ATLAS MOUNTAINS

atlas mountains, the general name for the mountain chains running more or less parallel to the coast of north-west africa. they extend from cape nun on the west to the gulf of gabes on the east, a distance of some 1500 m., traversing morocco, algeria and tunisia. to their south lies the saharan desert. the atlas consist of many distinct ranges, but they can be roughly divided into two main chains: (1) the maritime atlas, i.e. the ranges overlooking the mediterranean from ceuta to cape bon; (2) the inner and more elevated ranges, which, starting from the atlantic at cape ghir in sus, run south of the coast ranges and are separated from them by high plateaus. this general disposition is seen most distinctly in eastern morocco and algeria. the western inner ranges are the most important of the whole system, and in the present article are described first as _the moroccan ranges_. the maritime atlas and the inner ranges in algeria and tunisia are then treated under the heading _eastern ranges_. _the moroccan ranges._--this section of the atlas, known to the inhabitants of morocco by its berber name, idraren draren or the "mountains of mountains," consists of five distinct ranges, varying in length and height, but disposed more or less parallel to one another in a general direction from south-west to north-east, with a slight curvature towards the sahara. 1. the main range, that known as the great atlas, occupies a central position in the system, and is by far the longest and loftiest chain. it has an average height of over 11,000 ft., whereas the loftiest peaks in algeria do not exceed 8000 ft., and the highest in tunisia are under 6000 ft. towards the dahra district at the north-east end the fall is gradual and continuous, but at the opposite extremity facing the atlantic between agadir and mogador it is precipitous. although only one or two peaks reach the line of perpetual snow, several of the loftiest summits are snowclad during the greater part of the year. the northern sides and tops of the lower heights are often covered with dense forests of oak, cork, pine, cedar and other trees, with walnuts up to the limit of irrigation. their slopes enclose well-watered valleys of great fertility, in which the berber tribes cultivate tiny irrigated fields, their houses clinging to the hill-sides. the southern flanks, being exposed to the hot dry winds of the sahara, are generally destitute of vegetation. at several points the crest of the range has been deeply eroded by old glaciers and running waters, and thus have been formed a number of devious passes. the central section, culminating in tizi n 'tagharat or tinzar, a peak estimated at 15,000 ft. high, maintains a mean altitude of 11,600 ft., and from this great mass of schists and sandstones a number of secondary ridges radiate in all directions, forming divides between the rivers dra'a, sus, um-er-rabia, sebu, mulwiya and ghir, which flow respectively to the south-west, the west, north-west, north, north-east and south-east. all are swift and unnavigable, save perhaps for a few miles from their mouths. with the exception of the dra'a, the streams rising on the side of the range facing the sahara do not reach the sea, but form marshes or lagoons at one season, and at another are lost in the dry soil of the desert. for a distance of 100 m. the central section nowhere presents any passes accessible to caravans, but south-westward two gaps in the range afford communication between the tansift and sus basins, those respectively of gindafi and bibawan. a few summits in the extreme south-west in the neighbourhood of cape ghir still exceed 11,000 ft., and although the steadily rising ground from the coast and the prominence of nearer summits detract from the apparent height, this is on an average greater than that of the european alps. the most imposing view is to be obtained from the plain of marrakesh, only some 1000 ft. above sea-level, immediately north of the highest peaks. besides huge masses of old schists and sandstones, the range contains extensive limestone, marble, diorite, basalt and porphyry formations, while granite prevails on its southern slopes. the presence of enormous glaciers in the ice age is attested by the moraines at the atlantic end, and by other indications farther east. the best-known passes are: (1) the bibawan in the upper wad sus basin (4150 ft.); (2) the gindafi, giving access from marrakesh to tarudant, rugged and difficult, but low; (3) the tagharat, difficult and little used, leading to the dra'a valley (11,484 ft.); (4) the glawi (7600 ft.); (5) tizi n 'tilghemt (7250 ft.), leading to tafilet (tafilalt) and the wad ghir. 2. the lower portion of the moroccan atlas (sometimes called the middle atlas), extending north-east and east from an undefined point to the north of the great atlas to near the frontier of algeria, is crossed by the pass from fez to tafilalt. both slopes are wooded, and its forests are the only parts of morocco where the lion still survives. from the north this range, which is only partly explored, presents a somewhat regular series of snowy crests. 