ON THE MICROSCOPIC STRUCTURE OP SHELLS. 115 
rangemeDt, which is represented in figs. 53 and 54-. The former is a section 
of the compact portion of this formation, and shows the assemblage of large 
calcigerousshells ofwhich it is composed,which huveassumed a polygonalform 
from the pressure of their edges against each other. The latter is a section 
of the areolar or cancellated portion, in which vro. see an apj)carance as of 
aggn^ations of flattenetl or discoidal cells, of circular outline, partially over- 
lying <H)e another, very much after the manner of groups of blood-discs in 
a drop of Wood. This last condition is interesting, as showing, nndcr a new 
aspect what has been already pointed out as being probably the original form 
of the calcigerous cells in general. For tliese fiattened colls, when aggre¬ 
gated in regular piles, would give rise by their coalescence to long cylinders 
Barked Iw close transverse stri® j and these cylinders would bfcoine prisms 
when pressed against each other laterally, just as the diacoidal colls, which 
arc maid in the loose cancellated structure represented in fig. 54, are poly¬ 
gon^ when pressed together into a compact mass, as in fig. 53. 
9S. A singular variation of structure prcsetiU itself in Ilalioiu. The 
calcified lamina of this shell alternate with platisi of u Iirow n horny sub¬ 
stance, much resembling tortoise-shell in iht ap[)«‘anince; and when the cal¬ 
careous matter has been dissolved away by dilute uoid, these horny plates 
may be easily detached from each other, the hascnicnt-niembrann of llic ad¬ 
joining calcilietl lamina remaining adherent to one side of each of them. Of 
the|)€cu]iar disposition of this bafiement-merabrane, on which the iridescence 
of the calcified layers depends, 1 have givcu an account in my former Report 
(§ 25), In the particles of the liorny jilates themselves, I have notliecu able 
to detect any dcHuitc arningcmcnt; they are frwjuently marked, however, 
by impressions from the cells that lie upon them, which might be mistaken 
lor imlicatioDS of structure in themselves. In immediate contact with them 
a thin layer of large colls of very peculiar aspect, shown in fig. 55. The 
nam-ous lamin® of l/aliotis, when examined with a sufficiently high mag- 
nifyiDg power, present indications of a minute cellular structure, such as I 
have not observed in the nacre of Bivalves. The colls aro of a long oval form, 
their short diameter is not above l-5(KX)th of an inch. Their boundaries 
W i^y parts are very indistinct, or even disappear altogether; so that every 
pradatloncan ho traced, from ihc obviously-cellular arraugemeut shown in 
%■ oC, to the homogeneous aspect presented by the nacre of bivalve shells, 
teuce I should be disposed to draw the same inference with respect to the 
6'trucliire, as in regard to other forms of apparently-homogcneous 
* elkubsiancfl (§67), namely, that, like them, it was originally fomicd upon 
a oriluLu plau, but that the colls subsequently coalesced, their boundaries 
“‘sapp.jariiig. The appearance shown in fig. 5C, a is that which the cells 
present when the section passes through them transversely; at b the section 
traveled diem obliquely; and at c are shown tlieir natural terminations, 
where they crop out upon the surface of the shell. The same cellular struc- 
tire IS apparent in the decalcified membrane, whoii it is carefully examined 
“ndera high magnifying power; in fact, I know no better example of the 
BUnet cxUfence of cellular structure in n membrane which at first seemed 
Pfuccily homogeneous, and properly entitled to the designation of primari/- 
or tluan is alforded by the decalcified nacre-membrane 
"aoVrfis gplaideng, when submitted to the retjuisite tests. 
In the remarkable calcareous plater, or teeth, which arc imbedded in 
ae nimculo-tendiiious8tib8tao«:e of the gizzard of Bulla. I have found a 
and regular cellular arrangement, as shown iu fig. 57. 
ahl ^^‘uobranchuUa the shell does not present any consider- 
e aeparture from the ordinary type; the three layers being usually pre- 
l2 
