There were five treatments, namely, the unstressed control and
stress at the bolting, anthesis, seed formation, and seed maturation
stage, which refer to the growth stages at which water was withheld.
(I) Unstressed (Control): where plants were not subjected to
water stress. Irrigation in this treatment was frequent, once every
week during the whole period of plant growth and development.
(II) Stress at bolting stage: plants were subjected to stress when the
main flower stalk begins to appear. This stage was reached after 7–8
weeks from planting. (III) Stress at anthesis stage: Plants were subjected
to stress when the first florets of the main umbel started
to open. This stage was reached after 10–12 weeks from planting.
(IV) Stress at the beginning of seed formation: This stage was
marked by the presence of white watery seeds in the fruit carpel.
It was reached after 14–16 weeks after planting. (V) Stress at seed
maturation stage, when the color of seed coat turned black.
2.2. Watering intervals
During the first season, the plots were irrigated normally, once
every week throughout the season except at the time of water stress
application, where watering was withheld for an interval of 2 weeks
only once for every treatment “growth stage”. While during the
second season, the intervals of watering were extended to 3 weeks
instead of two. The quantity of water supplied was 1 m3/plot per
irrigation during the season. Ten plants were randomly selected in
each plot from the middle rows at the time when the plants were
reaching the designated stage of growth and development for the
water stress treatment. The first inflorescence from each selected
plant was tagged giving them a serial number from 1 to 10 with the
corresponding code of the treatment.
2.3. Irrigation control
The 90-V-notch weir method was used to measure and adjust
the water flow to the plots, as an equal quantity of water was supplied
using furrow irrigation over all experimental plots (Michael,
1978). Good tillage, plot and channels preparations were performed
to avoid outside drainage during irrigation. The 90-V-notch weir
was constructed across the approaching channel before starting
irrigation. The notch was 5 cm above the bottom of the channel and
the water was only allowed to pass through it. The height of water
over the 90-V-notch was measured in centimeters by using a ruler
after the constant and stable flow of water occurred. The amount
of water discharged was then calculated by measuring the time of
water flow into the plot in seconds with reading of the water height
over the 90-V-notch. Then return to the water discharge table to
know the discharge rate in liters per second. Then the total volume
of the water applied to each plot was counted from the discharge
table (Michael, 1978).