22 DR. FARADAY’S EXPERIMENTAL RESEARCHES IN ELECTRICITY. (SERIES XXII.) 
long, or oblique, or irregular, still the magnecrystallic force predominated and deter- 
mined the position of the crystal, and this happened whether pointed or flat poles 
were used, and whether they were near together or far asunder. The magnecrystallic 
axis is perpendicular, or nearly so, to two of the sides of the rhomboidal prism. I 
have some small prismatic crystals of which the length is nearly three times the width 
of the prism ; but when both the length and the magnecrystallic axis are horizontal, 
no power of the magnet, or shape, or position of the poles, will cause the length to 
take the axial direction, for that is constantly retained by the magnecrystallic axis, 
so greatly does it predominate in power over the mere magnetic force of the crystal. 
Yet this latter is so great as at times to pull the suspending fibre asunder when the 
crystal is above the poles (2615.). 
2547 . Sulphate of nichel. — When a crystal of sulphate of nickel was suspended in 
the magnetic field, its length set axially. This might be due, either to mere mag- 
netic force, or partly to magnecrystallic force. Therefore I cut a cube out of the 
crystal, two faces of which were perpendicular to the length of the original prism. 
This cube pointed well in the magnetic field, and the line coincident with the axis of 
the prism was that which pointed axially, and represented the magnecrystallic axis. 
Even when the cube was reduced in this direction and converted into a square plate 
whose thickness coincided with the magnecrystallic axis, it pointed as well as before, 
though the shortest dimensions of the piece was now axial. 
2548. The persulphate of ammonia and iron, and the sulphate of manganese, did not 
give any indication of magnecrystallic phenomena ; the sulphate of ammonia and 
manganese I think did, but the crystals were not good. The sulphate of potassa and 
nickel is magnecrystallic. All three salts were magnetic. 
2549. Thus it seems that other bodies besides bismuth, antimony and arsenic, 
present magnecrystallic effects. Amongst these are the alloy of iridium and osmium, 
probably tellurium and titanium, and certainly the sulphates of iron and nickel. 
Before leaving this part of the subject, I may remark that this property has probably 
led me into error at times on a former occasion (2290.). A mistake with arsenic 
(2383.) might very easily arise from this cause. 
^ V. On the nature of the magnecrystallic force, and general observations. 
2550. The magnecrystallic force appears to be very clearly distinguished from 
either the magnetic or diamagnetic forces, in that it causes neither approach nor 
recession ; consisting not in attraction or repulsion, but in its giving a certain de- 
terminate position to the mass under its influence, so that a given line in relation to 
the mass is brought by it into a given relation with the direction of the external 
magnetic power. 
2551. I thought it right very carefully to examine and prove the conclusion, that 
there was no connection of the force with either attractive or repulsive influences. 
For this purpose I constructed a torsion-balance, with a bifilar suspension of cocoon 
