8 Lovering and Bond on Magnetic Observations at Cambridge. 
clock for a much shorter time ; but it requires four observations and 
the accidental error of these may exceed that produced in the first 
case by the clock. Lastly, the third mode requires only two ob- 
servations and depends on the clock only for a short time; but it 
supposes the Tabular place of the star to be accurately known. A 
careful determination of the meridian must be based upon a combi- 
nation of the three methods; when this is done, the instrument is 
adjusted with great exactness. This reduction has been made out 
of the transits in the table, and the azimuth of the instrument when 
on the meridian mark, calculated from 18 different sets of stars 
selected according to the known conditions of the problem, gives as 
a mean result 
0.072 west of south, being the azimuth of the south end of the Transit instru- 
ment. 
The three successive transits of Polaris observed on the 14th and 
15th July give 
0.11 west of south. 
The final mean is 0/’.091 west of south. 
The octagon apartments to the west of the Transit-instrument con- 
tain a Gauss Magnetometer by which changes of magnetic declina- 
tion are observed. They are built of wood, with copper and zinc 
nails; the walls rest on wooden posts; iron, stone and all other 
substances known or suspected to exert magnetic influence having 
been carefully excluded from every part of the building. FD is 
the direction of the astronomical meridian and D C of the mean 
magnetic meridian. The three circles at D are the projections of so 
many wooden posts which are bound firmly together at the top 
and support the marble table on which a Variation-transit is placed. 
This instrument was made by Troughton and Simms, and is used 
in the magnetic observations. The larger interior circle represents 
