Functions which predict rate of leaf emergence and final number
leaves, used in a model to predict
the date at which growth stages occur, were tested in an experiment on
winter wheat cv. Mercia grown
with standardized husbandry at six sites in 1992/93, 1993/94 and
1994/95. A study of the number
of detectable nodes on the culm and leaf length was also made.
The predicted rate of leaf emergence was mostly within 5% of the observed
value. The difference
between observed and predicted final number of leaves was mostly less than
half a leaf but suspected
errors in leaf counts resulted in some differences of more than two leaves.
Variable extension of the basal internode impaired confidence in the
detection of nodes. The mean
number of detectable nodes differed significantly among sites and between
seasons from 3·7 to 4·8 but
could not be related to sowing date or final number of leaves. Further
information on factors affecting
extension of the basal internode is desirable to standardize node detection
and improve prediction of culm leaf appearance.
Culm leaves showed successively longer laminae up to the penultimate
There was a significant
relationship between length of the flag leaf and the final number of leaves,
but it was positive in
1993/94 and negative in 1994/95. This may have been due to greater
water stress in 1994/95.