88 Oliver . — On Physostoma elegans , Williamson , 
of the convex (abaxial) face of the tentacle that the hairs attain their greatest 
radial extension ; those on the flanks getting shorter and shorter as the region 
of contact between neighbouring tentacles is approached (PI. V, Fig. 7). 
The relations just described are shown on a larger scale in the sketch 
of a portion of a transverse section of a tentacle (PL VII, Fig. 26, where 
only the proximal portions of the hairs are given). It is usual to find about 
seven or eight hairs abreast on the transverse section of a tentacle in its 
thickest part, but where the section is obliquely cut (as in PL V, Fig. 8) this 
number is of course exceeded. 
Although in a majority of cases the adjacent hairs on a rib or tentacle 
have a tendency to stick together, it is improbable that there was real 
fusion for a greater distance from the place of insertion than about 7 5 / u. 
The form of the hairs is subject to some variations. In a fair number 
of specimens they are club-shaped — the hair tapering to its base. It seems 
possible that the clavate type corresponds more nearly with the form 
during life. 
As regards their distribution , the hairs are arranged in longitudinally 
running lines — on the tentacles and seed-body alike. This is clearly shown 
by an inspection of tangential or very oblique sections through the pili- 
ferous layer. In these the successive sections of the individual hairs follow 
one another in linear series, as shown in PL V, Fig. 11, and in PL VI, Fig. 1 6, 
for the tentacles, and in PL VI, Fig. 22, for the body of the seed. In this 
feature the hairs resemble the cells of the palisade-layer of the testa of 
Lagenostoma Lomaxii} 
With the tapering of the tentacles— as at the level of the summit 
of the pollen-chamber — first one and then another row of hairs dies out. 
Those that persist undergo no appreciable contraction as they pursue their 
way to the tip, though the length of the hairs diminishes. The fine 
points in which the tentacles end seem to have been without hairs (Text- 
fig. 5, three of the upper tentacles). 
As already stated, the hairs formed an ample covering to the seed, 
so that its surface during life would at best have been dimly visible along 
the lines of the grooves. 
The contents of the epidermal cells do not always show the same type 
of preservation. The non-elongated cells of the grooves on the seed-body 
and of the adaxial sides of the tentacles have, as a rule, blackened contents, 
but this is not invariable. The hairs, on the other hand, whilst occasionally 
showing blackened contents, are, in many cases, filled with an opaque 
yellow matter, which may be homogeneous or vacuolated, or fragmented 
into granules or larger masses. In the cases in which the secretion does 
not fill the hairs it occupies a peripheral position. The nature of the 
secretion that was present in these hairs must remain a matter of con- 
1 Oliver and Scott, On Lagenostoma Lomaxii . Phil. Trans. B., vol. 197, p. 205. 
