1: \begin{abstract}

2:    The number $\mc{F}(h)$ of imaginary quadratic fields with class number $h$ is of classical interest:  Gauss' class number problem asks for a determination of those fields counted by $\mc{F}(h)$.  

3: The unconditional computation of $\mc{F}(h)$ for $h \leq 100$ was completed by M. Watkins, using ideas of Goldfeld and Gross-Zagier; Soundararajan has more recently made conjectures about the order of magnitude of $\mc{F}(h)$ as $h \rightarrow \infty$ and determined its average order.

4:   In the present paper, we refine Soundararajan's conjecture to a conjectural asymptotic formula and also

5:   consider the subtler problem of determining the number $\mc{F}(G)$ of imaginary

6:   quadratic fields with class group isomorphic to a given finite abelian group $G$.  

7:   Using Watkins' tables, one can show that some abelian groups do {\em

8:     not} occur as the class group of any imaginary quadratic field

9:   (for instance $(\mbz/3\mbz)^3$ does not).  This observation is

10:   explained in part by the Cohen-Lenstra heuristics, which have

11:   often been used to study the distribution of the \emph{$p$-part}

12:   of an imaginary quadratic class group.  We combine

13:   heuristics of Cohen-Lenstra together with our refinement of Soundararajan's conjecture to make precise predictions about the

14:   asymptotic nature of the \emph{entire} imaginary quadratic

15:   class group, in particular addressing the above-mentioned phenomenon of ``missing''

16:   class groups, for the case of $p$-groups as $p$ tends to infinity.  Furthermore, conditionally on the Generalized Riemann Hypothesis, we extend Watkins' data, tabulating $\mc{F}(h)$ for odd $h \leq 10^6$ and $\mc{F}(G)$ for $G$ a $p$-group of odd order with $|G| \leq 10^6$.  The numerical evidence matches quite well with our conjectures.

17: \end{abstract}

18: