The growth and development of cotton was studied on cotton plants grown in pots in growth rooms under constant day/night temperature (C) and varying temperature regimes throughout the day and/or night (V) The V-treatments had a common mean temperature of 22°C. The objectives were to determine the thermal requirements of three cultivars and to observe the extent of genotype×thermal environment interactions throughout the entire growth period.

Vegetative growth was found to be almost exclusively time and temperature dependent, varietal differences being largely insignificant. Plant material was found to accumulate four times faster under the warmest C-regime, which was 7°C warmer than the coolest. The C-treatments caused variation in the number and size of lateral shoots and leaves, causing leaf area to be larger at the squaring stage in cool environments but at maturity in warm regimes. Relatively cool nights lowered the position of the first floral or fruiting branch, whereas warm days shortened flowering intervals and thus promoted earliness. However, the effect of temperature in altering the position of the first floral branch seems to be less important than its effect on the shedding of early squares.

The effects of fluctuating temperature (V) on vegetative growth and earliness were similar to those from constant temperature environments. Growth and development rates were low in the low minimum V-regimes, especially when the maximum temperature was also low. Temperature variation affected vegetative growth to a greater extent in the early than in the later stages of development. At squaring, leaf area and dry weight were lowest under the regime with the highest minimum and maximum temperatures. Later on, only leaf area at flowering and total dry weight at maturity significantly differed between treatments. High maximum or minimum temperatures produced effects similar to a higher or lower mean temperature, respectively. The low minimum raised the node number of the first floral or fruiting branch, whereas in the case of boll dry weight it acted in the same way as a further drop in temperature, decreasing the weight, regardless of the maximum temperature. Boll period was affected mainly by the temperature variation itself rather by than the type of variation.

The inverse of time to a certain stage and the corresponding mean temperature were linearly related and allowed threshold temperatures (T0) and thermal time requirements to be estimated. T0=12°C, except for the initial stage, when it was lower. A cotton growing season cooler on average by only 1°C will considerably delay maturity.