ii. 358), who makes Thule the farthest of the Orcades; by the

philosopher Petrus Ramus (de la Ramée); by Johannes Myritius, who rather cleaves to the end of Britain; by the learned Vossius, who prefers the Hebrides or Orcades; by Buchner (Ad Tacit. Agric., cap. 10); by Camden, by Gosselin, and others. Stephanus Byzantinus says: “Thule insula magna in oceano sub Hyperboreas partes, ubi æstivus dies ex viginti horis æqualibus constat, nox verò ex quatuor. Hyberna verò dies à contrario.” This calculation would place Thule three degrees south of the Polar circle, and would better suit the Færoe archipelago (N. lat. 61° 23´ to 62° 26´ 40´´). Forcellini understands Cellarius also to refer to the Færoes; De Kerguelen Tremarec (Voyages) opines for Iceland.

IV.

THULE = SCANDIA.

It has been seen that Pliny (Nat. Hist., iv. 16) apparently separates Norway from Thule; moreover, that Ptolemy (ii. 3) confirmed by Agatharcides and Stephanus Byzantinus (lib. i., in extremis), whilst pointing to North Britain and to Scandia, or Scandinavia, in his time held to be an island,[37] and little known to the civilised world, adds details which rather belong to Iceland. On the other hand, it is evident that during the later Roman empire, Thule was applied to Scandinavia.

Procopius, the Byzantine historian (nat. circ. A.D. 500), leaves no doubt upon this point. He devotes to it a considerable space (lib. ii., De Bello Gothico, c. 15), and his account will be little abridged. After relating how a party of Heruli, when conquered by the Longobardi, passed through the lands of the Slavini, the Varni (Οὐάρνοι, al. Harmi), and the Dani (Δάνοι, al. Dacæ), till they reached the ocean, he makes them take ship and settle at Thule:

“The island is ten times larger than Britain, and far to the north.[38] The greater part of it is desert. The inhabited region contains thirteen great peoples, each governed by its own king. A curious phenomenon is reported from that place: every year, about the summer solstice, the sun remains forty days above the horizon. Six months after this there is a night of forty days, a time of sorrow, when all intercourse and business are at an end. I (says Procopius) was greatly desirous of seeing this marvel for myself, but the opportunity was ever wanting. I therefore asked those who had been there how the sun rises and sets. They told me that for forty consecutive days, the sun lights the island; sometimes from the east, at other times from the west; but that when he returns to the same point where he appeared, a single day is counted. During the season of forty nights, time is measured by the moon. When thirty-five of these long and lasting nights have passed, some of the people ascend the highest mountains, and give warning to those below that after five days more they will see the sun. The Thulitæ rejoice over the good news, and celebrate in the dark a festival which in ceremony exceeds all their others. Although this happens every year, still it would appear the inhabitants apprehend a total desertion of the sun.

“Amongst the barbarian peoples of Thule, none are so savage as the Skithifini (Σκιθίφινοι, al. Scritifini). Like beasts,[39] they ignore clothes and shoes; they drink no wine, and they eat nothing which the earth grows. Both men and women, who will not take the trouble of cultivation, occupy themselves exclusively with hunting, and the forests and mountains supply them abundantly with game. They eat the flesh, and, being without flax and wool, they wear the skins, which they fasten with sinews, having no knowledge of sewing. Also, they do not bring up their offspring like other people. The children of the Thulitæ are fed upon the marrow of beasts, instead of being suckled by their mothers. When the woman has been delivered, she wraps her babe in a skin, secures it in another, places some brains in its mouth, and sets out with her man for the chase, in which both sexes equally excel. The Thulitæ adore several gods and demons, some of whom they believe to inhabit the sky, others the air; some are on the earth and in the sea, whilst others of the smaller kind, affect the rivers and springs. They often offer sacrifices and immolate all manner of victims, the most acceptable being the first man captured in war; he is sacrificed to Mars (Thor?), the most powerful of all their gods. On these occasions they do not simply slay the victim, they either hang him to a tree, or roll him over thorns, or put him to death in some other way, choosing the most cruel.

“Such are the customs of the _Thulitæ_, amongst whom are the Goths (Γαυτοί), a fecund people that gave land to the Herulian immigrants. The remnants of this race who lived amongst the Romans, after slaying their king, sent their chief worthies to the island of Thule, for the purpose of finding if any of the royal blood there remained. The deputies were successful, and chose out of many one who pleased them the most. But as he died on the way, they returned (to Thule) and brought with them one Todasius (Τοδάσιος, al. Datis); this man was accompanied by his brother named Aordus (Ἄορδς) and by two hundred youths of the island.”

This description of Thule is evidently great Scandinavia, not little Iceland. Hence Ortilius (Thesaurus sub voc.) D’Anville, who rejects Iceland; Farnaby, Schœnning (Von Nordich. Land in Neue Allg. Welt-Gesch, vol. xiii., p. 14, et seq.); Rudbeck, who understands Sweden; Murray (loc. cit.); Wedel (Alhandlung über die “Alt-Scandinavische Gesch.,” p. 32, et seq.); Schlözer (Allg. Nordisch. Gesch, pp. 14, 16), Parisot, and other geographers, have referred the descriptions of Procopius especially to the Norwegian canton still called “Tyle-mark,” or “Tile-mark.” Maltebrun (iii. 6) prefers Jutland, on the continent of Denmark, part of which, he hears, is still termed “Thy” or “Thy-land.” Calstron believed that all Scandinavia was meant. Celtes (Schardius, Basil ed., p. 59) makes Iceland “one of the isles of the ocean,” together with Scandia, Dania, Suecia, etc. Adelung (Mithridates) supports the claims of Norway. Others go as far as Lapland, and even Greenland has not been without claimants to the honour. Yet in the sixth century, Jornandes (De Origine Actuque Getarum Liber, p. 393, Basle edition of 1531), after mentioning the thirty-four Orcades, says, “Habet et in ultimo plagæ occidentalis aliam insulam nomine Thyle, de quâ Mantuanus, Italia, ‘tibi serviat ultima Thyle,’” and he carefully distinguishes it from the “ampla insula nomine Scanzia.”[40]

V.

THULE=ICELAND.

It has been shown that the accounts of Pytheas, supported by details from Pliny and Ptolemy, refer only to Iceland. They are confirmed by the following authorities. In Caius Julius Solinus (A.D. 230; 2 vols. fol., Traj. ad Rhenum, 1689), we find Thule five days’ sail from Orkney, and we cannot allow less than 100 knots for the δρόμος νυχθήμερος, or a total of 500 direct geographical miles; the run from northern Orkney to the south coast of Iceland being about this distance. The Polyhistor, held an oracle in the Middle Ages, adds (chap, xx., lll):

“Inter multas quæ circa Britanniam sunt insulas, Thylen ultimam esse commemorat. In quâ æstivo solstitio dicit esse noctem nullam. Brumali verò perinde diem nullum.”[41]

Orosius, whose history (London, 8vo, 1773) extends to A.D. 417, says:

“Tylen per infinitum à cæteris separatam undique terris in medio sitam oceano vix paucis notam haberi.”

Isidorus Hispalensis (A.D. 600-636; Orig. Seu Etym., xiv. 6; Opera Omnia, fol., Parisiis, 1601) appears to repeat Pliny:

“Thyle verò ultimam oceani insulam inter Septentrionem et occidentalem plagam,[42] ultra Britanniam sitam esse describit, à sole nomen habentem, quia in eâ æstivum solstitium sol faciat, et nullus ultra eam dies sit. Ultra Thylen vèro pigrum et concretum mare.”

The last sentence of the bishop being emphatically true in winter. Other authorities who identify Thule with Iceland, are Cluverius (Germ. Ant., ii. 39), Harduin and Dalechamp (Ad Plin.), Bougainville (c. 1, p. 152), Hill (Ad Dionys.), Penzel (Ad Strab.), Pontanus (Chorog. Dan. Descrip., p. 74), Isaac Thilo (Dissert., Lips., A.D. 1660), Gerhard Mercator, and Mannert (Geog., i., p. 78), to mention no others. Martin (Histoire des Gaules, i. 159) takes the Gauls to Iceland.

In the ninth century we have positive evidence that Thule had returned to its oldest signification, Iceland. The monk Dicuilus, who wrote in the year 825,[43] relates that thirty years before that date (A.D. 795) he had seen and spoken with several religious who had inhabited the island of Thule between February and August. He asserts that Iceland and the Færoes had been discovered by his countrymen; and his calculation of the seasons and the days at different times of the year, together with the assertion that a day’s sail thence towards the north would bring them to the Frozen Sea, shows that “Iceland, and Iceland alone, could have been the island visited by the anchorites.”

