/8E*KELEY\ 
 
 LIBRARY 
 
 UNIVERSITY Of 
 
 EAH7H 
 
 SCIENCES 
 LIBRARY 
 
THE GLACIAL NIGHTMAEE 
 AND THE FLOOD. 
 
 VOL. I. 
 
THE 
 
 GLACIAL NIGHTMAEE 
 
 AND THE FLOOD 
 
 V 
 
 A SECOND APPEAL TO COMMON SENSE 
 
 FROM THE EXTRAVAGANCE OF 
 
 SOME RECENT GEOLOGY 
 
 SIR HENRY H. HOWORTH, K.C.I.E., M.P., F.G.S., ETC, 
 
 AUTHOR OF "THE MAMMOTH AND THE FLOOD," "THE HISTORY OF 
 
 THE MONGOLS," " CHIXGHIZ KHAN AND HIS ANCESTORS," 
 
 ETC., ETC. 
 
 IN TWO VOLUMES 
 VOL. I. 
 
 LONDON 
 SAMPSON LOW, MARSTON & COMPANY 
 
 Limited 
 
 FETTER LANE, FLEET STEEEI, E.G. 
 
 1892 
 [All rights reserved'] 
 
SCIENCES 
 UBRAR.Y 
 
ZEo tbe /Ifoemocs of 
 CHARPEN.TIER, FORBES AND MURCHISON, 
 
 THREE TEACHERS FROM WHOM I HAVE LEARNT 
 MORE THAN I CAN TELL. 
 
 I wish it were possible for me to emulate their gifts. Their industry, 
 patience and accuracy, their sober and mature judgment, their scientific 
 courage, and above all, the possession of that eagle eye which enabled 
 them to overlook a great horizon without having their view distorted by 
 the smaller details of the landscape. Their thought and reasoning were 
 pursued by no mediaeval ghosts in the shape of scholastic formulae, and 
 they were not afraid to attribute great events to corresponding causes, 
 however much the induction might jar with the shibboleths of the orthodox 
 science of the day. If I cannot emulate their gifts I trust I may have 
 profited by their teaching and example, and that what follows would not 
 have been deemed by them an unworthy tribute from their disciple and 
 follower, 
 
 THE AUTHOR. 
 
PREFACE. 
 
 IT is a singular and a notable fact, that while most other 
 branches of science have emancipated themselves from the 
 trammels of metaphysical reasoning, the science of geology 
 still remains imprisoned in " a priori " theories. 
 
 It was many years ago that one of the teachers to whom I 
 am under great obligations, Adam Sedgwick, protested with 
 the gravity of a true philosopher against this method of reason- 
 ing. "The study of the great physical mutations on the sur- 
 face of the earth," he said, "is the business of geology. But 
 who can define the limits of these mutations ? They have 
 been drawn by the hand of Nature, and may be studied in the 
 record of her works, but they never have been, and never will 
 be, fixed by any guesses of our own, or by any trains of a 
 priori reasoning based upon hypothetical analysis. We must 
 banish all a priori reasoning from the threshold of our argu- 
 ment ; and the language of theory can never fall from our 
 lips with any grace or fitness, unless it appear as the simple 
 enunciation of those general facts, with which, by observation 
 alone, wehave at length become acquainted. . . . Anhypothesis 
 is indeed (when we are all agreed on receiving it) an admirable 
 means of marshalling scattered facts together. But by those 
 who differ from us, an hypothesis will ever be regarded with 
 just suspicion ; for it too often becomes, even in spite of our 
 best efforts, like a false horizon in astronomy, and vitiates all 
 the results of our observations, however varied, or many times 
 repeated/' l 
 
 With these words I most cordially agree, but if they were 
 
 wise and just in 1831, they are equally wise and just now, 
 
 when our science has become more professional and when 
 
 authority therefore has become more paramount. " That the 
 
 1 Address to the Greol. Soc. 
 
 260907 
 
viii Preface. 
 
 primary laws of Nature are immutable that all we now see 
 is subordinate to those immutable laws and that we can only 
 judge of effects which are past by the effects we behold in 
 progress " are truisms of all science. When this sound doc- 
 trine is perverted, however, to teaching men that " the physical 
 operations now going on are not only the type, but the 
 measure of intensity of the physical powers acting on tbe 
 earth at all anterior periods. This is to assume an unwarrant- 
 able hypothesis with no a priori probability." This is the 
 language of another great geologist, a distinguished observer 
 and generalizer. 
 
 The words were originally directed against the development 
 of the views of Hutton and Playfair, which we owe to the 
 brilliant and fascinating pen of Lyell. He had explained 
 with great wealth of illustration and a wonderful command of 
 limpid and picturesque language, the necessity of studying 
 Nature at work, if we are to unravel the history of Nature 
 from the ruins of her handiwork. ! He taught, as it had never 
 been taught before, that rain, snow, and frost, that river and 
 sea, that volcano and earthquake, that these and similar 
 instruments which Nature is continuously employing now, are 
 capable, if we allow a sufficient draft upon time, of producing 
 the largest effectsj So far no one quarrels with him, and so 
 far he was following modern methods in science, those 
 methods which are alone inductive and empirical. When he 
 went beyond this he travelled beyond the true limits of induc- 
 tion, and against the journey Sedgwick,Murchison, Hopkins and 
 others protested. The protest was, however, largely addressed 
 to an obstinate and perverse generation, which in its rebellion 
 against the old teaching whereby every difficulty was solved 
 by an appeal to the direct interposition of the Almighty, re- 
 fused to listen to any other voice than that of Uniformity. 
 Not uniformity in the sense that Nature has worked with 
 similar tools and with similar methods in all ages, but in the 
 sense that she has always worked with the same vigour and 
 intensity, or, as Mr. Conybeare translated the argument as 
 applied to geology, because a child grows two inches every 
 year, therefore, that is the normal growth of a human being 
 during the thiee score years and ten which are his allotted 
 pilgrimage. 
 
Preface. ix 
 
 To uniformity in its former sense we all adhere, against 
 uniformity in its latter and transcendental and metaphysical 
 sense the following work is meant to be a protest as its pre- 
 decessor was. 
 
 Misleading as I deem the arguments of Lyell and his- 
 scholars to have been when applied to the older beds, they 
 were much more so when applied to explain the superficial 
 mantle of gravel, clay, sand, etc., which covers the ragged and 
 ruined surface of the older rocks, and gives to the earth its 
 generally smooth and undulating outline. The former deposits 
 are for the most part arranged in regularly stratified beds,, 
 with a regular succession which can be studied in many places 
 in an undisturbed condition. It is very different with the 
 superficial soft beds, which are so incongruous and hetero- 
 geneous in structure, which mantle the country irrespective 
 of its contour, which often contain blocks of stone that have 
 travelled hundreds of miles from home, and which afford so 
 many puzzles to us all. Yet if there be a geological horizon, 
 which it is important that we should study on sound principles, 
 it is assuredly this one, for it enshrines the last completed 
 chapter in the history of the world, and, among many other 
 interesting riddles, contains the explanation of the mysterious 
 problem of the origin of our race. 
 
 To explain these beds has been the anxious effort of several 
 generations of geologists, and the difficulty of the task may be 
 measured by the divergent theories which have been forth- 
 coming about them, and by the further fact that they still 
 remain in many respects the despair of geology. 
 
 Forbes, in his chapter on the Geological agency of Glaciers, 
 writes as follows : " The occurrence of vast masses of primi- 
 tive rocks, apparently without any great wear and tear of 
 travelling, upon secondary or alluvial surfaces at great dis- 
 tances from their origin, has been one of the numerousjpppro- 
 bria of geology. It is peculiarly so because a thousand cir- 
 cumstances demonstrate that the deposition of these masses 
 has taken place at the very last period of the earth's history. 
 No considerable changes of surface have occurred since. 
 These blocks are superficial, naked, deposited upon bare rock, 
 which has received no coating of soil since, and are often 
 placed in positions of such ticklish equilibrium that any con- 
 
x Preface. 
 
 siderable convulsion of nature, whether by earthquake or 
 debacle, must inevitably have displaced them. A geologist 
 might therefore justly be asked, " If you cannot account for 
 these very latest and plainest phenomena of change and trans- 
 port on the earth's surface, whose various revolutions you pre- 
 tend to explain, how shall we follow you, when you tell us of 
 the metamorphoses of slates and the throes of granite ? " 
 
 This was written in 1843. In 1866 M. d'Archiac, one of 
 the most accomplished and trustworthy observers of recent 
 times, could still write : <f One of the epochs of Nature about 
 which most has been written in recent times, that which 
 immediately preceded our own and of which the duration was 
 apparently not very long is, nevertheless, that about which 
 we know the least, and which has given rise to the greatest 
 number of hypotheses. The slight development of marine and 
 lacustrine deposits, compared with the great predominance if 
 not universal distribution of the erratic depots, their small thick- 
 ness over wide extents of country, the entanglement of one 
 part, and the regular succession of the other, the absence of 
 regularity, symmetry and continuity in the disposition of the 
 beds, have caused the comparisons and relations which it has 
 been attempted to evolve from them to be incomplete and 
 uncertain." 3 
 
 I could easily quote passages from the geological writings of 
 men in the forefront of our science on both sides of the Atlantic, 
 to show that in the year 1892 we are still face to face with the 
 unsolved problem of rationally explaining the manifold and 
 perverse difficulties presented by the latest geological deposits. 
 
 I have thought it right in the following pages to begin 
 with an account at considerable length of the various theories 
 which have been propounded to explain a large series of these 
 beds. I have done so more especially, because my own con- 
 clusion largely involves a return to the views of those older 
 geologists, who wrote before the world was dazzled by the 
 extravagant development of the Glacial theory, and because 
 their careful labours have been recently ignored. 
 
 At one time most men agreed that these beds were assorted 
 and distributed by water, and the great conflict between the 
 Huttonians and their opponents, was as to whether the water 
 Forbes' Travels, 40 and 41. 3 Geologie et Paleontologie'693, 
 
Preface. xi 
 
 acted in anornial way, that is in the form of rivers, or the sea, or 
 in an abnormal way, that is by means of catastrophes or debacles, 
 as it was the fashion to call them. It must be said that the 
 Huttonians had found it hard to keep the field, for it seemed 
 impossible to realize that water acting in any normal way 
 could transport great blocks of stone for hundreds of miles, 
 and spread vast sheets of gravel and clay over whole continents. 
 
 Hence the stretching of the theory of Uniformity almost 
 to the breaking point by Lyell and Darwin, who in order to 
 meet the facts, postulated the submergence of whole con- 
 tinents, and the floating over them from some unknown Arctic 
 sea of fleets of icebergs which were supposed to have sailed 
 in and out of the Alpine valleys, across Germany and Russia, 
 etc. etc., and to have dropped their burdens as they went. 
 The sea bottom is then supposed to have risen again, and to 
 have disclosed what we now see. 
 
 The impossibility of maintaining this view in presence of the 
 facts, will, I think, be found to be satisfactorily shown in the 
 penultimate chapter of the following work ; although this was 
 scarcely needed, since the view itself was ultimately abandoned 
 by its principal champions. 
 
 At this stage Charpentier and Agassiz, qame to the rescue 
 of the sorely tried theory of Uniformity. The Swiss geologists, 
 who had studied the modus operandi and results of glacier 
 work, had also noticed signs and evidences in Switzerland that 
 the modern glaciers are only shrunken relics of much larger 
 glaciers which at one time existed there, and that the unmis- 
 takable work of these old glaciers could be traced far down 
 the valleys in the shape of polished and striated rocks, 
 transported debris, etc., etc. This led to inquiry being made 
 elsewhere, and it soon became patent that, unless we are to 
 ignore the clearest evidence, we must allow that in the last 
 period of the earth's history there was a development of glaciers 
 on a large scale in nearly all latitudes where high land existed. 
 
 This view, which was first propounded and elaborately 
 argued by Charpentier, has stood the test of criticism, and 
 seems to be established beyond question. 
 
 When ice was shown to have had a greater extension, and 
 consequently a greater potency in former times, than now, it 
 opened up possibilities as to its capacity to explain a great 
 
xii Preface. 
 
 deal more than had been dreamt of. The position was at once 
 seized by Schimper and Agassiz, and their arguments and 
 rhetoric took the world by storm. 
 
 It was plausible to argue that the same force which could 
 carry great masses of granite and other rock from the upper 
 Alpine valleys into the lower country, was competent when 
 working on a larger scale to scratch, furrow and smooth wide 
 extents of country many leagues away from any mountain chain, 
 or any sloping ground ; to remove masses of gravel, boulders 
 and clay, mixed with unweathered stone, over many degrees 
 of latitude and longtitude, and to explain all the varied and 
 widespread phenomena of the drift. Hence the invocation 
 of a Glacial Period to explain what were deemed to be plainly 
 glacial phenomena. 
 
 The Glacial theory as thus propounded by Schimper and 
 Agassiz, was naturally very welcome to the champions of 
 Uniformity, for it seemed to offer a way of escape from a 
 great difficulty without appealing to other than every-day 
 causes. 
 
 It is true that the magnitude and wide geographical distri- 
 bution of the facts to be explained involved unusual courage 
 on the part of those who held this view, and some of his 
 followers were startled and confounded by the lengths to 
 which Agassiz was prepared to go. For he not only invoked 
 a time when a portentous winding sheet of ice covered the 
 temperate zone, but he also took it into the tropics, and 
 argued that a similar icy mantle once filled the valley of the 
 Amazon in Brazil. 
 
 While this last extension of the theory met with con- 
 siderable opposition, the main conclusion for which he fought, 
 namely, that the drift was deposited by continental ice-sheets 
 conterminous with the drift itself, became the predominant 
 creed of a large and powerful school of geologists. 
 
 Some of Agassiz's eager scholars soon saw that the new theory 
 needed external support. If, as they argued, the doctrine of 
 Uniformity is conciliated by the prospect of a great ice age 
 and widespread ice-sheets, it also requires that, such an ice 
 age should be explained without a breach in the continuity 
 of Nature's laws. Heoce the fervid appeals by the Glacial 
 school of geologists to astronomy and meteorology to furnish 
 
Preface. xiii 
 
 a satisfactory basis on which their theory might stand, and 
 which should enable them to say that an ice age was not a mere 
 accidental occurrence, whose explanation is uncertain, but a 
 part of the great scheme by which the Universe is governed, and < 
 the natural result of the ordinary operation of Nature's laws. 
 
 The challenge was not neglected, and many elaborate 
 theories have been evolved by sharp-witted and ingenious 
 men, in order to satisfy the champions of the apparently 
 triumphant and unanswerable Glacial theory. A criticism of 
 these various theories will be found in the following pages, 
 where it is attempted to be shown that they have all com- 
 pletely failed, either to meet the facts of geology, or to 
 be consistent with the laws of physics, and it seems to be 
 as plain as anything can be that the dictum of the great 
 French geologist, Charles Martins, is still true, namely, " that, 
 if the Glacial theory be sound, its explanation is a secret which 
 the future alone can disclose." 
 
 One of the chief objects of the following book, however, is 
 to show that the Glacial theory, as usually taught, is not 
 sound, but that it ignores, and is at issue with, the laws which 
 govern the movements of ice, while the geological phenomena 
 to be explained .refuse to be equated with it. This is 
 partially acknowledged by the principal apostles of the ice 
 theory. They admit that ice as we know it in the labora- 
 tory, or ice as we know it in glaciers, acts quite differently 
 to the ice they postulate, and^pr^djic^sJiSereTrr^effects, 
 but we are bidden to put aside our puny experiments 
 w^ich can be tested, and to turn from the glaciers which can be 
 explored and examined, to the vast potentiality of ice in the 
 shape of portentous ice-sheets beyond the reach of empirical 
 tests, and which we are told acted quite differently to ordinary 
 ice. That is to say, they appeal from sub-lunary experiments to 
 a priori arguments drawn from a transcendental world. As- 
 suredly this is a curious position for the champions of Uni- 
 formity to occupy. In regard to it I will quote a fable 
 already utilized by Hugh Miller. ' A wolf/ says Plutarch, 
 " peeping into a hut where a company of shepherds were 
 assembled, saw them regaling themselves with a joint of 
 mutton. ' Ye gods ! ' he exclaimed, ' what a clamour they 
 would have raised if they had caught me at such a banquet.' " 
 
xiv Preface. 
 
 I hold that the Glacial theory, as ordinarily taught, is based 
 not upon induction, but upon hypotheses, some of which 
 are incapable of verification, while others can be shown to be 
 false, and it has all the infirmity of the science of the Middle 
 Ages. This is why I have called it a Glacial Nightmare. 
 Holding it to be false, I hold further that no theory of 
 modern times has had a more disastrously mischievous effect 
 upon the progress of Natural Science. It is not merely in the 
 domain of geology that its baneful influence has been felt. 
 We cannot take up a text book, in which the profounder 
 problems of Biology are treated, problems like the distribution 
 of animals and plants, the pedigree of life, the origin, and 
 beginnings of the human race, without being impressed with 
 its influence as a factor. 
 
 In all these and many other inquiries, the postulate of an ice 
 age forms a necessary element of current theories. The length 
 of the following argument is in fact the homage paid to the 
 widespread influence of the theory, which made it incumbent 
 in disputing it to do so with such care and completeness as I 
 can command, amid manifold duties, and. indifferent health. 
 
 What, then, is shortly the burden of the following 
 argument ? 
 
 I admit completely, that the position maintained by Char- 
 pentier in his work on Glaciers is unassailable, first, because 
 it makes no appeal to any occult and hidden forces underlying 
 the movements of ice, but proves the existence of greater 
 glaciers formerly by comparing and equating the ruins they 
 have left with the ruins made by existing glaciers. Secondly, 
 because it is consistent with all the geological facts that we 
 can summon to test it by. On the other hand, I not only dis- 
 believe in, but I utterly deny the possibility of ice having 
 moved over hundreds of miles of level country, such as we see 
 in Poland and Russia, and the prairies of North America, and 
 distributed the drift as we find it there. I further deny its 
 capacity to mount long slopes, or to traverse uneven ground 
 except when under the impulse of gravity. I similarly deny 
 to it the excavating and denuding power which has been 
 attributed to it by those who claim it as the excavator of 
 lakes and valleys, and I altogether question the legitimacy of 
 arguments based upon a supposed physical capacity, which 
 
Preface. xv 
 
 cannot be tested by experiment, and which, is entirely based 
 upon hypothesis. This means that I utterly question the 
 prime postulate of the Glacial theory itself. 
 
 The purport of the following pages, however, is not wholly 
 critical and destructive. If we sweep away the Glacial theory 
 in its extravagant form, we have not swept away the difficulties 
 which it was introduced to explain. It is our duty, it seems 
 to me, in view of the importance of the issue, to meet these 
 difficulties. How then are they to be met ? The answer to 
 this question I have tried to give in the last chapters of 
 this book. The answer is not one consistent with the modern 
 developments of the theory of Uniformity, and for those 
 who hold with Ramsay that Nature has not varied either in 
 the rapidity, the intensity, or the quality of her work since 
 she began to mould the world, there is nothing in those 
 chapters which can prove acceptable. My masters, Sedgwick 
 and Murchison, taught me a very different lesson which I have 
 seen no reason whatever to unlearn. They taught me that no 
 plainer witness is to be found of any physical fact than that 
 Nature has at times worked with enormous energy and 
 rapidity, and at others much more evenly and quietly, and that 
 the rocky strata teem with evidence of violent and sudden dis- 
 locations on a great scale. That these catastrophes were 
 aimless and lawless, I do not believe. On the contrary, they 
 were the result of law, but of a law whose tendency we have not 
 yet perhaps duly measured, and whose key we might perhaps 
 have discovered if we had not been pursuing the fantastic 
 shadows of metaphysical reasoning for so many years. 
 
 This conclusion is, in effect, the same as that arrived at by 
 some of the best writers on the subject. Let me quote from 
 one of them. 
 
 " No one appreciates more highly than I do," says Mr. Bab- 
 bage, " the labours of those older geologists," that is, Lyell and 
 his school, " who first taught an earlier generation to estimate 
 more truly the power and efficiency of certain forces which act 
 very slowly but continuously during long periods of time. The 
 error in scientific speculation against which they fought was 
 the error of laying exclusive or exaggerated stress upon 
 forces of a particular kind forces the existence of which they 
 did not deny, but which had worked only at comparatively 
 
xvi Preface. 
 
 distant intervals and in alliance always with other forces, the 
 operation of which is ceaseless. It is precisely the same error 
 in principle, though exhibited in a different form, which is now 
 exhibited by those geologists who attribute almost every- 
 thing to running water and to scraping ice. They are simply 
 catastrophists in a new dress. They attribute extravagant 
 power and stupendous effects to one form of force instead of 
 to another. There is, indeed, one difference, and it is a 
 difference in favour of the older school of catastrophists rather 
 than of the younger that whereas there never could be any 
 doubt of the adequacy of subterranean force to produce the 
 effects ascribed to it, there is the greatest doubt of the 
 adequacy of rain and ice to effect in any time, however long, 
 the stupendous changes ascribed to them by Mr. Geikie. On 
 the other, there is really nothing stupendous about these 
 effects when they are regarded in connection with the known 
 and visible effects of subterranean force. The highest ranges 
 of mountains we have are, relatively to the circumference of 
 the earth's crust, infinitely smaller than the puckers on an 
 orange-skin." 
 
 Again, the same writer says : " Magnitude is all relative. 
 The store of Time and the store of Force may be regarded as 
 both unlimited. But it does not follow that in accounting for 
 any given effect we are entitled to draw to an unlimited 
 extent either upon the one or upon the other. Extravagant 
 demands may as easily be made upon the one or upon the 
 other. The inventions and imaginations to which the 
 extreme glacialists resort are, beyond all comparison, more 
 violent than those which were common with the old convul- 
 sionists. Whole continents are built up upon the top of 
 the existing mountains, which there is no proof whatever 
 ever existed ; and then these continents are all ground down 
 by ice or washed away by ordinary surf, and yet so that not a 
 fragment shall be left behind. I venture to believe that I 
 shall have some support from the great leaders of geological 
 science, who, in power of intellect, are still young among us, 
 when I record nay dissent from the extravagant theories of the 
 younger glacialists." 5 
 
 With this reasoning I quite agree. The earth seems to me 
 4 Babbage, Journ. Geol. Soc. xxiv. 272 and 273. 5 Id. xxiv. 273. 
 
Preface. xvii 
 
 to be full of evidence of intermittent violence and repose. In 
 facing the solution of the Drift problem I must be taken 
 therefore to postulate, not merely the possibility of catastrophes, 
 but to maintain that they have occurred frequently in the 
 world's history. 
 
 Secondly, I would claim that, while the Glacial theory makes 
 demands upon the powers of ice which are inconsistent with its 
 proved qualities and cannot be made to fit in with the facts 
 which have to be explained, the power to which I appeal makes 
 no demands whatever upon any force but that of which we 
 can establish the competency, both by direct experiment and by 
 theoretical calculation, and that, so far as we know, it explains 
 all the facts. 
 
 Thirdly, this explanation is one which was deemed satis- 
 factory and complete by the Fathers of Geology, men who were 
 quite as keen observers and quite as keen critics as their 
 descendants, and who were also more independent, and less 
 dominated by official orthodoxy in Science. For a long time 
 some of the most brilliant masters of our science were advo- 
 cates of the Diluvial theory as an adequate explanation of the 
 facts. It is true that some of them attempted to explain too 
 much by this cause 3 and that until Charpentier published his 
 famous work the operations of land ice, on a much larger scate 
 than that in which it works now, had been overlooked. 
 
 It is true, also, that the views they maintained were in 
 some cases sophisticated by an appeal to untenable postulates, 
 but in the main the theory, of which I am an advocate, is a 
 return to that so strenuously supported by Hall and Cony- 
 beare, Von Buch, Sedgwick, Murchison, D'Archiac, Phillips, 
 and many others. 
 
 Lastly, I claim to have already established the necessity of 
 this appeal on entirely different grounds. In my previous 
 work on the " Mammoth and the Flood," I collected a great 
 mass of evidence which went to show that the Mammoth 
 and some of its companions, including so-called palaeolithic 
 man, were swept away in wide areas by a great flood of 
 waters which drowned them, and then covered them with 
 continuous layers of loam and gravel. The facts and the 
 arguments adduced in that book have been referred to many 
 times by many writers for their manifold character and. wide- 
 
xviii Preface. 
 
 reaching and cumulative effect, and I am pleased to think 
 that some of the first authorities in this country and else- 
 where in the realms of Physics of Natural History and 
 Archaeology, have assured me that they considered the argu- 
 ment unanswerable in so far as the particular facts referred 
 to are concerned, and that what is needed is tliat it should 
 be supplemented by a similar argument based on geological 
 grounds. On the other hand, in the many reviews of that 
 book which have appeared here in France and Germany and 
 in America, I know of none in which its arguments have been 
 met or traversed, while the great majority give a verdict of 
 suspense dependent on the production of adequate geological 
 evidence. 
 
 In the following pages, then, I claim to have shown that a 
 widespread flood, which seems to be a necessary postulate, if we 
 are to adequately grapple with the extinction of the Pleistocene 
 fauna, is an equally necessary postulate if the geological 
 facts are to be duly explained. At all events, I claim for 
 my arguments that they deserve adequate criticism. I am not 
 aware that I have omitted to notice a single fact or diffi- 
 culty which needs explanation, and I have tried to make the 
 book a fairly complete monograph on the controversies with 
 which it deals. That it contains many faults, both of com- 
 mission and omission, I know well, and those who have tried 
 to wade through the vast multitude of publications dealing with 
 the problems here discussed, and who realize also the many 
 and critical and necessary side issues which have had to be 
 sifted and criticized in the process of exhausting the problem, 
 will be tender and considerate to my method and my plan. 
 For the facts, arguments, and conclusions adduced, I ask for 
 no quarter. If they can be shown to be false and misleading, 
 they had better be swept away mercilessly. If they are true, as 
 I believe them to be, they cannot be ignored. Meanwhile, 
 I am ready to enter the arena and to fight hard for them. 
 
 The book does not exhaust the problem, nor does it pro- 
 fess to do so. When it was begun I hoped to have brought 
 together all the geological evidence bearing upon it, but the 
 necessity of entering at much greater length than I intended 
 into the history of the Glacial theory if I was to do justice to 
 the predecessors, upon whose shoulders I am privileged to 
 
Preface. xix 
 
 stand, and the necessity also of discussing at adequate length 
 the many theories of Ice motion and the various astronomical 
 and meteorological explanations of the Ice age, have made it 
 impossible to condense all my arguments in one book. The 
 work is, in fact, limited to the so-called drift beds. That 
 the same conclusion follows, from a consideration of the 
 distribution of the valley gravels, the brick earths and 
 loams, the loess, chernozem and Pampas inud, etc., I have 
 tried to show in papers printed in the Geological Magazine. 
 The facts and arguments contained in those papers I hope 
 to enlarge and deal with in another volume, in which the 
 geological evidence forthcoming from countries where the 
 true drift does not occur, will have a part, and in which 
 it may be possible perhaps to disc ass the extent and the 
 cause of the Pleistocene Flood, and the question of whether 
 it was one of a recurrent series of similar catastrophes, as I 
 am disposed to think it was, or an exceptional and unique 
 event. 
 
 At present I merely claim to have made good my promise to 
 produce adequate geological evidence in support of the theory 
 I advanced in my former work. This theory seems to me to 
 be overwhelmingly established, and to be an absolutely neces- 
 sary postulate if we are to treat our science as an Inductive 
 science at all. Being so, I would go a step further and not 
 only return to older opinions, but also return to older and 
 clearer and better nomenclature. The Pleistocene Flood, 
 though far from being universal, was certainly one of the 
 most widespread catastrophes which the world has seen. It 
 forms a great dividing line in the superficial deposits as was 
 maintained long ago, and as such, it is a very useful land- 
 mark which ought to appear in our nomenclature, and I do 
 not know of any better terms than ante-diluvian and post- 
 diluvian to mark the two great divisions of the post-Pliocene 
 beds. 
 
 Ante-diluvian times, so far as we can see, were marked in 
 the temperate latitudes 'of both hemispheres by accumulations 
 of ice in the shape of large glaciers on the high lands (then 
 probably much higher) of Western Europe, of North America, 
 of Australia, New Zealand, and perhaps South Africa. Along- 
 side of these glaciers, and in contact with them in all these 
 
 a 2 
 
xx Preface. 
 
 latitudes, were wide champaign and wooded districts in which 
 the Mammoth and the woolly rhinoceros were the most 
 prominent animals in North Asia and Europe, the Mammoth 
 and the Mastodon in North America, the Mastodon and 
 the great Sloths in South America, the various species of 
 gigantic Kangaroos and Wombats in Australia, the great 
 wingless birds in New Zealand. They lived and thrived 
 in the near neighbourhood of the ante-diluvian glaciers, 
 jusfc as the apteryx now thrives in the luxuriant forests 
 near the great glaciers in New Zealand, and just as the 
 tiger and the rhododendron thrive close to the Himalayan 
 glaciers. 
 
 Meanwhile, in Northern Asia and Western Europe, and in 
 North and South America certainly, and probably also in 
 Australia, ante-diluvian man lived alongside of and hunted 
 the ante-diluvian animals, and if we are to clear up his 
 pedigree we must find the true interpretation of these ante- 
 diluvian beds. 
 
 Presently came a tremendous catastrophe, the cause of 
 which, as I have tried to show in the Geological Magazine, was 
 the rapid and perhaps sudden upheaval of some of the largest 
 mountain chains in the world, accompanied probably by great 
 subsidences of land elsewhere. The breaking up of the earth's 
 crust at this time, of which the evidence seems to be over- 
 whelming, necessarily caused great waves of translation to 
 traverse wide continental areas, as Scott Russell, Hopkins, 
 Whewell, and Murchison argued they would, and these waves 
 of translation as necessarily drowned the great beasts and their 
 companions, including palaeolithic man, and covered them 
 with continuous mantles of loam, clay, gravel and sand, as 
 we find them drowned and covered. They also necessarily 
 took up the great blocks which the glaciers had fashioned, 
 and transported them to a certain distance and distributed 
 them, and the Drift associated with them, as we find them 
 distributed. This Induction, whose details are contained in the 
 following pages, seems to me to be complete, not only because 
 it adequately explains the facts, but because it is the only 
 theory that does so, and I know nothing against it, but the 
 almost pathetic devotion of a large school of thinkers to the 
 Religion founded by Hutton, whose High Priest was Lyell, and 
 
Preface. xxi 
 
 which, in essence is based on a priori arguments like those 
 which dominated mediaeval scholasticism and made it so 
 barren. 
 
 I must now say a few words in justification of the mode in 
 which this work has been put together. 
 
 A casual reader, on turning over its pages, will be apt to 
 say : Here is a compilation, a mere collection of opinions and 
 of facts gleaned from other sources. Perhaps if he will take 
 the pains to read a little more closely, he will find that this 
 would be a very inadequate and unfair view of the position. 
 Of course, like every other man who has written a scientific 
 book, I am dependent for a great deal on what others have 
 written and what others have discovered. If it had not been 
 for the scientific decalogue which prescribes, inter alia, that 
 the man who first makes an induction is entitled to the 
 credit of it, I could very easily have written the whole book 
 without giving a reference and without quoting a passage. 
 This also explains the laborious traversing of a vast number 
 of books dating from the earlier days of geological writing, 
 which are most of them forgotten, if justice was to be 
 done to the first rational explanations of the Drift phenomena. 
 Hence also the marshalling of many successive views in illus- 
 tration of the progress of a complete theory on the subject. 
 The survey is, I know, not exhaustive, for the subject is vast 
 beyond conception, but I trust that justice has been done to 
 some men at least, who lived before Agamemnon and whose 
 keen eyes and whose sound judgment it has been the fashion 
 to decry, while much .of what they have written has been 
 treated as obsolete. 
 
 This is my explanation of the large number of opinions 
 quoted, and of references made which perhaps unfairly disguise 
 the fact that I have tried, in view of what follows, to examine 
 the drift phenomena in many countries, and have frequently 
 worked out conclusions for myself for which I have quoted 
 others, because others had already been over the same 
 ground and reached the same results, and were therefore en- 
 titled to the credit of what they first described. The general 
 argument and the general theory, with a great deal of illus- 
 tration, is entirely my own. The men of water had been 
 long ago dispersed and scattered, the men who wonld dare to 
 
xx ii Preface. 
 
 appeal to cataclysms and catastrophes are few indeed* 
 Those who would venture to jeopardize their character for 
 scientific sobriety by reviving and extending the views of the 
 geologists of fifby years ago on great diluvial movements 
 are not to be found anywhere. Whether right or wrong in 
 his conclusion, it is perhaps well that, among the erratic 
 heretics who are careless of prestige and indifferent to con- 
 ventional opinion, one should occasionally be found to challenge 
 the dominant creed by assailing its foundations. 
 
 I hope I may have escaped one penalty of a good deal of 
 polemical writing, and that nothing contained in the following 
 pages will be found offensive to the living or unfair to the 
 dead. If I cannot agree with some of the conclusions of Lyell, 
 of the Geikies, and of Bamsay, I am well aware of the gap 
 that separates us in reputation and in knowledge, and of the 
 debt I owe them all, and no one has a profounder regard and 
 appreciation of what they have done than myself. They have 
 taught me a great deal of what I know. 
 
 I must now turn momentarily to protect myself against 
 misconstruction in another direction. This work, like its 
 predecessor, is written in the interests of no school of thought, 
 and is meant to be merely an inductive argument from the 
 facts. It may be used by some to support a position which 
 is not mine, and which, as] I have said elsewhere, is not a 
 conclusion of Science, but a dogmatic opinion, partly based 
 upon the primitive traditions of mankind and partly on 
 ingenuous speculations in cosmology made before the birth of 
 science. It has no other value. 
 
 No doubt if the catastrophe I have described really 
 occurred in the human period, as my contention is, the fact 
 must have impressed the imagination of the survivors, and it is 
 not unnatural to find it reflected in many scattered traditions. 
 
 This is a very different thing, however, from the conclusion 
 that the views here established support a narrative whose 
 dogmatic value depends upon its details, or give any counte- 
 nance to the notion that the postulated flood was universal, 
 that it destroyed all life save what was sheltered in one 
 vantage, or that the human race began its genealogy again 
 from one fortunate family. 
 
 The story in the Bible is interesting as an early example of 
 
Preface. xxiii 
 
 a widespread tradition, and nothing more. Whether directly 
 derived from Babylonia, or through some intermediate source, 
 we cannot with our present knowledge suppose that the narrative 
 was incorporated in the Pentateuch before the tenth and per- 
 haps not before the seventh century B.C., and that is the measure 
 of its value as a Jewish record. To suppose that religion or 
 morals are helped or furthered by trying to give some greater 
 authority to the story by factitious appeals to geology, must end 
 in disappointment and in failure, and it is right to say plainly 
 that it involves both a dangerous logic and a misleading hope 
 to continually try and equate notions which are incommensur- 
 able, while it must inevitably mislead those who ought not to 
 be misled. The grave and wise words of Sir Francis Bacon 
 ought assuredly to be ever present to us when we embark on 
 the impossible task of equating science and faith. " Tanto 
 magis haec vanitas inhibenda venit et coercenda quia ex 
 divinorum et humanorum male sana admixtione, non solum 
 eduntur philosophia phantastica sed etiam religio hseretica/' 
 These wise words are thus paraphrased by a great geologist, 
 who was also a church dignitary, my master, Sedgwick. 
 " This vanity merits castigation and reproof the more, as, 
 from the mischievous admixture of divine and human things, 
 there is compounded at once a fantastical philosophy and an 
 heretical religion." 6 
 
 It is not part of my subject, but I cannot help parentheti- 
 cally saying that Divines would do well to separate rather 
 more sharply than hitherto the annalistic and ritualistic from 
 the essentially religious books of the Bible. It is not Moses 
 and Samuel who among the ancient sears are very helpful to 
 us in these days in the realm of morals, or in presenting high 
 ideals and standards for our guidance. It is rather Isaiah, 
 Jeremiah and Job. 
 
 I have to thank my old friend Dr. Woodward for allowing 
 me to discuss some of the issues contained in the following 
 pages in the Geological Magazine. I have further to thank the 
 Council of the Manchester Literary and Philosophical Society 
 for permission to reproduce a long paper which I published 
 in its Transactions, dealing with the various theories of ice 
 motion, and I have also to thank a large number of geologists 
 6 Sedgwick, Address, Geol. Soc., 22. 
 
xxiv Preface. 
 
 from whose works I have derived infinite pleasure and profit, 
 and whose labours I have read and criticized. 
 
 Much of what is of value in the following work belongs 
 to them. In regard to the mistakes and blemishes which it 
 contains, and which are largely my own, I have to plead that the 
 book has had to be written under manifold disabilities and 
 especially the double burden of time being heavily mort- 
 gaged by other duties, and health and eyesight having 
 been equally uncertain. This has prevented the necessary 
 revision which the book would otherwise have had. I know 
 well that in many cases its style is slovenly and that many 
 verbal slips occur throughout it. These accidents will, I trust, 
 not entirely monopolize the attention of my critics, who will 
 perhaps do some justice to the long and conscientious labours 
 it enshrines and the important issue it tries to raise. I shall 
 feel indebted to any one who is disposed to comment on 
 what I have said, if he will let me have an opportunity of 
 profiting by the criticism, and will send me what he may write. 
 
 In introducing another work to the frigid welcome of a 
 world overloaded with books, I cannot forget the patience of 
 those who have been the uncomplaining partners of my long 
 toil at home. In conclusion, it is with genuine sympathy and 
 appreciation that I appropriate the words of the weary monk, 
 who, after long days in the Scriptorium wrote, <f Explicit, 
 expliceat ; ludere Scriptor eat." 
 
 THE ATHEN^IUM, 
 
 November, 1892. 
 
CONTENTS. 
 
 CHAPTER I. 
 
 THE CHAMPIONS OF WATER. I. 
 
 PA.GE 
 
 Abundance of boulders in North Europe Some remarkable 
 boulders Views of the unscientific about them The Swedish 
 writers Swedenborg, 1719-22 Tilas, 1740-42 Linnaeus, 1744 
 Bergmann, 1769 Writers on the Alps, &c. Bourguet, 1729 
 Ehrhart, 1735 Arduino, 1759 Ferber, 1773 Writers on 
 the erratics of Germany, Russia, &c. Guettard, 1762 
 Guldenstadt, 1768 Pallas, 1768 J. A. De Luc and his 
 explosion theory, 1779, &c. Silberschlag, 1780 De Luc the 
 younger Lariviere, 1829, and Dassen, 1835 De Saussure, 
 1779, &c. Dolomieu and the theory of fluvial erosion, 1793 
 Hutton, 1795 Play fair, 1802 Ebel, 1808 Conybeare The 
 fluvial erosion theory answered by De Luc Sir James Hall 
 and Charpentier The views of Mierotto, 1790 De la Metherie, 
 1795 Bertrand, 1797 Aikin, 1797 De Lasterye, 1798-9 . 1 
 
 CHAPTER II. 
 
 THE CHAMPIONS OF WATER. II. 
 
 Playfair, 1802 J.A. De Luc, 1809-1813 Von Buch, 1802-1818 
 Hausmann, 1807 Landt, 1810 Brochant de Villiers, 1819 
 Charpentier, 1841 Sir James Hall, 1812 Allen, 1812 Drake, 
 1815 Weaver, 1818 Imrie, 1818 Mitchell, 1818 Greenough, 
 1819 Escher Von der Linth, 1819 Fox Strangways, 
 1819-1821 Razoumofski, 1819 Henslow, 1820 Horace H. 
 Hayden, 1820 Bigsby, 1821-1823 Hitchcock, 1822 Grainger, 
 1822" N," 1823 Dr. Buckland, 1823 39 
 
 CHAPTER III. 
 
 THE CHAMPIONS OF WATER. III. 
 
 Chabrier, 1823 Hibbert, 1825 De Lasterye, 1826 Phillips, 
 1826-29 Sedgwick, 1826-31 Taylor, 1827 Jamieson, 1826 
 
xx vi Contents. 
 
 PAGK 
 
 "A," 1827 Hildreth, 1827 De Kay, 1828 Omalius 
 D'Halloy, 1828 Brongniart, 1828 Hausmann, 1829 Elie de 
 Beaumont, 1829 Pusch, 1830 De la Beche, 1831 Trimmer, 
 1831 Ware, 1831 Thompson, 1830 Lapham, 1832 Wood- 
 ward, 1833 Kloden, 1834 Strickland, 1 835-37 -Murchison, 
 1836 Home, 1835 Milne, 1837 Lartet, 1838 Sefstrom, 
 1838 Forchhammer, 1838 Jackson, 1838 Shepard, 1838 
 Hayes, 1839 Durocher, 1840 . 83 
 
 CHAPTEE IY. 
 
 THE CHAMPIONS OP ICEBERGS. 
 
 Tilas, 1470-2 Bergmann, 1769 St. Pierre, 1784 Dolomieu, 1793 
 Mierotto, 1790 Wrede, 1794 Yon Buch, 1808-1811 Bake- 
 well, 1815 Sir James Hall, 1815 Yenturi, 1817-18 Dick, 
 1819 Horace H. Hayden, 1820 Bigsby, 1821 Eaton, id. ' 
 Maclure, 1822 Hitchcock id. Dobson, 1825 Fischer, 1828 
 Lariviere, 1829 Kloden, id. Lyell, 1830-1845-De la Beche, 
 1833 Murchison, 1835-1841 BayHeld, 1837 Cordier, 1837 
 Darwin, 1839-1841 Wissman, 1840 Hays, 1843 . . .124 
 
 CHAPTEE Y. 
 
 THE CHAMPIONS OP GLACIERS. 
 
 Playfair, 1802-1816 Perraudin, 1815 Yenetz, 1822 Esmark, 
 1826 Gothe, 1829 Charpentier, 1835, etc. Agassiz, 1837, 
 etc. Le Blanc, 1838, etc. Eenoir, 1839-1841 Hogard, 1838 
 Studer, 1839 Conrad, 1839 Buckland, 1840 Murchison, 
 id. Whewell, id. Lyell, id. Fargeaud, 1839 Durocher, 1841 
 Boubee, 1842 Max Braun, 1843 . . . . . .162 
 
 CHAPTEE YI. 
 
 THE GROWTH AND CULMINATION OF THE GLACIAL NIGHTMARE. 
 EUROPE. 
 
 Bernhardi, 1832 Schimper, 1835-6 Agassiz, 1837-1843 Char- 
 pentier, 1842 C. Martins, 1846 E. Chambers, 1850-52 
 Murchison, 1853 Yon Post, 1856 Kjerulf, 1876, etc. 
 Erdmann, 1866 J. Geikie, 1873 Torell, 1875 Penck, 1875 
 Belgrand, Collomb, and Tardy, 1870 Eamsay, 1862 
 Jamieson, 1862-65 A. Geikie, 1863, Tiddeman, 1872 J. 
 Geikie, 1871 Croll, 1875 209 
 
 CHAPTEE YII. 
 
 THE GROWTH AND CULMINATION OF THE GLACIAL NIGHTMARE. 
 AMERICA. 
 
 North America: Agassiz, 1846-1850, etc., Dana, Winchell, jSTew- 
 berry, Morris Ice Cap in the Northern Hemisphere: Agassiz 
 and Dana Southern Ice Cap : Agassiz and Haast Tropical 
 America : Hartt, St. John, Belt, A. Wallace . ; . . 249 
 
Contents. xxvii 
 
 CHAPTER VIII. 
 
 ON THE ALLEGED RECURRENCE OF GLACIAL EPOCHS, AND ON SUPPOSED 
 INTER-GLACIAL BEDS. 
 
 PAGE 
 
 The Champions of former glacial periods Pre-Cambrian, Hicks, 
 Sir A. Geikie Cambrian, James Thompson Silurian, 
 Carrick-Moore, Lyell, J. Geikie, Dawson Devonian, Gumming, 
 Ramsay, Etheridge, Selwyn, J. Greikie Carboniferous, J. 
 Geikie, Haughton, G. Austen, Stirrup, Weiss, Orton, Dawson, 
 W. T. Blandford, T. Hughes, Waagen, Oldham Permian, 
 Ramsay, Hall, Sutherland, Stow Trias, Godwin Austen, 
 Ramsay, Wilkinson Oolite, Ramsay, Judd Cretaceous, 
 Godwin Austen, Escher von der Linth Eocene, Croll, C. 
 Martins, Lydekker Miocene, Gastaldi, Ramsay. Geikie, 0. 
 Martins, Stuart Menteath, Tardy Pliocene, Desor and 
 Stopanni Champions of inter-glacial beds, Venetz, Agassiz 
 Muhlberg, Bach, Tarancelli, Morlot, Chavenne, Scipion Gras, 
 Heer, Sir A. Geikie, Bennie, Searles Wood and Geikie, Penck, 
 Tornebohm, McGee, Whittlesey, Orton, Winchell . . . 280 
 
 CHAPTER IX. 
 
 APPEALS TO TRANSCENDENTAL PHYSICS AND ASTRONOMY. 
 
 Appeal of Geology to Physics and Astronomy-decrease in sub- 
 terranean heat : Lapperent, Laplace, Poisson, Hopkins, 
 Thompson, Phillips, Croll-^Different effects of subterranean 
 heat on land and sea: Frankland, Dawson, Croll Renoir's 
 theory ^variation in the sun's heat : Vicaire, Balfour Stewart, 
 Renoir Hotter and colder regions of space : Poisson and 
 Hopkins Variation in the transparency of space : Erman and 
 Babinet Variation in the transparency of the air Increased 
 cold alone will not produce a glacial epoch : Lecoq, Tyndall, 
 Frankland, Croll, Woeikof Shrinkage of J^e sun's disc : 
 Blandet, D'ArcEIac, Lapperent, S. Reinach Shifting of the 
 earth's axis : I. If the earth is rigid : Alessandro degli Alessan- 
 dri, E. Hill, Jukes, Thompson, Airy, G. Darwin, Croll, J. 
 Evans, Twisden II. If the earth contains a molten centre : J. 
 Evans, Thompson, G. Darwin, Airy, E. Hill Geological 
 witness against a shifting axis : Jukes, Haughton, Hutton, J. 
 Geikie, Boucheporn, Falsan Variation in the eccentricity of the 
 earth's orbit : I. Effect on the mean temperature of the earth : 
 Kepler, Croll and Carrick, Herschell, Haughton, Meech and 
 "Croft II. Its effects on the seasons : ]epler, Reclus, Lambert, 
 Meech, Croll, Arago, J. Geikie Variation in the obliquity of 
 the ecliptic": Herschell, Laplace, Stockwell, Ball, Drayson, 
 Belt, Meech, Croll, G. Darwin Effects of precession" and 
 nutation : Airy, Burr, Reclus, Meech, Humboldt, Price 
 Oeneral conclusion Croll, J. Geikie and Meech . . . 325 
 
LIST OF ERRATA. 
 
 Page 19, 
 
 21, 
 
 ., 42, 
 
 45, 
 
 ,, 47, 
 
 58, 
 
 64, 
 
 i. 81, 
 
 83, 
 
 ., 8*, 
 
 i. 87, 
 
 HI, 
 
 117, 
 
 118, 
 
 " 120, 
 
 122, 
 
 127, 
 
 ., 151, 
 
 17S, 
 
 192, 
 217, 
 
 297, 
 300, 
 309, 
 316, 
 319, 
 320, 
 339, 
 384, 
 401, 
 423, 
 
 432, 
 
 ,. 433, 
 
 , 461, 
 
 462, 
 
 494. 
 
 495, 
 
 line 37, insert and before La Sarra. 
 
 ,, 19, /or tois^s read toises. 
 
 14, for geog. read geol. 
 
 ,, 39, after ante insert 14. 
 
 13, /or St. read a. 
 
 ,, 11, erase comma after cover- 
 ing. 
 
 ,, 19, for is read an. 
 
 ,, 17, for Betws read Bettws. 
 
 24, insert with o/ter intervene. 
 
 ,, 22, for to read from. 
 
 ,, 4, insert and a/fer north. 
 
 ,, 25, /or Gimone read Gimont. 
 
 ,. 4, insert hyphens between por- 
 phyry like awd rock. 
 
 ,, 16, for state read State. 
 
 , , 26, for Klaadu read Koaadn. 
 
 ,, 19, for les read le. 
 
 ,, 24, for Silicia read Silesia. 
 
 ,, 22, insert of after south. 
 
 30, for Sedgewick read Sedg- 
 wick. 
 
 ,, 24, after he insert concluded 
 that. 
 
 ,, 4, for Hills. While read Hills, 
 while. 
 
 ,, 25, for glacial read glacier. 
 
 ,, 3, for glace read glace. 
 
 ,, 36, for deposited read caused. 
 
 21, insert and after Norway. 
 
 ,, 38, erase comma after sur- 
 face. 
 
 1 8, for Hall read Hull. 
 
 ,, 37, for Mantel read Mantell. 
 
 ., 25, for Ischer read Esc Her. 
 
 ,, 9, for whole read while. 
 
 31, for Faxo read Faxo. 
 
 ,, 2", for B read A.. 
 
 28, for 239 F read 239 F. 
 
 24, insert that is after and. 
 
 ,, 11, erase that. 
 
 3, erase comma after and. 
 
 ,, 12, for an astronomical read a 
 meteorological. 
 
 ,, 25, for Somervell read Somer- 
 vail. 
 
 20, 21, erase in the arctic re- 
 gions. 
 
 ,, 2 4, for their read these. 
 
 ,, 18, for have left read bear. 
 
 ,, 24, for north read south. 
 
 37, for is read are. 
 
 ,, 0, for Neu read Neue. 
 
 ,, 4, for Innspruck read Inns- 
 bruck. 
 
 ,, 18, 21, erase the inverted com- 
 mas before In and insert 
 them before variegated. 
 
 ,, 10, for Gereas read Geraes. 
 
 Page 498, line 39, for Schoaaberg read 
 Schomburgk. 
 
 516, ,, 36, insert the account of after 
 in. 
 
 ,, 529, 19, /or casue read cause. 
 
 ,, 549, 7, 8, erase the words from ThQ 
 in line 7 to that in line 8, 
 awd instead o/the scratch- 
 ing insert The in line 8. 
 
 ,) > >, 14, for it read a glacier. 
 
 ,, 551, 37, for stripes read strips. 
 
 556, 20, for Feriet read Ferret. 
 
 ,, 660, 8, for immediate.read inter- 
 mediate. 
 
 ,> >, M 15, /or glaciers read crevas- 
 
 562, ,, 3, for state-of the crevasses 
 
 read curves. 
 605, ,, 24, insert Memoirs of before- 
 
 U.S. 
 
 610, ,, 18, for Bell read Ball. 
 612, ,. 1, insert in after rule. 
 626, ,, 10, for its read the glacier's. 
 628, 4, for Dranse read Drance. 
 660, 32, for Moteratsch read Mor- 
 
 teratsch. 
 
 662, 18, for caused read causes. 
 
 663, 12, for inseparable read in- 
 
 superable. 
 
 671, 21, for luxurious read luxuri- 
 ant. 
 
 677, 34, for deep read steep. 
 
 681, 8, for green read granite. 
 ,, 18, for close read else. 
 ,, 30, for erratic read erratics. 
 
 684, 32, for it read them. 
 
 698, 15, for asas read asar. 
 
 739, ,, 43, for Id.- insert Q. J. G. S. 
 
 757, ,, 6, for St. read South. . 
 
 776, 27, for arar read asar. 
 
 779, ,, 7, for burst read bursts. 
 
 783, ,, 10, insert earth after brick. 
 ,, ,, 11, for earth read land. 
 
 786, ,, 37, for were read are. 
 
 832, 18, for Bell read Belt. 
 
 846, 30, for , as rad of. 
 
 848, 21, erase comma after Lowes- 
 toft. 
 
 856, 25, for Holmforth read Holm- 
 
 firth. 
 
 857, 35, for Kommeredge read 
 
 Kimmeridge. 
 860, ,, 37, for were read was. 
 
 864, 40, insert of the chan after 
 
 depth. 
 
 865, ,, 7, insert comma after soli- 
 
 tary. 
 
 887, 30, for Schomburgh read 
 Schomburgk. 
 
THE GLACIAL NIGHTMAEE. 
 
 CHAPTER I. 
 
 TJIE PHENOMENA OP THE DRIFT. 
 
 The Champions of Water (Part I.). Writers of the Eighteenth 
 
 Century. 
 
 " Frangitur enim pccurrentibus saxis, et per angusta eluctatus ubi- 
 cumque vincit, aut vincitur, fluctuat." SENECA, " Natur. Quass," lib. iv. 
 ch. 2. 
 
 Abundance of boulders in North Europe Some remarkable boulders 
 Views of the unscientific about them The Swedish writers 
 Swedenborg, 1719-22 Tilas, 1740-42 Linnaeus, 1744 Bergmann, 
 1769 Writers on the Alps, &c. Bourguet, 1729 Ehrhart, 1735 
 Arduino, 1759 Ferber, 1773 Writers on the erratics of Germanv, 
 Kussia, &c. Guettard, 1762 Guldenstadt, 1768 Pallas, 1768 
 J. A. De Luc and his explosion theory, 1779, &c. Silberschlag, 
 1780 De Luc the younger Lariviere, 1829, and Dassen, 1835 De 
 Saussure, 1779, &c. Dolomieu and the theory of fluvial erosion, 
 1793 Hutton, 1795 Playf air, 1802 Ebel, 1808 Conybeare The 
 fluvial erosion theory answered by De Luc Sir James Hall and 
 Charpentier The views of Mierotto, 1790 De la Metherie, 1795 
 Bertrand, 1797 Aikin, 1797 De Lasterye, 1798-9. 
 
 EVERY one who has travelled in the North of Europe, in 
 Sweden, Finland or North Kussia, must have had his atten- 
 tion arrested by the innumerable boulders and great rounded 
 masses of stone which strew the whole country. 
 
 In Wiaxall's Travels in the Baltic countries we find him 
 stating, in a letter written from Wyborg in June, 1 774, "that he 
 had seen no part of Sweden, except East Gothland, so free from 
 those vast stones which Nature in her anger scattered over 
 these kingdoms," and in describing the neighbourhood of 
 
2" s v The Glacial Nightmare. 
 
 Louisa, in Finland, he says : " It is difficult to depicture the 
 aspect of the country in which it stands. For the space of at 
 I east a league before I arrived there, and for more than two 
 leagues after I quitted it, the earth may almost be said to have 
 disappeared from my view, so completely was it covered with 
 stones ; or rather rocks : for many of them, from their magni- 
 tude, may well merit that appellation. It seems as if they had 
 fallen from the sky ; and Ovid, had he been acquainted with 
 this portion of the globe, might have here placed ' the Campi 
 Phlegraei ' where Jupiter overcame the Titans. The road, 
 compelled to respect these formidable impediments, performs a 
 thousand tortuous evolutions, in order to avoid them, and 
 serpentines beautifully for many miles. Neither cultivation 
 nor population can possibly exist among such a wilderness of 
 stones." 
 
 De Luc, speaking of North Germany, says " the blocks are 
 there accumulated in such numbers that it might be supposed 
 they had been brought together for the purpose of building 
 some large town." 
 
 Dr. Clarke says of these ruinous-looking landscapes : " The 
 soil, if it may bear the name of soil, is altogether incorrigible : 
 it consists of enormous loose fragments of bare granite, piled 
 together till they become mountains, and form steep precipices. 
 Upon these boulders there appears hardly a trace of any 
 vegetable earth or even of any kind of covering ; yet they are 
 thickly planted with forests of tall pines, birch and. juniper 
 trees which, in a marvellous manner, have found nourishment 
 for their roots in the interstices between the boulders/' * 
 
 These great stones, shaggy with grey mosses and weather- 
 beaten, the ruins of a world that has passed away, are a per- 
 petual source of glamour and of romance as we drive up 
 and down the forest roads of the north in the twilight, and 
 they form a fitting background to the adventures of giants and 
 dwarfs with which the imagination of the old-world folk peopled 
 them. 
 
 Nor are such stones confined to the North. As we move 
 further south, if they become more unfrequent, their strange 
 shapes and unexpected positions are even more obtrusive, and 
 
 1 Travels, part 3, sect. 2, 95. 
 
Some Famous Boulders. 3 
 
 especially is this so when they consist of some kind of rock 
 foreign to the country. 
 
 In the cultivated valleys few now remain ; they are not 
 convenient neighbours to the plough and harrow, and have 
 been largely broken up to make the roads or used to build 
 walls. It is otherwise with the wastes, especially the upland 
 wastes and commons. 
 
 It will not be uninteresting to call attention to a few 
 specimens of these stones, especially remarkable for their size. 
 The most famous of all, perhaps, is that called Pierre-a-Bot, a 
 boulder of fine-grained granite, planted on the Jura moun- 
 tains near Neuchatel, and which must have come from 
 Martigny in the Valais. It is 50 feet long, 40 feet high and 
 20 wide, thus having a cubic area of 40,000 feet. To reach its 
 present home from its original one it had to travel 60 to 70 
 geographical miles, and to cross the lake of Neuchatel. 
 Another one, near Zurich, is called the Pflug stein, and can be 
 traced to the Alps of Grlarus, whence it had to cross Lake 
 Zurich. It rises to 60 feet above the ground, and its cubic area, 
 according to Heer, is 72,000 feet with a weight of 4500 tons. 
 The Pierre des Marmottes, a granite boulder near Monthey in 
 the lower Valais, can be traced to the valley of Ferret, a distance 
 of 30 to 40 miles, and measures about 60 feet in length, 30 in 
 height and 33 in width, and has a cubic area of about 55,000 
 feet. The Pierre-du-tresor, a granite boulder near Orsieres, 
 measures over 90,000 cubic feet. Bloe-Monstre, near Devent, 
 a boulder of limestone from the valley of Avanzon, measures 
 about 55 feet in length, 43 in width and 60 in height,, 
 with an area of about 130,000 cubic feet. 2 
 
 A famous boulder, called the stone of the Markgraf, on the 
 Rauhe Alp, weighed 560,000 Ibs., and out of it has been 
 carved a vase nearly 23 feet in diameter. 3 The gigantic basin 
 before Uie New Museum at Berlin was wrought out of one 
 half of an erratic block of Swedish red granite, the other 
 half of which is still lying near Fiirstenwalde. 4 
 
 A block in the diocese of Flade, in the island of Mors, 
 is about 23 feet long, 18 wide, and of unknown depth. 5 Forch- 
 hammer describes a great pyramidal block, buried in the 
 
 2 Braun, Recreations in Popular Science, 39, 40. 
 3 Klee, Le Deluge, 143. 4 Braun, op. cit. 5 Klee, op. cit. 143. 
 
 B 2 
 
4 The Glacial Nightmare. 
 
 ground, 111.5 feet in circumference and llf feet above 
 the surface, which was presented to the King of Denmark 
 by a proprietor at Hesselagergaard in Fionia. The king had 
 it uncovered to a further depth of lOf feet, but its base 
 was not reached, and it went on increasing in size. 6 
 
 Visitors to St. Petersburgh have seen the granite base of 
 the statue of Peter the Great there. This vast block was 
 brought all the way from Carelia. " Being placed/' says Dr. 
 Clarke, "on balls of brass fifteen inches in circumference, 
 which rolled on sledges over a causeway raised for the pur- 
 pose, it was moved every day, by four hundred men, with the 
 assistance of pulleys and a windlass, over a space of ground 
 equal to about half a mile. From the coast it was brought on 
 a raft of a peculiar construction, to the city. The original 
 size of the rock was 36 feet in length, 20 in height, and as 
 many in breadth, but it was much reduced in size when 
 shaping it." 7 
 
 De Luc describes another of these famous stones, which was 
 found near Eutin. He tells us how he was shown at the gate 
 of the palace gardens there, four pillars about ten feet high, of a 
 fine grey granite, with their pedestals of the same piece, which 
 had been hewn out of a single block. The walls of the kitchen 
 garden were entirely built up out of fragments of the same 
 block, and they formed a quadrangle, having a gate in the 
 middle of each side, between two pillars seven or eight feet 
 high. Close by was a temple, supported by six columns, which, 
 with the steps round the pavement of the temple and its 
 cupola, had all been taken from the same block, and notwith- 
 standing this only one half of the boulder had been used. 8 
 
 The largest of all boulders known to me, however, are in 
 England. 
 
 They are of chalk and lie in the cliffs near Cromer in Nor- 
 folk, embedded in, and surrounded by, clay and sand. One of 
 them, according to the measurements of Mr. Searles Wood, is 
 more than 800 feet Ipng and 60 feet high. 
 
 If we turn to America, we meet with similar phenomena. 
 '* The size of some of our boulders," says Professor Hitchcock, 
 "is very great. Those from 10 to 15 feet in diameter are 
 
 8 Id. 7 Clarke's Travels, part 3, sect. 2, 49G. 8 Travels, i. 270. 
 
Homely Explanations of the Boulders. 5 
 
 very common. Those from 20 to 25 feet are less common. I 
 have seen several, as at Gay Head, and Bradford, in Mas- 
 sachusetts, and in Winchester, in New Hampshire, fully 30 
 feet in diameter ; and one of conglomerate existed a few years 
 ago, but is now a good deal blasted away, at Fall River, in 
 Massachusetts, which must have weighed not less than 5400 
 tons, or 10,800,000 pounds. I also recently measured one in 
 Antrim, in New Hampshire, whose horizontal circumference 
 is 150 feet." 9 
 
 Let us now turn to the various explanations which have 
 been forthcoming to account for these boulders and the phe- 
 nomena associated with them. 
 
 Unsophisticated philosophers like shepherds and other 
 contemplative students of nature who live much in the open 
 country, have not unnaturally invoked uncanny causes to 
 account for the strange stones. The Devil, or Robin Hood, or 
 some famous giant, or some still more potent god like Thor or 
 Odin, have been supposed to have brought these strangers 
 either through freak or for some good reason. Robin Hood's 
 bed, the Fairies' Stone, the Devil's Quoit or the Devil's 
 Marble, are names which recur in the mountainous dis- 
 tricts of England. To mention one famous case ; does not, 
 or ought not, everyone to know that the great stones on 
 Salisbury plain were brought thither by the magician Merlin 
 and ranged in that great double circle at Stonehenge to 
 afford speculation to antiquaries for all time ? while the very 
 rocks which the gods threw down upon the Titans, and with 
 which they crushed them, still remain in Greece to testify to 
 the story. 
 
 Professor Geikie, speaking of Scotland, says, fc Numerous 
 are the myths and legends connected with the great boulders 
 of our country. They are ( the giant's putting stones,' ' the 
 deil's burdens/ ' the witches' hearth stones ' of the fanciful 
 peasantry. Zealous antiquaries have occasionally claimed 
 them as monuments set up by man in some long-forgotten 
 age." 10 M. Falsan tells us how the strange outlines of many 
 of these erratic blocks (as De La Beche named them), have 
 arrested popular attention. In the midst of the white lime- 
 
 9 Phenomena of the Drift, 166. 10 Prehistoric Europe, 171. 
 
6 The Glacial Nightmare. 
 
 stones of Bugey, their deeper colour has especially aroused 
 notice. Hence they are called pierres Uses (bis or gris) ; 
 they are also called " blue stones/' " salt stones," " stones 
 which do not yield lime/' In the plains of Northern 
 Germany, as Credner tells us, they are called " Findlings 
 blocke " (i.e. foundlings). In other places their names recall 
 other notions, as Pierre du Bon Dieu ; Pierre du Diable, Pierre 
 de Samson ; Galet de Gargantua ; Pierre du Manage, etc. ; and 
 M. Ed. Piette, who has studied them from the point of view of 
 archaeology on the mountain of Espiaup in the valley of Oo, 
 in the Pyrenees, has shown that they have been made the 
 objects of a kind of religious cult, combining a number of 
 superstitions and customs, some bizarre and some licentious, 
 which are gradually dying out as education spreads. 1 
 
 The first scientific person, so far as I know, who referred to 
 these moss-grown wanderers, was that very remarkable man 
 Emanuel Swedenborg. The latter part of his life was so 
 entirely devoted to the mystical and visionary schemes which 
 culminated in the foundation of the Church of New Jerusalem, 
 that it will be news to many people to learn that in his earlier 
 days he was an acute and distinguished man of science, a 
 member of the Academies of Upsala, Stockholm and Peters- 
 burg, and that he filled the post of assessor of the School of 
 Mines in Sweden, and this at a time when Sweden was the 
 nursing mother of modern Natural History. 
 
 It was this official position that doubtless brought him 
 into contact with the geological aspects of the surface deposits 
 of the North, which he was the first to publish and to treat 
 seriously. Inter alia, he not only refers to the boulders which 
 are so conspicuous in Sweden, but he connects them with 
 other phenomena, such as the long whale-backed ridges of 
 sand and stones known to the Swedes as asar, and he assigns 
 a cause for both. 
 
 He describes these ridges as being very conspicuous in the 
 provinces of Westmannland, Dalekarlia and Upland. They 
 are occasionally of great size and as much as 150 Swedish 
 ells in height. A Swedish ell being one English foot and 
 li^g- inches. Some of them, he says, are formed of sand 
 
 1 Falsan, La Periode G-laciaire, 71, 72. 
 
Swedenborg. 7 
 
 intermingled with pebbles, the latter being rounded and 
 polished as if turned on a lathe. 
 
 Others of them, he remarks, are made up of pebbles only, 
 without sand, all most curiously turned. Many again seem 
 formed of enormous stones like fragments of mountains heaped 
 together. These last ridges, he says, consist entirely from 
 top to bottom of large fragments of stone, each probably 
 weighing from ten to forty Swedish maritime pounds 
 (skeffund), and he tells us he had often noticed them several 
 miles off, although they were not more than thirty to fifty 
 paces wide. Their sides are generally precipitous. The 
 open country is further strewn with vast boulders, like 
 ruins of mountains, even where there is not a hill in the 
 neighbourhood. They are found in the plains, at the bottom 
 of lakes, and even on the very summits of mountains. The 
 provinces of Helsingland and Orebro are full of them, spread 
 like buildings over the level country. He further noticed 
 that the ridges run north and south, that they have shelv- 
 ing sides whose slope depends on the materials of which they 
 are formed, those made of sand not being so steep as those 
 made of stones ; that the stones in question have been 
 rubbed and polished as if triturated by the sea, and that they 
 exist chiefly on the highest ground, as in the province of 
 Orebro. 
 
 Swedenborg argues that these ridges can be accounted 
 for only by invoking some very deep or universal diluvian 
 ocean, the former existence of which, he contends, is 
 proved by the mixture of substances of different kinds, 
 sand, clay, smooth pebbles, large stones and masses of 
 rock. Their slope proves that they were thrown up by 
 the sea into great accumulations, and so formed into 
 lengthened ridges, with sloping sides. The fact that these 
 ridges run north and south, he says, shows that the same winds 
 prevailed in the diluvian age as now, and it follows that the 
 ocean referred to must have been universal and stood at a 
 great height above the land, and that it had no limits, shores, or 
 straits to warp the winds ; and he appeals to the principles of 
 hydraulics to prove that the sea could have done this kind 
 of work. Its pressure depends on its depth and is more con- 
 siderable at greater than at lesser depths ; and as the bottom 
 
8 The Glacial Nightmare. 
 
 of the sea was then covered with different kinds of stones, sand 
 and clay, it is not wonderful that very deep waves should have 
 had sufficient power to tear enormous stones from their 
 foundations, to strew them about in heaps, pile them up in 
 mountains, and thus form ridges or cover whole provinces with 
 them. He explains the existence of so many stones on very 
 high ground on the principle that if the sea had sufficient 
 power to carry them to the highest parts of the district 
 they would stop there and travel no further, since the 
 force of the water would diminish with the depth. This is 
 shown by its frequently, when very deep, tearing up masses 
 of stone and overthrowing walls of masonry ; dykes, though 
 built of stone, being overthrown when the water rises 
 three or four ells higher : and he shows that just as the 
 wind, when it becomes a hurricane, can blow down trees 
 and houses, so can an ocean, when much agitated and in 
 gales of wind, alter the configuration of great sandbanks 
 and shoals. He also urges that this would be assisted by the 
 fact that although stone, when in the air, is heavier than 
 water, as 2 1 to 1, it is only 1 J to 1 when immersed in fresh water, 
 and still less in salt. On these grounds he concludes that the 
 phenomena of the boulders and the cisar is to be explained by 
 the operation of the diluvial waters. 
 
 Swedenborg's views, however crude in some respects, were 
 far in advance of his time. They were first published in the 
 years 1719-22, 2 
 
 Some years later, namely in 1740 and 1742, Daniel Tilas 
 drew attention to the great granite blocks existing in Sweden, 
 some distance from their parent mountains. These stones he 
 describes as of various kinds, some as big as houses or even 
 small mountains, and others quite small. He remarks that 
 the stones which have travelled the shortest distance are less 
 rubbed and have their angles sharper than the others. He 
 attributes their distribution to a great flood, adding that the 
 
 2 In 1719 he published a small memoir, entitled "Om vattrens 
 hojd och forra verldens starka ebb och flod. Bevis utur Suerige." 
 In 1721 he published a paper in the "Acta Literaria Suecia," i. 
 192-196, and in 1722 issued his "Miscellanea observata circa res 
 Naturales," Leipzig. These two last papers were translated in 1847 
 by Mr. C. E. Strutt, London. 
 
Tilas and Linnaus. 9 
 
 larger ones were possibly carried by floating ice which came 
 from the Pole, while all would be assisted by the specific 
 gravity of the stones being altered when immersed in water. 
 He says that stones in a rapid current do not sink to the 
 bottom but are forced to the sides. He mentions the rounded 
 and smooth outlines of the hills as evidence that they have 
 been planed down. He had remarked, he says, how much 
 more numerous the stones are on the high ground than 
 in the low valleys, diminishing in size as we get further 
 away from their source ; and that this is natural, since they 
 would be arrested in their progress by the ground ; and 
 he also noticed that on the low ground they are frequently 
 covered with clay, sand, etc. He tells us they are often 
 found tumbled together in heaps, sometimes a large stone 
 lying on smaller ones, and that they often occur on the rounded 
 and polished surfaces of the mountain tops. He also men- 
 tions the absence of these stones from the Russian and 
 Siberian steppes, and their being so much more common in 
 some districts than others. Thus he had seen hardly any 
 when travelling along the Ladoga canal and the Sweti rivei 
 towards the Onega lake, while in the Carelian uplands 
 between Kontzoffer and Petrof skoi they abound. Contrasting 
 their abundance further north with their infrequency as 
 we go south, he says the people of Scania have difficulty 
 in finding enough stones for their houses, while the Dutch 
 would freely exchange gold for the granite boulders of Central 
 Sweden. To the great flood which he invokes he also, like 
 Swedenborg, attributes the long sand ridges so conspicuous 
 in the north, which often traverse, he says, whole parishes and 
 even provinces. He also reports how these deposits some- 
 times occur as collections of mounds of sand and clay, which 
 he noticed consist of heterogeneous materials tumbled to- 
 gether and unstratified. 
 
 We next, namely in 1744, find Linnasus, whose quick eye 
 and keen judgment did so much for natural science, writing : 
 " There is, about forty miles from Hoburg, in Gothland, 
 an immense micaceous stone, heavier than any which 
 human power could move. It is not in its natal place, 
 for no such stone is found there, whence it may be con- 
 cluded that the sea brought it hither from Sweden or 
 
io The Glacial Nightmare. 
 
 Muscovy, when as yet Gothland was submerged under the 
 sea." 3 
 
 Erik Pontoppidan, whose knowledge of Scandinavia was so 
 profound, refers to the erratic blocks, and tells us the largest 
 known to him in Denmark was one near Herselagen in Fuhnen, 
 fifty-four ells in circumference, twenty-two ells long, and 
 fifteen wide. He also refers to the so-called Swan stone in 
 Moen, which is twenty-two ells in circumference. 4 
 
 T. Bergmann, writing in 1769, refers to the loose stones 
 scattered over Sweden, and says they cannot well be attributed 
 to earthquakes, since earthquakes do not occur there ; nor 
 again, he says, can we attribute them to the action of ice, since 
 while ice will no doubt lift up stones and deposit them on 
 higher ground, it could not have raised them to the tops of 
 the mountains where they are now found. He accordingly 
 argues that they were moved by water, and gives instances 
 to show that Scandinavia is rising from the sea at this 
 moment, a conclusion which has since been amply verified. 
 
 Elsewhere he mentions the so-called giant's cauldrons, and so 
 far as I know he was the first to try and explain them. 
 
 He refers them to the working of stones arrested by some 
 impediment and then made to turn round and round by water 
 until the holes have been bored. He mentions six of them 
 as occurring at Kokare Fiaiden in Finland : one of them, still 
 under water and a foot deep, was only thirty years old, and its 
 actual growth had been watched by Nordenskiold, and he 
 urges that similar ones elsewhere, out of the reach of 
 present streams, had the same origin. He mentions others as 
 existing in Jernt eland, in West Gothland, in Bohuslan and 
 in America under Cohoes Fall on the Hudson, and elsewhere 
 he remarks that they are always situated on the slopes of 
 mountains, where water may have flowed, and never on their 
 summits. They are sometimes two or three fathoms deep, 
 round or oval, always smooth and occasionally the stones which 
 made them still remain in them. 5 
 
 It will thus be seen that although these researches have 
 been almost entirely overlooked, the main features of the drift 
 
 3 De Telluris habitabilis incremento, sect. 39. 
 
 4 Kurzgefasste Nachrichten, etc., 1765. p. 52. 
 * Phys. Besch. der Erdkugel, 389-390, 403-404. 
 
Bourguet and Ehrhart. i T, 
 
 phenomena of the North, namely the scattering and distribu- 
 tion of the erratics, their occurring on strata of a different 
 character, and their connection with the mounds and asar, 
 the possible intervention of floating ice to assist the diluvial 
 waters in moving them, and the polishing of the rock surfaces, 
 had been recognized in the earlier part of the last century. 
 
 Meanwhile, similar observations had been made elsewhere, 
 and we will first turn to the Alpine and neighbouring lands. 
 
 In 1729 L. Bourguet, a native of Neuchatel, in his " Memoire 
 sur la Theorie de la Terre," speaking of the district of the Jura, 
 says that on mountains, hills and plains, there occur irregular 
 masses of granite, and of another kind of rock, very hard, and 
 composed of talcous scales, which are not attached to the solid 
 ground, nor have they tumbled down from a neighbouring 
 height, there being none such at hand ; and besides, they are 
 of an entirely different kind of stone to that found close by. 
 Some of these blocks are of an immense size, and their number 
 is so enormous in some places that they look as if they had 
 tumbled from the clouds. 6 
 
 In a letter, dated Meiningen, 10th May, 1735, and addressed 
 to Mr. C. Mortimer, the Secretary of the Royal Society, which 
 has been much neglected, Dr. Ehrhart calls attention to 
 certain observations he had made in the Tyrol. In this 
 memoir he tells us how at Meiningen there is on the moun- 
 tains a vast congeries of rounded stones which have evidently 
 been rolled by water, but in asmuch as no river reaches or could 
 ever reach them, it must have been a powerful torrent. He 
 goes on to say that these stones increase in size from Meiningen 
 as we near the Alps till they become three and four feet in 
 diameter ; in the opposite direction they gradually decrease to 
 the size of coarse sand. He further says that he had noticed 
 among the Tyrolese Alps stones in situ in a continuous layer 
 which were just like the scattered and broken stones occurring 
 between the Alps and the Danube. These scattered stones, he 
 says, are found of as many varieties as the rocks of the Alps. 
 The cause which broke the Alpine rocks, and rolled the frag- 
 ments into these rounded shapes, and distributed them over 
 all that part of Alemannia where he dwelt, he did not doubt 
 was the Noachian deluge. The waves of this flood, which he 
 
 6 L. Bourguet, Lettres Philosophiques, Amsterdam, 1729, pp. 198-9. 
 
1 2 The Glacial Nightmare. 
 
 says must have been twenty leagues in length and breadth, 
 must have flowed from the south, northwards. The fragments 
 which had been broken from the Alps (which were higher 
 before the Flood) were triturated and rubbed until they 
 acquired their present shape and size, whence those which are 
 nearest to the Alps are the largest and most intact, and they 
 diminish as we get away. The identity of the scattered 
 stones and the Alpine masses is unmistakable. He speaks 
 further of the Alpine erratics being scattered over Swabia as 
 well as Alemannia. 7 
 
 Our next authority is Arduino. 
 
 In 1759 we find that Italian geologist writing that among the 
 Tuscan Hills were strewn a large number of stones, some of 
 them foreign to the district, and which had come a long dis- 
 tance. Those, he says, which were derived from neighbouring 
 mountains, had still sharp edges, while the others were 
 rounded. 8 
 
 It will be seen from these extracts that the phenomena of 
 the drift had now been observed in the south as well as the 
 north of Europe. So far as I know, the only theory which 
 had, up to this time, been forthcoming to account for them, 
 was an appeal to the Noachian deluge. 
 
 Meanwhile, a revolt against this Diluvial Theory had com- 
 menced, and was largely fostered, if not created, by Voltaire, 
 in whose attacks upon the Bible it was found necessary to 
 discredit in every way the evidence of a general flood ; nor can 
 his efforts, although supported by some strange and fantastic 
 arguments, be considered as altogether fruitless. They 
 largely led to the discriminating of different strata, represent- 
 ing different epochs and different causes, and to the proof that 
 the great mass of fossil shells which had hitherto been traced 
 to the Deluge, must have had a very different origin, and 
 men's minds began to turn to more inductive methods of 
 research, and to the study of causes actually in operation 
 rather than to transcendental ones. It would seem that 
 among some of these students an idea prevailed that the 
 
 7 Philosophical Transactions, vol. xli. 549-551. 
 
 8 Giov. Arduino, Lettera seconda al Sig. Cav. Ant. Yallisneri, 
 written from Vicenza, March 30th, 1759, and published in Nuova Eacolta 
 d' Opusculi, etc., del P. Abb. Calogera, vi., clxv. and clxvi. 
 
Ferber. 1 7 
 
 \j 
 
 great blocks of stone to which we have referred were nothing 
 more than the ordinary debris of flooded rivers. This view 
 soon met with acute critics. 
 
 Ferber, in his u Letters on Italy/' published at Prague in 
 1773, says of the hill country about Venice : " The fields near 
 Gallio, Asiago, Campo di Rovere, and other places belonging to 
 the Sette Communi, are situated in the mountains much above 
 the level of the sea, and are covered with a great many detached 
 pieces of granite, quartz and other vein rocks, which appear to 
 be forced from, and belong to, the Tyrolian primitive moun- 
 tains. These ruins appear at the same horizontal elevation in 
 several other places of the Alps, as inFeltrino (Stato Veneto), 
 separated by the Brenta from the former; westward on the 
 same Alps from Astico to the Adige Eiver, and so on. Their 
 quantity and variety of size is remarkable at Tonezza and near 
 Bulgaria in the hills under the Austrian jurisdiction ; which, 
 being entirely calcareous and entirely destitute of primitive 
 strata, prove the more that these forced and blunted rocks, 
 pebbles and sands are adventitious. 
 
 " This phenomena is more curious because the debris arc 
 precisely like those washed and carried down by the Adige 
 and the Brenta from the Tyrolese Mountains, and the level of 
 these rivers being -at present one thousand feet lower, it is im- 
 possible to conceive how, in their course so deeply cut into the 
 Alps, they could have rolled and deposited these stones at 
 such a high elevation. But this and similar observations 
 force us to conclude that their channels and beds have been, 
 in former times, at the same elevations which are marked at 
 present by the position of their deposits. No water can have 
 carried, rolled and forced them up into their actual position, 
 whatever allowance be made for its impulse, and whatever 
 violent inundations we should please to suppose. This is fair 
 evidence of the great ruptures in the Alps and the prodigious 
 effects of the unwearied and never-ending course of rivers, and 
 points also to still greater ones which may have been occasioned 
 by violent ruptures and inundations in such places where 
 earthquakes first stopped, and afterwards entirely changed, 
 the course of rivers/' * 
 
 Let us now turn to another part of Europe. Tho boulders of 
 
 1 Op. cit. Engl. Trans. 43, 41 
 
14 The Glacial Nightmare. 
 
 North Germany must have early attracted attention, and it 
 must have been well-known to travellers that they existed over 
 a wide area. I have not, however, met with a reference to 
 them by a scientific man before Guettard. 
 
 In two papers on the geology of Poland, published by him in 
 the Memoirs of the French Academy for 1762, the immense num- 
 ber of boulders of granite, etc., found in the level plains between 
 the Baltic and the Carpathians, is continually referred to. He 
 mentions how these boulders are not distributed evenly over 
 the whole country, but abound in some districts and are rare 
 in others. In some places granite prevails, in others, quartz 
 and other stones, and the boulders are used for paving most of 
 the villages. M. Guettard says the boulders especially abound 
 near Grodno, between Otrock and Osieck, between Soroez and 
 Lukartof, and in the cantons cf Niesrietz and Pinczovia in 
 Lithuania, and he describes in detail the character of the 
 various stones of which they are formed. He attributes their 
 origin to the destruction in situ of high mountains chiefly 
 formed of granite. " The sand in which the boulders occur," 
 he adds, " is more like that which is formed by the decom- 
 position of granite than of sandstone rocks. 2 
 
 In 1768 Gulden stadt set out on his travels, the account of 
 which was afterwards written by Pallas. In this he calls at- 
 tention to the fact that in the neighbourhood of the city of 
 Twerza, in Russia, where the native rock is limestone, there 
 are found scattered in the bed of the Twerza and in the fields, 
 great stones, which are always rounded. Some of these are 
 of brown, others of green, and others of grey granite. There 
 also occurs red felspath. 3 Guldenstadt offers no explanation 
 of these stones. 
 
 In the same year Pallas himself commenced his famous 
 journey. In his account of it he tells us how a hill which 
 he saw at Bronitzoi Jam, not far from Old Novgorod, was 
 covered with boulders formed of a red quartz and black 
 blende. Many of them had been transported to Petersburg 
 for building purposes. The largest of these stones was three 
 toises in length. " Was it water," he says, " which rolled hither 
 these masses of stone ? We do not think it possible for 
 
 2 Op. cit. 234, 293, etc. * Op. cit. ii. 4GO._ 
 
J.A.DeLuc. 15 
 
 water to move such immense masses, and we think there must 
 have been some other * cause/ " 
 
 About this time another and more famous geologist devoted 
 himself to a critical study of the erratic blocks in many 
 countries, namely, J. A. De Luc. His long residence in 
 England makes him especially interesting to us, and he may 
 be looked upon as largely the founder of an English school of 
 geology. He also had an advantage over previous inquirers 
 in that his observations were made over a wide area, and he 
 consequently escaped some of those errors of induction which 
 are inevitable when a theory of the universe is based upon the 
 examination of a gravel pit or upon the traces of geological 
 revolution which are found in some small hamlet. If his theories 
 are sometimes fantastic they are easily separable from his ob- 
 servations, which are shrewd enough. It must be remembered 
 also that many of his observations were made in countries far 
 removed from high mountains and where the theories which 
 might avail to explain mountain phenomena are not tenable. 
 As early as 1779 he called attention to the granite blocks of 
 the Hartz, which he compared with the more numerous blocks 
 found on the Alps. 
 
 Having revisited the Hartz in 1786, his attention was 
 directed again to the blocks there, some of granite, others of 
 quartzose and other primordial rocks, but j^l foreign to the 
 beds on which they lie. He goes on to say that this pheno- 
 menon is a very general one and found on most continents. 4 
 
 His observations extended over the greater part of 
 Germany, and included the district of Luneburg, so fertile in 
 these stones. In one paper he refers to the fact that the granitic 
 part of the Hartz Mountains is the Blocksberg, against which 
 rests the Bruchberg, a schistose mountain separated from the 
 latter by a small hollow. On another side is the Rehberg, 
 which forms a lower branch of the Blocksberg. The Kehberg is 
 covered entirely with granite blocks, and it is, he says, of such 
 small extent and so little below the Blocksberg, that it is impos- 
 sible to conceive that these blocks could have descended from 
 thence. ' ' Whoever shall observe this scene of disorder, will 
 acknowledge that there is not any cause actually operating 
 
 4 Voyages, Fr. ed. i. 5-7. fc Lettres Physiques, etc., 337-3-10. 
 
1 6 The Glacial Nightmare. 
 
 which could, in any length of time, have produced such an 
 accumulation of these blocks/' Whereas the Eehberg is 1 
 covered with granite blocks, the Bruchberg, which is schistose 
 in its structure, is covered with blocks of a quartzose rock, no 
 stratum of which is to be found in any part of these mountains ; 
 and these blocks, as well as those of granite, are scattered also 
 on many mountains of calcareous stone, as well as in the plains, 
 and proceed from a class of strata whole mountains of which 
 are found in other countries, as in Yeteravia and in England. 
 The blocks of granite are traced upon the summits, on the de- 
 clivities, and at the bottom of the valleys of the schistose 
 ridges, as well as of the calcareous mountains which rest 
 against these last, and they abound in all the heaths of Lower 
 Saxony and Westphalia, on the hills, on the plains, and even 
 to the very coast of the sea. 6 
 
 He goes on to say that on the declivities of the calcareous 
 mountains of Hildesheim he had observed immense loose 
 and insulated masses of a calcareous stone resting on 
 strata totally different from the strata of the loftier parts 
 of those mountains ; and at the same time had found frag- 
 ments of granite, and other stones equally foreign to the 
 nature of these mountains, on their declivities and at the 
 bottom of their valleys, and that nothing is more common 
 than, on mountains, hills and plains, and within the loose 
 strata on the surface of the soil, to find blocks, or smaller 
 fragments and gravel, belonging to stony strata known to 
 exist in some mountains elsewhere, but nojie of which is found 
 in that country, either in eminences or under the soil of the 
 plains at any attainable depth. 
 
