172 



SCIENCE. 



[N. S. Vol. XVII. No. 422. 



A Determination of the Frequency of Alter- 

 nating Currents by the Automatic Ad- 

 justment of the Circuit to Resonance: 

 Carl Kjnsley, University of Chicago. 

 An electric current can be tuned to any 

 frequency within a wide range by varying 

 the self induction of the circuit. This 

 may be automatically accomplished by us- 

 ing a moving core in the coil giving self 

 induction. If the spring-held core is 

 slightly beyond the position of resonance in 

 the direction of too large self induction, 

 it will be retained in equilibrium between 

 the force of the spring and the pull of the 

 coil. The plunger will, therefore, rise and 

 fall in the coil as the varying frequency 

 requires a greater or less self induction for 

 resonance. 



Charts were used showing the mode of 

 operation of the method proposed. 



On Methods of Measuring Radiant Effi- 

 ciency: E. L. Nichols and W. W. 

 CoBLENTZ, Cornell University. 

 This paper deals with the visible and 

 infra-red spectrum of the light transmitted 

 by a Avater cell and by a water cell and 

 iodine cell in combination, for the purpose 

 of determining the nature of the correction 

 which it is necessary to apply in finding the 

 true radiant efficiency of sources of light. 

 It is shown that in the case of a water cell 

 one centimeter in thickness, at least five 

 sixths of the transmitted energy belongs 

 in the infra-red transparency of an iodine 

 solution which is opaque to the visible 

 rays is not such as to warrant the use of 

 this cell for the purpose of determining the 

 correction for the water cell. Comparison 

 was made between the value for the radiant 

 efficiency by the usual method of the water 

 cell, the same value corrected by integra- 

 tion of the curves for the transmitted 

 energy of the cell and the value of the 

 radiant efficiency obtained by direct inte- 



gration of the energy curve of the spectrum 

 of the source. The recent contention of 

 Angstrom that all determinations of 

 radiant efficiency by means of the water 

 cell thus far published give much too large 

 a value, is shown to be fuUy justified. 



The Infra-red Emission Spectrum of the 



Mercury Arc: W. C. Geer and W. W. 



CoBLENTZ, Cornell University. Presented 



by E. L. Nichols. 



While investigating the infra-red spec- 

 trum of the Aron's lamp, a grouping of the 

 emission lines was observed which is of 

 interest in connection with spectral series. 



The vertex of the arc was used for a 

 source of radiation. For this purpose a 

 side tube, having a window of fluorite or 

 rock salt, was placed at right angles to the 

 plane of the arc. The spectrum was pro- 

 duced by means of a mirror spectrometer 

 and rock-salt prism, while a Nicholas 

 radiometer was used to measure the dis- 

 tribution of energy. 



The spectrum was explored at every 

 minute, and at certain places every 20" 

 of arc of the spectrum circle up to 9 lu 

 It was found that the energy radiated con- 

 sists of a series of emission bands at 1 ij. 

 and 5 ij-, Avith a slight indication of a band 

 at 3 //. Nowhere in the spectrum was the 

 intensity of the radiation from the arc very 

 great, while beyond Qp- the deflections due 

 to the hot glass walls of the lamp were as 

 great as those due to the arc. The lamp 

 was kept in a water-bath. 



The width of the spectrum covered by 

 the radiometer vane at 1 ij. was about .13 /i. 

 The error at 1 p- is less than .01 //•, while 

 at 4/i the error is perhaps .03/-(. 



Since in the region at 3 p the radiometer 

 showed slight deflections at times which 

 were recorded as questionable, and since in 

 all other parts of the spectrum from 2 to 

 4 /i no such effect could be detected, one is 



