acceleration of the pendulum in different latitudes. 1 79 
sufficiently to carry on the experiments, and to keep it up 
during their course, with the requisite steadiness and unifor- 
mity, must have altogether failed. It may not be amiss to 
remark, that notwithstanding the ho'fcse was as effectual for 
the purpose as the utmost liberality in the supply of materials, 
with no labour spared in their application, could produce, a 
very little wind with so low a temperature abstracted the 
heat with such rapidity, that the influence of the stove was 
scarcely felt beyond its immediate vicinity ; and a thermome- 
ter placed in those parts of the inner room where the clocks 
would have stood, could not be kept above zero, with such 
fire in the stove as it would have been prudent to maintain. 
The clocks were therefore suffered to remain unpacked 
during the winter in the inner room, whilst the outer served 
a variety of useful purposes, which could not have been con- 
veniently effected on board ship. 
On the 24th of February, the matting with which the walls 
of the outer room were covered accidentally caught fire, and 
notwithstanding the endeavours of the persons who were pre- 
sent, the fire was communicated rapidly to the roof ; it was at 
length fortunately extinguished by the exertions of the offi- 
cers and men from the ships, before the clocks or any part of 
their apparatus had received injury ; the packing chest alone 
of one was slightly scorched : the only personal sufferer on 
the occasion was an artilleryman, who had accompanied me 
on the voyage, and who, in his anxiety to place the instru- 
ments out of danger, exposed his hands incautiously, and was 
in consequence so severely frost-bitten, as to render necessary 
the amputation of three fingers of the left hand, and two of 
the right. 
A a 
MDCCCXXI 