3. the anti-atlas or jebel saghru, also known as the lesser atlas, running parallel to and south of the central range, is one of the least elevated chains in the system, having a mean altitude of not more than 5000 ft., although some peaks and even passes exceed 6000 ft. at one point it is pierced by a gap scarcely five paces wide with walls of variegated marbles polished by the transport of goods. as to the relation of the anti-atlas to the atlas proper at its western end nothing certain is known. the two more or less parallel ranges which complete the western system are less important:--(4) the jebel bani, south of the anti-atlas, a low, narrow rocky ridge with a height of 3000 ft. in its central parts; and (5) the mountains of ghaiata, north of the middle atlas, not a continuous range, but a series of broken mountain masses from 3000 to 3500 ft. high, to the south of fez, taza and tlemcen. _the eastern ranges._--the eastern division of the atlas, which forms the backbone of algeria and tunisia, is adequately known with the exception of the small portion in morocco forming the province of er-rif. the lesser range, nearer the sea, known to the french as the maritime atlas, calls for little detailed notice. from ceuta, above which towers jebel musa--about 2800 ft.--to melilla, a distance of some 150 m., the rif mountains face the mediterranean, and here, as along the whole coast eastward to cape bon, many rugged rocks rise boldly above the general level. in algeria the maritime atlas has five chief ranges, several mountains rising over 5000 ft. the jurjura range, extending through kabylia from algiers to bougie, contains the peaks of lalla kedija (7542 ft.), the culminating point of the maritime chains, and babor (6447 ft.). (see further algeria.) the mejerda range, which extends into tunisia, has no heights exceeding 3700 ft. it was in these coast mountains of algeria that the romans quarried the celebrated numidian marbles. the southern or main range of the eastern division is known by the french as the saharan atlas. on its western extremity it is linked by secondary ranges to the mountain system of morocco. the saharan atlas is essentially one chain, though known under different names: jebel k'sur and jebel amur on the west, and jebel aures in the east. the central part, the zab mountains, is of lower elevation, the saharan atlas reaching its culminating point, jebel shellia (7611 ft. above the sea), in the aures. this range sends a branch northward which joins the mejerda range of the maritime atlas, and another branch runs south by gafsa to the gulf of gabes. here mount sidi ali bu musin reaches a height of 5700 ft., the highest point in tunisia. in the saharan atlas the passes leading to or from the desert are numerous, and in most instances easy. both in the east (at batna) and the west (at ain sefra) the mountains are traversed by railways, which, starting from mediterranean seaports, take the traveller into the sahara. _history and exploration._--the name atlas given to these mountains by europeans--but never used by the native races--is derived from that of the mythical greek god represented as carrying the globe on his shoulders, and applied to the high and distant mountains of the west, where atlas was supposed to dwell. from time immemorial the atlas have been the home of berber races, and those living in the least accessible regions have retained a measure of independence throughout their recorded history. thus some of the mountain districts of kabylia had never been visited by europeans until the french military expedition of 1857. but in general the maritime range was well known to the romans. the jebel amur was traversed by the column which seized el aghuat in 1852, and from that time dates the survey of the mountains. the ancient caravan route from mauretania to the western sudan crossed the lower moroccan atlas by the pass of tilghemt and passed through the oasis of tafilalt, formerly known as sajilmasa ["sigilmassa"], on the east side of the anti-atlas. the moroccan system was visited, and in some instances crossed, by various