 I have quoted these few paragraphs from De Luc's earlier 
 works to show how methodical and clear his observations 
 were. I will now describe the theory which he formulated to 
 explain the erratic phenomenon. 
 
 His view was that the contour of the earth's surface was 
 due to subsidences rather than elevations, and that this 
 necessitates our postulating the existence of subterranean 
 hollows or caverns, which, he says, must have been filled with 
 some expansible fluid, perhaps with atmospheric air. The 
 subsidenca of masses of great magnitude must have caused 
 
 6 Id. 30-32. 
 
/. A. De Luc. 1 7 
 
 great pressure in these caverns and as the compressed fluids 
 could only escape through the intervals of the disunited 
 masses, they must have rushed out with great impetuosity. 
 "Here then/' he says, " is a force of which we may form a suffi- 
 cient idea without comparing it with that which projected the 
 planets in their orbits. It is the same in its nature, though 
 far superior in degree, with the force that raises lavas to the 
 summits of the volcanoes of the Andes and by which rocks of 
 an enormous size are projected some miles into the air, an 
 effect which was observed during the formation of Monte 
 Nuovo near Naples. . . . By this force, therefore, which nothing 
 that was detached and loose could possibly resist, the frag- 
 ments broken off by the collision of the lower parts of the 
 strata were thrown up to the surface, in the manner of bombs, 
 notwithstanding the resistance of the water, under which these 
 explosions took place/' He goes on to argue why so many of 
 the blocks are of granite and other similar substances consti- 
 tuting the inferior strata. He further urges that, as the 
 revolution in question took place at the bottom of the sea, 
 there would be a tremendous conflict between the expanding 
 fluids and the water which would account for the fragments 
 being ground against each other. The combined effect would 
 be to triturate the fragments and also to afford a cause 
 adequate to the dispersion of gravel or rounded stones over 
 the surface of our present land. 7 A somewhat similar view 
 of the subterranean origin of the blocks occurred, it seems,, 
 independently to another observer. 
 
 In 1780 J. B. Silberschlag published a work entitled 
 " Geogenie oder Erklarung der Mosaischen Erderschaffung." 
 In this he discusses at some length the great deposits of sand 
 and the erratics of North Germany, especially in the valley 
 of the Elbe. He calls attention to their variety and wide 
 distribution, and urges that they are the products of sub- 
 terranean volcanic eruptions, and he professes to have found 
 traces and remains of these volcanoes in North Germany, a 
 conclusion which De Luc criticizes. 
 
 De Luc maintained his own position in many polemics, 
 and it was supported with considerable vigour by his 
 
 7 Lettres Physiques, iv. 152, 153 ; Elements of Geology, 338-341. 
 
 C 
 
1 8 The Glacial Nightmare. 
 
 nephew, the younger De Luc, in a paper published in the 
 eighth volume of the "Annales des Mines," but in his 
 later writings the latter considerably modified his theory 
 and supplemented it by the then more generally accepted 
 views of De Saussure and Von Buch. 8 In 1829_,Engelspach- 
 Lariviere published a paper on erratic blocks in which he attri- 
 buted their dispersion partially to the same cause, and as late as 
 1835 a similar theory was advanced by Dassen, in a paper on 
 the gravels of the province of Drenthe, in which he contended 
 that the Scandinavian peninsula was elevated when the strata 
 was still soft, except only its exterior envelope, which was 
 broken, and its fragments were scattered through the air. 9 
 
 De Luc's views were at once sharply challenged. Thus De 
 Saussure, writing in 1779, says, " Who has seen an example 
 of such explosions which would shoot through the air blocks 
 several cubic toises in extent for a distance of 12 or 15 leagues ? 
 To cause such explosions the heat must have been very great, 
 and would have vitrified or calcined the stones, and such 
 volcanoes must have poured out lavas or other distinctly 
 volcanic products, none of which results are present/' l Else- 
 where he urged that rocks of such a size shot through the 
 air over such a distance as that from the Alps to the Jura, would 
 have broken the rocks on which they fell or buried themselves 
 in the ground, while they are found on the surface and often 
 touching it only at a few points. 2 Von Buch also criticized 
 the theory, and argued that if explosions had scattered the 
 blocks they would have been found scattered promiscuously 
 over the whole range, and would not be absent, as they 
 are, from those places where other heights intervene. 3 De 
 Saussure and Von Buch further showed that diligent search 
 can always find the beds in situ whence the blocks have 
 been detached, and that we need not seek for them in the 
 bowels of the earth, but on the surface. That the problem 
 therefore is limited to finding some means by which they 
 could have been transported. 
 
 De Luc's explosion theory, I need not say, commands no 
 
 8 Considerations sur les Blocs erratiques, 27. 
 
 9 D'Archiac Histoire des Progres, etc., 59, 60. 
 
 1 Voyages dans les Alpes, i. 150. 2 Id. i. 168. 
 
 3 Abhandhugen., Berlin Acad., 1804. 
 
De Saiissure. 19 
 
 support now. We will next turn to his great contemporary, 
 the father of Alpine exploration, and, like himself, a careful 
 observer, De Saussure. In the first volume of his " Voyages 
 dans les Alpes," published in 1779, in discussing the gravels 
 and superficial blocks, he begins by protesting against 
 those who treated the pebbles in gravel, as they treated the 
 bones found in the ground, as lusus naturse, and then goes 
 on to show that the rounded character of the pebbles is 
 not their original form, as in the case of concretions, etc., 
 but that they were clearly rounded by the action of water, 
 and he appeals to what may be seen in most mountain torrents, 
 where the process of making such pebbles is continually going 
 on, and more remarkably so in the torrents underlying the 
 glaciers, and on the shores of the ocean and of lakes. 
 
 He then goes on to contrast this rounded gravel with the 
 blocks found on the glacier backs and on the flanks of the 
 mountains bounding the glacier, where not a stone, he says, 
 is to be found which has not its angles sharp and its edges 
 unrounded. 
 
 He next urges that the greater part of the boulders and 
 blocks found near Geneva are of granite, gneiss, etc., like the 
 Alpine masses, while the beds on which they lie are formed 
 of limestone, sandstone, etc., and therefore of an entirely 
 different character. 4 
 
 On the other hand, he points out that it is possible, by com- 
 paring these blocks with primitive rocks still in situ in the Alps, 
 to trace them to their parent beds. He adds further that they 
 are only found on those slopes of the Jura which face the Alps 
 (a view in which he was mistaken), and he concludes that they 
 had been transported from the Alps by some agency which had 
 rolled them and mixed them confusedly. 5 
 
 He next points out that the parts of the Jura which bear 
 the largest number of these blocks, and at the highest level, 
 are opposite the great Alpine valleys. Thus the most im- 
 portant deposits are opposite the Valley of the Rhone, and he 
 remarked how he had discovered prodigious masses of them 
 above Bonvillars, Grandson, La Sarra, to the N.W. and N.N.W. 
 of the opening of that valley ; while further south, above Nion, 
 
 4 Voyages dans les Alpes, i. 145-150. 5 Id. 148, 149. 
 
 C 2 
 
2O The Glacial Nightmare. 
 
 Bonmont, Thoiry, and Collonge, there were only very few such 
 blocks, since the exterior chain of the Alps, above Gingouph, 
 Meillerie, and Evian, forms an elevated and continuous barrier 
 which intercepted the passage of the fragments from the great 
 chain. 
 
 Similarly, Mount Saleve, opposite the Valley of the Arve, is 
 covered with blocks to a great height, and has itself formed a 
 barrier to their passage further into the Jura range. The 
 same applies to the Valley of St. Joire, opposite whose opening 
 at Montouse, on the southern flank of the Voirons, the blocks 
 abound ; while the northern flanks of the Voirons have none 
 of these erratics on them at any height, the continuous outer 
 buttresses of the Alps having prevented their passage. 6 
 Our author urges generally that wherever the country 
 is separated from the Alps by a continuously high buttress, 
 as the Valleys of Neuchatel and Franche-Comte, there the 
 blocks are not found, nor where the Jura intervenes, but 
 wherever gaps or valleys penetrate this barrier, opening up 
 routes to the Alps, in such valleys the case is different, and 
 he thus concludes that the blocks must have travelled to 
 where they now occur along these valleys. 7 This was clearly 
 an advance upon previous theories. 
 
 Having brought together these facts, and other similar ones, 
 De Saussure concluded that water alone could have taken up 
 and redeposited the boulders, and water alone could have 
 deposited the gravels, since the large and small stones are 
 often deposited in horizontal layers with mud intercalated. 
 
 But while water was the agent, he claimed that it was 
 water acting, not in its diurnal methods, either pluvial or 
 fluvial, "bait a great mass of water in violent motion. He 
 nowhere quite distinctly formulates a theory on the subject, 
 and we can only gather a general notion of what his view 
 was. 
 
 Inasmuch as the foreign stones have been found scattered 
 over Mount Saleve, and on the flanks of the Jura, to a height 
 of 800 or 400 toises above the level of the lake, he affirms that 
 the water which carried them must have risen at least to this 
 height. 8 
 
 6 Id. 152, 153. * id. 153, 154. 8 Op. cit. 150-152. 
 
De Saussure* 21 
 
 Writing in 1796, he speaks of the rolled pebbles of various 
 kinds of rock, granite, limestone, etc., and some of them of a 
 considerable size, which he found mixed with sand and 
 gravel on the very summit of the "Abbaye " ; all were rounded, 
 and he says " it cannot be doubted that these pebbles have 
 been deposited by a current of water on the flanks of the 
 mountain, and such a current must have been very large to 
 reach so high, and to have filled a valley more than half a 
 league in width. It is unnecessary to add that it could not 
 have been rain which produced a torrent sufficient to carry 
 along with it such large stones. The nature of the stones, 
 whose matrix is on]y found a long way off, also proves the 
 strength of the current." 9 
 
 He did not limit himself to explaining the rounded and 
 travelled stones by the agency of water. He also explained by 
 it, as Tilas had done long before in Sweden, the softened con- 
 tour of the country, and especially noticed the succession of 
 round-backed rolling rock surfaces, to which he gave the name 
 roclies moutonnees, from their fancied resemblance to flocks of 
 sheep lying side by side. 
 
 In addition to the blocks and roches moutonnees he also 
 mentions how, on Mount Saleve, there are remains of horizontal 
 fissures, four or five feet wide, twice this length, and one or two 
 feet deep, with their borders having rounded curves, such as 
 water makes, and he contrasts these fissures with those made by 
 rain ; the latter form excavations either perpendicular to the 
 horizon, or following the greatest inclination of the rocks, 
 while the others run along horizontally, and he attributes them 
 to the current which bore the Alpine erratics. He also refers 
 to rounded cavities in the rocks, two or three feet deep, whose 
 opening is towards the N.N.E.,and which he thinks were caused 
 by currents of water beating directly and impetuously against 
 the most salient and exposed parts of the rocks, namely, on 
 their vertical faces. These have their bottom and edges also 
 rounded. He very frankly admits that certain of the fissures 
 are distinctly in a direction inconsistent with that of the 
 current which he postulates, but he explains this by the fact 
 that, both in the case of the sea and of rivers, the water near 
 
 9 Op. cit. iii. 95. 
 
22 The Glacial Nightmare. 
 
 the border of the currents forms backwaters, etc., which often 
 move with considerable rapidity, and small whirlpools, whose 
 excavating power is considerable. The waves also have an 
 excavating force, and when driven by wind move in various 
 directions with the wind, and he appeals to the experience of 
 what occurs in ordinary rivers as a support to his theory. He 
 attributes to the same agency the deep pits, thirty feet deep 
 and eight or ten times longer, known as the Grottes de 
 1'Hermitage, as well as the carving out of the so-called gorge 
 of Monetier and the large gap separating the great and 
 little Saleve, since the beds of it on either side correspond, and 
 he could appeal to no other cause which could remove the 
 intervening rocks now wanting, 1 nor explain the portage 
 of the steep beds of white sand occurring on some parts of 
 Saleve. 2 
 
 Having appealed to water moving on a considerable scale, 
 De Saussure then goes on to discuss whence this water came, 
 and what gave it sufficient impulse to move such large masses, 
 He urges that long before the historic period the sea covered 
 the Swiss mountains to a considerable depth, when a violent 
 cataclysm occurred which disclosed great cavities in the ground, 
 previously empty, and caused the breakage of the strata. The 
 waters thereupon rushed into these gulfs, carrying with them 
 vast quantities of earth, sand, and fragments of different kinds 
 of rocks, choking the valleys with them. Presently the vio- 
 lence of the waters somewhat abated, whereupon the looser 
 materials were washed away, leaving the larger blocks behind. 3 
 He concludes quaintly that his explanation would suffice as 
 a demonstration for naturalists. " They know/' he says, "that 
 granite does not grow in the ground like truffles, and does not 
 grow like willows on calcareous rocks ; and if they have, as is 
 possible, different ideas to mine on the cause of the aqueous 
 movement which bore the debris so far, there will be few 
 who will not believe in a great debacle, or a current of con- 
 siderable violence and extent which transported and deposited 
 them where they are now found." 4 
 
 While De Saussure's view, that the blocks were dispersed by 
 water, gained many adherents, his theoretical explanation 
 
 1 Id. 164-166. 2 Id. 169. 3 Op. cit. 150-152. 
 
 4 Op. cit. 161-162. 
 
Dolomieu. 23 
 
 seemed very inadequate. For granting that the contour of 
 Switzerland was the same at the time of the disporsal of the 
 blocks as now, which is apparently De Saussure's view, and 
 that the general level of the sea was then much higher, it is 
 very difficult to understand how the water could have been 
 lowered. 
 
 As Charpentier says, we have no evidence that there exist 
 in the interior of the earth empty spaces capable of absorbing 
 such a quantity of water as would lower the general level of 
 the ocean by 11,000 feet, up to which point fossil marine 
 remains occur in the Alps; but, on the contrary, the evidence 
 points to the mountains in which these remains occur having 
 been raised violently above the present level of the sea. 
 
 Dolomieu, De Saussure's great contemporary, while accepting 
 his friend's appeal to a mass of moving water as alone explain- 
 ing the phenomena of the travelled stones, made the position 
 more tenable by claiming, as Ferber had done before him, that 
 since the stones were deposited the Alps had suffered great de- 
 nudation, so that what were once inclined planes down which 
 the water could roll the stones, were now uneven valleys, ne- 
 cessitating their travelling up and down hills. To quote his 
 own words, writing in 1793, he says " those who attribute 
 the foreign stones often found in valleys to the portage of the 
 rivers flowing through them, forget that many of these stones 
 are often ten thousand times larger than those which such rivers 
 roll along even when most impetuous. They also forget that 
 similar stones occur on the flanks of the valleys as those occur- 
 ring on some of the hills of Dauphine, a long way from the Alps; 
 also the rolled stones found in the plains of Verona, and among 
 them beautiful pieces of granite, and the various kinds of por- 
 phyry found in the Tyrol. These have been assigned to the 
 work of the Adige, because, in order to reach the place where 
 they are found from the Tyrol, they must have come through 
 the gorge through which the Adige flows. But they far exceed 
 in size any stones carried by the Adige even when most 
 flooded." 
 
 Turning especially to the plain of Lombardy, he calls 
 attention to the fact that it is covered with sands and trans- 
 ported stones which are foreign to it, and if they were 
 removed the Adriatic would overflow it right up to the Alps 
 
24 The Glacial Nightmare. 
 
 and the Appenines. Inter alia, he mentions now the hill 
 occupied by the hermitage of the Camaldoli at Turin is 
 formed of an immense mass of erratic blocks of granite and 
 other stone of great size, and much rounded, and imbedded 
 in shelly marls, the fragments of primitive rock being 
 agglutinated together with marine shells. This mountain, 
 three leagues long and one league wide, rises up isolated 
 between the valley of Aniers and the plain of Turin, and is 
 at a great distance from the mountains whence the stones 
 were derived. In between flows the Po, in a channel not more 
 than a hundred feet wide. There is no proportion, he says, 
 between the size of this river and the work it is supposed to 
 have done by those who attribute to it the excavation of 
 the valley. 5 
 
 The dispersal of the boulders, many of them thirty cubic feet 
 in size, and the mixture of the shells with the boulders, point 
 to the sea having been there ; it was no ordinary sea operating 
 through ordinary waves, which could not have reached up to 
 the height of 200 toises where the stones ocqur, nor carried 
 and mixed together debris of such size and of such different 
 origins. Like De Saussure, he was forced, he says, to appeal 
 to the sea agitated by a quite abnormal force, which swept up 
 over the Alpine summits, whence it broke off the masses of 
 rock, then rolled and deposited them where they are now 
 found. But, he adds, he does not demand that these waves 
 should have acted contrary to gravitation, and raised the 
 blocks more than 600 feet from a lower to a higher level, as 
 they must have been raised if the contour of the country had 
 been what it is now. He urges that the intervening valleys 
 have been since excavated, and that once a more or less 
 continuous plane joined the points whence the stones came 
 and those where they now occur. The intervening valleys 
 were previously occupied with debris, which was itself 
 swept away by the water, and it was over this that the big 
 stones rolled. 6 
 
 Dolomieu's views were virtually the same as those which 
 Hutton propounded in his famous " Theory of the Earth," 
 
 Observations sur la Physique, xlii. 54-57. 
 6 Observations sur la Physique, vol. xlii. (1793), 54-57, note. 
 
Hut ton and Play fair. 25 
 
 which caused a great revolution in geological reasoning. 
 The first edition of that work, published in 1788, did not 
 deal with the problem which at present interests us ; but, 
 in the second edition of it, published in 1795, he had to 
 face the facts which had been published by De Saussure, and 
 to reconcile them with his Uniformitarian views. After 
 quoting De Saussure's observations about the erratics, he 
 proceeds : 
 
 " These are facts which can only be explained in supposing 
 that the valleys have been hollowed out of the solid mass, by 
 the gradual operation of the rivers. In that case, stones, 
 travelled from afar, will be found at considerable heights, 
 upon the sides of the valleys at their under end, or where, 
 as our author says, they terminate in plains." 7 This ex- 
 planation was of course quite consistent with Button's 
 general premises that the operations of Nature have always 
 in times past been such as they are now. 
 
 The view propounded by Dolomieu and Hutton was 
 advocated by a considerable school of geologists. Thus 
 Playfair, in his " Illustrations of the Huttonian theory," 
 published in 1802, urges that when the blocks, etc., were 
 transported, the mountains were vastly higher than they 
 are now, and that the valleys between them have been since 
 cut out by the rivers and torrents from an immense ram- 
 part of solid rock. 8 Inasmuch as the blocks found on the 
 Jura have had to traverse the space now occupied by the 
 Rhone valley, he says, " Stones could not by any force that we 
 know of, be made to ascend over this height. We must 
 therefore suppose that when they travelled from Mont Blanc 
 to Jura, this deep valley did not exist, but that such a uniform 
 declivity, as water can run on with rapidity, extended from 
 the one summit to the other. . . . The horizontal distance 
 from Mont Jura to the granite mountains at the head of the 
 Arve, may be accounted fifty geographic miles, though we 
 suppose Mont Blanc and the rest of those mountains, to have 
 been originally much higher than they are at present, the 
 ridge of Jura must have been so likewise ; and though 
 probably not by an equal quantity, yet it is the fairest way to 
 
 * Theory of the Earth, ii. 249. 8 Playfair's Illustrations, 383. 
 
26 The Glacial Nightmare. 
 
 suppose tb.3 difference of their height to have been nearly the 
 same in former ages that it is at present, and it therefore may 
 be taken at 10,000 feet. The declivity of a plane, from the 
 top of Mont Jura to the top of Mont Blanc, would there- 
 fore be about one mile and three-quarters in fifty, or one 
 foot in thirty ; an inclination much greater than is necessary 
 for water to run on even with extreme rapidity, and more than 
 sufficient to enable a river or a torrent to carry with it stones 
 or fragments of rock, to almost any distance." 
 
 In answer to the obvious reply that such a notion would 
 not account for the fact pointed out by De Saussure, that the 
 blocks on the Jura are found in groups opposite the points 
 whence the central Chain of the Alps can be seen, he says that 
 this coincidence is quite accidental, because from whatever 
 cause the transportation of these blocks has proceeded, the 
 form of the mountains, especially of Mont Jura, must be too 
 much changed to admit of the supposition that the places on 
 it from which Mont Blanc is now visible, are the same from 
 which that mountain was visible when these stones were 
 transported hither. It may be, however, that the passes 
 which now exist in Mont Jura are the remains of valleys or 
 beds of torrents which once flowed westwards from the Alps, 
 and it is natural that the fragments from the latter mountains 
 should be found in the neighbourhood of those ancient water 
 tracks ; and, in order to explain how it comes about that the 
 granite of the Ornex has found its way across the inter- 
 vening valley of Orsiere and over the watershed into the 
 outlet of that of the Drance, he postulates that the valley of 
 Orsiere was not formed when the stones were transported. 9 
 The same view was urged in a work on geology, pub- 
 lished by Ebel in 1808. He adopts Dolomieu's view of an 
 inclined plane, and he further urges that the great debacle 
 flowed from SS.E. to NN.W., and argues that this is 
 proved by the direction taken by the erratics not only in the 
 district round Switzerland but in Darmstadt and Central 
 Germany. It is only, he urges, in Holstein, and the borders 
 of the Baltic, that the erratics seem to have come from another 
 direction. He urges that this southern origin of the flood is also 
 
 9 Id. 384-388. 
 
Conybeare and Phillips. 2 7 
 
 proved by the pebbles found in the Lower Rhine valley from 
 mountains near its sources. The gravel which is found 
 fifty French miles north of the Pyrenees points the same 
 lesson, and while it is true that drift occurs both to the 
 south of the Alps and Pyrenees, pointing to a smaller and 
 secondary flood in an opposite direction, this drift is much 
 less in amount and much less widely distributed than that to 
 the north of the mountains. 1 
 
 These same views were defended by Conybeare and Phillips 
 in their well-known " Outlines of Geology/' where we read: 
 "Another circumstance connected with the distribution of 
 these travelled fragments is, that we often find them in 
 masses of considerable size, accumulated in situations now 
 separated by the intervention of deep valleys from the parent 
 hills (if we may so speak) whence we know them to have been 
 torn. This appears to be a demonstrative proof that these 
 intervening valleys must have been excavated subsequently 
 to the transportation of these blocks ; for although we can 
 readily conceive how the agency of violent currents may have 
 driven these blocks down an inclined plane, or if the vis a tergo 
 were sufficient, along a level surface, or even up a very slight 
 and gradual acclivity, it is impossible to ascribe to them the 
 Sisyphean labour of rolling rocky masses, sometimes of many 
 tons in weight, up the face of abrupt and high escarpments." 
 
 It was not difficult to answer this theory, which not only 
 invoked a quite transcendental denudation in recent times, 
 but failed to meet the actual facts. The first person to 
 challenge it was the elder De Luc, one of whose principal 
 works is devoted almost entirely to a sharp criticism of it. 
 
 In the first of his letters to Blumenbach, dated from 
 Windsor, Sept. 1, 1792, he says, inter alia : "We are not 
 astonished to find great blocks of stone on the declivities 
 of mountains, and in their valleys when their summits are pre- 
 cipitous, because the first and most natural idea is, that they 
 have been detached from the upper parts. When afterwards 
 similar blocks are found in the channels of torrents and then 
 of rivers, still enclosed within the valleys, it may be thought 
 that they have descended along their declivities, covered with 
 
 1 Ebel, Bau der Erde, ii. 294, etc. 2 Op. cit. ed. 1822, xxix. 
 
28 The Glacial Nightmare. 
 
 other fractured stones, which, have yielded to their weight ; 
 when again we see those blocks even in the plains, carried 
 away by the idea that they must all proceed from the same 
 eminences, we are apt to forget the lakes, and so many deep 
 and tortuous channels, through which it was impossible for 
 the waters to propel them. But time is called in to supply 
 the want of means, and from the habit of thinking that no 
 other cause than running waters can have produced these 
 phenomena, impossibilities are overlooked. Neither the 
 enormous size of many of these blocks, nor the astonishing 
 dispersal of smaller blocks, not only on the plains but on the 
 hills ; nor the variety of their species in certain grounds, has 
 been noticed ; above all it has not been perceived that gravel 
 of the same kinds of stone is contained in many strata, mixed 
 with marine bodies. . . . When we are within the mountains of 
 granite, and find on their declivities great blocks of that stone, 
 down to the beds qf their torrents, and of their rivers, it seems 
 very natural to think that these blocks belonged originally 
 to the broken rocks observed in the upper parts of the 
 mountains ; and thus gradually to ascribe the same origin to 
 the blocks and pebbles of granite discoverable in the beds of 
 some rivers, even in the plains. But let us quit the granite 
 mountains and proceed to some remote chain whose strata 
 are of calcareous stone, as the chain of the Jura. In these 
 mountains, where the strata are in the same disorder, and the 
 heights present only broken rocks, we find also great blocks, 
 not only on the declivities, but in the channels of the torrents, 
 and of the rivers in the valleys : but do all these masses 
 proceed from the rocks above ? By no means : for, in the 
 first place, these rocks are of calcareous stone, and the greater 
 part of the blocks are of granite ; and besides, these blocks are 
 found on the summit of rocks of an entirely different nature, 
 and even sometimes in large heaps, which reduces the 
 hypothesis to an absurdity." 3 
 
 In his famous travels Be Luc devotes many pages to the 
 description and discussion of the boulders of the Baltic coun- 
 tries, and to the polemics in which he was involved about 
 them with Hutton and Playfair. He remarks on the great 
 
 3 Lettres, etc. 27-29. 
 
/. A. De Luc. 29 
 
 variety of species of granite heaped up as it were in the same 
 places, and mixed with a variety equally great of species of 
 porphyry, gneiss and other primordial stones scattered over 
 the hills and plains embraced by the courses of the Weser, 
 the Elbe and the Oder, and adds that " many observers as well 
 acquainted as himself with the mountains whence those rivers 
 proceed, knowing that they contain but few species of these 
 different genera of stones, have been convinced that the 
 scattered blocks and grovels could never have come from 
 these mountains, and that their source must be sought else- 
 where." 4 And he repeatedly refers to their being found aggre- 
 gated in masses, forming islands sometimes several acres in 
 extent, in the midst of a district otherwise more or less 
 free from them; that they are found on plains as well as 
 on hills. In some countries, he says, the blocks are more 
 generally disseminated ; but we always find their quantity 
 differs considerably in different places, without reference to 
 any particular point of the horizon. 
 
 In his travels De Luc states his objection to the running 
 water. theory. He says : "As the force of running waters 
 diminishes in proportion as the declivity decreases, the largest 
 masses should have remained nearest to the supposed moun- 
 tains whence the blocks came, and should get gradually smaller 
 and smaller. Now in reality nothing like this exists in the 
 dissemination of these masses ; besides, the blocks are found 
 in great quantities in many places surrounded by large tracts 
 of land containing scarcely any, and as for the size of the 
 masses it does not follow any regular order connected with 
 their situation." 5 
 
 He further urged, to use his own words, that " ridges of 
 mountains and hills, altogether foreign from the nature of 
 these fragments, intervene between the places where they 
 are dispersed, and every other mountain of their class." 6 
 
 De Luc says that Playfair, by appealing only to normal 
 causes, even granting that it is reasonable to suppose the 
 valleys have been hollowed out since the blocks were deposited, 
 fails in the first place to account for the blocks themselves ; 
 
 4 Op. cit. i. 56. 8 Travels (1810) i. 120-22. 
 
 6 De Luc, Elements of Geology, 109. 
 
30 The Glacial Nightmare. 
 
 and, secondly, fails to account for the necessary torrents of 
 water, since torrents cannot originate on the highest parts of a 
 country. 7 Nor again, he says, can any theory of inclined 
 planes account for the accumulations of blocks found on 
 level grounds, very remote from mountains. 8 
 
 The views of Ferber, Dolomieu, Hutton and Playfair were 
 also met by Sir James Hall, who says : " This hypothesis removes 
 the difficulty of the intervening valleys only ; for the trans- 
 portation in this case would be scarcely less difficult than it 
 would be to Geneva, as matters stand at this day/' He corrects 
 Playfair's calculation of the slope and shows it would be one in 
 twenty-three, since the blocks are not found at the summit of 
 the Jura, and then goes on to say that " the snowline is at 
 present 7500 feet below the summit of Mont Blanc. If a 
 sloping plane extended from that summit to the place where 
 the blocks occur on the Jura, the limit of perpetual snow 
 would extend along it horizontally for a distance of thirty-two 
 and a half miles. But granite is not found in situ beyond a 
 distance of ten miles from Mont Blanc at the Buet. Conse- 
 quently the whole of the granite in situ would be buried, deep 
 in snow, and could not well have been found on the surface of 
 the plane. Not only so, but inasmuch as the whole of the 
 snow covering the granite would be much above the snowline, 
 it is most improbable that streams could ever have run along 
 its surface, much less torrents and rivers competent to carry 
 huge blocks of stone." " To move a mass of granite of even 
 fifty or sixty cubic feet and to carry it a few yards, would require 
 the utmost efforts of the Rhone or the Arve, as they flow near 
 Geneva, in their highest floods, but that such blocks could be 
 conveyed by one of them along its whole course is contrary, I 
 conceive, to all experience, and still more when we consider 
 that these rivers are divided at their source from beneath the 
 glaciers into forty or fifty small streams. Yet from the 
 glaciers these blocks must have come ; and when we take into 
 account the magnitude of some of the granitic masses, it is 
 clear that the task is beyond the power of any river that 
 flows on the face of the earth. . . . These stones do not lie 
 merely in the beds of rivers, but occur all over the country, 
 
 i Elements, 269, 270. a Id. 272-3. 
 
Hall and Charpentier. 31 
 
 and on the summits of mountains, where rivers could least be 
 conceived to have flowed ; nor are they confined to that side 
 of the country, or to the side of the lake of Geneva which lies 
 next the Alps ; for we find them in particular on the face of the 
 Jura, which fronts the central ridge and the lake at an eleva- 
 tion of 2000 feet above the latter. A set of low hills also 
 intervene, which occasionally hide that central ridge from the 
 view ; and it is principally where the snowy summits are 
 visible from the face of the Jura by means of some depres- 
 sion in these intervening hills, that we find those travelled 
 masses/' 
 
 This argument, which seems conclusive, was further pressed 
 home by Charpentier. He urged that the mean distance of the 
 Alps from the Jura is about twenty-five leagues, that the highest 
 point where the erratics occur on the latter is about 4280 feet 
 above the sea level, and, granting that the places on the Alps 
 where the rocks occur in situ were 1100 feet higher, which is 
 considerably above the mean, the inclined plane would have a 
 slope of only 1 8' 50", much too slight to allow of the sliding 
 on it of the blocks as suggested. Inasmuch also as the upper 
 limit of the blocks on the Jura is marked by a convex line, 
 this also must have been the contour of the plane, which is 
 very improbable. The only means again by which such a 
 plane could be removed would be by the erosion of rivers, but 
 the valleys of the Alps are apparently all older than the deposit 
 of the blocks, nor are the greater part of them valleys of 
 erosion at all, but valleys formed by ruptures and fractures of 
 the strata. 1 
 
 These arguments make it clear that an appeal to the 
 diurnal action of rivers, whether flowing in their present 
 channels or flowing much higher up, before, as it has been 
 argued, they had worn down their present beds, is futile 
 as an explanation of the distribution of the boulders, and 
 that an explanation must be sought elsewhere. 
 
 Eeverting to the order of our story, we will now turn to an 
 inquirer whose memoirs on this subject have been somewhat 
 over-looked. . In 1790, J. H. L. Mierotto published a paper 
 in the Transactions of the Berlin Academy on the origin of the 
 
 9 Trails. Eoy. Soc. Edinburgh, vii. 142, 143. 
 1 Charpentier, Essai sur les Glaciers, 173-175. 
 
32 The Glacial Nightmare. 
 
 Baltic lands. In this he especially discusses the district watered 
 by the Oder. He urges that the external features of the boulders 
 found there, some of them rounded and some of them angular, 
 have special lessons for us. He also says that the sand in 
 which they lie is not composed of perfect crystals of quartz, 
 etc., but has had its angles and edges worn off as it would 
 have if one stone had been rubbed upon another as in a mill, 
 just as the rounded stones point to a similar friction as ex- 
 emplified in a sea strand. The sand dunes bordering the 
 Oder and the great number of rolled stones point, he says, to 
 the powerful action of water. If, as some argued, these 
 effects had been produced on the spot and the boulders, etc., 
 were the debris of mountains once existing here, then we 
 ought, in digging into the ground, to find larger and larger 
 blocks and eventually to come upon the unbroken strata; 
 but this is not so, the largest blocks occur on the surface. 
 If the sea had worn away some mountains of hard rock once 
 here we should have had some debris of the old rocks still left 
 in situ, and should also have bad a rugged coast instead of 
 a flat and sloping one ; and he consequently urges that the 
 debris referred to are not in situ but have been brought from 
 elsewhere. He then goes on to argue that in order to learn 
 whence they came we must examine the different kinds of 
 rock of which the boulders consist, and he gives a list of such, 
 and proceeds to argue that they may be traced to the 
 mountains whence the river springs and that they are the 
 debris of the weathering of these mountains subsequently 
 rolled and distributed by the flooded rivers and the result 
 of myriads of years of work. He also suggests that the 
 blocks of stone m iy possibly have been assisted in their trans- 
 port by being attached to masses of ice which bore them as 
 far as the water. This view of Mierotto he afterwards modi- 
 fied. I have not been able to find any memoir where he 
 published a revised theory, but both De Luc and Wrede refer to 
 it, and he must have published it before 1800, when he died. 
 Having apparently compared the boulders of North Germany 
 with the rocks to and in situ in the Silesian mountains and 
 discovered that they were not derived thence, he suggested 
 that they had come from Sweden across the Baltic, and had 
 been carried by ice, at a time when the sea was sufficiently 
 
De la Metherie. 33 
 
 high to cover the surrounding hills. This sea, he urged, had 
 been successively depressed, whence the large number of blocks 
 on the coast. De Luc says that in confirmation of his view 
 Mierotto had told him how blocks were sometimes seen en- 
 gaged in the ice and that when driven on the coast by storms 
 they remained there after the ice had melted. 2 I shall recur 
 to this theory in a later chapter; here it will suffice to 
 point out that Mierotto was the first to suggest that the 
 erratics of Germany came from Scandinavia, a view which 
 has been generally attributed to other inquirers. 
 
 De La Metherie, writing in 1795, devotes several pages 
 of his well-known work, entitled "Theorie de la Terre," to 
 the consideration of the travelled stones. He tells us that 
 so far as he could learn from travellers, granitic blocks 
 were not found on secondary mountains in Asia, nor in 
 Egypt, Mauretania, or Africa, nor yet in America (! ! !). In 
 Spain, France, outside the Jura range, England and Holland, 
 granitic pebbles, he says, are found only in valleys, plains, and 
 along the rivers, and he would limit the phenomenon to the 
 calcareous chains about the Alps, the Carpathians and the 
 Waldai mountains and generally over an area stretching from 
 Turin to St. Petersburg and Moscow. He speaks of the diffi- 
 culty of accounting for the dispersal of the blocks, chiefly 
 because we have to postulate their having travelled across 
 deep valleys, and because of the great distances to which they 
 have been borne. He urges in the first place that the blocks 
 must have been transported after the deposition of the 
 secondary rocks. Secondly, that they were transported by water 
 since they are all more or less rounded. Thirdly, that this water 
 must have moved at the height where the blocks occur. It 
 must also have included in its action the Alps, Carpathians, 
 and Waldai mountains, and its chief force was exerted against 
 the western faces of these mountains. He argues against the 
 possibility of any river or other such agent having excavated 
 the valleys, since the blocks were moved, as Dolomieu and 
 others urged, and also against a mere local flood such as would 
 be created by the bursting of an Alpine lake, and he attributes 
 the distribution of the stones to the action of the sea, especially 
 
 a De Luc's Travels (1810) i.57, & 219, 251. 
 
34 The Glacial Nightmare. 
 
 quoting the power of the tides at St. Malo to move large 
 stones and also to raise them up to a considerable height from 
 the floor of the Channel to the points where they occur. And 
 he says : " Granting that these waves were sufficiently great to 
 include the whole district from St. Petersburg to the Alps in 
 their action, their size and consequent force would be equal 
 to the work suggested. This being supplemented by the action 
 of earthquakes and other movements of the earth's crust which 
 sometimes have transported matters a considerable distance." 3 
 In 1797 M. P. Bertrand published his " Nouveaux principes 
 de Geologie," in which he traversed some of the positions 
 maintained by De La Metherie. In this work he urged that 
 the erratics are not confined to the environs of the Alps, Car- 
 pathians, and Waldai chain. He says he had seen similar 
 blocks from the Pyrenees, along the Garonne, and in the dis- 
 trict of Pau ; others from the Cevennes, along the Aveyron and 
 in Dordogne ; others from the Vosges at Bains. He further 
 urges that no such cause of transport as the waves of the sea 
 merely would account for the removal of such boulders as the 
 famous monster out of which the pedestal of Peter the Great's 
 statue at St. Petersburgh was carved. Its angles were still un- 
 rounded and its weight enormous. He argues that the only 
 cause competent to move the blocks, whose large numbers he 
 calls attention to, was a vast torrent of water consequent on a 
 general debacle or flood, and he argues, as Dolomieu had argued, 
 that this was perhaps assisted by the valleys having then been 
 tilled up so as to form more or less inclined planes, and he 
 quotes the. fact of the sharp angles of many of the stones 
 being against their having been worn by the sea, and to the 
 further fact that the sea really only acts upon the rocks and 
 stones over a depth of 200 feet about its margin and does not 
 raise them from its own depths and lift them 3000 or 4000 feet 
 upwards, as the facts necessitate it doing if it is to explain 
 the position of many of the boulders. 4 
 
 About this time, namely in 1798-9, M. De Lasterye made a 
 journey to Sweden and Norway. He apparently did not 
 publish the results of his investigations there till the year 
 1827, and it is not easy to separate what he discovered 
 
 3 La Metherie, Theorie de la Terre, ii. 251-203. - 
 4 Op. cit. 160470. 
 
De Lasterye. 35 
 
 during his voyage from wliat he afterwards added, but it 
 would appear the facts were mainly collected during his 
 travels. 
 
 He urges that the waters of the sea were driven from the 
 Pole towards the S. and S.W. of Europe, and that en route 
 they broke the mountain tops and carried with them the 
 debris and scattered them over the European continent and 
 the sea bottom. Among the debris were boulders which it 
 rolled, and others which were carried intact. The further one 
 gets from the Pole the finer do the materials of the loose 
 surface-deposits become. The mixed heterogeneous deposit of 
 this flood exists in Spain, Italy, England, France, and South 
 Germany. Further north the sands increase in quantity and 
 in coarseness of texture. This is so in Westphalia and North 
 Germany. In these sands rolled boulders often occur. In 
 Holstein and in Denmark the boulders increase in number 
 and size. In crossing the gulf of Flensborg the sea bottom 
 seems as if it had been paved by human hands, and the 
 rounded blocks of granite increase in numbers as we near 
 Elsinore. 
 
 In regard to Sweden, he tells us that its plains consist for 
 the most part of sterile sand and gravel, on which are 
 scattered blocks of granite whose angles show that they have 
 been less rubbed than the stones found further south. He 
 remarks that the outlines of the mountains are rounded, which 
 could only have been produced by the friction of the debris 
 carried along by water. He noticed further, that while 
 the northern faces of the mountains are rounded, the 
 southern ones preserve their rugged aspect and have their 
 angles intact, and he attributes this to the fact that the flood 
 of water laden with hard materials came from the N.E. and 
 flowed towards the S.W. He says he had noticed everywhere 
 on the mountains and the plains that the granite surfaces 
 were polished, wherever the superincumbent soft material was 
 removed, as perfectly as if they had been of marble and 
 polished by human agency. Everywhere over a space of 
 250 leagues, where the granite surface was not weathered it 
 appeared as brilliant as if it had just issued from an artist's 
 studio. In going north from Trolhatten the angular erratics are 
 found mixed with rolled pebbles, but the latter diminish and 
 
 D 2 
 
36 The Glacial Nightmare. 
 
 the former preponderate as we go northwards until they 
 occupy almost the entire surface. The two isolated mountains 
 of Hanneberg and Halleberg stand on a low granite plateau 
 showing the polish already named. On the summit o the 
 former are many angular erratics and many rounded boulders, 
 some of which are three metres in diameter, while the rock- 
 surface on the summit is also polished. Boulders of bhe same 
 kind of rock of which they are composed are scattered in 
 various places to the south and south-west. 
 
 In going from Halleberg to Christiania he noticed the same 
 polished surfaces and the same evidence of a smoothed contour 
 to the mountains on their northern faces and a rough one on 
 their southern ones, and that the rounded boulders are less 
 frequent where the angular ones prevail. 
 
 He noticed in regard to these polished surfaces on the tops 
 of the mountains that as far north as 62 N. lat., where the 
 surfaces had been exposed they had been weathered, but when 
 protected the polish had been preserved. After passing 
 the 66th degree of north latitude he had not found any of 
 these polished surfaces, although he noticed both rounded 
 boulders and sharp-angled ones. While rounded boulders 
 occur on the mountains, unrounded ones occur in the river 
 beds, showing that the rounding had not been caused by river 
 action, and that the rivers had not carried the unrounded ones 
 far. The east of Sweden, he says, presents similar testimony 
 to the west, and he argues that in going north to Lapland it is 
 probable the same testimony to a flood of waters coming from 
 the north would be available, and he argues from all the facts 
 before him, that there must have been two catastrophes, and 
 that the waters causing them flowed from the north towards 
 the south-west. 5 
 
 De Luc's repeated continental journeys, during which he 
 occupied himself so much with drift phenomena, naturally 
 opened his eyes to similar facts in England, where he lived 
 so long, and where, so far as I know, he was the first to 
 notice them. Had it not been that when here, he resided and 
 travelled almost entirely in the district south of the Thames, 
 he would no doubt have forestalled a good many later writers. 
 
 fc Journal des connaissances usuelles et pratiques, 2nd ed. T, 6-14. 
 
Aikin. 2>7 
 
 His only opportunities of seeing boulders was on his journeys 
 to the continent on the eastern coast, where he mentions 
 having noticed them, and also suggests that they possibly 
 came from Norway. He was apparently also the first to 
 refer to the Irish boulders. He tells us that his nephew, in 
 writing to him from Ireland in 1796, noticed, in constructing 
 an aqueduct at Kildare, that the materials, which were taken, 
 from one of two adjoining hills, about eight miles from Dublin, 
 to make an embankment between them, consisted of loose 
 debris, among which were rounded masses from two to six 
 feet in diameter, of a calcareous stone very different from 
 that of the strata of the hill. Similar rock appears at the 
 surface at various places some miles off, inter alia, at 
 Rahenny, four miles from Dublin, and Slane, twenty-four 
 miles from the same city. He also mentions how, in the 
 north of Clare, there is a plain thickly strewn with calcareous 
 blocks, some of immense size. 6 
 
 I only know of one other reference to the drift phenomena 
 in these islands in the literature of the last century. 
 
 Mr. Arthur Aikin, in his " Journal of a Tour in North 
 Wales/' published in 1797, refers to the beds of rounded 
 pebbles found on the highest parts of the slate moun- 
 tains. Their present position, he says, could never have 
 been that in which they were formed, for they consist almost 
 universally of porphyry, quartz, serpentine, and other sub- 
 stances which lie in large masses composing the primitive 
 mountains, their rounded shape too, like that of the pebbles 
 on the sea shore, seems to intimate that they have been carried 
 by the force of water to the places which they now occupy. 
 Another circumstance which appears to point out the quarter 
 whence they originally proceeded is, that, in proportion to 
 their vicinity to the primitive mountains, is their size : this 
 might be expected, since the further they were carried, the more 
 they would be rounded and comminuted. " Still, however, 
 there is a difficulty attending this hypothesis ; namely, by what 
 means could these rounded pebbles have been forced across 
 the many deep valleys that intersect the mountains in all 
 directions without first filling up the valleys themselves? 
 
 6 El. Treatise on Geology, 319-321. 
 
38 The Glacial Nightmare. 
 
 And, if this was the case, by what means were the valleys so 
 entirely cleared of them afterwards, as they appear now for the 
 most part to be ? " He then goes on to suggest that when the 
 slate mountains were formed under the water there were no 
 valleys, but the whole mass was one uniform bank, the valleys 
 being afterwards formed by the rivers as the waters sub- 
 sided. 7 
 
 In concluding this chapter two facts will be present to the 
 reader; first, that the superficial and commonplace pheno- 
 mena of the drift had been observed and recorded long be- 
 fore the time generally supposed, as judged by the current 
 memoirs on geology ; and secondly, that England and its science 
 had very little part in these first excursions into the un- 
 known land. The latter conclusion is a little unexpected 
 when we consider what an amount of work has been devoted 
 to elucidating the subject in later times by the geologists of 
 our own country. 
 
 7 Op. cit. 219, 220. 
 
39 
 
 CHAPTER II. 
 
 THE PHENOMENA OP THE DRIFT. 
 
 The Champions of Water (Part IL), 1800-1825. 
 
 " The Cornish tinners hold a strong imagination, that on the withdraw- 
 ing of Noah's flood to the sea, the same took his course from east to west, 
 violently breaking up and forcibly carrying with it the earth, trees, an. I 
 rocks, which lay anywhere loosely, near the upper surface of the ground. 
 To confirm the likelihood of which supposed truths, they do, many 
 times, dig up whole and huge timber trees, which they conceive at the 
 deluge, to have been overturned and overwhelmed." CAREW'S " Survey 
 of Cornwall," 1602, p. 7. 
 
 Playfair, 1802 J. A. De Luc, 1809-1813 Yon Buch, 1802-1818 Haus- 
 mann, 1807 Landt, 1810 Brochant de Villiers, 1819 Charpentier, 
 184L Sir James Hall, 1812 Allen, 1812 Drake, 1815- Weaver, 
 1818 Imrie, 1818 Mitchell, 1818 G-reenough, 1819 Escher You 
 der Linth, 1819 Fox Strangways, 1819-1821 Razoumofski, 1819 
 Henslow, 1820 Horace H. Hayden, 1820 Bigsby, 1821-1823 
 Hitchcock, 1822 Grainger, 1822" N., 1823 Dr. Buckland, 1823. 
 
 IN the previous chapter I collected some of the opinions of tho 
 geologists of the last century who first scientifically examined 
 the problem of the drift. I now propose to continue my 
 story and to discuss the views of students in the earlier 
 part of the present century who attributed the phenomena 
 in question to a diluvial origin. Whereas the names of 
 Englishmen are singularly absent from the former chapter, 
 they now become conspicuous, and the progress of discovery 
 in this difficult field is very largely traceable to our country- 
 men. The person who, as I have said, initiated a real interest 
 in the question here was De Luc, who lived so long in 
 England and who died at Windsor. It was the sharp 
 difference which separated him from the Huttonian school 
 which in a large measure led to his critical sifting of a 
 problem which afforded a fair touchstone of that theory 9 
 
40 The Glacial Nightmare. 
 
 In 1802 PI ay fair published his famous Illustrations of the 
 Huttonian theory. In this work he devotes a considerable 
 space to the travelled stones. He argues that they diminish 
 in size the further we get away from their source ; and he 
 further adds that many of the stones were already rounded 
 in a former geological period and have been washed out of 
 early conglomerates. For the most part, he quotes the facts 
 already collected by De Saussure and others, but he also refers 
 to a large granite stone on the sea shore of Arran, at least 
 three miles from the nearest granite in situ, and with a bay 
 of the sea intervening, and which must have come from Goat 
 Fell. 1 As we have seen, he objects to invoking a great debacle, 
 or torrent, and accounts for the transportation of the stones by 
 attributing a much higher altitude to the mountains formerly, 
 by the argument that the valleys have been cutout of immense 
 ramparts of solid rock by the rivers and torrents flowing 
 through them, and by the fact that when the stones were 
 transported, it was along inclined planes. 2 
 
 This work of Playfair's, as I have said, largely incited the 
 polemical spirit of De Luc, who, in his Treatise on Geology and 
 the several volumes which he devoted to an account of his 
 geological travels, adduced a large number of carefully 
 observed facts in regard to the distribution of the travelled 
 boulders in the country north of the Baltic. I have already 
 referred to the general line of argument which these facts 
 suggested to him, and will here content myself with mention- 
 ing the more important of his later conclusions. 
 
 His journeys round the Baltic enabled him to state very 
 positively that all its coasts, as also the isthmus joining 
 Denmark to the mainland, and the Baltic islands, including the 
 islands of Funen and Zeland, as well as the small islands of 
 Lsesoe, Anholt, and Samsoe in the Kattegat, and the island 
 of Rugen, are covered more or less with erratic blocks. 8 
 
 " In the heaps of blocks/' says De Luc, " found in West- 
 phalia, Lower Saxony, and Pomerania, there occur an immense 
 variety of species, not only of granite, but of porphyry, and other 
 stones of the same class, entirely unknown in the mountains 
 of Germany. This has induced some German mineralogists 
 
 1 Op. cit 393-4 2 Id. 383. 3 Travels, i. 384-386. 
 
De Ltic. 41 
 
 to argue that these blocks have proceeded from the mountains 
 of Sweden; the ice in the Baltic having served them as a 
 vehicle for traversing that sea of which, since then, the level 
 has subsided." * He elsewhere quotes Professor Pfaff as an 
 authority for the fact that in the mountains of the Black 
 Forest,, in passing over a horizontal tract a league or two in 
 extent, he had observed as great a variety of species of granite 
 as occurs among the blocks of Holstein and the countries 
 bordering on the Baltic. 5 
 
 Elsewhere he urges that the blocks are not geographically 
 arranged according to their size. They are generally found 
 associated with gravel formed of the same kind of rock, but 
 in some places the blocks are infrequent while the gravels 
 abound, and vice versa. The blocks again differ, in that some are 
 angular and others are rounded. In regard to the sand in 
 which many of them are buried in Westphalia, Lower Saxony, 
 Pomerania and further eastwards, it is not, he says, a detritus, 
 for it consists of grains of transparent quartz of irregular 
 figures, polished even in their smallest cavities. 6 He further 
 mentions how the blocks abound on the mounds occurring in 
 Lower Saxony in the form of rings. 7 All this, he urges, is 
 against the notion that the debris has been borne along by 
 running water, and in favour of its having been extruded 
 from below the ground by explosions. 8 Reverting to this 
 sand, he says, " When examined through a microscope its 
 grains are transparent and polished even in the hollow parts 
 of their irregular figures, while granitic sand not inferior 
 in fineness found on the sides of torrents among granite 
 mountains always consists of angular fragments without any 
 transparency, the grains having been rubbed against each 
 other/' This observation, he tolls us, had also been made by 
 De Saussure. 9 
 
 In answer to the notion of Playfair, he remarked, that 
 instead of the larger stones being always found nearest the 
 mountains from which they sprang, as they ought to do by 
 the diluvial theory, the district where he had observed 
 the largest of about twenty feet in diameter, was that most 
 remote from every granitic mountain, namely in Holstein. 
 
 4 Elements, 301. 5 Id. 339. 6 Id. 62. 
 
 7 Id. 64, 65. 8 Id. 65. 9 124-5. 
 
42 The Glacial Nightmare. 
 
 He also remarks, as contrary to that theory, that he had found 
 blocks of granite, porphyry, and other stones of different 
 kinds, accumulated in spots surrounded by tracts in which 
 few or none occur, forming as it were islands in tracts of land 
 free from them. 1 
 
 While De Luc was engaged in his widespread researches, 
 other inquirers were busy elsewhere. Leopold Von Buch, in 
 his " Geological results of travels in Germany, etc./' refers to 
 the boulders occurring in the country south of the Baltic, and 
 points out that they grow in numbers as we approach that sea ; 
 and, like Mierotto, he suggests that this points to their having 
 come from Scandinavia and not from the Silesian mountains. 2 
 In 1807 Haussman pressed the same view in his "Resultate 
 geog. Yersuch auf eine reise, etc." 3 
 
 In 1810 Landt wrote an account of the Faroes, from which 
 I take the following sentence : 
 
 " Besides the large collections of stones which are occa- 
 sionally found on the hills, there are seen sometimes in the- 
 valleys single stones six, eight, or ten feet in diameter, but in 
 places where it is impossible they could have fallen down from 
 the hills. Such stones are found also, here and there, at a 
 considerable height on the hills when there is no other emin- 
 ence in the neighbourhood from which they might have rolled 
 down," and he goes on to discriminate between these stones 
 and others which have obviously been detached from the 
 local rocks in situ. 41 
 
 On the 1st of October, 1811, Von Buch read a famous 
 memoir before the Berlin Academy, entitled "Ueber die 
 ursachen der Verbreitung grosser alpengeschiebe," in which 
 he collected fresh materials for solving the riddle of the 
 drifted stones, etc. 
 
 Following in the steps of De Saussure he not only showed 
 that these blocks occur on the Jura, opposite the outlet of the 
 great valleys and sorted according to their origin, but that 
 they are limited to the upper parts of the Jura, while similar 
 blocks are absent or very infrequent in the low ground 
 intervening between that range and the Alps. 
 
 1 Elements of Geology, 315 and 316. 2 Op. cifc. i. 132. 
 
 3 Gottingen gesell. anzeiger, 1807, 208 and 2074. 
 4 Landt, Desc. of the Faroes, 1810, 8. 
 
Von Buck. 43 
 
 Not only do they occupy the upper zone of country, but 
 they occupy it in a very curious way. They are not found at 
 the same level everywhere, but, on the contrary, at very 
 different levels. The culminating point, according to Von 
 Buch, is the mountain of Chasseron above Yverdun, where they 
 are first found at the village of les Bulets, at a height of 
 1900 feet above the sea level. 'At Chaumont, opposite 
 Neuchatel, they are first found at 2400 feet above the sea 
 level, on the hills bordering the Yal de Buz, they are first seen 
 at the villages of les Hautes-Genevays, Dombesson and St. 
 Martin, 1800 feet above the sea level, at Nods not far from 
 the lake of Bienne, at 1360 feet. On the mountain of Boudry, 
 opposite Neuchatel, at 1100 feet, at the upper end of the Yal de 
 Buz, at 840, 850 feet, and so on. The culminating point at 
 Chasseron is opposite the great Ehone valley. 5 
 
 The fact of the occurrence of the blocks, only in the upper 
 zone, and of their occurring in this way with a culminating 
 point whence the levels at which they are found are lower in 
 all directions, while they do not occur at all in the bottoms of 
 the valleys, led Yon Buch to the conclusion that only a 
 violent projection or impulse by a rapid current of water 
 would account for the facts ; and also that their dispersal was 
 due to one sudden effort, inasmuch as a continuous force acting 
 for some time would have carried the blocks outside the two 
 extreme limits of the embouchure of the Ehone. He urges 
 that the Alps were then considerably higher, and that the 
 force which impelled the stones was sufficient to overcome 
 their gravity. He compares the effects of this force to that of 
 powder on a cannon ball, and urges that in this way the 
 blocks were projected across the valleys and deposited on the 
 heights which formed barriers to them, and that if the Jura 
 had not intervened they would probably have been scattered 
 over Burgundy and Tranche Comte. 6 As evidence of the 
 violent transport of the blocks, he mentions how that frequently 
 the big ones are broken into two or three pieces and sometimes 
 surrounded with smaller ones whose angles correspond, 
 and which are clearly fragments which have been broken 
 from them. Among many others which he had seen he 
 
 6 Abhand Acad. Berlin, 1804-1811, p. 166, 167. 6 Id. 168, 169. 
 
44 The Glacial Nightmare. 
 
 especially cites a group on the way from Yanseyon to 
 Valangin, another at the entrance to the valley of Corcelles, 
 and others on the way to Sonceboz above Bienne. 7 
 
 With great pains Yon Buch traced the provenance of the 
 blocks. He showed that they consist of various kinds of 
 granite, among which the peculiar granite of Mont Blanc, 
 of the St. Gothard, of the Grimsel, and of Grindelwald can be 
 easily discriminated ; and, following up de Saussure's re- 
 searches, he showed how the blocks on the Jura and those of 
 theDrance come from the Ornex, which forms the most northern 
 promontory of the Mont Blanc range, which also supplied the 
 granite for the blocks in the valleys of Ferret, Champeix, 
 Entremont, Bagne, etc. At the other extremity of the Mont 
 Blanc range are the sharp-pointed rocks above Montjoie, 
 whence came the blocks in the valley of the Arve. Besides 
 granite, other blocks on the Jura are formed of a conglomerate 
 which he traced to Valorsine and Trient, on the line con- 
 necting Ornex with the Jura. The gneissic rocks on the same 
 range he traced to the mountains of Saint Branchier and 
 Martigny also on the same line ; the black limestone and slate 
 to the Dent du Midi and the Dent de Morcle, bordering the 
 Valais near Saint Maurice ; while the diallage and jade 
 came from the valley of Bagne near the Drance. He thus 
 showed that it was from the Rhone valley and its tributaries 
 that the blocks of the Jura came ; 8 those of the Aar, which 
 came by way of Thun, Berne and Soleure, are traceable to 
 the granites of Grindelwald and to gneiss like that of the 
 Eiger. Those opposite the 'Limmat which traverses the 
 greater part of the Canton of Zurich, are formed of the red 
 conglomerate so common in the Glaris range, and the deposits 
 opposite the mouth of the Arve below Geneva, of the granites 
 from the southern extremity of the Mont Blanc range, etc., 
 etc. 9 
 
 Yon Buch not only assigned a cause for the dispersal of 
 the erratics, but also calculated the intensity of the force with 
 which the torrent he postulated must have flowed. As the 
 summit of Ornex is distant from Chasseron about 356,117 
 feet, and the difference of their heights is about 5100 feet, and 
 
 * Id. 170. 8 Id. 170-183. 9 Id. 184, 185* 
 
Von Buck. 45 
 
 as the blocks must have traversed the distance at one bound, it 
 would take them the same time to go as it would for them to 
 fall perpendicularly to the ground by gravity, that is to say 
 eighteen seconds, so that they must have travelled at the rate 
 of 19,460 feet a second. If a mass of water 5100 feet deep 
 were suddenly to break loose, and if the lower blocks were 
 conveyed by it in a different time to the upper ones, the pace 
 at which they would travel would be 553 feet a second. To 
 get a speed of 19,460 feet would require a depth of water of 
 6,311,526 feet, hence it is clear no appeal to a divergence 
 of the Ehone over the Jura or the mountains of St. Maurice 
 would suffice, and we must appeal to a wider catastrophe. 1 
 
 Von Buch had now satisfied himself that the blocks found 
 scattered over Brandenburgh, Pomerania, Mecklenburg and 
 Holland came from Scandinavia, and he points out how a 
 continuous stream of similar stones may be traced across the 
 Danish islands into South Sweden where their number 
 culminates, and he urges that they crossed the depression of 
 the Baltic as those of the Jura crossed the intervening valleys. 
 The blocks found over Northern Europe, he says, differ from 
 the rocks of Saxony and Silesia. Besides which they become 
 scarcer as we go south, and disappear long before we reach 
 those districts. They occur infrequently near Leipzig, and 
 not at all about Weimar and Erfurt, and wherever any low 
 outliers of the Hartz intervene, there we find no blocks to the 
 south. If we trace their distribution we must draw a half 
 circle with Scandinavia for a centre. They are found in 
 Eastern England, at Antwerp, and scarcely reach Brussels. In 
 the woods of Breda there are many large blocks, as there are 
 in Groningen and Overyssel, Munster, Munden, Hildesheim, 
 the Hartz, Leipzig and the frontiers of Upper and Lower 
 Lusatia. In Poland they follow the Prussian boundary. 
 
 In Russia, Guldenstadt 2 had found masses of granite 
 scattered as far as the Torschok, not far from Twerza 
 near Tuer, but not farther south than Moscow, and Von Buch 
 urges that while the phenomenon of the dispersal of the 
 northern erratics is on a wider scale, it is of the same nature 
 as that of Switzerland, and was probably due to the same 
 
 1 Id. 183. 2 Tide ante. 
 
46 The Glacial Nightmare. 
 
 cause. He suggests that it may have been the same cause 
 which overwhelmed the great elephants (i.e. the mammoths). 8 
 
 In a letter to M. Brochant de Villiers (published in the 
 Annales de Chimie, vol. x.) Yon Buch supports his theory 
 lay the fact that other than granite blocks are found in the 
 Pays deVaud, whose original provenance was at a lower level 
 than Ornex, and he argues that this origin at a lower level 
 would cause them,, with the same impulse, to touch the ground 
 sooner. As to the extravagant calculation of the amount of 
 water and its speed combined, to secure the required effects, he 
 qualifies it greatly by showing that the immersion of the 
 blocks in water would greatly diminish their speed, and he cites 
 some experiments of Hook, on the rate in which objects rush 
 in water to that effect. By this means he increases the. time 
 employed in the trajectory from 18 to 1020 seconds, and cal- 
 culates the speed at 354 feet a second. This would be in water 
 simply, but water charged with mud and debris would have the 
 effect of still more reducing the speed to reasonable propor- 
 tions, and he quotes the experiences of Escher Von der Linth and 
 of Charpentier, that if we may take the famous debacle of 181 8, 
 at Bagnes^for our guide, the amount of solid debris was probably 
 seven-eighths of the whole flood. That famous debacle (an 
 account of which by Escher is contained in the Bibliotheque 
 Universelle for August, 1818) is described by Charpentier as a 
 torrent of mud preceded by a mountain of wood from the 
 forests, of debris of houses, dykes, etc., etc. According to M. 
 Escher this famous torrent flowed at the rate of thirty-three 
 feet a second at the higest point, and gradually diminished in 
 speed to 18'12, and lastly to six feet a second. 4 Yon Buch 
 nowhere, so far as I can find, specifically attributes this rush of 
 water to a distinct cause, but he seems to infer that it was due 
 to the sudden upheaval of the great granitic masses of: the 
 Alps. 
 
 The editor of these supplementary notes of Yon Buch, M. 
 Brochant de Yilliers, adds his own criticisms. In these he 
 admits with De Saussure, Dolomieu, Escher, etc., the Alpine 
 origin of the blocks as indisputable. He also admits the 
 probability of their transport by a violent debacle, but he 
 
 3 Id 186. 4 Annales de Chimie, x. 248-253. 
 
Brochant de Villiers. 47 
 
 cannot accept the rest of Von Buck's conclusions. He urges, 
 with reason, that the difference of level between Ornex and the 
 parts of the Jura where the blocks occur being 5100 feefc, gives 
 a slope of 0'014 only, while the torrent of Bagnes had a mean 
 inclination of 0'017. This would indicate a speed of only 
 15 feet a second, instead of one of several hundreds ; while, if it 
 was a torrent loaded with debris as Yon Buch suggests, it 
 would have to be at a considerable greater level than the 
 point where the blocks occur, and probably as high as the 
 crest of the Jura, which would diminish the inclination 
 unless the Ornex was itself much higher ; but in such a case it 
 would be difficult to account for the blocks existing on the 
 cirque of Pierre at Voie, at Liddes, etc., and which seem to fix 
 the limits of the height of the liquid mass near its origin. 
 How, again, was it that the torrent did not drop any portion 
 of its load of mud on lower heights and before it reached the 
 Jura, or cover the blocks instead of leaving them bare. It 
 would be difficult to believe, also, that such a torrent could 
 retain the blocks in the middle of its current, and that a 
 considerable number did not escape laterally and fall into 
 the intervening low ground. For these various reasons he 
 seems to incline to the notions of Dolomieu and others, namely 
 that a continuously inclined plane once joined the summits 
 of the Alps and the Jura, and he professes to find traces of 
 subsequent denudations. This plane would do away with the 
 necessity for Buch's immense watery trajectories, since it 
 would account for the removal of the blocks with the beds on 
 which they lay from the intervening valleys. He urges 
 that the fact of the blocks being found at different levels is 
 not an objection, as Yon Buch supposed, to this view, since 
 loaded currents have a tendency to deposit mud in their 
 beds. A section of such a torrent would have an arched 
 outline, and consequently when it met a barrier such a 
 current would deposit its load in an arched curve also. This 
 hypothetical torrent bed, our author apparently concludes, 
 was removed by subsequent denudations on a large scale. 5 
 
 Charpentier urges that if a huge flood was caused, as 
 seems to be implied, by the sudden upheaval of the ground, 
 
 5 Annales de Chimie, x. 254-264 
 
48 The Glacial Nightmare. 
 
 the blocks would not be projected horizontally only, so as to 
 occur in well marked zones, but many of them would be shot 
 into the air and be thus distributed indiscriminately on the 
 mountains. Again, if they were shot out horizontally, it is 
 difficult to understand how the uplifting of the granitic chain of 
 Ferret could so project the debris as to reach the Jura when in- 
 tervening ranges like those of Catogne, le Sex d'Armensier, le 
 Mont de Charavex, le Salantin, la Dent du Midi, la Montagne 
 de Foully and la Dent de Morcles stood in the way. 
 
 Again, the blocks found in the Rhone valley have como 
 not only from the Ferret, but from all the valleys of the 
 Valais, and it is impossible to understand how a shock such 
 as is postulated could give these blocks a horizontal move- 
 ment which should take them for a distance of sixty leagues, 
 and not in straight but in crooked lines. The stones are 
 found scattered over the low country, both south and north 
 of the Alps, so that the shock must have had a propelling force 
 in all directions at once. These blocks, again, have clearly 
 followed the waved courses of the valleys, thus the great 
 block of the Seeberg, which is 60,000 cubic feet in size, and 
 which came from the valley of Binnen, changed its course 
 during its progress of sixty leagues at least five times, two at 
 least of which were in the reverse direction to the others. 3 
 So much for Von Buch's special' theory. 
 
 We will now revert somewhat, and turn to Sir James Hall 
 whose remarkable memoirs on Chemical Geology formed an 
 epoch in the history of the science. In a paper on the 
 revolutions of the earth's surface, read on March the 16th, 
 1812, and published in the seventh volume of the Trans- 
 actions of the Royal Society of Edinburgh, he examines at 
 some length the question of the origin and distribution of 
 boulders. The paper is full of suggestive thought. He 
 framed his conclusion on a wide induction, and although 
 his principal data were collected in South Scotland, he 
 applies his theory to the boulders of the Jura, and refers us 
 for other examples to the granite boulders at Stainmore in 
 Cumberland, which he traced to Wast-dale-crag and to a great 
 block, four or five feet in diameter, in the streets of Darlington. 
 
 6 Charpentier, op. cit. 197-200. 
 
Hall. 49 
 
 He attributes the results named to the action of a great 
 diluvial wave. In the case of the largest blocks, he suggests 
 their having had an increased buoyancy by being attached 
 to ice. "The enormous masses already mentioned," he says, 
 " which are found near Geneva and at the Coteau de Boisy, 
 may now be accounted for ; and the same system will 
 apply also to the blocks on the Baltic, which may have 
 been brought to their present place, not by a permanent and 
 steady position of the ocean, varying by slow degrees, as has 
 been alleged by M. Wrede, but by a sudden diluvian wave 
 washing over some district, situated either at a sufficiently 
 high level or near enough to the Pole to be the seat of 
 glaciers." 7 
 
 Speaking of the vast deposit of sand in Holland and that 
 part of Europe, Sir James Hall says, " Its magnitude ap- 
 pears to me very far to surpass any deposition that could 
 reasonably be ascribed to the present rivers. All this sand 
 may be conceived to have been hurried along, by that mighty 
 stream, (id est the diluvial wave) and deposited when the 
 torrent began to spread, and lose its force by diffusion" 8 
 Hence why the blocks in Germany and the Baltic islands are 
 found on the hills of diluvial sand. After describing in detail 
 a number of boulders occurring near Edinburgh, he says, " I 
 have thus pointed out fifteen specimens, within a circle two 
 miles in diameter, surrounding Corstorphine Hill, each ex- 
 hibiting both the large and the small features which indicate 
 the action of water flowing with violence along the surface 
 and carrying large blocks of stone along with it. The number 
 of examples concurring in the same indications of direction 
 might easily have been greatly augmented, as any person will 
 see who examines the summit and western face of the hill." 9 
 
 Sir James Hall was, I believe, the first to describe in detail 
 the phenomena known as crag and tail. When an obstacle 
 whose height is equal or greater than its breadth, stands 
 in the way of a running stream, it generally causes a stagna- 
 tion on the side towards which the stream is flowing, and a 
 deposition is formed of the transported substances, consti- 
 tuting a tail or prolongation which extends in the direction 
 
 ? Op. cit. 158 and 159. 8 Op. cit. 159. 9 Id. 191-2. 
 
 E 
 
5<D The Glacial Nightmare. 
 
 of the stream, by a gradual descent, to the distance frequently 
 of eight or ten times the height of the obstacle, while a 
 hollow or depression is frequently caused on the other side of 
 the obstacle by the accelerated motion of the stream there. 
 The isolated hill of North Berwick Law has a conspicuous 
 tail running eastwards. A current of water acting upon sand/ 
 or of wind upon snow, would equally produce these effects. 
 He then applies this principle to the geological facts pre- 
 sented by the hills near Edinburgh, and shows how, in 
 each case, such a tail can be traced on one side, and on the 
 other frequently a hollow marked by a lake. In addition he 
 refers to the numerous parallel ridges occurring in that 
 district. Thus, he says, tf From Corstorphine Hill to the 
 eastward, the country embracing all the space between 
 Edinburgh and the sea presents one continued series of 
 ridges, upon one of which the New Town of Edinburgh 
 stands. It is an important circumstance that these ridges 
 maintain a very correct parallelism with each other, with the 
 tail of the Castle rock, and of the Calton Hill, and with the 
 alluvial prolongations that extend to the eastward from all 
 the eminences of this neighbourhood ; and a series of 
 parallel ridges occur also on the south side of Edinburgh, 
 extending from all the rocky eminences, as may be well 
 seen on the road leading to Dalkeith, which passes over 
 several of them; one of the most remarkable of which is 
 that on which the village of Libberton stands/' Hall urges 
 that these phenomena could not have been caused by the action 
 of diurnal waters flowing by the action of gravity, which would 
 have produced depositions at right angles to those under con- 
 sideration, and while diurnal waters are everywhere found in 
 the act of corroding and altering these forms, they are no- 
 where seen to produce them, and he insists that they are con- 
 sistent only with the action of a diluvian wave. 
 
 He further urges that small streams have a meandering 
 shape, while the long ridges alluded to have a short curvature 
 whose radius is of infinite length, whence we must infer that a, 
 cause very different from any now in activity, and far more 
 powerful, has exerted its influence on the spot, and that a 
 
 1 Id. 171, 
 
Hall. 51 
 
 current lias once flowed there capable of overwhelming and dis- 
 regarding objects by which the Nile and the Ganges 
 would have been turned out of their course. 2 
 
 Flowing water, he says, rounds the original angles of rocks 
 over which it flows, not only the prominent but also the 
 entering angles, the water excavating for itself a waving 
 groove more or less longitudinal. It also excavates the 
 surface into shallow depressions, which he calls scoopings, like 
 those which would be produced on a soft body by the oblique 
 blow of a spoon or scoop. These are probably caused by the 
 action of eddies of inferior force to the main stream, but 
 acting in connection with it in different and sometimes opposite 
 directions. These various corrosions going on together, have 
 each produced its particular effect, and most of them being 
 concave, their meeting has given rise to the set of waving 
 angular ridges which constitute the most unequivocal 
 feature of a water-worn rock. These angular forms differ 
 completely from those which occur in the broken surface of a 
 rock. These last are acute, rectilinear, and abrupt ; while 
 the others are continuous, flowing, and have their angles 
 very obtuse ; so obtuse, in some cases, as not to be visible, 
 unless the light strikes upon the rock in a particular 
 direction. 3 These forms, he says, occur on the surface of 
 water-worn rocks, and also on the surface of a mass of snow 
 acted upon by wind. The very rocks over which, according 
 to his theory, Hall says " torrents of water have flowed loaded 
 with sand and gravel and large stones, and accompanied by 
 streams of mud, are found to exhibit at their surface all the 
 characters of abrasion above mentioned, the rotundity and 
 flowing character, the excavation of hollows into the form of 
 waving grooves, the concave scoopings, and the obtuse-angled 
 and waving ridges." 4 
 
 Again he says, " It is obvious that the power of mud, when 
 flowing as a stream, in transporting heavy bodies, and in abrad- 
 ing assemblages previously formed, must bear some relation to 
 the resistance which it would oppose to any object forced 
 through it, and of course that its power in these respects must 
 have been much superior to that of pure water. I conceive also> 
 
 2 Id. 178, 179. 3 Id. 179, 180. Id, 180. 
 
 E 2 
 
52 The Glacial Nightmare. 
 
 that a deep stream must exert a greater power of transporta- 
 tion and corrosion than one which is shallow, flowing with the 
 same velocity." 5 
 
 He also attributes the grooves and scratches occurring on 
 the surface of certain rocks to stones of considerable bulk 
 carried along by torrents and acting as grinders, he 
 quotes the case of a sudden torrent of water loaded with 
 stones having assailed a house and gone through its lower 
 story, and he observed that every stone, as it passed the 
 house, left a rut or scratch behind it, upon the flags over which 
 it passed. 6 
 
 He then goes on to point out that when the mantle of 
 clay, etc., which covers some surfaces is removed, we find the 
 surface to resemble that of a wet road, along which a number 
 of heavy and irregular bodies have been dragged ; indicating 
 that every block that passed, and every one of its corners, 
 had left its trace behind it. These are rendered very dis- 
 tinctly visible when the surface is drenched with water. 
 " In some cases/ 7 he says, " these grooves have resisted all the 
 effects of the weather, and show themselves in rocks that have 
 been always exposed, sometimes for many yards ; occasionally, 
 single scratches, and parallel sets of them, deviate by five or six 
 degrees from the general direction, but such deviations are 
 rare, the majority agreeing in parallelism with each other and 
 with the general direction of the larger scoops and grooves, 
 and also of the ridges and larger features of the district/' 
 
 Lastly, he calls attention to the heterogeneous and un- 
 stratified character of the loose deposits as being consistent 
 only with a diluvial origin. He points out that a torrent, 
 such as he postulates, loaded with a multitude of blocks of 
 every size, shape and quality, and with a quantity of clay 
 (soon reduced to mud, through which the stones would be 
 irregularly and confusedly scattered), would deposit its 
 materials very much as we find them. The body of compact 
 blue clay which we find in many places, some of which he 
 specifies, is totally devoid of stratification, though frequently 
 of great thickness, and bears every indication of having 
 flowed as a mass into its present situation, together with 
 
 * Id. 174 6 Id. 182. 
 
Hall. 53 
 
 the mixed assemblage of blocks of all sizes and shapes and 
 qualities, some sharp and some round, confusedly scattered 
 through it. He urges that these facts are inexplicable by any 
 diurnal cause, and seem due to a torrent, some parts of which 
 encountered less of the strata of shale and clay, deposited the 
 comminuted sandstone in the form of sand and of gravel, 
 and gave to the external contour of the deposits a smoothness 
 and regularity forming a striking contrast with the abrupt 
 and irregular dislocation which very commonly occurs in the 
 solid mass within. 7 
 
 Sir James Hall was very emphatically of opinion that the 
 phenomena he thus describes were caused by one great catas- 
 trophe and not by several. " Thus/' he says, " all the diluvian 
 facts in this neighbourhood (i.e. Edinburgh) that have come 
 under my observation concur in denoting one inundation over- 
 whelming the solid mass of this district, which had been 
 elevated into its present position by some still more ancient 
 revolution of the same sort ; this inundation being the last 
 catastrophe to which it has been exposed .... my theoreti- 
 cal notions limit the action upon the hill under examination, 
 to the passage of a single wave, embracing a period of time 
 that could only be expressed in minutes ; but during that 
 short time, I conceive the water to have been urged forward 
 with such force, and to have carried with it so many power- 
 ful agents, that it has produced effects equal to the work 
 of ages under other circumstances." 8 In regard to the 
 extent of this flood of waters, he concludes, from the way 
 in which it seems to have passed over Corstorphine Hill 
 without being deflected or affected by it, that it must have 
 been incomparably superior in magnitude to the Nile or the 
 Ganges. "This hill/' he adds, " is stated to be four hundred 
 and seventy feet above the sea, and postulating that the stream 
 was double this depth, or 1000 feet, we should have a height 
 sixteen times that of the wave at Cadiz, which was 60 feet/' 
 The phenomena of the Alps, he says, indicate a magnitude 
 double of this. 9 In regard to the direction of the great wave 
 in the neighbourhood of Edinburgh, as indicated by the 
 medium result of a number of observations, he urges that it 
 
 * Id. 170-183. 8 Id. 105, fl Id. 192 and 193. 
 
54 The Glacial Nightmare. 
 
 appears to have been from 10 S. of W. to 10 N. of E. by its 
 true bearings taken with a needle, and allowing 27^ degrees 
 west of north as the variation. He had met with no case 
 deviating more than 10 or 12 from that average on either 
 side. 1 In other places this direction was not quite the 
 same. Thus, he says, in the case of the diluvian ridges called 
 Lowry's knolls, situated behind the promontory of Fast Castle 
 in Berwickshire, the direction is west 35 north, thus forming 
 an angle of 45 degrees with the general direction of the 
 estuary. 2 Near Stirling Castle, again, Hall noticed a set of 
 alluvial ridges of a true diluvial character which agree with 
 the surface marks of the rocks in having a direction no 
 less than 50 degrees north of west or south of east, and being 
 thus very different to that observed near Edinburgh. 3 
 
 The analogy of the events at Lisbon and Lima led him to 
 suppose, he says, that several similar actions may have taken 
 place in succession. At Lisbon five waves in succession followed 
 the first. In the present case the influence of all but the first, 
 would probably be confined to the western side ; or, if any of the 
 rest possessed elevation sufficient to surmount the ridge, still 
 the portion of its water which passed over, and that which 
 returned, would follow the same course with the correspond- 
 ing portions of the first wave. And there is reason to believe 
 that the action would be often repeated upon the side facing 
 the inundation. 4 
 
 In regard to the south-west of Scotland and Cumberland, 
 Hall noticed the absence of the straight and parallel ridges 
 and the forms of crag and tail, the diluvial debris being 
 there arranged in a perpetual succession of knolls equally 
 round at both extremities. To account for this contrast, he 
 postulates that the diluvian wave flowed from a westerly or 
 north-westerly direction and broke over the island, the great 
 body of its water crossing the ridge of country separating the 
 two coasts, overwhelmed the district near Edinburgh and 
 discharged itself into the German ocean. Another portion 
 of the wave being detained by that ridge of country, and 
 being left to follow the impulse of gravitation, when the 
 superior water flowed on, returned again upon the countries 
 
 1 Id. 183 and 184. 2 Id. 199. 3 Id. 200. 4 Id. 202, 203. 
 
Allen and Drake. 55 
 
 which it had just crossed, like the back draught of a common 
 wave. The eastern coast would thus be exposed to a single 
 diluvial action, while the western would undergo two in 
 succession, and nearly in opposite directions. " The first wave 
 would create the phenomena of parallel ridges and crag and 
 tail everywhere, the second and perhaps other subsidiary 
 waves would when they operated, break up the ridges into 
 successions of knolls, etc., and the crags would be filled up so as 
 to equalize the declivity on all hands. Thus the phenomena 
 in the west of Scotland would be accounted for, and this is 
 consistent with what we know of the grooves and furrows 
 there, which instead of keeping a uniform direction, follow 
 directions so irregular as to baffle all arrangement, or if any 
 such can be traced, it is found to be entirely local and to 
 depend upon the course of the valleys in the immediate 
 neighbourhood/' 5 
 
 In assigning a cause for this diluvian action, Hall appeals 
 to the elevation of the land, which according, to him, " was 
 performed by successive starts, similar to volcanic eruptions, 
 though far more rare and more powerful. The percussions 
 impressed by these starts upon the waters of the ocean were 
 such as to form waves, sometimes of a moderate force, as those 
 at Lisbon or Callao, sometimes of overwhelming magnitude, 
 and capable of producing the effects described in the Alps, 
 in Germany and in Prussia." 6 
 
 Some of the phenomena observed by Hall in Scotland 
 were also noticed in the Faroe islands in the same year, by 
 Allen, who speaks of the smooth surface of the mountains ; 
 and notices that near Eide the rock was scooped and scratched 
 in a very wonderful way not only on the horizontal surface, 
 but also on a vertical one of 30 or 40 feet high, which had 
 been opposed to the current. 7 
 
 The phenomena we are discussing were now beginning to 
 attract attention on the other side of the Atlantic. Thus we 
 find Drake writing in 1815 : " It is familiar to all persons in 
 any degree versed in geological science that granite gneiss, 
 mica slate and other rocks termed primitive are naturally 
 inferior in situation to all the strata that have been described. 
 
