Is string theory a theory of strings?

Recently a great deal of evidence has been found indicating that type IIA string theory compactified on K3 is equivalent to heterotic string theory compactified on T4. Under the transformation which relates the two theories, the roles of fundamental and solitonic string solutions are interchanged. We show that there exists a solitonic membrane solution of the heterotic string theory which becomes a singular solution of the type IIA theory, and should therefore be interpreted as a fundamental membrane in the latter theory. We speculate upon the implications that the complete type IIA theory is a theory of membranes, as well as strings.