Tab. Vindol. I and Tab. Vindol. II Addenda and Corrigenda

(from Appendix, A. K. Bowman and J. D. Thomas Tab.Vindol. III, British Museum Press 2003, 155-61)

The following notes correct and improve readings of texts published in Tab.Vindol. I and II. Most of the improvements are attributable to the superior facsimiles of the tablets which have been obtained by digital scanning and processing of the images. We here confine ourselves to what may be regarded as substantive changes. There are a significant number of minor changes as well as a number of texts which we think can be substantially improved with a good deal more work. It is our intention to present such further changes and improvements on the website. A bibliography is provided at the end of the document.

Individual texts

Renuntia 127 129 137 138 140 142 154 155 156 161 176 177 179 182 184 185 192 193 196 200 206 211 225 238 239 242 243 247 250 255 256 257 258 259 263 264 266 268 271 274 277 281 282 283 284 288 291 292 293 295 296 300 301 304 306 310 312 321 343 344 349 352 353 364 370 392 404 414 418 427 440 444 452 453 467 477 482 484 499 505 508 510 528 545 546 Bibliography

Renuntia (Nos. 127-153, see also notes on 453 and 545below).

We now accept that in all reports of this type we should read or restore qui debunt rather than q(ui) uidebunt, see Adams (1995a), 102-3 and cf. 628.ii.15. The word division in the one complete example (574.4-5) also clearly supports this.

127 . Tab.Vindol.I 12. Line 1: ]ias uacat can be read with confidence and there is room for a short date such as v K(alendas) Ma]ias. Line 2: there is a stroke over the numeral. Line 4: the descender of q is clearly visible. It is doubtful whether there is any ink after inpedimenta. Line 6: clearly ren]untiarunt [sic], and there is no uacat at the start. Line 9: read (centuriae) Exomni; the same century occurs in 575.5; see also 182.ii.13, 581.30 note.

129 . There are clear traces of a preceding line, presumably renuntiauit/- erunt etc. Line 2: detulit is clear, followed by a.....tius, suggesting Arcuittius (as in 574.8 and 128 ). Line 3: the centurial sign is clear. If the name Arquittius is to be read in line 2, we would expect Crescentis, as in 574 and 128; cres fits the first traces, but what follows is indeterminate.

137 . Line 1: read Idibus [. Line 6: perhaps (centuriae) ? V[.

138 . There are traces of a preceding line. Line 2: this could be ]..i, the end of an ablative name (perhaps ­ati or ­aci) preceded by excepto (cf. 575.4-5). Line 3: read ].. ? optio.

140 . Line 1: the reading ]us would of course allow Septembrib]us etc. but we now think that ].is is a more probable reading.

142 . Line 1: traces. Line 2: omnes ? ad loca. Line 3: et impedime[nta, indented.

154 . We have no reason to doubt that the figures are of the right order, although the scans suggest possible minor adjustments (additions or subtractions of not more than one or two individuals in any entry) to some of them. But these are not certain enough for us to suggest emending the readings. Lines 1-2: the scan gives us significantly more confidence in the readings proposed in the ed.pr. Line 10: we can now read tra prouinciam at the right before the numeral and we think [e]xtra more likely than [i]ntra, cf. RMR 63.ii.23-4 (the Moesian pridianum). It may be preceded by a participle, but it is not easy to read uestitum (cf. RMR 63.ii.18 and note 255.3-5, referring to someone having returned from Gaul and approved some clothing). Alternatively, we might read in s- at the start, but something like in statione is hard to reconcile with the traces; it looks like in scir..[... e]xtra prouinciam. Line 12: we can now read ]ecti in Gallia and this suggests [prof]ecti in Gallia[m] or Gallia. Line 14: stipendiatum looks very good and seems to be a supine of purpose meaning "to collect pay", an exact parallel to the supines in RMR 63.ii.18- 20 and 32. Tomlin (1996), 461 suggests [Ebura]co stipendiatum; the scan would justify .b...c. and the traces are compatible with Eburaco. The excavations of 2001 produced a fragmentary text which appears to be another such report of the First Cohort of Tungrians under the command of Verecundus.

