PRIVATIZATION OF AGER IN AFRICA FROM 123 TO 63 b.c.

Abstract Scholars have generally underestimated the level of Roman involvement in Africa in the period between the annexation of Carthage in 146 b.c. and Caesar's victory at Thapsus in 46 b.c., and the land in Africa which the Romans annexed has been conventionally called public land (ager publicus). This paper analyses the surviving text of the African provisions of the epigraphic lex agraria of 111 b.c. and notes that the term ager publicus is not attested in the provincial section of the law. The land in Africa appears simply as ager in Africa and the term ager publicus is confined exclusively to Italy in the law of 111. However, Cicero's references to Rullus’ agrarian proposal in 64/3 b.c. in the De lege agraria suggest that the term ager publicus was used to qualify land existing outside Italy in Rullus’ proposal. This paper argues that the concept of ager publicus as opposed to private land developed in Africa between 111 and 63 b.c., and that this was linked to privatization of ager in Africa in this period. The results of this study suggest a high degree of Roman exploitation of African land prior to the Caesarean and Augustan colonies in the 40s b.c.


INTRODUCTION
For the study of agrarian history after the Gracchi the principal source of information is the anonymous lex agraria dated to 111, the only Roman agrarian law from this period directly attested by an inscribed copy. 1 Since Crawford's new edition in 1996, the importance of the lex agraria of 111 has been increasingly noted in recent studies, particularly as the first half of the law sheds light on the situation regarding public land of the Roman people in the land of Italy (ager publicus populi Romanei in terra Italia) in the post-Gracchan period, for which our literary evidence is notoriously confusing. However, the latter part of the lex agraria of 111, which deals with land outside Italy (extra terra Italia [sic]) in Africa and Greece, has provoked relatively little discussion. This provincial section of the law is not seen as directly related to the issues of ager publicus in Italy in the studies of the Gracchan period. Too little is known of the Greek part of the law to allow analysis on its own. Lastly, those who look at Roman imperialism in Africa have generally played down the level of Roman interest and activities in Africa before the foundation of Caesarean and Augustan colonies following Julius Caesar's victory at Thapsus in 46. 2 The surviving text of the African provisions of the epigraphic law of 111 deals with the situation after the further implementation of the contentious colony of Junonia, the first full-scale Roman attempt at exploiting the African land under the lex Rubria, had been cancelled after Gaius Gracchus' death in 121. 3 Although the law of 111 is fragmentary, it is possible to compare the surviving text of the African provisions with some passages of Rullus' agrarian proposal in 64/3 which Cicero claims to quote in the De lege agraria. The result may provide new insights into Roman economic attitudes to the African land and their consequences in this period.
This study does not attempt to reconstruct the missing portions of the lex agraria; instead, it begins by looking at what is referred to in the law of 111 as ager in Africa (lex agr. line 68) which Scipio Aemilianus assigned to the Roman people (lex agr. line 89; Cic. Leg. agr. 2. 58) in 146. 4 This paper notes that the land extra terra Italia in Africa and Greece, which was made the property of the Roman people, is never called ager publicus in the epigraphic law of 111 as one might expect, and that the term ager publicus is used exclusively for land in Italy, whereas from the writings of Cicero we learn that the term ager publicus was used to qualify land outside Italy both legally and in practice in his day. Scholars such as Lintott and de Ligt have also noted that the expression ager publicus populi Romani is not used in the non-Italian section of the law of 111. 5 As we shall see, it is unlikely that the formula has been lost from the surviving text, but the significance of this has not been explored.
