ENDOGENOUS INFECTIONS 377 



It is certain that we do not have to do with embolisms of the larger 

 branches, for the large embolic masses would also settle down in the 

 other branches of the carotid. The eye rnetastases, however, are often 

 the only ones in the whole internal carotid region. The deposition 

 takes place more in the finest capillaries, and is due to infectious 

 material in the blood in the finest state of division. 



As the question appears to be one of the ernbolic plugging of the capillaries, we 

 must consider their calibre ; this factor alone cannot be the cause of their predilec- 

 tion, as Battler's capillaries in the uvea are just as fine. We must conclude that 

 septic conditions in the blood first cause toxic changes in the retinal vessels, which 

 react freely to many changes in the state of the body (e.g., retinitis albmninurica). 

 Fatt}- changes occur in the endothelium (Ponfick) ; hyaline and laminated clotting 

 results, leading to stasis, and thus the deposition of the bacteria from the circula- 

 tion. As there is a tendency in the retina to haemorrhages, it is obvious how, with 

 a toxic-septic bleeding taking place, the organisms can pass into the tissues. 



In a child dead from sepsis I was able to microscopically demonstrate a retinal 

 haemorrhage filled with Streptococci ; this would have formed the commencement 

 of a rnetastatic inflammation from the extravasation of blood containing cocci. 



All previously existing disturbances of the circulation, including traumatisms, 

 will act as localizing agents in the eye (see the researches of Panas 1 quoted above). 



Individual variations in the susceptibility are possible, as also are slight variations 

 in the action of different strains of the same bacterium. 



In individual cases it is difficult to decide which of the possibilities 

 discussed is the cause of any particular isolated rnetastatic ophthalmia, 

 where in one eye, or perhaps in both, an endogenous suppuration has 

 occurred, without any other part of the body being affected. From 

 the literature up to 1894 I have collected over thirty of such cases 

 (loc. cit.). 



In many of this latter class of case I have often noticed that the 

 general condition is very little disturbed ; perhaps it is quite normal 

 at the time of the ophthalmitis, so that the case appears to be one of 

 ' spontaneous panophthalmia.' 



The predilection of the non-purulent endogenous inflammations for 

 the eye is shown to a far higher degree. In these we are often unable 

 to demonstrate any other lesion in the body. The extent to which 

 these may be bacterial metastases has already been discussed on 

 p. 364. 



1 According to Leber and Erahnstoewer, necroses, followed by intense inflammatory 



symptoms and phthisis bulbi, can lead to the localization of circulating bacteria : such 

 destruction of tissue occurs in an iutra-ocular tumour, not uncommonly in conjunction with 

 vascular thrombosis. These endogenous infections will explain why sympathetic ophthalmia 

 is occasionally associated with such ati'ections. We have no cultural proof of this ; the 

 purely macroscopical findings which Deutschmanu records in these eyes are not absolutely 

 positive, and, moreover, they have no relation to sympathetic ophthalmia, whose cause 

 is not yet demonstrable by the microscope (cf. literature in Schottelius's Dissertation, 

 Freiburg, 1904 ; also Nesse, K. M.f. A., December, 1906, xliv.). 



