PLANTS JAVANICiE RARIORES. 585 



believed the whole family to agree in this respect with 

 Biynoniacem, next to which he had placed it ; and he had 

 at least the proof of the absence of albumen in Roxburgh's 

 figure of his Incarvillea parasitica, in which the structure 

 of ovarium as well as of seeds is correctly given. 



The genera referred by Dr. Jack to this new family are 

 Cyrtandra of Forster, Didymocarpus of Wallich, and Q07 

 two other genera for the first time proposed, namely, 

 JEscJiynanthus, to which Roxburgh's Incarvillea parasitica 

 belongs, and Loxonia. 



Nearly about the same time Mr. Don published his 

 Didymocarpece} a family consisting of two of Dr. Jack's 

 Cyrtandracece, namely, Didymocarpus and JEschynantJtus 

 (his Trickosporum), and a new genus Lysionotas. He in 

 like manner depends chiefly on placentation, which is not 

 very clearly described; and he also introduces into his 

 character the absence of albumen, the pendulous position 

 of seeds, and the undivided stigma. From this family he 

 excludes Cyrtandra, which has, he states, erroneously how- 

 ever, a copious albumen ; while Chirita of Dr. (Buchanan) 

 Hamilton, of which Dr. Jack's Didymocarpus includes at 

 least one species, is doubtfully referred to Scrophdarince, 

 chiefly on account of its bilamellar stigma. 



In 1826 Dr. Blume 1 refers Cyrtandracece to the natural 

 order Bignoniacece, distinguishing it as a tribe from true 

 Bignoniacece by its pendulous seeds, and subdividing it 

 into two sections ; the first, Tric/iosporte, with capsular 

 fruit, and seeds either winged or with some other form 

 of appendage; the second, Cyrtandrece, having baccate 

 fruit and seeds without appendage. To each of these sec- 

 tions he has added more than one new genus, but Loxotis 

 and Loxonia are not included in either of them ; though 

 Loxotis, as I have already stated, is probably his Bhi?icho- 

 glossum, referred by him to Bldnanthece ; and Loxonia, 

 notwithstanding his account of the fruit, may possibly be 

 his Loxopkyllum, which he has placed in Scrophularince ; 



1 'Edinburgh Philosophical Journal,' vii. p. 83, and 'Prodromes Three 

 Nepalensis' p. 121. 



2 ' Bijdragen? p. 759. 



