conception of development, but yet identical with those which formerly found expression in the theories of preformation and epigenesis.

By the term "embryonic development," in its ordinary acceptation, we understand the appearance of visible complexity. But when we speak of the visibility of the resulting complexity, we use a subjective term, the value of which is relative to the human eye. Going further into the matter, we must break up the conception into two parts, and distinguish between the actual production of complexity and the mere transformation of complexity from a condition invisible to us into complexity visible to our senses.

'The two kinds of development I have indicated bear a relation to each other that recalls the old opposing doctrines of preformation and epigenesis, the alternatives of a time when it was a task—perhaps the only possible task—to record the completed results of the stages in development as they became complete—in fact, to record the externally visible changes of shape. In this descriptive investigation of the development of external form, epigenesis, the successive formation of new shapes, gained a complete victory over evolution, the mere becoming visible of pre-existing details of shape.

'The closer investigation of embryonic development that is necessary in a search for causes brings us once more against the old alternatives, and compels us to a closer scrutiny of them.

'In this, if we still retain the old terms, epigenesis would mean not merely the building up of complicated form through the agency of a substratum, apparently simple, but perhaps with an extraordinarily complicated, minute structure, but, in the strictest sense of the term, the new formation of complexity, an actual increase of complexity. Evolution, on the other hand, would imply the mere becoming visible of pre-existing latent differentiation. Clearly, according to these general definitions, occurrences which outwardly exhibit epigenesis may be in reality partial or complete evolution. In fact, the deepest consideration leads us again to the original question: Is embryonic development epigenesis or evolution? Is it the new formation of complexity, or is it the becoming visible of complexity previously invisible to us?'

Thus, in our own days, after the controversy has been at rest for long, biologists are assembled in opposing groups, one under the standard of epigenesis, another under that of preformation.

Weismann[2] leads the van for preformation; for the last ten years he has occupied himself with the theoretical discussion of the questions set forth above; and now, in a recent treatise, The Germplasm, he has combined his views, already many times modified, in a coherent theory. Now he explains candidly that he has been driven to the view that epigenetic development does not exist. 'In the first chapter of my book,' he remarks, 'will be found an actual proof of the reality of evolution, a proof so simple and obvious that I can scarcely understand to-day how it could have escaped my notice so long' (Germplasm, p. 14). Elsewhere he writes: 'I believe that I have established that ontogeny can be explained only by evolution, and not by epigenesis.'

A mental process, which consciously or unconsciously plays a great part with evolutionists, and helps to determine their conclusions, is characteristic of the direction of their inquiries. They set out from the fact that the characters of the parents, often to the smallest detail, are transmitted to children by means of the germ or rudiment; they conclude that the active causes of all the complexity that arises must be contained in the apparently homogeneous germ, embryological differentiation being a spontaneous process. It follows that the apparent homogeneity is, in reality, latent complexity which becomes patent during the progress of ontogeny. Latent complexity implies a material substratum, consisting of actual particles for which many different names have been found. As our senses can give us no experimental knowledge of these particles, which are so small as to be invisible, modern evolutionists attempt to picture them, in imagination, by reflecting all the visible characters of the perfected organism upon the undivided egg-cell, so peopling that globule of yolk with a system of minute particles corresponding in quality and in spacial arrangement with the larger parts of the adult.

Weismann has practised this art in the true spirit of a virtuoso, and has elaborated it into a novel mode of biological investigation. Take an example;—'It would be impossible,' he says in The Germplasm (p. 138), 'for any small portion of the human skin to undergo a hereditary and independent change from the germ onwards, unless a small vital element corresponding to this particular part of the skin existed in the germ substance, a variation in this element causing a corresponding variation in the part concerned. Were this not the case, birth-marks would not exist.'

Thus, in a slightly altered fashion, we come again to the position of the evolutionists of last century, for whom the germ was an extremely small miniature of the adult creature. The new evolution, as Weismann in especial has established it, seems to me to differ from the old doctrine only in two important points; and these must be placed to the credit of the greater scientific knowledge of our century. The first point concerns the relative positions of the parts in the patent and latent conditions. The older evolutionists assumed that these were identical, that the germ was a true miniature. It is true that Weismann regards his almost countless germinal particles as being held together in an architectural structure of almost inconceivable complexity. For him the germ is an exceedingly complicated living being, a microcosm in the truest sense, in which every independently variable part that ever appears throughout the whole life is represented by a living particle, and in which each of the living particles is endowed with a definite, inherited position, a constitution, and the power of rapid multiplication. It is upon the qualities of these ultimate particles that he makes depend the qualities of the corresponding parts of the adult, the parts that are cells as well as the parts built of many cells. As, however, during visible development the parts of the embryo undergo many changes of position and metamorphoses, Weismann is compelled to make the assumption that the germ, as a micro-organism, is not simply a miniature of the adult, but that its minute particles have an arrangement totally different from that of the corresponding parts in the adult organism.

The second point is the origin of each new generation. To explain the continuity of development, the old evolutionists held that the generations lay enfolded one within another. Weismann avoids this difficulty by endowing his germs with divisibility, but he gives us no proof that division could possibly take place in the case of structures composed of innumerable particles built up into a definite and most complicated architectural system.

Although the new evolution differs from the old in the points mentioned above, the two theories obviously agree very closely in the nature of their arguments and conclusions. When, to satisfy our craving for causality, biologists transform the visible complexity of the adult organism into a latent complexity of the germ, and try to express this by imaginary tokens, by minute and complicated particles cohering into a system, they are making a phantasmal image which, indeed,