155 . Tab.Vindol.I 1. Line 1: read viii K(alendas) Maias [24 April] in officis etc. On officis for officiis see Adams (1995a), 92. The content would seem rather to require officinis (= "workshops", the sense conveyed by the original misreading fabricis), but this has definitely not been written. officium seems to have the same meaning as officina in CJ 11.12.1.2, ornamenta enim regia intra aulam meam fieri a palatinis artificibus debent, non passim in priuatis domibus uel officiis. Line 5: the reading in 155 was [a]d .ar.[, and we suggested  ad kar.[ in the note. This is not correct; we suggest serrari[(i), "saw-makers" or perhaps "sawmen", see  OLD s.v.; carrari[(i), "wagon-builders" (for vehicle construction see 309, 600), does not suit the traces at the beginning of the line so well. Line 6: probably s[t]ructor[es a]d ualetudinar[. Line 10: the traces suit ad capil.[, but not capill[ (cf. 596.i.7 and note). We could perhaps read capilu[ (for the form of capillus with a single l see TLL III 314.52). But we cannot explain the word in this context. Line 12: read .]. b (or l). Line 13: a new line, reading  ....[ ; renumber lines 13 and 14 as 14 and 15. Line 14: Perhaps ad u[.

156 . Line 1: read Martis. Line 2: read hospitium Marco medico, presumably " to build a guest-house for Marcus ". MLD s.v. cites 12th-century evidence for the word meaning "hospital" but there seems to be no precedent for this in earlier Latin; cf. Adams (1992a), 5. Lines 3-4: we might possibly have h(omines) before the number, rather than n(umero).

161 . Line 2: read Priuatus.

176 . Line 4: Coris is certainly correct, followed by ut possi[m]. Line 5: It is clear from the scan that the mark after what we read as fa is the top of a tall letter, b or l; therefore not fam[iliaribus or similar. We now think a better reading is: [ c.6] emere sal[em. At the start, ibidem  would be a suitable restoration.

177 . Line 2: Ru]fino should be discounted. ]etuo looks correct and could be the end of a name, e.g. Perp]etuo praefecto. Traces of another line below.

179 . Read cclxxiiii s(emis) (octans).

182 . Line 2: the symbols after xii are (octantem) (assem i). The octans symbol is normally a single horizontal tick (see 596 introduction) but this writer seems to use two ticks, one above the other, perhaps representing two asses (cf. 764). Line 3: the symbol at the end is the octans consisting of two horizontal strokes; so 8 and 16. Line 4 read:  xxxiix s(emissem) (octantem). Line 5: A.R.Birley (2002), 102 has suggested Troucisso instead of Ircucisso. This writer regularly leaves o open at the right and we agree that these are both possible readings. CIL III 4146 (Pannonia Superior) has the masculine name Troucissa (cited in AS III 1970). At the end of the line read (denarios) xiii (semissem) (octantem). Line 9: accipit, vi (octantem) (assem i). Line 15: at the end,  i.[. Back 1: ]..is. Line 6: Varia- looks good but Varianus is not long enough to account for all the traces. Thereafter a denarius symbol with at least i following.

184 . Line 3: in ecthesis. The name could be Tagarannis and it is followed by an erased centurial symbol. Line 14: (denarios) iii (quadrantem). Line 16: (denarios) i s(emissem) (quadrantem). Line 20: (denarios) v (quadrantem) (octantem). Line 24: traces of two more ascenders, further to the right. Line 27: he might have written Huetti[us for Vetti[us, or possibly [[li]] Vetti[us. Line 28: s is a correction from t. Line 30: l is very doubtful. Line 36: (denarios) ii (octantem). Line 38: probably just subarmal followed by a medial point. Line 39: (denarios) ii s(emissem)[. Line 40: perhaps just Aledus (NPEL cites Aledius and Alendus). Line 42: Perhaps there is room for Ver[ecundus. Possibly traces of indented writing above and below this line. Back: remains of 8 lines. Line 1: perhaps (centuria) .[…]lesi. Line 3: traces at the left; (denarios) ii (quadrantem) at the right. Line 4: tunic in larger letters, followed by a medial point, marking abbreviation; a subheading? Line 6: ].bu..is sagum; presumably a personal name before the noun. Line 7: ] traces sum[; perhaps sum[ma.