The standard view is that ager Romanus, as distinct from ager peregrinus, consisted of two types of land: ager publicus and ager priuatus. 6 The ager Romanus in this sense is not to be confused with the ager Romanus antiquus, ranging only a few kilometres beyond the pomerium, but indicates the Roman territory across the empire. 7 The ager  publicus comprised lands captured from defeated enemies and was the public property of the Roman people. 8 Following these categorizations, based primarily on the imperial jurists and writers, it has been claimed that some land in Africa became ager publicus in 146. 9 However, as we shall see, ager Romanus was confined to Italy by 210, and the legal status of ager publicus before the lex agraria of 111 is not always clear. 10 None the less, the concept of land in public ownership requires the existence of privately owned land. 11 In what follows, I will argue that, as a result of gradual privatization of annexed land and the development of private landowning in Africa, what was once officially a single category of ager in Africa was divided and became two categories, one of ager priuatus and the other of land which remained in public ownership. I will suggest that the term ager publicus, which seems to have been peculiarly Italian in legal language till 111, as we shall see, was extended to include the latter type of land in public ownership outside Italy in the period between 111 and 64/3, reflecting this development. In investigating how ager in Africa in the epigraphic law of 111 came to be termed ager publicus in Cicero's day, I will consider the lex agraria of 111 as a key factor which accelerated privatization of ager in Africa in this period.

AFRICA IN THE LAW OF 111
The structure of the lex agraria of 111 may be summarized as follows. The first half (lines 1-42) of the law is concerned with the ager publicus populi Romanei in terra Italia (line 4) affected by Tiberius Gracchus' law of 133. The gist of this Italian section is that certain categories of land in Italy that were ager publicus in 133 would be converted into private land and have the same status as other private property; this is followed by regulations for the land that would remain ager publicus. After a passage which seems to make provisions for a colony established by one Baebius in an unknown location (roughly lines 43-7), the second half of the law (lines 48-105) deals with the management of the land which the Romans had acquired extra terra Italia (line 49), first in Africa (lines 48-93) and then in Greece (lines 93-105). The reference to the starting point in time of the African section of the law is 146, when Cartago capta est (line 89), after which the lex Rubria proposed by Gaius Gracchus' colleague in 123 or 122 to found Junonia appears as an important reference point. 12 The general scholarly view is that the lex Rubria was repealed and that Junonia lost its legal status following the death of Gaius Gracchus. 13  were attempts to undo some or all of Gaius Gracchus' legislation during 121, and a tribune named Minucius Rufus proposed a rogatio to repeal the lex Rubria. 14 It is not evident, however, whether Minucius' proposal eventually passed into a law in 121. Minucius held a contio to advocate his proposal, while Gracchus and his followers were watching it. Gracchus' enemies succeeded in provoking violence and an attendant of the consul Opimius fell dead. The contio was dissolved over this incident. The following morning Opimius called a meeting of the Senate, where the so-called senatus consultum ultimum was passed for the first time; the deaths of Gracchus and Flaccus soon followed. It appears that Minucius' rogatio was not put to the vote in Gracchus' lifetime.
The phrase ex lege Rubria, quae fuit, which appears in lines 55 (restored), 59, 60 (restored) and 79 (restored) of the epigraphic law of 111, has usually been translated, or rather glossed, as 'according to the lex Rubria, which was repealed'. Literally, however, it means 'according to the lex Rubria, which was', confirming that the lex Rubria existed and is now to be replaced by the current statute. The law of 111 shows that the original colonists of Junonia retained their allotments (lines 58-61). In addition, a marble inscription from near La Malga on the outskirts of Carthage seems to bear the names of C. Sulpicius Galba, C. Papirius Carbo and L. Calpurnius Bestia. The text is heavily restored but may indicate an agrarian commission replacing Gaius Gracchus and Flaccus after their deaths. 15 Carbo poisoned himself in 119 when prosecuted by L. Crassus, and, according to the lex repetundarum lines 8-9, a trial concerning agrarian commissioners should not take place while they hold magistracies. 16 If the La Malga inscription is interpreted as representing a post-Gracchan agrarian commission, the implication is that the land allocation of the lex Rubria proceeded at least until the prosecution of Carbo in 119.
