Eight field experiments involving two sowing dates at each of the two contr
asting environments during 2 years were conducted to study the suitability
of potassium iodide (KI) to mimic drought stress in triticale (X Triticosec
ale Wittmack). A concentration of 0.3% of KI was applied 10 days after anth
esis on cv. Trujillo and three near-isogenic lines derived from it. Both dr
ought stress and KI caused a similar reduction in grain yield (42% and 47%,
respectively), number of kernels per spike (16% and 18%, respectively), an
d kernel weight (17% and 24%, respectively). The reduction caused by chemic
al treatment with KI on yield and yield components was greater under irriga
ted than under rainfed conditions and, in the absence of irrigation, it was
less for the late sowing. The senescence agent did not affect the number o
f spikes per m(2), but it caused a decrease in the number of kernels per sp
ike. The overall effect of the senescence agent on plant productivity was s
imilar to the reduction caused by drought. The ranking of the genotypes aft
er chemical treatment was basically the same than under drought conditions
for yield and yield components. Principal component analysis based on a set
of variables related to yield ability and other agronomic traits showed th
at the overall response of the plant to drought stress and to KI applicatio
n was similar. However, canonical discriminant analysis showed that kernel
weight on the main spike was more reduced by KI than by drought. It was con
cluded that potassium iodide applied over the plant canopy (including the s
pikes) may mimic the effects of drought, not only during grain filling, but
before anthesis. (C) 1998 Elsevier Science B.V. All rights reserved.