THE EMBRYOLOGY AND ETIOLOGY OF TUMOURS 73
somites may be shunted, actually or but physiologically, from the normal connection is purely hypothetical, and nothing of the kind has ever been witnessed. Rudimentary somites occur even in the trunk region in some animals, but these are rudimentary, and probably always disappear.
Wilms regards his germs as things destined in reality to form parts of “the embryo,” and therefore as belonging to this.* Under his views cases of five embryomata in one ovary require the shunting into this of five blastomeres during the early development; that is to say, in this instance the original fertilized egg must have been divided up in some way or other into at least six portions, one of which formed a normal embryo, while the remaining five retained at least the potentialities of each becoming an abnormal embryo or embryoma. It is open to doubt whether any upholder of epigenesis will admit the possibility of the course of events happening in this way. As it would seem, a new hypothesis is needed to account for each of the five embryomata, with an additional one to explain the continued normal character of the development after such a shaking and shunting.
Equally formidable difficulties are furnished by the well-known instances of multiple tumours, of various kinds, in one individual. Indeed, the doctrine of epigenesis as the mode of the development labours under quite sufficient insuperable intrinsic difficulties without having to bear the burdens imposed upon it by such
* As decisive against the origin of tumours from cells, or tissues, of the individual in which they develop, may be cited the facts that very many of them are encapsulated from the surroundings—thus, tumours of the kidneys, breast, and parotid; and that various observers—thus, Wilms and Borst—deny any passage or transition of normal tissues into them. The encapsulation of many tumours is of embryological interest, because many of the aberrant germ-cells exhibit this feature.