Let $G$ be a split semi-simple simply connected group over a global field $F$ and let $\omega$ be a top-degree differential form on $G$ without zeroes (defined over $F$). It is well known that $\omega$ defines a measure on the adele group $G(\mathbb A)$. The Tamagawa number formula states (if I understand correctly) that

1) If $F$ is a number field then the volume of $G(\mathbb A)/G(F)$ is 1

2) If $F$ is a functional field isomorphic to $\mathbb F_q(X)$ where $X$ is a projective curve over $\mathbb F_q$ then the above volume is equal to $q^{(g-1)\dim G}$ where $g$ is the genus of $X$.

My questions are the following:

a) Do I understand the statements correctly?

b) What is the reason why 1) and 2) look somewhat differently? Can one formulate the statement in a uniform way for all global fields?

Edit: In fact my understanding was wrong. In 1) one needs to multiply by the volume of $(\mathbb A/F)^{\dim G}$ which is equal exactly to $q^{(g-1)\dim G}$ in the functional case. I was confused by the case $F=\mathbb Q$ where the above factor is 1.

  • Side remark. Tamagawa number and susy: projecteuclid.org/euclid.cmp/1104201160 Quantum-mechanical calculations in the algebraic group theory M. A. Olshanetsky – Alexander Chervov Jan 31 '12 at 20:57
  • I think the Tamagawa number is still supposed to be 1 in the function field case. (Harder proved this for simply-connected Chevalley groups in "Chevalley groups over function fields and automorphic forms". See the last line in the paper.) Perhaps there are competing normalizations? – B R Jan 31 '12 at 22:30
  • Further, in Borel and Prasad's "Finiteness Theorems for Discrete Subgroups of Bounded Covolume in Semi-simple Groups" (1989), there is this line (at the end of Section 7.1): "The Tamagawa number of any simply connected group of inner type $A$ over an arbitrary global function field is known to be 1 (see [Weil's Adeles and Algebraic Groups]). However, whether this is the case in general over a global function field is not yet known." – B R Jan 31 '12 at 22:45
  • 1
    In the number field case this was proved using trace formula techniques by Kottwitz, following work of K.F.Lai and Langlands (see MR0942522). In the function field case, I believe the full result was recently (2011?) announced by Jacob Lurie. – fherzig Jan 31 '12 at 23:05
  • 1
    Actually, it's Gaitsgory-Lurie: web.me.com/teichner/Math/FRG-3.html – fherzig Jan 31 '12 at 23:09

Your Answer


By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Browse other questions tagged or ask your own question.