Magnetism, Cohesion, Adhesion and Viscosity. •ilS 



slight magnetism of the needle, the points of which are in- 

 serted in the straws, it preserves a definite position, without 

 the slight magnetism of the needle apparently interferino- with 

 its sensibility. It has likewise the advantage of being easily 

 set up or dismounted to protect it from injury, and is capable 

 of being contained in a very small box. I ought to remark 

 likewise that it is so susceptible, that if unprotected by a screen 

 it is kept in a sort of perpetual movement, by the otherwise 

 inappreciable currents of air in the room. 



Proper forms of the needles and magnets. 



After the trial of needles of various forms and lengths, I 

 have found none that answer so well as those which are em- 

 ployed in the compasses of commerce, set in mahogany boxes 

 two inches square, and to which I have already alluded. The 

 needles in these boxes are an inch and a half long, very slen- 

 der and pointed, weighing about three grains and a half. 

 Their caps are of brass, which in this case is an advantage ; for 

 needles with those of agate are much too tremulous from their 

 great sensibility, and the experiment will be almost sure to 

 fail if the point of the needle slips in the least from the spot 

 where it first settles, the circumference of its action being as 

 limited as the part of the point of the needle actually in con- 

 tact. Long needles, therefore, owing to their great oscilla- 

 tions, are particularly unfit for these experiments; while, on 

 the contrary, I have succeeded extremely well with a needle 

 not more than five-eighths of an inch long. 



The preceding directions simply concern the kind of mag- 

 netic needle that is to be employed ; but it is, as 1 have said, 

 by the aid of a bar-magnet that the needle is made to attach 

 itself to non-ferruginous bodies ; for by no other means could 

 it be caused to approximate itself by a slow, steady, and gentle 

 movement, and be left without disturbance when the motive 

 agent is withdrawn, a condition upon which the success of the 

 experiment entirely depends in the cases of low intensity ; but 

 for those of higher attraction, I shall presently show a sim- 

 pler mode of attaining the same end without a magnet. The 

 bar-magnet which I use is very convenient; it is six inches 

 and a half long, three-quarters of an inch wide, and rather 

 better than a quarter of an inch thick. If the magnet is long, 

 it is managed with difliculty; and if small, it requires to be 

 brought very near the object, and there is consequently danger 

 of disturbing the position of the needle when the magnet is 

 withdrawn, which had better likewise be wrapt in soft paper, 

 to enable it to be laid down on the table without causing ajar 



