If the horizon levels are not too finely graded, the difficulty of accepting such a time-scale is not excessive. Hence we might be prepared to accept the chronometric value of the division of fossiliferous rocks into Groups, such as the Palæozoic, the Mesozoic, and the Cænozoic, even though we are assured by Grabau that this time-scale is “based on the changes of life, with the result that fossils alone determine whether a formation belongs to one or the other of these great divisions” (“Principles of Stratigraphy,” p. 1103), but when it comes to projecting an elaborate scheme of levels or horizons into Pleistocene deposits on the dubious basis of index fossils and “industries,” our credulity is not equal to the demands that are made upon it. And this is particularly true with reference to fossil men. Man has the geologically unfortunate habit of _burying his dead_. Other fossils have been entombed on the spot where they died, and therefore belong where we find them. But it is otherwise with man. In Hilo, Hawaii, the writer heard of a Kanaka, who was buried to a depth of 80 feet, having stipulated this sort of burial through a special disposition in his will. His purpose, in so doing, was to preclude the possibility of his bones ever being disturbed by a plough or other instrument. Nor have we any right to assume that indications of burial will always be present in a case of this nature. We may, on the contrary, assume it as a general rule that human remains are always more recent than the formations in which they are found.