REPORT ON THE ANIMAL REMAINS FOUND AT THE VILLAGE SITE AT FIFIELD BAVANT.

By J. WILFRED JACKSON, M.Sc., F.G.S.

The various animal remains from the above site sent to me for examination by Dr. R. C. C. Clay consist of the bones and fragmentary skulls of several domestic species and of one or two wild forms. They are of interest in connection with the remains obtained in recent years at All Cannings Cross, Wilts, by Mrs. Cunnington. At this latter locality the age of the remains was proved by the presence of Halstatt pottery, and it is interesting to find that the present collection contains exactly the same forms of domestic animals as All Cannings Cross.

The animals represented are:-Celtic Pony, Celtic Ox, Sheep, Pig, and Dog. There is also one odd bone which may be referable to Goat. The wild animals consist of small forms only, viz., Polecat, Weasel, and Water Vole. The Fox may be represented also by two bones. A few bird-bones are included in the collection. There is an entire absence of the remains of Deer, but this animal was only sparingly present at All Cannings Cross.

HORSE (Celtic Pony). The remains of this animal consist of a few cannon-bones and three fragments of lower jaws. The cannon-bones indicate small-sized animals of the Exmoor Pony type, of 11 to 12 hands in height. One fragment of lower jaw is that of a male animal; another is that of a female; the third is too imperfect to be sure of the sex. These agree closely with the All Cannings Cross remains and with the Exmoor Pony.

Celtic Ox. Two imperfect skulls, three horn-cores, a lower jaw and a few limb-bones are referable to Oxen. One of the skulls consists of the frontals with short, somewhat flattened, horn-cores, and shows a notched occiput. It agrees exactly with two of the All Cannings Cross skulls. The other skull consists of the frontals from which the horn-cores have been broken. It possesses a notched occiput and is slightly larger than the firstmentioned. The three horn-cores are quite typical of Bos longifrons, and the lower jaw agrees with others referred to this animal from All Cannings

Cross, Glastonbury, etc. It possesses the normal six cheek-teeth. One of the limb bones (a metacarpal) is smaller than any from All Cannings Cross, but the other bones (metatarsals) agree closely.

There are several

SHEEP. The bones of this animal are not numerous. very young bones in the collection, as well as a number belonging to adult animals. They all agree very closely with similar remains from All Cannings Cross. Four young horn-cores are present, but these are too immature for diagnosis.

PIG. This animal is represented by two imperfect lower jaws with teeth, also by one canine and one incisor. They belong to the All Cannings Cross form-the "Torfschwein" of the Swiss Lake Dwellings.

DOG. A few limb-bones and a skull and lower jaws complete are referable to this animal. The skull is of the same general type as the example found at All Cannings Cross, but is slightly narrower across the head. The limbbones do not call for special mention.

GOAT? A single metacarpal bone may belong to this animal. It appears to be too robust for the sheep of this period.

POLECAT. A perfect skull and lower jaw, quite typical of the species, is present in the collection.

WEASEL. This animal is represented by a perfect skull.

WATER VOLE. This is represented by a skull, two left rami of the lower jaw, and a few limb-bones. Similar remains were found at All Cannings Cross and at Glastonbury.

Fox? Two limb-bones are referred to this animal on account of their small size.

BIRD-BONES.-A few bird-bones are present in the collection, some of which may belong to the Raven.

REPORT ON THE CEREALS. By Professor R. H. BIFFEN.

The material examined consisted of about 7 litres of grain washed out from the mass in which it was originally found. The grain, though completely carbonised, was in a good condition of preservation. It consisted almost entirely of wheat and barley in roughly equal proportions.

The wheat was mainly in the form of separate grains, but a search through the whole mass resulted in finding some seventy spikelets in a more or less whole condition (Plate XV., fig. 2). These spikelets were invariably twograined. The enveloping chaff had broken away to a great extent, but it persisted as a rule at the base of the spikelets. So resistant was it that a number of the paired bases of the empty glumes were found mixed up with the loose grain (Plate XVI., fig. 1). The bases of the glumes were rounded and practically keelless. No perfect apices of the glumes have yet been found. The grain was characterised by the flatness of the grooved surface, by a somewhat pointed base, and great variation in size (Plate XV., fig. 1). At first this variation in size was thought to indicate the presence of more than one kind of wheat, but further examination made it clear that the small grains were the produce of small spikelets from the tops of the ears. Two spikelets only showed any remains of the rachis-one a perfect basal spindle, the other part of one.

All of these characteristics point unmistakeably to the wheat belonging to the group Triticum dicoccum or Emmer.

The grains of barley (Plate XVII., fig. 1) were well preserved and in the majority of them the interesting paleæ still remained. They formed a fine well-grown sample, which, judging from the wrinkling of the grain coats, had ripened perfectly. The variation in the size of the grain indicated that the type was a six-rowed one, and final proof of this was given by the finding of several fragments of the ears still carrying the characteristic groups of three grains (Plate XVII., fig. 1). Fragments of the ear-stalk confirmed the view arrived at from the shape of the grain that the ears were of the dense type. By estimating the proportion of large, straight grain, to smaller, slightly-twisted grain it was found that the barley consisted of a six-row form only. It belonged to the group Hordeum hexastichum.

Mixed with the wheat and barley were a small number of oat grains preserved so extraordinarily well that some of them retained their stronglydeveloped dorsal awns (Plate XVIII., fig. 2). The outer paleæ, too, were frequently perfect, or if partially destroyed the nerves persisted in a fringe round the caryopsis. The thinner inner paleæ had, however, disappeared as a general rule. The grains were small, apparently two to each spikelet, and so tightly gripped by the paleæ that no naked caryopses could be found in the whole sample. The size and shape of the caryopsis made it clear that these were cultivated and not wild oats, the shape indicating that they belonged to the loose-panicled group, Avena sativa.

The bulk of grain was fairly free from weed seeds. So far the only examples met with have been seeds of two species of Bromus, namely B. sterilis and B. mollis (Plate XVIII., fig. 3). The investing paleæ of these had disappeared completely, leaving only the naked caryopsis. The spine-like shape of the former and the flattened barge-like shape of the latter made the identifications certain.

I don't know of any description of Roman or pre-Roman oats. But I have specimens from the Lake Dwelling sites at Meare, near Glastonbury. These have lost their paleæ almost entirely and consist of little more than naked grains. Your specimens are amazingly good. But why the chaff has persisted so whilst that of the wheat has disappeared to such an extent is puzzling. The grain has shrunk a good deal in carbonizing. But I think it would be safe to say that, as far as size goes, the wheat and barley were originally about the same as their modern counterparts. The oats were smaller.

REPORT ON A SKULL FROM THE ANCIENT VILLAGE SITE NEAR FIFIELD BAVANT, WILTS, OF LA TENE I. DATE.1

By SIR ARTHUR KEITH, M.D., F.R.C.S., L.L.D., Conservator of the Museum, Royal College of Surgeons, England.

There are no skulls of prehistoric England which anthropologists wish to

'The five drawings of the skull on Plates XXIV. and XXV. are all reduced tolinear from accurate full-sized drawings prepared under Sir A. Keith's supervision.