A chemically defined fluid medium—improved formate lactose glutamate—is described of which the cost is less than a quarter that of standard MacConkey broth and which is at least equally suitable for routine bacteriological examination of water. It has actual advantages in that the medium after fermentation is clearer and that, therefore, acid and gas production is more easily seen; moreover more isolations of both coliform bacteria and of Esch. coli are obtained with fewer false positive reactions. Trials with a solid version of the medium suggest that the relative freedom from false positive reactions arises from the failure of the medium to grow Cl. welchii—one of the commonest causes—and enterococci, another suspected participant. In fact, without any inhibitory agent the medium appears to be virtually specific for the coliform group and particularly favours Esch. coli. It remains to be seen whether certain aerobic spore-bearers found from time to time in surface waters may, to some slight extent, interfere with the almost complete specificity so far discovered.

The medium is recommended in place of MacConkey broth and other peptonecontaining media, not only for the prospect of better results, in that suppression of the coliform group does not occur, but also for cheapness and for freedom from the inevitable variability associated with media incorporating peptone and, particularly, bile salt. It is hoped that this medium will shortly become available commercially in dehydrated form at a price still considerably less than that required to make up MacConkey broth from its constituents.

Some labour-saving curtailments in the routine of water examination are also suggested.

It is hoped not only that the fluid medium will supplant MacConkey broth for the dilution method of water analysis but that the solid medium, without material alteration, will also prove suitable for water examination by membrane filtration.

I am deeply grateful to my laboratory staff, particularly Mr J. H. Evans, F.I.M.L.T., and Mr G. H. Lowe, F.I.M.L.T., for undertaking so much of the work when routine work was particularly heavy, and for many valuable suggestions. My thanks in addition are due to Mrs Peggy Johnson and Mrs Marilyn Childs, who made up the medium, and also to the directors of the participating P.H.L.S. laboratories not only for having the work carried out but for helpful comments.