%I #12 Jul 14 2019 18:01:30
%S 50,71,267,517,2274,1508,3884,1971,444,667,817,286,535,732,856,2926,
%T 541,736,858,293,542,233,483,695,834,2888,16995,41225,20304,4506,6713,
%U 2591,5091,22564,47502,21795,46686,21607,46484,215602,46433,68142,261041,510922
%N a(1) = 50 and for any n > 0, a(n+1)^2 is the smallest square that begins with a(n).
%C This sequence is similar to A308055.
%C The initial value (50) seems to be the first one for which the iteration of A018796 diverges; there are neither duplicates nor squares among the first 79000 terms.
%H Rémy Sigrist, <a href="/A309123/b309123.txt">Table of n, a(n) for n = 1..1000</a>
%H Rémy Sigrist, <a href="/A309123/a309123.gp.txt">PARI program for A309123</a>
%F a(1) = 50 and then a(n+1) = A018796(a(n)) for n > 0.
%e The first terms, alongside the square of a(n+1), are:
%e n a(n) a(n+1)^2
%e -- ---- --------
%e 1 50 5041
%e 2 71 71289
%e 3 267 267289
%e 4 517 5171076
%e 5 2274 2274064
%e 6 1508 15085456
%e 7 3884 3884841
%e 8 1971 197136
%e 9 444 444889
%e 10 667 667489
%e 11 817 81796
%e 12 286 286225
%e 13 535 535824
%e 14 732 732736
%e 15 856 8561476
%e 16 2926 292681
%o (PARI) See Links section.
%Y Cf. A018796, A308055.
%K nonn,base
%O 1,1
%A _Rémy Sigrist_, Jul 13 2019