Observe that the right-hand side can be written as =n(n+1)2=1+2...+n Using the result of problem 3.21of Sect. 3.3,we obtain n(n+1)2=∑d/1φ(d)+...+∑d/nφ(d).Now,forsomek,\displaystyle 1\leq k\leq n,letuscounthowmanytimes\displaystyle \varphi \left ( k \right )isatermofthesum\displaystyle \sum_{d/m}\varphi \left ( d \right )ifandonlyifkisadivisorofm,or,equivalently,misamultipleofk.So,\displaystyle \varphi \left ( k \right ) appearsasmanytimesasthenumberofmultiplesofkintheset1,2,...,nthatis,\displaystyle \left \lfloor \frac{n}{k} \right \rfloor.weconcludethatthesumcanbewrittenas\displaystyle \varphi \left ( 1 \right )\left \lfloor \frac{n}{1} \right \rfloor+\varphi \left ( 2 \right )\left \lfloor \frac{n}{2} \right \rfloor+...+\varphi \left ( n \right )\left \lfloor \frac{n}{n} \right \rfloor, giving the required result. Alternatively, we cound count in two ways the number of fractions ab with 1≤a≤b≤n where we do not insist that our fraction be in lowest terms, so for example 12 and 24would count as different fractions. First, this number is clearly n∑b=1=n(n+1)2since this is the number of such pairs (a,b).Second, we count fractions based on their reduced forms. We saw in problem 3.21 that the number of reduced fractions with denominator d is ϕ(d). The number of multiples of such a fraction with de-nominator at most n is ⌊n/d⌋ so the number is also n∑d=1ϕ(d)⌊nd⌋.