# [seqfan] Re: Likely an error in the sequences A046144, A046146 names

T. D. Noe noe at sspectra.com
Mon Dec 3 07:00:47 CET 2012

```Not an error.  IMHO, the original name was fine.  This is how experts talk
about it.  This is a rather esoteric part of number theory.  Without
restoring the old name and adding a comment that carefully describes the
sequence.

Best regards,

Tony

At 11:26 PM +0530 12/1/12, Raman wrote:
>In my opinion, the definition of the sequences A046144, A046146
>seems appearing to be incorrect. According to the definition of
>primitive root - 'x' is a primitive root (mod n) if the order of 'x'
>(mod n) is being equal to phi(n).
>
>So, if 'x' is a primitive root (mod n), then so is 'x+n', 'x+2n', 'x+kn',
>etc.
>So, if there is one primitive root of n (if n is being a member of the
>sequence
>A033948), then there are infinitely many, in this way. Otherwise there are
>none,
>(n belongs to the sequence A033949).
>
>In my opinion, the name of the sequence A046144 should be
>Number of primitive roots of n, <= n.
>OR
>Number of primitive roots (mod n).
>
>Similarly the definition of the sequence A046146 need to be
>Largest primitive root of n less than n, or 0 if no root exists.
>OR
>Largest primitive root (mod n), or 0 if no root exists.
>
>If everything is being in doubt, you can be able to clearly verify it with
>the
>sequence A071894 clearly stating
>Largest positive primitive root (< p) of the n-th prime p.
>
>Please avoid putting me into long waits, but be keeping on with doing from
>now onwards with whatever is
>being good enough for the OEIS with these corrections things, stuffs in
>these process !!!
>
>_______________________________________________
>
>Seqfan Mailing list - http://list.seqfan.eu/

```