ON THE ETHNOLOGICAL SURVEY OF CANADA. 



447 



It appears, therefore, that the result is not quite certain, since the 

 error is great as compared to the average difference. Since for later-born 

 children the errors of the average are very great, I have not carried out 

 the calculation. I have calculated the same differences, and their mean 

 errors, for the statures of girls : — 



Differences between the Average Statures of Girls and the Statures of 

 First-horn Girls, and their Mean Errors. 



+ 9± 6-7 



+ 4± 7-2 



- 8± 8-3 



+ 4 ±10-3 



+ 5-3± 1-9 



This result is much more certain than that obtained by means of the 

 measurements of boys. "When we combine both we find that the difference 

 of stature between the average of all the children and the average of the 

 first-born children is in favour of the latter. The amount is 3-6 mm., 

 with a mean error of 1'2 mm. It is therefore certain that first-born 

 children are somewhat taller than later-born children, but the amount of 

 the difference is not definitely known. 



It is of interest to investigate the constitution of families. I have 

 done so by recerding for each year the number of children according to 

 the order of their birth. 



Total Niimher of Children examined according to the Order of Birth. 



From these data we can obtain an insight into the constitution of 

 families in Toronto. The difference between the number of first- and 

 second-born children shows the numbers of mothers having one child 

 only ; the difference between the second- and third-born children gives 

 the number of mothers who have two children, &c. In this manner the 

 following table has been obtained : — 



It is of interest to compare the number of children according to the 

 order of their birth in various cities. I have tabulated for this purpose 



