Spiroplasma strains 23-6 and 27-31 from flowers of Liriodendron tulipifera L.(tulip tree) were indistinguishable from one another but were distinct from Spiroplasma citri and from members of serogroups other than the serogroup which contains S. citri. These two strains shared close to 100% deoxyribonucleic acid homology with one another, but only low homology (4%) with S. citrz and other Spiroplasma strains. The acrylamide gel electrophoretic patterns of the cellular proteins of strains 23-6 and 27-31 appeared to be identical to each other, but they were distinct from the patterns of S. citri and various other strains. On the basis of serological and biochemical characteristics, we propose that strains 23-6 and 27-31 belong to a new species, Spiroplasma floricola, with strain 23-6 (= ATCC 29989) as the type strain. This new species grew well at 30 and 37"C, in contrast to S. citri, which grows poorly if at all at 37°C. During log-phase growth in broth medium, S. floricola cells occurred as motile, helical, cell wallless filaments 0.15 to 0.2 pm in diameter and 2 to about 5 pm long. Classical "friedegg" Mycoplasma-type colonies were not observed on 1% agar medium. Instead, colonies had granular centers and diffuse borders and contained abundant, motile, helical cells. Growth was stimulated by cholesterol and inhibited by digitonin. Tests for catabolism of glucose, fructose, mannose, and arginine were positive. Tests for phosphatase activity were negative. The guanine-plus-cytosine contents of the deoxyribonucleic acids of strains 23-6 and 27-31 were close to 25 mol%.Spiroplasmas were first recognized in and cultivated in vitro from diseased plants (13, 14, 16, 29). It is now apparent that spiroplasmas are widespread in nature and that some strains associated with plants probably do not cause plant disease (7, 9, 10, 12, 22). Other Spiroplasma strains are associated with diseases in insects (6, 39,41) or are found in apparently healthy insects (17; R. E. McCoy, J. H. Tsai, and D. L. Thomas, Phytopathol. News 12: 217, 1978) and ticks (2, 5, 25, 30, 35, 36,43). Several distinct species are doubtlessly represented by the diverse strains now in culture in vitro, but only one species, Spiroplasma citri, has been recognized formally thus far (28).Among the spiroplasmas from varied habitats are strains 23-6 and 27-31, which were isolated from surfaces of flowers of the tulip tree (Liriodendron tulipifera L.) (9, 10, 12). Triply cloned strain 23-6 was shown to be serologically distinct from the honey bee Spiroplasma (strain AS 576) and from S. cztri, and Davis suggested that strain 23-6 represented a new species (9). Subsequently, a scheme for the classification of Spiroplasma strains into serogroups and serological subgroups on the basis of membrane antigen reactivity was proposed (12). Estimations of deoxyribonucleic acid (DNA) base compositions and DNA-DNA homologies confirmed that the classification scheme reflected phylogenetic relationships (18) and supported the proposal (12) that the serogroups represented separa...