254 
PROFESSOR FORBES ON THE EXTINCTION OF THE SOLAR RAYS 
of m, which we may call the coefficient of extinction, do not present any correspond- 
ence with the hygrometric variations. It is to be desired, however, that such curves 
should be extensively constructed. 
75. II. It has been assumed that the mass of air intercepted between Brientz and 
the Faulhorn was equal to the differences of the barometers multiplied into the secant 
of the zenith distance of the sun (Art. 33.). It does not appear, however, that the 
ratio of heat reaching the lower station, compared with that at the upper one, varies 
in a geometrical progression when the thicknesses vary arithmetically. But we can 
hardly thence argue against the hypothesis of uniform proportional extinction, be- 
cause the law of continuity is evidently not preserved. 
76. III. Are the variations of m from hour to hour to be considered as merely the 
result of errors of observation ? I apprehend that in some measure they may fairly 
be so considered, especially as resulting from a slight discontinuity in the obser- 
vations at Brientz before and after one o’clock, the former being made within doors, 
the latter without ; but the real analogy must evidently be of a somewhat compli- 
cated kind. A narrow inspection of the actinometric curves XII. and XIII., will 
illustrate this. It is one of the admirable results of graphical analysis, that we seize 
the slightest symmetry in the form of functions which might otherwise appear very 
dissimilar. 
77- Viewed generally, we observe in these curves, first, that they differ from the 
common diurnal temperature curves (which approach more or less to the curve of 
sines) by drooping more rapidly at each extremity ; secondly, that both curves have 
a morning and afternoon inflection before and after they attain their maximum ; 
thirdly, that the curve of intensities at the upper station lies wholly above the curve for 
the lower station ; fourthly, that the range of the former curve is greater than that of 
the latter ; fifthly, that the maximum is sooner attained in the former than in the latter 
case. 
78. Now the three geometrical characteristics last mentioned, make it plain that 
the law of the differences between the two curves must be a complex one. The ana- 
logy is very striking with the inquiry into the law of the decrement of temperature 
in the atmosphere at different hours and seasons, which I have fully considered in a 
paper in the Edinburgh Transactions*. I might, as in that case, reduce these curves 
to series of functions of sines, and show that the differential curve, having generally 
the same form, would admit of various maxima and minima in the course of the 
day, but I apprehend that for a single day’s observations the numerical results could 
not have much value. I am quite confident, however, that the jive peculiarities just 
mentioned of these curves will be found to be reproduced in every series made under 
equally favourable circumstances. If this be the case, the seeming irregularities which 
we are considering will be resolved into the more general consideration of the physical 
causes of the form of the actinometric curves. I shall make a very few remarks on 
each of the peculiarities above noticed. 
* Vol. xiv. p, 489. 
