similarity or uniformity of type; in other cases the diversity of types occurring in the same community is quite remarkable. We call a series the more variable, the more frequently deviating types occur in it; so that the average amount of differences between the individuals constituting the series and the most common type may be used as a measure of the variability of the series. These considerations are of prime importance in all attempts to compare different races. In some cases differences are found which are sufficiently fundamental to distinguish easily and definitely one from another. Thus the color of the skin, color and form of the hair, and configuration of lips and nose, distinguish the African negro definitely from the North European. When, however, we compare all the races and types of man, we find that innumerable transitions exist, which would make it difficult to state that any one particular feature belongs to all the individuals of one type, to the exclusion of all others. Thus it would not be difficult to find among members of the American race, for instance, lips and nose which approach in form those of the negro. The same may be said of color. This indefiniteness of distinctions between different types is due to the variability of the types, which has been described before, and to the comparatively small differences between the types. To give an instance. Negroes have thick lips. Nevertheless the thickness is not the same among all of them. In some cases it is quite small, in others very large. Europeans have thin lips, but we can find individuals whose lips have very considerable thickness. Thus it happens that there are some negroes whose lips deviate from the normal type in being unusually thin, and whose lips are therefore similar to those of Europeans whose lips are unusually thick. The less distinct two types are, the greater will be the number of individuals in both groups that are alike. It follows also, from what has been said, that the greater the variability of each type, the greater will be the probability that some individuals of the two types compared will be alike. We may perhaps best express this by saying that the varieties constituting each race overlap. In many cases, and in some of those that are most important for our inquiry, this overlapping is extended. Thus I have pointed out the differences in average brain-weight between different races. Brain-weights are, however, so variable, that a considerable overlapping occurs, and that even the average sizes of the brains of the white race are numerously represented among other races. Medium-sized brains of whites may be represented by the group of individuals having skull capacities of from 1450 cc. to 1650 cc. This group embraces 55 per cent of the Europeans, 58 per cent of the African negroes, and 58 per cent of the Melanesians. The same result appears when we compare the number of individuals having great cranial capacities. We find that 50 per cent of