¹ Birley, E., Actes du 2^e Congrès International d'Épigraphie Grecque et Latine, Paris, 1952 (1953), 226–38Google Scholar; esp. 228.

² Callies, H., ‘Die Fremden Truppen in Römischen Heer des Prinzipats und die sogenannten Nationalen Numeri’, BRGK xlv (1964), 130–227.Google Scholar

³ Baatz, D., Kastell Hesselbach und andere Forschungen am Odenwaldlimes, Limesforschungen 12 (1973).Google Scholar

⁴ Speidel, M.P.ANRW II, 3 (1975), 202–31.Google Scholar

⁵ Mommsen, T., Hermes xix (1884), 219–34.Google Scholar The title De Munitionibus Castrorum has been amended in Grillone's edition to De Metatione Castrorum.

⁶ von Domaszewski, A., Die Rangordnung des Römischen Heeres, BJ Beiheft 117 (1908), 59–61Google Scholar ; rev. ed. B. Dobson (1967).

⁷ ORL A II. 1. Die Wetteraulinie, 49.

⁸ Cheesman, G.L., The Auxilia of the Roman Imperial Army (1914), 90.Google Scholar

⁹ Ritterling-Stein, , Die kaiserlichen Beamten und Truppenkörper in Römischen Deutschland unter dem Prinzipat (1932), 233–73.Google Scholar

¹⁰ H.T. Rowell, Numerus in Pauly-Wissowa, , Real-Encyclopädie 17 (1936), 1327–1341Google Scholar ; 2537–2554.

¹¹ W. Wagner, Dislokation der Römischen Auxiliarformationen in den Provinzen Noricum, Pannonien, Moesien und Dakien von Augustus bis Gallienus (1938).

¹² Rowell, H.T., Yale Classical Studies vi (1939), 73–108.Google Scholar

¹³ Vittinghoff, F., Historia i (1950), 389–407.Google Scholar

¹⁴ Mann, J.C., Hermes lxxxii (1954), 501–6.Google Scholar

¹⁵ Mann, J.C. in Roxan, M., Roman Military Diplomas 1978–1984 (1985), 217–9.Google Scholar

¹⁶ op. cit. (note 2).

¹⁷ op. cit. (note 3).

¹⁸ op. cit. (note 8), 84.

¹⁹ Rowell, op. cit. (note 10), 1328.

²⁰ Oxyrh. Pap. 7, 1022, 6; Callies, op. cit. (note 2), 176–7.

²¹ Gilliam, J.F., Eos 48 (1975), 207–16Google Scholar , esp. 209–12.

²² Callies, op. cit. (note 2), 177. Rowell, op. cit. (note 10), 1328 concluded from these examples that centuries and turmae were called numeri, but ‘on the strength’ seems a better interpretation. ‘It is perfectly reasonable for ‘in numeros’ to have a distinct sense in official documents which is not to be identified with the use of numerus as part of a title for a unit’ (B. Dobson, pers. comm.)

²³ Rowell, op. cit. (note 10), 1328 suggested that numeri here refers more specifically to legionary cohorts, but this is perhaps unlikely. The officer in question had been centurion in three separate legions, and centurion then primus pilus in Leg. I Adiutrix, so Rowell assumed that in this instance the troops referred to must be legionaries too, but this is by no means certain.

²⁴ op. Cit. (note 5), 220.

²⁵ Callies, op. cit. (note 2), 178.

²⁶ Ritterling, , JRS xvii (1927), 31Google Scholar ; it was thought that qui a Moesia(e) inf(erioris) Monlan(ensi) praesidio numerum in Asia perduxit referred to a cavalry detachment raised in Moesia Inferior and sent to reinforce the Eastern army (JRS xvi (1926), 77–8). But Ritterling said that numerus ‘denotes in a quite general way the body of troops previously named’, i.e. Cohors I Claudia Sugambrorum.

²⁷ Speidel, M.P., TAPA cvi (1976), 339–48Google Scholar , esp. 347.

²⁸ Speidel, M.P., TAPA cxii (1982), 209–14.Google Scholar

²⁹ CIL vi 3341, Vet(eranus) ex num(ero) frum(entariorum) leg. IIII Flaviae; CIL vi 31139, ex numer(o) eq(uitum) sing(ularium) Aug(usli). See also AE 1983, 69.

