STOP Early Journal Content on JSTOR, Free to Anyone in the World This article is one of nearly 500,000 scholarly works digitized and made freely available to everyone in the world by JSTOR. Known as the Early Journal Content, this set of works include research articles, news, letters, and other writings published in more than 200 of the oldest leading academic journals. The works date from the mid-seventeenth to the early twentieth centuries. We encourage people to read and share the Early Journal Content openly and to tell others that this resource exists. People may post this content online or redistribute in any way for non-commercial purposes. Read more about Early Journal Content at http://about.jstor.org/participate-jstor/individuals/early- journal-content . JSTOR is a digital library of academic journals, books, and primary source objects. JSTOR helps people discover, use, and build upon a wide range of content through a powerful research and teaching platform, and preserves this content for future generations. JSTOR is part of ITHAKA, a not-for-profit organization that also includes Ithaka S+R and Portico. For more information about JSTOR, please contact support@jstor.org. THE "DRY LAND" IN PTOLEMAIC AND ROMAN EGYPT By W. L. Westbkmann The period of Egyptian history covered by the papyri written in Greek covers, roundly speaking, the centuries from 300 B.C. to 700 a.d. Throughout this thousand-year period- the taxes and rents obtained from the agricultural land of the Nile Valley were exacted by the governments, whether Greek, Roman, or Byzantine, according to the same general method of assessment. This method was one of rating the tax or rent upon the productive quality of the soil which, in its turn, depended largely upon the question of whether the land could be irrigated or not. For this reason the land surveys always reported the land under three large general classifications, flooded (usually called fiefipeyfxevT)), unflooded (&/3poxos), and dry (xep$ vdup or koB' vbaros, e/xppoxos, or kotA- |8poxos), in which case it could not be worked in that particular year. A protracted period of sogginess would make the soil eventually too salty or alkaline (dXjuvpts). In that case the land would be perma- nently virokoyov until brought back to a healthy state by drainage and proper working. It is entirely clear that such alkaline soil could be brought back to a state of partial productivity. 5 But the methods employed by the Egyptian farmer to this end are not appar- ent in the extant papyri. 1 Much confusion has reigned regarding the xkpaos, chiefly due to unclarity of statement arising from the lack of a strict definition of the agricultural-fiscal termi- nology. See Rostovtzeff's statement in "Studien zur Geschichte des rSmischen Kolonats," Leipzig, 1910, p. 3, where he contrasts the x*poos with the fltfjptyukvri as that land which is not flooded and therefore is not cultivated. The Introduction to P. Lond. II, 483, p. 324, proves clearly that the x^pcos is capable of production. Ros- tovtzeff, of course, understood this fact, as is proved by his further treatment in Kolonat, pp. 7 ff . 2 For the drainage canals and their importance see Westermann, Classical Phi- lology, XV, 126-27. 3 P. Teb. I, 616, 132-36. 4 Schubart's emphasis of the inroKoyop and yij iv dperg as being the two important divisions from the standpoint of production (Einfiihrung, p. 408), seems to me to introduce confusion into the subject. I think also that the distinction which he attempts to make (p. 413, with reference to Wilcken, Papyruskunde, Chrest., I, 2, No. 341) between iuflpoxos and ko.9' Maros is unsupported. He seems to have misinter- preted Wilcken's differentiation of the j3e/3pe7juewj and that portion of it from which the water has so far drained off that it could already be plowed. Wilcken means only to distinguish the ffe^pey/iivri which has not yet drained off, from the Ka0' vSaros , or water logged, portion of that year. 6 P. Teb. I 61b, 9-18 and 121-31, where we have 16J arourae of crown land which had been declared &\fivpis in the fiftieth year of Ptolemy Euergetes II and were rented out for reclamation in the fifty-third year. The crops obtained from them in the fifty-third year were wheat, barley, and fenugreek, 11. 