SPECIAL FORMS OF CONJUNCTIYAL INFECTION 137 



According to the older accounts, the frequency of this catarrh in 

 Egypt appears to increase in summer ; the trachoma, which was very 

 frequently also present, became more catarrhal. These exacerbations 

 were associated with the high Nile. The large numbers of flies then 

 present, perhaps, spread the infection more readily (L. Miiller, Lakah, 

 and Khouri). Lakah and Khouri decided that the maximum of the 

 Koch-Weeks infection is in the months of May and June (for gonorrhcea 

 of the conjunctiva from July to October). This has recently been con- 

 firmed by Meyerhof. The maximum, therefore, begins, in the case of 

 Koch-Weeks, before the rise of the Nile. The increase especially 

 affects children ; therefore it is not merely due to the temperature. 

 Perhaps the infective material may remain longer virulent in the warm 

 air. L. Miiller attributed the increase to the greater moistness of the 

 air ; there is some slight correspondence between the two conditions. 



We cannot yet definitely say which of all these circumstances is 

 determinative. The frequency of corneal complications in Koch-Weeks 

 conjunctivitis was greatest in the months of May and June. Meyerhof 

 recently stated that the actual temperature, not the moistness, sun, or 

 dust, is the important factor, and that the Koch- Weeks conjunctivitis 

 first becomes epidemic when the noonday temperature reaches 18 R. 

 (72-5 F.). 



Weeks records an increased frequency for New York in spring and 

 autumn. Morax, after years of observation, could not find for Paris 

 any seasonal variation. Chills which appear to favour a pneumo- 

 coccal conjunctivitis appear to have no such influence on Koch-Weeks 

 conjunctivitis. (There are no grounds for the assertion that the 

 epidemic spread of this form of conjunctivitis runs parallel with the 

 spread of epidemic influenza. The records of Koch- Weeks conjuncti- 

 vitis in Egypt, which is widespread, and still unchanged in frequency, 

 date from a time previous to the last wave of influenza. This is a 

 fact opposed to the identity of the Koch-Weeks and influenza bacilli.) 



Clinical Appearances. 



Descriptions, especially those of experimental inoculations on the 

 healthy human conjunctiva (Weeks, Morax, Weichselbaum, Miiller, 

 Hofmann, Luerssen), agree that the incubation period is relatively 

 short. The time is given by Weeks and Morax as thirty-six to forty- 

 eight hours ; Weichselbaum and Miiller generally observed the same 

 interval ; Hofmann found symptoms twelve hours after inoculation, 

 and in twenty-four hours the full clinical phenomena ; Luerssen 

 found flakes of pus in four hours, and in twelve hours a profuse 



