The sample sizes are quite large, so the central limit theorem applies. (It typically does as soon as the sample size exceeds 30 or so.) This means that the sample mean will be approximately normally distributed with the same mean 22 but standard deviation 3.1/√40 ≈ 0.4902.
Now, if the question is asking about the probability of the sample mean being an exact number, that probability would be zero.
But if you meant to ask something else, like "what is the probability that the sample mean is less than 21?" then we would have a non-zero probability. In this particular case, if Y is a random variable for the sample mean, then
Pr[Y < 21] = Pr[(Y - 22)/(3.1/√40) < (21 - 22)/(3.1/√40)]
… ≈ Pr[Z < -2.0402]
… ≈ 0.0207
