AMEEICAN MEN OF SCIENX'E 



805 



considered, there are practically no chil- 

 dren under five, but tho.se under twenty- 

 five who might die before reaching that 

 age make the death rate too low in this 



Table XVII. Deaths ix the Families of Scien- 

 tific Men 



group. It is difficult to see how more valid 

 statistics can be obtained than those sup- 

 plied by scientific men, the order of birth 

 and the dates of birth and death having 

 been given ; yet it is undoubtedly the case 

 that all deaths are not recorded, espe- 

 cially for the brothers and sisters who died 

 in infancy. This error would, however, 

 only emphasize the decrease in the deaths 

 of the children as compared with the pre- 

 ceding generation. The number of deaths 

 among a thousand children born is not 

 directly comparable with death rates per 

 thousand of the child population. When 

 there are seventy-five deaths under the age 

 of five among a thousand infants born to 

 the scientific men, the annual death rate is 

 somewhat larger than 15. 



The decrease in the infantile death rate 



Fall River to 22.4 in Seattle. In or about 

 19O0, the rates for the other nations were : 



'''^ssia 134 5 



^pain 104.1 



.\ustria gg ^ 



Germany y^ o 



Italy 74^8 



England ; 535 



France 47 g 



Sweden 37 g 



New Zealand 23.2 



'he table indicates a small selective 

 deth rate again.st the larger families. 

 "Wen there were one to six brothers or 

 sixers the percentage of deaths under five 

 yers was 14.2 as compared with 15.6 for 

 la?er families; when the number of chil- 

 dm was four or le-ss, the percentage was 

 7 as compared with 8 for families of five 

 o larger. This small difference is prob- 

 aiy due to the coinciding decrease in the 

 bth rate and in the deatli rate. The 

 I'ger families had a higher death rate 

 it because they were larger but because 

 ley were earlier when the death rate was 

 jjfher. Other stati.stics, such as those of 

 abin and Westergaard for Copenhagen, 

 :0w a selective death rate against the 

 rger families. In the statistics with 

 hich we are concerned poverty and ne- 

 ect are almost excluded. 

 In Table XVIII. is shown the relation 

 ?tween the size of the family from whicli 

 le scientific men come and the size of the 

 imily that they have, the scientific men 

 eing divided into three age groups. It 

 .ill be noted that the size of faniilv of the 



in recent j'ears is so remarkable as to be 

 almost incredible, and the existing differ- cientific men varies witli their age at the 

 ences in nations and social classes are ap- ii^^^, being 2.7 children when they are 

 palling. In the families from which our ibove 59 ; 2.3 when they are between 50 

 scientific men come, the death rate for '"f^ 59, and 1.9 when they are under 50. 

 children under five is somewhat over 30 I'hese differences are partlj' due to statis- 

 and in their OAvn families it is somewhat t'cal selection, tlie families of the younger 

 over 15. The rate in the registration area '^i^i^ being not absolutely- complete, and the 

 of the United States in 1900 was 51.9. In barren and smaller families being more 

 the present registration area it was 32.9l'kely to be included. They, however, in 

 in 1911, varying from 41.1 in Massachu- Part represent the declining birth rate 

 setts to 18.6 in Washington, from 74 in'^^ithin a period of twenty years. There is 



