The success of a doubled haploid breeding programme using wide hybridization depends initially on reliable seed setting between the 2 parents. Incompatibility has however been reported between H. vulgare cv. Vada × H. bulbosum (Pickering and Hayes, 1976) and many cultivars of T. aestivum × H. bulbosum (see for e.g. Snape et al., 1979). In the latter cross, tetraploid H. bulbosum genotypes appear to be more successful as a pollinator than diploid H. bulbosum (Barclay, 1975, Sitch and Snape, 1986), although there is no conclusive evidence as to whether this is simply due to higher ploidy level. In order to investigate the effects of H. bulbosum ploidy level per se on crossability, H. vulgare cv. Vada and T. aestivum cv. Chinese Spring were crossed with diploid and colchicine induced autotetraploid H. bulbosum genotypes Cb2920/4 and Cb2921/1 (for origin see Pickering, 1983). After colchicine treatment (see Pickering, 1980a for method) several cycles of splitting vegetative material and chromosome counting were carried out to obtain stable 28 chromosome clones.

Hybridizations were carried out as described previously (Pickering, 1980a) in 3 environments namely, 1) glasshouse- 21 ± 2 C 12hr/12 ± I C; 16h daylength using natural daylight supplemented when necessary with sodium and mercury vapour fluorescent bulbs. 2) cooled glasshouse - 17 ± 2 C 16h/12 ± 1 C (day/night) natural daylight only (October - December; i.e., early summer). 3) growth room - 18 ± 0.5 C 18h/14 ± 0.5 C (day/night); light supplied by cool white fluorescent tubes with an irradiance of 380 uE m^-2 sec^-1 at spike height. In order to compare the 2 pollen donors accurately, 'split- spike' pollinations were used (Pickering, 1980b) and results analyzed by paired sample t-tests (Table 1). For the duration of the experiment, control pollinators were carried out with the 4 H. bulbosum genotypes as pollen donors on H. vulgare cv. Emir, which is known to give seed sets of > 90% pollinated with diploid H. bulbosum. All seed sets from this cross were satisfactory.

In 3 of the 4 crosses seed sets were significantly increased when tetraploid H. bulbosum was used as pollen donor which supports the conclusions of Barclay (1975) and Sitch and Snape (1986). However diploid Cb2929/1 gave significantly more seeds on cv. Vada than autotetraploid Cb2929/1. The reasons for this remain unclear but it is evident that ploidy level itself is an important factor in the crossability of these 2 wide hybrids.

References:

Barclay, 1. R. 1975. High frequencies of haploid production in wheat (Triticum aestivum) by chromosome elimination. Nature 256:410-411.

Pickering, R. A. 1980a. Use of the doubled haploid technique in barley breeding at the Welsh Plant Breeding Station. Rep. Welsh Plant Breed. Stn. for 1979, pp. 208- 226.

Pickering, R. A. 1980b. Attempts to overcome partial incompability between Hordeum vulgare L. and H. bulbosum L. Euphytica 29: 369-377.

Pickering, R. A. 1983. The assessment of variation in two populations of Hordeum bulbosum L. for improving success rates in a doubled haploid barley programme. Euphytica 32:903-910.

Pickering, R. A. and J. D. Hayes. 1976. Partial incompability in crosses between Hordeum vulgare L. and H. bulbosum L. Euphytica 25:671-678.

Sitch, L. A. and J. W. Snape. 1986. The influence of the Hordeum bulbosum and the wheat genotype of haploid production in wheat (Triticum aestivum). Z. Pflanzenzuchtg. 96:304-319.

Snape, J. W., V. Chapman, J. Moss, C. E. Blanchard and T. E. Miller. 1979. The crossabilities of wheat varieties with Hordeum bulbosum. Heredity 42:291-298.