Re^2: Error-finding program

[Richard Mathar]

dr> From seqfan-owner at ext.jussieu.fr  Wed Oct 25 18:20:49 2006
dr> Return-Path: <seqfan-owner at ext.jussieu.fr>
dr> Date: Wed, 25 Oct 2006 10:13:43 -0600
dr> From: Don Reble <djr at nk.ca>
dr> To: Sequence Fans <seqfan at ext.jussieu.fr>
dr> Subject: Re: Error-finding program
dr> ...
dr>     ... have surprising trailing zeroes:
dr> A026468 A026469 A026470 A030755 A030795 A037533 A037534 A048060 A048070
dr> ...

The definitions of A026468 and A026469 are
 a(n) = least positive integer > a(n-1) and not a(i)^2 + a(j)^2 for 1<=i<=j<=n.
and
 a(n) = least positive integer > a(n-1) and not a(i)^2 + a(j)^2 for 1<=i<=j<=n.  	
which looks quite identical. So why can A026468/9 be two different
sequences? Is there a "not" too much or too little, or rather one of the 
"less-equal" signs to be corrected to "less" or a starting condition on a(2)
to be added?

--Richard