But an observation of the greatest importance readily occurs; that of any cluster, composing our idea of an object, the effects or consequents depend, in general, more upon one part of it than another. If a stone is hot, it has certain effects or consequences; if heavy, it has others, and so on. It is of great importance to us, in respect to those successions, to be able to mark discriminately the real antecedent; not the antecedent combined with a number of things with which the consequent has nothing to do. I observe, that other objects, as iron, lead, gold, produce similar effects with stone; as often as the name _hot_ can, in like manner, be predicated of them. In the several clusters therefore, hot stone, hot iron, hot gold, {315} hot lead, there is a portion, the same in all, with which, and not with the rest, the effects which I am contemplating are connected. This part is marked by the word _hot_; which word, however, in the case of each cluster, connotes also the other parts of the cluster. It appears at once, how much convenience there must be in dropping the connotation, and obtaining a word which, in each of those cases, shall mark exclusively that part of the cluster on which the effect depends. This is accomplished by the abstract or non-connotative terms, heat, and weight. Certain alterations, also, are observed in those parts of clusters on which such and such effects depend; which alterations make corresponding alterations in the effects, though no other alteration is observable, in the cluster, to which such parts belong. Thus, if a stone is more or less hot, the effects or successions are not the same; so of iron, so of lead; but the same alteration in the same part of each of those clusters, is followed by the same effects. It is true, that we know nothing of the alteration in the cause, but by the alteration in the effects; for we only say that a stone is hotter, because it produces such other effects, either in our sensations immediately, or in the sensations we receive from other objects. It is, however, obvious that we have urgent use for the means of marking, not only the alterations in the effects, but the alterations in the antecedents. This we do, by supposing the alterations to be those of increase and diminution, and marking them by the distinction of lower and higher degrees. But, for this purpose, it is obvious that we must have a term which is not connotative; because we suppose no alteration in any {316} part of the cluster but that which is not connoted; thus we can say, with sufficient precision, that a greater or less degree of heat produces such and such effects; but we cannot say, that a greater or less degree of hot stone, of hot iron, of hot any thing else, produces these effects. This then, is another use, and evidently a most important use, of abstract, non-connotative terms. They enable us to mark, with more precision, those successions, in which our good and evil is wholly contained.