No. 7.] HORIZONTAL INTENSITY. 71 
This table shows that the values found for « and C agree so well for 
the two positions of the double needle, P, and P,, that it may be considered 
immaterial in which position the double needle is used as deflected magnet. 
As mentioned in the introduction, determinations of constants were also 
attempted after the return, in Hamburg, in March, 1897; but as there are 
no simultaneous observations of absolute determinations, with any other 
instrument, of the value of the horizontal intensity, I have only been able 
to calculate wu, with the following result: 
0°11573 | 0°075845 
0°14007 | 0:091998 
0°11753 | 0:076812 
On the 7th March, only a series of deflections were taken, with deflector V 
and the small needle, and no vibration observations. I have therefore em- 
ployed for the calculation of w the time of vibration of magnet V, found on 
the 2nd March. In spite of the unfortunate circumstances under which these 
observations were made, the results, as it will be seen, agree fairly well with 
the values subsequently found in Wilhelmshaven. 
EMPLOYMENT OF THE OBSERVATIONS FOR THE VERIFICATION OF THE 
CONSTANTS. 
Although, as will be understood from the above, it may with tolerable 
certainty be taken for granted that the constants w and C, at any rate as 
far as magnet V is concerned, have remained unchanged throughout the 
expedition, I have considered ‘it worth while to attempt, as far as possible, 
to make use of the observations made during the expedition itself as a 
further check, the more so as it will also afford an opportunity for a more 
careful study of the temperature-coefficient «, whose determination in Hamburg, 
in 1893, was the result of observations made within comparatively narrow 
temperature limits. It is also expressly stated in Dr. NeumMayer’s manuscript, 
that there is reason to suppose that the temperature-coefficient has a different 
value in extreme temperatures such as those in which it is often necessary 