17-18. "Dry Land" in Ptolemaic and Roman Egypt 23 The unflooded land (afipoxos) I have explained as that which was not at any time covered over with water in the year in which it was declared to be &j8poxos. Though unflooded, it could be irrigated — by digging in deep ditches from the flooded portion or by using the water wheel. 1 Out of this category of land there would be no koltL- /Spoxos or 2/ij8poxoj. The only portion which would be highly produc- tive would be that which was eirtjvTKrip.evr], artificially irrigated by methods much more laborious than those required for irrigating the "flooded" land. The verb avrXeiv seems to be used in the papyri in the general sense of "irrigating" by any method, 2 and the £iravT\r]Tds yij of P. Rev. 24, which is planted to vineyards, was pre- sumably irrigated by use of the sakje. The persistent appearance of the participial form iirrivr'Xrjfji.ivr) only in connection with the appoxos yrj confirms my belief 3 that, in its technical meaning, it is strictly limited to that portion of the afipoxos which was artificially irrigated.* The outstanding documents in support of this view are: 1. Nine offers to lease government land icar' a{-iai>, "according to value " (i.e. according to actual production, as I hope to prove in a later study), following a decree of the newly installed Emperor Hadrian. 6 In eight of these the peasants insert a clause to the effect that they will not pay on any of the land at the rating for &Ppo%os yrj, but will pay one-half on that portion which is eirrjvTXruievrj. 6 2. Two lists of proprietors and tenants, chiefly of appoxos and iirrivT\rip,evr] yrj, published as P. Mendes Genev. and P. Ryl. 426. In the former important document 7 where the unflooded land and the irrigated portions thereof (iwrivrXrifievr]) are given, both are 1 Classical Philology, XV, 130-31. s I have been unable to determine any distinction in the papyri between avrktiv and TOTifeiv. For avrXtlv see P. Flor. I, 150; P.S. I. 60, 15, itrixavt/v avrKowrav; h.vThr)oubv /3o0j>, P. Flor. I. 16; dyrXiyriKa KT-hnara, P. Flor. 1, 148; avrXtiros 7?, P. Amh. 96. In P. Oxy. XIV, 1675, 11, iiravrKtiv is definitely used for water-wheel irrigation: f evyos fioQv els eiravrkeiv rb Krijfm. 3 Already expressed in Classical Philology, XV (1920), 131; XVI (1921), 184. 4 fcravrXijTos, however, may be used of the irrigated parts of the "dry" land. P. Rev. 24 and Wilcken, Papyruskunde, I, 1, 273. 5 P. Giss. 4-7 and P. Ryl. 96. 6 I suppose that this clause means that the peasants will pay one-half of the ifipoxos rate on that portion of the S/Spoxos which they have been able to irrigate. 7 Published in full in Stud. Pal., XVII, in so far as the painstaking and intelligent labor of V. Martin could restore it. 24 W. L. Westermann followed by references to the column of some other list in the local registry office where the peasant's declaration (&Troypar)) regarding the land could be found. 1 We have no declarations of hrqvTkruiivq yrj, but quite a number regarding the unflooded land. The extant evidence, therefore, leads to the conclusion that the "declarations" of the peasants here mentioned refer to the declarations of their afipoxos yrj. That is, the ewr>vT'\r)nevri was regarded as a part of the afipoxos yrj. 2 In the annual determination of the tax or the rent upon the unflooded land the first step was the declaration, by the peasant owner or by the lessee of government land, of his estimate of the amount of his farm parcels which should pay as "unflooded." 