³⁰ Mommsen, T., Hermes xxii (1887), 547–58.Google Scholar

³¹ Mommsen, op. cit. (note 5), 222 n.2; 226 n.3; 229. He had been careful to note that singulares were not ‘true’ numeri but he made an exception for the Britannici.

³² Stein, op. cit. (note 9), 234.

³³ Schallmayer, E., Fundberichte aus Baden-Württemberg 9 (1984), 435–71Google Scholar , esp. 451–5.

³⁴ Callies, op. cit. (note 2), 181–3.

³⁵ CIL xiii 6471, from Böckingen, which is not on the Murr. But CIL xiii 6454 from Benningen mentions the vicani Murrenses and this fort is opposite the Murr. The usual explanation is that the numerus was originally at Benningen whence it derived its name, but later went on to Boekingen where the inscription was set up, and the original unit name remained unchanged.

³⁶ CIL xiii 6490; 6498; and see note 33.

³⁷ The name Malvensis depends on a re-reading of CIL viii 9381 = ILS 2763, previously read as Mevensis; cf. Speidel, Dacia xvii (1973), 169–177, esp. 172, where he argues that the name derives from Malva in Dacia. But see also Speidel, Ant. Afr. xi (1977), 167–173, esp. 173 n.1 and 2 for the river Malva. This river is a very long distance from Numerus Syrorum, where the numerus was presumably based, but it is the longest and most important.

³⁸ An inscription found in Africa records a man who at some stage in his career was praef(ectus) n. Mauret. Tibiscensium (CIL viii 9368 = 20944). Rowell, op. cit. (note 10), 2544 argued contra Wagner that despite this inscription, numerus Maurorum and not Mauretanorum was the preferred reading for the Dacian units. Stein, op. cit. (note 9), 240 n.43 suggested that the units on CIL iii 1149 N.M. Hisp, and iii 1294 N. Maur. Hisp, would perhaps be better interpreted as numerus Maurorum Tibiscensium.

³⁹ Stein, op. cit. (note 9), 256.

⁴⁰ Callies, op. cit. (note 2), 186, quoting CIL xvi 108 and AE 1958, 30. He suggests one unit of Palmyrenes and one of Moors both later split up, and distinguished by additional titles. He does not seem to have considered the arrival of fresh units at a slightly later date.

⁴¹ Speidel, op. cit. (note 4), 205, n.17.

⁴² Maxfield, V.A., Military Decorations of the Roman Army (1981), 234.Google Scholar

⁴³ Holder, P.A., The Roman Army in Britain (1982), 126.Google Scholar

⁴⁴ Speidel, op. cit. (note 4), 211 n.36.

⁴⁵ Callies, op. cit. (note 2), 143–5.

⁴⁶ ibid., 164.

⁴⁷ ibid., 156; 166–7.

⁴⁸ E. Birley, Antiquitas, Reihe 4, Beiträge zur Historia-Augusta-Forschung Band 12 (1976), 65–73.

⁴⁹ op. cit. (note 30).

⁵⁰ Raeticae alae cohorlesque et ipsorum Raetorum iuventus sueta armis et more militia exercita, ‘the Raetian horse and foot and the young men of Raetia itself who were accustomed to arms and trained in warfare’. They were called out in A.D. 69 by Caecina against the Helvetii.

⁵¹ Birley made this connection: TCWAAS xxxv (1935), 56–60, esp. 59 note; Festschrift für Rudolph Eggers: Beiträge zur älteren ëuropäischen Kulturgeschichte 1 (1952), 175–88, esp. 183.

⁵² Stein, op. cit. (note 9). 243–5. The Raeti who came to Manchester were probably not irregulars: Birley, op. cit. (note 51) (1952). 183. Davies, R.W., Klio lx (1978), 363–70Google Scholar , esp. 367–8.

⁵³ Frere thinks it is Flavian: Britannia xi (1980), 51–60; Domaszewski dated it to the reign of Trajan in his 1887 edition of Hyginus: Birley argued for Marcus Aurelius' Danube wars. op. cit. (note 48), 67, and Corolla Memoriae Erich Swoboda Dedicata (1966). 54–67. esp. 57 n.6; Watson favoured a third-century date, see Oxford Classical Dictionary, article on Hyginus.

⁵⁴ Stein, op. cit. (note 9), 235, argued that they were unrelated; Rowell, op. cit. (note 10), 1330. thought that they were the same kind of troops. See also Cambridge Ancient History XI. 497 for the view that nationes and symmachiarii were synonymous terms and that they were the forerunners of the numeri.

