638 MINNESOTA BOTANICAL STUDIES. 



tions above the general surface of the spore can be detected, 

 there may yet be observed almost the usual appearance when 

 the spore is examined in face view. Instead of papillae, their 

 places seem to be supplied by translucent dots. Now the most 

 interesting outcome of this study is the observation that the 

 European specimens, while having what are always rated as 

 smooth teleutospores, yet show when looked at dry and in face 

 view, the same appearance of translucent dots and in the same 

 abundance and distribution as do American specimens. The 

 American form, therefore, simply accentuates characters that 

 are primitive in the trans-Atlantic form. 



The uredospores also have interesting, but less significant, 

 characters showing variability. As a rule they do not much 

 exceed in size those from European specimens, although the 

 tendency toward largeness is apparent. But in many American 

 specimens the walls are greatly thickened (compare uredospores 

 in Figs. 5 and 7), and give a striking appearance under the 

 microscope. These thick-walled uredospores are sometimes 

 small, and sometimes large. They occur on various species of 

 violets and range from the Atlantic coast to the Pacific. Fine 

 illustrations are found in material on Viola -primulcBfolia col- 

 lected at Isle au Haut, Me. (Fig. 8), the uredospores being 

 small, and on Viola glabella collected at Bingen, Wash. (Fig. 

 7), the uredospores being large. After all, the form is only oc- 

 casionally met with. If it is an adaptation to some particular 

 environment, it is difficult to see what that may be. The form 

 on Viola -primula/oHa from Maine was collected within a few 

 hundred feet of the open ocean among rocks, yet in the same 

 situation and intermixed grew Viola lanceolata with rust show- 

 ing uredospores having almost normally thin walls. 



It is possible that these interesting variations belong in some 

 way to obscure species, but our study has shown no morpho- 

 logical boundaries. It is more likely that they indicate races, 

 or possibly so-called biological species. A well-directed series 

 of cultures would undoubtedly yield important results. 



A few words regarding the American synonymy may be 

 helpful. Cooke described Puccinia hastattz in the third volume 

 of Grevillea from material collected in Maine by E. C. Bolles. 

 The host was Viola hastata. The uredo and teleuto stages are 

 described, but the only distinctive characters are the measure- 

 ments which are given as 20-22 // for the uredospores, and 