3 During the planting season the government officials conducted their inspection (iTfiffne\pi.s), checking up the actual amount of the flooded, unflooded, and dry, in preparation for the report of the village scribes upon the amount of revenue which was due the government in land taxes and rents from each village district. In line with the policy of the government, whether Ptolemaic or Roman, of forcing production of grain upon the ikPpoxos to the limit of its possibilities, 4 it was one of the duties incumbent upon the inspec- tion officials to see to it that the greatest possible amount of the pro- duction category called "unflooded" should be additionally irrigated (eirrjvT'Krifxivr}). P. Ryl. 82, I take to be a report of such inspection 1 P. Mendes Genev. 155-56, nai o(i.tik^s) a (apra/Si)*) ha(e) Bix«0eZ(s) 'CW W>pios), irnivr[\(iinkvTii) (apovp) .... (3d h.) eiBivnerplas) koKtijs) a, (1st h.) anoyp{a.(jx>iiiv). Cf. many similar passages in this document. 2 P. Lips. 105, republished by Wilcken, Papyruskunde, I, 2, No. 237; Stud. Pal, V, 119 recto, VII; P. Oxy. XII, 1459, 9, 36; P. Ryl. 82, 10-15. 3 There is no evidence to show that he declared the iirr)vTKrtii'ant, as is assumed by the editors of P. Ryl. 426. The peasant probably estimated the amount of his afipoxos on the basis of the experience of previous years. He could not know the amount of the Sj3poxos. Much less could he estimate in any way what portion of this uncertain amount he might be able to irrigate. Cf. Wilcken, Papyruskunde, I, 2, No. 237 where an inspection official makes a hurried preliminary report to a superior, in order to relieve his anxiety, that out of 1,850 arourae given in by the village scribe to the inspection officials as kmivrKruitin] and afipoxos, the subsequent inspection had shown only 127 arourae as being actually of this category. See Wilcken's excellent intro- duction. 4 Westermann in Classical Philology, XVI, 177. "Dry Land" in Ptolemaic and Roman Egypt 25 officials, who are here called ewi.fie\r)Tal tup eirl rod \ifivaafxov. 1 They take an oath that no aroura around their village district is unflooded (ixyriav Lpoiipav afipbx*<- v n^pi rijv ripJbv KUfwypanfiarelav) except 55 arourae. "We swear that we will cause all of these which can be additionally irrigated {&v rd Swafieva ewavr'KrjOrjvai. ewavajKa^eiv iwavT\eiadai.) to be additionally irrigated." the xlpvos The "dry" land in Egypt was that land lying in the irrigation basins along the Nile which in any given year was not reached and overspread by the Nile waters at the time when the sluices in the canals (diupvyes) were opened and the water from the canals was let in upon the fields. 2 Its difference from the "flooded" land GSe^pey- tievTj) is obvious. It was distinguished from the "unflooded" by the fact that it could not be irrigated, in the particular year in which it was "dry," by ditching in from the "flooded" section of any basin. 3 The condition of being " dry land " was not necessarily a constant one, although it would tend to persist in years of normal inundation. In years of abnormally high Nile those portions of each basin which were regularly x«P°"°s might become either "unflooded" or "flooded." 4 Conversely in years of low Nile some of the normally "flooded" and "unflooded" land in the basins might well, at the inspection, prove to be "dry." The xep^os must have had, of course, a top soil of silt just as the "flooded" and "unflooded" had. That the x«P°"°s could be made to 1 The title of the editors, "Declaration of Irrigation Officials," is, I think, incor- rect. 2 For the method of letting the water upon the land see Westermann, Classical Philology, XV (1920), 124-31, especially p. 124, notes 2 and 3, where I have recorded incorrect explanations of the x'po'os. 3 The xtpov-Ppoxos, of P. Ryl. 207a, has so far appeared only in this one place and I am unable to suggest any further explanation than that which the word implies. * B.G.U. Ill, 831 of 201 a.d., an offer to lease 18 arourae of government land which is ordinarily dry shore land. The lease stipulates that if any part of these arourae should become ftftooxos or ko6' Maros, 11. 