of truth, at least as much as those whose prejudices set in the contrary direction. But "if thine eye be single thy whole body shall be full of light," fearlessly pursue truth. It can never be inconsistent with itself, and the parts of it which may for a time seem to conflict, will assuredly ultimately be reconciled. Whether we may live to see this is far less important.

To return. We have in a great part of our county those formations, somewhat indeed different in feature, in which the earliest traces of men are believed to be found. It is the proper function of the local investigator to learn whether researches (for instance) in the gravel and peat of the Kennet may not be as fruitful as in those of the Somme.

1

I entirely agree with Sir J. Lubbock on the claims of pre-historic archæology to the rank of a science. These primæval antiquities too, the transition from geology to human history, are the bond between the two objects of our society. With him, "I care less about the facts than about the method. For an infant science, as for a child, it is of small importance to make rapid strides at first, and I care little how far you accept our facts or adopt our results, if only you are convinced that our method is one which will eventually lead us to sure conclusions," &c.

I am not deeply read in these things. But while many matters in this and in all science must be disputable, some may be taken as certain even by one whose knowledge is as superficial as my own.

1. The fact that investigators and reasoners, some of whom would be under no prejudice from any desire to adjust their facts to our understanding of the Mosaic Record, have greatly confirmed, by failing to find anything conflicting with it, the opinion that man is the final work of creation. The progressive character of creation, a progress from the inorganic to the organic, and a progress not indeed uniform, but very general, from the lower to the higher types of organism, is attested by the first chapter of Genesis, the testimony of the rocks, and (if we accept it) the Darwinian theory. 2. That man was co-existent with some of the large extinct pachyderms is proved. In this I see no difficulty whetever. The 1 1 Archæological Journal, 1866, p. 190.

conditions under which races which have long existed die out are obscure. But the fact is undoubted, and in the case particularly of large animals it may well be rapid. Their numbers in a given area are small. Their breeding places in a temperate and not very mountainous region must be discoverable and accessible. The more man in the hunting stage of society feels himself unequal to cope with the adult animal, the more surely will he, either for food or fear, direct his attacks upon the young.

The method of reasoning from instruments successively of stone, bronze and iron, and from the discoveries in kitchen rubbish heaps in Denmark, and lake villages in Switzerland, is certainly sound. But there is great danger of rash generalization leading to conclusions in which other ingredients than those of time are overlooked. We have distinct historical evidence in the sacred writings and in Homer, belonging to the transition from the bronze to the iron age, in the part of the world most advanced, and, from the Phonician communications extending from Greece to Egypt, likely to be the most advancing. I say distinct historical evidence. For whoever Homer may have been, and whether there be any truth in his narrative or not, no man can doubt that he was a painter of actual contemporary manners, whether more or less idealized; and in the arts of common life certainly an accurate painter. But though the civilized world has long discarded cutting instruments except of iron, we know not how long the earlier instruments may have continued in use among rude tribes, even at no great distance from those more advanced. The Bheels and Goonds (the latter retaining the very peculiar form of cannabilism, described by Herodotus as practised by wood tribes beyond the Indus) exist in our Indian empire. Lake dwellings like those of Switzerland are described by Herodotus as existing at no great distance, whether in actual space or in physical geography, from civilized Greece. The accumulations of deltas, gravel-beds, and the like are most important evidences of date. But here again caution is requisite. A single flood from the bursting of an ice-dam in the mountains not fifty years ago, produced changes near Martigny which might well be supposed to be the work of centuries.

In Sir John Lubbock's very interesting paper,' he tells us that "the antiquities referable to the Palæolithic age are found in gravel or loam, or as it is technically called loess, extending along our valleys, and reaching sometimes to a height of 200 feet above the present water level. These beds were deposited by the existing rivers when they ran in the same directions as at present, and drained the same areas."

Assuming this to be true as to direction and area, and proved by the material of the gravel and loam, yet something more than mere lapse of time, a much greater body of water than these rivers now contain, must have often been required to give the gravel its extent, both vertical and superficial. This last ingredient, however to be accounted for, is too much overlooked by some modern geologists.

