Handwriting is one of the pinnacles of the progressive achievements of civilization; the translation of human thoughts into graphic form, calling as this does upon many of the complex functions of both body and brain. For various reasons, a number of attempts have been made to design systems for identifying the handwriting of a particular person. In general these systems have been based on the principles applied to the natural sciences of ‘classification’ and of ‘measurement’. However, with a subject so intrinsically variable as that of handwriting and, by definition, so essentially consistent and reproducible as those of classification and measurement, these systems have failed in universal acceptance. This paper surveys, briefly, some of the major systems which have been designed, examines the basic principles employed, criticizes the methodology, and demonstrates reasons for the inevitable and predictable failure for the universal acceptance of such systems; at least in those exercises associated with the identification of persons in legal proceedings.