In the second article of this series, brief mention was made of a cave at Kastritsa, near Lake Ioannina, whose talus had yielded an industry reminiscent of the last phase at the shelter of Asprochaliko. The evidence for an increase in the size of Lake Ioannina during the Last Glaciation was also outlined. The purpose of this paper is to consider the concomitant environmental changes in relation to the evidence for human occupation.

Lake Ioannina lies at 469 metres above sea-level. To the north-east stands the Mitsikeli ridge, with a maximum elevation of 1,810 metres and separated from the Pindus range by the gorge of the upper Arakhthos (fig. 1). To the south-west lies the Tomarokhoria plateau, which in places rises to over 1,900 metres. The basin, a typical polje, is cradled by limestones which range in age from the Upper Triassic to the Upper Eocene. The Flysch of the Pindus borders it on the south-east. The lake lies along the axis of the Perama syncline, which runs north-west-south-east parallel to the Mitsikeli anticline and the Stavraki anticline; faulting is common. The basin would seem to owe its existence primarily to structural factors, although solution doubtless contributed to its present form.

When a team of the Archaeological and Anthropological Department of the Cambridge University, working under Mr. Eric Higgs, recently discovered a Mycenaean dagger at the foot of the Kastritsa Mountain (altitude 757 metres), a few metres west of the mouth of the paleolithic cave, it came as no great surprise: for, in the last 15 years, about 10 bronze swords and daggers of Mycenaean provenance have been found in Epirus.

On the other hand, the discovery of the paleolithic cave itself at the western foot of the Kastritsa Mountain, and the finds yielded by its excavation, were as unexpected as they were important. It is hoped that continued excavations of the cave with its deep deposits and clear stratification will throw light on some of the problems connected with Homo Sapiens: his origins, the time of his appearance in Europe, his relation to other groups in Epirus and outside it, the civilization he evolved, and the ecological conditions in which he lived in the last phase of the last Ice Age (Würm III).

A fresh and more extensive excavation of the prehistoric settlement north of the village of Kastritsa, at the eastern foot of the mountain, would not be devoid of interest either, for it would complete the cycle of prehistoric research and probably throw light on the mesolithic occupation of which little is yet known in Greece.

This article presents the evidence for the distribution of Late Minoan stone vases outside Crete. This class of evidence has not been discussed previously and as the amount of material involved is substantial, over a hundred vases, it is here set out as a contribution to the discussion of Minoan foreign connexions in the earlier part of the Late Bronze Age in the Aegean.

The vases were made in Crete in Middle Minoan III/Late Minoan I—Late Minoan III A 1, c. 1600–1400 B.C., as is shown by the contexts of parallel vases from sites in the island (see the Catalogue). Most of the exports occur in Aegean contexts of this period; for those from contexts later than c. 1400 B.C. it is most reasonable to believe that they too were exported from Crete at the time of their floruit, containing in use for a time afterwards. In any case most of them are fragments, which indicates that their time of arrival and period of use were earlier than the late date of their contexts. Moreover after c. 1400 B.C. Cretan influence abroad ceased (Desborough 1964, 166–7; cf. Furumark 1950, 263–4 for the end of Minoan contacts with Trianda and Phylakopi), nor is there evidence that stone vases were then made in the island.

For many years archaeologists interested in the study of the Palaeolithic in North Africa and Eurasia have been using cumulative percentage frequency graphs for the comparison of prehistoric artefact assemblages. For examples we refer to the references at the foot of this text. However, it is perhaps time to carefully review this technique and its future utility.

Statistical techniques and mathematical models are slowly infiltrating and reshaping the discipline of archaeology—increasing the power and depth of both analysis and synthesis. It is perhaps already possible to distinguish the cumulative advance of these techniques in archaeology from the initial role of demonstrative aids and methods of data display towards an increasingly powerful analytical role with a capacity for predictive inference. In the ranks of the first generation we have the early use of graphs, frequency polygons and histograms for mapping severely limited numbers of percentages or attribute ratios. In the second generation of statistical techniques the impact of the computer is felt for the first time and archaeology is developing an array of exploratory attempts to integrate the probing capacity of such methods as factor analysis, matrix analysis and principal component analysis.