185 . Lines 2-4: The small piece placed here on the plate probably does not belong here. Line 2: ]ouis is likely, suggesting Vin]ouis. Line 3 read: ]...[ c.7 ] s(emissem) (quadrantem). Line 4: read ]...[ c.6 ]dum (or duis); perhaps a date followed by a place-name. Line 18: read (octantem). Line 21: Perhaps read raedm. Line 26: read subuncolonis. Lines 27-8: after frumenti we can read mo(dii) lx m(odius) i, then further traces. In the first instance the modius abbreviation is marked by an o above the m (cf. 586; the second example may be the same but if so the suspended o is damaged and might have fouled the top of m). This suggests a unit price and the reading of the sum of 78 3/4 denarii in line 28 is plausible. From this we might deduce that there should be a unit price of 1 1/4 1/16 denarii at the end of line 27, but we find it difficult to reconcile the visible traces with these numerals and symbols; the real difficulty lies in reading the denarius symbol before the digits; if it was omitted, either the i  is oddly made or there is dirt on the surface of the tablet. Line 30: at the end perhaps ]viii. Back: abraded traces of a single line written along the grain.

186 .7-8. The cost of 100 nails was read as 2 asses, but the scan renders this reading very uncertain; the numeral might be x with perhaps one or two digits following.

192 . Line 5: read p(ondo) i a(sses) xi s(emissem) f(iunt).

193 . Line 4: read (octantem) (assem i).

196 .back 1-2. cerui[ / membra n(umero) [:  we should perhaps restore cerui[nae], see 587.1 note.

Back 3: read catacysem rather than catacysen.

200 . Line 9: the note is incorrect; it could be iii Nonas and Non(as) looks like a possible reading.

206 . Front: there may be some deletions. Line 1 read: ]. (denarios) ii  s(emissem) (octantem). Line 7: read (denarium) i (quadrantem). Back 6: read either ]xi or ] (denarium) i. Margin 1: fenicio is possible but perhaps better Senecio (or Senicio, which we would understand as a form of the same name), cf. 609.9. See Tomlin (1996), 462. Line 2: relicum looks a good reading.

211 . Line 1: cum looks possible, suggesting cum pr[imum, or cum plus a personal name. Line 2: the correct reading at the beginning of the line seems to be ad medias, followed by an interpunct. There is evidence for the use of Mediae  as a place-name, although not in this part of the world (AS II, 495, citing Itin. Burdig.). We wonder whether it might be ad medias Lugu[ualii (sc. partes), cf. perhaps OLD s.v. medius 2. Line 4: at the right we might again have ad me[di]a[s.

225 . Tab.Vindol.I 37. Line 1: Tomlin (1996), 462 suggests that Crispinus was surely a senator.

238 . Line 4: almost certainly explicata esset followed by a uacat.

239 . Line 1: after Flauius Cerialis read ..r[....] s[uo]´. Back: remains of an address on the back of the right-hand side. This is unusual in a letter written by Cerialis, so it must be either a fair copy left unsent, or something which was sent to someone outposted, who brought it back to Vindolanda with him (cf. 621 and 242, with the note below). Read ].rio´ or ].no´.

242 . Line ii.2: in view of 293.1 (see note below), we now think that this line begins with uale mi for uale mihi. What follows is not likely to be a personal name: perhaps feliciter, with frater restored after it, for which there would just be enough space. Line ii.4: at the start of the line, remains of about 8 letters, of which the first is certainly s; at the end we can see septembr, with traces of at least one more letter. Although the space is quite tight, we suggest that we might have scriptum c.3-5 K(alendis/-as) Septembri[bus or Septembre[s; after scriptum possibly something such as a(nte) d(iem) xiv (cf. note on 271.A.1 and 295, below), but we emphasise that this is very speculative. Back 2: read a Ceriale praef(ecto).

