TRANSACTIONS Of SECTION A. 575 



inquiries Are more consistent with the existence of a common cause, -which affects 

 the sun and the magnets on the earth. The question, however, was raised some 

 time since by Professor Garibaldi, and more recently by Sir Norman and Dr. 

 Lockyer, as to whether prominences may not supply the place of sun-spots in 

 those cases in which a great magnetic storm is unaccompanied by any spot. The 

 curve of frequency of great magnetic storms is exactly coincident with that of 

 solar prominences in high latitudes ; also, sun-spots do not occur in such latitudes. 

 Hence it would seem that to prominences should be attributed a special efficacy 

 in the causation of magnetic storms. But it may be argued, in the contrary sense, 

 that this coincidence is due to the fact that at times of solar-spot maximum 

 activity the disturbance on the sun is general, and extends to greater distances 

 froni the actual regions of the spots, both prominences and coronal streams 

 moving to higher latitudes with the advance of the sun-spot cycle. Moreover, it 

 is known that the profile area of prominences, and their curve of frequency, con- 

 forms to that of sun-spots. Hence it would appear that prominences cannot in 

 _ general be isolated from sun-spots as phenomena whicli are particularly active in 

 influencing terrestrial magnetism. But the question still remains whether particu- 

 larly large and violently eruptive prominences may not be effective in causing 

 magnetic storms in the absence of spots. The only method of answering this 

 question would be to make a detailed study of individual prominences that are in 

 any way noteworthy, and of the magnetic storms, to see whether any such relationship 

 subsists. As a contribution to such an investigation, two years have been dealt 

 with, for which very full observations of prominences have been published by 

 Father Fenyi, of Kalocsa, with detailed accounts of the larger and more note- 

 worthy eruptions. The years 1887-88 were also years of minimum sun-spots, 

 so that it would be easier to trace the connections, if any, which might exist 

 between prominences and magnetic storms in the absence of spots. A list of 

 forty-eight prominences was made from the observations, which were either 

 ■violently eruptive or distinguished by great displacements of spectrum- lines in the 

 line of sight, or attained a height of over 100". It was foimd that twenty-nine of 

 these were either immediately associated with spots or faculte, or occurred in the 

 spot-zones, this class including all the metallic prominences. As regards the 

 magnets, the maximum diurnal range of the declination was measured in each case 

 from the Stonyhurst curves. On only one of the dates on which a high prominence 

 was observed was there a magnetic storm, and, allowing three days before and 

 after each observation of a prominence, there were only nine active disturbances 

 of the magnets which occurred during such periods. In no single case can a 

 magnetic storm be with certainty associated with any given prominence, and great 

 prominences have occurred, with very large displacements of lines and violently 

 eruptive activity, without any answering swings of the needles. As with the spots, 

 so too with the prominences, the efficiency in the causation of magnetic storms, if 

 such exists, is exerted irregularly and capriciously. It is the general disturbance 

 of the sun and his surroundings which affects the earth's magnetism, and not any 

 particular manifestation of spot or prominence. 



6. Comparison of the Sj^edrum of Nitrogen and of the Aurora. 

 By Dr. A. Paulsen. 



On m'a fait I'honneur, pendant mon s^jour ici, de me demander de vouloir 

 presenter u la Section d'Astronomie et de M6t6orologie de la British Association 

 une photographie d'un spectre de I'aurore polaire que I'exp^dition danoise pour 

 explorer I'aurore polaire a pris pendant son seiour dans le nord d'Islande dans 

 I'hiver 1899-1900. 



Avant de vous montrer cette photographie je pense que peut-etre il pourrait 

 etre utile de faire quelques remarques sur la metiiode et les instruments que nous 

 avoDS employes et du caractere en g^n^ral du spectre. 



Sur ma demande le gouvernement danois m'a donn6 des moyens pour explorer 



