MOUNT WILSON SOLAR OBSERVATORY. 273 



We are greatly indebted to Professor H. S. Carhart for the pains- 

 taking construction of two batteries of Clark cells. The first, con- 

 sisting of 140 cells, supplies the constant voltage for charging the plates 

 of the electrometer. The second, of 110 cells, supplies the minute 

 current which passes through the photoelectric cell to the filament of 

 the electrometer. The cells consist of mercury, mercurous sulphate, 

 zinc sulphate, and zinc, sealed up in small glass tubes, each one giving 

 about 1.44 volts. They have been tested, and seem better adapted 

 to our purpose than any other form of battery. 



ELECTRIC-FURNACE SPECTRA. 



The electric-furnace spectrograms made by Mr. King, 170 in number, 

 have been mainly for a detailed study of the spectra of chromium and 

 vanadium at different temperatures, and for an examination of the effect 

 of hydrogen at varying pressures on the titanium furnace spectrum. 

 Photographs have also been made of the furnace spectrum of nickel and 

 of the ultra-violet region of iron, but are not yet reduced. Further 

 work with the tube-arc has been deferred until a more powerful trans- 

 former is installed. 



The range of wave-length covered in the investigation of chromium 

 and vanadium is X 3500— X 7500, within which 867 chromium and 

 1,092 vanadium lines have been classified as to the temperature at 

 which they appear and the rate of change in intensity as the tempera- 

 ture rises. The numerous lines in some portions of the spectrum and 

 the difficulty in bringing out weak lines in other regions have required 

 a variety of dispersions, combined with high resolving power, for which 

 the new concave grating has been highly useful. A decided gain in the 

 furnace spectra has resulted from the success attained by the Acheson 

 Graphite Company in supplying special tubes of very high purity, so 

 that no difficulty is experienced from blends with lines of foreign sub- 

 stances. Vanadium is found to resemble titanium in the richness of 

 its furnace spectrum at the higher temperatures, the lines comparing 

 in number closely with those of the arc, but showing large differences 

 in relative intensity. Among the lines which persist at lower tempera- 

 tures, many are faint in the arc spectrum. Chromium shows a wide 

 diversity at different temperatures and many deviations from the arc 

 spectrum. An important group of red lines, weak in the chromium 

 arc, is found to be strong in the low-temperature furnace. 



The effect of a hydrogen atmosphere on the enhanced lines of tita- 

 nium was studied to determine whether their strength in the chromo- 

 sphere and in some stellar atmospheres may be due to the presence of 

 hydrogen. The experiments failed to show any strengthening of the 

 enhanced lines, which appeared at low pressures with equal ease when 

 hydrogen was present and when the furnace contained only a residue 

 of air. But as the pressure of hydrogen was increased from 10 cm. 



