a1 University of Reading, Department of Agriculture, Plant Environment Laboratory, Cutbush Lane, Shinfield, Reading, RG2 9AD, England
a2 CSIRO Division of Tropical Crops and Pastures, The Cunningham Laboratory, 306 Carmody Road, St Lucia, Brisbane, Queensland 4067, Australia
a3 CSIRO Davies Laboratory, Private Mail Bag, PO Aitkenvale, Townsville, Queensland 4814, Australia
a4 Victorian Institute for Dryland Agriculture, Private Bag 260, Horsham, Victoria 3401, Australia
a5 Queensland Department of Primary Industries, Hermitage Research Station, via Warwick, Queensland 4370, Australia
a6 Department of Primary Industry and Fisheries, PO Box 1346, Katherine, Northern Territory 0851, Australia
a7 Western Australia Department of Agriculture, PO Box 19, Kununurra Regional Office, Western Australia 6743
Four genotypes of ‘desi’ and two of ‘kabuli’ chickpea (Cicer arietinum) were sown at six locations in Australia on various dates between 1986 and 1988, giving 22 combinations of site and sowing date with diverse photothermal environments. Times from sowing to first flowering (f) varied from 30 to 162 d, mean pre-flowering temperatures from 10.8° to 29.3°C and mean photoperiods from 11.3 to 15.6 h d−1. There was no evidence that any observation had been obtained in photoperiods shorter than the ceiling photoperiod (Pcc) or longer than the critical photoperiod (Pc). This suggests that, in typical agricultural environments, chickpea crops experience photoperiods (P) which should satisfy the condition Pcce < P < Pc. In one ‘kabuli’ and two ‘desi’ genotypes, 1/f was influenced by both temperature and photoperiod. The coefficient of determination (R2) for a linear, additive rate of development model ranged from 0.74 to 0.80, with no significant difference in either temperature sensitivity or photoperiod sensitivity among these three genotypes. In the remaining three genotypes, no significant response to temperature was detected between 10.8° and 29.2°C, so rate of progress to flowering was influenced solely by photoperiod. There was no significant difference in the sensitivity of 1/f to P among these three genotypes. The linear, additive rate model found here to be so satisfactory as a predictive tool for phenology, is also shown to have much wider general application.
Predicción del liempo de floración
(Accepted May 26 1993)
p1 Present address: South Johnstone Research Station, Queensland Department of Primary Industries, P0 Box 20, South Johnstone, Queensland 4859, Australia
p2 Present address: 11 Edward Crescent, Byford, Western Australia 6201.