In Fraenkel's Love and Math (and Richard Taylor's Modular Arithmetic IAS Post https://www.ias.edu/ideas/2012/taylor-modular-arithmetic), specifically in Chapter 8 Magic Numbers, page 88., Fraenkel mentions that Eichler was the first to discover the relationship between (number of solutions of) elliptic curves and (coefficients of Fourier expansions of) modular forms. He shows the concrete example that Eichler worked out.
Anyone has the reference where Eichler showed this example? I have his 1954 paper "Quaternäre quadratische Formen und die Riemannsche Vermutung fÜr die Kongruenzzetafunktion", but I don't understand german and the example doesn't seem to be stated there, at least explicitely.
And more interestingly, does anyone has any idea of how Eichler arrived at this discovery? Was there anything similar observed before? (specially considering infinite sequence of arithmetical information)