The Geographical Journal (Journal of the Royal Geographical Society): Volume XII, No. 2 [‎264r] (30/154)

OCEAN, AND ITS RELATION TO OTHER OCEANOGRAPHICAL PHENOMENA. 121
squares in the Indian ocean, from its northern limits to lat. 50° S., 83
per cent, contain observations in the two months of February and
August, 13 per cent, contain observations in one only of these months,
while merely 4 per cent, are blank as regards observations in these
two months.
Of the total number of 2° squares in the Atlantic ocean between
60° N. and 60° S., 67 per cent, contain observations in the two opposite
months, 19 per cent, contain observations in one only of these two
months, while 14 per cent, contain no observations in the two months
specified. As in the case of the Indian ocean, very few observations of
surface temperature are available in the Atlantic south of lat. 50° S.
Thus, of the total number of 2° squares in the Atlantic between 60° N.
and 50° S., 71 per cent, contain observations in the two opposite months,
21 per cent, contain observations in one only of these months, while
merely 8 per cent, are blank as regards observations in these two
months.
Of the total number of 2° squares in the Pacific ocean between
lat. 60° N. and 60° S., 46 per cent, contain observations in the two
opposite months, 28 per cent, contain observations in one only of these
months, while 26 per cent, contain no observations in these two months.
The zone of latitude from 50° to 60° S. contains relatively many more
observations than in the case of the Atlantic and Indian oceans, so that
the exclusion of that [zone does not alter the proportions in the Pacific
to any appreciable extent, the percentages for the Pacific between 60°
N. and 50° S. being respectively 45, 29, and 26 as compared with 46,
28, and 26 given above for the Pacific between 60° N. and 60° S.
It will thus be seen that, on the whole, the Pacific basin is the one
in which our knowledge of surface temperature is most deficient, and
therefore the one in which future observations may be expected to effect
the greatest modifications in the map showing the range of temperature.
In this ocean the greatest deficiency is within the tropics, the largest
number of blank squares occurring between 10° S. and 10° N., from
which zone the number of blank squares gradually decreases as far
as the 50th parallels north and south; by far the larger proportion
of blank squares lie within the six zones between 30° N. and 30° S.
On the other hand, in the Atlantic and Indian oceans the number of
blank squares is comparatively insignificant, and the blank squares are
not grouped together in any particular latitude.
When we consider the recorded temperatures at different seasons of
the year towards the arctic and antarctic regions north and south of the
60th parallels, our knowledge is very deficient; still a tolerably correct
idea of the range of temperature within the ice-covered seas can be
formed, when we remember that the minimum temperature must always
be somewhere about 28° Fahr., so that the summer temperatures which
have been recorded by explorers may be made use of in arriving at an