 5 Id. 203 and 204. 6 Id. 166-7. 7 Id. 265. 
 
56 The Glacial Nightmare. 
 
 In the western part of Ohio, these stones are found on the 
 surface of the ground, or partly embedded in the layers of soil 
 and loam. They are sometimes solitary ; at other times a great 
 number of masses may be seen collected together and piled 
 on each other, as in the township N.W. of the village of Eaton, 
 Preble county. They are of irregular shapes, and of various 
 sizes. The largest I have ever met with is in the town just 
 mentioned. It is composed of quartz and mica, and was 
 estimated by Mr. Jesse Embree, who ascertained the dimen- 
 sions of that part which rises above the ground, to contain at 
 least 300 cubic feet. The strata underneath are secondary 
 siliceous limestone. These fragments of primitive rocks are 
 said to be scattered extensively, over the State of Ohio, the 
 Indiana territory and Kentucky." In regard to the force 
 by which these blocks were scattered, Drake contests the 
 notion of Kirwan that they were ejected by volcanoes, since 
 the masses are too numerous and too large, their surfaces 
 too free from vitrification, and their distribution too much in 
 groups, nor does America contain any volcanoes or obsolete 
 craters. He favours the notion that at some former period 
 there were currents over this part of the continent from north 
 to south. " By these currents the masses of primitive stone 
 might perhaps have been brought down in cakes of ice, and 
 deposited where they are now found." 8 
 
 In the year 1818, Mr. Weaver read a paper before the 
 Geological Society of London on the Geological relations of the 
 East of Ireland. In this he- urges that it is impossible to reflect 
 upon the nature and disposition of the alluvial tracts on the 
 surface of the earth, without recognizing the powerful agency 
 of an agitated fluid in a state of retrocession ; inter alia he 
 mentions how Cronebane hill, which is made of slate, bears 
 upon its summit a boulder of granite (called Motty Stone), 
 9| feet high and 42 feet in circumference ; while the sides of 
 the hill are also strewed with boulders of granite, some of 
 which are not much inferior in dimensions. " How then did 
 these attain their present position ? The nearest granite 
 rock is that which extends from the eastern bank of the 
 Avonmore towards West Aston hill; but this is very dis- 
 
 8 Pictures of Cincinatti, by Daniel Drake, pp. 74 and 75. 
 
Weaver. 5 7 
 
 similar in aspect to the granite boulders on Cronebane. The 
 next granite, in point of distance, is that of BalKncarrig on 
 the banks of the Avonbeg ; yet in both these instances (with the 
 exception of West Aston hill itself), the granite rock is found 
 in a situation several hundred feet lower than the summit of 
 Cronebane, which stands nearly upon a level with the bottom 
 of G-lenmalur; but the interval between these two places 
 forms at present a curved line descending from that level 
 until it obtains a depth of nearly seven hundred feet below, at 
 the meeting of the two Avons." 9 Mr. Weaver goes on to urge 
 that these rocks were moved to and left in their present 
 position by the receding waters of a flood. Further on 
 he asks, ( ' Whence are the limestone, gravel and marl derived 
 which we find distributed along the coast of the counties of 
 Wicklow and Wexford ? The nearest visible limestone rock 
 on the northern quarter, is that which occurs at Williams- 
 town and Booterstown, on the southern side of the Bay of 
 Dublin, and to the southward the first rock of this description 
 that appears, is on the coast south of the town of Wexford. 
 .... It is worthy of remark that many of these deposits 
 of limestone-pebble, gravel and marl are situated at dis- 
 tances from two to ten miles from the nearest part of the 
 continuous calcareous tract ; and at an elevation reaching to 
 200, 300 and 400 feet higher than the existing surface of 
 the limestone rock itself ; as, for instance, in the glen of Imale, 
 and in the upper part of the Aghfarrel stream, the Liffey and 
 King's rivers. The occurrence of debris of red sandstone 
 conglomerate at an elevation considerably higher than the 
 actual fixed position of this rock is an analogous case." 1 
 
 " If we examine the borders of the mountain chains on the 
 west and the south, we observe similar relations, and in the 
 valleys which find their outlet in limestone plains calcareous 
 debris may be remarked intermingled with those of the 
 mountain tract, in some cases at the distance of two or 
 three miles from the plain itself, and considerably elevated 
 above it, while higher up they disappear altogether." 2 Mr. 
 Weaver elsewhere points out that in descending the declining 
 surface of the coal district of Lisnarnrock in Leinster, the 
 
 9 Trans. Geol. Soc. v. 294-297. l Id. 298-300. 2 Id. 300. 
 
58 The Glacial Nightmare. 
 
 hollows are found partly occupied by a deposit of limestone 
 gravel, mixed with debris of sandstone and shale, which 
 gradually increases in thickness as we approach the lower 
 grounds. In this manner, even large boulders of limestone 
 are found in the hollows at the distance of two miles from 
 the limestone flat ; 8 and summing up the result of his 
 observations, he says, "it is evident that the cause which 
 produced these phenomena was posterior to the formation of 
 all these rocks, since we find the debris of limestone not only 
 resting high up, on the borders of the older tracts, but also 
 intermingled with the debris of the newer, and covering, in 
 fact, the. more elevated surface of the latter. The limestone 
 field itself abounds in rolled calcareous masses, pebbles, 
 gravel, sand and marl, often raised into hillocks, or long 
 extended ridges ; which seem to owe their form to the action 
 of eddies, and opposite currents. There is scarcely any part 
 of the extensive limestone tract that is not more or less 
 marked by them. Sometimes these ridges appear like regular 
 mounds, the work of art forming a continuous line of several 
 miles in extent. That which passes by Maryborough in the 
 Queen's county, is a remarkable instance of the kind ; and 
 similar mounds, hillocks and ridges occur also in the counties 
 of Meath, "Westmeath, Kildare, Carlow and other portions of 
 the limestone field, in which the calcareous gravel and sand 
 frequently exhibit a stratified disposition, the alternate layers 
 being very distinct from each other. 4 
 
 Contemporaneously with the paper just cited, giving the 
 first details about the Irish drift phenomena, Colonel Imrie 
 published a paper on the Geology of the Campsie Hills in 
 Scotland, in the Transactions of the Wernerian Society. In 
 this, he says, inter alia, " In all situations of this district, 
 where the trap has disappeared, the vegetable or surface soil 
 rests upon a strongly tenacious clay, much mixed with water- 
 worn stones ; and this blue clay rests upon sandstone, 
 Among the water- worn stones embedded in the clay, I sel- 
 dom found specimens of the native rocks of the district ; 
 those which I examined, consisted mostly of rocks generally 
 deemed of the oldest formations, such as quartz, porphyries, 
 
 1 Id. 293. 4 Id. 300, 301. 
 
Imrie. 59 
 
 granite, etc. ; the native beds of which, are far distant to the 
 north and west of that part of the country/' 5 He attri- 
 butes the disappearance of the trap here mentioned to the 
 effects of the attrition of heavy bodies set in motion by a 
 great force of water in rapid movement, and says that in 
 certain places in the glens and narrow vales where it had 
 not entirely disappeared, he perceived upon its surface strong 
 indications and marks of attrition. " In some places the 
 surface of the trap was smooth and had evidently received 
 a considerable degree of polish ; and this polish was almost 
 always seen marked by long lineal scratches. In other 
 places, there appeared narrow grooves, apparently formed by 
 the rapid movement of large masses of rock having been 
 swept along its surface ; and I remarked that these striae or 
 scratches, were very generally, in a direction from west to 
 east, excepting when inequalities of the surface, and sudden 
 turns in vales had partially influenced the course of the 
 current. In the eastern part of the district, there occurs a 
 small elevated plain, slightly undulated. Here the surface 
 of the trap, in some places, had lost its covering of soil, and 
 was left bare for inspection. Upon this plain I again detected 
 some of these scratches which were in this situation, 
 generally in a direction from west to east, sometimes deviating 
 a little towards the south-west and south-east; but in taking 
 the average or medium of these directions, their general and 
 true course seemed to me to be from west to east. Upon 
 the surface there were scattered immense masses of trap, 
 which from their apparent weight, seemed perfectly capable of 
 forming the scratches and grooves above described, had 
 they been put in motion and impelled along the surface. 
 Upon examining some of these huge masses I found their 
 surfaces scratched and worn in such a way as to prove 
 sufficiently indicative to me, that they had been long sub- 
 jected to attrition in water, and I also observed that many of 
 them presented their principal or most projecting angle 
 towards the west, and sometimes towards the north-west, 
 which, according to my opinion, strongly implies the direction 
 of the current which left them in the position in which they 
 
 5 Mems. Wern. Soc. ii. 35. 
 
60 The Glacial Nightmare. 
 
 now are. . . That such currents as were capable of the effects 
 which I have endeavoured to describe, have flowed over the 
 surface of our globe, is to me clearly evident, and the 
 scratches and grooves here mentioned are some of the minor, 
 but clear proofs of its action." 6 
 
 In Dr. Mitchell's appendix to the American translation of 
 Cuvier's works, published in the same year, 1818, and entitled 
 " Observations on the Geology of N. America," we find him 
 arguing that the inland lakes of America are the remains of a 
 once vast and continuous sheet of water. This water, he urges, 
 burst its barriers at various points and thus distributed the 
 North American drift. Speaking of one of these breaches he 
 says, " The thousand islands and the whole of the scenery in 
 their vicinity, bear witness to the mighty rush of waters 
 which at some former period prostrated the opposing mounds 
 and left them as scattered monuments of the ruin. . . Great 
 masses of primitive rocks from the demolished mound or dam, 
 and vast quantities of sand, mud and gravel, were carried 
 down the stream to form the curious mixture of primitive 
 with alluvial materials in regions below/ 7 7 Speaking of an- 
 other breach, namely, that made by the Hudson, he quotes 
 Governor Chutons thus : " As you approach the falls you pass 
 through immense rocks, and actually of granite interspersed 
 with limestone. In various places you observe profound exca- 
 vations in the rocks, made by the agitation of pebbles in the 
 fissures. . . . The rocks are composed of granite and many of 
 them are 30 or 40 feet thick. You see them piled on each 
 other like Ossa on Pelion ; and in other places huge fragments 
 scattered about, indicating evidently a violent rupture of the 
 waters through this place as if they had been formerly dammed 
 up and forced a passage ; and in all directions you behold 
 great rocks exhibiting rotundities, points and cavities, as if 
 worn by the violence of the waves or hurled from their ancient 
 positions/' Dr. Mitchell similarly refers to breaches which he 
 argues were made through the barriers in question by the 
 other rivers of the States. He urges that another vast dam 
 confining the waters in former ages extended from the 
 extremity of the Cumberland mountain to the Missouri hills, 
 
 6 Id. 36, 37. * Op. cit. 334-5. 
 
Mitchell. 61 
 
 and that its demolition caused the vast tract behind it to be 
 drained, leaving lakes Erie, Huron, and Michigan as relics, 
 while the rivers of sand and soil were carried down the Missis- 
 sippi valley. 8 Later in the same work, after describing 
 the remains of shells and other marine debris found in 
 the loose beds, he reverts again to the travelled stones and 
 mentions the large number of loose rocks rolled and worn, 
 consisting mostly of granite and gneiss, of actinolite and black 
 schoerl, etc., etc., found abundantly on Long Island, New 
 York. Their great weight and bulk must, he says, have re- 
 quired extraordinary power to detach them from their primi- 
 tive beds and put them where they are. On the ridges of 
 hills traversing the island he noticed that while the north side 
 was occupied by great masses of these stones, not a pebble as 
 large as an egg existed on the south, and he suggests that they 
 were dispersed by the waters of the lake forcing its barriers, 
 " pouring over Joe's Hills, and carrying along theloose materials ! 
 sand and stones being urged to the greatest distance, moder- 
 ately weighty rocks not quite so far, and the heaviest ones 
 loitering in the rear or concealing themselves under the 
 water of the sound where they annoy ' navigators." 9 An- 
 other portion of the flood, loaded with rocks, stones, sand, and 
 water was, he suggests, apparently driven along the valley 
 between Bergen and Newark, and rolled up a huge pile of 
 materials on the north side of Staten Island ; while another 
 mass of dislodged materials appears beyond the Rariton in the 
 form of the Neversink Hills, on the summit of which are pieces 
 of sandstone, some many feet in length and of considerable 
 thickness, none of it in situ. 1 
 
 Later on he says : " The ground on which New York stands 
 is sand, gravel, rounded stones fit for paving, and loose rocks, 
 some of them of enormous magnitude .... the rounded 
 rocks are sometimes six feet in diameter .... they consist 
 of rolled or rounded masses of schoerl rock, quartz and schoerl, 
 stellated asbestos, granite and gneiss, and schist. These 
 boulders, etc., are similarly arranged as those already described 
 beyond the Highlands, and the boulders can only be viewed as 
 members torn by violence from the body to which they were 
 
 8 Op. cit. 332-345. 9 Id. 382. Id. 383-4. 
 
62 The Glacial Nightmare. 
 
 once attached." Dr. Mitchell, in accounting for the cause by 
 which the barriers of the great sea were broken, assents to the 
 view of a diluvial catastrophe, and speaking of America he 
 says : " What agent so capable or so likely to wash up the 
 sand and other materials into such ridges as our mountains 
 present? The impulse of an ocean upturned from its bed, 
 rolling impetuously over the land, and carrying everything 
 before it, may be supposed competent to such a work." 2 
 
 Mr. Greenough, then president of the Geological Society, 
 writing in 1819, calls attention to the wide distribution of 
 the drift phenomena. The boulders, he says, do not occur 
 only in Switzerland and the Baltic lands. " At Glenmalur, in 
 the county of Wicklow, huge masses of granite rest upon the 
 mica slate. Along the valleys of the Garonne and Gave de 
 Pau, you find granitic blocks derived from the Pyrenees ; 
 along the valleys of the Aveyron and Dordogne from the 
 Cevennes; at Bains from the Vosges. In the department of 
 Morbihan, the number of blocks is estimated at 4000, and 
 some of them are not less than 20 feet high. Near Turin, the 
 calcareous hills are covered by blocks of granite, some of them 
 of the size of 30 cubic feet, although no mountain of that 
 substance is found within many leagues. Gerenna in 
 Grenada, is famous for its boulders, which suggest the idea of a 
 shower of stones ... I am informed that blocks of granite 
 extend for more than one hundred miles on the south of lako 
 Huron, in North America, and appear in islands twelve miles 
 from its margin. Granite boulders, therefore, are not of 
 partial occurrence, nor is the theory tenable which supposes 
 those found in the north of Germany to have slid thither upon 
 the ice." . . . 
 
 After quoting with approval Hall's answer to Hutton in 
 regard to the incapacity of rivers to move the Alpine blocks, 
 Greenough adds, " The phenomena of the blocks which have 
 been traced to the northern parts of Europe, are still more 
 adverse to the hypothesis of Hutton. This, not merely on 
 account of the vastly greater distance to which they must 
 have been carried, but also on account of their lateral ex- 
 tension. They are scattered," he says, " over the continent 
 
 2 Id. 410. 
 
Grcenough and V.on der Lint/i. 63 
 
 from Holland to Petersburg and Moscow. It is perfectly 
 incredible, that these blocks, extending over so immense an 
 area, and found on the opposite sides of lakes and seas, should 
 be the waifs and estrays, either of a single river, or of any 
 number of rivers." 3 
 
 Greenough, after discussing the efficacy of seas and rivers to 
 produce the phenomena of the drift, and discarding them as 
 inadequate, says the only remaining cause to which these 
 effects can be ascribed is a Debacle or Deluge. 4 He 
 goes on to say, " The universal diffusion of alluvial sand, 
 gravel, etc., proves that at some time or other, an inundation 
 has taken place in all countries ; and the presence of similar 
 alluvial deposits, both organic and inorganic, in neighbouring 
 or distant islands, though consisting often of substances 
 foreign to the rocks of which the islands are respectively 
 composed, makes it highly probable, at least, that these 
 deposits are products of the same inundation. " 5 As to 
 the direction of this flood, he says, inter alia, " blocks from 
 the Cambrian mountains have travelled eastwards as far as 
 Pierce Bridge, and northwards as far as Staffordshire. Chalk 
 flints occur in alluvion on the north coast of Cornwall, and 
 even at the Land's End." 6 He elsewhere refers to the 
 parasitic gravel and soil of Malta and Gozo, the blocks of 
 primitive Norwegian rocks scattered over north Germany, 
 Russia, Holland, and occasionally met with on the east coast 
 of England, and pieces of granite on Staffa as evidences of 
 the non-existence at the time of their dispersal of the neigh- 
 bouring seas or lakes. 7 
 
 In 1819 M. Escher Von der Linth published a memoir on 
 the Natural history of the Erratics, which was printed in the 
 first volume of Leonhard's Taschenbuch, 631, etc. In this 
 memoir he developed the notion that the cause of the dispersal 
 of the blocks was the bursting of huge Alpine lakes ; one of 
 these, he considered, filled the Valais or upper Rhone valley to 
 a depth of 5000 feet, and was dammed in by a vast barrier 
 above St. Maurice. This, he says, would constitute a lake 
 containing 50,000,000,000 cubic fathoms, each 1000 feet in 
 extent. If such a lake were to empty itself at the rate of 
 
 3 Examination, etc., 137, 138. 4 Id. 149. 6 Id. 155. 
 
 6 Id. 159. 7 Id. 177-8. 
 
64 The Glacial Nightmare. 
 
 500 feet a second, it would discharge 25,000,000 cubic fathoms 
 per second, and take half an hour to empty. He compares 
 the results of such an outburst with that which followed the 
 breaking loose of the lake in the Bagnes valley in the Yalais 
 in 1818 by the giving way of a glacier, and urges that the 
 effects would be such as we see in the distribution of the 
 blocks and debris of the Alps. 8 
 
 In order to explain the fact that the debris from different 
 valleys is not mixed, but sorted separately, he urged that 
 the barriers which formed dykes to the several mountain 
 lakes burst simultaneously, and thus the current in each 
 valley prevented that in the adjoining valley from interfering 
 with it. Charpentier, in answering this view, argues that the 
 outer chains of the Alps which would form such barriers, are 
 calcareous, and as the lakes must have been very deep, they 
 would necessitate correspondingly deep and wide barriers, 
 which would leave immense masses of debris when dis- 
 integrated, a result which we nowhere find. The phenomena, 
 again, which we are discussing is not limited to the Swiss Alps, 
 but the Alps of the Tyrol, the Bavarian and Austrian Alps 
 present similar effects as do the Italian Alps. All of these must 
 have had barriers, which is a very improbable fact, and one not 
 supported by other evidence. How, again, can we account for 
 all these barriers having been ruptured at the same moment ? 
 Even if they were the result of the bursting of a barrier, the result 
 is very different to that which we find in the Alpine valleys. 
 When a breach occurs in such a barrier, the effects of the rush 
 of water are only traceable immediately in front of the breach 
 and for a short space. Within certain limits it sweeps every- 
 thing away, and carves out deep ravines, and when its force 
 diminishes it deposits debris along its course, but this is 
 outside the barrier. Inside the barrier, and within the lake 
 basin itself, there are no such results. There the only result is 
 the lowering of the level of the lake. All this was amply 
 shown in the case of the rupture of the glacier restraining the 
 lake du Gietroz in the valley of Bagnes on the 16th of June, 
 1818; the current of water which poured out was three 
 quarters of a league wide and 200 feet deep. The devasta- 
 
 8 Op. cit. 665, etc., etc. 
 
Fox Strangways. 65 
 
 tion caused by the progress of this column of water outside 
 the barrier was terrible, but inside, the basin of the lake it- 
 self suffered no denudation. Except close to the breach, the 
 water did not displace any of its bed, which had once been a 
 wide pasture, and the Chalets of Torrembey, built of rude 
 stones without mortar, and which the lake had submerged, 
 situated at a distance of two gun-shots above the breach 
 in a line with the direction of the current, were not des- 
 troyed. It is clear, therefore, that a rapid fall in the level 
 of a lake would have but a small effect on its floor or on the 
 basin in which it is situated. We cannot, therefore, account for 
 the distribution of the great erratics in this way. And if we 
 could do so, as M. Escher supposes in the Yalais, by pos- 
 tulating a barrier between the Dent de Morcle and the 
 Dent du Midi near Saint Maurice, this would not account for 
 the blocks in the valleys of Avan9on, of La Grand Eau, of 
 Illiez and de la Tiniere, which open into the Rhone valley 
 below St. Maurice. The famous block of Devens, 161,000 cubic 
 feet in extent, came out of the valley of Avan9on. 9 
 
 On the 16th of April, 1819, Mr. Fox Strangways read a 
 memoir before the London Geological Society on the Geology 
 of the district round St. Petersburg. In this he speaks at some 
 length of the so-called Diluvium of that district, a term which, 
 he says, expresses that superficial deposit which covers every- 
 thing and is composed of almost everything. He remarks 
 that it is usually thickest in the valleys or on the flat 
 summits of some of the hills, and is composed of two sets of 
 materials, one local and another of " a set of rocks totally 
 foreign to the cor.ntry, the analogues of some of which have 
 been recognized in situ at a vast distance, while others remain 
 yet to be identified. . . . All the primitive boulders, the 
 analogues of which have been recognized in situ, appear to 
 have been transported from the north. They are scattered 
 universally over hill and dale ; on the highest summits of the 
 hills, and on the shoros, and under the waters of the gulf ; 
 and are found not only on both sides of the Neva, but far to 
 the south, to the east and to the west. Their water-worn 
 surfaces, their enormous size, and the remoteness of their 
 
 9 Charpentier, op. cit. 202-207. 
 
6 6 The Glacial Nightmare. 
 
 ancient seat, seem to leave no other supposition to account 
 for their appearance in their present situation, than that of 
 their being gigantic pebbles, rolled and accumulated by the 
 greatest aqueous revolution to which this globe has ever been 
 exposed, the Mosaic deluge, of which they furnish the most 
 striking of the material evidences. . . . The environs of 
 Petersburg, as well as every other country, show those traces 
 of the superficial action of vast currents of water which can 
 only be attributed to the same cause; for instance, the 
 Kopsha and Crasnoe Celo denudations and the numerous 
 combes and valleys now destitute of water, along the edges of 
 all the escarpments. . . . One of the most remarkable circum- 
 stances respecting the boulders of this stratum, is, that rolled 
 pieces of the smaller-grained chamite beds are found on the 
 summit of a steep ridge forming the extreme point eastward 
 of the heights of Shulcowa. . . . The great elevation of this 
 ridge, the geographical distance from the nearest point where 
 the chamite beds appear in situ, and the great thickness of 
 the limestone in that part of the district, make it a question of 
 no easy solution, how rolled fragments of a rock three hundred 
 feet or more below that on which they are found lying, could 
 ever have arrived at their present situation." The limestone 
 boulders are never found north of the outcrop of that stratum 
 whence they are derived and rarely on the table-land itself. 
 They increase in numbers towards the south. On the surface 
 of the limestone plateau, particularly south and west of the 
 upper lake of Doudorof, are several low ridges, which appear 
 to be entirely composed of heaps of rolled pebbles of various 
 descriptions. They are on a large scale, exactly like what 
 are deposited by counter currents and eddies in running 
 water. 
 
 " At the quarries near Mishkina .... the surface of the lime- 
 stone, when scraped bare of superficial accumulations which 
 are ten feet thick, presents hollows and steps from one bed to 
 another, the edges of which are all rounded as if water-worn. 
 The diluvium in this instance is entirely free from stones. 
 
 " From the situation of the parent rock of such of the primi- 
 tive boulders as yet have been identified, and from the increase 
 in quantity of the secondary pebbles on approaching the south, 
 it seems that the great debacle, or stream of transportation, 
 
Razoumofski. 6 7 
 
 came from the north; while from the great height at which 
 the rolled blocks are found, it is evident not only that portions 
 of rock, whence they have been brought, once existed at a 
 higher level, but also that they were deposited in their present 
 situation before the excavation of the Gulf of Finland and the 
 vale of the Neva. These great denudations, together with 
 those smaller ones that open into them, were not hollowed 
 out till the diluvian waters began to draw off and drag the 
 slopes of what had been the bottom of the ocean. The heavy 
 blocks had already settled and the streams carried back with 
 them only the softer materials. This accounts for the seeming 
 incongruity in the direction of the diluvial currents ; just as 
 a great wave of the sea will transport pebbles large and small 
 in a great body in one direction ; then retreating, drop the 
 heavier, and scoop channels in the lighter materials which it 
 has itself just deposited, and in whatever is below them, into 
 some of which channels a few large pebbles fall back again/' 1 
 
 In giving a list of the^various kinds of rocks composing the 
 boulders found in the neighbourhood of St. Petersburg, 
 Strangways points out where the parent rock has been found 
 in siiu, thus granite of a large grained variety, containing much 
 black hornblende and red felspar, has been found in quarries 
 belonging to Baron Nicolai, near Wyborg ; a coarse-grained 
 gneiss, grey and red with iron, is found at Imatra and the 
 interior and western parts of Finland ; yellowish green felspar 
 rock with large garnets, is said to be found between Kexholm 
 and Cerdopol ; a red felspar thickly spotted with hornblende is 
 found at Borgo in Finland ; black mica slate is said to be found 
 near Cuopio in Finland ; and basalt is said to occur in the north 
 of Finland, Lapland and Norway ; and he adds : " By the 
 specification of the principal varieties which I have attempted 
 to give above, it will be seen that all the primitive boulders 
 the analogues of which have been found in situ, appear to 
 have been transported from the north." 2 
 
 In the same year Count Razoumofski published a memoir 
 entitled Coup d'oeil geognostique sur le Nord de 1'Europe 
 en general et particulierement de la Russie." In this he 
 presented very much the same views as those urged by 
 
 1 Trans. Geol. Soc. vol. v. 425-436. a Id. 430-433. 
 
 i- 2 
 
68 The Glacial Nightmare. 
 
 Strangways, founding his induction, however, on a wider 
 basis. He mentions how a careful examination of the various 
 blocks, etc., proves incontestably their transport from Scan- 
 dinavia and Finland and the country of the north. He states 
 how widely dispersed they are between St. Petersburg and 
 Moscow, on the Waldai mountains as far as Bronitza, i.e., 
 sixty-three French leagues from Wyborg, where similar 
 granite is found, and in other parts of Europe as between 
 Breslau and Berlin, and near Grossen, their furthest point, 
 which is 140 leagues from the south point of Scandinavia. 
 He mentions how the groups of stones are found in a kind of 
 parallel alignments, and that straight lines passed along the 
 iixes of such groups would include a number of them, like 
 beads on a string. He says the blocks of syenite found 
 between Breslau and Berlin have their parent rock in Sweden, 
 and those of primitive rock found in Hanover also in Scandin- 
 avia, and he assigns the blocks found in the neighbourhood of 
 St. Petersburg to an original home in Finland. He found be- 
 tween St. Petersburg and Riga, and between Teiletz and Riga 
 blocks of a red granite like that existing in situ in Ingria and 
 Finland, and similar blocks near Memel in Prussia. He noticed 
 that in Esthonia the boulders occur chiefly where the country 
 has a broken contour, and rarely when it is of a continuous 
 horizontal character; commonly on hills and rarely in the 
 plains, as if the stones had been in some way arrested. He 
 remarks that although the mountains and valleys of Scandin- 
 avia are on so much smaller a scale than the Alps, yet that 
 the rocks which have travelled thence have travelled a great 
 deal further. Thus he himeelf had found pieces of northern 
 granite between Straszof and Ivaniska, thirty-two leagues 
 from Cracow. He had also found at Memel enormous blocks 
 and pebbles of the quartzose sandstone, hard and red, which the 
 Germans call f< rothe todtliegende," and which occurs in situ 
 on the banks of Lake Onega, 245 French leagues off. He 
 also remarks on the absence of secondary rocks among the 
 boulders found near St. Petersburg, and that these secondary 
 rocks only begin to occur at Dorpat between Little and Rana 
 Poungern. 
 
 He lastly states that in one district, which he does not specify 
 but which he says is 130 French leagues in extent, there is 
 
Fox Strangways. 69 
 
 an immense lacuna in the general distribution of the blocks 
 and that those occurring near there, instead of showing a huge 
 current flowing from N.E. to S.W., show one from N.W. to 
 S.E., and he accordingly postulates the existence of two 
 currents, unequal in mass and size and starting from different 
 points. 3 
 
 Mr. Fox Strangways, writing again in 1821, says : 
 " Throughout the whole of Finland, the evident traces of 
 diluvial action are on the most astonishing scale. Without 
 dwelling on the stupendous size and universal distribution of 
 primitive boulders, it is impossible not to perceive, that the 
 top of every rock in situ, every tor, every hill and knoll of 
 granite, or primitive rock, from its first appearance in Carelia, 
 till it sinks beneath the Gulf of Bothnia, presents a surface as 
 much rounded, and as visibly water- worn, as the boulders or 
 colossal pebbles that lie round their bases. Where the rock 
 in situ does not rise high above the soil, and where the 
 boulders are at the same time thickly scattered and of vast 
 size, it is scarcely possible to distinguish one from the other. 
 This is particularly the case between Wyborg and Frederics- 
 haven, where they totally prevent the culture of the earth and 
 barely allow the passage of a carriage over a most tortuous 
 and narrow road. The islands off Abo possess the same 
 rounded character, and another fact connected with this sub- 
 ject, and well worthy of remark, is, that in the south of 
 Finland, where we recognize in situ the parent rocks of the 
 commonest boulders of the neighbourhood of Petersburg, we 
 find new varieties occurring in rolled masses, probably brought 
 from rocks existing still further north." 4 
 
 One of the most remarkable facts with regard to the central 
 parts of Russia, is the quantity of siliceous boulders which are 
 scattered over the government of Moscow, Vladimir, Tver and 
 the neighbouring countries. They are found on the Waldai 
 Hills, but rarely north of them, although they occur there in 
 situ ; a strong proof of the direction of the diluvian current 
 from north to south. . . . They are rarely found north of the 
 town of Waldai or east of Kostroma. Their boundary on 
 the south and west is unknown 5 
 
 8 Annales des ScieDces Naturelles, xviii. 133-147. 
 
 4 Trans. Geol. Soc. ser. ii. 1, 9 and 10. 6 Id. 18, note. 
 
7<D The Glacial Nightmare. 
 
 In Mr. Henslow's Observations supplementary to Dr. 
 Berger's account of the Isle of Man, read before the Geological 
 Society April 7th, 1820, we read: " Blocks of granite similar 
 to that of Slieu-ny-Clough lie dispersed for two miles over the 
 low plains between St. Mark's chapel and South Barrule, nor 
 have they the least appearance of resting on their birthplace, 
 so that no cause appears so likely to have placed them there 
 as that which scattered those of Dun. This disposition of the 
 granite boulders appears to indicate a current having set over 
 the country (at least on the eastern side of the central chain) 
 in a direction from N.E. to S.W." 6 
 
 In 1820 Mr. Horace H. Hayden published his " Geological 
 Essays," a remarkable work for its time, which is nowhere 
 referred to, so far as I know, in English geological literature. 
 In this work he tells us the surface deposits of the United 
 States offer stronger evidence than can be found elsewhere 
 of a general current flowing from N.E. to S.W. having 
 prevailed over the whole of that continent, and perhaps over 
 every other. 7 He contends that neither the sea nor 
 rivers working in their normal methods are competent to have 
 distributed the deposits in question, which must have been 
 deposited by a current flowing impetuously across the 
 continent. Inter alia he quotes the rolled or water-worn 
 pebbles and the wave-like undulating appearance of almost 
 every section of alluvial formation, whether perpendicular to 
 the surface or inclined. 8 " These pebbles are piled up into 
 hills 100 or 200 feet above the rivers near which they lie, 
 and sometimes spread over many square leagues of country 
 over which the current of no one river upon earth has ever 
 flowed. It is both morally and physically impossible that 
 such results could, by any means, be produced by any river 
 flowing through the district of country where they lie. 
 Nothing short of a universal current could have produced 
 such effects ; and it must have been of such extent and rapidity 
 as to have hurled them into motion with almost as much 
 facility as the leaves of trees are raised into the air by a whirl- 
 wind. The direction of the current is shown by the fact that 
 the pebbles are found in the greatest quantities on the W. or 
 
 6 Trans, Geol. Soc. v. 503. * Id. 3. 8 Id. 45. 
 
Horace Hay den. 71 
 
 S.W. banks of tliese rivers and streams and valleys." Again, 
 he remarks that most of the pebbles lie from three to five 
 miles, and sometimes more, in a S.W. direction from the 
 original gangiie or matrix. Of these facts he gives several 
 instances, and notably mentions Baltimore, where he says he 
 had counted among the pebbles in a gravel ridge upwards of 
 thirty masses of granite, micaceous schist and greenstone. 
 " Well defined granite does not occur," he says, " within a 
 mile to the north of where these masses now lie, and the green- 
 stone range does not occur on the north within three miles, 
 while to the south of the ridge neither granite nor green- 
 stone were ever known to exist in place within 500 or 600 
 leagues, whence it follows they must have been deposited by a 
 current from the north." Similar phenomena occur else- 
 where from New York to Georgia. At a little distance north 
 of the U. S. Bank in Washington is a circumscribed spot of 
 about an acre, covered with masses of rock and rolled stones of 
 various sizes. Among these Mr. Hay den says he discovered 
 in February, 1820, rolled masses of amygdaloid, and of 
 hornblende-porphyry, containing epidote, both peculiar to the 
 Blue Bidge or south mountains in Maryland and Pennsylvania, 
 and which cannot be found in place, within sixty miles of 
 Washington city. There were also masses of a kind of granular 
 quartz, weighing probably from two to five hundredweight, 
 which is not found nearer than Herkomer county in New York, 
 far beyond the north mountains in Pennsylvania. 9 
 
 To the current which moved the gravel he attributes the 
 clearing of the northern parts of America of its coating of 
 soil, leaving the surface, as Hearne describes it in his travels, 
 " a jumble of barren hills and rocks, destitute of every kind 
 of herbage except moss." The alluvial covering of a large 
 part of this area, he contends, was swept along by the great 
 current which deposited it in more southern latitudes. 
 
 Mr. Hayden also refers to the giant's cauldrons of America. 
 Thus Mr. Mackenzie, in speaking of the portage of Chaudiere 
 des Francois, says, " It must have acquired the name of Kettle, 
 from the great number of holes in the solid rock of a cylin- 
 drical form not unlike that culinary utensil ; at the bottom of 
 
 9 Id. 49 and 50. 
 
72 The Glacial Nightmare. 
 
 them are generally found a number of stones and pebbles. 
 These holes are upwards often feet above the present level of 
 the water at its greatest heigh t, and are to be seen along every 
 great river throughout this extended country.-" 1 Henry also 
 remarks that the phenomenon is not restricted to La Chaudiere 
 Francois, but is observable at almost every carrying place in 
 the Otaonais, evidencing the retirement of the waters from a 
 higher level. 2 On the Mohawk river these cylindrical ex- 
 cavations in the rocks are numerous and deep, and afford, says 
 Hayden, the most unequivocal proofs of the violent agitations, 
 of currents of water, at an elevation much above the present 
 bed of the river. 3 
 
 The flood of waters which distributed the American debris, 
 he urges, started in the Arctic regions. Following somewhat 
 in the steps of St. Pierre, he suggests that if the earth's axis 
 were changed from north to south, so that the sun should pass 
 immediately over the two poles, there would be " a certain and 
 inconceivably rapid dissolution of the immense hemispheres 
 of ice there/' and the water must accordingly have rushed in 
 wild confusion into the adjacent seas. " From the north pole 
 there are only two outlets, the one through the narrow 
 channel of Berings Straits, the other through a very wide one 
 into the Atlantic ocean, and by far the greater portion of the 
 rushing water must have flowed into the Atlantic, its tide 
 would consequently rise, accompanied by a current increasing 
 in power and force, which would flow over the adjacent 
 continents in riotous disorder. At first the current would 
 be divided by the craggy heights of Spitzbergen, a part of it 
 would be thrown into the White Sea, while the other, directing 
 its force against Lapland and the rocky cliffs of Sweden and 
 Norway, would be thrown back on the eastern and southern, 
 coast of Greenland ; from thence hi a S.W. direction until it 
 struck the S.E. coast of Labrador, along which it swept 
 through the Straits of Bell Isle across Newfoundland, Nova 
 Scotia and along the Atlantic coast into the Glulf of Mexico." 
 " In a short time," says our author, " the S. and E. coast of 
 Labrador, over which the current was urged with increasing 
 force, was desolated. Its soil was stripped off and carried 
 
 1 Mackenzie's Travels, 37. 2 Henry's Travels, 31. 
 
 3 Op. cit. 180. 
 
Silliman. 73 
 
 southwards, depositing the blue clay containing shells of 
 existing species and rolled stones of granite and gneiss with 
 barnacles adhering to them which in the State of Maine exists 
 at a depth of twenty feet below the present soil, perfectly 
 resembling that which is taken from the borders of creeks and 
 bays of salt water in odour and other properties. " 4 
 
 In reviewing Hayden's work, Professor Silliman, writing in 
 1821, says : " The almost universal existence of rolled pebbles, 
 and boulders of rock, not only on the margin of the oceans, 
 seas, lakes, and rivers ; but their existence, often in enormous 
 quantities, in situations quite removed from large waters; 
 inland, in high banks, imbedded in strata, or scattered, occa- 
 sionally, in profusion, on the face of almost every region, and 
 sometimes on the tops and declivities of mountains as well as 
 in the valleys between them ; their entire difference, in many 
 cases, from the rocks in the country where they lie, rounded 
 masses and pebbles of primitive rocks being deposited in 
 secondary and alluvial regions, and vice versa ; these and a 
 multitude of similar facts have ever struck us as being among 
 the most interesting of geological occurrences, and as being 
 very inadequately accounted for by existing theories. Pebbles 
 may, in given instances, be formed (possibly) by decomposi- 
 tion of the angular portions of a stone, by various chemical 
 agencies, aiding those of a mechanical nature, but an immense 
 number, and in our view the immensely greater number 
 of pebbles, present unquestionable evidence of having been 
 brought to their rounded form by attrition. The attrition of 
 the common waters of the earth, and even that exerted during 
 the comparatively short period of the prevalence of the deluge 
 of Noah, would do very little towards producing so mighty a 
 result, and we must assign this operation to the more recent 
 periods of the prevalence of the great chaotic deluge whose 
 existence is distinctly recorded in the first chapter of Genesis, 
 and equally admitted by all geologists." 5 
 
 In a letter dated March 21st, 1821, by Dr. Bigsby, published 
 in the third volume of Silliman's Journal, he says the bed of 
 Lake Huron is covered with the debris of distant countries, 
 its rocks are furrowed and abraded. And he explains these 
 
 4 Op. cit. 63-07. 5 Silliman's Journal, iii. 49, 50. 
 
74 The Glacial Nightmare. 
 
 and other facts by an enormous body of water having 
 rushed over these countries and swept from distant lands the 
 colossal fragments of rock so frequent in the lake. 6 " These 
 fragments/' he says, "are incredibly numerous in Lake 
 Huron, and may.be divided into two geological classes, the 
 foreign and the native. The former are the more plentiful 
 and are round and smooth. They are seen everywhere, but 
 are collected principally in the interior of the coasts and 
 islands, either in confused heaps, or in parallel ridges and 
 crowning the highest acclivities in great numbers, and the 
 fragments are of various dimensions. They belong almost ex- 
 clusively to the older orders of rocks, and are therefore of a 
 northerly origin, granites, gneiss, mica slato and porphyries 
 prevail, of kinds which I never saw in situ, although I have 
 skirted the north shore for 200 miles and have traversed the 
 wilderness to the E.N.E. for 600 miles. Mica slate I never 
 met with in a fixed state, excepting a few strata of the black 
 variety at the Falls des Chats, on the Ottawa." 7 
 
 In the spring of 1821 Dr. Bigsby read another paper before 
 the Geological Society of London, in which he says inter alia : 
 " The shores and bed of Lake Huron appear to have been sub- 
 jected to the violent action of a flood of waters and floating 
 substances rushing from the north. That such a flood did 
 happen is proved not only by the abraded state of the surface 
 of the northern mainland and scattered isles of the Mainton 
 line range, but by the immense deposits of sand and rolled 
 masses of rock which are found in heaps at every level, both 
 upon the continent and islands, and since these fragments are 
 almost exclusively primitive, and can in some instances be 
 identified with the primitive rocks in situ upon the northern 
 shore, and since, moreover, the country to the south and west 
 is secondary to a great extent, the direction of this flood from 
 the north seems to be well established. 
 
 " The boulders of granite, gneiss, mica slate (rare), green- 
 stone, porphyry, syenite, and various amygdaloids, are princi- 
 pally," he says, " of such rocks as I have not met with in situ, 
 either in the neighbourhood of Lake Huron, or in a journey of 
 COO miles which I made to the E. and N.E. of the lake, 
 
 6 Op. cit. 255. 7 Id. 256. 
 
Hitchcock. 75 
 
 through the forests of the river Ottawa. ... It can scarcely 
 be doubted that these rocks will be found in situ somewhere 
 on the northern shore of Lake Huron, between the Missassaga 
 and Pelletau's channel. It is there and on the isle of St. Joseph 
 that these boulders most occur. . . . Passing on to the 
 southern division of the lake, sixty-four miles south of Cabot's 
 Head > the limestone cliffs of the Manitouline range are suc- 
 ceeded by cliffs of clay. From this point beds of clay, covered 
 towards the upper part of the river St. Glair by thick beds of 
 sand, extend for 150 miles to Lake Erie, and thence along the 
 northern shore, which presents a series of clay cliffs and sand- 
 hills, to the N.B. extremity of the lake. The whole of the in- 
 tervening shores and woods are strewn with rolled blocks of 
 gneiss, porphyry, conglomerate, and greenstone, such as 
 prevail on the northern shore of Lake Huron. In a S.W. 
 direction, the clay beds prevail over the Michigan territory, 
 and the States of Indiana and Illinois, to an unknown distance. 
 In the two last named States (which I have not visited) rolled 
 blocks abound/' 8 
 
 Hitchcock, in a paper on the Geology of Connecticut, read 
 before the American Geological Society in 1822, tells us how, 
 along the Connecticut river, large boulders are frequently 
 found far removed from their original sites ; stragglers, he 
 says, may be found everywhere, and among all the rocks none 
 seem more scattered than the granite ones. Generally the 
 boulders, he says, are like the subjacent rock, but there are 
 many exceptions ; thus, a few miles west of New Haven, in 
 Woodbridge and Milford, the surface is covered with rolled 
 masses, sometimes quite large, of primitive and secondary 
 greenstone, mica slate, gneiss, granite, and almost every 
 other rock except that which is in place, viz. chlorite-slate or 
 argillet. The diameter of the fragments varies from an inch to 
 20 and 30 feet, and they are usually rounded, showing attrition. 
 Some of the highest of these boulders are found insulated on 
 the pinnacles of our mountains. Masses of greenstone are 
 found at a greater distance on the W. of the ridges than on 
 the E., and the boulders of Woodbridge and Milford have 
 evidently been brought from the N. He agrees with the view 
 
 8 Trans. Geol. Soc. 2nd ser. i. 204-206. 
 
76 The Glacial Nightmare. 
 
 of Hayden in his essays, that the alluvion of the middle and 
 S. states was formed by a current or currents that formerly 
 flowed across the continent from the N.E. to the S.W. 9 
 
 In the same year Mr. Grainger communicated to Silliman's 
 Journal an account of a grooved or fluted rock at Sandusky 
 bay, Ohio. The surface of the rock, which was of limestone, 
 was covered by several feet of earth. When this was bared it 
 was found to be fluted in a direction from east to west with 
 grooves differing in width and depth, perfectly straight and 
 parallel with each other. " It appears to have been polished as 
 if by friction ; and this polish is still retained in a considerable 
 degree. The grooves were traced over a distance of a mile." 
 The author, speaking of the fluting, says : " It has to me the 
 appearance of having been formed by the powerful and con- 
 tinued attrition of some hard body. It resembles, to some 
 slight degree, the sides of a saw-gate which has been for a 
 long time rubbing against the posts which confine and direct 
 it. ... the flutings in width, depth and direction are as regular 
 as if they had been cut by a grooving plane. This running 
 water could not effect, nor could its operation have pro- 
 duced that glassy smoothness, which in many parts it still 
 retains." l 
 
 In the year 1823 there was addressed, under the signature 
 " N," a letter to Professor Silliman on Boulders and rolled 
 stones. In this the author tells us how, in his younger days, 
 when rambling over the mountains and valleys of Connecticut, 
 he had been led to inquire why so many rocks of such various 
 sizes should have become so perfectly rounded ; why they 
 should be found on the highest mountains, as well as in every 
 valley ; why piled in such immense ridges, as well where no 
 stream of water was ever known to flow, or, according to every 
 appearance, ever had or could have flowed, as well as in those 
 places where larger and smaller streams still existed ? " Why, 
 again, the uncovered faces and angles of the granite and other 
 rock should bear the same marks of having been worn and 
 ground as the rounded rocks/' An answer was suggested, he 
 tells us, when he went down to the sea at Newport, R.I., after 
 a storm, and found the mighty process still going on ; and he 
 
 9 Silliman's Journal, vi. 83-85. J Id. 179, 180. 
 
"N." 77 
 
 concluded that the whole country must have been covered at 
 one time to a great depth with this now retired ocean. Proofs 
 of this fact he claims to exist everywhere. " Thus," he says, " if 
 we follow the Grafton Turnpike road, granite in place is first 
 found about three miles from the meeting-house in Lyme. 
 After that, it is everywhere seen in blocks, in rounded rocks, 
 as well as very often in place. . . . Presently the road crosses 
 a little brook ; following this up, you observe large hills of 
 gravel and of sand, apparently the residuum of what once filled 
 this valley to the depth of seventy or eighty feet. When you 
 arrive at the source of this stream, you have to ascend a steep 
 hill of perfectly rounded gravel to the height of from sixty to 
 eighty feet, and then pass on a level about thirty feet to a smooth 
 coarse-grained granite in place/' This gravel bore all the traces 
 of having been water- worn. The rocks also had their corners 
 rounded, while their hollows were filled with gravel, and he 
 concluded that only the ocean itself when in movement in the 
 valley could have produced these effects. He also remarked 
 that the rock was marked with circular cavities, three feet across, 
 such as are worn by water, and which were filled with gravel. 
 The cavities were in a line, and not more than twenty or thirty 
 feet apart. He noticed that the diameter of these cavities 
 diminishes as we go down, and in one instance that a piece of 
 rock which was harder than the rest still projected into the 
 hole. He urges that these cavities were worn by the action 
 of gravel and water. 
 
 Speaking of the rounded blocks and the smoothed rocks, he 
 says : " The rivulet that descends into the valley is but a 
 small mill stream, and never could have had any agency in 
 rounding, carrying down, and depositing such an immense 
 mass of rocks, gravel, and sand. . . . How could these blocks 
 become detached from their natural bed, how become so per- 
 fectly rounded, and that in immense quantities, in the course of 
 three or four miles, affording at the same time by their attrition, 
 in that short distance, hills of coarse and finer sand ? How 
 could they become deposited in alluvial hills, gradually dimin- 
 ishing in quantity as they recede from their natural bed, at 
 the same time becoming evidently more and more perfectly 
 rounded ? how could all this have been effected, and much 
 more, but by a slowly-retiring ocean ? " Our author then 
 
78 The Glacial Nightmare. 
 
 goes on to suggest that there may possibly be cavernous 
 spaces underground into which the water has escaped, and 
 he continues : " The fact being once established that the earth 
 was long covered with water, it would seem to account at once 
 for those rounded and other rocks found everywhere in ascend- 
 ing streams and hills, although no similar rocks should be 
 found in the higher situations immediately contiguous. Those 
 blocks, perhaps, were detached by ice from their native beds, 
 and being rounded and transported by the currents, might be 
 deposited anywhere at random." He tells us how he had 
 noticed that valleys shooting up between mountains have in- 
 variably at their outlet a deposit of gravel. This he attributes 
 to the flow of the water, which also smoothed the sides of the 
 valley. The gravel in one instance he names was spread out so 
 as to be almost 1000 feet wide on the western side of the ridge, 
 and nearly 200 feet high on the eastern side, reaching out 
 400 or 500 feet from the ledge. The mass is nearly of a semi- 
 circular form, and the eastern side of it was deposited in that 
 upright position which banks of sand or sand and gravel 
 always assume when deposited in still water . . . the mate- 
 rials of the gravel in the instance quoted must have originally 
 come from some distance, for they are composed almost 
 entirely of grey granite or gneiss. There is no rock like this 
 close by nor anywhere to the west, while it is found in situ 
 to the east ; and he concludes finally that only a retiring sea 
 aided by ice could explain these phenomena, which, as he 
 tells us, occur at heights of a thousand feet and over, above 
 the present sea-level. 2 
 
 In 1823 Dr. Buckland published his famous "Keliquise 
 Diluvianee/' a work written by a distinguished geologist 
 with the avowed purpose of proving an Universal Deluge 
 by means of evidence strictly scientific. Among the other 
 proofs adduced by him are the erratics. Thus he says of 
 the stones found in the so-called drift-beds of eastern 
 England : " The pebbles are of two classes ; (1) the wreck- 
 age of the adjacent inland districts of England; (2) large 
 blocks and pebbles of many varieties of primitive and 
 transition rocks which do not occur in England, and which can 
 
 2 Silliman's Journal, ix. 28-39. 
 
Dr. Buckland. 79 
 
 only be accounted for by supposing them to have been drifted 
 from the nearest continental strata of Norway, by a force of 
 water analogous to and contemporaneous with that which 
 drifted the blocks of Finland granite over the plains of 
 Russia, and the North of Germany. A diluvial current from 
 the north is the only adequate cause that can be proposed, 
 and it is one which seems to satisfy all the conditions of our 
 problem. The pebbles of iridescent felspar, like that of 
 Labrador, which are found on the coast near Bridlington, and 
 resemble similar fragments near Petersburg, can only be 
 referred to the primitive districts of the most northern parts 
 of Europe. Many of the other pebbles of the English coast 
 can be identified with rocks that are known to exist in 
 Norway, and must have been drifted hither at the time of 
 the deposition of the masses of clay and gravel through 
 which they are disseminated ; it is impossible to refer them to 
 any action of the present sea, because they occur on high 
 table-lands of the interior as well as on the coasts, and because 
 the cliffs themselves, being composed of clay mixed with the 
 pebbles in question, are undergoing daily destruction, and re- 
 ceive no addition from the action of the present waves. These 
 foreign and probably Norwegian pebbles on the coast of Eng- 
 land are mixed up with the wreck of the hills composing the 
 interior of each district respectively ; and the component 
 fragments of the latter are less rolled and more angular than 
 those which have come from the continent. ... In the dilu- 
 vium of the numerous valleys of Yorkshire, that unite to fall 
 into the Humber, there is a similar admixture of the debris of 
 strata composing the adjacent country with, rounded fragments 
 of distant rocks ; and in the county of Durham I collected 
 within a few miles on the north of Darlington, pebbles of more 
 than twenty varieties of slate and greenstone rocks, that occur 
 nowhere nearer than the Lake district of Cumberland. In the 
 street at Darlington, at the north end of the town, is a large 
 block of granite of the same variety with those at Shap, near 
 Penrith. Blocks of the same granite lie in the bed of the Tees 
 at Barnard Castle, and near the highest points of the pass of 
 Stainmoor Forest. Similar blocks are found also on the ele- 
 vated plain of Sedgefield, on the south-east of Durham : in all 
 these places they are mixed with blocks of greenstone. The 
 
So The Glacial Nightmare. 
 
 nearest point from which these blocks and pebbles could pos- 
 sibly have been derived is the Lake district of Cumberland, 
 and the only place in which this peculiar granite occurs in situ 
 is the neighbourhood of Shap, just mentioned, from which 
 they are at present separated by the lofty ridge and escarp- 
 ment of Cross Fell and Stainmoor Forest. If the difficulty of 
 transporting them over this barrier be thought too great, the 
 only remaining solution will be that they have come from 
 Norway, like the other pebbles before mentioned as abounding 
 in the diluvium of the whole east coast of England. I am dis- 
 posed myself to adopt the opinion of Sir James Hall that they 
 have come from Cumberland." 3 
 
 Buckland then goes on to describe the foreign rocks occurring 
 among the pebbles in the gravel-beds of Derbyshire and of 
 Central England, attested by several observers, such as Aikin, 
 Conybeare and Sedgwick, the last of whom discovered a joint of 
 a basaltic pillar between one and two feet long in the gravels of 
 Cambridgeshire and Bedfordshire. The same facts have been 
 observed in the gravels of the Thames valley, attesting, as 
 Buckland says, a violent rush of waters from the north which 
 has drifted pebbles of quartz rock from the plains of War- 
 wickshire and other central counties over the whole country 
 intermediate between them and London .... so that we have 
 pebbles of the porphyry arid greenstone of Charnwood Forest, 
 at Abingdon and Oxford ; and pebbles of the rocks near 
 Birmingham, at Maidenhead and in Hyde Park. It appears 
 then, he continues, ' ' that we have evidence, that a current 
 from the north has drifted to their present place, along the 
 whole east coast of England, that portion of the pebbles 
 there occurring, which cannot have been derived from this 
 country ; a certain number of them may possibly have come 
 from the coast of Scotland, but the greater part have ap- 
 parently been drifted from the other side of the German 
 Ocean. It appears also that there are proofs of a similar 
 current having passed over the central and south-eastern 
 parts of England; and if we examine its western side, we 
 find similar evidence of a violent rush of waters from the 
 north; in the pebbles and blocks of granite and syenite of 
 
 3 Op. cit. 192-194 
 
 
Buckland. 8 1 
 
 a very peculiar character, that have been drifted from the 
 Criffle mountain in Galloway, across Solway Firth to the 
 north base of the mountains of Cumberland, where I have 
 seen them at a place called Shalk, between Ireby and Car- 
 lisle j whilst pebbles and large blocks of another kind of 
 granite have been drifted in still greater numbers from 
 Ravenglass, on the west of Cumberland, over the plains of 
 Lancashire, Cheshire, and Staffordshire ; their course is 
 marked on Mr. Greenough's map of England, and they lie 
 in masses of some tons weight on the west of the towns of 
 Macclesfield and Stafford, and between Dudley and Bridge- 
 north." 4 
 
 Turning to Wales, Buckland quotes Mr. Underwood as hav- 
 ing noticed diluvial scratches and scorings on the surface of 
 slate rock where it is immediately covered with a very thick bed 
 of gravel, in a section of the new Irish road at Dyiias Hill, 
 about a mile east of Betws-y-coed. He says further that 
 the whole low country of South Wales from the shore 
 of the Severn is strewn over irregularly with large boulders 
 and beds of gravel derived by diluvial action from the 
 mountains that flank this district on the north. 5 In regard to 
 Scotland he merely quotes the evidence already cited from 
 Colonel Imrie and Sir James Hall. 
 
 Buckland, in referring to Mr. Weaver's paper already cited, 
 mentions how the distinct and lofty ridges of gravel mentioned 
 by him also exist in Holderness in Yorkshire. ' There they are 
 locally known as barfs, and are composed chiefly of rolled 
 chalk flints, and a few primitive pebbles (apparently Norwe- 
 gian), the most remarkable being near Brandsburton, north-east 
 of Beverley : it stretches Across the plain like a vast Chesil 
 bank on the sea-shore, being about fifty feet high and 100 
 broad at the base, and nearly a furlong in length, and has at 
 first sight the appearance of an artificial military mound of 
 enormous magnitude : it bears marks of having been applied to 
 military purposes, but is clearly of diluvial origin. 6 
 
 In another chapter he refers to the American evidence 
 collected by Dr. Bigsby in the paper already cited, and which 
 had not then been published, and goes on to tell us how he had 
 
 4 Id. 198, 199. * Eel. Dil. 206. 
 
 6 Eel. Dil. 209, 210. 
 
82 The Glacial Nightmare. 
 
 been informed by Sir Alexander Croke that the summits of 
 the highest hills in Nova Scotia, which are composed of slate, 
 are strewn over with large blocks of granite, the present 
 position of which can only be accounted for by supposing them 
 to have been drifted from the nearest granite districts by the 
 same rush of waters which transported those described by Dr. 
 Bigsby in the districts of Lake Huron and Lake Erie. Buck- 
 land also quotes a passage from Dr. Meade of Philadelphia, in 
 his account of the mineral waters of Ballaston and Saratoga in 
 the State of New York, where he says that the surface 
 of the ground, which is there composed of shale and lime- 
 stone, is covered with large insulated masses of stone, 
 consisting of large blocks of quartz and rolled masses 
 of other primitive rocks. " These," he adds, " must have 
 been transmitted from the neighbouring mountains, as they 
 are not attached to the rocks in situ, and have no connection 
 with them : they are found in every country, and only prove 
 the action of an extensive flood of waters. ... In this 
 dispersion of blocks of granite and beds of gravel in North 
 America we have evidence of a debacle by the diluvian 
 waters in the Western hemisphere, analagous to that we have 
 been examining in Europe. 7 
 
 With Dr. Buckland, the originator of the term diluvium 
 and the most famous champion of diluvial causes, we may fitly 
 bring this chapter to a close. 
 
 i Eel. Dil. 217, 218. 
 
CHAPTER III. 
 
 THE PHENOMENA OF THE DRIFT. 
 
 The Champions of Water (Part III.}, 1825-1840. 
 
 " It is in vain that a vast duration is ascribed to the influence of an 
 agent, unless it can be shown that its action has a tendency to produce 
 the alleged result. If it has a tendency to produce a different result, 
 that difference would be augmented in proportion to the duration of the 
 action. Now diurnal operations are everywhere found in the act of 
 corroding and altering the forms alluded to, but they are nowhere seen 
 to pi*oduce them. This class of facts, on the other hand, all conspire in 
 giving probability to the hypothesis of a diluvian wave, which affords 
 an easy explanation of all the large features of this country." SIR J. 
 HALL, Trans. Roy. Soc. Ed. vii. 178. 
 
 Chabrier, 1823 Hibbert, 1825 De Lasterye, 1826 Phillips, 1826-29 
 Sedgwick, 1826-31 Taylor, 1827 Jameson, 1826" A," 1827 Hild- 
 reth, 1827 De Kay, 1828 Omalius D'Halloy, 1828 Brongniart, 
 1828 Hausmann, 1829 Elie de Beaumont, 1829 Pusch, 1830 
 De la Beche, 1831 Trimmer, 1831 Ware, 1831 Thompson, 1830 
 Lapham, 1832 Woodward, 1833 Kloden, 1834 Strickland, 
 1835-37 Murchison, 1836 Home, 1835 Milne, 1837 Lartet, 
 1838 Sef strom, 1838 Forchhammer, 1838 Jackson, 1838 
 Shepard, 1838 Hayes, 1839 Durocher, 1840. 
 
 BEFOEE proceeding with my regular story, I must here inter- 
 vene a short notice of a work, the author of which had views 
 which were at least original. 
 
 In 1823 M. F. Chabrier wrote a memoir entitled " Disserta- 
 tion sur le deluge universelle," in which he denied that the 
 blocks found in North Germany came either from the Hartz or 
 from Scandinavia, or that they were distributed by water. 
 He declared them to be aerolites, and he treated them as 
 the debris of a small planet broken to pieces by a comet. 
 This rain of celestial stones was accompanied by a real 
 diluvial downpour from without the earth which constituted 
 the deluge. 
 
 G 2 
 
84 The Glacial Nightmare. 
 
 In 1825, Dr. Samuel Hibberfc read a paper before the Royal 
 Society of Edinburgh, on the Dispersion of stony fragments 
 remote from their native beds, as displayed in a stratum of 
 loam near Manchester. The bed referred to was situated 
 near Strangeways Hall, and traversed by the new red sand- 
 stone cliffs at the confluence of the Irk and Irwell. This clayey 
 bed contained innumerable fragments of rocks, some of them 
 several tons in weight, many of which belonged to primitive 
 or transitory beds foreign altogether to the neighbourhood 
 (which is underlaid by new red sandstone), which had 
 evidently been transported from a great distance. Granite 
 abounded, most of it containing hornblende. Also several 
 kinds of trap and "greenstone equally unknown in situ-, also 
 grauwacke slate, " loosened relics of the hills of far remoter 
 districts," and quartz. Dr. Hibbert affirmed his opinion that 
 these rocks had come from Westmoreland. He states that he 
 had never known any pieces of the porphyritic granite found 
 at Shap, and inferred that the fragments had therefore been 
 transported from a different place, perhaps from the neighbour- 
 hood of Dufton, near Appleby. Other specimens of newer rocks 
 also occurred, as, fragments of dark-coloured limestone dif- 
 ferent to that of Derbyshire, and which he supposes were 
 detached from limestone hills in the north. Also pieces of 
 sandstone shale and coal, which had probably come from the 
 extensive coal district of Lancashire, intermediate between the 
 Manchester red marl and the Westmoreland hills. These facts 
 pointed, he urged, to the various foreign rocks having been 
 transported from the north. Most of the transported frag- 
 ments appeared to be water-worn ; some fragments, however, 
 showed few or no marks of attrition. Dr. Hibbert concluded 
 that, as no rocks are found in situ like the older specimens 
 named nearer than eighty miles from Manchester, it gives no 
 small degree of support to the conclusion, that an overwhelm- 
 ing force, most probably from the north, far greater than any 
 which can be attributed to existing causes, transported these 
 Iboulders to a situation so very remote from the place whence 
 they were originally detached. 1 
 
 De Lasterye, writing in 1826, tells us he had noticed in 
 ^various parts of America, that when the soil was removed the 
 
 1 Ed. Journ. of Science ii. 208-212. 
 
Phillips. 85 
 
 surface below was found to be polished, the harder rocks 
 naturally preserving the polish better than the softer ones. 
 This effect he noticed on secondary as well as primary rocks, 
 and even in the conglomerates found near New York the sur- 
 faces were found thus polished, which, he says, can only be 
 explained by water having flowed over it ; and he adds that 
 several writers had remarked that currents of water had 
 traversed America from north to south, while in Asia they 
 flowed the other way. 2 
 
 On November 7th, 1826, Professor Phillips read an impor- 
 tant paper before the Yorkshire Philosophical Society, on the 
 direction of the diluvial currents in Yorkshire. In this he 
 cites some interesting facts, inter alia, that Shap granite can 
 be at once distinguished, and is different to all other kinds of 
 granite found in situ in Cumberland or in Scotland, its nearest 
 relative being that found at Ben Nevis. In the summer of 
 1826, he tells us, he noticed large boulders of this granite a 
 mile south of Carlisle, and on an elevated red sandstone fell 
 north-west of Kirk Oswald. Abundance of boulders of it lie 
 some miles north of their native site, near the villages of 
 Shap and Great Strickland. In a southerly direction such 
 large boulders lie on the hills of Kendal and Sedbergh 
 common in the Lune valley, and were found of a ton weight 
 in the vast heaps of diluvial matter cut through in the canal 
 south of Lancaster. 
 
 " Eastward of Shap fells this granite has been rolled by 
 the towns of Orton and Brough towards the hollow in the 
 great summit ridge at Stainmoor, on the top of which pass 
 the blocks remain to attest the direction and force of the 
 transporting waters. From this point the granitic boulders 
 seem to have been dispersed in different directions. We 
 find them in Teesdale, at Cotherstone, below Barnard Castle, 
 near Greta Bridge, and Darlington, at Scotton, south of 
 Richmond, at Catterick and Leeming, at Stoke sley and Thirsk, 
 and several places below the range of oolitic rocks between 
 Thirsk and Pocklington, and commonly in gravel pits and on 
 the surface of the country round York. I have observed them 
 of great size at several points on the Yorkshire coast between 
 Redcar and Scarborough, where they are very plentiful on the 
 
 2 Journal des Connaissances usuelles et pratiques, vol. iv. 242. 
 
86 The Glacial Nightmare. 
 
 shore., while they lie on the cliff at Flamborough Head above 
 100 miles from their native situation." 
 
 Professor Phillips urges that it is easy to understand how 
 the bursting of a lake, as Dr. Fleming had suggested, should 
 have forced the blocks into the vale of Eden or that of the 
 Lune, but he goes on to say that he cannot admit this as a 
 causa when we have to explain the transport of great blocks 
 across the deep valley of Eden, over the summit of Stain- 
 moor, and down the valleys and over the hills of Yorkshire 
 to Scarborough and Flamborough Head. " This case," he 
 says, " has always appeared to me decisive in favour of the 
 opinion, that the dispersion of gravel over large tracts is 
 attributable to the deluge/' From the facts cited he con- 
 cluded that the flood which moved the Shap boulders went 
 in an eastward or rather south-eastward direction, and that 
 the dispersion of fragments in other directions may have 
 been caused by minor currents. 
 
 Another rock easily recognized-, he says, is a white com- 
 pact felspar and large-grained hornblende, with here and 
 there some magnetic iron found in situ on the sides of Carrock 
 fell. To the north boulders of great size of this rock are 
 scattered between Hesket, Newmarket, and Bolton. Near 
 Kirk Oswald, on high ground west of the Eden, they abound 
 about Lowther Castle and in Durham. 
 
 Thirdly, a conglomerate formed of pebbles of light-coloured 
 mountain limestone united by red sandstone, and locally 
 called broc/cram, occurs in situ about Kirkby Stephen and at 
 Stenkrith Bridge. Phillips says he had " seen pebbles of it at 
 Scotton, south-east of Richmond, and at several places on the 
 Yorkshire coast, especially at Scarborough, Bridlington, and 
 south of Owthorne, in Holderness. These examples all point 
 to the current having flowed in an east or a south-east 
 direction, but there is also evidence of a current from the 
 north. The large blocks of limestone trap, millstone grit 
 and other sandstones, coal and magnesian limestone which 
 appear on the coast of Holderness may well be admitted to 
 have been derived from north-west Yorkshire, where such rocks 
 abound. But mica slate with garnets, which nowhere occurs in 
 Cumberland, is probably derived from Scotland ; the radiated 
 fetid limestone clearly claims origin from Building Hill, near 
 
Sedgwick. 87 
 
 Sunderland, and the lias fossils have been rolled from Whitby. 
 These effects seem to require the admission of a powerful 
 southward flow of water, and in the vale of York, lias fossils 
 from the north, chalk and oolite fragments from the north-east, 
 are mixed with the debris of rocks washed from the north-west 
 part of the county. " 3 
 
 I shall now devote a short space to the views of Professor 
 Sedgwick, whose grasp of phenomena was so firm and so 
 catholic. 
 
 Writing in 1826, he says, " Transported boulders of 
 considerable magnitude occur at the very top of several 
 parts of the Derbyshire chain, which overhangs the great 
 plain of Cheshire. For example, many large smooth boulders 
 of primitive or transition rocks lie scattered over the surface 
 of the ground on both sides of the high pass leading from 
 Buxton to Macclesfield. These facts speak the same language 
 with those which I have already quoted. They show the 
 generality of the causes which have produced the superficial 
 detritus." 4 
 
 Again, " Every part of the Yorkshire coast and the whole 
 neighbouring region, bears witness to the operation of similar 
 causes. The numberless valleys of denudation in the eastern 
 moorlands, the immense accumulation of transported materials 
 on the hills as well as in the valleys the whole contour of 
 the vale of Pickering the enormous cap of diluvium con- 
 taining rounded masses of primitive rocks many tons in weight, 
 and resting on the chalk near Flamborough Head the 
 external form of the Wolds and the continuous mass of 
 diluvium extending from Bridlington to Spurn Head, and 
 from the chalk downs to the sea, are so many monuments of 
 the gigantic powers which were let loose upon the world 
 during the epoch of the diluvial gravel." Again, speaking of 
 Holderness, he says, " The lower part of the cliffs, to the height 
 of about twenty feet, generally consists of a stiff bluish clay, 
 which in many places passes into a dark brown- coloured 
 loam. Through the whole of this mass are embedded 
 an incredible number of smooth round blocks of granite, 
 gneiss, greenstone, mica slate, etc., etc., resembling none of 
 
 3 Phil. Mag. ii. 138-141. 
 
 4 Annals of Philosophy. New Series, x. 23 and 24. 
 
88 The Glacial Nightmare. 
 
 the rocks of England, but resembling specimens derived from 
 various parts of the great Scandinavian chain. Irregularly 
 mixed with the preceding are found, in perhaps still greater 
 abundance, fragments of carboniferous limestone, of millstone 
 grit, of lias, of oolite, and of chalk, torn up from the regular 
 strata of the country, and driven into their present situation 
 by a great eastern current, which has left its traces on every 
 part of the neighbouring district. In regard to the embedded 
 fragments above mentioned, two things appear to deserve 
 notice. 1. They exist in equal abundance in the upper as well 
 as in the lower portions of the diluvial loam. This fact, 
 though difficult of explanation, has been remarked in other 
 similar deposits, and seems to prove the gigantic nature of the 
 forces by which the materials have been drifted into their 
 present position. 2. The boulders derived from distant 
 countries are rounded by attrition ; but those which are 
 derived from neighbouring rocks are little altered 'in form. 
 The hard Norwegian rocks are smooth and spheroidal, but the 
 fragments of oolite and lias, and still more the fragments of 
 chalk, are often sharp and angular. . . . The following con- 
 clusions may, I think, be fairly deduced from the facts above- 
 stated. 1. The diluvium of Holderness and of the whole 
 coast of Yorkshire, is due to a set of causes which have acted 
 over the western moors, and over all the great central plain of 
 the county. 2. The diluvial currents which produced the 
 gravel of Holderness were probably contemporaneous with 
 other more powerful currents which drove large masses of 
 primitive rocks from Scandinavia to the plains of Yorkshire. 
 And it seems probable that the same currents were contempo- 
 raneous with those mighty propelling forces which have driven 
 innumerable fragments of the Scandinavian rocks over the 
 great plains of Eussia, Poland, and Germany." 5 
 
 Again, "From the foot of Stainrnoor to Sol way Firth, through 
 the whole plain of the new red sandstone, the incoherent mate- 
 rials under the vegetable soil are spread over the greater part 
 of the surface, and are often of such an enormous thickness as 
 entirely to conceal all the subjacent strata. These accumu- 
 lations are not partial or irregular ; but seem to have been 
 rolled out over the surface of the country by an inundation 
 
 6 Id. 24-26. 
 
Sedgwick. 89 
 
 which acted at one moment over the whole district ; and like 
 all similar deposits, they contain an incredible number of 
 large boulders, principally derived from the neighbouring 
 mountains. . . . From St. Bees Head to the southern ex- 
 tremity of Cumberland, the region bordering on the coast is 
 formed of one almost continuous mass of diluvium, inter- 
 rupted here and there by low hills of blown sand, and by other 
 recent formations. In this part of the country the cliffs are 
 of a deep red colour, caused by the presence of innumerable 
 embedded fragments of the subjacent new red sandstone. 
 With these fragments, boulders of granite, porphyry and 
 greenstone, are scattered throughout the whole diluvial 
 covering ; sometimes in such abundance as to give it the 
 appearance of a true conglomerate. . . . Some of the granite 
 blocks embedded in the cliffs are of great magnitude. In the 
 diluvial cliffs near Bootle, I found one of a rude prismatic form, 
 which was twelve feet long, six feet wide, and five feet and a 
 half high. All specimens of this kind of rock have been 
 drifted to the coast from the granitic region which extends from 
 Wasfcdale foot through Muncaster fen to the neighbourhood of 
 Bootle, and occupies a part of Wastdale head and all the 
 lower parts of the valleys of the Mite and the Esk. If we 
 cross the estuary of the Duddon to the shores of Low Furness, 
 we find an exact repetition of the phenomena we have left 
 behind. All the country bordering on the western shore 
 is covered by an enormously thick deposit of red-coloured 
 diluvial gravel, containing innumerable rolled fragments of 
 rocks derived from every part of the lake mountains, and all 
 the neighbouring islands are composed exclusively of the same 
 materials. A rolled mass of Eskdale granite, which had been 
 embedded in the highest portion of the diluvial cliff near 
 Rampside, fell down upon the strand in 1822. It rose nine or 
 ten feet above the rubbish in which it was at that time 
 partially buried. At the base of the cliffs of the isles of 
 Barrow and Foudrey, among innumerable boulders of granite, 
 and of other Cumberland rocks, were some specimens of a 
 beautiful variety of compact felspar, which I afterwards 
 found in situ near the tojx of Scafell and Bowfell. . . . The 
 phenomena above described have obviously been caused by a 
 violent rush of descending waters. Whatever forces may have 
 
9O The Glacial Nightmare. 
 
 put these waters in motion, it is, I think, obvious, from the 
 facts already stated, that they have not acted partially, but 
 have swept over the whole cluster of the neighbouring 
 mountains." 6 He describes the syenite blocks which have 
 travelled from Carrock fell, and the granite ones from Shap 
 fell, one of the former, the Golden Rock on High Pike, being 
 twenty-one feet long, ten high, and nine wide ; the green slate 
 from Kentmeer and Long Sleddale ; masses of St. John's vale 
 porphyry found near Penruddock, and descending the valleys 
 into the Eamont ; boulders of Shap granite on the calcareous 
 hill of Appleby, some twelve feet in diameter. Shap granite, as 
 he points out, is peculiar and cannot be confounded with other 
 granites elsewhere, and blocks of it are found drifted over the 
 plain of new red sandstone right over the Pennine chain into 
 the plain of Yorkshire, embedded in tho detritus of the Tees, 
 and even carried to the eastern coast. 
 
 Sedgwick goes on to say, " I think it has already been 
 proved that diluvial torrents have swept over every part of 
 the Cumberland chain, because we find water-worn masses 
 derived from the highest elevations of the country, embedded 
 in the diluvial loam which covers almost all the neighbouring 
 plains ; and because we find large boulders of the same rocks 
 scattered over many parts of the mid regions of the moun- 
 tains, in situations to which they could never have drifted by 
 any less powerful agent than that to which they have 
 been ascribed." 7 As a general conclusion Sedgwick says 
 that " the floods which produced the diluvial detritus 
 swept over every part of England, that they were put in 
 motion by no powers of nature with which we are acquainted, 
 and that they took place during an epoch which was posterior 
 to the deposition of all the regular strata of the earth, and 
 prior to all known accumulations of alluvial matter. We 
 have evidence enough," he adds, " to justify us in extending the 
 same conclusions to every part of the European basin, and 
 there is some evidence which makes it probable that they may 
 be extended to the remotest parts of the earth's surface. In- 
 deed, the mighty disturbing forces which produced the accu- 
 mulations of diluvial detritus between the western extremities 
 of Europe and the central plains of Asia, must probably have 
 6 Id. 26-29. 7 Op. cit. 30. 
 
Taylor. 9 1 
 
 acted with sufficient energy to leave some traces of their power 
 over every quarter of the globe. On the continent of America 
 the succession of formations seems to be very nearly the same 
 with that of our own country, and over all the regular strata, 
 there occur in many places alluvial and diluvial formations in 
 every respect like those of Europe. It is, therefore, to say 
 the least of it, probable that the diluvial phenomena of Europe 
 and America belong to the same epoch/' ' 
 
 In some papers on the Geology of East Norfolk, published 
 in the Philosophical Magazine for 1827, Mr. R. C. Taylor was, 
 I believe, the first to notice some of the very remarkable 
 features of the drift deposits of East Anglia. He says, inter 
 aliob, " The crag has, at the last of the geological epochs, 
 been subjected to abrasion by diluvial currents. Portions, 
 probably from its western edges, have been swept away. 
 Their fragments, mingled with those of the chalk and pre- 
 ceding formations, piled in enormous heaps, form the cliffs 
 of Cromer and Trimingham, 250 or 300 feet in thickness 
 upon the original crag, which rests in situ at their base. The 
 proof of the disruption and transportation of more ancient 
 strata may be observed in the enormous detached masses of 
 chalk, in these diluvial cliffs, at various elevations above the 
 crag. Near the lighthouse hill at Cromer, one of these in- 
 sulated patches is 150 feet high, and has a kiln upon it, in 
 which lime of an excellent quality is burnt. Further on at 
 Eunton is a large mass eighty feet thick ; another rises to the 
 height of 100 feet ; and at Sherringham is another still higher. 
 In all these cases they rest upon the crag, proving alike the 
 breaking up of the older strata, and the continuity of the later." 9 
 In the same paper he refers to the variety of travelled rocks 
 occurring in the Cromer cliffs, such as blocks of several 
 varieties of granite, basalt, porphyry, trap and micaceous 
 schist ; sandstone of various kinds, chert breccia, beside lime- 
 stone, claystone, etc. The diluvial fragments of the later series 
 are from the chalk, plastic clay, London clay, green sand, 
 Kelloway's rock, the oolites, lias clay and marlstones. They 
 are of all sizes up to four tons in weight, one block on the 
 shore was observed to be six feet high, another standing six or 
 eight feet high was known to the fishermen as Black Meg. 1 
 
 8 Id. 32. 9 Phil. Mag. i. 288. a Id. 346. 
 
92 The Glacial Nightmare. 
 
 Turning to America, we find in the appendix to the 
 narrative of Parry's third voyage, a memoir by Professor 
 Jameson, dated April 21th, 1826. In this he mentions how 
 numerous boulders were strewn over the surface of some of 
 the islands in the Arctic ocean. Thus the surface of Igloolik, a 
 limestone island, is strewn with blocks of primitive rocks. 
 The island of Neerlo Nakto, principally composed of limestone> 
 is also strewn over with primitive blocks ; and in Ambers t 
 Island, in which greywacke and greywacke slate are almost 
 the only rocks, rolled masses or boulders of granite, gneiss 
 and quartz rock, are not uncommon. 2 According to Dr. 
 Neill's report, the whole limestone country, extending as far 
 north as Cape York, and south as far as Cape Fitzgerald, 
 was more or less covered with boulders of primitive rocks, 
 which were either rounded or angular. Similar boulders 
 were observed strewn over the limestone on the west side of 
 Prince Regent's Inlet in North Somerset. They included 
 granite, syenite, gneiss, indurated talc, hornblende-rock, acty- 
 nolite rock, a beautiful quartzy iron-glauce and brown hematite. 
 The first three were the most numerous and largest. Some 
 of the boulders near Port Bowen were upwards of fifty tons in 
 weight, and lay fully 400 feet above the level of the sea, and 
 Dr. Neill saw a boulder of granite, two or three tons in weight, 
 on the summit of the highest limestone hill near Port Bowen. 
 They were numerous on the slopes of the hills on both sides of 
 Prince Regent's Inlet ; everywhere resting on secondary lime- 
 stone, but on leaving the coast they gradually diminished in 
 number and size, and fourteen to sixteen miles from, the sea 
 were comparatively rare, and in general not larger than a 
 closed fist. These masses were most like the primitive rocks 
 of the Whale Fish Islands, and the nearest known fixed 
 primitive rocks were those at Cape Warrender, more than 100 
 miles distant. 3 Jameson argues that they afford evidence 
 of the passage of water across the country since the latest 
 tertiary times. 
 
 In a letter signed (( A," dated April, 1826, and published in 
 Silliman's Journal, reference is made to the fact that the 
 surface of the rocks in America, where it has not been 
 exposed to the atmosphere, is worn quite smooth, whether the 
 
 2 Op. cit. 144. 3 Id. 148. 4 Id. 15(X '' 
 
covering be shallow or deep, of whatever kind of rock the 
 mass may consist, or however unequal and irregular may be 
 the form it has assumed. The appearance is as if worn down 
 by some powerful but not very delicate agent. The harder 
 parts have in some instances, especially when forming veins 
 in a softer stratum, the feeling of being polished, but the 
 general character of the surfaces, although smooth to the eye, 
 is somewhat rough to the touch, with slight grooves or 
 channels, running in a uniform direction, very nearly north 
 and south, but from a little west of north to a little east of south. 
 The author connects these scratches with the masses of 
 rolled gravel often found at a great distance from any river 
 and often on mountain tops. He suggests that some sudden 
 change has sometime taken place in the velocity of the earth's 
 motion about its axis, when the whole mass of water on the 
 earth would rush forward with inconceivable violence. The 
 Pacific Ocean would thus sweep over the Andes and Alle- 
 ghanies into the Atlantic, which in the meantime would be 
 sweeping over Europe, Asia and Africa : a few hours would 
 cover the earth with one rushing torrent, in which the 
 fragments of disintegrated rock, earth and sand would be 
 carried along with the wreck of animal and vegetable life in 
 one all but liquid mass. The appearances, he says, favour this 
 conclusion, the outer more friable strata torn out from the 
 solid mass, rounded into boulders and pebbles of every size 
 or crushed into atoms, the surface of the harder strata 
 ground, pounded and worn. The distribution of the materials 
 favours the same view. The more elevated ridges would 
 arrest the more ponderous detached rocks, together with the 
 confused mass of pebbles and sand accompanying them, while 
 the lighter portions of the latter would be carried off to a 
 lower level. Thus we find the largest boulders, with immense 
 hills of coarse gravel, piled up in mountainous districts of 
 country, frequently nearly equal in height to the ridges of 
 rock about them, except when the latter shoot up into sharp 
 peaks. He remarks that he had noticed these phenomena in 
 all parts of New England, and he adds that the polished 
 surfaces are not confined to the primitive rocks but are 
 nowhere more striking and beautiful than on the pudding- 
 stones near Boston, the embedded pebbles being worn down 
 
94 The Glacial Nightmare. 
 
 and polished as in the columns of the Halls of Congress in 
 the Capitol. 5 
 
 Dr. Hildreth, writing in 1827, describes certain parts of Ohio 
 as consisting of a yellowish loam, twenty to fifty feet thick, based 
 on tough blue clay. ' ' But," he continues, " the most singular 
 feature of all is the abundance of detached fragments of blocks 
 of primitive rocks, with which this region is filled. All are 
 rounded or worn by attrition, and lie in that confused state 
 which they must be supposed to exhibit if brought there 
 by an immense current of water. I picked up pieces of 
 granite, hornblende, greenstone, gneiss quartz, limestone, etc. 
 Some blocks of granite are large enough to make a pair of mill- 
 stones, and are used for the purpose. ... In reflecting on 
 this formation, the impression is irresistible that it is the result 
 of an immense current or body of water pouring down from 
 the north, sweeping the south side of Lake Erie and all the 
 Scioto country and Miami valley, as those regions are said to 
 furnish specimens of the same primitive kind.'"' 6 
 
 In a paper by J. E. De Kay, published in vol. xiii. of 
 Silliman's Journal (1828), he speaks of the southern part of the 
 State of New York being composed of gneiss covered with sand. 
 Promiscuously scattered over the surface and embedded in the 
 sand are boulders of different sizes up to several tons in 
 weight. They are totally unlike any rocks in place in the 
 immediate vicinity, being chiefly greenstone and trap, 
 occasionally slaty, sometimes granitic, and not unfrequently 
 calcareous. He says, " Dr. Akerley has referred the schistose 
 rocks to the region above the Highlands, the greenstone to 
 the pallisade rocks on the Hudson, and others to rocks similar 
 in composition and structure nearest to the place where the 
 boulders occur." He himself believes that when the boulders 
 consist of secondary rocks and are scattered over primary 
 ones, that they represent the remains of strata disintegrated 
 and destroyed in situ ; when they are of primitive rocks, he 
 says, we need not look for their parent strata several hundreds 
 of miles away. We know that primitive rocks frequently 
 thrust themselves through all the superincumbent strata, and 
 if their peaks have been destroyed by some convulsion or by 
 
 6 Silliman's Journal, vol. xi. 100-103. 
 6 Silliman's Journal, xiii. 39. 
 
Brongniart. 95 
 
 time, and their origin concealed by the detritus, we can account 
 for their debris occurring in a region to which they are 
 apparently strangers. 
 
 Reverting to Europe; Omalius d'Halloy, writing in 1828, 
 refers to the granite blocks scattered over the district of Cam- 
 pine in Belgium, which, he says, are either found on the surface 
 or buried in the sand, and in immense numbers. He argues 
 that their shape is due to weathering, like that of the well- 
 known tors found in granitic districts, that they were 
 detached from beds of granite which lie below them, and that 
 the reason they are found at the surface is that, when a 
 mixture of unequally divided matter is tossed about, the big 
 pieces naturally come to the surface, and that earthquakes have 
 probably shaken the deposits in which they occur ; and he says 
 he is confirmed in this view, because the blocks only occur 
 in the sand and not in the clay which covers part of Frisia, 
 Groningen and Overyssel. 7 
 
 In 1828, M. Brongniart published a paper on the Swedish 
 erratics, in the Annales des Sciences Naturelles, vol. xiv. 
 p. 1. In this he mentions that although the blocks occur 
 isolated from each other, they are generally found in groups. 
 That sometimes they are found, as on the Alps and Jura, 
 planted on the base rock, at others buried in sand, as on the 
 plains of Westphalia ; sometimes their angles are quite sharp, 
 at other times their edges have been somewhat rubbed and 
 weathered. He remarks that the sandy plains of Westphalia, 
 Hanover, Holstein, and Zealand, Mecklenburg, Brandenburg, 
 the plains and shores of Pomerania and Prussian Poland as far 
 as the district between Warsaw and Grodno, and as far east as 
 the Neva, are covered with these blocks, in some places more 
 abundantly than in others. Thus between Lestadt and Ham- 
 burg there are a large number ; then follows a sterile zone, 
 then another group, and so on, and the same is the case in 
 Sweden, where more or less sterile zones alternate with zones 
 where the blocks are abundantly distributed. In East Friesland 
 and Groningen they are buried in the sands underneath the 
 turbaries, and are a fertile source of stone in those sandy 
 districts where stone is otherwise scarce. Here also they occur 
 
 7 Memoir pour servir a la description Geologique des Pays Bas et de 
 la France, etc., 205-6. 
 
g6 The Glacial Nightmare. 
 
 in zones. At the other end of the Baltic there are found 
 blocks of limestone containing trilobites, orthocerites, etc., 
 which are dug out and burnt into lime. These blocks are 
 apparently derived from beds with similar fossils in Sweden 
 and Norway. 8 He goes on to show that as the cal- 
 careous blocks are shown by their contents to have come 
 from Scandinavia, so M. Hausmann and others have shown 
 that the granitoid blocks found in Pomerania and Westphalia 
 are mineralogically identical with Scandinavian granites, and 
 contain, inter alia, Wernerite. Whence it is pretty clear that 
 the blocks of Northern Europe have been derived from 
 Scandinavia, and not from the Hartz or elsewhere, and the 
 Baltic depression which intervenes presents the same difficulty 
 in explaining their transport that the valley of the Aar does 
 in Switzerland. That they traversed the Baltic seems clear, 
 for we can trace them step by step. The sandy soil of Hoi- 
 stein is covered with them, so is that of Zealand. They abound 
 near Copenhagen, as far as Elsinore, and are used for building 
 purposes in these districts where sand abounds and stone is so 
 scarce. On crossing the narrow but deep Sound, we meet 
 with them again at Elsingborg on the other, side. The soil of 
 Scania is as much strewn with them as that of Zealand, but there 
 they are freer from sand and more exposed. Further north in 
 Sweden they abound still more, and are occasionally found 
 piled upon one another and intermixed with sand in peculiarly 
 shaped mounds called asar and sandasar by the Swedes, accord- 
 ing as the blocks or sand prevail. Brongniart says he found 
 these ridges more especially in Scania, Smaland, Sudermannia 
 and Upland. They are seldom a hundred metres high, are long 
 and narrow, and a little wider and higher at one end than the 
 other. These hillocks are widely distributed in Sweden, and 
 differ somewhat from each other. Generally, in the south they 
 are formed of gravel or sand, either quartzose or granitic, and 
 of granite blocks of a considerable size. Further north, especi- 
 ally north of Upsala, they are more abundant and more sandy. 
 A remarkable feature about them is that they all range from 
 N.N.E. to S.S.W., and are thus parallel over a large area. 
 They also remain continually of one height and width, so much 
 so that the roads in some places, as from Upsala to Wendel, 
 
 8 Op. cit. 1-10. 
 
Brongniart. 97 
 
 from Enkoping to Nora, from Hubbo to Moklinta, etc., run 
 along their crests. Brongniart compares them to the trains of 
 sand, etc., which are often found in streams behind some 
 obstacle, such as a large stone, etc., and he urges that the 
 basaltic hill of Kinnekulle on the south-east borders of Lake 
 Wenern, which has a tail of blocks detached from itself, that 
 have not travelled far, of sand, etc., ranging north and 
 south, forms an instructive example of the general law lie 
 would invoke, and he looks upon the asar as the trains formed 
 behind obstacles when the great watery debacle sped over the 
 land ; a view which, as we have seen, had already been urged 
 by Hall. 
 
 He attributes to the transport of the moving materials 
 over the rocks in situ their polish and mammillated contour, 
 and the fissures and grooves, etc., which mark them, and 
 which, as he says, are most frequently found in the pro- 
 vince of Gothenburg and on the borders of Norway and 
 Sweden, near Stromstadt, Hogdal, etc., and he mentions the 
 remarkably polished surfaces which are still to be found 
 wherever the covering is removed. He also remarks that 
 the grooves, like the asar, have a constant direction from 
 N.N.E. to S.S.W. He compares the polished surfaces to 
 similar ones found in Upper Egypt, in the United States, in 
 Cumberland and Westmoreland, etc. 9 
 
 He further claims to have traced the blocks to the granitic 
 and gneissic mountains of Scandinavia, and says that, as might 
 be expected, the point from which the force emanated which 
 dispersed them is the one where they are most infrequent, 
 while they are found most abundantly on high pieces of ground 
 which would naturally arrest their passage. M. Brongniart 
 records that he had found it universally true that the higher 
 one mounts the larger the blocks become. This, he says, he 
 tested on the Supergue, a mountain near Turin, in the Eastern 
 Jura, about Lausanne, and in the Apennines of Castel 
 Arquato, not far from Piacenza, and adds that M. Schultze, 
 in his History of the erratics of Pomerania, had also remarked 
 that che blocks are more abundant on the heights than in the 
 valleys. 1 
 
 In 1829, Phillips published his " Illustrations of the 
 
 9 Id. 10-17. a Id. 17-19. 
 
98 The Glacial Nightmare. 
 
 Geology of Yorkshire/' of which the second edition appeared 
 in 1835. In this work, Phillips was the first to describe in 
 detail the well-known boulder deposit of the Yorkshire coast. 
 He describes it as extending io a connected mass over nearly 
 all Holderness. " In the highest cliffs on this coast," he says, 
 "its thickness is not less than 130 feet. Its composition is 
 remarkably uniform. ' We everywhere observe it to be a solid 
 body of clay, containing fragments of many kinds of rocks 
 which vary in magnitude and in the degree of roundness 
 to which they have been reduced. The fragments are in 
 general not so numerous as to touch each other, but are 
 scattered through the clay as plums in a pudding. . . . The 
 rocks from which the fragments appear to have been trans- 
 ported are found, some in Norway, in the highlands of 
 Scotland, and in the mountains of Cumberland ; others in the 
 north-western and western parts of Yorkshire, and no incon- 
 siderable portion appear to have come from the sea-coast of 
 Durham and the neighbourhood of Whitby. In proportion to 
 the distance they have travelled is the degree of roundness 
 which they have acquired. All the fragments of granite, 
 porphyry, mica slate, and clay slate, which can be compared 
 with no fixed rocks nearer than those of Cumberland and 
 Westmoreland, are rolled to pebbles ; the angles are worn away 
 from every mass of limestone which has been drifted from the 
 north-west hills of Yorkshire, but those which have been brought 
 from the nearer points of the chalk range have yielded much 
 less to attrition. Some attention is required to the original 
 hardness of the stones ; we find solid masses of ironstone and 
 quartz, much less worn than granite ; limestone is less rounded 
 than millstone grit ; and flint retains uninjured angles, whilst 
 chalk and magnesian limestone have lost their original 
 surfaces. 
 
