ON THE MICBOSCOPIC STEUCTURE OF SHELLS. 
107 
tube becomes more and more elongated, and the new portion, like the bor¬ 
der of the mantle in an ordinary bivalve, will form both layers. But the por¬ 
tion already surrounded by a shelly tube seems only to throw out a very thin 
layer, which lines the interior of that previously formed; and thus we find, 
when we examine a transverse section of the older part of the tube (tig. 46), 
a very thick layer, b (the one first formed), ami a succession of very thin 
lamius, c, at its inner margin. Such a trausverae section sbovvs us a 
series of dwk, somewhat wavy lines, at considerable distances from each 
other, passing in a radiating manner from the iiuier to tlie outer margin. 
oelweeD these are more numerous fainter lines, having a directioo generally 
parallel to the preceding. The principal lines are evidently those along which 
the fracture fakes jilace when the tube is broken across ; and 1 consider them 
to be the borders of large cells of uii irregular prismatic form, made up by 
^ coalescence of smaller cells both in a longitudinal and lateral direction. 
The longitudinal coalescence is very apparent from the multitude of concen- 
tnc lines which cross the preceding at right angles, and which seem to be 
the exact representatives of the transverse strim on the prisms of Pinna 
regard being had to the circumstance that the shell is here cylindrical, 
whilit in the latter case it is nearly flat Each long irregular prismatic cell 
“CT, whose lateral boundaries are marked out by the fainter lines, appears to 
M formed (asin/*/?»««) by the coalescenceof a row of smaller cells directed 
jrom tlie centre to the circumference of the tube. But tlu^re is besides a 
lateral coaKcence of the Mualler prisms into larger, which is particularly 
rvjdent when a section is made in the direction of a chord to the circle, so 
to cut the prisms transversely (fig. 45). The relative character of the 
laimer and stronger linos is then apparent, the former being the boundaries 
m the smaller prisms, a number of which coalesce in a sort of huiidle to form 
he UM^er, wha-ic ootliues are marked out by th(5 latter. At the inner part 
^ the principal layer in fig. 46. the smaller prisms .nro seen (when examined 
t»itb a higher power) to remain isolated, and no coalescence appears to have 
^ thin lamiiuc subsequently formed, 
fif »r ntiantity of animal membrane wJiicb is left after dccalcitication 
tho shelly tube is extremely small, and no truces of cdls can be detected 
II I^ interpret the appearances presented by sections of 
• “ aid of the results previously detailed, I should have hesitated 
wh* “ cellular origin to the structure I have df^scriberl; luit those 
ho wfll take the trouble to follow me, step by step, through my previous 
entertain (I think) llttlo doubt of the. correctness of the view 
ve nuw presented. The result is very interesting from the light which it 
the structure of the calcareous sheath of tho Uelemuite •, the 
the fracture and by sections of which so closely 
able those just described, that I can have little doubt that the mode of 
wiw the same,— with this ditiircncc, that the solid cone was eri- 
tiii^ ^ additions to its exterior, whilst the hollow cylinder of Hi-ptaria was 
i**ternal deposit. The purpost- why the lamime formed aiibse- 
M«enuT to the first production «f the tube, should be of such extreme thin- 
_ IS obviously because the tube would soon be blocked up, if its whole 
bv ^ comijarahlc in thickncs.s to that produced 
prolonged portion, wducli is forming shell for the first time. 
XV’. Gasteropoda- 
tuw ^ same amount of diversity in the struc- 
liave in^r“* r different subdivisions of this group, as that which we 
'^ith among the Couchilerous AoephaUi. There is a certain typical 