It is unlikely that Minucius' rogatio was put to the vote in the time between the contio and the death of Gracchus, as we have seen, and, when the La Malga inscription is taken into account, Minucius' proposal does not seem to have been passed and carried into execution immediately following the suppression of Gracchus and Flaccus. 17 If the lex Rubria was indeed repealed, this may belong to 119 when Carbo was no longer an agrarian commissioner, that is, to quote Appian, 'about fifteen years after the law of (the elder) Gracchus', and the repeal was probably not by a lex Minucia. 18 In any case, by the phrase 'according to the lex Rubria, which was' the law of 111 seems to acknowledge the existence of a former law about Africa, and this reading would then be in line with the law's general confirmation of what had been done under the lex Rubria. Lintott

YEONG-CHEI KIM
Another question that arises is what Africa means in the epigraphic law. The law's understanding of Africa must have reflected Roman perception of it around 111. The first mention of Africa occurs in line 48, phrased as 'whatever land or piece of land that is in Africa' (quei ager locus in Africa est), and the introductory formula appears throughout the rest of the text, often simply as ager in Africa or Africa. 19 In the first place, Africa in the context of the lex agraria appears to refer to the Roman province as it was in 111, and the law introduces two main categories of the annexed land in Africa: land assigned to colonists under the lex Rubria before the further implementation of the colony was cancelled (lines 55-82); and land publicly sold off ( publice uenit) at auctions in Rome in 115 and 113 (lines 85-9), which could then become 'private land subject to taxation' (ager priuatus uectigalisque). 20 A large centuriation system (approximately 12,500 km 2 ) has been identified by aerial photography in north-eastern Tunisia. It was laid out to make 200 iugera from each centuria and may well be associated with the lex Rubria, later taken over by Julius Caesar and Augustus. date the bulk of the northern centuriation system to the Augustan period. These dates proposed for the centuriation systems have not been tested by correlative evidence from excavation or epigraphy. Yet the Gracchan date for the northern centuriation cannot be ruled out. Various sources suggest that the Gracchan commissioners surveyed a large territory in Africa (see n. 3 above). A letter by Fronto deserves particular mention in this context: iam Gracchus locabat Asiam et Karthaginem uiritim diuidebat (Ep. ad Ver. Imp. 2.1.13). If Gracchus here means Gaius, it may be that Gaius Gracchus effected both colonial and uiritim distributions in Africa. Perhaps the distributions were phased in. He also admitted 6,000 colonists for Junonia, which exceeded the number the lex Rubria allowed (App. B Civ. 1.24; Solin. 27.11). In addition, the La Malga inscription suggests that the Gracchan commission surveyed land very close to the ruined city of Carthage. It is relevant to note that when Copia was founded in 194/3 a third of the centuriated area was reserved for supplementary settlement: see Livy 35.9.7-8.
Scholars agree that the term prouincia before the first century meant the assignment or responsibility of a Roman magistrate, usually holding imperium, and was not a territorial term. 22 There are some indications in the law of 111 that the province of Africa was not established as a formal institution by then. The word prouincia appears only on two occasions in the law of 111. 23 The first is found in line 46, which seems to relate to land outside Italy but probably not in Africa, and the prouincia denotes the assignment of the quaestor who will be in charge of the treasury (line 46). The second occasion appears in the African section, and the law concerns a prefect or soldier in prouinciam (line 55). The prouincia here seems to have a territorial sense, although it remains a possibility that it means being on overseas service, as Gargola argues, owing to the many lacunae. 24 It seems that prouincia can refer to both a magistrate's duty and a territory according to context in the law of 111, if in prouinciam means 'into the province', and this perhaps suggests that the concept of the prouincia was taking on a territorial significance in the late second century. This is not to deny that the land annexed in 146 had geographical limits. The Romans previously approved the existing Carthaginian territory separated from the kingdom of Numidia by Punic ditches as part of the settlements at the end of the Second Punic War. 25 After the destruction of Carthage, Pliny (HN 5.25) states that Scipio Aemilianus and the heirs of Massinissa laid down a boundary ditch, later known as the fossa regia ('the royal ditch'), and it was used to mark the boundary between two Roman provinces, Vetus and Nova, in Pliny's day. The supposed course of the fossa regia from east of Thabraca to Thaenae generally corresponds to the western limit of the centuriation systems identified in Tunisia. 26 If this line is correct, the ditch can indeed be taken to indicate the original boundary between the kingdom of Numidia and the former Carthaginian territory in 146. 27 It is clear from the above that Rome had geographical conceptions of Africa and, whatever the range was, the fact of Roman territorial annexation in 146 of the finis Carthaginiensium, to use Sallust's words (Iug. 19.7), is undeniable.
