Aug. I, 1 872 J 



NATURE 



271 



below their crusts the whole land is vioiitoniiA: The abrasion by 

 ice is so uniform and so general that I found it difficult to trace 

 the direction of the abrading motion. Tliere seemed to be no- 

 where a distinct lee and strike-side among the hills. But, as I 

 grew more familiar with the appearance of the country, I became 

 satisfied, and succeeded in convincing others also, that the abrad- 

 ing niDvement had taken plice from the south northward — or, in 

 other worJs, had been connected with the climatic condition of 

 the Southern Hemisphere In Smyth's Channel there is no 

 possibility of mistaking the evidence. I know no more inte- 

 resting locality fur the study of glaciil phenomena than the 

 vicinity of Siuraarez Island. It shows in the most palpable 

 manner that glaciers only — that is, terrestrial masses of ice moving 

 upon solid ground — can liave produced these abrasions, that float- 

 ing icebergs cannot have been the cause. Their direction is such, 

 also, that no one could suppose the adjoining Cordillera to 

 have been in any way connected with the abrasion or plan- 

 ing of the rock, or with its grooving and furrowing. The country 

 has everywhere a gla;ier-worn aspect as far as the Gulf of Pen- 

 nas. Oa reaching Chiloe I noticed that the rounded knolls be- 

 came somewhat less marked, but yet the prominent trend of the 

 hills continued to be in a north-southerly direction. An observer 

 not familiar witli the character of glacial denudation may some- 

 times be perplexed by finding the seeming lee and strike-side of 

 the rocks in a position exactly the reverse of the general one. A 

 critical sciutiny shows that tliese appearances are due to a super- 

 ficial disintegration, often producing a rough side of a hill or 

 locky ledge where the observer of glacial phenomena would ex- 

 pect a smooth and polished surface. This is especially the case 

 here, where, from ihe character of the stone as well as from the 

 climate, the rock peels off and splits up very readily. One must 

 be careful not to blinded by local appearances to the more geniiral 

 phenomena. At the entrance of Corner's Cove, for instance, a 

 beautii'ul inlet trending east-west in Messier Channel, the roci.s 

 forming the southern and northern entrance might seem at list 

 sight to have been ground or smoothed by a local glacier, moving 

 out of the cove in an east-westerly direction. Seen however from 

 a certain distance, where lire local disintegration is merged in tlie 

 general aspect of the exposed surfaces, ihe direction of the main 

 abrasion fiomsouth-nortnward becomes as plain as daylight. Von 

 can trace rectilinear furrows upon the knolls both south and 

 north of ilie entrance of the cove, following not only the same 

 direction, but occupying the same identical level on both sides. 

 There can be no doubt that they were continuous. Darwin has 

 stated that the erratics, the only part of the ancient glacial pheno- 

 mena observed by him in these regions, loUow everywhere the 

 course of the main channels, and he believes this to be an evidence 

 of iceberg action. Valuable as are his results, being, indeed, 

 almost the only connected geological ob;ervations ever made in 

 this region, he is mistaken in his lacts upon this point. When- 

 ever we entered an inlet opening at right angles into the main 

 channel and intersecting several parallel ridges of hills, tiie roi/iis 

 moutonnces and all the accompanying glacial phenomena trended 

 iu a south-northerly direction ; as they did also in the main 

 channel. Before entering the oulf of Fennas, in Messier Chan- 

 nel, we passed an opening through whicli seven parallel ridges 

 could be seen on the eastern side and five toward the west, all 

 trending mainly northward, and plainly exhibiting glacier-worn 

 surfaces. 



Moreover, the Strait of Magellan itself has a main trend 

 from east to west, and yet there is no sign throughout tire 

 whole length of any transportation of erratics from east to 

 west, or from west to east. Dawson has made a similar mis- 

 take with reference to Switzerland. He supposes that the 

 erratics of the Jura were deposited by icebergs sweeping up 

 and dov\'n the great valley of Switzerland, from east to west 

 and from west to east. He seems not to know that the older 

 Esc'ner voa der Linth and Leopold von Bach had already clearly 

 demonstrated the line of their transportation across the valley of 

 Switzerland from south to northward ; and that Guyot, more 

 than twenty-five years ago, traced the different tracks of those 

 boulders separately through the chief valleys of Switzerland 

 northward across the very road which Dawson would have them 

 follow. 



