For a Normal Distribution, a 90% Confidence Level corresponds to a
Critical Z Value of zc = 1.645.
Guide on the relation zc = (X-bar − μ) ÷ σX-barand restate as zcσX-bar = (X-bar − μ).
This last translates to zc[σ/√n] = (X-bar − μ); σX-bar is replaced by [σ/√n] by presuming
that the sample size is less than 5% of the population size (or n < 0.05N).
Enumerate zc[σ/√n] = (X-bar − μ) to 1.645[61.8 ÷ √n] = 10 which goes to
0.1645 = √n ÷ 61.8.
Finally √n = 10.1661 gives n as 103.3495892; round to the highest integer
to obtain 104 as the sample size sought.