BOTANICAL GAZETTE. Ki'^ 



corolla and stamens fallen, a dehiscent capsule enclosed in a persistent im- 

 bricated calyx and surmounted by a persistent style. It was noticed in the 

 Flora Boreal i-Aum-kana, which was prepared by L. C. Richard from 

 Michaux's collections. Early in the year 1839, I found and examined this 

 specimen in Michaux's herbarium, and I received from t! e hand of M 

 Decaisne a drawing and some fragments of it. In a i)a])er treating of the 

 botany of these' mountains, contributed to this Journal in January, 1842, I 

 ventured to found a genus upon this plant, under the above name, trusting 

 thut the dilligent search pro^ecutsd by myself and by all botanists visiting the 

 region would duly bring it to light. The protracted failure of these endeav- 

 ors has thrown an a:r of doubt over the minds of my associates in the 

 search, as to the actual existence of any such plant. In 1868. 1 had the 

 pleasure of announcing in tliis Journal (Ser. II, xi, 402) the discovery of 

 this genus, not indeed where we were looking for it, but where experience 

 had led me to expect that any or every peculiar Atlantic States type might 

 recur, namely in Japan. That is, I identified the genus with the Schizoco- 

 floH uiiijionis A Maximowicz, which, singularly enough, was known only 



b}' specimens in tlie same condition /. ?., with calyx and gynoecium, but 

 neither corolla nor stamens. The patent relationship of these specimens to 

 Schizocodon soldaneUoides of Zuccarini gave ground for a conjectural res- 

 toration of the missing organs; and I ventured the opinion that Shortia 

 (of lH-12) and Schizocodon (1843). whether of one genus or two, were most 

 related to Diapensia. In the year 1870 (in Proc. Am. Acad., viii, 243) I 

 reconstructed the order D/apcnsiacece, referred to a separate tribe, Galarinere. 

 tlie genera Galax and SJiorfki, and adopted the idea of a probable identity 

 of ScJiizocodon with the latter. The next year Maximowicz decided tha^ 

 the two genera should be distinct, founding this conclusion upon the close 

 seed-coat (confirmed in the Japanese Shorfi<( nniflora) and the campanulate 

 corolla, with lobes undulate-crenate instead of fimbriate, and upon some 

 characters in the stamens, all these taken from a rude figure in the Japan- 

 ese 6'oo Bokf., iv. fo!. 8, which is supposed to represent S. nniflora, although 

 the leaves would (as Maximowicz rightly observes) refer it rather to *S'. (jala- 

 cifolia, these being all represented as acute or in one dubious case subcor- 

 date at base, instead of reniform-cordate. The identification as to genus is 

 doubtless correct; but the analysis of the flower is too rude for reliance as 

 to all relating to the stamens and the squamulae. Happily I can now give 

 the characters from an actual blossom. 



For 1 have now received, at first indirectly from Mr. J. W. Congdon, 



