lyO AN INTRODUCTION TO MODERN GENETICS 



chromosomes and cytoplasm and are not of a very far reaching nature 

 (p. 67). But in an autopolyploid, types of heterozygosity are possible 

 which could not occur in a diploid; e.g. a factor A may in a tetraploid 

 occur simplex Aaaa, duplex AAaa, or triplex AAAa, as well as in the 

 homozygous aaaa (nuUiplex) or A AAA forms. The new balances 

 between the dominant and recessive factors may give new intermediate 

 phenotypes for the Aaaa and AAAa classes, or the simplex type may 

 be fully dominant. No general rule can be given, the result depending 

 on just how hypomorphic the recessive is. Examples have already been 

 given (p. 57) in the section on characters affecting the triploid endo- 

 sperm in one of which the simplex form showed the dominant char- 

 acter, in the other the recessive. Wettstein^ has described a very com- 

 plete series of types in the moss Funaria hygrometrica where polyploids 

 can be produced artificially. 



In allopolyploid hybrids clear cases of balance between the genes 

 from the two parents may be found. A well known example is Kar- 

 pechenko's^ radish-cabbage hybrid. Other good cases are known in 

 hybrid mosses (p. 67). 



In polysomics the unbalance may give very, or only sUghtly, abnor- 

 mal phenotypes. In such cases, the balance is between whole blocks of 

 genes, and the units involved (whole chromosomes or large parts of 

 chromosomes) have a general effect on nearly every part of the organ- 

 ism, usually with very htde tendency for a main effect on one character : 

 one is dealing as it were with a lump of the genetic background, not 

 with a specific gene. It might have been thought that fairly large blocks 

 of genes, such as whole chromosomes, would have nearly the same 

 residual effect as the whole genotype and would therefore, when 

 redupHcated, have less influence on the general balance of the geno- 

 typic miheu than would a small fragment containing just a few isolated 

 genes. This is not so as a rule. The effect of a dupHcation is usually 

 roughly proportional to the amoimt of redupUcated chromatin. The 

 explanation is probably that most genes are working in the horizontal 

 part of their dose-effect curve, and that an extra dose has Httle effect. 

 The alteration caused by adding a new block of genes is dependent 

 only on those comparatively few genes which are not at their maximum 

 of effectiveness, and the number of these is roughly proportional to 

 the total number of genes added. 



The very large and varied series of polysomics synthesized by 



Blakeslee^ and his collaborators in Datura provide perhaps the best 



^ Wettstein 1927. ^ Karpechenko 1924, 1928. ^ Rev, Blakeslee 1934. 



