41 



NOTES UPON SOME NORTHWESTERN CASTILLEIAS 

 OF THE PARVIFLORA GROUP. 



By Merkitt L. Fernald. 



A recent study of large collections of Washington and Oregon 

 plants has made very emphatic the confusion, which has long pre- 

 vailed in Castilleia parvifiora of American authors ; and the attempt 

 to identify very diverse forms of the group has made it necessary to 

 devote more than the usual time to a study of this complex of species. 

 Though there yet remain a number of puzzling forms, which only 

 fresh material and close study can clear up, yet it has been possible 

 to identify nearly all the species, which have been confused with 

 Castilleia parvifiora, and to show; that they are all probably quite 

 distinct from that Alaskan plant. 



The group of species, of which Castilleia parvifiora forms the 

 type, consists of rather low perennials ; the usually simple stems 

 clustered from a caudex, and pilose or pilose-hispid at least above: 

 the upper leaves at least and the bracts mostly cleft (rarely entire), 

 the bracts generally colored crimson or scarlet (sometimes yellowish' 

 or white): the calyx equally or subequally cleft before and behind : 

 the galea somewhat exserted, much longer than the very small 

 lower lip which is not protruded beyond the calyx. 



Castilleia parvifiora was described by Bongard* in 1833 from the 

 Island of'Sitka; and specimens from the original collection of Mertens 

 distributed by Bongard, are deposited in the Gray Herbarium ' 

 Although only one of these specimens shows young flowers it is 

 confidently identified with recent excellent material collected at 

 Disenchantment Bay, Alaska, by Frederick Funston (No. 89)- and 

 from this fresher material it has been possible to secure for exami- 

 nation well-preserved flowers. Castilleia parvifiora is slender with 

 very thin pectinate-laciniate leaves, which dry much blacker than 

 those of related species. From them, too, it is distinguished by it* 

 very short calyx (12 mm. or so long); and from the old species C. 

 angustijoha, C hispida, etc, with which it has been confused' it 



*Meua. Acad. Petersb., ser. 6, ii (J833), 158. 

 Ekythea, Vol. VI, No. 5 [31 May, 1898]. 