european travellers carried into slavery by the salli rovers, and was traversed by rene caille in 1828 on his journey home from timbuktu, but the first detailed exploration was made by gerhard rohlfs in 1861-1862. previous to that almost the only special report was the misleading one of lieut. washington, attached to the british embassy of 1837, who from insufficient data estimated the height of mount tagharat, to which he gave the indefinite name of miltsin (i.e. _mul et-tizin_, "lord of the peaks"), as 11,400 ft. instead of about 15,000 ft. in 1871 the first scientific expedition, consisting of dr (afterwards sir) j.d. hooker, mr john ball and mr g. maw, explored the central part of the great atlas with the special object of investigating its flora and determining its relation to that of the mountains of europe. they ascended by the ait mizan valley to the tagharat pass (11,484 ft.), and by the amsmiz valley to the summit of jebel tezah (11,972 ft.). in the tagharat pass mr maw was the only one of the party who reached the watershed; but from jebel tezah a good view was obtained southward across the great valley of the sus to the anti-atlas, which appeared to be from 9000 to 10,000 ft. high. dr oskar lenz in 1879-1880 surveyed a part of the great atlas north of tarudant, determined a pass south of iligh in the anti-atlas, and penetrated thence across the sahara to timbuktu. he was followed in 1883-1884 by vicomte ch. de foucauld, whose extensive itineraries include many districts that had never before been visited by any europeans. such were parts of the first and middle ranges, crossed once; three routes over the great atlas, which was, moreover, followed along both flanks for nearly its whole length; and six journeys across the anti-atlas, with a general survey of the foot of this range and several passages over the jebel bani. then came joseph thomson, who explored some of the central parts, and made the highest ascent yet achieved, that of mount likimt, 13,150 ft., but broke little new ground, and failed to cross the main range (1888); and walter b. harris, who explored some of the southern slopes and crossed the atlas at two points during his expedition to tafilalt in 1894. in 1901 and again in 1905 the marquis de segonzac, a frenchman, made extensive journeys in the moroccan ranges. he crossed the great atlas in its central section, explored its southern border, and, in part, the middle and anti-atlas ranges. a member of his expeditions, de flotte rocquevaire, made a triangulation of part of the western portion of the main atlas, his labours affording a basis for the co-ordination of the work of previous explorers. (see also morocco, algeria, tunisia and sahara.) authorities.--vicomte ch. de foucauld, _reconnaissance au maroc 1883-1884_ (paris, 1888, almost the sole authority for the geography of the atlas; his book gives the result of careful surveys, and is illustrated with a good collection of maps and sketches); hooker, ball and maw, _marocco and the great atlas_ (london, 1879, a most valuable contribution, always scientific and trustworthy, especially as to botany and geology); joseph thomson, _travels in the atlas and southern morocco_ (london, 1889, valuable geographical and geological data); louis gentil, _mission de segonzac, &c._ (paris, 1906; the author was geologist to the 1905 expedition); gerhard rohlfs, _adventures in morocco_ (london, 1874); walter b. harris, _tafilet, a journey of exploration in the atlas mountains, &c._ (london, 1895), full of valuable information; budgett meakin, _the land of the moors_ (london, 1901), first and last chapters; dr oskar lenz _timbuktu: reise durch marokko_, vol. i. (leipzig, 1884). atmolysis (gr. [greek: atmos], vapour: [greek: lyein], to loosen), a term invented by thomas graham to denote the separation of a mixture of gases by taking advantage of their different rates of diffusion through a porous septum or diaphragm (see diffusion). atmosphere (gr. [greek: atmos], vapour; [greek: sphaira], a sphere), the aeriform envelope encircling the earth; also the envelope of a particular gas or gases about any solid or liquid. meteorological phenomena seated more directly in the atmosphere obtained early recognition; thus hesiod, in his _works and days_, speculated on the origin of winds, ascribing them to the heating effects of the sun on the air. ctesibius of alexandria, hero and others, founded the science of pneumatics on observations on the physical properties of air. anaximenes made air the primordial substance, and it was one of the aristotelian elements. a direct proof of its material nature