Let $D$ be a definite quaternion algebra over $\mathbb{Q}$ ramified at $p$ and $\infty$ for simplicity. Then an automorphic form on $D^{x}/F^{x}$ has several different interpretations. First, through the conjugation action of $D^{x}$ on totally imaginary quaternions, this group is isomorphic to $O(3)$ split away from $p$ and $\infty$. Secondly, through the theory of the class group, a level $\mathfrak{R}$ modular form is associated to a function on the genus of an Eichler order of level $\mathfrak{R}$. Finally, Jacquet-Langlands tells us it is an ordinary modular form, via the trace formula.

Now using the theta series for the Eichler order, we can compute what turns out to be a classical modular form. Does this modular form agree with the one Jacquet-Langlands would give us? It seems to me the answer is yes by strong multiplicity one. Furthermore, if we take the form on $O(3)$, we can take its theta series to get a weight $3/2$ modular form, and apply the Shimura correspondence to get a weight $2$ modular form. (Although Jacquet-Langlands is used in the proof by Waldspurger of this fact, and we do have a nontrivial kernel for some of these maps). Is this the same as the one given by theta series on the Eichler order? I think the answer is no, because many sources say so, but I've not found or constructed a good example yet.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.