The Domesday Book of the north, the “Landnámabók,” whose lists of 1400 places and 3000 persons were drawn up by various authors in the twelfth century, supported, according to Mr Blackwell (note, p. 189), “by other ancient Icelandic documents,” simply states (Prologus, p. 2), “Before Iceland was settled by the Northmen there were men there called by the Northmen Papæ. These men were Christians, and are thought to have come from the west, for there were found Irish books, bells (biöllur), staves (baglar), and various other things, whence it is thought that they were Westmen,” Irishmen--a name still preserved in the Vestmannaeyjar archipelago. Moreover, we learn that these relics were found in Papey (the Isle of the Papæ), a rock off the eastern coast, which still bears the same name, and at Papyli, in the interior; and finally, that “the Christians left the country when the Northmen settled there”[44]--the latter being pragmatical pagans.

Mr Blackwell concludes that these people were probably fishermen from the north of Ireland and the Western Isles of Scotland, who may annually have frequented the northern seas, and made Papey one of their winter stations. Mr Dasent (i., vii.) more justly identifies them with the Papar or Culdees (?), a class of churchmen who have left their traces in almost every one of the outlying islands of the west. Under the name of “Papar” we find them in the Orkneys and Shetlands, the Færoes and Iceland; “and to this day the term ‘Papey’ in all these localities denotes the fact that the same pious monks who had followed St Columba[45] to Iona, and who had filled the cells at Enhallow and Egilsha and Papa, in the Orkneys, were those who, according to the account of Dicuil, had sought Thule or Iceland that they might pray to God in peace.”[46] These Culdees were not likely to spread, as they carried no women, but they left traces of their occupation in their cells and church furniture.

The simple story told by Dicuil is eminently suggestive. Thus Thule became, probably for a second time, one of the “Britanniæ,” the Isles of Britain; and we may consider the discovery a rediscovery, like the central African lakes, whence Ptolemy derived the Nile. When the rude barks of the eighth century could habitually ply between Ireland and Iceland, we cannot reject as unfit the Roman galleys, or even the Phœnico-Carthaginian fleets. The Periplus of Himilco was not more perilous than the Periplus of Hanno, and the Portuguese frequented the northern seas long before they had doubled Cape Horn. Bergmann had evidently no right to determine that Iceland was not “Ultima Thule,” _because_--(1.) The Romans were bad sailors; (2.) They were in the habit of writing “Rome--her mark” wherever they went, whereas no signs of their occupation are visible in Iceland; and (3.) Because Iceland was probably raised from the sea at the time when the Vesuvian eruption buried Herculaneum and Pompeii.

It is true that Roman remains have not yet been discovered in Iceland, but this is a negative proof which time may demolish; moreover, the same absence of traces characterises the Papar occupation which we know to have been a fact. On the other hand, Uno Von Troil speaks of a ruined castle near “Videdal” (Viðidalr), some 200 perches in circumference, and smaller features of the same kind on the glebe of Skeggestað, near Langanes. Mr Henderson[47] declares of Hrutur’s cave, or rather caves--a vast apartment 72 feet long by 24 broad and 12 high, within which is a small recess 15 feet by 9, apparently a sleeping place--that both “are said to have been cut by people in former times.”

We are, then, justified in concluding that we need no longer question with Synesius, if such a place as Thule exists, or doubt with Giraldus Cambrensis, whether it has yet been discovered. We may follow A. W. Wilhelm (Germanien, etc., 1823), and believe with the Teatro Grande Orteliano, “Islandia insula, veteribus Thyle dicta, miraculis si quæ alia clarissima.” We may agree with Mannert that Iceland might have been discovered by Pytheas the Phocæan, and even by the Carthaginians. We may even support what appears to be rather an extreme opinion:

“Pytheam præterà increpat Strabo ut mendacem, qui Hiberniam et Uxisamam (Ushant) ad occidentem ponit à Galliâ, cum hæc omnia, ait, ad Septentrionem vergant. Itaque veteres geographi Hiberniæ situm definiunt meliùs quam scriptoris seculi aurei Augusti, Himilco et Phœnices meliùs quam Græci vel Romani” (Rer. Script. Hib., prol. i., xii.).

Moreover, it appears certain that the old tradition of Thule, though different ages applied the word differently, was never completely lost; and that the Irish rediscovered the island before the eighth century, if not much earlier, when the official rediscovery dates from the ninth, and the earliest documents from the eleventh and twelfth.

The Venerable Bede (eighth century) speaks of Iceland under the name of Thyle, more than a hundred years before its official discovery by the Scandinavians; and Alfred (ninth century), in his translation of Orosius (p. 31), assures us that the utmost land to the north-west of Ireland was called Thila, and that it was known to few on account of its great distance. Yet even after the occupation of Iceland by the Northmen, we find in the literary world the same vagueness which prevailed in earlier ages. For instance, Isaac Tzetzes (twelfth century), in his notes on Lycophron, calls the fabled Fortunate Islands of the Greeks “the Isle of Souls, a British island between the west of Britain and Thule towards the east,” which is impossible. But in the fifteenth century Petrarch has left us a valuable notice of the knowledge then familiar to men of letters (De Situ Insulæ Thules, epist. i., lib. iii., De Rebus Fam., vol. i., pp. 136-141, ed. 1869, J. Fracassetti, Le Monnier. Florentia). In reply to his own “Quæro quiânam mundi parte Thule sit insula?” he quotes Virgil, Seneca, Boethius, Solinus, Isidore, Orosius, Claudian, Pliny, and Mela. He could obtain no information from “Riccardo, quondam Anglorum regis cancellario”--Richard de Bury was probably too busy for such trifles. He learned something, however, from the “Libellus de Mirabilibus Hiberniæ, à Giraldo (Cambrensi) quodam aulico Henrici secundi, regis Anglorum.” And after quoting this “scriptorum cohors,” he thus ends with “pointing a moral”--“Lateat ad aquilonem Thyle, lateat ad austrum Nili caput, modò non lateat in medio consistens virtus,” etc.[48]

Icelandic Thule was advocated by Saxo Grammaticus; but his opinion was strongly opposed by his commentator (Johannis Stephanii, Notæ Uberiores in Hist. Dan. Sax. Gram. Soræ, ed. 1644, fol.). The words of the latter’s preface are--“Ex opinione magis vulgari, quam rei veritate _Thylenses_ ubique nominat Saxo, qui Islandi rectius dicerentur;” but he relies chiefly upon the controvertible arguments of “Arngrimus Jonas.” Iceland was opposed by Gaspar Peucerus (De Terræ Dim.), by Crantzius (Præfatio in Norvagiam, borrowed from Nicolaus Synesius, epist. 148); by Abraham Ortelius (Theatrum Orbis and Thesaurum Geographicum), and by Philippus Cluverus (Germania Antiqua). The globe of Martin Behaim (A.D. 1430-1506) shows a certain knowledge of details: “In Iceland fair men are found who are Christians. The custom of its inhabitants is to sell dogs at a very high rate; while they willingly part with some of their children to merchants for nothing, that they may have sufficient to support the remainder. Item.--In Iceland are found men eighty years old who have never tasted bread. In this country no corn grows, and in lieu of bread dried fish is eaten. In Iceland it is the stock fish is taken which is brought to our country.”

THULE (ETYMOLOGY OF).

Perhaps the origin of “Thule” is ground more debatable and debated than even its geographical position.

“Some,” says Sibbald, “derive the name Thule from the Arabic word Tule (طول = Túl), which signifies ‘afar off,’ and, as it were with allusion to this, the poets usually call it ‘Ultima Thule;’ but I rather prefer the reason of the name given by the learned Bochartus,[49] who makes it to be Phœnician, and affirms that it signifies ‘darkness’ in that language. Thule (צל) in the Tyrian tongue was ‘a shadow,’ whence it is commonly used to signify ‘darkness,’ and the island Thule is as much as to say, an ‘island of darkness;’ which name how exactly it agrees to the island so called at the utmost point to the north is known to everybody.”

Others find Thule in the Carthaginian צל = “obscurity;” the Hebrew has צלל, and the Arabic ظل = obscuravit.”

After using or abusing the Semitic tongues, we come to Greek, which puts forth three principal claimants: θόλος = fuscus color, caligo; τέλος, a goal; and τηλὲ, procul. Meanwhile Isidorus (Orig. Seu Etym., lib. xiv., 6) derives Thyle, as has been shown, from the sun and its solstice. In the twelfth century, Suidas (Lex. sub voc.) makes Thulis (θούλις) a king who reigned over Egypt and the isles of the ocean, one of which was called after his name.

Etymologists presently applied themselves to the Gothic languages and their derivatives; and they did not reject geographical resemblances. Pontanus (loc. cit., i., p. 746) asserts that the islands about the Norwegian coast were generally called Thuyle. Ortelius (Thesaur. and Theatr. Orbis, p. 103), relying upon Ptolemy’s latitudes and longitudes, declares that “Thilir” was the term applied to the people of Norwegian “Tilemark;” the latter word is also written “Thulemarchia” (Johannes Gothus); “Thielemark,” “Thylemark” and “Tellemarck” (Pontanus).[50] Not a few writers refer “Thule,” as has been said, to “Thy” or “Thy-land,” the extreme point of Jutland. The commentator on Saxo Grammaticus, before referred to, records a derivation of “Thule:”

“Quod vel instar _Tholi_, cujusdam orbis terrarum sit imposita, vel quod eo navigantes ad ploratum (tothülen Belgæ dicunt) proficiscerentur.”

