A study of Iraq with a view to carrying out prehistoric investigations into the earliest settlements of the northern plain, suggested that one large area offered good possibilities. Not only does it stand away from those regions where, for various reasons, it is impracticable to carry out field work, but it is unknown territory as far as archaeology is concerned. That part of the Hatr a which surrounds Hatra, particularly the region to the west, is rather daunting countryside. Many thousands of square kilometres of slightly rolling, treeless and salty steppelands with seasonal salt lakes and watercourses, brackish springs and waterholes, not to mention an annual rainfall of only about 200 mm. do not seem conducive to early settlement. Nevertheless, one small Hassuna/Samarra settlement, Umm al Dhiabba, situated about 15 km. west of Hatra, had long been known. It seemed reasonable to suppose that more sites would be found in the area. Furthermore, Jebel Sinjar is only three days' walk at most N.W. from Hatr a for an unencumbered man, but moving with sheep the same journey today takes five days. The Sinjar then is close, but more important, beyond it lies the Khabur, the probable diffusion route for that conjectured third region of early cultures lying between those of the Zagros and the Levant-with-Anatolia, and perhaps located somewhere in the Mardin area. Some early settlers fanning out from the Khabur might well have established themselves along the watercourses running down into the Hatr a from Jebel Sinjar as they did across the fertile plain to the north of the area under discussion, as the large Hassuna site at Yarim Tepe and others in the Tell Afar region would seem to indicate. Finally, prehistoric research in Iraq has been largely confined to the Zagros foothills and Tigris banks; the settlements of the central northern plain were unknown. Accordingly the north central Hatr a was chosen for a short preliminary survey using Hatra as a base.

Since the present writer could not go to Iraq and himself pick out samples for investigation of the plant food in the above described site excavated by Diana Kirkbride, a small series of samples was brought to Copenhagen by the excavator. Although this may be described as an emergency measure, the slender results justify the labour involved. A widened investigation would probably improve the picture quantitatively, but hardly very much qualitatively. The question was whether that meagre tract was exploited for agriculture in the Neolithic, and the findings prove that this was the case.

The samples consisted of a small collection of house debris, plaster and clay, as also some earth samples suitable for notation.

Two different categories of house material were examined: a crude, sandy plaster, probably made up of gypsum and clay with many small pebbles in it; and a fine-grained levigated (wall ?) plaster of a yellowish-white colour, to an extent of c. 80 per cent soluble in acetic acid and thus mainly calcareous.

The coarse material was copiously tempered with husk and straw fragments (tibn), but gave few usable imprints. The yellow plaster, heavily tempered with the same material, produced a small number of determinable imprints of spike parts of cereals. All imprints in this material are characterized by a thin rusty-coloured coating representing the decomposed vegetable matter.

BM 135586 = 1971–7–5, 1, until recently in private possession, is a hitherto unknown Neo-Assyrian letter belonging to the royal correspondence of Nineveh, and indisputably an important specimen of its genre. In 39 completely preserved lines, it contains a message to the Assyrian king Esarhaddon (680–669 B.C.) from his son Šamaš-šumu-ukīn, who at that time was the crown prince of Babylon. So far only three very short and fragmentary letters sent by Šamaš-šumu-ukīn in this office had been known, and they are now completely overshadowed by the new text, which is here published by permission of the Trustees of the British Museum. It is of dark grey-brown colour and measures 41 by 82 mm. For all further details concerning the tablet and the script I refer to the photographs and copy published on Plate XIX.

Not having seen the Mandali region for himself, the present writer is not able to give a proper ecological-topographical description of the Choga Mami area. Therefore this introduction must needs be superficial, based upon details published by Joan Oates, supplemented by information kindly supplied by her in personal communication.

Mandali lies roughly 125 km. E.N.E. of Baghdad as the crow flies, not far from the Iranian frontier. It is a staging post on the ancient caravan route to Persia and beyond. The excavation area is fairly near to the town in a north-westerly direction, a flat plain on the borderline of the alluvium in the marginal rainfall zone. The terrain is backed towards the N.E. by two low mountain ranges a few kilometres away, locally known as Jebel Kunah and Jebel Kaum a Sang. Phyto-geographically it is in the middle of what is described as the Persian Foothills Region.

The main source of water is and has always been the small river Gangir that from the N.E. cuts through a gap between the two mountain ranges, supplemented by other, less reliable, seasonal watercourses. Although, geographically speaking, this area is an extension of the Diyala plain, neither the Gangir nor any of the other watercourses of the region reach the Diyala drainage system, but are swallowed up in seasonal mudflats in the intervening desert S.W. of Choga Mami. Even now irrigation agriculture in the district is fed mainly by the Gangir, but on its way into the desert the water gets increasingly brackish and becomes useless for agriculture.

