Evidence of radius inflation in stars approaching the slow-rotator sequence

Context. Average stellar radii in open clusters can be estimated fromrotation periods and projected rotational velocities under theassumption that the spin axis has a random orientation. These estimatesare independent of distance, interstellar absorption, and models, buttheir validity can be limited by lacking data ( truncation) or data thatonly represent upper or lower limits (censoring).Aims. We present a new statistical analysis method to estimate averagestellar radii in the presence of censoring and truncation.Methods. We used theoretical distribution functions of the projectedstellar radius R sin i to define a likelihood function in the presenceof censoring and truncation. Average stellar radii in magnitude binswere then obtained by a maximum likelihood parametric estimationprocedure.Results. This method is capable of recovering the average stellar radiuswithin a few percent with as few as about ten measurements. Here weapply this for the first time to the dataset available for the Pleiades.We find an agreement better than approximate to 10 percent between theobserved R vs. M-K relationship and current standard stellar models for1 : 2 >= M / M-circle dot >= 0 : 85 with no evident bias. Evidence of asystematic deviation at 2 sigma level are found for stars with 0 : 8 >=M = M-circle dot >= 0 : 6 that approach the slow-rotator sequence. Fastrotators (P < 2 d) agree with standard models within 15 percent with nosystematic deviations in the whole 1 : 2 greater than or similar to M =M-circle dot greater than or similar to 0 : 5 range.Conclusions. The evidence of a possible radius inflation just below thelower mass limit of the slow-rotator sequence indicates a possibleconnection with the transition from the fast-to the slow-rotatorsequence.