206 RICHARD THRELFALL. 
leave the iron surface normally, then the pressural forces are tan- 
gential, and we get the formula we have been using, and similarly 
if the tubes of induction are tangential (2.e. when the infinitesimal 
air gap separates similar poles), the pressures operate alone, and 
we have a repulsion equal to the former attraction, as in the 
elementary theory. If the tubes of induction leave the iron at 
any angle to the surface between 0 and 7/2 we must consider the 
effect of the pressural forces. 
To calculate these effects, it is convenient and perhaps correct 
to assume, that just as the internal stresses of the iron do not 
affect the forces which are the expression of the external ether 
tensions, so they do not affect the forces corresponding to the 
hydrostatic pressures. If, therefore, we consider a line of force 
running out of iron into air and making an angle 0 with the 
normal, we can estimate the direction and magnitude of the 
magnetic forces at once, thus : 
Let A B be the trace of a plane boundary between air and iron, 
and O N a normal drawn outwards into air. Let O Pbea vector 
in the plarte of the paper representing the tensional force on an 
infinitesimal area about O. Draw O Q perpendicular to O P and 
in the plane of the paper. Then the pressural forces lie in a semi- 
circle of which O Q is a radius and whose plane contains O Q, 
Since the pressures are symmetrical with respect to O Q, O Q is 
