You will use the t-distribution for this test because you don't know the population standard deviation.
The degrees of freedom = 104-1 = 103.
For a two-sided test need to look at 0.01/2 = 0.005 tail value as the critical value which is 2.624.
So critical value = +/- 2.624.
t = (77.2 - 73.1)/(19.9/sqrt(104)) = 2.101.