250 CARNEGIE INSTITUTION OF WASHINGTON. 



4 cm. of the record represent the first stage, when energy is passing into 

 the vapor from the oscillations of the condenser, while the remainder of the 

 record shows the second stage, when the vapor is cooling by radiation, no 

 energy being given to it from the circuit. It is in this second stage that 

 the velocity of sound is determined. If a sound-wave passes through the 

 vapor in a direction parallel to the slit, it will be registered on the photo- 

 graph, since the adiabatic compression in the sound-wave elevates the tem- 

 perature momentarily, causing a momentary increase of radiation. The 

 result is a bright line inclined at an angle to the axis of the band of light, 

 which is found to be easily measurable. Since the linear speed of the slit- 

 image with respect to the photographic plate, the lateral magnification, 

 and the value of this angle are all known, it is a simple matter to calculate the 

 velocity of sound in the vapor. The source of sound is the explosion itself, a 

 cylindrical or spherical reflector placed a few centimeters from the wire return- 

 ing the wave to the explosion. By varying the distance between the reflector 

 and the wire, the sound-wave may be made to strike the cylindrical mass of 

 vapor at any desired time after the beginning of the explosion. 



Since the velocity of the sound-wave gives T/m (T = absolute temperature 

 of the vapor, m its molecular weight), it is clear that this quantity may be 

 determined at any instant of time during the cooling of the vapor. And 

 since m is now pretty accurately known, as a result of Saha's theory, the 

 temperature may be calculated, and hence the rate of cooling by radiation. 

 Moreover, by comparing the intensity of the light at any point with the 

 intensity of a black body of known temperature photographed with the same 

 apparatus, the rate of emission of energy may be determined. Finally, by 

 comparing the rate of fall of temperature with the rate of emission of energy, 

 it should be possible to deduce the apparent specific heats of the vapors. 



New Apparatus. 



The rotating-mirror camera is being modified so that it will be better suited 

 for accurate work. A two-stage mercury diffusion pump of the latest type 

 has been installed in order to make possible a study of the explosions and other 

 phenomena in very high vacua. Glass plates for a new high-voltage con- 

 denser have been ordered, and this additional unit should be available during 

 the coming year. 



COMBINED EFFECT OF ELECTRIC AND MAGNETIC FIELDS ON RADIATION. 



The experiments mentioned in the last report on the combined Zeeman 

 and Stark effects have been continued by Mr. Sinclair Smith, and good results 

 have been obtained with the hydrogen lines. Although several special forms 

 of quartz tubes have been tried, no satisfactory results have yet been obtained 

 with the metallic lines in which we are especially interested. It is hoped, 

 however, that means will yet be found of producing the brilliant spectra in 

 tubes strong enough to withstand the long exposures which are essential in 

 this work. 



ABSORPTION OF OXYGEN AND WATER-VAPOR BANDS. 



Observations by Mr. King in the laboratory of the bands A and B of the 

 solar spectrum, ascribed to terrestrial oxygen, and of the water-vapor band a, 

 show that these may be seen as absorption bands with a much smaller thick- 



