218 



CARNEGIE INSTITUTION OF WASHINGTON. 



WAVE-LENGTHS IN SKYLIGHT AND IN THE SPECTRUM OF VENUS. 



Mr. St. John and Mr. Nicholson have continued their observations of the 

 spectrum of Venus with a view to determining the cause of the apparent dis- 

 placements of the spectral lines found by Evershed. From observations 

 made in 1919-20 they concluded that the differences between the wave-lengths 

 in skylight and in sunlight reflected from Venus could be represented best 

 by an empirical formula based on the consideration that atmospheric dis- 

 persion at the low altitude at which the observations necessarily are made 

 when the phase angles are small produces an unsymmetrical illumination of 

 the slit of the spectrograph. This is due to the fact that the photographs are 

 made in the violet portion of the spectrum, while the guiding is done upon the 

 visual image. 



As a test of this question, a series of 85 spectrograms was made in 1920-21, 

 41 with Venus east, with phase angles 43° to 92°, and 44 with Venus west, 

 phase angles 107° to 25°. A blue ray filter was used in the guiding telescope, 

 so that the observer was able to keep the photographic image symmetrically 

 placed upon the slit. These spectrograms were measured and the results 

 compared with those from 66 similar spectrograms of skylight. In the earlier 

 series taken with the observer guiding upon the visual image, the wave-lengths 

 from photographs obtained at low altitudes had been found to be systemat- 

 ically shorter than those from photographs at high altitudes. In the present 

 series this difference is practically zero. The mean wave-length derived from 

 20 lines on 37 spectrograms, taken at average altitude 14°, is 0.001 a larger 

 than that from 48 spectrograms taken at altitude 28°, a quantity within the 

 limits of errors of measurement. 



A definite test of the variation of wave-length with the angle Venus-Sun- 

 Earth is afforded by the following comparison of observations made at nearly 

 constant altitude, but with different values of this angle: 



The results offer strong evidence against a shortening of wave-length in the 

 sunlight reflected by Venus from the solar hemisphere turned away from the sun. 



The material collected in this investigation provides the data for a deter- 

 mination of the solar parallax and a study of the rotation period of Venus, and 

 will be used for these purposes when the measurements have been completed. 



ABSENCE OF OXYGEN AND WATER-VAPOR LINES FROM THE SPECTRUM 



OF VENUS. 



Reference was made last year to the results of Mr. St. John and Mr. Nichol- 

 son which point to the absence of the absorption lines of oxygen and water- 

 vapor from the planet's atmosphere. Since that time Mr. King has found 

 through measurements in the laboratory that a column of air 39.5 meters in 

 length, equivalent to 8 meters of oxygen, will show the lines of the B band 

 of oxygen with an intensity comparable to that of lines of intensity 1 on Row- 



