ee eae ee ee SC lhc ee ee Pp oS ~ ee 
eee ee ele 
5O8 ROBINSON. | 
agreement, that until Conocephalus is very critically studied, Conocephalus gran- 
difolius Warb. may be held to represent OC. erectus (Blanco) F.-Villar. 
Proeris grandis Wedd. was based on Cuming 1730, from Samar, but wrongly 
localized as from New Guinea until Rolfe’s correction:® the specimens were 
staminate. Both from the description, and from the comparisons made at Kew 
by Mr. Merrill, it has been very strongly suspected of being a Conocephalus, and 
this opinion has been confirmed by a recent examination of the type, made at 
Kew, though no opinion is expressed on the reductions here suggested, owing to 
the lack in that herbarium of type material of the other species. 
2. Procris vioLacka Blanco FI. Filip. (1837) 706. — 
Conocephalus violaceus Merr. in (Philip.) Bur. Govt. Lab. Publ. 27 (1905) 80. 
CO. ovatus Tréc. in Ann. Sci. Nat. Bot. IIT 8 (1847) 88. 
C. suaveolens F.-Vill. Noviss. App. (1880) 203, non Blume Bijdr. (1825) 484. 
The second name should stand for this species. 
’ 9, ELATOSTEMA Forst. + 
The original generic description of Hlatostema apparently was drawn, 
according to the statement of the authors themselves,** from one plant, 
and the genus was placed in Monoecia Pentandria. They describe or 
perhaps rather indicate, at least they name two species. The first is 
Elatostema pedunculatum, the second F. sessile, and the only distinction 
given, apart from that contained in the specific names, is that the former 
is pentandrous, the latter tetrandrous. By all authors of the last half- 
century and more, the former is regarded as a Procris, the latter has 
only too much history in Hlatostema as now understood. The generic 
description, “Cor. quinquepartita . . . Stam. Filamenta quinque,” un- 
fortunately leaves no doubt as to which of the two must be considered 
the type of the genus. The figure of the staminate flower shows it as 
pentamerous: those of the pistillate flowers are quite inconclusive, being 
correct for the ovary of either Procris or Elatostema, but showing no 
perianth, which both possess; the figure of the flowering pistillate recep- 
tacle is perhaps better for Hlatostema, that of the fruiting is surely taken 
from the Procris. The name, drawn from the elastic stamens, would be 
appropriate for either. On any logical interpretation of generic types, 
Hlatostema is typified by Procris pedunculata (Forst.) Wedd., which 
would necessitate a transfer of the comparatively few species of Procris 
to Llatostema, while for the genus now known under the later name it 
would be necessary to take up Langevaldia Gaudich. The generic names 
are here retained according to traditional usage. 
Some notes as to the characters of specific importance in the genus may 
be useful, the limits having previously been discussed under Llatoste- 
matoides. Historically, venation has probably played the most important 
part, and there can be no doubt of its value. The great .difficulty in its 
use is the extent to which it may vary upon the same plant, never 
* Journ. Bot. 23 (1885) 215; Journ. Linn. Soc. Bot. 22 (1886) 229. 
*® Forster, J. R. & G. Char. Gen. Pl. (1776) 105, 106, pl. 58. 
Ve 
ere 
Tre 
> AS 
