EVOLUTION 1031 



L. serpa only. But an insulated mainland mass or the centre of a 

 large island that had given off peripheral islets, would contain only 

 L. fiumana. This sharp isolation would be facilitated by the lizard- 

 less neutral zones already referred to, and the insulation of some 

 of these would account for islands without any lizards at all. 



(c) If a land-complex inhabited by L. fiumana only gave origin 

 to islands, the same would be true of them. 



{d) But in some parts of the mainland (North Dalmatia to the 

 mouth of the Narenta) L. serpa seems to have subsequently estab- 

 lished itself, although previously separated off islands had contained 

 only L. fiumana (c) . And conversely (b), an island may contain only L. 

 serpa, though the nearest tract of mainland contains only L. fiumana. 



The details are intricate, but the first point is that the occurrence 

 of L. serpa and L. fiumana on the Dalmatian Islands can be 

 accounted for in terms of their geological history. The second point 

 is that variation is going on, and that "each island has a form (a 

 Phenotype) which is different from the forms of all the other 

 islands and mainland- areas". When the distinctive features of the 

 islet forms are not demonstrable in the young, they are seen in the 

 adults; and, conversely, when the adults are on the average almost 

 the same, the young forms are sometimes distinctive. 



But when Kammerer speaks of there being a distinctive "form" 

 on each small island, he does not mean that a specialist could from 

 an individual specimen at once tell from what Dalmatian island it 

 came. He means that the lizards on each island are in a phase of 

 saltatory variability, which requires a curve of frequency for its 

 expression. The point is that on each island there is a distinctive 

 variety with its own range of variations at present continuing. 

 Many of the varieties are not yet fixed; they are in statu nascendi; 

 but in some cases they have passed beyond endless flux towards stabi- 

 lisation. Thus there are six named and described varieties of L. serpa. 



Kammerer finds that the observed variations are what he calls, 

 not very happily, bipolar. That is to say, they seem to occur along 

 definite lines in opposite directions, plus and minus. This agrees 

 well with what we have spoken of in another section as the anabolic 

 and katabolic alternatives of variation. In any particular island the 

 variational movement is usually in one direction, but the opposite 

 movement may be seen in a different one. It may be useful to con- 

 trast these extremes in parallel columns : 



[Plus.] [Minus.] 



Total Melanism. Lightening of Colour. 



Gigantism. Nanism. 



Thick stumpy tail. Elongated thin tail. 



Very small granular scales. Enlarged scales. 



Slowness of movement. Agility. 



Tameness. Shyness. 



