Comparison of Different Hill‐plot Designs for Evaluating Quantitative Response to Oat Crown Rust1

Abstract

Four hill‐plot field designs and a rod‐row design (plan 5) were compared for evaluation of the quantitative responses of five oat (Avena sativa L.) cultivars to two races of crown rust (Puccinia coronata Cda.) each at two levels of infection. In hill‐plot plan 1, test hills were planted in rows in a solid block. In plan 2, test hills were planted in rows alternating with rows of susceptible rust‐spreader hills. In plan 3, test hills were surrounded by barley (Hordeum vulgare L.) hills. In plan 4, single oat plants were grown in barley hills.Evaluation of response to rust was based on how the performance of rusted plots compared with the performance of corresponding unrusted control plots. Rust severity was highest in plan 1; intermediate in plans 2, 3, and 5; lowest in plan 4. A combined analysis of data from plans 1, 2, and 3 showed that cultivars, plans, rows, and levels of infection all significantly affected both yield and seed weight in terms of reduction due to rust infection. The plan ✕ cultivar interaction was significant for reduction in seed weight, but not for reduction in yield. However, in the four hill‐plot plans and the rod‐row plots, the cultivars ranked the same in order of reduction in seed weight. All correlation coefficients among the different plans for yield and seedweight reductions were significant. Coefficients of variation for these traits were lowest for the rod‐row plots; intermediate for plans 1, 2, and 3; and highest for plan 4. Plan 1 was most desirable for general use.