§ 1. Let me begin by borrowing an example from Bain (_Logic_: B. III. c. 6). The North-East wind is generally detested in this country: as long as it blows few people feel at their best. Occasional well-known causes of a wind being injurious are violence, excessive heat or cold, excessive dryness or moisture, electrical condition, the being laden with dust or exhalations. Let the hypothesis be that the last is the cause of the North-East wind's unwholesome quality; since we know it is a ground current setting from the pole toward the equator and bent westward by the rotation of the earth; so that, reaching us over thousands of miles of land, it may well be fraught with dust, effluvia, and microbes. Now, examining many cases of North-East wind, we find that this is the only circumstance in which all the instances agree: for it is sometimes cold, sometimes hot; generally dry, but sometimes wet; sometimes light, sometimes violent, and of all electrical conditions. Each of the other circumstances, then, can be omitted without the N.E. wind ceasing to be noxious; but one circumstance is never absent, namely, that it is a ground current. That circumstance, therefore, is probably the cause of its injuriousness. This case illustrates:-- (I) THE CANON OF AGREEMENT. _If two or more instances of a phenomenon under investigation have only one other circumstance (antecedent or consequent) in common, that circumstance is probably the cause (or an indispensable condition) or the effect of the phenomenon, or is connected with it by causation._ This rule of proof (so far as it is used to establish direct causation) depends, first, upon observation of an invariable connection between the given phenomenon and one other circumstance; and, secondly, upon I. (a) and II. (b) among the propositions obtained from the unconditionality of causation at the close of the last chapter. To prove that A is causally related to _p_, suppose two instances of the occurrence of A, an antecedent, and _p_, a consequent, with concomitant facts or events--and let us represent them thus: Antecedents: A B C A D E Consequents: _p q r_ _p s t_; and suppose further that, in this case, the immediate succession of events can be observed. Then A is probably the cause, or an indispensable condition, of _p_. For, as far as our instances go, A is the invariable antecedent of _p_; and _p_ is the invariable consequent of A. But the two instances of A or _p_ agree in no other circumstance. Therefore A is (or completes) the unconditional antecedent of _p_. For B and C are not indispensable conditions of _p_, being absent in the second instance (Rule II. (b)); nor are D and E, being absent in the first instance. Moreover, _q_ and _r_ are not effects of A, being absent in the second instance (Rule II. (d)); nor are _s_ and _t_, being absent in the first instance.