§82 



SCIENCE. 



homogeneous whole. The channel is rarely so regular 

 as to allow the ice to be altogether free from crevasses, 

 however. Such fissures are invariably at right angles to 

 the line of greatest tension, and greatly facilitate melting 

 by increasing the exposed surface ; and when the rills 

 formed by superficial melting flow into them they 

 may be converted into cylindrical shafts, or moulins, ex- 

 tending to the base of the ice. Thus both crevasses 

 and moulins remain practically stationary, or rather, 

 when either has passed beyond the obstruction or irreg- 

 ularity of the channel which produced it, it gradually 

 closes, and another forms in the same place, with respect 

 to the valley and not to the moving ice, as that which it 

 originally occupied. Aside from the longitudinal medial 

 moraines the surface of the ice is often indistinctly 

 marked by depressed transverse bands within which 

 wind-blown sand or dust accumulates (known as dirt- 

 bands) ; which bands curve downward medially more 

 and more toward the debouchure of the glacier, and thus 

 attest the differential motion of the various parts of the 

 ice-stream. There are, moreover, occasional scattered 

 blocks of stone and small pebbles lying upon the surface 

 of the ice. The larger blocks prevent superficial melting 

 of the ice on which they rest, and hence become appar- 

 ently lifted on columns of ice, forming glacier-tables, 

 which sometimes reach a height of some feet ; while the 

 smaller pebbles, on the other hand, facilitate melting, 

 and thus gradually sink into miniature wells perhaps 

 several inches in depth. 



Glaciers of the alpine type are supplied by the perpet- 

 ual snows accumulating in the elevated valleys and plains 

 intervening between the highest peaks. Over these 

 snowfields — the nivi or firn — the snow is generally- 

 granular and contains much air, especially near the sur- 

 face ; though where it is thick its basal portions may ap- 

 proximate true ice in structure. It is only when the nevS 

 passes over the considerable declivity generally separ- 

 ating the snow-field from the ice-stream proper, and des- 

 cends below the snow-line, however, that it becomes 

 compacted, deprived of its air, and diminished in volume, 

 so as to constitute a veritable glacier. A glacial area 

 may accordingly be divided into two distinct regions on 

 this basis alone ; — the neve, the locality of no melting but 

 of constant addition ; and the glacier proper, the locality 

 of constant decrease. In polar regions the glacial phe- 

 nomena are more varied. Thus, in Greenland, the 

 transition may be observed from glaciers of the charac- 

 teristic alpine type to those of the characteristic polar 

 type, in which the snow-line is at the sea level and the 

 ice is essentially identical with the Swiss nev6, though of 

 vastly greater thickness. It is only slow-moving glaciers 

 of the polar type that give origin to ice-bergs — the term- 

 inal portion extending into the sea " until the buoyancy 

 of the ice causes a mass to break away from its attach- 

 ments, rise to the surface, and float away, (p. 28,) scat- 

 tering the debris frozen to its base over the sea-bottom 

 as it gradually melts ; for the bergs, as the neve of which 

 they are formed, are generally destitute of superficial ac- 

 cumulations of earth and stones. 



Since in the circumpolar regions the snow-line descends 

 to the sea-level, the ice of winter may not be melted dur- 

 ing the succeeding summer, but may remain in situ for 

 years or ages, as in the paleocrystic sea seen by Nares. 

 Now such a sea might be itself overspread with snow to 

 such a depth as to depress the ice to the sea-bottom and 

 to convert the whole mass into nev6 similar to that of 

 northern Greenland, or into a true continental glacier. 

 Indeed,— "it ssems probable that the so-called antoretic 

 continent is nothing but an immense sheet of ice such as 

 this paleocrystic sea would become if it were to increase 

 in depth until it fastened on the bottom of the sea." 

 (P. JI.) 