was given by galileo, who weighed a copper ball containing compressed air. before the development of pneumatic chemistry, air was regarded as a distinct chemical unit or element. the study of calcination and combustion during the 17th and 18th centuries culminated in the discovery that air consists chiefly of a mixture of two gases, oxygen and nitrogen. cavendish, priestley, lavoisier and others contributed to this result. cavendish made many analyses: from more than 500 determinations of air in winter and summer, in wet and clear weather, and in town and country, he discerned the mean composition of the atmosphere to be, oxygen 20.833% and nitrogen 79.167% the same experimenter noticed the presence of an inert gas, in very minute amount; this gas, afterwards investigated by rayleigh and ramsay, is now named argon (q.v.). the constancy of composition shown by repeated analyses of atmospheric air led to the view that it was a chemical compound of nitrogen and oxygen; but there was no experimental confirmation of this idea, and all observations tended to the view that it is simply a mechanical mixture. thus, the gases are not present in simple multiples of their combining weights; atmospheric air results when oxygen and nitrogen are mixed in the prescribed ratio, the mixing being unattended by any manifestation of energy, such as is invariably associated with a chemical action; the gases may be mechanically separated by atmolysis, i.e. by taking advantage of the different rates of diffusion of the two gases; the solubility of air in water corresponds with the "law of partial pressures," each gas being absorbed in amount proportional to its pressure and coefficient of absorption, and oxygen being much more soluble than nitrogen (in the ratio of .04114 to .02035 at 0 deg.); air expelled from water by boiling is always richer in oxygen. various agencies are at work tending to modify the composition of the atmosphere, but these so neutralize each other as to leave it practically unaltered. minute variations, however, do occur. bunsen analysed fifteen examples of air collected at the same place at different times, and found the extreme range in the percentage of oxygen to be from 20.97 to 20.84. regnault, from analyses of the air of paris, obtained a variation of 20.999 to 20.913; country air varied from 20.903 to 21.000; while air taken from over the sea showed an extreme variation of 20.940 to 20.850. angus smith determined london air to vary in oxygen content from 20.857 to 20.95, the air in parks and open spaces showing the higher percentage; glasgow air showed similar results, varying from 20.887 in the streets to 20.929 in open spaces. in addition to nitrogen and oxygen, there are a number of other gases and vapours generally present in the atmosphere. of these, argon and its allies were the last to be definitely isolated. carbon dioxide is invariably present, as was inferred by dr david macbride (1726-1778) of dublin in 1764, but in a proportion which is not absolutely constant; it tends to increase at night, and during dry winds and fogs, and it is greater in towns than in the country and on land than on the sea. water vapour is always present; the amount is determined by instruments termed hygrometers (q.v.). ozone (q.v.) occurs, in an amount supposed to be associated with the development of atmospheric electricity (lightning, &c.); this amount varies with the seasons, being a maximum in spring, and decreasing through summer and autumn to a minimum in winter. hydrogen dioxide occurs in a manner closely resembling ozone. nitric acid and lower nitrogen oxides are present, being formed by electrical discharges, and by the oxidation of atmospheric ammonia by ozone. the amount of nitric acid varies from place to place; rain-water, collected in the country, has been found to contain an average of 0.5 parts in a million, but town rain-water contains more, the greater amounts being present in the more densely populated districts. ammonia is also present, but in very varying amounts, ranging from 135 to 0.1 parts (calculated as carbonate) in a million parts of air. ammonia is carried back to the soil by means of rain, and there plays an important part in providing nitrogenous matter which is afterwards assimilated by vegetable life. the average volume composition of the gases of the atmosphere may be represented (in parts per 10,000) as follows:-- oxygen 2065.94 ozone 0.015 nitrogen 7711.60 aqueous vapour 140.00 argon (about) 79.00 nitric acid 0.08 carbon dioxide 3.36 ammonia 0.005 in addition to these gases, there are always present in the atmosphere many micro-organisms or bacteria (see bacteriology); another invariable constituent is dust (q.v.), which plays an important part in meteorological phenomena. reference should be made to the articles barometer, climate and meteorology for the measurement and variation of the pressure of the atmosphere, and the discussion of other properties. atmospheric electricity. 