Let us now turn to the undoubted works of man so abundant on our downs. They afford a most interesting field for speculation; and the facts are to a great extent known, though not universally nor accurately. It is disputed whether Silbury Hill is not on the line of Roman Road, and therefore posterior to the Roman occupation. It has been examined whether, where the Roman Road coincides with Wansdyke, the excavators of the dyke used the road or the roadmakers used the dyke. A theory that does not rest on an accurate investigation of this fact must be unsatisfactory. A theory which does not allow time for progress from the rude masses of Avebury to the squared and fitted stones of Stonehenge is unsatisfactory. A theory which deals with onr Wiltshire monuments alone without embracing the many smaller kindred works widely dispersed, and the greater kindred works of Carnac, is unsatisfactory. A theory which attributes Stonehenge to Romanized Britons without accounting for the entire absence of moulding, so near the finished Roman work at Bath, is unsatisfactory. I would not deprive the local observer either of the mental pleasure or of the aid to memory derivable from stringing his facts on a speculation. His guesses also may be of farther value. But he must

1

Archæological Journal, 1866, p. 190.

2 This has been disproved by excavations made since the meeting.

remember, first to record his facts with judicial fidelity and impartiality; and secondly to hold his speculations as merely tentative and provisional, subject to be displaced by a wider induction.

We proceed now to something nearer to what we usually understand by history. Our downs have been the battle-fields and our forests the fastnesses of various races, Dr. Guest gave, when the Archæological Institute met at Bath, a most interesting account of the acquisition of the Cotswold District by the West Saxons, and its subsequent loss by their own dissensions. This explains how that part of Gloucestershire, though comprised in Mercia, speaks our dialect. He told us that the intermediate vale country, from the Thames to Trowbridge, continued Welsh. I do not know that Welsh names, except that of the river Avon, which is common to many others, survive there. Lydiard has rather a Celtic sound, but I am not etymologist enough to know anything more about it. But the Roman Road across this district from the downs westward ceases to be a highway, just where, if it were to be traversed within the limits of the same society, an artificial road would be most valuable. Much of the district also is, or was, for much has been disafforested, a string of forests-Bradon, Pewsham, Woolmore. Now forest has nothing to do with wood, except that woodlands will usually be the last to be settled. But it means Out Land. To compare small things to great, nearly what the Americans would call a territory. Land not yet, at the time when the law assumed its consistency, absorbed into the social system, and therefore, when it came to be settled, governed by perogative.

Much of it was early settled

In architecture our county is rich. and we have good building stone. The result has been that we have a multitude of small churches in which parts of very pure and graceful early structure still remain. Perhaps the oldest and most curious edifice is a small building at Bradford on Avon, which it seems impossible to assign to any period subsequent to the Conquest, though with all its rudeness there is a feeling in it almost classical, which it is difficult to assign to the Anglo-Saxons, and yet we can hardly think of a still higher antiquity.

The earthworks of Old Sarum and Ludgershall, though by their

Parliamentary representation down to our own day, the tradition of their being the habitations of men has been kept up, can hardly be called architectural. Yet the regular frame of Old Sarum has much interest as an early fortress adapted by art, on a site suggested by nature.

Later we have two instances of the foundation of medieval towns, both interesting, and affording examples of contrary currents of progress-the fortified town of Devizes, and the open town of Salisbury.

At Devizes the escarpment of the green-sand is very steep and deeply cut into, by ravines, two of which so nearly meet at their heads, as to leave a peninsular eminence, with steep sides, and only a very narrow attachment to the high ground behind. This was an admirable site for a Castle: and on that high ground, grew up a town, in the form of a semi-circle, the diameter of which abutted on the steep descent on each side of the approach to the castle, and the curve was, and is clearly marked by the line of New Park Street, and Bridewell Street. The castle was held in the reign of Stephen, by Bishop Roger, and the town, both from the regularity of its form, and from the Norman Architecture of St. John's Church, would seem to be of the same date, or nearly so, and a part of the same plan. St. John's Church, of which the chancel, transept, and intersection with its tower, remain entire, with unimportant additions and insertions, is a very characteristic specimen of somewhat advanced Norman architecture. The nave, to meet increased population, has been re-built with aisles added in the 15th century. Outside the walls is another church, St. Mary's, which for the most part, dates from that age; but shows some Norman work, though later than St. John's.

We have no very great Norman work in this county except the remains of the Abbey at Malmsbury. But several churches show Norman proportions though the existing structures are in various degrees modernized. Of these Westbury is perhaps the most remarkable.

Of the next style we have much, There are many fragmentary parts of Early English work, particularly in the chancels of small