328 AN INTRODUCTION TO MODERN GENETICS 



Thus there are very few homozygous dominants and nearly all the 

 individuals showing the character will be heterozygotes. If such an 

 individual marries a normal, his offspring will show a i : i ratio of 

 aflfecteds (showing the character) to normals. This requirement is 

 sufficient to identify some characters, e.g. diabetes insipidus, brachy- 

 dactyly, as due to autosomal dominants. 



With rather commoner genes, the homozygous dominants become 

 more important and cannot be neglected. If we know the frequency of 



2pq + q^ 



the character in the population, i.e. — we can calculate p and g, 



P 

 and from them the proportions of different kinds of matings and finally 

 the proportion of affecteds which should be expected among offspring 

 of matings between affecteds and normals. The agreement between the 

 theory and the actual data is quite good for some common genes, such 

 as those for the blood groups or taste-capacity, but is sometimes not so 

 good for other semi-rare genes, when an excess of affecteds is found to 

 occur. This is probably because the assumption of random mating 

 throughout the entire population does not really apply. Rare genes do 

 not immediately spread throughout the entire population but tend to 

 be concentrated in certain localities, owing to some degree of inbreeding 

 within local groups. In such a group, the concentration q of the gene is 

 higher than it is in the population as a whole, and there are therefore 

 relatively more homozygotes and fewer heterozygotes among the 

 affecteds than would be suggested by a calculation based on the 

 frequency of the gene in the whole population. 



Characters dependent on dominant genes will clearly run in families, 

 recurring generation after generation; they will be "hereditary" in the 

 usual medical sense, whereas characters due to recessive genes, as we 

 will see, tend to occur in several of a group of brothers and sisters and 

 then to disappear, a type of behaviour which is called "familial." Quite 

 often, a "hereditary" trait occurs sporadically, the proportion of 

 affecteds among offspring of affecteds being much less than a half, and 

 the trait sometimes skipping a generation altogether. The suspicion 

 may arise that we are dealing with a rare recessive, and this may some- 

 times be the case. But the gene may also be a dominant of low pene- 

 trance. In this case we can identify it by another test. Consider a family 

 tree containing a rare dominant. Very nearly all matings of affecteds 

 with normals will be Aa x aa (since AA is very rare), and it is easy to 

 see that, if the penetrance is constant, the proportion of affecteds to 

 normal will be the same among the parents, sibs or offspring of an 



