( 453 ) 



similar cases. This inductive method of reasoning may often lead to wrong 

 inferences, as the correctness of the latter depends first ou the premiss that there 

 are cases which are really proved to be trne, and secondly that the cases which we 

 believe to belong to the same category as those proved are really similar cases. 



If we apply this to our question as to specific or non-specific distinctness, it is 

 evident that the conclusion of the diagnosticist can be correct only under the 

 condition that the specific or non-specific distinctness of some forms is proved by 

 experiment, and that he is so well acquainted with the morphology of the forms 

 in question that he can with great probability of correctness decide whether the 

 required similarity is actual or superficial. If in a given group of forms the 

 specific distinctness of any form is not proved, we have to resort to a proved case 

 in an allied group of forms ; of course, the more dissimilar the forms referred to 

 for comparison are, the more it becomes probable that our inference is not correct. 

 In most cases it is, therefore, circumstantial evidence we have to judge from, and, 

 as many an innocent man has been condemned by a competent judge on the ground 

 of circumstantial evidence, we cannot very well expect to be always right in our 

 judgment of the specific value of the difl'erences of forms. Thongli the special 

 evidence furnished by morphology and biology is to be carefully considered in every 

 single case, there are nevertheless some general arguments which apply to a 

 multitude of cases. The question as to specific identity or non-identity concerns first 

 forms which occupy the same area, or whose areas overlap, and secondly forms 

 which inhabit localities separated from one another by districts that are not in- 

 habited, or not inhabitable, by them. 



We have above tried to show that a species can develop into more species only 

 with the help of isolation of the varietal forms. If, therefore, two allied s])ecies are 

 found to inhabit the same district, no matter whether the areas are totally or only 

 partly the same, it is obvious that at a former period, when the species in question 

 were not yet so far advanced in divergent development, they must have occupied 

 separate areas. From the fact of cohabitation (in a wide sense) the farther inference 

 must necessarily be drawn that the possibility of cohabitation without fusion is due 

 to the forms having become so divergent that they are indift'erent to one another. 

 The time which has elapsed since the two forms now living together became 

 specifically different must therefore be much greater than that elapsed since the 

 formation of the geographical representatives of those two species. If comparative 

 anatomy and morphology are of any value as to the judgment of the phylogeuy of 

 species, the morphological difl'erences between a species and an ally which branched 

 off at an early period must be greater than the differences between the same species 

 and its younger geographical representative species, and still greater than the 

 differences between the geographical forms of the species. If in a given case we 

 have to decide whether A and B, which live together, are two different species, or 

 two forms of one species, the morphological characters of A compared with those 

 of .Band the geographical representatives of i? will have to guide us in our judgment. 

 There are three possibilities resulting from the comparison. First, the morphological 

 differences between A and B are greater than those between B and its representa- 

 tives ; in this case A and B must be considered sjiecifically distinct, until experiment 

 proves the reverse. Secondly, the morphological differences between A and B are 

 less great than those between B and any one of its representatives ; in this case 

 xl and B are specifically identical. Thirdly, the differences between A and B are 

 e(|ual in morphological value to those between B and any one of its representatives ; 



