582 



APPENDIX. 



[No. III. 



The second general class of Aurora are those which propagate themselves 

 from different points of the compass, between north and west, towards the 

 opposite points ; sometimes also originating in the S. E. quarter, and extend- 

 ing themselves towards the N. W. They may be subdivided, like the former, 

 into the distant arches, which pass to the southward without much visible 

 change in their appearance ; and those which discover beams, and separate at 

 intervals into wreaths, flashes, and irregular segments, exhibiting all the phe- 

 nomena described above. In explaining the mode by which the two general 

 classes of Aurora are conducted into the horizon, I shall call the motion of the 

 arches (which is in a plane seldom deviating more than two points from the 

 magnetic meridian) the direct motion ; and that by which the Aurora propa- 

 gate themselves nearly at right angles to the magnetic meridian, the lateral 

 motion. Let us suppose a mass of Aurora to be modelled at its birth, in a 

 longitudinal form, crossing the meridians at various angles ; the whole to be 

 impelled with a direct motion towards the magnetic south, but the parts 

 having different velocities, and each extremity continually removing itself, by 

 a lateral motion, from the centre, so as to increase the length of the mass. If 

 the centre enter the northern horizon, it will appear like an arch, the real ex- 

 tremities being invisible ; and its direct motion will carry it to the southward 

 in that form. But if one extremity first enter the horizon, it will extend itself, 

 by its lateral motion, to the opposite point, passing, at the same time, by its 

 direct motion to the southward. 



Of the unequal velocities of the parts, we had many instances, in the direct 

 motion of the arches, by which the centres were often carried from the zenith 

 60° or 70° to the southward, while the extremities did not materially alter 

 their positions. Nor can this be accounted for by any application of the 

 rules of perspective, because arches exactly similar sometimes changed the 

 bearings of their extremities in proportion to the advance of their centres ; 

 and at Cumberland-House, on two occasions, the extremities of arches arrived 

 at the east and west points of the compass, while their centres remained only 

 10° above the northern horizon. 



There are two other forms of the Aurora which cannot be classed in either 

 of these descriptions, and the occurrence of which is not frequent. 



The first is a small mass, much elevated above the horizon, sometimes ex- 

 hibiting brilliant beams, and then disappearing or passing to the southward. 



