90 DR J. HALM ON 
“lag” appeared to increase continually towards the poles. So far, then, as the general 
character of this peculiarity goes, the question appears to be empirically settled. But 
there remains still another, and, as I think, not less important question. Are we 
allowed to suppose that the surface rotation of the sun remains unaffected by the 
periodic changes of solar activity ? Judging the question from a purely logical point of 
view, we are almost bound to answer it in the negative. It seems to me difficult to 
imagine that such violent disturbances of the normal conditions of convection as we 
perceive in solar eruptions and spots, and the consequent displacements of matter im the 
solar globe, should have no influence on the distribution of the rotational velocities at the 
surface. A careful study of the behaviour of the solar rotation during a cycle of 
activity may probably teach us far more about the causes and the seat of these solar 
ay eb? 3 4 
Spectrum of receding 
limb. 
} Solar limbs. 
Spectrum of approach- 
ing limb. 
Fie. 1. 
Group of lines as seen in the viewing telescope. (1 and 8 are telluric, 2 and 4 solar lincs. ) 
clisturbances than the whole array of statistical facts regarding the periodic displays of 
dynamical phenomena at the surface which are now in our possession. Professor 
Dunkr’s observations, covering a period of three years, during which next to no change 
took place in the activity of the sun, cannot give an answer to this question. I there- 
fore thought it a promising venture, the success of which seemed to me in some way 
guaranteed by the great accuracy and consistency of Dun&r’s results, to extend these 
observations over a time of more pronounced changes of solar activity. The following 
contains a description of the results so far obtained, and of the instrument employed in 
my investigations. 
The observations were begun in August 1901, and so far carried on to the end of 
1903. Although this interval of time is not greater than that covered by Dunir’s work, 
there is this essential difference, that the year 1903 was characterised by an abrupt and 
violent increase of solar disturbances after a pronounced and persistent calm during 
1901-2. This fact, as we shall see, has an important bearing on the results, which 