243 . For the position and shape of the notches, compare 645. Line 1: the name is spelt Cereali, cf. 261.1.

247 . Tab.Vindol.I 25. Back 2: this could  be read a B[ro]ccho, followed by traces which ought to be coll(ega).

250 . Tab.Vindol.I 22. Line i.2: there is no sign of su]o. Line i.5: commendarem corrected from commendaret. Line i.8: the name should be read as Questori. LC cites one occurrence, CIL VIII 2554. Line ii.11: is corrected from e. Back: there are probably traces of the name of the sender in a third line at the bottom left.

255 Line 1: ]odius: o is certain. On Super see 629 introd. Lines 10-14: the suggestions made by Tomlin (1996), 463 are ingenious and persuasive and we accept them, with minor adjustments. Read: tu[nicas] q[uin]/que. scis certe hoc me uor-/sute impetrare cum sim an-/nonarius et iam adepturus / translationem; "you know that I am smart in getting hold of this since I am the commissariat officer and am now on the point of acquiring transport". In lines 11-12 Tomlin suggested u[er]/sute, but there are traces visible at the end of 11 and the spelling uorsute is cited by OLD s.v. In line 13 the spacing and perhaps the traces favour the future participle over Tomlin's adep[tus]. We do not think that the reading of A.R.Birley (2002), 101, cum sim / nonanus, can be correct. Line 16: the traces are compatible with desideratis-/sime.

256 . Line 1: A.R.Birley (2002), 128 suggests that the sender"s name may have been Gentilis rather than Genialis. It is true, as he says, that the letters between Gen and lis are not entirely clear, but we are satisfied that Gentilis cannot be read and that Genialis  is a reasonably secure reading. Lines i.3-5: as a tentative restoration and interpretation we suggest: ex con[pluri]b[u]s reos quod penes te / rem[issio est] facio tibi aut / Genito[ri se]ptem, "I am indicting out of a larger number seven men before you or Genitor because it is in your power to release them". At the end the traces suit tem better than cem (d]ecem) and septem better suits the size of the gap, though still a little short. Line ii.2: read et rather than eo. Line 5: there is a deletion at the right after domine. Line 6: perhaps the tops of letters from frater] karissi[me.

257 . Lines 3-4: we are now confident that the small fragment to which we referred in the note reads ].ere [ and does not belong in this line (there is a blank space after re suggesting it may form the end of a line below line 6). We find the reading suggested by Tomlin (1996), 462 attractive and would now accept this, reading: domin[e per [pos-/teritat[e]m tuam. This in turn would guarantee that nothing has been lost at the right in lines 4-5.

258 . Lines 1-3: Tomlin (1996), 462 suggests: [NN et ...-]/.um ut conten[tiore]s / praestarem. Back 3-4: read: a Floro dec(urione) alae / Petrianae.

259 . Line 3: notam is incorrect. Read tiliam; this is interesting as the first clear evidence for the term used to to refer to these ink tablets. It also occurs in 589.a.ii.1,  643.a.ii.4 and 707.2.

263 . Tab.Vindol.I 23. Lines ii.1-2: ]etuó might be the end of a name, e.g. Perpetuo, which is common (cf. note on 177 above). Thereafter we might tentatively restore com-/m[ iliton]e. Line 3: there might be an insertion above the line to follow quas. At the end acceperas looks preferable. Lines 5-6: si is in ecthesis and we took it to belong to the left-hand column (as m below it does); but the traces at the end of 5 suit mi so we should perhaps read mi-/si and restore [re]mi-/si (so CEL 91) or another suitable compound. Line 7: benefi looks plausible, perhaps a reference to a beneficiarius. Lines 8-9: perhaps dominam tuam a me salu-/ta but everything between dominam and ­ta is more restoration than reading. After ta, o looks clear and at the end of the line we probably have ualeas (or ualere); we could envisage opto (or perhaps optamus which would fill the space better) bene ualeas, or opto te bene ualere. Back 3-4: the correct reading is clearly a Vitale / dec(urione) alae / Aug(ustae). This is important as it appears to be the earliest clear attestation of this unit, found later at Chesters and Old Carlisle, see Jarrett (1994), 40. RIB 606, probably late first century, from Lancaster may refer to it; there is no indication of its whereabouts in 263.

