34 MERRILL. 



in Java. and. accoidinfj to Praiii, in Indo-China, wild in SiHiet. Cacliar, and 

 Chittagong. 



Inga timot-iana DC', was reduced by Bentham to Par Ida roxburghii G. Don, 

 and following the principles of priority, the earliest specific name must be 

 adopted. In order to be sure of the identity of the Philippine plant with 

 DeCandollo's species, material comprising flowers, fruits, and leaves of the 

 Philippine plant, as well as fragments of two species cultivated in the Botanic 

 (iarden at Buitenzorg, labeled I'arkia intermedia. Hassk., and P. roxburghii 

 (>. Don, was sent to M. C. DeCandoUe for comparison with the tj'pe of Inga 

 timoriana DC. 1 am indebted to him for the following statement: "I have 

 entrusted to M. Buser the comparisons you desired to be made of three specimens 

 of Parkia with Inga timoriana DC. and Parkia Roxburghii Don, and of the latter 

 with what we have here under P. intermedia Hassk., in view of ascertaining if 

 they are distinct species. M. Buser has submitted to me his following conclusions 

 in which I entirely concur. 



" 'Taking for the type of Parkia intermedia Ilas.sk. the plant distributed under 

 this name by Zollinger (n. 3580) there exists a complete identity with intermedia 

 for the plant '"ex Hort. Bot. Bogor. cult." under the name of P. Roxburghii, but 

 not for the plant labeled, ibidem, P. intermedia Hassk. 



" 'P. intermedia Hassk. ( =Zollinger n. 3586, n. 736) and P. Roxburghii G. Don 

 (Wall. Cat. 5288) are certainly two distinct species (see leaflets and floral 

 characters ) . 



" 'Inga ? timoriana DC. is the same plant as Barnes 323 = P. intermedia Hort. 

 Bogor. cult., and quite different from true P. intermedia Hassk. In a broad 

 sense it may be identified witli P. Roxburghii Don, as done by Bentham; in a 

 more restricted specific conception it may be regarded as a species of secondary 

 order. 



'•'Roxburghii: rhachide rotundato-angulata, foliolis uirinque glaberri- 

 iiiis. inargine adprcsseiciliatis, siibconcoloribus, costa tenui, nervis secun- 

 darii.^ in((»iis]ii(\iis, rliachilla tenuioro. Corollae segmcntis extus liirsutis. 



"'Timoriana: rhachide qiiadrangulari, foliolis utrinque, supra prao- 

 sertiiii. ])lus iiiinus pilosis, subtus ])allidioribus, eosta latiuscula, nervis 

 secmidariis supra subreticulatc-prominulis. rhachilla latiore; corollae seg- 

 7nentis (Barnes 89) glaberrimis.' " 



The specimens sent for comparison were For. Bur. .i.i,i liarnes (leaves and 

 fruits), with flowers of For. Bur. 89 Barnes from the same locality (Lamao 

 River, Province of Bataan. Luzon), and two specimens from trees cultivatfed in 

 llie Botanic Garden at Buitenzorg, Java, one labeled ''Cult, in Hort. Bog. I, B, 51. 

 I'arkia- intermedin Hassk.," which is not Ilasskarl's species, but is Parkia 

 timoriana, and the other labeled "I, H, 4 = 48 = 50, Parkia Roxburghii Don," 

 wliich is not Don's species but is P. inlermcdia Hassk. Prain "' who has worked 

 over the species of Parkia occurring in the Malay Peiiinsula, also expresses the 

 o|)inioii that /'. roxburghii Don, and P. intermedia Hassk.. are distinct. Com- 

 parative .studies with a full series of specimens of typical P. roxburghii G. Don, 

 and P. timoriana may show the distinguishing characters indicated above to 

 be constant, and tlie two species wortliy of spocitic rank, a ]ioiiit tliat is left 

 for some future monographer to decide. 



. I. .111 II. As. ,S()c. Hrui:. 66' (iS'.tT) 240. 



