342 STUDIES ON PATHOLOGIC OVA. 



if extended through prenatal life back to conception, as done by Busch and Moser 

 and by Pearson, would tell the truth. Indeed, it could do so only by the merest 

 chance, for the curve of postnatal mortality is based upon a totally different set of 

 conditions. Besides, it undoubtedly is true that the rate of mortality varies from 

 month to month in prenatal much as it does from decade to decade in postnatal 

 life, although probably in a totally different way. It could fail to do so only if a 

 perfect uniformity of conditions obtained throughout the period of gestation. 

 Since this is not the case, the curve of postnatal mortality is of course based upon a 

 totally different set of conditions. 



Upon theoretical grounds, Pearson (1897) decided for a mortality of 37.6 per 

 cent, or 1 abortion in every 2.7 pregnancies, an estimate which Mall (1917 C ) regarded 

 as too low. This opinion of Mall would seem to be confirmed by Taussig (1910), 

 who, from data obtained in 201 gynecological dispensary patients, concluded that 

 there was 1 abortion to every 2.3 pregnancies, a mortality of 43.4 per cent; and 

 also by the present series of almost 700 cases, which indicates a prenatal mortality 

 of 58 per cent. 



If we take the statement of certain social workers or propagandists, who 

 allege that a conservative statement of the total number of criminal abortions 

 annually performed in the United States is 250,000, then on the basis of Pearson, 

 1 in every 6, and on the basis of Mall's earlier estimate, 1 in every 3 interrupted 

 pregnancies, is terminated criminally. However, Taussig (1910), on the basis of 

 histories obtained from 293 patients at the St. Louis gynecological dispensary, 

 reported that only 36 out of 371, or approximately 10 per cent, of the abortions 

 in these women were admittedly mechanical. The histories in the Carnegie Col- 

 lection present similar evidence, but these percentages undoubtedly are too low. 



It should not be overlooked, however, that the surprisingly high percentages 

 of prenatal mortality in the above women undoubtedly do not represent the actual 

 life conditions of the whole population. They merely represent the conditions in 

 women who have aborted. How much the inclusion of all those women who never 

 had aborted would have lowered these percentages it is impossible to say, but one 

 scarcely can doubt that the lowering would be considerable. After a fuller con- 

 sideration of the literature, Schultz, page 183, estimated the prenatal mortality 

 among the general population at 22.0 per cent, a figure somewhat higher than 

 Mall's earlier but considerably below his later estimate. 



Since the women in the present series do not constitute a dispensary group, 

 but very largely also represent cases in private practice, one can not contemplate 

 the amazing prenatal mortality in these women without the profoundest concern, 

 not alone because of its significance upon the birth-rate, but also because of its 

 relation to the wellbeing of these women and the effect of such practices upon 

 public morals. Nor can one be quite certain that the indicated antenatal mor- 

 tality is on the decrease or that it is high in these women alone, for Malins (1903), 

 for example, believed that abortion is more common among the economically 

 more fortunate classes. This opinion seems to be shared also by others. Nothing 

 even remotely like it seems to be known in the case of the domestic animals, 