264 . In line 4 correct quo to quod.

266 . In line 1 correct uolo to uolt (cf. 720.3).

268 . Line ii.4: probably resscri[ (cf. 645.ii.11), perhaps preceded by ext[emplo er-]/go or ro-]/go.

271 . Side A, line 1: read quae ante diem xviii K(alendas) Febr(uarias). We now reject the reading suggested for Side B, which is not necessarily an address, and there is thus no reason to connect this text with the Cerialis archive.

274 . Back: traces of Ceriali, below that a Broccho coll(ega), not slanting.

277 . The right-hand portion of a diptych with four lines of a letter. At the left, traces of the ends of lines from the left-hand column. Col.ii reads: traces / si mariti sumus non / ideo uobis alieni sumus / se.enam   traces . Line 4: perhaps the personal name Serenam, although the n would be made differently form that in line 1. The meaning is presumably something like " although we are married, that is no reason to cut ourselves off from you (plural)." Back: we now reject our earlier reading and there is thus no reason to connect the text with the Cerialis archive. Read: traces / alae Petrian(ae). Line 2: see note on 258 (above) and 281, 284 below). This appears to belong with the name of the addressee rather than the author. The evidence for the movement of units and parts of units is sufficient to account for an officer from this ala (which might have been stationed at Coria, see 611) receiving (and/or disposing of) a letter at Vindolanda. There is a mark after n which may or may not be ink; if it is, perhaps an abbreviation mark.

281 . Line 1: read bellos, probably the adjective (cf. 632), preceded by a noun ending with ­s. Back 2: read .[ c.2 ]. viiii Batauor(um); the last word must have been abbreviated unless it ran over the fold. The numeral has a superscript bar. Before the break at the left there appear to be traces of ink which, if they are part of the address, rule out praef(ecto). Possibly just a centurial symbol followed by [co]h, in which case the letter will not have been sent to Cerialis. Lines 3-4: read a Cluuio Floro / dec(urione) a]l(ae) Petri(anae). See notes on 258 and 277, above and cf. A.R.Birley (2001b), 15.

282 . Line i.4:  -nusquam or -amus quam. Line ii.7: at the end et. Back 1-2: read: Lucio dec( urioni) / coh(ortis) viiii Bat(auorum). This must surely be the same person as a the recipient of 299 and 300 and the text must therefore be removed from the Cerialis archive and reassigned to the small archive of Lucius the decurion.

283 . Tab.Vindol.I 40. Line 6: we now reject tabulas; read quod nisi nomen ..[. Back: read Fla]uio Ceriali (the cognomen  is certain on the scan).

284 . Back 2: after Vero, read dec(urione) alae; what follows is very hard to make out. We think it most likely to be Petrianae (no doubt abbreviated), see notes on 258, 277, 281 above; but we would not rule out Augustae (again, probably in an abbreviated form), for which see note above on 263.back 3-4.

288 . We can now state confidently that this is a letter from Brocchus to Cerialis; see 623introd.

291 . Line 9: there is no visible trace after meus on the scan; 635.1 strongly suggests that we should supply [te], despite the unexpected word order. Line 17: probably a Cl(audia) ? Seuerá.

