THE MAKERSTOUN OBSERVATIONS. 5 
during the adjustments; it was vibrated through small arcs only. Many of the 
apparent anomalies which I have observed in this instrument, and described in 
the Edinburgh Transactions, were probably due to the facts, that the needle rests 
upon knife edges, and that every care being taken in the management of the instru- 
ment, the consequences of such a mode of support were exhibited under the most 
favourable circumstances. 
I have referred to the imperfections of the inclinometer, in the introductions 
to the various volumes; one error I believe to be due to the presence of iron in 
the copper circle or basement of the instrument. 
The barometer, a standard by Mr Newman, was fixed to the south wall of 
the west ante-room, not the best possible position in the observatory, as it was 
subjected to greater variations of temperature than it would have been if placed 
to the north. The thermometers were placed on a revolving frame in front of one 
of the north windows; but the case was removed to the west in the morning, and 
to the east in the evening, when the sun shone near to, or north of the prime 
vertical. The anemometer is sufficiently well-fitted for the determination of the 
laws of variation, but there is considerable uncertainty as to its value for giving 
absolute results, as there is with most of the anemometers yet in use. From 
various causes, the working of the anemometer became less trustworthy in 1848 
and 1849. 
Times of Observation.—The ordinary observations were made at certain pre- 
viously arranged hours and minutes of Gottingen mean time. When the observer, 
from any cause, was too late for his observations, he was requested to note the 
exact minutes when the observations were made; this was done in all cases, and 
the true minutes will be found in connection with the printed observations. No 
check-clock was used, and none was necessary. Indeed, it is my opinion, where 
little confidence is placed in some portions of the labours of assistants, they will 
generally, in consequence, be found to deserve little trust in others; and this is 
especially true in the work of an observatory, requiring such a various fidelity 
and care. Besides the regular observations, large masses of observations were 
made during magnetic disturbances and other occasional phenomena. 
Reductions.—Before the reductions of the magnetical observations could be 
completed, certain factors were requisite, representing the effect of one degree 
Fahrenheit on the positions of the bifilar and balance magnets, and the values 
of a given change of angular position in terms of some known or easily found 
unit. It is not necessary that I should do more than refer to the fact, that I 
have proved the insufficiency of the processes originally prescribed for the 
accurate determinations of those co-efficients and have substituted new ones, of the 
accuracy of which there can be no doubt, as they have been verified in every possible 
B 
