No join can be established between these two fragments. The larger (a) contains the remains of six lines of an account or a list written across the grain. The smaller (b) has exiguous remains of two lines which may well be in the same hand and might supply the end of the account. The list appears to contain foodstuffs and may concern the domestic organisation of the praetorium rather than the military unit as a whole.
_mandat[
.a...[
lent.s [
ligust..[
amuli [
]...[
. . .
. . .
]uri
uacat
]mma [
. . .
"Items ordered (?):
beans (?)
lentils
lovage
meal
...
butter (?)
total ... (?)"
The reading does not present any serious palaeographical difficulty and is suitable for the heading of an account or shopping-list, cf. [512].1. A possible alternative reading would be ama..a.[, which would suggest a plus the name of the supplier of the goods (cf. [192].1, [207].1).
The traces are very abraded and only a is clear. We think it possible to read fabae (for the collective singular cf. [192].3, [302].1) which fits the context very well (see line 3 and note and for beans and lentils in the military diet Davies (1989), 199), but the reading is highly conjectural.
There is a mark before l, but we take this as belonging to the line above. The penultimate letter is most easily read as a but lentas seems unlikely; we think it possible to read lentis (lentes is less likely), which we would regard as a collective singular (cf. OLD, s.v.1 and the references for fabae in the previous note) and assume that it was followed by a quantity or a sum of money. For lentils at military sites in Britain and Germany see Dickson (1989) and cf. Davies (1989), 199.
The traces are compatible with ligustic[. On ligusticum see Pliny NH 19.165, 20.187; in the latter passage he notes that it is an ingredient in eye-salves which were perhaps in some demand at Vindolanda (see [154].24 note). For a culinary use see Apicius 7.4 where it appears with amulum (below, line 5) in a recipe for cooking ofella (for which see [203].2).
The last three letters are certain and m is probable. The trace of the first letter is compatible with a.
The form of r is very distinctive. ]uri could well be the end of but]uri, for which see Dannell and Wild (1987), 69.
su]mma [ looks an attractive restoration and suggests that this might be the end of a cash account, with the sums of money for individual items lost at the right.