On the subject of the boundaries of Europe there is still divergence of opinion. While some authorities take the line of the Caucasus as the boundary in the south-east, others take the line of the Manych depression, between Boundaries. the upper end of the Sea of Azov and the Caspian Sea, nearly parallel to the Caucasus. Various limits are assigned to the continent on the east. Officially the crest of the Caucasus and that of the Urals are regarded in Russia as the boundaries between Europe and Asia on the south-east and east respectively,[2] although in neither case does the boundary correspond with the great administrative divisions, and in the Urals it is impossible to mark out any continuous crest. Reclus, without attempting to assign any precise position to the boundary line between the two continents, makes it run through the relatively low and partly depressed area north of the Caucasus and east of the Urals. The Manych depression, marking the lowest line of this area to the north of the Caucasus, has been taken as the boundary of Europe on the south-east by Wagner in his edition of Guthe’s Lehrbuch der Geographie,[3] and the same limit is adopted in Kirchhoff’s Länderkunde des Erdteils Europa[4] and Stanford’s Compendium of Geography and Travel. In favour of this limit it appears that much weight ought to be given to the consideration put forward by Wagner, that from time immemorial the valleys on both sides of the Caucasus have formed a refuge for Asiatic peoples, especially when it is borne in mind that this contention is reinforced by the circumstance that the steppes to the north of the Caucasus must interpose a belt of almost unpeopled territory between the more condensed populations belonging undoubtedly to Asia and Europe respectively. Continuity of population would be an argument in favour of assigning the whole of the Urals to Europe, but here the absence of any break in such continuity on the east side makes it more difficult to fix any boundary line outside of that system. Hence on this side it is perhaps reasonable to attach greater importance to the fact that the Urals form a boundary not only orographically, but to some extent also in respect of climate and vegetation,[5] and on that account to take a line following the crest of the different sections of that system as the eastern limit between the two continents.[6] Obviously, however, any eventual agreement among geographers on this head must be more or less arbitrary and conventional. In any case it must be borne in mind that, whatever conventional boundary be adopted, the use of the name Europe as so limited must be confined to statements of extent or implying extent. The facts as to climate, fauna and flora have no relation to any such arbitrary boundary, and all statistical statements referring to the countries of Europe must include the part of Russia beyond the Urals up to the frontier of Siberia. In such statements, however, in the present article the whole of the lieutenancy of the Caucasus will be left out of account. As to extent it is provisionally advisable to give the area of the continent within different limits.

The following calculations in English square miles (round numbers) of the area of Europe, within different limits, are given in Behm and Wagner’s Bevölkerung der Erde, No. viii. (Gotha, Justus Perthes, 1891), p. 53:—Europe, within Extent. the narrowest physical limits (to the crest of the Urals and the Manych depression, and including the Sea of Azov, but excluding the Caspian Steppe, Iceland, Novaya Zemlya, Spitsbergen and Bear Island) 3,570,000 sq. m. The same, with the addition of the Caspian Steppe up to the Ural river and the Caspian Sea, 3,687,750 sq. m. The same, with the addition of the area between the Manych depression and the Caucasus, 3,790,500 sq. m. The same, with the addition of territories east of the Ural Mountains, the portion of the Caspian Steppe east of the Ural river as far as the Emba, and the southern slopes of the Caucasus, 3,988,500 sq. m. The same, with Iceland, Novaya Zemlya, Spitsbergen and Bear Island, 4,093,000 sq. m. In all these calculations the islands in the Sea of Marmora, the Canary Islands, Madeira, and even the Azores, are excluded, but all the Greek islands of the Aegean Sea and the Turkish islands of Thasos, Lemnos, Samothrace, Imbros, Hagiostrati or Bozbaba, and even Tenedos, are included.

