72 MB. BENTHAM ON LOGANIACEjE. 



with some species of Lisyanthus, the chief ordinal distinction 

 consists in the greater development of the placentae and fleshy 

 fruit, and the habit is by no means dissimilar. 



The known species of Fagrcea are all Asiatic or Polynesian. 

 They are all thick-leaved trees or shrubs with a more or less 

 tendency to pseudo-parasitism, or to a somewhat climbing habit ; 

 the inflorescences and flowers are apt to be thick and succulent, 

 so as to be difficult to dry ; many of the species run much one 

 into another, and herbaria specimens are very unsatisfactory for 

 distinguishing them. That an erect or climbing habit is not in this 

 case a good specific difference, we have not only the presumption 

 derived from the inspection of specimens, but the positive evidence 

 of the late Col. Champion, a most careful observer, who found both 

 the common Ceylonese species to vary as stunted shrubs, weak 

 trees, or woody climbers, or perhaps rather trees with sarmentose 

 branches. Blume, who has with perfect justice included Cyrto- 

 phyllum and Ficrophloeus, enumerates thirty -three species, besides 

 seven others contained in DeCandolle's Prodromus, the F. coro- 

 mandeliana since published by "Wight, and three new ones which 

 I now propose. But it is probable that several of the above will 

 have to be reduced when better known. Some are described from 

 specimens in leaf only, others in fruit without flowers, or from 

 manuscript descriptions and figures, and there appears reason to 

 believe that the characters derived from the leaves are not more 

 constant in this than in other genera. 



The division proposed by Blume into three groups according 

 to the inflorescence is a very good one. The first, with few- 

 flowered terminal cymes or corymbs and large flowers, contains 

 the greatest number of species. These may be subdivided, or 

 rather arranged, according to the length of the tube of the corolla 

 before it expands into a campanulate throat. In F. carnosa, Jack, 

 of which we have specimens from Moulmeyn, gathered by Lobb, 

 it is near 5 inches long ; F. tubulosa, Blume, is said to be very 

 near that one ; F. zeylcmica of Thunberg, very well figured by him 

 in the * Stockholm Transactions,' and by Blume in the ' Bumphia,' 

 although often confounded by others with a short-flowered 

 species, has the tube full 3 inches long. In the Carissa grandis 

 of Bertero, from the Society Islands, which is an unpublished 

 Fagrcea, and in F. lanceolata, Blume, the tube is shorter, but still 

 it does not expand till above the middle ; so it is also in F. lanceo- 

 lata, Wall., a Penang plant with smaller flowers, which DeCan- 

 dolle had on that account placed in Cyrtophyllum, but which has 