⁵⁵ op. cit. (note 4). 207–8.

⁵⁶ op. cit. (note 30), 550–2.

⁵⁷ Rowell. op. cit. (note 10). 1330.

⁵⁸ Richmond, I.A., PBSR xiii (1935). 1–40.Google Scholar

⁵⁹ Dobson, B.. Epigraphische Studien 8 (1969). 122–4.Google Scholar but it could still be Trajanic as previously interpreted: B. Dobson, pers. comm. See AE 1985, 719. where it is redated to Trajan's Second Dacian War.

⁶⁰ Birley, op. cit. (note 48). 68.

⁶¹ Richmond, I.A.. JRS xxxv (1945). 15–29.Google Scholar

⁶² ibid., 15 n.12.

⁶³ Mann, op. cit. (note 14). 502.

⁶⁴ Rowell, op. cit. (note 10). 1339.

⁶⁵ Callies. op. cit. (note 2). 165; 199 n. 396.

⁶⁶ CIL xvi 68; M. Roxan. Roman Military Diplomas 1954–1977 (1978). nos. 17. 27. 28.

⁶⁷ Mann. op. cit. (note 15), 217–9. esp. 219 no. 10.

⁶⁸ Callies, op. cit. (note 2). 182. Speidel, op. cit. (note 4). 210. says that place names do not necessarily denote association with that particular location; see also note 35.

⁶⁹ Rowell, op. cit. (note 10), 2546.

⁷⁰ From Optatiana: Aelius Bolhas. CIL iii 907 = 7963; P. Aelius Septimius Audeo, CIL iii 1471. From Tibiscum: Aelius Habibis, CIL iii 7999; Aelius Boraeas, CIL iii 14206. From Porolissum: Aelius Them(arsa?). AE 1979. 495. No unit is named on this last example, but it is probable that it was the numerus Palmyrenorum Porolissensium.

⁷¹ Callies, op. cit. (note 2), 191.

⁷² Mann, op. cit. (note 15), 217. See also Kennedy, D., Limes: Akten des XI Int. Limeskongresses 1976 (1977), 521–31. esp. 527, for Parthian regiments, where local recruitment seems to be the normal pattern.Google Scholar

⁷³ Pius: Albertini, E., Revue Africaine lxxv (1934). 37Google Scholar ; Commodus: Carcopino, J., Syria vi (1925), 122Google Scholar ; Severas: Picard, G.C., Castellum Dimmidi (1944), 104.Google Scholar

⁷⁴ Baradez, J.. Fossatum Africae (1949), 102–4.Google Scholar An inscription was found quoting trib. pot. X of Hadrian, A.D. 126–7 (AE 1950, 58). Baradez thought there must have been a ‘provisional’ fort at Gemellae built by Cohors I Chalcidenorum, because the principia was not built until A.D. 132 by Legio III Augusta (AE 1950, 59), but Fentress points out that this inscription dates only the principia, not the fort; see Numidia and the Roman Army BAR S53 (1979). 84.

⁷⁵ Cagnat, R., L'Armée Romaine d'Afrique (1913), 578.Google Scholar

⁷⁶ Callies, op. cit. (note 2), 201 n.407. It is possible that the Palmyrenes arrived along with the oriental auxiliaries who came to Africa during the Moorish revolt. In A.D. 145 a vexillation of Legio VI Ferrata from Syria, or possibly from Judaea, built a road across the Aurès behind Lambaesis (ILS 2479), and Palmyrenes may have arrived with the legionaries. It is sometimes argued that they must have arrived by AD 150 because a Palmyrene, Mocimus, son of Simon, died at Lambaesis in that year (CIL viii 3917 = 18202). but his gravestone merely quotes his name and no military unit is mentioned. He may have been a civilian, perhaps a trader.

⁷⁷ Libyca ii (1954), 178–81. I am grateful to Alan Rushworth for this reference. See also AE 1980. 954 = AE 1941, 156.

⁷⁸ Albertini, op. cit. (note 73), 30. It is probable that the centurion was indeed praepositus numeri (though the inscription does not actually say so); the dating is dependent upon a restoration of the name A. Iulius Piso, who was legate in A.D. 177–8. See also AE 1980, 953.

⁷⁹ Carcopino. op. cit. (note 73). 123.

⁸⁰ Picard, op. cit. (note 73), 105 n.57; 199 no. 23.