 " Few substances, originally soft, are carried by water to a 
 great distance in a solid form. The sandstones of the western 
 and north-western parts of Yorkshire are plentiful in the 
 gravel of the vale of York ; but only the hard ' galliard ' of 
 Leeds and Bradford, and the solid millstone grit of the moors, 
 can be recognized in the clay of Holderness. . . ." Speaking 
 of the clay, he continues, "We might have expected to find 
 the finer particles at the top, and the solid fragments of 
 
Haitsmann. 99 
 
 rocks lying beneath, according to their individual magnitude 
 and weight. As nothing of this kind is observable, we must 
 suppose the flood to have moved with such extraordinary 
 violence that its spoils, when heaped together, were little 
 influenced in their arrangement by the direct form of gravita- 
 tion." ' In vol. ii. p. 193 of the same work, published in 1836, 
 Phillips says : " The extraordinary phenomena of the dis- 
 persion of the Cumbrian rocks to the east, will require the 
 operation of great and violent oceanic currents excited by 
 sudden subterranean movements." 
 
 Hausmann, in his dissertation entitled " De origine Sax- 
 orum per Germ, septentrionalis regiones arenosas despersorum 
 commentatio," after passing in review the various theories 
 accounting for their origin, goes on to discriminate certain 
 blocks which have come from local strata, including also flints, 
 and which may generally be distinguished by their sharper 
 corners from the great mass of stones which have clearly come 
 from Scandinavia, and he very elaborately describes the 
 various localities in Sweden where the parent rocks are found 
 in situ, and concludes, inter alia, that the blocks have come 
 from the north, some from Scandinavia, others from the 
 borders of the Baltic ; that they were translated at one 
 time, and by one effort, and by the agency of water ; that 
 the phenomenon was a general one. extending over the greater 
 part of the north, and that the sand in which many of the 
 blocks lie is very largely the result of the disintegration of 
 tertiary beds. 3 
 
 M. Elie de Beaumont, writing in 1829, argues that it was 
 by the sudden melting of great masses of Alpine snows, caused 
 by the outburst of gases, such as originated the dolomites 
 and gypseous rocks of the Alps, that great diluvial currents 
 were constituted whose effects are seen in the dispersal of the 
 boulders, etc. 4 
 
 Charpentier urges against this view, that even if it explained 
 the phenomena of the Alps, it would not explain those of the 
 Pyrenees, for it depends on the fact that the Alps were a con- 
 
 2 Op. cit. id. 1835, vol. i. 20 and 21. 3 Op. cit. 3-34. 
 
 4 Recherches sur quelques unes des revolutions de la Surface dn 
 globe, 285, note. 
 
 H 2 
 
ioo The Glacial Nightmare. 
 
 siderable range, and capable of nursing reservoirs of snowbefore 
 their present contour was finally settled, a fact which is probable, 
 but in the case of the Pyrenees there is no evidence that there 
 was any range at all, but, according to M. Elie de Beaumont 
 himself, they were raised by one effort from the surrounding 
 horizontal strata. Again, the elevation of the Pyrenees took 
 place between the deposition of the chalk and tertiary times, 
 which is too early for an explanation of the diluvian phenomena. 5 
 I would add that no invasion of hot gases would melt the 
 glaciers sufficiently quickly to cause the transcendent floods 
 necessary to explain the transport of the boulders, etc. 
 
 In a paper on the Geology of the South of Poland, pub- 
 lished by Pusch in 1830, he adduces fresh facts to show that 
 the majority of the blocks found in that country have not 
 come from the Carpathians, but from Scandinavia. Their size 
 and number diminishes as we leave the Baltic and go south- 
 wards. Their southern limit extends along the environs of 
 Czenstochau, by Pizedborz by the foot of the mountains of 
 Sandomirz, and descending the Kamiona by the Vistula upon 
 Lublin and Lubartof across Southern Lithuania and into 
 Central Eussia, that is a long way from the Carpathians, nor 
 are they formed of materials found in the Carpathians, but 
 they are in fact identical with those found in Sweden, Finland, 
 and Southern Russia. Between St. Petersburg, the Duna and 
 Niemen they consist of a red granite found at Wiborg, a 
 felspathic granite found in Ingria, a red quartzose sandstone 
 found on the borders of Lake Onega and the limestone of 
 Esthonia and Ingria. In Eastern Prussia and in Poland, 
 between the Vistula and the Niemen, of three kinds of granite 
 which are found in Finland, at Abo and at Helsingfors, of an 
 amphibolite either from Central Finland or from Sweden, of a 
 quartzite exactly like the Fjall sandstone occurring between 
 Sweden and Norway, and of porphyry from Elfdalen in 
 Sweden. From Warsaw westwards towards Kalisch and 
 Posen the red granite of Finland becomes scarcer, amphi- 
 bolite, gneiss, and porphyry increase in numbers, there are few 
 specimens of Finland and many of Swedish rocks. These facts 
 agree with those forthcoming from North Germany, the Low 
 Countries and the East of England, whose blocks have also come 
 6 Charpentier, Essai sur les Glaciers, 209-213. 
 
De la Beche. 
 
 from Scandinavia, and he concludes that the evidence forth- 
 coming from Prussia, Livonia, Courland and the North of 
 Russia, concurs in proving a debacle coming from the N.N.B. 
 to the S.S.W., which also carried the blocks into the country 
 N. of the Carpathians. 6 
 
 De la Beche, writing in 1831, calls attention to the fact, that 
 on Great Haldon Hill, about 800 feet above the sea, there 
 occur in the superficial gravel pieces of rock which must 
 have been derived from lower levels, and he says he " found 
 pieces of red quartziferous porphyry, compact red sandstone, 
 and a siliceous rock, not uncommon in the graywacke of the 
 vicinity, where all these rocks occur at lower levels than the 
 summit of Haldon, and where certainly they could not have 
 been carried by rains or rivers, unless the latter be supposed 
 to delight in running up hill. . . . Not only have gravels 
 been brought from various distances, but even huge blocks, 
 the transport of which by actual causes into their present 
 situations seems physically impossible." After referring to 
 various facts quoted by different authors, he continues : " The 
 probability, therefore, as far as the above facts seem to 
 warrant, is, that masses of water have proceeded from north 
 to south over the British Isles, moving with sufficient velocity 
 to transport fragments of rock from Norway to the Shetland 
 Isles and the eastern coast of England, the course of such masses 
 of water having been modified and obstructed among the 
 valleys, hills and mountains which they encountered ; so that 
 various minor and low currents were caused. De la Beche says, 
 referring to Elie de Beaumont, " The appearances are precisely 
 what would have been expected from the passage of a rock- 
 charged mass of waters down the respective channels, the largest 
 fragments having been transported the shortest distances, 
 being most angular, while the smaller and most rounded have 
 been carried the furthest. Thus in the valley of the Durance, 
 the transported substances become more angular and of 
 greater volume as we proceed from the great mass of pebbles 
 called the Crau to the mountains beyond Gap, whence the 
 debris, judging from its mineralogical characters, have very 
 clearly been derived. Similar phenomena will be observed up 
 the valley of the Drac,- which proceeds by another course to 
 6 Journal de Geologic I. 251, etc. 
 
The Glacial Nightmare. 
 
 the neighbourhood of the same mountains, the two streams of 
 debris not mingling till they join in the Crau. De la Beche 
 then goes on to refer to his own observations near lakes Como 
 and Lecco, and notably on the north face of Monte San Primo, 
 which, he says, presents a bold front to any shock coming from 
 the north, leaving open passages to the right and left of it, 
 one down the south part of Lake Como and the other down 
 that of Lecco. Not only in front, facing the high Alps, but 
 also round the flanks and shoulders of the mountain, and 
 even behind it, where the eddy current would have trans- 
 ported them, blocks of granite, gneiss, mica slate, and others, 
 from the central chain, of various sizes, and often accompanied 
 by smaller fragments and gravel, are seen in hundreds, nay 
 thousands, scattered over the dolomite limestone and slate of 
 the mountain, and nearly filling up a previously existing 
 valley which faces the north, the direction whence the rock- 
 charged fluid descended. Proceeding down the side valleys, 
 partly occupied by the lower lake of Como, and the lake of 
 Lecco, we find the evidences of such a current in the presence 
 of blocks occurring, as they should do, where obstacles were 
 opposed to its course, or in situations where eddies would be 
 produced behind the shoulders of the mountains. One very 
 remarkable instance of such occurrence is behind, or on the 
 southern side of, the Monte San Maurizio, above the town of 
 Como, where numerous blocks are accumulated on the steep 
 flank of the mountain, precisely where a body of water, 
 rushing down the great valley, would produce an eddy at its 
 discharge into the open plains of Italy. The blocks, though 
 no doubt many have descended from their first positions in 
 consequence of the long-continued action of atmospheric agents, 
 occupy an elevated line, and occur also on other but lower 
 heights in the vicinity, which opposed more direct obstacles 
 to the debacle ; seeming to show that the blocks floated near 
 the surface of the fluid mass, and were whirled by the eddy, 
 at nearly the same level, against the steep sides of the 
 calcareous mountain, as well as thrown against the more 
 direct obstacle of a range of conglomerate hills." 7 
 
 Having referred to these facts, he proceeds : " Therefore, 
 proceeding south, the course of the waters seems to have con- 
 
 7 Id. 194-6. 
 
Trimmer. 1 03 
 
 tinued in that direction, over the low districts of Germany, 
 to the Netherlands, depositing huge blocks in their passage ; 
 these blocks prove by their mineralogical composition to 
 have been derived from rocks known to exist in the northern 
 regions. Such a movement as this over part of Europe 
 would, if the supposition of a mass of waters were correct, be 
 observed in other northern regions, for the waters thrown 
 into agitation would cause waves around the centre of dis- 
 turbance, and he accordingly refers to the great masses of 
 such erratics in America, and says that, as they also point one 
 way, we can scarcely refuse to admit that the course of the 
 disturbance or disturbances was towards the north, the un- 
 dulations of the waters having been caused by some violent 
 agitation, perhaps beneath the sea in those regions, for it is 
 by no means necessary that it should have been above its level. 
 
 " A convulsion or convulsions of this magnitude, reasoning 
 from the analogy of those minor agitations which we term 
 earthquakes, would be felt over a considerable portion of the 
 globe, and the waters over a large surface would be thrown 
 into agitation. A part of the earth would be greatly disturbed, 
 and we should expect fractures and faults produced in strata 
 where the convulsion was most felt, as similar minor effects 
 are produced at present from the exertion of a less intense 
 force." 
 
 In June, 1831, in the first volume of the Geological Proceed- 
 ings, Mr. J. Trimmer describes the mantle of gravel covering 
 large parts of Carnarvonshire. " A large proportion of this 
 gravel," he says, " is composed of pebbles and blocks of various 
 sizes derived from rocks that occur in Carnarvonshire, many 
 of which are less rolled than pebbles of another class, that are 
 mixed with them and which have come from a greater 
 distance, and must have been drifted upwards by some 
 violent inundation, in a direction contrary to that of the 
 rivers which descend from Snowdonia into the Menai. 
 Among these pebbles are several which can be identified with 
 the granite, syenite, greenstone, serpentine and jasper of 
 Anglesea : other granite pebbles agree with no rock in 
 Anglesea or Wales, and resemble the granite rocks of 
 Cumberland ; some may have come from Ireland or the south- 
 west extremity of Scotland. There are also somefliats which can 
 
1 04 The Glacial Nightmare. 
 
 have come from no other source than the chalk of Antrim. 
 The diluvium occurs in great thickness in the lower region of 
 the valley of the Ogwen, usually from sixty to 100 feet. . . . 
 Beneath the diluvial deposit of this district, when the surface 
 of slate rock is laid bare, it is found to be covered with 
 scratches, furrows and dressings, like those observed by Sir 
 James Hall on the summit of the Corstorphine and other hills 
 near Edinburgh. These furrows and dressings were noticed 
 some years ago by Mr. Underwood. They are referred to the 
 action of the diluvial currents which overspread the country 
 with gravel ; some of the larger blocks amid the gravel have 
 also deep scratches upon their surface." 8 
 
 In a paper by Dr. S. Ware on the Direction of the Diluvial 
 Wave in the Shetland Islands, published in 1831, he remarks 
 on the absence there of the large diluvial deposits found in 
 England and Scotland, and which are represented there merely 
 by detached boulders on the sides of declivities or strewed on 
 the plains. He found on Papa Stour, at a place where the 
 rock in situ was sandstone and secondary porphyry, numer- 
 ous fragments of hornblende schist and actinolite schist which 
 is nowhere met with in the Shetland Islands but at Hills wick 
 Ness, at a distance when measured across the Bay of Saint 
 Magnus of at least twelve miles, that is in a direction about north 
 47 east, a line pointing to the north-west boundary of Norway. 
 He also noticed some large blocks, the largest twenty- three 
 feet high and ninety-six in circumference, situated near the 
 mansion of Lunna, on the east of Shetland, named the stones 
 of Stefis, which seemed to have been moved a mile or two 
 by diluvial currents in a south-west direction. On the east 
 coast of the Island of Yell, which is composed of gneiss, were 
 many large fragments of serpentine and euphotine, which 
 have evidently been drifted from Unst and Fellar. On 
 Roeness Hill, which consists of red granite and is 1447 feet 
 high, were an immense number of boulders of a primary 
 greenstone or trap which seem to have come from a site two 
 or three miles off, and to have been rolled in a south or south- 
 west direction up a gradient ascent of three or four miles on 
 the summit of Hillswick Ness. He also noticed a surprising 
 granite block, mantled with grey moss, removed from a 
 
 8 Proc. Geol. Soc. I. 331-2. 
 
Thompson. 105 
 
 rock the nearest site of which is about two miles north. It is 
 many feet in dimension. "This Thor stone, as it was 
 anciently called, evidently owed its original displacement to 
 diluvial torrents/' He also noticed in Shetland " fragments 
 of stones which are strangers to the country, and which have 
 probably been transported from foreign shores, as in the case 
 of some detached boulders in Soulam Voe, an unsheltered 
 harbour open to the Northern ocean ; one of them, three or four 
 feet high, is of a kind of granite quite unknown in the country, 
 and popular imagination affirms that it was thrown here by 
 the devil as he stood on some high hill in a neighbouring 
 parish ; and he concludes that the great currents which deluged 
 the British Islands as well as some parts of the continent had 
 in Shetland a north-east origin and a south-west direction, 
 and that the evidence of the fragments of rocks from Cum- 
 berland found in Lancashire and Cheshire point to the current, 
 after encountering the headlands of Scotland, England and 
 Ireland, having had its direction changed from a south-west to 
 a south or south-east course." 9 
 
 Reverting again to America, Mr. Thompson writes, Dec. 22nd, 
 1830, in regard to the Alleghany range in Sullivan County, N.Y., 
 " For twenty years past I have been forcibly struck with the fact 
 that wherever the earth has been removed, leaving the horizontal 
 graywacke bare, scratches and deep grooves are observed 
 running a few degrees north of a due east course. I have 
 observed this fact in more than fifty places where the earth has 
 been removed in the construction of turnpikes, common roads, 
 mill works, etc. But they scarcely ever appear on rocks which 
 have been exposed to air, rain, etc. When these scratches 
 first attracted my attention ... I could not resist the 
 inference, that they were made by heavy rocks or boulders, 
 driven over the surface of the upper layers of graywacke by 
 the waters of the deluge. I was even inclined to infer that the 
 scratches indicated the direction of the oceanic waters. I 
 supposed that the general direction was everywhere the same, 
 but that the particular configuration of mountains, valleys, etc., 
 might give local variations to the course of the mighty move- 
 ments of water many miles in depth/' 10 
 
 9 Ed. Journ. of Science, New Series, IV. 85-91. 
 18 Silliman's Journal, xx. 125. 
 
io6 The Glacial Nightmare. 
 
 In a paper on the primitive and other boulders of Ohio, by 
 D. and E. A. Lapham, published in 1832, they describe the 
 gravel of greenstone and granite found in the Ohio valley and 
 not in the neighbouring hills. " It is found mixed with clay 
 and sand, forming banks andhills of moderate elevation, entirely 
 beyond the influence of the Ohio or any other existing current 
 of water. As we travel north the gravel predominates over the 
 clay, and the fragments grow larger. In general, when they 
 exceed the size of one's head, they are called boulders by 
 geologists, but there is no distinct mark by which they are 
 separated, they gradually pass into each other. Proceeding 
 further north the boulders become more numerous, are of 
 larger dimensions, and less worn. From all these facts it is 
 evident that we must look beyond the Great Lakes for the 
 origin of our primitive fragments." 
 
 " The surface of the diluvion everywhere presents a wave- 
 like undulated appearance. These swells are generally more 
 abrupt about Circleville than north of it. One elevation near 
 that place has been mistaken for an ancient artificial mound. 
 . . . The boulders occur more frequently in the beds of ravines 
 and of rivers and the tops of hills. They seldom occur on 
 the surface of the diluvion where it remains entire ; only when 
 the earth has been washed away. . . . The largest boulder we 
 have seen was on the summit of a hill near Lancaster, and 
 was about six feet in length. We are convinced that a great 
 flood has at some remote period swept down from the north 
 and inundated the whole country." 1 
 
 In Woodward's Geology of Norfolk, 1833, he assigns as the 
 wreck of the Deluge the transported gravel, numerous boulders 
 of many tons' weight, and disintegrated portions of all the 
 superstrata of the antediluvian period. The beds of g*ravel he 
 describes as ten to twenty feet thick, consisting of small 
 water-worn pebbles, in some places cemented by oxide of iron. 
 The blue clay constituting the cliffs from Hasboro' to Cromer, 
 is evidently, he says, the wreck of the blue lias brought from 
 the interior of the kingdom by the diluvial waters. 2 
 
 In 1835, Strickland wrote a paper on the deposits of trans- 
 ported materials usually termed diluvium, which exist in the 
 counties of Worcester and Warwick. In this he discriminated 
 
 -. y ' * 
 
 1 Silliman's Journal, xx. 300-303. 2 Op. cit. 14-16. 
 
Murchison. 107 
 
 between what he called the local &&& fluviaiile drifts and the 
 general drift. The latter, he says, has been produced by an 
 agency far more powerful and extended than that which has 
 carried the other two varieties. 
 
 The general drift, again, he divides into a quartzose and a 
 flinty variety. Of the former he says : " From the northerly 
 position of the rocks which have supplied fragments to the 
 quartzose drift we may infer that the current which has placed 
 these gravel beds in their present situation came from the 
 northward. This is further proved by the fact noticed by Sir 
 Roderick Murchison, that as we proceed from Worcestershire 
 towards Cheshire, the boulders are found gradually to increase 
 in size." In regard to the flinty drift, he brings the current 
 which brought this gravel into Warwickshire from some point 
 between N.E. and S.E. 3 
 
 On February 3rd, 1836, Sir R. Murchison read a paper 
 before the Geological Society, in which he pointed out that in 
 South Wales and Siluria the surface is completely free from the 
 far-transported rocks constituting the diluvium of other dis- 
 tricts, and this area is covered with local debris only, which, 
 however, is spread over the country irrespective of the drain- 
 age. This, he supposes, was deposited when the land was 
 generally submerged under the sea. Differing entirely from 
 this local debris is the deposit spread over Lancashire, Cheshire, 
 and North Shropshire, which consists of granites, porphyries, 
 and other hard rocks which have been derived from the moun- 
 tains of Cumberland, and a few perhaps from Scotland. This 
 drift also contains much sand and clay. Its distinguishing 
 feature is the recurrence at intervals of large blocks or 
 boulders of northern origin, some of which have been propelled 
 as far as the edge of the Silurian rocks south of Shrewsbury. 
 They prevail in vast quantities in the high inland district 
 between Wolverhampton and Bridgnorth, from which latitude 
 they begin to diminish in size ; but coarse gravel, composed of 
 the same materials, is prolonged southwards like the tail of a 
 delta, through Worcestershire until it dies away in the fine 
 silt and gravel of the vale of Gloucester. Not a fragment of 
 any such detritus enters into the region of Welsh and Silurian 
 drifts. 
 
 3 Memoirs, 91. 
 
io8 The Glacial Nightmare. 
 
 From the occurrence of remains of recent shells in so many 
 places associated with this drift, he infers that the sea must have 
 formed a strait separating Wales and Siluria from England, 
 and that the drift in question was not driven over the land by 
 diluvial waves but accumulated under the sea, and was brought 
 to its present position by powerful marine currents. Against 
 the notion that the larger boulders were in this district borne 
 by ice rafts, he refers to the rounded and worn exterior of the 
 boulders, and to the fact that they diminish in size and quan- 
 tity in going from north to south. 4 
 
 To these views Strickland gave his adhesion. 5 In a second 
 paper, read before the Brit. Assoc. in 1837, Strickland says : 
 a A considerable portion of the surface of this island is covered 
 by patches of gravel and clay transported from distant 
 localities by forces which are no longer in operation. On 
 examining these masses of drifted material at various points, 
 many remarkable distinctions of character are perceived. The 
 presence or absence of certain varieties of rocks affords a clue 
 to the direction of the forces which have brought them to their 
 present sites. In some cases the pebbles and boulders are 
 derived from the immediate vicinity, in others their origin 
 must be sought for at a distance of many hundreds of miles. 
 Sometimes these beds are found wholly unstratified ; in other 
 cases they are finely laminated, and a violent or tranquil state 
 of the transporting current is inferred accordingly. Some 
 varieties of drift occupy the summits of hills, and are indepen- 
 dent of the present configuration of the surface ; others occur on 
 the sides or bottoms of valleys, with a constant relation to the 
 present line of drainage. The former he distinguishes as 
 marine drift, and he definitely concludes that the great mass 
 of the erratic gravel occurring in England has been brought 
 by a northerly current at a time when all or the chief part of 
 England was under the sea. 6 
 
 In the account of Berwickshire by Mr. A. Home, in the New 
 Statistical Account of Scotland, published about 1835, we are 
 told that across the moor in the upper part of the parish of 
 Greenlow, there runs an irregular ridge called the Kaimes. 
 
 4 Proc. Geol. Soc. II. 232, etc. ; the pagination jumps from 232 to 
 333 "* 
 
 6 Memoirs, 102. 6 Id. 109. 
 
Milne. 109 
 
 It is upwards of two miles long, about fifty feet broad at the base, 
 and between thirty and forty high. " In accounting for it we 
 must remember," says Mr. Home, <( that the stones scattered 
 over fche fields towards the Tweed consist principally of gray- 
 wacke which must have been detached and carried thither from 
 the rocks of the Lammermoor hills by the agency of a current 
 of water setting in from the north towards the south. As the 
 Kaimes are composed of similar materials reduced to a very 
 moderate size, their formation is to be ascribed to the same 
 cause. . . . The waters subsiding at the universal deluge or 
 some more partial inundation, rushing with great impetuosity 
 from the Dorringfcon Laws, which are about two miles north- 
 west from the Kaimes, would carry all the wood and under- 
 wood with them till the collected materials would form a kind 
 of dam or weir through which the waters could not penetrate, 
 and would force up against the dam or weir the gravel and 
 sand which form the ridge in the same manner as is frequently 
 done by rivers in certain parts of their courses during great 
 floods. The materials of which the ridge is composed, and its 
 shape, which is somewhat like a horseshoe with the hollow 
 towards the hills, favour this explanation. 7 
 
 In 1837, Mr. D. Milne published an elaborate paper on the 
 Geology of Berwickshire. In this he says : " Over almost 
 every part of the surface of Berwickshire, and particularly in 
 the eastern part of the county, extensive and deep accumula- 
 tions of loose sand and gravel are found, consisting of water- 
 worn fragments of rocks of every age and description. These 
 gravel and sandy deposits are very generally in the form 
 of high ridges running in an E.W. direction. Good ex- 
 amples of such may be seen at Hardacres, in the parish of 
 Eccles ; and at Kaimknow, betweeen Newton Don and Kelso. 
 Sometimes they form well-rounded knolls, as between Lam- 
 berton Shiels and Marshall Meadows ; these accumulations are 
 also more frequent on the east side of the rocky hills in the 
 district than on the western side. On the estate of Milne 
 Graden, in Coldstream parish, there is a bank of gravel and 
 sand, which runs for about a mile in an east and west direc- 
 tion, and is about half a mile in breadth. The sand lies over 
 the gravel, and is between fifty and sixty feet thick. The 
 7 New Stat. Ace. of Scotland, II. 41 and 42. 
 
I io The Glacial Nightmare. 
 
 fragments embedded in the sand and gravel vary from small 
 gravel to large blocks many tons in weight, the hardest are 
 the largest, but all are more or less worn down to a globular 
 shape. Large blocks of mica slate (a rock which does not 
 occur in the south of Scotland at all, unless perhaps in Gallo- 
 way) have been found near Dunse Castle, at Blanerne (east 
 side), and at Coldstream. Blocks of granitic greenstone may 
 be seen at Billie and at Blackburn, four miles off. These must 
 have come from Cockburn Law, about eight miles W.N.W. 
 from Billie. The characteristic yellow trap of the Dor- 
 ringtons is found in Langton Burn, which is five miles to the 
 E.S.E. The Kyleshill trap has been found at Pitten- 
 hill Quarry, six miles to the E.S.E. The felspar por- 
 phyry of Cowden knows-hill is strewed all over the plains 
 which slope down from it to the eastward. Blocks of green- 
 stone about two centuries ago covered all the country on the 
 east side of Dunse. 8 They have long since been carried off. 
 Similar blocks are spread over the country at Gordon, and can 
 be seen sticking out of the moors to the N. and N. W. of Green- 
 law. Blocks of cherty sandstone are found in the rivulet at 
 Blackburn, while none of this rock exists in situ except at 
 Cumledge, seven miles W.S. W. of Blackburn, and at Hordens, 
 four miles further in the same direction. 
 
 " These facts," says Milne, " afford irrefragable evidence of 
 the nature of the agent by which these alluvial debris have 
 been formed ; that agent could be no other than a great accu- 
 mulation of waters, which moved chiefly from the westward. 
 Other facts point to this not having been the only current. 
 Thus fragments of coal a good deal water-worn are found at 
 Coldstream and on the Tweed near Milne Graden House. 
 The coal nearest this place is six miles to the south, the next 
 nearest localities being Berwick, the Carter and East Lothian. 
 The probability therefore is that these pieces of coal came 
 from the south. Again, the northern face of a hill, two miles 
 north of Spottiswood, is entirely free from soil, while on the 
 other side it is covered with boulders and deep alluvial debris. 
 
 8 In the border history an account is given of a rencontre between one 
 of the Homes of Wedderburn, and a Frenchman of some celebrity who 
 was slain by him near the spot. The Frenchman had fled ; but the 
 historian relates that his horse was impeded and obstructed by numer- 
 ous blocks of stone, which lay scattered over this part of the country. 
 
Lartet. 1 1 1 
 
 In a small valley near St. Abb's Head pieces of red sandstone 
 occur in the debris, which have probably been swept from the 
 northward, where they occur five miles off. These facts seem 
 to warrant the inference that, whilst the general movement of 
 the diluvial waters was from the westward, there were many 
 minor or accidental currents which had different directions." 9 
 In a later page Milne, referring to the Kaimes already 
 noticed, mentions there being more sand at their eastern 
 extremity, where the ridge is lowest, than elsewhere. The 
 highest parts consist of the coarsest gravel. " No one/' he 
 says, " can doubt the aqueous origin of these deposits, and that 
 the water came from the west is shown by the gravel consist- 
 ing almost entirely of graywacke." 
 
 In a letter addressed in 1838 by M. Lartet to M. Arago, and 
 published in the Comptes Rendus, he describes the diluvian 
 deposits of the Lower Pyrenees and the adjoining low country. 
 This diluvium, which overlies the cretaceous beds, he derives 
 from the Pyrenees themselves, and says it could only have 
 been deposited by a great inundation which probably coincides 
 with the upheaval of the ophitic system at the feet of the 
 Pyrenees. He refers to this inundation as the Pyrenean 
 flood. Its violence differed in different localities. It was 
 most marked on the G-ers (Auch) and the Baise (Mirande), 
 where it tore up and denuded the calcareous beds, less so in 
 the valley of the Gimone (Simorre), and still less so on the 
 Save (Lombez). There is not a single granite pebble in the 
 gravels which are found in these valleys. These gravels are like 
 those on the plateaux, and have clearly been derived from the 
 northern faces of the Pyrenees. In the valleys of the Adour, 
 the Garonne, the Save, etc., which drain the central part of the 
 chain, granite pebbles abound. The dispersion of the granite 
 pebbles he attributes to a later flood, which was limited in 
 extent by the mountains bounding the valleys last mentioned. 
 In regard to the origin of the water of this flood, M. Lartet 
 agrees with M. Elie de Beaumont in attributing it to the rapid 
 melting of large masses of snow ; and he accounts for the former 
 larger deposits of snow by the increased elevation of the 
 mountains in old days, urging that this Pyrenean flood 
 
 9 Prize Essays of the Highland Society, v. 226-228. 
 1 Id. 230-233. 
 
H2 The Glacial Nightmare. 
 
 completely destroyed all the animals living in the district 
 which it reached. 2 
 
 In 1838, Sefstrom published an important paper in the 
 Transactions of the Swedish Academy on the furrows occurring 
 on the Scandinavian mountains. This was translated into 
 English by Mr. Thomas. He had noticed, he says, at 
 Fahlun, where the mountain -slopes are largely uncovered and 
 the smoke from the calcining furnaces prevents all vegetation 
 of mosses, that the furrows have all the same principal direc- 
 tion from north to south. This agrees with the fact that frag- 
 ments of the porphyry of Elfdalen, north of Fahlun, are to be 
 met scattered about Fahlun, and that the blocks of granite 
 and limestone met with in Pomerania and Brandenburg bear 
 marks of having come from the Scandinavian granite moun- 
 tains and the limestone beds of West Gothland. Sefstrom 
 accordingly determined to see how far this was a general law^ 
 and he tabulates a large number of careful observations in vari- 
 ous parts of Sweden, and the result was to show that there is 
 a very considerable divergence in the direction taken by the 
 furrows, and this divergence is not at haphazard but has a 
 certain method in it. 
 
 He had noticed, like De Lasterye before him, that the 
 northern sides of the Scandinavian hills are all rounded and 
 destitute of peaks and angles, while on the southern side 
 sharp corners and peaks still remain, and he had named the 
 former, the opposing side, stofcsidan, and the latter, the lee 
 side (lasidan). He was now able to correlate the furrows 
 with the other phenomena and to show that they run over the 
 mountains between the two slopes, as if the steep northern part 
 of the mountain had offered a certain resistance to the flood 
 by which it was smoothed down, while the south side had been 
 protected. 
 
 The flood to which he appeals he thus describes. <( A mass 
 of large and small stones, sand and gravel, by the action of 
 water have been rolled and washed forward over the already 
 exposed surface of the earth, by which these stones rolling 
 against each other in their course, have produced the pebbles 
 which lie collected in heaps on the extensive hills we call 
 asar (ridges) and of which the heaviest, less subject to roll 
 
 2 Coraptes Uendus, vi. 377, etc. 
 
Sef strom. 1 1 3 
 
 under the pressure of the weight of the masses of stone lying 
 above them, have slidden over and around the surface of the 
 mountain and furrowed it in the same manner as a polished 
 surface of marble is furrowed by grains of sand under the 
 pressure of the finger when drawn briskly over it. This 
 flood he named a Boulder flood (Rullsteiisflod) from the 
 boulders which show signs of rolling and are quite different 
 from the erratic blocks with sharp angles seldom or never 
 rolled, which he says appear to have been conveyed by an 
 agency quite different, and he further suggests as a scientific 
 name for this flood the Petridelaunian flood, from ireTplSiov a 
 little stone, and eXaww I roll forward. 
 
 In regard to the date of this flood he urges that it 
 must have been later than the rocks which are furrowed- 
 On the other hand, the fact that Napoleon's expedition 
 found the quarrymen's marks still fresh on the quarries 
 whence the stone was got for the Egyptian pyramids, shows 
 how long-lived such scratches may be. The fact, again, that 
 at the great and little falls of Avesta, uncommonly beautiful 
 furrows exist which make an angle of from 75 to 86 
 to the present direction of the Dalelfven, which itself 
 passes over them carrying sand, stones and gravel without 
 having succeeded in abrading them, justifies us in attributing 
 them to no very recent period. That the contour of Scan- 
 dinavia has not altered since the boulder flood, is shown by the 
 subordinate or side furrows agreeing with the present slopes of 
 the mountains, and if any change has taken place, several 
 combined localities must have been elevated above their 
 former level, for neither large nor small fissures or faults are 
 to be found. Older fissures which are filled with trap often 
 appear, but the trap is furrowed equally with the softer rock. 
 
 Sefstrom not only attributed the distribution of the boulders 
 to his flood, but also the crushing to a fine sand of enormous 
 masses of hard rocky substances, such as gneiss and granite, 
 with which wide districts of Europe are covered, often to a 
 considerable depth, and also the formation of the so-called 
 giants' cauldrons. Hence he postulates that the flood must 
 have been long continued. 
 
 That the boulder flood was of great violence he argues from 
 certain facts, such as the removal up hill of large rocks at 
 
 i 
 
U4 The Glacial Nightmare. 
 
 Blekinga from the beds where they once rested and the 
 shifting of such an enormous stone as that near the church of 
 Pelarne in Smaland, which, it is calculated, weighs seven 
 millions of pounds, and of the various large stones in 
 Germany and Denmark which have come from Scandinavia. 
 The fact that the upper layers of the solid rocks now covered 
 by loose materials are so often found shivered and broken, 
 while in other cases they have been removed and reduced to 
 fragments, is further evidence of the same fact. " It is besides," 
 says Sefstrom, " very commonly to be seen that the boulders 
 are conveyed upwards over the mountain slopes, which have 
 from 33 to 40 inclination, to effect which not a little force 
 must certainly have been required/' 
 
 Since the furrows have been found twenty-one feet below 
 the level of the sea at Carlskrona and more than 1 500 feet 
 above it near Sarna in Dalecarlia, the depth of the boulder 
 flood must have been very considerable. 
 
 In regard to the direction of the flood he points out that 
 the evidence is contradictory. The direction of the furrows 
 round Fahlun, Uddevalla and Westervik, seem to show the 
 flood came from the highest land and flowed down on 
 all sides. Snahattan, N.W. of Christiania, is the highest 
 land in Scandinavia, the furrows at Christiania, however, go 
 to the S.W. instead of the S.E. At Gefle and on the 
 northern coast of Uppland again, the furrows run to the 
 S.W. and not down hill but ascending, and at many places 
 over very steep slopes and hills. The same is the case east of 
 Carlskrona. These facts point to the flood having had a 
 generally N.E. direction. When we examine mountains such 
 as Omberg, Taberg, Billingen, Kinnekulle, and Stora Tiveden 
 (east of Finn rodja church), we find that while on the summits 
 the flood, as measured by the furrows, has followed the general 
 direction, near the foot of the mountains they follow the 
 irregularities of the country. 
 
 In regard to the asar, Sefstrom says the lesser ones are oftenest 
 found on the lee side of some rock. They generally follow 
 the line of the furrows, but are constant in direction, whereas 
 the latter diverge and converge, and he illustrates some of their 
 abnormal shapes by those of the sand and gravel banks continu- 
 ally forming under our eyes in streams and by great floo'ds. 
 
Sef strom. 115 
 
 In regard to the cause of the flood, he states that until 
 further observations have been made it is not possible to 
 say whether the boulders have themselves been impelled 
 against the solid land by a vis a tergo, or whether (by the 
 stoppage or alteration of the earth's revolutionary movement 
 about its axis) it was the earth which moved against the 
 boulders, but he seems to favour the latter view. 
 
 Sefstrom did not content himself with an examination of 
 his own country. In 1836 he travelled through Germany and 
 Switzerland, making observations on the erratic phenomena. 
 He mentions the abundance of drifted flints in Pomerania, 
 while at Montzberg, near Dresden, he remarked that the solid 
 rock seemed much battered by the boulder stones, but there 
 were no furrows. So far as could be judged from the phenomena 
 of the lee side and the stoss side in the mountains of Saxony, 
 notably of that of Winterberg, south-east of Schandau, the 
 boulder flood went from north to south there. So it was in the 
 country higher up the Elbe, from Teschen to Toplitz, where 
 the boulders of sandstone, basalt, gneiss and porphyry are found 
 to the south of the same rocks in situ. Bound Prague, and as 
 far south as Budweis and Linz, are boulders evidently, he, 
 says, of a northern origin, notably blocks of pure quartz 
 which have travelled from the high mountain of Jeschen, south- 
 west of Keichenberg, in the extreme north of Bohemia, but 
 neither there nor in the Saxon Switzerland did he notice any 
 furrows. 
 
 In the Austrian Alps he found both the granitic and 
 calcareous hills with rounded tops, and he makes the shrewd 
 remark that the reason why the Alps in landscape drawings 
 have always a craggy appearance arises from the artists gene- 
 rally choosing a deep valley for their station, from which they 
 cannot see the plateau ; and he also remarks that the 
 Archduke John held that these plateaux'are distinctly furrowed 
 on their rounded surfaces, often to the depth of the cart- 
 ruts of a public road. In the Taunus mountains he also 
 noticed smoothed and rounded surfaces, except on the higher 
 peaks. 
 
 As a proof that the boulder flood had traversed the Hartz, 
 Professor Hausmann quoted to him the fact of certain rents 
 occurring in the shell limestone, which he attributed to it. At 
 
 i 2 
 
1 1 6 The Glacial Nightmare. 
 
 Berlin it was mentioned to him by Professor Rors, how, at the 
 limestone quarries of Riidersdorf, east of the city, on removing 
 the covering of earth, the upper surface of the limestone beds 
 was furrowed and polished. 
 
 Lastly, in regard to Finland, he tabulates some observations 
 made by M.M. HiriakofF and Eochette in 1836, showing that 
 the furrows there, instead of following the general course of 
 those in Sweden, follow the lines of the Finnish water- 
 courses, both those which fall into the Gulf of Finland and 
 those which fall into the Gulf of Bothnia. 
 
 The Danish geologist, Forchhamm er, divided the drift de- 
 posits of his country into those formed of clay and those of 
 sand, both containing stones. In both cases, he tells us, the 
 beds contain marine debris. In the former of deep-sea, and in 
 the latter of littoral species, and he argues that this can only be 
 accounted for by their having been deposited by a greatly 
 agitated sea. Hence he also explains the irregularity and 
 frequent interruptions of the beds, the longitudinal slope of 
 the hills or kames, as that between Heslov and Lystrup in 
 North Zealand, a similar one between Nestved and the hamlet 
 of Mogenstrup, the hillocks of Refsnoev in the northern part 
 of Samsoe, and of Helgenoes in Jutland. He thus explains 
 why the boulders prevail in the Swedish ridges, which are 
 nearer their original home, while sand prevails in those of 
 Denmark. 
 
 Elsewhere he points out that in some countries the largest 
 blocks are found near the sea, while in others they diminish in 
 size as we leave the mountains. He points out how areas occur 
 several leagues in extent in which they do not occur while all 
 around they are abundant, and that while most of them are 
 rounded, a larger number, especially those which are buried, 
 have their angles sharp ; and he is very emphatic that all the 
 blocks found in Denmark and Germany came from Sweden and 
 Norway, those of Prussia and Poland from Finland, while 
 those of North Russia, as far as the Niemen, came from either 
 Finland or the high land about Lake Onega. He says that 
 there is no doubt some of the blocks in Yorkshire came from 
 .Norway. " Others," he adds, " came from Labrador, while 
 others came across the English Apennines/' 
 
 In Holland are sand hills fifty metres high above the sea- 
 
Jackson. 117 
 
 level. In the sands as far as the Scheldt are found numerous 
 erratic blocks which came from the north. About Amersfoort 
 and in the province of Utrecht are blocks of granite and 
 porphyry like rock found in situ in Scandinavia. They in- 
 crease in size in going towards East Friesland, as they do in 
 going from Scania to Sinaland. About Mooker, on the other 
 hand, are blocks of pyromacic silex which could not have come 
 from Scandinavia. These blocks are not dispersed, he says, 
 along definite lines, but seem as if they had been distributed 
 like the spray from a fountain in semi-circular lines. 
 
 In regard to the asar or sandasar, he says, that while 
 in Sweden they generally range north and south, in Fin- 
 land their direction is more variable and sometimes they 
 cross each other. He attributes to the action of stones 
 borne along by water the polishing and striating of the rocks in 
 Sweden, the striations being generally parallel, and from north 
 to south. In other countries they go from west to east, and 
 sometimes affect different directions on the same mountain. 
 All the heights between 40 and 43 of long, and 60 
 and 62 of lat. are rounded and deprived of salient 
 angles, and in the case of granite are rounded like the suc- 
 cessive waves of the sea, and when protected by gravel are 
 polished so as to reflect the sunlight, and marked by scratches. 
 
 Leonhard and others had argued that these phenomena 
 were caused by violent invasions of the sea, which after 
 inundating Scandinavia and Finland, transported the debris 
 far and wide. Forchhammer somewhat modified this view in 
 supposing that the rounded outlines of the Scandinavian 
 hills are due to the ordinary action of the waves when 
 the country was submerged, but he appeals to violent up- 
 heavals and subsidences such as are evidenced by the 
 faulted and broken strata of the isle of Moen, the inclined 
 beds of Halgoland, etc., as setting in motion the necessary 
 currents to transport the blocks as we find them. 
 
 Let us now turn for a short space to America. Dr. 
 Jackson, in his remarks on the geology of Maine, writing in 
 1838, says, "How much I should be delighted could any of 
 those geologists in Europe, who feel sceptical on the subject, 
 visit with me the numerous and most decided proofs which are 
 presented in Maine, demonstrating the general direction and 
 
1 1 8 The Glacial Nightmare. 
 
 power of the diluvial current. I think that no reasonable 
 person would venture to stand in opposition to the facts which 
 we can present in illustration of this doctrine. ... I have 
 found strong diluvial marks extending across the whole state 
 of Maine, and the transported boulders are always traced to 
 their origin in the direction of the furrows worn on the rocks. 
 .... The direction of the diluvial current in Maine is proved 
 to have been from N. 10 to 20 W., to S. 10 to 20 E., the 
 mean being 1ST. 15 W. to S. 15 E." 3 
 
 In an elaborate geological report on Maine published in 
 1839 by the same geologist, he says, inter alia : "In various 
 sections of this report may be seen recorded the proofs of 
 diluvial transportation of rocks, far from their > parent beds, 
 and we have every reason to believe that this removal was 
 effected by a tremendous current of water, that swept over 
 the state from north 15 W. to south 15 E. ; and we have 
 adduced in testimony that such was the direction of that 
 current, numerous grooves, furrows or scratches upon the 
 surface of the solid rocks, in place, and have shown conclu- 
 sively that the rocks that we find thus transported were 
 portions of ledges situated to the north of the localities 
 where their scattered fragments are found/ 7 Dr. Jackson 
 then goes on to urge that the fact of rocks only weighing- 
 half as much when immersed in water makes the transporting 
 of big blocks much easier to realize, and he continues : " From 
 the observations made upon Mount Klaadu, it is proved, that 
 the current rushed over the summit of that lofty mountain, 
 and consequently the diluvial waters rose to the height of 
 more than 5000 feet. Hence the ancient ocean which rushed 
 over the surface of the State was at least a mile in depth . % . 
 we find that the whole mass of loose materials on the surface 
 has been removed southwardly ... all the markings on the 
 surface of the rocks, and the scattered boulders of granite and 
 gneiss which abound in that soil, indicate its origin to have 
 been in the north 15 or 20 W. . . . Diluvial grooves on 
 the rocks are exceedingly common in Maine, but I know of 
 few localities where they are so distinct as at Hope and Apple- 
 ton. Here they may be observed running in a N.W. and S.E. 
 direction, while they are very deep and perfectly defined. 
 
 8 Silliman's Journal, xxxiv. 71-72. 
 
Shepard. 119 
 
 Their direction, it will be remembered, does not coincide with 
 that of the stratification of the rock, and could not have 
 resulted from disintegration of the different strata. Three- 
 quarters of a mile S.E. from a hill in Appleton, they may be 
 seen forming deep channels in the rocks, to the depth of, a 
 foot, and six inches in width. Since the direction and appear- 
 ance of these grooves correspond with those observed in other 
 parts of our country, I feel no hesitation in attributing them 
 to a similar origin. They are certainly the results of an 
 aqueous current, which once prevailed over New England, and 
 probably over the whole world. The current, from 'similar 
 grooves seen in other places, appears to have proceeded from 
 N. to S., or from N.W. to S.E." 4 
 
 In regard to Illinois, Dr. Shepard, writing in 1838, says : 
 " The occurrence of boulders on the rolling prairie had often 
 been mentioned to me, under the significant and original 
 appellation, bestowed upon them in this region, of ' lost rocks/ 
 Their abundance, however, surpassed my expectation.^ They 
 first attracted attention soon after leaving the twelve-mile 
 house from Chicago, and appeared to form a belt between a 
 quarter and half a mile in width, whose direction was N.W. 
 and S.E. In crossing this belt it was uncommon to pass 
 many rods without encountering a boulder. In general they 
 were rather more than half buried in the soil. They varied in 
 diameter, from ten inches up to three feet, and belonged to 
 the following species of rocks ; granite, granitic gneiss, and 
 trap. A few detached boulders only were noticed between this 
 deposit and Ottawa. Soon after leaving this place, another 
 batch of them was passed ; they were scattered over the Illinois 
 bottoms so plentifully, as to prove objects of no inconsiderable 
 annoyance in the road. . . . What serves materially to heighten 
 one's interest in these boulders is the consideration that they 
 must have been transported over a distance of between two 
 and three hundred miles, since the southern shore of Lake 
 Superior is the nearest region affording rocks of a similar 
 character, in situ." 5 
 
 From a report on the geology of Indiana, published in the 
 same year, we gather that boulders of primitive rock are 
 
 4 Silliman's Journal, xxxvi. 153-4. 5 Id. 141, 142. 
 
I2O The Glacial Nightmare. 
 
 numerous in that State, where they are called by the expressive 
 name of lost rocks, also grey heads, and negro heads. 6 
 
 In 1839 Dr. Hays published a letter in Silliman's Journal, 
 which accompanied some specimens of rocks showing grooves. 
 In regard to one of these he says : " You will perceive, that 
 wherever a nodule of chert projects above the surface, a 
 ridge of the softer limestone has been protected, in some 
 measure, from friction, which invariably at this locality, as at 
 the Black Kock quarry, one and a half miles distant, points in 
 a southerly direction .... the nodules of chert often have a 
 semicircular depression worn into the rock on their northern 
 
 sides, opposite to the projecting ridge Can proof be 
 
 more conclusive, that these marks and scratches were produced 
 by gravel stones and boulders, swept over the surface of the 
 rocks by currents, tides or waves, which flowed from the 
 north ? " 7 
 
 Eeverting once more to Europe: on the 10th of August, 1840, 
 M. Durocher presented a famous memoir to the French Academy, 
 entitled <f Observations sur les phenomene diluvien dans le nord 
 de 1'Europe." In this he summed up the results of long and 
 careful researches in Kussia, Poland and Finland. He especially 
 examined the furrows and fissures on the northern rocks, from 
 Altea on the Arctic sea to the south of Finland, and from the 
 gulf of Bothnia to Lake Ladoga. These striae, he says, are found 
 everywhere in the north, where the'rocks are sufficiently hard and 
 fine-grained to have preserved them. They vary in Finland 
 from a length of a few inches to a foot or more. They are often 
 parallel and the larger grooves are often themselves traversed 
 by finer ones. Sometimes on the same rock two sets of stria3 
 are visible with an angle to each other of ten to twelve degrees. 
 Sometimes there are deep striae in one direction and fine ones 
 in another. Sometimes the deep striae intersect one another. 
 The direction of the strise is quite independent of the stratifi- 
 cation. They occur on unstratified as well as stratified rocks 
 and often make an angle with the dip of the strata. They 
 occur less frequently and less regularly on irregular ground 
 than when the surface is more or less regular. On perpendi- 
 cular faces of rock the grooves are sometimes horizontal, as 
 
 6 Silliman's Journal, xxxiv. 194. 7 Id. xxxv. 191. ** 
 
Durocher. \ 2 1 
 
 on the borders of Lake Ladoga, a fact previously observed 
 by De Saussure on mount Saleve. The mammillated rocks of 
 Finland are entirely covered with scratches, those on the 
 summits of the hillocks having a normal direction from 
 N.N.W. to S.S.E., while the others have a tendency 
 to follow the sinuosities of the surface. The surface of 
 these rounded rocks is always polished, and they generally 
 have a curved outline elongated in the direction of the striee. 
 It is only 011 the low hills of Finland that the whole surface is 
 striated. On the higher hills above 100 feet in height, as 
 Sefstrom and De Lasterye showed, the faces on the S.S.E. 
 have preserved their original roughness. Although the 
 normal direction of the striae is pretty constant they occasion- 
 ally and in certain localities, as notably between Brakestad 
 and Gamle Carleby in Finland, diverge considerably, a 
 fact also noticed by Bohtlink. The normal direction of the 
 strioe is very nearly that of the valleys and lakes, and this 
 both in Sweden and Russia. When the erratic detritus ceases 
 to lie in continuous mounds and is broken up into hillocks, 
 etc., the alignment of these in Russia corresponds in direction 
 with the normal direction of the striae, as Brongniart had 
 shown that they do in Sweden. In Finland and Lapland 
 the loose deposits are mostly unbroken and spread in a 
 continuous mantle, in many places thirty and forty feet 
 thick. 
 
 In regard to the erratic blocks, he remarks that all through 
 the north of Europe they occur not only on the surface but 
 buried in the sand, and when so buried their axes are often 
 inclined to the horizon or even vertical, as if they had 
 fallen all at once and been buried in the sand below. They 
 occur most frequently on hillocks or high ground ranged in 
 lines or bands. The larger blocks are never rounded ; some 
 of them have their angles exceedingly sharp, and when they 
 are blunted it may be due to atmospheric erosion. In 
 Finland and Lapland the blocks are of local origin, but north 
 of Lake Ladoga there are blocks of a rock not occurring in 
 situ at less than a distance of fifteen leagues further north. 
 M. Durocher discusses the provenance of a large number of 
 blocks found in Poland and North Germany agreeing with 
 the conclusions already cited. He says that as we travel 
 
122 The Glacial Nightmare. 
 
 westward from Poland the blocks from Finland decrease and 
 those from Sweden increase, and that on the zone between 
 Tilsit and Berlin they are much mixed. In Denmark the 
 blocks are generally of Norwegian and Swedish origin. 
 M. Durocher remarked that everywhere large blocks of 
 granite exceed greatly in number those of gneiss, although 
 gneiss is the prevailing rock ; and he accounts for the fact by 
 the tendency of gneiss to split up into fragments. He re- 
 marked further that, taking the original site of each kind of 
 stone out of which the erratics are formed as a centre, the 
 present distribution of the latter occupies a third of the cir- 
 cumference of the generating circle, or even one half, so that 
 some of the stones have moved at right angles to others. He 
 quotes as an example that the rapaikiwi (a peculiar granite) of 
 Wiborg is found from Kostroma in Russia as far as Pome- 
 rania. 
 
 The granite blocks have travelled further than the rest, and 
 the calcareous blocks the least. The blocks are ranged not 
 only in a fanlikeway but also in concentric zones. Generally 
 speaking, the range of the blocks may be measured by a sec- 
 tion of a cirqle with a line joining Stockholm and Moscow as a 
 radius and the former place as the centre. As we near the 
 circumference the blocks diminish in number both in Eussia 
 and Poland, while in Saxony and Silicia, where the curve is 
 broken and indented, they stop short more suddenly. This 
 Durocher accounts for by the fact that in the former 
 district the transporting agent gradually exhausted itself, 
 while in the latter it was abruptly stopped by the mountains. 
 The blocks over the vast area of Russia, North Germany and 
 Poland are not all far-travelled erratics, some are derived from 
 the rocks in situ. 
 
 When the travelled blocks cease the local blocks continue, 
 as do also the beds of sand and soft material in which they 
 are embedded, so that there is a zone encircling the outer 
 margin of the transported drift composed of local materials 
 otherwise continuous with it. 
 
 In regard to the cause of the varied phenomena thus 
 described, M. Durocher distinguishesjbetween the agent which 
 transported the boulders and that which made the striae and 
 gave the asar their shape and alignments. The former he 
 
Durocher. 123 
 
 attributes to ice, the latter he assigns to the operation of a 
 diluvial current starting to the north of Scandinavia and 
 sweeping over Lapland, Sweden and Finland in the direction 
 marked by the normal direction of the strias, etc., and explains 
 the local variations by the deflecting influence of mountains 
 and other barriers. His reason for putting the focus of dis- 
 turbance to the north of Scandinavia is, that the striae which 
 further south have somewhat different directions, when we 
 reach the 70th parallel in Lapland and in Finland take 
 almost the same direction, and that they can be traced not 
 only on the uplands but on the northern face of the rocks 
 which border the polar sea. 8 
 
 I do not propose to carry this part of my story any further, 
 nor, in fact, would it be easy to do so, for at the epoch we 
 have reached there arose another school of writers, who would 
 have nothing to say to diluvial theories, and whose vigorous 
 blows and acute rhetoric caused a very rapid revolution in the 
 whole course of geological thought. 
 
 It will be noticed that among the names quoted in this and 
 the previous chapter there occur nearly all those who are 
 remembered as the fathers of geology ; yet in the course of 
 half a dozen years the views they had in various ways main- 
 tained in regard to the drift phenomena were almost entirely 
 displaced, and one at least among them, the very standard- 
 bearer of the diluvial school, went over to the enemy and 
 became an advocate of the new views. Among those who re- 
 mained staunch to the old creed the names of Sedgwick, Phillips, 
 Von Buch, Durocher, and Sefstrom are the best known, but they 
 preached in vain to the younger students of geology. The 
 champions of water had to give place to the champions of ice. 
 How this came about will be described in the succeeding 
 chapters. 
 
 8 Comptes Kendus, xiv, 78410. 
 
124 The Glacial Nightmare. 
 
 CHAPTER IV. 
 
 THE PHENOMENA OP THE DRIFT. 
 
 The Champions of Floating Ice. 
 
 " Saxa singula, petris primordialibus constituta, per Poloniam septem- 
 trionalem dispersa, incolis lapides Suecicos appellari ; quam notitiam 
 auditor! studiosissimus L. Zeuschner, Polono, debeo. Utrum Poloni in 
 Suecia, an Sueci in Polonia Saxorum illorum cum petris Suecicis 
 similitudinem agnoverint, decernere non possum; non improbabile 
 autem videtur, denominationem illam, a veritate geologica non abhor- 
 rentem, e bellis, inter nationes illas gestis, origin em suam traxisse." 
 HAUSMANN, Ac. G. A. Comm. Eec. Vll. 15-16. 
 
 Tilas, 1740-2 Bergmann, 1769 St. Pierre, 1784 Dolomieu, 1793 
 Mierotto, 1790 Wrede, 1794 Yon Buch, 1808-1811 Bakewell, 
 1815 Sir James Hall, 1815 Venturi, 1817-18 Diet, 1819 Horace 
 H. Hayden, 1820 Bigsby, 1821 Eaton, id. Maclure, 1822 
 Hitchcock id. Dobson, 1825 Fischer, 1828 Lariviere, 1829 
 Kloden, id. Lyell, 1830-1845 De la Beche, 1833 Murchison, 1835- 
 1841 Bay field, 1837 Cordier, 1837 Darwin, 1839-1841 Wissman, 
 1840 Hays, 1843. , 
 
 SCIENTIFIC men naturally shrink from appealing to abnormal 
 causes. It is of the nature of science that we should 
 endeavour to explain the operations of Nature by invoking 
 phenomena and laws which we see at work around us rather 
 than by remote or occult causes. When it began to be seen, 
 therefore, that it is impossible to account for the transport 
 and distribution of the boulders and to explain their associated 
 phenomena by a reference to water acting in its usual way,, 
 and that especially in North Germany and in other countries 
 remote from mountains and impetuous rivers we must, if we 
 appeal to water at all, have recourse to its action in a quite 
 abnormal way, men began to turn elsewhere to find some 
 
St. Pierre. 125 
 
 explanation which should not involve other than forces 
 actually at work and working in a normal way. It was 
 natural that those living near the Baltic and in North 
 America, who had opportunities of noticing the tremendous 
 force exercised by shore ice and by floating icebergs, should 
 turn to these causes and supplement them by theories of 
 wide-spread submergence, which seemed also warranted by 
 the finding of marine shells at a considerable elevation in 
 many places. 
 
 Among the first to record such observations was the Swede, 
 Til as, who suggests that some of the larger stones found in 
 Sweden were probably carried by floating ice which came 
 from the Pole. 1 
 
 T. Bergmann, writing in 1769, speaks of the way in 
 which ice being frozen hard to the ground detaches from 
 it, occasionally, earth and stones, and when itself loosened 
 bears them away and deposits its burdens elsewhere. Thus 
 he mentions how a big stone, eight to ten feet high and nine 
 feet thick, which lay near the church of Kalajoki a foot above 
 the water, was taken up by the ice, moved over the land, and 
 then so deposited on some smaller stones that a man could creep 
 under it. In the year 1760, the ice turned over a stone twelve 
 to fourteen feet in length, at Breitgrund, and the year follow- 
 ing it was moved still further. 3 While he points out how ice 
 can occasionally move great stones, he urges that we cannot 
 explain by its action the raising up of huge blocks and their 
 deposition on the summits of high mountains. 
 
 St. Pierre, in his " Etudes de la Nature/' published in 1784, 
 argues that the tides of the ocean are largely due to the 
 alternate thawing of the ice at each pole of the earth, 3 and 
 he urges that if this melting were to take place suddenly or ' 
 rapidly it would cause a general inundation, just as rapid 
 melting of the snows on the Andes gorges the Amazon and 
 the Orinoco, and the similar melting of the snow on the 
 Mountains of the Moon floods the Senegal, the Nile,, etc. 
 He argues that at the time of the great deluge, the sun left 
 the ecliptic and advanced further north and south, and in 
 consequence the cupolas of ice at .either pole and the great 
 
 1 Ante, p. 9. 2 Phys. Besch. der Erdkugel, 399. 3 Op. cit. i. 240. 
 
125 The Glacial Nightmare. 
 
 glaciers on the high mountain chains rapidly melted, and 
 innumerable torrents, escaping by the various outlets of the 
 polar basin, swept into the Atlantic charged with masses of ice, 
 carried with them wreckage from the land and sea, broke 
 down mountains and transported rocks, sand and gravel. To 
 continue in our author's own words, " Ce fut alors que tous 
 les plans de la nature furent renverses. Des iles entieres de 
 glaces flottantes, chargees d'ours blancs, vinrent sechouer 
 parmi les palmiers de la zone torride, et les elephants de 
 FAfrique furent roules jusques dans les sapins de la Siberie, 
 ou Ton trouve encore leurs grands ossements," etc. 4 
 
 Dolomieu, in discussing the question of the transport of 
 erratics, tells us how he had seen ice transport to great dis- 
 tances stones which running water could not move. 5 
 
 De Luc reports Mierotto as saying he " had had oppor- 
 tunities of seeing in winter, on the coast of Pomerania, that 
 while the sea is at its lowest level, the ice found on its shores 
 encloses some of the blocks. Afterwards the level of the water 
 rising and the ice breaking on the borders, detached masses 
 of it containing these blocks are cast up on the strand, where, 
 when the ice melts, the blocks remain. He argued from this 
 that formerly the whole north of Europe was covered by a 
 sea which was kept at a high level by a succession of 
 barriers of rocks, forming a kind of dike around it .... 
 that in consequence of ruptures in the dike, portions of this 
 sea rushed down with impetuosity into lower basins, the dikes 
 of which, in their turn, were also broken and that the blocks in 
 question are fragments of these dikes .... the fragments 
 were brought down by the rush of waters, having been carried 
 to the borders of each basin by ice in the manner above 
 described." 6 Elsewhere he reports the same geologist as 
 urging that " the masses of stones scattered in low situations 
 were fragments of dikes, whereby superior seas had originally 
 been retained in successive basins, one lower than another, 
 each of which had been emptied in its turn by the rupture of 
 its dikes ; that this having extended all the way into Sweden, 
 masses of the granites and porphyries of that country had 
 
 4 Id. 245-248. 5 Obs. sur la Physique, xiii. 59, note. 
 
 6 De Luc's Travels, 1813, i. 420, 421. 
 
Wrede. \ 2 7 
 
 been brought over to the lands on the south of the Baltic, by 
 the ice of that sea, at its successively lower levels. 7 
 
 Mierotto was presently followed by Dr. Wrede, the author 
 of two remarkable works, one published in 1794, entitled 
 "Geologische'^Resultate aus Beobachtung ueber einen Theil der 
 Siidbaltischen Lander," Halle, 1794 ; and a second, entitled 
 " Geognostiche Untersuchungen ueber die Siidbaltischen 
 Lander." 
 
 His views were based chiefly on the facts observed in 
 Scandinavia and north Germany. He was of opinion that 
 a change has taken place in the centre of gravity of the 
 earth which has been moving southwards and has caused 
 a corresponding movement of a large part of the watery 
 envelope of the globe to the Southern Hemisphere. He 
 urges that formerly the Baltic was a much more extensive 
 sea than it is at present, and quotes with approval the notion 
 of Von Buch that the flat country between the Oder and the 
 Volga, between the Baltic and the Carpathians, and between 
 the Black Sea and Finland was once a vast sea. 8 Having 
 postulated this sea he goes on to discuss in great detail the 
 various modes in which the erratics reached their present 
 position in the country south the Baltic, and continues on page 
 71 of his first work in these words : " In our experience we 
 know of only one means of by which these effects can have been 
 produced, namely by ice, in which granite blocks of this size 
 were frozen, which were then carried away until the ice was 
 pounded into fragments and eventually sank with its load." 
 In support of this view he quotes a statement of H. Gotze, 
 vol. ii. p. 237, that annually blocks of granite, sometimes of 
 twenty hundredweights, are detached in the Hartz mountains, 
 etc., and are carried in the ice, in which they are fast, and re- 
 moved for several miles. Plouquet, in his Swiss Travels, 1786, 
 tells us the district between the lake of Brienz and that of Thun 
 is strewn with debris similarly deposited by the Lutschine, and 
 argues tbat both those lakes will eventually be filled up ; the 
 Rhine similarly carries blocks into the Boden lake and the Rhone 
 into the Genser lake. 9 Dr. Wrede urges that if it be argued 
 that the erratics are too large to have been carried by such ice- 
 bergs as are now detached from the mountains of the Hartz, of 
 ? Id. ii. 43. 8 Geog. Unter., etc., 39. 9 Op. cit. 71, 72. 
 
128 The Glacial Nightmare. 
 
 Switzerland or Norway, that the higher mountains of former 
 days would have greater torrents which would bear larger 
 masses of ice. He calculates that inasmuch as the specific 
 gravity of ice to salt water is as 6 to 7, and to fresh water 
 as 8 to 9, the mean being as 7 to 8, that a block of granite 
 weighing 389,078 Ibs. might have been carried by a raft of ice 
 55,683 cubic feet in extent, and he shows that a depth of 
 water of forty-four feet would amply suffice to float it. 10 If the 
 stones, instead of being more or less cubical in shape, were 
 spread out so as to be about the same thickness as the 
 ice, then a much smaller depth would suffice to float the 
 bergs with their loads. He urges that if only one of each 
 group of blocks of ice which crowd the present streams in 
 spring were to carry a stone, there would speedily accumu- 
 late a considerable deposit. He also attempts to show 
 how, under the conditions he names, large stones would 
 sometimes be rounded and polished. His argument does 
 not imply that the erratics south of the Baltic came from 
 Scandinavia specially, but from the various districts in the 
 north where rocks like those from which the boulders were 
 separated are found. 
 
 Similar views, it would seem, had also been propagated 
 elsewhere. Yon Buch tells us expressly that, according to 
 those most worthy of credit in Switzerland, the blocks of stone 
 had fallen from their original position upon icebergs which 
 floated in the sea then covering that country, which had been 
 driven by currents to the Jura, where they deposited their 
 loads. Others deemed this view of immense icebergs inadmis- 
 sible, and invoked natural rafts of trunks of trees like those 
 sometimes seen on tropical rivers as the transporting agents. 
 Von. Buch urges the impossibility of supposing that such ice- 
 bergs or rafts would all float in the same direction. They 
 again would have deposited their loads at one level, namely, 
 that of the then sea-level, and not at different elevations. 
 
 Bakewell, in his " Introduction to Geology/' second edition, 
 published in 1815, says vast " masses of rock near the sea-shore 
 are sometimes enveloped in fields of ice, and raised up and 
 transported to distant countries. Ice is specifically lighter than 
 water; every cubic yard will support a sfcone of 100 Ibs. weight ; 
 
 10 Id. 71, 72. 
 
Dick. 129 
 
 hence we need not be surprised at the insulated rocks of granite, 
 that are sometimes found in situations far remote from primary 
 mountains. These blocks have been floated over the ocean, 
 and their angular points and edges defended from attrition 
 during their passage by the surrounding ice. In this manner 
 large fragments of granite, and other primary rocks, may have 
 been brought upon our coasts from Norway and Greenland." ] 
 
 Sir James Hall, writing in the same year, says he is induced 
 to retain part of the system proposed by Wrede, involving the 
 transport of blocks in some cases by ice. Having referred to 
 the case, to be presently cited, of the Fitly stone, he goes on to 
 say that a wave sweeping over the Alps would meet with 
 many stones thus embedded in ice, which it would move 
 along. 2 
 
 He elsewhere urges that the fact of the blocks in Germany 
 and the Baltic islands being found on hills of sand and not 
 on the low ground, points to " these hills then constituting 
 shallow sandbanks, which would afford the first resting-place 
 for the floating bergs of ice which, grounding upon them, 
 would accumulate to very numerous assemblages, and there 
 deposit their granitic charges, while all other similar blocks 
 flowed onwards in the deep water between." 3 
 
 In 1817 and 1818 the Cavaliere Venturi published some 
 papers in Brugnatelli's " Giornale de Fisica," in which he dis- 
 cussed the mode of transport of the Alpine blocks, and after 
 quoting the views of Wrede and Hall with approval, he says 
 that the main cause of the transport of the greater number of 
 them was ice. When the sea was several thousand feet 
 higher and bathed the feet of the Alps, the glaciers coming 
 down the Alpine valleys carried stones ; when these glaciers 
 reached the sea they broke off into icebergs, which were blown 
 along by winds and taken by currents, and thus deposited 
 primitive blocks from the higher Alps on other strata." 
 
 Mr. Thomas Lauder Dick read a paper in 1819, before the 
 Wernerian Society, on a famous boulder known as the Travelled 
 stone of Fitly, measuring about four to six feet in height and 
 six to seven in thickness and height, and weighing about eighty 
 tons, which, up to the 19th of February, 1790, served as the 
 
 1 Op. cit. 55 2 Trans. Roy. Soc. Ed. vii. 157, 158. 
 
 3 Op. cit. 160. 4 Op. cit. decad. 1. vol. x. 297. 
 
 K 
 
1 30 The Glacial Nightmare. 
 
 march stone between the properties of Castle Stuart and 
 Culloden. It lay just within the flood mark, near where a 
 little stream empties itself into a shallow sandy bay. There 
 had been a long-continued frost, which had frozen the upper 
 part of the bay, and a thickness of about eighteen inches of 
 ice clasped fast the stone. On the evening of the 19th of 
 February, a hurricane arose and broke up the ice. In the 
 morning the miller at the mill of Fitly missed the stone, and 
 he called out to his wife that the " meikle stane was awa," and 
 there remained but a hollow pit in the sand with a long 
 shallow furrow stretching from the pit towards the sea. 
 Presently, when the tide ebbed, the stone was found lying out 
 on the sands, 780 feet from its old position, and around it was 
 still a cornice of ice. "In its new position/' says Hugh 
 Miller, writing in 1859, " the stone still lies ; and only a few 
 years ago, mayhap still, a wooden post which marked the 
 point where the two contiguous properties met, marked also 
 the spot from which, after a rest of ages, it had set out on its 
 short voyage." ! 
 
 The same graphic writer refers to a second instance of a 
 similar kind which occurred at the beginning of the century on 
 the eastern coast of Sutherlandshire. " Near the small hamlet 
 of Toridal, on the upper part of Loch Fleet, there stood/ 5 he 
 says, " about fifty years ago, a rude obelisk of undressed stone, 
 generally regarded as Danish. It stood in a swampy hollow, 
 which was frozen to the bottom. On the evening before the 
 incident to be related, a drunken Highlander, returning home 
 put his bonnet on the top of it and left it there. The next 
 day the thaw came, the river rose, the ice- cake round the 
 obelisk floated high above the level, wrenching up the obelisk 
 along with it, and both floated down the loch to the sea. In 
 the morning report spread that there was a man standing in 
 the middle of the loch on the floating ice, being swept out to 
 sea. A boat put out from the shore, when the supposed man 
 was found to be the obelisk with a bonnet on it, and bonnet 
 and obelisk were left to find their way to the German Ocean, 
 where both now lie/ 7 
 
 Horace H. Hayden, in his " Geological Essays/' published 
 in 1820, while assigning a great current as the cause pf the 
 
 L Geological Sketch-book, 69, 70. 6 Id. 71, 72. 
 
Horace H. Hay den. 131 
 
 distribution of the American drift, also invokes the assistance 
 of ice. Thus, he says, " it is well known that large masses of 
 rocks of different kinds are often enveloped in ice, and by the 
 freshes of rivers or tides of the sea are raised and transported 
 to great distances, before they are disengaged." Tilloch had 
 observed that masses of stone are sometimes transported by 
 cakes of ice in which they happen to be enveloped, and Hay- 
 den goes on to say that just as white bears have been known 
 to be transported to Iceland by icebergs, so might great masses 
 of rock be similarly transported. 7 He elsewhere describes the 
 floods which took place on the Connecticut after the severe 
 winter of 1780, when great masses of ice were scattered over the 
 country on either side of it. " When this was accomplished, 
 the following season large pieces of rocks and heaps of rolled 
 pebbles were left exposed to view on an alluvial surface, on 
 which, before, a stone could not be found for its weight in gold. 
 These rocks and stones, from their characters, were known to be 
 the same as those which composed the bed of the river many 
 leagues above." And in discussing the great blocks found on 
 the surface of Massachusetts, he urges that they were borne 
 along by water when encased in ice, " for such is their form 
 that if embraced in a mass of ice of sufficient size, and elevated 
 from the earth by water, they might be floated over the ocean 
 without the chance of escape, except by the melting of the ice." 8 
 The intervention of ice in carrying the blocks he urges 
 as another reason for bringing his diluvial currents from 
 the north, " for/' he says, " it would be absurd in the 
 extreme, to suppose that masses of ice could have been 
 found in the torrid zone, or even within the temperate zone, 
 of sufficient magnitude to have transported those rocks from 
 the south to their present position. . . . This opinion re- 
 ceives additional support from the following circumstance : 
 the huge fields and mountains of ice, which every spring 
 float down the Atlantic Ocean, seldom reach the latitude of 
 from 39 to 40 (which is about the latitude in which those 
 rocks lie), before they become so weak and rotten as to be 
 incapable of supporting or retaining any considerable weight 
 that might be attached to them. It therefore follows as a 
 natural consequence, that by the time the masses of ice 
 
 ? Op. cit. 86. 8 Op. cit. 89-91. 
 
 K 2 
 
132 The Glacial Nightmare. 
 
 had reached those latitudes, they must necessarily have dis- 
 charged the entire balance of their weight, if they contained 
 any." 9 
 
 " In the spring," says Dr. Bigsby, writing in 1821, " the ice 
 occasionally removes fragments of great size. During the 
 winter it surrounds those which are placed in the shallows, 
 and on being broken up in April by rnild weather and a 
 casual rise of water, it carries them to some other shore. 
 Remarkable instances of this are found on the islets near 
 the south end of St. Joseph, where a few yards from the 
 water, and little above its level, are deposited rolled 
 stones some yards in diameter, with a furrow extending 
 from them to the water, most probably tracing the last 
 steps of the route to their place of rest/' * 
 
 Professor Eaton mentions how on one occasion the ice of 
 the Hudson, with gravel on it, in breaking up, pressed against 
 the shores, and a large cake of it, pushed by others from behind, 
 was squeezed up to a height of thirty-four feet above the 
 water, when it melted and " left the gravel which it had trans- 
 ported from the northern countries on the bank." 
 
 Dr. Hitchcock states how he had known masses of rock, 
 from one to ten feet in diameter, raised from the bottom of a 
 river by ice and removed down stream one or two miles. 3 
 
 Maclure writes, in July, 1822 : " The large masses of granite, 
 some of them weighing tons, scattered over the secondary rocks 
 between Lake Erie and the Ohio, while there is not an atom 
 of granite in place nearer than the north side of the lake, 
 would seem to point at the only mode by which they could 
 have been transported ; by supposing the lake extended thus 
 far and that the large pieces of floating ice from the north 
 side might carry those blocks attached to them and drop 
 them as the ice melted in going south; few or none being 
 found south of the Ohio, shows that the southern sun melted 
 the ice before it got so far." 4 
 
 In Hitchcock's paper on the geology of the district adjoining 
 the Connecticut river, I find him saying : " At the outlet of the 
 Connecticut through the mountains below Middletown, and 
 
 9 Id. 95. Silliman's Journal, iii. 256. 
 
 2 Id. v. 21. 3 Id. vi. 37, 38. 
 
 4 Id. vi. 102. 
 
Lariviere. 133 
 
 600 or 700 feet above the present bed of the river, I saw 
 rounded masses of old red sandstone, several inches in 
 diameter, mixed with the fragments of the rocks in place. 
 Such a fact I never noticed at any other place in the primitive 
 region along the river ; certainly not on the east side of 
 it. I was led irresistibly to the conclusion, that they were 
 conveyed thither by the ice of the ancient lake, which would 
 be floated to the ocean through this inlet." : 
 
 In a letter from Peter Dobson to Professor Silliman, written 
 on Nov. 21st, 1825, we read : " I have had occasion to dig up 
 a great number of boulders of red sandstone and of the 
 conglomerate kind, in erecting a cotton manufactory, and it 
 was not uncommon to find them worn smooth on the under 
 side, as if done by their having been dragged over rocks 
 and gravelly earth, in one steady position. On examination, 
 they exhibit scratches and furrows on the abraded part. , . 
 These boulders are found not only on the surface, but I have 
 discovered them a number of feet deep in the earth, in the 
 hard compound of clay, sand and gravel. One block, of more 
 than thirty hundredweight, washed and worn as above 
 described, was dug out of a well, at the depth of twenty-four 
 feet. 
 
 "Boulders with these marks upon them I have observed 
 not only in this town (i.e. Vernon, Con.), but in Manchester, 
 Ellington and Wilbraham. I think we cannot account for 
 these appearances, unless we call in the aid of ice along 
 with water, and that they have been worn by being suspended 
 and carried in ice, over rocks and earth, under water." 
 
 Fischer, in a paper published in 1828, entitled, " Ueber die 
 Gerollen des Nordens," attributes the portage of the erratics 
 largely to ice. He cites Scoresby's observations on icebergs, 
 and urges that in former times, when the climate was colder, 
 such bergs were floated in a great flood of water, from the 
 north or otherwise, on to the flat lands of Germany, and he 
 also mentions how ground ice moves great stones in the 
 Eastern Baltic. 7 
 
 Engelspach Lariviere, in a memoir entitled " Consi- 
 derations sur les Blocs erratiques," published in 1829, 
 
 5 Id. vii. 18. 6 Id. x. 217, 218. 
 
 7 Kastner's Archiv. xiv. 421-423. 
 
134 The Glacial Nightmare. 
 
 describes at some length the erratics found in the Low Countries. 
 He attributes them, and those found elsewhere, to more than 
 one cause, but the principal one, he says, which brought the 
 blocks from foreign countries to Germany and Holland, was 
 enormous icebergs; and he mentions how, when he was at 
 Memel in 1821, his companion, Baron Chaumont, called 
 his attention to a mass of ice twenty-eight to thirty feet 
 long, stranded on the shore, embedded in which was a 
 triangular mass of granite like that of Finland, and on 
 returning to see it again it had disappeared, having doubtless 
 moved on. 8 
 
 In the same year K. F. Kloden published a work, entitled 
 "Ueber die Gestalt und die Urgeschichte der Erde," in 
 which he endeavours to explain the contour of the present 
 surface of the world by a theory of a change in the position 
 of its axis. This, he goes on to say, would cause, if effected 
 suddenly, a great rush of water, and he thus accounts for the 
 rush from the N.E., which Bruckner had postulated as 
 necessary to account for the surface deposits of Mecklenburg, 
 Brandenburg, Pomerania and Prussia. This great current 
 would bear along with it masses of ice which would carry 
 great blocks of stone. Hence he accounts for the transport 
 of blocks from the Scandinavian mountains, the icebergs 
 dropping some of them in Sweden, Finland, Denmark, 
 Prussia and Germany, as far as the Polish mountains, 
 Lusatia, the Erzgebirge, the Thuringenwald and the Hartz, 
 while some were also carried to the Netherlands and to 
 Eastern England. 9 
 
 In the first edition of his " Principles of Geology/' pub- 
 lished in 1830, Lyell says: "Marine currents are sometimes 
 instrumental in the transportation of rock and soil, by floating 
 large masses of ice to great distances from the shore. When 
 glaciers in northern latitudes descend the valleys burdened 
 with alluvial debris, and arrive at the shore, they are 
 frequently detached and float off. Scoresby counted 500 
 icebergs in lat. 69 and 70 N. . . . Many of these 
 contained strata of earth and stones, or were loaded with 
 beds of rock of great thickness, of which the weight was 
 conjectured to be from 50,000 to 100,000 tons. A&, the 
 8 Op. cit. 29 and note. 9 Op. cit. 347-350. 
 
De la Bee he. 135 
 
 mass of ice below the level of the water is between seven 
 and eight times greater than that above, these masses may 
 sometimes take the ground in great numbers in particular 
 parts of the sea, and may, as they dissolve, deposit such 
 masses of matter on particular parts of the bottom of the 
 deep, or on the shores of some isles, as may offer perplexing 
 problems to future geologists/' l 
 
 Applying these facts to Switzerland, Lyell says : " Those 
 naturalists who have seen the glaciers of Savoy, and who have 
 beheld the prodigious magnitude of some fragments conveyed 
 by them from the higher regions of Mont Blanc to the valleys 
 below, to a distance of many leagues, will be prepared to 
 appreciate the effects which a series of earthquakes might 
 produce in this region, if the peaks or ( needles,' as they are 
 called, of Mont Blanc were shaken as rudely as many parts of 
 the Andes have been in our own times. The glaciers of 
 Chamouni would immediately be covered under a prodigious 
 load of rocky masses thrown down upon them. Let us then 
 imagine one of the deep narrow gorges in the course of the 
 Arve, between Chamouni and 01 use, to be stopped up by the 
 sliding down of the hillside (as Rossberg did in 1806), and a 
 lake would fill the valley of Chamouni, and the lower parts of 
 the glaciers would be laid under water. The streams which 
 flow out of arches, at the termination of each glacier, prove 
 that at the bottom of those icy masses there are vaulted 
 cavities through which the waters flow. Into these hollows 
 the water of the lake would enter, and might thus float up the 
 ice in detached icebergs, for the glaciers are much fissured and 
 the rents would be greatly increased during a period of earth- 
 quakes. Icebergs thus formed might, we conceive, resemble 
 those seen by Captain Scoresby far from land in the Polar seas, 
 which supported fragments of rock and soil conjectured to be 
 above 50,000 tons in weight. Let a subsequent convulsion 
 then break suddenly the barriers of the lake, and the flood 
 would instantly carry down the icebergs, together with their 
 burden, to the low country at the base of the Alps/' ! 
 
 "Ice," says De la Beche, writing in 1833, " seems to offer 
 a possible explanation of the transport of many masses, for 
 the glaciers which descend the valleys of high northern regions 
 1 Op. cit. i. 299. 2 Op. cit. iii. 1833, 149, 150. 
 
136 The Glacial Nightmare. 
 
 are, like those of the Alps, charged with blocks and smaller 
 rock fragments, which have fallen from the heights. Waters 
 rushing up or down such valleys would float off the glaciers, 
 more particularly as northern navigators have shown that they 
 project into the sea. It is considered that the huge masses of 
 ice known as icebergs, are the projecting portions of these 
 boreal glaciers, which having been detached from the parent 
 mass, are borne into more temperate climes, in some cases 
 transporting blocks and smaller fragments of rock. The 
 debris will, as Mr. Lyell has observed, be deposited at the 
 bottom of the seas over which they pass ; and therefore, if such 
 bottoms were raised so as to become dry land, blocks might be 
 discovered scattered over various levels of that land, presenting 
 appearances that might be mistaken for the action of diluvial 
 currents. If the present continents bore evident marks of 
 long submergence beneath an ocean immediately previous to 
 their present appearance, and if the blocks were merely 
 scattered here and there, this explanation would by no means 
 be without its weight ; but there are too many circumstances 
 tending to other conclusions to render it probable. The 
 supposition of masses of ice, covered by blocks and smaller 
 rock fragments, borne southwards with violence, though it may 
 account for some appearances, does not, it is confessed, seem 
 applicable to all, more particularly where blocks can be traced 
 to their sources at comparatively small distances. Supposing a 
 wave or waves discharged over Europe and America from the 
 northwards, many phenomena would depend on the time of 
 the year at which the catastrophe or catastrophes took place ; 
 for in the winter, waters rushing from that quarter would 
 transport a larger quantity of ice, and many superficial blocks 
 and gravels, bound by ice together, might be torn up and 
 carried considerable distances, from the possible small specific 
 gravity of the mass ; for even in the case of rivers, it has been 
 found that large masses of rocks have occasionally been trans- 
 ported from places, when encased in ice and acted on by the 
 stream. In Sweden and Russia it is more than probable that 
 many blocks would be thus encased during winter, and there- 
 fore a flood of waters passing over them would cause them 
 to rise, to float, and to be borne onwards until, the ice melting, 
 the blocks would sink and be finally brought to rest. t?pon 
 
Murchison. 137 
 
 the hypothesis of a convulsion or convulsions, in the North, the 
 effects would become less as we receded from the centre of 
 disturbance, and finally, all traces of them would be lost. 3 . . . 
 It has been frequently remarked that the Alpine erratic blocks 
 frequently occur in groups. ... It may be asked, as a mere 
 conjecture, whether masses of floating ice charged with blocks 
 and other detritus, rushing down the great valleys, into the 
 more open country of lower Switzerland, might not be whirled 
 about by the eddies, and that groups of rocks would afterwards 
 be found where the whirlpools had existed. Masses of ice 
 charged with blocks and pent up for the moment within such 
 basins as might be formed between the Alps and Jura, might 
 also be carried at certain levels against the sides of the oppo- 
 site mountains, such as the Jura, and be there deposited in 
 groups and in lines of level." 4 
 
 Murchison, writing in 1835, says : " Once let it be granted 
 that large frozen masses, like those now periodically liberated 
 from the polar regions, were drifted to certain distances and 
 in given directions by currents dependent on former configura- 
 tion of the land, and we are furnished with an adequate agent, 
 each ice-floe as it dissolved, might have dropped its load of 
 stones, at intervals, upon a submarine surface of gravel, sand 
 and shells." And he goes on to suggest that when England 
 and Wales were separated by a strait, the distribution of land 
 and sea may have been such as to have permitted the production 
 of icebergs, which, being dislodged from the shores of Cumber- 
 land, might have been drifted into the straits of the sea then 
 existing to the south of Bridgnorth. And in order to explain 
 the generally rounded state of the boulders in Central England, 
 he suggests that they may have been carried down by streams 
 to the shores, and have been long bowldered there "previous 
 to their insertion in the ice. Secondly, they may have been 
 fragments falling from the adjacent rocks which were exposed 
 to the action of water on the shores before their transport by 
 ice. Thirdly, that granite is so prone to desquamation, that 
 nearly all granitic chains are topped with rounded masses which, 
 though in xitu, have often the appearance of being boulders, 
 and these, if dislodged from cliffs and embedded in ice-floes, 
 
 8 De la Beche, a Geological Manual, ed. 1833, 192, 193. 
 4 Id 197. 
 
138 The Glacial Nightmare. 
 
 would at once present the appearance objected to, though they 
 had never been rolled under water. Lastly, if transportation 
 by ice be supposed, we can account rationally for the blocks 
 occupying for the most part the surface or upper portions of 
 the drift, for we know by modern analogies cited by Captain 
 Bayfield, that ice-floes, in narrow bays or straits, are generally 
 stranded on coasts or shallow shores." 
 