In p. 175 he becomes still more vague:

“Rectius itaque Velljus nostro, juxtà ac M. Christiernus Petri, primus Saxonis interpres, reddidere Blend aff Telløe vel Blend aff Tyløe. Quænam verò iste sint insulæ, juxtà scimus cum ignarissimis.”[51]

Prætorius (De Orbi Goth., iii. 4, § 3) deduces “Thule” from the Gothic “Tiel,” “Teule,” or “Tuole” (= τέλος, finis), meaning a long distance, and denoting the remotest land; he doubts the existence of the place, with D’Anville (Mem. de Paris, vol. xxxvii., p. 439). Reinerus Reineccius (Reinech, Historiæ tam Sacræ quam Profanæ Cognitio, Frankf. et Lipsiæ, 1685, and Methodus Legendi, etc., Historiam tam Sacram quam Profanam, Frankf. 1670) advocates the Saxon “Tell,” meaning a limit--limes septentrionis atque occidentis. Dr Charnock compares the Saxon “Deel,” a part or portion, and quotes Wachter (Gloss. Germ.), who gives amongst other meanings of “Teil” (hod. Theil), pars, portio, segmentum, and “teilen,” _i.e._, dividere in partes.

Torfæus (Hist. Norwegiæ, i. 5, p. 12) proposes a variety of derivations. Wilhelm Obermüller (Wörterbuch, etc., Williams and Norgate, Lond. 1872) would explain “Thule Procopiana,” by Dal (a dale), or “Tulla,” also written “Tolin” and “Tullin,” a meadow or pasturage; and he remarks that Norwegian “Tellemark” or “Thilemark,” is of the same descent. The Thracian Kelts had a kingdom of Tyle, which here probably signified “Dail,” a fortress. When Pliny makes men sail from Nerigos to “Thule,” the latter might have meant _Du-ile_, “the little island,” or perhaps “the dark (‘dubh,’ cloudy and wintry) isle.”

Even the orthography of “Thule” is disputed, and there are sundry variants--Thula, Thyle, Thile, Thila, Tyle, and Tila. The popular Greek form adopted by Strabo, Ptolemy, Agathemerus, Isidorus, Jornandes (De Reb. Get., cap. 1, 1), Procopius (De Bell. Goth., ii. 15) and Stephanus Byzantinus, is θούλη, which in Romaic would be pronounced “Thúle;” the ethnic being θουλαῖς (Thulæus), and θουλίτης (_plur._ θουλίται). The Latins (Mela, Pliny, Tacitus, Anonymus Ravennæ, Martianus, Solinus, etc.) seem to have preferred “Thule;” and Cluverius (Germ. Ant., iii. 39) rejects all others as barbarous. The learned and humorous Salmasius (in Solin., cap. xxii.) declares that “Thyle” ought never to be written, despite many good codices of Virgil, Pliny, Jornandes, Isidore, the Anon. Ravennæ, and others, which give Thyle and even Tyle, θύλη and θυλίτης; Æthicus (in Cosmog., p. 730), borrowing from Orosius, has “Tilæ;” Boethius (xx. 11), “Tile” and “Ðyle.”

We here conclude the subject of Thule, “celebrata omnium litteris insula.” To do it full justice, and especially to quote from the “cohort” of modern writers, would require a volume.

SECTION II.

PHYSICAL GEOGRAPHY OF ICELAND.

§ 1. GENESIS AND GEOLOGY.

“Iceland owns its existence wholly to submarine volcanic agency”--such is the statement generally made by travellers and accepted by readers. The genesis of this “Realm of Frost and Fire;” this “fragment of earth white with snow, black with lava, and yellow with brimstone;” this “strange trachytic island, resting on an ocean of fire in the lone North Sea,” where the “primary powers of nature are ever at war with one another,” is compared with the efforts, vastly magnified, which in 1811 threw up from the waters Azorean Sabrina to a height of 480 feet above sea-level. And many have assumed as its exemplar the three-coned Nyöe (Nýey) that rose during the Skaptár eruption (1783), some thirty miles south-west of Reykjanes, and sank into a subaqueous reef before the end of the same year.[52]

This is true, but not the whole truth. The basis of Iceland was recognised by Baron Sartorius Von Waltershausen to be the Palagonite[53] which forms the foundation of volcanic tufas on Etna, the Azores, Tenerife, the Cape Verds, and other Plutonic regions. It is known to the people as “Mó-berg,” the _saxum terrestre-arenosum_ of Eggert Olafsson, translated by the dictionaries Clay-soil, but generally used in contradistinction to Stuðlaberg,[54] hard stone, the basalts, basaltites, dolerites, and others of their kind. By the older travellers, as Henderson, it is termed sandstone, and _conglomérat-basaltique_, while not a few have confounded it with trachyte. In Iceland this mineral substance, rather than mineral, is a far more important feature than even in Sicily.

By virtue of its composite character and different colour, this hydrosilicate of alumina is a Proteus; massive and amorphous; crystalline, muddy, sandy, and ashy; friable, porous, and spongy like lava and pumice; granular, silicious, and arenaceous; heavy and compact like slatey clays; vitreous and semi-vitreous with the lustre of pitch-stone. It is as various in tint as in texture; usually ferruginous brown, dark brown or dun yellow; grey and slate-coloured; dark with hornblendic particles; pure white where it is converted into gypsum, clay marl, and limonite with the aspect of chalk, by exposure to the action of sulphurous acid; green tinged with olivine; garnetic-red; ochreous, the effect of iron; and at times showing a ferreous coat of pavonine lustre. Palagonite lava is often “of so deep a brick-red colour that it resembles an iron slag, were it not for its superior lightness.”

Here, this Palagonite degrades to the yellow sand which contrasts so remarkably with the black Plutonian shore; there, in the lowlands it shows fissile strata horizontal like sandstone, and at times marly couches. It paves the soles of valleys and the floors of rivers; and it rises on the surface of the loftiest Heiðar (highland heaths), where earth is worn down to the very bone by rains, snows, and winds. Now it towers in huge cliffs and scaurs, irregular masses of rock overlying or underlying the traps; then it bulges into high belts of country, sierras and detached mountains, like Herðubreið and others which will afterwards be mentioned. Consolidated and in places crystallised by heat and high pressure, this produce of submarine volcanoes was elevated by the long continued action of quietly working forces, but it still displays its subaqueous origin. Firstly, it is a hydrate containing 17 to 25 per cent. of water; secondly, it is stratified as if formed of hardened ashes and modified lavas; and, thirdly, it contains broken mollusks[55] of marine types still existing, and the silicious skeletons of infusoria: a negative proof is that we never meet with it among volcanic tuffs subaërially deposited. In places it becomes an acute-angled breccia, enclosing basalts and lavas varying from the size of a pin’s head to that of a man, or rounded conglomerates suggesting that the foreign matter was deposited in a shallow sea. The fresh appearance of the shells and the presence of infusoria also tend to prove that it was deposited in a heated, at least not in a gelid sea.

Professor Tyndall finds in Palagonite the first stage of the fumarole: “If a piece be heated with an excess of aqueous sulphuric acid, it dissolves in the cold to a fluid, coloured yellow-brown by the presence of peroxide of iron. On heating the fluid, the peroxide is converted into protoxide; a portion of its oxygen goes to the sulphurous acid, forming sulphuric acid, which combines with the basis of the rock and holds them in solution.” But the resultant springs show no trace of oxide of iron which has been dissolved and has disappeared. “The very rock from which it was originally extracted, possesses the power of re-precipitating it, when by further contact with the rock, the solution which contains it has its excess of acid absorbed, and has thus become neutral. In this way, the aqueous sulphurous acid acts as carrier to the iron, taking up its burden here, and laying it down there; and this process of transference can be clearly traced to the rocks themselves.”

Upon this Palagonite floor, the “Protogæa,” or oldest formation, were laid immense tracts of sand and stratified ejections of “trap.” According to Macculloch, “the word is a cloak for ignorance which saves the trouble of investigation.” But it is still a general term for the older, lighter, less earthly and basic, and more crystalline forms than the basalts, containing intercalated pumice-tuffs deficient in shells, whilst the cavities abound in zeolites and amygdaloids.[56] Concerning the strike and dip of the trap-strata, which rise sheer from the sea, in grades and layers, steep, angular, and bare, and which outline the mural copings and stepped cones of the old coast and the jaws of the river-gorges, there are many conflicting opinions. Some hold that the strata all incline gradually and quaquaversally, more or less, towards the centre of the island; whilst others find that as a rule, they are horizontal. The expedition led by Prince Napoleon (1857) recognised convergence, and often a slope of 15° towards the grand foci of eruption that form the respective systems; for instance, the _inclinaison rayonnante_ towards Snæfellsjökull. The author could lay down no rule, except that the steps, viewed in profile, especially from the gashes and torrent-beds, appear to recede rather than to project, to dip inland rather than seawards. The strata vary in number to a maximum of fifty; they are perpendicular courses separated by débris, and sometimes footed by déblai and humus, disposed at the natural angle--this regularity again suggests submarine deposition, and everywhere attracts the stranger’s eye.