⁸¹ Maximus Zabdibolus CIL viii 2505; Suricus Rubatis viii 2515; Heranus = Hairan viii 2510 = 18006; 2511; 2512; and Albertini, , Revue Africaine lxxii (1931), 213Google Scholar no. 18. Bilingual inscriptions: CIL viii 2515; and Albertini, 220 no. 29.

⁸² Mann. op. cit. (note 14), 505 NO. 1 says that the Palmyrenes at El Kantara ‘seem to have formed a hereditary and mainly non-Roman group’.

⁸³ Picard, op. cit. (note 73), 110. contra Cagnat, op. cit. (note 75), 206, who had said that there was no African element in the numerus.

⁸⁴ Rowell. op. cit. (note 10). 2550.

⁸⁵ Speidel, op. cit. (note 4). 210; and see Mommsen, op. cit. (note 5), 228. n. 1.

⁸⁶ Cataniciu, I.. Evolution of the System of the Defence works in Roman Dacia BAR S116 (1981), 76Google Scholar , n. 181.

⁸⁷ It is not certain whether the Palmyrenes were organised as a fully-fledged numerus when they first arrived in Africa. Rowell. op. cit. (note 10), 2552 thought that the Palmyrenes were already formed as a numerus when Agrippa commanded them, though he does not decide about the date of the command. Elsewhere he also stated that it was a completely independent unit, not attached to Cohors I Chalcidenorum: op. cit. (note 12). 77.

⁸⁸ No fort is known at El Kantara. The theory that this was the headquarters of the numerus with detachments from it sent to man outposts must remain unproven. See Marrou, , MEFR 1 (1933), 42–86CrossRefGoogle Scholar ; Carcopino, op. cit. (note 73)- 136.

⁸⁹ Callies. op. cit. (note 2). 161; Speidel. op. cit. (note 4). 211.

⁹⁰ Speidel. op. cit. (note 4), 212 n.38.

⁹¹ Wagner, op. cit. (note 11), 207.

⁹² Speidel, op. cit. (note 4), 209–10.

⁹³ Stein, op. cit. (note 9), 243; Mann. op. cit. (note 15), 218.

⁹⁴ Jones, A.H.M., JRS xxvi (1936). 230–1Google Scholar ; Thomas, J.A.C., Textbook of Roman Law (1976), 408.Google Scholar

⁹⁵ Mann. op. cit. (note 15), 218.

⁹⁶ Wagner, op. cit. (note 11), 207; Speidel, op. cit. (note 4), 209–10.

⁹⁷ Cataniciu, op. cit. (note 86), 44.

⁹⁸ op. Cit. (note 4), 206–7; 211.

⁹⁹ Saxer, R., Epigraphische Studien 1 (1967). 54Google Scholar ; 131.

¹⁰⁰ Albertini, E. and Massiera, P., Revue des Etudes Anciennes xli (1939). 223–44.CrossRefGoogle Scholar esp. 242–3, argue that the term praetendere as used on several inscriptions from Messad implies rather less stability than other terms such as morare. See also ILS 569; 2660; 2767. where the term is used for collections of troops which do not seem to be permanent garrisons.

¹⁰¹ Mann. op. cit. (note 14). 502.

¹⁰² Birley, A., Septimius Severus (1971), 265.Google Scholar

¹⁰³ MacKendrick, P., The Dacian Stones Speak (1975). 204–5.Google Scholar

¹⁰⁴ Collingwood, R.G., Roman Britain and the English Settlements (2nd ed, 1937). 145–6.Google Scholar

¹⁰⁵ Gillam, J.P. in Richmond, I.A. (ed.), Roman and Native in North Britain (1958). 60–90.Google Scholar

¹⁰⁶ Jobey, G., AA⁵ ii (1974), 17–26.Google Scholar

¹⁰⁷ Gillam, J.P. in Miket, R. and Burgess, C. (eds.). Between and Beyond the Walls (1984), 287–94.Google Scholar

¹⁰⁸ A study of the pottery at Hesselbach revealed that the foundation of the fort was ‘hardly possible before A.D. 90′; Baatz opted for c. A.D. 95, op. cit. (note 3), 94. More recently Barbara Pferdehirt re-examined the pottery from Hesselbach and Seckmauern, and concluded that they were not established until A.D. 100–101: Jahrbuch des Römische-Germanischen Zentralmuseums xxxiii (1980), 221–320. esp. 279 and 307–12. The other Odenwald forts yielded few finds, but since they are so alike in size and shape it is probably safe to assume that they were founded at about the same time and garrisoned in the same way.

