SURFACE AVATERS OF THE NORTH ATLANTIC OCEAN. 
bo 
about the same in January as in Jiily. In the latter month the isothermal curve of 
7°'5 C. runs from Newfoundland far toward the north, l)eyond tlie whole of Europe ; 
it corres})onds in general with the January curve of 2'^‘5 C. ; the amplitude of tlie 
Gulf Stream, therefore, the fluctuation Ijetween its maxiniTun and minimum 
temjjerature (July to January, or August, and February) would be, on the average, 
only about 5° C. 
This general description was amplified in many directions l^y the additional data 
collected by special expeditions, and by the discussion of surface oljservations made 
on board merchant vessels, during the period 1870-90. Amongst the former are to l)e 
noted the contributions of H.M.S. “ Challenger” (1873-76) (2), which made soundings 
in the western part of the Atlantic basin ; of the Norwegian North Atlantic Expedition 
(1876), which explored the whole eastern part of the basin up to Spitsbergen; of 
H.M.SS. “ Knight Errant ” (1880) and “ Triton ” (L882), chiefly in tlie Faeroe-Shetland 
Channel; the different expeditions to the Polar Seas ; the expeditions of the Danish 
ships “ Fy 11a ” and “ Ingolf” (1877-79); of the United States Coast and Geodetic 
Survey (3), especially within the region of the Gulf Stream properly so called; and 
amongst the latter, the meteorological and hydrographical services of Denmark, 
Germany, Great Britain, and the United States. 
Stated in the most general way, the eflect of the increased information was to 
reduce the relative importance of the Gulf Stream current; it was recognised that 
while in the first instance most of the stream moving northward near the eastern 
seaboard of the United States was derived from the I'egion outside the West Indies, 
and did not pass through the Strait of Florida, that stream did not cordhiue as such 
much beyond the south-eastern extremity of the Newfoundland Banks (4). At the 
same time, observations at the higher latitudes, while defining more fully the general 
northward movement of the waters in the upper layers of the Eastern Atlantic, 
brought out the unexpected magnitude of the Polar streams moving southward both 
at the bottom and at the surface. Both results led, after much discussiou, to 
increasing belief in the direct frictional action of wind as the })iime factor in oceanic 
circulation, the gravitational influence of inequalities of specific gravity being 
relegated to second place, except in the greater depths : this doctrine was filially 
established by the observations of Murray (5) in Scottisli loclis. 
Up to this point little or no attempt was made to ascertain the limits within which 
the circidation in different parts of the ocean was liable to periodic or other varia¬ 
tions, notwithstanding the emphasis with which Petermann had insisted on the need 
of it. The seasonal variations in the strength of the Gulf Stream, recognised by 
Petermann, and indeed the seasonal variations of all currents except in the monsoon 
regions, were practically ignored, either because their existence was disbelieved in, or 
because they were assumed to be so small as to be Iieyond investigation with the 
available observations, and therefore too small to prevent any observations being com¬ 
parable, The only exception to this was the preparation by Meteorological Depart- 