Professor Bunsen divides the rocks of Iceland, and probably those of most other volcanic systems, into two great groups: (1.) _Normal Pyroxenic_, the basalts and dolerites, whence silica is almost absent; and (2.) _Normal Trachytic_, abounding in that mineral. The basalts[57] are of two kinds, the true, rich in, and the basaltite, which notably wants, olivine. Both are either honey-combed with drusic cavities, or perfectly compact and fine-grained; the water-rolled pieces are soft, and smooth as marble. The basalts pass by almost imperceptible degrees into dolerites (green-stones) coloured by admixture of chlorite, and often containing iron pyrites. Of less importance as a geological feature, are the masses, veins, and crests of trachyte which pierce the Palagonites, the traps, and the basalts. The rock which is compared with the chain of the Puys (Auvergne), occurs, however, in an altered form at many places unsuspected by old travellers, and every explorer adds to its importance. From Reykjavik appear two gold-yellow and white-streaked peaks, associated with jasper and other forms of quartz. The Snæfellsjökull peninsula is also for the most part trachytic. The celebrated Baula (the cow), a cone rising 3000 feet high, contrasts the mechanic neatness of its whitey-grey pillars[58] with its red neighbour, Little Baula, and with the surrounding chaos of darkness; and heat-altered trachytes are found about Hekla and the Geysir. The green trachyte of Viðey, apparently tinted by chlorite, was found to contain silica, alumina, iron, and traces of magnesia. Daubeny, and a host of writers, assumed that a trachytic band, disposed upon a rectilinear fissure 200 kilometres long, bisects the island from south-west (Reykjanes) to north-east (Langanes), and represents the original Iceland, as the Longmynd and Stiper Stones are the nucleus of England. Moreover, the great centres of eruption, igneous and aqueous, were disposed upon this diagonal, flanked by the earlier Plutonic masses. Lastly, the modern volcanic chimneys were all theoretically opened in the old and new trachytic domes. M. Robert (1835) especially sought and failed to find the “trachytic band,” and, since Von Waltershausen’s visit, it has been determined that the material is the Palagonite floor traversed by the Geysir and by most of the active volcanoes.

The peculiar contrasts of the island are thus noticed by an old writer: “The king of Denmark is still master of Iceland, which is supposed to be the _Ultima Thule_ of the ancients. The surface, though it is covered with snow, nevertheless contains burning mountains, whence issue fire and flames, to which the Iceland poets compare the breasts of their mistresses. It has also smoking lakes, which turn everything thrown into them to stone, and many other wonders which render this island famous.” Iceland, like Tenerife, owes its present general contour to subaërial volcanic action of the post-Tertiary period, the secular growth of the detached regions overlying the pockets and foci of eruption, as explained by Von Buch, together with the gradual accretion, the gift of exit-chimneys and dejections from the Plutonic cauldrons. The normal pyroxenic was followed by the felspathic formations, trachytic, acid and pumiceous, which, though comparatively modern, still date from immense antiquity. The distribution into fire-vents (true volcanoes) and sand-vents (pseudo-volcanoes), will be noticed in a future page.

The lava is composed of trachytic (silicious) and doleritic (basic) ejections, varying in weight;[59] the stone averages about half the specific gravity of granite, and in a molten state it flows at the rate of 50 to 100 yards per diem. When first cooled, the ejections are lamp-black; they are then tarnished by oxygen to brown; they become grey with lichens; and finally, the lapse of ages converts them into humus. To the latter process, Brydone, on Etna, assigned 14,000 years, and greatly scandalised our grandsires, who held sound opinions upon the date (B.C. 4004) empirically assigned to creation. We can hardly forget poor Cowper’s poor verse, and poorer sense:

“Some drill and bore The solid earth, and from the strata there, Extract a register, by which we learn That He who made it, and revealed (!) its date To Moses, was mistaken in its age.”[60]

The following is a list of the principal orographic features, Jökulls,[61] Fells (mountains), volcanoes, masses of Palagonite, snow-peaks, and true glaciers, which are rare. Gunnlaugsson’s astronomical positions are given in Danish feet, and the former are reduced to the meridian of Greenwich by assuming Copenhagen to lie east 12° 34´ (Rafn, 12° 34´·7). The Danish foot is calculated at 12·356 inches English, or about 67:69.

The north-eastern quarter numbers fifteen points, ranging from 1000 to 3000 Danish feet, and the following ten exceed the latter:

Dan. Eng. feet = feet. N. lat. W. long. (C.)=Greenwich.

Lambafell, 3459 3562 64° 58´ 28´´ 26° 39´ 19´´ 14° 5´ Herðubreið, 5290 5447 65° 10´ 39´´ 28° 58´ 55´´ 16° 25´ Gagnheiðarhnúkr, 3009 3098 65° 13´ 35´´ 26° 53´ 42´´ 14° 20´ Beinageitarfjall, 3517 3621 65° 27´ 37´´ 26° 42´ 2´´ 14° 8´ Dyrfjöll, 3606 3713 65° 31´ 20´´ 26° 35´ 17´´ 14° 1´ Smjörfjall, 3859 3973 65° 36´ 40´´ 27° 24´ 6´´ 14° 50´ Heljarfjall, 3991 4109 65° 48´ 26´´ 31° 31´ 56´´ 18° 58´ Rimar, 4020 4139 65° 52´ 45´´ 31° 7´ 33´´ 18° 33´ Ólafsfjarðarfjall, 3272 3369 65° 58´ 34´´ 31° 31´ 8´´ 18° 57´ Kaldbakr, 3699 3810 66° 0´ 24´´ 30° 48´ 58´´ 18° 15´

In the south-eastern quarter, nine heights range from 1000 to 3000 Danish feet, and eleven rise higher, viz.:

Dan. Eng. feet = feet. N. lat. W. long. (C.)=Greenwich.

Stórhöfði, 4509 4643 63° 55´ 34´´ 29° 17´ 7´´ 16° 43´ Staðarfjall, 3782 3894 63° 57´ 55´´ 29° 12´ 51´´ 16° 39´ Öræfajökull,[62] 6241 6426 64° 0´ 48´´ 20° 20´ 16´´ 16° 46´ Thverártindsegg, 3668 3776 64° 11´ 14´´ 28° 46´ 12´´ 16° 12´ Birnudalstindr, 4300 4428 64° 14´ 54´´ 28° 34´ 1´´ 16° 0´ Bakkatindr, 3316 3414 64° 20´ 50´´ 28° 50´ 22´´ 15° 47´ Afrèttartindr, 3842 3956 64° 31´ 4´´ 27° 33´ 54´´ 15° 0´ Búlandstindr, 3388 3488 64° 41´ 54´´ 27° 3´ 4´´ 14° 31´ Snæfell,[63] 5808 5964 64° 48´ 1´´ 28° 11´ 43´´ 15° 38´ Kistufell, 3499 3602 64° 51´ 18´´ 27° 11´ 16´´ 14° 47´ Lambafell, 3459 3561 64° 58´ 28´´ 26° 39´ 19´´ 14° 5´

In the north-eastern quarter, twenty points range from 1000 to 3000 Danish feet, and only three rise higher, viz.:

Dan. feet=Eng. feet. N. lat. W. long. (C.)=Greenwich.

Illviðrahnúkr, 3476 3579 66° 8´ 14´´ 31° 37´ 4´´ 19° 4´ Hvammsfell, 3785 3897 65° 39´ 18´´ 31° 48´ 21´´ 19° 14´ Mælifellshnúkr, 3476 3579 65° 23´ 30´´ 31° 59´ 10´´ 19° 25´

In the south-western quarter, thirteen points range from 1000 to 3000 Danish feet, and again only three rise higher, viz.:

Dan. feet=Eng. feet. N. lat. W. long. (C.)=Greenwich.

Snæfellsjökull, 4577 4713 64° 48´ 4´´ 36° 25´ 8´´ 23° 51´ Hekla,[64] 4961 5108 63° 59´ 2´´ 32° 19´ 2´´ 19° 45´ Eyjafjallajökull,[65] 5432 5593 63° 37´ 2´´ 32° 16´ 18´´ 19° 42´

From these tables we see that the north-eastern and south-eastern quarters contain not only the greatest number of heights, respectively twenty-five and twenty, exceeding 1000 Danish feet, but also the apex of Iceland. The north-western, though generally a high level, has only three master peaks, and the traveller’s eye soon determines the south-western to be the lowest of all. It may here be remarked that the islanders have names for the mountains, peaks, and even blocks, as well as for the valleys, whereas the Arabs, as a rule, name only their wadys.

Upon the points above named,

“Nix jacet et jactam nec sol pluviæque resolvunt Indurat Boreas, perpetuamque facit.”