¹⁰⁹ Baatz. op. cit. (note 3), 14; 71.

¹¹⁰ Scott, J.G., GAJ iv (1976), 29–41.Google Scholar

¹¹¹ Gillam, op. cit. (note 107), 293.

¹¹² Frere, S.S., Britannia v (1974), 495Google Scholar ; Britons may have been stationed on the Taunus at the end of the first century, but the evidence is slight, consisting of British brooches of late first-century to early second-century date from the forts at the Saalburg, Zugmantel and Hüfingen, and an undated graffito, CIL xiii 11954a, con(tubernium) Brittonis, from the Saalburg.

¹¹³ Birley, E., Roman Britain and the Roman Army (1953), 22.Google Scholar He thought that the cohort was formed c. A.D. 84, writing at a time when it was held that Mons Graupius was fought in the same year. But it is still possible that it was formed by A.D. 83 and was present at the battle.

¹¹⁴ Hartley, B., Britannia iii (1972). 13Google Scholar ; Breeze, D.J., Northern Frontiers of Roman Britain (1982). 65Google Scholar ; Hanson, W.S. and Maxwell, G.S.. Rome's North West Frontier (1983). 45.Google Scholar

¹¹⁵ Baatz, op. cit. (note 3), 67; 70; 73.

¹¹⁶ Todd, M.. Roman Britain 55 B.C.-A.D. 400 (1981). 122.Google Scholar The place name Anava is equated with the river Annan in Dumfriesshire; others have disagreed with this equation because the Annan was outside the province, but Rivet, and Smith, . Place Names of Roman Britain (1979). 249–50 restated the case for the Annan.Google Scholar

¹¹⁷ The Historia Augusta (Hadrian. 5.2) says that on Hadrian's accession the Britons could hardly be kept under control, but it was a commonplace to mention trouble in Britain at the beginning of several Emperors' reigns. More reliable is Fronto's letter to Marcus Aurelius mentioning heavy casualties in Judaea and Britain in Hadrian's reign (Fronto, 2.22). He gives no indication of date, but if there was any fighting it was probably all over by A.D. 119. quelled by Pompeius Falco; commemorative coins were issued in that year with the legend Britannia. See Birley, A.R., Fasti of Roman Britain (1981), 99Google Scholar , and RIC ii 577 a and b.

¹¹⁸ Gillam. op. cit. (note 107). 290.

¹¹⁹ Barthel, W., BRGK vi (1910–1911). 149.Google Scholar Stein, op. cit. (note 9), 256.

¹²⁰ Rowell. op. cit. (note 12), 105.

¹²¹ Hassall, M. in Wacher, J. (ed.). The Roman World (1987). 698.Google Scholar The name Alexandriani need not be connected with Severus Alexander. It could have been tacked on to an existing unit formed some years previously.

¹²² Sherwin-White, A.N., JRS lxiii (1973), 86–98Google Scholar , esp. 98. The literature on this subject is enormous: see also Jones, A.H.M., Studies in Roman Government and Law (1968). 129–40.Google Scholar

¹²³ Alföldy, G., Epigraphische Studien 6 (1968), 79Google Scholar ; 219–20.

¹²⁴ Rowell, op. cit. (note 10), 1333–4.

¹²⁵ Stein, op. cit. (note 9), 241–4.

¹²⁶ praefecti: CIL iii 1149: viii 9368 = 20944; xiii 6814; 11979. Tribuni: viii 9381; 11343; 20945. Domaszewski. op. cit. (note 6). 108; Rowell, op. cit. (note 10), 1336; Cagnat, op. cit. (note 75), 210.

¹²⁷ Davies, R.W., Epigraphische Studien 12 (1981), 183–214Google Scholar , esp. 184.

¹²⁸ Birley, E., Latomus xlii (1983). 73–83Google Scholar , esp. 83.

¹²⁹ Smith, R.E., ZPE xxxvi (1972), 263–78Google Scholar , esp. 264.

¹³⁰ Saxer, op. cit. (note 99), 122–3.

¹³¹ Smith, op. cit. (note 129), 123.

¹³² Callies, op. cit. (note 2), 187, n.324.

¹³³ Saxer, op. cit. (note 99), 130–1.

¹³⁴ Rowell, op. cit. (note 10), 1336–7.

¹³⁵ Cailies, op. cit. (note 2). 189.

