REPORT ON THE MAGNETICAL RESULTS. 3 



knowledge of the changes going on in the magnetic elements in the regions of 

 previous observation. 



In presenting the accompanying charts of the magnetic elements for 1880, 

 numbered I. to IV., it is thought that they will not only be acceptable to magneticians 

 as showing the distribution of magnetic force and direction for that year, but, when 

 compared with Sabine's, to indicate the general tendency of the secular change for the 

 previous forty years. But before comparing these later charts with their predecessors, 

 and before their value or otherwise can be duly determined, it seems necessary that the 

 various steps in their construction should be given in detail. Let the large share of 

 data contributed by the Challenger be first considered. 



An ideal vessel for carrying out a magnetic survey at sea is one in which there is 

 no iron used in her construction, or at least with the iron so distributed as to have little 

 or no effect at the position of the magnetical instruments used for the observations, and 

 further that she should be an easy vessel at sea under ordinary circumstances. 



The Challenger can hardly be said to have come up to this ideal in either 

 respect, for she was seldom at rest from pitching or rolling motion at sea, and although 

 north of the magnetic equator the errors of the compass and Fox dip and intensity 

 apparatus were moderate and could be eliminated by occasional " swinging " of the 

 ship, the errors caused by the vertical component of her magnetism were large, and, 

 although quite manageable, necessitated a frequent comparison with normal values on 

 land. This magnetic condition of the ship was not without its compensation, for south of 

 the magnetic equator the hard and soft iron which had previously combined to produce 

 errors in the observed values of the magnetic elements, had now opposite signs, and 

 when near the Antarctic circle and far from a point of comparison on land, had but 

 small effect. 



It may be urged that the differences observed between the results on board the 

 ship and those on land might not be a true measure of the effects of the iron of the ship 

 on account of possible local magnetic disturbance at the land station selected. In some 

 places this was no doubt true, but from a lengthened discussion of observations made 

 in numerous places in both hemispheres, where no traces of such disturbance could be 

 found, the magnetic condition of the ship could be ascertained at any period of the 

 voyage. This knowledge was not only fruitful as a means of reducing the observations 

 made at sea to normal values, but during the visits of the vessel to the neighbourhood 

 of places either known to be or suspected of being affected by local magnetic disturbance, 

 the amount of such disturbance could be measured with considerable accuracy. 



Local magnetic disturbance in the solitary islands of the great oceans was especially 

 open to this method of detection, for the vessel could be brought sufficiently near them 

 to avoid errors due to difference of geographical position, and yet in sufficiently deep 

 water to be free from magnetic effects in the land. To illustrate this, some results 



