134 Mr. J. Miers on Diclidanthera. 



flores alterni breviter pedicellati ; pedicellus breviusculus, e no- 

 dulo 3-tuberculato ortus, in imo articulatus et 3-bracteatus ; 

 bractcse parvee, acuta, valde dcciduce ; corolla ochroleuca, sicci- 

 tate purpurascens ; fructus cerasiformis. 



1. Diclidanthera pendidiflora, Mart. Nov. Gen. ii. 140, tab. 196 ; 

 Flor. Brasil. fasc. xvii. 11, tab. 4. fig. 1 ;— ramis pendulis, 

 ramulis subpuberulis * foliis oblongis vel lanceolato-oblongis, 

 utrinque acutis, supra lucidis, hinc ad basin petiolum versus 

 2-glandulosis, costa media inferne rufescente, glabris, utrin- 

 que valde reticularis, petiolo tereti, pubescente; racemis in 

 ramulis novellis terminalibus, folii floriferi fere longitudine; 

 sepalis extus fere glabris, sub lente sparse retrorsum puberulis, 

 corolla? glabra? ochroleucse tcrtia parte longitudinis. — Brasilia 

 tropica, ad fluv. Amazonas — v. s. in herb, meo, — prope Santarem 

 (Spruce). 



Prof. Martius describes it as a tree 10 to 15 feet high, with 

 loose pendulous and subscandent branches, the leaves being 

 4 to 6 inches long and 12 to 18 linos broad. In Spruce's speci- 

 men they measure 4 inches in length, and 15 to 17 lines in 

 breadth, on a petiole 3 lines long. In the size of the tree, its 

 habit, in the form of its leaves, their size and colour, this spe- 

 cies is scarcely to be distinguished from the following one, the 

 only valid specific difference being in the two basal glands 

 constantly seen on the upper side of the leaf, at the point of 

 junction of its blade with the petiole, which are always wanting 

 in D. laurifolia. The corolla is a little more slender and rather 

 longer in regard to the calyx, and when dried, of a brighter 

 crimson colour ; its pedicel is only half a line long ; the sepals, 

 which become reddish in drying, are 4 lines long and 1 line 

 broad; the length of the corolla is 10 or 11 lines, including the 

 border of 3 lines, the tube being only a line in diameter. 



Var. /3. penceantha, Mart. loc. cit. D. Martii, P'dpp. MSS. in 

 herb. 2883 ; — " foliis brevioribus, latioribus, ovato-oblongis, 

 basi subrotundatis, petiolis evidentius puberulis; racemis 

 depauperatis (2-3-floris) petiolo 2-plo longioribus ; calycibus 

 evidentius pubentibus." — Fluv. Amazonas, Ega (Poppig). 



This variety I have not seen. In the form of its leaves it seems 

 to approach more to D. elliptica ; but the presence or absence of 

 the basal glands on the upper surface of the leaves will deter- 

 mine to which species it belongs. The shortness of its racemes, 

 which are only half an inch in length, and its being only two- 

 or three-flowered, if a constant character, would suggest the 

 possibility of its being accompanied by other marked specific 

 differences. 



