The line with 18 chromosomes of gamagrass was a trisomic for chromosome 4 or 5 of corn; its genome was 2n=21Zm + 18Td. The line with 49 chromosomes had the genotype 2n=40Zm + 9Td, where a set of 9 small chromosomes of gamagrass was added to 40 chromosomes of corn - minimally necessary to maintain apomictic development.

As pollinators, diploid forms were used: 1) Montana White; 2) commercial F1 hybrid Wilson (Seeds Inc.); 3) line with a T6-9 translocation (wx 18A); and 4) tetraploid N 102 A (Maize Genetic Cooperation Stock Center).

The results obtained by us are presented in Tables 1 and 2.

Table 1. Effects of gamagrass chromosome number and pollinators on hybrids 2n = 21Zm +18Td kernel weight.
 

Male	n	x	± m	d	V,%	min	max	% Montana White
Montana White	19	0.038 a	0.0062	0.0270	71.4	0.010	0.128	100
T6-9	67	0.021 b	0.0020	0.0167	78.8	0.002	0.077	55
Wilson	39	0.083 c	0.0071	0.0446	53.9	0.010	0.154	218
Tetraploid N 102 A	90	0.069 c*	0.0028	0.0269	38.7	0.007	0.131	182

Table 2. Effects of gamagrass chromosome number and pollinators on hybrids 2n = 40Zm + 9Td kernel weight.
 

Male	n	x	± m	d	V,%	min	max	% Montana White
Wilson	91	0.106 d	0.0027	0.0258	24.4	0.010	0.153	279
Tetraploid N 102 A	305	0.105 d	0.0016	0.0283	27.0	0.024	0.160	276

The results marked by the letters a, b, c, d reliably differ with P = 0.05. The values c and c* differ with P = 0.1. For comparison, the mean kernel weight of T. dactyloides = 0.03 ± 0.001 g and of the maternal tetraploid corn = 0.215 ± 0.003 g.

As is seen from Table 1, the diploid pollinators differ largely in the quantitative effect of imprinting. It is notable that the F1 hybrid (Wilson) is superior to even the tetraploid in this parameter though in the latter's offspring the ratio of genomes in endosperm will be 2 : 1, compared with 4 : 1 in the former.

The results adduced in Table 2 suggest that the hybrid is not inferior to the tetraploid in the "efficiency" of imprinting, though the difference in the ratio of maternal and paternal genomes is still more radical (8 : 1 in the former and 8 : 2 in the latter). It is remarkable that the mean kernel weight from pollination by the F1 hybrid of the 49-chromosome form increased by 84 %, the maximum weight remained as it was (Table 2) whereas both the mean and maximum weight increased from pollination by the tetraploid.

In this connection it may be supposed that one of the elements of the combining ability of lines producing heterosis is epigenetic modification specific to them, leading to an increase in kernel weight in the F1.

The analogous results from dependence of endosperm formation on pollen source and its ploidy are obtained on aposporous Paspalum notatum (Quarin, CL, Sex. Plant Reprod. 11:331-335, 1999). The suppression of imprinting is likely to be characteristic of all pseudogamous apomicts irrespective of their type as well as the value of endosperm development depends on pollinator.

Completing this section of our report we can conclude:

1) the size of kernels in the hybrids of corn with gamagrass shows dependence on both their ratio of corn male and female genomes and the type of pollinator;

2) the kernel weight trait in the corn - gamagrass hybrids has high genetic variability and can be increased by genetic and "selection" methods.

The authors express appreciation to Dr. E. Coe for fruitful discussions. The research was supported by the Netherlands Organization for Scientific Research Grant No. 047.007.019 and Russian Foundation of Basic Research Grant No. 00-04-49542.
 
 

Return to the MNL 75 On-Line Index
Return to the Maize Newsletter Index
Return to the MaizeGDB Homepage