His thought is not connected. Thus he will say of the Tasmanians that they had no Neanderthaloid characters, and then remark that they represent a Neanderthaloid stage in the evolution of true man. He will tell us that one cannot go very much by the measurement of skulls because the shape of the skull sometimes changes rapidly. There he is quite right. But a few pages on he repeats all the modern fashionable nonsense about a Nordic race and a Mediterranean race and an Alpine race and the rest of it—which is based almost entirely on the idea that the skull formation is permanent. He traces religion to certain past terrors and offensive habits in a bestial type prior to Man, but won’t allow it in palæolithic man long after; then he suddenly recurs to it in neolithic man far later still. How familiar it all is! Meanwhile, throughout the whole of the work there appears that wretched _a priori_ bias which is the bane of all insufficient or false religions: the determination to squeeze facts to any shape, however unnatural, if only they may thus be forced to fit theories; the effect of a fixed mythology to which evidence must correspond, or be neglected, or so distorted as to become wholly unnatural, and under the influence of which the absence of evidence is supplied by sheer fiction. We have seen how this could lead so intelligent a man to affirm that men with capacity to fashion instruments, use fire and bury their dead with ceremony and gift, could not speak a consecutive language. It also makes him say of men who hunted the reindeer in an Arctic climate and were remarkable artists, that “perhaps” they had no clothing. It leads him to postulate as necessary an exceedingly slow and unceasing process for all development—right against known evidence—and thus bring him to deny or neglect or forget plain record, such as armament, ornament, burial.