608 JNTRODUCTION, 



galvanometer, by placing an insulated copper plate or ball above 

 the mast-head, or in an elevated position above the Observatory, 

 with an insulated wire connecting it to the galvanometer, the 

 other galvanometer- wire being connected with a metal plate buried 

 at some depth in the ground or in contact with water. For this 

 purpose and for detecting Earth currents, a delicate thick- wire 

 galvanometer would be required. 



As far as lines of equal cold can be laid down from the observa- 

 tions made in the most extreme latitudes on the temperature, the 

 dryness and the direction of the winds, it would seem that a pole 

 of greatest cold may possibly be met with in the region to the 

 north-west of Melville Island and Wellington Channel ; and sup- 

 posing that the tides which meet in Wellington Channel are two 

 bra.nches of the Baffin's Bay tide, it seems probable that in the 

 neighbourhood of the pole of greatest cold there is also a pole or 

 region where the three tides meet which enter the Arctic regions 

 by Behring Strait, Baffin's Bay, and the North Atlantic or Spitz- 

 bergen Sea. What relation these facts may have to the supposi- 

 tion that cold upper currents descend to the Earth's surface at the 

 pole of greatest cold remains to be considered. Of true currents 

 independent of the tides there seem to be few. The Gulf Stream 

 flows to the north on the west and east of Spitzbergen, and from 

 the north or north-east come two currents divided by North 

 Greenland, the westerly current probably flowing through Robeson 

 Channel and Lady Franklin Strait westward, beyond which we 

 know nothing of it ; but we find in Barrow Strait and Bellot Strait 

 and Lancaster Sound a return current flowing from the west into 

 Baffin's Bay. 



As these currents seem to have nothing to give rise to them but 

 the Gulf Stream and the Siberian rivers, may not the current on 

 the north of Greenland flowing westward be regarded as the 

 continuation of this stream, which, flowing till it strikes against 

 Kellett Land, is reflected back to the south of Melville Island ? 



There is one very remarkable fact with regard to the direction 

 and temperature of the winds. On the east and north of Spitz- 

 bergen it has been found by Lieutenant Weyprecht and others 

 that the prevailing winds are from the south-east and east, driving 

 the ice round the north of Spitzbergen. As we go westward to 

 Polaris Bay and Kennedy Channel, and almost everywhere imme- 

 diately west of Greenland except at Van Rensselaer, the prevailing 

 winds are from the north-east and east, except in summer, and 

 they are invariably warm winds loaded with vapour. On the east 

 of Greenland the prevailing winds, except in summer, are from 

 the north ; they also are the warmest winds, and are loaded with 

 vapour. This seems to show that the prevailing winds to the east 

 of Spitzbergen, blowing over the region of the Gulf Stream, are 

 warmed by its heat, and take up vapour ; retaining their warmth 

 and vapour, they become the north-east warm winds of the regions 

 west of Greenland and the north winds of the East Greenland 

 coast. 



