APICAL DOMINANCE - VARIABILITY AMONG COTTON GENOTYPES AND ITS ASSOCIATION WITH RESISTANCE TO INSECT HERBIVORY

We investigated variability in apical dominance among cotton genotypes (Gassypium hirsutum, G. barbadense) and the relationship between degr ee of apical dominance and resistance to insect herbivory. Two groups of genotypes were compared: a 'normal' group, including genotypes with naturally occurring concentrations of tannins and terpenoid aldehydes (characteristics of the tribe Gossypiae), and an 'enhanced' group, in cluding experimental lines selected for improved defences. The degree of apical dominance (AD, dimensionless) and the apparent time required for the activation of axillary buds after the removal of the apical b ud (t, days) were measured in a glasshouse experiment including 24 gen otypes and two damage treatments, i.e damaged plants, in which the api cal bud was removed 40 days after sowing, and undisturbed controls. To quantify AD we assumed that the stronger the apical dominance, the gr eater the distance (d) between the plant apex and the nearest vegetati ve branch; we measured d and plant height (h) in control plants, and c alculated AD = d/h. To calculate t, we counted periodically the leaves in the axillary bud closest to the apex of damaged plants, fitted lin ear regressions between leaf number and time, and took the x-intercept of the fitted functions as an estimate of t. Between 12 and 19 of the genotypes used in the glasshouse were grown in held experiments invol ving two treatments: control, protected with insecticides, and unprote cted crops exposed to naturally occurring herbivorous insects, includi ng Helicoverpa spp. (Lepidoptera: Noctuidae) and sucking mirid bugs (H emiptera: Miridae). The ratio between the yield of unprotected and pro tected treatments for each genotype, R, was taken as an empirical meas ure of herbivory resistance. Degree of apical dominance ranged from 0. 35 to 0.80, the height of undamaged plants from 0.68 to 1.0 m, and t f rom 1 to 7 days. Within the normal group, genotypes with strong apical dominance were taller, and normally had axillary buds that required m ore time for activation after release of apical dominance. A significa nt, non-linear relationship between R and AD was found for this group of genotypes, which indicates a broad optimum and a consistent decline in resistance with apical dominance increasing over a threshold (appr oximate to 0.6). None of these relationships was found for genotypes w ith 'enhanced' defences, partially because this group had a narrower r ange of apical dominance. We conclude that a large variability in the degree of apical dominance among cotton genotypes exists that can infl uence resistance to herbivory, but interactions between degree of apic al dominance and defences cannot be disregarded. (C) 1997 Elsevier Sci ence B.V.