MR. F. N. WILLIAMS ON THE GENUS SILENE. 9) 
genus, or the species. Thus the presence of chlorophyll is 
essential to the structure of the higher plants, but green leaves 
do not constitute an essential character in the natural order or 
the genus. There can be, therefore, constant characters, which, 
in spite of their constancy, are not essential to the appreciation 
of affinities, or for the grouping of species and of genera, and 
which in fact need not imply that they are in any way related 
to one another. Generally we can, as Nageli very rightly 
remarks, say that there is no character, however important it 
may seem to be, whether morphological or physiological, which 
is of necessity of systematic value. In each grade we must 
rather decide by analogy and by demonstration, what characters 
in this particular grade of subdivision are of systematic value 
in their relation to the grouping of genera or of species. Let 
us then apply the methods of these theoretical “ points d’appui” 
to the group of genera comprised within the subtribe of 
Silenoidee (i.e., the tribe Lychnidew of Braun). If we look 
about, in various series of species which seem to fall into more 
or less natural groups, for that character in the floral organs 
which, constant in certain groups of species, can be used for the 
larger grouping of genera, the one that more especially suggests 
itself, after examining the points of resemblance in several 
series of species, is the structure of the fruit. The following 
points are obvious and worthy of notice :— 
(1) The nature of the fruit; whether a berry (Cucubalus), or 
a capsule. 
(2) The relative position of the carpels to the segments of 
the floral envelope; whether alternate with the lobes of the 
calyx (Uebelinia and Agrostemma, in which also the indicated 
segments of the capsule are isomerous and never oligomerous), 
or whether they are opposite to them. 
(3) The manner of dehiscence of the capsule ; whether circum- 
scissile somewhat in the form of a pyxidium (Drypis), or by 
teeth as many (Coronaria, Viscaria), or twice as many (Silene, 
Melandryum), as the styles. 
(4) The internal structure of the capsule; whether without 
(Lychnis, Coronaria), or with dissepiments (Eudianthe, Silene). 
To these may be added that the carpophore is very short in 
Drypis and UVebelinia, and absent in Agrostemma, and that 
Uebelinia is further distinguished from the latter in the claws 
of the petals being without lamellated margins. 
