November 29, 1907] 



SCIENCE 



753 



Dr. L. A. Bauer presented papers on " The 

 Apparent Alterations of the Mass of a Magnet 

 as disclosed by Weighings in Various Mag- 

 netic Fields " and " The Local Magnetic Pole, 

 Treadwell, Alaska." Both papers were treated 

 together and were illustrated by lantern slides 

 and diagrams showing instruments used and 

 giving results so far obtained. 



At the meeting of the American Physical 

 Society held last April in Washington, the 

 speaker had made a preliminary announce- 

 ment that the reduction of the carefully ex- 

 ecuted observations on the magnetic survey 

 yacht Oalilee indicated that there might not 

 necessarily be everywhere an exact " balan- 

 cing " or " coupling " in the magnetic forces 

 acting on the two ends of a suspended magnet. 

 If the action of the earth on a magnetic 

 needle does not consist solely in a couple, then, 

 contrary to our usual ideas, the needle, if free 

 to move, would be subject not only to a motion 

 of rotation but also to one of translation. 

 Accordingly a compass needle might be caused 

 to press against its pivot, or a dip needle exert 

 an additional pressure on the surfaces on 

 which the pivots rest. 



As far as known no further experiments to 

 test this hypothesis have been made since the 

 historic ones of Robert Norman, about the 

 year 1576. 



To have only a couple acting on the magnet, 

 it is necessary that the following three condi- 

 tions be fulfilled exactly: (a) that the forces 

 acting on the two ends or " equivalent poles " 

 be opposite, (b) that they be equal, (c) that 

 they be parallel. 



The non-fulfillment of any of these condi- 

 tions will give a resultant force besides that 

 of a couple. It is possible to assume magnetic 

 systems acting on the needle in which even 

 the first condition is not fulfilled, viz., that 

 the acting forces on the two ends shall be 

 opposite. And as to non-fulfilment of the 

 second and third conditions, this occurs in 

 every magnetic field not strictly homogeneous 

 or uniform. An obvious case under this head 

 would be the earth's field disturbed either 

 artificially by the too close proximity of iron 

 structures or of magnetic impurities in the 

 instruments used, or naturally by magnetic 



iron ore deposits near the surface. A case of 

 a magnetic system in which the forces acting 

 on the two ends of a magnetic needle are not 

 opposite but in the same direction is that due 

 to a vertical electric current passing through 

 the pivot of a compass needle. Any of these 

 cases may occur in nature or in practise. 



The hypothesis has been subjected by the 

 writer during the past summer to a series of 

 preliminary tests, involving careful weighings 

 of a magnet using a non-magnetic Becker 

 analytical balance, embracing the region of 

 the earth from Washington, D. C, to Sitka, 

 Alaska. 



The magnet was weighed in two horizontal 

 positions (noi'th end towards magnetic north 

 and next reversed so that north end was to- 

 wards magnetic south) and also in two vertical 

 positions (north end up and north end down). 

 The weighings were made at Sitka (Alaska), 

 Victoria (British Columbia), Baldwin (Kan- 

 sas) and Washington, D. C. ' At these places 

 the earth's magnetic field is approximately 

 uniform, i. e., no marked local disturbances 

 are known to exist. 



The magnet was invariably weighed in both 

 scale pans and the weighings in general were 

 extended over two days at each station. In- 

 variably for each station, the result for mag- 

 net, north end south, was greater than for 

 north end north, the average difference being 

 nearly 1/1,000,000 part of the weight of the 

 magnet (about 33.6 grams). The differences 

 in the weighings for the two vertical positions 

 (JJ-B) were sometimes positive and sometimes 

 negative, the average, regardless of sign, being 

 of about the same order of magnitude as for 

 the horizontal positions. 



Weighings were also carried out in the in- 

 tensely locally disturbed area at Treadwell 

 Point, Douglas Island (Alaska). Here the 

 average difference for four points of observa- 

 tion was 0.07 mg. or 1/500,000 part, in the 

 ease of the two horizontal positions of magnet 

 (weight for north end south being again 

 greater) and for the two vertical positions, 

 0.25 mg. (weight north end down being the 

 greater or nearly 1/100,000 part of the weight 

 of the magnet). The differences are thus 

 more pronounced — as was to be expected — - 



