CHANGES IN NON-PREGNANT UTERUS 95 



These " sub-epithelial hsematomata " have been noticed especially in 

 the procestrous bitch. The walls of the vessels in the muscular 

 layers do not give way. 



Eventually the extravasated blood corpuscles (or, at any rate, 

 the majority of them) make their Way into the cavity of the uterus, 

 and thence to the vagina, where external bleeding is observed. This 

 is especially noticeable in the case of the bitch," with which, as already 

 mentioned, external bleeding may last for as long as ten days. The 

 bleeding is accompanied by an increase in the mucous secretion. At 

 about the same stage goblet-shaped cells are frequently observable in 

 the glandular epithelium, and it is suggested that these are in some 

 way connected with the secretory activity of the glands. 



It is probable that destruction of the superficial epithelium occurs 

 normally to a greater or less extent both in the bitch and in the 

 ferret. Epithelial cells have been observed lying free in the uterine 

 cavity, while, in some sections, places have been noticed where the 

 stroma presented a raw edge, having been stripped of its epithelial 

 covering. In the bitch a layer of flattened stroma cells may some- 

 times be seen in close attachment to the epithelium during the 

 process of denudation. In the ferret it would appear that the 

 destruction may occasionally be severer, but it is thought that this 

 is exceptional. It has been pointed out, however, that a comparison 

 between the thickness of the uterine wall (and conversely the size 

 of the uterine cavity) in ferrets killed at the commencement of the 

 recuperation period and during the period of rest, is very suggestive 

 of a definite removal of stroma as well as of epithelium in the 

 process of destruction. 



Polymorph leucocytes have been observed in abundance at this 

 stage in the bitch's uterus, both in the stroma and also in the cavity, 

 and large mononuclear leucocytes (hyaline- corpuscles), containing 

 pigment derived doubtless from the extravasated blood, have also 

 been seen to pccur. Large cells, with faintly staining nuclei of very 

 considerable size and conspicuous nucleoli, have been noticed at rare 

 intervals lying in spaces in the stroma tissue of the procestrous bitch. 

 The origin and significance of these cells are not known. 



There is no blood-clot formed in the uterus, either in the bitch 

 or in the ferret. 



(4) Period of Recuperation. — The new epithelium in the bitch is 

 first seen as a layer of flattened cells which bear a resemblance to 

 the cells of the stroma. Its manner of formation is an open question, 

 but it would seem probable that it is derived mainly, if not entirely, 

 from the remaining cells of the old epithelium, or from those of the 

 glands. It is just possible, however, that in certain places the 

 epithelium may be renewed from the underlying stroma tissue, as 