14—15, the lessee will not be com- pelled to pay on the fi/3poxos portion the rate of rent for that type of land, and pre- sumably not at all upon the na$' iSaros. The presence of land which was nod' vSaros necessarily implies that a portion of the farm had been flooded (flefipeynkvti) in that year; cf. P. Oxy. VI, 918, col. v, 1. 16, where we find xkpaos in a hollow which is now under water; P. Lond. II, 350, p. 192; P. Pet. Ill, 99. 26 W. L. Westermann produce a revenue, in grain or otherwise, to the owner or renter, should be clear to anyone who has used the papyri. 1 It obtained remission of the regular tax or rent only by official action, granted for a definite and accepted reason. 2 As was the case with the land of the flooded and unflooded categories, but much more easily, the dry land might become unproductive (inroXoyov) . In that case it was clearly designated as vroXoyov, or sometimes "dry bringing in no revenue" (xfycros a4>opos). s The explanations given for reversion to an unproductive condition are interesting. The land became unpro- ductive of course if it was not sown, 4 or if it were covered with sand 6 or with a low growth of bushes, 6 or if it lay too close to other unpro- ductive parcels. 7 The general impression left by the papyri is that production upon the "dry land" required greater effort and gave distinctly less return for the effort expended than production upon the "flooded" or "unflooded" land. The continual struggle required to keep the dry land in productive condition by copious use of the water wheel and constant working is shown by the special name given to some of the dry land — x*P ff oOpvi.Tis* — dry land overgrown with rushes. It is obvious that the first type of the arable land in Egypt which would lie uncultivated would be the x«pcos for the very reason that it gave 1 P. Teb. I, 74, 21-30: Menches, the village scribe, has been able to lease out some of the x^P^os. The grants of land to guards and catoecio cavalry are regularly from the xlpvos, as in P. Teb. I, 61a, 41. Cf. B.G.U. Ill, 831, an offer to lease dry shore land, aird x*pi> at-yiaXoO: B.G.U. II, 571, xkpaos which had paid a rent of 2 drachmas; P. Oxy. IV, 740, 46, where the a/3poxos and x^paos are sown. 2 P. Oxy. XII, 1434, 19, where some x£P<*°s ' s mentioned in an official \6yos KovoTt\eidv without statement of the reason therefor. 3 P. Lond. II, 267, p. 135, 1. 149. 4 B.G.U. II, 560, 26 cunropov x*P<">v\ P. Ham. 12, 9; P. Teb. I, 616, 144 &avbpov Kextpaewivys, cf. 74, 29; P. Cair. Preisigke 47, 7-9, and note. 6 P. Oxy. VI, 988 x«po-<*MM<>u; P. Ham. 12. 6 P. Lond. II, p. 139 cb-6 p,vpmSiv (tamarisk growth) p.t)5tv bwankvoiv kpa.v. 7 P. Teb. I, 74, 56, of 114-13 B.C., an accounting of the unproductive land made by the village scribe Menches, x^P"ov SiA. rd irapantioOai. rrji. Xowriji cufrcipce; cf. p. 75, 1. 74, where we have the same reason given, and written in above the line, koi SiA &tloi)i " because grown over with roughage." 8 P. Oxy. X, 1347; P. Flor. I, 64, 22. Cf. 71781(0?) x*p™8pbia Wilcken, Griech- ische Ostraka, Leipzig, 1899, II, 1224, and xepvov toXXi)? naldpiiov in Stud. Pal., V, 7, 16. It is to be noted that even the wild growth, such as dpiop, could be used. See the bank receipt for rushes (Opba) and reeds (icakapios) in P. Pet. II, 26, 4 and 5 (cf. Ill, 64). "Dry Land" in Ptolemaic and Roman Egypt 27 the lowest returns upon the greatest expenditure of effort. If it were not worked, just as with the fitftpey pivti and the afipoxos, it would grow up with reeds and roughage. So strong was the tendency on the part of the renting farmer to neglect the dry areas of the plot he had leased that many of the leases of private land contain a pro- vision under which the lessee agrees to return the land "free from dry land, roughage and dirt of all kinds." 