II.] ON HEREDITY. ' 103 



time, for the members of the species already possessed necks 

 of varying length, and the variations which form the material 

 for natural selection were already in existence. Now all the 

 organs of every species vary in size, and any one of them will 

 undergo constant and progressive increase, as soon as it ac- 

 quires exceptional usefulness. But not only will the organ 

 fluctuate as a whole, but also the parts composing it will be- 

 come larger or smaller under given conditions, will increase or 

 diminish by the operation of natural selection. I believe that 

 qualitative variations always depend upon differences in the 

 size and number of the component parts of the whole. A skin 

 appears to be naked, when it is really covered with a number 

 of small fine hairs : if these grow larger and increase in number, 

 a thick covering is formed, and we say that the skin is woolly 

 or furry. In the same way the skin of many worms and 

 Crustacea is apparently colourless, but the microscope reveals 

 the presence of a number of beautiful pigment spots ; and not 

 until these have increased enormously does the skin appear 

 coloured to the naked eye. The presence or absence of colour 

 and its quality when present are here dependent upon the 

 quantity of the most minute particles, and on the distance at 

 which the object in question is observed. Again, the first 

 appearance of colour, or the change from a green to a yellow 

 or red colour, depends upon slight variations in the position 

 or in the number of the oxygen atoms which enter into the 

 chemical combination in question. Fluctuations in the chemical 

 composition of the molecules of a unicellular organism (for 

 example) must continually arise, just as fluctuations are always 

 occurring in the number of pigment granules in a certain cell, 

 or in the number of pigment cells in a certain region of 

 the body, or even in the size of the various parts of the 

 body. 



All these quantitative relations are exposed to individual 

 fluctuations in every species ; and natural selection can 

 strengthen the fluctuations of any part, and thus cause it to 

 develope further in any given direction. 



From this point of view, it becomes less astonishing and less 

 inconceivable that organisms adapt themselves — as we see that 

 they obviously do — in all their parts to any condition of ex- 

 istence, and that they behave like a plastic mass which can be 



