In this study cost and returns structure of irrigated and un-irrigated farmers, producing cotton are framed. For this purpose the collected data have been analysed for cost and returns structure including various cost components cost function, production function, input output structure. The Regression model was estimated by the method of least squares for irrigated, un-irrigated and overall farmers cultivating cotton cost separately. There is a positive relationship between the total cost of cotton cultivation and various factor costs. The multiple regression model was estimated by the method of least square. It is found that the independent variables in the regression model are jointly responsible for 93.70 per cent (R2) variations in the total cost of cotton cultivation irrigated farmers in the study area. The independent variable human labour is positively related to the total cost of irrigated cotton cultivation in the study area. It means that an additional unit made in this variable may lead to the increase of 1.213 per cent with the total cost in cotton cultivation. It is inferred from the analysis that the variable total cost of irrigated cotton cultivation has a greater influence on the human labour in irrigated area cultivation. The F value (705.510) shows that the model fitted is statistically significant at 5 per cent level. In the case of un-irrigated farmers R2 value indicated that about 94.10 per cent of variations in the total cost. It was also found that the human labour and fertilizer had a greater influence on the determination of cost of cotton cultivation. The impact of the variable, yield per acre was found to be higher in the case of un-irrigated farmer. Thus it may be concluded from the analysis of cultivation of cotton, cultivation of land and yield of cotton was found to be significant variables in the case of irrigated and un-irrigated farmers and total cost found to be significant in overall farmers. Total cost is considered as an important variable for overall farmers.