More than 110,000 young adults—Navy recruits, college students and student nurses—from nearly all parts of the country were tested simultaneously with coccidioidin and histoplasmin between the years 1945 and 1951. Results for persons classified as lifetime residents of a single county provide material for mapping geographic variations in the prevalence of coccidioidin and histoplasmin sensitivity, based on the conventional 5 mm. criterion for a positive reaction.
Correlations of reactions to coccidioidin and histoplasmin are studied in three groups of Navy recruits, lifetime residents of five northwestern states, of eight eastcentral states and of four southwestern states. Just over 1 per cent of recruits from the northwest reacted to either coccidioidin or histoplasmin. More than 60 per cent of recruits from the Histoplasma endemic area (eastcentral) reacted to histoplasmin, 1.5 per cent to coccidioidin and 1.2 per cent to both. Eleven per cent of those from the Coccidioides endemic area (southwest) reacted to coccidioidin, 8 per cent to histoplasmin and just over 4 per cent reacted to both antigens.
Reactions to the homologous antigen in the endemic area—presumably specific reactions—were appreciably larger, on the average, than reactions in the non-endemic areas. Moreover, as the mean size of specific reactions does not appear to vary with variations in the prevalence of sensitivity, the smaller reactions found in the non-endemic areas are interpreted as nonspecific. Relatively few of the small, presumably nonspecific reactions to coccidioidin were found in lifetime residents of the eastcentral area, and most of them were probably cross-reactions with Histoplasma. Some of the presumably nonspecific reactions to histoplasmin in the southwest can likewise be ascribed to cross-reactions with Coccidioides. But the substantial frequency of small histoplasmin reactions found in coccidioidin non-reactors probably represent infection with some other fungus, or fungi, which may be as prevalent as Coccidioides in the southwest.
The criterion for a positive (specific) reaction is shown to vary with the prevalence of both specific and nonspecific sensitivity. The use of the conventional 5 mm. criterion can therefore be misleading, as it overestimates the percentage of specific reactions in some localities and underestimates it in others. Careful measurement of the sizes of reactions, with detailed study of frequency distributions of the measured sizes, can provide much needed information on the complex and little understood nature of specific and nonspecific sensitivity.