On a conjecture on Pythagorean numbers. III (Q1327782)

[For Parts I, II cf. Proc. Japan Acad., Ser. A 69, No. 7, 252-255 (1993; Zbl 0796.11009) and ibid. No. 8, 287-290 (1993; Zbl 0796.11010).] In this paper, the author considers the following diophantine equation on \(l,m,n\in \mathbb{N}\) \[ (4 a^ 2- y^ 2)^ l+ (4 ay)^ m= (4 a^ 2+ y^ 2)^ n, \tag{1} \] where \(a,y\in \mathbb{N}\) with \((a,y)=1\), \(2a> y\), \(y\equiv 1\pmod 4\). He gives several sufficient conditions for \((l,m,n)= (2,2,2)\), e.g. let \(a\) be odd, \(y=p\) an odd prime, and \(p\equiv 5\pmod 8\) in (1), if \(m\) is even, then \((l,m,n)= (2,2,2)\).