III. Distribution of the existing glacier.— In the 

 Scandinavian mountains there are the large snow-field 

 in the gostedal highland with many scattered glaciers of 



considerable interest, and, in lat. 70°, a vast snow-field 

 with an immense ice-stream descending to the sea level ; 

 while on the opposite side of Russia the Ural range is 

 without glaciers. In the Pyrenees the glaciers are much 

 shrunken, and mainly confined to moist northern slopes, 

 though about one hundred in number. In the Alps 

 there are over a thousand glaciers, occupying, with the 

 neve, about one-seventh of the mountainous alpine re- 

 gion. Eastward there are no glaciers until the Cau- 

 casus is reached, where a considerable snowy range, 

 with ice-streams on both slopes, is found. A few scat- 

 tered glaciers are known in Asia Minor, one in Persia 

 (on the volcano Demarend,) and many on Hindu Kush ; 

 though these have been but imperfectly described. In 

 the Himalayas the glaciers are of remarkable size and 

 extent, though as yet but partially known. In the South- 

 ern Alps of New Zealand the glaciers are also of con- 

 siderable extent and of great interest. On the western 

 hemisphere glaciers occur along the western border of 

 South America as far north as Upper Chili, where they 

 mainly disappear, and are but meagerly represented 

 along the Andes and Cordilleras until Oregon and Wash- 

 ington Territory are reached. Those occurring within the 

 United States are of little prominence, however ; but they 

 increase in size and number northward, until at Mount 

 St. Elias the ice reaches the sea level. In both Arctic and 

 Antarctic regions there are also immense bodies of mov- 

 ing ice or neve, constituting glaciers of the polar type. 



It thus appears that glaciers are mainly confined (a) 

 to regions of great cold and considerable precipitation, 

 (b) to mountain ranges along western coasts outside of 

 the trade- wind zones in regions of heavy and frequent 

 precipitation, and (c) to interior ranges of great height 

 and considerable snow-fall ; while (a) broad arid areas — 

 even though " the ground is frozen to the depth of sev- 

 eral hundred feet " (p. 36), — (b) interior ranges of lim- 

 ited snow-fall, and (c) regions having a hot and dry 

 summer, are generally free from glaciers. The essential 

 conditions for glaciation are hence, 1st., cold 6f consider- 

 able intensity ; 2nd., considerable snow-fall ; and 3rd., 

 the absence of a dry season of sufficient length to melt 

 the winters' snow. 



IV. Distribution of ancient glaciers. — " The most 

 remarkable fact that has been discovered by geologists 

 during this century is, that at various times in the earth's 

 history the glaciers, which now cover but a very small 

 space on the earth's surface, certainly not over about one 

 hundreth of its area of land, have been extended until 

 they occupied a very large part of land and sea "( p. 38 )." 

 The glacial records are, however, so ephemeral that none 

 save the last ice-period can ever be well known to us. 

 During this period the accumulation of ice was most ex- 

 tensive in regions where glaciers yet prevail, or where the 

 various meteorological conditions at least approach those 

 which existing ice-fields indicate to be essential for glacia- 

 tion, as in the Alps, the Pyrenees, and Scandinavia, over 

 northern Europe, and in the Himalayas and New Zealand, 

 on the eastern hemisphere ; and over much of the north- 

 ern portion of North America and a lesser area in the 

 southern extremity of South America as well as isolated 

 localities along the Andes and Cordilleras, in the western 

 hemisphere. Over the plains of Switzerland an ice-sheet 

 more than 4,000 feet thick swept its debris to the flanks of 

 the Jura, a hundred miles away ; but on the northern slope 

 of the Alps the extension was less. Here the direction 

 of motion was everywhere determined or at least modified 

 by local topographical features. In the Pyrenees, the Ap- 

 penines, the volcanic mountains of central France, and 

 the Jura, in the Vosges, and in Corsica, the accumulation 

 of ice was little more than the development of an exten- 

 sive system of local glaciers ; and north of the Alps there 

 is little evidence of glaciation within inland Europe. The 

 most complete testimony concerning European glaciation 

 in the Quaternary, is furnished by Scandinavia and Great 

 Britain. " Stretching from Scandinavia across the North 



