Summary: X-ray diffraction data on serpentine minerals are discussed. There arc three structural varieties of chrysotile: ortho-chrysotile, clino-chrysotile, and para-chrysotile. Methods are described for the estimation of the proportions of the first two of these in mixed specimens, and their distribution in nature is reviewed. The variations observed in powder photographs of serpentine minerals are interpreted in accordance with this classification. The criteria proposed by Selfridge for distinguishing between chrysotile and antigorite are shown to be unreliable, but valid criteria are presented. The powder photographs also reveal the existence of a third serpentine mineral, distinct from chrysotile and antigorite, for which the name lizardite is proposed. Four samples of bastite are shown to consist of chrysotile or lizardite rather than antigorite. All the serpentine minerals examined consist either of antigorite or of one or more of the group comprising lizardite and the chrysotile varieties.