6 University of California Publications in Botany [Vol. 5 



what lower stature and the spreading leaves at the base give it 

 a sort of pyramidal (or conical) shape (cf. plate 3). The leaves 

 are long, broad in the middle and tapering very rapidly to each 

 end. At the apex is a fairly long point curved to one side, while 

 the base is narrow, to expand at the junction of the stem into 

 two broad clasping auricles. The panicle is ample as compared 

 with that of other members of this section. The flowers are 

 very light pink, with slender tube and infundibulum and with 

 the limb broadly but deeply lobed. The lobes have slender in- 

 curved points. It would be classed by both Comes and Anas- 

 tasia as a combination form under Nicotiana Tabacum var. vir- 

 ginica. It comes near to the plant figured by Anastasia (1906, 

 opp. p. 30) and by Comes (1899, pi. V) but is not identical with 

 either. It seems best to call it "Maryland." The original 

 Nicotiana virginica Ag. (1819, p. 18 ), as represented by the 

 type-specimen in the Herbarium of the University of Lund, is of 

 the same general type, as is also the type of Nicotiana Tabacum 

 L. of Linnaeus 's herbarium in London. The type-specimen of 

 N. fruticosa in the Linnaean Herbarium seems also to belong 



here but may possibly, however, represent the plant referred 



71 

 by Anastasia (cf. above under U. C. B. G. 05) to var. brasili- 



ensis. The plant in Hb. Agardh is decidedly narrow-leaved and 



may be near to what Comes (1899, p. 10, pi. IV) has called 



var. lancifolia. From what could be seen of the flower, it 



seemed to be broader-lobed than that represented by Comes for 



this latter variety. 



Nicotiana Tabacum var. calycina 



no 

 V. G. B. G. 05. — This is the plant known in botanical gar- 

 dens as Nicotiana Tabacum var. calycina. The seed was re- 

 ceived from the Botanic Garden of the University of Cambridge 

 in 1905. It has remained constant in its peculiarities, ever 

 since, under conditions of pure-line breeding. In habit, it is 

 peculiar, as shown in the photograph reproduced in plate 4. The 

 lower laterals soon come to equal the main axis or even slightly 

 to overtop it, so as to obscure it as a main axis. The leaves are 

 large, approaching in shape the members of the virginica-gromp 



