CONTRIBUTIONS TO BOTANY. 
81 
ticis, basi obtusis, apice acuminatis et longe cuspidatis, lucidis, 
prominenter reticulatis, margine cartilagineo subreflexo, sub- 
tus pallidioribus, nervds venisque prominulis ; racemis termi- 
nalibus, longiusculis, densifloris, tomentosis. — Brasilia, v. s. 
in herb. Lindl. (Forrest, 26). 
Species valde pecuHaris, A. marginato, DC., proxima: inter- 
nodia 4 poll, dist.; foliola firma, obscure viridia, nitentia, 
poll, long., Ij-l^ poll- cl. Cham, folia inferiora multo 
raajora, 9 poll, long., 3^ poll, lat.) ; petiolus 6-7 lin., petioluli 
tenuiores, 5 lin. long., superne canaliculati ; cirrhus simplex, 
tenuiculus, 6 poll. long. ; racemi rachis pulverulenta, 4 poll, 
long., fere e basi dorifera, circiter 30-flora; pedicelli oppositi, 
4 lin. long, (plerique delapsi), imo bractea 4 lin. long, donati; 
calyx tubulosus, fulvido-pulverulentus, 5-dentatus, infra dentes 
glandulis 20 biseriatim dispositis instructus, 4 lin. long. ; co- 
rolla in alabastro fulvo-tomentosa; maturae in specim. desunt, 
sed sec. cl. Cham. 2-pollicares, extus (lobis utrinque) dense 
farinoso-tomentosae, intus glabrae. 
In the herbarium of the British Museum I find a plant, in 
fruit, from the neighbourhood of Rio de Janeiro, the seeds of 
which difier from those last described : the specimen has no 
flower, so that it cannot be determined to be a species of Ade- 
nocalymna*. The capsule is compressed, not cylindrical as in 
A. scansile, and the valves are proportionally thinner. The 
seeds are uniserial and much more compressed ; the central disk 
is testudiniform, one of its sides, that of the hilum, being straight, 
the other rounded and emarginated, while the extremities are 
broadly expanded into a rather thin but opake wing, which is 
considerably narrower than the disk ; the hilum, instead of being 
broad, is veiy narrow, linear, and marginal, corresponding with 
the cicatrices seen along the margins of the dissepiment. This 
appears to conform with the brief description by DeCandolle of 
his genus Pachjptera, no specimen of which I have seen. The 
internal structure of the seed is somewhat different, though ap- 
proaching that I have given of Adenocahjmna (see pp. 4.5, 
72) : the discoid portion, although coriaceous, is not nearly so 
* I take this opportunity of confirming what I formerly stated (see 
p. 68) concerning the little dependence to he placed on the calyx as a con- 
stant and unerring test for generic discrimination. Perhaps no genus in the 
family offers a more striking feature than Adenocalymna, in its peculiar 
calyx, which gave origin to its name ; but I find in Gardner’s colleetion a 
plant, allied to Dolichandra, with an entire, tubular, coriaceous, pidverulent 
calyx, marked with polished glands placed biserially helow the margin, 
just as in Adenocalymna-, and yet it is far removed from that genus on 
account of the difference of its hahit, of its coroUa, in the structure of its 
anthers, its ovarv', its thick flat siliquose capsule, and its seeds. 
VOL. II. M 
