IT IS COMMON for students in different branches of science to exert themselves in the effort to obtain new and positive results from their researches; and misconceptions often arise precisely because of the deep-rooted idea that they ought to be 'proving', or giving a definite status to, something, whereas in reality they can only state what they have found and not what they may have set out to find. To have reached a 'proof,' or to have arrived at some clearcut conclusion, is far more gratifying to an all-too-human sense of the fitness of things than the simple but factual statement that one still does not know with any reasonable degree of certainty the answers to the problems one has investigated and that one's findings must therefore be entered to a suspense account. In this paper, then, I have chosen to err on the side of caution and make no attempt to 'prove' or 'disprove' particular points relating to its subject-matter. Indeed, in the present state of knowledge, it is evident that hardly anything definite can be said about the secondary sexual characters of fossil hominid forms that do not belong to the species Homo sapiens. At the present time I am engaged in an investigation of sexual differences in the innaominate bone (os coxae), having at my disposal a very substantial number of specimens - over six hundred - from the parish church of St Bride, Fleet Street, London ECI, for the preliminary work. Since the pelvis furnishes unquestionably the most reliable guide to the sex of a skeleton, some first-hand acquaintance with the diagnostic features of the innominate is a good starting-point for any study in which the sexing of individuals represented only by skeletal remains is called for; and the value of the conclusions reached must be greatly lessened if the assignment of sex to specimens in any given group is not reasonably accurate. This is sometimes true even when one is sexing unidentified modern material but is nevertheless able to verify one's results, to some extent, by reference to a control series of comparable data of known age and sex, as is the case in the St Bride's skeletons now being studied at Cambridge. In view, therefore, of the difficulties with which the anthropologist may be faced in almost ideal situations such as the one just mentioned, I have often felt that we have perhaps been over-hasty in assigning a sex to palaeanthropic remains from scattered localities; 1 above all because the diversity of characters they *The present paper, which is part of a programme of research in human palaeontology suggested by Dr P. Martinez del Rio, Dr Juan Comas, and Dr M. Maldonado Koerdell, has been written during the tenure of a British Council scholarship in the United Kingdom. Some of the fossil material discussed in it was examined in the course of a brief visit to Brussels and Paris in 1954, made possible by financial assistance from Corpus Christi College, Cambridge, and from my father