The anemone Metridium senile occurs in a number of striking colour varieties. These are to be found side in nature. The colours are inherited at least during asexual reproduction and do not depend upon special food. The colour varieties are founded on three systems of pigment, black granular melanin in the endoderm, brown diffuse melanin in the ectoderm and red to orange lipochrome in fat droplets in both endoderm and ectoderm. In any variety each system may be present or absent. If all are absent we have a white form, if one is present we get red, brown or grey forms, if two or more are present we get other varieties. In combinations each pigment retains the distribution it has in the corresponding simple form. The black endodermal pigment is a true melanin. The brown ectodermal pigment is also a melanin but is less stable. Even the white variety can be shown to possess a complete tyrosinase system when the tissues are finely ground. It is perhaps comparable to a case of dominant albinism. The red and orange colours are due to a variety of lipochromes. The pigment of the red variety is chiefly the acidic carotenoid described by Heilbron, Jackson and Jones. It is related to astacene and we have named it 'metridene'. Yellow-orange anemones of the red series may contain considerable quantities of xanthophylis in addition to metridene esters. The only pigment of the white form is contained in the gonads and appears to be esters of astacene. Many of the carotenoids are present in too small quantities to influence the appearance of the animal. The assortment of carotenoids present varies considerably from one lot of Metridium to another. There is evidence that the occurrence of a variety of xanthophylls and carotene is more common in varieties which develop melanin. But there is no obvious correlation between the presence of a particular assortment of carotenoids and a particular distribution of melanin pigments. Flavines are not present in any of the colour varieties. Coloured purine compounds are not present. The presence of significant quantities of uric acid was demonstrated and of this purine alone. It is present in the mesenteric filaments, in the yellowish excreted mucus, and granules of uric acid form white bands in the endoderm of the tentacles. In this position they contribute their whiteness to the total colour effect. Haematins do not contribute to the colour. Those that are present are derived from the ordinary intracellular respiratory systems. The colours of the varieties are not an adaptation to the external environment as in warning coloration etc. nor is there evidence as yet that the colour as such is of direct physiological importance to the animal. The varieties appear to illustrate Poulton's category of 'non-significant colours', in which colour is a by-product of biochemical processes utilized for other purposes.