96 Rhodora [June 



tains, New Hampshire, A. W. E.\ Crawford Bridle Path, White 

 Mountains, G. E. Nichols. These stations are all near or above the 

 timber line. As Miiller points out the deep purple color is the only 

 important feature distinguishing S. Joergensenii from S. nemorosa, 

 and this feature by itself seems insufficient to justify a specific separa- 

 tion, 



The second segregate, S. Austin-it', 1 was based on No. 18 of Austin's 

 Hep. Bor.-Amer., distributed as " Seapania nemorosa, var. 3" and 

 described as very common "in shady places, on rocks and on the 

 ground," no definite localities being cited. This specimen is referred 

 to S. nemorosa without question by Miiller, but Warnstorf criticises 

 him for doing so and compares his S. Austinii with S. curta (Mart.) 

 Dumort., rather than with S. nemorosa. The plant under considera- 

 tion is frequent in southern New England, where it grows on shaded 

 banks in woods, and at first sight looks very different from the typical 

 form of S. nemorosa on moist rocks. It is characterized, according 

 to its author, by its smaller size, by the small number of teeth on the 

 leaf-lobes, and by the absence of wings on the keels. In the writer's 

 opinion these features are associated with unfavorable environ- 

 mental conditions and S. Austinii represents a juvenile condition 

 of S. nemorosa, in which certain pecularities of the species fail to 

 manifest themselves. It may be added that inconspicuous keels 

 are occasionally present and that the gemmae, which are abundantly 

 produced, are yellow, pyriform or elliptical, and unicellular, thus 

 agreeing with those of S. nemorosa rather than with those of S. 

 curta. Warnstorf 's species, therefore, should be regarded as a synonym 

 of S. nemorosa or perhaps as a variety. 



The third segregate, S. recurvifolia, 2 was based on another specimen 

 distributed by Austin in his Hep.-Bor.-Amer. This specimen is 

 No. 10 and was designated "Seapania nemorosa var. 1." It was 

 described as common on the "margins of rivulets, swamps, &c," 

 and was likewise referred to S. nemorosa without question by Miiller. 

 Warnstorf admits the close relationship of his species to S. nemorosa 

 but emphasizes, as differential characters, its recurved and entire 

 dorsal leaf-lobes and its thin-walled leaf-cells with poorly developed 

 trigones. Here again the distinctive features, even if they were 

 constant, are very slight and appear to be associated in some way 



1 Hedwigia 63: 79. 1921. 



2 lledwigia 63: 115. 1921. 



