But there is the reply that the places occupied by this cartilage being places at which the bone lengthens, the non-ossification is in part apparent only--it is rather that new cartilage is formed as fast as the pre-existing cartilage ossifies; and there is the further reply that the slowness of the ultimate ossification of this part, is due to its non-vascularity, and to mechanical conditions which are unfavourable to its acquirement of vascularity. Once more, there is the demurrer that in the epiphyses ossification does not begin at the surface but within the mass of the cartilage. Explanation of this implies ability to follow out the mechanical actions in a resilient substance which, like india-rubber, admits of being distorted in all ways by pressure and recovering its form, and it seems impossible to say how the more superficial and more deep-seated canals traversing it will be respectively affected. Of course it is not meant that this osseous development by direct equilibration takes place in the individual. Though it is a corollary from the argument that in each individual the process must be furthered and modified by the particular actions to which the particular bones are exposed; yet the leading traits of structure assumed by the bones are assumed in conformity with the inherited type. This, however, is no difficulty. The type itself is to be regarded as the accumulated result of such modifications, transmitted and increased from generation to generation. The actions above described as taking place in the bone of an individual, must be understood as producing their total effect little by little in the corresponding bones of a long series of individuals. Even if but a small modification can be so wrought in the individual, yet if such modification, or a part of it, is inheritable, we may readily understand how, in the course of geologic epochs, the observed structures may arise in the assigned way.