292 . Line a.4: we now reject the reading corde, and we cannot improve on our earlier suggestion to suppose a scribal error before it, reading mihi ta. After it cor is certain and appears to be followed by ie or te before semp[er. After this we suggested li]citum, which we still think is possible though by no means certain. corte would be acceptable as a form of cohorte but we do not see how it would construe. corie too would be acceptable as for Coriae, but the locative of Coria is always Coris elsewhere in the tablets (but cf. 613.a.3 note on the locative of Vinouia). We have tried to read Coris in 292 but it is very difficult to read the last letter as s. If Coris or Cori(a)e is correct, the passage could be interpreted to mean that Brocchus was in effect replying negatively to Severa "No, you cannot travel to Vindolanda; you can always meet at Coria"; Severa then goes on to say that she will try to get to Vindolanda nevertheless. This suggests the possibility that Brocchus and Severa  might have been stationed at Coria, Brocchus perhaps as praefectus of the ala Petriana in succession to Haterius Nepos (see 611 introd. and 622 introd.), though we could not exclude the possibility that Coria was somewhere between Brocchus' base and Vindolanda. Line b.iii.4: for nobis read tibi. Line c.v.2: there is probably no ink to the left of ra. Line b.back 1: read ual]e mi soror, see 635.3-4 note.

293 . Line 1: read uale mi (no further traces), see 635.3-4 note.

295 . Tab.Vindol.I 30. Line 3: perhaps read Crispum et Pe[.... mili]tes coh(ortis). Line ii.8: we now think a reference to Lindum very unlikely. For the possibility of a date at the end cf. note on 242, above.

296 . Line 1: instead of Conianus read Fontanus (and cf. 325).

300 . Line i.1: At the beginning perhaps read ]..s Super, although it is not easy to read the letters before per, or conceivably cer, as Su (cf. note on back, below). The author of 255 and 629 is Clodius Super. Lines ii.9-10: read hoc ut scires scripsi / tibi (cf. notes on 306.3 and 4, below). Back 1: read Lucio decur(ioni). Line 2: the numeral viiii is fairly clear, with traces following, presumably Bat(auorum). There is no trace of coh preceding. Line 3: traces of writing on a slant, perhaps ...pero or cero, with perhaps some further traces. Possibly a Supero (see note on i.1, above).

301 . Line i.3. For the explanation of souxtum as a Celticised spelling of su(m)ptum see Adams (1996), 238. Back: read Candido Genialis / a Seuero / corniclarió. We can see no sign of praef(ecti) on the scan. There is a diagonal stroke at the end of corniclario which may be an apex; but it is equally possible that it marks an abbreviation and we should read it as corniclar(io). It is very interesting indeed that a soldier of this rank should write to a slave addressing him as frater.

304 . Line 2: instead of administret ? unam read administretur iam.

306 . As the address on the back shows, this belongs to the archive of Cerialis. Lines 1-2 perhaps suggest that the subject-matter is concerned with fighting. Line 1: the traces at the right can be read as pugna[, perhaps preceded by ]uit. Line 2: read ].nis  iacula[. Line 3: read ].s quod ut s...e.[ (perhaps scires[). Line 4: read ].i tibi m2 uale mi domine [; at the start perhaps scrip]si (for the phrase hoc ut scires scripsi tibi see note on 300.ii.9-10, above). Lines 5-6: read et frater caris-/sime. Back 1: Flauio Cer[iali. Line 2: pra[ef(ecto). Line 3: read a Suces[..]re. The name Successor is well attested and we suggest that we should read Suces[so]re here; there are traces following which might be c[o]l(lega).

310 . Line 17: the correct reading is probably Velbutenam, a female name, rather than Velbuteium.

312 . Line 7: there are more traces of numerals at the end after x, at least xiii. Line 11: perhaps read Viriccium, or Virucium. Viricius is attested. Back 2: we have no confidence in the reading of the cognomen as Nigrino, but we cannot suggest what it should be.

321 . Line 3 ends with uale and there is no line 4.

