MUCH recent work on the systematics of the Notostraca (Crustacea, Phyllopoda) has included comment on the great range of the variation in each systematic character in comparison with other animal groups, and the consequently unsatisfactory nature of a classification which must be based upon such characters; Linder (1952) has reviewed a number of these comments and has analysed the extent of the variation in many of the important characters. Data derived from the culture of a pure line of hermaphrodite Triops cancriformis has been presented by Longhurst (1955a) and it was thereby demonstrated that much of the recorded variation can be of little significance systematically or, probably, adaptively. In a percentage of specimens variations occur which fall far outside the statistical normal distribution of the character for the remainder of the sample. For instance, in a sample of 12 specimens of Triops cancriformis simplex from Kebili in Tunisia (MNHNP)1 there is one individual in which the sulcus spines are completely absent; in the remainder, though these spines are rather small, they are quite normal in number for the subspecies, having a mean of 33.8 ? 3.9. In the same category are the cases of abnormal body segmentation recorded by Barnard (1929) and Linder (op. cit.) where the normal sequence of exoskeletal annuli is disturbed by the formation of incomplete rings or by spiral growth. These abnor-