286 



DR. H. GADOW ON EVOLUTION 



[Mar. 20, 



" lumpers," who, rightly allowing more amplitude of variation 

 in their conception of a species, commit the following error. They 

 think that intergrading of two species is the same as continuity 

 from one extreme to the other. For instance, if the overlap of all 

 the available characters should occur in one and the same specimen, 

 then presumably the two supposed species would be the same, 

 but not — and this is the usual procedure — if the overlap of the 

 characters occurs only in a whole number of specimens taken 

 together. 



The following diagram illustrates an important point. Let «, h, c, 

 d be 4 different character's, each of which can vary from, let us say, 

 small to large ; and let us assume that character a (for example 

 the scutellation of the forearm) is the quickest, the rrrost sus- 

 ceptible, to change. Let species A change towards B, arad let B 

 change towards A, by gradually assrrmiug the respective characters. 

 Therr it will be found that the two changing series will overlap com- 

 pletely or- coincide in all their forrr characters, only wherr all these 

 characters have arrived at a medium condition, and again wherr they 

 all have arrived at the other extreme end. The diagram shows 

 moreover' that, although the results are the same, at the terminus 

 and irr the rrriddle, the A and B series of evolution are different at 

 every stage. 



Small. 



Species A =a h c d 

 changing 



towards 15-^ a 



Medium. 



Large. 



d 



c d 

 bed 

 a h c d, 



d 

 d a 

 d a h 

 d a h c=a h c 



A lias changed into a form in 



which all the characters are 

 large; A resembling B. 



a b c d = Species B changing into, or 

 towards A. 



d 



d a 



d a b 



dab c = a b c d ' 



Diagram illustrating the overlapping of characters. 



Species B resembles A, but is not genetically the same, since the combinations 

 dab small, or a b c medium and d small, &c. occur nowhere in the series 

 which represents the changes from A towards B. 



In this paper I have employed a great rrrrmber of specific &c. names, 

 often using trinomials, in fact as many as the greatest of " splitters" ; 

 birt this has been done for the sake of convenience, for shortness 

 of expression, and having done my best to diagnose the groups, 



