30 GUNNAR DAHLBERG 



figures are far from exact. The conclusion reached by Weinberg 

 that the inheritance of the capacity of twin-birth from one ovum cannot 

 be proved by this method must be considered as well founded. It is 

 to go too far, on the other hand, to assert that the method has proved 

 that an hereditary capacity does not exist. 



Weinberg has again discussed this problem in a later work pub- 

 lished in 1909. He makes use of about the same material and adds 

 no new ideas bearing upon this question. To avoid repetitioHi I shall 

 not discuss this paper here. 



Of later investigations into these problems only those made by 

 Bonnevie and Davenport are of any interest from the point of view of 

 heredity. Other genealogical investigations have had too little material 

 to work with and have been of a more or less casual type; in general no 

 difference has been made, or could be made, between twins born from 

 one ovum or from two. 



Bonnevie has examined the Ringebu families and has found that 

 in certain branches twin-births appear in 7,7 per cent of the births 

 (1151 births with 87 cases of more than one birth at a time). Bonnevie 

 has further found that, according to Weinberg's differentiation method, 

 about 80 per cents of these twin-births arise from two ova and about 20 

 per cent from a single ovum. (The exact figures are not given). »Thus», 

 Bonnevie writes, );it is the twins born from two ova which characterize 

 these branches of the family as »burdened»; it is the two-egg twin- 

 births which we first and foremost must examine in regard to inheri- 

 tance». Bonnevie sitates later on that »Weinberg has also, in his 

 statistical work, abandoned all thought of the one-egged twin-births 

 being hereditary.» 



Twin-births from two ova occur, according to Weinberg, in 0,95 

 per cent of the births in Norway, and twin-births from one ovum 

 in 0,38 per cent of the births (in Weinberg's table the headings are 

 reversed and the decimal point has been dropped. Arch. f. Rassen- 

 biol. 1909, page 333). 



According to Bonnevie 20 ])er cent of the 7,7 per cent of twin- 

 births in the Ringebu families are one-egg twin-births; in other words, 

 1,54 per cent of the births are of this nature. Furthermore, 80 per 

 cent of the 7,7 per cent of twin-births, that is 6,i6 per cent, are from 

 two ova. Thus in the Ringebu families there occur 6,i6 per cent two- 

 egg births as compared with 0,95 per cent characteristic of the popu- 

 lation in general, i. e. something more than 6 times as many. 



With regard to one-egg births 1,54 per cent are found in the 