Cremated remains, usually interred in urns, most commonly occur in cemeteries dating from the latter part of the Bronze Age, the closing phase of the Pre-Roman Iron Age and the Pagan Saxon period. Although such remains are fragmentary, they are now recognized as providing a basis, after appropriate treatment and anatomical study, for assessing the numbers of individuals cremated, together with their age and sex (Gejvall, 1947). Sometimes, additional information can be gleaned about endemic pathological conditions (Lisowski, 1955).

The physical characteristics of the cremated remains vary from site to site, depending, as discussed by Gejvall (1963), on such factors as the method used in cremation, the firing temperature, and the extent of any subsequent pulverisation—natural or artificial. But despite the fact that there is always a large proportion of unidentifiable fragments, there are invariably certain parts, often of appreciable size, which can be recognized. It is from these fragments, when separated, that information about the number of individuals present, together with their sex and age can be obtained.

Agriculture and stock rearing moulded the life of the English peasant during the Middle Ages. Together, they absorbed the greater part of his energy, time and skill. In good seasons there was plenty, and in times of dearth, the threat of starvation. Tracts were written on the most effective means of stock management and soil utilization, and farming methods varied with regional changes in topography and environment.

There is no reason to suppose that the same does not hold for the prehistoric peasant: yet the economic prehistory of England, indeed of Europe as a whole, remains unwritten. This paper will consider the possibility of discovering the basic economic framework of a number of prehistoric societies through the examination of both faunal and floral remains. The sites in question are restricted to the Alpine Foreland, due to the excellent local conditions for the survival of floral, faunal and artefactual material.

The importance of attempting to reconstruct diet from a study of biological data from archaeological sites has been stressed by earlier workers, both in Britain and the United States. The present study is based on materials recovered from the excavation of a midden deposit at Galatea Bay, on an offshore island of the North Island of New Zealand, by Mr J. E. Terrell, whose complete report (No. 1) is to appear in the Transactions of the Royal Society of New Zealand. The site overlooks a sheltered area of sea from the mouth of a small and now dry valley (see pl. IX) and the excavation revealed an initial, specialized cooking area which was replaced by dumps of food waste and then abandoned. The most characteristic elements of the midden were marine shellfish valves and a much smaller quantity of vertebrate bones. However, the apparent frequency of the shellfish could not be taken as a sure guide to their importance over the vertebrate animals because the nature of the animals and their surviving evidence makes it impossible to directly compare them. Shellfish, though individually small, are found in large numbers and are thus best studied by means of estimating populations derived from samples. Vertebrate populations on the other hand can be measured by more direct means, though it must be recognized, as Mr Terrell points out, that such an excavation is itself only a sample. In his report Mr Terrell examined the effects of dehydration, size and sifting upon the shell samples and then compared the archaeological population with those of living shellfish, so enabling him to demonstrate the relative influence of natural and cultural causes on the archaeological population. These findings provide basic evidence which will be referred to during the course of the present paper.

Over a century ago Theobald (1860) described the Pleistocene deposits of the middle reaches of the Narmada river and noted the presence of Pleistocene fauna and man-made tools of stone. Eighty years later the Yale-Cambridge expedition in 1935 carried out the first systematic work under the leadership of De Terra and Paterson (1939) whose major concern was with deposits of the glacial and peri-glacial zones. But their brief survey of the Pleistocene deposits of central and southern India was very valuable, even though their extension of glacial chronology to the sub-tropical zone was mainly based on lithic-tool typology. Since 1947 considerable prehistoric research has been carried out in Peninsular India but without much success in establishing Pleistocene chronology on firm grounds. During the last decade a number of observations and researches have been carried out by several workers in the Narmada valley (Joshi, 1958, Subbarao, 1958, Malik, 1959a, Sen, 1961, Khatri, 1961, 1962, and Sankalia, 1963).

