1022-7954/01/3712- $25.00 © 2001
Russian Journal of Genetics, Vol. 37, No. 12, 2001, pp. 1382–1387. Translated from Genetika, Vol. 37, No. 12, 2001, pp. 1642–1649.
Original Russian Text Copyright © 2001 by Chalyk, Rotarenco.
Recurrent selection is an important method of maize
selection. Its use can increase the frequency of favor-
able genes in synthetic populations.
Chase was ﬁrst to use doubled haploid lines (DHLs)
for recurrent selection . He suggested that DHLs
would allow more precise selection for favorable genes
and elimination of harmful ones from synthetic popula-
tions. In spite of the fact that each selection cycle with
the use of DHLs takes a year more than in the case of
conventional methods, Chase suggested that DHLs
would increase selection efﬁciency.
The efﬁciencies of various recurrent selection
schemes involving DHLs were estimated with the use
of mathematical modeling . It offers advantages over
common selection procedures, if the average increase
in productivity per selection cycle is taken into account.
The ultimate result of selection depends on the duration
of each cycle. Successful use of haploid methods in
recurrent selection is determined by shortening the time
required for obtaining DHLs and the duration of a sin-
gle selection cycle.
Instead of DHLs, as in Chase’s method, we pro-
posed to use haploid plants . This considerably
shortened the time required for one cycle. In our
scheme, one cycle is performed for only two agricul-
Matroclinous haploid maize plants are highly via-
ble, and this is of principal importance for their use.
Several kernels appear in almost each ear of haploid
plants pollinated with pollen obtained from diploid
ones [4, 5]. Almost all seeds obtained from haploid
plants have normal diploid embryos. This allows their
use in genetics and breeding. In addition, the phenotype
of a haploid plant completely corresponds to its geno-
type [6, 7].
The use of haploid plants for increasing the fre-
quency of favorable genes in synthetic populations
offers several advantages over diploids. First, all genes,
both dominant and recessive, are expressed in haploid
plants. Allelic interactions are eliminated. This facili-
tates selection of favorable genes and elimination of
plants carrying harmful ones.
Second, matroclinous haploids develop from an
unfertilized ovum. Segregation in haploid plants corre-
sponds to gamete segregation. Therefore, the emer-
gence of a desired gene set in haploids is more likely
than in diploids .
Third, natural selection is an excellent tool for
cleansing populations or strains from harmful recessive
genes when using haploid plants. Plants carrying harm-
ful, lethal, or semilethal genes are either nonviable or
sterile and, hence, are eliminated from the breeding
Fourth, haploid plants can be used for rapid devel-
opment of homozygous strains from synthetic popula-
tions to be improved [4, 9, 10].
As haploid plants can be applied for maize improve-
ment, we used them in a recurrent selection scheme.
The purpose of our study was to investigate the possi-
bility of using matroclinous haploid plants in recurrent
selection of maize and to evaluate the efﬁciency of this
MATERIALS AND METHODS
Experiments were performed with two synthetic
populations, SP and SA. The synthetic population SP
was obtained by crossing four strains (P346C, ms1334,
The Use of Matroclinous Maize Haploids
for Recurrent Selection
S. T. Chalyk and V. A. Rotarenco
Institute of Genetics, Academy of Sciences of Republic Moldova, Chisinau 2002, Moldova;
fax: (3732) 55-61-80; e-mail: firstname.lastname@example.org
Received April 20, 2000
—Two cycles of recurrent selection were performed in maize with the use of matroclinous haploids.
Two synthetic populations, SP and SA, were improved. Each cycle consisted of two stages: (1) isolation of hap-
loids from the synthetic populations and (2) growth of the haploids, pollination with pollen from diploid plants,
and selection. The selection was performed for ear size in haploid plants. The mean gain in productivity in the
synthetic populations SP and SA the per cycle was 16.48 and 20.98%, respectively. It is suggested that the high
value of this index is related to the fact that haploid plants reveal useful genes with additive and epistatic effects.
Natural selection may have played a part, too. The combination of artiﬁcial and natural selection in haploids
resulted in a considerable gain in productivity in the synthetic populations to be improved.