Neither typescript draft is a carbon, and there is no direct relationship between them. MS.6642 appears to be the earlier: the names of the characters differ from those given in MS.6643 and in the printed version as found in ‘The definitive edition of the plays of J M Barrie', page 761, so that 'Sir James Ings' is equivalent to 'Sir Harry Sims', 'Lady Ings' to 'Lady Sims', 'Marion' to 'Kate', and 'Findlater' to 'Tombes'. There must have been at least one corrected version beween MS.6642 and MS.6643 (possibly a carbon duplicate of MS.6642), since, as well as occasional corrections in MS.6642 (which has very few alterations) appearing as typescript in MS.6643, there are differences in the typescript of MS.6643 not indicated by any revision in MS.6642. Some corrections appear as such in both drafts, so that if a carbon of MS.6642 was the typist's copy, Barrie must have kept MS.6642, and worked on it at a later stage, as was sometimes his habit (cf. notes to MSS.6625-6635, 6636-6641).