The snow-line above the tableland (1500 to 2000 feet) varies according to position and formation of ground from 2000 to 3500[66] feet over sea-level. The mean has been laid down at 2830 feet. Iceland, as far as it is known, contains few true glaciers. The best known of the Skriðjöklar, _glaciers mouvants_, the “vacillating jökuls” of Henderson (i., pp. 237, 265), protruded by the thrust from behind and above, are the southern offshoots of the great Klofajökull. Two have been often described--the Skeiðarárjökull and the Breiðamerkrjökull. Concerning these ice masses, which are confined, as far as is known, to the southern and the south-eastern shores, and which slope gently to the sea, it is generally believed in Iceland that the congealed tracts are diminishing. Professor Tyndall observed the same in the Mer de Glace, and Mr Freshfield on the Caucasus, where the excess of consumption over supply threatens to make the “gletchers” mere spectres of their former selves.

We now approach the modern formations, the volcanic tracts which overlie the plateaux of Palagonite, trap, and trachyte, and the valleys of elevation and erosion which cleave their masses. As usual throughout the world, the fire-vents are confined to the neighbourhood of the sea and lakes: the centre of Iceland is the Sprengisandur (bursting sand),[67] a black “Ruba’ el Kháli.” In many places the trap terraces have become a wall, over which great gushes of modern lavas have poured down towards the ocean--stone models of the waters which stream down the valleys, and which spring in cataracts from step to step.

Again, it is asserted, with premature generalisation, that the volcanic vents trend, as a rule, from north-east to south-west--a corollary of the “trachytic-band” theorem. The principal systems, which are the following, do not bear out this disposition, and it is probably true only of the south-western part of the island, which was first examined by travellers. Beginning from the north-west, we have the following list of eight great systems.

1. The Dranga[68]-Glámu system in the great palmated projection, the former lying north-east of the latter.

2. The Leirhnúkr, Krafla, and Heiðarfjall, near the Mý-vatn Lake. They anastomose, by the Ódáða-hraun, with the Vatnajökull and the Skaptár--the direction being north to south.

3. The Snæfellsjökull (Western Jökull) runs distinctly from west to east, ending at the sea-shore.

4. The Hofsjökull, including the Arnarfells branch to the east, and the Blágnýpujökull to the south-west. Occupying the centre of the island, it approaches the Túngnafellsjökull, an outlier of the Vatnajökull system to the south-east; and westward, it almost touches the north-eastern extremity of the long Reykjanes line.

5. The Hekla system, which the old theory of fissures connected with Etna. It lies on a parallel, a Palagonite ridge about 2000 feet high, extending from west to east through the Torfajökull, to the banks of the Skaptá.

6. The Vatnajökull, whose apex is Öræfa, the whole measuring some 330 miles in circumference, and occupying an area of 3000 to 4000 square geographical miles: stretches upon a parallel, and is connected by a meridian of lava-run with No. 2.

7. The Katla, or Kötlu-gjá system, again, is not linear, but disposed in a group at the southern extremity of Iceland. The principal items are the Mýrdals, Eyjafjalla, Merkr, Goðalands, and Tindfjalla Jökulls. This great mass is generally known as the Eastern Jökull, opposed to the Western or Snæfells.

8. The Reykjanes system is apparently the only diagonal which extends from the Fire Islands north-eastwards to Skjaldbreið, and to the snow mountains, whose northernmost point is Eyriksjökull. Its items are the Láng, the Ball, the Bláfells, the Geitlands, and the Ok.

Mr Keith Johnston, sen., and other authorities, give the following list of volcanic eruptions which have occurred during the present century.[69]

1. Aust-Jökull (an indefinite term for the great Eyjafjalla system), in December 1820 to June 1822, and January to June 1823.

2. Mýrdals Jökull (or rather Kötlu-gjá) in 1823, from 26th June, covered about a hundred square miles with sand and ashes.

3. Skeiðar Jökull began to erupt February 13, 1827, and did considerable damage. No record of this outbreak is to be found.

4. The submarine eruption off Cape Reykjanes took place in 1831.[70]

5. Hekla, in September 2, 1845 (-46), broke out the twenty-sixth time, according to popular writers, throwing up ashes, which fell in the Orkneys, and which gave the first intelligence of the event.

6. Kötlu-gjá again was slightly active, vomiting ashes and water in May 1860, its thirteenth eruption.

7. It has been generally assumed that on March 23, 1861, the Öræfajökull broke its long rest, and the smoke is said to have tarnished silver at the distance of fifty miles. But Mr Jón A. Hjaltalín, who was in Iceland during that year, denies having heard of any convulsion, nor was it mentioned by the island papers. He adds, “What is spoken of in Metcalfe’s book was a ‘Jökul-hlaup.’”

An ash-eruption from Trölladýngjur is recorded in 1862, but accounts of it greatly vary. Mr Keith Johnston chronicles nine eruptions extending through nearly five centuries and a half--namely, the submarine volcano in the middle of Breiði Fjörð (A.D. 1345), Trölladýngjur (1510), Herðubreið (1716-17), “Krabla” (1724-25), Leirhnúkr (1730), Síðu Jökull (1753), Öræfajökull (1755), Hnappafellsjökull (1772), and Skaptárjökull (1783). And he further informs us that two great groups are active--Leirhnúkr, “Krabla,” Trölladýngjur, and Herðubreið,[71]--all nearly on a parallel of latitude to the north-east; and Hekla, Aust Jökull, Mýrdals, and Öræfa, placed in a right-angled triangle to the south.

Concerning the unvisited volcano in the snows of the Vatnajökull, all procurable details will be found in the Journal. The author was surprised to find that not one of the known centres was in a state of activity, although every preconceived idea suggested that the summer of 1872 would be one of unusual perturbation.[72] Two days before the outbreak of Vesuvius (January 1, 1872), shocks began in the north-east of Iceland. On the afternoons of 16th and 17th April, Húsavík, a small comptoir to the east of Skjálfandi Fljót, suffered severely, as will appear in a future page. This immediately followed the fearful cyclone at Zanzibar (April 15), a phenomenon unknown in former times, which destroyed part of the town, and which sank most of the foreign and native craft,[73] doing damage estimated at £2,000,000. The earthquake at Húsavík also took place only thirteen days after the earthquake at Antioch (morning of April 3), which shook down two-thirds of the houses, and killed nearly one-third of the people. Moreover, shocks were reported at Accra on the Gold Coast, a town which had been almost destroyed some ten years before.[74] Followed (May 1) by the cyclone at Madras, which breached the pier, severely injured the city and suburbs, and wrecked eleven merchantmen, drowning many of the crew. Lastly came the report that the unseen crater in the untrodden snows of the Vatnajökull, whose smoke was first seen in August 1867, had again begun to “vomit flames.”

Meanwhile the eruptions of Vesuvius continued till April 26, when a new crater built a hill in the Atrio del Cavallo, where only a fissure before appeared. Professor Palmieri, who stuck staunchly and gallantly to his observatory on the banks of the new Styx, reported that the mountain was sweating fire at every pore, and that after the showers of ashes and red-hot stones, and the discharges of lava and “boiling smoke,” storms not less dangerous had begun to rage. These meteors, as a rule, occasion great floods, which, sweeping down the ashes and _rapilli_ that cover the slopes, complete the ruins of the lands spared by the lava. During this eruption, a report was spread that the crater of Vesuvius had become an electric pile; that strong currents, generated by the violent ejections of the crater, showed themselves in lightnings, flashing with a dry and hissing sound from the great trunk of smoke and ashes; and, finally, that an earthquake might at any moment shake Naples to its foundation. This abnormal electricity may explain the meteorological peculiarities of the spring of 1872, even in England, where May behaved itself with the leonine violence of March. The great Pacific earthquake (August 1867) and the tremendous and unusual storm which simultaneously visited the eastern coast of South America, to quote no other instances, showed that, whilst similar effects usually are of limited extent upon solid ground, they stretch to great distances at sea, and they may influence the atmosphere in the furthest regions of the world. Though we may accept only as provisional the geological theory which places volcanoes upon fissures or solutions of continuity in the earth’s surface,[75] we must remember that on October 17, 1755, a fortnight before the earthquake which shook down Lisbon, the Kötlu-gjá fissure began the terrible eruptions that lasted for a year: at the same time the waters of Loch Ness were agitated; the British Isles were rocked by repeated oscillations, and shocks extended to Asia and to America. Again, in 1783, the Upper Calabrian earthquake (February 5 and 7, and March 28) was closely followed by the fearful phenomena of the Skaptárjökull.[76] Thus Nature appeared to have made in the summer of 1872 every possible arrangement for a grand pyrotechnic display; yet the author can positively assert that during the whole of his stay in Iceland not one of the twenty-seven to thirty great vents showed a symptom of activity. Indeed, only one was ever reported to be in existence, and that one has never been visited.

Professor Bunsen has shown that active volcanoes whose temperature is high, discharge sulphurous acid, whilst the dormant give forth sulphuretted hydrogen; hence the irregular and simultaneous appearance of these two gases which play a most important part in Iceland. “Let a piece of one of the igneous rocks be heated to redness, and permit the vapour of sulphur to pass over it. The oxide of iron is decomposed; a portion of sulphur unites with the iron which remains as sulphuret; the liberated oxygen unites with the remaining sulphur, and forms sulphurous acid. Let the temperature of the heated mass sink just below a red heat, and then let the vapour of water be passed over it: a decomposition of the sulphuret before formed is the consequence; the iron is reoxydised, and the liberated sulphur unites with the free hydrogen to form sulphuretted hydrogen. Thus the presence of two of the most important agents in volcanic phenomena is accounted for. These are experimental facts capable of being repeated in the laboratory, and the chronological order of the gases thus produced is exactly the same as that observed in nature.”

