312 Gwynne- Vaughan. — On Polystely 
with that of the stem-stele from which the leaf-trace arose. 
In P. Auricula and P. Palinuri each leaf is supplied with 
several meristeles which arise from so many different steles 
in the stem. In other cases a single, meristele arising from 
one of the stem-steles is given off to each leaf, and this 
meristele generally branches as it passes outwards so that 
several are to be found in the petiole. The structure exhibited 
by such petioles may therefore be fitly termed ‘ meristelic. 5 
The median meristele is always the largest and the most 
curved, sometimes forming as much as two-thirds of a com- 
plete circle (P. obtusifolia ). The lateral meristeles become 
smaller and less curved as they are the more distant from 
the median one. 
Van Tieghem describes the petioles of all the species of 
the genus Primula as being astelic throughout their whole 
length 1 ; that is to say, they contain either a single meristele 
or several separate and distinct ones. The only exceptions 
to this statement that I found were in the petioles of 
P. japonic a and P. denticulata , where certain structures are 
invariably present, which in all essential respects are entirely 
similar to normal and perfect steles. These steles were found 
in all the specimens examined of both species, and they 
extended from a point near its base throughout the petiole, 
and often for a long distance along the midrib of the lamina 
also. Of the several vascular strands in the petiole, only the 
three median ones form steles, the lateral strands being 
meristeles similar to those described above. 
These petiolar steles are circular in transverse section, and 
consist of a ring of phloem surrounding a ring of xylem. 
A considerable amount of parenchyma is scattered among 
the elements of the xylem, but separate vascular bundles are 
not distinguishable either here or in the phloem (Fig. 6 ; 
cf. Fig. 5 ). The centre of the stele is occupied by a sclerotic 
parenchyma, scattered among the peripheral cells of which 
are the elements of the protoxylem, generally rather crushed 
1 Ann. d. Sc. Nat. Bot., 7 ser., T. Ill, p. 294; and also Bull. Bot. Soc. de 
Fr., XXXIII, p. 95. 
