| Clearly, differences between the families occurred for the three characters
studied and were highly significant in the case of the number of grains
per ear. Correlation coefficients between the characters demonstrate that
there is some evidence that height and the number of ears per plant are
controlled either by the same genes or by associations of dissimilar genes
due to linkage. This follows from the significant correlation coefficient
for these two characters r = 0.8552 + or - 0.1636, P = < 0.001. There is
no evidence for a correlation of either of these two characters with the
number of grains per ear. Certain qualitative differences were also apparent,
in particular, the shapes of the ears varied considerably, so that some
families were extremely compactoid, whereas in others a more extended spike
was observed, It is possible that some of the differences observed may reflect whole chromosome changes which resulted from non - disjunctional behaviour or from compensatory chromosomal changes associated with the nulli - 5B genotype. The wide range of variation observed, however, suggests that such possibilities are unlikely to be the main source of segregation and that the more probable cause of most of the variation is homoeologous recombination. The initial pre - requisite, namely, that of available variation, would appear to be satisfied by the results described here. Further requirements are, however, obviously necessary before techniques of this nature can be used in plant breeding. Thus two questions, which are vital to the success of the method have still to be tested and answered. The first concerns the problem of stabilizing the genotype of a desired phenotype either by the re - introduction of chromosome 5B or by alternative genetic means. The second involves the question of whether an economic improvement, by alterations within a genetic system that is already finely adjusted, can be achieved by drastic measures of this kind. In this connection it is of interest to recall that the often quoted advantage of inbreeding with allopolyploidy is that it allows heterozygosity to be fixed. The recombination produced by the use of nulli-5B techniques, could destroy the advantages of any homoeologous heterozygosity which may exist within the wheat genotype. If this is so, improvements in economic fitness will be difficult to achieve. On the other hand, a better understanding of the genetic architecture of wheat, a subject which at present requires some refinement, may demonstrate that such difficulties may not arise and new levels of genetic balance within hexaploid wheat may be possible. |
| <-- Back |