A great deal has been said about the differ-ntiates of expansion of different bodies, but le rates of contraction seem to have been ather overlooked. Now to obtain a motive rawer by the means of heat, contraction and xpansion are equally necessary. To double le volume of a body is only one half of the work, and to bring it back to its original con-ition, constitutes the other Lhalf. Different >odies are held together by different rates of ohesive power, and in expanding we work gainst the cohesive power, and in contract-ng we work with it. Let z=amount of cohesive power. Let a=a certain amount of heat. Then a.c=the expansive power. a+x=the contractive power. And az+a+z=whole amount of heat sed. It will readily be perceived that whatever may be the value of x the whole amount o( leat used will always equal 2a. This may e further exemplified in the case ot the pis-on of a steam engine with an upright cylinder. If an extra amount of power is required ;o overcome the weight of the piston in the upward stroke, just so much the less power is equired to bring it down again, so that nothing is lost or gained by the weight of the piston. It follows theni that whether we apply the heat to air, water, carbonic acid gas hydrogen, mercury, or any other substance the result will be the same. A pound of coa contains a certain definite amount of heat just as it has a certain definite weight, and i is an error to suppose that by artificial mean we can increase either. C. Paterson, N. J.