While it is thus conceivable that all organisms may resemble each other, and all differ, owing to the similarities and diversities in the “conditions of existence” (and among those conditions that of descent is of wide range), it is not very readily conceivable how advantage in the external struggle could have determined the varieties of form and function, because many differentiations give no superiority in the struggle. As Mr. St. George Mivart urges, “Natural Selection utterly fails to account for the conservation and development of the minute and rudimentary beginnings, the slight and infinitesimal commencements of structures, however useful those structures may afterwards become.”[78] And this is undeniable on the supposition that Natural Selection is an agency not identical with the variations of growth, but exclusively confined to the accumulation of favorable variations. 141. In estimating the two hypotheses--First, of Descent from one primordial germ, and the modifications _due_ to Natural Selection, or, as I should say, _expressed_ in Selection; and Secondly, of Descent from innumerable germs having initial differences, which differences radiated into the marked modifications, there is this superiority to be claimed for the first, that it is more easily handled as an aid to research, and is therefore more decidedly useful. The laws of Organic Affinity are at present too obscure for any successful application. I only wish to point out that the theory of Descent is an imaginary construction of what _may have been_ the process of species-formation, not a transcription of the process observed. It constructs an imaginary Type as progenitor of a long line of widely different descendants. The annelid which is taken as the ancestor of the vertebrates is not any annelid known either to zoölogists or geologists, but a generalized and imaginary type. So daringly liberal is the imagination in endowing the ancestor with whatever may be required for the descendants, that Mr. Darwin thinks it _probable_, from what we know of the embryos of vertebrates, that these animals “are the modified descendants of some ancient progenitor which was furnished in its adult state with branchiæ, a swim-bladder, four simple limbs, and a long tail, all fitted for an organic life,” (p. 533); and Dr. Dohrn conceives the original type to have contained within itself all that has been subsequently evolved in the highest vertebrate, the other and less elaborate organisms being mere degradations from this type.[79] This use of the imagination, although not without advantages, is also not without dangers. It may direct research, it must not be suffered to replace research. PROBLEM II. THE NERVOUS MECHANISM. “All the functions of the nervous system are as dependent upon its structure and nature, as the accurate indication of time upon the construction of the chronometer.”--PROCHASKA.