Martin (1992: 537):
Barthes's sequence, which is equivalent to the notion of activity sequence used here, is defined as follows (his nuclei are roughly equivalent to the clause rank nuclear structures proposed in Chapter 5):
A sequence is a logical succession of nuclei bound together by a relation of solidarity (in the Hjelmslevian sense of double implication: two terms presuppose one another): the sequence opens when one of its terms has no solidary antecedent and closes when another of its terms has no consequent. To take another deliberately trivial example, the different functions order a drink, obtain it, drink it, pay for it, constitute an obviously closed sequence, it being impossible to put anything before the order or after the payment without moving out of the homogeneous group 'Having a drink' (Barthes 1977: 101).
 In terms of SFL theory, Barthes' notion of 'sequence' corresponds to a sequence of figures that lack cohesion with the surrounding co-text. However, Martin's notion of 'activity sequence' corresponds to a sequence of figures without regard to (non-conjunctive) cohesion. Further, in Martin's model, such semantic sequences are misconstrued as context (culture-as-semiotic), which, in turn, is misconstrued as register (subpotential of language). This does, however, raise the question as to whether Barthes' 'sequence' is the unacknowledged source of Martin's 'activity sequence'.
 In terms of SFL theory, Barthes' notion of 'nuclei' corresponds to semantic figures. However, in Martin's model, such semantic figures are misconstrued as 'clause rank structures' (lexicogrammar), which, in turn, are construed as (discourse) semantics. This does, however, raise the question as to whether Barthes' 'nuclei' are the unacknowledged source of Martin's 'nuclear structures'.
 It is worth pointing out that, as stated, Barthes' notion of 'sequence' does not survive close scrutiny. For example, if it only requires one of its terms to have no antecedent or consequent, then the integrity of a sequence is "only as strong as its weakest link". Further the claim that it is 'impossible to put anything before the order or after the payment without moving out of the homogeneous group Having a drink' is clearly false, since the sequence could be preceded by offering to buy for friends, and followed by thanking the bartender, to name just two possibilities.