In Europe, as in North America, the Permian is developed in two very distinct facies. Southern Europe, Sicily, and the Alps have almost purely marine rocks and faunas, which resemble those of the Texas Guadalupian series. In central and western Europe the disturbances which, in many places, occurred at the end of the Carboniferous and, in those areas, produced a marked unconformity between the Carboniferous and Permian, resulted in the formation of a great inland sea, extending from Ireland to central Germany. In this great salt lake were deposited masses of red sandstones, shales, and marls, a predominant colour which strongly suggests desert conditions, though the coal-beds of France and Bohemia and central Germany might seem to contradict this. Occasionally the ocean broke into this closed basin, but the invading marine faunas soon perished. In Germany the Permian is in two very strongly marked divisions, the Rothliegendes, or Lower Permian, and the Zechstein, or Upper Permian, whence the period is, in that country, frequently called the Dyas. The very interesting discovery has quite lately been made in the lower Rothliegendes of Westphalia in Germany of undoubted and characteristic glacial moraines, resting upon a polished and striated pavement of Upper Carboniferous rocks. (G. Miiller, 1901.) These moraines are not thick, about four feet, and suggest local rather than general glaciation, but in the Midlands of England are boulder beds which apparently show ice-action, though here the evidence is less conclusive, for no striated glacial pavements have been found, and the glacial hypothesis is not accepted by most geologists.

The Lower Permian of Europe is remarkable for the great masses of volcanic rocks, lava flows and tuffs, which it contains, and which occur in Great Britain, France, Germany, and the Alps. This is in strong contrast to the corresponding American series, which gives no evidence of vulcanism.

Renewed disturbances at the end of the Lower Permian shifted the boundaries of the inland sea and changed its position, so that the Zechstein extends beyond the Rothliegendes and overlaps upon older rocks, and at the same time brought it into communication with the ocean, permitting the ingress of marine animals, but the conditions of life were evidently unfavourable, for the fauna is a curiously limited one, though a few species are individually abundant, and in striking contrast to the varied faunas of the truly marine facies. Later, the sea withdrew, leaving salt lakes, in which enormous bodies of rock-salt were formed in north Germany, including layers of the salts of potassium and magnesium, already referred to in a previous shapter. (See p. 225.) Smaller deposits of salt extend over central Germany to Russia. Upper Permian beds with gypsum occur in England, but not in France, which has only the Lower.

The Permian of Russia covers a very large area, but aside from typical Zechstein limestones on the Baltic coast, is quite different from that of western Europe; the principal area extends along the west side of the Ural Mountains to the Arctic Sea, and into Nova Zembla and Spitzbergen. The transition from the underlying Carboniferous is gradual, and the lower stage, the Artinsk, corresponds to the Wichita of Texas. Non-marine beds follow, which are again succeeded by a limestone with fossils like those of the German Zechstein, registering another invasion of the sea. The series ends with the Tataric stage, which is a sequence of red marls, passing upward without any apparent break into the overlying Triassic. The fossils of the Tataric stage are of peculiar interest and will be referred to again. Gypsum and salt in the non-marine beds testify to the aridity of the climate.

In Asia the two facies of the Permian are again met with. The marine facies occurs along a line which extends the course of the Mediterranean eastward, in Armenia, Persia, northern India, Tibet, China, and the island of Timor. In the central Himalayas Upper Permian rests upon the upturned and eroded Lower Carboniferous, and is conformably overlaid by Triassic beds. The Salt Range of northwestern India has a very full succession of the Middle and Upper Permian, the Productus limestone, resting upon continental formations of the Lower Permian. At the base of the latter is a boulder clay of glacial or iceberg origin, which is an outlier of a great ground moraine that covers large areas in central India, and in places reaches a thickness of 2000 feet. The smoothed, striated and characteristically glacial pave ment upon which the boulder clay rests, has also been observed. It is certainly very remarkable to find glacial deposits formed on such a scale within the tropics and evidently at no great height above the sea-level. The boulder-clay (Talchir) forms the base of the Gondwdna system, a succession of continental deposits, with much coal, laid down by apparently unbroken sedimentation, and including the Permian, Triassic, and Jurassic systems.

