148 BOTANICAL GAZETTE. 



great handfulls of the flowers. I think thej r must have opened 

 about a week ago, as the racemes had set fruit helow. There were 

 many budding clusters also. I account for this display by the 

 fact that the swamp in question was reduced to dusty dryness by 

 the longf-continued drouth; this acted on the plants much as 

 their normal winter rest. When followed by rain — flooding the 

 marsh — and then by an extended period of warm weather, they 

 burst forth into flower. A natural consequence, I should suppose, 

 would be a dearth of my favorites next spring. — W". Whitman Bai- 

 ley, Brown University. 



Tllbei'S. — That the scales on the tubers represent leaves of the 

 of aerial stems is well known. The study of the phyllotaxy of these 

 subterranean leaves is quite as interesting as that of ordinary leaves. 

 In examining all tubers of cultivated and wild plants that I can 

 obtain, I find that a plant has the same arrangement of foliar or- 

 gans on tubers that it has on the stem; and where two plans exist, 

 the one at the base of the stem is the one followed by the tuber. Good 

 examples are found in the potato, in which both leaves and scales 

 are alternate; and in Helianthus doronicoides, L., and tuberous, L., 

 where the leaves are opposite below and more or less alternate 

 above, the scales on the tubers are opposite. In the latter species 

 scales frequently subtend "knobs", the tuber branches, 

 which are then opposite and themselves bear scales — the leaves of 

 the branch. The dimerous whorls decussate on tubers as well as 

 do those of the stem. Another interesting fact is the completion 

 of growth as to length in the lower in^ernodes of the tuber 

 while the upper are still quite small — a characteristic of the stem. 

 — Aug. F. Foerste, Dayton, Ohio. 



Notes. — Teachers of botany may be interested in knowing, if 

 they do not know already, that the now common Japanese Ampelcpsis 

 presents an excellent instance of a uni-foliolate compound leaf. The 

 three-lobed, or sometimes barely lobed and dentate leaflet has all 

 the appearance of a simple leaf, but falls off by a distinct articula- 

 tion from the top of the extremely long petiole, which is apt to per- 

 sist sometime thereafter. 



In the analysis of Heterocentron roseum, of the order Melas- 

 tomacece, students complain that they can not ascertain the name 

 by the key in the School and Field-book of Botany. I find that 

 the difficulty is in the statement that in that family the calyx is 

 coherent with the ovary. In Heterocentron, so far as I have exam- 

 ined specimens, it is distinctly free. The ordinal characteristics 

 given by LeMaout and Decaisne, give the alternative of free or co- 

 herent. — W. W. Bailey, Brown University. 



