1. A retrospective survey has been made to determine the amount of cross-infection caused by Staphylococcus aureus of one particular phage-type, 83 A, in four main hospitals of a provincial hospitals group, over a period of 2½ years.

2. The total number of strains considered was 1131. Of these, 776 were resistant to tetracycline and therefore the most likely to be of significance in cross-infection. Of these 776 strains, about one-third belonged to phage-type 83 A. In the hospital where 83 A strains were most prevalent, they formed about one-half of the tetra cycline-resistant strains; in that where they were least prevalent, about one- quarter.

The clinical sources of the tetracycline-resistant strains are described.

3. It was found that type 83 A strains were the most likely to show resistance to erythromycin, chloramphenicol or novobiocin.

Of a series of 214 type 83 A strains sensitivity tested against methicillin, using a disk-plate method, none was found to be resistant.

My thanks are due to Prof. R. E. O. Williams, St Mary' Hospital Medical School, for his helpful criticism of this paper and, in his previous capacity as Director of the Staphylococcus Reference Laboratory, Central Public Health Laboratory, Colindale, to him and his staff for supplying me with typing phages, and instructing me in their use. I should also like to thank Dr M. T. Parker, Director of the Cross Infection Reference Laboratory, Colindale, for a continued supply of these phages; the Hon. the Viscount Waverley, F.R.C.P., for allowing me to quote from the case history of a patient under his care; and Mr K. Allen, F.I.M.L.T., and other members of my technical staff for their assistance.