132 G. King — Materials for a Flora of the Malayan Peninsula. [No. 1, 
fleshy granular albumen, and small conical embryo.— Distrib. A single 
Malayan species. 
Cardiopteris lobata, Wall. Cat. 8033 A, and in Benn. PI. Jay. 
Rar. 246, t. 49. Stems twining, terete, striate, pale when dry. 
Leaves membranous, ovate-cordate, acuminate, more or less lobed, 
glabrous ; main nerves 7 to 9, radiating from the apex of the petiole ; 
length 2*25 to 5 in., breadth 1*5 to 2'75 in., petiole 2 to 5 in. long. 
Panicles 2 to 4 in. long, solitary, axillary, on long peduncles with 
a few long slender recurved few-flowered branches. Flowers distant, '1 
in. long, on short puberulous pedicels. Calyx puberulous ; corolla slight¬ 
ly longer than the calyx, pale. Anthers emarginate. Fruit *8 to 1*25 
in. long, by *6 to *8 in. broad, the calyx persistent and very slightly 
accrescent. Seed sub-cylindric, compressed : Miq. FI. Ind. Bat. I, Pt. 1, 
799. H. Brongn. in Adansonia, X, 280 : Baillon in DC. Prod. XVII, 
26. C. moluccana , Blume, Rumpliia, IV, 207. t. 177, f. 2. C. javanica , 
Blume 1. c. Ill, 206, IV, t. 177. Peripterygium quinquelobum , Hassk. 
Cat. PI. Hort. Bogor. 351. Olus sanguinis , Rumph. Herb. Amb. V, t. 
4S2. 
Perak: King’s Collector.— Distrib. British India, Siam, Malayan 
Archipelago, New Guinea. 
A widely distributed plant, varying considerably as to the form of 
its leaves. On these diversities of shape, four varieties have been 
founded. 
The position of Cardiopteris has given rise to considerable discus¬ 
sion. Robert Brown was the first to suggest its affinity to Phytocrene 
and lodes. By Wallich (its original founder) the genus was placed in 
Sapindacese ; Hasskarl placed it in Euphorbiacese ; and Blume regard¬ 
ed it as forming a separate family near Boraginese and Verbenacese . 
There is no doubt that the gamopetalous character of the corolla and 
the absence of a disk are characters at variance with those of the majo¬ 
rity of the species which are grouped in the family of Olacinese ; but 
the ovulation and structure of the fruit of Cardiopteris are quite un¬ 
like those of either of the gamopetalous families suggested as allies by 
Blume. The single character which, in my opinion, suggests a relation¬ 
ship with Euphorbiacese is the milky nature of the juice. The stamens 
and ovary, as Brown pointed out, are essentially those of lodes , from which 
genus however this differs in calyx and in fruit. The genus with which, 
as it appears to me, there is by far the closest relationship is Pteleocarpa , 
Oliver. In fact the only characters which separate Pteleocarpa from 
Cardiopteris are that Pteleocarpa is a tree, while Cardiopteris is a scandent 
milky-juiced herb ; and that Pteleocarpa has a 2-celled ovary and 2-celled 
fruit. And even this latter distinction is to a great extent neutralised by 
