476 MIss T. L. PRANKERD ON THE STRUCTURE OF 
show the central cone touching the outer wall (Pl. 28. figs. 11-15, and 
particularly fig. 17), and in such cases it is a matter of great difficulty to 
determine whether there is mere contact or actual continuity of tissue. 
The subject is referred to later (p. 484). 
As in L. Боташ, the wall of the lagenostome is bevelled to correspond 
with the erenulations of the canopy, but this bevelling is not found either on 
the neck or at the base of the organ. The wall is one layer of cells thick, 
and, when well preserved, exhibits a very interesting structure. 
А beautiful transverse section (К. 755) * from Owen's College, Manchester, 
kindly lent by Prof. Weiss, shows that the radial walls are enormously 
thickened in a manner reminding one of an exaggerated endodermis (РІ. 24. 
fig. 29). The cells measure 16-18 р tangentially, and the thickenings are 
about 12 р, sometimes reaching as much as 16 д broad, so that they are 
often laterally in contact. The thin outer walls of the cells are for the most 
part strongly eurved (usually outwards, but sometimes inwards), so that the 
possible tangential extension of the cells is considerably greater than the 
above. These curved outer walls are well shown in transverse section in 
PI. 24. fig. 30, and in longitudinal section in РІ. 28. fig. 9. The inner walls 
are rarely present. Ву careful focussing it may be seen that the tongue- 
shaped radial thickenings are frequently double, and a surface view of the 
wall, obtained from the specimen 1435 (2) in the Williamson Collection 
(text-fig. 3, p. 477), explains this appearance. It is here seen that the cells are 
elongated, with tapering ends fitting closely into one another, and that their 
walls are strengthened by thick bars which do not quite reach the middle 
partition. This double nature of the bars (obliterated in many sections, but 
indicated in В. 45), together with the shape of the cells, explains why the 
thickenings so frequently seem two-lipped in transverse section, and also 
their variability i in size and proximity, which is largely due to the varying 
levels at which the cells are eut. One of the most interesting points in the 
structure of the seed is the fact that the epidermis of the | lagenostome is 
directly continuous with, and of the same character as, that of the plinth. 
This forms a circular belt about } mm. broad, curving slightly from the base 
of the lagenostome to that of the c 'anopy, but is not hev elled as in L. Lomaæii. 
Owing to its small size it is not so fr 'equently cut as many parts of. the seed, 
and when it is included in the plane of section this is not usually the best 
to exhibit its structure. 
The only surface view of the plinth is in R. 20 b (2), Но. 19. В. 49 
(Pl. 24. fig. 32) is a very good preparation for demonstrating that the 
epidermis of the plinth was similar to that of the lagenostome. In the first 
place, it is easy to be sure of the exact plane of this section, because it is 
* This specimen gave the key to the interpretation of the structure described, but it mav 
be partially made out in several of the U niversity College, London, slides, e. g. В. 206 (1), 
fig. 16. 