¹³⁶ Baatz, op. cit. (note 3). 28–31.

¹³⁷ Callies, op. cit. (note 2), 190.

¹³⁸ Rowell, op. cit. (note 10), 1337.

¹³⁹ Domaszewski, op. cit. (note 6), 2nd ed. B. Dobson (1967), xviii. 60–1.

¹⁴⁰ Baatz, op. cit. (note 3), 74 n.189. Graffiti were found at Arzbach. Ems, Kernel. Heftrich. Feldberg. Kapersburg, Inheiden, Altenstadt, Walldürn. It should be noted that numeri are not definitely attested in all of these so-called ‘numerus’ forts.

¹⁴¹ Mann, op. cit. (note 14), 502 n.15.

¹⁴² Albertini, op. cit. (note 81), 212. no.17.

¹⁴³ ibid., 214, and see Mann's comments, op. cit. (note 82).

¹⁴⁴ Speidel. op. cit. (note 4), 208 n.30; Wagner, op. cit. (note 11). 167.

¹⁴⁵ Albertini, op. cit. (note 81), 201–3.

¹⁴⁶ ORL B 39. 16 nos. 5 and 6; Tafel 3 nos. 8 and 9. Baatz, D., SJ xxxv (1978), 61–107Google Scholar, esp. 107.

¹⁴⁷ Cheesman, op. cit. (note 8), 88; MacMullen, R., Klio lxii (1980), 451–60Google Scholar , esp. 452.

¹⁴⁸ Rowell, op. cit. (note 10), 1337–8.

¹⁴⁹ Wagner, op. cit. (note 11), 208.

¹⁵⁰ Speidel, M., Dacia xix (1973), 169–77Google Scholar , esp. 171.

¹⁵¹ Marichal, R., CRAI (1977), 436–52Google Scholar , esp. 440.

¹⁵² Baatz, op. cit. (note 3), 36; 59.

¹⁵³ ibid., 76.

¹⁵⁴ Stein, op. cit. (note 9), 263.

¹⁵⁵ Ritterling, , BJ cxx (1911), 177Google Scholar ; Rowell, op. cit. (note 10). 2543: Stein, op. cit. (note 9), 263.

¹⁵⁶ It is not certain how long the fort at Heddesdorf remained in occupation after Niederbieber was built. It is possible that Cohors II Hispanorum equitata which garrisoned Heddesdorf in the mid-second century was moved to Niederbieber, but there is no proof.

¹⁵⁷ Mann, op. cit. (note 15), 217.

¹⁵⁸ Albertini, op. cit. (note 81), 213 no. 18; 214 no. 19.

¹⁵⁹ Mann, op. cit. (note 14), 503.

¹⁶⁰ ibid., 505.

¹⁶¹ Rowell, op. cit. (note 12), 106; Mann, Heer und Integrationspolitik. Passauer Historische Forschungen 2 (1986), 187–9.

¹⁶² Callies, op. cit. (note 2), 189–98; Stein, op. cit. (note 9), 237–8. Baatz, op. cit. (note 3), 73 n.187 takes issue with the description ‘barbarie’, since it is not precisely defined: to the Romans the King of Parthia was a barbarian. Baatz prefers to judge by the degree of Romanization, though it is difficult to measure.

¹⁶³ Wiegels, R., Epigraphische Studien 12 (1981), 311–31Google Scholar ; esp. 312–9.

¹⁶⁴ Stein, op. cit. (note 9), 268.

¹⁶⁵ ORL A Strecke 7–9, 178.

¹⁶⁶ Stein, op. cit. (note 9), 271. See also Fishwick, D., JRS lvii (1967). 142–60.Google Scholar

¹⁶⁷ Mann, op. cit. (note 14), 505–6.

¹⁶⁸ Wiegels, op. cit. (note 163).

¹⁶⁹ Baatz, op. cit. (note 3), 74 suggested that the Triputienses dropped the title Brittones and called themselves exploratores because by the time they arrived in Miltenberg there were no Britons left in the unit. By analogy then, this should mean that units which kept the title did contain Britons; but this does not readily apply in auxiliary units.

¹⁷⁰ Stein, op. cit. (note 9), 247–50.

¹⁷¹ Cagnat, op. cit. (note 75), 263–7.

¹⁷² Stein, op. cit. (note 9), 241.

¹⁷³ Rowell, op. cit. (note 10), 1333.

¹⁷⁴ Picard, op. cit. (note 73), 124.

