18 MEMOIRS FROM THE DEPARTMENT OF BOTANY OF COLUMBIA UNIVERSITY 



Pickeringii. It is true that H. argyrocoma differs from the three Horkeliae mentioned by 

 the number of stamens, which is 20 ; but there are two species of true Horkeliae with 

 numerous pistils that also have 20 stamens, viz., H. piorpurascens and H. pinetorum. 

 The characters which have been used to separate Horkelia and Ivesia, viz., the number 

 of stamens and pistils and the dilated filaments in Horkelia, fail altogether. The nu- 

 merous and more or less imbricated leaflets generally found in the Ivesiae, is a charac- 

 ter lacking in I. Baileyi, but found in H sericata and H. Howellii, which are in other 

 respects good Horkelias. I have, therefore, come to the conclusion that it is best to 

 unite the two genera. The characters that have been used to separate Ivesia from Po- 

 tentiUa, viz., the fewer number of pistils in the former, is, of course, of no value. Ivesia 

 pygmaea Gray has very often as many as 20 pistils, while such species as Potentilla 

 Coloradoensis has often not more than 1 2 or 15, and Ivesia Lemmonii, which is a good 

 Potentilla and should be transferred to that genus, seldom has that many.^ There is, 

 however, another character that can serve very well to distinguish Ivesia as well as 

 Horkelia from Potentilla. The character was partly used in the original description of 

 Horkelia, in which the dilated stamens were not referred to. In Potentilla the stamens 

 are inserted in the bottom of the hypanthium, very near the base of the receptacle, and 

 at their bases there is a more or less distinct, annular and somewhat glandular thick- 

 ening of the tissues. In Horkelia (with Ivesia included), the insertion of the stamens is 

 farther out from the center of the hypanthium, so that there is a large open space be- 

 tween the base of the receptacle and the stamens, and there is no trace whatever of a 

 thickening (except perhaps in H. saxosa from Lower California). As Horkelia {Ivesia 

 included) is a genus characteristic of the drier regions of the North American Pacific 

 slope, it is well to keep it separate from Potentilla, even if the distinctive characters are 

 not very prominent. 



LIMITATION OF SPECIES. 

 It will be seen from the following that my opinion as to the limitation of species 

 differs widely from that expressed in our manuals and from that of Dr. Watson in his 

 Revision.^ As there is a tendency to acknowledge many and small genera, so is there 

 also a tendency among botanists to admit many more species than was customary only 

 a few years ago. My conception of a species agrees in the main points so far as Potentilla 

 is concerned with that of Lehmann, as expressed in his Revisio of 1856. In Horkelia 



1 p. Newberryi Gray (/oesta gracilis T. & G.) and P. Arizoniea Greene (/. pinnatifida Wats.), are typical Potmtillae, and 

 were described in the wrong genus. 

 2Proc. Am. Acad. 8 : 549-573. 



