﻿ON THE BOTANY OF JAPAN. . 415 



in the genus. The anther is extrorse in its attachment, but the line of dehiscence 

 slightly introrse. This is the case, in a more decided manner, in Medeola. It be- 

 comes abundantly evident that the insertion of the anthers, and the partial or complete 

 separation of the styles, are too artificial and (through gradations) too indefinite char- 

 acters for warranting the ordinal separation of the Uvulariea; from the Convallariea:. 

 It would appear that the order Liliacea: must be opened, according to Mr. Bentham's 

 indications, to receive not only these plants, but the Trilliaceo', Slelanthacece, &c., and 

 I suspect even S?nila.v itself, notwithstanding its orthotropous ovules.* 



Streptopus amplexifolms, DC, was gathered at Cape Soya. This is a truly northern 

 plant in the New World, extending from Hudson's Bay, Xe^vfoundland, and Xew 

 England to our northwestern coast and islands ; thence to Kamtschatka and Japan. 

 It has not been detected elsewhere in Asia, nor is it known in Europe north of Saxony 

 and Silesia, whence it ranges southward to the Pyrenees, the mountains of Calabria, and 

 those of Hungary. So its geographical position in Europe is analogous to that of the 

 Lily of the Valley in the United States of America. Our Streptopus roseiis, Michx. 

 doubtless inhabits the northern part of Japan, since it occurs in the Aleutian Islands 

 on the one side, and in the Okotsk district on the other, where it is clearly Ledebour's 

 Smilacina streptopoides ! 



Although Lilium maculatum of Thunberg, which I have not seen, would appear to 

 represent our L. Canadense and its near allies, yet most of the Japanese species are 

 of European or Himalayan types. A Lily which was abundantly met with on the 

 coast of Jesso may be equally well referred to L. spectabile, Link., or to the Linnsean 

 L. bulbiferum, of which the former is apparently a mere variety. A single specimen, 

 and that with an unopened flower-bud only, was collected of a Lily, so well marked in 

 character that it may be named and described, even from such incomplete materials.f 

 Drs. "Williams and Morrow gathered, at Simoda, a specimen of Gagea triflora, E. & S., 

 hitherto collected only by Tilesius, — the habitat unknown to Ledeboui-, probably on 

 the eastern coast of Siberia, or in Kamtschatka. The plant connects Loydia with 

 Gagea : flowers apparently white, destitute of folds or pits at the base of the segments, 

 and with only about six ovules in each cell. 



* Ledebour and some other botanists have adopted Endlicher's error in considering the ovules of most 

 ConvaUarinets to be orthotropous. 



t Lilium ? medeoloides (sp. nov.) : glabrum ; bulbo granulato ; caule simplicissimo longe nudo ad apicem 

 folia plura (pseudo-)verticiUata gerente et peduncnlo superne bracteato unifloro terminato ; perigonii in ala- 

 bastro parvi phyUis oblongis dorso carinatis nudis apice calloso intus barbulatis. Genitalia omnino Lilit. 

 Hakodadi. 



