APPENDIX, 



DEALING WITH CERTAIN CRITICISMS. 



ONE way of estimating the validity of a critic's judgments, is 

 that of studying his mental peculiarities as generally displayed. 

 If he betrays idiosyncrasies of thought in his writings at large, 

 it may be inferred that these idiosyncrasies 'possibly, if not 

 probably, give a character to the verdicts he passes upon the 

 productions of others. I am led to make this remark by con- 

 sidering the probable connexion between Professor Tait's 

 habit of mind as otherwise shown, and as shown in the opinion 

 he has tacitly expressed respecting the formula of Evolution. 



Daily carrying on experimental researches, Professor Tait 

 is profoundly impressed with the supreme value of the experi- 

 mental method; and has reached the conviction that by it 

 alone can any physical knowledge be gained. Though he calls 

 the ultimate truths of physics " axioms," yet, not very con- 

 sistently, he alleges that only by observation and experiment 

 can these " axioms " be known as such. Passing over this in- 

 consistency, however, we have here to note the implied propo- 

 sition that where no observation or experiment is possible, no 

 physical truth can be established; and, indeed, that in the 

 absence of any possibility of experiment or observation there 

 is no basis for any physical belief at all. Now The Unseen 

 Universe, a work written by him in conjunction with Professor 

 Balfour-Stewart, contains an elaborate argument concerning 

 the relations between the Universe which is visible to us and 

 an invisible Universe. This argument, carried on in pursu- 

 ance of physical laws established by converse with the Uni- 

 verse we know, extends them to the Universe we do not know: 

 the law of the Conservation of Energy, for example, being 

 regarded as common to the two, and the principle of Con- 

 tinuity, which is traced among perceptible phenomena, being 

 assumed to hold likewise of the imperceptible. On the strength 

 of these reasonings, conclusions are drawn which are consid- 



575 



