194 
PROCEEDINGS OP SECTION A. 
It would appear from result VII. that the error at Kurrachee 
vanishes in the sum of the two intervals Ispahan-Kurrachee and 
Madras- Kurrachee ; in which case the results VI. and VII. compare 
very well indeed, considering that the unknown personal equation 
(Fritsch-Pogson) is involved in VII. I think, however, that this 
latter value may not be used for any further purposes at present. 
It would be difficult to do proper justice to it, even if favourable 
assumptions were made, which is always a dangerous course. 
( t ) Madras- Singapore . — Observers : Dr. J. A. C. Oudeman, Sur- 
veyor-General of Java, at Singapore ; Mr. Pogson, at Madras. This 
measurement was made in July, 1871, by the exchange of galvanic 
signals through the cable, on the evenings of the 24th, 25th, 26th, 
and 28th. Mr. Pogson observed with the transit circle of the 
observatory and clock, but had to carry a mean time chronometer to 
the cable offices for exchange of signals at a distance of four miles. 
Dr. Oudeman made his time determinations on the 24th by observing 
zenith distances of two stars with a universal instrument. On the 
25th and following dates the observations were made with a “ broken 
transit instrument” — viz., one of the form in which the eyepiece is at 
one end of the horizontal axis. lie also had to carry his chronometer 
to the cable offices for exchange of signals at a distance of three- 
quarters of a mile. Observations at both stations were made by eye 
and ear — no chronographs being used. The personal equation of tile- 
observers was not determined — (13) page 11, and (15) page 69. The 
point to which Dr. Oudeman referred his longitude was the position 
of the flagstaff on Fort Canning in 1871 — seo (15) page 69, and (14) 
page 2 LI. 
This result was deduced by Dr. Oudeman — (14) page 214. 
(/,) Madras-Sing apore. — Same operations as in ( t ). Kesult given 
by Mr. Pogson— (13) pages 11-24. 
(f tt ) Madras- Singapore. — This determination was made by Lieut. 
CommaDder C. M. Davis, U.S.INr., at Madras, and Lieut. John A. 
Norris, U.8.N., at Singapore, in 1882 ; the operations being repeated 
on the five nights of January 20, 21, 23, 26, and 27. These officers 
made their time observations with the so-called “ broken transit 
instruments/ * which offer the great advantage that the observer 
remains in the same position during observations of stars at all 
altitudes — a condition greatly favouring the constancy of personal 
equation. They exchanged galvanic signals directly from their huts, 
thus avoiding the danger of having their chronometer rates accidentally 
disturbed, and errors of comparison. They had chronographs upon 
which their observations were recorded, and their personal equation 
was continuously tested by “ absolute personal equation instruments,” 
each observer being provided with one. This equation, however, was 
not introduced in the results, on account of its being always very 
small, and probably no greater than its possible variations. Cable 
signals were observed by reflecting galvanometers. The observers 
were especially well trained for that class of work, having made 
together many longitude determinations in various parts of the world. 
Their plans were all prearranged and methodically carried out, and 
the excellence of their results is shown in the agreement of values 
deduced from each night’s observations. The discrepancy between 
the values (t) and (t u ) is 0*71 seconds, and that between (t z ) and (t tl ) 
