scientific theory of evolution, the moral sense has been developed in ourselves[28]. The truth is that in this matter natural theologians begin by assuming that the First Cause, if intelligent, _must_ be moral; and then they are blinded to the strictly logical weakness of the argument whereby they endeavour to sustain their assumption. For aught that we can tell to the contrary, it may be quite as 'anthropomorphic' a notion to attribute morality to God as it would be to attribute those capacities for sensuous enjoyment with which the Greeks endowed their divinities. The Deity may be as high above the one as the other--or rather perhaps we may say as much external to the one as to the other. Without being supra-moral, and still less immoral, He may be un-moral: our ideas of morality may have no meaning as applied to Him. But if we go thus far in one direction, I think, _per contra_, it must in consistency be allowed that the argument from the constitution of the human mind acquires more weight when it is shifted from the moral sense to the religious instincts. For, on the one hand, these instincts are not of such obvious use to the species as are those of morality; and, on the other hand, while they are unquestionably very general, very persistent, and very powerful, they do not appear to serve any 'end' or 'purpose' in the scheme of things, unless we accept the theory which is given of them by those in whom they are most strongly developed. Here I think we have an argument of legitimate force, although it does not appear that such was the opinion entertained of it by Mill. I think the argument is of legitimate force, because if the religious instincts of the human race point to no reality as their object, they are out of analogy with all other instinctive endowments. Elsewhere in the animal kingdom we never meet with such a thing as an instinct pointing aimlessly, and therefore the fact of man being, as it is said, 'a religious animal'--i.e. presenting a class of feelings of a peculiar nature directed to particular ends, and most akin to, if not identical with, true instinct--is so far, in my opinion, a legitimate argument in favour of the reality of some object towards which the religious side of this animal's nature is directed. And I do not think that this argument is invalidated by such facts as that widely different intellectual conceptions touching the character of this object are entertained by different races of mankind; that the force of the religious instincts differs greatly in different individuals even of the same race; that these instincts admit of being greatly modified by education; that they would probably fail to be developed in any individual without at least so much education as is required to furnish the needful intellectual conceptions on which they are founded; or that we may not improbably trace their origin, as Mr. Spencer traces it, to a primitive mode of interpreting dreams.