1897.] D. Prain — Some additional Legiiminosee. 515 



8. Albizzia stipdlata Boiv. 

 Add to localities of F. B. I. :— 

 Andamans ; E. H. Man! Nicobars ; Kurz ! 



Two varieties may be easily distinguished in the field, viz., var. typica with large 

 stipules, and var. Smithiana {Mimosa Smithiana Roxb.) with small stipules. They 

 cannot easily be separated in the herbarium as the stipules are someAvhat deciduous 

 in both ; as they grow the two trees are wonderfully unlike and it would not be a 

 matter for surprise to find that Roxburgh was justified in separating them. The 

 typical A. stipulata is well known as the Sao in Assam and in Sikkim. 



9. Albizzia myriophylla Benth. 

 Add to localities of F. B. I. : — 



Kedah ; Curtis ! Penang ; Curtis ! Kunstler ! Perak ; Scortechini ! 



This species is perhaps most easily recognised by the pulvinus enlarging into a 

 recurved hook just below the leaf -base ; it is always a climber. 



Another species which has a similarly enlarged pulvinus is the Chinese Albizzia 

 Millettii Bth. the oldest name for which is Mimosa comiculata Lour. (Fl. Cochin-Chin. 

 800). Why it deserves our attention here is because of its having been introduced to 

 India and of its having been long cultivated in gardens under the name " Acacia 

 Careyana Hort." Mr. Kurz in a manuscript note in the Calcutta Herbarium, by 

 way of criticism of the F. B. I., has expressed the opinion that Acacia Careyana. is 

 the true " Mimosa elata " of Roxburgh. This is not the case ; Roxburgh has left a 

 coloured drawing of his M. elata which shows that his tree does not have the pul- 

 vinus developed into a spur. Moreover Roxburgh has with his own hand written on 

 the drawing " Mimosa elata considerably less than natural size "; consequently, the 

 leaflets of Acacia Careyana, which are the size of those in the drawing, are consider- 

 ably less than those of Mimosa elata. As has already been explained under A. 

 procera, the writer is not only convinced that Mr. Baker is right in referring Mimosa 

 elata to Albizzia procera, but is strongly of opinion that there is no variety " elata" 

 really distinguishable from A. procera proper, the leaflets are not smaller in the 

 variety than in the type and both kinds of leaflets (consequently both " varieties ") 

 can be collected from the same tree. 



A. Millettii has been collected in Tonkin by Balansa (nn. 1283 and 1290, both 

 issued as A. procera) and recurs in Borneo whence it has been sent by Haviland, 

 (nn. 57 and 2909). 



Mr. Kurz has described in the Society's Journal xlv. 2. 299 and again in For. 

 Flor. Brit. Burm. i. 428 a Siamese tree under the name Albizzia Teysmanni. This has 

 alternate leaflets and has no glands on the rachis and does not bear much resem- 

 blance to any Albizzia. Most probably it belongs to the suborder Caesalpinieas ; his 

 only specimen is in such a condition that Kurz was not justified even in suggesting 

 a genus for it. 



131. PITHECOLOBIUM Mart. 



4. Pithecolobium bigeminum Mart, in Flora xx. 2. Beibl. 115 in obs. 



5. Pithecolobium affine Bah. 



Add to localities of F. B. I. : — Burma ; Hills east of Tonghoo, 

 Brandis ! Perak ; Kunstler! Singapore; Ridley! Distrib. Borneo. 

 J. n. 65 