In the later part of the African section of the law of 111, however, Africa seems to extend beyond the territory owned by the Romans at the time. Africa in the context of the law appears to be composed of various categories of territory created as a result of the Third Punic War. It includes 'land which was within the boundaries of the free peoples', that is, Rome's allies in Africa; land previously given to perfugae, those who deserted from Carthage to the Romans; land 'granted or assigned to the stipendiarii', instructs a Roman official (duumuir) to make an equivalent restitution if any land of free peoples or the perfugae had been granted to a Roman citizen (lines 74-7) and to ensure that land granted or assigned to the tribute-payers was to be recorded on the public maps (in formas publicas, line 78). It appears from the above that Africa in the law of 111 is conceived of juridically rather than in a strictly territorial sense. 28 It may then be defined as the area of Africa where the Romans exercised power, as Lintott suggests. 29 This picture of Africa emerging from the law of 111 is consistent with the current view that the concepts of prouincia and Roman imperialism developed gradually in the second century. 30

THE EXTENSION OF THE TERM AGER PVBLICVS TO OUTSIDE ITALY
As noted above, the law of 111 speaks of a type of land in Africa called 'private land subject to taxation' (ager priuatus uectigalisque), and it recognizes that a quantity of land in Africa would become private land by this law (quei ager ex h(ac) l(ege) priuatus factus erit, line 80). It also seems that, apart from certain types of land, the rest of the land in Africa would be recorded on 'public maps' (line 82). 31 Whatever the status of 'private land subject to taxation', the distinction between the land to be treated as private and the land to be entered in 'public maps' is clear, and the land to be recorded on 'public maps' cannot mean anything other than Roman public land. No such land in Africa, however, is specifically termed ager publicus in the law; it appears simply as ager in Africa without qualification. Even the land allotted to the colonists under the lex Rubria is described as ager in Africa, not as ager publicus (lines 59-60). Likewise, under the same law, colonies were founded in Africa, not specifically on ager publicus in Africa (line 61). It is also relevant to note that, where the law deals with land in Corinth, the land is termed 'the land or piece of land, which belonged to the Corinthians' (ag[r]um locum, quei Corinthiorum [ fuit], line 96). In short, the land extra terra Italia with which the lex agraria of 111 is concerned in the latter half of its text is never specifically qualified as 'public', and this is in contrast to the earlier Italian section of the law, where the land concerned is carefully specified as ager poplicus populi Romanei in terra Italia in the introductory line 1, and is abbreviated consistently as ager publicus throughout the text. The specific reference to 'the land or piece of land, which belonged to the Corinthians' (ag[r]um locum, quei Corinthiorum [ fuit], line 96) cited above suggests that the absence of the formula ager publicus in the provincial section of the law was not the product of a careless omission by the engraver.