The most remarkable features of the island, after the volcanic, are the Fjörðs,[77] or firths proper, conducting streams and admitting the sea; opposed to Víks and Vágrs, bights and bays, mere indentations of the coast. Though of igneous origin, they are compared with the granitic features of Norway, where a volcano is unknown, and yet where the shape becomes that of an _arête_, a fish’s dorsal bone with regular ribs on both sides: this flat snow-capped ridge is “the keel” of the maritime population. The popular theory (Students’ Manual of Geology, Jukes and Geikie, Blacks, Edin. 1872) is that the Fjörðs are glens once submerged, raised above water, and hollowed out by glaciers and by the various influences which come under the name of “weather.” Glacial action is, we must own, distinctly traced in most parts of the island. But in many places, Berufjörð for instance, there is no room at the head of the dwarf amphitheatre for a glacier of any magnitude. As in the Færoe archipelago, these ravines are the rents and fissures which divided and fractured the first upheaval; and in Iceland they were bound together by the action of earthquakes and eruptions, ice and snow, wind and rain. The greater gorges are found chiefly on three sides of the island. The south-western shore, like that of Ireland, is digitated by gales, currents, and Greenland ice, and it abounds in “Út-ver,”[78] the narrow-necked peninsulas of Norway. The Síða, or sea-“side” to the south-east, is a long, narrow strip of habitable land between the mountains and the waters: here the Fjörðs were obliterated by the combined action of the Jökulls. Under the name “Fjörðs” are also included immense bays, as the Faxa Fjörð, sixty-five miles across; the Breiði Fjörð, forty-five miles wide; and the Húnaflói, into which the Arctic Sea sends its unbroken swell, running forty-six miles deep and twenty-seven in diameter. The western features are, as a rule, broad, with shallow sag: here, according to some,[79] was deposited the Surtarbrand[80] or lignite, and, like the driftwood of Kerguelen Island, it escaped incineration by subsequent eruptions from causes analogous to the operation of charcoal burning. The northern firths are long and deeply indented, and the eastern are sharp and narrow, encased in walls of Palagonite, trap, and basalt.

The archipelagoes and solitary islands outlying Iceland are invariably small; and in places, as will be seen, the “stacks” and “drongs” form a “skerry-guard,” almost a false coast.

Concerning a common feature of the interior, the Gjá (pron. _Geeow_, or like _ow_ in fowl), rent, chasm, or fissure, details will be given in the course of the Journal. Here it may be mentioned that it perfectly resembles the “Ka’ah” of the Lejá and the Haurán, and the Lava Fields in the Far West of North America, which lately sheltered the “Indians,” and gave so much trouble to the Federal troops.

The surface of Iceland, where free from snow, and over which men travel, may be reduced to four general formations.

1. Loose, volcanic ashey sand, grey above and black below; often mixed with pulverised Palagonite; barred with white lines of salt and potash, and either erupted subaërially or formed under water, as the rolled stones and pebbles show. This feature is found best developed in the central and the north-eastern parts of the island; the Sprengisandur and the Stórisandur (Sahará or Great Sands) being the great examples. The hills and terraces are utterly barren, because they will not hold water: the lower levels, fed by percolation, bear the normal growth, and especially the wild oat.

2. Stone; chiefly Palagonite, trap, basalts, trachyte, lavas, and obsidians, the Μαῦρα λιθάρια of the modern Greeks. It is, however, far safer travelling than the polished limestone of the Libanus, and an hour’s ride over calcareous Kasrawán is more troublesome than a day in Iceland. Its greatest inconvenience is perhaps the sun: during a clear day it becomes, in Icelandic phrase, “hot enough to make a raven gape.” A fair specimen of the stone-country may be found between Reykjavik and Krísuvík.

3. Clay and humus, the former generally disposed in horizontal strata, the latter deposited by decayed vegetation upon the surface. These formations, the Geest-lands of Denmark, mostly extend round the hill feet, dividing them from the deeper levels of bog. They form essentially “rotten” ground; drilled with holes by frost, rain, and sun, and cut by gullies of all sizes, a plexus of wrinkles or gashes and earth-cracks, radiating from the highlands to the lowlands. When the path becomes a hollow way, sunk too deep for riding, rut-tracks straggle, as in the Brazil, over wide spaces and, after the vernal thaws, the traveller will find the “corduroys” of America and the “glue-pots” of Australia; whilst in places scattered stones are so many traps for careless horses. Yet these clays and humus are the best paths and, after the sands, give the fairest chance of a gallop.

4. Bog in Iceland clothes the hill-sides, as well as the bottoms and the “flats,” that is, any low alluvial land: it is easily discovered from afar by the dull-red tint of iron-rust and the snow-white spangles of cotton-grass. There are two forms of profile: one lumpy, tussocky, and what one traveller calls “hassocky,” like the graves of a deserted churchyard; the other a plane, the swamp pure and simple; often flooded after rains, and in the dries provided with two or three veins, into which animals plunge, struggle, and fall. These channels change so frequently that none but local guides are of use, and often the best path leads to the place which has lately become the worst. Instinct and experience do something, but not much, for man and beast: both naturally prefer running water to stagnant, and when the foremost is bogged, the followers seek a better place either higher up or lower down. On frequented lines the impassable places are provided with “Brúr,” dykes or causeways of peat or stone, traversed by rude arches and flanked by shallow ditch-drains.

The Heiði, or high divide separating two river-valleys, is a “dry-land wave” (κῦμα χερσαῖον), varying from 1500 to 2000 and even 3000 feet in altitude. These ridges, especially during the mist and fog, snow and hail, wind and rain, are the horror of native travellers, and few venture upon the passage in foul weather. The profile is a harsh caricature of our Scotch and Irish moors and mosses, bogs and swamps, combining all the troubles of sand, stone, clay, and slush; whilst the marshes and drains are most troublesome to cross. “Carlines,” or old women (Vörður and Kerlingar),[81] are built in places where transit must be made at all seasons; but they are often useless, as the streams shift their bottoms, and permanent paths cannot be traced on what is neither water nor good dry land. At the beginning and end of the travelling season, snow-_fonds_ and veins, based upon compressed ice, streak the slopes and dot the hollows, whilst natural arches and bridges, under which savage torrents gnash and foam, must be crossed on horseback. Concerning the behaviour of the snow, details will be found in the course of the Journal.

Roads are made in Iceland, like those of Syria, by taking off, not as in Europe by putting on, stones. In the more civilised parts of the island they are represented by horse-paths, which are occasionally repaired, and by sheep-paths, which are left to themselves: they humbly suggest the “buffalo” track of the prairie, and the elephant tunnel of the African forest. Not a few show worse engineering and tracery than those of olden Austria; hence we find upon the map such pleasant titles as Höfða-brekka[82] (head-brink or slope), Hálsavegr (neck-or-nothing way), Íllaklif (evil cliff), and Ófæra or Úfæra, Úfærð (the untravellable)--the latter often applied to short cuts over the sea-sands where the wayfarer is exposed to a cannonade from the heights.

§ 2. HYDROGRAPHY.

The hydrography of Iceland has several peculiarities. A glance at the map shows that the Sprengisandur is the keystone of the flattened arch, which, averaging 2000 feet in altitude, forms the centre of the island. From this point the main lines diverge quaquaversally, except to the south-east, where the huge white oval, denoting the Vatnajökull, bars the way, and forms a drainage-system of its own. Hence none of the streams are navigable above the mouth, and their magnitude, as well as the dimensions of their basins, are out of all normal proportion to the area of the island. The four head rivers--Hvitá,[83] Thjorsá, Jökulsá (western), and Skjálfjandifljót (shivering or waving flood)--range from 100 to 160 miles in length. The Thjorsá is 150 miles long, and falls 2000 feet in twenty leagues, carrying more water than the Hudson of New York. “White River” is a common local name, the effect of glacier detrition giving the milky aspect familiar to every traveller in Switzerland, and hence, probably, the muddy White Nile, as opposed to the clear Blue River. A more unusual feature is the Fúli-lækr (foul or stinking stream); the iron pyrites, where the stones are ground to powder, part with their sulphur, and the latter, uniting with the hydrogen, accounts for the unsavoury name. The Jökulhlaup, or “Snow-mountain leap,” is the sudden débâcle and exundation which spring from the congealed masses, often with the irresistible might and the swift destruction of the true avalanche.

