INHERITANCE OF ALLERGY 315 



type of intelligence (see below), the information obtained is looked upon 

 as very trustworthy. 



We realized at the outset the necessity of getting full pedigrees of both 

 maternal and paternal sides of the families. The work was done as far as 

 possible with methods commonly employed in genetics. This is in contrast 

 with methods sometimes used in studies of this nature in which records 

 are kept only of the affected parents of the individuals being studied, omit- 

 ting the contribution to the inheritance made by the parent who may not 

 happen to show the character. 



Data as complete as could possibly be obtained were collected from fifty- 

 five families, both maternal and paternal sides of each pedigree being in- 

 cluded. Forty-one of these pedigrees gave a sufficiently complete history 

 for genetic use. We have retained only the pedigrees which gave fairly 

 complete information about two and usually three generations. Most of 

 the people to whom questionnaires were sent have suffered from migraine. 

 The preliminary report concerning hay fever was based upon a study of 

 twelve families. Including the pedigrees of the earlier paper, therefore, 

 fifty-three out of sixty-seven different family histories complete enough for 

 genetic study form the basis of this report. 



In the preliminary report we found that hay fever and other forms of 

 allergy may act as a dominant character but that their dominance is irregu- 

 lar, for it is often incomplete and sometimes an unaffected individual may 

 transmit the ability to become sensitive to several or to all of his or her 

 children. In the earlier paper the various forms of allergy were considered 

 as different expressions of a single gene, although the possibility of multiple 

 factors as a cause was not overlooked. The genetic evidence then available 

 led to this conclusion, but it is to be noted that it was quite in harmony 

 with the clinical behavior of the diseases. 



After studying allergic families for several years we have found that if 

 allergy is present in a strain, it is extremely rare (families 2 and 23) for a 

 generation to occur without some member of the family being affected. 

 We have only five cases of this kind. Unaffected individuals may pass 

 on the allergy to their children, but usually these individuals have allergic 

 brothers or sisters even though they themselves have no allergic diseases. 

 In no case which we have studied does allergy skip two generations; that is, 

 allergic individuals have never been found to appear after the family has 

 been negative for two generations. Occasionally one generation may be 

 entirely skipped, but we have no record of two negative generations fol- 

 lowed by a third allergic one, and we have sought carefully for cases bear- 

 ing upon this point. For these reasons we have come to regard a family as 



