intense interest, especially in its bearing on the question so much discussed at the present day, of the amount of mixture which the Anglo-Saxon race has undergone with the indigenous races of this island.

Now

But I have said that the discovery is also a puzzle, and the puzzle is this. Near one of these villages lie a number of round barrows, and the excavation of these barrows is related in General Pitt-Rivers' second volume. These barrows were found to belong to the bronze-using, brachycephalic people who by some ethnologists are supposed to be the Kelts, and by others to be of Finnish origin. Whoever they were, it is generally believed that they came later into Britain than the long-headed, stone-using race, overcame them by their superior weapons, and afterwards intermarried with them, thus producing the medium type of skull which is so often found in the round barrows. underneath the site of a house in the Romano-British village which lay near the barrows, was found a grave containing a skeleton of the usual large stature and round skull of the Bronze race, with an unmistakeable Bronze Age food-vessel by his side. It looks as though the people of the village had cleared away his barrow to make a site for their huts, without finding the interment below. This cool indifference to the graves of their supposed conquerors is surprising in itself; but still more puzzling is the question, Where are the conquerors? What has become of them? Has their type become lost in that of these dwindled neoliths, or did they retreat to other parts of Britain before the Roman sword? If ever excavations as careful as those of General Pitt-Rivers are accomplished on as large a scale in other parts of Britain, we shall perhaps get answers to these questions. In the meantime, we shall do well to suspend our desire for the solution of problems until we have a wider range of facts to deal with.

The third volume of General Pitt-Rivers' work contains an account of his excavations in Bokerly Dyke and the Wansdyke, the latter a famous earthwork, as long as the Roman Wall, stretching from Portishead on the Severn to Chisbury Camp in Wiltshire, the latter a shorter length of earthwork, in the main parallel with the Wansdyke, but further to the south. The late Dr. Guest, of whose great learning such scholars as Dr. Freeman and Mr. J. R. Green

always spoke with bated breath, thought he had determined the origin of these earthworks, which he called the "Belgic ditches.' But the spade has little respect for great reputations. It has tossed Dr. Guest's theories into the air, for in the hands of General Pitt-Rivers it has conclusively proved that Bokerly Dyke was not earlier than the reign of Honorius, and Wansdyke, though the terminus a quo of its possible date is not so decisively fixed, is certainly Roman or postRoman. These facts were placed beyond the possibility of a doubt by coins of Honorius, which were found in the rampart of Bokerly Dyke at a depth to which they could not possibly have been carried by drainage from the surface, and by pieces of red Samian pottery which were found on the original surface of the ground underneath the Wansdyke. It will be seen at once that these discoveries are of great importance to early English history. Both the Wansdyke and Bokerly Dyke have their ditches to the north, showing that it was from the north that the enemy was expected. Within the protection of these dykes a large population appears to have been crowded at the period represented by the Romano-British villages already excavated by General Pitt-Rivers. "Within a radius of six or seven miles from Rushmore," he says, "I have counted twelve or thirteen places in which Roman remains have been found, some of them apparently villages of equal size to those which I have excavated." Calling to mind the brief facts recorded by the Anglo-Saxon Chronicle concerning the West-Saxon conquest, which have been elaborated by Mr. Green into a complete history of that conquest, it is plain without laying down all the details as clearly as Mr. Green has related them, that the West-Saxons did not make any advance over the frontiers of what is now Dorsetshire for more than a hundred years after the time of Cerdic, while Gildas relates that they received a severe check at Mons Badonicus, a place which has never been identified, but which Dr. Guest supposed to be Badbury in Dorsetshire. Comparing these statements with the evidence of archæology, and also with some significant names of places to be found in that neighbourhood, General Pitt-Rivers is disposed to believe that the Wansdyke and Bokerly Dyke represent two different lines of boundary raised at different periods by the Romano-Britons of the sixth century against the encroach

ments of the West-Saxons, and behind which they maintained a precarious independence until the conquests of Cenwealh in 658.

General Pitt-Rivers has had a unique experience in the exploration of earthworks, whether boundary lines or forts, and the record of his discoveries is encouraging, for it teaches us that it need no longer be considered hopeless to find any clue to these mysterious traces of human activity. "I have seldom or never failed," he says, "to find something in a rampart capable of throwing light on its construction. "I believe there is not one of the enormous number of camps scattered over the country, the date of which might not be fixed by sections cut through the rampart." Yorkshire abounds so much in earthworks of every kind, that these words have a peculiar interest for Yorkshire readers. Some of these earthworks have already been subjected to scientific exploration, as, for example, the so-called Danes' Dyke, near Flamborough Head, which General Pitt-Rivers has excavated himself, and which he has proved to belong to the early Bronze Age: the same age as the barrows on the York Wolds, explored by Canon Greenwell. I am not aware that the so-called Roman Rig, the dyke which runs along the hills from Sheffield to Mexborough, has ever been made the subject of a scientific excavation. Another type of earthwork is very abundant in Yorkshire: the moated mound with the moated and ramparted platform attached. This form of fortification is supposed by Mr. G. T. Clark to be AngloSaxon, but the evidence is not conclusive; and in the only case I know of where such a camp has been cut into-Cæsar's Camp, near Folkestone, excavated by General Pitt-Rivers some years ago-it proved to be Norman.

