182 BOTAKY OE THE VOYAGE OF H.M.S. nEKALD. 
* 
In endeavouring to find out characters for a sectional subdivision of the Order Orescentiacea , my atten- 
tion was drawn to an interesting paper in the 21st volume of the * Linnean Transactions' (p. 141), in which 
Mr. !Miers describes Oxycladus, a new genus of plants, which he associates with Bignoniacefs, an Order 
which, according to the views of that author, comprises both the true Bignoniacece and the Otescentiacea . 
The structure of Oxgcladus, however, is so entirely different from all other genera of Bignoniacea; , that Mr. 
Miers was obliged to make a new division of that Natural Order, called OxycladecB. I should have been 
glad to adopt that learned botanist's news, had they not so widely differed from those I hold on the same 
subject. It was in order to bring our mutual views into harmony -with each other that I began to make a 
critical investigation of the paper alluded to, the result of which was that I felt convinced Mr. Miers's third 
division of Bignoniacea, viz. the tribe OxycladeeB, was not tenable, because the genus Oxgcladus, Mers, has 
nothing to do with BignoniaccfS, even in the widest sense, but belongs, I think, to Myoporacece. 
I have therefore made an attempt to distribute the true Crescenfiacece into tivo sectional subdivisions : 
the firnt tribe, for which I adopt the name Tanceciece, includes aU those genera with a persistent, regular, 
five-cleft calyx (viz. Colea, BariUema, Bhgllarthron, Tamscium, and Tripinnaria) . The second tribe, 
called by me CracentietB, comprises those genera which have a deciduous, irregular, spathaccous, or 
biparted calys (viz. Barmentiera, Creseentia, Kigetia, and perhaps Sotor). 
I may add, that all the plants belonging to the Order CresccjitiacecE have a tendency to form winged 
petioles ; and it is by no means unlikely that the simple-leaved Crescentieee will in future be looked upon as 
plants with abortive leaflets and highly developed petioles {plgllodid). I may also remark, that all Crescen- 
tiacew have a parietal placentation, and a truly unilocular fruit, as an examination of the ovary will show ; 
but that, when the placenta? meet, as they generally do when the fruit approaches maturity, the placentation 
appears to be asile, and the fruit bi- or quadri-locular. 
w 
Tribus I. TAXiEciE^j Beem. 
Cahjx persistenSj regularis, 5-merus, 
778. Taxjectum Ulacimm, ScQm.Sc/ilegelia UJacina, Mic^., Pe Caucl. Prodr. vol. ix. p. 564. 
Cupica, Daricn. 
w 
There is no generic distinction between BcKlegeVm and Tanmcium; the berrj is black, about a quarter 
of an inch long, and has a sweet taste- 
Tribus II. Crescentie^, ^ecm. 
Calyx deciduus, irregularis, spathaceus vel bipartitus. 
779. Parmentiera cereifera, Seem. (Tab. XXXII.) ; inermis, foliis sparsis, pefiolis alatis, 
foliolis membranaccis obovato-oblongis acuminatis, corolla alba, fructu cylindrico ecostato bisulcato 
glaberrimo, seminibus subcordatis.— Nomen vemacul. '^Palo dc velas." Near the viUages of Gor- 
gona, Cruces, and San Juan, Province of Panama. 
The tree is confined to the valley of the river Chagres, where it forms entire forests. In entering them 
a person might almost fancy himself transported into a chandler's shop. Prom aU the stems and lower 
branches of the trees hang long cyHndrical fruits, of a yellow wax-colour, so much resembling a candle as to 
have given rise to the popular appellation, Palo de velas (CaUdle-tree). The fruit is generally from two to 
throe, but not unfrequently four, feet long, and about an inch in diameter. The tree itself is about twenty 
feet high, and has opposite trifoliolated leaves, and largo white blossoms, which appear throughout the year, 
but are in greatest abundance during the rainy season. Hitherto only one species of Parmentiera, P. eduUs, 
De Cand., was known to exist. The fruit of the latter, called Quauhxilotl, is eaten by the Mexicans, while 
that of the B. cerelfera serves for food to numerous herds of cattle. Bullocks especially, if fed with the 
