2S KANSAS UNIVERSITY QUARTERLY. 



snows in early winter, one, and one mentioned the probability of 

 a greater dissemination of germs in dry weather. Five had seen 

 no connection between meteorological conditions and the disease. 

 It is seen from these reports that the greater number believe that 

 both cold and wet weather favor the prevalence of diphtheria, but 

 there is by no means a unanimity of opinion in the matter. 



There seems to be a still greater difference of opinion in regard to 

 the effect of dampness of locality. Fifteen believed that nearness to 

 streams and bodies of w^ater, and damp surroundings in general, 

 favor the occurrence of diphtheria; while fifteen others had noted 

 no influence of such environment on the disease. 



When we come to compare the death rate in different localities 

 in this state and in other states, and consider the question from the 

 point of view of geographical location and climate, we get addi- 

 tional light on the influence of topography and weather. For a 

 summary of the mortality in different counties of this state the 

 reader is referred to the map accompanying this article. The data 

 for this map were all obtained from the State Board of Health Re- 

 ports, and cover the years from 1886 to 1897 inclusive. The upper- 

 most figure in each county gives the total number of deaths from 

 diphtheria reported during the twelve years. The lower figure 

 gives the average annual death rate per 1,000 persons of school 

 age. This rate was obtained by dividing the total number of 

 deaths reported from a given county during the twelve years by the 

 number of years in which the county has given a classified report 

 of the deaths from infectious diseases, and this result was divided 

 by the average number of thousands of persons of school age in 

 the county. The figures from which the average population was 

 estimated were taken from the reports of the superintendent of 

 instruction, and all the years from 1886 to i8g6 inclusive, were 

 averaged. The 1897-98 report was not at hand, but an estimation 

 showed that the population in 1897 does not vary enough from the 

 average to make a material difference in the result. The school 

 population was used instead of the total population because the 

 figures were more accessible, and in the study of a disease prevail- 

 ing largely among children, the school population forms a fair 

 basis of comparison. Besides, the school population bears a 

 nearly constant ratio to the total population when we average a 

 considerable number of years. The fact that the annual average 

 was obtained by dividing the whole number of cases by the number 

 of years the county reported, instead of by the entire twelve years, 

 explains why two counties reporting the same number of cases and 



