Premium
Heterosis and Gene Action in Cotton, Gossypium hirsutum L. 1
Author(s) -
Meredith William R.,
Bridge R. R.
Publication year - 1972
Publication title -
crop science
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.76
H-Index - 147
eISSN - 1435-0653
pISSN - 0011-183X
DOI - 10.2135/cropsci1972.0011183x001200030015x
Subject(s) - lint , heterosis , biology , epistasis , gossypium hirsutum , hybrid , fiber crop , horticulture , agronomy , botany , genetics , gene
Heterosis (F 1 — MP), useful heterosis (F 1 — ‘Deltapine 16’), and gene action estimates were obtained at four locations from the six groups of populations involving ‘Deltapine 16’ ( Gossypium hirsutum L.) crossed with six cultivars (‘Q 1 ,’ ‘Del Cerro 183,’ ‘FTA 263‐20,’ ‘Acala 3080,’ ‘Mo277‐396,’ and ‘Stoneville 603’). Heterosis for lint yield was detected in each cross. F 1 yields ranged from 92 to 115% of that of Deltapine 16, while F 2 yields ranged from 86 to 107%. Increases in boll size were partially responsible for the increased productivity of the hybrids. Parental lines originating from outside the Delta in general produced hybrid‐by‐location mean squares greater than those from Delta‐derived parents. These interactions were related to the higher relative performance of Deltapine 16 at the locations with higher yield. The potential benefits from utilizing heterosis were especially evident when all traits were considered simultaneous. Four of the six F 1 's produced yields equal to or greater than Deltapine 16 while also having longer, stronger, and finer fiber. Heterosis for each of the nine traits studied was determined primarily by dominance gene action. Of the 54 possible tests additive‐by‐additive epistasis was significant in only four instances, and even in those four cases the mean squares for epistasis were much smaller than were those for additive or dominance effects. Additive effects predominated for lint percentage, seed index, fiber strength, fiber elongation, and fiber fineness. However, dominance was greater for boll size. Both were involved for lint yield and fiber length. In every case there were some crosses that did not fit the general trend. The interaction of additive effects with locations was usually of greater magnitude than the dotninance‐bylocation interaction. There were 28, 13, and 1 instances of significant additive, dominance, and additive by additive epistatic by location interactions, respectively. Interactions with locations were the greatest practical limitation for lint yield selection and they were less important for the other traits. The interactions with locations were particularly large for the cross involving Acala 3080. Breeding implications are discussed.