I'm currently working through introductory functional analysis from kreyszig, and I don't quite understand one of the proofs. this is it:

My question is why it is necessary to go from $k = 1, 2, ...$ to the infinite sum. Why can't we just let $n\to\infty$ directly in (3)?