In receptor autoradiography, the tissue distribution of receptors is v
isualized in the form of ligand bindings. However, this visualization
is indirect because we can only observe and measure nonspecific bindin
gs and total bindings. The distribution of receptors is obtained by su
btracting the nonspecific bindings from the total bindings. In recepto
r autoradiography, it is important to compare the distribution of rece
ptors between two groups of samples obtained from two different region
s of tissues and/or organs. However, no appropriate method of statisti
cal analysis has yet been used to compare the distribution of receptor
s of two different groups. In this paper, we present a statistical met
hod for this purpose and apply this method to an example.