Earlier this was a section; I made it into a subsection,
since it is short. -AmodWe present two sets of examples in which the Manin constant
is not  .
.
I rewrote this subsection, and kept a copy of William's
original version after it. Feel free to pick the one you like. 
-AmodUsing results of [Kil02], Adam Joyce [Joy05] proves 
that there is a new optimal quotient of  with Manin
constant
 with Manin
constant  .
Joyce's methods also produce examples with Manin
constant
.
Joyce's methods also produce examples with Manin
constant  at levels
 at levels  and
 and  .
For the convenience of the reader, we breifly discuss his
example at level
.
For the convenience of the reader, we breifly discuss his
example at level  . 
There are exactly two
elliptic curves
. 
There are exactly two
elliptic curves  and
 and  of prime conductor
 of prime conductor  , and
, and 
 as subvarieties of
 as subvarieties of  , so
, so 
 is an
optimal quotient of
 is an
optimal quotient of  attached to a saturated ideal
 attached to a saturated ideal  .  If
.  If
 is the newform corresponding to
 is the newform corresponding to  , then one finds
that
, then one finds
that 
 , and so
, and so 
![$ g = (f_1 - f_2)/2 \in S_2({\bf {Z}})[I]$](img215.png) . However
. However  is not
in the image of
 is not
in the image of 
 . Thus the Manin constant
of
. Thus the Manin constant
of  is divisible by
 is divisible by  .
.
As another class of examples, one
finds by computation for each prime 
 that
that  does not leave
 does not leave 
 stable.
Theorem 3.5 (with
 stable.
Theorem 3.5 (with  ) 
then implies that the Manin constant of
) 
then implies that the Manin constant of
 is divisible by
 is divisible by  for these values of
 for these values of  .
.
William Stein 2006-06-25