Breeding Potentials of Noncultivated Cottons. V. Productivity of Cultivars, Race Stocks, and F1 Hybrids among Them in Long- and Short-Day Environments

Twelve primitive race stocks of cotton (Gossypiura hirsutum L.), four Upland cultivars, and 24 race stock ✕ cultivar F₁ hybrids were grown in three replicated tests in the summer at Phoenix, Ariz. and in the winter at Isabela, Puerto Rico. Objectives of this research were to compare productivity, to study hybrid expression of agronomic properties, and to determine the effects of cultivar ✕ race stock and genotype ✕ environment interactions on parents and their F₁ hybrids when grown in long- and short-day environments. Six of the 12 race stocks produced no bolls at Phoenix and few boils at Isabela. Hybrids of four of the six also produced no boils at Phoenix, but produced more bolls than either parent at Isabela. The other six race stocks, representatives of race latifoliura (from which Upland cotton was derived) produced from 2 to 29 bolls/plant at Phoenix. Five of the latter six produced more bolls at Isabela than at Phoenix; two of these five produced more bolls than the cultivars at the Puerto Rican location. None of the latifolium race stock ✕ cultivar hybrids produced more boils than the cultivars at Phoenix, but 5 of the 12 produced a larger number at Isabela. Genetic variability among the hybrids was caused by differences among the race stocks. Lack of cultivar ✕ race stock interactions suggested that the genetic variability may be attributed largely m additive effects. Numerous instances of heterosis, however, also suggested the presence of nonadditive genetic effects. for hybrids from 8 of the 12 race stocks, location ✕ race stock interactions were significant for all six agronomic characters studied, indicating the inconsistency of hybrid performance for those traits between the two locations. For the other four hybrids, however, the location ✕ race stock interaction was significant only for lint yield/plant, but not for its components. Significant location ✕ cultivar interactions were much more infrequent than location ✕ race stock interactions. Positive heterosis for F₁ vs. cultivar comparisons was shown by 26% of all hybrid/character combinations at Isabela and by 9% of those combinations at Phoenix. Several race stocks, potentially useful for specific characters, were noted for future breeding efforts.