 Speaking of the numerous icebergs referred to by Bennett, 
 Murchison says : " The multitude of these ice-floes sailing to- 
 gether for such a great distance from the sources of their 
 origin before they are dissolved, not only teaches us that any 
 stones of the polar regions which they might be transporting, 
 must now be part of submarine deposits, even in inter-tropical 
 regions, but also explains how a vast number of such stones 
 might be collected in owetrack very remote from the parent rock, 
 as in the south of Staffordshire." 5 In a later paper he urges 
 that in some cases " the different heights at which boulders 
 occur may be explained by the fact that the bed of the ocean on 
 which the stones were dropped was itself full of inequalities, 
 which, when desiccated, would present the undulating outline of 
 our hilly districts, and in other cases it might well be that grea,t 
 dislocations and disruptions took place when the land was 
 raised above the water, when the boulders would be raised with 
 the beds on which they lay." 6 He illustrates his position by 
 the case of the Tyrolese mountains" thus. "In one of the 
 loftiest courbes of the Alpine Limestone, called Hogelsand, at 
 a height of more than 3000 feet above the valley of the Inn, 
 there is an accumulation of blocks of gneiss, granite, porphyry, 
 chlorite-slate, etc., materials which can only have been derived 
 from the Central chain of the Alps, from which they are now 
 completely cut off by a deep valley/' He explains this by 
 postulating that there was a period when the blocks in the 
 valley and those on the summits of the limestone mountains 
 were deposited under water, when the limestone peaks were 
 probably at no very different level from those now lying on 
 the banks of the Inn. After the drift was deposited, powerful 
 movements took place, heaving up the outer shores and raising 
 
 5 Silurian System, 541, note, in which he says the preceding remarks 
 are printed from a paper presented to the Geological Society in 18ti5. 
 
 6 Id. 544, 545. 
 
LyelL 139 
 
 the pre-existing bottom of the sea, estuary or lake, to great 
 heights, the blocks and loose materials being thus left at 
 different heights, and in the detached positions we now find 
 them. 7 
 
 Lyell, writing again in 1835, and speaking of the boulders 
 covered with lichens on the coast at Gefle, in Sweden, and on 
 the Isle of Lofgrund close by, points out their size and sharp 
 edges. One of the fishermen remarked that the ice might have 
 brought them, and he undertook to show him much .larger 
 blocks which had been recently stranded on different parts 
 of the skar, He accordingly went to a small island called 
 Hustgrund, where he <f saw blocks of red granite five or six feet 
 in diameter, perfectly free from lichens, amidst other blocks 
 of various sizes which were coloured grey, white and black, by 
 a coating of these plants. The sailors named other spots 
 where stones similarly bare or only beginning to be covered 
 were to be seen among thousands which, having probably 
 lain for a great many years at the same height above the sea, 
 had entirely changed their colour. They declared they knew 
 the exact date of the arrival of some of these blocks. 8 
 
 Again, Mr. Westbeck, of Mar strand, for thirty years em- 
 ployed by the Swedish diving company, had opportunities of 
 witnessing the extraordinary power of ice to lift up from the 
 bottom of the sea and remove to ' a distance very heavy 
 masses. In two instances the ice collected round sunken 
 vessels which were under his charge, and having frozen 
 round them, floated them off with their cargo and ballast from 
 shallow into deep water. 9 Again, " I saw some considerable 
 boulders overlying the deposits of recent shells at Capell- 
 backen near Uddevalla, a phenomenon analogous to that 
 described near Upsala, where these huge erratic blocks repose 
 upon the sand hills, characterized by fossil shells of Baltic 
 species. These rocky fragments continued therefore to be trans- 
 ported after the period when the modern shelly formations of 
 both coasts were accumulated, and it may be inferred that the 
 drifting of such blocks may be going on by means of ice every 
 year. The largest blocks are found on the highest part of 
 the ridges, and if these ridges were originally sand-banks in 
 the sea, as the sea-shells found in some of them incline me to 
 7 Id. 545, note. 8 Phil. Trans. 1835, 20, 21. 9 Id. 30, 31. 
 
140 The Glacial Nightmare. 
 
 believe, the summits of such banks would have arrested the 
 ice islands, which might transport fragments of rocks as 
 suggested." 1 
 
 In his address to the Geological Society in 1836, Lyell 
 mentions, inter alia, how Captain Belcher, in 1815, fell in with 
 an ice-field near St. John's Harbour in Newfoundland, in which 
 there were muddy streaks, gravel, and even stones ; and he 
 goes on to suggest that in Scandinavia the blocks having been 
 separated from their source by fissuring, one of them after another 
 might be buoyed up and floated off on the rise of the Baltic. 
 Turning elsewhere, he says : " Let us imagine a sunken reef in 
 Baffin's Bay, in about 75 north, divided into fragmentary 
 masses, and these masses becoming year by year involved in 
 packed ice. In a few months they may be drifted more than 
 1800 miles to the southwards, through the Straits of Belle 
 Isle, to lat. 48 N. After a repetition of these operations 
 for thousands of years, the uneven bed of the ocean far 
 to the south may be strewed over with drift fragments which 
 have either stranded on shoals or have dropped down 
 from melting bergs. Suppose the floor of the ocean, where 
 they alight, to be on the rise as gradually as the bottom of the 
 Baltic in our own times. ... At length a submarine ridge, 
 covered with the travelled fragments, emerges, and first 
 constitutes an island, which at length becomes connected 
 with the main land, in time perhaps the site of a university 
 of Upsala." 2 
 
 In a letter to Sir Charles Lyell, dated November, 1837, 
 Captain Bayfield says : u The bed of the St. Lawrence 
 below Lake St. Peter is full of immense boulders of 
 primary rocks, most of them (but not all) rolled or water- 
 worn or with their edges worn by atmospheric agency. . . . 
 I have seen a granite boulder fifteen feet in length 
 by ten in width and depth, transported many yards along 
 a meadow by the ice, and last spring I watched the lake ice 
 (lake St. Peter), which takes two or three days to pass 
 Quebec every spring, and had the pleasure of observing several 
 boulders of considerable size and many small stones, sand, 
 earth, reeds and plants, on their way down the river, drifting 
 at a rate measurable by the excess of every ebb tide over the 
 1 Id. 32. 2 Proc. Geol. Soc. 1836, 384-5. 
 
Darwin. 141 
 
 preceding flood. . . . Any boulders thus transported are 
 liable to be dropped at various points along the bed of the 
 river, as the ice gives way to the increasing temperature of 
 the air and water in the spring of the year." He then goes on 
 to report how large boulders, and also deep stakes which he had 
 caused to be placed in certain positions, were removed and 
 replaced by other boulders, while in the same season a mass 
 of granite containing 1500 cubic feet, which he had particu- 
 larly marked, was transported several hundred yards from its 
 observed position. From these facts he infers that the older 
 erratic blocks, from whatever source derived, have been 
 dropped from time to time (from ice-floes) on the bed of the 
 tertiary sea. 3 
 
 M. Cordier, in his instructions to the voyage of the 
 Astrolabe,* referring to the Anglo-American expedition 
 of 1830, tells us that the reports of that voyage state that 
 the shores of New Shetland are covered with blocks of 
 granite, and consequently of a rock foreign to the country, 
 and Mr. James Eights, naturalist to the expedition, does not 
 hesitate to attribute the transport of these blocks to the portage 
 of masses of ice which annually come there and deposit their 
 loads. 
 
 In the year 1832 Darwin set out on his famous travels on 
 board the Beagle, and during the next four years he accumu- 
 lated a large part of the materials for his later works. It was 
 during this voyage that he for the first time described in 
 detail the erratic phenomena of South America. I shall quote 
 largely from his notices. 
 
 He says he met with no boulders on the eastern plains of 
 South America until he arrived on the banks of the Santa 
 Cruz, in lat. 50 10' S., nor did they occur near the coast, 
 but were first noticed in ascending the river about 100 
 geographical miles from the Atlantic, and sixty-seven from 
 the nearest slope of the Cordillera. Twelve miles further west, 
 in long 70 50' W., they were extraordinarily numerous, consis- 
 ting of compact clay slate, feldspathic rock, a quartzose 
 chloritic schist and basaltic lava. They were generally of 
 angular form, and many of them resembled fragments at the 
 foot of a precipice. Many were enormous ; one, which was 
 3 Silurian System, 733. 4 L'Institut, 1837, p. 283. 
 
142 The Glacial Nightmare. 
 
 rounded, was sixty feet in circumference, and stood six feet 
 above the ground. The place where they lay was about 1400 
 feet above the sea level. They lay on the surface of, and 
 embedded in, a shingle certainly, says Darwin, of submarine 
 origin, and within the period of existing sea shells which are 
 found with them. 
 
 <c I met," he says, "with erratic boulders nowhere else in 
 Patagonia." Captain King, however, states in his " Sailing 
 Directions/' that the surface of Cape Gregory, a headland of 
 about 800 feet in height, on the northern shore of the Straits 
 of Magellan, is strewed with great fragments of primitive 
 rocks. 
 
 In the Straits of Magellan and on Tierra del Fuego, boulders 
 both rounded and angular abound. One of them, composed of 
 syenite, and shaped somewhat like a barn, was forty-seven 
 feet in circumference. There were many others half this size, 
 and they must all have travelled at least ninety miles from 
 their parent rock. The boulders there have all travelled from 
 mountains to the west, which is in accordance with the 
 present currents about Cape Horn. In some situations there 
 is no trace of stratification in the long lines of cliff sixty feet 
 high, while the clay contains boulders of all sizes hetero- 
 geneously mixed, and similar rocks do not exist in situ nearer 
 than sixty miles off. In other places they are sometimes regu- 
 larly interstratified with beds of shingle. " I entered in my 
 note-book," says Darwin, " that a vast debacle appears to have 
 been suddenly arrested in its course:" And he goes on to urge 
 against Lyell's view of the deposition of sand and blocks from 
 melting drift ice, in a tranquil sea of mud, how, near the eastern 
 mouth of the Strait of Magellan, where the finest sediment has 
 been arranged in horizontal laminae, and the coarse shingle in 
 beds, that it appears strange that stratification should be so 
 entirely and often suddenly absent in the till alone. It 
 should be remarked that boulders both rounded and angular 
 occur both in the till and in the finest laminated matter. 5 
 
 Darwin tells us that going north from the extremity of the 
 continent along the west coast he did not land s'outh of lat. 
 47, but inasmuch as a fragment of granite was seen floating 
 on an iceberg between lat. 49 and 50, we may be sure that 
 
 5 Journ. Geol. Soc. vi. 417-423. 
 
Darwin. 143 
 
 erratic blocks occupy this space. Between lat. 47 and the 
 south extremity of the island of Chiloe he landed several times; 
 but saw no boulders, which he accounts for by the coast being 
 separated from the Cordillera by intervening high land. Chiloe 
 is an island 100 miles long between lat. 41 46' and 43 26', 
 about thirty miles from the foot of the Andes. Boulders 
 occur on it in extraordinary numbers on both the eastern and 
 western beaches and up to a height of 200 feet on the land. 
 They also abound on the islets between it and the coast. The 
 boulders are of syenite and granite, the latter alone occurring 
 on the south of the island. They all probably came from the 
 Cordillera. The larger boulders were all angular, and re- 
 sembled fragments at the foot of a steep mountain, one block 
 at Chacas was a rectangular oblong, 15 feet by 11, and 9 
 feet high, another was pentagonal, quite angular, and 11 
 feet on each side, it projected 12 feet from the sand, with one 
 point 10 feel high. There were many others smaller in size. 
 The smaller ones were often more rounded. The land has 
 risen here in post-pliocene times, and there are remains of 
 numerous old straits along which the boulders were probably 
 floated. Most of the boulders were in stratified gravel, but 
 in some cases in a hardened unstratified mud like the Scotch 
 till. This till often passes into finely laminated beds with 
 shells. Darwin says that north of Chiloe in lat. 41 47' he 
 met with no fragments of far-transported rock which can be 
 classed with the erratic boulders just described, their great size, 
 frequent angularity and complete separation by wide valleys 
 or arms of the sea from their parent source, being taken as 
 their distinctive characters. 6 
 
 In regard to the origin of this phenomenon, Darwin, writing 
 with the views of Agassiz and others before him, says: "Neither 
 the till beds of Tierra del Fuego, which pass into and are 
 regularly interstratified with a great formation of horizontally 
 laminated sandstone containing marine remains, nor the 
 stratified gravel and till which form low plains on the shores 
 of Chiloe, and cross in regular beds the tertiary strata, can have 
 been produced by ordinary moraines. I am led to the same 
 conclusion with respect to the till of southern Tierra del 
 Fuego. . . . The boulders on the lower levels at the head of 
 
 6 Id. 423-426. 
 
1 44 The Glacial Nightmare. 
 
 the Santa Cruz Eiver are strewed on land which certainly has 
 been modelled by the action of the sea. Those on the 1400 
 feet plain are sixty-seven miles from the Cordillera, of which 
 the highest pinnacle is only 6400 feet and the general range 
 considerably lower ; this little inclination of the surface, with 
 the absence of mounds or ridges on it,, and the angularity of 
 the fragments are opposed to the notion that the blocks have 
 been pushed to this great distance by glaciers. Hence I 
 conclude, that in the two first mentioned districts it is quite 
 certain, and in the latter three highly probable, that the 
 boulders were transported by floating ice. The fact of many 
 of the blocks on the northern end of Chiloe being different 
 from those thirty miles southwards, where there must 
 anciently have been a channel across the island, he explains 
 by the direction of the tidal currents as controlled by the 
 seaward channels. According to the situation of the spot 
 whence the iceberg with its cargo of rock was first launched, 
 so would it be swept towards one or the other channel." The 
 circumstance of the boulders on the higher and lower plain of 
 Santa Cruz being of different kinds of rock, he explains by 
 the fact that as the land rose some of the glaciers which 
 formerly debouched on the coast would cease to do so, and that 
 rocks hitherto submerged beneath the sea, would become 
 exposed and their fragments, falling on the glaciers, would be 
 transported with the icebergs. " Masses of ice by which 
 fragments of rock are conveyed are produced in two ways, 
 by the breaking off of icebergs from glaciers descending 
 into the sea, and by the actual freezing of the surface of the 
 sea or its tributary streams. Great boulders can be enclosed in 
 ice by this latter means, only (with rare exceptions) where the 
 winter is extremely cold, as in the Gulf of Bothnia and on 
 the shores of North America. A large proportion of the 
 fragments thus enclosed will generally have been exposed 
 to the wearing influences of the sea beach, and from the ice 
 being in a sheet they will be liable to be repeatedly stranded 
 in shallow places, and thus to become still more worn. The 
 other method of transportal, namely, by the descent of 
 glaciers to the sea-level and the production of icebergs, is 
 far from necessarily requiring an extremely cold winter. 
 Blocks are not embedded in glaciers, except high up at their 
 
Darwin. 145 
 
 sources, but rest on their backs, and from being exposed to 
 much abrasion remain angular, hence only loose angular 
 blocks of rock, as was the case with those on the floating ice 
 in Sir Gr. Eyre's Sound, can be transported by icebergs detached 
 from the glaciers of temperate climates, and such icebergs 
 must be floated off perpendicularly and in large masses or 
 the loose fragments would be hurled into the sea. . . . The 
 angularity of the blocks at Chiloe and Santa Cruz accords 
 with their transportal by icebergs, but it is not improbable 
 that in Tierra del Fuego and the southern parts of the 
 continent, where the boulders frequently show signs of 
 attrition, as if they had been worn on a sea beach, the 
 other agency, namely, the freezing of the sea, may also have 
 come in ; the shape of the fragments and their position enable 
 us to discover whether they had been embedded in sheet ice or 
 carried on the surface of deeply floating icebergs." 7 
 
 In the Journal of the Geographical Society for 1839, p. 528, 
 Darwin describes an angular block of rock embedded in the 
 perpendicular face of an iceberg twenty feet above the sea 
 level, in lat. 61 S. and long. 163 E., which was seen by 
 Mr. Macnab, mate on board the Eliza Scott, when in the 
 Antarctic seas. The iceberg in question was between 250 
 and 300 feet high. The same man about a week later saw on 
 the summit of a low flat iceberg, a black mass which he 
 thought was a fragment of rock. He had also frequently seen 
 at considerable heights on the bergs, both reddish brown and 
 blackish brown ice. The first of the icebergs above mentioned 
 was 450 miles distant from Sabrinaland (if such land exists), 
 and 1400 miles from any certainly known land, although land 
 may possibly exist 100 miles due south, as the seas have not 
 been explored there. An iceberg with a considerable block 
 on it was met to the east of South Shetland by Mr. Sorrell, 
 when in a sealing vessel, and it is to be noted that Cordier 
 reports the shores of South Shetland as covered with erratic 
 boulders which were supposed to have been brought there on 
 ice. 8 
 
 " Although," says Darwin, " the glaciers I saw were quite 
 clear, many of the icebergs described by Mr. Kirke in Sir 
 
 7 Trans. Geol. Soc. vi. 429-30. 8 Journ. Geog. Soc. ix. 528-9. 
 
 L 
 
146 The Glacial Nightmare. 
 
 G. Byre's Sound were dark coloured, and on the surface of 
 one several blocks of granite and serpentine were found." 9 
 
 In the appendix to his " Journal of Discovery/' published in 
 1839, the same writer says, " As the numerous masses of ice, 
 which fall from the glaciers at the head of the sounds on the 
 South American coast, are slowly drifted outwards (owing to 
 the fresh water flowing in from the foot of the glaciers), and in 
 the more open channels are left to be acted on by the winds 
 and currents ; so must it have been with the icebergs from the 
 glaciers of the Alps, situated in the same latitude, and under 
 similar conditions. These icebergs would in most cases be 
 driven on some part of the surrounding shore ; but from 
 floating deep they would ground a little way from the beach, 
 and then being packed together, and driven to and fro, as 
 the winds changed, and as the tides rose and fell, would they 
 not, like a glacier on the land, though in a lesser degree, 
 bruise and grind down everything and polish the surface on 
 which they reposed ? In the rapids of the North American 
 rivers, over which large bodies of ice are driven, carrying with 
 them pebbles and fragments of rocks, I am informed by Dr. 
 Richardson, that the primitive rocks are scooped and hollowed, 
 and have their surfaces polished and glossy. Dr. Richardson 
 is not prepared to say, however, whether this is caused by the 
 passage of the ice or of the pebbles. In the Arctic seas, he 
 tells me, great icebergs are packed together and are driven 
 with such force against the shore, that they push up before 
 them, to the height of several feet, every pebble and boulder 
 which lies on the bottom ; and consequently the submarine 
 ledges of rock are kept absolutely bare. If a fragment were 
 to be wedged beneath one of these mountain-masses of ice, 
 when forced upwards with such overwhelming power, it is im- 
 possible to doubt that the underlying surface of solid rock 
 would be deeply scored. As it is known that the shingle on 
 most beaches has a tendency to travel in one direction, so must 
 the icebergs, and hence we may conjecture, that the grooves 
 would generally be slightly oblique to the line of coast, and 
 parallel to each other. 
 
 " Although the icebergs might be drifted from side to side of 
 the sound, if they were moved after having grounded, it would 
 
 9 Trans. Geol. Soc. vi. 428. 
 
Darwin. 147 
 
 be along the shore by the set of the currents or winds, and 
 perhaps slightly up and down by the tidal changes. Would 
 not the necessary effect of this be, that the scratches formed by 
 the sand grating between the rocks and the bottom of the ice- 
 bergs should be, with some irregularities, longitudinal or (from 
 the effect of the tidal movement) oblique ? And as the moun- 
 tains slowly emerged during ages, every part would be thus 
 acted on ; and consequently the whole surface would be marked 
 by longitudinal scratches. As the icebergs on the South Ameri- 
 can coast sometimes transport angular fragments to the dis- 
 tance of many miles from the glacier whence they were 
 detached, and as the winds and tides generally are strong 
 enough to drive them ashore .... the blocks would be gene- 
 rally landed on the shores of the channels between the Alpine 
 ranges and not dropped in the intervening spaces. Occasion- 
 ally, however, these loads would be launched into the deep, 
 and this would account for the extraordinary position of 
 immense single blocks, as referred toby Charpentier, which are 
 sometimes found planted vertically in the soil, in the valleys 
 as on the sides of a mountain, and split up through their 
 whole length from top to bottom, which would go to show, as 
 Charpentier urges, that they had fallen perpendicularly from 
 a certain height on the spots we now find them, and had been 
 rent asunder by the fall." Charpentier argues that they fell 
 down crevasses, but Darwin claims with justice that his theory 
 explains them as well. 1 " Again, if a berg were arrested on its 
 way by some pointed rock so near the surface that when thus 
 charged it would ground on it, the block would, when the ice 
 melted, be there left. . . . From what I have seen when passing 
 in boats through the channels of Tierra del Fuego, and from 
 frequent examination of the armings of lead used in sounding, 
 I feel nearly sure that absolutely bare submarine rock is not 
 very common. Where matter is depositing near a shore, the 
 finer the particles are the further they are drifted. Thus, 
 when the coast is rising, a l#yer of sand will be covered with 
 a layer of small pebbles, and these by larger pebbles, and as 
 icebergs would seldom be driven up on the beach of a sheet 
 of water if, like the channel between the Jura and the Alps, it 
 were protected from the open sea, any fragments of rock 
 
 1 Id. 620, note. 
 L 2 
 
148 The Glacial Nightmare. 
 
 transported by them would have been dropped some way out- 
 side, and therefore, when upraised with the whole country, 
 they would be found in most cases reposing on beds (where 
 the loose matter had not subsequently been removed), 
 characterized by the order of superposition just de- 
 scribed." 
 
 In replying to Sir James HalFs diluvial theory, Darwin says 
 that " the boulders embedded in the blue clay near Edinburgh 
 are, according to Mr. James Hall and Mr. Smith of Jordan- 
 hill, themselves marked with parallel lines having one direction, 
 which shows that they were held fast while drifting across the 
 country, and were not rolled over and over, like a pebble in 
 a stream .... What then would be the effect of tides and 
 gales of wind, driving packed icebergs with irresistible force, 
 through channels, and over rocky shoals; each part of the 
 surface being exposed for centuries, as the country was elevated, 
 to this action ? Would not the fragments of rock embedded in 
 the ice grate in a direct path over the surface, regardless of 
 minor inequalities, and would not the fragments themselves be 
 grooved and scored in one direction ? Can we for one moment 
 believe it possible that boulders, either in water or in the 
 thickest mud, could be driven over a rugged surface, or along 
 a perpendicular face of solid rock with such enormous velocity 
 as with their points to groove and scratch it, and nevertheless 
 not to be rolled over and over, like a stone descending a moun- 
 tain, but be marked with parallel lines of abrasion, equally 
 with the fixed, underlying mass ? . . . Travellers in arctic 
 regions tell us that the drift ice, with its irresistible power, can 
 force up the gravel and sand into mounds, 2 and drive before it 
 great boulders, and even ships and masses of ice, high and dry 
 on the beach. What then would be the effect of a few pebbles, 
 or a single fragment, between such masses of ice and a steep 
 coast-wall of rock ? Would not scratches, ' horizontal, or 
 nearly so/ be formed, indicating (to use Sir James Hall's 
 words) that ' grinders had been pressed against the rock,' as 
 if ' independently of their gravity ' ? " 3 
 
 In a paper by Wissman, on the erratic blocks of Switzer- 
 land, etc., published in 1840, he devotes several pages to a de- 
 
 ,v^> 
 2 See Geog. Journal, vii. 221, 3 Darwin's Journal, Addenda, 615-625. 
 
Lyell. 1 49 
 
 scription of the phenomena, and concludes that they can be 
 best explained by invoking floating ice. 4 
 
 In 1840 Sir Charles LyelL wrote a paper in the " Philoso- 
 phical Magazine," on the Boulder formations of Norfolk. In 
 this he describes the drift on the East Anglian coast as at- 
 taining a thickness of over 300 feet of clay, loam and sand, in 
 some places stratified, in others wholly devoid of stratification, 
 and as having interspersed in it pebbles and large boulders of 
 granite, porphyry, greenstone, lias, chalk and other transported 
 rocks. Pure and unmixed white chalk rubble, and even huge 
 fragments of solid chalk, being also associated with it in certain 
 localities. Lyell says in this paper : (t In no other parts of 
 our island, or perhaps of Europe, are there evidences of local 
 disturbance on so great a scale, and of an equally modern 
 date, for there are proofs of the movement both downward and 
 upward of strata several hundred feet thick for an extent of 
 many miles, together with most complicated bendings and 
 foldings of the beds and the intercalation of huge masses of 
 chalk, and what is no less perplexing and difficult of explana- 
 tion, the superposition of contorted upon horizontal and un- 
 disturbed strata." 5 
 
 He goes on to say that the boulder formation of Norfolk is 
 strictly analogous to that of Denmark and Sweden, Holstein, 
 Hamburg, Bremen, Osnabruck and Holland, and he urges 
 that " it has all been accumulated almost exclusively on ground 
 permanently submerged beneath the waters. . . . The entire 
 want of stratification in the till, whether of Scandinavia, 
 Scotland or Norfolk, implies some peculiarity in its mode of 
 origin ; yet in all these countries some of the till has accumu- 
 lated contemporaneously, and apparently in the same body of 
 water as much of the accompanying stratified gravel, sand and 
 clay. Moreover, the stratified drifts are often identical in 
 composition with the till, the distinction consisting merely in 
 the mode of arrangement. I have seen no kind of deposit 
 now in progress precisely similar in character to the till, except 
 one, namely, the terminal moraines of glaciers which are ac- 
 cumulated without the influence of currents of water, by which 
 the materials would be sorted and arranged according to their 
 relative weight and size. . . . There can be no doubt that 
 
 4 Neues Jahrbuch, 1840, 314-325. 5 Op. cit. 3rd ser. xvi. 346. 
 
1 50 The Glacial Nightmare. 
 
 similar accumulations must take place in those parts of every 
 sea, when drift ice, into which mud, sand, and blocks have 
 been frozen, melts in still water and allows the denser matter 
 to fall tranquilly to the bottom. The occasional intercalation 
 of a layer of stratified matter in the till, or the superposition 
 or juxtaposition of the same, may be explained by the existence 
 or non-existence of currents, during the melting of the ice, 
 whether successively in the same place or miscellaneously in 
 several places. 
 
 " It is, I believe, a common error of those who are not un- 
 willing to admit the agency of ice in reference to the larger 
 fragments of transported rock, to forget that what carries 
 heavier masses from place to place must unavoidably convey a 
 much larger volume of lighter and finer materials/' 6 
 
 On a later page he reverts to " that most perplexing 
 phenomenon, the superposition of bent and folded beds upon 
 others which appear to have undergone no dislocation." 
 Speaking of the cliffs between Bacton gap and Mundesley, he 
 says : " The curves of the various coloured beds of loose sand, 
 loam, and gravel are so complicated, that not only may we 
 sometimes find portions of them which maintain their 
 verticality to a height of ten or fifteen feet, but the replica- 
 tion is often such that a continuous seam of fine loose sand 
 between two layers of gravel or loam might be pierced three 
 times in one perpendicular boring. As it is clear that some 
 of the underlying horizontal beds, and apparently the till 
 also, of which the surface is so even, have not participated in 
 these movements even in the smallest degree, we are com- 
 pelled to suppose that some lateral force has been exerted 
 against the upper masses of drift which has not been applied 
 to the lower ones. Yielding beds having a thickness of at 
 least fifteen or twenty feet must in some cases have been 
 subjected to this sideway pressure and moved bodily ; and it 
 is impossible to conceive that any original irregularity in the 
 mode of depositions, or any shrinking or settling of r/he 
 materials, or anything in short but mechanical violence, 
 could have produced such complicated folds/' 7 
 
 In discussing the well-known outliers of chalk at Triming- 
 ham, and the drift in which they stood, he says he%was 
 
 6 Id. 348. i Id. 351-2. 
 
Lyell. 151 
 
 fully confirmed in his opinion that " both the chalk and 
 incumbent formation, for the thickness of several hundred 
 feet, must have been subject to some common movement, 
 whether sudden or gradual, by which the strata of both 
 have been tilted." 8 Not only are the beds of drift tilted, 
 but the drift is in certain places found to underlie the chalk 
 Lyell urges that the outliers were brought into their present 
 position after the deposition of the greater part of the drift 
 which has been subjected to precisely the same movements 
 and abuts in some cases in vertical beds against the wall of 
 displaced chalk, and that the derangement is a very modern 
 event. 9 Near Cromer the cliffs contain much more chalk, and 
 sometimes huge masses of it are intercalated in a remarkable 
 way. Some of the enormous masses of transported chalk not 
 only contain layers of undisturbed flints, but also sand-pipes 
 in the middle of thin or cylindrical cavities filled with sand 
 and gravel, such as are found penetrating the chalk at various 
 depths from the surface in the interior of Norfolk. These 
 pipes seem to imply that such masses of chalk were once at or 
 near the surface of emerged land and have been brought 
 bodily into their present position. The intercalated masses of 
 unregenerated chalk are sometimes horizontal and sometimes 
 vertical. 1 
 
 The boulders which are contained in the Cromer cliffs, 
 some of large size, seem to imply, says Lyell, " that while a 
 great proportion of the mass of the till may have been derived 
 from neighbouring regions, part, at least has come from a 
 great distance. ... In different parts of the interior of 
 Norfolk, boulders weighing several tons have been found in 
 blue clay or till. . . . Professor Sedgewick informs me that 
 in the unstratified brown clay or till of certain parts of 
 Cambridgeshire, large angular blocks of lower greensand and 
 chalk, with fossils of the Oxford clay and lias, occur. The till 
 alluded to attains at some points a thickness of 300 feet it 
 resembles that in the Norfolk mud cliffs and has been traced 
 over many of the adjoining counties." ; 
 
 In regard to the origin of these deposits, Lyell says : cf As I 
 believe coast ice and icebergs to have been instrumental in 
 transporting much of the Scandinavian drift, so I presume 
 8 Id. 356. 9 Id. 359-60. J Id 365-6. 2 Id. 374-5. 
 
152 The Glacial Nightmare. 
 
 that at the same period, the effects of the same agency was 
 extended to the British seas, although on a smaller scale. 
 But while some of the Norfolk erratics may be of northern 
 origin, other portions of the associated drift may have been 
 brought from neighbouring regions, and perhaps in an opposite 
 direction, just as we now observe that some granitic 
 boulders are floated on ice from the distant shores of Labrador 
 into the Gulf of St. Lawrence, while other fragments of rock, 
 together with much gravel and sand, are firmly frozen into 
 ice and carried down every winter by various rivers into the 
 same gulf." 
 
 Lyell's views would favour Dr. Mitchell's notion, that many 
 of the boulders may have come from the destruction of strata 
 which once occupied the site of the German Ocean. " We 
 ought always," he says, " to bear in mind that fragments of 
 chalk, greensand, oolite and lias embedded in the drift of 
 Norfolk and other countries may not have come from the 
 westward, where those formations now crop out, but possibly 
 from the N.N.E., like the erratic blocks, if some of these be of 
 Scandinavian origin." 3 Lyell argues that the present con- 
 torted and dislocated condition of the strata in the Norfolk 
 cliffs can be explained by ordinary subterranean movements, 
 by landslips and slides, and lastly by the stranding of icebergs 
 and large masses of packed ice, a view to which he inclines. 
 This, he urges, may explain the great foldings and contortions 
 of the upper beds while the lower ones are undisturbed. He 
 says coiled and folded beds of loam, gravel and sand are 
 frequent in different parts of Scotland, Sweden, Norway, and 
 probably everywhere in Europe where drift occurs between 
 lat. 50 and 70 N/ 7 He mentions how Dean and Simpson, 
 in lat. 71 N. and long. 156S W., found alow spit named 
 Point Barrow, composed of gravel and coarse sand, in "some 
 parts more than a quarter of a mile broad, which the pressure 
 of the ice had forced up into numerous mounds, that, viewed 
 from a distance, assumed the appearance of huge boulder 
 rocks ; 4 and he cites what he had himself seen in the Baltic 
 and the Gulf of St. Lawrence, as proving that lateral pressure, 
 exerted under favourable circumstances by drift ice on banks 
 of stratified and incoherent sand, gravel and mud, is an 
 
 3 Id. 376. 4 Journ. Geog. Soc. viii. 221. 
 
Murchison and VerneuiL 153 
 
 adequate cause for producing considerable flexure and dis- 
 location. (t The banks on which icebergs run aground occa- 
 sionally, between Baffin's Bay and Newfoundland, are many 
 hundred feet under water, and the force with which they are 
 struck will depend not so much on the velocity as the 
 momentum of the large floating islands. The same berg is 
 often carried away by a change of the wind, and then driven 
 back again upon the same bank, or in other cases it is made 
 to rise and fall by the waves of the ocean, and may thus 
 alternately strike the bottom with its whole weight, and then 
 be lifted up again until it has deranged the beds over a wide 
 area. On these beds new and undisturbed strata may be 
 afterwards thrown down. In other cases, where banks of mud 
 and sand forming the top of a shoal have been made to assume 
 various shapes by the lateral pressure of icebergs, the bed of 
 the sea may subside, and then the disturbed beds may be 
 overspread by horizontal strata, which may never afterwards 
 be deranged by similar mechanical violence/' 5 
 
 In a paper by Murchison and Verneuil on the Geology 
 of Kussia, published in the Bdin. New Phil. Journal for 1841, 
 those authors urge that " the Eussian detritus, the blocks 
 of which have all been derived from the north, is shown to 
 have been deposited under the sea or on a sea bottom, since it 
 covers beds with marine shells. ... In endeavouring to 
 account for the immense distances to which these blocks had 
 been transported, the authors express their belief that they 
 had been floated on former icebergs ; these, breaking loose 
 from ancient glaciers which they suppose have existed in 
 Lapland and the adjacent tracts, were dislodged upon an 
 elevation of the northern chain, and impelled northwards into 
 the sea of that period in which the post pliocene shells were 
 accumulated ; and they compare the conditions to those in 
 Central England, where a similar collocation of blocks and 
 shells was accounted for, by ( supposing that the northern 
 blocks were borne thither in vessels of ice, which in melting 
 dropped them upon what was then a sea bottom." 6 Later in 
 the same paper, after criticizing Agassiz's views, they go on to 
 offer as a probable explanation of the chief phenomena of the 
 
 6 Phil. Mag. 3rd ser. xvi. and 379, 380. 
 6 Op. cit. 136. 
 
154 The Glacial Nightmare. 
 
 north of Russia, that currents strongly determined in given 
 directions by the elevation of the northern continental masses 
 might dislodge and set in motion ice-floes and detritus, which, 
 grating upon the bottom of a sea, may have produced the 
 parallel strias. This is confirmed by the fact that the longer 
 axes of the lakes and stony ridges in north Russia have 
 generally the same direction, which is consistent with the 
 supposed icebergs and land detritus having been borne in that 
 direction also. 7 
 
 Durocher, in his admirable report on the Geological work 
 done during the voyage of the Recherche, argues that there 
 is evidence of two causes, distinct in date, for the erratic 
 phenomena. (1) The breaking up of the Northern or frozen 
 ocean, by which a current loaded with ice was sent over the 
 partially submerged country, abrading the rocks, producing 
 striae and boulders, and carrying them to the south-east. (2) 
 The forming of icebergs on the coast which carried off each 
 summer their load of rocks and earth, and deposited them to 
 the south-east. In the meantime the land was gradually 
 rising and brought the detritus above the surface. 
 
 He says that he had very seldom noticed stones on the 
 Spitzbergen icebergs, and never large blocks, only gravel; and 
 he urges that the greater portion of the blocks from the north 
 have been derived from low lands where glaciers cannot exist, 
 and cannot become the mothers of icebergs. He rather 
 invokes ice, acting as it does on the shores of the Baltic, 
 where it often holds masses of granite, etc., in its grip when it 
 breaks up, and transports them when it moves off in the 
 spring. He cites Dr. Beck as having frequently seen such 
 occurrences on the coasts of Denmark, and he refers especially 
 to two instances communicated by Baer to the St. Petersburg 
 Academy, where blocks had been raised up on the Gulf of 
 Finland. The ice of rivers has a similar carrying power; thus 
 at Uleaborg there was a block of stone, three or four feet in size, 
 which was carried down the Ulea Elf on an ice-raft. He 
 further cites M. Dubois, who described how, at Rabelle in 
 Kurland, the low ground by the River Neiss had, according to 
 
 ' Id. 137. 
 
 8 M. Eugene Eobert tells us that on the 18th of June, 1838, every one 
 on board the Recherche saw floating ice covered with pebbles and sand. 
 (Bull. Soc. Geol. xi. 299.) 
 
Hays. 155 
 
 the testimony of some old men, been once strewn with great 
 boulders, but they had been taken up by the winter ice and 
 deposited high above their former place on the banks. A 
 similar thing had been noticed on the low ground near 
 Pokroie. Durocher argues that at this moment the float- 
 ing ice in the Baltic is depositing on its bottom various 
 erratics, and reproducing in some measure the drift beds of 
 north Germany and Russia. 9 
 
 Darwin did not limit his geological speculations to South 
 America. In a paper published in the Edin. Phil. Journal 
 for 1842, while furnishing numerous proofs of the former ex- 
 istence of glaciers in North Wales, he argues that the till 
 between Bethesda and Bangor was accumulated in a sloping 
 sheet beneath the waters of the sea, and he explains the 
 irregularly stratified gravel and boulders, with layers of 
 coarse yellow sand, on Moel Tryfan in the same manner. As, 
 he says, " there are no marks of the passage of glaciers over 
 this mountain, we must suppose the boulders were transported 
 on floating ice ; and this accords with the remote origin of some 
 of the pebbles, and with the presence of the sea shells. . . . 
 May we not conjecture that the icebergs, grating over the 
 surface, and being lifted up and down by the tides, shattered 
 and pounded the soft slate rocks as they appear to have con- 
 torted the sedimentary beds of Eastern England ? . . . The 
 drifting to and fro and grounding of numerous icebergs 
 during long periods near successive uprising coast lines, the 
 bottom being thus often stirred up and fragments of rock 
 dropped on it, will account for the sloping plain of unstratified 
 till, occasionally associated with beds of sand and gravel, 
 which fringes to the west and north the great Carnarvonshire 
 mountains. . . . te As far/' he says, " as my very limited observa- 
 tions will serve, one of the best criterions between the effects 
 produced by the passage of glaciers and icebergs is boss 
 or dome-shaped rocks." 10 
 
 A report by Mr. Hays on the probable influence of ice- 
 bergs upon drift, read before the Association of American 
 Geologists and Naturalists in April, 1843, comprised a 
 large mass of information collected from over eighty people 
 who had seen icebergs. Inter alia, the report states that 
 9 Op. cit. 132. 10 Edin. New Phil. Journ. xxxiii. 352-363. 
 
156 The Glacial Nightmare. 
 
 blocks of rock and earth are found on the surface and in the 
 interior of the glaciers in the Arctic and Antarctic regions, 
 and instances were cited where some of these glaciers or fixed 
 icebergs in the Antarctic regions were strewn with stones trans- 
 ported from a distance, which stones had afterwards been 
 covered by new deposits of snow and ice ; where large rocks 
 were found in the perpendicular face of the glacier overlook- 
 ing the sea ; and where they had been covered with piles of 
 sand and volcanic scoriae. The motion of icebergs was shown 
 to be very slow and steady, and in the direction of the great 
 undercurrents of water tending from the poles to the equator. 
 Facts were also adduced to show the great depth of water in 
 which icebergs sometimes ground and remain grounded many 
 years, and the limit of transport of icebergs was shown to be 
 about 40 of N. lat. and 36 of S. lat. Many original facts 
 were stated as to the transportation of boulders, from a small 
 size to the diameter of many feet, and from the examination of 
 the positive and negative evidence it was inferred that icebergs 
 are rarely seen charged with foreign materials except near 
 their source. From the facts exhibited the following inferences 
 were drawn. 
 
 1. The steadiness of the movement in icebergs from great 
 undercurrents in our northern hemisphere, causes which must 
 have prevailed in ancient as well as modern seas, favours 
 the theory that icebergs with gravers of rock in their lower 
 portions, on pressing the sand and gravel with their immense 
 weight, along the surfaces of the rocks in the bottoms of the 
 ancient oceans, might have scored and grated along the rocks, 
 grinding off their salient points, and leaving their surfaces 
 smoothed and striated in the fixed southerly direction in 
 which they now occur. 
 
 2. The immense magnitude of the icebergs in our present 
 seas, and the evidence as to their present mechanical power, 
 when moved by powerful currents, warrant the conclusion that 
 they must have exerted a powerful influence in pushing and 
 crowding along the sand and gravel which formed the 
 bottoms of the ancient seas, and in this way forming accumula- 
 tions analogous to the moraines of the glaciers. 
 
 3. The length of time during which icebergs may remain 
 aground, even when swept by rapid currents which might 
 
Redfield. 1 5 7 
 
 surround them with sand and mud, or sweep away the loose 
 materials, leaving hills or banks upon spots protected by the 
 stranded bergs, favours the idea that this agency had an 
 influence in giving the present form to our drift. 
 
 4. The formation of glaciers upon the present coasts under 
 such circumstances that fragments of rock and detritus from 
 the land become attached to them, the constant advance and 
 separation of the glaciers from the land, and their floating 
 into the sea as icebergs with loads of earth and rock, lead to 
 the conclusion that icebergs breaking off from the shores of 
 ancient seas were important agents in the transportation of 
 rocks and earth from their parent beds. The existence of 
 immense fragments of rock in situations where they could not 
 have been carried by water alone, as on the sides of hills with 
 valleys intervening between them and their parent beds, but 
 where they might have been stranded by icebergs, favours 
 this conclusion. 
 
 5. The fact that icebergs are for the most part broken to 
 pieces and dissolved near their source, and that they are 
 seldom seen with blocks on them far away, leads to the 
 inference that the same cause limited the transportation of the 
 boulders and larger fragments of the drift to within the com- 
 paratively small distance from the parent rocks at which they 
 now occur. 1 
 
 Mr. Eedfield urged at the same meeting, that the natural 
 and the deflected course of the currents issuing from the 
 Arctic Ocean and carrying icebergs into the North Atlantic, 
 correspond severally in direction with each of the two systems of 
 striae found in North America. While he argued that the 
 phenomena of single icebergs in the open sea would fail to pro- 
 duce conviction of their efficiency as a cause of drift pheno- 
 mena, those who had attentively considered the narrative 
 of Back's last voyage might be satisfied that the movements 
 in mass of such vast packs of ice and icebergs as those in 
 which his ship was enclosed for many months would be an 
 efficient cause. They were probably moved slowly a great 
 distance to the southward by the force of the great current ; 
 while the agitation produced by the vast floes and the base of 
 the bergs, upon the incoherent portions of the shores and the 
 
 1 Silliman's Journal, xlv. 317-319. 
 
158 The Glacial Nightmare. 
 
 subaqueous topography, and the grooving of the faces of the 
 coherent rocks, were causes which, acting without stint of 
 time, were sufficient to produce most of the phenomena of the 
 drift formation. 2 
 
 In describing the American drift in his American 
 Travels, published in 1845, Lyell urges that it was 
 deposited during the successive submergence of a region 
 which had been previously elevated and denuded, and which 
 had already acquired its present leading geographical 
 features and superficial configuration. " At South Brooklyn," 
 he says, " I saw a fine example of stratified drift, consisting 
 of beds of clay, sand and gravel, which were contorted and 
 folded as if by violent lateral pressure, while beds below of 
 similar composition, and equally flexible, remained horizontal. 
 These appearances, which exactly agree with those seen in the 
 drift of Scotland or the North of Europe, generally accord 
 well with the theory which attributes the pressure to the 
 stranding of ice islands, which when they run aground are 
 known to push before them large mounds of shingle and sand, 
 and must often alter greatly the arrangement of strata form- 
 ing the upper part of shoals or mud banks and sand banks in 
 the sea, while the inferior portions of the same remain un- 
 moved/' 3 
 
 Referring to Mr. Mather's observations, that the boulders on 
 Long Island are methodically arranged and occur opposite the 
 places on the other side of the sound where the parent rocks 
 are respectively found in situ, Lyell remarks that this reminds 
 one of the similar arrangement of the Jura boulders, which 
 have crossed the intervening great valley of Switzerland at 
 right angles. " The sound/ 5 he says, "which separates LoDg 
 Island from the mainland, is from five to twenty-five miles 
 broad. The fragments have doubtless been transported by 
 ice, but we must suppose them to have been floated by ice 
 islands in the sea, as there are no high mountains in this part 
 of North America from which glaciers can have descended 
 after the continent had acquired nearly its present shape and 
 altitude/' 4 
 
 In regard td the origin of the polished and grooved surfaces 
 near lakes Erie and Ontario, he urges that "the land was 
 submerged and that large islands and bergs of floating- ice 
 2 Id. 326. 8 Op. cit. i. 239, 240. 4 Id. 241-2. 
 
LyelL 159 
 
 came from the north, which, as they grounded on the coast 
 and on shoals, pushed along all loose materials of sand and 
 pebbles, broke off all angular and projecting points of rock, 
 and when fragments of hard stone were frozen into their lower 
 surfaces, scooped out grooves in the subjacent solid strata. 
 The sloping beach, as well as the level bottom of the sea, and 
 even occasionally the face of a steep cliff, might all be polished 
 and grooved by this machinery; but no flood of water, how- 
 ever violent, or however great the quantity of detritus, or 
 size of the rocky fragments swept along by it, could produce 
 straight, parallel furrows such as are everywhere visible in the 
 district underconsideration. . . . After the surface of the rocks 
 had been smoothed and grated upon by the passage of innumer- 
 able icebergs, the clay, gravel and sand of the drift were 
 deposited, and occasionally fragments of rock, both large and 
 small, which had been frozen into glaciers, or taken up by 
 coast ice, were dropped here and there at random over the 
 bottom of the ocean, wherever they happened to be detached 
 from the melting ice. During this period of submergence 
 the valleys in the ancient rocks were filled up with drift, and 
 finally, during re-elevation, the ridges so conspicuous between 
 lakes Ontario and Simcoe were formed and certain valleys were 
 re-excavated." 5 
 
 Again, speaking of the valley of the St. Lawrence, Lyell 
 remarks that " wherever the drift or superficial clay and gravel 
 have been removed, the rocks are worn, smoothed and 
 furrowed, the furrows being clearly defined on the sandstone. 
 The direction of all the straight and parallel grooves is nearly 
 N.E. and S.W., differing uniformly from those traced by Hitch- 
 cock and Percival through New England, where they usually run 
 from N.N.W. to S.S.E. In both places the blocks have been 
 transported southwards along the same lines as are marked out 
 by the direction of the furrows. ... I know of no theory that 
 can account for both of them with any plausibility, except 
 through the agency of large islands of floating ice, which by 
 their buoyancy and enormous weight supply the carrying 
 power and pressure required to scratch, polish and groove the 
 solid floor of the ocean, and to convey stones of all sizes, 
 firmly fixed and frozen into the ice, to great distances. . . . At 
 the bridge above the falls, at Montmorenci, over which a large 
 6 Op. cit. ii. 98-101. 
 
] 60 The Glacial Nightmare. 
 
 quantity of ice passes every year, the gneiss is polished and 
 kept perfectly free from lichens, but neither there nor else- 
 where, where ice is now at work in this loose way, are any long, 
 straight grooves observable ; " and Lyell urges that any degree 
 of freedom of motion in the rocky fragments forced along by 
 small pieces of ice, or by a flood of water, would be quite 
 incompatible with the mechanical effects exhibited in what are 
 called glacial furrows. 6 
 
 On a later page he describes some remarkable striae 
 in Nova Scotia, occurring on a recently formed ledge of rock, 
 straight and about half an inch broad, some nearly parallel. 
 On inquiring, he learnt that the place in winter was covered 
 with packed ice in which fragments of hard stone coated 
 with quartz crystals were frozen ; and he adds, " I have no 
 doubt that the hardness of these gravers, firmly fixed in 
 masses of ice, which, although only fifteen feet thick, are often 
 of considerable horizontal extent, have furnished sufficient 
 pressure and mechanical power to groove the ledge of soft sand- 
 stone. 7 In our speculations, therefore, on the carrying power 
 of ice, we ought always to remember that besides gravel and 
 large fragments of rock, it transports with it the finest mud. 
 Dr. Harding informed me that the surface of mud banks along 
 the estuaries near Wolfville, are often furrowed with long 
 straight and parallel ruts, as if large waggons had passed over 
 them. These conform in their general direction to the 
 shore, and are produced by the projecting edges of irregular 
 masses of packed ice, borne along by the tidal current." 8 
 
 Elsewhere Lyell tells us that in Canada boulders are usually 
 found on the surface ; at some points, they have been found 
 in sinking wells, thirty feet deep in the clay and sand. 
 He also says the American drift is more stratified than 
 that associated with large boulders in Scotland. It is 
 generally barren, but in some places molluscs abound, and, as 
 on the other side of the Atlantic, they are of Arctic types. 
 Captain Bayfield, he says, had called his attention to 
 the fact that boulders accompany the shells in such a 
 manner as to imply that they had been dropped from melt- 
 ing icebergs to the bottom of a sea in which the mollusca lived 
 and died. 9 
 
 6 Id. 135-137. 7 Id- 173-175. 8 Id. 175. 9 Id. 138 and 146. 
 
Got he. 1 6 1 
 
 One more reference to the work of floating ice will suffice. 
 " Off the western part of Shippagan Island within the bay of 
 Chaleur, there is an extensive shallow flat, extending nearly 
 two miles from the land, called the ' Grand Batture.' On this 
 flat are numerous large blocks or boulders firmly imbedded- 
 These boulders are brought over from the wild and mountainous 
 shores of Caspe, directly across the bay, by the huge masses of 
 floating ice driven over by the northerly gales, which ground 
 and melt, adding the rocks they bring to those already 
 deposited. Mr. Wilson mentioned that there was formerly a 
 very large rock directly in front of his landing-place, at Miscow, 
 which was much in the way of his boats, but the severe 
 winter of 1848 and 1849 caused the ice to attain an unusual 
 strength and thickness near the shore of Miscow ; and when 
 it moved off in the spring, it carried off this large rock to 
 deposit it where he hoped it would be less troublesome." l 
 
 I do not propose to carry this part of my subject further at 
 present. Unlike the diluvial theory, that of floating icebergs 
 as an explanation of the so-called glacial phenomena still 
 maintains its hold upon some inquirers, notably upon those of 
 America and Northern Europe, who have to face the obvious 
 evidence of submergence involved in the stratified drift and the 
 presence of sea shells. 
 
 It will be convenient, however, to stop at this point while 
 we trace the history of another ice theory, which has acquired 
 a wider fame and numbers a larger army of supporters, namely 
 the theory of glaciers, 
 
 1 Porley's Report on the Fisheries of New Brunswick, 26 ; Silliman's 
 Journal, N.S, xi. 425. 
 
 [In writing the previous chapter I overlooked a passage which the fame 
 of its author justifies me in adding in a note. Speaking of Scan- 
 dinavia, Goethe describes masses of ice passing through the Sound and 
 carrying with them pieces of rock torn from its shores, and he suggests 
 that the blocks found, in North Germany, and which have been traced to 
 Scandinavia, may have been detached from rocks which once bordered 
 the Baltic and have since been denuded. He goes on to say : " If a great 
 cold covered the greater part of North Germany with ice, when the water 
 was of a depth of 1000 feet there, one can imagine that the collision of 
 glaciers would cause great destruction, and storms coming from the 
 north-west and the east would carry icebergs with blocks on them further 
 south." 2 ] 
 
 2 Gothe, Gesamm. Werke, i. 129. 
 
1 62 The Glacial Nightmare. 
 
 CHAPTER Y. 
 
 THE PHENOMENA OP THE DE1FT. 
 
 The Champions of Glaciers. 
 
 " Zuletzt wollten zwey oder drei stille Gaste sogar einen Zeitraum 
 grimmiger Kalte zu Hiilfe rufen, und aus den hochsten Gebirgsziigen 
 auf weit ins Land hingesenkten Gletschern, gleichsam Riitschwege fur 
 schwere Ursteinmassen bereitet, und diese auf glatter Balm fern und 
 ferner hinausgeschoben im Geiste sehen. Sie sollten sieh, bei eintre- 
 tender Epoche des Aufthauens, niedersenken und fiir ewig in fremden 
 Boden liegen bleiben." GOTHE, " Wilhelm Meister," 2nded. (1829), vol. 
 ii. ch. x. 
 
 Playfair, 1802-1816 Perraudin, 1815 Venetz, 1822 Esmark, 1826 
 Gothe, 1829 Charpentier, 1835, etc. Agassiz, 1837, etc. Le 
 Blanc, 1838, etc. Eenoir, 1839-1841 Hogard, 1838 Studer, 1839 
 Conrad, 1839 Buckland, 1840 Murchison, id. Whewell, id. 
 Lyell, id. Fargeaud, 1839 Durocher, 1841 Boubee, 1842 
 Max Braun, 1843. 
 
 So far as I know, the first person who appealed to glaciers 
 to explain the distribution of the old erratics was Playfair. 
 In 1802, before he had visited Switzerland, he published his 
 " Illustrations of the Huttonian Theory." On page 126 of that 
 memoir, as reprinted in the first volume of his collected 
 works, speaking of the transporting power of water, etc., 
 he says : " Hence fragments of rock, from the central chain, 
 are found to have travelled into distant valleys, even where 
 many inferior ridges intervene : hence the granite of Mont 
 Blanc is seen on the plains of Lombardy or on the sides of the 
 Jura, and the ruins of the Carpathian Mountains lie scattered 
 over the shores of the Baltic." Later on, in the^ same 
 volume, he has the following pregnant sentence. " For the 
 
Play fair. 163 
 
 moving of large masses of rock, the most powerful agents, 
 without doubt, which nature employs are glaciers, those 
 lakes or rivers of ice which are formed in the highest valleys 
 of the Alps, and other mountains of the first order. These 
 great masses are in perpetual motion, undermined by the 
 influx of heat from the earth, and impelled down the 
 declivities on which they rest, by their own enormous weight, 
 together with that of the innumerable fragments of rock with 
 which they are loaded. These fragments they gradually 
 transport to their utmost boundaries, where a formidable wall 
 ascertains the magnitude, and attests the force, of the great 
 engine by which it was erected. The immense quantity and 
 size of the rocks thus transported have been remarked with 
 astonishment by every observer, and explain sufficiently how 
 fragments of rock may be put in motion, even where there is 
 but little declivity, and where the actual surface of the 
 ground is considerably uneven. In this manner, before the 
 valleys were cut out in the form they now are, and when the 
 mountains were still more elevated, huge fragments of rock 
 may have been carried to a great distance ; and it is not 
 wonderful if these same masses, greatly diminished in size 
 and reduced to gravel or sand, have reached the shores, or 
 even the bottom of the sea." ] This was written in 1802, before 
 Play fair had visited Switzerland. In the notes of his journey 
 thither in 1816, we find some interesting sentences. Thus we 
 read that, " On entering the Val de Travers in Mont Jura, I 
 met with a phenomenon, curious in itself, which had often 
 engaged my attention, namely, the existence of loose blocks 
 of granite, gneiss and mica slate, on the surface of a chain of 
 mountains entirely calcareous. They first appeared within 
 the French frontier, on the western declivity of the Jura, and 
 numbers are scattered all the way through to Neuchatel. 
 The largest and most striking of thBm, the Pierre a-Bot (so 
 named from the farm in which it is situated), lies concealed in 
 a wood upon the rapid slope of a hill, at an elevation of at 
 least 700 feet above that town, and measures 64 feet in length, 
 32 in breadth, and 16 in height. It contains therefore 32,768 
 cubic feet, which (allowing 13 cubic feet of granite to a ton) 
 gives a weight of 2520 tons. When we consider that the 
 
 1 Huttonian Theory, Playfair's Works, i. 385. 
 M 2 
 
164 The Glacial Nightmare. 
 
 nearest point where granite is to be found in its native 
 place, is at a distance of seventy miles, it will appear no easy 
 matter to assign a conveyance by which this block could have 
 performed such a journey over intervening hills and valleys 
 without considerable injury. A current of water, however 
 powerful, could never have carried it up an acclivity, but would 
 have deposited it in the first valley it came to, and would, 
 in a much less distance, have rounded its angles, and given 
 to it the shape so characteristic of stones subjected to the 
 action of water. A glacier which fills up valleys in its course, 
 and which conveys rocks on its surface free from attrition, is 
 the only agent, we now see, capable of transporting them to 
 such a distance, without destroying that sharpness of the 
 angles so characteristic of these masses." 2 
 
 This conclusion, which Playfair was the first scientific man 
 to publish, had probably been reached by the unsophisticated 
 natives of Switzerland before. Thus, Charpentier tells us 
 that in 1815 he passed the night in the hut of a famous 
 chamois hunter, named J. R. Perraudin, who lived in the 
 hamlet of Lourtier, in the valley of Bagnes. The conversa- 
 tion having turned upon glaciers, he told Charpentier that in 
 the mountains where they were, the glaciers had formerly been 
 much more extensive than at present. " The whole valley," 
 he said, " as far as a great height above the Drance (i.e. the 
 torrent running in the valley), was once occupied by a vast 
 glacier which extended as far as Martigny, as is proved by the 
 blocks of stone which are found near that town, and which are 
 too big for water to have carried them.-" 3 
 
 M. Charpentier reports another conversation he had had 
 with a woodman from Meiringen, when travelling along the 
 valleys of Hasli and Lungern in 1834, when, having passed a 
 great block of granite, his companion said to him, " There are 
 many of these stones here, but they have come a long distance. 
 They all came from the Grimsel, for similar stone is not found 
 nearer, 1 ' and he went on to explain that the glacier of the 
 Grimsel had brought them and deposited them on both sides of 
 the valley, for, he said, that glacier once extended as far as 
 Berne. Water could not have carried them without filling up 
 
 7 Playfair's Works, i. p. xxix ; Forbes's Travels, 42, 43. 
 3 Charpentier, Essai sur les Glaciers, 241, 242. 
 
Venetz. 165 
 
 the lakes (i.e. the lakes of Brienz and Thun). M. Brard 4 also 
 mentions a person at Chamounix who attributed the trans- 
 port of the blocks of protogine which occur on certain heights 
 in that valley, and the parallel scratches there, to the agency 
 of glaciers. 
 
 The induction of the peasants was assisted no doubt by the 
 fact, now perfectly ascertained, that glaciers have periods of 
 extension and periods of shrinking, in consequence of which 
 serious damage has often occurred. The Glacier de Getroz, 
 in the upper Ehone valley, having outstepped its limits in 
 April, 1818, precipitated a large portion of itself down the 
 valley and formed a barrier which dammed back a new lake. 
 In fear that this dam would break down suddenly and 
 that the valley would be flooded, an engineer named Venetz 
 was appointed to devise some means of escape. This led him 
 to study glaciers more closely, and in the course of the year 
 1822, he read a paper before the Helvetic Society of Natural 
 History, in which he showed that when we first begin to have 
 historic records about Switzerland, the glaciers were not so 
 extensive as now, and that they have gradually grown to their 
 present size during the last eight centuries. This he showed 
 from the evidence of old roads and Alpine passes, bridges, 
 chapels, pastures and forests which once existed in places now 
 covered with ice, also from the records of old vineyards, 
 cherry orchards and walnut plantations. This evidence 
 was supplemented by that of tradition and by the names of 
 places, 
 
 On the other hand, he went on to argue that long before the 
 historic time the glaciers, instead of having been smaller, had 
 been much larger than now, and for evidence he turned to a 
 large number of cases of old moraines existing much below 
 the present glaciers in the valleys of Piedmont and Valais, 
 which pointed to the temperature having once been much 
 lower. 5 Forbes tells us that although M. Venetz did not 
 publish any more memoirs on the subject, he was the first 
 person publicly to maintain in Switzerland the former exten- 
 sion of glaciers to the Jura, as the transporting agents of 
 
 4 Dictionnaire des Sciences Naturelles, torn. xix. p. 16. 
 6 Bibliotheque Universelle des Sciences, etc. Section of Sciences and 
 Arts, vol. 21, Geneva, p. 77. 
 
1 66 The Glacial Nightmare. 
 
 the erratics ; and he adds that he was introduced to him in 
 1832, as the originator of a speculation which, although it had 
 not then probably another advocate, was acknowledged to be 
 novel, ingenious and bold. 6 
 
 In the Edinburgh New Philosophical Journal for October, 
 1826, is the translation of a paper in the Christiania Journal 
 by Professor Esmark, in which he discusses the erratics of 
 Norway and attributes their portage to ice. In this paper 
 we read, inter alia, " There are not wanting many facts to 
 confirm the conclusion that ice has worked prodigious 
 changes on mountain masses, and conveyed from them large 
 rocks into regions where now no perennial ice is to be found, 
 at a great distance from the mountains from which they must 
 have come." Going on to speak of the erratics on the Jura 
 and elsewhere, our author says, " A few suppose that these 
 masses have been brought to their present situation by ice. 
 According to Von Buch, this is a very general opinion in 
 Switzerland. It is not only in mountainous regions we find 
 this phenomenon, but also in flat alluvial districts, where 
 these rocky masses lie upon gravel and sand, a circumstance 
 which cannot be explained in any other way than by their 
 having been brought thither in combination with masses of 
 ice. It cannot be admitted that they could be brought to 
 such situations by torrents of water, for the same torrents 
 which could have been capable of bringing such masses of 
 rock, must at the same time have carried off the gravel and 
 sand on which they rest. One needs only travel through the 
 plains of Denmark to perceive how improbable the supposi- 
 tion is, that such masses could be placed where they are 
 by water. In short, in every country, whether it be moun- 
 tainous or flat, we shall find similar traces of the operation of 
 masses of ice. The prominent conglomerations to be found in 
 many districts, may be easily accounted for in the same 
 manner. But it is particularly in Norway I have found 
 many proofs of the operation of immense masses of ice which 
 have now disappeared/' 7 
 
 1. "As in other countries, we find large loose rocky masses 
 lying spread over pretty level plains ; for example, in travel- 
 ling from Marstuen on the Miosen to Lenton in Hedemark. 
 
 6 Forbes's Travels, 44. 7 Op. cit. 114, 115. 
 
Esmark. 167 
 
 These must have been brought from a great distance, for 
 there are in the neighbourhood no mountains of the same 
 character as these masses. 
 
 2. ft In no other satisfactory way than by the operation of 
 ice can we explain how those prodigiously large loose stones, 
 sometimes with sharp corners, have been brought up to the 
 ridges and tops of mountains, which are found in such 
 numbers in the province of Christiansand. . . . By the assistance 
 of immense masses of ice it is easy to conceive how they could 
 have been brought from a great distance, and pushed high 
 up on the mountains. 
 
 3. In travelling over our mountainous districts, especially 
 in Osterdal, it will be frequently found that the slope of the 
 mountain towards the valley is covered with very large loose 
 stones, mixed with a great quantity of loose sand and gravel, 
 and that this covering extends to a considerable height over 
 the bottom of the valley." Our author urges that this " could 
 not have been deposited by a current of water, which would 
 have first deposited the big rocks and have carried the 
 gravel and finer sand to places where the influence of the 
 current was less powerful/' Nor could it have been the effect 
 of two currents, for if a mighty current had first brought 
 down the big stones, they would have rested on each other 
 without anything interposing, and the gravel, etc., brought by 
 the subsequent current would merely have filled up the 
 cavities, etc., whereas the big stones are separated from one 
 another, and surrounded by sand and gravel, " which cannot 
 be explained in another way than by supposing that the 
 whole has formerly been filled up with ice which has pushed 
 the whole mixed mass up the slope of the mountain. The 
 water of the ice, afterwards thawing, carried off by its rapid 
 streams a part of the stones and gravel, which were then 
 heaped together, deeper down in the valleys : these heaps 
 resemble entirely those which glaciers carry before them." 
 
 4. Our author refers to the evidence presented by a large 
 level sandy plain near Fossan in the Stavanger Fiord. " Its 
 upper end," he says, " is bounded by a glacier-dike or rampart, 
 extending across the whole valley. This dike is remarkable, 
 and," he continues, " so far as I know, the only one of the kind 
 lying close to the level of the sea, in a district where you find 
 
1 68 The Glacial Nightmare. 
 
 only a few heaps of perpetual snow in hollows of the 
 mountains. The length of the dike is 2250 feet, and its per- 
 pendicular height 100 ; a break about 200 feet long, where 
 its highest point is only 12 feet above the plain, exists at the 
 end nearest the mountain. It consists of coarse gravel and 
 sand mixed with a great number of immense blocks of gneiss, 
 which is the prevailing kind of rock in the mountain. This 
 gravel and sand is not only heaped up across the valley, but 
 pushed up in great quantity on the opposite side of the dike 
 to the length of 1400 feet towards the mountain. The 
 bottom of the valley is filled with a lake, which is diked in fact 
 by the rampart. . . . From this description," says M. Esmark, 
 " it will be easy to see that this dike could have been formed 
 only by masses of ice, which must have filled up the whole 
 valley, and by their spreading and pressure have hollowed 
 out its bottom. . . . Not only the dike itself, but the whole 
 horizontal surface, exhibits proofs that there has been a 
 glacier here, for the plain exactly resembles those which I 
 found adjoining to the glaciers at present existing between 
 Londfiord and Lomb in Gudbrandsdal." Esmark says 
 so striking is the resemblance that a young mineralogist, 
 named Tank, who accompanied him to both places, on seeing 
 the latter, immediately exclaimed, without a hint from himself, 
 that the dike they had seen at Stavanger must be a glacier 
 dike. 8 
 
 Esmark goes on to say that these facts prove the Nor- 
 wegian mountains must have been covered with ice down to the 
 level of the sea, and he adds that inasmuch as the containing 
 mountain precipices are often 3000 to 5000 feet high and the 
 valleys also several thousand feet wide, it is a matter of 
 astonishment to think of such valleys being filled with ice to 
 the extent of several miles. This ice in lower districts must 
 have stretched a long way out into the sea, and, on its 
 thawing, large masses must have broken loose, and gone 
 out to sea, as we find takes place now in the polar 
 regions. 
 
 Our author goes on to say that the part of Norway 
 separated from the islands of Gule a-nd Eastern Lule, consists 
 of a solid conglomerate, composed of boulders from the size 
 
 8 Id. 115-118. 
 
Go the. 169 
 
 of a pea to that of a man's head, consisting chiefly of gneiss, 
 quartz, and clay slate, which are involved and bound together 
 in a mass so solid that it is difficult to find out what the 
 binding material is in the interstices between the large 
 stones. This rock, he says, presents proofs of the powerful 
 operation of ice. The precipices on the side of the mountain 
 next the sound are several feet high and perfectly perpen- 
 dicular, and though composed of boulders cemented together 
 are perfectly smooth. If the precipices had been caused by 
 rents attended by successive masses falling down, each 
 boulder would have adhered to one mass and left a cup 
 or cavity in the corresponding mass. This is not the case, 
 however, but the face of the rock is worn perfectly smooth 
 and has the appearance as if these boulders had been cut 
 across by a knife, and he urges that he can only explain the 
 fact by the supposition that large masses of ice have pressed 
 through the Sound and worn down the boulders and their 
 matrix to a common level. 9 
 
 Our next witness is Gothe. I will translate the heading of the 
 present chapter, which comes from his famous novel. " Finally 
 two or three quiet guests called to their aid a period of 
 intense cold, and saw in their mind's eye glaciers descending 
 from the highest mountain like gliding roads, far into the low 
 country, upon which, as on an inclined plane, heavy primary 
 blocks were slid forward and farther onwards. So that, at 
 the period of thawing of the ice, they sank down and 
 remained permanently on the foreign soil/' 1 
 
 In vol. ii. of his collected Works, page 129, after rejecting the 
 theory of a violent transportation of the erratic blocks in Savoy 
 and Switzerland, he says : ' ' We state that there was a period of 
 great cold when the sea covered the continent to a depth of 1000 
 feet, and the Lake of Geneva was in communication with the 
 North Sea. The glaciers of Savoy descended much lower than 
 now, and the moraines and long lines of boulders extended as 
 far as the valleys of the Arve and the Drance, where the rocks 
 were transported with their angles intact/' 
 
 In 1835, M. Jean de Charpentier published a memoir, 
 entitled " Notice sur la cause probable du transport des blocs 
 
 Ed. Phil. Mag. ii. 107-120. 
 
 1 Wilhelm Meister, 2nd ed. (1829), vol. ii. ch. x. 
 
1 70 The Glacial Nightmare. 
 
 erratiques de la Suisse." In this memoir he argued strongly 
 in favour of the conclusions of M. Yenetz. He urged that 
 since the erratic blocks are found mixed together of all sizes, 
 from that of a grain of sand to that of several thousand cubic 
 feet, and this on the Jura as well as the Alps, the transporting 
 cause could not have been water, which would have sorted 
 them according to their weight. Water, again, would stratify 
 its deposits, while the great masses of erratic debris are 
 unstratified, and it is only very sporadically and exceptionally 
 that slight beds of sand, gravel, etc., are found thus arranged. 
 The fact, again, of the angles of the boulders in many cases 
 notbeing rounded, and the absence of traces of rolling, precludes 
 an appeal to running water, which would have had the effect of 
 so rounding them. Then, again, the deposits in question are 
 found ranged in elongated lines like ramparts or in pointed 
 hillocks, sometimes isolated, sometimes in rows, and never 
 in continuous layers or mantles. These alignments are 
 planted horizontally to the feet of the mountains and along 
 their flanks, generally one behind the other in parallel 
 rows, and in the direction of the valleys. Occasionally they 
 approach one another and unite, terminating in one or several 
 ramparts. The highest point where they have been traced in 
 the valley of the Rhone is about 1100 or 1200 feet above the 
 river in the environs of Bex, and about 2400 feet in those of 
 Sion. The ground on which they are placed is always the 
 solid rock in situ. Water, urges M. Charpentier, could not 
 have arranged its deposits thus. Nor, again, could water have 
 carried the stones across the lakes which so often are found at 
 the feet of the valleys ; nor would it have left them in many 
 cases planted vertically on the flanks of the mountains or on 
 the plains, and occasionally broken or split from top to bottom 
 as if they had been dropped upon their present sites. 
 
 Again, the blocks coming from the lateral valleys seldom or 
 never mix with those of the main valley, or with those coming 
 from the valley opposite. Thus the feldspathic and talcous 
 rocks of the valley of Hereus, which form considerable depots 
 near Sion, are not found mixed with the calcareous stones 
 coming from the valleys of the Sionne and the Lierne. The 
 ramparts or lines of blocks coming from these valleys, which 
 join the main valley of the Rhone nearly opposite to one 
 
J. de Charpentier. 171 
 
 another, are perfectly separate and distinct. Water flowing 
 rapidly would have mixed these stones, and especially would 
 it have done so when the current abutted with its load upon 
 the Jura. 
 
 Again, the valleys in which the blocks occur have their un- 
 weathered surfaces polished. If water carrying a load of stones 
 rushed along those rocks, it would scratch and plough them up 
 like some gigantic emery paper, but could not polish them ; and 
 this is confirmed by the fact that the polished surfaces do not 
 occur above the line where the highest ramparts of erratics 
 are found. Above that line the rocks are scored and 
 weathered and rough, but not polished. The same conclusion 
 follows from the fact that the vaults of the so-called barmes or 
 balmes of the mountaineers are similarly polished, and also 
 the surfaces under rocks which jut out, and which would 
 preserve them from a torrent bearing stones. The fact also 
 that the higher we mount in the Alps, the more pronounced 
 does this polishing become, while it becomes less and less as 
 we get into the lower valleys, is precisely opposite to the 
 effect which running water with a cargo of stones would 
 produce. 
 
 M. Charpentier goes on to confess that at first his pre- 
 judices were strongly aroused against the view of M. 
 Venetz, which in lieu of water summoned glaciers on a great 
 scale to explain these ancient moraines, glaciers which once 
 filled all the Alpine valleys, and spread out like huge fans at 
 their feet, which spread as far as the Jura, and even climbed 
 that range and went over to the other side of it ; but he was 
 obliged to concede that glaciers and glaciers alone could 
 explain the facts. The deposits left by existing glaciers are 
 ranged tumultuously and without stratification, great blocks 
 and minute particles all mixed together and all carried to 
 similar distances. The moraines of existing glaciers fre- 
 quently form dykes at the bottoms of small lateral valleys, which 
 intercept the drainage and cause small lakes to be formed, into 
 which sand, gravel, etc., are washed and arranged in beds, 
 thus accounting for occasional patches of stratified materials 
 found in the ancient moraines. Glaciers still bear along on 
 their backs great blocks whose angles are intact. The form 
 of moraines which glaciers are still making is precisely that 
 
j 72 The Glacial Nightmare. 
 
 of the ancient moraines ; they are either like ramparts or 
 conical hills, sometimes isolated, sometimes in lines, and when 
 several occur they are always parallel to one another. So that 
 both the interior and the exterior configuration of the 
 moraines is precisely that of the deposits of erratic blocks. 
 Glaciers, again, never deposit moraines in the shape of 
 mantles and continuous beds. In advancing they plough 
 through the soft materials down to the bare rock, and push 
 before them the debris which they move ; hence why the'lakes 
 at the feet of glaciers are not choked by debris, which, if 
 water had carried it, would have filled them. Hence M. 
 Charpentier accounts for the extremely few erratics found 
 in the plains at the foot of the Alps, where the ancient 
 glaciers formerly rested, and for their abundance in places 
 which once formed the edges and borders of the glaciers. 
 They abound on the steep flanks of the Rhone valley, and 
 are much less frequent in the plain of Yalais and those at 
 the feet of the Alps. 
 
 Water would have deposited them in a different fashion, 
 and spread them out on the valleys. The stones found split 
 vertically and still standing upright, he explains as having 
 fallen down the crevasses. 
 
 As M. Saussure showed, the moraines of converging glaciers 
 do not mix when both reach the main valley, but continue to 
 march separately, which explains the fact already named, that 
 among the deposits of erratics the same separation into different 
 kinds of stone is found. Occasionally a mass of angular frag- 
 ments, which once apparently formed parts of one great block, 
 are found together. These M. Charpentier explains as resulting 
 from the fall of some stone avalanche on the hard back of the 
 glacier where it was shattered. Glaciers, again, rub, wear 
 and polish the rocks with which they are in contact, and 
 struggling to dilate, they follow all their sinuosities, and 
 press and mould themselves into all the hollows and 
 excavations they can reach, polishing the surfaces, even those 
 which are overhung, which a current of water could not accom- 
 plish. This, again, accounts for the evidence of rubbing and 
 polishing being more frequent in the upper valleys, where 
 the glaciers have been longer at work than lower^down. 
 Finally, it is possible to follow lines of erratics from the Jura 
 
Agassis. 173 
 
 right to the feet of modern glaciers, where similar blocks are 
 being laid down in the modern moraines. M. Charpentier 
 urges that it is not necessary, in explaining the presence of 
 erratics on the Jura by means of glaciers, to suppose that the 
 Whole intervening valley was filled up with ice to the height 
 of the Jura. He affirms that when a glacier deploys into a 
 valley at right angles, or nearly so, it is sometimes known 
 to cross it, and, without becoming thicker, to climb 
 the opposite mountain to a certain height, which depends 
 on the volume of the glacier and the steepness of the 
 slope. 2 
 
 I have quoted at considerable length from this excellent 
 memoir, because I look upon it as having definitely applied 
 inductive methods to this question, with results which are for 
 the most part sound and unanswerable, and as having formed 
 a notable epoch in the discussion. It finally established 
 for the Alpine country and for Switzerland, the fact that 
 glaciers were formerly much more extensive than now, and 
 that the erratic blocks and debris are largely the wreckage 
 of former moraines. 
 
 The next name which occurs in the history of the theory is 
 that of Agassiz. In 1837, Agassiz opened the session of the 
 Helvetic Society of Natural Sciences with a memoir on glaciers, 
 moraines and erratic blocks. In this he accepted the results of 
 MM. Venetz and Charpentier, and supported them by a number 
 of fresh facts. Thus he mentions how, at various distances from 
 existing glaciers, there are found at different altitudes moraines 
 perfectly like those now being formed. These are also concentric 
 and follow the inequalities on the flanks of the valleys. They 
 are also found at various stages, the highest being at an 
 elevation of several hundred feet above the bottoms of the 
 higher Alpine valleys, where glaciers no longer exist. But 
 on descending to the lower valleys they occur at 12 or 15 or 
 even 18 hundred feet above the bottoms of the valleys. Near 
 St. Maurice, in the Valais, in the Rhone valley, they occur at 
 2000 feet above the level of the Rhone, and may be followed 
 on the two flanks of the lake of Geneva. They also occur very 
 high above Yevey and in the neighbourhood of Lausanne, 
 where they correspond with those on the south side of the 
 2 Annales des Mines, 3rd ser. vol. viii. 219-232. 
 
1 74 The Glacial Nightmare. 
 
 lake. These old moraines differ entirely from the talus formed 
 
 by an avalanche, which is always angular, and from the 
 
 mounds of debris left by rivers, in being unbroken and of 
 
 one continuous form, in having a triangular section and 
 
 occurring in parallel rows along the flanks of the valleys 
 
 instead of being spread out in continuous sheets. They are 
 
 also formed of rounded stones. The lateral valleys present 
 
 the same evidence as the main ones. He also urges the fact 
 
 of the existence of polished rocks, wherever the surfaces have 
 
 been protected from weathering, at a long distance from the 
 
 present glaciers ; as, for instance, on the borders of the valley 
 
 of the Khone, and on the banks of the lake of Geneva. 8 He says 
 
 the southern flanks of the Jura, facing the Alps, from the lake of 
 
 Bienne to beyond the Orbe, present polished surfaces called 
 
 laves by the natives, which are quite independent of the 
 
 stratification of the beds and of the direction of the mountains. 
 
 They traverse all kinds of rocks, and are found in the valley 
 
 as they are on the highest summits, and are polished almost 
 
 like a mirror wherever the protecting mantle of gravel or clay 
 
 has been recently removed. Sometimes these surfaces are 
 
 flat, sometimes undulating, and often traversed by fissures more 
 
 or less deep and sinuous or marked by longitudinal bosses, 
 
 much rounded, but which never occur in the same direction as 
 
 the slope of the mountain but obliquely to it. The polished 
 
 surface runs uniformly, irrespective of the hardness or softness 
 
 of the rocks, and when they contain shells these are smoothed 
 
 down like those in polished marble. Often the surfaces are 
 
 finely scratched like a glass scratched by a diamond point. These 
 
 scratches, again, are arranged obliquely. In many places there 
 
 are deep excavations and pits. All these effects are consistent 
 
 only with glacier action and quite different from the effects of 
 
 water or of strata sliding over one another. 
 
 Agassiz cites several localities where these effects can be well 
 studied, and points out that in some places there are deep pits 
 and excavations which could only be produced by cascades 
 falling down crevasses. He remarks that the polished 
 surfaces are not found in the fissures which run longitudinally 
 in the Jura, nor on scarped rocks facing the Jura, but occur only 
 on the surfaces facing the Alps, and he shows how the laves 
 3 Actes de la Societe Helvetique, 1837. 
 
Agassiz. 175 
 
 differ from polished surfaces caused by faults in the rocks or 
 by the sliding of strata on one another. The former are very 
 local, while the latter show marks of erosion running along the 
 slopes and not independently of them like the laves. He also 
 distinguishes the polished surfaces caused by running water 
 or by floating bergs, and concludes that those in question 
 are consistent only with glacier action. He goes on to show 
 that as the ancient moraines occur on the banks of the lake of 
 Geneva, on both sides of it, at the same height, there must 
 have been a time when that lake was frozen to its bottom 
 while the ice rose considerably above its present level, and he 
 concludes that these old moraines were left behind by the 
 glaciers when they began to "retire and to shrink to their 
 present dimensions. 4 
 
 In answer to the champions of water, Agassiz urges that 
 since the blocks are found on all sides of the Alps and in all 
 the Alpine valleys, if the Alpine chain was in existence when 
 the currents were in motion, they must have flowed from their 
 summits in different directions with sufficient depth and 
 speed to have hurried the blocks along, so that they should 
 not fall into the valleys but be transported right across to 
 the heights where they are now found on the Jura. But how 
 can we invoke currents several leagues in length, able to 
 sustain such blocks at a height of 1000 feet ? The fact that 
 the blocks in the different valleys differ from each other and 
 are found spread out below like a fan, also excludes the action 
 of violent torrents on this scale or a great diluvial wave pass- 
 ing over Switzerland. If the torrents carried the blocks 
 before the elevation of the Alps, how is it that when the moun- 
 tains rose they did not destroy the continuity of the old 
 moraines, which still exist at the same level on both sides of the 
 valleys and follow their sinuosities ? Water, again, has a 
 different force according to the nature of its bed; ice alone has 
 the same energy at all points. 5 
 
 Agassiz then turns to LyelFs theory of floating icebergs, 
 and shows that icebergs act generally at high water 
 mark, and, if grounded, the marks they make trend in 
 various directions, owing to the rocking or rotating move- 
 
 4 Actes de la Societe Helvetique, 1837, vii.-xviii. 
 6 Id. xviii. and xix. 
 
i 76 The Glacial Nightmare. 
 
 ment of the iceberg. Again, he urges, that before the drift 
 phenomena and the polished and engraved surfaces can be 
 attributed to floating ice and icebergs, it must be shown that 
 these appearances have been produced by some agency moving 
 from the seaboard towards the land, and extending up to the 
 very summits of the mountains, or else that all the countries 
 exhibiting glacial phenomena have been sunk below the ocean 
 to the greatest height at which glacier marks are found, and 
 have since gradually emerged to their present level . . . All 
 the evidence of the kind we have, at the period under con- 
 sideration, indicates only a comparatively slight change of 
 relative level between sea and land within a narrow belt along 
 the shores. . . As to the supposition that the motion pro- 
 ceeded from the sea towards the land, all the facts are against 
 it, since the whole trend of these phenomena is from inland 
 centres toward the shore, instead of being from the shore 
 upward. Certainly no one familiar with the facts could 
 suppose that floating ice or icebergs had abraded, polished, 
 and furrowed the bottoms of narrow valleys as we find them 
 worn, polished and grooved by glaciers. 6 
 
 It was natural that these conclusions should, in the first in- 
 stance, have been derived from the Alps and their attendant 
 satellite, the Jura, where glaciers still remain close by and 
 where it is only a small step to postulate their having been 
 once much larger. It was not long, however, before the sam e 
 conclusions were drawn from similar evidence elsewhere, 
 where glaciers no longer exist, and notably was this the case 
 in the Vosges. 
 
 In 1838, M. Le Blanc, in discussing Agassiz's views, 
 noticed how in the valleys of Giromagny, Saint Amarin, and 
 St. Maurice, in the Vosges, there are masses of debris, which 
 he identified with old moraines, the work of glaciers which once 
 descended from Giromagny and Drumond, and he added that 
 the people of Wisserling, who are Swiss by origin, had 
 eighty years before given the name of moraine to one of these 
 masses of debris situated to the east of them. 7 A year later M. 
 Renoir noticed that the unstratified heap of sand and rolled 
 pebbles on which Wisserling, in the valley of St. Amarin, is 
 
 6 Geological Sketches, ser. 2, 37-40. 
 
 7 Bull. Soc. Geol. de France, ix. 410. 
 
Renoir and Hogard. 177 
 
 built, was in all probability a terminal moraine. Other similar 
 deposits were noticed at the entry of the vale of Urbey (Urbes) 
 as well as above Oderen, below Gruth, and near the castle 
 of Wildenstein. M. Renoir also noticed polished surfaces on 
 the right bank of the Thur, and striated ones on the southern 
 slope of Giromagny, while ancient moraines are well pre- 
 served at the foot of this mountain, such as the three 
 terminal moraines in front of the village of Puy. The upper 
 part of the village of Giromagny is built on a similar mass of 
 debris. Similar phenomena occur in the valley of the Moselle. 
 
 In regard to these morainic deposits of the Vosges, M. 
 Eenoir remarks that their absolute height is greater in the 
 middle than towards the rocks encompassing the valley, a 
 form peculiar to true moraines, and which is the very opposite 
 of that which deposits formed by rapid currents would assume. 
 In the valley where the village of Giromagny lies, the deposits 
 lie in parallel mounds like lateral moraines which follow the 
 sinuosities of the valley, as we see in modern moraines, and 
 are not accumulated on transverse heights as they would have 
 been if laid down by water. 
 
 In the Vosges as in the Alps the detritus is spread out in a 
 fanlike shape from the mountain summits, which is contrary to 
 the action of water and like that of glaciers. 8 M. Hogard, in 
 the same year, published fresh observations confirming those 
 just cited, and describing how the debris opens out fan-like 
 from the mountain summits, and when it reaches the wide 
 valleys is either spread out like a sheet, or in raised rounded 
 dikes, or in long strings of debris. He argued that while the 
 lateral moraines have been largely preserved intact, the 
 terminal ones have often been broken up and re-arranged by 
 subsequent torrents, to which the more or less stratified beds 
 at the entrance of the valleys are to be also attributed. He 
 also argued that the erratic blocks so often found isolated are 
 the remains of old lateral moraines from which the gravel 
 and sand have been washed away. He especially called 
 attention to the phenomena about Kochesson, Gerardiner, 
 Retournemer, etc. 9 
 
 8 Ed. Phil. Journ. xxix. 280-295. 
 
 9 Ann. de la Soc. d'mulation du Dep. des Vosges, iv. 91. 
 
 N 
 
178 The Glacial Nightmare. 
 
 In another paper he refers to such lakes as those of 
 Longerner, de Retournemer, de Foudrome and others now 
 dried up, which, he says,, have always one bank much steeper 
 than the other. He urges that the debris found on the sides 
 of these lakes lacks stratification, and forms barriers to the 
 water in the way moraines do. If these dikes had been 
 deposited by water, the lake basins would have been filled 
 up. 1 
 
 M. Constant Prevost mentions how he had seen on the 
 way to Chambery calcareous surfaces deeply furrowed and 
 boulders foreign to the country, notably a block of green schist 
 fifteen feet in diameter, and which he attributed to the action 
 of glaciers. 2 
 
 M. Le Blanc, at the same meeting, reported that M. 
 Fargeaud, professor of Physics at Strasburg, had discovered 
 similar phenomena in the Black Forest and the Pyrenees, 
 which he explained by the theory of glaciers. 
 
