3
$\begingroup$

The paper An update on Hurwitz groups by Marston Conder seems to suggest that the Chevalley group $G(2,5)$ of order $5859000000$ is a quotient of $G := \langle a, b \ | \ a^2, b^3, (ab)^7, [a,b]^{10} \rangle$. Can we check this, and if it is true, will it help with finding more quotients of the group?

$\endgroup$

2 Answers 2

5
$\begingroup$

Yes, $G(2,5)$ is a quotient of your group $G$. -- We can find generators $a$ and $b$ of $G(2,5)$ satisfying the relations in a few seconds with GAP (so in particular there is no need to buy Magma for this!):

gap> G25 := SimpleGroup("G2(5)");
G2(5)
gap> cl2 := G25.1^G25;; # conjugacy class of involutions
gap> cl3 := G25.2^G25;; # one of the 2 conjugacy classes of elements of order 3
gap> repeat             # find the desired generators 
>      a := Random(cl2); b := Random(cl3);
>    until Order(a*b) = 7 and Order(Comm(a,b)) = 10 and G25 = Group(a,b);
gap> Order(a);
2
gap> Order(b);
3
gap> Order(a*b);
7
gap> Order(Comm(a,b));
10
$\endgroup$
1
  • 1
    $\begingroup$ You could also do without charge with the Magma calculator! $\endgroup$
    – Derek Holt
    Nov 21, 2013 at 11:58
3
$\begingroup$

Yes, it's not hard to check! Just choosing random elements of $G_2(5)$ of order 2 and 3 in Magma, I very quickly found two matrices that satisfy the relations and generate $G_2(q)$:

[4 0 0 0 0 0 0]
[0 4 0 0 0 0 0]
[0 0 1 0 0 0 0]
[0 0 0 1 0 0 0]
[0 0 0 0 1 0 0]
[0 0 0 0 0 4 0]
[0 0 0 0 0 0 4]

[0 0 4 3 1 0 0]
[3 1 4 3 1 3 1]
[4 3 3 1 2 1 2]
[1 2 0 4 1 0 0]
[3 3 0 4 1 0 2]
[1 3 3 4 4 0 4]
[1 4 3 1 1 1 2]

A good reference for these types of calculations is W. Plesken and D. Robertz, Representations, commutative algebra, and Hurwitz groups, J. Algebra 300 (2006), 223-247. They find all characteristic zero representations of the Hurwitz group up to degree 7.

They say there that there are two 7-dimensional representations of $\langle a,b \mid a^2=b^3=(ab)^7=[a,b]^{10}=1 \rangle$ defined over quadratic extensions of the number field ${\mathbb Q}(\sqrt{5})$. That means that there are finite 7-dimensional representations over fields of order $p$ or $p^2$ or $p^4$ for almost all primes $p$. I don't know exactly what the images would be - probably linear or unitary groups for most primes - I can try and find out.

But in any case, there are definitely infinitely many finite simple images of the group.

Added: I have managed to compute a presentation of $G_2(5)$ by introducing a third generator $z$ and two long relations involving $y$ and $z$. It won't fit your idea of a simple presentation, but that's a limitation of computer calculations! The subgroup $\langle y,z \rangle$ has order 744000 and index 7875, so I got the presentation by first getting one for the subgroup and then doing coset enumeration over the subgroup.

< x,y,z |  x^2, y^3, (x*y)^7, (x,y)^10, z=x*y*x*y^-1*x,
  y^-1*z*y*z*y^-1*z*y*z*y*z*y*z*y^-1*z*y^-1*z*y*z*y^-1*z*y^-1*z*y^-1*z*y*z*
    y*z*y*z*y^-1*z*y^-1*z*y^-1*z*y*z*y^-1*z*y*z*y^-1*z* y*z*y^-1*z,
  y^-1*z*y*z*y^-1*z*y*z*y^-1*z*y*z*y^-1*z*y*z*y^-1*z*y*z*y^-1*z*y^-1*z*y^-1*
    z*y^-1*z*y*z*y*z*y^-1*z*y*z*y^-1*z*y*z*y^-1*z*y*z*y^-1*z*y^-1*z*y*z*y*z*
    y*z*y*z >;

Added: I have calculated explicitly the two 7-dimensional representations over number fields now, and tried reducing modulo some finite prime powers. The finite images all seem to be $G_2(p^e)$ for $e \le 4$. As well as $G_2(5)$,I've found $G_2(3^4)$, $G_2(7^2)$, $G_2(11^2)$, $G_2(13^4)$.

$\endgroup$
4
  • $\begingroup$ Is there a simple presentation of the group as a quotient of G? $\endgroup$
    – Thomas
    Nov 21, 2013 at 10:57
  • $\begingroup$ I don't know. The problem is that the group is too big to easily test computationally whether a candidate presentation is correct. $\endgroup$
    – Derek Holt
    Nov 21, 2013 at 11:28
  • $\begingroup$ So, you're saying that G(2,49), G(2,81), G(2,121), and G(2,28561) are all quotients of G? $\endgroup$
    – Thomas
    Nov 22, 2013 at 1:48
  • $\begingroup$ Yes, that's right. It looks as though $G(2,p^k)$ is a quotient of $G$ $k \le 4$ for all primes, possibly with one or two of exceptions (I don't think it works for $p=2$). $\endgroup$
    – Derek Holt
    Nov 22, 2013 at 8:13

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.