1 Given the proper facilities in respect to wells, reservoirs, and water wheels, and the requisite labor to make use of these, there was no reason why "dry" land should remain in that category, from the fiscal standpoint of being relatively unproductive. But if the use of these irrigation aids were neglected the land would quickly revert to a dry condition. Such a process is admirably illustrated by the contract of sale presented in P. Oxy. XII, 1475. 2 This document records the sale of a plot containing 11 arourae of grain-producing land and another unit consisting of two houses in the village of Paimis, some bare ground, and an orchard covering ff of an aroura. All of this smaller unit when taken as a whole (vrb p,iav 5\(/ip, 1. 22) abutted upon the northeastern corner of the 11 arourae grain plot. I have located the irrigation tank (or cistern) 3 and water wheel necessarily in that part of the farm which was farthest removed from the irrigation ditches, that is, in that part which would most quickly go dry. The well (piap) has been arbitrarily located near the two houses. The tank was out of repair when the farm was sold, the water wheel partially so. 4 This no doubt explains why the orchard was recorded in the deed of sale as "now dry." 6 1 P. Ryl. 166, 27, nadapop awd x^poov, aypixrTtws, Skttjs ir&mjs; B.G.U. I, 197, 28, xipaou &Ypd>s SUrijs itckttjs $piov, as properly restored by Preisigke, Berichtigungs- liste der griechischen Papyrusurkunden aus Aegypten, Strassburg, 1913, p. 26. 2 The date is 267 a.d. 3 For Xdiotoj as "cistern," frequently with the water wheel (6pyavoii) , see A. Calderini in Aegyptus, I, 310. 4 P. Oxy. XII, 1475, 16-17. 6 Ibid., 1. 20. Cf. P. Lond. Ill, p. 260, a lease of 3 arourae of dry land planted to vineyard which also contained an irrigation cistern, a watering system, and a water machine; and note further the paradoxical phrase "the irrigated dry land" in a loan contract of 14 a.d. (B.G.U. IV, 1132, 13), and the lease of an entire island, "except of the dry land as much as they may irrigate according to the survey," in P.Hib. 90, 8, of 222 B.C. 28 W. L. Westermann Several special types of dry land appear, distinguished from the ordinary x&pvos as "dry wood land" xepow ^uXirts, 1 "dry vine land" XepAms is marked as unproduc- tive, 4 which would imply that it was customarily productive of some revenue to the state. This conclusion is strengthened by an official Ptolemaic list which gives the record of loans of croton seed by the government to the peasants, in which two Egyptians have been granted croton seed for planting "in the wood land," around an Isis shrine. 5 Other suggestions that a tax or rent was levied on wood land are to be found in the %vKiKr), which was probably an import duty on wood from Lycia, 6 and the remission by Euergetes II in 118 B.C. of the penalties imposed upon those who had illegally cut down wood on their own properties. 7 The apcLKoxepvos is sufficiently explained by its name as dry land planted to aracus. Aracus is a leguminous fodder plant, 8 which in the reports of the Ptolemaic village scribe, Menches of Kerkeosiris, appears along with the hays (xopros and xop'rovop.a.'C) among the "green stuffs" (xXwpa), whereas clk6s (lentils) appears with wheat and barley among the "foodstuffs" (alros). 9 With comparative cer- tainty we may call it alfalfa. The only data available upon this " dry- aracus land," in addition to the x*P)s) occur, and a very small amount of vine and garden land; see 1. 22. 2 41J arourae, col. I, 2; 8J arourae, col. Ill, 7; 32 arourae, col. Ill, 15; 40 odd arourae, col. Ill, 18. 'Translated by Viereck in Deutsche Rundschau, Vol. CXXXVII (1908), 105. If this document is otherwise published I have not found it. 4 P. Lond. II, 483, pp. 323-29. 