A throat swab from a 9 year old girl with pharyngitis yielded a non-toxigenic strain of Corynebacterium diphtheriae var mitis and Streptococcus group G. C pseudodiphtheriticum was isolated from the throats oftwo ofher four brothers. In each case the isolate was sent to the reference laboratory before fill identification. The growth was found to be mixed for one brother; the other isolate being a toxin producing C diphtheriae var gravis. The child was asymptomatic and the case proves that all colonial types on the Hoyles plate should be identified. (7 Clin Pathol 1992;45:1036-1037 We have already suggested that routine screening for Corynebacterium diphtheriae should not be abandoned in the United Kingdom despite the current lack of staff and resources.' We describe a family outbreak that would not have otherwise been detected and which illustrates a potential pitfall in diagnosis. The family were then requested to attend the general practitioner to have throat swabs taken and they were treated with erythromycin. The parents and two of the brothers were asymptomatic, but the 2 year old and the 4 month old brother had fever without a sore throat. The family had been in the United Kingdom for 13 years since leaving Malaysia. One month earlier, the parents and the youngest child had been on a pilgrimage to Mecca. The other children had been looked after by a grandmother who was visiting from Malaysia.No pathogens were isolated from the pharyngeal cultures of the parents, the youngest child, or from one of the asymptomatic brothers. However, the throat swab from the 2 year old showed a heavy growth of a group G Streptococcus and, on Hoyles medium, of Gram positive bacilli, later identified by API Coryne as C pseudodiphtheriticum.Culture of the throat swab of the asymptomatic 7 year old produced six colonies of a coryneform on the Hoyles plate. A subculture was made and the original plate was sent to the reference laboratory. The subcultured isolates did not produce a halo on a Tinsdale plate, and were later identified as C pseudodiphtheriticum. However, subcultures from the original plate at the reference laboratory showed two colonial types, one of which produced a halo on the Tinsdale medium. This isolate was identified as C dtphtheriae var gravis and was shown to produce toxin both by the Elek test and by guinea pig inoculation. It differed from the strain isolated from the sister by producing acid from glycogen and not from sucrose. The second colony type was C pseudodiphtheriticum.Re-examination of the original culture of the sister's throat confirmed the presence of only non-toxigenic C diphtheriae var mitis.Five days later, the mother complained of headache and malaise and was found to have tonsillitis and a nasal discharge. The father and all the children were well. The mother and three of the children were admitted to an isolation hospital and, with the exception of the youngest child, they were treated with antitoxin and another course of erythromycin. All throat swabs were negative and the mother m...