72 MR. BENTHAM ON CARYOPHYLLE.E, 
characterized by the styles, whether 4 or 5, always isomerous with 
the sepals, and alternating with, not opposite to them as in Ceras- 
tium, and as in the old Sagina (or Mænchia) erecta, now transferred 
to Cerastium, of which it has also the capsule. The pentamerous 
Saginas have been proposed by some German authors as a distinct 
genus under the name of Spergella; but the character is purely 
artificial, and not always constant in the same species. 
The four remaining small genera of exstipulate Alsinez have 
each some remarkable peculiarity of structure which has occasion- 
ally suggested their respective removal to some other order, but 
their general affinities are clearly with Alsinez, and I have already 
alluded to the insufficiency of these peculiarities for their removal. 
Thus, Colobanthus has no petals, and the stamens, of the same 
number as the sepals, are alternate with, not opposite to them ; but 
the remaining characters and habit are those of Sagina: Thyla- 
cospermum only differs from the low tufted alpine Cherlerioid 
Arenarias by the calyx forming an obconical tube at the base, 
round the margin of which are inserted the stamens, which are 
thus, exceptionally, very perigynous : Schiedea, with the habit and 
most of the characters of some Stellarias, is distinguished by the 
remarkable position of the petals (or staminodia ?) opposite the pe- 
tals, already alluded to: and Queria, very near some of the smaller 
annual Arenarias, has only one ovule and no petals, and thus passes 
into Paronychiacee ; but as the fruit is a three-valved capsule and 
not a utriculus, I have preferred retaining it among Alsinee. 
The stipulate Alsinee comprise two genera of three or four 
species each—Spergula, Linn., and Spergularia, Pers. ( Lepigone, 
Fries)—differing by their stipules only, the first from Sagina, the 
second from Arenaria. This character, admitted as ordinal by some 
botanists who transfer these genera to Paronychiacez, is rejected 
by others even as generic, as being derived from vegetative organs 
alone. Estimating its value from its practical relation to habit, 
we are induced in this instance to consider it as generic, placing 
Spergula and Spergularia in the tribe of Alsinem, of which they 
have the free styles, rather than with the other stipulate Caryo- 
phyllez which form our tribe of Polycarpez. 
Balardia, Cambess., is a South-American Spergularia with re- 
duced petals and stamens, and has been correctly referred to that 
genus by Fenzl and others. 
The Caryophyllez of our third tribe, PonycanPEx, with free 
sepals and the styles more or less united, are almost all stipulate. 
