424 THE VOYAGE OF H.M.S. CHALLENGER. 



and the density 1"0240. Station D, the most southerly where serial waters were collected, 

 shows the distribution of density in the neighbourhood of the pack. The comparatively 

 light stratum of water at the surface is of but little thickness, at 50 fathoms the density 

 has already risen from 1*02452 (its value at the surface) to l - 02527; at 140 fathoms it 

 is T02542, and at 300 fathoms 1 '02553. While thus the density steadily increases with 

 the depth, at least as far as 400 fathoms, the temperature of the water falls from 32° at 

 the surface to 29°'2 at 50 and 29°"0 at 100 fathoms; at 300 fathoms it has risen to 

 33°"8. At Stations B and C, both approximately on the same parallel, the rise of density 

 with increasing depth is less marked, and there is no such irregularity in the distribution 

 of temperature. At Station C, which is 14° of longitude to the west of B, the temperature 

 at all depths is very decidedly lower than at the corresponding depths at B, while much 

 farther east, at Station H, the water is much warmer and denser than even at B, though 

 the position is 5° farther south. No serial waters were obtained on the way south 

 between the 47th and the 64th parallel. At Stations D and E, waters and tempera- 

 tures were taken in the vicinity of the pack. It will be seen that the temperature 

 falls to a minimum at about 100 fathoms, while the density rises to about T0255 at that 

 depth, and remains nearly constant at greater depths, while the temperature rises to 33° 

 or 34° at 300 fathoms according to the latitude. The density of the bottom water was 

 usually from T0254 to 1"0256. The observations at B, D, and E are exceptions. At B 

 and E the densities observed are almost identical with those of the surface water at the 

 same locality, and at D it is identical with that of water between the surface and 50 

 fathoms. The only risk attending the collection of bottom water is that due to the 

 possibility that the water-bottle may close at or near the surface, and thus enclose surface 

 water which it would take to the bottom and bring back again. Although there is no 

 reason for believing that this took place at each of these three Stations, it is possible that 

 it may have done so, and the results may be considered as doubtful, and the bottom water 

 may be assumed to have an average density of l - 0255. It must be observed that the 

 collection of the intermediate water is attended by no such danger. At the bottom 

 the temperature was, owing to the nature of the thermometers, uncertain, but there can 

 be little doubt from the indications which it was possible to have, that it was lower than 

 at 300 fathoms, though it may not have reached the minimum of 29° observed in the 

 superficial water. In regions free from ice the temperature of the bottom water was 

 found to be somewhat above the freezing point of fresh water, namely, 33°, and as this 

 temperature persists at the bottom without sensible alteration as far as equatorial regions, 

 it is probable that the bottom water in the deeper regions of the Antarctic Ocean is due to 

 a mixture of water cooled to a low temperature in these regions with water drawn in from 

 a lower latitude and with a higher temperature. This will be easily understood if the 

 effect which will be produced on a sea when its surface is frozen be considered. For 

 this purpose some knowledge of the nature of sea water ice is necessary. 



