Rydberg : Notes on Rosaceae 49 1 



Manitoba: Rapid City, 1896, Macoun 14447 and I 445°>' 

 Fort Ellis, 1906, Macoun & Herriot 68930. 



Saskatchewan: Milk River, 1881, Dawson 34348; The Holes, 

 1885, Macoun 10447; Cypress Hills, 1894, Macoun 4538; Herzel, 

 1906, Macoun & Herriot 69827; Park Bay, 1896, Macoun 14447; 

 Silver City, 1885, Macoun 635 and 7284. 



British Columbia: Slotch-oot-a Lake, 1876, Dawson 7282; 

 Nicola Valley, Macoun 7229; Revelstoke, 1890, Macoun 7287. 



In 1901, while visiting northern Europe, I found in the Botan- 

 ical Garden at Upsala specimens of Potentilla pulcherrima Lehm., 

 which had been cultivated for several generations since Lehmann's 

 time and kept unchanged the characters of pinnate leaves, etc. 

 Side by side were growing also specimens of the so-called P. 

 gracilis from Colorado with its digitate leaves. The latter is of 

 course not the true P. gracilis but the P. pulcherrima of my 

 monograph or P. pulcherrima communis Th. Wolf. Seeing these 

 two forms together, the suggestion came to me that they might 

 not be one species. While collecting in Utah, I found P. pul- 

 cherrima in the same canon where Dr. Watson had rediscovered 

 it, viz., in the Big Cottonwood Canon, southeast of Salt Lake 

 City. Here it was growing alone. Numerous specimens were 

 seen in an open place along the river, but no digitate-leaved 

 specimens were seen. I have come to the conclusion that it is 

 not a parallel case to P. diver si folia, in which the basal leaves are 

 either pinnate or digitate or both on the same plant. In all speci- 

 mens of P. pulcherrima proper, the basal leaves are all pinnate, 

 while in P. pulcherrima communis they are all digitate. In 1901, 

 I described a supposed new species under the name of P. filipes, 

 which Dr. Wolf reduced to a variety of P. pulcherrima. I have 

 found that the characters separating P. filipes and the so-called 

 P. pulcherrima of Colorado do not hold. I therefore united the 

 two into one species under the name of P. filipes in the North 

 American Flora. A depauperate high-mountain form of this 

 species is P. pulcherrima condensata Th. Wolf. 



Professor Nelson in the New Manual of Botany of the Central 

 Rocky Mountains makes not only P. pulcherrima and P. filipes 

 but also P.fastigiata and P. Blaschkeana synonyms of P. gracilis, 

 a consolidation which goes altogether too far. 



