6 
THE VOYAGE OF H.M.S. CHALLENGER. 
A summary of the results is given in that paper at page 42, where it is shown 
that at a depth of 2500 fathoms, or nearly 3 miles, the mean correction of all the 
thermometers examined is only 0 o, 27 Fahr., the largest correction of any one of these 
thermometers being 0 o, 41 Fahr. 
The observations of deep-sea temperatures strictly as observed, no correction being 
thus shown to be needed, are published in the Challenger Reports, together with 
the same temperatures represented by curves on 258 plates. 1 The curves 
are drawn, libera manu, through the observations, which are fully detailed at 
the top of each plate, allowance being thus made for such errors of observation as 
10°, 5°, 1°, &c. The figures adopted in constructing the accompanying maps of deep-sea 
temperature are taken directly from the curves. In dealing with the other observations 
employed in this discussion, the observations as read are alone employed, any correction 
for pressure on the bulbs of the thermometers being in no case allowed for. As regards 
the important series of observations made by Commander Belknap, U.S.S. “ Tuscarora,” 
it is stated in his Report, p. 18, what corrections w 7 ere applied to the thermometers for 
depths of 100, 300, 500, and 700 fathoms, but no further explanation is given. An 
examination, however, of the observations themselves at all depths as printed makes it 
sufficiently plain that Captain Davis’ correction of half a degree Fahr. for each mile 
of depth had been applied throughout. Before, therefore, entering the temperature 
observations of the “ Tuscarora ” on the maps, they were increased by the amount by 
which they had manifestly been diminished, before publication, by the application of 
Captain Davis’ rule. It was interesting to note that when the “ Tuscarora’s ” course 
was in the vicinity of the course of the Challenger, “ Egeria,” &c., the additions made to 
the published readings brought them into accordance with the readings of these other ships. 
As this discussion deals only with depths at which the temperature is virtually 
uninfluenced by season, no depths less than 100 fathoms are examined. The 
temperature maps which accompany this Report are for depths of 100 fathoms, and 
then, in succession, for each additional 100 fathoms down to 1000, then for 1500, for 
2200 fathoms and another for bottom temperatures. It is plain that in this discussion 
no temperatures of the surface of the sea are admissible except good mean annual 
temperatures of the surface, it being these and these alone that rule deep-sea 
temperatures. Hence it is impossible to overestimate the value of Lieut. Baillie’s map 
showing the mean annual distribution of the temperature of the surface. 
In order to arrive at a correct knowledge of the mean temperature of the ocean 
considered as a whole at each of the depths selected, all the temperatures at each depth 
were averaged, excluding the temperatures which are from so-called “ closed areas.” 
The mean temperature thus arrived at is entered on each map, and in drawing the 
isothermal lines, all temperatures above this mean are represented by red lines, and 
1 See Ante, No. 1, footnote, p. 3. 