6 Ibid., 11. 50-51. 6 Ibid., 11. 56-58. 7 Ibid., 1. 15. 8 See 11. 42-43, Kal X&kkous ivupi^ai. 9 Ibid., 11. 77-79. 30 W. L. Westehmann payment of two angia (10 pints) of new wine. There are no pro- visions about the grazing rights upon the dry land. They pertain consequently to the lessee. P. Amh. 94, a contract of partnership in the lease of some state land, 1 is unfortunately mutilated, but the designation of the arourae concerned as " dry hay land " (x o P Tev X^P ffov o-povpuv line 6) strengthens the conviction of the use of the xfyvos f° r grazing purposes. To this is to be added the certainty of the assignment of a cleruch holding of 100 arourae in the Fayum district in the thirty-second year of Ptolemy Philometor from dry land which is also distinctly designated as "pastures." 2 The weight of the evidence at present extant upon the dry land emphasizes also its adaptability and its use, under proper irrigation, for intensive viticulture. In the Ptolemaic period the well-known decree of Euergetes II, dated 118 B.C., granted five years' tax exemp- tion, and four years' additional lightening of the customary tax, to all those who would plant vines or gardens in the " irrigated and dry land." 3 This dry land, which was either cultivated or cultivable with vines, is called anTeKlns x&poos,* or more commonly xepcaM 71 "*- Xos. 5 The latter usage is found in documents dating from the time of Augustus to the seventh century of our era. From the Arab period we have a list of individuals paying land and poll taxes in which an entire district appears under the name Xepaapirehos. 6 Where ajuireXiKa xwpia appear they stand in close proximity to dry land. 7 For this constant close relationship of vine culture to the xepvos a Ptolemaic list of temple-lands (quoted above) is help- ful. 8 The entire amount of the land of Magdola as given in the 1 The date is 208 a.d. 2 P. Teb. I, 79, 7-9, &t6 vop&v [t5>v krAs /»ur]S(W£as xpds xoXwiv otoiKov/iivav [ ] x^poov opoupoi p. Possibly also in 79, 29-30. 3 P. Teb. I, 5, 93-99, robs ytapyovvTas koto tvv x&P'"' IV" iffreKirw )} xapaSeUrovs 8s 8.V KaTOApVTtixjwnv iv rrji KaTaKfkKvaixkvrii. Kai KtxtpwP&W" 4 P. Teb. I, 82, 4; B.G.U. IV, 1158, 8. 5 B.G.U. IV, 1158, 8; P. Oxy. Ill, 506, 25; P.S. I, III, 240, 8; P. Ryl. 222, 12; B.G.TJ. IV, 1031, 9; Stud. Pal., XVII (P. Mendes Genev.) 1. 488, xvajuTrtXoj otpou; P. Cairo Cat. 67097, recto 1; 67104, 7, 11. 6 P. Lond. IV, 1419, 104, 699, 973 (here tottos X«Xs with vine culture is further illustrated by B.G.U. 177 of 46-47 a.d., in which the boundaries of four "Dry Land" in Ptolemaic and Roman Egypt 31 synopsis was 170 arourae, of which 150 arourae were held by the temple of the god Suchus. These 150 arourae are classified under the general caption "dry vineland," &/MreX?m x^os. It is evident that this must mean dry land adapted to vineculture, which however is not necessarily or exclusively planted to vines. For this "dry vineland," as the report of the inspection shows, contains "vine- bearing," 1 olive-bearing, dry, 2 and hay parcels, as well as a small acreage in wheat and barley. 3 A vineyard maintained upon irrigated dry land reverts, if neglected, to the condition of "dry available as vine land," as in P. Oxy. Ill, 506, where two sisters mortgage l£f arourae which had formerly been an ifiirekiKdv Krrjua, but was at the time of the loan contracted by these sisters merely "dry land suitable for a vineyard." 4 If constant care and effort were not expended upon this xepffttMT^os the vines dried out. The result was xepcrajuireXos inr6\oyos. Conversely, if irrigated, such dry vine land might again be restored to productivity. 6 The x^pcos was also used for truck gardening, 6 for palm growing, 7 and even for grain production. 8 The soil could be used for brick. 