Vol. 52.] ROCKS OP THE LIZARD DISTRICT. 25 



their paper. 1 It amounts to this : that in an extensive district, 

 where the rocks exhibit many difficulties and complications, each 

 section or limited group of sections can be regarded by itself and be 

 made the basis of induction, independently of all the others. But 

 here, unless I mistake, lurks a fallacy, the nature of which will be 

 readily perceived by considering a particular instance as an illustra- 

 tion. Suppose the question to be whether an igneous rock is intrusive 

 into or contemporaneous with a group of sedimentaries. Suppose, 

 further, that in 11 out of 12 sections the evidence is inconclusive, but 

 that the remaining one gives a decisive answer on the one side or the 

 other. Then there would be no room for further doubt. The uncertain- 

 ties of the eleven would count for nothing against the positive evidence 

 of the one section. Suppose, however, that, of the 12 sections, the 

 evidence of four or five is rather more favourable to one conclusion ; 

 that of two makes any other highly improbable, while that of one 

 appears to accord better with the contrary conclusion ; and the 

 remainder give no answer at all. We cannot regard the testimony 

 of this one unfavourable witness, as if the others were non-existent. 

 Certainly no judge, probably no jury, would act upon such a principle. 

 The apparent conflict of evidence undoubtedly calls for a careful 

 cross-examination of each witness, but in such a case as we have 

 mentioned it is surely far more probable that the one section 

 should have been misunderstood than that a contradiction should 

 exist. The more specious interpretation does not always turn out 

 to be the true one, and we must proceed in our attempts to learn 

 the secrets of Nature as men who are seeking to decipher inscriptions, 

 where the language is dead and the characters are obsolete. 



It may be said, however, that the ordinary laws of inductive 

 reasoning do not apply to this particular case, because we are 

 dealing with an ' igneous complex.' The term undoubtedly is an 

 attractive one ; there is a certain mystery about it, but there is also, 

 as in all mysteries, a certain vagueness. ' Complex ' means a tangle, 

 and thus is a term which fails to explain anything ; to fall back upon 

 it, is a confession of defeat and an admission that when Nature 

 plays the Sphinx we decline to take the part of (Edipus. In the 

 case before us, however, the authors' meaning may be inferred from 

 certain conclusions which they have formulated. 2 These, briefly stated, 

 amount to this : that in the cliff and on the adjoining shore, east of 

 the Lion Bock, the serpentine is traversed by basic dykes, which pass 

 locally into hornblende-schist, and besides this put on, in the thicker 

 portions, appearances which are characteristic of the ' granulitic 

 group ' ; also that there is a wedge-shaped mass of this granulitic 

 rock, which exhibits a structure incompatible with the hypothesis 

 that the serpentine was intruded after the mass itself became 

 solid. 



In regard to these conclusions, I may venture at the outset to 

 make the old comment Dolus latet in generalibus. {a) ' The dykes 

 pass locally into hornblende-schist.' Yes, but into what horn- 



Op. cit. p. 200. 



2 Ibid. p. 210. 



