Gel filtration of lyophilized rat pancreatic juice has been carried out on Sephadex G-100 at pH 5.8 and 8.0, both in the presence and absence of sodium taurodesoxycholate, to determine whether bile salts affect the gel filtration behaviour of pancreatic juice proteins or the enzymes hydrolyzing emulsified triolein (triolein lipase) and micellar mono-olein (mono-olein lipase). At pH 5.8 and at pH 8.0 sodium taurodesoxycholate, in concentrations above its critical micellar concentration, changed the elution volume of triolein lipase only slightly. It did, however, significantly increase the recovery of applied activity in the column eluent. This increased recovery appeared to be due to the elution of a lipase activator with the lipase in the presence of the bile salts. In the absence of sodium taurodesoxycholate this activator was separated from the lipase with consequent loss of activity. At pH 5.8 sodium taurodesoxycholate changed the protein elution pattern markedly, protein being eluted earlier, as a single peak instead of a multicomponent complex. At least some of this change was probably due to desorption of amylase from the column. Mono-olein lipase was also eluted earlier in the presence of sodium taurodesoxycholate. At pH 8.0, proteins were eluted earlier than at pH 5.8 and the effect of sodium taurodesoxycholate was less marked. Mono-olein lipase was eluted with the same K av at pH 8.0 as at pH 5.8, and showed a similar change in elution volume in the presence of sodium taurodesoxycholate at both pH's. Studies with 35S-labelled sodium taurodesoxycholate failed to demonstrate any binding of this bile salt to proteins or either lipase. It is concluded that sodium taurodesoxycholate does not change the molecular size or configuration of triolein lipase sufficiently for this to be detected by gel filtration. It does, however, promote the elution of a lipase activator together with the enzyme, resulting in an improved yield of lipase after chromatography.