The laser-Doppler blood flow and erythema index were assessed in 16 patch test reactions caused by irritants (1% aq. sodium lauryl sulphate and 1% aq. benzalkonium chloride) and in 13 varied allergic reactions, at 2 or 4 days. Both irritant and allergic responses produced statistically significant increases in laser-Doppler flow index and erythema index compared to control sites (p less than 0.05 or less, using the Wilcoxon rank-sum test). A disproportionately greater increase in erythema index than in laser-Doppler flow was seen in mild irritant reactions, compared to allergic, though the two could not be reliably distinguished by these tests. 2 non-allergic nickel sulphate sites produced an increase in both blood flow and erythema without any clinical change. Petrolatum alone produced no significant change in either measurement. The laser-Doppler flow showed an overall correlation with the erythema index (product moment method; r = 0.55, p less than 0.001), but there was little correlation between these indices and patch test reactivity as judged clinically by conventional scoring. Generally, allergic and irritant patch test reactions could not be differentiated on the basis of laser-Doppler flow or erythema index. However, the disproportionate increase in erythema index over laser-Doppler flow index for mild irritant responses warrants further study in other irritant models.