In According to Stephenson, Rome's new conquests in the early Republic kept their ancient names while their lands were distributed to the citizens individually (uiritim) or through colonial foundation or rented out; unassigned part of subjugated territories would then become part of ager publicus. 34 Similar processes may well have occurred outside Italy. Rathbone argues that the concept of ager publicus emerged from the late fourth century, when the Romans began large-scale conquest in Italy, to differentiate the territory annexed from neighbouring states from the original Roman territory. 35 The idea that the concept of ager publicus emerged because of the need to categorize different types of land is worth exploring. In the provincial cases, the emergence of private property out of the annexed land may have caused the extension of the term ager publicus to overseas provinces.
Although land in Africa was made the property of the Roman people on annexation in 146, the fact that it is referred to as a transient geographical name without the qualification of ager publicus in the law of 111 may reflect a relatively low level of state control and private exploitation of African land, primarily through land distributions, until this point. Since the colony of Junonia was stopped, so was the conversion of ager in Africa into private property. Notably, the allotments under the lex Rubria were not confirmed until 111, and measures for exploiting African land after Junonia did not take place until 115 and 113 (lines 86 and 89). Although the clauses are less clear, it appears that these two previous arrangements also needed authorization from the law of 111. 36  previously-was not formally recognized until 111, implying that a large amount of private land in Africa was unauthorized up to this time. As we shall see, prior to 111, private land transactions seem to have taken place. However, the extent of private ownership of African land was perhaps limited by the administrative conditions described above, and there may have been little recognition of the need to treat publicly owned land separately from private land in legal limbo. The conspicuous absence of ager publicus in the provincial section of the lex of 111, where it clearly refers to the annexed land extra terra Italia, may then be explained if the law could treat the annexed land overseas as a single category. The author of the law may have found it unnecessary to specify whether the law dealt particularly with publicly owned land if the extent of privatization of land overseas was insignificant or perceived as such. Literary evidence suggests that ager publicus existed only in terra Italia in Roman legal thought until that time. Konrad draws attention to two important passages in Livy and Dio with regard to the status of ager Romanus in the Republican period. 37 According to Livy (27.5.15), when the consul in Sicily proposed to appoint a dictator in 210, the Senate maintained that any such appointment could not occur extra Romanum agrum-eum autem Italia terminari. This episode shows that by 210 the rules were that ager Romanus was confined to Italy, and that any territory located extra Italiam could not be converted to ager Romanus. If the Senate insisted that ager Romanus was exclusively Italian in 210, it is hard to imagine that while doing so it recognized some land that existed outside Italy as ager publicus. It is possible that ager publicus likewise was confined to Italy until the late second century, especially given that the first systematic attempt at the exploitation of overseas territories through land distributions, the foundation of the first overseas colony of Junonia, began only in 123/2. Dio (41.43.2-3) suggests that a small plot of land in Thessalonica was converted to ager Romanus for the auguries in 49, and that it was under unusual circumstances. These two passages show, first of all, that it would be anachronistic to think that ager Romanus comprised of both Italy and Rome's overseas territories under the Republic. They further suggest that the rules about the qualification of ager Romanus had been relaxed between 210 and 49, and that the concept had extended to outside Italy. This adds weight to the possibility that the annexed land overseas was not counted as part of ager publicus before Cicero's day, and that ager publicus likewise was confined to Italy.
