326 APPENDIX. 



rose ratlier above the centre of gravity of the box and watch ; so that 

 they could not be displaced unless the ship were upset. The shelves, 

 on which the sawdust and boxes were thus secured, were between 

 decks, low down, and as near the vessel's centre of motion as could 

 be contrived. Placed in this manner, neither the running of men 

 upon deck, nor firing guns,* nor the running out of chain-cables, 

 caused the slightest vibration in the chronometers, as I often proved 

 by scattering powder upon their glasses and watching it with a 

 mao-nifying glass, while the vessel herself was %'ibrating to some 

 jar or shock. 



All the watches were in one small cabin, into which no person 

 entered, except to compare or wind them, and in which nothing else 

 was kept. The greater number were never moved fi-om their first 

 places, after being secured there in 1831, until finally landed at 

 Greenwich in 1836. 



Durino- eight years' observation of the movements of chronometers, 

 I have become gradually convinced that the ordinary motions of a 

 ship, such as pitching and rollmg moderately, do not affect tolerably 

 good timekeepers, which are fixed in one place, and defended from 

 vibration as well as concussion. Frequently employing chronometers 

 in boats, and in very small vessels, has strengthened my conviction 

 that temperature is the chief, if not the only cause (generally speak- 

 ing) of marked changes of rate. ITie balances of but few watches 

 are so well compensated as to be proof against a long continuance of 

 higher or lower temperature. It often happens that the air in port, 

 or near the land, is at a temperature very different from that over 

 the open sea — in the vicinity ; and hence the difference sometimes 

 found between harbour and sea rates. The changes so frequently 

 noticed to take place in the rates of chronometers moved from the 

 shore to the ship, and the reverse, are well known to be caused partly 

 by change of temperature and partly by change of situation.! In 

 the Beagle we never found the watches go better than when their 

 boxes were bedded in saw-dust, and they themselves were moving 

 freely in good gimbals. 



Suspending chronometers, as on board the Chanticleer, not only 

 alters their rate, but makes them go less regularly ; and when fixed 



» The Beagle's guns were long six and long nine pounders, of brass : 

 tbey were only fired from the foremost ports, 

 t This may be connected with magnetism. 