 In 1839 Studer, who had been a champion of the diluvial 
 theory, read a paper before the French Geological Society, in 
 which he confessed that after examining the glacier of 
 Gorneren, the lower part of which is formed by the union of 
 the glaciers of the Cima de Jazi, Monte Rosa, and Breithorn, 
 and on climbing the Eiffel, now raised about 500 feet above 
 the glacier, he says : u The height of this place above the glacier 
 prevents us supposing that the glacier ever rose so high, since 
 the commencement of the present epoch ; and yet we saw the 
 surface of the serpentine rocks polished like a mirror, and 
 covered with furrows and strise nearly horizontal, and in every 
 respect resembling those in contact with the glacier itself. 
 The supposition of currents of water carrying along stones, to 
 which this condition of surface may be attributed, is in like 
 manner rendered a very improbable explanation, in conse- 
 quence of the isolation of the crest between two very deep 
 valleys of ice and the proximity of the summits of all this 
 group of mountains. At a league distant from this place, below 
 Zermatt . . . we again find on granitic gneiss the same 
 polished surfaces, farrowed and striated, and those rounded 
 
 1 Bull. Soc. Geol., 2nd ser., ii. 249 ; see also a paper by M,,Collomb 
 in the same volume, p. 506, etc. 
 
 2 Bull. Soc. Geol. de France, xi., 65. 
 
Studer. -179 
 
 prominences which Saussure calls moutonnees.. On the 
 southern slopes of the Alps we have observed the same 
 phenomenon, both in the immediate proximity of the glaciers 
 and at continually increasing distances from them, to which 
 they cannot reach in the existing physical conditions. On 
 the route from Cormayeur to the Col de la Seigne, for 
 example, we find one of these surfaces opposite the glacier of 
 the Brenva, and if this glacier may have extended so far in 
 present times, this cannot be admitted in regard to the 
 polished and striated rocks which we have noticed between 
 Macugnagna and Pesterena, nor in regard to those of Val- 
 Quarrazza, at the northern base of Monte Turlo ; and still less 
 in relation to the distinctly marked ones which we observe 
 near St. Vincent, in the vale of Aosta, on coming from the 
 defile of Mont Jo vet/' Our author then goes on to describe 
 as a remarkable feature the larger bank or range of hills 
 separating the basin of Ivica from that of Biella, extending to 
 the foot of the Alps over a length of five or six leagues, as far 
 as Santja in the plain of Piedmont. " The height of this range, 
 on the road from Mongrande to Bolengo, cannot be estimated 
 at less than a thousand feet above the neighbouring plain ; it 
 then diminishes in height as it recedes from the ridges of the 
 Alps, in such a manner that the latter, when seen at a distance, 
 appear like a very uniform talus, which had been cut on its 
 two sides by the rivers which it separates. . . . The entire 
 mass of this hill seems to be composed of Alpine debris. 
 Blocks of gneiss and other rocks of all sizes, many of them 
 from fifteen to twenty feet in length, with their edges little 
 worn, are scattered over the whole surface, and wherever the 
 soil itself is disclosed, nothing is to be seen but unstratified 
 gravel and sand, enclosing a great number of these same blocks. 
 If we traverse the plain of Ivica and ascend Lessolo to the 
 plateau of Yico and Brosso, we perceive a like number of 
 Alpine blocks covering the whole of this rather steep declivity, 
 which rises to a height nearly equal to that of the opposite 
 edge above the soil of the plain.-" Studer goes on to say : " I 
 am almost disposed to believe that the partisans of new 
 doctrines respecting moveable rocks or boulders, will see in 
 these hills a classical example of the colossal grandeur of 
 ancient moraines, and that, if they happen to examine the 
 
 N 2 
 
I So The Glacial Nightmare. 
 
 roches moutonnees, and striated surfaces in the neighbourhood 
 ot" Ivica, they will declare their conviction that the ancient 
 glacier, of which we discovered traces at St. Vincent, had 
 extended as far as the edge of the great plain of Piedmont/' 3 
 
 Studer, while to some extent a convert to the theory of 
 glaciers, did not subscribe absolutely to it as explaining all the 
 facts. Writing in 1848 about the glacial phenomena of Scot- 
 land, we find him still saying : " To the assumption of former 
 glaciers, I feel quite favourably disposed ; although it appears 
 difficult to believe in so unlimited an extension of them 
 as must be assumed in this case, as well as in many others, 
 if we attribute the whole phenomena of erratics to this 
 agency/' 4 
 
 Meanwhile the same theory had been applied to the 
 American phenomena. In some notes on American geology 
 by T. A. Conrad, published in Silliman's Journal for 1839, 
 he writes : ' ' The phenomena of the ' diluvial epoch ' have long 
 attracted peculiar attention, from the many curious and 
 highly interesting facts which they embrace, and the great 
 difficulty of reconciling them with existing hypotheses. 
 Enormous angular masses, transported perhaps a hundred 
 miles from the parent rock, and reposing on sand or gravel 
 which even a mill-stream would have swept away, bid 
 defiance to the mighty currents which so long flourished in 
 the imaginations of certain geologists. Whence came these 
 floods and whither did they go ? Such gigantic movements 
 would soon have restored the equilibrium of the waters ; and 
 truly they should have been busy during their short reign on 
 earth, to grind down mountains into sand, roll into smoothness 
 myriads of siliceous pebbles, plough deep trenches in the 
 solid rocks, and polish their surfaces with sand. 
 
 (f The boulders rest usually on sand, gravel, or the natural 
 soil, which would naturally have been swept away, had 
 currents transported these huge fragments, leaving them 
 in every instance reposing on indurated strata. The 
 hypothesis of ice floes bringing them from the north, floating 
 on the waters of an ocean, and depositing them where they 
 
 3 Bull. Soc. Geol. Feb. 1840 ; Ed. New Phil. Jour. xxix. 274>279. 
 
 4 Ed. Phil. Jour. xlvi. 171. 
 
Conrad. 1 8 1 
 
 are now found, has been supported by some of the geologists 
 of the present day ; but this is in direct opposition to 
 another theory of these same geologists, that a higher mean 
 temperature prevailed over the northern regions of that period 
 than now reigns in temperate climates. This would not have 
 been the case, all other things being equal, if the northern 
 half of the continent had been nearly all formed by the ocean, 
 notwithstanding the mean temperature is greatly modified in the 
 same parallel of latitude, by the presence or absence of large 
 bodies of water, rising with the former and falling with the 
 latter condition of the globe. Whence then this immense body 
 of ice, which has scattered boulders over so vast a country ? 
 Appearing too at an epoch subsequent to the extinction of 
 the mastodon and other mammalia, which evidently lived in 
 this region and enjoyed an equatorial climate anterior to the 
 icy period. Nothing can reconcile this apparent contradiction 
 but the admission of a fall of temperature far below that 
 which prevails in our day, freezing the enormous lakes of 
 that period, and converting them into immense glaciers, which 
 probably continued undiminished during a long series of 
 years. At the same time, elevations and depressions of the 
 earth's surface were in progress, giving various degrees of 
 inclination to the frozen surface of the lakes, down which 
 boulders, sand and gravel would be impelled/ to great distances 
 from the points of their origin. This in some cases might 
 result from gravity alone ; but in others, during the close of the 
 epoch when the temperature had risen, and avalanches began 
 to descend from the mountain tops and from numerous less 
 elevated places, there occurred on a vast scale the same 
 phenomena which are now familiar to the travellers on the 
 Alps. . . . M. Agassiz attributes the polished surfaces of the 
 rocks in Switzerland to the agency of ice, and the ' diluvial 
 scratches' as they have been termed, to sand and pebbles 
 which moving bodies of ice carried in their resistless course. 
 In the same manner I would account for the polished surface 
 of the rocks in Western New York. . . . The scratches and 
 grooves, Mr. Hall informs us, on the rocks bordering the 
 Genesee river, have a direction N.N.E. and S.S.W., and they 
 would probably therefore follow the dip of the stratum down 
 which the ice moved. ... If a glacier had previously occupied 
 
1 82 The Glacial Nightmare. 
 
 this area, the uplifts would have produced a symmetrical line in 
 the ice, and pebbles and boulders thus brought from opposite 
 directions. Mr. Hall has noticed this phenomenon, but attri- 
 butes it to the agency of opposing currents. 
 
 " While granite boulders have been removed to surprising 
 distances from the rocks in situ, those of transition limestone 
 and sandstone seem never to have been far removed from the 
 parent mass, a fact which harmonizes with the theory of 
 refrigeration. The vast thickness of granite, and its corre- 
 sponding uplift from the force of crystallization, has protruded 
 its naked summits through the overlying strata, and from 
 these peaks, rising to a great altitude, replete with parallel 
 fissures, and split and rent by the upheaving power, large 
 masses would necessarily fall, and when coming in contact 
 with the surface of a glacier, however slightly inclined from 
 the horizon, many of the boulders might of course traverse the 
 extreme limit of the slope, and without losing their angular 
 form ; but the limestone fragments being imbedded in the 
 bottom of the glacier would be only affected in position by 
 contraction and expansion of the ice, and the more extensive 
 movements caused by its breaking up in melting, which would 
 have ample power to wear down the angles of these fragment- 
 ary rocks/' Lastly, our author says, " I think it impossible 
 to explain the breaking up of rocks to a distance of many 
 feet below the surface except by the agency of intense cold, 
 freezing the water which filled the fissures and thus forcing 
 the rocks into tubular fragments and disturbing their position 
 by the lateral and upward pressure/'* 5 
 
 Early in the summer of 1840 Agassiz visited England with 
 the express intention, as he tells us, of finding traces of 
 glaciers in Great Britain, and was advertised in Punch as on 
 a sporting tour in search of moor-hens (moraines). He 
 travelled through the north of England, Scotland and Wales 
 with Buckland, who had, on a previous visit to Switzerland, 
 been much impressed with his presentation of the glacial 
 phenomena of Switzerland. The result of this visit was 
 certainly very remarkable. 
 
 " The first valley I entered in the glacial regions of Scotland," 
 
 - 
 6 Silliman's Journal, xxxv. 240-243. 
 
Agassiz. 183 
 
 says Agassiz, " was barred by a terminal moraine ; and through- 
 out the north of England, as well as in Scotland and Ireland, 
 I found the hill sides covered with traces of glacial action, as 
 distinct and unmistakable as those I had left in my native 
 land. Not only was the surface of the country polished, 
 grooved and scratched, as in the region of existing glaciers 
 presenting an appearance, corresponding exactly to that 
 described elsewhere, but we could trace the path of the 
 boulders where they had come down from the hills above, and 
 been carried from the mouth of each valley down into the 
 plains below/' 6 Again he says : " Some of the finest roches 
 moutonnees I have seen in Scotland are to be found at the 
 entrance of the valley of Loch Treig. ... I have observed 
 many other roches moutonnees in Scotland, especially about 
 the neighbourhood of Loch Awe, Loch Fyne and Loch Etive. 
 In fact, they may be found in almost all the glens of Scotland, 
 in the lake-region of England, and in the valleys of Wales 
 and Ireland." 7 
 
 On November 4th M. Agassiz read his famous paper on 
 the former existence of glaciers in Scotland, Ireland and 
 Eogland. In this paper he tells us he had arrived at the con- 
 clusion that great masses of ice, and subsequently glaciers, 
 existed in these portions of the United Kingdom at a period 
 immediately preceding the present condition of the globe. 
 He urged that " the distribution of blocks and gravel, as well 
 as the polished and striated surfaces of rocks in situ, do not 
 indicate the action of a mighty current flowing from N.W. 
 to S.E., as the blocks and masses of gravel everywhere 
 diverge from the central chains of the country, following the 
 course of the valleys. Thus in the valleys of Loch Lomond 
 and Loch Long, they range from N. to S., in those of Loch 
 Fine and Loch Awe from N.W. to S.E., of Loch Etive and 
 Loch Leven from E. to W., and in the valley of the Forth 
 from N.W. to S.E., radiating from the great mountain masses 
 between Ben Nevis and Ben Lomond." Ben Nevis in the 
 north of Scotland, and the Grampians in the south, are con- 
 sidered by the author to constitute the great centres of dis- 
 persion in that kingdom and the mountains of Northumberland, 
 
 6 Geol. Sketches, 30, 31. y Id. 63-64. 
 
1 8 4 The G lacia I Nigh tmare . 
 
 Westmoreland, Cumberland and Wales, as well as those of 
 Ayrshire, Antrim., the west of Ireland and Wicklow, to be 
 other points from which blocks and gravel have been dispersed, 
 each district having its peculiar debris, traceable in many- 
 instances to the parent rock, at the head of the valleys. 
 Hence, M. Agassiz, says, " it is plain the cause of the transport 
 must be sought for in the centre of the mountain ranges, and not 
 from a point without the district." The Swedish blocks on the 
 coast of England do not, he conceives, contradict this position, 
 as he adopts the opinion that they may have been transported 
 on floating ice. 8 
 
 Agassiz attributes the parallel roads of Glenroy to the 
 damming up of a main valley by glaciers coming from lateral 
 valleys, thus forming lakes which gave rise to stratified 
 deposits and parallel roads or beds of detritus at different 
 levels. 9 He also seems to argue that similar barriers may 
 have existed at the point of contact with the sea, whence the 
 formation of salt marshes inhabited by animals the remains 
 of which are found in the clays superimposed on the Scotch 
 till. 
 
 In his paper on the parallel roads of Glenroy, Agassiz 
 enlarged on these arguments with his customary graphic force. 
 "Tides or currents," he says, "driven powerfully and con- 
 stantly against a rocky shore, and bringing with them hard 
 materials, may produce blunt, smooth surfaces, such as the re- 
 peated blows of a hammer on stone would cause; but they 
 never bring it to a high polish because the grinding materials 
 are not held steadily down in firm, permanent contact with the 
 rocky surfaces against which they move, as is the case with a 
 glacier. On the contrary, being dashed to and. fro, they strike 
 and rebound, making a succession of blows, and never a con- 
 tinuous, uninterrupted pressure and friction. The same is 
 true of all the marks made on rocky shores against which loose 
 materials are driven by water currents. They are separate, 
 disconnected, fragmentary; whereas the lines drawn by the 
 hard materials set in the glacier, whether light and fine or 
 strong and deep, are continuous, often unbroken for long dis- 
 tances and rectilinear. Indeed, we have beneath a glacier a 
 
 8 Proc. Geol Roc. iii. 328, 329. 9 Id. 332. 
 
Buckland. 185 
 
 complete apparatus adapted to all these results. In the softer 
 fragments, ground to the finest powder under the incumbent 
 mass, we have a polishing paste ; in the hard materials set in 
 that paste, whether pebbles, or grains of sand, we have the 
 various graving instruments by which the finer or coarser 
 lines are drawn. Not only are these lines frequently uninter- 
 rupted for a distance of many yards, but they are also parallel 
 except where some change takes place in the thickness of the 
 ice, which may slightly modify the trend of the mass, or where 
 lines in a variety of directions are produced by the intermit- 
 tent action of separate glaciers, running successively at differ- 
 ent angles over the same surfaces. The deeper grooves 
 sometimes present a succession of short staccato touches, just 
 as when one presses the finger vertically along some surface 
 where the resistance is sufficient to interrupt the action with- 
 out actually stopping it, a kind of grating motion, showing 
 how firmly the instrument which produced it must have been 
 held in the moving mass. No currents or sudden freshets 
 carrying hard materials with them, even moving along straight 
 paths down hill-sides or mountain slopes, have ever been 
 known to draw any such lines. They could be made only by 
 some instrument held fast as in a vice by the moving power. 
 Something of the kind is occasionally produced by the drag of 
 a wheel grating over rocks covered with loose materials." 1 
 
 The 18th of November, 1840, was a memorable day in the 
 history of geology in England. On that day Dr. Buckland 
 read a memoir on the evidences of glaciers in Scotland and the 
 north of England. In it the great champion of diluvian 
 methods and the author of the " Keliquise diluvianaa," confessed 
 that he had been won over by the arguments of Agassiz, who, 
 in October, 1838, had urged upon him that the polished, 
 striated and furrowed surfaces on the S.E. slope of the Jura 
 near Neuchatel, and also the transport of boulders to the Jura 
 were the effects of ice. Buckland tells us that, after devotin^ 
 some days to the examination of actual glaciers in the Alps, he 
 had acquiesced in Agassiz's theory relative to Switzerland. He 
 goes on to say that, as early as 1811, he had observed on the 
 head rocks on the left side of the gorge of the Tay, near 
 
 1 Geological Sketches, 2nd ser. 35, 37. 
 
1 86 The Glacial Nightmare. 
 
 Dunkeld, rounded and polished surfaces; and in 1824, in com- 
 pany with Lyell, grooves and striae on granite rocks near the 
 east base of Ben Nevis. About the same time Sir George 
 Mackenzie pointed out to him, in a valley near the base of 
 Ben Wyvis, a high ridge of gravel laid obliquely across, in a 
 manner inexplicable by any action of water, but in which, 
 after his examination of glaciers in Switzerland, he recognized 
 the form and condition of a moraine. In his present memoir 
 Dr. Buckland brought together a number of effects which he 
 had noticed in Scotland, and which he assigned to glacier 
 action. Inter alia, he mentions (1) a long terminal moraine, 
 like the vallum of an ancient camp, covered with grass, about 
 400 yards long, 100 feet broad at the base, and from 20 to 30 
 feet high, and chiefly formed of a mass of unstratified pebbles, 
 in the ravine of Creekhope Linn, near Dumfries. (2) He 
 considers the gravel and sand covering the greater part of 
 the granite table-land from Aberdeen to Stonehaven, as the 
 detritus of moraines, the larger insulated tumuli and tortuous 
 ridges of gravel occupying 100 acres near Forden, a mile 
 east of Achinbald, he looks upon as terminal moraines, and 
 the blocks, large pebbles and small gravel spread over the 
 level portions of the valley of the North Esk, after emerging 
 from the Grampians, as the residue of moraines re-arranged 
 by water. 
 
 To glacier action, modified subsequently by water, he assigns 
 the cones and ridges of gravel at Cortachy and Piersee in 
 Forfarshire ; the polish and striae on a porphyritic rock im- 
 mediately above these; the vast longitudinal and insulated 
 ridges of gravel extending for two or three miles up the valley 
 east of Blair Gowrie, and the transverse barriers forming a 
 series of small lakes in the valley of Lunanburn ; the lofty 
 mounds comprising the ornamental grounds near Dunkeld 
 Castle, and the detritus in the valleys of the Tay, the Tummel 
 and the Garrie. He refers to the vast congeries of gravel 
 and boulders on the shoulder of the mountain exactly opposite 
 the gorge of the Tummel; the mammillated, polished and 
 striated slate rocks about a mile above the falls of the Tummel; 
 and the rounded outline and polish on veins of quartz, which 
 project eight or ten inches above the weathered masses of 
 mica slate near the same locality. Similar mammillated masses 
 
Buckland. 187 
 
 of mica slate retaining strige and flutings are visible, lie says, 
 at Bohaly, one-and-a-half miles east of Tummel Bridge. 
 
 Dr. Buckland pointed out the rounded, polished, and striated 
 surfaces, many of them recently exposed by the making 
 of roads, on the shoulders of Schiehallion, and also polished 
 and striated surfaces on dykes of porphyry recently exposed 
 in the valley called the Braes of Fors descending from 
 Schiehallion and on the intervening space, where the slate and 
 quartzite are rounded, polished, grooved, and striated. The 
 strige, as in the case of the porphyry dykes, are parallel to the 
 direction a glacier would naturally take. 
 
 Two lofty ridges of gravel crossing the park at Taymouth 
 Castle, Dr. Buckland considers to be moraines or the detritus 
 of moraines, and, like the deeply scored and fluted boulders 
 of hornblende near Fortingal, to have been distinctly due 
 to glacier action. 
 
 On the highlands separating the Tay and the Bran are a 
 remarkable group of moraines, and between the sixteenth and 
 the fourteenth milestone are thirty or forty round-topped 
 moraines, from thirty to sixty feet high, crowded together like 
 tumuli. These mounds, composed of unstratified gravel and 
 boulders, cannot be referred to the action of water since they 
 are placed precisely where a current descending from the 
 adjacent high lands would have acted with the greatest 
 velocity, and they exactly resemble, according to Dr. Buckland, 
 some of the moraines in the valley of the Rhone between 
 Martigny and Lock. 
 
 In Strathearn are ridges and terraces of gravel, the detritus 
 of moraines ; near Comrie in the same valley hard slaty rocks of 
 the Devonian era have been rounded and striated. 
 
 Near there Dr. Buckland marked down spots where, if 
 glaciers formerly existed, traces ought to occur, and he found 
 them accordingly on a hill above the so-called Devil's Cauldron, 
 where there are rounded surfaces of greenstone covered by 
 moraines. At Kenagart, near the same place, is a small 
 cluster of moraines, easily separable into lateral and ter- 
 minal. Two miles higher up the valley is a medial moraine. 
 He mentions numerous other traces in the same valley. 
 
 At Loch Earn Head, at the junction of Glen Ogle and Loch 
 Earn, is a group of conical moraines. They are at the very 
 
1 88 The Glacial Nightmare. 
 
 point where, had they been brought by a rapid current, they 
 would have been propelled into the loch, but nevertheless in 
 the exact position where the terminal moraine of a glacier 
 would be deposited. 
 
 Near Callender, the Dean says, "Moraines are stated to 
 cover more or less the valley of the Teith from Loch Katherine 
 to Callender, and the lofty terraces flanking the valley from 
 Callender to Doune are considered to be the detritus of 
 moraines, modified by the great floods which accompanied the 
 melting of the ice. One of them has been mapped as the 
 vallum of a Roman camp, and the greater part of the first 
 table-land on the right bank of the Teith between Callender 
 and Doune is composed of re-arranged glacial detritus. 
 
 Having thus surveyed the Highlands, Dr. Buckland moved 
 into the lowlands. He points to striae and furrows still 
 existing on the hills of Stirling and Edinburgh, and assignable 
 only to ice action. Some of these may have been caused by 
 stones projecting from the side or bottom of floating masses 
 of ice, bat this could not be the case at the Blackford Hills, 
 two miles south of Edinburgh. On the south face of this hill, 
 at the base of a cliff of trap, is a natural vault filled with 
 gravel and sand, cemented by recent infiltration of carbonate 
 of lime. The sides and roof of this vault are highly polished 
 and covered with striae, irregularly arranged with respect to 
 the whole surface, but in parallel groups over limited extents. 
 He shows that these marks could not be referred to pebbles 
 moved by water, or fixed to floating ice. It is, however, 
 easy to explain the polish by the long-continued action 
 of fragments of ice forced into the cave laterally from the 
 bottom of a glacier descending the valley, on the margin 
 of which the vault is placed, and the irregular grouping of 
 the parallel striae by the unequal motion of different frag- 
 ments of ice, charged with particles of stone firmly fixed in 
 them, like the teeth of a file. The cave is not more than 300 
 feet above the level of the sea, and justifies the belief that 
 glaciers may also at that period have covered Calton Hill and 
 the Castle Hills of Edinburgh and Stirling. 2 
 
 South of Edinburgh Dr. Buckland drew attention to a 
 
 Proc. Geol. Soc. iii. 332, 337. 
 
Buckland. 189 
 
 longitudinal moraine in the valley of the North Tyne, about a 
 mile east of Haddington, to others near Dunbar, and others 
 again dispersed in terraces in the high valleys at the eastern 
 extremity of the Lammermuir Hills. While some round and 
 oblong mounds of gravel are to be seen on a slope of a hill 
 300 to 400 feet above the sea-level, on the left bank of the 
 Tweed, three miles north of Berwick. 
 
 Mr. C. Trevelyan had pointed out to the Dean in 1831 a 
 tortuous ridge of gravel at North Charlton, between Belford 
 and Alnwick, supposed to be an inexplicable work of art, but 
 which he in 1838 became convinced, from an examination of 
 the upper glacier of Grindelwald and that of Rosenlaui, is a 
 lateral moraine. Enormous moraines are said to cover a tract 
 four miles from N. to S. and two from E. to W., immediately 
 below the mountains of the eastern valleys of the Cheviots. 
 The high road winds among cultivated ones near Waller, 
 through North and South Middleton, and by West and East 
 Lillburn to Rosedean and Wooperton. 
 
 Turning to Cumberland and Westmoreland, he pointed out 
 extensive moraines loaded with enormous blocks of porphyry 
 and slate near the junction of the Eden with the Eament and 
 the Lowther, as the remains of glaciers which formerly 
 descended from Helvellyn, Martindale, etc. Similar moraines 
 exist near Eden Hall, four miles east of Penrith. Immediately 
 below the gorge through which the Kent descends from the 
 valleys of Kentmere and Long Sledmere, many hundred acres 
 of the valley of the Kendal are covered with large and insulated 
 piles of gravel, and smaller moraines or their detritus nearly 
 fill the valley from Kendal to Morecombe Bay. A group of 
 moraines occupies a space of about 200 acres at an elevation of 
 500 feet on the shoulder of a mountain in front of the valley of 
 Long Sleddale, on the high road from. Shap to Kendal. South 
 of Kendal the high roads from Burton and Milnthorpe to 
 Lancaster pass for the greater part over moraines or their 
 detritus. 
 
 Dr. Buckland considers that many of the conical hills laid 
 down on Fryer's large map of Cumberland, in the valley of the 
 Duddon, at the south base of Harter Fell, are moraines ; that 
 some of the hillocks on the same map on the right of the Esk, 
 at the E. and W. extremities of Muncaster Fell, are also 
 
1 90 The Glacial Nightmare. 
 
 moraines formed by a glacier which once came down from Sea 
 Fell, and that many of the hillocks near the village of Wastdale 
 were formed by moraines descending westwards. Extensive 
 medial moraines exist on the shoulder of the hill called the 
 Braw Top at the junction of the valley of the Greta with that 
 of Derwent Water. 
 
 He found polished and striated surfaces on a recently 
 exposed surface of greywacke at Fox House near Ambleside, 
 also in a slate quarry at Kydal, and on newly-bared rocks by 
 the side of the road ascending from Grrasmere to the pass 
 of Wythburn. Many of the round and mammillated rocks 
 at the bottom of the valley leading from Helvellyn to Winder- 
 mere, he urges, also owe their form to glacial action. 
 
 The remarkable assemblage of boulders of Criffle granite at 
 Shalkbeck, between Carlisle and Cockermouth, he thinks, were 
 conveyed across the Solway Firth on floating ice, as similar 
 boulders are said to have come from Scandinavia. 
 
 Dr. Buckland concludes his paper by an examination of the 
 critical case of Shap and its granite boulders. The fact that 
 these boulders have spread north, east and south seems only 
 consistent with glaciers descending the valleys radiating from 
 Shap Fell. " There are abundant proofs," says Dr. Buckland, 
 " of the existence of such a glacier in large mud and boulder 
 moraines, in the ascent of the gorge between Shap Fell and 
 Birbeck Fell, and in the furrows and strias as well as the 
 mammillated forms of the rocks at the portals of the gorge, 
 particularly on the northern side." In the physical structure 
 of this neighbourhood, Dr. Buckland pointed out other con- 
 ditions which would have facilitated the accumulation of glaciers, 
 as the lofty mountains of Yardale Head, which overtop Shap 
 Fell on the N.W., and the still higher mountains to the W., 
 whose snows must have nourished enormous glaciers ; and 
 he concludes by stating that Professor Agassiz, during an in- 
 dependent tour, arrived at similar conclusions respecting the 
 mode by which the Shap boulders were distributed. 3 
 
 The reading of this paper by Buckland led to a dramatic 
 scene. Murchison at once took up the cudgels. " The day 
 will come/' he sneeringly said, " when Highgate Hill will be 
 regarded as the seat of a glacier, and Hyde Park and Belgrave 
 
 8 Id. 34-5, 348. 
 
Whewell. 
 
 Square will be the scene of its influence." He protested 
 against having- the question prejudged by giving the name of 
 moraines to the diluvian mounds. "There ought to be a 
 co-ordinate relation in the phenomena. But in the Highland 
 mountains, not one-third the elevation of the Alps, we have 
 moraines two or three times the magnitude of any known in 
 Switzerland. . . . These grooved and striated surfaces and 
 heaps of boulders are also to be found in Scandinavia, on the 
 east of the Gulf of Bothnia, all proceeding from the north and 
 north-west. Have these crossed the gulf on ice? In Russia 
 too we shall find them where there are no mountains, and if 
 we look to the remains of marine shells found in beds elevated, 
 differing in no respect from those in our present seas, except 
 that they are called Pleistocene, we have proof of a lower 
 elevation at the very time when these glaciers should be 
 introduced. On these accounts I am still contented to retain 
 our old ideas, that when a mountain was elevated or a body 
 of water passed over a series of elevations, the diluvium would 
 descend with the strike, and be disposed in mounds and terraces 
 according to the direction of currents, etc. 
 
 On the same evening Dr. Whewell objected to the glacier 
 explanation, inter alia, (1) Because of the extent of the 
 diluvium over such wide areas in countries of opposite physical 
 structure, surface, climate, etc. ; (2) From the marine remains 
 of the glacial period showing the continents to have been sub- 
 merged ; (3j Because of the physical conditions under which 
 glaciers now exist. <( We find them universally stretching out 
 from lofty mountain chains which take their rise in warm coun- 
 tries, so as to allow of the downward motion and the retiring in 
 summer. . . . Where shall we obtain mountains as fulcra for 
 glaciers, stretching many leagues into the plains, producing 
 such results as are ascribed to their action in Scotland? " 
 
 In reply Agassiz said that grooves are to be found in the 
 valley of the Aar glacier seven leagues (22 miles) below the end 
 of the present glacier. On soft rocks the grooves are annually 
 renewed in winter and removed by the atmospheric action in 
 summer. He said he had been for many hundred feet under the 
 glacier of Monte Rosa, and found the quartzose sand forming 
 a bed beneath, and acting like emery on the rocks. " Moraines 
 differ from all other accumulations, the lateral moraines are ex- 
 posed to constant friction with the rocks with which they are 
 
192 The Glacial Nightmare. 
 
 brought in contact, and their terminations are passed over by 
 the whole mass of the glacier, so that they become rounded and 
 striated ; while the medial moraines, remaining on the surface, 
 continue angular. When the glacier retreats in the summer, 
 the medial moraine, composed of angular fragments, is spread 
 out over the surface of the lateral and terminal moraines, 
 composed of rounded fragments. There are moraines in the 
 Alps 200 feet wide, composed of boulders several feet in 
 diameter." 
 
 Lyell, at the same meeting, said a glacier had been known 
 to retire half a mile in a single summer ; a number of moraines 
 might have been left in succession, and in severe winters might 
 be driven successively into one by the downward motion of 
 the glacier. 4 
 
 At the same time that Dr. Buckland was giving up his 
 diluvian views in favour of the glacier origin of the 
 phenomena of the drift, etc. etc., Lyell was similarly sur- 
 rendering to some extent his theory of drifting ice. This 
 partial change of view was embodied in papers read before the 
 Geological Society on November the 18th, and December 
 the 2nd, 1840, on the former existence of glaciers in Forfar- 
 shire, in which he goes over very much the same geological 
 reasoning which was pursued by Buckland. I do not think it 
 necessary to state his views in detail. 
 
 Ly ell's final conversion to the glacial theory did not take 
 place, however, till his visit to Switzerland in 1857. 
 
 As was to be expected, the adherence of two such doughty 
 champions as Dr. Buckland and Mr. Lyell to the glacier 
 theory caused a great revolution in geological opinion 
 in this country, and the champions of water, as the agent 
 by which the so-called diluvial phenomena had been 
 caused, had to beat a retreat. Some still held out, but 
 virtually the battle was over, and the army of enthusi- 
 astic workers in the still green and largely fresh pastures 
 of geology betook themselves with wonderful patience and 
 pertinacity to map out and to survey in detail the effects of 
 former glaciers in many countries. 
 
 In January, 1841, M. Leblanc mentioned having seen 
 
 4 Notes of a discussion at the Geological Society, taken by JJr. S. P. 
 Woodward, "Midland Naturalist," October, 1883.' 
 
Renoir and Durocher. 193 
 
 erratic blocks in the Atlas range in Africa, notably those on the 
 mamelon of Minirch, near Blidah, where they were arranged 
 in the manner they ought to be if a glacier had occupied the 
 valley of Wed Sidi el Kebir, and if Mirnich had been a kind 
 of projecting boss on the glacier forming a so-called " jardin " 
 on its mer de glace* 
 
 In the Bulletin of the French Geol. Soc. for the same year, M. 
 Renoir illustrates the erratic phenomena of northern Russia by 
 an appeal to glacier ice. He thus explains the fact mentioned 
 by M. Robert that while on the left bank of the Neva as it comes 
 from lake Ladoga and on the margin of the lake at the same 
 point primitive blocks rolled and rubbed occur in great'numbers, 
 none occur on the other bank. He explains them as forming a 
 moraine which directed the course of the Neva andformed a dyke 
 for it on its issuing from the lake. He similarly accounts for 
 the line of blocks between Wol-Raculskaia andCopachefskaia on 
 one of the banks of the Dwina, while the other bank is free from 
 them. A great flood, he says again, spreads debris evenly and 
 has not a tendency to create small hills. The numerous mounds 
 and small hills occurring in north Russia, he urges, are the 
 remains of moraines ; the fact of their occurring in parallel lines 
 is another evidence of the same fact. M. Robert had described 
 the whole coast of Finland from Helsingfors to Abo with the 
 adjacent islands, as being perfectly polished for some distance 
 into the interior. M. Renoir urges that this perfect polish is 
 clearly the work of masses of ice moving immediately over the 
 surface of the rocks. 6 
 
 The first to explain the erratics of the Pyrenees by means of 
 glaciers was, as I have said, Fargeaud, who, however, seems to 
 have published nothing on the subject. 7 In 1841 M. Durocher 
 mentions having found in the Pyrenees all three types of 
 phenomena which characterize the Alpine drift, namely polished 
 and striated surfaces, erratic blocks and stretches of gravel on 
 the flanks of the valleys. The striae follow the valleys on both 
 sides of the chain, while the polished surfaces occur wherever 
 the rocks are sufficiently hard to preserve them, as in the valleys 
 
 Bull. Soc. Geol. de France, xiii. 134. 
 
 Op. cit. p. 68. 
 
 Yideante, 178, see Bull. Soc. Geol. de France (1839) xi. 65. 
 
194 The Glacial Nightmare. 
 
 of La Tet, Audoire, Segre, Ariege, Yicdessos, Luchon, Essera, 
 Gedre, Bareges, Cauterets and Assan." 8 
 
 On the 2nd of May, 1842, M. Boubee read a paper before the 
 French Geological Society, in which he confessed that against 
 his will he had been obliged to appeal to glaciers to explain 
 the erratic phenomena which he had observed in the Cevennes 
 and the Pyrenees. He had spent nearly two years in making 
 observations in the latter chain and found on both flanks 
 of it polished and striated rocks, undoubtedly due to 
 the action of glaciers long before the historic period, and 
 also great moraines beyond the limits of the mountains 
 themselves. He noticed polished and striated surfaces in 
 the valleys of la Pigne, Lys, Larboust, Aran, Venasque, 
 Louron, Gavarine, etc. Moraines existed in these valleys 
 and in others besides, as those of Garen, near Bagneres 
 de Luchon, and those met in descending the valley of 
 Yenasque on the Spanish side of the chain. Some are extra- 
 ordinarily preserved as that of Labroquere, near St. Bertrand 
 de Comminges, that of Tibiran near the cave of Gargas, 
 those of Gripp, of Argelis, of the valleys of Gavarine and Heas, 
 that of Arudy near Pau, that at the meeting of the Taule, 
 Tarascon and Orgeix in the Ariege, that of Vernet at the foot 
 of the Canigu, etc. ; and he is very emphatic iin the conclu- 
 sion that precisely the same evidences of former ice action 
 which are found in the Alps occur also in the Pyrenees. 
 Although he had not examined the Cevennes with the same 
 care, he felt that the evidences he had there seen, proved the 
 same fact. 9 
 
 Lastly, Max Braun * found in all the Pyrenean valleys 
 polished and striated rocks, roches moutonnees and moraines, 
 but he found no erratic blocks on the southern flanks of the 
 Oorbieres of the Black mountain, nor of the Agenais, which 
 points to the fact that the Pyrenean glaciers did not extend, 
 beyond the chain itself. 
 
 The new theory was now making rapid way and was the subject 
 of fierce discussion in every scientific circle in Europe. Yon 
 Buch, Sedgwick, Sefstrom, Durocher, the younger De Luc, 
 
 8 Compte Rendu, xiii. 902, and Yoyage of the Recherche, 140. 
 
 9 Bull. Soc. Geol. de France, xiii. 317-18. 
 1 Neu. Jahr (1843), p. 80. 
 
Charpentier. 1 9 5 
 
 and others fought it bravely, but the close reasoning, the careful 
 and minute observation and the generally sober judgment of its 
 two great advocates, Charpentier and Agassiz, were irresistible. 
 
 In 1841 Charpentier published a second memoir on the 
 dispersal of the erratics by glaciers, entitled " Essai sur les 
 Glaciers," in which he, in the most methodical manner, pro- 
 ceeded first to demolish every other theory up to that time 
 suggested to explain the dispersal of the erratics, and then 
 stated in greater detail the positive case in favour of his own 
 theory of the former great extension of glaciers. I have 
 already incorporated some of his criticisms. 
 
 In this paper he sums up in several pages his general objec- 
 tions to the transport of the blocks by water. He urges that 
 in the famous flood in the Vallee de Bagnes in 1818, only five 
 large blocks of stone were moved, and then only for a distance of 
 1800 feet, in the narrowest part of the torrent, and they were 
 borne along its bottom and not carried along the top or the 
 borders, while the other materials were deposited directly the 
 torrent began to widen and spread out. The transport in this 
 case was also assisted by the great masses of debris in the shape 
 of chalets, trees, etc., etc., which were also borne along. Along 
 the streets of the village and the town of Martigny, where the 
 current was nine and a half feet deep and sufficiently powerful 
 to carry along houses, etc., it left, no gravel or even sand, 
 but only fine mud. He urges that this being so, it is im- 
 possible to conceive how a block 100,000 cubic feet in extent 
 could be deposited on the mountain Plan-y-Beuf near 
 Ferret, 2700 feet above the level of the valley, a valley, 
 be it remarked, only three and a half leagues in length. If 
 a great current, as in the case of Bagnes, deposits its debris 
 when it spreads out, why did that coming down the Valais 
 not leave its burden in lower Switzerland instead of carrying 
 it along to the flanks of the Jura ? 
 
 2. The fact that the rocks are not sorted according to their 
 size and weight, requires not only that the current should 
 have been exceedingly powerful, but should also have been 
 suddenly arrested. If it abated gradually it would sort its 
 contents, but the debris are not so sorted. 
 
 3. The fact that on the Jura the greater number of the 
 blocks are found opposite the outfall of the great Rhone 
 
 o 2 
 
196 The Glacial Nightmare. 
 
 valley, shows, according to M. Charpentier, that they were 
 not deposited by a current, for, he urges, when a torrent 
 suddenly comes upon a transverse barrier it throws its contents 
 to the right and left and does not leave them directly in 
 front of it. 
 
 4. If the blocks were transported in great trajectories the 
 ruin they would cause in striking such a barrier as the Jura 
 would be visible in many places, and the blocks themselves 
 would be broken to fragments, but this is not the case, and 
 the exposed beds of Chasseron, which must have borne 
 the chief shock, are no more weathered than those of the rest 
 of the Jura, while the blocks which lie on it are no more 
 broken than those elsewhere. 
 
 Again, if, as M. Studer urged, the current was so charged 
 with mud as to make it more easily carry the great blocks, 
 how could it acquire the necessary speed ? One necessity here 
 seems to destroy another. As a matter of fact, blocks which 
 sometimes appear to be floating along a muddy current are in 
 reality rolling over other debris which supports them, but 
 there are no remains of such debris left in the valleys on 
 which the great blocks could thus roll. If there had ever 
 been such debris it must have filled up the lakes. 
 
 Certain valleys peculiarly situated, like that of Vallorbe, 
 ought to have contained large quantities of debris if it had 
 been deposited by water, but they are singularly free. 
 
 It is also impossible to understand how great blocks of 
 60,000 cubic feet in extent should have preserved their angles 
 sharp if they had been rolled by water. How, again, could 
 water deposit the debris in the form of bands and continuous 
 dykes running in parallel lines along the flanks of the moun- 
 tains and corresponding to one another on each side of the 
 valley ? Why are these dykes only found on the flanks of the 
 mountains and not in the open valleys ? How could a torrent, 
 again, deposit its debris in some places heterogeneously and in 
 others in a stratified way ? How, again, by this theory can we 
 explain the groups of similar rocks found together in so many 
 places, and which occur as the granite groups above Monthey, 
 those of Plan y Beuf, Steinhof, etc., and the calcareous 
 group of Devens and the Bois de la Chaux, unmixed with 
 other kinds of rock ? If, as has been supposed, these groups are 
 
Charpentier. 197 
 
 the debris of vast masses broken to fragments, we must suppose, 
 as in that of Monthey, that such a mass was three-quarters 
 of a mile long and several hundred feet wide. How, by such 
 currents, can we explain the suspended and poised blocks, 
 and those which are split from top to bottom ? 2 
 
 In regard to the champions of icebergs, Charpentier says 
 those who have invoked icebergs in Switzerland have done 
 so in one of two ways, in one they postulate a vast sea or lake 
 as enveloping the whole country, in the other that the bergs 
 were brought by currents. The former views are referred to by 
 Von Buch, and also formed the main position of Darwin. Char- 
 pentier says frankly that this theory accounts for the moving 
 of great blocks without rounding their angles, for the occurrence 
 of erratics in the valleys and in Lower Switzerland, as well 
 as on the flanks of the mountains, and for the non-occurrence 
 of blocks above a certain level; but he urges the great 
 improbability of a great lake or , sea occupying Switzerland 
 after the general elevation of its surface so as 10 reach the 
 feet of the glaciers. If we admit such a sea as possible, its 
 border ought to correspond with the limit at which the 
 erratics occur, but such a limit must have been horizontal, 
 unless the surface of the sea was not horizontal, whereas the 
 upper limit of the debris is in some cases, as on the Jura, 
 marked by a curve, and in others the upper limit is many 
 hundreds of feet higher in some spots than in others. The 
 notion of a curved or irregular surface to such a sea or lake is 
 contrary to the laws of physics. 3 
 
 Again, if Darwin's view were right, we must admit one of 
 three things, (1) either Switzerland acquired its present relief 
 under the level of the sea, if so, glaciers could not be formed, 
 for they cannot grow under water, and without glaciers there 
 can be no icebergs; (2) after the present configuration 
 of the country was attained, the sea must have invaded 
 Switzerland, until its depth between the Alps and the Jura 
 was at least 4428 feet, which is incredible ; or (3) Switzerland 
 and the rest of Europe must have been buried under water 
 up to the level of the highest erratics, which is also incredible. 
 
 But granting the possibility of any one of these solutions, 
 
 2 Charpentier, op. cit. 214-228. 3 Essai sar les Glaciers, 178-181. 
 
198 The Glacial Nightmare. 
 
 where do icebergs deposit loads simulating moraines so 
 closely as do those in many parts of Switzerland ? How is 
 it that the only bergs which reached the Jura were those 
 which came from the valleys of the Rhone and the Arve ? 
 How was it none reached there from the valleys of the Sarine, 
 Aar, Reuss, Limmat or Rhine ? or how is it that the bergs 
 coming from any one of these valleys never invaded another ? 
 Why did no bergs take the contrary direction and remove 
 loads from the Jura to the Alps? How is it the bergs coming 
 down the lateral valleys of " the Yalais " did not proceed right 
 across the main trunk and deposit their load on the flank 
 of the mountains immediately opposite ? Why are not the 
 largest accumulations found where the valleys change their 
 course, and where, consequently, the majority of the bergs 
 would be stranded ? How explain the mounds whose position 
 in the valleys is exactly that of terminal moraines, being per- 
 pendicular to the enclosing mountains ? How explain the 
 vertical furrows and erosions ? How is it this sea has left no 
 traces of its recent sojourn in Switzerland, no traces of wave 
 action, no marine debris, no shells, no sand or shingle like 
 that we find on the sea shore ? 4 
 
 Another theory urges a vast current as coming from the 
 upper Alps, which detached and carried off masses of ice-bear- 
 ing rocks like ice-rafts. This theory overlooks, Charpentier says, 
 the very sinuous character of the Alpine valleys and also the 
 fragile nature of ice. The current must have been at least 2500 
 feet deep, with a fall of one in three. How could bergs on such 
 a torrent arrive safely in lower Switzerland or on the flanks of 
 the Jura ? Many of them coming from the upper Yalais must 
 have been broken to pieces against the Col du Trient near 
 Martigny, and left their loads there, but we may look in 
 vain for them ; not a trace of rocks from the upper Valais has 
 been found at that very frequented place. The same argu- 
 ment applies elsewhere. How could such rafts, again, have 
 safely navigated defiles such as that by which the valley of 
 Binnen opens into the Rhone valley, or the defile of 
 Entremont, etc., without being broken and depositing their 
 loads ? When the lower glacier of Geetroz was destroyed by 
 
 * Id. 181-183, note. 
 
Ch a rpen tier. 199 
 
 M. Venetz, blocks of ice 10,000 to 40,000 cubic feet in size 
 were detached which fell into the Drance, but they were 
 broken to pieces so quickly that when they reached the 
 village of Lourdier, two leagues from the glacier, only 
 small fragments remained/' 
 
 Another theory still remains, namely that by which it is 
 sought to explain the dispersal of the erratics by the supposi- 
 tion that the ice covering the Alpine lakes (in which fragments 
 of rock were embedded) was broken to pieces, which floated 
 away with their loads. This theory invoked considerable 
 lakes as formerly filling the Alpine valleys, one of them occu- 
 pying the plain of lower Switzerland, and opening a com- 
 munication between the Alps and the Jura. These lakes, 
 frozen in the winter, were melted by the summer sun, 
 which broke their covering into fragments that floated away 
 with their loads of rock from one lake to another. Char- 
 pentier urges that there is no evidence of the former existence 
 in Switzerland of such a system of lakes at different levels, 
 nor would such a series explain the contour-line formed by 
 the erratics on the flanks of the mountains ; nor, unless we 
 invoke transcendental cold, can we suppose that in one winter 
 the lakes would be sufficiently frozen to bear great masses 
 of stone, several thousand cubic feet in extent, and to create ice 
 rafts sufficiently large to bear them away. 6 Having thus 
 negatived other theories, M. Charpentier proceeds to press 
 reasons in favour of his own view that the erratics were dis- 
 tributed by huge glaciers. 
 
 In addition to the arguments already quoted, he showed 
 that the erratic fragments deposited in each of the great 
 Swiss valleys comprise fragments of all the rocks found in 
 the mountains bordering that valley, just as is the case in a 
 modern moraine. The blocks are mixed, some with sharp 
 angles and some rounded also, as in modern moraines, where 
 the former have been carried down on the back of the glacier, 
 and the latter rolled by its underlying stream. Hence why 
 the largest blocks are the best preserved, because when they 
 fell on to the glacier they would travel nearer its medial line 
 and thus escape the rubbing they would have met with if 
 
 6 Id. 184-188. Op. cit. 189, 190. 
 
2OO The Glacial Nightmare. 
 
 smaller and deposited near its flanks. 7 No block is so big that 
 a glacier would not move it. Thus the Blaustein, a mass of 
 serpentine containing 244,000 cubic feet, has been moved by 
 the actual glacier of Matmark or Schwarzberg, and there were 
 old men living at Saas in 1821 who stated how they had heard 
 their fathers tell they had seen this block on the glacier. 
 The largest erratic known to Charpentier was that of Devens, 
 containing 161,000 cubic feet, and therefore considerably 
 smaller than the Blaustein. 8 The mode in which the debris is 
 arranged, again, is exactly that of modern glaciers. In some 
 cases the fragments are strewn sporadically. We may, in 
 illustration of this, compare the abandoned bed of the recent 
 glacier of 1'Arpellaz or Salene, near the hamlet of Pratdefort, 
 in the valley of Ferret, with the surface of certain sterile dis- 
 tricts of the canton of Fribourg, of the plateau of Jorat, of the 
 environs of Cossonay, la Bretonniere, Premier, etc. On the 
 other hand, when the deposits are in masses on a flat surface, 
 they occur generally in the form of long mounds or dykes, just 
 as a glacier deposits its moraine. If the slope is very great 
 it forms bands, single hills or rows of hills. 
 
 The occurrence of long and parallel bands of deposit at the 
 same height on both sides pf the valleys is exactly paralleled 
 by modern glaciers, and proves that the retreat of the glacier 
 was not continuous but intermittent and by a series of oscilla- 
 tions. M. Charpentier accounts for the absence of piled-up 
 deposits in the Yalais, except in the neighbourhood of Lausanne 
 and of the plateau of Jorat, by the almost horizontal character 
 of the valley from which this great glacier retired continuously 
 and without oscillations. 9 The occurrence of occasionally 
 stratified portions of sand and gravel he explains by the occa- 
 sional forming of lakes in the lateral valleys where the glacier 
 in the main valley barred them and prevented drainage, as 
 is the case now. In regard to the finding of fragments of 
 one kind of rock together, he mentions that if the famous 
 landslip of the Diablerets, which occurred in 1749, and that of 
 the Rorsberg in 1806, had fallen on glaciers they would have 
 formed two groups of fragments, one entirely composed of 
 calcareous rock and the other of gompholite. 1 In some 
 
 7 Id. 250, 252. 8 Op. cit. 252, 253. 9 Id. 253-256. Id* 261. 
 
Charpentier. 201 
 
 cases blocks are found singularly perched one on another, 
 as in the case of the well-known Pierre a Dzo, which is perched 
 on another block, and the Pierre a Bot, resting on one of its 
 smallest faces ; others again, like the Pierre a Mourguets, split 
 horizontally by having struck against another rock when 
 rolling, the Pierre Bassa and other stones similarly split, but 
 vertically. These are paralleled by similar rocks in modern 
 moraines formed by retreating glaciers, and are apparently 
 explainable in no other way than by an appeal to glacier 
 action. Charpentier shows how a glacier descending a main 
 valley intercepts all the debris coming down the lateral 
 valleys and prevents it from traversing the main trunk, thus 
 accounting for the fact that when two lateral valleys abouch 
 opposite one another the debris of one valley is not deposited 
 in the mouth of the other. The size of former glaciers is to be 
 measured by the height at which they left their debris in any 
 place. Thus from Aernen to Brigue, in the Haut Yalais, the 
 glacier must have reached 2800 feet above the Rhone. At 
 Brigue the valley widens and the glacier must have spread 
 out, and consequently we find the debris only reaches 2500 feet 
 above the Rhone. For seventeen leagues the valley remains 
 fairly even in size and the debris remain at about 2500 feet. 
 Between Martigny and St. Maurice, however, it contracts and 
 the glacier must also have contracted, and we consequently 
 find the debris between these towns occurring as high as 
 3000 feet. Below St. Maurice the valley again widens and 
 keeps its new width as far as the basin of Geneva, consequently 
 we find the height at which the debris occurs falling to 2300 
 feet. Beyond this the glacier would meet the plateau of Jorat, 
 which would impede it and cause it to spread out laterally until 
 blocked towards the south by the chain of Verreaux. Hence 
 why we find debris on the mountain of Playau, 2600 feet 
 above the Rhone. Towards the west there was no obstacle, 
 for there is the lake, hence we find that beyond Thonon the 
 line of debris comes down to the level of the lake itself. 
 Having traversed the Jorat, the glacier would meet with no 
 barrier until it reached the Jura, where an immense accumu- 
 lation of ice naturally followed, and hence we find the remains 
 of its frontal moraine at a height of 3000 feet, at the point 
 where the pressure was greatest, but as it relieved itself 
 
2O2 The Glacial Nightmare. 
 
 laterally, the depth of ice would diminish, and thus leave a 
 curved line of debris on the flanks of the Jura, just as we 
 find to be the case ; and inasmuch as a glacier moves fastest 
 along its medial line, it would deposit more debris where this 
 medial line abutted on the barrier, as we in fact find to be 
 the case. The effect produced by a mountain barrier would 
 be imitated where another glacier intervening prevented the 
 forward flow of the ice and caused it to thicken vertically. 
 At the Chalets du Boule, near Givrins, the debris is found at a 
 height of 2300 feet, for there the glacier coming from the 
 Rhone valley must have encountered that coming from the 
 Arve. 2 Just as the beds of the torrents flowing under modern 
 glaciers are free from moraines, so we must not expect to 
 find, and do not find, that kind of deposit where we have 
 reason to believe the beds of the ancient glaciers lay, of 
 which that of the Rhone valley was much the largest in 
 Switzerland. 3 
 
 Perhaps it will be well to state Charpentier's general 
 conclusion in his own words : " It goes without saying that 
 not only all the valleys of the Valais were filled with ice up 
 to a certain height, but that all lower Switzerland, in which 
 we find the erratic debris of the Rhone valley, must have been 
 covered by the same glacier. Consequently all the country 
 between the Alps and the Jura, and between the environs 
 of Geneva and those of Soleure, has been the bed of a 
 glacier." 4 
 
 While Charpentier battled bravely for his theory, he was 
 helped by a still more famous champion, namely Agassiz. 
 To use the graphic words of the latter, " Every terminal 
 moraine is the retreating footprint of some glacier, as it 
 slowly yielded possession to the plain, and betook itself to 
 the mountains. Wherever we find one of these ancient 
 semicircular walls of unusual size, there we may be sure the 
 glacier resolutely set its icy foot, disputing the ground inch 
 by inch, while heat and cold strove for the mastery. . . . By 
 these semicircular concentric walls we can trace the retreat 
 of the ice as it withdrew from the plain of Switzerland to the 
 fastnesses of the Alps. It paused at Berne, and laid the 
 
 2 Id. 269 273. 3 Id. 275. 4 Id. 266. 
 
Agassi z. 203 
 
 foundations of the present city, which is built on an ancient 
 moraine ; it made a stand again at the lake of Thun, and 
 barred its northern outlet by a wall which holds its waters 
 back to this day. 5 The great mountain chain of the Alps 
 has been a central axis from which immense glaciers at one 
 time descended in every direction, not only to its base, 
 beyond which the lowlands extend in flat undulations, but 
 to a greater or less distance over the adjoining plains, while 
 at present they are confined to the higher valleys." 6 
 
 Elsewhere, speaking of the glacial strise in the State 
 of Maine running invariably due north and south, and 
 never being deflected except by local causes, he asks : " How 
 is it possible to suppose that floating icebergs would advance 
 over such an uneven country with this steadfast, straight- 
 forward march ? Instead of ascending the hills, they would 
 be caught between them in the intervening depressions, or, if 
 the land were completely submerged, floated over them. . . . 
 Equally impossible is it to suppose that anything so 
 unstable as water has produced such straight and 
 continuous lines. 7 ... In the State of Maine I have followed, 
 compass in hand, the same set of furrows, running from 
 north to south in one unvarying line, over a surface of 180 
 miles, from the Katahdin Iron Range to the sea shore. These 
 furrows follow all the inequalities of the country, ascending 
 ranges of hills varying from twelve to fifteen hundred feet 
 in height, and descending into the intervening valleys, only 
 two or three hundred feet above the sea, or sometimes even 
 on a level with it. I take it to be impossible that a floating 
 mass of ice should travel onward in one rectilinear direction, 
 turning neither to the right nor to the left, for such a 
 distance. Equally impossible would it be for a detached mass 
 of ice, swimming on the surface of the water, or even with its 
 base sunk considerably below it, to furrow in a straight line 
 the summits and sides of the hills, and the beds of the valleys. 
 It would be carried over the depressions without touching 
 bottom. Instead of ascending the mountains, it would remain 
 stranded against any elevation which rose greatly above its 
 
 5 Agassiz, Geol. Sketches, 6, 7. 6 Id. 20. 
 
 7 Geol. Sketches, ser. ii. p. 150. 
 
204 The Glacial Nightmare. 
 
 own base, and if caught between two parallel ridges would 
 float up and down between them. 8 
 
 Elsewhere Agassiz points out that we cannot attribute the 
 so-called terminal moraines extending across the valleys to 
 the action of water. " Any current/' he says, " powerful 
 enough to bring the boulders and debris of all sorts of which 
 these walls are composed to the places where they are found, 
 would certainly not build them up with such regularity, but 
 would sweep them away or scatter them along the bottom of 
 the valley. That this is actually the case is seen in the lower 
 course of the valley of the Rhone where there are no trans- 
 verse moraines, while they are frequent and undisturbed in 
 the upper part of the valley/' He goes on to argue that the 
 lake which formerly occupied the ground here, swept away 
 and totally obliterated the traces of former glaciers. 9 
 
 Agassiz urges that the longitudinal moraines found in the 
 Alpine valleys differ from glacier-detritus remodelled or 
 spread out by water, in being disposed in ridges with a double 
 talus, one flank of which is presented to the glacier and the 
 other to the side of the valley ; and their continuity and 
 parallelism at the same height easily distinguish them from 
 the debris disposed along the bottoms of valleys by currents. 1 
 He contrasts the stratified deposits made by water with those of 
 glacier detritus. The materials of the former are much smaller, the 
 pebbles also are elongated, and fine gravel and sand generally 
 form the upper beds. In the latter, large and small materials 
 are arranged without order, the largest blocks being often in 
 the upper part; and where very large angular blocks occur, they 
 rest on the surface. Referring to polished and striated surfaces, 
 without denying the power of water to produce such results, 
 he says he has sought for them in vain on the borders of 
 rivers and lakes, and on sea coasts ; and that the effects pro- 
 duced by water are sinuous furrows, proportioned to the hard- 
 ness of the rocks ; not even, uniform, polished surfaces, such as 
 those presented by the rocks under discussion, and which are 
 independent of the composition of the stone. The parallel striae 
 in the general direction of the movement which the glaciers 
 
 Id. 159. 9 Geol. Sketches, ser. ii. 55. * Id. 330. 
 
Agassiz. 205 
 
 must have taken, and the so-called roclies mo utonnees are other 
 phenomena to which he refers as due only to ice, and which he 
 says he has traced under the glacier of the Aar, in the valley 
 of the Ehine and of Chamounix, and in Scotland on the banks 
 of Loch Awe and Loch Leven, and he adds that they are very 
 remarkable near Kendal. The most striking points in the dis- 
 tribution of the striae, are their diverging at the outlets of 
 the valleys, and their being oblique, and never horizontal on 
 the flanks, which they would be were they due to the agency 
 of water, or floating masses of ice. 2 
 
 Agassiz explains the Alpine lakes of Switzerland and north- 
 ern Italy as largely due to their lower extremities being closed 
 by transverse moraines. " The greater part of them/' he says, 
 11 have such a wall built across one end, and but for this 
 masonry of the glacier there would have been nothing to 
 prevent their waters from flowing out into the plain at the 
 breaking up of the ice period. We should then have had open 
 valleys in place of all these sheets of water/' Turning from 
 the plains to the mountains, he speaks of the traces of old 
 lateral moraines as uncertain and fragmentary, depending on 
 the existence of some ledge or terrace on the wall of the valley 
 affording support for them, and thus marking the former 
 limit of the ice ; but, he continues, " thousands of feet above the 
 present level of the glacier, far up towards their summits, we 
 find the sides of the mountains furrowed, scratched and 
 polished, in exactly the same manner as the surfaces over 
 which the glaciers pass at present. . . . Above the lines at 
 which these indications cease, the edges of the rocks are 
 sharp and angular, the surface of the mountain rough, un- 
 polished, and absolutely devoid of all marks resulting from 
 glacial action." 3 
 
 Later on he contrasts the effects of water and of ice in 
 their abrading power, and says : te On any surface over which 
 water flows we shall find that the softer materials have 
 yielded first and most completely. Hard dykes will be left 
 standing out, while softer rocks around them are worn away, 
 furrows will be eaten into more deeply, fissures will be 
 widened, clay slates will be wasted, while hard sandstone or 
 
 2 Id. 3;tt. 3 Agassiz, Geol. Sketches, '25-27. 
 
206 The Glacial Nightmare. 
 
 limestone and granite will show greater resistance. Not so 
 with surfaces over which the levelling plough of the glacier 
 has passed. Wherever softer and harder rocks alternate, 
 they are brought to one outline ; where dykes intersect softer 
 rock, they are cut to one level with it ; where rents or fissures 
 traverse the rock, they do not seem to have been widened or 
 scooped out more deeply, but their edges are simply abraded 
 on one line with the adjoining surfaces. Whatever be the 
 inequality in the hardness of the materials of which the rock 
 consists, even in the case of pudding stone, the surface is 
 abraded so evenly as to leave the impression that a rigid 
 rasp has moved over all the undulations of the land, ad- 
 vancing in one and the same direction and levelling all before 
 it." 4 
 
 Turning to the roches moutonnees (so-called, as we 
 have seen, by De Saussure, from their resemblance to the 
 rounded backs of sheep resting on the ground), Agassiz 
 urges that " they are the result of the prolonged abrasion of 
 rocks separated by deep indentations wide enough to be filled 
 up by large glaciers, overtopping the summits of the inter- 
 vening prominences, and passing over them like a river, or 
 like tide currents flowing over a submerged ledge of rock/' 
 He notices two circumstances specially striking about this 
 phenomenon. " The hummocks have frequently glacial marks 
 on one side only, the other being rough and unequal like a 
 hill slope not acted upon by ice. The polished side was clearly 
 the one turned towards the advancing glacier which the ice 
 pressed in its onward motion, while it passed over the other, 
 the lee side as we may call it, without coming in immediate 
 contact with it, bridging the depression and touching bottom 
 again a little further on. The lee side of the knolls often has 
 on it accumulations of the loose materials left by the glacier 
 in its progress. These effects only occur when the knolls are 
 sufficiently high and abrupt to allow a rigid body to bridge over 
 the space in passing from summit to summit. When the knolls 
 are low and slope gently downwards in every direction, they 
 present the characteristic glacier surfaces equally on all sides. 5 
 Another feature of the roches moutonnees is that the scratches 
 
 * Id. 40, 41. 5 Id. 41, 44> 
 
Agassis. 207 
 
 upon them ascend the slope in a direct line on one side while 
 they deviate to the right and left on the other side of the 
 knoll, more or less obliquely according to its slope. The 
 summits of the knolls are often again occupied by boulders 
 which would have been swept away by currents, but which, on 
 the supposition that the ice where they rested melted away, 
 would remain poised on prominent peaks or ledges." f 
 
 Agassiz forcibly contrasts the stratification of deposits from 
 water with the unstratified deposits from glaciers. The 
 debris of which the drift consists is thrown together pell 
 mell without any arrangement according to size or weight, 
 larger and smaller fragments being mixed so indiscriminately 
 that the heaviest materials may be on the very summit of the 
 mass, and the lightest at the bottom in immediate contact 
 with the underlying rock, or the larger pieces may stand at 
 any level in the mass of finer ones. Impalpable powder, 
 coarse sand, rounded, polished and scratched fragments of 
 every size, are mixed together in a homogeneous paste, in 
 which the larger materials are embedded, to use a homely but 
 expressive comparison, like raisins and currants in a pudding. 
 The adhesive paste holding all these fragments together is, 
 no doubt, the result of the friction to which the whole was 
 subjected under the glacier, and which has worked some of the 
 softer materials into a kind of cement. The mode of aggrega- 
 tion of water- worn materials is very different. Examine the 
 shingle along our beaches ; we find it so distributed as to 
 show that the fading tide-wave has carried the lighter materials 
 farther than the heavier ones, and the successive deposits 
 exhibit an imperfect cross-stratification, resulting from 
 changes in the height of the tide and the direction of the 
 wind. Moreover, in any materials collected under water we 
 find the heavier ones at the bottom, the lighter on the top. . . . 
 Another distinction between water-worn deposits and drift 
 consists in the fact that the former are washed clean, while 
 the latter always retain the mud gathered during its journey 
 and spread throughout its mass." 
 
 I do not propose to carry this part of my subject further, 
 nor, in fact, beyond applying the arguments to fresh localities, 
 
 6 Id. 44, 45. 7 Id. 51.54. 
 
208 The Glacial Nightmare. 
 
 the accumulation of minute observations, such as those em- 
 bodied in the works of Guyot, Desor, Maclaren, and many 
 others, or such as Hooker's extension of the theory to the 
 Lebanon and the Himalayas, Hector's and Haast's extension of 
 it to New Zealand, etc., etc., could it be carried much further 
 than it was carried by Charpentier and Agassiz. 
 
 It is plain to those who would look without prejudice that 
 the rounded and mammillated surfaces, the scratched, 
 polished and grooved rocks, and a great number of the 
 phenomena which accompanied the distribution of the boulders 
 and the drift, are consistent only with the fact that in the last 
 geological age there was an immense development of glaciers 
 which occupied not only the high ranges of the Alps and the 
 Dovrefelds, but the secondary ranges and lower heights 
 of the continent of Europe and North America. This con- 
 clusion seems supported by every form of converging evidence, 
 and is apparently beyond the reach of cavil. So far there is 
 no question at issue. 
 
209 
 
 CHAPTER VI. 
 
 THE GROWTH AND CULMINATION OF THE GLACIAL NIGHTMARE. 
 
 Part I. Europe. 
 
 " When the rivers that are called elivagar had flowed far from their 
 sources, "replied Har, " the venom which they rolled along, hardened, as 
 does dross that runs from a furnace, and became ice. When the rivers 
 flowed no longer, and the ice stood still, the vapour arising from the 
 venom gathered over it and froze to rime ; and in this manner were 
 formed in Ginningagap many layers of congealed vapour, piled one over 
 the other. That part of Grinningagap," added Jafnhar, " that lies towards 
 the north, was thus filled with heavy masses of gelid vapour and ice, 
 whilst everywhere within were whirlwinds and fleeting mists." " The 
 Prose Edda," Mallet's Translation, 401. 
 
 Bernhardi, 1832 Schimper, 1835-6 Agassiz, 1837-1843 Charpentier, 
 1842 C. Martins, 1846 R Chambers, 1850-52 Murchison, 1853 
 Yon Post, 1856 Kjerulf, 1876, etc. Erdmann, 1866 J. Geikie, 1873 
 Torell, 1875 Penck, 1875 Belgrand, Collomb, and Tardy, 1870 
 Eamsay, 1862 Jamieson, 1862-65 A. Geikie, 1863, Tidde- 
 man, 1872 J. Geikie, 1871 Croll, 1875. 
 
 THE diluvial theories of the earlier geologists had now been 
 largely abandoned and the struggle had lapsed into a fight 
 between the champions of land ice and icebergs a struggle 
 which still continues. 
 
 It was not long, however, before men, while accepting the 
 glacier theory of Charpentier or the iceberg theory of other 
 writers as explaining a great deal in the distribution of the 
 drift deposits, began to have serious doubts as to whether 
 they explained everything, and these doubts began very 
 early. 
 
 It was perhaps natural that students living in the plains of 
 Germany, far away from the sea and from any mountain ranges 
 which seemed capable of nursing large glaciers, should, while 
 appealing to ice to account for the phenomena they desired to 
 explain, have appealed to it in some other form than that of 
 
 p 
 
2 1 o The Glacial Nightmare. 
 
 icebergs or glaciers. There arose an evident necessity of en- 
 larging the induction which had been so ably made by Char- 
 pentier if the distribution of the erratics and drift over the 
 plains of Russia and Poland, many hundreds of miles away 
 from any mountain chain, the similar phenomena occurring in 
 America, the existence of striae and polished surfaces and great 
 boulders not only in the valleys but on the very crests of the 
 hills, etc., etc., were to be duly explained. 
 
 Hence there arose another form of appeal which took two 
 directions. On the one hand, a resort to continental ice moving 
 over the earth's surface in spite of gravity and by an inherent 
 quality of the ice itself ; and, secondly, the postulating of huge 
 ice caps about the two polar areas extending far into the temper 
 ate regions and even into the tropics. It is not quite easy to 
 separate the history of the two theories, and one great inquirer, 
 Agassiz, successively urged them both. He has been 
 treated as the inventor of the new views, but as a matter of 
 fact he derived his inspiration from others. The first person 
 known to me to advocate them was A. Bernhardi, professor at 
 the Forest Academy at Dreissigacker, who, writing in 1832, 
 modestly urges that he lived away from the centres of culture 
 and that the latest scientific publications were therefore not 
 accessible to him. He formulates his conclusion in these 
 words : " Once the polar ice reached as far as the southern limit 
 of the district which is still marked by the erratics. This ice 
 in the course of thousands of years shrank to its present pro- 
 portions, and these deposits of erratics must be identified with 
 the walls or mounds of rock fragments which are deposited by 
 all glaciers large and small, or in other words are nothing less 
 than the moraines which this vast sea of ice deposited in 
 its shrinkage and retreat." * This is assuredly a remarkable 
 generalization to have been made by a virtually unknown 
 student long before Agassiz published his theory. 
 
 A similar view had meanwhile occurred to another 
 German student, and as the story is of some interest to 
 others besides geologists, I shall enlarge somewhat upon it. 
 
 Carl Schimper, a botanist, and known as an authority upon 
 mosses, was born at Mannheim on February 15th, 1803. He 
 
 1 Leonhard and Brunn, Jahrbuch, 1832, p. 258, 259. 
 
Schimper. 2 1 1 
 
 was a precocious boy, and at Munich became the favourite 
 pupil of Schelling. In his various rambles in Bavaria he had 
 "been attracted, as many others have been, by the mysterious 
 boulders scattered over the country, which were a double 
 object of interest to him since they were covered by lichens 
 and mosses which were strange to him. According to the 
 testimony of Dr. Gustav. Bezold, he gave a series of lectures 
 in the winter of 1 835-36 at Munich to a society of old friends 
 there, including Schinzlein, afterwards professor at Erlangen, 
 Meyer the poet, Dobner, afterwards professor at Aschaffen- 
 burg, etc., etc. Of these lectures Dr. Bezold took notes. In 
 them Schimper distinctly urged that the blocks found in the 
 Isar and Wiirm valleys and on the Starnberg, " foundlings " 
 he called them, had been transported thither not by rivers or 
 water floods, but by ice, and he invoked a long period of cold 
 during which they were distributed. 
 
 In July, 1836, Schimper took part in a gathering of Swiss 
 naturalists at Solothurm, when he met Charpentier and Hugi, 
 who had both written about the Swiss glaciers. Agassiz was 
 also there, but he was busy with his work entitled " Poissons, 
 Echinodermes, et Mollusques fossiles," and had merely gone to 
 see Charpentier' s collections. Schimper stayed at Bex with 
 Charpentier from September to December, 1836. Agassiz, 
 who was there also, was busy with his shells, and although he 
 listened to the discussions of the two friends, he only once ac- 
 companied them on a glacier exploration to the Col de Balme 
 and the Trient Glaciers, which was conducted by Schimper. 
 During this time Schimper pressed upon Charpentier the fact 
 that the ice action to which he had appealed was in no sense 
 a res domestica, and limited in its operations to Switzerland, 
 but was, in fact, a widespread effect of a mantle of ice cover- 
 ing a much wider area. On the 19th of December, 1836, 
 Schimper discovered the famous glacier marks in the chalk of 
 the Jura, near Landeron. This fact he communicated to 
 Agassiz, whose imagination was at length fired by the impul- 
 sive and picturesque conversations of his friend, and he 
 determined for the first time to appear in the character of a 
 geologist, and advertised a course of lectures to be delivered 
 to his fellow-citizens of Neuchatel which were duly announced 
 in the Courier Neuchatelois of January 24th, 1837. In order 
 
 p 2 
 
212 The Glacial Nightmare. 
 
 to carry out his purpose more effectually he asked Schimper 
 to lend him the notes of the lectures the latter had given at 
 Munich the winter before. As these were locked up, Schimper 
 wrote to his friend and pupil Bezold to forward him his own 
 notes. 
 
 These were duly sent. Schimper acknowledged their 
 receipt in a letter dated the 5th of February, 1837, in which 
 were enclosed some verses on the lake of Neuchatel. In this 
 letter he again urged that the granite blocks had not been 
 transported by rivers, but by glaciers before the upheaval of 
 the Alps and the Bavarian mountains, and that, in the penul- 
 timate geological period, masses of ice several thousand feet 
 thick covered, not only Switzerland, but all Europe; and that 
 the then existing lakes were also filled with ice, whence it 
 was that the same lakes were not filled up with alluvion 
 and drift. He goes on to urge that the great blocks must 
 have originally been scattered on the ice, and been carried 
 along by it according to the physical laws which govern 
 ice, and that they were thus carried over lakes, valleys, and 
 mountains, and deposited on the high ridges bounding 
 them, etc., etc. 
 
 The lectures above referred to proceeded the next month, 
 but in the very first one, Agassiz 's knowledge of physics, 
 especially of the physics of ice, proved so slight, that 
 Schimper offered to help him in the subsequent ones ; 
 which offer was gladly accepted. Schimper, to commemorate 
 Galileo's birthday, issued a poem on the 15th of February, 
 1837, entitled "Die Eiszeit : fur Freunde gedruckt am 
 Geburtstage Galilei, 1837." It was signed Dr. K F. 
 Schimper, Neuchatel, February 15th, 1837, and was dis- 
 tributed by Agassiz among his auditors. This is the first 
 time the term " Ice Age " occurs, and Dr. Volger dates the 
 birth of the glacial theory from its publication ; so does 
 Charpentier. 2 
 
 In July, 1837, the Helvetic Society of Natural Sciences met 
 at Neuchatel, and Agassiz addressed to it his famous 
 " Discours preliminaire/' which is said to have startled his 
 hearers, and which has been generally accepted as the first 
 
 - .. ^ 
 - See Essai sur les Glaciers, 228. 
 
Schimper. 2 1 3 
 
 public pronouncement on the subject of the glacial theory. 
 In view of this meeting, Schimper, who was then at Carls- 
 ruhe, wrote him a letter, in which he asked "his brother 
 Agassiz " to bring, in his stead, before the meeting this vast 
 and important truth more quickly to the notice of the learned 
 world. In the letter he complains naively how his discovery 
 had already brought him some troubles, since it had aroused 
 the inveterate prejudices of the " desiccating - water " and the 
 " cooling-fire " men, and ran counter to the traditional notion 
 of a merely mechanically progressive diminution of the earth's 
 temperature, 3 and he concludes that the polished surfaces in 
 the Jura, and the erratic blocks, are the result of the action 
 of a great field of ice, " ein grosses Eisfeld." 4 He extended 
 his researches elsewhere than to the Jura, and tells us erratic 
 blocks are scarce in the Black Forest, but he had found some 
 of them, in 1826, near Lake Tili above Hollenthal, and he 
 had also found them about Carrigou, in the Pyrenees, in 
 1825. 5 
 
 This letter clearly shows who was the originator and who 
 the mere disseminator of the new views. Agassiz himself, in 
 his address, speaks of his theory as a fusion of his own views 
 with those of Schimper, and refers to an unpublished memoir 
 by his friend. Even this scanty acknowledgment, however, 
 was itself very shortlived, and when Schimper wrote to com- 
 plain that the press were attributing to Agassiz what had 
 been the work of his own brain, he replied that he did not 
 read the newspapers, and was not aware of their contents, but 
 that the printed report of the society's proceedings would put 
 everything right. When Agassiz published his well known 
 " Etudes sur les Glaciers," in 1840, it will hardly be credited 
 that he did not even refer to his friend, and the process of 
 ignoring him, in spite of protests, he carried out to the end. 
 
 Braun was a friend both of Agassiz and Schimper, and 
 cognisant of the facts. He was too timid to interfere, how- 
 ever, although a letter is extant which he wrote to Professor 
 Roper, at Rostock, dated February 22nd, 1840, in which he says, 
 speaking of Schimper, " Agassiz und Charpentier, die in 
 
 3 Id. 5, see also Actes de la Societe* Helvetique, 1837-38, etc. 
 4 Id. 46. * Id. 49, 50. 
 
214 The Glacial Nightmare. 
 
 dieser Sache jetzt das Meiste thun, sind, wenigstens was die 
 allgemeine Ausdenhung der Ansicliten und die tief eren physio- 
 logischen Gedanken dabei betriift, seine Schiller." Agassiz 
 himself, in a letter to Braun, accompanying a copy of the 
 work last cited, and subsequently included in Braun' s " Life/' 
 published by his daughter,, writes the following extraordinary 
 sentence: "Du darfst Dich nicht wundern, den Namen von 
 Schimper darin nirgends genannt zu finden. Ich wollte so 
 seine Anmassung strafen. Alles, was er hatte sein nennen 
 konnen, wenn auch nur von ferae, habe ich unberuhrt 
 gelassen, selbst wenn ich beistimmen miisste." Mr. G. P. 
 Evans says of this, "Wherein consisted this presumption 
 which Agassiz wished to punish by a policy of utterly ignoring 
 the achievements of a colleague, in a manner which, in the 
 interests of true learning and to the honour of human nature, 
 one would gladly think is rare in the annals of scientific 
 research ? Merely in the modest expression of a desire to 
 have his name publicly mentioned in connection with a theory, 
 of which, as is now clearly shown, he was the real and only 
 author." 6 Well may Mr. G. Evans say, " Whatever glory 
 emanated from the new doctrine haloed round the brow of 
 Agassiz as its public expounder, and naturally enough he 
 soon grew fond of the easily-won fame. The nimbus of the 
 saint is a covetable head-gear, provided one is not compelled 
 to win it by the thorny crown of martyrdom." 7 
 
 The result has been what might be expected. Schimper' s 
 name has almost dropped out of geological literature, and 
 is not even referred to in such works as Professor James 
 Geikie's " Ice Age " and " Prehistoric Europe," whose main 
 thesis is that first maintained by the German botanist. 
 
 While it is clear that Agassiz has got credit for inventing a 
 theory which he only appropriated from Schimper, there can be 
 small doubt that it was the persistent, picturesque and seductive 
 preaching of the new doctrine by one so famous as a naturalist 
 as himself that made it acceptable in many quarters where scien- 
 tific idolatry is maintained chiefly by the size of the idol, and he 
 certainly did not shrink either from disseminating his views or 
 pressing them to their logical conclusion. He first had a polemic 
 
 6 G. P. Evans, North American Keview, cxlv. 96, 97. * Id. 96. 
 
Agassiz. 2 1 5 
 
 with Charpentier, against whom he argues that the remains 
 could not have been those of the glaciers in the present Swiss 
 valleys, however large we choose to make them, but a vast ice 
 sheet covering Switzerland and reaching higher than the 
 summits of the Jura. This is proved, he says, by the presence 
 of old moraines 2500 metres above the sea level on the borders 
 of the lake of Geneva ; and he urges that since the polished 
 surfaces have been found on the summit of " the Pilgrim," 
 which dominates Yevey, and is 3301 feet above the sea level, 
 the bottom of the ice sheet must Lave been at least at 
 this height. He contests Charpeiitier's notion that the 
 glacier ice from the Alps crept along the valleys and mounted 
 the Jura to its summit, and urges that this is disproved by the 
 fact that the blocks on the Jura are generally both larger 
 and less rounded than those in the moraines which are being- 
 deposited at this moment, from which it follows that they 
 cannot have been subjected to the very long process of 
 trituration involved in travelling up and down the inequalities 
 of the intervening land on a glacier since they would have 
 been ground into a smaller size. This observation,, however, 
 about the relative size of the boulders on the Alps and Jura, 
 seems at issue with that of M. Elie de Beaumont made in the 
 valley of the Durance. 8 
 
 Agassiz urged that the whole space between the opposite 
 chains of the Alps and the Jura was once filled with ice ; that 
 this mass of ice completely covered the Jura, with the excep- 
 tion of a few high crests, rising island-like above it, and 
 mounted to a height of some nine thousand feet upon the 
 Alps, while it extended on the one side into the northern 
 plain of Italy, filling all its depressions, and on the other down 
 to the plains of Central Europe. 9 
 
 He further argues that while Charpentier's theory may ex- 
 plain the portage of the angular blocks, it cannot explain the 
 other facts without having recourse to currents, and as against 
 currents he quotes the view of Schimper, that in traversing 
 the valleys they must have filled up the lakes with debris and 
 have worn down their angular and steep banks, and however 
 
 Actes de la Societe Helvetique, 1837, vii.-xviii. 
 G-eol. Sketches, 28. 
 
2io The Glacial Nightmare. 
 
 violent and powerful the currents, they must have eventually 
 relaxed and left in their last efforts portions of their burdens 
 behind, while very few blocks indeed are found between the 
 Alps and the Jura. How, again, was it that the softer and 
 more intangible debris, such as the sand, etc., which such 
 currents would have carried just as much as the blocks, did 
 not fill up the inequalities of the valleys ? Sand does occur in 
 small quantities 011 the polished surfaces of the Jura, so does 
 gravel, and it is generally on a quantity of one or the other, 
 from a few inches to a few feet thick, that the blocks are 
 found, the sand being below the gravel and the latter 
 topped by the blocks. Sand would have been swept off these 
 polished surfaces by a violent rush of water, which again would 
 have deposited the big boulders at the bottom and overlain 
 them with the finer debris, and not deposited them in the 
 reverse order ; * and he concludes that the existence of the 
 polished surfaces, the fact that these are immediately covered 
 with gravel and sand, and that on these again rest the angular 
 blocks, is inconsistent with every explanation forthcoming 
 about the blocks, previous to his own. 2 
 
 The dispersal of the mantle of gravel which is found so 
 much developed in Switzerland, and of the loess which he 
 attributes to the denudation of the tertiary rocks there, he 
 accounts for by invoking debacles, due to the melting of the ice 
 sheet, which may also have borne along with them ice-rafts 
 loaded with stones. 3 His general conclusion is formulated in the 
 following sentence : " Ce serait done une grave erreur de con- 
 fondre les glaciers qui descendent du sommet des Alpes, avec 
 les phenomenes de 1'epoque des grandes glaces qui ont precede 
 leur existence." 4 
 
 In 1833 M. Godet made an excursion to Sweden, and in 
 1837 communicated to the Helvetic Society an account of his 
 observation of the erratic blocks of Scania, upon which 
 M. Agassiz made some remarks, attributing them to a similar 
 ice sheet to that he had invoked in Switzerland. 5 
 
 In his famous memoir, entitled " Notice sur les Glaciers," 
 published originally in 1839, he says : " Je conclus de 1'ensemble 
 
 1 Id. xix.-xxii. * Id. xxii. * Id. 
 
 4 Id. xxix. * Id. 122. 
 
Agassiz. 2 1 7 
 
 de ces phenonaenes et des stries analogues observees en Suede 
 par M. Sefstroem, qu'a une certaine epoque PEurope entiere 
 s'est couverte de glace : que cette epoque est celle de la dis- 
 parition des grands mammiferes que Fon trouve deposes dans 
 les graviers glaces du Nord." 6 
 
 Such was the beginning of what I choose to call the Glacial 
 Nightmare. A far more experienced geologist than I can 
 ever claim to be has said of this famous and epoch-making 
 memoir : " Dans ce premier developpement trop rapide de 
 1'hypothese, le celebre zoologiste depassa le but, e.t ce fut a la 
 difficulte de le restreindre au domaine de 1'observation 
 rigoureuse des faits comme des inductions rationnelles que 
 1'on doit une grande partie des recherches auxquelles on s'est 
 livre depuis/" 7 
 
 In his paper on glaciers, read before the Geological Society 
 of London on November 4th, 1840, Agassiz says that if the 
 analogy of the facts which he had observed in Scotland, 
 Ireland, and the north of England with those of Switzerland be 
 correct, it must be admitted, that not only glaciers once 
 existed in the British Islands, but that large sheets (nappes) 
 of ice covered all the surface. He again urged that glaciers did 
 not descend from the mountain summits into the plains, but that 
 they were the remaining portions of the sheets of ice which at 
 one time covered the flat country. " It is evident/' he says, 
 "that if the glaciers descended from the high mountains and ex- 
 tended forward into the plains, the largest moraines ought to 
 be the most distant, and to be formed of the most rounded 
 masses, whereas the actual condition of the detrital accumula- 
 tions is the reverse, the distant materials being widely spread, 
 and true moraines being found only in valleys connected with 
 great chains of lofty mountains. It must then be admitted 
 that great sheets of ice resembling those now existing in 
 Greenland once covered all the countries in which unstratified 
 gravel is found." 8 
 
 Elsewhere he says : " Scandinavian boulders, scattered upon 
 English soil and over the plains of Northern Germany, tell us 
 that not only the Baltic Sea, but the German Ocean also, was 
 
 6 Op. cit. 15, 16. 
 
 7 D'Archiac Histoire des Progres, etc., ii, 240. 
 
 8 Proc. Geol. Soc. iii. 331. 
 
218 The Glacial Nightmare. 
 
 bridged across by ice on which these masses of rock were 
 transported. In short, over the whole of Northern Europe, 
 from the Arctic Ocean to the northern borders of its southern 
 promontories, we find all the usual indications of glacial action, 
 showing that a continuous sheet of ice once spread over nearly 
 the whole continent, while from all the mountain ranges 
 descended those more limited glacial tracks terminating fre- 
 quently in tranverse moraines across the valleys, showing that, 
 as the general ice sheet broke up and contracted into local 
 glaciers, every cluster or chain of hills became a centre of 
 glacial dispersion, such as the Alps are now, such as the Jura, 
 the Highlands of Scotland, the mountains of Wales and Ireland, 
 the Alps of Scandinavia, the Hartz, the Black Forest, the 
 Vosges, and many others have been in ancient times." 9 
 
 It will be noted that Agassiz's conclusions as here formu- 
 lated, although they were supported by a wider generalization, 
 did not carry the theory a step further than it had been 
 carried by Bernnardi and S chimp er. The memoir last quoted 
 was nevertheless a great landmark in the development of the 
 Glacial Nightmare, which at this stage of its history involved 
 an appeal to an ice sheet co-extensive with the distribution of 
 the drift deposits. 
 
