22 
REPORT— 1854 . 
seen, were much greater in the low than in the high temperatures. It appear!, u 
before stated, that about 5 per cent, of the chronometers which have passed through 
the observatory are altogether unfit to be used for nautical purposes. With the menu 
which we possess fortesting, there is no difficulty in detecting such chronometers in 
a trial of two or three weeks. With regard to the remaining 95 per cent,, it appears 
that very great dependence may be placed on them if we take into consideration th 
change of rate due to change of temperature. Unfortunately, we have no record n 
the temperature to which the chronometers that have passed through the obwmton 
have been exposed during the voyage j but we find that the average of the 
of chronometers employed in the North American trade agrees with the rate* of tic 
same chronometers on shore in a temperature of about 60°, and that the average of tU 
sea rates of chronometers which have been exposed to a tropical cliraute daring ereii 
part of the voyage, agrees with the rates of the same chronometers on »hore ;r i 
temperature of about 80 °. The regularity with which we find this to take pi** " 
such as to leave no doubt as to the advantage which an intelligent navigator wom 
derive by correcting the rate of an imperfectly compensated chronometer for chur 
of temperature when the voyage is long and the climate changeable. Ui'-Jj-' 
however, have never been instructed to do this ; such a method of proceeding - 
no part of their education; and even in the examination which all ore nowcoinpt 
to undergo before they take the commaud of a ship, their attention!* not even o’ 1 
to the possibility of the rates of chronometers being affected by change of 
llie rates, as before stated, arc assumed to be uniform in nil climate*. Tk 1 
owners of the 300 chronometers, the variable rates of which arc shown in t e 
were made acquainted with this imperfection for the first time when they*™ 
chronometers to the Liverpool Observatory. I am not aware of the ellsttn J*. 
other establishment in the world in which captains of merchant *hip* canob 
information relative to the variation of the rates of their chronometer* 10 * 
emperatures. Attention appears to have been exclusively directed to 1 
vnrE miC c m . eans of det0Cm ' 11 the rate. Now, our records show thatit* ^ 
j q „°p !' a,c canted by chnnging the temperature bo small a <1 .. 
J f F «hrenheit, is very mSeh greater than the probable error a * 
' ® r . om a,i y imperfection in our astronomical means of determinmg ■ ^ 
JET- tbe f mte t0 affected by no other cause, the variation arising J 
l l a r r 18 s,) 8 rCat > that unless we take it. into account, the 
thf* bffr aUC i s , ea ' rafL ' mu8t be the result of mere accidental coincidence. Qndp n ^ 
Ihl Z f t0 b f thc Professional raters make themselves acq ” , 
Sra?? ° f thc V8rious chronometers which pass through their htt**, j: 
Bea ratl ia Sh ° re ‘ T™ to them * The «««» of this variation between ^ 
of vnrinV 8 8 t Dera .7 ftU Pl )08ed not. to be known. The duty of Ending o , j; y 
and discussing the sea- and shore-rate* »«“£***. 
of temnpr * ratCr ' and ca Ptaiu8 are kept in ignorance of its existence , ^ L b 
ra£Tri m T M an r ti,uc be caS ofthe chronometerdWJJJ ct ,: 
there cannor ir pt ?’ D ; barters do occasionally 
captains Ti? * d ,° ubt; and lfc » quite certain that they might be amide - rf ^ 
chronometers 8 U1 T C ac, l l,aint:£ ' d with the variation of rate pecatuu- *2* c**!** 
Ration Wift,’r° r Jy “ a *f ra removing the fault by effecting a wore P® obfe: at^ 
provided wirh tf pect to tbc former method, our records show that he «rb'- : 
O f ™te Z t h i necewar y a PParatus, it is quite practicable to •***»£+» ; 
makinirfh 0 change of temperature, and to supply captains wi 1 
■bM bin VVi,1 ‘ "*«nl to to. latter, the Miration ofebtof™' 1 b ,„ bt' 
patented ? tbe sul) j<?ct fur tuanv years; numerous me ^° d bv thf P- S 
lished rates of H^i t0 effcct a more Perfect‘compensation, and .^ e . nn uailF fo,l: - 
tliat three fourtl, cb r oaometer8 » which arc now sent to Greenwich rrr 
are ^ have some improvement a*** ^ 
tables which I now ‘ . , neetl only to appeal to the res tbo5 e c^' . 
Acting the compLSn aZo«'S Jiy W t^ouwTcTT& t***' ‘ 
