in the Genus Primula . 
3*5 
returns to its collateral structure, and enters into the vascular 
circle as a horseshoe-like meristele. Later, passing down 
the stem, the now cauline portion of the leaf-trace once more 
develops internal bundles, and in connexion with the cauline 
portions of other leaf-traces with which it has fused, it becomes 
once more a complete and perfect stele. Finally, it divides 
into two branches, which eventually fuse with the leaf-traces 
of the lower leaves as they take their places in the vascular 
circle. 
Bearing in mind the manner in which the leaf-trace enters 
the vascular ring between two steles already present, as 
described for P. denticulata and P. involucrata , and also the 
general course of a leaf-trace as followed out above, it seems 
possible to regard the whole vascular system of these plants 
as consisting of the cauline portions of the leaf-traces. They 
pass in, arrange themselves in a circle, and having passed 
some distance down the stem, divide into two branches, which 
insert themselves on the entering leaf-traces of the lower 
leaves. The fact that the leaf-trace has become a stele during 
the cauline portion of its course may be regarded as incidental 
only. For such an incident may, and does, take place at 
other points of its course where no very great importance 
can be attached to its occurrence. Moreover, the examination 
of P . japonica and P. obtusifolia has shown that whether the 
trace becomes a stele or not, even in the stem itself , is just as 
much a matter of chance, or rather of convenience. 
Regarded in this light, the existence of the imperfect steles 
in P. japonica , obtusifolia , involucrata , and denticulata is easily 
understood. For the appearance of the internal bundles on 
the inside of the cauline portion of the leaf-trace may either 
be more or less delayed (P. involucrata , P. denticulata :), or 
they may sometimes be altogether dispensed with (P .japonica^ 
P. obtusifolia ). Again, in the two latter plants their presence 
in certain varying regions of the stem, and their absence in 
others, is also to be expected, for the leaf-trace may be 
regarded as potentially capable of producing internal vascular 
tissue at any point throughout its course. Their actual 
