190 AN INTRODUCTION TO MODERN GENETICS 



chance to obtain this information is by investigating identical twins, 

 which are formed from one fertilized ovum and therefore have identical 

 geneiical constitutions (p. 337). 



(b) Penetrance, — The penetrance is the frequency, measured as a 

 percentage, with which the gene shows any effect at all. Most of the 

 genes usually worked with in genetical experiments are chosen for 

 having a high or complete penetrance; every organism homozygous for 

 a recessive factor shows it. But for many genes, which tend to get 

 rather negleaed in experiment, this is not so. Only some of the homo- 

 zygotes show the effect. The frequency of effectiveness depends both 

 on the environment and on the genotype. A very good case of en- 

 vironmental influence is the gene giant in D. melanogaster} In a stock of 

 this mutation, the percentage showing the effect is dependent on the 

 amount of food; under conditions of severe larval competition very few 

 giants emerge. The effect, as its name impHes, is an increase in size, 

 caused by a delay in pupation; Gabritschevsky and Bridges suggest the 

 analogy with castes- in social insects, where again there are two sharply 

 distinct phenotypes formed as a response to different quantities of food. 



In the giant stock, the expressivity hardly varies at all; all giants are 

 of much the same size relative to normal. With other genes expressivity 

 and penetrance may both vary, but they often do not vary together. 

 Thus lines with high penetrance may have low expressivity and vice 

 versa. 



Penetrance is a statistical idea, and it presumably expresses a statis- 

 tical variation in developmental processes. It seems easiest to picture it 

 if we suppose that a gene with low penetrance does not determine the 

 whole course of the developmental reaction producing the organ or 

 character which the gene affects, but only acts for a certain short 

 period. Then we can think of penetrance in the same terms as were 

 used in discussing the Minutes (p. 186). The rest of the genotype may 

 be taken to define a "landscape" along the main valley of which the 

 character-producing reaction moves; but there are side valleys, and a 

 gene acting at the right time may push the reaction out of its normal 

 course into one of these. A gene with low penetrance provides a push 

 which only occasionally carries the reaction out of the main valley; one 

 with high penetrance is usually or always successful in this. 



According to this suggestion, the potentiality for the altered pheno- 

 type, for instance, the development of giant larvae in the example 

 quoted above, is already given by the normal phenotype. We are 

 ^ Gabritschevsky and Bridges 1928. - Wheeler 1937. 



