CERTAIN DEMOGRAPHIC BACKGROUNDS FOR POPULATION STUDIES 295 



about the mortality relations within a popu- 

 lation when ages of the components are 

 taken into account. Con\ entionally also a 

 life table starts with a certain sized group, 

 usually 100,000 or 1000, at its time of 

 birth and tabulates the events to which that 

 cohort is subjected, although it may have 

 other variants. Tliis tabulation takes the 

 following form (adapted from Pearl, 1940) : 



is based on the mortahty statistics for white 

 males in continental United States, 1929 to 

 1931. 



Several obvious points can be derived 

 from Table 22 about the age distribution of 

 mortality. The death rate, or more specifi- 

 cally the infant mortahty rate, is high dur- 

 ing the first year of hfe. As shown in the 

 9 1 column, the probabiUty is that of each 



X 1. 



Age in appropri- The number sur- 

 ate units, stated viving at the be- 

 as an interval ginning of the 

 age interval stated 

 in the x column 



dx Qx ex 



The number dying The number dying Life expectation. 



within the age in- in the age inter- Mean length of 



terval stated in val divided by life remaining to 



the X column the number of each organism alive 



survivors at the at the beginning 



beginning of the of the age interval 



interval. The rate 



of mortality 



Other columns are sometimes used, but 

 these are by and large the ones of greatest 

 applicabihty. The plan seems self-explana- 

 tory. The X column first states age; l^ tabu- 

 lates the survivors remaining after death 

 takes its toll; d^ shows the actual number 

 of deaths; ^^ is the rate at which the deaths 

 occur and is usually expressed as a rate per 

 1000 population, or 1000^^, and e^ denotes 

 the fife expectation, mean "after lifetime," 

 remaining once an organism has attained a 

 certain age.* The computations, although 

 tedious in an extended table, have the vir- 

 tue of essential simpficity. 



Illustrations 



Human Population. There is in the Uter- 

 ature a profusion of hfe tables for human 

 populations for many countries and other 

 poUtical units and under many socio-eco- 

 nomic conditions. An unabridged life table 

 is a formidable creation in terms of length. 

 For our own illustration we have chosen a 

 table reported in extenso in Pearl (1940), 

 which we have considerably condensed. 

 This condensation is effected by reporting, 

 after the first ten years of hfe which are 

 taken up year by year, only the last year 

 of a five year span; e.g., 14 to 15, 19 to 20, 

 and so on. This table (Table 22), originally 

 taken from Hill's (1936) study of the 

 United States Bureau of the Census data, 



• These items, especially l^ and d^, were 

 dealt with in a somewhat different connotation 

 in the discussion of physiological and ecological 

 longevity (p. 273). 



1000 births, 62.32 will die before the first 

 birthday. After this initial and rigorous 

 eUmination, with its selective significance 

 (p. 640), there is a consistent drop in the 

 death rate until the age interval ten to 

 eleven years is attained. At this time q^ 

 is at its lowest (1.47). Thereafter it rises, 

 at first slowly and later with increasing 

 rapidity. After age fifty the rate accelerates. 



A hfe table for females from the same 

 sample shows that their mortahty relative 

 to age is lower, or, conversely, their lon- 

 gevity greater, than that of the males. The 

 e, figure for baby girls (0 to one year) is 

 62.67 years, as contrasted with 59.12 years 

 for the males. In their fifteenth year of hfe 

 the females had a hfe expectancy of 53.92 

 years; the males, 51.29 years. In the fiftieth 

 year the figures are 24.19 and 22.25 for 

 females and males, respectively. This is al- 

 most universally true for the human species. 

 It is just as real a sexual difference in popu- 

 lation terms as, say, secondary sexual 

 characteristics are in terms of an individual 

 organism. Nor is it hmited to Homo sa- 

 piens. As the data accumulate, it seems to 

 hold equally true for other animal popula- 

 tions. 



One of the most useful features of the 

 life table is that its columns lend them- 

 selves readily to graphic representation. 

 Conventionally, three graphs are derived 

 from the table: the /, curve, the d^ curve 

 and the lOOOq^ curve. These are plotted on 

 the ordinate against age on the abscissa. 

 The graphs are illustrated in Figures 89, 90, 



