196 Correlation of Cancer and Diabetes Deathrates 
of diagnosis of either disease in this recent period, and the general sweep of the 
plotted curves seems to show that the source of the increase whatever its nature 
has been continuous since the middle of last century. 
In the following investigation I have correlated the deaths from cancer and 
diabetes for the male population only. I had intended to deal also with the female 
population, but the results for the male were so conclusive, that I did not think 
it needful to go further. But the idea of dealing with the female deathrates led 
me to select as standard population the total male and female population of 
England and Wales as given by the 1901 census. Of course the choice of the 
Diagram V. Associated Inflexion Points in Corrective Factors and % Males under 45. 
1871 
1911 
Year 
Diagram illustrating that the inflexion point in the corrective factor diagram (IV) corresponds 
to actual changes in the age distribution of the population. 
standard population to obtain a corrected deathrate is arbitrary, but if female 
deathrates are to be compared with male, we must take a common standard for 
both, and this is why the combined male and female age groups of 1901 were taken 
as the standard population classes. 
A word must be said as to the manner of correcting the crude deathrate for 
age groups. Actually we only know the age groups accurately for the census 
years ; the age groups for intervening years are more or less guesses not beyond 
suspicion. Accordingly it was considered best to obtain the corrective factors 
for the crude deathrates in each census year, then to plot these to the year, draw 
a continuous curve through the plotted points by aid of a spline and read off from 
