158 H. Crew — Rotation of the Sun. 



contrary, it will be seen that, taking either of the observations 

 made during the month of March, or those made in July, there 

 appears to be a gradual increase of daily angular motion with 

 latitude. For the values of v'-v", given in column 6, Table 

 I, are equatorial values, reduced from higher latitudes of obser- 

 vation, on the assumption that the sun rotates as a solid body ; 

 they are therefore proportional to the angular velocities in their 

 respective latitudes of observation. From column 7, Table 

 I, it is seen that the differences, between each particular value 

 of the equatorial velocity of rotation and the mean of them all 

 are, for the lower latitudes, mostly negative ; for the higher 

 latitudes, mostly positive. If there is any physical meaning to 

 be attached to this, it would seem that while the sun-spot layer 

 (or photosphere, if they be the same) is accelerated in the 

 neighborhood of the equator, the layer, which by its absorption 

 gives rise to the Fraunhofer lines, tends to dag behind, having 

 here a smaller angular velocity than in higher latitudes. I have 

 drawn through the observations the straight line which most 

 nearly represents this change of angular velocity with latitude, 

 and find, by the method of least squares, its equation to be 



v = l-15S cos x° (l+0'00335x°) 



where v is the linear velocity in miles per second of any point 

 in latitude %° of the sun's reversing layer. This gives for the 

 daily angular motion of any point on the reversing layer 



6=194:' (1+0-00335^°) 



while from sun-spots, Carrington* obtains for the photosphere 



6=865' (1—0-191 sin *j°) 



As will be seen from Table I, the greatest irregularities in 

 the value of v'-v" occur between the latitudes 15° and 25°. 

 May this not be connected with the fact that this is the region 

 most favored with sun-spots, the zone royale? 



Two neighboring linesf in the solar spectrum are often so 

 differently affected by disturbances on the sun's surface as to 

 indicate that the absorbing layers to which they respectively 

 belong are situated at widely different heights. 



That locus of absorption which is highest will, if we assume 

 the sun a solid, rotate with the greatest equatorial velocity, and 

 one might think that the values of v'-v" for different lines 

 should therefore arrange themselves in the same order as their 

 corresponding metals in the sun's atmosphere. But with a 

 tangential slit, as used in these measurements, it will be seen 

 that the section of the solar sphere made by a plane passing 

 through the slit and line of sight cuts each layer (for any given 

 latitude) in a different heliocentric longitude ; so that however 

 * Carrington: Observations on Solar Spots, p. 224. f Young: Sun, p. 100. 