The present project carried out jointly by the authors was primarily concerned, not with collecting stone tools, but with understanding the stratigraphy and chronology of Peninsular India.

Current research on the prehistory of the Near East is proceeding at a pace unmatched since the ‘golden era of archaeology’ immediately preceding World War II. Iran, because of its great archeological potential and its stable political situation, has become a prime target for new investigations. Expeditions from the United States, Great Britain, Denmark, France, and Japan have all made substantial contributions to knowledge of Iranian prehistory in the last five years.

One recurrent problem in the new research, however, is a lack of coordination and communication between expeditions, which means that many opportunities for useful cooperation are lost. For example, in 1961 we surveyed in some 15 valleys in western Iran, plotting the distribution of pre-Uruk settlements. Unbeknown to us, at least one American and one Danish survey team crossed our path, duplicated many of our survey runs, and excavated one of the sites found on our project. We later learned that we, in turn, had duplicated several survey runs made previously by a British team, who were interested mainly in later sites but would have been happy to share their data on sites in our range of interest. Had all of us known what our colleagues were doing, we could have saved ourselves many hours of duplication.

This situation is worsened by the fact that many surveys and test-excavations remain unpublished, and diagnostic artifacts remain undescribed. With regard to pottery sequences, the urgency of the problem has recently been stressed by one of our colleagues (Young 1966: footnote 28).

Rainsborough is 1 mile South of Charlton village, in the parish of Newbottle, S. Northants, 20 miles North of Oxford, (SP 526348). The camp lies at c. 480′ OD, and the area enclosed is c. 6·25 acres.

It lies on the edge of a plateau: the approaches to it are flat on the north-east, east, south and south-west, but a gentle slope to the north, north-west and west gives it a wide view across the Cherwell valley, towards Madmarston and Tadmarton (see map, fig. 1 and also pl. XXV). The defences are bivallate: the inner bank stands to 10 feet above the interior, and there is a drop of about 15 feet from the crest into the inner ditch; the second bank is very much lowered by ploughing, but still reaches a height of about 4 feet on the south side, where a hedge line has protected it; the outer ditch is nowhere visible on the surface, except on the west, when it carries a higher growth of weeds. The defences are covered with turf: the inner bank has also trees, bushes and the stumps of large beeches felled c. 1950. The bank is riddled with tree roots, and the sandy character of the core has attracted rabbits: recent attempts to dig and smoke out the warrens have slightly damaged the profile of the bank. A small dry stone wall is visible part way up the outer slope of the inner bank in several places.

This study of iron currency bars began as a note, prepared at the request of the late Professor Sir Ian Richmond, for inclusion in his publication of the British Museum excavations at Hod Hill. It was intended to accompany a similar account of the Celtic coins found at Hod Hill, for which he had previously asked. Hod Hill is the only site where coins and bars have been found in comparable numbers. However, as one thing led to another, the original note grew out of scale for a contribution to an excavation report; hence this separate paper. A summary will be included in Mr J. W. Brailsford's volume on Hod Hill.

The term ‘iron currency bar’ has long been in use to describe certain elongated iron objects found in late iron age sites in Britain. The earliest records are of a hoard found at Meon Hill, Warwickshire, in 1824. Bars became well known after the discovery of an important hoard at Salmonsbury, Gloucestershire, in 1860. The first serious study was made by Charles Roach Smith, when in 1864 he published an earlier hoard of bars from Hod Hill. The classical account of them was given by Reginald Smith in 1905 and developed in a number of subsequent articles, which have been the basis of nearly all that has been written since. It is the theories of Reginald Smith which require reconsideration today.

Between 1,100 and 1,500 iron bars relevant to this study have been found at more than 20 sites in Britain. Whether they are all iron currency bars depends on the definition of currency bar adopted. The first task was to trace as many bars as possible; I have found examples in more than 30 collections and there are, no doubt, more which I have failed to find. What has come to light, however, summarized in the Appendix, has been enough to show that the surviving material needs to be analyzed on fresh lines.

