RvDiJERG : Notes on Rosaceae 40.') 



//. tiiiuisecta was better known, but the auth(jr mistook it for 

 //. teiu-lla (//. fusca tcnclla S. Wats.). A reexamination of the type 

 of the latter proved it to belong to a species which the writer 

 thoutjlit un described. //. ttnuisccta is therefore the same as //. 

 tenclla, in greater part, of the writer's monograph. To it belong 

 the following specimens : 



Wa.shington : Falcon Valley, July 28, \'^)S2, Suksdorf 24^2 

 (type), and 1896,^0; Ice Cave, Trout Lake, Aug. 5, 1894, /''. 

 E. Lloyd. 



To the true //. toiella belong the following : 



California: San Joaquin River, l-'resno Co., 1891, Coville 

 & Funston iSj6 ; Hogg Ranch, Yosemite National Park, 1902, 

 Hall & Babcock 3378 ; Morgan, Tehama Co., 1903, .^jp^/ Can- 

 nell Meadows, Tulare Co., 1904,57/7. 



Horkelia hispidula is related to H. sericata but lacks the silvery 

 pubescence characteristic of that species. In habit it resembles 

 H. Micheneri also and may easily be mistaken for it, if the floral 

 characters are overlooked, the sepals and bractlets being very dif- 

 ferent. H. hispidula is known only from the type locality. 



H. Broivnii is related to H. tenella and H. pati'iflora. It differs 

 from the former in the dense pubescence and from the latter in the 

 deeply divided leaflets and the open inflorescence. To it is to be 

 referred the following specimen, besides the type : 



California : Mt. Shasta, 1892, E. Palmer 2448a. 



H. integrifolia is related to H. tridcntata but differs in the entire 

 leaflets and the appressed pubescence of the stem and the petioles. 

 It is known only from the type locality. 



H. pnlchra is related to H. congesta but the leaflets are 13-17 

 instead of about 9 and they are deeply cleft into lanceolate divi- 

 sions. It also is known only from the type station. 



One change of name was necessary. H. beniarditia Rydb. 

 was substituted for H. Parryi Rydb., there being an older H. 

 Parryi Greene. H. Parryi Rydb. was based on H. Bolanderi 

 Parryi S. Wats. It may be that Dr. Greene had the latter in mind, 

 when he proposed the new species, but there is no evidence that 

 he did, for he cited no synonym, nor in any other way referred to 

 Watson's variety. The types of both H. Bolanderi Parryi S. 

 Wats, and H. Parryi Greene were collected by Parry in southern 

 California, but at different localities and in different years. 



