At 
Farther, the questions whether there is a Creator, and whether 
there have been creative causations, enter into this argument, 
not as theological, but as natural questions. In their rela- 
tions to the inductive problem, they are as purely physical 
questions, as the question whether a given rock is the result 
of fusion or sedimentary deposition from water. A moment’s 
reflection will show the justice of this statement. And hence 
it follows that an a posteriori analogical argument on this topic 
is entirely fragmentary and inconclusive, until the claims of 
this parole-witness are entertained and adjusted. The his- 
torical and the physical parts of the argument cannot be ° 
thus rent asunder and legitimately pursued apart. 
The second rule of induction which applies to show this 
reasoning invalid, is that pointed out on p. 10. If there may 
be two antecedents, either of which is competent efficiently to 
produce an effect (naming one of them A, and the effect X), 
the closest possible induction can only prove that all A’s will, 
ceteris paribus, produce X; but cannot prove that all X’s are 
produced by A. Now, until atheism is demonstrated, another 
competent cause for natural structures may be supposed as 
possibly existing in the existence and action of a God. ‘And 
whatever is the strength of the probable or demonstrative 
evidence that there is a God, from whatever valid quarter 
drawn, there is just so much probability of error in the 
attempted induction, which assigns a natural origin to all 
structures. ‘To attempt to exclude the divine cause by the 
force of this @ posteriori analogy is to reason in a circle ; 
because the validity of the analogy depends wholly on the 
prior exclusion of the divine cause. Second, a wise Creator 
must have had some final cause guiding his action. We should 
not be so presumptuous as to surmise in advance what par- 
ticular final cause prompted a given creative act, but when his 
own subsequent action has disclosed it we are on safe ground. 
It is always safe to conclude that the object for which a wise 
and sovereign Creator produced a given thing is the object to 
which we see him devoting it. When, therefore, we see him 
in his subsequent: providence subjecting all things to the reign 
of natural law, we may safely conclude that, when he created 
them, he designed to subject them to natural law. But that 
which is to be ruled by natural law must needs be thoroughly 
natdral in traits. Hence this Creator must have made the 
first structures, which in their origin were supernatural, in 
their properties entirely natural. Whence it follows that the 
inference from naturalness of qualities to a natural origin 
would be, as to those structures, wholly worthless. Let it be 
repeated also: that whatever probability or certainty there is 