 The theory thus formulated still largely maintains its hold 
 upon the scientific world, and soon attracted adherents who 
 wrote with weight and authority. Among these none was 
 more remarkable than Charpentier himself. 
 
 In the year 1842 he communicated to the Bibliotheque 
 Universelle of Geneva a paper in which he examined the 
 erratic phenomena of Scandinavia and the north of Europe- 
 In this paper he contests the notion that mountains are neces- 
 sary to the existence of glaciers, except that they sometimes 
 favour the accumulation of snow by the wind. " It is only,' 7 
 he says, " their cold snowy and rainy climate which causes the 
 formation, development and movement of glaciers, and if these 
 conditions existed in a flat country, even at the level of the sea, 
 there would be nothing to prevent glaciers being formed and 
 developed." He also urges that, according to his theory of 
 dilatation, glaciers can and do move over level plains. In 
 
 9 Geol. Sketches, ser. ii. 76. 
 
Charpentier. 219 
 
 order to account for the former large development of glaciers, 
 he postulates that the winters were more snowy and the 
 summers more rainy and colder, with a smaller difference in the 
 mean temperature of summer and winter than exists at present; 
 such a climate now exists in Tierra del Fuego and the Straits 
 of Magellan. He further urges that the ice of glaciers is only 
 partially derived from the neves whence they spring, a large 
 part of it accumulating en route, and he urges that there is 
 therefore " no need to suppose that all the ice of the diluvian 
 glacier of the north came from one single point. On the con- 
 trary, that glacier would be constantly increased by the rain 
 and the snow falling directly upon it, and its increase must 
 have gone on augmenting in proportion as it acquired a larger 
 surface." 
 
 Charpentier does not shrink, therefore, from the conclusion 
 that a large glacier once occupied Finmark, Lapland, Norway, 
 and the greater part of Sweden and Finland; and he adds, 
 " as it crossed the Baltic, and extended to the north of Ger- 
 many, Prussia, and the plains of Russia as far as Moscow, and 
 perhaps the east of England, there is nothing extraordinary in 
 supposing that the erratic formation really reaches to Moscow, 
 Stezyka, Oppeln, Leipzig, etc/' 
 
 He says the phenomena of Scandinavia are like those of the 
 Alps. The widely scattered debris of rocks are sometimes 
 spread out, sometimes collected in bands and mounds, fragments 
 of all sizes being found mixed together pell-mell, many having 
 their prominent surfaces well preserved, while the rocks, as in 
 the Alps, have marks of wear and rubbing, smooth surfaces, 
 striae, furrows, and vertical erosions in the form of cauldrons. 
 
 In order to account for the true diluvium, that is, the deposits 
 containing sea- shells and traces of stratification attributable 
 to the action of water, he postulates the former great extension 
 of the Baltic and the existence of icebergs supplementing the 
 ice sheet. 
 
 Answering Durocher, he refuses to believe that the marks of 
 rubbing and of wearing on rocks is due to a great current from 
 the north, which must have risen on the coast of Finmark to 
 the height of 2500 feet, since they occur on the summit of 
 Raipas and on the high plateau of Norwegian Lapland, and since 
 water by itself cannot polish and scoop out rocks, it must have 
 
22O The Glacial Nightmare. 
 
 been a flood of water charged with other materials. He urges 
 that such a torrent would have filled up the Baltic and other 
 hollows which it, traversed and would not have allowed the 
 quiet or stratified deposit of the true diluvium. He argues 
 that a large proportion of the erratic debris are not the 
 remains of what the glacier carried on its back, but that 
 glaciers have the power of raising such fragments from their 
 bed to their surface, a view to which we shall presently revert. 
 He says further that the external shape of the asar is much 
 more consistent with ice action than with that of a current or 
 with floating ice, and so with the fine striae occurring on the 
 surfaces of hard rocks and also the giants' cauldrons. 
 
 He argues that the moraines, etc., are never permanent in 
 a glacier which is growing, since it must overthrow and 
 destroy them in its progress until it reaches its maximum 
 extension where a terminal moraine would exist, if not swept 
 away by torrents or buried under true diluvium. Every 
 oscillation of a glacier gives rise to new accumulation of debris. 
 He finds the remains of such frontal moraines in the so-called 
 Steindamme stretching E, and W. in Prussia. As the ice-sheet 
 retreated beyond the Baltic and shrank into its original limits, 
 it would be lacerated or broken up into fragments forming 
 separate glaciers which would each deposit moraines. These 
 he finds in the existing asars. Thus also he accounts for the 
 striae sometimes crossing each other. (( The localities where 
 the striae cross have been covered at two different times by the 
 ice. The first time they have been invaded by the great 
 ice- sheet, which has scratched them in the general directionfrom 
 north to south ; and the second time they have been so by 
 partial glaciers, whose action has there produced striae, in 
 some degree anomalous, which cross the first in various 
 directions." 
 
 When two partial glaciers united, they would create 
 a superficial moraine, such as one of the asar, whose upper 
 extremity rests against the rock or eminence forming the 
 termination of the chain of heights, which, by separating 
 the two glaciers, has given rise to two deposits of the super- 
 ficial moraine. That the glacier moved from north to south 
 is shown by the direction of the striae, etc., whose plight 
 deviations were caused by the slope and inequalities of the 
 
Charpentier and Agassis. 22 i 
 
 surface. This direction is also shown by the phenomena of 
 the so-called stoss side and lee side. "In the Alps when a 
 glacier encounters a rock or an eminence in its passage, the 
 flank turned towards the side whence the glacier proceeds is 
 always more rounded and rubbed than that turned towards 
 the opposite direction." He similarly argues that the fine 
 strise on the rocks, noticed by M. Kobert, which are 
 generally parallel, run in the direction of the valleys and 
 cross all the strata indiscriminately, without regard either to 
 their direction or hardness, and show that they have been traced 
 by hard bodies moving in concert at fixed distances, that 
 is to say, attached to the same solid body, which prevented 
 them yielding to any obstacles they might encounter; 
 and he further argues that the polish on the surfaces left by 
 the ice in the northern regions being better preserved than in 
 other climates is due to the ice having begun to melt there at 
 a later period, and therefore to the weathering having been 
 less. 
 
 He urges further that while the advancing ice-sheet would on 
 uneven ground, where its progress was impeded, upturn the 
 soil, and pierce to the solid rock ; where it could extend freely 
 it must have stretched over " the true diluvium " without 
 displacing it. In regard to the reason why the great 
 northern ice-sheet moved southward, he argues that the Polar 
 region, from lat. 60 to the Pole, was covered with snow. 
 Between 60 and 70, this snow was converted into a glacier, 
 which in its dilatation could not extend in any other but a 
 southern direction, because in other directions it had to 
 encounter the resistance arising from snow and ice them- 
 selves. 1 
 
 In a memoir entitled " A Period in the History of our 
 Planet," published .by Agassiz in the Edinburgh Phil. Journ. 
 for 1843, we find him further developing his views. Thus he 
 says : " The earth had already assumed its present contour, 
 with the exception, perhaps, of the principal range of the 
 Alpine chain, and of the mountains which rose simultaneously 
 with it. ... Then numbness seized the light f sailors of the 
 atmosphere,' the clouds and vapours ; icy winds drove them 
 
 Bibl. Univ. xxxix. 327-346. ; Ed. New Phil. Journ. xxxiv. 55-74. 
 
222 The Glacial Nightmare. 
 
 in a solid form to the earth, and like a huge winding-sheet, they 
 enveloped the polar regions, the north of Europe and of Asia. 
 The British Islands, Sweden, Norway and Kussia, Germany 
 and France, the mountainous regions of the Tyrol and 
 Switzerland, down to the happy fields of Italy, together with 
 the continent of northern Asia, formed undoubtedly but one 
 ice-field, whose southern limits investigation has not yet 
 determined. And as on the eastern hemisphere, so also on 
 the western, over the wide continent of North America, there 
 extended a similar plain of ice, the boundaries of which are 
 in like manner still unascertained. The polar ice which at 
 the present day covers the miserable regions of Spitzbergen, 
 Greenland and Siberia, extended far into the temperate zones 
 of both hemispheres, leaving probably but a broader or 
 narrower belt around the equator, upon which there were 
 constantly developed aqueous vapours, which again condensed 
 at the poles ; nay, if Tschudi's observations in the Cordilleras 
 and Newbold ; s at Seringapatam shall be confirmed, and to 
 these we may subjoin those made by earlier travellers upon 
 the Atlas and Lebanon chains, the whole surface of the earth 
 was, according to all probability, for a time one uninterrupted 
 surface of ice, from which projected only the highest 
 mountain ridges covered with eternal snow. The limits 
 which would seem to be indicated by the various phenomena 
 which we shall afterwards treat of, are very probably refer- 
 able to a subsequent epoch, when the universal ice crust had 
 already begun here and there to disappear, and particular 
 tracts to emerge like oases in the immense icy desert." 
 Having gone on to postulate that the present North Sea was 
 filled to the bottom with solid ice, and that it reached to at 
 least 8000 feet above the sea level on the Alps, he goes on 
 to say that, " where there prevailed such an uncommonly low 
 temperature as was necessary to envelop the earth to such 
 an extent in a frozen covering, there would be little or no 
 moisture and little or no melting and freezing, so that, in the 
 temperate zones at least, even at the sea level there existed 
 only snow or neve but no ice, the temperature standing 
 almost constantly below 32 Fahr." Hence, he urges, "there 
 could be no moving of ice-fields in any direction during^ that 
 period. . . . During the glacial period there was no motion ; 
 
Agassiz. 223 
 
 not a brook, nor a rill furrowed the surface of the snowy 
 covering, to remind by its purling that all life had not yet 
 become torpid . . . According to our present state of 
 knowledge, there could be no such thing as motion of the 
 glacier ice, until the warmth of the surrounding media had 
 become such as to cause the melting of the superficial 
 strata. . . . Motion demands warmth as the medium and 
 condition of its existence. It was not, therefore, until a 
 return of warmth and the ice began to soften that all those 
 grand phenomena could take place of which the surface of 
 our present earth is witness. What, therefore, the majority of 
 geologists have hitherto believed to be the result of immense 
 floods and currents, and a few to have been caused by the 
 increase and the progressive motion of glaciers, namely the 
 transportation of erratic blocks, of alluvial boulders and the 
 polishing and channelling of rocks these are to me mani- 
 festations of the retreat of the glacial period, phenomena 
 which denote the moment when an alteration in the climate 
 of our earth began to confine the cold within those narrow 
 boundaries which it now occupies phenomena which denote 
 the places whence torpid winter began her slow retreat 
 towards the rocks of our lofty mountains, and to the ice- 
 bound continents of the north. ... It was the plains which 
 first freed themselves from their torpid covering. . . . The 
 snowy covering first began to dissolve, and by the trickling 
 in of the water, and its recongelation in the interstices of the 
 looser snowy strata, to change into more compact ice. With 
 this transmutation of snow into ice began the movement of 
 the masses of ice in the direction of the superficial slope, and 
 their effects upon the solid ground on which they rested. . . . 
 clefts opened under the expansive force of the solar heat, and 
 thus furnished the waters which the melting of the super- 
 ficial strata called forth in copious abundance. . . . The 
 waters thus formed dug their mutable beds out of the 
 extensive surface . . . undermined deep beds in the rolled 
 matters, and thus produced valleys of denudation the 
 direction of which is frequently inconceivable with reference 
 to the present configuration of the surface. . . . Thus began 
 the retreat. The centres towards which its course was 
 directed were, on the one hand, the far north ; on the other 
 
224 The Glacial Nightmare. 
 
 hand, the lofty mountain ranges of Central Europe, and from 
 these regions descended the rude witnesses of the extension of 
 the icy covering, those huge blocks which lie scattered in the 
 plains of Germany, etc. . . . Torn from the ridges of the 
 Scandinavian chain, these erratic blocks were moved forward 
 on the surface of the ice coverings over the Baltic Sea, whose 
 basin was filled with ice instead of salt water, and deposited on 
 the plains of Northern Germany on the skirts of the ice-field." 
 So also, argues Agassiz, was it the case with the Alps, " from 
 which radiated the erratic blocks skirting them all round. . . . 
 More and more was the ice mass compelled to recoil, more and 
 more was the land denuded to become covered with verdure. 
 The plains of Germany, Eussia, France and Italy, were 
 rescued from their frozen state, the Baltic was set free ; the 
 North Sea, for the most part, was anew covered with waves. 
 But the more considerable chains of mountains, presenting 
 by their elevation above the sea level a more secure resting- 
 place to the ice-covering, still retained the eternal winter 
 upon their summits and in their valleys, and thus there arose 
 detached glacier ranges, which no longer formed a connected 
 whole, like their common ancestress, but separate groups, each 
 of which belonged to a particular mountain chain, etc. " 
 
 Charles Martins, writing in 1846, also championed the cause 
 of a continental ice in Scandinavia. Arguing against Durocher. 
 he urged that a glacier moves on a much smaller slope than any 
 solid body, that of the Aar glacier, for instance, moving on a 
 slope of 1*30 only. He also argues that a glacier can move 
 along an undulating surface, and cites the glaciers of Alalein, 
 Tschingel and Griinberg, while among ancient glaciers the 
 roches moutonnees of the valley of the Aar, such as those of 
 the two Baerenbuhls, those forming the contrefort of 
 Naeglisgraelli, the Spitalnollen and the Kirchet, which are 
 covered with undoubted glacial striae. In regard to certain 
 instances, such as that of Areskutan, cited by Durocher, where 
 the glacier must have mounted a steep of 222 metres, he 
 urges that it is as difficult to see how water could have 
 mounted this as a glacier, and he argues that the levels 
 of the country have altered since the glacial age. Again, 
 
 * Ed. New Phil. Journ. xxxv. 17-29. 
 
R. Chambers. 225 
 
 answering Durocher, who doubted the capacity of ice to 
 traverse a depression like the Gulf of Bothnia, he says that 
 he found a greater difficulty in postulating the same move- 
 ment for a current of water. As a matter of fact, the Swiss 
 glaciers cross depressions and even lakes over which they pass 
 because the specific gravity of ice is less than that of water. 
 Thus the Aletsch glacier crosses the lake of Moirel and the 
 Spitzbergen glaciers bridge over the sea for some distance in 
 the same way, and he concludes that it is not at all improbable 
 that the Swedish glaciers crossed the Gulf of Bothnia as they 
 crossed the Swedish lakes. We can prove, he says, by the 
 lateral moraines left by these northern glaciers that their base 
 did not rest on the floor of the lake, but corresponded to the 
 level of the water in it. 3 
 
 In the year 1850 Mr. Robert Chambers read a paper before 
 the British Association in the abstract of which he asks some 
 pointed questions : " How could ice," he says, " move over 
 so large a portion of the North American continent, in a 
 direction admitted to be tolerably uniform, allowing for slight 
 deviations easily explicable as owing to irregularities in the 
 original surface, and this without any mountain chain to give 
 it forth ? (2) How was this ice capable of ascending slopes 
 and topping mountains of considerable height. (3) How, in 
 such a valley as that of the Forth, could there be an ice- 
 torrent of undeviating flow for many miles and deep enough 
 to envelop hills many hundred feet high." 4 
 
 In 1852 the same acute observer read a remarkable memoir 
 before the Royal Society of Edinburgh on glacial phenomena 
 in Scotland and parts of England, which has been too little 
 noticed. In this he begins by urging, against the diluvial 
 theory that he had seen what he considered an ice-polished 
 surface crossed by a small runnel of water carrying minute 
 gravel, and it was clearly observable where the water crossed, 
 that the surface \vas changed, became rougher and dimmer, 
 something like the difference between chased and polished 
 goldsmith's work, and that he had examined the beds of many 
 mountain streams accustomed to bring down all sizes of 
 
 3 Bull. Soc. Geol. de France, iv. pt. i. pp. 99-102. 
 
 4 Report, p. 79. 
 
226 The Glacial Nightmare. 
 
 gravel, but never found in any instance those flowing outlines 
 of abrasion which we see in the so-called glacial surfaces. 
 There is also a palpable enough difference between the 
 confused masses of mud and sand mingled with rounded 
 blocks, which are found in connection with polished surfaces, 
 and the sorted materials, gravel, sand, and clay, which are 
 indisputably attributed to watery action. He similarly argues 
 that the wide extent of abraded surface in Norway and 
 Sweden and the uniformity of striation over such wide areas 
 is inconsistent with the light, partial and irregular action of 
 floating masses of ice. Passing on, he accepts the evidence for 
 the former existence of local glaciers in mountain clusters now 
 free from them as overwhelming, and especially commends the 
 researches of Maclaren and C. Martins on the subject. He 
 goes on to urge, however, that local glaciers will not explain 
 all the facts. How Maclaren himself had shown that the 
 eminence between Loch Long and the Gairloch, 600 feet high, 
 is as perfectly smoothed along the top as is the bottom of 
 either of the two valleys. How Professor Fleming had found 
 strias identical in direction with those of the plains on the 
 summit of the Pentlands, 1400 feet high. How Maclaren, 
 again, had shown that the striae which cross the valley of 
 Westwater, 800 to 900 feet above the sea, at the western 
 extremity of the Pentlands, and which runs north and 
 south, cross it from west to east, thus persevering in the 
 normal direction in circumstances where, if anywhere, a 
 divergence was to be expected ; and he asks how any group of 
 Highland mountains, 3000 to 3500 feet high, could have sent 
 out glaciers reaching Mid Lothian, 70 miles off, in such volume 
 and depth as to envelop a range of hills to the depth of 1400 
 feet, and in such unyielding force as to cross a minor 
 valley 800 or 900 feet above the sea without diverging in the 
 least from its course, and he concludes that these results are 
 not consistent with such a cause. 
 
 He then proceeds, in great detail and in clear language, to 
 describe how very general the marks of abrasion are in Scotland. 
 " I have, since August, 1850, found them/' he says, " along 
 the whole range of the coast north of Arg-yleshire, namely 
 in Inverness-shire, and the counties of Koss, Sutherland, and 
 Caithness. I have found them in the Isle of Sl^ye, in 
 
R. Chambers. 227 
 
 situations independent of the Cucliullin hills ; in the island of 
 Mull ; all along Loch Lomond, and Loch Katrine ; even the 
 picturesque eminences which constitute the Trosachs being 
 roches moutonnees with abraded faces to the west. I have 
 likewise found them in Perthshire, Fife, and Aberdeenshire. 
 They are reported from Ayrshire. I have found a large 
 lateral moraine near Maxwellton House, in Kircudbright, and 
 seen fine smoothings with striation on the surface of 
 Corncockle Muir in Dumfriesshire." And he urges that we 
 should find traces of ice everywhere except on the highest 
 summits, if all the rock surfaces were exposed, and if all those 
 actually exposed had been equally capable of retaining the 
 impressions made by ice. These facts, and especially the 
 direction of the striae, impelled him to postulate two indepen- 
 dent stages of glaciation. tf First, a general sweeping of the 
 surface of the district by a deep flow of mobile ice, one great 
 cause, if not the principal, of that enormous denudation which 
 has been referred to, but of which the spoils, from the 
 universality and power of the agent, were in a great measure 
 swept away. Secondly, local and certainly subaerial glaciers 
 occupying certain valleys in the higher mountains and 
 producing moraines, composed of brown clay, sand and 
 blocks." And he quotes several instances as settling the 
 question of two systems and epochs of glaciation, where 
 the strong normal streaks athwart the hill from the north- 
 west, a direction in which no local or limited mass of ice 
 could move, are clearly chequered with fainter streaks, 
 produced by simple down-hill movement, which happen to 
 be from W.S.W. ; and he points out further that this 
 second glaciation has in certain cases obliterated the traces 
 of the earlier one. And he especially quotes, as utterly 
 destructive of the idea of exclusively local glaciers, the 
 striation ascending obliquely out of Loch Katrine at 
 Stronachlachar, passing over a high jutting hill promontory, 
 reappearing under compact clay, in low ground, at some 
 distance from the loch, and everywhere maintaining pre- 
 cisely one direction and that from N.N.W. To all appear- 
 ance, the agent which produced these impressions came over a 
 lofty range of hills from Balquidder and passed on to cross a 
 scarcely lower range and descend into the valley of Loch Ard. 
 
 Q 2 
 
228 The Glacial Nightmare. 
 
 Turning to Scandinavia, lie mentions how Bohtlinck had 
 noticed that the marks of striation in the intermediate valleys 
 are at an angle of 50 degrees with those on the intervening 
 heights. He himself had observed that on a summit of 4000 
 feet of elevation at Jerkind and in the immediate neighbour- 
 hood of Sneehatte, the southern slope is abraded and polished 
 almost to the top with strige between north-east and south- 
 west, a line totally irrespective of all the great mountains of 
 the district,, such as Sneehatte, which it sweeps laterally 
 There is no higher ground from which the agent could 
 descend to this spot, and it must have crossed the chain again 
 on the col over which the road passes from Lavanger to 
 Sundsvall in crossing the frontier between Sweden and 
 Norway. " The whole of the space composed of rocks- of 
 chlorite schist, is abraded by an agent which has been able to 
 shear sharply through the upturned edges of the strata, 
 leaving clear striae to mark its course. Surprising to say, 
 that course has not been across from west to east nor from 
 north to south in the line of the little ridges, but it is 
 athwart both of these lines of hollow from north-west to 
 north-east, and thus clearly has been independent of the 
 form of the country." 
 
 Chambers explains the immense mass of detrital deposits 
 which exists 011 the lee side of the Dovrefelds on descending 
 from the bare cols, compared with the paucity of similar remains 
 in Sweden, by the chain having been involved in a flow of ice 
 which pressed hard upon and swept bare the hither side, but 
 passed comparatively lightly over what lay beyond, leaving 
 there little of the solid matter with which it was charged. 
 
 Again, after referring to the polished surfaces on the shores 
 of the Gulf of Bothnia (in Finland) at Stockholm, and at 
 Gothenburg, he says : " We are many hundreds of miles from 
 the presumed centre of action, and the intervening space 
 presents an infinite number of minor obstructions, all of which, 
 however, have been swept over by the agent, whatever it was. 
 How could any such agent, descending from hills only half as 
 high as the Swiss Alps, travel over twenty times the space, in 
 a condition, too, necessarily attenuated through the wideness of 
 the country over which they must have spread ? " 
 
 Turning to America he says : " There are proofs all over 
 
Murckison and Von Post. 229 
 
 Canada and to a point far south in the United States, as well 
 as around Lakes Huron and Superior, of an abrading agent- 
 for the most part from the north-west. Mountains of 2000 feet 
 in height bear on their sides and tops striation in that direction, 
 while to the north-westward no mountains of greater elevation 
 to serve as gathering places for glaciers can be pointed out." 
 
 Chambers finally concludes that the more general glaciation 
 to which he refers points to a wide extension of the circum- 
 polar ice and a southern movement of that envelope, in the 
 course of which the surface was abraded and the detritus 
 produced/' 5 
 
 In 1853 Murchison, who was such a staunch supporter of 
 other views, was momentarily induced to countenance the 
 theory of ice-sheets. He says : ' ( There are many lofty tracts 
 in Scotland, as well as in Norway and other countries, where 
 the striation seems to be quite independent of the outline of 
 the ground, thus indicating* a grand and general movement of 
 ice. It is to countries which present such phenomena, that the 
 memoir of Dr. Rink forcibly applies ; and it leads us to 
 imagine, that there was a period when Scotland, particularly 
 all the Highlands, was analogous to what Greenland is now, 
 and when an icy mantle extended itself from higher plateaux 
 into the fiords or friths on its sides." G 
 
 Meanwhile the general adoption of Forbes' conclusions in 
 regard to the motion of ice apparently checked the spread of 
 views about continental ice which were necessarily based on its 
 motion being independent of gravity. It was apparently the 
 researches of Rink in Greenland, and the papers he wrote on 
 its great ice-fields, that caused a reversion to the ice-sheet 
 theory. 
 
 Already in 1856 Hampus Yon Post argued that local 
 glaciers were not adequate to explain the phenomena of the 
 Swedish surface beds. 7 He was followed a few years later by 
 Kjerulf, facile princeps in his knowledge of these beds in 
 Norway. 
 
 Kjerulf had been much struck by Dr. Rink's reports about 
 the condition of Greenland, as affording an explanation of the 
 
 5 Ed. Phil. Journal, liv. 229-281. 
 
 6 Murchison, Address to the Geographical Soc., May 1853, 29. 
 Penck, Vergletschernng der Deutschen Alpen, 15, 16. 
 
230 The Glacial Nightmare. 
 
 glacial phenomena of the north. "From his description/' he 
 says, " it is obvious that here, as in our own land, we have an 
 enormous stretch of country ; that this country is completely 
 covered with ice, which, on its west coast, is ever slowly 
 pressing outwards to the sea, where it ' calves '; that immense 
 loads of these ice ' calves ; are every year floated away by 
 currents in the same constant direction ; and finally that this 
 ice-cake attains a thickness of no less than ] 000 feet. 
 
 f ' Here all the conditions of our problem are provided for. 
 Let us figure to ourselves a universal ice-covering, a complete 
 glaciatioii instead of mere individual glaciers, and the 
 phenomena of friction takes its place harmoniously with all 
 the other phenomena of the glacial epoch. In the existing 
 state of Greenland is the very analogy we needed to justify our 
 supposition ; a gigantic cake of ice slowly moving outwards, 
 and with the might of its tremendous pressure stripping, 
 polishing and striating the rocks over which it glides, just 
 like a common glacier, only in so much greater a proportion as 
 the ice-cake of Greenland surpasses any glacier of Switzerland. 
 . . . From an examination then of the whole upper covering 
 of our country, we learn that the older part of this formation 
 has a markedly arctic character, and that the oldest masses of 
 all are moraines. We are entitled, therefore, I think, to con- 
 ceive of Norway, at the close of the Tertiary period, as beiog 
 in a true glacial condition. Its ice -cove ring had an outward 
 movement like the present inland ice of Greenland. By this 
 ice, ' rolled stones 9 and debris were borne outwards to 'the 
 sea shore, where large and small blocks of stone were floated 
 off and carried away on ice-floes. Hence arose those huge 
 moraine masses (far too vast to have been produced by mere 
 separate valley glaciers), which cover the whole flat land near 
 the coast, and which may be traced on both sides of the 
 Christiania Fiord. From this universal covering of moving 
 land ice resulted the first great network of scorings and 
 scratches, as well as the moraine masses extending out to the 
 edge of the sea. Then followed a diminution of the glacial 
 intensity. Instead of a continuous covering of ice, there were 
 separate glaciers, which laboured down from the various 
 mountain slopes into the open valleys at their feet. All the 
 loose material which lay in their way they bore off as moraines, 
 
Erdmann and Geikie. 2 3 1 
 
 either along their sides or in front, or where two valleys con- 
 verged as medial moraines. The process of striation thus 
 went on anew in all these valleys, and in this way two different 
 sets of scratches, lying across one another, might sometimes 
 be produced." 8 
 
 Kjerulf s researches in Norway were supplemented by 
 those of his equally famous compatriot Torell in Sweden/ 
 and also by those of another Scandinavian geologist, Axel 
 Erdmann, who, writing in 1868, speaks of " a glacial 
 epoch during which the surface of our country was covered 
 with an immense sheet of ice" as being established " by un- 
 doubted evidence." J He thus describes the state of things : 
 '' The Scandinavian peninsula, surrounded by a Polar sea, was 
 covered with ice, except its highest points and its loftiest 
 chains. Glaciers of several hundreds, or rather thousands, of 
 feet thick filled the valleys and the depressions of the upper 
 regions, extended over the low lands and reached the coasts, 
 breaking, triturating, polishing and striating the surface over 
 which they passed, and carried to the sea masses of stones, 
 gravel, sand, and clay." 
 
 Professor Geikie has described in felicitous language, which 
 I am tempted to quote, the features of a Norwegian landscape. 
 a No grander display," he says, (C of ice action could one wish 
 to see than that which the fiords and fiord valleys of Norway 
 present. The smoothed and mammillated mountain slopes, 
 the rounded islets that peer above the level of the sea like the 
 backs of great whales, the glistening and highly polished 
 faces of rock that sweep right down into the water, the great 
 perched blocks, ranged like sentinels on jutting points and 
 ledges, the huge mounds of morainic debris at the heads of 
 the valleys, and the wild disorder of the crags and boulders 
 scattered over the former paths of the glaciers continue to 
 make a picture which no after amount of sight-seeing is likely 
 to cause a geologist to forget. The whole country has been 
 moulded and rubbed and polished by one immense sheet of 
 ice which could hardly have been less than 6000 or 7000 feet 
 thick." " The Gulf of Bothnia appears to have brimmed with 
 
 8 Ed. New Phil. Journ. xviii., new series, 
 
 9 Formations Quaternaires de la Suede, 21. 
 1 Id. 19, 20. 
 
232 The Glacial Nightmare. 
 
 ice, which pressed up against and even insomeplaces overflowed 
 the lofty Norwegian frontier, through the valleys of which it 
 found its way into the North Sea. Geikie quotes M. Horbye 
 for the fact that the striae cross the watershed between the two 
 countries. "Mr. Tornebohm," he further says, "of the geo- 
 logical survey of Sweden, informs me that the glacier-carried 
 erratics of Jemtland clearly show that the ice has passed from 
 oast to west that is, right against the slope of the land ; and 
 according to Keilhau, similar blocks, which could only have 
 come from Sweden, are now found inTrondjems fiord. The most 
 remarkable circumstance in connection with some of these 
 blocks consists in the fact, that they occur at a considerably 
 greater height than the rock from which they have beeu 
 derived. Thus at Areskutan, Tornebohm found blocks at a 
 height of 4500 feet which could not possibly have come from 
 any place higher than 1800 feet." Mr. Geikie then quotes 
 Nordenskiold for the conclusion that the Scandinavian ice- 
 sheet once filled the Gulf of Bothnia and occupied the whole 
 area of the Baltic sea, overflowing the Aland islands, Gothland, 
 Oland, Bornholm and Denmark, and passing south-east over 
 Finland into Kussia across Lake Onega, Lake Ladoga and the 
 Gulf of Finland. 2 
 
 While these views prevailed in Scandinavia, the geologists 
 of Germany had not, apparently, been attracted by them, and we 
 first find them advocated there by the Swedish explorer 
 Torell. In November, 1875, he communicated to the German 
 Geological Society a paper on the results of a journey he 
 made with Berendt and Orth to the Rudersdorf mountains, 
 where he found the rocks marked by stria3, and adduced them 
 as evidence of the former extension thither of the Scandina- 
 vian ice-sheet, and argued against the iceberg theory. 3 
 Similar groovings were met with by Credner at Klein Stein- 
 berg in Saxony, and by Penck on the mountain of JDeortz 
 near Taucha. Writing in 1879, the Norwegian geologist 
 Helland argued that these facts could best be explained by 
 the theory that the Scandinavian ice-sheet had deposited them. 
 This he argued from the presence of Scandinavian erratics ; 
 from the fact that the soft beds of North Germany are precisely 
 
 2 Great Ice Age, 403-405. 
 
 3 Zeitschrift der Deutschen Geol. Gesell. xxvii. 961. 
 
Tore II. 233 
 
 like moraine matter, in that the stones in them are of different 
 sizes, heterogeneously mixed, and are many of them unrolled, 
 etc., etc. ; and he argues that since these beds are like a ground 
 moraine in structure, their further limit marks the ice-foot 
 of the ice-sheet. 
 
 He then proceeds to give these limits. He says that neither 
 in France nor Belgium have the northern stones occurred. In 
 Holland they abound, and the most southern specimen 
 recorded is one described by Herr Winkler from Oudenbosch 
 in North Brabant, near Breda. They do not occur, apparently, 
 west of the Rhine, but have been found at Wagennigeii on that 
 river, and at Maaren, east of Utrecht. Turning to England, 
 M. Hellaud claims, after microscopic and other examination, 
 as certainly of Norwegian origin, the boulders of zirkon 
 syenite and rhombic porphyry found in Holderness, the former 
 he derives from Fredrikswarn and Laurwig in Norway, and the 
 latter from Tornebohm. He also recognized granite like that 
 of Christiania. These boulders have been found at Withernsea 
 and Hornsea in Holderness, and as far north as Carnelian Bay. 
 It is curious that these stones should have come from south- 
 east Norway, Christiania, and not from its western coast. As 
 these stones have not been found south of the Thames, 
 Helland puts the limit of the Northern ice sheet at a line 
 joining the mouth of the Thames and the Rhine. Turning 
 to the Continent, he draws the frontier at a line passing 
 through Jevennaar in Holland, Rheinberg, Essen, Dort- 
 mund, Unna, Werl, Soest, Paderboru, then northward to 
 Lemgo, Bodenwerden, south to Hildesheim, then along the 
 Hartzrand by Harzburg and Blankenburg, west of Hartz- 
 gerode, east of Stolberg; then the line makes a great loop by 
 Nordhausen, Muhlhausen, Langensalza, Erfurt as far as 
 Saalfeld, Gera, south of Zwickau, Chemnitz, Perna, Schlu- 
 kenau, Warnsdorf, Reichenberg in Bohemia ; then by the 
 Kiesengebirge and the Sudetes to Teschen in Silesia, Lublin, 
 Kief, Woronesch, Nishni Novgorod and up to within a short 
 distance of the Ural mountains. 
 
 He goes on to urge that these blocks were all deposited by 
 one great ice-sheet, which has also left its traces in the strias of 
 Saxony and of Denmark. That the beds are a ground 
 moraine is proved by their containing striated blocks from 
 
234 The Glacial Nightmare. 
 
 North Germany as well as from Scandinavia, the moving ice 
 having gathered them all on its way. Such striated blocks 
 have been found at Teutschenthal near Halle, and at Ruders- 
 dorf and Mischutz in Saxony. 
 
 He claims that the ice-sheet flowed over Norway and 
 Sweden, overwhelmed the Danish islands, went over Scania, 
 Bornholm, Gland and Gothland, and spread into North 
 Germany until stopped by the Hartz and the Silesian moun- 
 tains. On the Hartz the northern blocks are found, according 
 to Lessen, as high as 450 metres, while Credner only puts them 
 at- 407 metres. Herr Orth has found them in the Silesian 
 mountains at 439 metres high, and as they occur on the Scandi- 
 navian mountains at a height of 1700 metres, this gives a 
 good idea of the mass of ice which must have occupied the 
 intervening district. The mean slope of the Swiss glaciers 
 is about twenty centimetres in twenty-four miles. With this 
 slope the boulders would take 7000 years to travel from 
 Scania to Saxony. 4 
 
 In another memoir, published in the same volume, by 
 Penck, marked by the same learning and knowledge, we find 
 him subscribing to the same views. He urges that the 
 Scandinavian land ice crossed both the Baltic and the North 
 Sea, overflowing North Germany on the one hand, and reach- 
 ing as far as England and Holland on the other. In Germany 
 it flowed on till it was stopped by the Hartz, the Erz and 
 Riesengebirge mountains, while further east it flowed over the 
 flat lands of Central Russia. Northwards it united with the 
 polar ice. This ice-sheet carried no stones on its back. The 
 stones it carried were underneath it in a huge ground 
 moraine, hence why true medial and side moraines do not 
 exist in Germany. It was only when the ice-sheet had shrunk 
 and retired to Scandinavia that it began to deposit these 
 tokens of true glaciers. 5 
 
 Turning from Germany for a moment to France, we find 
 that a considerable discussion has ensued there as to whether 
 glacial striae occur in the neighbourhood of Fontainebleau. 
 M. M. Belgrand and Collomb professed to find them there and 
 also at La Padole and at Champcenil. M. Collomb urged that 
 
 4 Zeitschrift der Deutschen Geologischer Gesell. xxxi. 63-lGo. 
 5 Id. 150, 151. 
 
Tardy and Rarnsay. 235 
 
 the absence of lateral and frontal moraines there is what might 
 be expected ; they can only exist in mountainous districts, 
 and not when the ice traverses a flat district and has no limit- 
 ing barriers, and the striae point towards Orleans and Poitiers, 
 in which direction there are no mountains. He urges again, 
 that the stria3 are at an angle with the contour of the country, 
 the valleys, rivers, and denudations of the plateau of La Brie 
 point to the north-west, while the striae point to the north- 
 east, in a direction almost at right angles, whence it follows 
 that the valleys could not have existed at the time the strias 
 were made, and he dates the latter at the very beginning 
 of the quarternary if not in the pliocene period. 
 
 M. Tardy, addressing himself to the same problem, solved 
 it in another way. He says of these striaB at Fontainebleau, 
 (t Si nous cherchons 1'origine de ce glacier, nous recoimaitrons 
 que les seuls montagnes dont il pouvait descendre sont les 
 Alpes Scandinaves ; ce qui affirme le grand diluvium 
 Scandinave signale par M. E. de Beaumont, tout en le 
 transformant en uu grand glacier continental." ' He thus 
 brought the Scandinavian ice-sheet into central France. 
 Hogard had, many years before, 7 virtually described the ancient 
 glaciers of the Yosges as forming a covering mantle of 
 ice. 
 
 Let us now revert to England. In 1862 Ramsay, in a 
 paper on the glacial origin of lakes, abandoned the theory of 
 icebergs to which he had adhered, and says : "' I am con- 
 strained to return, at least in part, to the theory, many years 
 ago strongly advocated by Agassiz, that in the period of ex- 
 tremest cold of the glacial epoch, great part of North America, 
 the north of the continent of Europe, great part of Britain, 
 Ireland, and the Western Isles were covered by sheets of true 
 glacier ice in motion, which moulded the whole surface of the 
 country and in favourable places scooped out depressions that 
 subsequently became lakes. 1 " He then goes on, as Chambers 
 had done, but without reference to him, to speak of two "periods 
 of glaciation ; one, later, of local glaciers and the other earlier, 
 when great original sheets of ice converted the north of 
 
 6 Bulletin Soc. Geol. de France, xxvii. 562. 
 
 7 Coup d'Oeil sur le terrain erratique des Vosges, 1851. 
 
236 The Glacial Nightmare. 
 
 Europe and America into a country like North Green- 
 land . . . the change extended alike over the lowlands of the 
 Northern and the Southern Hemispheres." 8 
 
 Not only did Eamsay explain the origin of many lakes 
 which are apparently basins scooped out of the rock by 
 glacier erosion, but he tried to prove that the hollow occupied 
 by the Menai Straits was scooped out by the great Scotch ice- 
 sheet. 9 
 
 Kamsay's views as to the excavation of lake basins by 
 glaciers were supported by Mortillet in France, Gastaldi in 
 Italy, and J. Geikie and others in England. I shall 'return 
 to them in a later chapter. 
 
 In the same year Jamieson read a paper on the glaciation 
 of Scotand, full of interesting and valuable observations. 
 From these he concludes that " nothing short of ice filling the 
 valleys up to the brim, and covering the country in one great 
 winding-sheet will meet the requirements of the case. . . . 
 Land ice moving in a volume like that seen in Greenland or on 
 the Antarctic Continent, explains these and many other facts 
 better, in my opinion, than any other theory yet proposed, 
 and so far as I can see, the only strong objection against it is 
 the extraordinary climate for this latitude that it requires. . . . 
 We cannot account for such a development of ice in this 
 country without supposing the whole of the atmospheric 
 moisture, or nearly so, to have fallen in a frozen or snowy 
 condition, and to have had to find its way off the land in the 
 shape of solid ice ; and such a climate in the latitude of Scot- 
 land cannot with any probability be supposed without some 
 great changes in the physical geography of the northern 
 Hemisphere, for it comes to this, that the whole of Scotland 
 must, during the period of greatest glaciation, have been 
 within the snow line. 1 
 
 In a later paper, written in 1865, the same writer says : 
 " It was, therefore, not in the form of narrow glaciers like 
 those of the Alps that the ice existed at this time, but as a 
 thick cake like that of North Greenland, enveloping both hill 
 and dale, and flowing off not so much on account of the 
 
 s Quart. Journ. Geol. Soc. xviii. 204. 
 Id. xxxii 116. i Id. 178-181. 
 
Jamieson and A. Geikie. 237 
 
 inclination of the bed on which it rested as owing to the 
 internal pressure exerted by the immense accumulation of 
 snow over the whole interior of the island, somewhat in the 
 way that a heap of grain flows off when poured down on the 
 floor of a granary. The floor is flat and therefore does not 
 conduct the grain in any direction ; the outward motion is 
 due to the pressure of the particles of grain on one another ; 
 and given a floor of infinite extension, and a pile of grain of 
 sufficient amount, the mass would move outwards to any 
 distance ; and with a very slight pitch or slope, it would 
 slide forward along the incline. The want of much inclina- 
 tion in the surface of a country, and the absence of great 
 Alpine heights, are therefore objections of no moment to the 
 movement of land ice, provided we have snow enough." 
 Turning to the glacial markings on Schehallion and Morven, 
 he urges that these two cases are excellently adapted for 
 showing us to what a prodigious extent the snow and ice must 
 have accumulated. 2 
 
 Sir A. Geikie, who, until the year 1861, had been an ad- 
 vocate of the Iceberg theory as explaining the Scotch phe- 
 nomena, now abandoned that view. In his memoir on the 
 glacial drift of Scotland, published in 1863, he writes : "The 
 fact that the striations and groovings are not confined to the 
 Highland glens, where actual glaciers may well have existed, 
 but extend far away over the Lowlands from sea to sea, 
 seems sufficient proof that, although such markings are un- 
 doubtedly traces of the course of a moving body of land ice, 
 this ice could not have been in the form of a mere ordinary 
 glacier. For the central valley is of greatly too large an 
 extent ever to have been filled by any glacier which could 
 descend from the Highlands. Moreover the same markings 
 occur high on the sides and summits of the Highland moun- 
 tains, which no glacier, in the usual sense of the term, could 
 ever have reached. ... If these markings are in any one 
 place evidence of the former existence of a stream of ice, 
 their universality must also prove that the ice existed not as 
 mere local glaciers descending the chief valleys, but as one 
 wide sheet covering the whole, or nearly the whole, of the 
 
 2 Jouvn. Geol. See. xxi. 106. 
 
238 The Glacial Nightmare. 
 
 country. . . . The former extension of a vast icy mantle over 
 the surface of our island accounts well for the character of the 
 roches moutonnees and for the diversion of the striations from 
 the main ridges of the country. We see at the same time 
 how these markings should not be confined to the valleys, but 
 run along and over the sides and summits of the hills. We 
 see how masses of ice, many hundred feet in thickness, must 
 have accumulated in the Highland glens, overflowing into 
 the adjacent valleys and seeking egress to the sea, by cross- 
 ing over ranges of high ground, as in Gareloch, Bute and 
 Cantyre, where, it will be remembered, the striations on the 
 rocks run up and over the ridges and are as clearly shown on 
 the hill-tops as in the valleys. . . . Down the whole of the 
 west coast, from Cape Wrath to the Mull of Cantyre, one 
 long expanse of ice filled up the fiords, and stretched out into 
 the Atlantic. From the uplands of Wigton and Galloway, 
 the icy stream swept down into the valley of Solway, 
 and onward for Ireland. From the bills that border 
 the lovely valley of Liddesdale far away into the blue 
 Cheviots, the same universal mantle of ice threw its folds 
 athwart the hills and dales of the north of England. . . . 
 The whole country was thus probably sealed in ice, save 
 perhaps some high cliffs and precipices. . . . The rounded 
 and flowing outlines of the hills, the smoothed contour 
 of the knolls, and the long banks of boulder clay, all tell of 
 the passage of the great ice flow. ... It appears that the 
 island was covered, as large tracts of Greenland are, with a 
 wide sheet of snow and ice which, constantly augmented by 
 fresh snow falls, pushed its way outward from the mountain 
 chains towards the sea." 
 
 In a paper by Mr. S. V. Wood, jun., read before the Geological 
 Society in 1866, on a section at Litcham, he argues that, 
 "when the masses of chalk in the Norfolk cliffs were de- 
 posited, the land was occupied by a great glacier, from whose 
 seaward termination they were carried and dropped into the 
 
 marine sediment In speaking of a glacier, I do not 
 
 refer to any such as those occurring in mountain regions, but 
 to one enveloping the whole surface of the country, as de- 
 
 3 Op. cit, 76-96.- 
 
/. Geikie. 239 
 
 scribed by Dr. Sutherland, and now covering the land around 
 Baffin's Bay. . . . Much of the centre and south-wssb of 
 Norfolk was, during the Lower Drift period, the seat of this 
 capping glacier, and its principal seaward termination was 
 in the direction of Cromer, where the bulk of its moraine was 
 spread out under the sea/' 4 
 
 Writing in 1871, Mr. James Geikie says : '' The only theory 
 which seems to satisfy all the requirements of our present 
 knowledge, is that so long ago advanced by Agassiz. Every 
 part of Scotland, with the possible exception of a few peaks or 
 tips of some of the loftier mountains, has certainly been 
 buried under snow and ice. This mer de glace must have 
 levelled up the valleys and occupied all the fiords, sounds and 
 shallow seas around our island. Below this deep sea of ice 
 hill slopes were ground and polished, valleys were deepened 
 and smoothed, and the wreck and rubbish resulting from all 
 this work gathered unequally below the ice, according as the 
 direction and pressure of the superincumbent mass deter- 
 mined/' 5 
 
 In a paper read before the Geological Society in 1872, on 
 the Ice-sheet in. North Lancashire, etc., Mr. Tiddemann says : 
 ' The ice producing these effects must have come from the 
 north, and not from the south. It is clear that local glaciers 
 going down hill have not made them ; for in this case the ice 
 has not been travelling in the direction of the greatest fall, but 
 considerably across it. Nor can they have been produced by 
 icebergs grounding, as might be suggested ; for then there 
 would be some indication of coasting, whereas these markings 
 are all nearly parallel, and going up the hill instead of along 
 it. Nor will an ice-foot explain it ; for then we should have 
 scratches radiating from the higher ground. In fact I can 
 conceive of no arrangement but that of land ice pushed up 
 over this hill (i.e. Claughton Fells) by pressure from behind, 
 and obliged to maintain a definite course by ice to the right 
 and left of it." 
 
 Again, speaking of the scratches, etc., at Ingleborough, he 
 says : " They certainly show no signs of having been formed 
 by any small local glaciers which had their birth on Ingle- 
 
 4 Quart. Journ. Geol. Soc. xxiii. 84-87. 
 
 5 Geological Magazine, viii. 547. 
 
240 The Glacial Nightmare. 
 
 borough ; for they are transverse to the course of any such 
 glaciers." Again, speaking of similar marks at Bowland 
 Knotts : u If these were produced by a local glacier, what were 
 its gathering grounds ? Not the Fells immediately west ; for 
 then the scratches would be at right angles to their present 
 course. Looking' north, the nearest ground equalling this in 
 height is seven and a half miles distant, and between this and 
 that lies a broad valley, excavated to the depth of about 1000 
 feet. Ingleborough, with its additional 900 feet, could hardly 
 fill up that valley, and make itself felt at that distance. 
 There is no ground equally high nearer than Pendle Hill, 
 twelve miles off, and across two broad valleys. The 
 height at which these scratches occur across this ridge is 
 greater than the height of some of the passes across the Fells 
 to the west of it. There can be little doubt that the ice 
 stream passed entirely over these Fells. " Speaking of certain 
 scratches running in directions transverse to each other, he 
 recalls a suggestion of Professor Ramsay, that it is possible 
 that there may have been undercurrents in the ice-sheet 
 caused by the form of the ground. Again, he urges " that 
 such an agent as is needed to explain the facts is to be found 
 in a great ice-sheet pushing on from its northern gathering 
 grounds, recruited by the greater elevations on its course, but 
 overriding the lesser, grinding down and smoothing by its 
 weight and friction rocks presenting but a gentle incline, 
 tearing' up and turning over the basset edges confronting its 
 approach. ... I know not any phenomenon connected with 
 the glacial period which gives so vivid a picture of a resistless 
 force working in an undeviating course over hill and dale, 
 across the ordinary drainage channels of the country as this 
 wreck and ruin of all opposing obstacles.'''' Our author 
 further argues that a barrier of ice filled up the Irish Sea 
 between the Lake district and the Isle of Man, which diverted 
 the drainage of the Lune and Kibble from their natural outflow. 
 This barrier, he says, was but the line of junction of the ice of 
 the Pennine chain with that from the Lake distinct, and to 'the 
 eye they must have presented the appearance of one great 
 sea of ice, and this barrier must have been supported and 
 shouldered up by other ice coming from portions of the Lake 
 district still further west. By this ice stream he explains the 
 
/. Geikie. 241 
 
 fact of the striae on Anglesea being from the N.N.E. and not 
 radiating from the mountains, and that you get foreign blocks 
 from the hills of Cumberland there. Similarly Mr. Close has 
 shown that on the opposite coast of Ireland the glaciation is 
 along the coast and not from the land. Professor Phillips 
 had observed the same thing at Bray Head. 6 
 
 In his " Great Ice Age/' first published in 1873, Professor 
 Geikie says : " On every hand we are furnished with abundant 
 proof of the former existence of a great mer de glace in 
 Scotland. From the tops of some of the higher mountains 
 down to the edge of the sea, no part of the country has 
 escaped abrasion. . . . Nay, most of the islands which lie off 
 the coasts plainly indicate by striations and other glacial 
 markings that ice has swept over them. They are smoothed 
 not from the centre to the circumference, as would have been 
 the case if they had supported separate glaciers of their own, 
 but the striations go right across them from side to side. It 
 cannot be doubted, therefore, that the ice actually crossed from 
 the mainland over what now forms the bed of the sea. In Lewis 
 I found the northern portion of Long Island glaciated across its 
 whole breadth from S.E to JST.W. The land ice that swept over, 
 this tract must have come from the mountains of Ross-shire, a 
 distance of not less than thirty miles, and have filled up the 
 whole of the North Minch (sixty fathoms in depth) and over- 
 flowed Lewis to a height of 1300 feet at least, for I found 
 Suainabhal, a conspicuous mountain rising to the south of Loch 
 Roag, was merely a huge roche moutonnee, strongly glaciated 
 from base to summit by ice which had passed over it from 
 the S.E. . . . In like manner the island of Bute has been 
 scored and smoothed from end to end by a mass of ice which, 
 streaming from the highlands of Argyleshire, filled up the 
 Kyles and then passed southward over the whole island to 
 occupy the bed of the Frith of Clyde between Bute and 
 Arran. Many of the islets scattered along the western 
 coasts tell the same tale." After describing the edge of the Ant- 
 arctic continent he says : " And such must have been the aspect 
 presented by the margin of the old ice sheet which in the 
 early stages of the glacial epoch mantled Scotland and its 
 
 6 Journ. Geol. Soc. xxviii. 471-491. 
 
242 The Glacial Nightmare. 
 
 numerous isles, filling up the intervening straits and channels 
 of the sea, and terminating far out in the Atlantic Ocean in a 
 flat-topped vertical cliff of blue ice." 7 
 
 Similar views are developed at greater length in the same 
 author's " Prehistoric Europe." 
 
 " Thus are we driven to conclude," he says, " that during 
 the climax of the Glacial Period all Scotland was drowned in a 
 widespread mer de glace, which coalesced in the north and 
 east with a similar sheet that crept outwards from Scan- 
 dinavia. To the west the Scottish ice meeting with no im- 
 pediment to its course overflowed the Outer Hebrides to a 
 height of 1000 feet, and probably continued on its path into 
 the Atlantic as far as the edge of the 100 fathom plateau, where 
 the somewhat sudden deepening of the sea would allow it to 
 break off and send adrift whole argosies of icebergs. The 
 height reached by the upper surface of the ice that overwhelmed 
 the Outer Hebrides enables us to ascertain the angle of slope 
 between these islands and the mainland. This was one in 
 211, that is to say, the inclination of the surface of 
 the ice-sheet was about twenty-five feet in the mile, an 
 inclination which would appear to the eye almost like a 
 dead level. 8 , . . Glacial strise and boulder clay have been 
 followed over all Ireland. . . . Our sister island was buried 
 under an ice-sheet hardly less extensive than that which over- 
 whelmed Scotland. . . . The ice flowed off Ireland in all 
 directions save to north-east in Antrim, upon the coast of 
 which it encountered the Scotch mer de glace, which forced it 
 to turn away to north-west and south-east ; but along the 
 whole western and southern shores, where no obstacle to its 
 passage intervened, it seems to have swept in one broad and 
 continuous stream, out probably as far as that of Scotland, 
 into the Atlantic. The thickness attained by the ice that 
 flowed into the Irish Sea from Scotland, where it coalesced with 
 the mer de glace coming from the eastern seaboard of Ireland, 
 and also with that creeping from England and Wales, makes it 
 quite certain that the area now occupied by that sea must 
 at that time have been filled with glacier ice. 9 Despite 
 the pressure exerted by the massive sheet that made its 
 
 7 Op. cit. 77-81, 8 Prehistoric Europe, 139> 
 
 9 Id. 189, 190. 
 
/. Geikie. 243 
 
 way outwards from the Pennine chain, the English ice could 
 not escape into the basin of the North Sea, and conse- 
 quently we find stones from Scotland, Northumberland and 
 Durham plentifully present in boulder clay all along the 
 eastern maritime districts of England. More than this, when 
 we get as far south as the Humber, and follow the spoor of the 
 ice as indicated by the carry of the boulder clay stones, we are 
 led across Lincolnshire into the Midland Counties by Notting- 
 hamshire and Leicestershire. . . . The North Sea was filled 
 with a massive mer de glace, continually advancing in a general 
 S.S.W. direction, the presence of which is distinctly traceable 
 in the remarkable deflection of the glaciation all along the 
 sea-board of Scotland, from Stonehaven southwards." Our 
 author urges that the greater elevation and extent of the 
 Scottish mountains alone protected the invasion of Scotland by 
 this alien ice-stream, while in England, where the ground was 
 lower, "it overflowed bodily and crept south-west across the 
 Midland tableland on its way to the valley of the Somme and 
 the Bristol Channel. This is proved," he says, " by the carry 
 of the local stones and by the appearance in the till at 
 Cornelian Bay and Holderness of boulders of two well-known 
 Norwegian rocks, which were recognized by Mr. Amund 
 Helland, etc." l 
 
 " The ice," he continues, " which would thus appear to have 
 streamed transversely across England, eventually coalesced 
 with that which overflowed from the basin of the Irish Sea 
 south-east through Cheshire together with that which streamed 
 east from the Welsh uplands, and the united mer de glace 
 thereafter made its way into the Bristol Channel. Here it 
 joined the thick ice that flowed out to sea from the high 
 grounds of North "Wales. ... In the south-eastern counties, 
 so far as we know at present, the ice-sheet at the climax of 
 the Glacial Periqd did not extend farther than the valley of the 
 Thames " z 
 
 " Thus in time," Mr. Geikie says elsewhere, " all Scotland 
 became enveloped in ice that flowed west to break off in deep 
 water beyond the Hebrides, and east to meet the Scandinavian 
 mer de glace which had all the while been creeping outwards 
 
 1 Prehistoric Europe, 192, 193. * Id. 193. 
 
 R 2 
 
244 The Glacial Nightmare. 
 
 into the basin of the North Sea. The North of England was 
 also shrouded in ice, part of its sheet coalesced with the 
 united Scottish and Scandinavian ice-sheet that filled up the 
 German Ocean. How far south in England that ice-sheet 
 flowed, still remains to be more rigorously determined. A 
 broad belt of ice overflowed from the basin of the Irish Sea, 
 and uniting with the glaciers that descended from Wales, 
 spread in the direction of the Severn Valley. In like manner 
 the ice that flowed eastward from the Pennine Chain to 
 coalesce with the mer de glace of the North Sea appears to 
 have advanced into Lincolnshire. But in the high grounds of 
 Derby the ice-flow may have been more or less independent, 
 as it appears also to have been in Charnwood Forest. Ireland 
 with its lofty hills and humid climate, was, like Scotland and 
 Wales, more or less buried in snow and ice, and its immense 
 glaciers, uniting with those of Scotland and England, must 
 have filled up the Irish Sea." 8 
 
 We will next turn to Mr. Croll, who, in his well-known work 
 " Climate and Time," says : " At one time not only an arctic 
 condition of climate prevailed in our island, but the greater 
 part of the temperate region down to comparatively low lati- 
 tudes was buried under ice. 4 Without the conception of 
 continental ice, the known facts connected with glaciation 
 would be perfectly inexplicable. It was only when it was 
 found that the accumulated facts refused to be explained by 
 any other conception, that belief in the very existence of such 
 a thing as continental ice became common. But although 
 most geologists now admit the existence of continental ice, yet 
 nevertheless, adequate conceptions of its real magnitude are 
 by no means so common. Year by year, as the outstanding 
 facts connected with glaciation accumulate, we are compelled 
 to extend our conceptions of the magnitude of land ice. Take 
 the following as an example. It was found that the transport 
 of the Wastdale Crag blocks, the direction of the striae on the 
 islands of the Baltic, on Caithness, and on the Orkney, Shet- 
 land and Faroe Islands, the boulder clay with broken shells 
 in Caithness, Holderness and other places were inexplicable 
 on the theory of land ice. But it was so only in consequence 
 
 3 Id. 351, 352. 4 Climate and Time, i. 7. 
 
Croll. 
 
 : 4S 
 
 of tlie inadequacy of our conceptions of the magnitude of the 
 ice." 5 Speaking of the glacial period, Mr. Croll says : " All the 
 northern seas, owing to their shallowness, must, at that period, 
 have been blocked up with solid ice, which displaced the 
 water and moved along the sea-bottoms the same as on dry 
 land. In fact, the northern seas, including the German 
 Ocean, being filled at the time with glacier-ice, might be re- 
 garded as dry land/' 6 
 
 Again he says : " It may now be regarded as an established 
 fact that during the severer part of the glacial period, Scotland 
 , was covered with one continuous mantle of ice, so thick as to 
 bury under it the Ochil, Sidlaw, Pentland, Canipsie and other 
 moderately high mountain ranges. For example, Mr. J. Geikie 
 and Mr. B. 1ST. Peach found that the great masses of the ice from 
 the N. W". Highlands came straight over the Ochils of Perthshire 
 and the Lomonds of Fife. In fact, these mountain ridges were 
 not sufficiently high to deflect the icy stream either to the right 
 hand or to the left ; and the flattened and rounded tops of the 
 Campsie, Pentland, and Lammermoor ranges bear ample 
 testimony to the denuding power of ice. ... It is therefore 
 evident that the great mass of ice entering the North Sea to 
 the east of Scotland, especially about the Firths of Forth and 
 Tay, could not have been less, and was probably much more, 
 than from 1000 to 2000 feet in thickness. 7 We must consider 
 that the ice from Scotland and England was but a fraction of 
 that which entered the North Sea. . . . The Scandinavian 
 ice, before it could break up, would thus, like the Scottish ice, 
 have to pass the North Sea and into the Atlantic. It 
 could not pass to the north, or to the north-west, for this 
 ocean in these directions would be blocked up by the polar 
 ice." After urging that the deep trough along the coast of 
 Norway would be rapidly filled up, he continues : " Thus 
 the only path for the ice would be by the Orkney and Shetland 
 Islands. . . ; In all probability, had Scotland been a low, flat 
 island, instead of being a high and mountainous one, the ice 
 would have passed completely over it. But its mountainous 
 character, and the enormous masses of ice at the time proceed- 
 ing from its interior, would effectually prevent this, so that the 
 
 5 Id. 385, 386. 6 Id. 438. 7 Id. 442. 
 
246 The Glacial Nightmare. 
 
 ice of Scandinavia would be compelled to move round by the 
 Orkney Islands. Consequently, these two huge masses of 
 moving ice the one from Scotland and the much greater one 
 from Scandinavia would meet in the North Sea, probably not 
 far from our shores, and would move side by side northwards 
 into the Atlantic as one gigantic glacier. . . . The median line 
 of the Scandinavian and Scottish ice-sheets would be situated 
 not far from the east coast of Scotland. The Scandinavian ice 
 would press up as near to our coast as the resistance of the ice 
 from this side permitted, the enormous mass of ice from 
 Scotland, pressing out into the North Sea, would compel the . 
 Scandinavian ice to move round by the Orkneys and would also 
 keep it at some little distance from Scotland. Where, on the 
 other hand, there was but little resistance offered by ice from 
 the interior of the country, and this might be the case along 
 many parts of the English coast, the Scandinavian ice might 
 reach the shores and even overrun the country for some dis- 
 tance inland." 
 
 Turning to the South Swedish ice-sheet, Mr. Croll says: 
 "All this mass of ice must have gone into the shallow 
 Cattegat, and thence passed over Jutland into the North Sea. 
 After entering the North Sea it would be obliged to keep 
 between our shores "and the ice coming direct from the 
 western side of Scandinavia. ... A very large proportion of 
 the Scandinavian ice would pass into the Gulf of Bothnia ; after 
 passing down the Baltic, a portion of the ice would probably 
 move south into the flat plains in the north of Germany, but 
 the greater portion would keep in the bed of the Baltic, and 
 of course turn to the right round the south end of Gothland 
 and then cross over Denmark into the North Sea. . . . The 
 Gulf of Bothnia and the entire Baltic were filled with one 
 continuous mass of land ice, derived from the drainage of the 
 greater part of Sweden, Lapland, and Finland. In fact, the 
 whole glacial phenomena of Scandinavia are inexplicable on 
 the hypothesis of local glaciers. . . . The icy stream, affcer 
 passing Karlskrona, bent round to the west along the present 
 entrance to the Baltic and again invaded the mainland, and 
 crossed over the low headland of Christianstadt, and thence 
 passed westwards in the direction of Zealand. This immense 
 Baltic glacier would in all probability pass over Denmark, and 
 
Croll. 
 
 247 
 
 enter the North Sea somewhere to the north of the river Elbe, 
 and would then have to find an outlet to the Atlantic through 
 the English Channel, or pass in between our eastern shores 
 and the mass from Gothland and the N.W. shores of Europe. 
 . . Is it possible that the entire Atlantic, from Scandinavia 
 to Greenland, was filled with land ice ? Astounding as this 
 may at first appear, there are several considerations which 
 render such a conclusion probable. The observations of 
 Chambers, Peach, Hibbert, Allan and others show that the 
 rocky face of the Shetland and Faroe Islands has been ground, 
 polished and striated in a most remarkable manner ; that this 
 could not have been done by ice belonging to the islands them- 
 selves is obvious, for these islands are much too small to have 
 supported glaciers of any size, and the smallest of them is 
 striated as well as the largest. Besides, the uniform direction 
 of the strias on the rocks shows that it must have been effected 
 by ice passing over the islands. . . . That the whole of these 
 islands have been glaciated by a continuous sheet of ice passing 
 over them was the impression left on the mind of Robert 
 Chambers after visiting them. This is the theory which alone 
 explains all the facts. 8 The glacial phenomena of Scotland are 
 wholly inexplicable upon any other theory than that, during at 
 least a part of the glacial epoch, the entire island from sea to 
 sea was covered with one continuous mass of ice of not less 
 than 2000 feet in thickness." 
 
 Such have been the drafts upon space invoked by the more 
 extreme advocates of " omnipotent ice." They have made 
 corresponding drafts upon time. Thus it was calculated, from 
 the average progression of glaciers, that it would require 
 10,000 years to carry a boulder 100 miles. Now, as the erratics 
 from Scandinavia have been found in Russia, at least 700 miles 
 from Sweden, as Murchison testifies, this means that these 
 erratics were actually in process of transportation for 70,000 
 years ; but this calculation is outdone by others. M. Mortillet, 
 the well-known editor of the " Materiaux pour 1'Histoire de 
 1'Homme/' says in his recent work on Prehistoric times : " The 
 length of the great Alpine glaciers varies from 110 to 280 
 kilometres. Eighteen observations made upon the rate of 
 
 8 Op. cit. 451. 
 
248 The Glacial Nightmare. 
 
 progress of modern glaciers gives a mean rate of 62 m '66 
 per annum. An erratic block, therefore, would take 4468 
 years to travel from the superior limit of the glacier to its 
 inferior one; but this number is much too small. These 
 eighteen observations were made among the mountains, where 
 the slope of the ground is very great, and as frozen water 
 follows the same law as liquid water, its speed and thrust 
 diminish rapidly with the diminished inclination." M. 
 Mortillet then cites in proof of this the well-known observa- 
 tions of Agassiz, and says : " The slope of the Quaternary 
 glaciers was at least five times less than that of existing 
 glaciers, whence the rapidity of flow was also five times less, 
 which would transform the 4468 years just cited into 22,340 
 years. As the Quaternary glaciers quitted the narrow 
 mountains and spread themselves out into the plains, it v^ould 
 require a proportional increase in the time required, and we 
 must at least double the number of years here indicated. 
 The quaternary glaciers, again, must have remained a long time 
 at their position of greatest extension, as is proved by the 
 enormous moraines forming series of hills at their termination. 
 Lastly, the retreat of the glaciers, which must have required as 
 long a time as their advance. This advance and retreat, again, 
 was not continuous, but subject to oscillations which would 
 absorb a considerable time. One is therefore below the truth in 
 assigning 100,000 years as the term of existence of the Glacial 
 Period." 9 Applying M. Mortillet's figures to the Eussian 
 boulders, and replacing 280 kilometres by 700 miles, we 
 shall have a measure and test of the courage of some 
 prominent champions of the Glacial theory. 
 
 9 Le Prehistorique, 625, 626. 
 
 
249 
 
 CHAPTER VII. 
 
 THE GROWTH AND CULMINATION OF THE GLACIAL NIGHTMARE. 
 
 II. The New World. 
 
 " In thrilling regions of thick-ribbed ice." " Measure for Measure," 
 Act III. sc. 1. 
 
 North America: Agassiz, 1846-1850, etc., Dana, Winchell, Newberry, 
 Morris Ice Cap in the Northern Hemisphere: Agassiz and 
 Dana Southern Ice Cap : Agassiz and Haast Tropical America : 
 Hartt, St. John, Belt, A. Wallace. 
 
 IN the previous chapter I have traced the progress of what I 
 have ventured to call the glacial nightmare in the Old World. 
 I now propose to extend and complete the survey by describ- 
 ing the application of the theory to the New, and I will begin 
 with North America. 
 
 In 1846 Agassiz went to America, and lie tells us how, on 
 landing in a new continent full of promise for him, he sprang 
 on shore at Halifax and started at a brisk pace for the heights 
 above the landing. On the first undisturbed ground, after 
 leaving the town, he was met, he says, by the familiar signs, 
 the polished surfaces, the furrows and scratches, the line en- 
 graving of the glacier, so well known in the Old World. 1 
 
 In 1850 Agassiz published a work on Lake Superior. In 
 this he devoted a chapter to the erratic phenomena of that 
 district, in which he says, inter alia: "I maintain that the 
 cause which has transported these boulders in the American 
 continent must have acted simultaneously over the whole 
 ground which these boulders cover, as they present throughout 
 the continent an uninterrupted sheet of loose materials, of the 
 
 1 Geol. Sketches, ser. II. 77. 
 
250 The Glacial Nightmare. 
 
 same general nature, connected in the same general manner, 
 and evidently dispersed at the same time." * 
 
 <f It is well known/' he says elsewhere, <f that the usual cha- 
 racteristic marks of glaciers extend over the face of the land in 
 the eastern half of the continent, from the Atlantic shores to 
 the States west of the Mississippi, and from the Arctic Sea to 
 the latitude of the Ohio, in its middle course, while within the 
 range of the Alleghanies they stretch as far south as Georgia and 
 Alabama. In no other region where these traces have been 
 observed do they extend over such wide tracts of country in 
 unbroken continuity, this being of course owing to the level 
 character of the land itself." 3 After describing the phenomena 
 as they present themselves in America with his usual graphic 
 force, and urging that there are few traces of local and 
 sporadic glaciation, he continues : " In short, the ice of the 
 great glacial period in America moved over the continent as 
 one continuous sheet, overriding nearly all the inequalities of 
 the surface. . . . The polished surfaces stretch continuously 
 over hundreds and hundreds of miles; the rectilineal- 
 scratches, grooves, and furrows are unbroken for great 
 distances; the drift spreads in one vast sheet over the 
 whole land, consisting of an indiscriminate medley of clays, 
 sands, gravels, pebbles, boulders of all descriptions, so 
 uniformly mixed together that it presents hardly any difference 
 in its composition, whether we examine it in New England, 
 New York, Pennsylvania, Ohio, Michigan, Indiana, Illinois, 
 Wisconsin, in Iowa beyond the Mississippi, in the more 
 northern territories or in Canada." 4 Speaking of the fact that 
 in America the boulders are for the most part rounded, while 
 there are few erratics like those of Europe which have preserved 
 their angles intact, he says : " Ifc is evident from this that 
 the ice overtopped the rocky inequalities of the land and that 
 the detached fragments remaining beneath the icy covering 
 underwent the same action from friction and pressure to 
 which the whole mass of drift was subjected." 5 The few 
 angular blocks that are found he attributes to the period when 
 on the retreat of the ice some of the higher portions of the 
 rock were laid bare, and he thus accounts for their greater 
 
 2 Silliman's Journal, x. 83, etc. a Geol. Sketches, 78f?9. 
 
 4 Id. 82. 5 Id. 83, 84. 
 
Agassiz. 251 
 
 frequency in New England, where the mountain elevation is 
 greatest. 
 
 He urges again that the mineralogical character of the loose 
 materials forming the American drift leaves no doubt that the 
 whole movement, with the exception of a few local modifications 
 easily accounted for by the lay of the land, was from north to 
 south, all the fragments not belonging to the localities where 
 they occur being readily traced to rocks in situ to the north 
 of their present resting-places. " The farther One journeys 
 from their origin, the more extraordinary does the presence 
 of these boulders become. It strikes one strangely to find 
 even in New England fragments of rock from the shores of 
 Lake Superior; but it is still more impressive to meet with 
 masses of northern rock on the prairies of Illinois and Iowa. 
 One may follow these boulders to the 40th. degree of latitude, 
 beyond which they become more and more rare, while the 
 finer drift alone extends farther south. " 
 
 Again, in favour of the same conclusion, he cites the roches 
 moutonnees of America. " Whenever," he says, " the natural 
 surface of any hill, having a steep southern slope, is exposed, 
 the marks are always found to be very distinct on the northern 
 side and entirely wanting on the southern one, showing that 
 .... the mass moved up the northern slope, forcing its way 
 against it, grinding and furrowing the northern face of the hill as 
 it moved over it, but bridging the opposite side in its descent 
 without coming into contact with it. This is true not only of 
 hills, but of much slighter obstacles which presented them- 
 selves in the path of the ice. Even pebbles embedded 
 in masses of pudding stone, but rising sometimes above 
 the level of the general surface, often have their northern 
 side polished and scratched, while the southern one remains 
 untouched. 3 '' 6 
 
 Again, speaking of Maine, he says : " The glacial traces over 
 the State of Maine, and especially between Bangor and the 
 sea-coast, afford the means of estimating approximately the 
 thickness of the ice-sheet which once moved over the whole 
 land, as well as its limitations during a later period, when it 
 had begun to wane. In order to advance across a hilly 
 
 (i Geol. Sketches, ser. II. 85, 86. 
 
252 The Glacial Nightmare. 
 
 country and over mountainous ridges rising to a height of 
 1200 and 1500 feet in the southern part of the State, and to a 
 much higher level in its northern portion,, the ice must have 
 been several times thicker than the height of the inequalities 
 over which it passed ; otherwise it would have been encased 
 between these elevations, which would have acted as walls to 
 enclose it. We are therefore justified in supposing that the 
 ice-fields, when they poured from the north over New England 
 to the sea, had a thickness of at least five or six thousand 
 feet." 7 
 
 Elsewhere he gives us another measure of the postulated 
 ice-sheet in depth and extent. He urges that "the slopes of 
 the Alleghany range wherever examined are glacier-worn 
 to the very top with the exception of a few points, but these 
 points are sufficient to give us data for the comparison. 
 Mount Washington, for instance, is over 6000 feet high, and 
 the rough, unpolished surface of its summit, 'covered with 
 loose fragments, just below the level of which glacier marks 
 come to an end, tells us that it lifted its head alone above the 
 desolate waste of ice and snow. In this region then the thick- 
 ness of the sheet cannot have been much less than 6000 
 feet, and this is in keeping with the same kind of evidence in 
 other parts of the country, for wherever the mountains are 
 much below 6000 feet, the ice seems to have passed 
 directly over them, while the few peaks rising to that height 
 are left untouched ; and while we can thus sink our plummet 
 from the summit to the base of Mount Washington and 
 measure the thickness of the mass of ice, we have a no less 
 accurate indication of its extension in the undulating line 
 marking the southern termination of the drift. . . . We may 
 trace the southern border of our ancient ice-sheet on this con- 
 tinent by the limit of the boulders ; beyond that line it 
 evidently did not advance as a solid mass, since it ceased to 
 transport the heavier materials." 8 
 
 The views here formulated were fully accepted by some of 
 the most prominent American geologists. Thus Professor 
 Dana, whose authority as their mouthpiece none will dispute, 
 says : " The surface of the glacier of America must have been 
 
 ,v^ 
 7 Id. 151. s Id. 98, 99. 
 
Dana. 253 
 
 of unblemished whiteness ; for from New England west to the 
 Mississippi there was not a peak above its surface, excepting 
 the White Mountains, and these probably had their cap of 
 snow. . . The direction of the scratches and the extent of 
 country they cover appear to show that over New England and 
 Eastern Canada there was a gradual rise in the surface of the 
 glacier towards the north-west and over western New York and 
 Western Canada towards the north-east, and that the ice- 
 summit or the region towards which the directions converge 
 and therefore the opposite slopes rise was over the Canada 
 water-shed nearly north of Montreal. From this, its southern 
 portion, the broad ice-range stretched northward and north- 
 eastward, for this is proved by the south-west direction of the 
 drift and scratches over the country from Lake Huron to and 
 beyond Lake Winnipeg. Since evidences of the great southward 
 moving glacier fail over the region west of a line passing from 
 a few degrees west of Winnipeg, south-eastward through 
 Western Minnesota and Iowa, near the meridians of 98 to 100, 
 and all the way westward to the borders of California and 
 Oregon, if not to the Pacific coast, the ice thinned out towards 
 the interior of the continent, and was mostly absent except 
 about the higher parts of the Rocky Mountains. . . . The 
 lower limit of the glacier of New England may have been over 
 the shallow border of the ocean, perhaps as far out as the fifty 
 fathom or one hundred fathom line of soundings, which on 
 the east passes outside of St. George's Shoal, and off Sable 
 Shoal east of Nova Scotia. Long Island and the other islands 
 off New England have been made by Clarence King and some 
 other writers the course of a part of the terminal moraine, 
 and a range of boulder hills covering New Jersey, west of 
 Perth Ambay, have been pointed out by Professor Smock as 
 another portion of it. In Wisconsin the Green Bay Valley 
 is bordered on the east, west and south by a range of stony 
 hills called the Kettle Eange, and this range is described as 
 of moraine origin by Chamberlin and Irving, and as directly 
 connected with the Green Bay ice during part of the era of 
 ice. At the time of its greatest extension the glacier reached 
 to Southern Iowa and the border of Missouri, where occur 
 boulders and even native copper derived from the Lake Superior 
 region, and even into Kansas. A mass of copper found in 
 
254 The Glacia I Nightmare . 
 
 Lucas Co., Iowa, travelled 460 miles, if it came from Keneenaw 
 Point, the probable source/ 3 ! 
 
 He estimates the ice in the northern part of the glacial 
 districts of America to have had a thickness of from 5000 to 
 6500 feet, and in the south an average of 2700 feet. The 
 pressure, he says, must have been immense, 6000 feet corre- 
 sponding to at least 300,000 pounds to the square foot. 1 
 
 Professor Winchell, who is an ardent champion of the views 
 of Agassiz, writes thus in regard to the ice-sheet in North 
 America. " In regions far to the north, the eye probably would 
 not be able to discern any object except that of universal ice. 
 The surface of the ground would be thousands of feet below 
 the traveller, if we may be permitted to presume so hardy a 
 human being. Like Dr. Kane exploring the great Hum- 
 boldt Glacier of Greenland, he would meet with countless 
 obstacles and dangers. But those obstacles would consist of 
 hummocky ice, or crevassed ice, or perpendicular ice-walls. 
 He would see no soil, no rocks, no vegetation, no animal life. 
 The winds would whistle, storms would rage, snow would be 
 drifted about, and the ineffectual sun would rarely venture to 
 smile on the dreary waste. Farther to the south, the ex- 
 plorer would find isolated spots of bare ground. He would 
 see about them the accumulated debris of boulders, gravel, 
 and dust, from constant winds, spread more or less over the 
 ice-field, staining its painful whiteness, and showing the more 
 grateful aspect of earth and stones. Another hundred miles 
 farther south, and he finds the 'evidences of the dissolution of 
 the ice- sheet multiplying. Occasional streams of water run 
 on the surface of the ice, or plunge into some of its openings. 
 Deep gorges reveal multitudes of fragments of rock frozen 
 into the ice, and occasional bands of dirt and gravel embraced 
 in the solid ice. The surface is everywhere dirty, or perhaps 
 muddy, from the wasting away of the surface of the glacier. 
 He meets frequent openings, in which generally water may be 
 seen or heard. In these gorges the debris slides down the 
 sloping sides, increasing the insecurity of his footsteps. Still 
 farther south, the general surface is covered with a pulpy 
 earth, mingled with stones and boulders. The ice is evidently 
 much attenuated. The areas of firm, uncovered terra. Jtrma 
 
 9 Dana, Manual of Geology, 537, 538 x G-eol. Mag. x. 277 
 
Wine he! I. 255 
 
 are wonderfully increased in size and frequency. The ice 
 itself is crowded into the valleys, or, if it be in a broad, level 
 tract, like the State of Minnesota, the surface is covered with 
 the debris of the conflict of ice with earth, the ice itself being 
 visible only in those places where crevasses reveal it, or where 
 deep gorges are worn by running streams. Travelling still 
 farther south, the explorer would come upon large areas in 
 which he would not be able to know whether the glacier 
 underlay the superficial drift or not. If he were to stop on 
 one of those wide areas, and make his latitude and longitude 
 certain, by a series of astronomical observations, he might 
 find, to his surprise, after a few years' residence, that his 
 observatory and apparatus had been bodily carried, by an 
 imperceptible motion, some rods to the south. If he were to 
 penetrate the earth on which his foothold seems so steadfast, 
 he might find, equally to his surprise, that he was still riding 
 on the surface of a vast ice-sheet, the earth and soil of which 
 may have furnished him annual crops of potatoes and barley. 
 In other places in the same latitude he would find the ice laid 
 bare over considerable areas, washed clean by the drainage 
 incident to the dissolution of the glacier. The turbid streams 
 would be vastly larger than those which occupy the same beds 
 to-day. They would run with tenfold more violence. The 
 drift-materials would be freed from the clayey portions, and 
 be spread along their channels in curious and varying assort- 
 ment. In some places the thickness of the whole sheet of drift 
 would be brought under this washing and stratifying process. 
 In others, the ice gently dies out, and lets it down on the 
 rocky surface without any change from the condition in which 
 it lay on the glacier. 
 
 11 There is still one important point in this discussion that 
 must not be omitted. It is plain to see that, in some parts of 
 the north-west, the advance of the continental glacier would 
 be up gentle slopes, instead of descending an incline. These 
 slopes, of course, present obstructions to the movement of the 
 ice in those directions. It is true al^o, that the continental 
 glacier would tend to level the country and obliterate such 
 northward slopes. But, in the later part of the ice period, the 
 valleys would be the last relinquished, and would be deeper dug 
 by isolated branches or spurs from the main ice-sheet, which 
 
256 The Glacial Nightmare. 
 
 would conform in their direction to the contour of the valleys 
 they might occupy. . . . All the phenomena of the drift in the 
 north-west are, hence, attributable to the approach, long 
 duration, and slow disappearance of the glacier-ice of Professor 
 L. Agassiz." - Elsewhere the same geologist says : " The 
 marks of this stupendous glacier are yet visible. The northern 
 limit was chained by eternal frost to its rocky bed. The 
 southern only was free to move, and the whole expansion 
 would be developed along the southern border." 3 " From 
 season to season, and from year to year, the mighty mass 
 marched irresistibly forward, mowing down the forests, 
 crushing tree-trunks, or burying them with the rubbish of 
 the rocks, from ten to sixty feet below the surface. The 
 great glacier moved onward. It bridged Niagara River; it 
 bridged Long Island Sound, and bathed itself in the mild 
 waters of the ocean beyond/' 4 " It made a tabula rasa. 
 River channels were filled up ; in Ohio and Indiana these 
 buried river-beds are of frequent occurrence. The ancient 
 gorge of the Niagara River was filled by the obliterating 
 agency of this continental glacier. It wiped out the 
 trifling furrow, . . making it necessary for the river to begin 
 anew its work." 5 " It bore rocks southward twenty, fifty, 
 and even five hundred miles. These fragments have been 
 transported over lakes, sounds, and seas." The glacial 
 agency is not recognized south of the Ohio River. 6 The 
 south was spared ; nay, greatly favoured ; for except the 
 chains of mountains, the material ploughed up by the ice was 
 carried by the waters to make the Cretaceous seaboard, the 
 Gulf States, and many more of the Mississippi and Missouri 
 valleys." 
 
 Dr.Newberry,in the Ohio Geological Report for 1873, speaks 
 no less emphatically. " We have abundant evidence," he says, 
 " that during what is called the drift period the climate of our 
 continent had changed from the all-pervading warmth of the 
 Tertiary period to an all-pervading arctic cold. While in the 
 former age the climate of our southern states was carried to 
 
 2 Winchell, The Drift Deposits of the North- West, 293-295. 
 
 3 Winchell, Sketches of Creation, 214. 4 Id. 218, 215. 
 5 Id. 219, 220. 6 Id. 220, 221.' 
 
Morris. 
 
 257 
 
 Greenland, in the latter, the present climate of Greenland was 
 brought as far south as the Ohio. The continent of Greenland is 
 now nearly buried under snow and ice. Precisely such must have 
 been the condition of much of the North American continent 
 during the Glacial period, for we find evidence that glaciers 
 covered the greater part of the surface, down to the latitude 
 of 38 or 40 degrees." 7 
 
 f( In America," says Mr. Morris, " the ice-sheet was of 
 enormous proportions. Whether it descended from Green- 
 land, or from the central region of British North America, is 
 still a debated question, but its lower limit is somewhat dimly 
 marked out by the line of moraine which it left behind 
 it. This lower limit extends along the Atlantic coast to 
 Long Island. . / . . The limiting line extends through 
 New Jersey in a N.W. direction, and enters Pennsylvania 
 south of the Water Gap. It passes through this State in 
 a somewhat devious course, and leaves it at a point north 
 of Pittsburg. Thence it passes through Ohio, touching the 
 river east of Cincinnati, and extending for a few miles into 
 Kentucky, whence it passes through Middle Indiana and the 
 States farther west as far as Dakota. Beyond this point the 
 moraine disappears. . . . The movement of this great glacier 
 was not due south. Its direction was S.E. over the Atlantic 
 slope. Farther west it was southward and south-westward. Its 
 lines of motion converge northward toward a region N.W. of 
 Montreal, which formed its more immediate centre of motion. . . 
 The thickness of the Atlantic glacier was in due proportion to 
 its extent. It seems to have pushed southward with little 
 conformity to the valley lines, as if every outline of elevation 
 was buried beneath its enormous depth. Indications of its 
 presence are found 5500 feet high on the White mountains, 
 nearly 3000 feet high in southern Massachusetts and the 
 Catskill region, and 1000 feet high in southern Connecticut. 
 These indicate the lowest limit of its depth, which was pro- 
 bably considerably greater, and can scarcely have been less 
 than 6000 or 6500 feet in Upper New England." After de- 
 scribing the phenomena observed all over the district, our 
 author continues : " The conditions are such as would be pro- 
 
 ' Op. cit. 85, 86 . 
 
258 The Glacial Nightmare. 
 
 duced by an ice plane of enormous weight, a mile or more 
 in thickness, with numerous sharp stones frozen into its bottom 
 and pushed onward with irresistible force over the rocks and 
 earth beneath. . . . The lines of scratches observed pursue a 
 nearly regular course over high and low lands alike, without 
 regard to valley slopes, except in the case of the great valleys 
 of the country, thus proving that no mountain range was 
 sufficiently high to seriously distort the course of the ice 
 flow." 8 
 
 When a vast ice-sheet had been postulated for North 
 America, corresponding to that of the Old World, it was a 
 very small step to combine the two, and to make out of them 
 a continuous ice-cap or ice-mound, spreading outwards from 
 the Pole, where it had its focus, in different directions, and this 
 was the next advance that was made. Here also Agassiz was 
 to the front. Thus, he says, inter alia, " As to the southward 
 movement of an immense field of ice extending over the whole 
 north, it seems inevitable, the moment we admit that snow 
 may accumulate around the Pole in such quantities as to 
 initiate a pressure radiating in every direction. Snow, 
 alternately thawing and freezing, must, like water, find its 
 level at last. A sheet of snow ten or fifteen thousand feet 
 in thickness, extending all over the northern and southern 
 portions of the globe, must necessarily lead, in the end, to the 
 formation of a northern and southern cap of ice, moving 
 towards the Equator." 
 
