Twelve stock and 15 freshly isolated strains of Mycoplasma pulmonis were examined and compared using four serological techniques. Complement fixation tests and inhibition of growth on solid medium showed that the strains were all M . pulmonis, although, with some strains, only one of these techniques gave an unequivocal result. These tests were confirmed to be specific only at a species level. Gel diffusion and metabolic inhibition techniques revealed heterogeneity of the strains at the subspecies level. The relationship between some strains was statistically significant and stable in vitro and some small groups could be defined.These serogroups bore no relationship to groups formed on the basis of protein patterns produced on polyacrylamide gel electrophoresis of strains of Mycoplasma pulmonis.

Strains of Mycoplasma pulmonis, examined by polyacrylamide-gel electrophoresis of proteins extracted in a phenol-acetic acid-water solvent, gave patterns with reproducible differences. Some strains could be placed into groups, but a number could not be associated with any group. In all but one case a 'basic M. pulmonis pattern ' was easily recognizable. This exception suggests that occasionally one might fail to recognize a strain of M.pulmonis by relying solely on the electrophoretic technique. I N T R O D U C T I O N Fowler, Coble, Kramer & Brown (1963) suggested that the electrophoretic separation of proteins of mycoplasmas be used as a new method of identifying members of this species. Their observations were extended by Rottem & Razin (1967). The patterns obtained by these workers were distinct for the species of mycoplasmas examined. Sometimes there were differences between the electrophoretic patterns of strains within a species; for example, Myco-plasma yulmonis (Razin & Rottem, I 967 ; Razin, I 968). In the present investigation 19 strains of Mycoplasma pulmonis were examined by poly-acrylamide-gel electrophoresis to find out to what extent the electrophoretic patterns of their proteins differed and to see if a typing scheme could be formed on the basis of these patterns. M E T H O D S Organisms. The strains of mycoplasmas and the sources from which they were obtained are given in Table I. The strains included reference strains and recent isolates. Preliminary identification of recently isolated strains was made by the growth inhibition test (Clyde, 1964) employing at least two antisera to Mycoplasma pulmonis plus antiserum to the other rodent mycoplasmas M. neurolyticum and M. arthritidis. Strains which were inhibited were regarded as members of the species M. pulmonis. Clones. All strains were cloned at least three times. Where an original culture showed several colonial forms, representative colonies were cloned six times. Media and growth conditions. The media used for the cultivation of the mycoplasmas were modified from that described by Chanock et al. (1962). Thallous acetate was omitted and ampicillin at a final concentration of 0-1 yo was used in place of penicillin (Fallon, 1969). Medium which contained I % (w/v) glucose and 0'002 % phenol red was inoculated from a logarithmic phase culture and incubated aerobically at 37 "C until a colour change in the medium showed a drop of 0.5 pH unit, which took approximately 48 h. For solid medium I yo (wlv) Ionagar no. 2 (Oxoid) was added.

SUMMARYSerological evidence is presented that Sabin's murine type C mycoplasma belongs to the species Mycoplasmapulmonis and not to a separate species, M . histotropicum.Between 1938 and 1941, several apparently distinct mycoplasmas were isolated from mice by Sabin (1938, 1939 a, b, 1941). These organisms were designated types A culture of type C, which had undergone 60 passages since recovery from Sabin's lyophilized material, was received from Dr J. G. Tully (National Institutes of Health, Bethesda, Md., U.S.A.) and examined at the Lister Institute in 1964. The culture was cloned by selecting a single colony at each of two successive subcultures, and tested by complement-fixation and gel-diffusion (Lemcke, 1964(Lemcke, ,1965 and by an agar growthinhibition technique (Clyde, I 964) against antisera to I 7 serologically distinct mycoplasmas (Lemcke, 1964). It reacted only with antisera to two strains of Mycoplasma