Northern Asia has a widespread area of continental deposits, which are presumably Permian, and the Artinsk of Russia reappears in central Asia, but the continent was mostly above sea-level and the great seas of the Carboniferous had withdrawn.

South Africa has a Permian development so closely parallel to that of India, that a direct land connection between the two regions may be confidently inferred. Extending almost across the continent from east to west in Cape Colony and Natal is the thick (1000 feet) glacial boulder clay of the Dwyka, or Lower Permian. Part of the Dwyka is shale of subaqueous origin, but most of it is a mass of boulders, striated and faceted, embedded in a fine unstratified matrix. (See Figs, 111, p. 231, 115, p. 233.) The formation extends northward, growing thinner on the way, into the Transvaal and perhaps into Rhodesia, and in the more northerly areas the underlying ice-worn pavement of older rocks is exposed in a state of wonderful freshness (see Figs. 70 and 275) entirely comparable to the recently abandoned beds of the shrinking glaciers of the Alps. In South Africa, therefore, the remarkable phenomenon of a continental glaciation in and near the tropics presents itself as well as in India, with the additional difficulty that in the former region the movement of the ice was from the Equator polewards.

Following the Dwyka boulder clay are the continental strata of the Karroo system, the lower part of which is Permian and in the Transvaal coal-bearing, and which corresponds to the Indian Gondwana in character, in geological date, and in the contained fossils.

Fig. 275. - Roche moutonnee, exposed by removal of Dwyka boulder clay. - Riverton, Cape Colony. (R. B. Young).

The eastern portion of Australia and Tasmania, which had been land during the Upper Carboniferous, was largely submerged during the Permian, but Victoria on the south coast remained above sea-level and was glaciated, with the formation of the familiar ice-pavements and thick boulder clays, interstrati-fied with sandstones and shales. The Upper Permian is coal-bearing, as it is also in New South Wales, where the glacial series is divided into two distinct parts by coal measures. The glacial beds occur interstratified with marine strata and some of the ice-made layers themselves contain marine fossils, which leads to the inference that the great blocks were deposited by icebergs rather than directly by glaciers. In Queensland and in northwestern Australia only the upper boulder succession is found, and the overlying coal measures are there also. The Australian ice movement was from south to north, as would naturally be expected in the southern hemisphere, but this makes the direction of movement in South Africa only the more inexplicable.

In New Zealand the Permian, which is reported to be 7-10,000 feet thick, contains neither coal nor evidence of ice-action, but includes lava-flows.

In South America Permian beds of continental origin are found in Argentina and southern Brazil. In the latter is a glacial boulder clay, followed by a great series of strata which resemble those of the South African Karroo system.

The distribution of the Permian rocks and fossils leads to the inference that at that period the continents were so grouped as to form two great land masses, a northern including North America, most of Asia, and Europe, and a southern comprising India, equatorial and southern Africa, Australia, and South America. The existing southern continents were probably then connected by comparatively narrow land bridges across the site of the present Atlantic and Indian oceans. ' Between the northern and southern land-masses was the great continuous mediterranean, a sea which has been named "Thetys" (Suess) and of which we have found indications in Texas, Sicily, the eastern Alps, Asia Minor, northern India and southern China to the Pacific. That land communication was occasionally, at least, established between the northern and southern lands is evident from the distribution of the land animals and plants of both regions.

Climate

The plain and obvious inferences from the character of the Permian rocks are so remarkable and inexplicable that they were long received with incredulity and they offer a series of fascinating problems for which no solution can yet be found. The earliest Permian in the southern hemisphere was a time of vast glaciation and of rigorous climate, as is convincingly shown by the boulder clays and ice pavements of Australia, South Africa, and South America. In the northern hemisphere the glaciation was extensive in peninsular India, apparently local in Germany and perhaps in England. In several regions, as in West Virginia, France, Germany, and Bohemia, there are workable coal measures in the Lower Permian, but arid conditions established themselves over all parts of the northern hemisphere where Upper and Middle Permian rocks are found, western North America, Texas, Kansas, England, Germany, and Russia. The extreme conditions of desert climate are registered in the great bodies of gypsum and rock-salt which characterize so much of the Permian areas.

What can have caused these climatic vicissitudes and especially the development of continental ice-sheets so near the Equator and so little above sea-level, is a problem for which many solutions have been propounded, but none of them is convincing.