THE AURORA. 65 
which, however, no observer can usefully attack without previous 
experience in spectroscopic matters. An observer who has had such 
experience will probably know what to do, but attention may be 
drawn to the extreme importance of obtaining photographic records 
of the spectrum of the aurora, especially if a satisfactory reference 
spectrum can be added to the plate. Possibly the solar spectrum 
(obtained from the sky or moon) may be sufficient, but in case a small 
induction coil is available a selection of Geissler tubes would prove 
useful. 
The best summary of existing observations connected with 
aurore and their connection with other phenomena is contained in 
relation to ‘Les Aurores Polaires, by A. os (Felix Alcan, Paris). 
a 
