AS AN INDEX OF AGF. 45 



aggregations of ordinary connective-tissue cells. By degrees the number and 

 size of the cells increases, and there arise between them thin connective-tissue 

 fibrils. Sections seem to show, without any doubt, that these fibrils of con- 

 nective tissue in the scale pocket are directly formed at the expense of cells 

 of the dermis, and are their immediate elongation. The developing connective 

 tissue of this intervening part grows, as it were, between the epidermis and 

 the anterior end of the scale, the horizontal position of the last gradually 

 changing into an oblique position ; the posterior end of the scale cuts into 

 the epidermis, and envelops itself in this as in an envelope. The large 

 development of connective tissue, the formation of the scale pocket, is thus 

 the cause of the change in the position of the scale from a horizontal to 

 an oblique position. The connective-tissue pocket itself appears as a newly 

 developed connective-tissue layer, which lies between every two scales, the 

 layer surrounding the scale on all sides (at least on the lower two-thirds). 

 Owing to their oblique position in the skin of the fish, the scales can extend 

 themselves in all directions without hindering one another in their growth. 



When the first layer of the scale and the beginning of the connective-tissue 

 pocket have already been formed, a layer of transparent, quite homogeneous 

 substance appears at the border of the dermis underneath the scale ; this layer 

 contains pear-shaped nuclei, which increase very rapidly during the develop- 

 ment of this layer; nucleoli are, as a rule, not observed, Klaatsch regarded this 

 layer simply as cells of the dermis. According to Ussow, however, they are 

 the lower elements of tlie papilla still remaining behind ; during the entire 

 development of the first layer of the scale they retained the characteristic 

 spindle-shaped form of the basal elements of the papilla ; but at this time 

 they commence to increase in size and number, and between them a trans- 

 parent intervening substance comes into appearance ; further, one finds in 

 longitudinal sections that this developing second layer enters into close con- 

 nection with the first layer. 



In this second layer one also finds nuclei whicli are plainly distinguishable 

 by their size and pear-shaped form. No striation, no fibrils, are at first to be 

 seen in the layer. At a later stage, however, we find a scale, which now con- 

 sists of two distinct layers, the upper already known as the hyalodentin layer, 

 and the lower without cells, but with fine stripe parallel to the upper surface of 

 the scale. How does this striation originate 1 From what are those longi- 

 tudinal fibres of the second layer formed 1 Klaatsch says the following : " In 

 order to understand the structure of the second layer of the scale, one must 

 know the formation of the entire dermis." According to Klaatsch, the dermis 

 at the commencement of development consists of a layer on whose inner side 

 lie the formative cells. The breaking down of the dermis into fibrillw does not 

 depend on those cells, whicli only penetrate into it much later. In the develop- 

 ment of scales, the process which took place in the development of the entire 

 dermis repeats itself, but in lesser degree. But the possibility of a direct 

 appearance in the scale of the dermis of the connective-tissue pocket, already 

 ditrerentiated into fibrils, Klaatsch denies: he asks, "Where then do the 

 connective-tissue cells disappear, for one finds no cells in the second layer 

 of tlie scale (in the trout)." Ussow says, " I think my preparations show 

 fairly clearly that the second layer develops itself anew, and is not merely 



