— ME 
registering by day and the 6 hour-cylinder at night; besides, 
on May 18 the 6 hour-cylinder was introduced during 
eclipse-time. For tbe Horizontal Int. the change from one 
registration to the other proceeded without any difference. 
The eye-observation of the second variation-apparatus had 
to procure the correction of transition, but on account of 
the failure of that observation, the determination of this 
correction was impossible as well. 
The thermometer at the H.-instrument was only read 
before and after eclipse-time on May 18“; this trifling 
uncertainty about the variation ot temperature was, however, 
amply made up for by the advantage of the magnetograph 
remaining wholy undisturbed. 
The temperatures read are: 
May 18, rn DRA 26.°29 
7-26 26. 18 
10 49 26. 70 
Ae Sap mn. 27. 20 
4 39 27. 54 
8 0 27. 54 
With these readings a very plausible graphic-curve of the 
presumable temperature-variation can be drawn, from which 
curve the temperatures below have been tabulated. 
In an astronomical camp the time was of course known 
with more than sufficient accuracy; moreover a time-signal 
being given on May 18" both at the beginning and the 
end of the registration on the 6 hour-eylinder, and also at 
2.33 p.m. With respect to the Hor.-Int. the magnetogram 
shows a clear and well-defined curve permitting an easy 
reading to tenths of millimeters. The curve for the Deeclin- 
ation was somewhat less distinct, but, considering its great 
smoothness, the tenths of millimeters could by careful reading 
be estimated with sufficient accuracy. 
In the following table the variations dH and HdD 
are given in absolute units: H being taken to be = 0.367. 
The zero-point has been chosen arbitrarily. 
Lt aai A 
