856 



NOTES ON SAPOTACE.E.—ll. 



By Marcus M. Haetog, M.A., B.Sc, F.L.S. 



[Read at the Meeting of the British Association at Sheffield, 1879.] 



This paper is a continuation of my notes on SapotacecB published 

 in the 'Journal' last year. It is intended to form part of a series 

 of prolegomeua to a descriptive monograph of all the species of the 

 Order, which I trust to be able to work up in due time. The order 

 of these notes has, unfortunately, to be decided by that in which I 

 am able to study the material. 



In the paper above referred to (' Journ. Bot.,' 1878, pp. 65-72) 

 I proposed a division of this Order into three tribes, founded on 

 what ai^peared to me the most constant characters, i. e., the 

 fertility of the alternipetalous stamens, or their reduction to 

 staminodes or abortion, and the presence or absence of stipular 

 appendages to the true petals. My first section, BassiecB, has 

 exstipulate petals and all the stamens fertile; the second, 

 ChnjsophijUea:, has exstipulate petals and the alternipetalous 

 stamens sterile ; the third, Mimusojjem, has stipulate petals and the 

 alternipetalous stamens sterile, except in two species. The present 

 paper is devoted to a critical examination of the genera of this last 

 section. The genera recognised by Bentham and Hooker are as 

 follows : — Bumelia and DipJwlis, both pentamerous, but the one 

 with albuminous, the other with exalbummous seeds ; Mimusops 

 with a 2-seriate, 3-merous, or 4-merous calyx, the outer whorl 

 valvate, and a 6- or 8-merous corolla ; Imhricaria differing from 

 Mimmops essentially only in the deep lobing of the stipular 

 appendages to the petals ; Lahramia marked by the laciniate 

 stipular appendages of Imhricaria, but 3-merous and with the 

 staminodes reduced (said to be absent), and a diiDlomerous 

 gynaeceum. To these I have brought under the name of Muriea 

 two other species, formerly" ascribed to Labour donnaisia, which, 

 however, possess the flowers of Mimusops, but with both whorls of 

 stamens fertile. Finally, there remains another genus, with 

 several whorls of stamens or staminodes, represented by a sohtary 

 specimen (Coll. Blackburn in Herb. Hort. Kew), of which the 

 locality is not known, and the flowers are too imperfect to determine 

 then- exact symmetry. 



Bumelia is one of Swartz's genera. Dipholis was founded by 

 A. DeCandolle in the 8th volume of the * Prodromus.' Both these 

 genera are found side-by-side in the New World, to which they are 

 equally confined. They show no real diversity in structm*e or 

 habit. The presence of albumen is a very inconstant character in 

 this Order, and loses its practical importance from the rarity of 

 fruiting specimens. The total number of species in the two 

 genera is not large, and the argument from convenience, forcible 

 in the nearly parallel case of Lucuma and Sideroxylon, is wanting 

 here. Hence, in the absence of any strong reason to the contrary, 

 I propose to abolish iJijJiolis, even as a distinct section. 



