364 APPENDIX V.

return journey, in August 1878, Dr. Emin carefully compared Mr. Felkin's " Newton " with his own " Naudet," and it was then found that the correction which had to be applied to the former, in order to reduce it to the " Naudet " as a standard, amounted to 0'32 in. Dr. Emin's aneroid is no doubt an excellent instrument, but no opportunity has occurred hitherto for trustworthily determining its index error. At Eubaga, where Dr. Emin observed during four months, as shown in Table III., his recorded readings are O'lOl in. higher than those of Mr.

Wilson's " Browning " ; whilst, as compared with Dovyak's mercurial barometer readings at Gondokoro, they are O'll in. too low for the whole year, and 0'07 too low for the month of August. It results from this that if we would reduce Mr.

Felkin's readings to Dovyak's barometer, 0'25 in.

(0-32 --07) must be deducted from them. The altitudes then found by reference to Dovyak's observations at Gondokoro, are shown in the columns marked a in the following table. Altitudes computed on the assumption that a correction of only '02 in. should be applied fall about 250 feet short of the figures given.

We have, in addition to this, corrected Mr.

Felkin's readings on the assumption that the error, futility of deducing trustworthy corrections from observations limited to a few days, and made with instruments the errors of which, under variable conditions, are not thoroughly well known.