NO. 2 METHOD OF REACHING EXTREME ALTITUDES 21 



system would return to within one millimeter (usually much less 

 than this) of the zero position after being displaced. In order to 

 avoid any such displacement as that just mentioned, an eccentric 

 clamp K, plate 7, figure i, was employed to' keep the suspended 

 system rigidly in its zero position during assembling and dismount- 

 ing the apparatus. 



This clamp consisted of an eccentric rod, K, free to turn in a hole 

 in the cap C, the lower end being held in a bearing in the yoke Y. 

 Through the upper end of this rod was pinned a small rod K', at 

 right angles to K. The surface of the rod K was smeared with a 

 mixture of bee's wax, resin, and Venice turpentine ; and the hole in 

 the cap through which K projected was rendered air-tight by wax of 

 the same composition. 



The suspended system was assembled while the cap C was held by 

 a support touching its under side. When the assembling was com- 

 plete, the wax was heated by a small alcohol blow torch until it was 

 soft, then a rubber band was slipped around the rod K' and the outlet 

 pipe E. A trial showed that the cap could now be put in place on the 

 pipe and removed, without the suspended system moving appreciably. 

 After the cap C was in position on the pipe, the rubber band was 

 removed, and the wax heated until the rod K could be turned out of 

 engagement with the holders H, H'. After a shot had been fired, the 

 clamp was again placed in operation until the system had been taken 

 from the 3-inch pipe and the smoked glass removed. 



The circuit which carried the electric current to ignite the charge 

 consisted of the insulated wire, W, plate 7, figure i, which passed 

 through a tapered plug of shellacked hard fiber, in the cap, C, thence 

 through a glass tube to the yoke Y, to which it was fastened. Below 

 the yoke it was wrapped with insulating tape, except at the lower end 

 where it was shaped to hold the 0.24 mm. steel wire, attached to the 

 fine copper wire from the wadding. From the chamber the current 

 passed up the rods R, R and out of the cap, around which was 

 wrapped a heavy bare copper wire, V, which together with W, con- 

 stituted the terminals of the circuit. It should be mentioned, in 

 passing, that a small amount of black powder, B, figure 2(a) , placed 

 over the platinum fuse-wire on the wadding, was found necessary as 

 a primer in order to ignite dense smokeless powders in vacuo. 



In order to make the joint, between the cap and the pipe, air-tight 

 during a determination, the following device was adopted. The out- 

 side of the cap, C, and also a lock nut, were both turned down to the 

 same diameter. The lock nut was made fast to the pipe. These were 



