The origins of this work can be traced back to the spring of 1964, when I was a teaching assistant at the University of California at Berkeley. I was giving a class on the topic of consumer demand to a group of students taking an introductory economics course. In the process of my presentation I attempted to show the link between the demand function and consumer satisfaction by resorting to the use of a consumer surplus diagram. I had thought this to be a useful pedagogic device, particularly in that it enabled, I believed, the student to better understand marginal analysis. It was easy to show that the consumer would increase consumption up to the point where marginal utility in terms of the surplus just equaled price. The loss of satisfaction due to a price rise could then be illustrated in a straightforward manner. However, simplicity has a cost, as I was soon to learn. Near the end of the discussion, one student inquired as to how the consumer surplus analysis could be utilized if more than one price was varied. If two commodities were involved, then the demand curve for each would shift according to the degree of complementarity or substitutability. My answer consisted of two parts. First, I argued that my analysis was purely partial and was presented as a pedagogic device.