There are two reasons for the regrettable shortage of radiocarbon dates for prehistoric Mesopotamia. One is that much of the relevant excavation took place before the development of such techniques; the second, the inconvenient fact that the prehistoric farmers of Mesopotamia were uncommonly tidy and swept clean their house floors, leaving for posterity little in the way of carbonized or any other remains. Excavations like that at Ras al ‘Amiya failed to reveal any recoverable quantity of charcoal in spite of a diligent search, while at Choga Mami, where every scrap of such material was preserved, many of the samples have proved too small for dating purposes. In this situation it is perhaps useful to publish quickly and briefly the first date from the latter site, remembering at the same time the unreliability of a single unsupported date from any site:

The charcoal sample which produced this date came from what has been termed, perhaps inadequately, the “Transitional Samarra/Hajji Muhammad phase”, so designated because the ceramic material included pottery comparable, and in some cases identical, with ceramic types of these two otherwise distinct assemblages. Although some of the pottery is indistinguishable from that found in true Samarran context, there is no doubt that this phase at Choga Mami clearly post-dates classical Samarra.

In approaching the problem of Archive keeping at Mari, it might be profitable to inspect, at the outset, the archaeological evidence. On doing so, one is immediately struck firstly by the heterogeneous arrangement of objects in storage and secondly by the apparent lack of a noticeable system in organizing the archives. A spot check of other second millennium palaces which produced appreciable quantities of texts reveals this observation to be generally applicable. In almost every major segment of the Mari palace, a cluster of chambers, really store-rooms, contained a collection of tablets, varying in size. In addition, a text or two would be uncovered here and there at unexpected quarters. For this scattering, Hammurapi's victorious troops were often blamed. This accusation is contradicted, however, by the evidence, documented with archival tags, that after its initial victory, Babylon attempted to preserve Mari's archives under a modicum of order.

In the south-east section of the palace, dubbed by Parrot as the “Eastern” and “Workshops and Store-rooms” Quarters, at least twelve chambers sheltered tablets, many of which dated to the days of Yaḫdun-Lim and his (son (?) and) successor Sūmū-Yamam. On the opposite end, tablets were found in four rooms and in one large chamber of the Administrative Quarters. More to the centre of the palace, and around the vast open courtyard no. 106, the largest proportion of tablets were stored. Rooms 115 to the east, 110 to the north-west, and 108 to the west came the closest to being strictly archival in character.

The exceptional weather during much of the 1971 season at Umm Dabaghiyah provided an unusual opportunity to observe the steppe-country of the Jazira in flood conditions. Series of violent though short-lived storms produced an effect on the land surface which was the more pronounced because of the drought, and consequent de-vegetation, of the previous twelve months. The difficulties of travel in the water-logged countryside in the short time available prevented any systematic survey, nor was the party equipped for geomorphological measurements, but even the limited observations possible revealed much of interest, both from the point of view of present-day land-forming processes, and concerning the possible geomorphological history of the area.

Umm Dabaghiyah lies within the dry steppe zone of the Jazira. The Lower Fars series dips gently to the NNW, and the surface levels consist of redeposited silts and clays, more or less cemented by gypsum, lying upon gypsiferous rocks and gravel. The topography is of shallow valleys between low, smooth hills, often with rock outcrops near their summits. The differences in elevation are usually only of some few metres. There appears to be no coherent system of drainage within the area; storm-water drains from the hills in the form of sheetwash to form playa-lakes in the lowest-lying valleys. Only exceptionally are there signs of a permanent system of linear wadis debouching into the valleys or connecting one playa with the next. Presumably this lack of incised drainage results from the small differences in elevation and the consequent low energy potential of the system.

The Iraq Museum has acquired by confiscation a number of Sumerian statues. Their provenance unfortunately is unknown, but their many similarities to statues from the Diyala sites suggest that this new group may have come from the same region. For dating these new pieces, therefore, I have relied on Frankfort's publications of the Diyala group.

The first statue (Plate XXIIa, b) portrays a youth. The head has been broken but repaired. The front of the body is damaged, presumably where it lay face to the ground. The eye inlays are missing and the nose is damaged. The hands are broken, so we cannot see clearly how they were folded. The hair is rendered in horizontal ridges and is divided into two symmetrical halves falling down over the shoulders and chest. The short kilt is in shape a truncated cone, with double overlapping tassels at the bottom. The legs are complete and are supported by a stone slab resting on a “T”-shaped base.

The back of the head is broken on the left-hand side, but the parting of the ridged hair is clear (Plate XXIIb). The tasselled girdle has been carefully worked and carved in relief. There are two statues from the Diyala region similar to this figure. One (no. 33) is poorly made and does not bear close comparison with the new piece; the other (no. 32) is roughly worked and has been damaged by salt, whereas that of Plate XXIIa, b is more carefully carved and can be ranked with the best works of the period.