Here's a radioactive binding assay: 1. Mix radioactively labeled ligand of known concentration with normal ligand (assuming radioactivity does not affect the ability to bind). 2. Add the mixture of labeled and unlabeled ligand into the solution of receptor. 3. Separate bound ligand from unbound (by centrifugation). 4. Measure radioactivity of the receptor (which should be ligand bound). By comparing with the known concentration of labeled ligand, the fractional saturation can be calculated and a binding isotherm can be plot. I don't understand why they have to mix the labeled ligand with normal ligand in the first step. Why can't they just use a solution of radioactively labeled ligand? You still get the portion of ligand bound by comparing with the known concentration in the beginning?