Berthollet in this book points out the quantity of each base necessary to neutralize a given weight of acid, and he considers the strength of affinity as inversely that quantity. Now of all the bases known when Berthollet wrote, ammonia is capable of saturating the greatest quantity of acid. Hence he considered its affinity for acids as stronger than that of any other base. Barytes, on the contrary, saturates the smallest quantity of acid; therefore its affinity for acids is smallest. Now ammonia is separated from acids by all the other bases; while there is not one capable of separating barytes. It is surprising that the notoriety of this fact did not induce him to hesitate, before he came to so problematical a conclusion. Mr. Kirwan had already considered the force of affinity as directly proportional to the quantity of base necessary to saturate a given weight of acid. When we consider the subject metaphysically, Berthollet's opinion is most plausible; for it is surely natural to consider that body as the strongest which produces the greatest effect. Now when we deprive an acid of its properties, or neutralize it by adding a base, one would be disposed to consider that base as acting with most energy, which with the smallest quantity of matter is capable of producing a given effect. This was the way that Berthollet reasoned. But if we attend to the power which one base has of displacing another, we shall find it very nearly proportional to the weight of it necessary to saturate a given weight of acid; or, at least those bases act most powerfully in displacing others of which the greatest quantity is necessary to saturate a given weight of acid. Kirwan's opinion, therefore, was more conformable to the order of decomposition. These two opposite views of the subject show clearly that neither Kirwan nor Berthollet had the smallest conception of the atomic theory; and, consequently, that the allegation of Mr. Higgens, that he had explained the atomic theory in his book on phlogiston, published in the year 1789, was not well founded. Whether Berthollet had read that book I do not know, but there can be no doubt that it was perused by Kirwan; who, however, did not receive from it the smallest notions respecting the atomic theory. Had he imbibed any such notions, he never would have considered chemical affinity as capable of being measured by the weight of base capable of neutralizing a given weight of acid.