At Gondokoro, in March, 1862, it was lent to Mr. Baker, who made all his observations with it, and brought it back safe: it has, therefore, been in use about 4 and 3/4 years.
On November 9th, 1865, Mr. Baker returned it to the Royal Geographical Society, and it was immediately taken to Mr. Casella, who tested its accuracy by trying its boiling-point, in nearly the same manner as Mr. Baker had made his observations. The result by two independent observers was that the boiling-point had increased in its reading by 0 degree point 75 in 4 and 3/4 years, or 0 degree point 172 yearly.
On November 23d the thermometer was again tested by Mr. Baker at the Kew Observatory. The observation was made under the same conditions as those near the Albert N'yanza, as nearly as it was possible to make it. (By immersion in boiling water.) The result gave the thermometer 0 degree point 80 too much at the boiling-point.
The readings of the thermometer have, therefore, been TOO MUCH; and by REDUCING the readings, it ELEVATES all positions at which observations were made.
Table No. 1.—In this Table the error obtained at Kew Observatory has been treated like that of a chronometer, the error being assumed increasing and regular.
Table No. 2 is to correct the height, computed by Mr. Dunkin, using the quantity taken from Table No. 1.
Table No. 3 is the final result of the observations for height, corrected for instrumental error.
TABLE No. 1.
Table for Increased Reading of Thermometer, using 0 degrees 80 as the
Result of Observations for its Error.
Month. 1861. 1862. 1863. 1864. 1865.
January. . . — 0'143 0'314 0'487 0'659
February . . — '157 '328 '501 '673
March . . . 0'000 '172 '344 '516 '688
April . . . '014 '186 '358 '530 '702
May . . . . '028 '200 '372 '544 '716
June . . . . '043 '214 '387 '559 '730
July . . . . '057 '228 '401 '573 '744
August . . . '071 '243 '415 '587 '758
September . . '086 '257 '430 '602 '772
October . , . '100 '271 '444 '616 '786
November . . '114 '285 '458 '630 0'800
December . . 0'129 0'300 0'473 0'645 —