FLORA OF SOUTH AMERICA. 631 
exterioribus parvis, angulis ovarii oppositis, 3 interioribus 
duplo majoribus faciebus ovarii arete adnatis. Ovarium tri- 
quetrum. Stigmata 3, sessilia, brevia, lobata. ^ Perigonii 
fructiferi lacinic interiores valde auctæ membranaceo-foliaceæ 
fructui acutissime triquetro arcte adnate. Pericarpiun tenue. 
Semen triquetrum, ruminato-rugosum ? — S. paniculata, 
species unica. Arbor habitu Triplaridi affinis. Ramuli ju- 
niores ferrugineo-tomentosi, mox glabrati. Petioli dilatati, 
breviter vaginantes, semipollicares ad pollicares. Folia ovalia 
V. oblongo-elliptica, 6-10 poll. longa, 2-4 poll. lata, apice 
obtusa, margine integerrima, basi obtusa v. leviter cordata, 
glabra, rigide membranacea, costa media venisque parallelis 
subtus prominulis, rete venularum vix conspicua. Panicula 
mascula pedalis, divaricato-ramosa, ferrugineo-tomentella. 
Flores parvi, secus ramos graciles per 3-5 glomerati, glome- 
rulis inter se distantibus bracteis minutis suffultis. Pedicelli 
perigonio breviores. Perigonii laciniæ interiores vix lineam 
longe. Stamina ultra 20, perigonium subæquantia ; antheræ 
majuscule, breviter oblongæ, biloculares. Panicula foeminea 
mari brevior, confertior. Inflorescentia eandem legem se- 
quitur, pedicellis tamen proportione longioribus. Perigonii 
laciniæ interiores ab ovario non separabiles, 1 lin, longe. 
Fructus (nondum maturus) cum perigonii laciniis interioribus 
auctus, in specimine 7 lin. longus, triquetro-pyramidatis, 
angulis subalatis, faciebus prope basin 4-5 lin. latis, laciniis 
Perigonii summo apice brevissime liberis ceterum arcte 
adnatis ; laciniæ exteriores sub angulis horizontalibus persis- 
tentes et iis dimidio brevioribus. Pericarpium tenue stig- 
matum vestigiis coronatum. Semen jam formam illius Tri- 
Dlaridis sumere incipit. 
THYMELEACEÆ. 
. The genera of this order, as usually characterized, were in 
the greatest confusion, very unnaturally marked out by 
characters often of very little importance. Dr. Meissner has 
lately reduced to their proper limits the South African and 
some of the Asiatic genera, and it is to be hoped that he will 