The streams in the south-eastern corner are the shortest and the most perilous, rising full grown from the glaciers, and sweeping down fragments and miniature floes of ice. Henderson is the first English traveller who forded and described the Skeiðará and the network called the Gnúpsvötn. We may here acquit him of excessive exaggeration: the natives of the eastern coast, when travelling to Reykjavik, prefer the immense round by the north to the short cut along the southern shore; and when asked the reason, they invariably allege the dangers of the snow-drains. In the course of the Journal we shall cross two of the four head streams, and observe a water-power amply sufficient for the wants of a first-rate European people. The principal cataracts are the Oxará, the Seljaland Foss, the Goða Foss, and the Dretti Foss, first visited by Baring-Gould. All have been described by travellers, and the highest is the Hengi Foss which we shall pass on the road.

Of the lakes (Vötn), we shall inspect the two largest, the Thingvalla-vatn[84] and the Mý-vatn; and we shall sight a multitude of tarns and ponds, single and grouped. One peculiarity is noticed in many of the minor waters. In Iceland it is emphatically untrue that lakes without drains are salt or briny--a rule apparently applicable only to the temperate and tropical zones. Whether the phenomenon in the north arises from subterranean drainage through the fissures of the bed, or if it be due to absence of saline matter in the area of drainage, which is often modern lava too hard to be sensibly degraded, we have no means of determining: perhaps there is a union of both causes.

A remarkable feature is the abundance of warm water laid on by the hand of Nature; the map shows upwards of two hundred; and here perhaps the hottest springs of the Old World are found. Suffice it to say at present that they are divided into two main groups. The acidulous and acid-silica, which redden litmus-paper, depositing gypsum and sulphur, do not erupt: these are the “Öl-keldur” (ale springs) mentioned in the “Royal Mirror” of the twelfth century, and they are still locally and popularly distributed into three species. Some, like “martial” waters, inebriate from the abundance of carbonic acid gas; others when allowed to stand, part with their stimulating property; and others again when filled in rise elsewhere. The second class is the alkaline-silica, which restores the colour of litmus paper; it is often explosive, and it contains chiefly sodium and silica. In the valley of the Yellowstone River the springs are either (1.) Calcareous (alkaline), depositing carbonate of lime with sulphates of magnesia and soda, chloride of calcium, and a little silica; or (2.) Silicious (acid), containing 85:100 silica, chloride of magnesium, and only a trace of lime.

The Geysir (gusher)[85] is a spouting spring; the Reykirs (reekers) give forth steam; the Laug is a warm fountain which may serve as a bath; the Náma[86] (hole of hot water) is sulphurous and gaseous; the Hverr (cauldron), like its smaller congener the Ketill (kettle), is a tranquil, hot, and even boiling well or pool, it is also applied to mud springs; and the Makkaluber (the Italian “Salsa,” or “Hofetta,” and the American “Mud-puff”) is a miniature volcano of hissing, boiling bolus. Further details concerning the names and natures of these features will be given in the Journal.

§ 3. CLIMATE.

The “cold of Iceland” is as proverbial as the “deserts of central Africa,” and both sayings are equally based upon unfacts. “Iceland, where the cold and winter are perpetual, and the cold scarce to be endured,” is what we read. But those who travel in the island find--(1.) that even in winter the temperature is rarely severe; (2.) that there are two distinct climates, on the north coast and in the southern country; and (3.) that the air, however unpleasant, is exceptionally wholesome.

1. The isotherms by no means follow the circles of latitude. The cold lines swerve away from, instead of passing through, Iceland, and show none of that severity which characterises Greenland and the northern parts of British America. As has long ago been observed,[87] the isotherm of F. 32°, the freezing point of water, which is that of Akureyri, varies 14° between southern Asiatic Russia (N. lat. 56°) and northern Norway (N. lat. 70°).

The mildness of the insular climate, and that of the easterly winds, which are too clear to come from warmer waters, are popularly attributed to the “great Gulf Stream.” This sea-river, we are told, “sweeping up from the south, brings with it a store of heat to bless the islanders, and so materially affects the island that in the south of Iceland the winter is not more severe than in Denmark.” The Gulf Stream is generally supposed to strike the south-western angle, and to flow along the southern shores; while others make it bifurcate off Reykjanes, hence one part subtends the north-western point or Land’s End of Iceland, where it meets the Polar and Arctic current, the other half embraces the southern shore, and both meet in the north Atlantic arm separating Iceland from Norway. Dufferin’s map shows the popular belief: the true Florida current, sweeping past the southern shore of Iceland, forks about Spitzbergen, sending off a branchlet to the west, and ends south of Novaya Zemlja. On the other hand, Dr Carpenter contends that the real “River in the Ocean” dies out in the mid-Atlantic. According to Dr Joseph Chavanne of Vienna (Mittheilungen, No. vii., 1874), the northern arm of the Gulf Stream, which flows between Bear Island and Novaya Zemlja, touches the northern coast of Asia, and eastward of the New Siberia Islands joins the western drift of the Kurosiwo. The other northern branch, which subtends the western coast of Spitzbergen and the Seven Islands, is submerged between the Polar currents, to reappear at the surface farther northward, and thence to lave the shores of the Arctic continent: the latter is thus washed by two warm streams, rendering the existence of perennial ice a sheer impossibility.

We may fairly question the existence of the Gulf Stream along the southern Icelandic shore, and doubt its bifurcation and subsequent reunion. This is not the place to discuss the subject of ocean circulation, a “discovery equal to that of the circulation of the blood,” first made by Professor Lenz of St Petersburg in 1845, based upon the second voyage of Kotzebue in 1823-26, and independently by Dr Carpenter during the cruise of the “Porcupine” (1869). Their aqueous movement corresponding with the aërial; and the mass of thermal equatorial waters travelling towards the poles, whilst the counter current sets in the inverse direction, would account for many phenomena yet unexplained, but it is still _sub judice lis_.[88] We may remark that the comparatively shallow seas between the British Islands and Iceland must accumulate heat, and that this fact perhaps suffices for what has been attributed to the Gulf Stream and to the general circulation. Thomas Bartolin (Acta Medica Havn. ad annum 1673) mechanically explains away the necessity of the former: “Aqua Insulas Ferroenses allabens, quamquam per se frigida sit, salsitudine tamen suâ, ex perpetuo motu, plerumque producit hyemem temperatam.” Hence the waters of Niagara are colder above than below the falls, and the ocean is warmer after a storm.

Practical men, especially mariners, in Iceland vigorously deny the existence of the Gulf Stream.[89] Captain Tvede, an intelligent and observing Dane whom we shall meet in the eastern regions, considers that the theory, like judicial phrenology and a host of pseudo-sciences, became popular because it generalises, formalises, and simplifies facts. He declares that a Gulf Stream, if it existed, would entangle the Greenland icebergs, and carry them to the southern coast of Iceland, which never happens. He asserts that a few miles south of Ingólfshöfði the Sea River is still warm, but that instead of striking the shore it trends directly north-eastwards to western Norway, sweeps round the continental North Cape, and here meets the icebergs from Spitzbergen and Jan Mayen. He has found himself in an ice-dock floating in water which showed 35° F.

Captain Tvede kindly gave me the following series of observations:

1. June 19, 1867: thermometer in water 46° F. outside of Hrollaugseyar, 6 miles east of Ingólfshöfði, 48° F. 3 miles south-east of ditto, and 47° F. 20 miles west of ditto.

2. June 20: thermometer 47° between Portland and the Vestmannaeyjar, 47° F. 12 miles west of the Vestmannaeyjar.

3. June 23: thermometer 46° in the Breiði Fjörð, off Stykkishólm.

4. June 24: 43° outside of the Dýrafjörð, and 43°-43°·50 outside of the Ísafjörð.

5. June 25: 38° off the Húnafljói, and 43° off Cap Nord.

6. July 1: 40° off the Axarfjörð.

7. July 4: 39° off the Langanes (north-eastern point of Iceland).

8. July 6: 40° off Viðivik, and 42° outside of Borgarfjörð.

9. August 4: 46° 16 miles south-east of Langanes.

10. August 6: 42° in the Testilfjörð, western side of Langanes.

11. August 10: 38°·50 off Hornnes, and 39° same day off Gerpir, 4 miles south of Hornnes.

12. August 19: 44° off Dalataur, entrance of Sydisfjörð.

13. August 21: 44° off Héradsflói.

14. August 22: 42° to north, with Kollumúli bearing south-west, 44° at sea.

15. September 1: 41° off Berufjörð.

The subjoined figures are the means of observations taken every fourth hour on board the “Jón Siggurðsson” steamer, in which the author voyaged (June 26 to August 5, 1872) between Hafnarfjörð and Grafarós:

Air. Water. 1. 12° (C.=53°·6 F.) 10° (C.=50° F.) at Reykjavik. 2. 11° (C.=51°·8 F.) 8°·5 (C.=47°·3 F.) at Flatey. 3. 13° (C.=55°·4 F.) 9° (C.=48°·2 F.) at N. lat. 66°30´, W. long. (G.) 24°. 4. 9° (C.=48°·2 F.) 5°·8 (C.=42°·4 F.) at N. lat. 66°10´, W. long. 23°12´. 5. 14°·5 (C.=58°·1 F.) 8°·8 (C.=47°·8 F.) at Borðeyri. 6. 14°·5 (C.=58°·1 F.) 8°·3 (C.=46°·9 F.) at Grafarós. 7. 11° (C.=51°·8 F.) 6°·8 (C.=44°·2 F.) at Cap Nord. 8. 11° (C.=51°·8 F.) 8°·5 (C.=47°·3 F.) at N. lat. 65°8´, W. long. 23°24´.

