I'm studying the Jacobi inversion problem. Could you help me with the next question?
Let us consider an algebraic curve $C$ and it's Jacobian $J(C)$. For a initial point $P_0$ we can construct the Abel map $\mathcal{A}\colon C \mapsto J(C)$.
We denote by $K$ the vector of Riemann constants of $\mathcal{A}$. These constants depend on the choice of initial point.
One can define the Theta-divisor $\Theta$ in $J(C)$ as the set of zeros of $\theta(z-K)$. Here $\theta(z)$ is theta-function of $C$.
Is it true that the image $\mathcal{A}(D)$ of any special (not general) divisor $D$ belongs to $\Theta$?
Thanks.
There is a relationship between $\Theta$-divisors and global sections of line bundles on $Pic_{g-1}(C)$. But the relationship is not obvious and may not be what you wanted.
Let $C$ be a Riemann Surface, let $Pic_{g-1}(C)$ be the Picard variety of line bundles of degree $g-1$ on $C$. Let $E$ be a divisor of degree $r+g-1$. Then $O_{X}(D)\in Pic_{g-1}(X)$ can be viewed as $O_{X}(E-P_1\cdots -P_r)$. Therefore by fixing $E$, we can define a map $$ \phi: X^{r}\rightarrow Pic_{g-1}(X)\cong Jac(X) $$ And the pull-back of $O_{J}(-\Theta)$ via $\phi$ is precisely the line bundle whose fibre defines the so-called "Faltings volume" of $\det (O_{X}(D))$. A rather dense proof can be found in Lang's book, Chapter V.