Soluble sugars have been analyzed from field-cotton (Gossypium hirsutum L.) leaf tissue samples taken from plants grown in various locations across the U.S. Cotton Belt during 1989-1991. The leaf starch content in these samples was found to be relatively constant but the ratio of leaf hexoses (fructose + glucose) to leaf sucrose was found to be significantly lower in those samples collected in the arid southwest than in leaf samples from the eastern part of the country. To determine the effect of leaf ontogeny upon this pattern, sugars were also determined in leaf samples taken from sympodial leaves at the 10th node of the plant over the course of a growing season in Phoenix, Arizona and Fayetteville, Arkansas. Leaf respiration, photo-synthesis and malate content were also determined as a function of leaf development during these seasons. Differences in cotton varieties could not explain the observed differences, since the same pattern was observed in leaves from Stoneville 506 grown at five locations across the cotton belt and in Coker 100A grown in both Phoenix and North Carolina. Differences in weather might be invoked as an explanation for the observed differences, especially the difference in night temperature between these two cotton-growing regions. The daytime leaf temperature of leaves in these two regions also differ, since in contrast to cotton grown in the southeast, leaves on irrigated plants in the relatively dry southwest are able to effectively reduce their temperature relative to the air by transpiration. This observed difference in leaf soluble sugar content is important for explaining source-sink behavior in this plant, since during daylight hours, the rate of carbon export from cotton leaves is directly proportional to their sucrose content. Leaf carbohydrate content also serves an important role as a basic parameter for some recently developed computer models which simulate cotton growth and yield.