Both series tend to show the capricious variation of temperature (from 38° to 48° F., and from 48°·2 to 58°·1 F.), where the summer sea is subject to the influx of a little snow-water, and none of the regularity which might fairly be expected from a “gulf-stream.”

2. Every book of travels from Horrebow and Mackenzie to the present day, has given notices of the climate of Iceland.[90] The mean temperature of the Iceland year between 1828 and 1834, has been laid down at 3°·42 Reaumur (= 39°·7 F.). The annual average of Copenhagen is assumed at 46°·8 (F.); the maximum, observed in the shade, being 94°, and the minimum about 19° (F.). That of Montreal stands at 6°·30 Reaumur (= 46°·2 F.). The winters in Iceland are colder than in Montreal in October and November (both included); warmer from December to March, and again cooler from April to December. Eyjafjörð (N. lat. 65° 40´) is more genial than Cumberland House (N. lat. 53° 57´), and much warmer than any place in its own parallel. The almost nightless summers from June to August, which must affect the respiration of plants, gather caloric, and the sun at that season fails to heat only at a very obtuse angle, when the rays are intercepted by a thicker column of air. The equatorial current which prevails in occidental England for eight or nine months during the year, as the south-wester in Iceland, must greatly modify the climate. Old travellers assure us that the sub-surface is frost-bound throughout the year; this takes place only after a succession of hard winters and ungenial summers--even the cellars are rarely frozen in winter if care be taken to close the doors. Mr Vice-Consul Crowe (first Report on Iceland, 1865-66), asserts that “the average temperature of the earth is about 4½° Reaumur all the year round.”

Reykjavik, the capital of Iceland (N. lat. 64° 9´), enjoys a more genial climate than any place whose temperature is recorded between the parallels of 55° and 85° (N. lat.), except only St Petersburg (N. lat. 59° 56´) and Sitka Sound (N. lat. 57° 3´). The mean of the year is but 1° (F.) less than that of St John’s, which lies 16° farther south. The winter corresponds with that of Illukuk, 10° to the south, and the summer is much hotter. Humboldt’s mean temperature, 40° F., is generally adopted, although some reduce it to 39°·4, and even to 39°. He makes February, the coldest month, average 28°·22, and July, the hottest, 56°·3--a difference of over 28°, which others reduce to 27°. He fixes the winter mean at 29°·1; the spring at 36°·9; the summer at 53°·6 (in Berghaus’ Atlas, 50°); and the autumn at 37°·9. Dillon (pp. 167, 168), during the severest season of half-a-century, saw the mercury as low as 10° (F.), in February; and Pliny Miles (p. 55) declares that the thermometer seldom falls below 12° or 18°.

It will be remembered that the annual mean of climates, where civilisation is highest, represents in Europe 52° (F.), and the zone is 15° north and south of N. lat. 40°, an undulating belt of 30° arching towards the equator and the poles. Including its protraction eastward and westward, it contains 95/100ths of the white races, and almost all the greatest development.

Certain valuable “notes on the distribution of animals available as food in the Arctic regions,” compiled by Herr Petermann, and published in the Journal of the R. Geog. Society (vol. xxii.), enable us to compare the thermometer in the south and in the north of the island. “Reykiavig” (N. lat. “64°·08´´) is placed between New Herrnhut and Fort Reliance, whilst Eyjafjörð (N. lat. 66° 30´), stands between Fort Hope and Winter Island.

The figures are as follows:

Annual Difference Spring. Summer. Autumn. Winter. mean. Sum. & Wint. 1. New Herrnhut, 26°·15 39°·28 26°·50 14°·30 26°·83 24°·48 _Reykjavik_, 37°·04 53°·54 37°·94 29°·18 39°·43 24°·36 Fort Reliance, ” 12°·21 ” -16°·97 16° (?) “

2. Fort Hope, -4°·73 39°·59 13°·93 -25°·09 5°·96 64°·68 _Eyjafjörð_, 28°·10 45°·80 34°·46 20°·84 32°·30 24°·96 Winter Island, 6°·35 31°·80 17°·58 -20°·47 8°·82 52°·27

Ranged according to seasons and months, the figures stand:

SPRING.

New Herrnhut in February (coldest), 22°·10 in March 21°·65 in April 24°·80 _Reykjavik_ ” 28°·31 ” 29°·86 ” 36°·46 Fort Reliance ” -18°·84 ” -6°·14 ” 8°·23 Port Hope ” -26°·68 ” -28°·10 ” -23°·95 _Eyjafjörð_ ” 18°·50 ” 20°·66 ” 27°·50 Winter Island ” -23°·99 ” 10°·72 ” 6°·48

SUMMER.

New Herrnhut in May, 32°·0 in June, 40°·10 in July, (hottest), 40°·33 _Reykjavik_ ” 44°·80 ” 51°·58 ” 56°·19 Fort Reliance ” 36°·03 ” ” ” “ Fort Hope ” 17°·88 ” 31°·38 ” 41°·46 _Eyjafjörð_ ” 36°·14 ” 43°·52 ” 46°·94 Winter Island ” 23°·29 ” 23°·17 ” 35°·36

AUTUMN.

New Herrnhut in August 37°·40 in September, 34°·03 in October, 32°·90 _Reykjavik_ ” 52°·86 ” 46°·45 ” 36°·91 Fort Reliance ” ” ” ” ” 20°·70 Fort Hope ” 46°·32 ” 28°·57 ” 12°·56 _Eyjafjörð_ ” 46°·94 ” 43°·16 ” 34°·34 Winter Island ” 36°·86 ” 31°·61 ” 13°·25

WINTER.

New Herrnhut in November, 15°·80 in December 11°·75 in January 9°·05 _Reykjavik_ ” 30°·45 ” 29°·41 ” 29°·82 Fort Reliance ” 13°·44 ” -17°·07 ” -25°·00 Fort Hope ” 0°·68 ” -19°·27 ” -29°·32 _Eyjafjörð_ ” 25°·88 ” 18°·32 ” 25°·70 Winter Island ” 7°·88 ” -14°·24 ” -23°·17

Dr Joseph Chavanne, before alluded to, gives the following table of the wind temperature at Reykjavik, showing the deviations from mean:

WINTER.--Mean Temperature -1·8. N. N.E. E. S.E. S. S.W. W. N.W. Max. Min. Diff. 3·6 -2·2 +1·3 +4·1 +3·7 +1·1 -1·4 -2·9 E. 68, S. +4·4 N. -3·6 8·0 SUMMER.--Mean Temperature +11.0.

0·0 +0·5 +0·1 +0·2 +0·3 -0·7 -1·0 -1·3 E. 30, S. +0·7 W. 35, N. -1·6 2·3

Thus the climate of southern Iceland is insular and not excessive. We have a notorious instance of the same disposition in England. With us Devonia represents the south-western coast of Iceland, and justifies Carrington’s high praise:

“Thou hast a cloud For ever in thy sky; a breeze, a shower For ever on thy meads. Yet where shall man, Pursuing spring around the globe, refresh His eye with scenes more beauteous than adorn Thy fields of matchless verdure?”

The northern climate of Iceland, distant only 3° or 180 direct geographical miles, is distinctly continental; the difference ranging between 14° and 17° (F.). This is easily accounted for by the Arctic current, by the proximity of Polar ice, and by the prevalence of northern and north-western winds, which, in south Iceland as in Palestine, drive away rain. Whatever discrepancy of opinion there may be concerning the Gulf Stream, there can be none about the cold drift which, between Greenland and Iceland, measures some fifty miles in breadth, and many hundred feet in depth. Hence the north-western digitations are more subject to floes and bergs than the Breiði Fjörð, which again is oftener invested than the Faxa Fjörð, the latter being rarely beset more than once during the century. According to Uno Von Troil, the sea-ice, now so rare, came regularly in January with the north-eastern gales, and was never far from the north-east coast. At present the season is about April and even later.

In the north, according to Metcalfe (p. 152), the winter is much keener, and the summer is proportionally milder than in the south; some observers deny the truth of the latter part of the proposition, and make the hot months average about the same figure. The snow often begins with October and lasts till mid-May when the temperature stands at a mean of 35° (F.). For Akureyri Baring-Gould (quoting the Almanak um Ár 1863), gives the year as 32° (F., freezing point = Eyjafjörð), the winter as 20°·7, and the summer 45°·5. He therefore determines that, while the mean of Reykjavik is very nearly that of Moscow, Akureyri almost corresponds with Julianshaab in Greenland.

At Stykkishólm on the mid-west coast (N. lat. 65° 4´ 44´´, and W. long. (G.) 22° 43´ 17´´), observations have been taken by Hr A. O. Thorlacius for nearly thirty years. The gross results are given in the following table, taken from the Journal of the Scottish Meteorological Society,