738 THE AURORA BOREALIS. 



turn in a vertical plane about an axis at the top of the tube and 

 in a horizontal plane, by the sliding of the two tubes one in the 

 other. To this plate another plate is fixed by a vertical steel axis 

 about which it can turn, and the two are pressed together by a 

 screw and a circular spring. The upper plate carries a hori- 

 zontal arm of brass. Two prisms having a refracting angle 

 of 60° are fixed on this plate. Their bases are placed against a 

 vertical plate of metal with metal corners so that their bases form 

 an angle of about 166° '5. The prisms are enclosed in a square 

 box, which is fixed by screws. On one side of this box is a tube 

 carrying a slit, and on the other a tube with a telescope. An 

 arm of brass projects through the side of the box on the side of 

 the telescope, and a micrometer screw is attached at the end of it. 

 The rays of light entering by the slit fall on the lens and into the 

 tube, and are then refracted and dispersed by the first prism, and 

 afterwards internally reflected at the base of the second prism. 

 They then enter the telescope. The bases of the prisms are 

 inclined at an angle of 166° '5 to one another. By turning a 

 micrometer screw the prisms are turned slowly so as to bring any 

 ray into the middle of the field. The head of the screw is 

 graduated in hundredth parts, and the scale in intervals corre- 

 sponding to one turn. A reading of the screw determines the 

 place of a ray of the spectrum, and a comparison with other 

 scales or with wave-lengths may be made by fixing the relative 

 positions and taking the readings for a great number of known 

 lines of the solar spectrum. 



The readings of the scale and the positions of the lines may be 

 determined by alternately illuminating and obscuring the scale 

 during the observation, so that the lines of the spectrum and their 

 positions in the scale are blended together. 



Spectroscopic Observations. 



In the first Swedish Expedition, 1868, some very remarkable 

 observations were made on the appearance of luminous beams 

 around the tops of mountains, and by the spectroscope M. Lem- 

 strom has showed that these displays were of the same nature as 

 the Aurora. 



He mentions also that on the 1st of September, on the Isle of 

 Amsterdam, in the Bay of Smeerenberg, there was a light fall of 

 snow, and the snowflakes were observed falling obliquely ; all at 

 once there appeared a luminous phenomenon, which, starting from 

 the earth's surface, shot up vertically, cutting the direction of the 

 falling snowflakes, and the phenomenon lasted for some seconds. 



The characteristic yellow line of the Aurora is seen in the light 

 on the tops of the mountains, and its intensity is constantly 

 changing, and its variations are such as to show that the light is 

 discontinuous. This line of the spectrum was also observed, but 

 was of feeble intensity when the slit was directed towards a lake 

 covered with snow, or towards a roof covered with snow, and 

 even in the snow all round the observer. 