9 arourae of dry vine land are given as follows: north and south dry land, east a vine- yard and dry land. Cf. Stud. Pal., V, 7, col. II, a vineland farmstead of 22 arourae surrounded by "much dry land and rushes" and supplied with the customary cistern (Xokkos) and a small reservoir (iSpoSoxeiov). 1 Ibid., 1. 10, 4/«reA6<£opos. 2 In the case of irrigated dry land (the Karaictichvanivr) M&'iJ yn of P. Teb. I, 5, 93-98) a subclassification of x*p<«>s under the general heading must, I think, refer to portions which cannot be reached by the irrigation methods used, just as the "unflooded" (fiftjoxos) which is not hrnivrk-rmkyri is still recorded in the land lists as Sfipoxos. 3 I accept the resolution K(pi9ri) in P. Teb. I, 82, 28, as correct. 4 P. Oxy. Ill, 506, 25, rd irplr &inrt\i.Kod KTrmaros vvvl St xep; also P.S. I, III, 240, 8, where we have xepv&pn&os which is cnrcKpopiKV, "grain-producing." 6 See the second-century land register, P. Ryl. 222, II. 8 and 12. 6 P. Teb. I, 79, 22; B.G.U. IV, 1120, 31. Cf. the close connection of the reservoir, dry land, and garden plots in P. Teb. I, 86, col. Ill, 43, 50. 7 P. Amh. 31; P. Oxy. XIV, 744, djuoiws iv xip<"# 6pov, II. 14-15 (first century after Christ); P. Oxy. IV, 740, 46 (about 200 a.d.). 9 B.G.U. IV, 1031, pbvr\aov row Troiijoat itfxiv irKaadrjvai. pvpiadas irXlvSov iv /liv f jj Xcp&aiiTe\ip. 32 W. L. Westermann The constant effort of the Ptolemaic government to bring the greatest possible acreage of the "dry" land under cultivation has long since been known through the publication of the land surveys from Fayum district. 1 As was the case in the official treatment of the "unflooded" land, 2 the interest of the. Ptolemaic government lay in the filling of the basilikon with revenues, both in money and in kind. To this end the land which was potentially productive but lying idle (ager rudis), was assigned to the members of the territorial army or to peasants in general, under leases which were not limited in time, but emphyteutic. 3 For this official purpose the land which was most available was the xep ff os, which by reason of its position in the basins of irrigation, paid the least return to the actual farmer for his economic outlay in seed and labor. In such an agricultural situation we find plentiful evidence of the constant tendency of the producing farmer to abandon the cultivation of that land which brought the lowest margin of profit in comparison with the effort expended. 4 1 have already indicated something of the equipment, in cisterns, pumping machines and reservoirs, which was necessary for successful cultivation of the x^po"°s> except when it was allowed to grow up in rough pasturage. More detailed information is available, however. A lease made out in 103 b.c. of an allotment of land near Kerkeosiris which was divided into three separate plots 5 is most instructive. One Ptolemaeus had taken a sublease of this allotment from a sword- bearer named Horion. Horion was a middleman who held his lease on the property from a certain Maron, the owner. The lease was to run for five years. 6 The sublessee's annual rent was fixed at a lump sum of 120 artabas of wheat for the whole area. The middleman, Horion, agreed to pay to the actual farmer four talents and 3,000 » P. Teb. I, 60-88. 2 Westermann in Classical Philology, XVI, 174, 177. s Rostovtzeff, Rom Kolonat, 30 ff.; Wilcken, Papyruskunde, I, 1, pp. 280-81. 4 P. Teb. I, Appendix I, 3, pp. 555-56. 6 P. Teb. I, 105. 6 This lease was not carried through for the specified five years. Two years later, in 101 B.C., Ptolemaeus, who is also called Petesuchus, rented the same property directly from the owner, Maron, P. Teb. 106. Horion, the middleman, had disappeared from the transaction. "Dry Land" in Ptolemaic and Roman Egypt 33 drachmas (copper) on consideration that Ptolemaeus would break up the "dry" land in the allotment, presumably in each of the three plots. An exception was made of some of the xtyvos in the first section of the farm, where the "dry" land abutted upon the holdings of Thoonis and Pausiris. This was not to be broken up. 1 The sub- lessee had already received an advance payment of 2 talents and 3,000 copper drachmas for this breaking up of the "dry" land. The verb used for "breaking" up the land is xepvoicoweip. 