Does anyone know why in IFG4 beta clauses at initial position are not considered marked Themes any longer (cf. IFG2 and IFG3)? I don't seem to find any reference to the previous analysis there. I've checked both the chapter on Theme and the chapter on clause complexes.
Perhaps as beta clauses are not elements of the clause, they are no longer considered thematic in the alpha clause structure.
 In IFG, beta clauses in regressive sequences have never been 'considered thematic in the alpha clause structure'. The beta clause is thematic in the clause nexus.
 The beta clause in a regressive sequence is neither marked nor unmarked Theme. The features unmarked or marked Theme — not both — are systemic options that are realised in clause structure. The misconstrual of beta clauses in regressive sequences as marked Themes can be traced to Martin (1992: 445).
 In IFG4, the discussion of Theme in clause complexes occurs in section 7.6 The Clause Complex as Textual Domain (pp549-53). That is, in IFG4 the discussion has been made more prominent by being given its own separate section — in IFG3 it was included in the section 7.3 TAXIS: parataxis and hypotaxis (pp383-95), and in IFG2 it was included in the section 3.6 Clauses as Themes (pp54-8).