3 o 8 G wynne- V aughan . — On Polystely 
in a paper written by Kamienski h Believing that these 
variations may also be of some value for the purpose of 
indicating the manner in which the anomalous Primulas came 
by their remarkable structure, I purpose describing such 
cases as appeared in the species that I have examined. 
To begin with, P. japonic a, obUisifolia , denticulata , and 
involucrata are described by Van Tieghem as being gamo- 
stelic with a few steles only, and these arranged in a single 
circle 2 . The steles are two to four in number, elongate or 
arcuate in form, and they anastomose into a network, the 
meshes of which correspond to the points of insertion of 
the leaves. Although it is quite true that such a structure 
is frequently to be met with in these plants, yet at the same 
time it is by no means universally present ; indeed, in the two 
first species, it can hardly even be described as generally 
characteristic. For instance, in the ten or more mature 
specimens of P. japonica that I examined, I did not find 
a single case in which the so-called steles maintained, 
throughout the plant, the essential features of true and perfect 
steles. On the other hand, both in P. japonica and in P. obtusi- 
folia , the steles present in the stem are usually found to be 
defective or imperfect. These imperfect steles are at once to be 
distinguished from perfect ones by the more or less complete 
absence of those internal vascular bundles which are invariably 
to be found on the inner side of every perfect stele. In fact, 
they consist merely of a number of collateral vascular bundles 
with normal orientation, fused together laterally into so many 
groups each of which is surrounded by its own endodermis 
and pericycle (Fig. i). The endodermis is clearly demon- 
strable right round the group, on the inner side as well as the 
outer. The pericycle on the inner side of the group is 
represented by 1-3 layers of parenchymatous cells directly 
intervening between the protoxylems of the vascular bundles 
and the internal endodermis ; on the outer side it is greatly 
complicated by the development of a radiciferous network 
1 Abhandl. der Naturf. Gesellschaft zu Halle, XIV, p. 143, 1878. 
2 Ann. Sc. Nat. Bot., 7 ser., T. Ill, pp. 300 and 305. 
