reproduction. If males should ever appear and, if they were functional, I should expect the progeny from a mating to produce variations equivalent to the species of the parthenogenetic ring. On the other hand, if males are entirely suppressed then we should expect these species to be fixed within certain limits and incapable of any great morphological variation from their norms, and this appears to be the case in these New Zealand species. Fig. 1.—Diagrammatic representation of parthenogenetic variations in Acanthoxyla. This continual variation round a norm can be depicted as a circle or ellipse. If the variations of these seven parthenogenetic species are plotted around the circle (1–7) we obtain a diagram as shown in Fig. 1, in which each species varies on either side of the axes AA1, BB1, etc. As each species varies irregularly, it cannot be represented by a circle, but plots, instead, an elliptical figure and the inward limits of these seven figures subtend a further circle A-G, which should demarcate the limits of variation of the originating form. In the case of Acanthoxyla, this originating form from which the parthenogenetic species presumably arose, gives us a concept of the characters of the genus. This diagrammatic concept of variation might be applied to other groups of variant species, not necessarily parthenogenetic, with equally interesting results.