The most northern point of the mainland area is Cape Nordkyn in Norway, 71° 6′ N.; its most southern, Cape Tarifa in Spain, in 36° 0′ N.; its most western, Cape da Roca in Portugal, 9° 27′ W.; and its most eastern, a spot near the north Extreme points. end of the Ural Mountains, in 66° 20′ E. A line drawn from Cape St Vincent in Portugal to the Ural Mountains near Ekaterinburg has a length of 3293 m., and finds its centre in the W. of Russian Poland. From the mouth of the Kara to the mouth of the Ural river the direct distance is 1600 m., but the boundary line has a length of 2400 m.

Two of the most striking features in the general conformation of Europe are the great number of its primary and secondary peninsulas, and the consequent exceptional development of its coast-line—an irregularity and development which Coastline. have been one of the most potent of the physical factors of its history. The total length of coast-line was estimated by Reuschle in 1869 at 19,820 m., of which about 3600 were counted as belonging to the Arctic Ocean, 8390 to the Atlantic, and 7830 to the Black Sea and Mediterranean. This estimate, however, does not take into account minor indentations. Reclus’s estimate, including the more important indentations, brings the coast-line up to 26,700 m., and that of Strelbitsky up to 47,790 m. (smaller islands not included), or 1 m. of coast for about 75 sq. m. of area. Rohrbach[7] calculated the mean distance of all points in the interior of Europe from the sea at 209 m. as compared with 292 m. in the case of North America, the continent which ranks next in this respect. It must be pointed out, however, that such calculations are apt to be very misleading, inasmuch as the commercial value of the relations thus determined depends not merely on the existence of natural harbours or the presence of facilities for the construction of artificial harbours, but also on the presence of natural facilities for communication between such harbours and a productive interior.

The consideration just mentioned gives great significance to the fact that while the coast-line of Europe is in its general features very much the same as it was at the beginning of the true historic period, it has undergone a number of important Changes of coast-line. local changes, some at least of which are due to causes that are at work over very extensive areas. These changes may be conveniently classified under four heads: the formation of deltas by the alluvium of rivers; the increase of the land-surface due to upheaval; the advance of the sea by reason of its own erosive activity; and the advance of the sea through the subsidence of the land. The actual form of the coast, however, is frequently due to the simultaneous or successive action of several of the causes—sea and river and subterranean forces helping or resisting each other. That changes in the coast-line on the shores of the Gulf of Bothnia have taken place within historical times through elevation of the land seems now to be generally admitted. The commune of Hvittisbofjärd north of Bjorneborg on the Finland side of that gulf gained about 2¼ sq. m. between 1784 and 1894, an amount greater than could be accounted for by the most liberal estimates of alluvial deposit, and the most careful investigation seems to show that on the Swedish coast of that gulf a rise has taken place in recent years on the east coast of Sweden from about 57° 20′ N. increasing in amount towards the north up to 62° 20′ N., where it reaches an average of about two-fifths of an inch annually.[8] Our information is naturally most complete in regard to the Mediterranean coasts, as these were the best known to the first book-writing nations. There we find that all the great rivers have been successfully at work—more especially the Rhone, the Ebro and the Po. The activity of the Rhone, indeed, as a maker of new land, is astonishing. The tower of St Louis, erected on the coast in 1737, is now upwards of four miles inland; the city of Arles is said to be nearly twice as far from the sea as it was in the Roman period. The present St Gilles was probably a harbour when the Greeks founded Marseilles, and Aigues Mortes, which took its place in the middle ages, was no longer on the coast in the time of St Louis (13th century), but Narbonne continued to be a seaport till the 14th century. At the mouth of the Hérault, according to Fischer,[9] the coast advances at least two metres or about 7 ft. annually; and it requires great labour to keep the harbour of Cette from being silted up. The Po is even more efficient than the Rhone, if the size of its basin be taken into account. Ravenna, which was at one time an insular city like Venice, has now a wide stretch of downs partly covered with pine forest between it and the sea. Aquileia, one of the greatest seaports of the Mediterranean in the early centuries of the Christian era, is now 7 m. from the coast, and Adria, which gives its name to the sea, is 13. The islands on which Venice is built have sunk about 3 ft. since the 16th century: the pavement of the square of St Mark’s has frequently required to be raised, and the boring of a well has shown that a layer of vegetable remains, indicating a flora identical with that observed at present on the neighbouring mainland, exists at a depth of 400 ft. below the alluvial deposits. A little to the south of Rovigno on the Istrian coast on the opposite side of the Adriatic a diver found at the depth of about 85 ft. the remains of a town, which has been identified with the island town of Cissa, of which nothing had been known after the year 679.[10] At Zara ancient pavements and mosaics are found below the sea-level, and the district at the mouth of the Narenta has been changed into a swamp by the advance of the sea. A process of elevation, on the other hand, is indicated along nearly all the coasts of Sicily, at the southern end of Sardinia, the east of Corsica, and perhaps in the neighbourhood of Nice, while the west coast of Italy from the latitude of Rome to the southern shores of the Gulf of Salerno has undergone considerable oscillations of level within historical times. About the time of the settlement of the Greeks the coast stood at least 20 ft. above the level of the present day. Depression began in Roman times, though then the land was still 16 ft. higher than now. A more rapid depression began in the middle ages, so that the sea-level rose from 18 to 20 ft. above the present zero, and the coast began gradually to rise again at the close of the 15th century.[11] Passing eastward to the Balkan peninsula, we find considerable changes on the coast-line of Greece; but as they are only repetitions on a smaller scale of the phenomena already described, it is sufficient to indicate the Gulf of Arta and the mouth of the Spercheios as two of the more important localities. The latter especially is interesting to the historian as well as to the geologist, as the river has greatly altered the physical features of one of the world’s most famous scenes—the battlefield of Thermopylae.

