156 Lovering on Magnetic Observations at Cambridge. 
observations were made on that day, and not to the other hours. 
No account has been made of this difficulty in the monthly means; 
but, in making up the yearly mean, two columns are placed at the 
left hand of the Table on page 141, the second of which omits 
the months of January, February, and August, when observations 
were made only every even hour; the other retains them. Of 
course, the hourly means derived from the whole time will give 
a more accurate mean daily curve of declination changes when 
taken from the extreme right column, than when drawn from the 
one next to it on the left. Similar remarks apply to the baro- 
metric means of page 142. 
Concluding Remarks, explanatory of the Tables. 
The moment of observation has been given in mean aslronom- 
ical time, reckoned from mean noon at Gottingen. It is apparent 
that so many instruments as were under observation could not 
all be registered at the same moment by one person. ‘The time, 
given in the first column of the Tables which contain the ordi- 
nary observations, belongs strictly only to the observations on the 
Declination. The Horizontal-Force Component was observed two 
and a half minutes before, and the Vertical-Force Component two 
and a half minutes after that time. The attached Thermometers 
were observed after their respective instruments. The Barometer 
and the external Thermometer may, with sufficient accuracy, be 
assigned to the same moment as the Vertical-Force Magnetom- 
eter. The three lines that are inclosed in braces make what 
is called a Triple Observation. This consists of nine magnetic 
observations, three upon each of the three instruments. The 
moment when each of these observations was made will appear 
