MR. JOHN MIERS ON THE GENUS CRESCENTIA. 163 



genus. Burmann (in 1757) collected and edited the posthumous notes and drawings of 

 Plumier, among which we find a valid but neglected species, C. latifolia, described and 

 figured in tab. 109. Jacquin (in 1763) in his 'Stirp. Amcr.' 175, tab. ]11, published a 

 drawing, accompanied by the most copious details on record, of Linnaeus's typical species ; 

 and Linnaeus in the same year repeated the short character of his type, in which he 

 placed together, as its varieties, most of the previously recorded forms, at the same time 

 expressing his doubt whether they might not prove to be specifically distinct. Swartz 

 (in 1791) described fully his Crescentia cucurbitina, a plant which corresponds with Lin- 

 naeus's var. 8, and with Browne's Crescentia ]S'o. 4. Kunth (in 1818) established his species 

 O. alata and C. acuminata. Gardner (in 1840) described his C. euneifolia. l)e Candolle 

 (in 1845) enumerated seven species of the genus in his 'Prodromus,' from which two 

 have since been excluded; he there repeated the errors in his references into which all 

 other botanists had fallen. Mr. Bentham (in 1844) described and figured his C. oborota. 

 Dr. Seemann (in 1854) gave the details of his C. macrophylla, which was figured in the 

 ' Botanical Magazine,' and which will presently be seen not to belong to Crescentia. 

 Dr. Seemann has written several memoirs on the Crescentiacem, the last being a monograph 

 of the family upon the basis he had proposed, which was communicated to the Linnean 

 Society (in 1859) and published in its Transactions; he there reduced the number of 

 species of Crescentia to two (if we exclude his C. macrophylla and C. alata, II. B. K.); 

 he justified this reduction solely upon the shape of the leaves, altogether ignoring the 

 more discriminating characters furnished by the structure of the flowers and by the 

 form and size of the fruit, characters which had been sufficiently Avell described by pre- 

 ceding botanists ; in this manner he made C. euneifolia, Gardn., and C. acuminata, II. B. K., 

 identical with C. cujete, aud he declared C. obovata, Benth., and C latifolia, Plum., to be 

 the same as C. cucurbitina, Linn. I will, however, presently denote the differential 

 features which give validity to all these several species. 



In regard to Crescentia macrophylla, Seem., already alluded to, the valuable observa- 

 tions of M. Bureau upon its living flowers, confirmed by those of M. B illon, show 

 (Monogr. Bign. p. 118) that its ovary is decidedly 2-locular, with the ovules arranged 

 longitudinally in four or six series in each cell, two or three being on the left, and the 

 same number on the right-hand side of the dissepiment, as in Kigelia, Colea, Sec. M. 

 Bureau kindly sent me these analyses, which prove categorically, according to my view, 

 that the plant in question does not belong to Crescentia, but will form the type of a 

 distinct genus, to which the name Amphitecna may be given, and which will class in the 

 same tribe as Colea, Phyllarthron, and some others. 



M. Bureau has also favoured me with his analyses of Colea involucrata, Boj., C.jlori- 

 bunda, Boj., Phyllarthron Bojeriana, all from Madagascar, and P. Comorense from 

 Mauritius : in all of these the ovary is manifestly 2-locular, with ovules arranged upon 

 the dissepiment, in the same manner as \xl Amphitecna and Kigelia, — a structure quite 

 incompatible with that of Crescentia. 



The genus Schlegelia, erroneously confounded with Tanaec'unn by Dr. Seemann, and 

 included in his family of the Crescentia ce w, offers many very peculiar characters. M. 

 Bureau has examined the ovary of Schlegelia parasitica in the living state, which he 