The fine round barrow described below was the second of two excavated on behalf of the Ministry of Public Building and Works during the summer of 1961. The site, which is on Upper Chalk with Flint, is the eastern outlier of a linear barrow cemetery on Earl's Farm Down, and lies at 400 feet on the western slope of Beacon Hill (fig. 1) south of the main Amesbury-Andover road (O.S. National Grid: SU/184419, Sheet 167).

For over 40 years South Africa and for over 25 years East Africa have been yielding fossilized remains of creatures identified on their bony structure as lowly members of the Hominidae or human family. From five sites in the Republic of South Africa (Taung, Sterkfontein, Kromdraai, Makapansgat and Swartkrans) and from three sites in the Republic of Tanzania (Garusi, Olduvai and Peninj) have emerged a considerable number of hominid fossils of the Lower and Middle Pleistocene. Most of them have been classified as members of an extinct hominid genus called by R. A. Dart, the discoverer of the first specimen, Australopithecus. Although variable among themselves, the structural features which aligu these fossils with the Hominidae, rather than with the family of the apes or Pongidae, are as follows:—

(1) Australopithecus showed modifications of his skeleton—especially the pelvis, femur, ankle and foot—which permit us to infer that he walked upright, even though such modifications had not progressed as far as in hominids of the genus Homo.

(2) The calvaria or brain-case of Australopithecus was characterized by a number of hominid features, such as marked flexion of the axis of the cranial base; forward displacement of the occipital condyles by which the skull articulates with the vertebral column; a small, low nuchal area at the back of the base of the cranium, for the attachment of those muscles which tether the back of the cranium to the trunk—the last two features suggesting a different, more man-like poise of the cranium on the spine; the consistent development, early in life, of a pyramidal mastoid process as in man, and unlike the apes in which this process develops inconsistently and then only later in life (Schultz 1950).

(3) The canine teeth of Australopithecus, like those of Homo, lacked the enlargement and marked interlocking which characterizes the upper and lower teeth of the apes—correspondingly, Australopithecus lacked the diastemata or gaps which, in apes, lodge the projecting tips of the enlarged canines.

(4) Another dental characteristic shared by Australopithecus with Homo was the bicuspid structure of the first lower premolar tooth: in apes, this tooth does not possess two sub-equal cusps but is a cutting or sectorial tooth with one predominating cusp, like a canine tooth.

(5) Although the brain-size of Australopithecus (as inferred from the volumetric capacity of the brain-case) was no bigger than that of some largerbrained apes, the shape or morphology of the brain, as preserved in endocranial casts, approached more closely to that of early Homo than to that of the apes. We have, however, no direct evidence whether these external observable differences between the brains of Australopithecus and of apes were paralleled by internal, microscopic, structural differences such as those which have been demonstrated between modern man and modern apes.

(6) Many other detailed morphological features of the skull, the teeth and other bones of Australopithecus showed human resemblances.

The primary object of the present article is to offer a review of the current state of knowledge of those flange-hilted swords of bronze which characterize the earliest phase of the Hallstatt period, and of some derivative varieties; and to support it by as complete an inventory of the available material as lies within the writer's power. It aspires, accordingly, first and foremost to be a factual survey of a clearly defined group of weapons, which over a significant portion of the area treated were the last of their kind to be made in bronze.

More specifically, it is hoped that the topographical, typological, and chronological aspects of this essay may be found acceptable, in and for themselves, and as a basis for further study. But the interpretation of the character of the Hallstatt sword, and the socio-historical situation presented as a background to that, inevitably lie in a more speculative field, and are offered solely as a possible explanation of the genesis of the type. This explanation, if correct, would indeed make good sense of a number of otherwise disconnected factors. It remains, nevertheless, a hypothesis propounded to account for the origins of the material under review; and can be jettisoned, in whole or in part, without at all impairing the validity of the treatment of that material, once it had come into being.

Though the contemporary swords of iron are certainly overdue for comprehensive and up-to-date treatment, they fall outside the scope of this study.