It is not easy to understand the history and origin of the different peoples of today's world inasmuch as scientific data are partial and seemingly contradictory. These roughly fall into three categories:-prehistoric data are remains of cultures and human skeletons. They allow us to affirm that such and such a region was inhabited in such and such an epoch. Their absence, however, means nothing, and they hardly permit the attribution of a biological origin to the peoples of the past because (...) of the rapid evolution of the forms and dimensions of the skeletons during the course of time;-present genetic data give an interesting picture of the biological relationship of peoples and of the way they diverged throughout history. Their interpretation is delicate, however, especially when it concerns small populations;-“biometric” data involve the measurement of pigmentation, dimensions and proportions of the body and organs. They are susceptible to very rapid variations during the course of history and mainly give an indication of the adaptation of the people to a certain kind of environment: tropical, temperate, cold or arctic, plain or altitude, forest, savannah, prairie, steppe or desert. (shrink)

A similarity index allowing comparisons of human populations has been defined as the common “Percentage of Isoactive Genes” or PIG, which can be calculated from any gene frequency distribution characterizing two populations. The complement to one of this value has been proved to be a distance, a measure which can be used in most techniques of cluster analysis as well as in usual representations of multivariated data (dendrograms, etc...). Furthermore, the formula can be generalized to a set of populations. From (...) a biological point of vue, PIG values are particularly meaningful, whereas other previously defined indices which tend to measure similarity or difference between populations neither have such an advantage, nor can they be expressed by a single and clear number like a percentage. Moreover,comparisons using PIG indices depend at first on the fluctuations of the most frequent genes of a distribution; on the other hand, different measures principally reflect the variations of low frequency genes, which unfortunately are nearly always poorly estimated, due to weak samples of populations.Interestingly, PIG values can be applied to any frequency distribution of any kind of objects. They can also extend to a whole set of distributions by using a mean value. When applying these measures to a series of polymorphic genetical systems such as ABO, Rhesus, MNSs, Gm and HLA, one can account for the variability or the homogeneity particular to the different human groups, with simple and objective criteria. (shrink)