178 The Ottawa Naturalist. [Noverrber 



are of special interest in reg"ard to the supposed synonyms, and 

 Pallas' work (1. c.) gives an excellent description and figure of his 

 G. ciliafa, which is known now as G. barbata Frcel. A very com- 

 prehensive treatise of the various sections of Gentiana is presented 

 by N. Kuznezow in Eng-ler and Prantl's Natuerliche Pflanzen- 

 familien. But as we have stated above, the diagnosis of G. serratu 

 Gunn, does not seem to have been fully appreciated, and more- 

 over there are some salient points in its floral structure which 

 have not been mentioned by Scandinavian authors. The diagnosis 

 may be written as follows : 



Gentiana serrata, Gunn. 



Annual or biennial, glabrous ; stem erect, quadrangular, 5 to 

 16 cm. high, branched from the base: leaves mostly crowded near 

 the root, obovate-lanceolate or the upper linear-lanceolate, acute : 

 peduncles long and quite stout, i -flowered : calyx about 2 cm. 

 long, unequally cleft to near the middle, 4-lobed, the longer lobes 

 lanceolate, the shorter ovate, all acuminate with membranaceous 

 margins, but none carinate : corolla deep blue, 3 to 4 cm. long, 

 4-lobed, cleft to about ^3 of its length, the lobes nearly erect, 

 oblong, erosely denticulate across the obtuse summit, mostly 

 without lateral fringes and destitute of basal nectariferous glands*: 

 stamens 4 with slender filaments : ovary fusiform, stipitate with 

 an almost sessile 2-lobed stigma : mature capsule longer than the 

 corolla: seeds scabrous from short papilla. 



Said to bloom in July or August, and has been collected on 

 the sea-shore of Norway from 66° 10' to 70° 50' N. lat., and on 

 the west coast of Greenland at 61^ N. lat., where Vahl first col- 

 lected it. The plant is also said to be frequent in the northern 

 parts of Iceland, but we have seen no specimens from there, and 

 are, therefore, not certain whether the Icelandic plant is identical 

 with the Norwegian, the former having been described by Rott- 

 boell as G. detonsa. ^ " 



* Hartman (1. c.) describes the flower as tetramerous or, but seldom, 

 pentamerous. None of the Scandinavian authors mention nectariferous 

 g-lands in this species, and they were totally absent in our material from Nor- 

 way and Greenland. 



10 Acta Acad. Hafn., Vol. 10, p. 435. (Not seen.) 



