1905 - 6 .] Studies in Immunity : Theory of an Epidemic. 507 
prevent the development of any autumnal epidemic, both of which 
conditions are observed to exist. I do not mean to say that the 
assumption of high-grade infectivity need always occur at the 
same season of the year — that would not be true ; but in the great 
majority of cases it does take place in the early summer. So far 
as my investigations go, this seems the only means of explaining 
the facts. 
If this theory is true, certain conclusions are justified. If the 
number of cases or deaths be given for each week of the year, and 
if an average of a large number of years be made, the amount of 
endemic disease should be represented by a straight line, while the 
epidemic portion should appear on the surface of this in the form 
which represents the characteristic course of such an outbreak. 
On the other hand, on the hypothesis that the seasonal maxima 
and minima are due to different epidemics running into one 
