Index | Search | Home | Table of Contents
Andrews, D.J., B. Kiula, and J.F. Rajewski. 1993. The use of protogyny to make hybrids in pearl millet. p. 122-126. In: J. Janick and J.E. Simon (eds.), New crops. Wiley, New York.

The Use of Protogyny to Make Hybrids in Pearl Millet

David J. Andrews, Barnabas Kiula, and John F. Rajewski

  1. METHODOLOGY
  2. RESULTS AND DISCUSSION
  3. CONCLUSION
  4. REFERENCES
  5. Table 1
  6. Table 2
  7. Table 3
  8. Fig. 1
  9. Fig. 2
While cytoplasmic male sterility (CMS) facilitates hybrid seed production, the necessity of breeding CMS seed and restorer pollen parents which combine specifically, lengthens hybrid development time and restricts the possible combinations. In maize, these restrictions can be overcome by making hybrids by detasselling. In pearl millet [Pennisetum glaucum (L.) R. Br.], the protogynous nature of flowering can be used to make hybrids, termed pro-hybrids to distinguish them from CMS hybrids. The duration of complete stigma emergence, the protogyny period (see Fig. 1A) before anthesis (Fig. 1B) on pearl millet heads may vary between genotypes from 1 to 5 days. While choice of parent lines and careful management of pro-hybrid production fields will reduce self pollination in the chosen "seed parent," some may still occur. Prior research in hybrid/inbred mixtures in tall forage hybrids (Burton 1948,1989) showed that up to 50% inbred line seed did not significantly reduce hybrid performance. Tall grain hybrids made in West Africa by protogyny using dwarf populations as females with tall varieties as pollinators showed as much as 45% heterosis over check variety and less than 5% female selfing (Lambert 1982). Our research investigated, using one top-cross (TC) and 3 single-cross (SC) hybrids, what effect various levels of simulated "seed parent" selfing would have on dwarf pro-hybrid performance.

METHODOLOGY

Experiments 1 [68A x MLS (TC)] and 2 [554 x 556 (SC)] were grown in 1989 and repeated in 1990. Experiments 3 [68A x 60051 (SC)] and 4 [205 x 319 (SC)] were added in 1990. All were grown at the UN-L Farm, Mead, Nebraska, in experiments with 4 or 5 replications, 4 m plots, 2 row or single row plots. Treatments were hybrid seed mixed with 0, 20, 40, and 60% inbred seed and both parents. Seeds were mixed on the basis of live seed number, and no thinning was done. In Experiment 1, seed of 68B was used in mixtures while in Experiment 3 only late parent line 60015 was used. In Experiments 2 and 4, mixture treatments were made separately with both parents.

RESULTS AND DISCUSSION

Yields of the 80:20 hybrid/inbred mixtures were not significantly different from the control (100% hybrid) in either the TC (Fig. 2) or two out of three SC hybrid tests (Tables 1 and 2). Actual yield reductions were 4.4, 4.1, and 6.3%, respectively. In the remaining SC hybrid (Table 3), the yield of 60:40 mixtures were not significantly different (-14%) from the control but, for no discernable reasons, the 80:20 mixtures (-22%) were significantly lower yielding. Heterosis levels were 20% in the TC hybrid over the open-pollinated cultivar parent, and 275, 70, and 260% over the best inbred in the respective SC hybrids. Experiments are underway with marker traits to determine the actual hybrid/self seed percentages occurring in pro-hybrid seed production plots, but selfing is expected to be less than 20%.

CONCLUSION

The use of protogyny in pearl millet appears to have potential for making hybrids, thus permitting more rapid hybrid development and a wider choice of parent combinations compared to CMS systems. The presence of up to 20% inbred parent seed reduced grain yields only 4 to 6% in top-cross or 2 out of 3 single-cross hybrids, probably because inbred plants offered little competition to hybrid plants. The use of parent combinations which produce a dominant hybrid phenotype compared to that of the "seed parent" will reduce the effect of any possible selfing.

This research was conducted using dwarf hybrids. Where tall grain cultivars are preferred, as in Africa were 12 million ha of pearl millet are grown for food, and where susceptibility to ergot is associated with hybrids made with CMS seed parents, this method of hybrid seed production provides opportunities for the rapid development of top cross hybrids using existing tall cultivars as males and adapted semi-dwarf or dwarf lines as protogyny seed parents.

REFERENCES

Table 1. Grain yields and plant height of hybrid parents and mixtures of SC hybrid 554 x 556 (Expt. 2, Mead, Nebraska, 1989)z.

Treatments (% mixtures) Yield
Hybrid
554 x 556		 Inbred 1 554 Inbred 2 556 (kg/ha) (% of pure hybrid) Plant height (cm)
100 0 0 3330 100 83
80 20 0 3130 94 85
80 0 20 3120 94 84
60 40 0 2890 87 86
60 0 40 2860 86 80
40 60 0 2230 67 80
40 0 60 2180 65 82
0 100 0 1210 36 69
0 0 100 1150 35 83
LSD .05 406 6.6
zExpt. 2 in 1990 was discarded following chinch bug damage.

Table 2. Grain yields and plant height of hybrids, parents and mixtures of SC hybrid 68Ax 60015 (Expt. 3, Mead, Nebraska, 1990).

Treatments (% mixtures) Yield
Hybrid
68A x 60015		 Inbred 60015 (kg/ha) (% of pure hybrid) Plant height (cm)
100 0 3220 100 93
80 20 3090 96 90
60 40 2550 79 86
40 60 2390 74 87
0 100 1890 59 77
0 100 (68B) 1820 57 90
LSD .05 671 1.1

Table 3. Grain yields and plant height of hybrids, parents and mixtures of SC hybrid 205 x 319 (Expt. 4, Mead, Nebraska, 1990).

Treatments (% mixtures) Yield
Hybrid
205 x 319		 Inbred 1
205		 Inbred 2
319		 (kg/ha) (% of pure hybrid) Plant height (cm)
100 0 0 2730 100 84
80 20 0 2080 76 87
80 0 20 2180 80 81
60 40 0 2410 88 82
60 0 40 2240 82 76
40 60 0 1830 67 86
40 0 60 1840 67 83
0 100 0 630 23 67
0 0 100 1040 38 74
LSD .05 520 1.8

Fig. 1. (left). Pearl millet heads. (A) Protogynous heads with stigmas fully exerted prior to anthesis; (B) Anthesis following the period of protogyny.


Fig. 2. Experiment 1. pearl millet top cross hybrid 68A x MLS. Grain yields (2 year means) of hybrid/seed parent mixtures.
Last update September 10, 1997 aw