380 Mr. Barlow on the effects produced in the rates of 
Practical deductions from the results of the preceding experiments. 
The first general conclusion which may be drawn from the 
foregoing experiments, is, that the rate of a chronometer is 
undoubtedly altered by its proximity to iron bodies. 
Secondly ; it appears that it is by no means a general case, 
that iron necessarily accelerates the rate of a chronometer, as 
would appear from Mr. Fisher's observations ; for five out 
of the six chronometers which I have made use of, were ob- 
viously retarded in every situation in which they were placed. 
In one instance only, viz. chronometer No. II, there is an 
indication of acceleration in one situation ; but it is more 
doubtful than the retardation in all the other five. 
It is also very obvious from the experiments on Nos. IV. 
and V., that much depends on the direction of the balance 
with respect to the iron : thus, No. IV. lost nearly 2" per day 
when its 12 o’clock hour mark was turned to the South, and 
only seven tenths when it was placed to the East ; but as 
soon as the chronometer was returned to its old direction, the 
loss again became 2"'i daily. The same occurred in the 
case of No. V., which lost 3"* 6 per day in one direction, and 
gained o"-5 in another at right angles to it ; and on returning 
it again to its former direction, the losing rate became 4"*i per 
day, viz. rather stronger than at first. It must be admitted, 
however, that the same striking difference in the rate, as de- 
pending upon direction, was not observed in another instance, 
when a similar experiment was repeated on the same chro- 
nometer. Speaking generally, it also appears, that the greatest 
effect is produced in those instances where the change in the 
magnetic intensity is the greatest *, but there does not seem 
