ON SOLAR RADIATION. 243 
corresponding to equilibrium between the torsional force at the moment 
and the deflecting force belonging to the thermo-electric current at the 
same moment. A special experiment showed, however, that such was not 
the case, and that the difference between the actual position and that 
corresponding to equilibrium was practically insensible, provided a very 
few seconds were allowed to elapse after any sudden change of the nature 
of letting on or cutting off the sun’s light. 
Tt follows from this that the simple differential equation mentioned in 
the report of the Committee for 1892 may be used in this case as well as 
when the solar rays actuated an internal thermometer. The integral of 
the differential equation shows that the curve is a logarithmic curve, the 
parameter of which, or index of the exponential, is constant, provided the 
constant relating to cooling, the q of that report, is the same under all 
circumstances. 
Measurement of the curves obtained showed that they agreed extremely 
well with logarithmic curves. Any two logarithmic curves, of which the 
parameter is the same, may be superposed by moving one relatively to the 
other without disturbing the parallelism of their axes, but not if the 
parameters are different. In general the curves obtained seemed to be 
sensibly superposable, indicating that gy was not merely constant during 
an exposure made under given circumstances, but even when the circum- 
stances were different ; as, for example, when one diaphragm was replaced 
by another of twice the area. In one case, however, it seemed that the 
coefficient was slightly but sensibly smaller. The constancy or otherwise 
of gis a point still under examination. Should it prove to be sensibly 
different when the circumstances are changed, the expression may still be 
obtained from three observations, that is, from three points of the curve. 
If, for the sake of simplifying the formula, the intervals of time are 
chosen equal, the expression (Av)?/(—A?w) has merely to be multiplied 
by q, the expression for which need not at present be written down, in 
order to obtain a measure of the radiation. It may be well to remark 
that this measure is merely relative ; the question of obtaining the radia- 
tion in absolute measure is one into which the Committee have not as yet 
entered. 
Bibliography of Spectroscopy.—Report of the Convmittee, consisting of 
Professor HERBERT McLeop, Professor W. C. Roperts—AUSTEN, 
Mr. H. G. Manan, and Mr. D. H. Nace. 
Tue work of collecting and arranging the titles of papers on subjects 
related to spectroscopy has been continued up to the present date. 
In the Report presented by the Committee last year it was proposed 
to discontinue the work at the end of 1895, and it was suggested that the 
four sections of the list of titles should be printed as a separate publica- 
tion. In view, however, of the meeting of the International Committee 
to consider the preparation of a catalogue of scientific literature, the 
Committee now proposes to continue the work up to the year 1900, so as 
to complete the list up to the time when the International Bureau will 
commence its labours. 
The Committee, therefore, asks for reappointment. 
