Most manuscripts (followed by most editors) place Persius’ fourteen choliambic lines before the six hexameter satires, but two of the three best (AB = α) place them after all six, while the third (P) omits them entirely. The lines themselves fall into two halves of seven lines each, with a very abrupt transition. Some have therefore argued that they are two separate poems, though this opinion is now out of fashion.(1) A bit of ring-composition – the first and last lines both contain allusions to Pegasus – makes it hard to separate them.
There is one other possible arrangement I think worth outlining, since, so far as I can tell, no one else has suggested it before. (If I am mistaken, please correct me in the comments.) I don’t believe it myself, but hope that it may prove a stimulus to further thought.
Is it conceivable that the choliambs are in fact Persius’ prologue and epilogue, that lines 1-7 should be placed before Satire 1 and lines 8-14 after Satire 6? That would explain their uncertain position in the manuscripts: faced with two bits of verse in the same non-hexameter meter, a scribe might well have been tempted to combine them in one place, and either end of the corpus would have done as well as the other. The lack of connection between lines 7 and 8 would obviously not be a problem if they originally had 660+ hexameters between them, and the Pegasean ring-composition would work just as well as a device to unite an entire book, rather than a single short poem.
The first half without the second would arguably work as a lead-in to Satire 1. It could almost be printed with a colon at the end: “I’m bringing a poetic offering to the Holy Rites:” . . . and here it is, in a new paragraph and a different meter. So far, so good. However, the second half doesn’t work for me at all as an epilogue. It starts out as an interesting new poem, but whether we punctuate the last line with a full stop (editors) or with a question mark (as Harvey suggests, ad loc.), it doesn’t seem to provide anything in the way of closure. That is why I do not believe my hypothesis. I could modify (and complicate) it by arguing that something is missing at the end of 8-14, but that would ruin the matching lengths of the hypothetical Prologue and Epilogue, so I’m not willing to go that far.
(1) Dessen (16 and n4) attributes it to F. Leo, “Zum text des Persius und Iuvenal”, Hermes 45 (1910) 48. I will look it up next time I’m in a research library, since Google Books refuses to tell me whether they have even scanned it, much less whether I can read it, no matter how carefully I tailor my search terms.