SECT. 1] 



THE MAGNETIC FIELD OVER THE OCEANS 



189 



(d) 







w 



20 



40 



60 



Miles 



Fig. 9. Magnetic profiles and bottom topography in the North Atlantic. 



(a) 37° 48'N, 25° 57'W to 38° 50'N, 26° 15'W. 



(b) 44° 52'N, 27° OO'W to 45° 58'N, 27° 28'W. 



(c) 46° 20'N, 20° 24'W to 46° 25'N, 19° 12'W. 



(d) 45° 40'N, 06° 45'W to 45° 20'N, 05° 40'W. 



46|°N from 17°W to 23° W. The dashed curve shows the spectrum that would 

 be obtained if all wave numbers were equally represented at the sea bottom, 

 that is, if the spectrum at the sea bottom were "white". The observed spectrum 

 is, on the average, steeper than the dashed curve ; thus measurements at the sea 

 bottom would show a spectrum that falls towards the shorter wave-lengths. This 

 might be due to the sources of the disturbance being buried beneath the ocean 

 floor, or to the short wave-lengths being absent in the source. 



The loss of detail is also shown in Figs. 9a, b, c and d, which show the total 

 field over 70-mile sections of ship's track after the removal of a regional trend. 

 Fig. 9a shows a section of track in comparatively shallow water over the Mid- 

 Atlantic Ridge just north of the Azores ; the field changes by nearly 1000 y in a 

 few miles and the changes are correlated with the bottom topography. Fig. 9b 

 shows the field over the ridge in deeper water further to the north ; the fluctua- 

 tions are smaller and not so sharp. Fig. 9c shows the field over the approaches 

 to the ridge in a depth of about 2500 fm (this is part of the fine from which the 

 spectrum shown in Fig. 8 was obtained). Fig. 9d shows a section over the Biscay 

 abyssal plain where the bottom is flat and the sources are presumably buried 

 beneath the sediment. 



If the irregularities in the field are partly due to the inhomogeneities in the 

 crust and partly to motions in the fluid core of the earth, it is to be expected 

 that the spectrum would fall into two parts, a part due to the core with wave- 

 lengths of a few thousand kilometres and a part due to the crust with wave- 

 lengths up to some tens of kilometres. The intermediate wave-lengths might 

 be expected to be of smafl amplitude since the mantle at depths below about a 

 hundred kilometres will be above the Curie point and therefore non-magnetic. 

 The faU at long wave-lengths seen in the spectrum in Fig. 8 may show the short- 



