432 V. BUNAK 



problem. But whatever the true SR may be in view of the above stated 

 male overmortality an increase of the per cent of stillbirths and abortions 

 in a given group must inevitably decrease the SRi. Simple calculation 

 shows that if the per cent abortions increase from to 8 and to 16 or if the 

 per cent of dead male foetuses increase from to 4.9 and 8.9 the SR of the 

 remaining foetuses in a given group must decrease the initial value (let it 

 equal 125) by 3 units in the first case and by 6 units in the second case. The 

 variation of about 6 units is large enough to explain the fluctuations of the 

 SRi with which we are dealing. If the true SR were exactly known we could 

 decide whether the variations of the SRi are a mere consequence of a single 

 cause of varying prenatal mortality or whether in addition we ought to give 

 consideration to the variability of the male overmortality itself which may 

 diminish the SRi from 115 to 102 when the per cent of overmortality in- 

 creases from 2 to 5. At the present time I see no objection to the assump- 

 tion that the first principle, the variation of the prenatal mortality in gen- 

 eral, is of far greater importance. 



5. But are the variations of early mortality in reality great enough to 

 justify this assumption? An examination of statistical works in this field 

 confirms this fully. The per cent of stillborn infants varies in different 

 countries from 2 to 6 (G. v. Mayr). The percentage of abortions ranges 

 from 9 to 17 (Prinzing) but this number is undoubtedly much greater in the 

 early period when the abortions remain unregistered and are even frequently 

 not noticed by women. 



6. Thus as percentage variations of abortions and stillbirths are really 

 very great and as they perhaps together with the variations of the male over- 

 mortality must inevitably lead to variations in the SRi all other factors which 

 have any connection with the SRi have it only indirectly thereby influencing 

 the prenatal mortality. If the SRi is lower in groups of greater natality, 

 among illegitimate children, older mothers, in urban districts (as compared 

 with rural districts), among the poor classes (as compared with the well-to- 

 do groups), in the period of peace (as compared with after- war and famine 

 times), in certain epochs and nations, — we may conclude that all these fac- 

 tors increase the percentage of prenatal deaths and therefore decrease the 

 SRi. The presumption of a higher mortality in all the above mentioned 

 groups is fully confirmed by statistical investigations made by Lehnhossek, 

 Firks, Manschke, Heape, Tshuprov, Holmes, Hartmann and others. 



7. However it would be a mistake to attempt to generalize on the mean- 

 ing of the above enumerated factors. As the conditions influencing prenatal 

 mortality are not the same in different analogous groups it is utterly fruitless 

 to discuss whether it be rural or urban population, race crossing or normal 



