THE NGINDO ARE AN EAST AFRICAN PEOPLE with a segmentary social system. Small groups of them, each numbering perhaps twenty souls in all and comprising mainly agnatic kinsfolk, go to form the residential units which I call 'cells'. Ngindo settlements, isolated in vast expanses of forest, number anything from one to a dozen such cells. Thanks to its remoteness, the settlement is transcended by few events in the Ngindo calendar. As an exception come the annual initiation ceremonies, attracting several hundred people from a radius of fifty miles around; this alone gives an indication of the importance the Ngindo accord to initiation and to its decisive element, male circumcision. In reply to the question 'What is your religion?', a number of informants from central Ngindoland, which is nominally Muslim throughout, said simply 'It is the initiation'. Nevertheless the ceremonies themselves are seldom on a super-settlement scale nowadays, and the big attendances come about through the gathering of outside kin; the individual sessions held in the various settlements are often staggered so as to permit assemblies of this sort. Such does not seem to have been the case in the past, when twenty or thirty youths might be circumcised jointly, whereas at the present day it is rare to see more than half a dozen initiates together. Of course the old-time concentration of the rites was achieved not only by inter-settlement cooperation, but also to some extent by skipping a few seasons in any one settlement until the number of candidates for initiation reached a sizeable total. To-day, on the contrary, no such restraint is exercised, and the ceremonies will take place even if no more than a single candidate is available. This particular tendency towards dispersal is consistent with a general Ngindo tendency in that direction, associated primarily with individual and cell-group status drives; that is, cell leaders, cell members collectively, or limited alliances of cells seek to acquire prestige by mounting their own independent ceremonies. Thus, even within an Ngindo settlement, two or more sets of ceremonies may occur in a given season, more especially in the case of the relatively small initial circumcision dances, less so in the case of the bigger conclusory 'release' rites. Another marked change has been the drop in age of male initiates. A century ago, so it is said, they were mostly adult men, and all certainly past puberty. Now on the contrary the initiates are mere children on whom many of the ritual lessons, and in particular the formal instruction in such matters as sex, are altogether lost. In the main the change appears to have been an economy measure, as the prosperity of Ngindoland has declined over the period. At least this is the reason commonly advanced by the natives. But it is noteworthy that Ngindo girls still undergo the initiations shortly after their first menstruation, much as they are alleged to have done in former times. The explanation may lie in the fact that parental discipline over males is evidently less rigid than it was, while new mechanisms such as schools, paid employment, and the like, have now come into existence which help define their position in society. Women on the other hand remain insulated from external influences, which in turn prompt the menfolk to impose strict control for fear of their getting out of hand. To-day the central Ngindo observe male circumcision as a universal custom, though it is open to doubt whether it has held good