¹⁷⁵ Baatz. BJ clxvi (1966), 194–207, esp. 201–3. and n.49; and oP. cit. (note 3), 113.

¹⁷⁶ Calwell, E., Small Wars 3rd ed. (1909). 210.Google Scholar

¹⁷⁷ Contamines, P., War in the Middle Ages (1984). 212.Google Scholar

¹⁷⁸ Mommsen. op. cit. (note 5), 223.

¹⁷⁹ Speidel, , JRS lx (1970), 142–53Google Scholar , esp. 148 n.69 (quoting P. Dura 100 and P. British Museum 2851 = Hunt's Pridianum).

¹⁸⁰ ibid., 148, n.75.

¹⁸¹ ibid., 148.

¹⁸² Its exact whereabouts are unknown; it may have formed the garrison of Passau, but Dietz, , Germania lx (1982), 183–91Google Scholar , esp. 184, suggested Ruffenhoffen. It appears on diplomas for Raetia between AD 140–168, and one inscription was found at Weissenburg (CIL iii 11918 and ORL B 72, 45) with the title exploratores.

¹⁸³ Davies, R., GAJ iv (1976), 103–7Google Scholar , esp. 104.

¹⁸⁴ Dietz, op. cit. (note 182), 191 n.63–66.

¹⁸⁵ Mommsen, op. cit. (note 5), 228 n.2; Cagnat, op. cit. (note 75), 249–50.

¹⁸⁶ Gostar, N., Germania 1 (1972), 241–7.Google Scholar

¹⁸⁷ Speidel, op. cit. (note 179), 148.

¹⁸⁸ Stein, op. cit. (note 9), 264–5; 267–8.

¹⁸⁹ Rankov, N.B., Britannia xviii (1987), 243–9CrossRefGoogle Scholar , esp. 246–7.

¹⁹⁰ The name Halicenses or Halicanenses has been linked to both the Altkönig (ORL B 10. 16) and to Halicanum in Pannonia: Baatz, D. (ed.). Die Römer in Hessen (1982), 269.Google Scholar

¹⁹¹ Schönberger, H., Kastell Künzing-Quintana, Limesforschungen 13 (1975). 112–4.Google Scholar

¹⁹² Beckmann, B., Limes: Akten des XI Int. Limeskongresses 1976 (1977). 113–22Google Scholar , esp. 116.

¹⁹³ Gillam, J.F., Yale Classical Studies xi (1950). 169–252Google Scholar , esp. 217 and 226–7.

¹⁹⁴ Stein, op. cit. (note 9), 261.

¹⁹⁵ Schleiermacher, W., Historia ii (1953–1954), 103.Google Scholar

¹⁹⁶ Stein, op. cit. (note 9), 261.

¹⁹⁷ Kiechle, F., Historia xi (1962), 171–92.Google Scholar

¹⁹⁸ Richmond, I.A., History xliv (1959), 1–16CrossRefGoogle Scholar , esp. 13.

¹⁹⁹ Rankov, op. cit. (note 189), 248.

²⁰⁰ Mann, J.C., GAJ iii (1974), 34–42Google Scholar , esp. 38–40.

²⁰¹ Davies, op. cit. (note 183), 105 n.34; and see Not. Dig. Oce. 5.45 for sagittarii venatores.

²⁰² ILS 2628 = CIL viii 9059 T. Ioincatius Sabinus of the numerus Divitiensis G(ermania) S(uperior), who died at Auzia in Mauretania; ILS 9187 Ulpius Qu(i)etus of the numerus exploratorum Germanicianorum who died at Ain-Temouchent in Mauretania; ILS 9187A = CIL viii 21668 Iulius Adventus Spiorator Bataorum (sic), who also died at Ain-Temouchent.

²⁰³ Speidel, M., Epigraphische Studien 13 (1983), 63–78.Google Scholar

²⁰⁴ ibid., 66 n. 19.

²⁰⁵ ibid., 67–73.

²⁰⁶ ibid., 74.

²⁰⁷ Speidel, op. cit. (note 179), 148 n.69.

²⁰⁸ Moramsen, op. cit. (note 5), 227 n.3; 232.

²⁰⁹ Jones, A.H.M., Late Roman Empire (1964), 99–100.Google Scholar

²¹⁰ Varady, L., Acta Antiqua Academiae Scientiarum Hungaricae ix (1961), 360.Google Scholar

²¹¹ Jorns, W., SJ xxiv (1967), 12–32.Google Scholar

²¹² Baatz, op. cit. (note 3), 75.

