18 REPORT — 1851. 



Having liberated the crystal it was boiled in hydrochloric acid, so as to remove any 

 impurity it might have contracted by contact with the iron. It was again suspended 

 between the poles and completely verified the foregoing anticipation. The line of 

 compression, that is, the magnecrystallic axis of the crystal, which formerly set from 

 pole to pole, set now equatorial. The experiment was then repeated with a common 

 vice ; various pieces of bismuth, protected by plates of copper, were placed within its 

 jaws, and pressed to the thickness of a shilling. The plates thus obtained, when 

 suspended from their edges in the magnetic field, exhibited one unvarying result : 

 the line of compression stood always equatorial, and it was a matter of perfect in- 

 difference whether this line was the magnecrystallic axis or not. In these cases no 

 gum was used, and not only was a ' predominance ' of axes present, but they all 

 worked together ; they were further assisted by the great mechanical advantage 

 offered by such plates to diamagnetic repulsion ; the line of compression, neverthe- 

 less, triumphed over all and determined the position of the crystal. 



The author concludes his paper as follows : — 



"Whoever denies the influence of proximity will have to answer the following 

 questions : — How is it possible that a greater differential action can be exhibited by 

 a cube of bismuth dough than by the crystal itself? What is it which causes the 

 magneciystallic axis to forsake its usual position, and to set equatorial when the 

 crystal is compressed in the direction of that axis ? He must further assume a cry- 

 stalline structure on the part of wax, flour, shale, and the pith of fresh rolls ; for in 

 all these substances the line of compression determines the position of the mass in 

 the magnetic field." 



Dr. Tyndall introduced an experiment in thermo-electricity with the monothermic 

 pile recently invented by Prof. Magnus of Berlin. It is well known if two small 

 bars, one of bismuth and the other of antimony, be united at their ends, the other 

 ends being connected by a wire, that on heating the place of contact of the bars an 

 electric current is evoked, which passes from antimony through the connecting wire 

 to bismuth, and from bismuth across the place of contact to antimony. These two 

 metals are mentioned because the phsenomenon is exhibited by them with peculiar 

 energy ; but any two metals will answer the purpose ; and even the same metal, 

 under certain circumstances. Thus Becquerel found that by knotting a wire, and 

 heating it in the vicinity of the knot, a current was developed ; it has even been 

 thought that a difference of diameter was sufficient to produce the effect, while some 

 have attributed it to a difference in the radiative power of the metals employed. M. 

 Magnus has been unable to substantiate these views. On taking a wire three lines 

 thick,and reducing a portion of it to the thickness of half a line, when the point of junc- 

 tion of thick and thin was heated, no current was produced ; and on taking a length 

 of polished wire, and roughening one half of it with a file or with coarse sand-paper, 

 when the point of junction of rough and smooth was heated, there was no current, 

 althotgh the radiative powers of both portions were very different. With a knot 

 upon the wire no current was obtained up to a temperature of 212° Fahrenheit, It 

 was necessary to heat the wire to redness, and by this means alter its molecular 

 structure before a current could be obtained. The principle of the monothermic pile 

 depends on this molecular change brought about by heating. A length of hard brass 

 wire was taken and heated to redness in alternate lengths of 6 inches ; these por- 

 tions became permanently soft, while the intervening portions remained hard. The 

 wire was coiled round a little wooden frame, and was of the shape of a rectangle. One 

 of the sides of this rectangle was composed entirely of soft wire, and the side oppo- 

 site of hard wire, while the centres of the remaining two sides were the junctions of 

 hard and soft. When the hard side of the wire rectangle was heated there was no 

 current, and this was also the case when the soft side was heated. But when the 

 junction of hard and soft was taken between the finger and thumb, the mere heat of 

 the hand was sufficient to cause a deflection of 50° by the needle of a galvanometer. 

 The investigation goes to prove that to the production of a thermo- current the contact 

 of different metals is virtually necessary, and that if the same metal exliibit the phse- 

 nomenon it is because its various parts are in different states of molecular aggregation. 



