318 
MR. A. W. RtrCKER AND DR. T. E. THORPE OH A MAGHETIC 
supposed to be situated, the more difficult is it to account for the fact that their 
effects are so local. 
A more conclusive answer is the fact that our observations show clearly that the 
needle is attracted to certain lines or points round which currents would have to 
circulate in the same direction with the hands of a watch) in order to account for 
the facts. If the magnetic field of the earth is due to currents flowing from east to 
west, it is easy to imagine that they might be deflected by layers of more or less 
than average conductivity ; but why should they form local eddies, in which the flow 
is always in the same direction ? They must circulate similarly round the Heading and 
Wash peaks, round Kells, probably also round the Antrim basalt and the Toad-stones 
of Derbyshire. In accordance with Commander Creak’s investigation, they must flow 
round magnetic islands in opposite directions in the two hemispheres. They must 
flow in difterent directions on the two sides of the palaeozoic ridge in the southern 
counties, of the Malverns, of the Yorkshire ridge, of the Scotch coal-field, and of the 
Caledonian Canal. It may be possible to imagine physical causes which would account 
for such a state of things, but it does not appear easy to frame an hypothesis which 
shall be more probable than that involved in the theory of magnetic rocks. 
As regards Dr. Naumann’s view, that geological faults determine local magnetic 
action, it is disputed in its application to Japan by Dr. Knott, who has superintended 
the recent magnetic survey of that country. We find &t Malvern a mass of crystalline 
rock bounded by a fault. Either the rock or the fault may be supposed to be the 
cause of the attraction toward the ridge which is undoubtedly exerted on the needle; 
but whereas the rock is susceptible of magnetisation, and its effect on the magnet is 
precisely such as would be produced if it were magnetised by induction in the earth’s 
field, there is no shred of direct evidence to prove that the fault is capable of causing 
or deflecting earth-currents, so as to make them flow in opposite directions on its 
opposite sides. It is more probable that the action of faults is due to the displacement 
of crystalline rocks in their neighbourhood ; and that a fault is, at all events, not a 
necessary cause of regional attractions is proved by our observations in Scotland. Two 
great fault lines traverse that country, one coinciding with the Caledonian Canal and 
the other running from Stonehaven to the mouth of the Clyde. The first of these is, 
and the second is not, associated witli a locus of attraction. 
Taking the next argument in order, we agree that mountains which exercise . 
considerable influence on the magnet when close to them produce no effect at distances 
which are small relatively to the range of regional disturbances. It must, howevei’, 
be remembered that the depth from the siuTace to magnetic rocks concealed by over- 
lying strata may not exceed the horizontal distances at which the Malverns and the 
Wash affect the Declination, and that their influence on the Vertical Forces might 
extend over vast areas. If, as would often be the case, they were not horizontal, but 
approached the surface by a gentle slope, the magnet would certainly tend to turn 
