336 CARNEGIE INSTITUTION OF WASHINGTON. 



been overcome there remains the fact that the "activity" as measured depends 

 on the specific velocities of the radioactive carriers as well as on the true radio- 

 active content of the air. For this reason a new method has been developed 

 by the author. The general principle involved in this method has been used 

 in one form or another by other investigators. It consists in drawing air 

 between two concentric cylinders, the central one of which is charged negatively 

 to such a high potential that it catches all of the active carriers entering the 

 concentric cylinders. The saturation current produced in an ionization 

 chamber by the active deposit collected in a given time, combined with a 

 knowledge of the air-fiow during the collection of the deposit, makes it possible 

 to estimate the amount of active material per cubic meter of air, if one assume 

 a knowledge of the nature of the deposit, which can be obtained from the 

 form of the decaj^ curve. 



The collecting apparatus, as at present employed, consists of a copper 

 cylinder 25 inches long and 12 inches in diameter, with an anemometer at one 

 end and a fan at the other. The central system consists of an insulated wooden 

 cylinder 4.7 inches long, supported by a rod passing through its axis and insu- 

 lated from it by sulphur. The surface of the wooden cylinder is covered with 

 copper foil held on by rubber bands, and it is on this foil that the deposit is 

 collected. Earthed metal caps attached to the central rod fit over the top and 

 bottom of the central cylinder without touching it, and insure that the deposit 

 of the active material is confined to the copper foil. 



A large air-current is necessary if a large amount of deposit is to be obtained, 

 and in order to have saturation with a reasonably low potential on the central 

 system, it is necessary that the central cylinder shall be large. A large central 

 system when afterwards introduced into the ionization chamber, so as to form 

 the central system there, would, however, on account of the large capacity, 

 reduce the sensitivity in the ionization-chamber measurements. For this 

 reason the central system of the ionization-chamber is formed of a thin rod, 

 and the foil, after removal from the inside cylinder, is bent over and made to 

 line the walls of the ionization chamber with the active surface facing inwards. 

 In this way the foil does not contribute to the capacity of the system. The 

 height of the ionization-chamber is about twice the height of the foil cylinder, 

 so that the latter occupies only the central portion of the wall of the chamber. 

 In this way difficulties are avoided arising from the uncertainty which would 

 otherwise be introduced in the fraction of the alpha particles traveling any 

 assigned portion of their range in the vessel. The uncertainty referred to 

 would be great in the case where an appreciable fraction of the alpha particles 

 strikes the top or bottom of the vessel, since the distribution of active deposit 

 on the foil is by no means uniform. The central system of the ionization- 

 chamber is attached to a single fiber electroscope adjusted to a sensitivity of 

 about 5 to 10 divisions per volt, and the potential is applied to the outer ves- 

 sel, the whole being mounted on a gimbal. 



The number of ions produced per cubic centimeter per second in a closed 

 vessel is measured by observing the saturation current produced in a copper 

 vessel of about 27 liters capacity. Here again the central system is connected 

 to a single-fiber electroscope, and the potential is applied to the outer system, 

 the whole instrument being mounted on a gimbal. The rod of the central 

 system is insulated from the outer vessel by an amber ring surrounded by a 

 brass collar which fits into a hard-rubber ring. The latter fits into the opening 

 in the vessel, and by maintaining the brass collar at zero potential, leak from 

 the outer vessel to the central system is avoided. The principle of using a 

 sensitive single-fiber electroscope for measuring the alteration of potential of 

 the central system, while the potential is applied to the outer vessel, secures, 

 of course, great sensitivity and reduces leakage errors to a minimum. 



