i-2 



In spite ul' the siuiibu-ity uf the liowers and I'niil, it is very diffi- 

 cult to bring oneself to class all the plants included as varieties 

 by Hiern as varieties only. In life no plants of a genus could 

 look more dissimilar than the tidal swamp bush (var. 

 racfiHosa) and the tree (jlahra. 



('ONNAKACE.E. 

 44'. ROUREA CONCOLOK, Bl. 



Telom Forest, on a fallen tree. Specimens nearly jjerished and 

 in very bad condition but apparently this species, which 

 occurs als(.> in the low country and in Sumatra and Borneo. 



I>EGUMIXUSyT<:. 

 45. Desmodium laxum, Be. 



A slender wiry plant, hardly more than a herb, with pink 

 flowers. Telom, in a Sakai clearing, scanty, also obtained in 

 Ulu Bataug Padang by Wray, and occurring in India, China 

 and the Malay islands. 



4(3. D. SC'ALPE, I)c. 



A blender plant like the last but with orange scarlt't flowers. 

 Telom Camp, not previously recorded from the Peninsula. 

 Distrlb. — Africa and India. 



47. D. MEGAPHYLLU3I, Zoll. 



A beautiful bush about 8 feet tall, with violet flowers. Ulu 

 Bataug Padang, collected in the same district by Wray, but 

 not known elsewhere in the Peninsula. 



Di'6'///6.--Tenasserim and Java. 



48. Erythkina, sjA, probably lithosperma, Miq. 



One or two big trees of an Erythrina in flower, but too lofty to 

 obtain specimens from, were seen in the forests of the Ulu 

 Batang Padang, luidoubtedly wild. The occurrence of any 

 s])ecies of the genus wild in the Peninsula has been extremely 

 doubtful except in the case of the Lankawi plant, perhaps. 

 E. .siiherosa, E. indiea, E. strida and E. ovcdlfolla are planted, 

 and the specimens referred to in the "Materials,"' as well as 

 at least most, if not all, of the E. lithosperma are also from 

 remains of cultivation. 



40, Bauhinia, sj). 



A large climbing species, with red and orange flowers, was seen, 

 drapiug a mass of trees from the lower slopes of Gunong 

 Berumban, and seedlings were seen scattered through the 

 Telom woods, but the flowering plants were quite iuacces- 

 sil)le, and the species could not l»e identilied, 



