561 
Mecea, however, it had a daily rate, in the mean — 0.78, which 
differed much from that at Jidda and was moreover very irregular, 
while during the journeys themselves the rate seems to have had 
about the same value as at Jidda. 
The following results were derived from this watch: 
1st by interpolation between the nearest determinatons of time 
at Jidda. 
March 11 + 2m 32510 
5 he 32.38 
se erie 34.52 
ee 25: 35.33 
3 33.77 
Je pee 33.98 
Mean + 2m 33:68 
Ind by means of quadratic formulae. Using also the observations 
of March 2 and 3, which are further away from the time of the 
journey, or leaving them out, we obtained 
+ 2m 34872 (+ 1:03) 
or 32.88 (2 1.08) 
3rd We obtained in the same two manners by means of linear 
formulae : 
+ 2m 33°08 (4 1227) 
or 33.64 (=, 4:02) 
Of the results by the quadratic formulae the mean of the two 
has been adopted; of those by the linear formulae it seemed best 
to adopt the second. 
Giving finally equal weight to the results of the 3 methods, the 
final result becomes; 
+ 2m 33°70 
differing rather much from those yielded by the other watches. 
g. General result from the 6 watches. 
For the derivation of the difference of longitude according to each 
of the employed watches, given in the preceding paragraphs, different 
methods of calculation were followed, which had all of them special 
advantages and disadvantages, and in most cases the mean of the 
results found by these different methods was adopted. Naturally in 
this procedure some arbitrariness could not be avoided; its influence 
on our final results, however, will not be great. 
37 
Proceedings Royal Acad. Amsterdam. Vol. XV. 
