SUMMARYPurified teichoic acids and extracts with group-specific activity have been examined with specific antisera for a number of lactobacilli. The wall teichoic acids from groups D and E lactobacilli have been identified as the group-specific substances, whereas in group A the intracellular teichoic acid showed group-specificity. It is possible that group-specificity is associated with intracellular teichoic acid in group F.

I N T R O D U C T I O NIn work on Cheddar cheese flavour, methanethiol-producing coryneform bacteria were isolated from milk and cheese and their characteristics have been described (Sharpe, Law & Phillips, 1976). As the optimum temperature for growth of these strains was 30 to 37"C, and they tolerated 15 % NaCl, these authors suggested that the strains might originate from the skin. Investigations of coryneform bacteria from human skin have shown that organisms with a similar cell-wall pattern are common members of the diphtheroid skin flora (Pitcher, 1976). Preliminary work showed that many of these organisms produced methanethiol from L-methionine.This paper describes characteristics of some of the methanethiol-producing dairy and skin strains of coryneform bacteria having meso-diaminopimelic acid as their cell-wall diamino acid. Strains of Brevibacterium linens were included, as many characteristics of these organisms (Yamada & Komagata, 1970, 1972a were similar to those of our strains. METHODSThe 24 strains examined in detail are listed in Table I. In addition, four orange-pigmented isolates from Limburger and Meshanger cheeses and three orange-pigmented strains from sea fish, described by Mulder et al. (1966) and Crombach (1974a, b) (provided by W. H. J. Crombach), and one strain of Brevibacterium linens ~~1 0 7(from NIZO, The Netherlands) were examined only for methanethiol production and for tyrosine breakdown.Organisms were cultured in Oxoid broth, Oxoid brain-heart infusion broth or on Oxoid nutrient agar. Because B. linens strains grew better at 22 "C, whilst most of the others grew better at 30 "C, tests were incubated at both temperatures unless otherwise stated.Growth temperatures, survival at 60 "C for 30 min, proteolytic effect on gelatin and milk, level of tolerance to NaCl, utilization of acetate and lactate, and morphological change from rods to cocci were determined as described by Sharpe et a/. (1976).Orange pigmentation. To determine pigment production in the presence and absence of light (Mulder et al., 1966), cultures were grown on nutrient agar at room temperature. The presence of a carotenoid-like pigment was determined as described by Jones, Watkins & Erickson (1973).Tyrosine decomposition. Cultures grown for 48 h in nutrient broth or infusion broth were inoculated on nutrient agar containing 0.5 % (w/v) L-tyrosine suspended evenly in the medium. Plates were incubated and inspected for clearing for up to 4 weeks. DNAase activity. Cultures grown for 48 h in nutrient broth or infusion broth were inoculated on DNA agar (Oxoid) and incubated for I week. Activity was detected by flooding the plates with 0.1 M-HCI, depolymerization of DNA being indicated by zones of clearing.Urease activity. Oxoid Christensen medium was used, tests being read at intervals up to I week. Presence of mycolic acids. The method of Minnikin, Alshamaony & Goodfellow (1975) was used.

S U M M A R YAn antiserum prepared against a strain of Lactobacillus acidophilus gave a reaction of identity with HCI extracts of strains of lactobacilli, some pediococci and leuconostocs and Staphylococcus aureus. The antigenic determinant, isolated from a strain of L. casei RO-94 was identified as a membrane glycerol teichoic acid. Purified membrane lipoteichoic acids from Lactobacillus serological groups A, B, C, D and F and streptococcal polyglycerophosphate also reacted with the antiserum. The antigen-antibody reaction was partially inhibited by a-glycerophosphate and strongly inhibited by glycerol-phosphoryl-glycerol-phosphorylglycerol, but was not inhibited by glucose. It was concluded that the antiserum was reacting with the common polyglycerophosphate backbone of the membrane teichoic acids in the reacting strains.

SUMMARYWhen serological type strains of Streptococcus faecalis from different workers were compared with those of Sharpe & Shattock (1952) by using precipitin and reciprocal absorption tests, several of these types were found to possess the same type antigens. The location of these type antigens was in the cell wall and they were polysaccharide in nature. Some strains had two of these antigens. The distribution of S. faecaEis serotypes is discussed and a typing scheme applicable to all recognized types of group D streptococci is suggested.

SUMMARY: A streptococcus reacting with both group D and group N specific antisera was found to belong to group D ; the reaction with the group N serum was shown by reciprocal absorption tests to be due to a common type antigen. Two group D strains were encountered having this antigen ; both were identified as Streptococcus faecium. The group N strain possessing this same antigen was Streptococcus Zactis. It proved to have a t least one other type antigen, also found in another strain of Strep. Zactis which was not serologically identical. Two antigenic variants were present in cultures of this organism ; one possessed all the type antigens characterizing this strain the other had only one. Attempts to cause further dissociation or reversion were not successful.Streptococci of different Lanceficld groups having common type antigens have been previously reported; Bliss (1 93'7) found a group F type to be serologically identical with a group G type, and Maxted (1949) described four strains of streptococci, one belonging to group B, two to group C and one to group G , each having the R i antigen of group A, type 28. When a streptococcus was encountered which reacted with both a group D and a group N specific antiserum in the precipitin test, using forniamide extracts, it seemed likely that one of the reactions might have been due to a common type antigen. This was found to be so, and the following account is given of a group D and a group N streptococcus having a similar type antigen. Later another group I) strain was found with this same antigen. MATERIALS AND METHODSThe streptococcus strain 748, an extract of which reacted with a group D and a group N serum, was sent to this laboratory by Prof. M. Seelemann, Institut fur Milchhygiene, Kid. It was isolated from the blood of a case of endocarditis. Prof. Seelemann was also kind enough to send samples of his group D and group N sera, and the strains group D 7 5 3