ZOOLOGY AND BOTANY^ MICEOSCOPY, ETC. 37 



bundles cannot be regarded as taking any active part in the expansion 

 of the cbromatophores. Their importance lies in their being the 

 agents for the fixation of the pigment-cells in the layers which they 

 occupy. The contraction of the chroma tophore is due rather to the 

 elasticity of the capsule and the contraction of the basal cells ; the 

 expansion of the cell seems to be due to its protoplasm. 



The layer of iridocysts is formed of a series of plates formed from 

 the primitive connective-tissue-cells ; they have a central nucleus, 

 and are made up of a number of rods. Where the iridocysts are only 

 arranged in one layer they are more closely packed. 



In the developmental history of the layer of cbromatophores the 

 first point is the conversion of certain cells into pigment-cells ; around 

 these other cells become grouped ; the intermediate cells then increase 

 in number, or form fresh pigment-cells, and new " common cells," 

 which, in their turn, are capable of proliferation. The pigment-cell, 

 once constituted, grows rapidly, though the nucleus remains of the 

 same size all through the period of growth. The limiting cells 

 divide, and so increase in number till there are from twenty to thirty 

 of them. The radial bundles are formed by the striation of the cells 

 of the peripheral reticulum. 



Development of Gills of Cephalopods.* — L. Joubin describes the 

 gills of Sepia officinalis as commencing under the form of two small 

 rods placed symmetrically in relation to the antero-posterior plane, 

 and in the middle of what will become the posterior wall of the 

 pallial cavity. The bud, which is primitively due to an outgrowth of 

 the epithelial layer, soon elongates, becomes rounded at its tip, and 

 attached by a wide base. The bud then flattens, and its hinder face 

 becomes applied to the visceral mass, while the anterior is still 

 covered by the mantle. 



One and then a second fold, and afterwards others, appear on the 

 bud, and these form depressions on one surface which correspond to 

 elevations on the other. Although these folds increase in number 

 they do not occupy the whole face of the young gill ; along its edge 

 there remains a space, and in the anterior of these an efferent vessel 

 is developed, and in the posterior the special branchial gland. 



Any one elevation may be regarded as a semicircle formed of 

 three parallel arcs of cells. If these be fixed at their extremities, 

 and if the arcs were to grow equally, we should soon have a large, 

 deep, and more or less conical cul-de-sac. This is not, however, what 

 happens. The cells of the median layer increase in number, and 

 push forwards the epithelium of the convex surface, while that of the 

 concave remains unaltered ; the middle layer soon forms a stratum 

 invested on either side by the convex epithelium ; the cells of this 

 stratum, which at first touched one another, soon become separate, 

 and give rise to intermediate lacunae and vessels. 



Each of the layers thus formed gives rise, in its turn, to a series 

 of transversely disposed elevations, which soon form hollow out- 

 growths, this time on either side. Finally, in the adult there is a 



* Comptes Kendus, xcvii. (1883) pp. 1076-8. 



