SKELETO-TROPHIG TISSUES AND COXAL GLANDS. 135 



arranged in rows of three^ six, or even eight, parallel with the 

 adjacent lines of fibrillation. Some of the cells are isolated, or 

 iu pairs, whilst in the deeper parts of the tissue, in place of 

 rows of such cells, we find irregular clusters, or groups of cells 

 (four to eight). The relative abundance of matrix and cells is 

 shown in PI. VII, fig. 3, which represents a piece of the tissue 

 where the fibrillation is not very strongly marked, and not all 

 in one plane. 



The nuclei of the cells staia deeply when a piece of the 

 tissue is removed from alcohol and placed in borax carmine ; 

 at the same time the matrix also is strongly, but not so deeply, 

 stained, whilst the protoplasm surrounding the nuclei does not 

 stain at all. 



Owing to the fact that the entosternite is a sort of central 

 tendon in which the tendons of several muscles converge, it is 

 possible to obtain sections of difi'erent character from different 

 parts of it. By cutting the divergent processes parallel with 

 their long axes, we can obtain sections showing strong fibrilla- 

 tion of the matrix parallel with the long axis of the process, and 

 with rows of cells all strictly parallel to the same lines 

 (PI. IX, fig. 2). Such sections most closely resemble Vertebrate 

 tendons, such as those of the rat's tail, but the cells do not 

 appear to be flattened, as in Vertebrate tendon, and the matrix 

 is not fissured so as to form distinct bands, but rather indicates 

 a tendency to such fissuring in the fact of its fibrillation. A 

 comparison with white fibrous cartilage of mammalia is, per- 

 haps, more nearly justified. 



Again, by taking sections more deeply, we may obtain pieces 

 of tissue in which the lines of fibrillation from various processes 

 of the entosternite are converging and crossing one another. 

 Very complicated grouping of cells and fibres results from this 

 convergence. 



Lastly, in the more central parts of the entosternite we find 

 the matrix no longer showing a tendency to fissuring so as to 

 suggest the descriptive term " fibrillated," but we find actual 

 branching fibres developed in the hyaline matrix, the branching 

 fibres being of denser substance than the rest of the matrix 