The lex of 111 suggests that after the Romans annexed territory overseas it was not immediately termed ager publicus, a phrase confined to Italy if the above reconstruction is accepted. It seems that the annexed land was called for some time simply the ager in a certain location or named after its previous owner. This is not to say that the annexed land was still treated as ager peregrinus; it was made the property of the Roman people (Cic. Leg. agr. 2.58). However, the type of land termed ager publicus may well have existed only in Italy by the formulation of the epigraphic law, as ager Romanus did at least by 210. There was a juridical distinction in the law of 111 between land in Italy, where the division between public land and private land was well established, and the relatively newly acquired lands overseas, which seem to have been conceived more as conquests. Where the law of 111 deals with land in Italy, it concerns both public land and the private land of citizens and allies affected under the Gracchan agrarian reforms; therefore, a clear distinction has to be made between the two categories of land. The public category has to be further specified-namely, the land that existed as public in 133. As with the land in Greece, on the other hand, the law of 111 directed a duumuir to ensure that part of Corinthian land now be measured out and boundary stones erected (lines 96-7). Saturninus' law in 100, if not in 103, was the first known attempt at founding transmarine colonies after the lex Rubria, and Achaea was one of the areas for colonization, along with Sicily and Macedonia. 38 It is tempting to relate 'the land which belonged to the Corinthians' in the lex of 111 to Saturninus' choice of Achaea for colonization in 100. 39 No colonies, however, resulted from Saturninus' law, according to Cicero. 40 More progress was made on African land before 111. The founding of the colony of Junonia was evidently under way before it was halted. 41 Furthermore, the lex agraria provides a valuable glimpse into economic activities in Africa and how ager in Africa was being exploited on the eve of the Jugurthine War (112-106). There are also indications that the market in land and the privatization of ager in Africa as a result would now be stimulated by the effect of the law of 111, as will be discussed in the following section.

THE DEVELOPMENT OF PRIVATE LANDOWNERSHIP IN AFRICA AND DISCUSSIONS AT ROME
As noted above, the lex agraria of 111 dealt with two main categories of land in the African section: land assigned to colonists under the lex Rubria, legislated in 123 or 122 (lines 48-61), and land publicly sold by the state ( publice uenire) in 115 and 113 (lines 85-9), which could then become ager priuatus uectigalisque. The lex implies that, after land had been assigned as a colonial allotment or publicly sold, private sales of both types of land had been going on for some time (lines 61-2, 63-5, 68-70). 42 The lex not only confirms the allotments of the original colonists of Junonia, but also sees to it that the interests of the buyers in previous transactions will be looked after in various contexts (lines 58-61, 61-2, 63-5, 66-70). It appears that the buyers of African land sometimes risked investing in land whose legal status was not incontestably certain: a duumuir was to compensate the buyer with an equivalent amount of land if he bought land which in fact was not a colonial allotment (lines 68-70). Where the law seems to 38  concern the uectigal imposed on ager priuatus uectigalisque, it orders that the money is payable from the next round of the tax year (lines 70-2), allowing a reasonable time for the payers to prepare the money due. A general consensus is therefore that the uectigal imposed on ager priuatus uectigalisque was not a token rent but a real sum. 43 If this is correct, the buyers of African land must have expected that the return from ager priuatus uectigalisque would be higher than the payment of uectigal. Cicero speaks of Romans and Italians owning large landholdings in Africa for cash-crop agriculture in the late second/first century. 44 It seems that the law of 111 is a strong testimony to Roman interest in African land in the period between the lex Rubria and the Jugurthine War, an interest which may have been linked to the creation and growth of large estates such as those mentioned by Cicero.
The numismatic evidence shows that the fall of Carthage brought a significant change in the precious metal currency of Africa: in addition to a complete disappearance of gold coinage, three of the four hoards found in what can be perceived as the Roman province after 146 consist exclusively of Roman denarii, indicating that Punic silver was entirely replaced by denarii. 45 That the earliest of these hoards (at Henchir-Djebel-Dis) dates to about 126 is worth a moment's attention. The Romans did not mint coin in Africa until 47, and therefore the denarii must have come from Italy. It is possible that some of the denarii were sent to support the Roman army during the Jugurthine War and the Civil Wars. However, from what can be adduced from the above reconstruction, one hypothesis might be that these silver coins are suggestive of Roman/Italian investment in large estates in Africa from the late second century onwards and in settlements of veterans in later stages.
