If repeating an experiment many times with different samples what values for the sample means would be possible?

The answer depends on the population and is described by the sampling distribution of the mean.