476 Proceedings of the Royal Physical Society. 



this living relic is the least specialised form of the whole 

 group, and perhaps may owe its persistence to this cause. 



The intensity of specialisation attained to by those groups 

 of Mesozoic reptiles that underwent extinction will be 

 only too familiar to all palaeontologists. The Ichthyosaurs, 

 Plesiosaurs, Dinosaurs, and Pterosaurs seem literally to have 

 taken possession of the land, air, and water, and underwent 

 a marvellous degree of specialisation along many separate 

 lines. The same remark applies to the Anomodonts, which, 

 although so generalised in their affinities, were really forms 

 highly specialised for several entirely different methods of 

 life, as is seen in the Pariasaurs, the Theriodonts, and the 

 Dicynodonts. 



The extinct Pythonomorpha, which had a world-wide dis- 

 tribution in the Cretaceous, and did not survive that period, 

 represent, according to Dr Woodward, an early stage in the 

 evolution of the Squamata. Early forms of the latter, how- 

 ever, appear to have already been coexistent with these 

 mighty reptiles, and one would rather look on them as a 

 highly specialised branch of the early squamate stock. 



If close in-breeding amongst living forms tends to produce 

 sterility and also a weakness in relation to the surroundings, 

 it is possible that the same cause which is at the bottom of 

 the ill effects brought about by in-breeding is also the cause 

 which produces the weakness which appears to be the result 

 of long-continued specialisation within narrow limits. 



In the first place, then, what is this cause, and the only 

 answer is that we don't know. It is evident that if we 

 wish to get further developments along a certain line of 

 specialisation, the method is to breed true, or to pick out 

 forms to breed from that do not deviate from the type of 

 specialisation. It is also evident that what is termed true 

 breeding has effects in horses and dogs and men that 

 gradually merge into those that are the result of in-breeding. 

 In the breeding of cattle and sheep, the very constant choice 

 of rams and bulls of high reproductive power would tend at 

 least to prevent the occurrence of sterility. 



Now we suppose that all life has originally sprung through 

 a one-celled condition, or something which we might com- 



