BBITISH ASSOCIATION : ZOOLOGICAL SECTION. 399 



geneous lot change orfchogenetically (this term has been translated 

 into the far less expressive " rectigrade ") in one direction, and if 

 there be no lagging behind, they all reach precisely the same end. 

 This would be a case of transmutation (true mutations in Waagen's 

 and Scott's sense), producing new species without thereby increasing 

 their number, whilst divergence always implies, at least potentially, 

 increase of species, genera, families, &c. 



If for argument's sake the mutations pass through the colours of 

 the spectrum, and if each colour be deemed sufficient to designate a 

 species, then, if all the tenth generations have changed from green to 

 yellow and those of the twentieth generation from yellow to red, the 

 final number of species would be the same. And even if some lagged 

 behind, or remained stationary, these epistatic species (Eimer) are 

 produced by a process which is not the same as that of divergence or 

 variation in the usual sense. 



The two primary factors of evolution are Environment and 

 Heredity. Environment is absolutely inseparable from any existing 

 organism, which therefore must react (Adaptation) and at least 

 some of these results gain enough momentum to be carried into the 

 next generation (Heredity). 



The life of an organism, with all its experiments and doings, is its 

 Ontogeny, which may therefore be called the subject of Evolution, 

 but not a factor. Nor is Selection a primary and necessary factor, 

 since, being destructive, it invents nothing. It accounts, for instance, 

 for the composition of the present fauna, but has not made its 

 components. A subtle scholastic insinuation lurks in the plain 

 statement that by ruthless elimination a black flock of pigeons can 

 be produced, even that thereby the individuals have been made 

 black. (But of course the breeder has thereby not invented the 

 black pigment.) 



There can be no evolution, progress, without response to stimulus, 

 be this environmental or constitutional, i. e. depending upon the 

 composition and the correlated working of the various parts within 

 the organism. Natural selection has but to favour this plasticity, by 

 cutting out the non-yielding material, and through inheritance the 

 adaptive material will be brought to such a state of plasticity that it 

 is ready to yield to the spur of the moment, and the foundation of 

 the same new organs will thereby be laid, whenever the same 

 necessity calls for them. Here is a dilemma. On the one hand the 

 organism benefits from the ancestral experience, on the other there 

 applies to it de Rosa's law of the reduction of variability, which 

 narrows the chances of change into fewer directions. But in these 

 few the changes will proceed all the quicker and farther. Thus 

 progress is assured, even Hypertely, which may be rendered by 

 " over-doing a good thing." 



Progress really proceeds by mutations, spoken of before, ortho- 

 genesis, and it would take place without selection and without 

 necessarily benefiting the organism. It would be mere presumption 

 that the seven-gilled shark is worse off than its six- or five-gilled 

 relations ; or to imagine that the newt with double trunk- veins suffers 



