PROPORTION CHANGES 329 



most readily available source of successful variation in all organisms. 

 Its further analysis is therefore a most pressing problem. 



If single genes or blocks of genes are freely liable to reduplication 

 within the haploid set, as these observations suggest, we must 

 suppose that characteristic haploid sets wiU probably contain a 

 proportion, perhaps a majority, of their genes represented two, 

 three or more times. The gametic set of an autotetraploid will 

 have all its genes present an even number of times and odd 

 numbers will be restored to it by mutation, until it reaches an 

 allopolyploid condition which will not be distinguishable from that 

 of a diploid. 



Haldane (1932 a) has suggested that chromatin-diminution in 

 Ascaris requires localisation of certain genes ; such a condition 

 might be produced by the extensive reduplication of all or most of 

 the genes and their later selective elimination. 



This conception makes the occurrence of forms with different 

 degrees of qualitative differentiation as between different chromo- 

 somes intelligible. If one species has very little reduplication, 

 all its chromosomes must be differentiated in the highest degree 

 compatible with random assortment of the genes among the 

 chromosomes. If another form has a great deal of reduplication 

 with most of its genes represented six or a dozen times it will 

 have a very low differentiation with the same random assortment. 

 Thus it is not necessary to assume that organisms with low 

 differentiation have individual genes of any special kind but 

 merely that they are reduplicated more frequently either at one 

 stroke with polyploidy, or segment by segment with structural 

 change. 



4. INERT CHROMOSOMES AND GENES 

 Individuals of many species of plants and animals have 

 chromosomes that produce no specific or general effect on their 

 growth. Such chromosomes are said to be inert. They do nothing, 

 and can therefore be lost entirely or reduplicated almost without 

 limit and the organism will develop in just the normal way. They 

 are never regularly members of the complement in the species, for 

 since they can be lost without effect they frequently are lost. 



