The Whitehead group of a group G, Wh(G), is a functor from the category
of (finitely presented?) groups to the category of Abelian groups. A
reference is _A Course in Simple Homotopy Theory_ by Cohen.

If I have a short exact sequence of finitely presented groups 1 -> K -> G
-> Q -> 1, and I take the Whitehead torsion of the sequence, forming Wh
(K), Wh(G), and Wh(Q), when is it true that Wh(G) -> Wh(Q) is onto? It it
true if G is a semi-direct product of Q and K?

In any case when Wh(G) -> WH(Q) is onto, are any results known about the
kernel of Wh(G) -> Wh(Q)? Are there any nice situations when the kernel
is Wh(K)?

Thank you in advance for any assistance you can provide.