Quantitative structure activity relationships (QSARs) have been derive
d relating skin corrosivity data of organic acids, bases and phenols t
o their log(octanol/water partition coefficient), molecular volume, me
lting point and pK(a). Datasets were analysed using principal componen
ts analysis; plots of the first 2 principal components of the above pa
rameters, which broadly model skin permeability and cytotoxicity, for
each group of chemicals showed that the analysis was able to discrimin
ate well between corrosive and non-corrosive chemicals. The derived QS
ARs should be useful for the prediction of the skin corrosivity potent
ial of new or untested chemicals.