Rydberg : Studies on the Rocky Mountain flora 469 



mentioned should be regarded as such. The short characteriza- 

 tion was, however, drawn principally from my number 2^6^ from 

 Lima, Montana, and this is marked in our herbarium as the type. 

 Since more specimens have been seen, both of the Montana plant 

 and of that from the Black Hills, it has become evident that they 

 are not exactly the same. As the Montana plant is marked as 

 the type, I now limit my G. cephaloidea to it. If Jenney's plant 

 or my 886 were to be regarded as the type of G, cephaloidea, this 

 would become a synonym of G. congesta and the Montana plant 

 should have a new name. As it is, the G. cephaloidea of Brand's 

 monograph should become G. congesta Hooker, and Brand's 

 G. congesta is G. iberidifolia Benth. 



Brand divides the Gilia spicata group in two divisions: one 

 containing G. spicata, G. glohularis and G. trifida, with the corolla- 

 lobes (in dry state) dark purple ; and G. cephaloidea and G. congesta, 

 with lobes of the corolla (in dry state) whitish. The dark purple 

 color is simply due to poor drying. Dr. Brand also describes the 

 corollas of G. spicata as purple. In fact they are greenish white. 

 We have specimens of G. spicata which still retain the greenish 

 white color. Such a distinction is scarcely scientific. 



Under Gilia congesta iberidifolia, Dr. Brand gives as synonyms 

 G. spergnlifolia Rydb. and G. roseata Rydb. Evidently Dr. Brand 

 had not seen a specimen of G. roseata. This is perhaps closer 

 related to his own G. glohularis than to G. iberidifolia, except that 

 the stems are branched and bear several heads. He had seen a 

 specimen of Baker 334, which I referred to G. spergulifolia. When 

 doing so I had in mind only the specimen in the herbarium of the 

 New York Botanical Garden. I do not know by what this number 

 may be represented elsewhere. However, I can not distinguish this 

 from Nelson 5430, which Dr. Brand refers to the var. crebrifolia, 

 evidently not knowing that the var. Merrillii {G. Merrillii A. Nels.) 

 is the original G. crebrifolia Nutt. A duplicate of the type is in 

 the Columbia University herbarium. The synonomy of this 

 group of Gilias is therefore very mixed. In order to straighten out 

 the matter I give the following synonymy: 



Gilia spicata Nutt. Jour. Acad. Nat. Sci. Phila. II. i: 156. 

 1848. 