1. it was not until the middle of the 18th century that experiments due to benjamin franklin showed that the electric phenomena of the atmosphere are not fundamentally different from those produced in the laboratory. for the next century the rate of progress was slow, though the ideas of volta in italy and the instrumental devices of sir francis ronalds in england merit recognition. the invention of the portable electrometer and the water-dropping electrograph by lord kelvin in the middle of the 19th century, and the greater definiteness thus introduced into observational results, were notable events. towards the end of the 19th century came the discovery made by w. linss (6)[1] and by j. elster and h. geitel (7) that even the most perfectly insulated conductors lose their charge, and that this loss depends on atmospheric conditions. hard on this came the recognition of the fact that freely charged positive and negative ions are always present in the atmosphere, and that a radioactive emanation can be collected. whilst no small amount of observational work has been done in these new branches of atmospheric electricity, the science has still not developed to a considerable extent beyond preliminary stages. observations have usually been limited to a portion of the year, or to a few hours of the day, whilst the results from different stations differ much in details. it is thus difficult to form a judgment as to what has most claim to acceptance as the general law, and what may be regarded as local or exceptional. 2. _potential gradient._--in dry weather the electric potential in the atmosphere is normally positive relative to the earth, and increases with the height. the existence of _earth currents_ (q.v.) shows that the earth, strictly speaking, is not all at one potential, but the natural differences of potential between points on the earth's surface a mile apart are insignificant compared to the normal potential difference between the earth and a point one foot above it. what is aimed at in ordinary observations of atmospheric potential is the measurement of the difference of potential between the earth and a point a given distance above it, or of the difference of potential between two points in the same vertical line a given distance apart. let a conductor, say a metallic sphere, be supported by a metal rod of negligible electric capacity whose other end is earthed. as the whole conductor must be at zero (i.e. the earth's) potential, there must be an induced charge on the sphere, producing at its centre a potential equal but of opposite sign to what would exist at the same spot in free air. this neglects any charge in the air displaced by the sphere, and assumes a statical state of conditions and that the conductor itself exerts no disturbing influence. suppose now that the sphere's earth connexion is broken and that it is carried without loss of charge inside a building at zero potential. if its potential as observed there is -v (volts), then the potential of the air at the spot occupied by the sphere was +v. this method in one shape or another has been often employed. suppose next that a fixed insulated conductor is somehow kept at the potential of the air at a given point, then the measurement of its potential is equivalent to a measurement of that of the air. this is the basis of a variety of methods. in the earliest the conductor was represented by long metal wires, supported by silk or other insulating material, and left to pick up the air's potential. the addition of sharp points was a step in advance; but the method hardly became a quantitative one until the sharp points were replaced by a flame (fuse, gas, lamp), or by a liquid jet breaking into drops. the matter leaving the conductor, whether the products of combustion or the drops of a liquid, supplies the means of securing equality of potential between the conductor and the air at the spot where the matter quits electrical connexion with the conductor. of late years the function of the collector is discharged in some forms of apparatus by a salt of radium. of flame collectors the two best known are lord kelvin's portable electrometer with a fuse, or f. exner's gold leaf electroscope in conjunction with an oil lamp or gas flame. of