Although the Junonian scheme could not be implemented further, surveying and centuriation were carried out in Africa on a vast scale under the direct supervision of Gaius Gracchus, creating favourable conditions for the sale and leasing out of land in the future. This should be linked to the auctions at Rome in 115 and 113, which the lex of 111 mentions. It does, therefore, seem that wealthy Romans or Italians in the late second century were keen to buy up purchasable land in Africa, making use of the Rubrian land divisions. It also seems that they were willing to accumulate ager priuatus uectigalisque despite the uectigal, where they could create new large estates for commercial farming, ushering in the vast estates for which Africa was famous in the later Imperial period. 46 That said, however, the fact that the allotments at Junonia were only reviewed and confirmed as late as 111 and the errors and confusions that the epigraphic law intends to address suggest low administrative activity in Africa since the Junonia scheme ceased. It seems that the task of dealing with the Gracchan legacy had been left with little intervention before the enactment of the law of 111. 47 This is where the epigraphic law marked a crucial turning point in the exploitation of African land and the development of private landowning in Africa. The lex of 111 must have had the Jugurthine War in sight, and the effect of its measures was to encourage future land purchases. 48 It establishes a formal African landholding register on declaring war on Jugurtha, and thereby safeguards the interests of previous buyers of the land. It is also worth noting that the law declares old Punic roads which existed before 146 and which seem to have been connected to 'the limites between the centuriae' public (lines 89-90). These centuriae were in all probability laid out under the lex Rubria. The law of 111 again confirms an existing system and tries to pre-empt the debate on the status of the ownership. In short, the lex agraria seems to imply a situation in which considerable amounts of Roman capital had been invested in ager in Africa after the lex Rubria, if not earlier. It is reasonable to suppose that, promoted by the law of 111, private ownership was steadily established in Africa out of the land annexed in 146 through the cycle of sale, inheritance and, in some cases, abandonment. The pace of the development of private ownership may have accelerated after the conclusion of the Jugurthine War, thereby resulting in the need to specify what became private property and what remained in 'public maps', which seems to have been first compiled under the lex agraria of 111. 49 As noted above, the Gracchan commission admitted 6,000 families at Junonia, more than the law initially allowed. That the lex Rubria assigned colonists 200 iugera of allotments is drawn from the lex agraria (line 60), and Hobson estimates that the territory of Junonia would have been 3,035 km 2 without subseciua, assuming that the 6,000 colonists received equal plots of 200 iugera. 50 Previously, there were occasions where the colonists of citizen colonies received an equal but very small land. 51 However, the foundation of Copia in 194/3, which is recorded in more detail, shows that the normal practice was to assign land according to the recipient's military rank. 52 The more helpful point of comparison for the uniquely large Junonia is the size of the allotments in larger Latin colonies established in the early second century and, in most of the known cases, equites received plots at least twice as large as those of pedites and their numbers were roughly a tenth of pedites. 53 Lines 59-60 of the epigraphic law of 111 contain two closely related clauses reviewing the size of the allotments to colonists with parallel phraseology, which may reflect that different amounts of land were assigned under the lex Rubria. 54 The 200 iugera in line 60 of the lex agraria seem to indicate the maximum allotment under the lex Rubria, assigned to the members of the equites, 'the most respectable citizens'-to quote Plutarch (C. Gracch. 9.2)-who were presumably expected to serve as the governing class of the colony. 55 The sum of the colonial allotments may have been therefore smaller than 3,035 km 2 , although this does not preclude that the total area of centuriation, including subseciua, was much larger than the allocated units. 56 It is worth reviewing Saturninus' agrarian law in 103 in this context, which assigned Marius' veterans 100 iugera of allotments in Africa. 57 The 100 iugera allocated under Saturninus' law in 103 should also be taken to mean the maximum allotment for the equites, not that the veterans received allotments each of 100 iugera, as often presumed. 58 It is important to note that the 100 iugera in Saturninus' law was half of the 200 iugera of allotments under the lex Rubria. Also relevant is that 'colonies' claiming a Marian foundation have been identified in Numida, in the fertile valley of the middle Bagradas, outside the territory of the Roman province. 59 It may be that, encouraged by the agrarian law in 111, most of the land in the province of Africa had already been sold or tied up by lease by around 103, perhaps except for the urban centre of Carthage. 60 It is also attractive to conjecture that the price of land in the province of Africa rose progressively in this period, stimulated first by the founding of a colony and pushed up subsequently by rises in sales, especially after the Jugurthine troubles were over. The limit of allotments in Saturninus' law of 103 to 100 iugera, half the size of the previous Junonian allotments, and the puzzle of Marian settlements within the kingdom of Numidia could be explained by assuming a shortage of available land for distribution in the province of Africa at the end of the second century. Saturninus' second agrarian law to found colonies in Sicilia, Achaea and Macedonia in 100 also deserves comment. 61 It is easy to identify Gaius Gracchus' influence on Saturninus' legislative programme in general and, above all, Gaius Gracchus' idea of creating a Roman colony overseas was significantly expanded in Saturninus' second agrarian law to establish multiple colonies outside Italy. Saturninus' scheme met with fierce opposition. Yet it may have generated discussion of Roman land extra terra Italia as a whole and contributed to the expansion of the concept of ager publicus to outside Italy in the Roman legal thinking.
