Determination of age at the time of death based on the observation of cranial sutures has led to numerous studies with sometimes contradictory results. The initial hypothesis being that suture closure is part of an age-related physiological process, the conflicting results have been interpreted by various authors as secondary to the choice of sutures, under the co-existing influence of pathological factors or genetic factors, or even independent of age. Despite these differences, macroscopic methods remain much used in anthropology and in forensic medicine. In our work, we evaluated the value of the degree of closure of the frontosphenoidal suture in estimating age at death of mature subjects, with the secondary objective of establishing a linear regression which could be used in routine practice. The study concerned bone specimens from individuals whose age, sex and medical history were known. Macroscopic observation was carried out on the ectocranial and endocranial sides according to four stages of closure previously defined. 290 sutures were taken from a population of whom two-thirds were men. The method can be repeated and reproduced and the regression established shows the confidence range for average error to be 1.5 years. While this result is of interest in terms of precision in prediction for a group of people, the prediction range is too great ( 23 years) to be applicable to a single individual as part of a forensic procedure.