84
it is doubtful whether it would be noticed on the basis of phenotype. Due to the
inevitable limits to population size and to practical considerations regarding the
possibility of selection, most breeders use small population size and a large number
of combinations, rather than the other way round. In practice, plant selection usually
means ear selection, because in most cases it is impossible to separate the plant from
the population, while ears are easy to handle and contain enough seeds to allow the
sowing of ear progeny rows in the following year. Plant selection is more frequently
applied when plant selection follows bulk selection. This has the advantage that the
whole yield of a near-homozygous plant allows a greater volume of testing and
more rapid multiplication than is possible after ear selection.
In the course of traditional breeding, visual selection is carried out on the basis
of phenotype in populations consisting of a few thousand plants. In this stage, selec-
tion for monogenic/oligogenic traits is effi cient for traits such as earliness, plant
height or disease resistance. Plants with favourable traits can be labelled during the
vegetation period and given priority during the fi nal selection process. Selection can
best be made from populations with great variability and where the plants can be
fairly easily identifi ed, i.e., the stand is neither too dense nor too thin. In a dense
stand it is diffi cult to decide whether an ear in a low position is the main ear of a
short plant or the side ear of a tall plant, while in a thin stand the growth habit of the
plants is not a reliable prediction of how they would behave in the dense stands typi-
cal of farm production.
The third stage of breeding is testing. Yield level, yield stability and end-use
quality can only be identifi ed after several years of experiments in a range of
environments.
3.3 Alien Gene Introgression into Wheat
Crosses between wheat and related wild or cultivated species have been carried out
ever since breeding was begun. The fi rst sterile interspecifi c wheat × rye hybrid was
reported by Wilson in 1875, after which Rimpau developed similar hybrids in 1891
(cit. Lelley and Rajháthy 1955 ). Similar research was initiated worldwide, but for a
time the results were not utilised in practical plant breeding. The method used for
transferring genes from related species to wheat depends greatly on the evolutionary
distance between the species involved.
Species belonging to the primary gene pool of common wheat share homologous
genomes. Gene transfer from these species can be achieved by direct hybridisation,
homologous recombination, backcrossing and selection (Friebe et al. 1996 ). The
secondary gene pool of common wheat includes polyploid Triticum / Aegilops spec
ies that have at least one homologous genome in common with T. aestivum. Gene
transfer from these species is possible by homologous recombination if the target
gene is located on a homologous chromosome. Species belonging to the tertiary
gene pool are more distantly related. Their chromosomes are not homologous to
those of wheat. Other strategies need to be employed, because gene transfer from
Z. Bedő and L. Láng