stations occupied by the latter, we can identify Danum as Doncaster, and Lavatræ as Bowes. Veterum, or Veneris, is certainly a corruption of Verteræ (Brough under Stainmore), and Braboniacum is probably Kirby Thore. These last three stand in this order on the list, and occur in the same order on the great road from York to Carlisle, a coincidence which suggests that this list also is arranged per lineam. If so, we must look for the unidentified sites in regular order.

The list begins with Præsidium, occupied by a troop of Dalmatian cavalry. We have no clue to its position, but a likely site is Brough, on the north shore of the Humber. A Roman road led from York to this place, which is opposite the termination of the road running north from Lindum; and it would be strange if, while guarding the Saxon shore, the Romans had no garrison on the Humber, which in later times was so often the landing-place of invaders.

This brings us by a natural sequence to Danum (Doncaster), occupied by the Equites Crispiani, whose duty probably was to watch the neighbouring fens and marshes. We know from the exploits of Hereward the Wake what shelter such country could afford to fugitives and outlaws. Next, if we follow up the Don for about fourteen miles, we come to Templeborough. This may be Morbium, where the Equites Catafractarii were stationed. Originally Catafractariï were men in complete heavy armour, and horsemen so burdened might seem out of place in the Sheffield district; but probably by the end of the third century they were catafractarii only in name. We learn from Vegetius' that in the time of Gratian (A.D. 367-383), even the infantry of the legions asked to be relieved of the cataphracta (body.armour), and then of the shield; and probably a corps of auxiliary and possibly irregular cavalry had unburdened itself at a much earlier date.

Next on the list is Arbeia, a name with a curiously eastern sound, garrisoned by a numerus of Barcarii Tigrisienses. About twenty miles north-west from Templeborough, in the neighbourhood of Huddersfield, we

[blocks in formation]

find two forts—at Slack and at Almondbury. The former is usually identified with the Cambodunum of the Second Iter, and Almondbury may be Arbeia. About twenty miles further north we find an undoubted Roman site at Ilkley, which may accordingly be Dictis, the station of the Numerus Nerviorum Dictensium. A further distance of twenty-eight miles in much the same direction brings us to Wensleydale, where we find the remains of a Roman fort at Bainbridge near Askrigg This may be Concangium, where a Numerus of Vigiles was posted. Fifteen miles further, and we come to Bowes, which from the Itinerary we know to be Lavatræ, and Lavatræ, with a Numerus of Exploratores, stands next on the Votitia list. So we proceed over Stainmore to Verteræ (Brough), where a Numerus of Directores was stationed; and then to Braboniacum (Kirby Thore), where the Notitia places a Numerus of Defensores.

We have now four stations left-Maglova, held by the Numerus Solensium (possibly Cypriotes or Cilicians), Magæ, held by the Numerus Pacensium (these may have come from Pax Julia in Lusitania, from Forum Julii, also called Pacensis Colonia, or from a place of the same name in Thrace), Longovicum, held by the Numerus Longovicariorum, and Derventio, held by the Numerus Derventionenses. Now, Longovicum is probably Lanchester, in Durham. Some years ago an altar was found there, with an inscription mentioning a vexillatio Sueborum Lon. Gordiana. Lon. seems to stand for Longovicaria ; and though the expansion is not free from uncertainty, if the list is arranged systematically, Longovicum must be somewhere in this neighbourhood.

We must now try to find sites for Maglova and Magæ somewhere between Kirby Thore and Lanchester. If we follow the Maiden Way from Kirby Thore over the hills to the South Tyne valley, we reach the Roman station of Whitley Castle, which may be Maglova. From this place an ancient road led over the hills into Allendale, and seems to have continued in an easterly direction till it joined the Dere Street (the road of the First Iter) at the bridge end opposite Corstopitum. Near this road, at a place called Old Town, in Allendale, the traces of a Roman fort have been observed, and this may be Magæ. The country here is very wild and rugged, and possibly there was a mining industry to be protected. From Old Town, Lanchester is twenty-five miles distant across the hills, and occupies a site well suited for a corps controlling the wild country of West Durham. Only Derventio remains, and one is strongly tempted to fix it at Ebchester, on the Derwent, a few miles north of Lanchester; for this linea would thus terminate at a point only fifteen miles from Wallsend, where the Item per lineam valli section begins. However, in that case we are forced to imagine a change of name, since the Itinerary shows Ebchester to have been Vindomora. It is safer to suppose that Derventio was on the Yorkshire Derwent. It may be the same as the Derventio of the First Iter, which seems to have been in the neighbourhood of Stamford Bridge. In this way we round off the circle : we began at York, and at York we finish.

[ocr errors]



(Read May 16th, 1906.) HERE is one class of antiquities in England

of a philological nature, which is almost unique in Europe : I mean, the relics of an ancient language. Nowhere, except in England, I believe, can we even fix any death-place of a language. One of

the reasons for this is that languages die so hard : the only European language, besides the Cornish, that has died out in modern times is the Prussian, and I question if we can fix the time or place of its expiring

As to the dying languages of Europe, none expired in the nineteenth century, and I rather question if any will die in the twentieth : unless, at least, the strong feeling of nationality which now prevails in the smaller nations of Europe, and which seems to be growing rather than declining, should suddenly yield and die out. Most of these lesser languages are more vigorous now than in the middle of the nineteenth century. A hundred years ago, one might have prophesied that they would expire before the twentieth century, but now they are very vigorous. Political reasons and the spirit of nationality have much to do with this.

1. The Lithuanian.—I lately had in my house a ballticket in Lithuanian, used in London. Even in this metropolis, Lithuanian is used by some hundreds of persons. In its own country it is used by tens of thousands, and a literature is rising in it.

2. As for Wendish, it has its literature, magazines, and




newspaper. National spirit supports this Slavonic language in the heart of Germany.

3. As for Manx, only lately there was a petition to the Education Department for grants for teaching Manx in the Isle of Man schools. The title of the laws decreed by the House of Keys are read in Manx at the Tynwald.

4. As for Irish, it is far more vigorous than some years ago. In London, I see Irish bills in the old Erse alphabet.

5. Mordvinian is lively still, although the Russian Government seeks to stamp it out.

6. Servian and Bulgarian are safer than ever, as being now the languages of established and independent nations.

We had a lively reminder of the vigorous life of the Celtic languages of Western Europe at the Pan-Celtic Congress of 1904. There were gathered in the charming old Welsh town of Carnarvon / the chief eagle of the eagles of Snowdonia”) the representatives of the six Celtic nationalities of Western Europe

1. The Welsh ; 2. The Breton ; 3. The Cornish—of the Brythonic group.

4. The Irish ; 5. The Scottish Highlanders; 6. The Manx-of the Goidelic

group The Cornish had not been recognised at the Dublin Congress, but was accepted at Carnarvon, and will be noted among the six races of the Celts.

I may say that, in my opinion, this is not an exhaustive list of the Čeltic races of Europe.

1. The old Gauls of France were true Goidels, although now Latinized and mixed with other races. The greatest Celtic nation of Europe is really the French. In Cornwall one is often reminded of this : the country-folk, especially the agriculturalists, are in many ways and customs, and in physical aspect, like the French peasants, i.e., those of true Gallic descent.

2. The Walloons of Belgium are Celts, but their nationality is obscured. The Walloon tongue is like French, and is being superseded by it.

3. The Cumbrians are a vestige of the Strathclyde Celts, but now nearly absorbed in the English.

All this shows the interest we have in the last relics of

« ForrigeFortsett »