Survey of Northeast Greenland. 129 
do not seem to have become any surer by means of the more 
circumstantial method of observation, which is perhaps due to the 
circumstance that the rate of the watch at the low temperature 
becomes less reliable, the longer the observation lasts. Consequently 
when undertaking the final computation I have assigned the weight 1 
to all of the twelve observations indicated below. The approxi- 
mated value of the longitude has in each single case been I, — 
1514730. 
In the table I have under "Number of transits” given the number 
of moon transits on either side of the meridian. The temperature 
is quoted in order to give an idea of the conditions under which 
the observations were made. | 
Date emp. | Number å LEE | > | v? 
1907 | oftransits + | + | 
| | | | | 
Sept. 24 . | + 4°C 1 | Be | + 44 | 19.36 
u Inn ZE 1 UNSS г 351225 
№ i SG ca LE CR | BB 
Оса... | +14 10-1208 | | +122 | 14884 | 
RN | +14 1 5.3 | + 50 | 2500 | 
SPL LEONE | +17 il FAR 2 AEE Or) 0.81 | Two stars 
Novl7… НОТ 1 |e) | MOT | 292.0, || 7144.00. 
И: | 5294 О ЕЯ ER a еее 
= => 2 33 | EN 300 
2) REG HE: DE ITS | 104. | +107 | 11449 | 
BDA... = 2A 2 | MO END SO GER] 
— 25 29 2 11,384 | = 10 2109.01 
| | | 429 | 397 | | 782.18 | 
495 9 —_ $ 
en 292887 pu 4053. 
12 
Mean error of a single observation: m — aoe — 85.4 
m 
Mean error of the mean: м = Lies 25,4 
Final result: 2 = 1514™50s.3 + 254 w. of Grw. 
It may reasonably be assumed that the twelve determinations 
used give a fairly reliable value of the mean error of the mean, 
p — 2°.4. This error of the mean may be considered an expression 
of the accuracy which may be obtained by determination of longitude 
by means of the moon azimuth under the particular conditions. It 
must be borne in mind that the mean error when expressed in time 
or arc must necessarily increase, according as we approach the 
9* 
