Bergman : comments on malva rotundifolia l. 439 



The characters here enumerated, for the most part, are to be 

 found in descriptions as given in European manuals. These have 

 been checked by careful study of a large number of specimens from 

 dififerent parts of the world. 



Referring to American manuals, we find in Gray's Manual, 7th 

 edition, that M. rotundifolia is the only species mentioned. Britton 

 & Brown, Illustrated Flora, ed. 3, add M. parviflora. 



Specimens of M. borealis have been collected from New Jersey, 

 Pennsylvania, and IMichigan as indicated below, and it seems that 

 this species must have been introduced and have become established 

 at other points in the east. A single specimen, too young for cer- 

 tain identification, but apparently M. borealis, was collected at Ames, 

 Iowa, by J. R. Campbell. M. borealis occurs in the Red River Val- 

 ley in North Dakota and probably also on the Minnesota side. M. 

 borealis has been collected at other points in North Dakota, as has 

 also M. parviflora. From this it is evident that both manuals used 

 generally in northeastern United States should include all three 

 species. 



In the Rocky Mountain region, Rydberg, in the Flora of Colo- 

 rado, includes M. rotundifolia and M. parviflora, while Nelson, in 

 the Nezv Manual of the Rocky Mountains, describes only the 

 former, although Colorado is within the region covered. Both M. 

 parviflora and M. borealis occur within the range of Nelson's man- 

 ual, and should be included therein. The confusion of M. borealis 

 with M. rotundifolia may account for the failure to include the 

 former. M. borealis occurs in Montana, Wyoming, and Colorado. 

 M. parviflora has been collected in Wyoming, and is reported also 

 from ^Montana in Rydberg's Flora of Montana. 



References to the manuals of the Pacific Coast region reveal 

 some interesting things. Brewer and Watson in the Geological Sur- 

 vey of California, 83, describe M. borealis Wallm., but it is apparent 

 from the description that the plant described is not M. borealis of 

 Wallman but M. parviflora L. The erect habit, enlarged calyx, and 

 transversely rugose-reticulate carpels are peculiar to the latter 

 species. M. borealis is seldom erect, the calyx does not become en- 

 larged in fruit, and the carpels are not transversely rugose- 

 reticulate. 



Howell, in the Flora of Northzvesfern North America, 100, 



