MB. D. OLIVER ON Ar/RANTIACEJ:. 13 



marked section of the genus. I have never, or very rarely indeed, 

 found the calyx of A. monophylla perfect. I have seen very many 

 specimens of it in flower, but the flowers appear to open simul- 

 taneously, and I have hardly had the opportunity of examining a 

 sufficient number of young buds to understand clearly its original 

 form. At the time of flowering it is invariably irregularly split, 

 and its margin scarious and eroded. The ovary is 4-, very rarely 

 5-locular. In other Atalantias the calyx is perfect at flowering, 

 as in all other species of the order, and usually 4- or 5-lobed or 

 partite ; the ovary being, in the species nearest to A. monophylla, 

 usually bilocular. A. Hindsii, as observed supra, is a rather ano- 

 malous member of the genus. It is not improbable that further 

 discoveries in Eastern Temperate Asia may lead to its union with 

 other forms partaking more fully than Atalantia of the characters* 

 of Citrus. 



SCLEEOSTTLIS. 



Blume published this genus in his 'Bijdragen' in 1825. I have 

 not seen authentic specimens of any of the five species which he 

 describes ; but from an examination of his descriptions, and my 

 acquaintance with Javanese species oi Aurantiaceoe , I feel satisfied 

 his plants may be referred, part to the genus Atalantia, part to 

 Glycosmis. Of his S. (?) macrophylla, he says himself, " An Gly- 

 cosmis pentaphylla ?" If it should appear needful to separate from 

 Atalantia the species which retain a perfect calyx at the time of 

 flowering, the diagnosis of Sclerostylis might be, I think, suitably 

 modified for their reception. The species referred by Dr. Wight 

 to Sclerostylis I include under Atalantia. Miquel divides the 

 genus between Limonia and Olycosmis, following Sprengel (Syst. 

 Veget. iv. 161). 



Tripuasia. 



This genus, proposed by Loureiro in his 'Flora Cochin Chinensis,' 

 was based on a species previously described iu Linnaeus' ' Man- 

 tissa' (1771) as a species of Limonia (L. trifoliata). Triphasia 

 trifoliata is probably entitled to generic distinction, although I 

 believe the ternary symmetry of its flowers alone would by no 

 means suffice to base the genus exclusively upon. It sometimes 

 occurs with tetramerous flowers, though rarely; the ovary and 

 berry are also occasionally 4-locular, and the leaves 1-foliolate. It 

 is difficult to say to which genus Triphasia is most nearly allied, — 

 probably to Limonia. T. monophylla, from Timor, is a trimerous 



