The inadequacy of our ideas of kinship as applied to lower societies has been often felt by those ethnographers who wished to enter deeper into the problems of kinship among a given people. They have found the greatest difficulty in conveying to a European reader the meaning of different terms of relationship. While warning the reader to put aside our (the modern European) ideas of kinship, they have hardly succeeded in giving any definite and clear concept instead. The reasons for this failure are simple: our ideas of kinship are defined by certain facts which are not to be found in the given primitive society. In order to define kinship so as to fit the latter, the author ought to bring forward a series of facts, playing a part analogous in the given society to that played by the essential defining elements amongst us. But it is by no means easy to know among which facts to look for such analogous, defining elements. And here again arises the necessity of a general definition of kinship, one which would afford indications in what direction to search for social facts giving a right idea of kinship in any given society. Such a general definition would be like an algebraic formula, having its constant and its variable terms; if for the latter special data be inserted (in this instance the special conditions proper to the given society), the special value for any given case is obtained (namely the special concept of kinship proper to the given society). And it should also be indicated within what range the variables should be taken; in other words, in what facts the elements which specifically determine kinship in the given society must be looked for. The practical value of such a general definition of kinship is obvious. On the one hand it indicates the constant elements in kinship common to all societies; on the other hand it indicates the general character of the variable elements, and the way in which they must be looked for and worked into the general formula.