FROCEEDINGS OF SECTION 0. 



527 



was 122,713, of whom 67,331 were males and 55,382 were 

 females. Arranged according to age the particulars are as 

 follows : — 



Tahle I. 



These figures represent the total number of deaths of children 

 under five years of age for the period under review, and if the 

 number of births per annum had been constant for a series of say 

 fifteen years, the figures liere given, in conjunction with the num- 

 ber of births would furnish all that is needed for constructing a 

 life table for the first five years of life. As, however, the number 

 of births has varied during the pei'iod, it is necessary to obtain a 

 birth basis iu which these variations have been taken into account. 



4. Birth Basis for Computation of Death Rates. 



In accordance with w^liat lias already been stated in section (1) 

 above, the birth basis for age for the decenniura 1901-10 will 

 be half the births in 1900, plus the births for the years 1901 to 

 1909 inclusive, plus half the births in 1910. The division of the 

 number of deaths at age during the decennium by the total so 

 obtained gives the rate of mortality for age 0, that is, gives the 

 proportion of children born alive who would die before reaching 

 age 1, on the basis of the experience of the decennium. 



For determining the rate of mortality for age 1 a new birth 

 basis is required. This niay be obtained from the consideration 

 that the number of deaths at age 1 represents those who have sur- 

 vived the first year of life, but have failed to survive the second. 

 The birth basis for computation in this case will thus extend a 

 year farther back than in the case of deaths at age 0, and wiir be 

 obtained by adding to half the births for 1899 the whole of the 

 births for 1900 to 1908 inclusive and half the births for 1909 A 

 similar procedure will be required for each of the other ages at 

 death, the basis for deaths at age 4 extending as far back ah the 



