 

 him on that occasion my inability to form any satisfactory conception of 6". 

 subbicolor. He very kindly gave me a local specimen for comparison, but this 

 specimen agrees so closely with 5. magellanicum that one can refer it to it with- 

 out a qualm of conscience, in fact I do not see how one can well dispose of it 

 otherwise. With careful study of their aggregate characters I think these 

 forms can in all cases be referred satisfactorily to one or the other of the two 

 species and that the two (5. palustre and S. magellanicum) are specifically suf- 

 ficiently distinct. Facts of distribution would tend to confirm this conclu- 

 sion, e. g., the fact that 5. magellanicum is widely distributed in South America, 

 where S. palustre (and it may be added, S. subbicolor) does not occur. 



5. pseudomedium Warnstorf, 1891, originally accredited to Central America, 

 its author has now ascertained^ to have come from Australia or New Zealand 

 or thereabouts, where 6'. magellanicum also occurs. 



Our species is widely distributed in North America: from Labrador south- 

 ward to Florida and Alabama, in Michigan and Minnesota, from California 

 northward to Alaska. It is entirely lacking in the West Indies, but occurs 



^ Cf. also Loeske, Ziir Morphologic iind Systematik der Laubmoose 46. 1910. 

 ^ Kryptogamenflora der Mark Brandenburg 1 : 343. 1905- 

 ^ Pflanzenreich 51: 486. 1911. 



