LOGIC 311 



often so arrange the conditions of bodies as to make 

 experiments with them. The latter process is especially 

 useful in studying the cause of any action, or condition, 

 of bodies. 



The cause (or causes) of anything must be sought 

 amongst its invariable antecedents or concomitants. 



The circumstances in which all the instances in which 

 anything takes place, alone agree, must be a cause. This 

 is called the method of agreement. 



If two instances have all their circumstances alike 

 save one, and the event (the cause of which is sought) 

 only occurs in that single case, then that one must, at 

 the least, be closely related to its cause. This is the 

 method of difference. 



If two instances in which Y occurs have X, and two 

 instances in which Y does not occur, have nothing in 

 common but the absence of X, then X is the cause of 

 Y. This is the joint method of agreement and difference. 



If we subtract from a given effect, all that is due to 

 certain causes, then the residue is the effect of the rest of 

 the causes. This is the method of residues. 



If X and Y increase, decrease, and vary together, 

 then one is the cause of the other, or closely connected 

 with such cause. This is the method of concomitant 

 variations. 



One effect may have several causes, therefore the 

 method of agreement is uncertain, but not so " the joint 

 method of agreement and difference." 



A cause is generally suggested by analogy, or resem- 

 blance, from cases in which the connection of a cause and 

 effect is better known. 



A cause will, as a rule, be the more easily discovered, 

 the greater the number of forms of the effect which are 

 examined. 



