Connective Tissue, Nerve, and Muscle. 519 



far greater number of cells than can be seen in the most successful gold 

 preparations. The figures illustrating the structure of tendon usually 

 given by investigators account for only a portion of the cells whose exist- 

 ence is thus proved that portion, namely, which consists of the rows of 

 cells occupying the stellate spaces, and which colour easily in gold and 

 carmine. In longitudinal sections, prepared by the method I have above 

 described, not only are the nuclei much crowded together, but two are 

 frequently seen on the same level, and applied to the opposing surfaces 

 of contiguous bundles. In transverse sections a similar arrangement 

 is found. The nuclei between the bundles are very numerous ; two are 

 often found together on opposite bundles ; and in one stellate space three 

 and four nuclei can often be found at the same level. 



This is clearly a condition to which the so-called division of the 

 nucleus is not applicable. 



If we believe that each of these nuclei represents a cell, the 

 conclusion is inevitable that, in addition to the cells hitherto de- 

 scribed and occupying the centre of the stellate spaces, there exists 

 another and very numerous class of cells applied to the surface of the 

 bundles. 



This effect of hsematoxylin and subsequent action of acetic acid on 

 sections of tendon is perfectly analogous to that similarly produced in 

 sections of cornea. 



The treatment of tendon by the potash solution has seldom yielded me 

 satisfactory results ; but when it has succeeded, I have found confirmation 

 of the inferences I draw from the effect of the saturated solution of 

 haematoxylin. A reference to figure 7 (Plate IX.) shows that while many 

 of the cells isolated by the potash correspond to those found on the sur- 

 face, others are similar to the long narrow cells that cover the fasciculi 

 of fibrillary tissue in the cornea, and do not resemble in shape, even 

 approximately, the superficial elements defined by nitrate of silver. Al- 

 though I have not succeeded, as in the case of the cornea, in reducing 

 tendon to a mass of these cells, I consider it a fair inference that the 

 long narrow cells I have seen are samples of cells that invest the fas- 

 ciculi of fibrillary tissue. 



The comparative difficulty in successfully treating tendon by potash 

 is probably due to the denseness of its structure. 



It is in regard to the branched cells, which I hold to be the analogues 

 of the branched cornea-cells (corpuscles), that the important fact demon- 

 strated by Spina, that it is on the surface of the cells that the fibres of 

 elastic tissue are formed, specially applies. 



In the centrum teudineum of the rabbit the continuity of the flat 

 cells, which in silver preparations are considered to indicate lymphatic 

 vessels, with cells covering the fibrillary substance can be shown to a 

 greater or less extent, according to the success which has attended this 

 difficult manipulation. That it often succeeds in patches, is shown by 



