40 THE PEEIODIC GROWTH OF SCALES IN GADID^E 



commence on the inner surface. The scale has here received, hy the forma- 

 tion of a scale pocket, a connective tissue foundation. It appears that this 

 lower layer gradually differentiates itself so that it becomes similar to the 

 deep lamellar dermis layer, but that close to the scale a layer of cells persists, 

 which continues in an indifferent condition, in so far as the tissue surround- 

 ing it still shows no fibrillar structure. This cell material on the floor of 

 the scale pocket becomes a matrix for the so-called lower scale layer. In 

 its fibrillar structure and the lamellar layering of the fibrillse the lower 

 scale layer agrees with the dermis tissue ; by the total absence of cells it 

 difi'ers from that tissue. AVe cannot, however, assume from the first factor 

 that the lower scale layer owes its origin to a development of connective 

 tissue fibrils, for this is contradicted by the second factor and by the genesis 

 of the layer. If previously differentiated dermis tissue of the scale pocket 

 were included directly in the substance of the scale, there must exist a con- 

 nection between scale and scale pocket in order that both of these may pass 

 directly the one into the other. Secondly, the cells already occurring in the 

 dermis tissue must be found again in the interior of the scale after the in- 

 clusion of the tissue in the interior of the scale. Neither of these occurs ; in 

 the interior of the lower scale layer there is no trace of cells or cell remains, 

 and the scale is separated from the dermis tissue by an indifferent zone. 



From the previous histogenetic facts one gains the following ideas as to the 

 histological structure of scales : — 



The outer layer consists of bony tissue. This layer is homogeneous and 

 is deficient in any special structure, except for a slight lamellar layering 

 (see, for example, Williamson, 1851, plate xxviii., fig. 9, of the Carp). 

 The chemical composition consists of amorphous phosphate of lime and 

 carbonate of lime. The formative cells of this layer are situated chiefly on 

 the upper surface. They represent that which Williamson has described as a 

 membrane, on which the growth of the layer depends. The scleroblasts form 

 the superficial relief of scales. If they become enclosed in their own secretory 

 product, then bone corpuscles are found in a great variety of conditions as 

 regards number and arrangement. On the addition of hydrochloric acid the 

 entire layer dissolves, but somewhat slowly. There is no difi'erence in re- 

 action with this acid between a piece of fish bone from the internal skeleton 

 and the external scale layer; both develop at first rapidly and then more 

 slowly carbon di-oxide. 



The lower scale layer consists of fibres united into bundles, the fibres all run- 

 ning parallel within the bundles. The bundles of one layer again lie fairly 

 parallel to one another, and cross those of the next higher and deeper layers at 

 acute angles. There are usually three different systems to be distinguished in 

 a scale, which cut one another at similar angles. This tissue agrees with tense 

 connective tissue, and especially with that of the deep dermis layer. Con- 

 sequently it appears right, as most authors do, to regard the lower scale layer 

 as the connective-tissue part of the scale, yet no one has placed the peculiar- 

 ities of this tissue in a clear light. This should be done in two directions : 

 firstly, in regard to the adjoining connective tissue of the scale pocket; 

 secondly, in regard to the external scale layer. As regards the first point, 

 development has taught that the lower scale layer does not represent pre- 



k 



