Section 13 — Plant Genetics and Breeding 



weight, and fiber strength. These data were re- 

 corded for each season only at the end of the 

 seven year program. The results showed that 

 some improvement was obtained in number of 

 bolls per plant with an average increase of 4 

 per cent per year. There was no improvement in 

 the other characteristics. This may be due to 

 lack of genetic variability for these factors and 

 due to the fact that the progenies were selected 

 only on the basis of their field appearance. The 

 latter may explain why the only success was in 

 increasing number of bolls. Variability was 

 large for number of bolls; fair for lint percent; 

 and small for lint index, boll weight, and fiber 

 strength. After seven years the variability re- 

 mained practically constant for each character. 

 This method of breeding showed to be ineffi- 

 cient for making steady improvement once there 

 is a genetic population which has reached a 

 certain stage of progress. Procedures that can 

 utilize new recombinations and that can measure 

 accurately the genetic variability are suggested. 



must be exploited in order to realize the ful 

 yield potential of Upland cotton in Uganda. 

 1. Comstock, R. E. and Robinson, H. F. 

 Chapter in Heterosis, Iowa State College Press, 

 Ames, Iowa. 



13.40. The Average Degree of Dominance of Genes 

 determining Components of Lint Yield in the 

 BP52 Upland Cotton Variety. J. T. Walker 



(Namulonge, Uganda). 



1 . Estimates of the average degree of domi- 

 nance of components of lint yield of Upland 

 cotton were obtained from a field trial based on 

 a design by Comstock and Robinson (1952.)( 1 ). 

 The F2 populations of hybrids between inbred 

 lines of the BP52 variety were used in back- 

 crossing and had previously been evaluated 

 in diallel crossing trials. 



2. Only additive genetic variance was found 

 for the traits loculi per boll and lint per seed. 



3. Genotypic interaction due to specific geno- 

 types caused dominance for seeds per loculus, 

 seeds per boll and earliness as measured by the 

 number of days to first flower. 



4. The characters lint percentage and seed 

 weight showed generalized genotypic interaction 

 and apparent overdominance with transgressive 

 segregation, the interaction variance being dis- 

 tributed evenly throughout the material. 



5. Findings are discussed in relation to pre- 

 vious inferences about the inheritance of yield 

 in cotton made as a result of selection index and 

 diallel crossing studies. It is suggested that in- 

 teraction at the genetic level may be involved 

 in the expression of lint yield heterosis. 



6. It is concluded that non-additive effects 



13.41. Genetics of Internode Length in a Wide Cross 

 of Pisum sativum L. Donald L. Mahoney 

 and Jules Janick (Lafayette, U.S.A.). 



The inheritance of internode length was 

 analyzed in reciprocal Fi, F 2 , F3, and backcross 

 progenies of crosses between the field pea 

 (Black-eyed Susan) and the garden pea (Wiscon- 

 sin Perfection). Associative characters (total 

 number of nodes, total height, and number of 

 nodes to the first flower node) were also analyzed 

 and correlated with internode length. 



The Fi exhibits positive hybrid vigor in height 

 and average internode length significantly greater 

 than either parent (BES — height, 156 cm, average 

 internode length, 6.3 cm; WP— 90 cm, 4.9 cm; 

 Fi— 186 cm, 8.9 cm). The total number of nodes 

 appears to be controlled by partially dominant 

 factors; in most cases, the Fi lies closer to the 

 fewer noded parent ( WP). The greater height of 

 the Fi is therefore attributable to heterosis in 

 internode length. Analyses indicate that the 

 heterotic expression of internode length in the 

 Fi through the F 3 is best explained by the 

 dominance rather than the overdominance 

 hypothesis. Lines were obtained in the F3 whose 

 average internode length was equivalent to the 



Fi. 



Tetraploidy decreases the average internode 

 length in both Black-eyed Susan and Wisconsin 

 Perfection. Fertility in tetraploid peas is very low, 

 and crosses involving tetraploids and diploids 

 were unsuccessful. 



13.42. The Somatic Variability of Plant Clones. 



G. Haskell (Dundee, Great Britain). 



Variability, the capacity for variation, is a 

 property of living matter. It may be genetic, 

 producing observable differences between indi- 

 vidual organisms, and somatic, embodying dif- 

 ferences in the same individual at different life 

 stages (Raymond Pearl). 



Biometrical variability of the plant clone, a 

 group of individuals arising vegetatively from one 



223 



