MOUNT WILSON SOLAR OBSERVATORY. 165 



(d) The enhanced lines show no distinctive behavior as regards type or 

 magnitude of separation. The stronger enhanced lines of titanium are 

 almost without exception triplets for which the separation is moderate and 

 not related closely to the normal interval. 



(e) No direct connection between magnetic separations and pressure dis- 

 placements has been found. There is, however, a general agreement between 

 the mean results for large numbers of lines. 



(11) A comparative study of the Zeeman effect for 14 metals for which 

 results are available has shown that in all cases the triplets are similarly dis- 

 tributed with respect to their separations in a standard magnetic field. From 

 a combination of the measures for about 2,900 lines a value was found of 

 the ratio e ^m in excellent agreement with that determined by other methods. 



(12) An investigation of the displacements of the spectrum lines in the 

 arc and spark spectra of iron and titanium under pressures between 2 and 16 

 atmospheres has been carried on in the laboratory. The results indicate : 



(a) That the displacements in the arc spectrum are directly proportional 

 to pressure and amount, on the average, at A 4500 to 0.0045 Angstrom per 

 atmosphere for iron and to 0.0042 Angstrom for titanium. 



(&) That the displacements in different parts of the spectrum are closely 

 proportional to the third power of the wave-length in the case of iron, and 

 to the second power of the wave-length in the case of titanium. 



(c) That the enhanced lines present in the arc spectrum of titanium are 

 displaced on the average considerably more than the majority of the other 

 lines, and reverse much less readily under pressure. 



(d) That in the same region of the iron-spectrum groups of lines may be 

 selected having average displacements differing widely from one another, 

 with which the displacements of groups in other parts of the spectrum are 

 connected by the relationship of the cube of the wave-length. 



(e) That there is no definite evidence of the existence of a relationship 

 1:2:4 between the displacements of the lines in the spectrum either of iron 

 or titanium when all of the lines are included in the discussion, such as was 

 suspected by Duffield from his investigation of the iron spectrum. 



(13) An investigation of the effect of pressure upon the furnace spectra 

 of iron, titanium, and vanadium indicates : 



(a) That the displacements of the lines in the furnace spectra are con- 

 sistently larger than those in the arc spectra under equal pressures, at least in 

 certain regions of the spectrum, in some cases the differences amounting to 

 100 per cent. The difference does not depend on the nature of the com- 

 pressed gas, 



(b) The relative displacements of the furnace lines follow closely, in 

 general, those of the arc lines. 



(c) The displacements are closely proportional to pressure. 



