52 Transactions. — Miscellaneous. 



Art. II. — An Investigation into the Bates of Mortality in 

 Neic Zealand during the Period 1881-91. 



By C. E. Adams, B.Sc, A.I. A., F.S.S., Lecturer on Applied 

 Mathematics, Lincoln College. 



[Read before the Philosophical Institute of Canterbury, 6th May, 1S96.] 



Plates I.-IV. 



The following tables, showing the rates of mortality in New 

 Zealand during the period 1881-91, are deduced from the 

 censuses for 1881, 1886, and 1891, and from the deaths for each 

 year during the period. 



Generally it may be said that the final tables show a com- 

 parison of the numbers living at each age with the deaths 

 occurring at that age. It would have been possible to have 

 computed the tables from one census and the deaths in that 

 census year, but it was considered preferable to use average 

 results, and for this purpose the average population as given by 

 the three censuses has been employed, and the average num- 

 ber of deaths has also been used. In adopting this method 

 there is a greater chance of the final results exhibiting cor- 

 rectly the general mortality of the colony than there would 

 have been had the figures relating to one year only been 

 employed. 



The census has never been taken in the middle of a calen- 

 dar year in New Zealand. In 1881 it was taken on the 3rd 

 April, in 1886 on the 28th March, and in 1891 on the 5th 

 April. This necessitates an assumption being made as to the 

 population living on the 1st July in each year, for the numbers 

 living in the middle of the year have to be compared with the 

 deaths during the year. It was assumed that the population 

 on the 1st July was the mean of the populations on the 1st 

 January and the 31st December. The numbers living in each 

 age-group, as given by the census, were increased in the same 

 proportion as the whole population had increased from the 

 date of the census to the 1st July. This adjustment was 

 made for each of the three censuses, and the total for each age- 

 group found. One-third of these totals gives the average 

 number of persons living in each of the age-groups, as shown 

 in Table A. 



No adjustments were necessary for the deaths. One- 

 eleventh of the total number of deaths in each age-period for 

 the eleven years 1881-91 was taken for the average number of 

 deaths, and the results are given in Table A. 



