Petronius' Satyrica 119.11-12, in Konrad Müller's Teubner edition (1995), reads:
hinc Numidae †accusatius†, illinc nova vellera Seres,
atque Arabum populus sua despoliaverat arva.
What reasons could the editor have had, to read accusatius here? To me, that does not seem to work for two reasons:
Firstly, it breaks the metre: de bello civile is in dactylic hexameters. If we read accusatius, the line scans:
Hinc Numi | dae -ccū | sātius | illinc | nōva | vellera | Sēres
That makes seven feet. The only way to fit this line in a dactylic hexameter, is by ignoring the length by nature of the ū in accūsatius, and by making the first syllable in illinc short (which does have the advantage of restoring the short o in nova):
Hinc Numi | dae -ccusa | tīus / ill | inc nova | vellera | Sēres
Secondly, I do not see how "accusatius" forms a coherent sentence. These lines list things stolen by, or forcibly given to, the Romans, as do the verses before them. The Seres, known for their silk, yield nova vellera and the Arabum (gen. pl. of Arabs) populus yields the spoils of sua arva. But what comes from the Numidae?
We can read accusatius as a perfect passive participle of accuso, in the neuter comparative nom/acc/voc, which would mean the Numidae yielded some single thing which was accused more than some other thing. Alternatively, Morpho suggests it could be a comparative adverb, which makes even less sense.
The critical apparatus has this to say about these lines:
accusatius O: accusant L: crustas Scaliger
L and O are two different collections of manuscripts, from which Müller constructed most of the text of his edition.
Reading crustās instead of accusatius, as the Loeb edition (Gareth Schmeling, 2020) does, perfectly fixes the metre:
hinc numi | dae crus | tās / ill | inc nova | vellera | Sēres
Furthermore, it perfectly fits the sense of the passage. L&S list "inlaid, chased, or embossed work on walls or vessels" under crusta and the Romans knew of marmor numidicum. North Africa, then, would have yielded marble, alongside silk from China and grain from Arabia.
The only argument in favour of accusatius seems to be lectio difficilior, but this lectio is so difficilis that it seems vix Latina.
What reasons could Müller have had to read accusatius here, instead of crustas, or what steps can I take to find an answer?