The growth temperature of fruit on Malus domestica trees of two cultivars (`Queen Cox' and `Golden Delicious') was altered during an eight-week period shortly after anthesis. This was achieved by moving potted trees in a reciprocal transfer experiment within two temperature controlled glasshouse compartments. The trees were initially grown at either 15 degreesC or 20 degreesC. After about three weeks, half the trees at each temperature were moved to the other temperature compartment. Throughout the growing season fruit size was measured non-destructively on selected fruit. Measurements of fruit size were used to determined temperature-induced changes in relative growth rate. Measurements were also made of cortical cell size to determine if fruit cell number had increased. Complementary measurements, at harvest time, were made of total fruit weight, number of fruit and mean fruit size per tree and seasonal tree growth. The results of this work show that the fruit setting potential of `Queen Cox' and `Golden Delicious' was severely reduced at the higher temperature. Fruit set was highest for ambient temperature treatment, but this treatment had a significantly reduced mean fruit size at harvest. The largest fruit were obtained from `Golden Delicious' trees initially growing at 20 degreesC. However, fruit set was the lowest on these trees. Measurements of fruit growth showed that treatment differences in `Golden Delicious' fruit size at harvest were unlikely to be explained by initial differences in fruit number per tree. Despite higher temperatures producing larger fruit, measurements of cortical cell size showed that increasing the temperature induced larger cells not more cells per apple.
