440 REPORT—1883. 
obtained from two sets of apparatus differing in size, but otherwise similar in con- 
struction, and on contrasting these with the currents registered by the galvano- 
meter; and lastly aad most importantly the absence of any comparisons of the so- 
called aurora-line with other spectra, for the purpose of elucidating the still obscure 
problem of the real nature of the auroral discharge, such discharge having hitherto 
proved not to accord with the spectrum of any artiticial electric discharge yet pro- 
duced. His object is rather to invite careful watch and look-out at the present time 
for auroral displays, in order that, so far as the less favourable conditions of this 
climate will permit, some at least of the conclusions arrived at by Pref. Lemstrém 
may be tested. 
For the purpose it will be necessary for observers to employ— 
1. A spectroscope of large field and low dispersion, but provided with some 
means for measuring the positions of the lines seen. 
Owing to the faint character of the light obtained, the ordinary filar micro- 
meter is not applicable, and both bright and dark points or lines, which are some- 
times substituted for the micrometer wires, have each a disadvantage. 
In fact, the approach of extraneous light (which is generally needed in such cases) 
during the primary examination of an auroral spectrum is undesirable. 
On the whole the author thinks the simplest and best form of micrometer is 
obtained by the expedient of dividing the slit plate longitudinally into two halves, 
and maling the upper half traverse the lower by a suitable micrometer movement. 
In this way no artificial illumination of the field is resorted to, the citron line. 
the brightest of the group, being used to measure the fainter ones, the position of 
this in the lower spectrum being previously arrived at and indicated by a dark fixed 
point or index. 
2. A galvanometer and pointed collecting apparatus should be employed, which 
may follow the lines of Professor Angstrém’s apparatus, as recently described in 
‘Nature,’ so far as the circumstances of the locality and its surroundings will 
ermit. 
4 3. To this latter it would be desirable to add some form of Professor Thom- 
son’s quadrant, or portable electrometer, for the examination of the condition of 
atmospheric electricity pending the discharges. 
4, Some simple form of auroral transit instrument for obtaining heights and 
parallaxes of beams and arches, so that another big floating beam may not again 
catch observers unprepared, as in November last, and leave them with a good deal 
of guess-wwork as to altitudes and compass-points. 
Lastly, whether the aurora examined be one ranging freely above, or one, so 
to speali, held captive by a‘streaming’ apparatus, it is most necessary to obtain 
direct comparisons of its spectrum (after first securing as close position-measure- 
merts of the lines seen as may he) with other spectra of an appropriate nature. It 
is not possible in the short limits of this paper to indicate the direction of these 
comparisons, 
As they have so signally failed in result hitherto, a wide field is still open for 
experiment ; and if anything like a permanent auroral display can be secured by 
the electrician, upon the chemist will then fall the task of finding a spectrum which 
will aptly compare with it. If an aurora can be truly brought down to the earth’s 
habitable surface, unattainable conditions of pressure and temperature can no longer 
be set up as excuses for failure, and it should be strange if the spectrum of the 
aurora remains much longer a mystery and a puzzle. 
7. On some Indefinite Integrals, that contain the Elliptic Integrals E and ¥. 
By Dr. D. Birrens pe Haay. 
In a Memoir in the Transactions of the Amsterdam Academy of Sciences, ‘ An 
Appendix to the Tables of Indefinite Integrals,'1 the author gave the reduction- 
formule for the general indefinite integrals 
’ Verhandl. Koninkl. Akad. Wetenschap., Vol. xxii. (1883). 
