«3y Mil. 0. BKNT1IAM ON TEENSTBffiMIACEiE. 



in their ovules, and generally in habit, differ in their stamens (free 

 in the former, 5-adelphous in the latter) and in the capsule 

 (dehiscent from the apex downwards in Bonnetia, from the base 

 upwards in Archyteea). In these respects, as well as in every 

 other, except the free styles, Ploariwm, Korth., agrees so well 

 with Archyteea, that Choisy is quite justified in uniting them. 



To the above genera we woidd add, as somewhat anomalous 

 members of the tribe of Bonnetiece, two with opposite leaves : — 

 Marila, which has the ovules and anthers of Mahiirea, but with a 

 less perfectly convolute corolla and a long narrow capsule ; and 

 Ilaploclathra, a genus we would propose for the two opposite- 

 leaved Caraipas. Choisy indeed considers the leaves as falsely 

 opposite ; but in all the specimens I have seen they are as truly so 

 as in Marila. These two species differ likewise from Caraipa in 

 their long linear erect anthers without terminal glands, in their 

 single erect ovules, and in the endocarp of the capsular valves not 

 separable from the epicarp. 



To the two "families principales" into which M. Choisy dis- 

 tributes the above four tribes, he appends the three " families 

 secondaires," Visneacca?, Ixionanthea, and JPyrenariece. I do not 

 clearly see the precise meaning he attaches to the term " families 

 secondaires," nor the exact position he claims for them in the 

 series of Natural Orders ; but I presume they are intended to be 

 either groups of genera having most of the characters of the 

 " famille principale" to which they are appended, but differing in 

 some one character considered as essential although artificial, or 

 groups naturally distinct from the "famille principale," but too 

 small or too nearly connected to take up an independent position 

 in the linear series. If these views are correct, the course we 

 adopt is in the first case to embody such artificially separated 

 genera with the " famille principale" as exceptional or anomalous 

 members of it, and in the latter case append them to the Order 

 as " genera affinia." 



The first of the above groups, the Visneacece, consists ofAnneslea 

 with two species, and Visnea with one. These are Ternstrcetnieai 

 proper, in their habit, foliage, inflorescence, flowers, fruit, and 

 seeds, with the single exception of the ovary being more or less 

 immersed in the thick torus, which enlarges with the base of the 

 calyx, so that the fruit becomes more or less inferior. This cir- 

 cumstance would seem indeed to remove these plants altogether 

 from Tlialamijlorea ?; but on examination it will be found to be of 

 less importance than il appears at first sight. It may be com- 