The erratic pebbles and boulders from the e. stern to the 

 western coa^t of Patagonia, judging from my observations at 

 Montevideo, iu the Bay of San Malhias, in Possessicn Bay, at 

 Sandy Pcint, in all the ports of the Straits of Magellan which 

 we have visited, at Shell Bay, on entermg Smyth's Channel, 

 throughout Smyth's Chamiel itself, and upon the shores of 



Chiloe, have the same character. They consist of a mixture of 

 plutonic and metamorphic rocks, among which the hardest sili- 

 ceous rocks prevail. Their geological identity is further shown 

 by the unfailing presence of a very hard, compact, epidotic rock, 

 never absent from these erratic materials, yet never found in place, 

 as far as I know, over the whole c.vtent of country through 

 which I have traced them. Vou will remember that I mentioned 

 it among the loose pebbles of San Mathias Bay ; nor did I lose 

 sight of it until we left San Carlos, at the northern end of Chiloe 

 Island, where I found it again, and as you wiU soon see, in still more 

 interesting juxtaposition. This fact is of great significance, inas- 

 much as it shows that the drift phenomena in this region cannot 

 have been due to the enlargement of the present glaciers, other- 

 wise the drift would consist mainly of the rocks in place, and 

 differ from one locality to the other. And yet their glacial origin 

 is unmistakable, since a considerable proportion of these pebbles 

 and boulders are polished, scratched, grooved, and funowed, like 

 the erratics of the United States and of Northern Europe. It is 

 this uniformity in the character of the drift which has led me 

 from the first to discriminate between the glaciers as they exist, 

 and even as they once existed in their greater extension, in short, 

 between all the phenomena conriected with local glaciers, and 

 those belonging to what 1 have called the glacial period, during 

 which the two liemispheres must have been capped with a sheet 

 of ice of enormous thickness and extent. The equatorial limit of 

 this ice- sheet, both in the northern and southern hemisphere, is 

 part of the problem upon which we have thus far fewest facts 

 in our possession. In South America I have now traced the facts 

 from the southernmost point of the continent uninterruptedly to 

 37° S. latitude, on the Atlantic as well as the Pacific coast. Even 

 here at Talcahuano, large erratic boulders and rociics jnoulonnas 

 exist at the mouth of theBiobio on the hills of Hualpen. 



In San Carlos de Anend, at tlie northern end of Chiloe Island, I 

 have observed a fact which introduces a new element in the study 

 of the glacial period. The ground upon which San Carlos is 

 built is volcanic : the promontory of San Carlos consists of a 

 volcanic breccia, the precise age of which I had no means of 

 determining. From its mineralogical character, it must belong 

 to the .age of volcanoes proper. Now, erratic materials, small 

 pebbles, and large boulders, among which some exhibit unmis- 

 takable glacial polish, rest in considerable quantity upon this 

 volcanic ground. It is therefore plain that the glacial period in 

 this part of the world, at least, has followed the older volcanic 

 eruptions. Among these erratic materials the green epidote 

 which I had followed so far was still to be found. The facts ob- 

 served Ijy me at San Carlos, taken in connection with Pourtales's 

 discovery of a great many extinct craters near Possession Bay, 

 point to the possibility of climatic changes in this region, which, 

 should similar facts be found elsewhere, may account for the 

 glacial period. At all events, it shows a direct connection be- 

 tween the glacial period and volcanic phenomena. Since finding 

 drift upon volcanic ground at Aneiid, I have been watching for 

 erratic pebbles and boiilders of volcanic rocks along the coast of 

 Chili. Their presence near the shore would piove that the 

 glaciers of the Andes formerly reached the sea-levtl, after cross- 

 ing the coast ranges in the temperate, and perhaps also in the 

 tropical zone. Thus far I have failed to find anything of the 

 kind. Darwin assumes that the erratics of western Patagonia 

 have descended from the Andes, and he compares the outlying 

 islands, such as Chiloe, in their relation to the Corderillas, with 

 the chain of the Jura in its relations to the Alps. But the erra- 

 tics of Chiloe have the same character as those of the Strait of 

 Magellan and of San Mathias Bay and the two latter, and those 

 of the two latter can hardly be referred to this source. Neither 

 did I see any indicalicn of very large glaciers coming down 

 from Ihe Andes in a westerly direction, though I have no doubt 

 that I shall find them farther north. Evidently we are not yet 

 sufficiently advanced in our journey from the southern extremity 

 of the continent northward that the influences of altitude should 

 outweii_h those of latitude in the increase and decrease of those 

 climatic conditions upon which the extension of glaciers has de- 

 pended in former ages. During the waning of the glacial period, 

 the glaciers of the Cordillera have unquestionably been much 

 more extensive than now, and I shall not be surprised to find, 

 upon a more careful survey, that the glacier of Snowy Bay in 

 Smyth's Chai.nel and those of Eyre Sound, and perhaps some 

 of the other parts of the Cordilleras, once crossed tl.e main chan- 

 nel and reached the opposite island. But I doubt that they e\er 

 reached the shores of the Pacific Ocean. It is at all events cer- 

 tain that the local glaciers of the present time have never had 



