HUMAN GENETICS 33I 



ment backwards, we can see that affected individuals tend to be the 

 offspring of related parents. The calculation required to put this 

 argument on a quantitative basis is rather complicated, and so many 

 factors enter in that exaa predictions can hardly be made. But qualita- 

 tively we may say that if the parents of affecteds show an excess of 

 inbreeding over that characteristic of the population as a whole, that is 

 strong evidence that we are dealing with a recessive gene. This test is 

 particularly useful in distinguishing between recessives and characters 

 due to two complementary genes, which otherwise show rather similar 

 behaviour. 



4. Sex-linked Genes 



In man, the male is the heterogametic sex. Rare sex-linked genes can 

 therefore show in males, but in females tend to be covered by their 



D-r-O 



"3 6-r-n 

 fi~yT^ kk^k i 6 i 6 6 6 



Fig. 137. A Pedigree of Red-Green Colour Blindness.— (Males squares, 

 females circles, affected individuals black.) The trait is inherited as a sex-linked 

 character. Note that it is transmitted by females, but is only shown by males, 



(After Holmes, from Groenouw.) 



normal allelomorphs. It is often concluded without more ado that a 

 character which is found in males but is also transmitted through 

 normal females must be due to a sex-linl<ed factor. But clearly this 

 condition is also fulfilled if the gene is autosomal but its expression is 

 Hmited to the male sex. Additional criteria of sex-Unked inheritance, 

 sufficient to distinguish it from sex-limited inheritance, are (i) that the 

 trait is not handed on from father to son (except in the rare case where 

 the female is heterozygous for the gene), and (2) that at least one 

 female showing the character is known. For many of the characters 

 usually listed as sex-linked neither of these conditions is fulfilled by the 

 data as yet available. 



Still further information is required before it can be decided whether 

 a sex-linked factor is dominant or recessive. Clearly the criterion should 

 be the appearance or non-appearance of the character in a heterozygous 

 female. In a random mating population, the frequencies of the various 

 genotypes should be pA : qa in males and p^ AA : 2pq Aa : q^ aa in 



