A process used to manufacture paint yields, on the average, 70 tons of paint each day. Yields, however, vary from day to day due to changes in raw materials and plant conditions. Suppose, it is established that the daily yields are normally distributed, and that the variability from one day to the other is more or less independent, and the variance is nine tons. Because of increasing demand for this type of paint, certain modifications are suggested, and we are interested in estimating the mean yield of this modified process. How many sample observations do we have to obtain if we want to be 98% confident that our estimate is within 1 ton of the true, but unknown mean?