It may also be relevant to note Cicero's reference to a treaty ( foedus) made under the consul C. Aurelius Cotta in 75 (Cic. Leg. agr. 1.11, 2.58). It granted Hiempsal II of Numidia possession of some agri publici in the coastal region, which 'P. Aemilianus had assigned to the Roman people'. It is tempting to link these coastal agri publici in possession of Hiempsal II in 63 to the lands which Scipio Aemilianus allowed Massinissa's family to 'have or exploit' according to the usual restoration of the text of the lex of 111 (line 81). The treaty arranged by Cotta, however, was not ratified by the Roman people. Cicero is our sole source for this event, and we do not know its terms and conditions or how references to these coastal lands were phrased in it. Yet Cicero suggests that this treaty was favourable to the Numidian king, so was controversial at the time and was discussed both in the Senate and before the Roman people. We may at least assume that the discussion of the treaty with Hiempsal II renewed attention to the status of ager in Africa and extra terra Italia in general, the ownership of which was in hands of the populus Romanus. This may have been another occasion for prompting the legal formalization of the conquests overseas at Rome, if the term ager publicus had not extended to Africa by 75. Cicero's writings suggest that by the time Verres was prosecuted, and more clearly when Rullus proposed his rogatio, a concept of ager publicus extra terra Italia was well established at Rome. It is conceivable that the land in Africa recorded on 'public maps', complied under the lex agraria of 111, was one of omnes agri publici of which the consul Cicero was aware. That the Numidian king was eager to obtain formal recognition of his possession of a part of ager publicus may reflect the tighter grip of the Romans on African soil by the first half of the first century.

CONCLUSION
The lex agraria of 111 has been primarily used as evidence for post-Gracchan Italy. While the law of 111 was seen as a winding up of the Gracchan agrarian reforms, the forward-looking aspects of the law might have been overlooked. I have sought to argue that private landowning had been steadily developing in Africa before the first century out of the land which was annexed in 146 to the degree that it became necessary to specify for administrative purposes what remained in public ownership as ager publicus as opposed to mere ager in Africa. Two agrarian laws from the period between the death of Gaius Gracchus and the battle of Thapsus, the epigraphic law of 111 and the rogatio Seruilia in 64/3, show that the term ager publicus, which seems to have been confined to Italy by 111, was extended to include all lands in public ownership outside Italy. Privatization of overseas lands which the Romans had annexed, as reconstructed above with the case of Africa, may have come to influence the scope of the term ager publicus. The shift in formulae from ager in Africa on the eve of the Jugurthine War to quicquid igitur extra Italiam quod publicum populi Romani factum sit in 64/3 for the former Carthaginian territory suggests vigorous private exploitation of African land and consolidation of the Roman rule on the region in this period.