A. O. POWYS 
245 
so that the expectation beyond 45 of those surviving age 46 is 24'34 years ; the 
expectation at age 60 is 14'10 and at age 75, 6'68 years*. The complete expecta- 
tions at similar ages computed by the present writer from the New South Wales 
census experience of 1891 are 26"17, 14"37, and 7 02 years respectively. The 
average durations of life determined from the experience of married women in 
the present investigations are 23"05, 13"82, and 7'22 years respectively. It is 
seen that at the first two ages the averages for married women are less than 
the expectations according to both the English and New South Wales life tables 
for the general population of females, although in excess at the later period. At 
first sight it might appear that the married condition was somewhat unfavourable 
to longevity ; but this is not so. We are able to compare, not the actuarial 
expectation, but the mean duration of life of married and celibate men and 
women upon exactly the same basis. The ages at death of each sex are tabulated 
in quinquennial age groups, and by subtracting the married, the numbers of 
bachelors and spinsters have been obtained as shown in Table XII. 
TABLE XII. 
Number of Deaths of each Sex, Single and Married. 
Deaths of Females (1898 to 1902) 
Deaths of Males (1901 and 1902) 
Lge at Death 
All Females 
Married 
Single 
All Males 
Married 
Single 
45 
and under 50 
1181 
1050 
131 
939 
616 
323 
50 
55 
1093 
972 
121 
851 
588 
263 
55 
60 
1218 
1126 
92 
931 
652 
279 
60 
65 
1484 
1362 
122 
1083 
750 
333 
65 
70 
1609 
1491 
118 
1382 
937 
445 
70 
75 
1513 
1423 
90 
1276 
871 
405 
75 
80 
1356 
1281 
75 
830 
606 
224 
80 
85 
1132 
1080 
52 
589 
433 
156 
85 
90 
579 
551 
28 
310 
271 
39 
90 
95 
219 
206 
13 
129 
98 
31 
95 
100 
69 
65 
4 
41 
33 
8 
100 
105 
14 
14 
13 
9 
4 
Totals 
11,467 
10,621 
846 
8374 
5864 
2510 
Note: — The numbers given in the columns headed "Married" do not agree with those given 
in Tables I. and II., as in this table are included those mothers whose ages at death were known 
but the numbers of whose offspring were unknown. The latter were of course excluded from 
Tables I. and II. 
From the data in this table the average durations of married and celibate 
life were determined, the central age of each age group being adopted as the 
mean age of that group. The results for females were as follows"}* : 
* I cannot tell from the table referred to whether the expectations given are complete or curtate, 
t It will be noticed that the averages determined by the adoption of central ages of a quinquennial 
group differ but slightly from those determined by the adoption of single years. 
