The new methods and devices in gas warfare which were developed by the great corps of scientists and laboratory experts attached to the American University Experiment Station were given practical trials at Lakehurst, where they were tested under conditions approximating as nearly as possible those of actual warfare. Here experiments were carried out to determine the value of gas-shells bursting in the air instead of by impact, the value of mixing toxic or lachrymatory gas with shrapnel, the value of 14-inch naval shell filled with a combination of high explosive and toxic substance, and the value of clouds of poison-smoke. Had the war continued, I imagine that the results of some of these experiments would have given the Germans the surprise of their lives. Though the gas production of Edgewood Arsenal from August to November, 1918, increased from 450 to 675 tons a week, and though the filling-plant had a weekly capacity of nearly 1,000 tons, less than 100 tons of gas was actually filled into shell weekly. This unfortunate state of affairs was due to the failure of the Ordnance Department to supply enough, or nearly enough, shell and boosters to keep pace with the production of gas. In other words, there was far more gas than there were shell to put it in, and far more shell than there were boosters for them. During the early summer of 1918, large quantities of this surplus gas were shipped overseas and there loaded into shell, but later instructions were received to stop all shipments in bulk except a limited amount of chlorine. From that time on, the production of gas was limited by the number of shell and booster available, because it is impossible to store toxic-gases in any large quantities. In fact, at all times after the manufacture of poison-gases began in the United States, the supply of such materials was not only in excess of the supply of shell and booster, but the gas-plants could not be operated to their full capacity because there was no way of utilizing the maximum output.