343 . Lines 40-1: coriatione quem: we now think that quem = que[m], with que =  quae n.pl.( a favourite misspelling of Octavius), and -m hypercorrect. Adams has reminded us of a hypercorrect -m in Eunus, see Adams (1995a), n.21 and that Octavius, has at least one hypercorrect spelling (mae = me). He points out that a plural seems essential with quinos and that coria  is near enough to be understood as the antecedent to quae. He adds that licere, meaning "have for sale" (cf. TLL VII.2 1357.71) needs a direct object; pro coriatione cannot serve as direct object, but would make a nice complement to the direct object que(m) = quae. He offers the translation " I hear that F.I. has for sale at a high price for leather- making (the things) which he bought here for 5 denarii apiece". This, as he points out, converts coriatio into a verbal noun, instead of a concrete noun, which is how it would have to be understood in the original interpretation.

344 . Line 7: the repositioning of a small fragment, if correct as it surely must be, makes it certain that the last letter before praefecto is e and makes the reading Procle almost certain. This is close to the reading Proc[ul]e which A.R.Birley (1990a), 18 originally suggested, i.e. the appeal is to a high-ranking official with the name Proc(u)lus. Subsequently Birley, accepting our reading prout, suggested that the text was a draft appeal to the emperor Hadrian ((2002), 116, cf. A.R.Birley (1997), 276). If Procle is correct, this Proc(u)lus may have been the provincial governor, and might indeed be the Flavius Proculus who turns up as correspondent of Genialis in 219, 444 (see note below) and 612, and is mentioned in 611.i.b.7; on this Proculus see A.R.Birley (2002), 125.

349 . Lines 6-7: read con-]/tubernales cum quibus opt[o. Margin: read bene ua]leas. ual(e). There is no second line in the margin. Back: A.R.Birley (2002), 116 and n.10 suggests that the letter may be from Ti. Claudius Fatalis. It is clear from the scan, however, that the ductus of the letter before io rules out d.

352 . Back 1: the reading of the cognomen as Veloci is not convincing: we now suggest Vegeto, cf. 614 and Tab.Vindol.II, pp.364-5. Line 2: there is an ascender under the first o of Cocceio, which would suit co]h but the traces before it do not suit praef. A stilus tablet from London published by Tomlin (2003) and assigned to the Domitianic/Trajanic period attests an imperial slave named Vegetus. If our re-reading of 352 is correct, we wonder whether M.CocceiusVegetus might be the same man, manumitted by Nerva (the Vegetus of 614 will probably be a different person). This suggests the further possibility that the addressee of 645, Cocceius Maritimus (for the gentilicium see note on Back 25), might also be a freedman of Nerva, especially since Caesariani are mentioned in lines 8-9). 

353 . Back: we had no photograph of this when the edition was prepared. The scan reveals the name of the sender which can be read as a Gnauorige mag(istro) / [co]hortis. The name Gnauorix is not attested, but seems plausible as a Gaulish or Germanic name. For the magister cohortis see Domaszewski (1967), 59, "Exerziermeister der equites sagittarii dieser Cohorte" (sc. cohors miliaria Hemesenorum, CIL III 10307)).

364 . Line 1: possible traces of bottoms of letters. Line 2: read simum hibernum (for the substantival use of hibernum as "winter" see LS s.v. II). Line 3: read: m2 opto frater felicis[- . Back: traces of 1 or 2 lines, presumably the name of the sender of the letter.

370 . Line 1: we are now confident in reading ] Felici fr[atri.

392 . We now read s]tabulo before hiber and assume that the second word qualifies the first.

404 . Line 1: read: habeas. Back: Simili suits the traces well (cf. 235 for a letter to Similis).

414 . Line 2: read: qua die profic[. Back 1-2: we read ab Attic[o / corniclario; the same person occurs in 586.i.4.

418 . Presumably the beginning of a letter. Read Primigenius [ (and see 347).

427 . Read ]fectos, e.g. re]fectos instead of ficias.

