The attraction of moisture by gunpowder, is known to be very great: according to Rees's Cyclopedia, upwards of 16 per cent. has been absorbed, and that the removal, simply, from near the fire to the corner of the room, produces a considerable change in its weight. I presume, therefore, that the lime, which in its caustic state has also a great affinity to water, attracts a portion of it from the powder, and leaves it in a state of dryness best fitted for inflammation. But if the lime should remain too long mixed with the powder, it would probably attack the water of crystallization of the saltpetre, and, according to Count Rumford's idea, destroy a great part of the power. If also left exposed, attractions of moisture would take place from the atmosphere, the gunpowder would remain surcharged with humidity as before, and the lime would be only an inert mass. The examination of this subject led me to consider the increase of the power of gunpowder in various situations, and of its use in the field. It is well known that after a few discharges a cannon becomes heated, and the range is much greater, as well as the recoil. The charge of powder is therefore reduced about one quarter, to produce the original effect. As I have not heard or seen any explanation of this fact I shall take this opportunity of mentioning, that it appears to arise from the same cause as the first explained, viz. the desiccation of the powder. No person will dispute the heat acquired by a cannon, or even a musket, after repeated discharges; and this heat must volatilize or destroy a great portion of the moisture combined with the powder, assist its speedy inflammation, and perhaps add to its power, by causing a more perfect combustion of the inflammable parts of the gunpowder. This would cause a much greater volume of gas to be produced, and the high temperature would also greatly augment its elasticity; and it is well known that the effects of gunpowder depend upon the rapid production and high degree of elasticity of a great quantity of aeriform fluids or gases. ART. XX. _On The Connexion between Magnetism and Light._ By Col. GIBBS. _Extract from a Letter to the Editor._