ON COMPARING AND EEDUCING MAGNETIC OBSERVATIONS. 99 



is a woi'k of considerable labour — tbe results would hardly be accepted 

 -as final until observation had done its work in every navigable sea and 

 on every shore open to the explorer, in proof of the theoretical results. 



It has already been remarked that we know far more of the earth's 

 magnetism from observation in Arctic regions, where the approximate 

 position of the noi"th magnetic pole has been determined, than in the 

 Antarctic regions where the position of the south magnetic pole is yet 

 indeterminate. Thus it appears that great advantage to the science of 

 terrestrial magnetism would be derived from a new magnetic survey of 

 the southern hemisphere extending from the parallel of 40° S. as far 

 towards the geographical pole as possible. 



For carrying out such a survey we have many advantages over our 

 predecessors in the Erehus and Terror, besides the benefit of their 

 experience. At Melbourne there is a magnetic observatory equipped with 

 all the most modern apparatus which would form an admirable base sta- 

 tion, whilst subsidiary base stations might be formed at the Cape Observa- 

 tory, and Sandy Point, Magellan Strait, for the use of the portable abso- 

 lute instruments. The survey, too, must in a great measure again be 

 carried out on board ship at sea, and here we have a powerful aid in steam 

 which would enable an observer in calms and moderate weather to obtain 

 excellent results by the process of swinging the ship. The observations 

 at sea might be accompanied with considerable advantage by observations 

 made with the portable absolute instruments on ice as frequently as pos- 

 sible. Ice is specially mentioned as being free from the local magnetic 

 disturbance which is common in islands and on i-ocky shores of igneous 

 formation. 



But the valuable aid of steam, which enables the seaman and observer 

 to handle his vessel with ease and precision, involves a large increase of 

 iron in the ship in the form of engines, boilers, &c., and a corresponding 

 increase of trouble to the magnetician. Those who have read Sabine's 

 detailed account of the errors of the compass, due to iron in the ships 

 Erebus and Terror, will have found that the deviation of 3° or 4° in 

 the compass at Hobart Town, became 50° in the high southern latitudes, 

 which must have all but annihilated the earth's directive force on certain 

 courses and rendered the compass useless. 



Experience derived from the magnetic results of H.M.S. Challenger 

 and other ships of the Royal Navy points to a means of avoiding much 

 of this difficulty, as well as to the selection of a suitable vessel, and above 

 all to a proper position on board — considered magnetically — for the 

 instruments. The importance of this latter point will be appreciated 

 when it is remembered that the errors of the observed magnetic elements 

 due to the direction of the ship's head can generally be eliminated by 

 swinging the ship, whilst those proceeding from vertical magnetic forces 

 are constant for every direction of the ship's head when upright — variable 

 when she is inclined at different angles of heel, and requiring frequent 

 references to a base station to ascertain their amount. 



On all accounts, therefore, it is necessary that directly a vessel is 

 selected for a magnetic survey, positions for the compasses and relative 

 magnetic instruments used at sea should be determined after careful 

 experiment, and all iron within 30 feet of them removed if possible. 



Subject to these precautions, a magnetic survey of the Antarcfic seas 

 might be made with satisfactory precision and great advantage to the 

 science of terrestrial magnetism. 



h2 



