346 STUDIES ON PATHOLOGIC OVA. 



In all groups except the first two, composed of women having none or but 

 one child, the average number of abortions lies between 2.9 and 3.6 per woman. 

 Moreover, this ratio is practically the same in the groups having borne 2, 4, and 

 even 6 children, but since the total number of cases involved in this table is only 

 585, the number in each group is necessarily small, being over 100 in the first 

 three groups only. 



Upon comparing the total number of previous abortions suffered by 697 

 women with the total number of children borne by them, we find that there was 

 0.84 previous abortion for every child. However, if the 697 abortions represented 

 by the specimens which brought these women to our notice be included, then the 

 ratio becomes 1.3 instead of 0.84 abortion per child; yet Malins, on the basis of 

 2,000 selected private and hospital cases, found but 1 abortion to every 5 children. 

 A similar proportion is recorded also by Hellier, who, on the basis of 6,974 births 

 and 1,288 abortions in 1,800 married women, found one abortion for every 5.5 

 children. Keyssner (1895), on the basis of 9,381 births and 1,194 abortions, found 

 a ratio of but 1 abortion to every 8 births. Although the statistics of Keyssner 

 were taken from the clinics, polyclinics, and gynecological journals, and those of 

 Malins from selected private and hospital cases, one is at a loss to explain the 

 great disparity between them and those in the present series. 



In this series of 697 women with 1,351 children and 1,843 abortions, there were 

 1.3 abortions for every child or 1 abortion for every 1.7 pregnancies. This result 

 differs somewhat from that recorded by Taussig for the cases in the St. Louis 

 Gynecological Clinic, which was 1 abortion for every 2.3 pregnancies. The lack 

 of correspondence between the estimate made by Taussig and that in the present 

 series is not surprising, for the Carnegie series is fairly representative, being com- 

 posed to a considerable extent of material obtained from the general practitioner. 

 However, it is surprising to find that this ratio of children to abortions is lower in 

 these women than in the cases from a dispensary, unless we accept the opinion of 

 those who hold that abortion is more common among the economically more 

 favored classes. 



The relative constancy in the ratio of abortions to children in families with 

 3 to 7 children seems to imply that whatever the factors responsible for the inter- 

 ruption of pregnancy, they act with unexpected regularity in women of widely 

 differing ages and with decidedly different reproductive histories. This would 

 seem to imply that in these women there is no tendency to limit the family to any 

 particular number of children through interference with the gestation, for were 

 such the case abortion should be more frequent in connection with the particular 

 number to which it is attempted to restrict the size of the family. This could fail 

 to be true only if we could assume that this supposed limitation in the size of the 

 family were due to causes other than interference with the gestation. 



Only 29.1 per cent of the 607 women whose ages were given were less than 25 

 years old, but, as shown in table 17, 56.1 per cent were less than 30 and 77.4 per 

 cent less than 35 years. In the series of Stumpf this was true of 23.3, 51.7, and 

 71.8 per cent, respectively. Upon considering the relation of the different age 



