It has been shown by several workers that the colloidal gold test is positive in a proportion of cases of rheumatoid arthritis; in 76 per cent. of a series of twenty-four by Carter and Maclagan (1946), in 71 per cent. of a series of eleven by Rennie and Rae (1947), and in 61 per cent. of a series of a hundred by Fraser (1948). Although the latter performed 133 tests in the hundred cases, none of these workers made repeated observations on the same cases over a period to see whether the test altered with the passage of time or with changes in clinical condition. This present investigation was undertaken in order (a) to confirm the observation that the gold sol reaction was indeed positive in a proportion of cases of rheumatoid arthritis; (b) -to see whether the test varied over a period of time or with a change in the clinical condition; and (c) to try to discover the significance of the test in the light of clinical and other observations.
Material and TechniqueThe patients in this series of cases all suffered from rheumatoid arthritis according to the accepted diagnostic criteria for that disease; sixty-five were in-mpatients, and five were out-patients of the Rheumatic Unit of the Northern General Hospital, Edinburgh, and two were from the general wards of the same hospital.Serum for the test was obtained by centrifuging a specimen of venous blood which had been allowed to clot and theui stored overnight in a refrigerator. The 'test was performed at weekly, fortniightly, or monthly intervals, depending whether the patient was in the wards or attending the follow-up clinic. The method of performing the test and the gold sol ursed were those recom-