RECENT ADVANCES IN SCIENCE 201 



tions of prominences, but Mr. Evershed's observations leave 

 little room for doubt ; when a filament of wide extent is disap- 

 pearing at the limb, the latitude of the prominence observed 

 agrees with that of the end of the filament and the two change 

 together. 



Magnetic storms on the earth are most frequent when 

 prominences reach the poles. As this, however, occurs at the 

 time of sun-spot maxima, the two phenomena cannot be directly 

 correlated. It can only be said that both are indications of 

 great solar activity. It seems, however, that a sun-spot may 

 only be definitely assigned as the cause of a magnetic storm 

 when associated with an eruptive prominence. It is interest- 

 ing to note also that the solar corona at the stage in solar 

 activity when prominences reach the poles contains long 

 streamers right round the disc, whilst when the high-latitude 

 prominence zones die away there are only small tufts at the 

 poles. 



It was first remarked by Mrs. Maunder that sun-spots con- 

 sistently predominate on the eastern hemisphere of the sun 

 as compared with the western, and this was tentatively attri- 

 buted to an extinguishing effect of the earth on solar activity. 

 The same result is found to hold for prominences, and not only 

 are the prominences on the east more numerous, but they are 

 also more active and denser. The only cause to which it 

 seems possible to attribute this is an earth influence, and the 

 presence of such is apparently supported by the fact that 

 the line-of-sight movements in prominences are more fre- 

 quently away from the earth than towards it. The remark- 

 able fact is that no influence of any other planet can be detected. 

 The eastern preponderance recently has not been quite so 

 marked, but for thirty years was very persistent ; it remains 

 to be seen whether there is a gradual oscillation or not. An- 

 other result found to hold both for sun-spots and prominences 

 is a greater activity in the southern hemisphere than in the 

 northern. 



There is not much information available as to the speed 

 of rotation of prominences. The indications at present are 

 that they rotate at a faster rate than the chromosphere, and 

 that the speed of rotation is slower in high latitudes than in 

 low ; the first result is supported by a determination of the 

 speed of rotation of the corona made by Bosler from plates 



