The production of tritium, even in small amounts, will nevertheless be a formidable process. As we have seen, it takes eighty times as many neutrons to produce any given amount of tritium as to produce a corresponding amount of plutonium. Since the lithium will have to compete with uranium 238 (parent of plutonium) for the available supply of neutrons, and since the number of atoms of U-238 per given volume is nearly forty times greater than the number of lithium atoms, the rate of tritium production would be very much slower than that of plutonium. On the other hand, even if it took as much as two hundred times as long to produce a given quantity of tritium, the handicap would be considerably overcome because of the relatively small amounts that may be required. If, for example, we should need only 30 to 150 grams of tritium per bomb, it would take our present plutonium plants only six to thirty times longer to produce these quantities than it takes them to produce one kilogram of plutonium. A hypothetical plant such as the one mentioned in the official Smyth Report, designed to produce one kilogram of plutonium per day, would thus yield 30 grams of tritium in six days. How much tritium would be needed for an adequate stockpile of H-bombs? Since our primary reasons for building it are to deter aggression, to prevent its use against us or our allies, and as a tactical weapon against large land armies, it would appear that as few as twenty-five, or fifty at the most, would be adequate for the purpose. On the basis of the larger figure (assuming 30 to 150 grams of tritium per bomb), it would mean an initial stockpile of only 1.5 to 7.5 kilograms of tritium, which would entail the sacrifice of about 120 to 600 kilograms of plutonium. Once this initial outlay had been made, however, our plutonium sacrifice would be reduced annually to only one twenty-fourth of the original respective amounts—namely, 5 to 25 kilograms a year—just enough to make up for the decay of the tritium at the rate of fifty per cent every twelve years. One of the major problems to be solved, in addition to the main problem of designing the assembly, arises from the fact that the deuterium and the tritium booster will have to be in liquid form. Liquid hydrogen boils (that is, reverts to gas) at a temperature of 423 degrees below zero Fahrenheit under a pressure of one atmosphere (fifteen pounds per square inch). To liquefy it, it is necessary to cool it in liquid air (at 313.96 below zero F.) while keeping it at the same time under a pressure of 180 atmospheres. To transport it, it must be placed in a vacuum vessel surrounded by an outer vessel of liquid air. This would point to the need of giant refrigeration and storage plants, as well as of refrigerator planes for transporting large quantities of liquid deuterium and its tritium spark plug.