liquid collectors the representative is lord kelvin's water-dropping electrograph; while benndorf's is the form of radium collector that has been most used. it cannot be said that any one form of collector is superior all round. flame collectors blow out in high winds, whilst water-droppers are apt to get frozen in winter. at first sight the balance of advantages seems to lie with radium. but while gaseous products and even falling water are capable of modifying electrical conditions in their immediate neighbourhood, the "infection" produced by radium is more insidious, and other drawbacks present themselves in practice. it requires a radium salt of high radioactivity to be at all comparable in effectiveness with a good water-dropper. experiments by f. linke (8) indicated that a water-dropper having a number of fine holes, or having a fine jet under a considerable pressure, picks up the potential in about a tenth of the time required by the ordinary radium preparation protected by a glass tube. these fine jet droppers with a mixture of alcohol and water have proved very effective for balloon observations. table i.--_annual variation potential gradient._ +---------------------------------+-----+-----+-----+-----+-----+-----+----+----+-----+-----+-----+-----+ | place and period. | jan.| feb.|march|april| may | june|july|aug.|sept.| oct.| nov.| dec.| +---------------------------------+-----+-----+-----+-----+-----+-----+----+----+-----+-----+-----+-----+ | karasjok (10), 1903-1904 | 143 | 150 | 137 | 94 | 74 | 65 | 70 | 67 | 67 | 87 | 120 | 126 | | sodankyla (31), 1882-1883 | 94 | 133 | 148 | 155 | 186 | 93 | 53 | 77 | 47 | 72 | 71 | 71 | | potsdam (9), 1904 | 167 | 95 | 118 | 88 | 93 | 72 | 73 | 65 | 97 | 101 | 108 | 123 | | kew (12), 1898-1904 | 127 | 141 | 113 | 87 | 77 | 70 | 61 | 72 | 76 | 96 | 126 | 153 | | greenwich (13), 1893-1894, 1896 | 110 | 112 | 127 | 107 | 83 | 71 | 76 | 84 | 83 | 104 | 104 | 139 | | florence (14), 1883-1886 | 132 | 110 | 98 | 84 | 86 | 81 | 77 | 90 | 89 | 99 | 129 | 125 | | perpignan (15), 1886-1888 | 121 | 112 | 108 | 89 | 91 | 92 | 89 | 82 | 74 | 99 | 122 | 121 | | lisbon (16), 1884-1886 | 104 | 105 | 104 | 92 | 91 | 93 | 87 | 92 | 100 | 99 | 115 | 117 | | tokyo (17), 1897-1898, 1900-1901| 165 | 145 | 117 | 86 | 62 | 58 | 41 | 59 | 59 | 97 | 134 | 176 | | batavia (18)(2 m.), 1887-1890 | 97 | 115 | 155 | 127 | 129 | 105 | 79 | 62 | 69 | 79 | 90 | 93 | | " (7.8 m.) 1890-1895 | 100 | 89 | 103 | 120 | 98 | 103 | 85 | 99 | 73 | 101 | 117 | 112 | +---------------------------------+-----+-----+-----+-----+-----+-----+----+----+-----+-----+-----+-----+ 3. before considering observational data, it is expedient to mention various sources of uncertainty. above the level plain of absolutely smooth surface, devoid of houses or vegetation, the equipotential surfaces under normal conditions would be strictly horizontal, and if we could determine the potential at one metre above the ground we should have a definite measure of the potential gradient at the earth's surface. the presence, however, of apparatus or observers upsets the conditions, while above uneven ground or near a tree or a building the equipotential surfaces cease to be horizontal. in an ordinary climate a building seems to be practically at the earth's potential; near its walls the equipotential surfaces are highly inclined, and near the ridges they may lie very close together. the height of the walls in the various observatories, the height of the collectors, and the distance they project from the wall vary largely, and sometimes there are external buildings or trees sufficiently near to influence the potential. it is thus futile to compare the absolute voltages met with at two stations, unless allowance can be made for the influence of the environment. with a view to this, it has become increasingly common of late years to publish not the voltages actually observed, but values deduced from them for the potential gradient in the open in volts per metre. observations are made at a given height over level open ground near the observatory, and a comparison with the simultaneous results from the self-recording electrograph enables the records from the latter to be expressed as potential gradients in the open. in the case, however, of many observatories, especially as regards the older records, no data for reduction exist; further, the reduction to the open is at best only an approximation, the success attending which probably varies considerably at different stations. this is one of the reasons why in the figures for the annual and diurnal variations in tables i., ii. and