THE PLACENTAS IN THE TRIBE ALSINEJX. 497 
The development of the capsule of Stellaria Holostea closely 
resembles that of S. media; but the young pocket-like loculi are 
wide and open, the central column is short and thick ; the ovules 
are four in each loculus, and, when fully developed, seem to arise 
from the upper extremity only of the slightly enlarged central 
column. As their funiculi are short, the cluster of seeds is con- 
fined to the base of the large inflated capsule. 
The capsule of Stellaria uliginosa during flowering is enclosed for 
one third of its length by the calyx-tube. It is long and narrow, 
and contains about seventeen ovules; the slender central column 
occupies less than half the length of the capsule; from its apex to 
the roof of the chamber extend the loose-textured dissepiments, 
which have become detached from the lateral walls of the capsule. 
The development of the capsule of Cerastium quaternellum is 
much like that of Stellaria media (Pl. XXXIV. figs. 9-14). The 
ovules appear on the upper part of the central elevation when 
the walls of the capsule have risen to about two thirds or four 
Jifths of its height. The carpels are usually four; the ovules 
vary from four to twelve in each loculus. 
In Cerastium triviale and C. glomeratum the central column is 
very conspicuous as soon as the staminal whorls have been formed, 
and grows rapidly after the formation of the five carpellary pro- 
minences. These acquire the shape of five very shallow, open 
pockets. The dissepiments are strongly developed ; at the base 
of the pockets they are broad, but become gradually narrow and 
less marked, till they cease near the apex (Pl. XXXIV. fig. 2). 
When the first ovules appear between the uppermost ridges of 
the dissepiments, the capsule-walls have not risen to half the 
height of the central column. At a later stage they rise above 
it, and the dissepiments project into the upper part of the cavity 
of the capsule as five free plates (Pl. XXXIV. figs. 4&6). About 
sixteen ovules develop in each loculus. 
Payer, in his * Organogénie Végétale, regards the placentas of 
all Phanerogams as formed from the axis, or from a branch of the 
axis. The margins of the carpellary leaf he considers derive their 
power of bearing ovules from being overlain with outgrowths of 
the branched floral axis. Holding this view, he would have no 
difficulty in explaining that, in the tribe Alsinex of the order 
Caryophyllex the ovules are developed entirely from the floral 
axis itself; and that in the tribe Silenez the ovules in the lower 
part of the capsule are developed from the axis itself, and in the 
: 