²¹³ Baatz, D., Der Römische Limes (2nd ed, 1974), 187.Google Scholar

²¹⁴ Birley, E., AA⁴ lx (1932), 205–15Google Scholar , esp. 212–3.

²¹⁵ Collingwood, op. cit. (note 104), 139.

²¹⁶ Breeze, D. and Dobson, B., Hadrian's Wall (rev. ed., 1978), 40Google Scholar ; Frere, S.S., Britannia (3rd ed., 1987), 121–2Google Scholar ; 125 n.26.

²¹⁷ Breeze and Dobson, op. cit. (note 216), 22.

²¹⁸ Frere, op. cit. (note 216), 495.

²¹⁹ Dobson, B., Britannia xiv (1983), 362–4CrossRefGoogle Scholar , esp. 363.

²²⁰ Richmond, op. cit. (note 198), 12.

²²¹ Potter, T.W., Romans in North West England (1979), 355Google Scholar , fig. 144.

²²² Breeze and Dobson, op. cit. (note 216), 142; Frere, op. cit. (note 216), 167. Rivet and Smith, op. cit. (note 116), 212, dismiss the whole idea of loca, but Frere, Britannia (1980), 422–3 points out that the practice is compatible with frontier policy elsewhere in the Empire.

²²³ Shotter, D.C.A., Britannia iv (1973), 206–9CrossRefGoogle Scholar , esp. 209.

²²⁴ Frere, op. cit. (note 216), 226 n.5; 344; 352 n.45.

²²⁵ Alföldy, op. cit. (note 123), 95.

²²⁶ Böhme, A., SJ xxix (1972), 55.Google Scholar

²²⁷ ORL B 4, 18.

²²⁸ ORL A 1, 55; 119.

²²⁹ ORL B 8, Tafel 28 nos. 31 and 40.

²³⁰ Schönberger, H., Die Kastelle in Altenstadt, Limesforschungen 22 (1983), 61–4Google Scholar ; 160.

²³¹ Schönberger, H. and Hartley, B.. SJ xxvii (1970). 21–30.Google Scholar

²³² Baatz, D. (ed.). Die Römer in Hessen (1982), 457–9.Google Scholar

²³³ Filtzinger, P.et al. (eds.). Die Römer in Baden-Württemberg. 3rd ed. (1986). 600Google Scholar ; Planck, D. Archäologische Ausgrabungen in Baden-Württemberg (1984). 128–38Google Scholar, esp. 129–32; ibid. (1985), 128–36.

²³⁴ Filtzinger, op. cit. (note 233), 508–9; Heiligman, J. in Studien zu den Militärgrenzen Roms III (1986), 175–81Google Scholar, esp. 179.

²³⁵ ORL A 5, 93.

²³⁶ ORL A 1, 121.

²³⁷ Dietz, K., Chiron xiii (1983), 497–536Google Scholar , esp. 518.

²³⁸ ibid., 533 11.207; Dietz. op. cit. (note 182). 184: Speidel. op. cit. (note 203). 74 n.50.

²³⁹ See Cataniciu, op. cit. (note 86), 22; 75 n.181 for discussion.

²⁴⁰ Rowell, op. cit. (note 10), 2550.

²⁴¹ Tudor, D., Actes du IX Congrès International d'Études sur les Frontieres Romaines 1972 (1974), 235–46Google Scholar, esp. 243.

²⁴² Vladescu, C.M., Limes: Akten des XI Internationalen Limeskongresses 1976 (1977), 353–63Google Scholar, esp. 359.

²⁴³ Cataniciu, op. cit. (note 86), 31; Tudor, op. cit. (note 241). 245.

²⁴⁴ Fentress, op. cit. (note 74), 90–91, no. 22. The inscription was not found on the site, but one kilometre away where there are no ruins, so it is assumed that it was removed there at some date.

²⁴⁵ C.M. Daniels (pers. comm.).

²⁴⁶ Picard, op. cit. (note 73), 103; 199.

²⁴⁷ ibid., 87.

²⁴⁸ Speidel, M.P., Dacia xvii (1973), 173.Google Scholar

²⁴⁹ Salama, P., Limes: Akten des XI Int. Limeskongresses 1976 (1977), 577–95Google Scholar, esp. 587.

²⁵⁰ Baatz, op. cit. (note 3), 73.