440 . The first two lines are deleted. Line 1 ends with ferrum. Line 2: possibly Veldedius fab[; there is some ink to the left of the name, possibly a check-mark of some kind. For the unusual name Veldedeius see 310.back 2; since we know from 310 that Veldedeius was an equisio consularis its seems unlikely that fab[ should be restored as fab[er (which occurs in 160); since other commodities occur in this text it is more likely to be some case of faba. Line 3: read sudari[um, cf. 184.10 etc.

444 . Back 1: the first letter is certainly f and the traces thereafter do not exclude Flauio Ge[niali. The hand which wrote the text on the front could well be the same as that of 219. Line 2: the scan confirms a Proculó (see 219, 611-612, note on 344.7, above).

452 . This fragment has the first line of Vergil's Aeneid I, on both front and back in two different, but both good, hands. This strongly suggests writing exercises. Compare 118. For an example of Aeneid I.1-3 in a military context see O.Claud. 1.190, with citation of other occurrences, and cf. Horsfall (1984), Hoogma (1959). Side A: arma] uirumque [c]anó Tro-/ iae qu]i  primus ab oris. Side B: arma [ui]r[umque cano / Troiae qui primu[s ab oris.

453 . Two fragments preserving parts of 2 lines of a renuntium. Line 1: coh(ortis) [viiii] Batauorum. Line 2: omnes] qui debunt [.

467 . Read Flauius Cerialis.

477 . Read: traces per- /   portaui usque Vind[o-/ landam qu traces. Lines1-2: per-/portaui is possible; the verb occurs in 642.ii.5 where see note.

482 . Col.i: ]it /  ]xviii / ]ti ?i. Col.ii:  de ? gustis ?ceruesa- / rum. ii.1-2. gustis: the word is occasionally used in the plural and there is testimony for the neuter form gustum, see TLL VI 2368-9. A reference to beer-tasting?

484 . uacat simo salutem / e]pistulas quas `mihi´ misis- /  ti.].....ó in quibus. Line 1: this is probably the second line of an address which read "X to Y fratri karis-/simo salutem". This occurs elsewhere in the tablets only in 670, though cf. 310 and 311; it occurs also in CEL 77 and 141 (with plurimam); there is no problem with the line being inset. Line 3: only the tops of letters survive at the left. These would suit Ulucio, a place-name which occurs in 174.4 and perhaps 311.back 1 (see note); it is not clear whether there is room for ab before this, but there letters could of course have been sent either to or from the place.

499 . Read ]s Seráno Cornelian[o. This must the be the first line of a letter. Serano would appear to be the cognomen of the sender whose gentilicium precedes, but, although NPEL cites a few cognomina ending with ­ano, we have not found any such name attested.

505 . We are now confident that the back reads: a Masculó dec(urione), cf. 586, 628, 367.

508 . It is clear that both pieces belong to a right-hand column and have the beginnings of lines, which should be renumbered consecutively. Line 7 (= ii.4): qui or quia. Line 8 (= ii.5): habeo. The back has an address to a person with a name ending ­cio.

510 . Tab.Vindol.I 8. Line 6: read s(emissem) (quadrantem) (octantem) (assem i).

528 . Tab.Vindol.I 57. Line 1: we now read ut beneficio tu[o.

545 . Tab.Vindol.I 76. Read ren]untium coh[ortis. This is certainly the same hand as 139 and perhaps part of the same document.

546 . Tab.Vindol.I 77. The back reads dec in address script and is therefore a letter to a decurion.

Bibliography

J.N.Adams (1996), "The interpretation of souxtum at Tab. Vindol. II.301.3", ZPE 110: 238

A.R.Birley (1997), ""Supplying the Batavians at Vindolanda", in Groenman-van Waateringe et al. (ed. 1997): 273-80

A.R.Birley (2001b), "The Anavionenses", in Higham (ed. 2001): 15-24

A.R.Birley (2002), Garrison Life at Vindolanda. A Band of Brothers. Stroud

R.S.O.Tomlin (1996), "The Vindolanda tablets (review of Bowman and Thomas 1994; Bowman 1994; Birley et al. 1994)", Britannia, 27: 459-63