Strawberry (Fragaria × ananassa) seedlings were evaluated for yield, fruit weight, and commercial appearance in two field trials established in 1985 and 1986. Genetic analyses for unbalanced diallels were performed to quantify genetic, environmental, and interaction variance for each trial separately, and for crosses common to two locations in a single year. When data from crosses common to two test locations were analyzed simultaneously, narrow-sense heritabilities (h2) averaged 0.35 (±0.11), 0.21 (±0.07), and 0.08 (±0.06) for yield, fruit weight, and appearance score. Broad-sense heritabilities (H2) were 0.35 (±0.11), 0.27 (±0.12), and 0.21 (±0.11) for the same traits, respectively. These estimates do not differ significantly from heritabilities estimated from the ancestral breeding population 20 years ago. Estimates of H2 for single-location analyses were biased upwards by dominance × location interactions for all traits. Additive × location interactions were detected for appearance score and contributed a small bias to single-location estimates of h2. Use of biased estimates in predicting genetic gain could lead to errors in choice of appropriate selection strategy.
Received for publication 21 Mar. 1988. The cost of publishing this paper was defrayed in part by the payment of page charges. Under postal regulations, this paper therefore must be hereby marked advertisement solely to indicate this fact.