2 The meaning of this word, "breaking up the dry ground" by plowing, is made certain by the analogy of xepaonoirTeiv in a farm account of the period of Vespasian, where a yoke of oxen was used for the purpose. 3 An indication of the usual rough condition of the " dry " land appears in the lease of Ptolemaeus. The lessee binds himself, as frequently occurs in the extant leases, to return the land at the expiration of the lease "levelled and diked up and free from rushes, reeds, coarse grass, and other roughage, except the before mentioned x«p0W-" 4 This exception must refer to the portion of the x«P' ttiv h> rm Kkftput xkpaov itKitv rijs yaTvi^xrqs riji Qo&vios nal Qavalpws yiji. 3 P. Lond. I, 131, pp. 176-77, 11. 225, 241, purS&s po&v .... xepv) of the Ptolemaic royal domain paid generally a rent of 1 artaba of wheat to the aroura, though lands newly reclaimed might pay as low as | artaba per aroura. 3 I judge that dry land would be used much more than the flooded or Hfipoxos as hay fields. One type of pasturage, probably rough growth on "dry" land, was given out by the state without lease for a definite annual money return. 4 A land survey of the principate of Antoninus Caesar shows an aroura of xepcros rjireipos on which the rate of rent had formerly been two drachmas 6 and another aroura of x^pow paying 1 drachma 3 \ obols. 6 x^P^os found in a land register of the early fourth century paid its rent in money. 7 A land register from Aphroditopolis in Upper Egypt of uncertain date, but of the Arab period, 8 gives a summary of the "flooded" (here /cciflapos = |3e/?peY/ueVT?) and "dry" land in a certain assessment group. 9 The "flooded" paid at the rate 1 Wilcken, Papyruskunde, I, 1, p. 170. 2 Ibid., 171; Griech. Ostraka, I, pp. 199 ff., Wilcken explains the acceptance of vegetables in rents and taxes under the Ptolemaic system as restricted to the require- ments of the provisioning of the Court, ibid., p. 201. I judge that he would include under the Verproviantirung des Hofes the current needs of the army. 3 P. Teb. I, Appendix I, 5, p. 563. 4 P. Teb. I, 60, 41; 61a, 154; 79, 7-9 avb voii&v [tGiv eicroj )ii]aBaaeopos). Several types of x«pow are distinguished in the papyri: dry reed land, dry wood land, dry vine land, and aracus-dry land. The extant data warrant the conclusion that the xepvos, when irrigated, was used for grains, for vine culture, hay growing and truck garden- ing; when unirrigated for rough pasturage. The government, both 1 Ibid., 1. 4, KaOap&s apovpai. pvy vdfiuxfia vyy' , -xkpoos apovpai qe vopnapa ufJV, ef. Introduction to 1419, p. 170. 2 In each case only one aroura of the x^pcos is concerned, but Bell has made the calculation on the basis of four arourae for convenience of comparison. 3 P. Lond, 1419, Introduction, p. 171. 4 For the Ptolemaic period see Papyruskunde, I, 1, pp. 170-71; for the Roman period, p. 187. 36 W. L. Westermann Ptolemaic and Roman, desired to bring as much of it as possible under cultivation, especially in grain sorts, granting the idle fields of xqwos to soldiers for reclamation (chiefly by irrigation). There is but little evidence extant upon the rent and taxes exacted from the x*P a °s- These were probably lower than on the "flooded " and "unflooded" land, because of the greater labor required on the part of the farmer to make the "dry" land produce effectively. The evidence is sufficient to dispel the preconception that the xepcos was necessarily, or largely, land of poor quality and low production. It needed only water and careful tilling. Cornell University