In 1778 Mr. Thompson was with Lord G. Germain at his house, Stoneland Lodge, Sussex. Whilst there Thompson made experiments on testing gunpowder, and on a new method of determining the velocity of projectiles. The results were sent to the Royal Society, in 1781, and were published at great length in the 'Philosophical Transactions.' One good observation is now of great interest. 'Being much struck with the accidental discovery of the great degree of heat that pieces acquire when they are fired with powder without any bullet, and being desirous of finding out whether it is a circumstance that obtains universally, I was very attentive to the heat of the barrel after each of the succeeding experiments, and I constantly found the heat sensibly greater when the piece was fired with powder only than when the same charge was made to impel one or more bullets.'
In order to pursue these experiments he went in 1779, on board of the 'Victory,' of 110 guns, commanded by his friend Sir Charles Hardy. He passed the whole of the campaign on board of the fleet, and the result of the observations that he then made furnished the materials for a chapter which he contributed to Stalkart's 'Treatise on Naval Architecture.' He added to it a code of signals for the navy, which was not published. In his paper on gunpowder, read in 1797 to the Royal Society, he says:
During a cruise which I made, as a volunteer, in the 'Victory,' with the British fleet, under the command of my late worthy friend Sir Charles Hardy, in the year 1779, I had many opportunities of attending to the firing of heavy cannon; for though we were not fortunate enough to come to a general action with the enemy, as is well known, yet, as the men were frequently exercised at the great guns and in firing at marks, and as some of my friends in the fleet, then captains (since made admirals), as the Honourable Keith Stewart, who commanded the 'Berwick,' of 74 guns,--Sir Charles Douglas, who commanded the 'Duke,' of 98 guns,--and Admiral Macbride, who was then captain of the 'Bienfaisant,' of 64 guns, were kind enough, at my request, to make a number of experiments, and particularly by firing a greater number of bullets at once from their heavy guns than ever had been done before, and observing the distances at which they fell in the sea,--I had opportunities of making several very interesting observations, which gave me much new light relative to the action of fired gunpowder.