Actinometrical observations from Greenland. 373 
of-day in hours." The last pair of curves shows the annual increase 
and decrease of the mean actinometer number. For the construc- 
tion of these curves the numbers from Table IV were used, but in 
a somewhat modified form. By the modification the maxima were 
somewhat diminished and the minima somewhat more increased. 
The vertical lines show the limits of the months. The different 
latitudes express the real number of days of the month in question. 
The dotted vertical lines denote equinoxes, the lined ordinates the 
actinometer numbers, and the dotted ordinates the annual mean of 
both places. 
I have to thank Mr. CARL PETERSEN, the architect, for drawing 
up the table of curves. 
Some remarks on the Steenstrup-method. 
In the method of light-measurement by means of sensitive paper, 
introduced by WIESNER, the difficulty lies, as is well-known, in the 
production of standard, normal tones; as regards the STEENSTRUP- 
method, some tracing-pavers and sensitive papers are prescribed 
which are in the market and easily obtainable and it is assumed 
that these kinds of paper always remain constant within the limit 
of accuracy of the method. STEENSTRUP has found the tracing- 
papers to be constant, and he considers the “Solio”-paper used by 
him to be very durable also. But we know from practical photo- 
graphy that ‘it is not possible to produce two pieces of paper of 
exactly the same sensitiveness. According to WIESNER and others, 
we can at most presume that all the tracing-papers contained in 
the same packet are equally sensitive for the same emulsion; also 
the sensitiveness is altered, in spite of all the care that may be 
taken, by keeping. Therefore in carrying out the STEENSTRUP- 
method it is urgently necessary to adopt some method of control. 
STEENSTRUP has proposed to expose the photometer containing the 
paper to the light of a magnesium ribbon of a stated length for a 
stated time. An actinometer number will thereby be obtained, and 
later experiments, especially whenever a new packet is opened, may 
be made for comparison. It is much to be wished that the inventor 
of the method at any rate would, after experiments of this nature, 
publish a number so ascertained. These control numbers would 
be of particular importance to those who employ the method in 
far-off places like Greenland, where fresh paper is obtainable only 
1 The curves are calculated without considering refraction, according to the well- 
known formula: H = 90° — ¢ + 0; cos Т = —tan ¢g-tan д, where H is the 
altitude of the sun in the meridian, T the hour-angle of its setting, & the 
geographical latitude, aud © declination, + northern, — southern. 
