HEREDITARY TYPE OF CHARACTER STUDIED. 49 



even in the case of a character so easily controlled as the one here 

 described, it is superfluous further to emphasize how much more 

 necessary this method of personal examination of all family members 

 is when far more vague characters are studied. This is especially true 

 when applied to such characters as physical and mental defects where 

 the diagnosis is difficult even for a trained specialist, not to mention 

 cases where imponderabilia (such as mental abilities, etc.) are the 

 object of the research. 



The consequence is that in human material, where one observer will 

 not be able to examine more than three, or in especially favorable 

 cases four, generations it is in general much more valuable to carry 

 out an exact examination of as many members as possible of these 

 few generations than to try to follow the character farther backwards. 

 The only source of information is here in most cases the more than 

 questionable memory of individuals now living, and the reliability of 

 the data obtained is inversely proportional to the length of the pedigree 

 studied. The information obtained will be more hkely to decrease 

 than to improve the value of the investigation. 



It is unnecessary to mention that this reasoning is especially appli- 

 cable to the study of dominant characters in man. Where recessive- 

 ness is involved the situation is of course different. But the working 

 out of the heredity of recessive characters in man wiU in general be 

 possible only where special sources of information are at hand — for 

 instance, in the case of Lundborg's investigation of the Myoclonus 

 epilepsy (1913). 



2. THE RATIO BETWEEN AFFECTED AND NORMAL INDIVIDUALS. 



As for the numerical ratio between affected and normal individuals 

 in the families here studied, the expectation is that half of the children 

 of the heterozygous individuals should be brachyphalangous, provided 

 no intermarriage of heterozygotes takes place. 



In many pubUcations on human material too much attention is 

 paid to the question, whether the actual ratio is in harmony with the 

 Mendelian expectation. It is obvious that it will be a mere chance if 

 distinct "Mendelian ratios" are found in material where the frater- 

 nities and the number of examined individuals are so small as is the 

 case in this sort of heredity work. Not the numerical ratios, but "the 

 fundamental principle of segregation, is the essential feature of Mendel's 

 discovery." (Morgan, Sturtevant, MuUer, and Bridges, 1915.) 



In this publication elaborate information has been given concerning 

 all members of 15 families, born alive, in which one of the parents 

 was heterozygous for brachyphalangy, the other normal. The numeri- 

 cal ratio between affected and normal individuals in the offspring are 

 shown in table 1. 