If we proceed to the Atlantic seaboard we observe, as we might expect, great modifications in the embouchures of the Garonne and the Loire, but by far the most remarkable variations of sea and land have taken place in the region extending from the south of Belgium in the neighbourhood of the Straits of Dover to the mouth of the Elbe and the west coast of Schleswig-Holstein. Here there has been a prolonged struggle between man and nature, in which on the whole nature has hitherto had the best of the battle. While, as is well known, much land below sea-level in the Low Countries has been protected against the sea by dikes and reclaimed, and the coast-line has been, on the whole, advanced between the Elbe and the Eider,[12] there has been a great loss of land in the interior of Holland since the beginning of the Christian era, and on the balance a large loss of land north of the Eider since the first half of the 13th century.[13] In the 1st century A.D. the Zuider Zee appears to have been represented only by a comparatively small inland lake, the dimensions of which were increased by different inroads of the sea, the last and greatest of which occurred in 1395. Among the local changes of European significance within this area may be mentioned the silting up towards the end of the 15th century of the channel known as the Zwin running north-eastwards from Bruges, which through that cause lost its shipping and in the end all its former renown as a seat of commerce.

The Baltic shores of Germany display the same phenomena of local gain and loss. In the western section inroads of the sea have been extensive: the island of Rügen would no longer serve for the disembarkation of an army like that of Gustavus Adolphus; Wollin and Usedom are growing gradually less; large stretches of the mainland are fringed with submerged forests; and at intervals the sites of well-known villages are occupied by the sea. Towards the east the great rivers are successfully working in the opposite direction. In the Gulf of Danzig the alluvial deposits of the Vistula cover an area of 615 sq. m.; in the 13th century the knights of Marienburg enclosed with dikes about 350 sq. m.; and an area of about 70 sq. m. was added in the course of the 14th. The Memel is silting up the Kurisches Haff, which, like the Frisches Haff, is separated from the open sea by a line of dunes comparable with those of the Landes in France. The so-called strand or coast-lines at various altitudes round the Scandinavian peninsula, though belonging for the most part to glacial times, speak also of relative changes of level in the post-glacial period.

([Click to enlarge left side.])

([Click to enlarge right side.])