120 



AGE-INCIDENCE OF CAUSES OF INFANT MORTALITY 



are stated by five-year periods, falls with even greater emphasis 

 on the first year when deaths are stated by single years. In the 

 same table the deaths of infants, in the years 1900, 1901 and 

 1905, are given by single months, and here again we find the 

 emphasis of the first year falling on the first month. American 

 statistics do not permit any finer distinction of time than this, 

 but the need of minuter measurements is now generally recog- 

 nized, and a few tables are available containing the required 

 data. 



The Report of the Registrar-General gives the infant mor- 

 tality of England and Wales, in 1907, for the first day of life, 

 for the following six days, for each of the next three weeks 

 of life, and by months to the end of the first year. The state- 

 ments are made in the ratio of the dead to the number of children 

 born alive during the year. From this table I have attempted 

 to construct separate statements of mortality for the first 20 

 weeks and for the first 20 days of life. 



Whatever corrections actual statistics may apply to these fig- 

 ures in future, I do not believe that the significance of first-day 

 (birth-day) mortality will be diminished. In these figures we 

 see the event of birth as an occasion of death. When we consider 

 deaths by separate causes, as we shall do later, we shall find 

 that the antenatal causes of death in early infancy include a sig- 

 nificant number of casualties belonging to parturition, and other 

 numbers attributable to antenatal pathology. We can see per- 

 haps some necessity for studying still births more carefully, and 

 certainly we must be convinced that the gross measurements of 

 time which suffice for the study of adult mortality do not suffice 



