12 Art. 9. -T. Teracla: 



ponds nearly to 17*8 sec. Closely in front of the drum a fine 

 horizontal slit and a cylindrical lens are introduced, hoth of the 

 same length as the drum. During the wet season, a desiccator 

 dish containing calcuim chloride was placed under the drum. 



For giving time-marks on the record, a clockwork was placed 

 in front of the lamp to interru})t the light for a few mimutes at 

 the end of each liour. 



The three magnetic instruments are arranged in succession as 

 shown in Fig. 1. For the Z-instrument an auxiliary lens with the 

 focal length of 2 m. was placed in the path of the incident beam 

 of light, by means of a special holder fixed to the pedestal. 



To save the considerable breadth of the photographic paper 

 required, owing to the remarkable sensibility of the instruments, 

 the following device was adopted Avith success. Two long strips 

 of thick plane glass plate (breadth 10 cm., length 3 m.) were 

 placed along both sides of the long pedestal Pi (Fig. 1, T). These 

 were held firmly by means of special holders, with their refiecting 

 surfaces vertical and parallel to each other, so that when the 

 luminous beam reflected from the magnetic instrument is deviated 

 just beyond the limit of the horizontal slit of the photographic 

 apparatus, it is caught by one of the glass plates, or ' ' optical 

 traps" as we have called them, and reflected back to the slit. 

 For a still greater deviation the opposite glass caught the l:)eam 

 reflected from the first one, and so on. Except in the extremely 

 damp season, even the third reflection produced a luminous spot 

 intense enough to affect the photographic paper; By properly 

 adjusting the position of the three instruments and also the mean 

 direction of the reflected beams produced l:)y them, an uninter- 

 rupted record of the three components could be ol)tained, even in 

 the case of remarkable magnetic storms. 



The mutual magnetic influences of tlie three instruments 

 were tested by mechanically disturbing each instrument, while 

 the photographic record was being taken. No sensible effect was 

 noticed. 



Since the drum carrying the photographic paper revolved 

 once per o'76 hours, paper of a length of about 4*85 m. was 



