78 W. H. Finlay.—Preliminary Result fer Longitude, §c. 
The discordance between the East and West observations, is easily 
explained by a small error in the adjustment of the theodolite: but. 
the mean of the East and West results should be nearly free from. 
the effects of this error. 
hm s 
The disappearance of the star was noted at 8°9°41 by the watch,,. 
h m 8 : 
or at 8°11°34°3 local mean time. Assuming the tabular quantities for 
the moon’s place given in the Nautical Almanac and an approximate: 
h m 8 
longitude of Bloemfontein equal to—1°44°53 from Greenwich, I get 
the equation. 
Al == ~-— 0°64 — 2°34Aa+ 2°65A6 — At 
Where Af = correction to assumed longitude in seconds of time 
Aa = 3 » R.A. of Moon Hs are 
ND = ” ” Dec. ” ” ” 
Np Se rv 3 local time Be time 
The deduced correction, therefore, to the assumed iongitude is 
very small, if the moon’s place as given in the Nautical Almanac: 
is fairly correct. On this point however I have as yet no control.. 
As I said before, it was a cloudy night at the Cape: an observation 
here of the same occultation would have made the result differential,. 
a very different thing to an absolute determination. Possibly when 
the observations of the moon made at other observatories are published 
I may find an observation on this date which will give me the means. 
of determining the values of Aa and Ad, and of arriving at a more 
definite result. But I do not wish to lay any stress on this. Before 
long a very accurate longitude of Bloemfontein will without doubt 
be detérmined by telegraphic signals with the Cape Observatory : 
but Ido want to lay stress on the accuracy that can be secured in 
remote places by these observations compared to the rough results 
obtained by travellers trusting to the rates of a pocket chronometer 
over long periods. All that is wanted is a watch that will keep. 
going fairly well for an hour, an intelligent observer and a fine night 
and a result can be relied on for longitude to a mile or two 