In fact, one of the great lessons of General Pitt-Rivers' work is that the spade, and the spade only, can decide the date of an earthwork or a barrow. The spade, moreover, must be an intelligent and instructed spade, or it will only destroy the evidence it is seeking to reveal. An amateur, cutting into a barrow or an earthwork, does not know what the questions are which have to be answered, and so he obliterates the answers to them. Minute observation is one of the most essential conditions of successful exploration.

The spade, unfortunately, is an expensive instrument, and for this reason we may have to wait long for the data on

which a science of archæology must depend. But, in the meantime, if some of our Yorkshire landed proprietors would follow the example of General Pitt-Rivers, and cause the earthworks on their estates to be excavated by competent hands, the cause of knowledge might be greatly advanced. In Germany, such things are deemed worthy of the attention. of Government, and the Pfahlgraben or Limites Romani, built by the Romans as a frontier against the Germans, has just been excavated by a government commission.

Archæological science is still in its infancy in England, and it must be confessed that the infancy has been a prolonged one. How is it that a boy who collects butterflies, or a girl who goes out with her botanical case, is supposed to be pursuing a scientific taste, while a man who cares for such things as earthworks and old stones is still generally regarded with a half-smile as a foolish though harmless trifler? Is not anything which concerns the evolution of man the noblest of all subjects of knowledge? Surely the reason of this indifference is that the realm of archæology has so long been the realm of guess-work, the happy huntingground of those who took little heed to observe, and not much to read, the observations of others, but whose delight it was to sail paper-boats of theory over the ocean of the unknown. The instance above alluded to of Dr. Guest and the Belgic ditches shows that even the rise of the new historic school was not accompanied with the rise of an archæological school founded on exact observation. slowly archæology in England is learning that it must crawl before it can fly. When it does this, it can get facts as solid as those of any other science. A piece of Samian ware, found in a dyke in such a position that it cannot have been carried in by silting, is evidence as positive for the date of that earthwork as an ammonite in a piece of rock. Pottery is well called by General Pitt-Rivers "the human fossil," on account of its special value as evidence of the period when it was buried. There is no point in which the newer school of archæology differs more from the old than in the attention it pays to trifles, or what were formerly thought trifles. The archæologists of thirty or forty years ago dug chiefly to get objects for their collections; the archæologists of to-day (the genuine ones, that is) dig to get knowledge. A shard of pottery, or an iron nail, is as precious to them as a gold

But

torque, because it furnishes irrefragable information as to the date of the monument in which it is found. It is accurate data of this kind, extending over a wide range, which are needed to build up a real science of archæology in England. The labours of Dr. Thurnam, Canon Greenwell, and Professor Rolleston, which have established the existence of two very different races, distributed over the whole of this island before the coming of the Romans, form the most important step which has yet been taken in England in the direction of such a science. But many problems concerning these primitive races yet remain to be answered, and can only be answered by excavations made on the same scale, and with the same care, as those of General Pitt-Rivers.

While strongly asserting that the spade only can certainly decide in individual cases the date of an earthwork, it was inevitable that General Pitt-Rivers should have been led, after the investigation of so many of these remains, to certain general conclusions which are of great value, coming from one who adds special military knowledge to wide experience of ancient fortifications. General Pitt-Rivers regards the circular or irregularly-shaped earthworks which are so common on hill-tops as strongholds, constructed by primitive tribes as places of refuge, to which they fled when some invading host was scouring their country. He dismisses the idea of "chains of fortified posts," which has been such a favourite one with many of our local antiquaries, for with the imperfect military resources of those times, such forts would have been quite inadequate for the defence of a district, as an invading army could have slipped between them. "They imply a low state of civilization, before the inhabitants of any large district had attained to such organization as was necessary for combined defence." He thinks, therefore, that it was only when a more advanced stage of civilization had been attained that long ramparts with ditch and bank, probably headed with a palisade, were built to defend large tracts of territory. It is therefore probable that all the long entrenchments are more recent than the hill-forts. We might therefore expect to find that such a hill-fort, for example, as that of Wincobank, near Sheffield, is the work of an older people than the so-called Roman Rig which runs so near it. Bearing in mind the post-Roman origin of Bokerly Dyke, it seems not impossible that the