As part of an overall program to evaluate cotton germplasm for traits that may be associated with drought tolerance, selections of exotic, as well as commercial cottons taken from field (rainout shelter) studies that indicated a relative improvement in dry matter production under drought conditions were grown in the glasshouse for further evaluation. Twenty-five cotton (Gossypium sp.) strains were grown under two glasshouse environments (different planting dates) in large containers (37.8 L) filled with soil with sufficient water added to bring the soil water content to field capacity. The plants were allowed to use the available water without any loss due to evaporation (pots were covered with plastic) until they had reached permanent wilting as determined by the inability of the third true leaf from the top of the plant to regain turgor at predawn. When permanent wilting occurred the above ground portion of the plant was harvested and dried at 80 C. The pots were weighed to determine the total amount of water used by the plant. The root systems were then washed free of the soil and also dried at 80 C to obtain both the root and shoot contribution to the total biomass produced as well as determine the changes in root-shoot relationships between the strains.

The environment for the second planting date was altered by increasing the air movement across the plants during the day with large fans. Thus the boundary layer at the leaf surfaces was reduced increasing the rate of water transpired and creating a more stressed environment. Differences did not occur between the two planting dates for shoot dry weights or leaf dry weights at permanent wilting. Significant differences did occur, however, between the two planting dates for root dry weights with an increase in root dry weight occurring for the strains grown under the second, more stressed environment. There was a significant increase in the root/shoot ratio for the same planting date since the change in the environment favored an increase in root growth with no difference in shoot growth. Significant differences occurred among the strains for shoot weight, leaf weight, root/shoot ratios. G. herbaceum had the highest shoot weight while the strains (Tl84 and T141) representing the race mariagalante had the lowest. In the case of the roots, the maria galante strains had the highest root weights while the experimental strain Lubbock Dwarf had the lowest. The strain T1236 showed the greatest increase in root/shoot ratios which may be related to the observed increase in salt tolerance of this strain. There were no interactions between the strains and planting dates for shoot weights, leaf weights, and root weights indicating that those strains that had large shoots for example, in the first environment, also had large shoots in the second environment. There was however, a significant interaction between the strains and the planting dates for root/shoot ratios due to the magnitude of the changes in the values for root/shoot ratios between the two environments. There was apparently no correlation between shoot weights and root weights between the strains for both planting dates, thus providing an opportunity for genetic improvement of productivity by selection of desirable traits in the plant root system apart from selection of favorable shoot growth characteristics.