It has been shown that some of these reductions—like the retention of outer incisors—are hereditary, and thus liable to perpetuate themselves. To a certain extent this explains the observation that certain variations occur with greater frequency among primitive tribes than in civilized man. Most primitive tribes are very small in numbers, or for long periods, during which they increased in numbers, have had little intercourse with foreign people. If in such a group any of the original families showed a certain peculiarity, it must now be found more frequently than in any other tribes. A case of this kind is the frequency of supernumerary vertebræ among the Indians of Vancouver Island, and probably also the frequency of the _torus palatinus_ among the Lapps. It may be left an open question, whether the frequent occurrence of the _os Incæ_ among the Pueblo Indians and Peruvians may be explained by the same consideration. Therefore it may be that the greater variability of certain races, in regard to these phenomena, is not an expression of a lower or higher degree of development of the whole group, as the case may be, but of the presence of a great number of members of a family which possessed the peculiar character. We do not deal in these cases with spontaneous variations, but with their hereditary re-appearance. In other words, if we are to admit the conclusion that greater variability means a lower or higher stage of development, it will be necessary first to prove that these variations appear spontaneously in any member of the group, and do not belong to certain families in which the feature is hereditary. Otherwise it will be necessary to prove that in larger groups of mankind the families exhibiting the particular anomaly had a greater probability of surviving than others. However this may be, the occurrence of these variations shows that man cannot be assumed to have a stable form. It is of course an entirely open question, how long a time may be required to fix any one of the variations that we are here discussing. The general tendency of anthropological inquiry has been to assume the permanence of the anatomical characteristics of the present races, beginning with the European races of the early neolithic times. Kollmann, the most pronounced advocate of this theory, claims that the oldest remains of man found in the neolithic deposits of Europe represent types which are still found unchanged among the modern civilized population of the continent. He has tried to identify all the varieties found in the neolithic prehistoric population with those living at the present time. All studies of the distribution of head-forms and of other anthropometric traits have shown uniformity over considerable continuous areas and through long periods; and the natural inference has been that heredity controls anthropometric forms, and that these are therefore stable (Deniker).