MOUNT WILSON OBSERVATORY. 183 



does not appear to be so consistently defined. However, the dis- 

 turbing conditions existing in complex spot-groups, together with 

 other causes, frequently prevent the certain determination of the 

 direction of the whirl in the Ha flocculi. 



Our first investigation of the general magnetic field of the sun has 

 been essentially completed. The results include the discovery of the 

 existence of the field, the detection of thirty lines showing magnetic 

 displacements, the determination of the sun's magnetic polarity, 

 and the measurement of the strength of the field at different levels in 

 the solar atmosphere, the inclination of the sun's magnetic axis, and 

 the period of its revolution about the solar axis of rotation. Certain 

 anomalies encountered in this research remain to be studied, but it 

 will be impossible to gather observational material for further investi- 

 gation of the general magnetic field before the approach of the sun-spot 

 minimum, because of the disturbing effect of the powerful magnetic 

 fields in sun-spots. 



The spectrographic measurement of the solar rotation has led in 

 recent years to curiously discordant views among the various astro- 

 nomers who have undertaken it. At Mount Wilson the variations in 

 rotational velocity observed over the period 1914 to 1918 show no 

 evidence of periodicity, but seem to depend upon just such temporary 

 conditions in the sun's atmosphere as our knowledge of solar phenom- 

 ena would lead us to expect. An important confirmation of previous 

 conclusions has resulted from a comparison of the rotational velocities 

 given on the same photographs by the high-level H and K Unes of 

 calcium and the low-level lines of the cyanogen band. The velocities 

 for calcium are from 10 to 15 per cent greater than for the cyanogen 

 below, and these differences increase for higher latitudes. 



The determination of stellar parallaxes has been pushed forward 

 with the 60-inch reflector, using both the trigonometric and spectro- 

 scopic methods. By the former means 26 new parallaxes have been 

 determined, for the most part of stars of small proper motion and 

 advanced spectral type. The list also includes Nova Persei No. 2, 

 Barnard's star of great proper motion, and three nebulae. The absolute 

 magnitudes of the central star of three planetary nebulse determined 

 here are -f6.1, +8.1, and +9.4, giving a mean value of +7.9. It is 

 interesting to note, though it may be merely a coincidence, that this 

 value is in close agreement with the absolute magnitude +6.9 found 

 for the three novae, at minimum brightness, of which the parallax is 

 known. 



The spectroscopic method of parallax determination, which yields 

 results of high precision at minimum cost of time, has been effectively 

 utilized for the measurement of the absolute magnitudes of several 

 hundred stars during the year. Peculiarities detected in the spectra 

 of the Cepheid variables, including exceptional intensity of the hydro- 



