occurs that the most important part of the recapitulative phases are absent from the developmental history of an animal. The egg proceeds very rapidly to produce the adult form, and all the wonderful series of changes showing the animal's ancestry are absolutely and completely omitted; that is to say, all those stages which are of importance for our present purpose. Just as certain bodies pass from the solid to the liquid state at a bound, omitting all intermediate phases of consistence, but giving evidence of "internal work" by the suggestive phenomenon of latent heat-so do these embryos skip long tracts in the historically continuous phases of form, and present to us only the intangible correlative "internal work" in place of the tangible series of embryonic changes of shape.

Now I want to put this case-a supposition-before the reader who has so far followed me in these pages. Suppose, as might well have happened, that the Barnacles, one and all, instead of recapitulating in their early life, were to develop directly from the egg to the adult form, as so many animals do; should we have ever made out that they were degenerate Crustaceans? Possibly we should: their adult structure still bears important marks of affinities with crabs and shrimps; but as a matter of fact before their recapitulative development had been discovered they were classed by the great Cuvier and other naturalists with the Molluscs, the mussels and snails.

Suppose again that all the existing Ascidians, as many of them actually have, had long ago lost their recapitulative history in growth from the egg: suppose

that no such a thing as an Ascidian tadpole existed, but that the Ascidian's egg grew as directly as possible into an Ascidian, in every living species of the group. This might easily be the case. Then most assuredly

we should not have the least notion that the Ascidians were degenerate Vertebrates. We should still class them where they used to be classed before the Russian naturalist Kowalewsky discovered the true history and structure of the Ascidian tadpole. I believe that I shall have the assent of every naturalist when I say that the vertebrate character of the Ascidians and the history of their degeneration would never have been suspected, or even dreamed of, had the Ascidian tadpoles ceased to appear in the course of the Ascidian development at a geological period anterior to the present epoch.

This being the case, it must be admitted that it is quite possible—I do not say more than possible—that other groups of animals besides parasites, Barnacles, and Ascidians, are degenerate. It is quite possible that animals with considerable complexity of structure, at least as complex as the Ascidians, may have been produced by degeneration from still more highly organised ancestors. Any group of animals to which we can turn may possibly be the result of degeneration, and yet offer no evidence of that degeneration in its growth from the egg.

Accordingly, wherever we can note that a group of organisms is characterised by habits likely to lead to degeneration, such as I have enumerated, viz. parasitism or immobility, or certain special modes of

nutrition, or again, by minute size of its representatives -there we are justified in applying the hypothesis of degeneration, even in the absence of any confirmatory evidence from embryology. When we so apply this hypothesis we find in not a few cases, in working over the details of the organisation of many different animals by the light which it affords—that much becomes clear and assignable to cause which, on the hypothesis either of "balance" or of "elaboration," is quite hopelessly obscure. As examples of groups of animals which can thus be satisfactorily explained I may cite first of all the Sponges as only somewhat less degenerate, we have all the Polyps and Coral-animals, also the Starfishes. Amongst the Mollusca-the group of headless bivalves, the oysters, mussels and clams, known as the Lamellibranchs, are, when one once looks at their structure in this light, clearly enough explained as degenerated from a higher type of head-bearing active creatures like the Cuttle-fish; whilst the Polyzoa or Moss-polyps stand in precisely the same kind of relation to the higher Mollusca1 as do the Ascidians to the higher Vertebrates: they have greatly degenerated, and become minute encrusting organisms which, like some of the Ascidians, build up colonies by plant-like budding growth. The Rotifers, or wheel animalcules, I have already mentioned as best explained by the supposition that they are the descendants of far larger and more fully-organised animals provided with loco

1 Not to the higher Mollusca in all probability, but to some higher worm-like form. The ancestry of the Polyzoa is still a profound mystery! December, 1889.

motive appendages or limbs: they have dwindled and degenerated to their present minute size and curiously suggestive structure.

Besides these there are other very numerous cases of animal structure which can best be explained by the hypothesis of degeneration. A discussion of these, and a due exposition of the application of the hypothesis of degeneration to the various groups just cited, would involve a complete treatise on comparative anatomy and embryology, and lead far beyond the limitations of this little volume.

All that has been, thus far, here said on the subject of Degeneration is so much zoological specialism, and may appear but a narrow restriction of the discussion to those who are not zoologists. Though we may establish the hypothesis most satisfactorily by the study of animal organisation and development, it is abundantly clear that degenerative evolution is by no means limited in its application to the field of zoology. It clearly offers an explanation of many vegetable phenomena, and is already admitted by botanists as the explanation of the curious facts connected with the reproductive process in the higher plants. As a further example of its application in this field, the yeast-plant may be adduced, which is in all probability a degenerate floating form derived from a species of mould. In other fields, wherever in fact the great principle of evolution has been recognised, degeneration plays an important part. In tracing the development of languages, philologists have long made use of the hypothesis of

degeneration. Under certain conditions, in the mouths and minds of this or that branch of a race, a highly elaborate language has sometimes degenerated and become no longer fit to express complex or subtle conceptions, but only such as are simpler and more obvious. (See Note D.)

The traditional history of mankind furnishes us with notable examples of degeneration. High states of civilisation have decayed and given place to low and degenerate states. At one time it was a favourite doctrine that the savage races of mankind were degenerate descendants of the higher and civilised races. This general and sweeping application of the doctrine of degeneration has been proved to be erroneous by careful study of the habits, arts, and beliefs of savages; at the same time there is no doubt that many savage races as we at present see them are actually degenerate and are descended from ancestors possessed of a relatively elaborate civilisation. As such we may cite some of the Indians of Central America, the modern Egyptians, and even the heirs of the great oriental monarchies of præ-Christian times. Whilst the hypothesis of universal degeneration as an explanation of savage races has been justly discarded, it yet appears that degeneration has a very large share in the explanation of the condition of the most barbarous races, such as the Fuegians, the Bushmen, and even the Australians. They exhibit evidence of being descended from ancestors more cultivated than themselves.

With regard to ourselves, the white races of