The first kilocurie cobalt-60 therapy machines were produced and installed in Canada in the Fall of 1951, one at Saskatoon, Saskatchewan, and one at London, Ontario. Since that time there has been a steady and continuous increase in the number of machines in service, and hence in the number of patients receiving treatment. This expansion in experience with cobalt machines has produced numerous papers and much healthy discussion concerning the best type of design and the best techniques of treatment. It has also produced speculation on the probable extent of future requirements for cobalt machines. This speculation is desirable, both on the part of radiologists and physicists who will use the machines, and on the part of manufacturers who will produce them. An exchange of opinions should help to keep things in balance. It will be recalled that the balance between supply and demand was out of phase for several years due to a world shortage of high-specific-activity cobalt 60. As one associated with the production of cobalt 60, I shall try to examine the extent of past and probable future uses of these machines with particular reference to the adequacy of expected supplies of the isotope. During the early stages, production of cobalt in the United States was small and production in Canada depended entirely on the NRX reactor. I suppose some of us in Canada thought we had stolen a march on our good neighbors in the United States and other countries. On Dec. 12, 1952, however, it was reported that a “pin hole” had developed in the NRX reactor, necessitating a shutdown for repairs. This “pin hole” proved to be of a very cantankerous nature, since it required a year or so for repair. Production in the United States was still on a modest scale, and consequently many hospitals had to wait for long months to get their cobalt machines installed; others may have wondered where a replacement source would come from if they went ahead. There may be those who now wonder whether this is the kind of thing that could happen again. I shall try to produce some facts and assumptions which should provide an answer to this question. There are a number of methods of approach to this problem of trying to determine whether the future supply of cobalt 60 will be adequate to the needs of therapy. One could simply plot the number of machines installed each year and attempt to extrapolate the graph into the future. I have collected (as best I could) such figures and these are shown in Figure 1. Would anyone dare to extrapolate that curve for four or five years without some other considerations? Perhaps I should explain that in all of these discussions on the amount of cobalt required, and the expected production, I have in mind the world total with the exception of the USSR and satellite areas. There is some information available on Soviet production, but I do not know whether it is reliable.