 Again he urges that Greenland and the Arctic regions have 
 preserved for us the remains of the ice-sheet. " Their shrunken 
 ice-fields, formidable as they seem to us, are, to the frozen 
 masses of that secular winter, but as the patches of snow and 
 ice lingering on the north side of our hills after spring has 
 opened. Let us expand them in imagination till they extend 
 over half a continent, and we shall have a sufficiently vivid 
 picture of this frozen world, and a temperature which would 
 bring the climate of Greenland down to the fortieth degree 
 of latitude would not only render the field of ice far more 
 extensive, but thousands of feet thicker than it is at 
 present. . . . How can we doubt that when, under the same 
 
 8 Encyclopaedia Americana, iii. 221, 222. 9 Id. 160. 
 
Agassiz and Belt. 259 
 
 latitude, Norway, Sweden, Scotland, England and Ireland 
 were covered by sheets of ice many thousand feet in height, 
 the ice-fields of' Greenland must have shared in the same 
 climatic influences, and have been much thicker and far more 
 extensive than they are at present ? " 
 
 Elsewhere he says that in the two hemispheres we have direct 
 reference to the polar areas and the extent of the local boulders 
 from their centre of distribution reaches levels, the mean 
 annual temperature of which corresponds in a surprising manner 
 with the mean annual temperature of the southern limit 
 of the northern erratics, which points to the phenomena of 
 local mountain erratics in Europe, and of northern erratics in 
 Europe and America, having probably been produced by the 
 same cause. The chief difficulty is in conceiving the possi- 
 bility of the formation of a sheet of ice sufficiently large to 
 carry the northern erratics into their present limits of distri- 
 bution. 1 
 
 Here also Agassiz is largely supported by Dana, who, speak- 
 ing of the southern border of the ice-cap in America, says : 
 " The glacier of the Northern Hemisphere was, therefore, not 
 a great enveloping ice- cap extending southward about the 
 sphere. On the contrary, the polar ice had a broad and lofty 
 southward prolongation toward the eastern border of the 
 continent ; another narrow and shorter one on the western 
 border ; and also a southward extension through local glaciers, 
 along the higher portion of the Eocky Mountains." 
 
 Again, Mr. Belt says : " At the present sea-level the ice 
 extended in the northern hemisphere from the Pole to lat. 39 
 in America, to about the valley of the Thames in England, to 
 lat. 50 in Central Europe, and to lat. 52 in north-western 
 Asia ; along the high lands of America it reached to the 
 tropics, and in Central America all the land lying over 2000 
 feet above the sea supported glaciers. I do not contend that 
 the present low lands of tropical America were ever covered 
 with ice, and it is on the mountain chains of that continent 
 alone that I believe it nearly reached the equator." 2 
 
 Having completed the ice-cap theory in the northern hemi- 
 sphere, let us now turn to the southern, and especially to 
 
 1 Silliman's Journal, x. 83-101. 2 Nature, x. 25. 
 
 s 2 
 
260 The Glacial Nightmare. 
 
 South America, where Darwin had long ago described the 
 boulders and other debris of ica action in Patagonia, attribut- 
 ing them, as we have seen, partially to glaciers in the Andes, 
 and partially to portage by icebergs. The country was 
 examined with great patience and care by Agassiz when he 
 was in charge of the Haslar expedition, and I shall quote from 
 his reports to the head of the United States Survey. 
 
 After describing in some detail the phenomena he observed 
 in the Straits of Magellan, he says : " It was not till we 
 rounded Cape Froward that I felt confident that the range of 
 hills immediately in sight along the channel we followed had 
 assumed their present appearance in consequence of abrasion 
 by ice. Now, however, that I have seen the whole length of 
 the Straits of Magellan, have passed through Smyth's Channel, 
 and visited Chiloe, I am prepared to maintain that the whole 
 southern extremity of the American continent has been 
 uniformly moulded by a continuous sheet of ice. Everywhere 
 we saw the rounded undulating forms so well known to the 
 students of glacial phenomena as roches moutonnees, combined 
 with the polished surfaces scored by grooves and furrows 
 running in one and the same direction ; while rocks of unequal 
 hardness, dykes traversing other rocks, slates on edges were 
 cut to one level. In short, all the surface features of the 
 Straits of Magellan have much the same aspect as the glaciated 
 surfaces of the Northern Hemisphere. Whenever the furrows 
 and scratches were well preserved their trend was north- 
 wards/' 3 
 
 Our author proceeds to point out in detail the various places 
 where he had observed the phenomena in the Straits, urging 
 that the evidence points to the planing machine having moved 
 south and north, and that the strike-side of the glacier agency 
 was upon the southern slope, and the lee-side upon the northern, 
 while nothing indicates the fitful action of glaciers. 4 In 
 sailing up the western coast of America, he tells us how " the 
 hills in the Guia narrows were everywhere distinctly moutonnees 
 to the height of about 2000 feet .... that the south side was 
 here also the strike-side was equally apparent from the facts 
 that all these marks were either wanting or less distinct on 
 
 * Nature, vi. 230. 4 Id. 
 
Agassis. 261 
 
 the north side of the islands. Had any abrading agent 
 advanced from the north, all appearances must have been 
 reversed in these narrow channels, or they must have crossed 
 them at right angles had the action come from east or west. 
 Floating ice is out of the question where the furrows are not 
 horizontal, and here in the narrowest part of the channel, west 
 of Saumarez Island, there is a track where the scratches and 
 furrows are distinctly ascending on the west side of the 
 channel, and horizontal on the eastern side opposite, showing 
 that the pressure of the ice-sheet must have been from S.E. 
 to N.W. 3 Looking south, after passing Saumarez Island, the 
 scenery appears totally different, from the fact that this is the 
 lee side of the glacier action. . . . That local glaciers, however, 
 existed, and extended much beyond their present range, may 
 be plainly seen in many of the inlets crossing the main channels 
 in an east-westerly and west-easterly direction. It is true 
 that general and local glacial phenomena are so interwoven 
 throughout this region that it is at times difficult to appreciate 
 their true connection, but there are also many localities where 
 the difference is quite obvious. ... as in Switzerland, there 
 seems to be a level above which the ice-sheet has never risen ; 
 at least, there is a line above which the mountain ridges 
 remain jagged and abrupt, while below their crests the whole 
 land is moutonnee. ... I know no more interesting locality 
 for the study of glaciers than the vicinity of Saumarez Islands. 
 It shows in the most palpable manner that glaciers only that 
 is, terrestrial masses of ice moving upon solid ground can 
 have produced these abrasions ; that floating ice cannot have 
 been the cause. Their direction is such that no one could 
 suppose the adjoining cordillera to have been in any way con- 
 nected with the abrasion or planing of the rock, or with its 
 grooving and furrowing. The country has everywhere a 
 glacier-worn aspect as far as the Gulf of Pennas. . . . The 
 erratic pebbles and boulders from the eastern to the western 
 coast of Patagonia, judging from my observations at Monte 
 Video, in the Bay of St. Matthias, in Possession Bay, at 
 Sandy Point, in all the ports of the Straits of Magellan which 
 we have visited, at Shell Bay, on entering Smyth's Channel, 
 
 5 Id. 270. 
 
262 The Glacial Nightmare. 
 
 throughout Smyth's Channel 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 
 siliceous 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 extent 
 of country through which I have traced them. . . . This fact 
 is of great significance, inasmuch as it shows that the drift 
 phenomena in this region cannot have been due to the enlarge- 
 ment of the present glaciers, otherwise 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 furrowed, like the erratics of 
 the United States and of Northern Europe. It is this uni- 
 formity 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 connected with local glaciers 
 and those belonging to what I have called the glacial period, 
 during which the two hemispheres 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 hemispheres, 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. lat., on the 
 Atlantic as well as the Pacific coast. Even at Talcahuano, 
 large erratic boulders and roches moutonnees exist at the mouth 
 of the Biobio on the hills of Hualpen." 6 Again, he says : "It is 
 certain that the local glaciers of the present time have never had 
 the power in their greatest extension, or lasted long enough 
 to obliterate or even obscure the phenomena of the glacial 
 period. To refer the latter to an enlargement of the present 
 glaciers is simply absurd. . . . After seeing the dispersion 
 of the drift in a south-northerly direction over this part of 
 the South American continent, and observing the relation of 
 
 Id. 271. 
 
Agassis. 263 
 
 the local to the general glacial phenomena, I protest anew 
 against the confusion introduced into the subject by those who 
 imagine that what I have called the glacial period was pro- 
 duced by the gradual enlargement and subsequent shrinking 
 of the glaciers now in existence." 7 
 
 In a subsequent letter, describing his journey north from 
 Talcahuano to Santiago, Agassiz describes the great valley of 
 Chilian, extending from the Gulf of Ancud or Port Montt to 
 Santiago and farther north. "This great valley/ 5 he says, 
 " extending for more than 25 of latitude, is a continuous 
 glacier bottom, showing plainly that for its whole length the 
 great southern ice-sheet has been moving northwards on it. 
 Nowhere," he adds, " could I find any indication that glaciers 
 descending from the Andes had crossed this valley and reached 
 the shores of the Pacific." 8 
 
 Agassiz was generally fearless in his logic, and he now pro- 
 ceeded to apply the theory by which he had explained the 
 drift phenomena of the northern hemisphere by a polar ice 
 cap to the southern hemisphere. Thus he writes : " After 
 having traced what seems to me palpable evidence of an ice 
 mantle overspreading once the southern part of the continent, 
 the effect of which I have seen from Monte Video on the 
 Atlantic to Talcahuano on the Pacific coast, the question 
 naturally arises how far the southern extremity of Africa as 
 well as New Zealand and Australia, were involved in the 
 extension. I hope 'I may live to see younger naturalists in- 
 vestigate these regions with the same object. I believe that 
 whenever such an investigation is undertaken by a competent 
 observer, it will be found that over and above all local 
 glacial phenomena, and side by side with them, there is also 
 evidence of a southern circumpolar glacial agency." 9 
 
 Writing in 1865 he says : "During this period two vast caps 
 of ice stretched from the northern pole southward and from 
 the southern pole northward, extending in each case far 
 towards the Equator ; and ice-fields such as now spread over 
 the Arctics, covered a great part of the temperate zones." 1 
 
 I have now described how Agassiz applied his ice-sheet 
 
 Id. 272. 8 Id. 414. ' Id. 272. 
 
 1 Geological Sketches, 212. 
 
264 The Glacial Nightmare. 
 
 theory to the southern as well as the northern hemisphere, 
 and must finally turn to the application he and others made of 
 it to the Tropics. In doing so I shall have to revert somewhat 
 in time, since his arguments in this behalf were published some 
 years before his visit to Patagonia. 
 
 He first announced his discovery in a letter to Professor 
 Peirce, dated the 27th of May, 1865. In this he says : " Yes- 
 terday was one of the happiest days of my life, and I want to 
 share it with you. Here I am at Tijuca, a cluster of hills 
 about 1800 feet high, and some seven or eight miles from Rio, 
 in a charming cottage -like hotel, from the terrace of which 
 you see a drift hill with innumerable erratic boulders as 
 characteristic as any I have ever seen in New England. I had 
 before seen sundry unmistakable traces of drift, but there was 
 everywhere connected with the drift itself such an amount of 
 decomposed rocks of various kinds, that though I could see the 
 drift and distinguish it from the decomposed primary rocks in 
 place on account of my familiarity with that kind of deposit, 
 yet I could probably never have satisfied anybody else that 
 there is here an equivalent of the northern drift, had I not 
 found yesterday, near Bennett's hotel at Tijuca, the most 
 palpable superposition of drift and decomposed rocks with a 
 distinct line of demarcation between them." After describing 
 the decomposed rocks which remain distributed and stratified 
 as they originally were, except in being disintegrated into a 
 soft paste, "a sort of rotten crust of their own substance," 
 he goes on to say that they consist of small particles of 
 primitive rocks, but without a trace of larger or smaller 
 boulders in them, while the superincumbent drift, consisting of 
 a similar paste, does not show the slightest sign of the in- 
 distinct stratification characteristic of the decomposed meta- 
 moiphic rocks below it, nor any of the decomposed dikes, but 
 is full of various kinds of boulders of various dimensions. " I 
 have not yet traced the boulders to their origin, but the 
 majority consist of a kind of green stone, composed of equal 
 amounts of a greenish black hornblende and feldspar." 
 Agassiz says an engineer had told him that the iron mines of 
 Minas Geraes are worked in stone of this kind. He goes on 
 to say that besides this drift he had not seen any trage of 
 glacial action, properly so called, if polished surfaces and 
 
Agassis. 265 
 
 scratches and furrows are especially to be considered as such. 
 Later on we read in Madame Agassiz's narrative " that it is 
 hardly possible to hope to find traces of striae and furrows where 
 the rock is so perishable and disintegrates so fast ; but this 
 much is certain, that a sheet of drift covers the country, com- 
 posed of a homogeneous paste without trace of stratification, 
 containing loose materials of all sorts and sizes embedded in it, 
 without reference to weight, large boulders, smaller stones, 
 pebbles, and the like. This drift is very unevenly distributed; 
 sometimes rising into high hills, owing to the surrounding 
 denudations ; sometimes covering the surface as a thin layer ; 
 sometimes, and especially on steep slopes, washed away com- 
 pletely, leaving the bare face of the rock ; sometimes deeply 
 gullied, so as to produce a succession of depressions and 
 elevations alternating with each other. ... A number of 
 localities exist where the drift rests immediately above 
 stratified rocks, with the boulders protruding from it, the 
 line of contact being perfectly distinct/' Our authoress then 
 goes on to point out several localities where the phenomena can 
 be well studied. 3 Going north from this to the mouth of the 
 Amazon, Agassiz verified the existence of this drift along the 
 whole length of coast, and notably at Bahia, Maceio, Pernambuco, 
 Monteiro, Parahyba doNorte, Cape St. Roque, Ceara, Maranham 
 and Para. " The sheet of drift," says the graphic reporter of 
 his journey, " which he has thus followed from Rio de Janeiro to 
 the mouth of the Amazons, is everywhere of the same geological 
 constitution. It is always a homogeneous clayey paste of a 
 reddish colour, containing quartz pebbles ; and, whatever be 
 the character of the rock in place, whether granite, sandstone, 
 gneiss or lime, the character of the drift never changes or 
 partakes of that of the rocks with which it is in contact. This 
 certainly proves that, whatever be its origin, it cannot be 
 referred to the localities where it is now found, but must have 
 been brought from a distance." 4 
 
 In going up the Amazon, we are told, the drift became more 
 conspicuous, and Agassiz devotes a long chapter to its dis- 
 cussion in that river valley, in which, he says, it differs some- 
 
 2 Agassiz, A Journey to Brazil, 86-88. 
 3 Op. cit. 99-101. 4 Id. 146, 147. 
 
266 The Glacial Nightmare. 
 
 what from the deposit farther south, in that the boulders are 
 less frequent and in showing occasional signs of stratification. 
 At one point, namely, at the Serra of Erere, which rises to 
 about a height of 900 feet on a side arm of the Amazon named 
 the Gurupatuba, Agassiz found some interesting evidence of 
 a glacial character. " On the northern flank of the Serra," 
 he says, " I found the only genuine erratic boulders I have 
 seen in the whole Amazonian valley from Para to the frontier 
 of Peru, though there are many detached masses of rock, as 
 for instance at Pedreira, near the junction of the Rio Negro and 
 Eio Branco, which might be mistaken for them, but are due to 
 the decomposition of the rocks in place. The boulders of 
 Erere are entirely distinct from the rock of the Serra, and 
 consist of masses of compact hornblende/' Then, again, he 
 makes the very interesting remark that Alpine lichens were 
 growing among the cacti and palms and a crust of Arctic crypto- 
 gamous growth covered rocks between which sprang tropical 
 flowers. 5 These erratics of Erere are sunk in the unstratified 
 mass of clay, 6 and this Eio drift, as he calls it, covers the 
 country everywhere. Agassiz says : " It may truly be said that 
 there does not exist on the surface of the earth a formation 
 known to geologists resembling that of the Amazons. Its 
 extent is stupendous ; it stretches from the Atlantic shore, 
 through the whole width of Brazil, into Peru, to the very foot 
 of the Andes. Humboldt speaks of it { in the vast plains of 
 the Amazons, in the eastern boundary of Jaen de Bracamoros,' 
 and says, ' This prodigious extension of red sandstone in the 
 low grounds stretching along the east of the Andes, is one of 
 the most striking phenomena I observed during my exam- 
 ination of rocks in the equinoctial regions/ 7 When the 
 great natural philosopher wrote these lines, he had no idea 
 
 5 Id. 418. 6 Id. 421. 
 
 7 Id. 422. Bohn's edition of Humboldt' s Personal Narrative, p. 134. 
 " Humboldt alludes to these formations repeatedly : it is true that he 
 refers them to the ancient conglomerates of the Devonian age, but his de- 
 scription agrees so perfectly with what I have observed along the banks 
 of the Amazons, that there can be no doubt he speaks of the same thing. 
 He wrote at a time when many of the results of modern geology were 
 unknown, and his explanation of the phenomena was then .perfectly 
 natural. The passage from which the few lines in the text ai^ taken 
 shows that these deposits extend even to the Llanos." 
 
Agassis. 267 
 
 liow much these deposits extended beyond the field of his 
 observations. Indeed they are not limited to the main bed of 
 the Amazons ; they Lave been followed along the banks of its 
 tributaries to the south and north as far as these have been 
 ascended. They occur on the margins of the Huallaga and 
 the Ucayale, on those of the la, the Hyutahy, the Hyurua, the 
 Hyapura and the Purus. On the banks of the Hyapura, where 
 Major Coutinho has traced them, they are found as far as the 
 Cataract of Cupati. I have followed them along the Eio 
 Negro to its junction with the Rio Branco ; and Humboldt not 
 only describes them from a higher point on this same river, 
 but also from the valley of the Orinoco. Finally they may be 
 tracked along the banks of the Madeira, the Tapajos, the 
 Xingu, and the Tocantins, as well as on the shores of the 
 Guatuina, the Trombetas and other northern affluents of the 
 Amazons. The observations of Martius, those of Gardner, and 
 the recent survey above alluded to, made by my assistant, 
 Mr. St. John, of the valley of the Eio Guruguea and that of 
 the Eio Paranahyba, show that the great basin of Piauhy is 
 also identical in its geological structure with the lateral 
 valleys of the Amazons. The same is true of the large island 
 of Marajo, lying at the mouth of the Amazons. And yet I 
 believe that even this does not cover the whole ground, and 
 that some future writer may say of my estimate, as I have 
 said of Humboldt's, that it falls short of the truth ; for, if my 
 generalizations are correct, the same formation will be found 
 extending over the whole basin of the Paraguay and the Eio 
 de la Plata, and along their tributaries, to the very heart of 
 the Andes. 
 
 " Such are the facts. The question now arises, How were 
 these vast deposits found ? The easiest answer and the one 
 which most readily suggests itself, is that of a submersion of 
 the continent at successive periods to allow the accumulation of 
 these materials, and its subsequent elevation. I reject this 
 explanation for the simple reason that the deposits show no 
 sign whatever of a marine origin. No sea-shells nor remains 
 of any marine animal have as yet been found throughout 
 their whole extent, over a region several thousand miles in 
 length and from five to seven hundred in width. It is con- 
 trary to all our knowledge of geological deposits to suppose 
 
268 The Glacial Nightmare. 
 
 that an ocean basin of this size, which must have been sub- 
 merged during an immensely long period in order to accumu- 
 late formations of such thickness, should not contain numerous 
 remains of the animals formerly inhabiting it. ... It is my 
 belief that all these deposits belong to the ice period in its 
 earlier or later phases, and to this cosmic winter which, 
 judging from all the phenomena connected with it, may have 
 lasted for thousands of centuries, we must look for the key to 
 the geological history of the Amazonian valley. I am aware 
 that this suggestion may appear extravagant. But is it, after all, 
 so improbable that when Central Europe was covered with 
 ice thousands of feet thick ; when the glaciers of Great Britain 
 ploughed into the sea, and when those of the Swiss mountains 
 had ten times their present altitude ; when every lake in 
 Northern Italy was filled with ice, and these frozen masses 
 extended even into Northern Africa ; when a sheet of ice, 
 reaching nearly to the summit of Mount Washington in the 
 White Mountains, that is having a thickness of nearly 6000 
 feet, moved over the continent of North America is it so im- 
 probable that in this epoch of universal cold, the valley of the 
 Amazons also had its glacier poured down into it from the 
 accumulations of snow in the Cordilleras, and swollen laterally 
 by the tributary glaciers descending from the tablelands of 
 Guiaica ? The movement of this immense glacier would be 
 eastward, and determined as well by the vast reservoirs of 
 snow in the Andes as by the direction of the valley itself. It 
 must have ploughed the valley bottom over and over again, 
 grinding all the materials beneath it into a fine powder, or 
 reducing them to small pebbles, and it must have accumulated 
 at its lower end a moraine of proportions as gigantic as its own ; 
 thus building a colossal sea wall across the mouth of the valley. 
 I shall be asked at once whether I have found here also the 
 glacial inscriptions, the furrows, striae and polished surfaces 
 so characteristic of the ground over which glaciers have 
 travelled. I answer, not a trace of them; for the simple 
 reason that there is not a natural rock surface to be found 
 throughout the whole Amazonian valley. The rocks them- 
 selves are of so friable a nature, and the decomposition caused 
 by the warm torrential rains and by exposure to the'-burning 
 sun of the tropics so great and unceasing, that it is hopeless 
 
Agassis. 269 
 
 to look for marks which in colder climates and on harder 
 substances are preserved through ages unchanged. With the 
 exception of the rounded surfaces so well known in Switzerland 
 as the roches moutonnees, heretofore alluded to, which may be 
 seen in many localities, and the boulders of Erere, the direct 
 traces of glaciers as seen in other countries are wanting here. 
 I am, indeed, quite willing to admit that, from the nature of 
 the circumstances, I have not here the positive evidence which 
 has guided me in my previous glacial investigations. My 
 conviction in this instance is founded, first, on the materials 
 in the Amazonian valley, which correspond exactly in their 
 character to materials accumulated in glacier bottoms ; 
 secondly, on the resemblance of the upper or third Amazonian 
 formation to the Kio drift, of the glacial origin of which there 
 cannot, in my opinion, be any doubt ; thirdly, on the fact that 
 this fresh-water basin must have been closed against the sea 
 by some powerful barrier, the removal of which would 
 naturally give an outlet to the waters, and cause the extra- 
 ordinary denudations the evidences of which meet us every- 
 where throughout the valley. 3 ' 8 
 
 In regard to this Kio drift, Agassiz says : ( f As I stated in the 
 beginning, I am satisfied that the unstratified clay deposit of 
 Rio and its vicinity is genuine glacial drift resulting from the 
 grinding of the loose materials interposed between the glacier 
 and the solid rock in place, and retaining to this day the 
 position in which it was left by the ice. Like all such 
 accumulations it is entirely free from stratification. On the 
 other hand, the ochraceous sandy clay of the Amazon valley, 
 while it was originally ground by glaciers, in the upper part 
 of the valley, these materials have subsequently been spread 
 throughout the whole basin and actually deposited under the 
 agency of water." 9 
 
 Under the low temperatures of the ice period, he urges, the 
 climatic conditions necessary for the formation of land ice 
 existed in the valley of the Amazons, and it was actually 
 filled with an immense glacier, whose terminal moraine, he 
 argues, closes the Amazon valley on the east towards the sea. 
 " When these fields of ice yielded to a gradual change of climate 
 and slowly melted away, the whole basin, then closed against 
 8 Id. 425, 426. 9 Id. 426, 427, note. 
 
2 7O The Glacial Nightmare. 
 
 the sea by a huge wall of debris was transformed into a vast 
 fresh-water lake. The first effect of the thawing process must 
 have been to separate the glacier from its foundation, raising 
 it from immediate contact with the valley bottom, and thus 
 giving room for the accumulation of a certain amount of water 
 beneath it, while the valley as a whole would still be occupied 
 by the glacier. ... It is at this time and under such circum- 
 stances that I believe the first formation of the Amazonian 
 valley, with the coarse pebbly sand beneath and the finely 
 laminated clays above, to have been formed, when the greater 
 part of the ice was melted and the basin was filled with water. " 
 It was while thus partially or wholly filled with water that 
 Agassiz supposes the boulders of Erere to have been " carried 
 by floating ice when nothing remained of the ice-fields except 
 such isolated masses ice-rafts, as it were, or perhaps by ice- 
 bergs dropped into the basin from glaciers still remaining in 
 the Andes and on the edges of the plateaus of Guiana and 
 Brazil." 
 
 He lastly urges that the gigantic moraine which closed the 
 valley was broken through either by some violent movement 
 inside or by the sea outside, whereupon the waters rushed 
 through, denuding and sweeping away a good deal of the 
 deposit and leaving some outliers alone remaining. 
 
 Agassiz was not content with visiting the Amazon valley ; 
 he went back to the province of Rio to try and find traces 
 of glaciers in its mountain ranges or serras, and also traces of 
 the southern lateral moraine marking the limit of ice which 
 he supposed once filled the Amazonian basin. Inter alia, he 
 learnt from a land surveyor, named Dr. Felice, that there is a 
 wall of loose materials, boulders, stones, etc., running from 
 east to west for a distance of some sixty leagues from the Rio 
 Aracaty-Assu to Bom Jesu in the Serra Grande. From his 
 account, says Agassiz, this wall resembles greatly the " Horse- 
 backs " in Maine, those remarkable ridges accumulated by the 
 ancient glaciers, and running sometimes uninterruptedly for 
 thirty or forty miles. The horsebacks are, however, covered 
 with soil and turf, whereas Dr. Felice describes this wall as 
 rough and bare. 1 Agassiz was not able to visit this moraine, 
 
 1 Id. 447, 448. 
 
Agassiz. 2 7 1 
 
 but lie did visit more than one serra in the province of Eio. 
 Thus he says of the Serra of Monguba : " All the valleys 
 have had their glaciers, and these valleys (? glaciers) have 
 brought down from the hillsides into the plains boulders, 
 pebbles, and debris of all sorts. " 2 Of the Serra of 
 Aratauha, he says : " The glacial phenomena are as legible 
 as in any of the valleys of Maine or in those of the 
 mountains of Cumberland in England. It had evidently 
 a local glacier, formed by the meeting of two arms, which 
 descended from two depressions spreading right and left on 
 the upper part of the serra, and joining below in the main 
 valley. A large part of the medial moraine formed by the 
 meeting of these two arms can still be traced in the central 
 valley. One of the lateral moraines is perfectly preserved, 
 the village road cutting through it; while the village itself 
 is built just within the terminal moraine, which is thrown up 
 in a long ridge in front of it." 3 Agassiz tells us he specially 
 examined the right lateral moraine of this serra to ascertain its 
 real character. " Everywhere/' he says, "in the ridges encircling 
 the depressions on the serra, the loose materials and large 
 boulders are so accumulated and embedded in clay and sand 
 that their morainic character is unmistakable. Occasionally, 
 where a ledge of the underlying rock crops out, in places where 
 the drift has been removed by denudation, the difference 
 between the moraine and the rock decomposed in place is 
 recognized at once. It is equally easy to distinguish the 
 boulders which here and there have rolled down from the 
 mountain and stopped against the moraine. The three things 
 are side by side, and might at first be easily confounded ; but 
 a little familiarity makes it easy to distinguish them. Where 
 the lateral moraine turns towards the front of the ancient 
 glacier near the point at which the brook of Patacuba cuts 
 through the former, and a little to the west of the brook, there 
 are colossal boulders leaning against the moraine, from the 
 summit of which they have probably rolled down. Near the 
 cemetery the front moraine consists almost entirely of small 
 quartz pebbles ; there are, however, a few larger blocks 
 among them. The medial moraine extends nearly through the 
 
 2 Id. 454. Id. 456, 457. 
 
272 The Glacial Nightmare. 
 
 centre of the village, while the left-hand lateral moraine lies 
 outside of the village, at its eastern end, and is traversed by 
 the road leading to Ceara. It is not impossible that eastwards 
 a third tributary of the serra may have reached the main 
 glacier of Pacatuba. I may say that in the whole valley of 
 Hasli there are no accumulations of moraine materials more 
 characteristic than those I have found here, not even about 
 the Kirchet ; neither are there any remains of the kind more 
 striking about the valleys of Mount Desert in Maine, where 
 the glacial phenomena are so remarkable, nor in the valleys 
 of Lough Fine, Lough Augh, and Lough Long in Scotland, 
 where the traces of ancient glaciers are so distinct. In none 
 of these localities are the glacial phenomena more legible than 
 in the Serra of Aratanha." 4 
 
 Agassiz tells us that two of his subordinates, Mr. F. C. 
 Hartt, whom he sent to the province of Spiritu Santo, the 
 valley of Rio Doce, and that of Mucury, and Mr. 0. 
 St. John, who went into the interior in the north and visited 
 the valleys of the Rio San Francisco and the Bio das Velhas, 
 both experienced geologists found similar phenomena. 5 
 Mr. Hartt, who revisited Brazil the next year, wrote an 
 elaborate work on the " Geology and Physical Geography of 
 Brazil." In this, while differing from Agassiz as to the strati- 
 fied beds of the Amazon valley, he is careful to explain " that 
 these views do not affect Agassiz' s theory of the former exist- 
 ence of glaciers under the tropics, down to the present level 
 of the sea ; a theory which," he adds, ' ' I hold as firmly as he." 
 This testimony is the more valuable as he says : " I desire to 
 record the fact that I began my studies of the Brazilian drift 
 with a conviction that Professor Agassiz was wrong, and I feel 
 much gratified that my independent observations have so fully 
 confirmed the results of his own." G 
 
 Speaking of the valley of Tijuca, he says : "If we follow 
 down the valley we shall find the soil full of boulders, and 
 some of them are many feet in diameter. As we descend the 
 valley still further these boulders are seen lying bare, not only 
 in the brook, where the water has washed away the loose 
 material, but on the hillsides. I think no geologist familiar 
 
 4 Id. 463, 464. 5 Id. 404, 405. 6 Op. cit. 493. 
 
Hartt. 273 
 
 with drift phenomena, who should suddenly find himself in 
 this valley, would have even the slightest suspicion of there 
 being anything else than the most ample testimony of the 
 former prevalence of glacial action over the region. ... If 
 one descends the valley towards the Cascate Grande, he will 
 see that the valley is heaped with a confusion of immense 
 boulders tumbled one upon another ; masses of greenstone, 
 weighing hundreds of tons, piled up with those of gneiss of all 
 qualities. Where these are bare they are all rounded, but I 
 believe this is referable, to a large extent, to a concentric and 
 even decomposition of the surface ; but there is no resisting 
 the conclusion that we have here a inorainic deposit from a 
 glacier which anciently occupied the valley. . . . Descending 
 to the plain below the Cascate, one traces the drift clays and 
 boulders quite down to the plain, where they end abruptly, and 
 the flat lands are seen to be of alluvial origin resting on sea 
 sands. . . . From Belem to the Parahyba River the same red 
 clay entirely covers the surface, lying even on very high slopes. 
 Nowhere is there the slightest trace of stratification, and it is 
 sharply defined from the alluvial deposits of the river. The 
 same pebble sheet is seen almost everywhere, though in cutting 
 after cutting it may sometimes be wanting. Boulders are 
 rare, and are almost invariably so decomposed as to be seen 
 only in fresh cuttings. The rock on which the drift rests is 
 always smoothly and evenly rounded down." 7 Speaking of 
 the surface deposits of Bahia and the Sao Francisco, he 
 says : " I must insist upon the fact that the unarranged 
 materials are precisely like our unmodified drift in the north, 
 and that the surface of the rock on which they lie has the 
 moulding of the surfaces on which our northern drift lies ; 
 and that if we refer the northern drift to the action of glacier 
 ice, we must do the same thing for the Brazilian surface 
 detritus ; contrary as it is to all our preconceived opinions of 
 the distribution of the drift. The fact that neither Professor 
 Agassiz nor myself, nor any one else of our expedition, has been 
 able to discover glacial striae in Brazil is of very secondary 
 importance. The drift itself exists all over the country, and 
 it cannot be explained away. ... I can offer but one ex- 
 
 7 Id* 29, 31. 
 
274 The Glacial Nightmare. 
 
 planation of the formation of the pot-holes observed by 
 Mr. Alien,, and that is that they were formed by glacial 
 cascades in the same way as the pot-holes seen so often on the 
 surface of ridges in the north have been formed during the 
 drift ; for, according to the testimony of Mr. Allen, the pot- 
 holes of the province of Bahia occur on the gneiss plains, far 
 away from any present obstacle over which the water may 
 have flowed. Mr. Allen describes them as being exceedingly 
 well preserved, and having smooth sides. ... I believe that 
 during the time of the drift the country stood at a much higher 
 level than at present, and that it was covered by a general 
 glacier. 3 ' 8 
 
 The case for the former extension of a glacial age to the 
 tropics was further pressed by Mr. Belt in his well-known work 
 " The Naturalist in Nicaragua." In it he speaks of unstratified 
 beds of gravel being exposed in numerous natural sections on 
 the banks of the river about a mile below Depilto, in Nicara- 
 gua. " These beds/ 7 he says, " deepen as the river is descended, 
 until at Ocotal they reach a thickness of between two and 
 three hundred feet. . . . These unstratified deposits consist 
 mostly of quartz sand with numerous angular and subangular 
 blocks of quartz and talcose schist. Many of the boulders were 
 very large, and in some parts great numbers have been accu- 
 mulated in the bed of the river by the washing away of the 
 smaller stones and sand. Some of these huge boulders were 
 fifteen feet across, the largest of them lying in the bed of the 
 river two miles below Depilto. Most of them were of the 
 Depilto quartz rock and gneiss, and I saw many in the un- 
 stratified gravel near Ocotal fully eight miles from their parent 
 rock. Near Ocotal this unstratified formation is nearly level, 
 excepting where worn into deep gulches by the existing 
 streams. The river has cut through to a depth of over 200 
 feet, and there are long precipices of it on both sides, similar 
 to those near streams in the north of England, that cut through 
 thick beds of boulder clay." 1 
 
 Elsewhere Mr. Belt tells us that some time after leaving San 
 Eafael, he found a steep ridge fully 1200 feet high, composed 
 entirely of boulder clay. " This clay was of a brown colour, 
 
 8 Hartt, Geology and Physical Geography of Brazil, 570-573, 
 1 Op. cit. 259, 260. 
 
Belt. 
 
 275 
 
 and full of angular and subangular blocks of stone of all sizes 
 up to nine feet in diameter. . . . This boulder clay had 
 extended all the way from San Kafael, and ranges of hills 
 appeared to be composed entirely of it. The angular and sub- 
 angular stones that it contained were an irregular mixture of 
 different varieties of trap, conglomerate and schistose rocks. 
 In the northern States of America it would be unhesitatingly 
 ascribed to the action of ice ; but I was at the time unprepared 
 to believe that the glacial period could have left such memorials 
 of its existence within the Tropics, at not greater elevations 
 above the sea than 3000 feet. 2 . . . The evidences of glacial 
 action between Depilto and Ocotal were, with one exception, 
 as clear as in any Welsh or Highland valley. There were the 
 same rounded and smoothed masses of rock, the same moraine- 
 like accumulations of unstratified sand and gravel, the same 
 transported boulders that could be traced to their parent rocks 
 several miles distant. The single exception was, I am con- 
 vinced, one of observation and not one of fact. ... I saw no 
 glacial scratches on the rocks ; but geologists know how rare 
 these are on natural exposures in districts that have certainly 
 been glaciated, and will not be surprised that, in a hurried 
 visit of only a few hours, I should not have discovered any. 
 Glacial scratches are seldom preserved on rock surfaces exposed 
 to the action of the elements. Even in Nova Scotia, where 
 scratches and grooves are met with wherever the rock surface 
 has been recently laid bare, I do not remember having ever 
 seen any on natural exposures. It is only where protected by 
 a covering of clay or gravel from the action of the elements, 
 that they have been preserved through the ages that have 
 passed since the glacial epoch ; and as I did not see any rock 
 surfaces near Depilto that had been recently bared, it is not 
 surprising that, notwithstanding the other proofs of glacial 
 action, I should not have seen any ice scratches or grooves. 
 
 " I could no longer withstand the evidence that had been 
 gradually accumulating of the presence of large glaciers 
 in Central America during the glacial period, and these 
 once admitted afforded me a solution of many phenomena 
 that had before been inexplicable. The immense ridges of 
 boulder clay near San Rafael, the long hog-backed hills near 
 
 2 Op. cit. 247, 248. 
 T 2 
 
276 The Glacial Nightmare. 
 
 Toblason, the great transported boulders two leagues beyond 
 Libertad on the Juigalpa road, and the scarcity of alluvial 
 gold in the valleys of Santo Domingo, could all be easily 
 explained on the supposition that the ice of the glacial period 
 was not confined to extra tropical lands, but in Central 
 America covered all the higher ranges, and descended in great 
 glaciers to at least as low as the line of country now standing 
 at 2000 feet above the sea, and probably much lower." 3 
 
 Mr. Belt then goes on to account for the scarcity of alluvial 
 gold in the valleys of Nicaragua as in the glaciated valleys of 
 Nova Scotia and North Wales by the potency of glaciers in 
 scooping out all the contents of the valleys. " When," he says, 
 " the denuding agent was water, the rocks were worn away and 
 the heavier gold left behind at the bottom of the alluvial 
 deposits ; but when the denuding agent was glacier ice the 
 stony masses and their metallic contents were carried away, or 
 mingled together in the unassorted moraines. 
 
 He then proceeds : " That the transportation of boulders in 
 Nicaragua was due to glaciers, and not to floating icebergs, 
 may be argued on zoological grounds. The transported 
 boulders near Ocotal, are about 3000 feet above the 
 sea, those near Libertad about 2000 feet. The low pass 
 between the Atlantic and the Pacific oceans, through the 
 valley of San Juan and the lake of Nicaragua, is less than 200 
 feet above the sea, and to allow for the flotation of 
 icebergs at the lower of the two places named, a channel of 
 more than 1800 feet in depth would have connected the 
 two oceans. This supposition is negatived by the fact that 
 the mollusca on the two coasts separated by the narrow 
 Isthmus of Darien, are almost entirely distinct, whilst we know 
 that since the glacial period there has been little change in the 
 molluscan fauna, nearly if not all the shells found in glacial 
 deposits still existing in neighbouring seas. . . . No zoologist 
 of note believes that there has been any submergence of the 
 land lying between the Pacific and the Atlantic since the 
 pliocene period, and icebergs would not have floated without 
 such submergence, so that in the cases I have mentioned the 
 boulders, if ice-borne, have been carried by glaciers and not 
 by floating ice." 4 " While I thus found evidences of -the ice 
 
 3 Op. cit. 261, 262. 4 Id. 263, 264, 
 
Belt. 277 
 
 of the glacial period reaching in the Northern Hemisphere to 
 within the tropics, in the Southern Hemisphere Professor Hartt 
 has found glacial drift extending from Patagonia all through 
 Brazil to Pernambuco, and Agassiz has even announced the 
 discovery of glacial moraines up to the Equator. I have 
 myself seen, near Pernarnbuco, and in the province of 
 Maranham, in Brazil, a great drift deposit that I believe to be 
 of glacial origin." 5 
 
 In answer to the natural question of what became of the 
 tropical fauna and flora of America at this time, such, for in- 
 stance, as the Heliconinae and Morphinse, groups of butterflies 
 peculiar to tropical America, containing many distinct genera 
 which, on any theory of descent from a common progenitor, 
 must have originated ages before the glacial period " How 
 is it," says Mr. Belt, "that such peculiar tropical groups 
 were not exterminated by the cold of the glacial period, 
 or, if able to stand the cold, that they did not spread into 
 temperate regions on the retreat of the ice ? I believe the 
 answer is that there was much extermination during the glacial 
 period, that many species and some genera, as for instance the 
 American horse, did not survive it, and that some of the great 
 gaps that now exist in natural history were then made, but 
 that a refuge was found for many species on lands now below 
 the ocean, that were uncovered by the lowering of the sea 
 caused by the immense quantity of water locked up in frozen 
 masses on the land." 
 
 Elsewhere he says, " The existence of a glacial period, how- 
 ever much derided when first announced, is now a recognized 
 fact. The divergence of opinion respecting it is limited to a 
 question of extent ; and after my recent journey in the Amazons, 
 I am led to add a new chapter to the strange history of glacial 
 phenomena, taken from the Southern Hemisphere, and even 
 from the Tropics themselves. I am prepared to find that the 
 statement of this new phase of the glacial period will awaken 
 among my scientific colleagues an opposition even more violent 
 than that by which the first announcement of my views on this 
 subject was met. I am, however, willing to bide my time ; 
 feeling sure that, as the theory of the ancient extension of 
 
 5 Id. 264, 265. 
 
2 78 Ths Glacial Nightmare. 
 
 glaciers in Europe has gradually come to be accepted by 
 geologists, so will the existence of like phenomena, both in 
 North and South America, during the same epoch, be recog- 
 nized sooner or later as part of a great series of physical 
 events extending over the whole globe. Indeed, when the ice 
 period is fully understood, it will be seen that the absurdity 
 lies in supposing that climatic conditions so different could 
 be limited to a small portion of the world's surface. If the 
 geological winter existed at all, it must have been cosmic ; and 
 it is quite as rational to look for its traces in the Western as 
 in the Eastern hemisphere, to the south of the equator as to 
 the north of it. Impressed by this wider view of the subject, 
 confirmed by a number of unpublished investigations which I 
 have made during the last three or four years in the United 
 States, I came to South America, expecting to find in the 
 tropical regions new evidences of a bygone glacial period, 
 though, of course, under different aspects. Such a result 
 seemed to me the logical sequence of what I had already 
 observed in Europe and in North America." 6 
 
 Such, then, is a sample of the evidence for the existence in 
 the Tropics of phenomena which in the Northern Hemisphere 
 are held to prove the existence of a glacial age. These are not 
 the casual opinions of dilettanti, but the matured views of 
 experienced geologists, who have spent months in investigating 
 them. Mr. Alfred Wallace, himself among the foremost 
 authorities upon tropical natural history, thus speaks of the con- 
 clusions of Agassiz and Hartt about the Brazilian drift. " This 
 is truly a startling conclusion, and one which has hitherto 
 been received in this country with some incredulity. Professor 
 Agassiz was thought to be glacier-mad ; but if we separate his 
 theories from his facts, and if we carefully consider the addi- 
 tional facts and arguments adduced by Mr. Hartt, we shall be 
 bound to conclude that, however startling, the theory of the 
 glaciation of Brazil is supported by a mass of evidence which 
 no unprejudiced man of science will ignore merely because it 
 runs counter to all his preconceived opinions." Again he says, 
 '*' It can hardly be maintained that the discoverer of glacial 
 phenomena in our own country, and who has since lived in 
 
 ^*w - 
 6 Id. 398, 399. 
 
Agassiz. 279 
 
 such a pre-eminently glaciated district as the Northern United 
 States, is not a competent observer ; and if the whole series of 
 phenomena here alluded to have been produced without the 
 aid of ice, we must lose all confidence in the method of 
 reasoning from similar effects to similar causes which is the 
 very foundation of modern geology. . . : The facts as stated 
 by two careful observers, both thoroughly experienced in 
 glacial phenomena, are undoubtedly such as to warrant the 
 main conclusion drawn by them, and it is to be hoped that 
 geologists will not ignore the facts because the conclusions 
 seem improbable, as they so long ignored facts proving the 
 antiquity of man for no other reason." 7 
 
 It is not wonderful that in view of generalizations like these 
 our breath is taken away, and that Agassiz had doubts about 
 his contemporaries and made appeals to the verdict of posterity 
 to justify him. Thus, in a letter already quoted which he wrote 
 to Professor Benjamin Peirce, he says, " You may think that I 
 have given you too many details. I have done so purposely, that 
 no one may accuse me of basing theories on imperfect observa- 
 tions. I am well aware that my results will be questioned, and 
 I shall be thought fanciful by geologists of all schools, as I have 
 been at every step of my glacial researches. But a*n old 
 hunter does not take the track of a fox for that of a wolf. I 
 am an old hunter of glacial tracks, and I know the footprint 
 whenever I find it." 8 
 
 7 Nature, ii. 511, 512. 8 Id- 272, 
 
280 The Glacial Nightmare. 
 
 CHAPTER VIII. 
 
 ON THE ALLEGED RECURRENCE OF GLACIAL EPOCHS, AND ON 
 SUPPOSED INTER-GLACIAL BEDS. 
 
 "From the Laurentian epoch down to the present day, all the 
 physical events in the history of the earth have varied neither in kind 
 nor intensity from those which we now have experienced." RAMSAY, 
 Address to the British Association, 1880. 
 
 The Champions of former glacial periods Pre-Cambrian, Hicks, Sir A. 
 G-eikie Cambrian, James Thompson Silurian, Carrick-Moore, 
 Lyell, J. Geikie, Dawson Devonian, Gumming, Eamsay, Btheridge, 
 Selwyn, J. Geikie Carboniferous, J. Geikie, Haughton, G. Austen, 
 Stirrup, Weiss, Orton, Dawson, W. T. Blandford, T. Hughes, 
 Waagen, Oldham Permian, Eamsay, Hall, Sutherland, Stow 
 Trias, Godwin Austen, Eamsay, Wilkinson Oolite, Eamsay, Judd 
 Cretaceous, Godwin Austen, Escher von der Linth Eocene, 
 Croll, J. Martins, Lydekker Miocene, Gastaldi, Eamsay. 
 Geikie, J. Martins, Stuart Menteath, Tardy Pliocene, Desor and 
 Stopanni Champions of inter-glacial beds, Yenetz, Agassiz, 
 Muhlberg, Bach, Tarancelli, Morlot, Chavenne, Scipion Gras, Heer, 
 Sir A. Geikie, Bennie, Searles Wood and Geikie, Penck, Tornebohm, 
 McGee, Whittlesey, Orton, Winchell. 
 
 THE champions of the new views in regard to the Glacial 
 epoch were also, for the most part, champions of Uniformity. 
 Champions, that is to say, of the doctrine embodied in the 
 heading to this chapter, that Nature has always used precisely 
 the same tools and employed them in precisely the same way. 
 It was natural, therefore, that they should begin to look round 
 and try and find traces of former periods when ice was wide- 
 spread and when the results of its handiwork were similarly 
 widely scattered. It was contrary to much modern teaching 
 to suppose that an event like the glacial epoch was solitary 
 and detached and not a link in a great chain in which the 
 astronomical and other forces which influence the earth 
 produced recurrent phenomena. 
 
 Especially in late years has the evidence been collected to 
 prove, first, that evidences of other glacial epochs can be found 
 in the older rocks, and, secondly, that the glacial epoch itself 
 
Hicks. 281 
 
 was no uniform period of continuous cold, but was broken by 
 intervals during which temperate conditions existed. I do 
 not propose to collect an exhaustive series of the examples 
 which have been thus adduced, but only to bring together 
 some samples of them, and I think it better to go strati- 
 graphic ally through the different beds rather than to give a 
 chronological account of the successive discoverers. 
 
 In 1876 Hicks published a paper in the Geological 
 Magazine on the Northern Palaeozoic Kocks. In this, speaking 
 of the Pre- Cambrian land, he says : " This condition of 
 continents of great size in high latitudes with elevated 
 plateaux and high mountains, would lead us to believe that 
 they were covered with ice and snow in their higher parts, 
 and that the plains and valleys had therefore much loose 
 material strewn over them. In the Cambrian rocks in Wales 
 boulders of considerable size, sometimes a foot in diameter, 
 are often found in beds of conglomerate, and the frequent 
 alternations of conglomerate grits, and sandstones, seem to 
 prove that, at least in the earliest epochs, an abundance of 
 loose material must have been near and ready at hand to be 
 easily denuded off as each part became submerged." After 
 arguing against this having been caused by marine currents, 
 he says that some additional force must have been at work, 1 
 and he concludes that the early Cambrian was a cold epoch, 
 as evidenced from the fact that the Pre-Cambrian continents 
 occupied very extensive areas in the higher latitudes, and that 
 they were traversed by mountainous ranges attaining in some 
 cases to great heights. At no time since, unless in the glacial 
 period, does there seem to have been so much high land in 
 the higher latitudes, and it is, therefore, reasonable to suppose 
 that in the earlier stages at least of the epoch the climate was 
 one of great cold. 2 
 
 In 1880, in a paper on Pre-Cambrian Volcanoes and Glaciers, 
 the same author says: " Evidences showing that glacial 
 phenomena prevailed in pre-Cambrian times were pointed out 
 by me in 1876, and there can, I think, be little doubt that 
 glaciers did, even at that early period, spread over some of the 
 higher lands." As evidence he points to the regular character 
 
 1 Geol. Mag. N.S. id. 157. 2 Id. 252. 
 
282 The Glacial Nightmare. 
 
 of the masses in the Cambrian rocks which rest immediately 
 on the old land in the Highlands of Scotland. "... In Wales 
 also the Cambrian conglomerates, when they rest on the old 
 floor, contain very large angular masses, though I have never 
 recognized blocks as large there as those in the angular 
 breccia which is presumably of Cambrian age, near Gareloch." 
 And he finally concludes that glacial phenomena have prevailed 
 since at least later pre-Cambrian times in proportion as the 
 geographical changes have been favourable or otherwise. 3 
 
 In 1880 Ramsay quotes a letter from Sir A. Geikie, in which 
 he informed him that he had discovered mammillated moutonnee 
 surfaces of Laurentian rocks passing underneath the Cambrian 
 sandstones of the north-west of Scotland at intervals all the 
 way from Cape Wrath to Loch Torridon, for a distance of about 
 ninety miles. " The mammillated rocks are as well rounded off 
 as any recent roche moutonnee, and in one place these bosses 
 are covered by a huge angular breccia of this old gneiss (Lau- 
 rentian) with blocks sometimes five or six feet long. ... In 
 the Gareloch district this breccia is utterly unstratified, the 
 angular fragments standing on end and at all angles, just as 
 they do in many modern morainic mounds wherever large 
 glaciers are found." 4 In a paper in Nature, published in the 
 same year, 5 Geikie argues that these mammillated and polished 
 surfaces, like the accompanying breccia, belong distinctly 
 to Pre-Cambrian times ; and he concludes thus : " On these 
 far northern shores, then, there still remain fragments 
 of the surface on which our oldest sedimentary accumula- 
 tions were deposited. These fragments are found to 
 bear on their smoothed hummocky contours a striking 
 resemblance to the surface which geologists now always 
 associate with the action of glacier ice. There can be no 
 doubt at least that they are denuded surfaces. The edges of 
 the vertical and twisted beds of gneiss and schist have been 
 smoothly bevelled off. These rocks, however, would never 
 have assumed such a contour if exposed merely to ordinary 
 sub-aesial disintegration. They would have taken sharp 
 craggy outlines like those which are here and there gradually 
 
 8 Id. N.S. vol. vii. 490, 491. 
 
 4 Eamsay, Address to the British Association. y ^ '- 
 
 5 Nature, vol. xxii. p. 400, etc. 
 
Thompson and Car rick- Moore. 283 
 
 replacing the ice-worn curves of the roches moutonnees. They 
 have certainly been ground by an agent that has produced 
 results which, if they were found in a recent formation, would, 
 without hesitation be ascribed to land ice. The breccia, too, 
 is quite comparable to moraine stuff/' 6 
 
 In 1870 Mr. James Thompson described certain pebbles and 
 boulders of granite in schistose rocks in Islay in Scotland, 
 which he assigned to Cambrian age. These fragments were 
 some of them angular but mostly rounded, of all sizes, from 
 mere particles to great boulders of granite, resembling the 
 granite of the Isle of Mull. Similar rocks do not occur in 
 Islay, and Mull is at a considerable distance to the north, 
 while there is a deep sea between the two islands, and he 
 suggested the probability of the granite having been trans- 
 ported by the agency of ice. These deposits, he says, 
 resemble the boulder drifts of more recent times, in the 
 following respects : 1. In the absence of stratification in one 
 part of the section, which in another shows signs of regular 
 deposition. 2. In the close proximity of fragments of 
 far-transported rock, varying in size from minute fragments 
 to large boulders. He suggests that the mass had been 
 deposited in a tranquil sea of mud, sand, and blocks from 
 melted drift ice. " The discovery," he says, " of fragments 
 and boulders of granite embedded in these deposits furnish 
 adequate proof that glacial action was not limited to any 
 special geological period/' 
 
 Turning from the Cambrian beds to the next horizon, 
 we find that as far back as 1848 Carrick- Moore described 
 certain conglomerates occurring in the Silurian beds of 
 Wigtonshire in a way which made subsequent writers, such as 
 Dr. Croll, assign them to a glacial origin. Thus he wrote : 
 " These vertical and highly inclined beds near the Corswall 
 Lighthouse are very remarkable from their containing beds of 
 conglomerate of a coarser nature than any I have ever seen 
 described as occurring in so old a formation, with the excep- 
 tion of those of the Potsdam sandstone, described by Mr. 
 Lyell as Old Silurian. The fragments generally vary from the 
 size of one inch to a foot in diameter, but, in some of the beds, 
 
 6 Nature, p. 403. 7 Brit. Assoc. Eep. 1870, p. 88. 
 
284 The Glacial Nightmare. 
 
 boulders of three, four, and even five feet diameter occur. 
 They are well rounded, and principally consist of red quartzi- 
 ferous porphyry and a large-grained grey syenite ; but serpen- 
 tine, red jasper and other rocks occur, and I have found one 
 or two instances of large angular fragments of greywacke. 
 There are no rocks in the neighbourhood, as far as I know, 
 from whence any of these rounded fragments could have been 
 derived." 8 
 
 In regard to the boulders in the Potsdam limestone just 
 referred to, Lyell says : " The upper part of the Potsdam 
 limestone at the falls of Montmorenci is remarkable for con- 
 taining boulders of enormous size, the largest I can remember 
 to have seen in any ancient stratified rock. I measured some 
 of them which were eight feet long. 1 " 9 
 
 In his " Great Ice Age," Mr. J. Geikie says : " In the 
 Lower Silurian of the south of Scotland (Glen App and 
 Dalmellington), we find large blocks and boulders (from 
 one foot to five feet in diameter) of gneiss, syenite, granite, 
 etc., none of which belong to the rocks of that neighbour- 
 hood. Indeed, no such rocks of older age than the Silurian 
 occur nearer than the Laurentian rocks of the north-western 
 Highlands and islands. Possibly the boulders may have 
 come from some ancient Atlantis ; and considering the great 
 size of the blocks and the considerable distance they may 
 have travelled, it is not unreasonable to conjecture that ice 
 may have had something to do with their transport. Similar 
 boulder-beds have been found in Ireland." * 
 
 Principal Dawson records a somewhat similar instance in 
 the Lower Silurian of Maimanse, Lake Superior, where a con- 
 glomerate occurs with boulders two feet in diameter. Again, 
 in the Upper Silurian of Nova Scotia, the same author has 
 detected beds of angular stones and chips, " the materials of 
 which seem precisely similar to that which is at present pro- 
 duced by the disintegrating action of frost on hard, and 
 especially schistose and jointed rocks." 2 
 
 Passing on again to another horizon, namely the Devonian, 
 Mr. Gumming, in his " History of the Isle of Man," pub- 
 
 8 Quart. Journ. Geol. Soc. v. 10. 9 Lyell's Travels, ii. 125. 
 
 1 Great Ice Age, 568. 
 
 2 Great Ice Age, 568 ; Canadian Naturalist, ii. 6 ; vi. 416. 
 
Ramsay, Taylor^ and Etheridge. 285 
 
 listed in 1848, compares the old red sandstone conglomerate 
 there to a consolidated ancient boulder clay formation, and 
 he says, inter alia : " Was it so, that those strange trilobite 
 looking fishes of that sera (the Coccosteus, Pterichthys and 
 Cephalaspis), had to endure the buffeting of icy waves and to 
 struggle amidst the wreck of ice-floes and the crush of 
 bergs ? These are questions which we may venture perhaps 
 to ask, but which we dare not hope to have solved till we 
 know something more than at present we know of the history 
 of the boulder clay formation itself." 3 
 
 In regard to the Old Red Sandstone of Kirkby Lonsdale 
 and Sedbergh, Ramsay writes : " We have found many 
 stones and blocks distinctly scratched, and on others the 
 ghosts of scratches nearly obliterated by age and chemical 
 action, probably aided by pressure at a time when these rocks 
 were buried under thousands of feet of carboniferous strata. 
 In some cases, however, the markings were probably pro- 
 duced within the body of the rock itself by pressure, accom- 
 panied by disturbance of the strata ; but in others the longi- 
 tudinal and cross striations convey the idea of glacial action. 
 The shapes of the stones of these conglomerates, many of 
 which are from two to three feet long, their flattened sides 
 and rectangular edges, together with the confused manner 
 in which they are often arranged (like stones in the drift), 
 have long been enough to convince me of their ice-borne 
 character ; and the scratched specimens, when properly inves- 
 tigated, may possibly convince others." 4 
 
 At several places in Victoria, Australia, Messrs. Taylor and 
 Etheridge junior, found a conglomerate consisting of a 
 mixture of granite pebbles and boulders of various colours and 
 textures, porphyries, indurated sandstone, quartz and a peculiar 
 flint-coloured rock in a matrix of bluish-grey very hard mud 
 cement. Rocks similar to the pebbles and blocks composing 
 the conglomerate do not occur in the intermediate neigh- 
 bourhood, and from the curious mixture of large and small 
 angular and water-worn fragments, it was conjectured that it 
 might possibly be of glacial origin. Scratched stones were 
 not observed, although a careful examination was made* 
 
 ' Op. cit. p. 89. 4 Eeader, Aug. 17, 1865. 
 
286 The Glacial Nightmare. 
 
 From similar mud pebble beds on the Lerderderg River, 
 Victoria, Mr. E. Daintree obtained a few pebbles grooved 
 after the manner of ice-scratched blocks. 5 Mr. Selwyn 
 classed the deposit provisionally as Devonian. 
 
 In 1851, Professor Haughton brought forward at the 
 Geological Society of Dublin, a case of angular fragments 
 of granite occurring in the carboniferous limestone of the 
 county of Dublin ; and he explained the phenomena by the 
 supposition of the transporting power of ice. 6 
 
 " Great beds of conglomerate," continues Geikie, " occur 
 at the bottom of the carboniferous, in various parts of Scotland, 
 which it is difficult to believe are other than ancient morainic 
 debris. They are frequently quite unstratified and the stones 
 often show that peculiar blunted form which is so characteristic 
 of glacial work. They are confusedly huddled together in a 
 dull, tough arenaceous matrix; and the whole deposit will 
 sometimes continue to exhibit these appearances over a 
 wide area. Ever and anon, however, we detect traces of 
 water action, the stones become more rounded and are spread 
 out in more or less regular layers or beds. The coarse 
 unstratified portions so closely resemble till, that only a 
 practised eye can distinguish the difference at a glance, and 
 many geologists who had not previously tried their hammers 
 on the deposit might well mistake it for a post- tertiary 
 glacial accumulation. It is typically developed in the 
 Lammermuir hills district, and appears also in Ayrshire, 
 Arran and other regions in the west of Scotland. Among 
 these conglomerates my brother has noticed many striated 
 stones." 7 
 
 In 1856, Mr. Godwin Austen, writing about the French 
 carboniferous beds, says : " Whether from local elevation, 
 or from climatal conditions, there are certain appear- 
 ances over the whole which imply that at one time the 
 temperatures must have been very low, as glacier action can 
 alone account for the presence of the large angular blocks 
 which occur in the lowest detritalbeds of many of the southern 
 coal basins ." 8 
 
 6 CrolPs Climate and Time, 295. 6 Climate and Time, 296, 
 
 ? Great Ice Age, 568, 569. s Quart. Journ. Geol. Soc. xii. 58. 
 
Weiss and Or ton. 287 
 
 In 1888, my friend, Mr. M. Stirrup, read a paper before 
 the Manchester Geological Society on Foreign Boulders in 
 Coal Seams. In this he refers to several such discoveries in 
 Lancashire, the first of which was described by Mr. T. W. 
 Binney in 1851. As a rule these boulders are of a hard 
 siliceous grit or quartzite, are smooth, often polished, with 
 the corners rounded off by abrasion. In form they are 
 roughly quadrangular, irregularly ovoid or elliptical, and 
 occasionally globular, and all have been evidently water-worn, 
 but they are neither scratched nor furrowed. According to 
 Mr. Plant, the miners report them to be infrequent, but when 
 found it is chiefly in groups of five, six, or a dozen. They 
 range in size from two inches in diameter to twenty. In 
 addition to the quartzite boulders, others have occurred of 
 quartz and occasionally of granite. Mr. Stirrup argues that 
 transport by ice, or floating ice-bergs in a summer sea would 
 perhaps better than any other theory explain the position 
 and often isolation of these boulders, in strata which indicate 
 quiet deposition and which are singularly free from any 
 admixture of extraneous matter. 9 
 
 In 1884, C. E. Weiss of Berlin published a paper on 
 Boulders from coal seams in Upper Silesia, in which he 
 refers to a list of such boulders contained in a paper by 
 Herr D. Stur of Vienna. These continental boulders, like 
 the English ones, are rounded and polished, except some of 
 the granite ones, which are somewhat rough, due to the 
 projection of certain crystals of garnet, the result of 
 weathering. They consist of gneiss, granite, quartz, 
 porphyry, granulite, quartzite, etc., and vary in size from 
 one to seven pounds. In one case a boulder of typical 
 granulite, lenticular in shape, from Kattviertz, weighed 121 
 English pounds. In another there was found a group of 
 ten boulders at the same place, one weighing 176 pounds. 
 Weiss argues that these boulders were derived from the 
 neighbouring mountains of Bohemia. 
 
 According to Professor Orton, similar boulders occur in 
 the American coal fields. Thus, in the 5th volume of the 
 " Geology of New York," there is a notice of a boulder of 
 
 Trans. Manch. Geol. Soc. xix. 420. 
 
288 The Glacial Nightmare. 
 
 metamorphic sandstone apparently derived from the Cambrian 
 rocks of the Appalachian border. This weighed 200 pounds. 
 " The surface o the boulder/' says the report, " appears 
 glaciated, and ice transport would seem almost a necessity 
 for such a block." Boulders of quartzite very similar to 
 those from the Lancashire coal field and similarly worn and 
 smoothed have been described from the Zaleski coal beds in 
 Ohio. 1 A notice of several similar boulders from the 
 Anthracite Mammoth seam in Pennsylvania is given in a 
 later volume of the same journal. 2 
 
 Principal Dawson also mentions the occurrence in the 
 Lower Carboniferous of Nova Scotia, of angular fragments, and 
 chips of various hard rocks cemented together, which he thinks 
 may fairly be regarded as evidence of somewhat intense 
 winter cold, in the same way as the angular debris detected 
 by him in the Upper Silurian of Nova Scotia. He describes 
 also a gigantic esker of carboniferous age, " on the outside of 
 which large travelled boulders were deposited, probably by 
 drift ice ; while in the swamps within, the coal flora flourished, 
 and fine mud and coaly matter were accumulated." 3 
 
 In the district of Cuttack, in Peninsular India, there occurs 
 a series of coal-bearing beds known as the Talchir beds, whose 
 age has been much discussed, but which have apparently 
 been proved to be carboniferous. 4 Speaking of these 
 beds, which were examined in 1855-6, Mr. W. T. Blandford 
 says : " The lowest bed which we find resting on the 
 gneiss is most generally ' the boulder bed/ which occa- 
 sionally assumes the form of a coarse conglomerate. It 
 consists essentially of boulders of granite and gneiss, those 
 of the former comparatively small and the latter of much 
 larger size, frequently from four to five feet in diameter, 
 embedded in a matrix which varies from a coarse sandstone to 
 the very finest shale. The question arises how these enormous 
 blocks of stone, manifestly requiring a great force to abrade 
 and transport them, are found mixed with a sediment so fine 
 that in any except a very sluggish current it must have been 
 swept away, and could not have been deposited ? Should any 
 
 1 Trans. Manch. Geol. Soc. xix. 417, etc. 2 Vide vol. xxi. 71* 
 
 3 Great Ice Age, 569. American Naturalist, vi. 416. 
 
 4 See Waagenj Becords Geol. Surv. of India, xix. 34. 
 
Blandford and Hughes. 289 
 
 evidence hereafter accrue allowing the inference that these beds 
 may have been formed in a lake on a high table-land, when the 
 winter temperature was sufficiently low to admit of ice reaching 
 the waters of the lake without melting, then an adequate 
 explanation of the phenomenon may be given, as it resembles 
 exactly the effects of the action of ground ice which, enabling 
 boulders to be carried down by a sluggish current, would un- 
 doubtedly produce such an intermixture of large rounded 
 masses of rock and of fine silt, as is seen in the present case.""' 
 
 In a series of blue shales belonging to the same series and 
 occurring in the river Brahmini, there occur rounded pebbles 
 or small boulders of gneiss. Mr. Blandford suggests that 
 their scarcity might induce at first sight their attribution to 
 the portage by roots of trees like the boulders in the chalk, 
 but the total absence of wood or carbonaceous matter seems to 
 preclude this. They rather confirm the view already urged in 
 regard to the boulder bed, that they were due to ground ice. 6 
 Mr. Williams found a similar boulder bed, composed of enor- 
 mous boulders of granite, some measuring from three to ten 
 feet in diameter, mixed with large boulders of white quartz 
 chlorite rock, hornblende and mica schist, under the lowest 
 coal beds in the bed of the Boodu Nullah in the Bamghur coal- 
 field in Orissa, and he describes it as being the lowest member 
 of that coal field. 7 
 
 Mr. T. Hughes says of the Talchir boulders that some of 
 them attain an enormous size, and he measured one forty-two 
 feet in circumference. 8 In regard to the boulders in the 
 river south of Ranchi and Partabpur, Mr. Ball says they 
 differ from the underlying rocks and were derived from the 
 granitic gneisses three miles north, one rock of pink por- 
 phyritic granite occurring in situ north of Tarki seems to 
 have been a prolific source for them. In the country west 
 of Eehr many of the boulders consist of a reddish quartz 
 sandstone, probably of Yindyan age; and Mr. Ball adds 
 that if that supposition be correct, they must have been 
 transported some distance. This could only have been 
 effected through the agency of ice. 9 
 
 5 Records Geol. Surv. of India, i. 49, 50. 6 Id. 55,. 56. 
 
 7 Jd. 79. 8 Id. vol. vi. art. iii. p. 7. 
 
 9 Id. vi. 28. 
 
2 go The Glacial Nightmare. 
 
 In the Olive group in the Salt range, which was once classed 
 as Cretaceous and now as Carboniferous, Dr. Waagen has 
 described a boulder conglomerate made up of boulders, often 
 of enormous size, true erratic blocks, which are embedded in a 
 dark-coloured shale, intermixed often with gravel or coarse 
 sand. The rocks of which the boulders consist are all of a 
 very hard nature, beautiful granites of different colours, 
 syenites, different porphyries, greenstones, quartz rock, etc., 
 but no rocks of a softer nature, like sandstone, etc. " When I 
 visited the Salt range I observed," says Dr. Waagen, " in these 
 conglomerates, a great number of striated pebbles, and recently 
 Dr. Warth has also sent a large number of them to Europe. 
 Such scratched pebbles and boulders embedded in a soft 
 shale are always indications of the influence of ice action, and 
 thus we must admit, for the formation of these boulder 
 conglomerates, the collaboration of ice on a grand scale." 1 
 Similar boulder beds, which Dr. Waagen assigns to the 
 same carboniferous age, occur in the western Salt range as well 
 as in the Trans-Indus continuation of the range, and he says : 
 " That ice was really greatly concerned in the formation of 
 these beds appears beyond doubt, partly from the numerous 
 striated and scratched stones that are found in them, partly 
 from the huge erratic blocks which are off and on to be met. 
 ... If thus all the boulder beds of the Salt range belong to 
 one and the same group, we have all at once a large glacial 
 formation stretching through the whole Salt range, during the 
 time of the coal measures. In speaking of a glacial formation, 
 I do not mean that real glaciers were concerned in the 
 formation of these beds ; it was probably floating ice coming 
 out of the mouths of rivers which brought the boulders 
 with it and deposited them along the sea shore." Speaking 
 of the evidence of a large southern continent, including 
 India, South Africa and Australia, in Carboniferous times, 
 Dr. Waagen continues : " This continent had evidently a 
 well-developed river system, and at the Salt range there was 
 probably the mouth of a great river. Down these rivers large 
 masses of ice floated while in other parts of the world the coal 
 measures were being formed; and these ice masses drifted 
 
 1 Becords Geol. Surv. of India, xix. 30. 
 
Waagen. 291 
 
 along the shore and, as they melted away, deposited large 
 boulders, gravel and fine silt at different places. . . . The 
 enormous development of boulder beds that have been formed 
 under the influence of ice action on this ancient southern 
 continent makes the supposition of very low temperatures 
 during those times on that continent an absolute certainty. 
 These low temperatures were not of local occurrence only, 
 but spread over the whole continent, thus indicating a true 
 glacial period ; a glacial period, however, that was in the 
 beginning restricted to the southern hemisphere and only later 
 on spread also to the northern one/' 2 
 
 From a general examination of the fossils, the same author 
 concludes that in the earlier times of the Carboniferous period 
 a rather high mean temperature must have prevailed on the 
 southern continent, as luxuriant forests of carboniferous plants 
 were thriving then of which remains have been preserved to us 
 both in Australia and in South Africa. All of a sudden a 
 considerable lowering of the temperature took place, ice began 
 to be formed in South Africa and in India and all the carboni- 
 ferous flora was destroyed in these countries, as well as in 
 Australia, by this low temperature. In the meantime, in 
 Australia a new flora began to appear a flora that was suited 
 to support moderate or low temperatures. Notwithstanding 
 the ice that covered part of the southern continent, the flora 
 spread slowly westward from Australia. This flora was of 
 mesozoic types. During the time this went on in the 
 Southern hemisphere, the coal measures were deposited in the 
 Northern one. . . . This state of things continued for a certain 
 time until, in the Permian period, again a change took place. 
 Now the cold had spread to the Northern hemisphere, where 
 traces of it occur, and the result was the extinction of the 
 carboniferous flora also in Europe, and elsewhere in the north, 
 and the flora of a moderate climate, i.e. the mesozoic flora, 
 succeeded it ; and as there were no warm currents which could 
 then reach the Northern hemisphere, the Palaeozoic fauna 
 perished there except a few genera. 
 
 In the Southern hemisphere the intense cold returned during 
 the Permian times and after. In Australia, in the Hawkes- 
 
 2 Records Geol. Surv. of India, xix. 35, 36. 
 u 2 
 
292 The Glacial Nightmare. 
 
 bury beds and the Bacchus March sandstone, which are pro- 
 bably of Permian age (? vide infra), glacial boulder beds occur. 
 The flora endured the low temperature, bub the marine fauna 
 was destroyed by the currents of cold water which reached the 
 Salt range towards the end of the Permian period. In con- 
 clusion Dr. Waagen says : " The cardinal point that is demon- 
 strated beyond doubt is that towards the end of the Palaeozoic 
 times a great glacial period occurred similar in extent and 
 more grave in its effects than the one that took place at the 
 end of the Tertiary era, and that thus the earth has passed 
 already through two severe periods of cold." 3 
 
 Mr. Oldham, writing in 1886, describes the great tract 
 between the Aravalli mountains and the Indus, and known as 
 " the Desert," as containing, near the town of Pokran, an old 
 land surface showing glacial groovings and striae In the 
 boulder beds which occur close by, and are, without doubt, he 
 says, " of the same age as the glaciated land surface, there may 
 be found facetted blocks which, like those of the Salt range, 
 would not be ascribed to anything but glacier action. More- 
 over, this land surface is covered in places by a boulder bed 
 with a hard, intensely tough matrix, differing from the 
 stratified boulder beds of the neighbourhood much in the same 
 manner as ' the till ' of Scotland differs from the marine 
 boulder clays of the Midland counties. If the hypothesis that 
 the toughness is due to its being a ' Grund moraine ' be 
 accepted, it follows that the same explanation will account for 
 the toughness of the boulder beds of Pokran, and we have yet 
 another proof of the existence of glaciers in this old land 
 surface. The boulder beds of the Desert have been traced for 
 60 miles to the N.E. of Pokran ; in the vicinity of the old land 
 surface the boulders are almost exclusively of porphyry and 
 syenite derived from it, but further north blocks of gneiss of 
 the peninsular type become common; and in N. lat. 27 30', 
 E. lat. 72 30', there is a block of very coarse-grained granite, 
 of which 10 ft. x 7J ft. x 3 ft. is exposed above ground. The 
 nearest source from which this block could have been derived 
 is in the Aravallis, fully 150 miles away. The age of these 
 boulder beds appears to be the same as that of the Talchirs. 
 
 3 Eecords Geol. Surv. of India, xix. 38. 
 
Oldham. 293 
 
 . . . Their extent, combined with the distance from which some 
 of the blocks have been transported, as well as their position 
 on the western margin of the peninsular area, point to thecon^ 
 elusion that they are of marine origin ; so that here we have 
 evidence of glaciers having descended to the sea in a district 
 now less than 27 degrees from the equator." 4 
 
 We will now turn to the Permian beds, which form a 
 classic ground for the champions of former glacial periods. 
 
 In 1854, Professor Ramsay communicated a paper to the 
 British Association on the former probable existence of 
 Palaeozoic glaciers. In this he produced some evidence that 
 glaciers and icebergs existed in our latitudes during a part 
 of the Permian epoch, and that there were also probable 
 traces of the same actions in some of the beds of the Bunter 
 sandstone. 
 
 Thus, in speaking of the Permian breccias in South Staf- 
 fordshire and at Enville in Worcestershire, he says : " The 
 fragments found in the breccia are not derived from the Per- 
 mian beds lower in the series on which it rests in South Staf- 
 fordshire and near Enville, nor yet from the Coal measures, 
 Old Red Sandstone, and Silurian strata on which it lies in the 
 Malvern, Abberley and Bewdley districts ; nor is it probable, 
 taking the physical structure of the country into account, that 
 they were formed from the waste of rocks now concealed 
 beneath the upper and lower New Red Sandstones of these 
 or of neighbouring areas. In lithological structure they can 
 generally be identified with the Cambrian rocks of the 
 Longmynd and the Lower Silurian slates and igneous rocks, 
 etc., that overlie these between the Stiper stones and Chirbury 
 in Montgomeryshire. They are enclosed in a hardened paste 
 of fine red marl, probably analogous to the boulder clay of 
 Pleistocene deposits. The included fragments and blocks 
 consist of greenstone, felstone, felstone porphyry, fels- 
 pathic ash, hard ribboned slate such as occurs north of 
 Shelve, quartz rock, purple and green sandstone and slate, 
 black slate and sandstones and limestones containing Silurian 
 fossils. Very few of the stones are well rounded. Many of 
 them are angular, most of them are subangular and have 
 
 4 Geol. Mag. N.S. iii. 303. 
 
294 The Glacial Nightmare. 
 
 those flattened surfaces common in fragments found in 
 
 moraines. They are frequently well polished and occasionally 
 
 scratched. They are of all sizes up to four feet in diameter. 
 
 If they came from the Longmynd district, in places they 
 
 have travelled at least fifty miles, and there is evidence from the 
 
 outcrops that even now in their fragmentary state they occupy 
 
 an area of from twenty-five to thirty-five miles in diameter, 
 
 while there is perhaps much more concealed beneath the 
 
 surface. Had they formed sea beaches they would probably 
 
 have been rounded by the waves, and there is nowhere any 
 
 sound proof that they were scattered abroad by great tidal 
 
 marine floods. Blocks two feet and more in diameter are not 
 
 moved along sea bottoms by tidal currents ; and, considering 
 
 their size, angularity, polish, occasional scratchings in 
 
 straight and sometimes parallel lines, and also the matrix 
 
 that encloses them, Ramsay believes they were deposited from 
 
 icebergs, derived from glaciers that rose in the ancient 
 
 land of the Longmynds and the overlying Lower Silurian 
 
 strata. 
 
 Traces of a similar glacial action are said to occur in 
 the Bunter sandstone on two ' horizons, first in a part of the 
 pebble beds near Bewdley and later at the base of the white 
 and brown sandstones between Stourport and Abberleys. 
 
 In a paper specially devoted to these beds and their contained 
 boulders, published in the Journal of the Geological Society for 
 1855, Ramsay enters at greater length into the question. He 
 says, inter alia; " Every where, in spite of exceptional fragments 
 in the Malvern district, the embedded fragments seem to be 
 derived from one set of rocks ; they are all enclosed in the 
 same red marly paste, and they are mostly angular or sub- 
 angular. A well-rounded water-worn pebble is in places of 
 rare occurrence. The surfaces of a great majority of the 
 pebbles are much flattened, numbers are highly polished, and, 
 when searched for, many of them are observed to be dis- 
 tinctly grooved and finely striated. The strias in some are 
 clear and sharp, and run parallel to or cross each other at 
 various angles ; while in others, though you see their remains, 
 age and surface, decomposition have impaired their sharp- 
 ness and roughened the original polish of the stone. 
 
 5 British Assoc. Keport, 1854, 93, 94. 
 
Ramsay. 295 
 
 " If lithological character be any guide, the fragments, with 
 rare exceptional pieces, seem to have been derived from the 
 conglomerates and green, grey and purple Cambrian grits of 
 the Longmynd, and from the Silurian quartz rocks, slates, 
 felstones, felspathic ashes, greenstones and upper Caradoc 
 rocks of the country between the Longmynd and Chir- 
 bury. The south end of the Malvern Hills is from forty 
 to fifty, the Abberleys from twenty-five to thirty-five 
 miles, Enville from twenty to thirty miles, and South Staf- 
 fordshire from thirty-five to forty-five miles distant from 
 that country. The question then arises, by what process 
 were so many angular and subangular fragments trans- 
 ported so far; many of them being a foot, and some two 
 to three, or even four feet in diameter; the whole in places 
 forming a deposit of several hundred feet in thickness ? 
 Why, also, are they angular and not well-rounded, like the 
 pebbles of the great conglomerate beds of the Bunter Sand- 
 stone ; and why have they flattened sides, and often polished 
 and striated surfaces ? " After examining in turn the various 
 possibilities of water portage, as by rivers, marine currents, 
 etc., and rejecting them in turn, Ramsay concludes that ice 
 alone can have moved them. 
 
 In proof of this he quotes, (i.) the great size of many of the 
 fragments, the largest observed weighing (by a rough esti- 
 mate) from half to three-quarters of a ton ; (ii.) their forms. 
 Bounded pebbles are exceedingly rare. They are angular 
 or subangular, and have those flattened sides so peculiarly 
 characteristic of many glacier fragments in existing moraines, 
 and also of many of the stones of the Pleistocene drifts and 
 the moraine matter of the Welsh, Highland, Irish and Vosges 
 glaciers, (iii.) Many of them are highly polished, and others 
 are grooved and finely striated, like the stones of existing 
 Alpine glaciers, and like those of the ancient glaciers of the 
 Vosges, Wales, Ireland, and the Highlands of Scotland ; or 
 like many stones in the Pleistocene drift. He urges that he 
 had never seen stones thus scratched in other breccias, and 
 that an experienced eye would have no difficulty in recog- 
 nizing the peculiar characteristics of glacial scratching in the 
 Permian fragments. In true water-deposited breccias like 
 
296 The Glacial Nightmare. 
 
 the New Red Conglomerate, whose component stones are often 
 from three to nine inches in diameter, these stones are all 
 beautifully rounded, and where they touch in the rock they 
 are not scratched, but indent each other at the points of 
 contact, due to lateral pressure in loose lying pebbles. 
 These marks rarely occur in the Permian breccias because, 
 instead of the pebbles being mingled with sand, they are 
 separated by a hardened cementing mass of red marL 
 
 " I conceive, therefore," he continues, " that the peculiar 
 forms, polish and markings of many of the stones indicate 
 that these characteristics have been produced by the agency of 
 ice of the nature of glaciers, for mere coast ice would have 
 picked up and drifted away numerous rounded pebbles from 
 the beach and not a great majority of angular flattened stones, 
 such as form the breccias wherever they occur. If this con- 
 clusion be correct and if the parent rocks whence the stones 
 were derived be properly identified, then it follows that the 
 ancient territory of the Longmynd and the adjacent Lower 
 Silurian rocks, having undergone many mutations, at length 
 gave birth to the glaciers which, flowing down some old 
 system of valleys, reached the level of the sea and, breaking 
 off into bergs, floated away to the east and the south-east 
 and deposited their freight of mud, stones and boulders in the 
 neighbouring Permian seas." He further suggests that " the 
 occasional fragments of granite and syenite may be accounted 
 for by the floating bergs having occasionally stranded on or 
 grazed along some of the higher Malvern hilJ-tops that, as 
 islands, dotted the Permian sea and thus picked up a few 
 fragments to be mingled and deposited with the foreign 
 material wherever they chanced to melt. . . . There is one 
 point of resemblance between these Permian breccias with 
 their associated strata and the Pleistocene drift deposits 
 worthy of note. In the latter, fossils are much scattered, and 
 in most of the beds of rare occurrence. They are still more 
 scarce in that part of the Permian series with which the 
 breccias are associated. I have thought, that, in like manner, 
 this paucity of life may be connected in these latitudes with 
 the glacial phenomena of the Permian and the Bunter periods ; 
 and I no sooner mentioned this to Professor L. Forbes than 
 he suggested that it might also be connected with the great 
 
Hall and Sutherland. 297 
 
 break in life that has taken place between Palaeozoic and 
 Secondary times." Kamsay concludes with stating that near 
 Wribbenhall, in part of the pebble beds of Bunter date, there 
 are breccias strikingly like those of Permian strata, and also 
 near Astley, a little S.W. of Stourport, and probably at Alfrick 
 at the base of the White Sandstones; and although it is 
 possible they may have been reconstructed from the waste of 
 the older Permian, yet he felt more disposed, after examining 
 them, to attribute them to direct glacial action, and to explain 
 them as Godwin Austen had explained similar boulders in the 
 same beds. 6 
 
 According to Croll, Mr. James Thompson found a bed of 
 conglomerate which he considered of Permian age, and 
 probably of glacial origin. This conglomerate enclosed 
 angular fragments of various schistose, volcanic and limestone 
 rocks and contained carboniferous fossils. 7 
 
 Writing in 1873, in regard to the Lower Permian 
 beds of Armagh, Professor Hall says of certain boulder 
 beds there : " The pebbles and stones of the boulder bed 
 sometimes reach two feet in diameter, and consist of hard 
 greenish and purple grits, probably of Lower Silurian age ; 
 vein-quartz, slate, carboniferous limestone, diorite, etc., some 
 of which have probably been transported from the region 
 lying twenty or thirty miles to the north-west of this locality. 
 Considering the large size of these boulders, it is difficult to 
 conceive that they could have been carried to their present 
 position by any other agency than that of floating ice." 8 
 
 In 1870 Dr. Sutherland communicated a paper to the 
 Geological Society on an ancient boulder clay of Natal. 
 He classes this bed as of Permian age, and says it rests 
 generally on old Silurian sandstones, while upwards it passes 
 first into newer shales and through them into the sandstones 
 and shales which are associated in Natal with Carboniferous 
 deposits. The deposit itself, he says, consists of a greyish- 
 blue argillaceous matrix, containing fragments of granite, 
 gneiss, graphite, quartzite, greenstone and clay slate. 
 These embedded fragments are of various size, from the 
 
 c Joum. Geol. Soc. xi. 185-205. 
 
 i Climate and Time, 299. 
 
 8 Explanation of sheet 47 Geol. Surv. of Ireland, p. 12 and 13. 
 
298 The Glacial Nightmare. 
 
 minute dimensions of sand grains, up to vast blocks measur- 
 ing- six feet across and weighing from five to ten tons. They 
 are smoothed as if they had been subject to a certain amount of 
 attrition in a muddy sediment, but they are not rounded like 
 boulders that have been subjected to sea breakers . . . the 
 rocky fragments embedded in the boulder clay formation are 
 all of them of the character of rocks that are contemporaneous 
 with, or inferior to, the Silurian Sandstone. Fragments of a 
 higher series are never found in the matrix. In some 
 instances, very ponderous rock-fragments are found as much 
 as fifty or sixty miles away from the sources of their deriva- 
 tion. . . . Dr. Sutherland inclines to think that the transport 
 of vast massive blocks of several tons weight, the scoring of 
 the subjacent surfaces of sandstone and the simultaneous 
 deposition of minute sand grains and large boulders in the 
 same matrix all point to one agency as the only one which 
 can be rationally admitted to account satisfactorily for the 
 presence of this remarkable formation. He believes that the 
 boulder-bearing clay of Natal is of analogous nature to the 
 great Scandinavian drift, to which it is certainly intimately 
 allied in intrinsic mineralogical character ; that it is virtually 
 a vast moraine of olden time, and that ice in some form or 
 other has embedded fragments in the amorphous matrix. 1 
 
 Ramsay, in his address to the British Association in 1880, 
 says : " Mr. G. W. Stow, of the Orange River Free State, has of 
 late years given most elaborate accounts of similar Permian 
 boulder beds in South Africa. These great masses of moraine 
 matter not only contain ice-scratched stones, but on the banks 
 of rivers where the Permian has been removed by aqueous 
 denudation, the underlying rocks, well rounded and mam- 
 millated, are covered by deeply incised glacier grooves 
 pointing in a direction which at length leads the observer 
 to the pre-Permian mountains from whence the stones were 
 derived that formed these ancient moraines. 
 
 In February, 1850, Mr. E. Godwin Austen speaks of the 
 great blocks of porphyry of the middle beds of the New Red 
 Sandstone in the west of England included in sands and marls 
 indicating no great moving power, as seeming to require some 
 
 1 Quart. Journ. Geol. Soc. xvi. 515, 516. 
 
Wilkinson and Ramsay. 299 
 
 such agent as that of floating ice to account for their posi- 
 tion. 2 
 
 Ramsay, speaking from his own observations of the 
 Rothliegende in the north of Germany, says that though 
 marls and gravels are interstratified with it, these, as in our 
 post-pliocene drift, are exceptional, and the main character- 
 istic of this vast formation (2600 feet thick) in the Thiiringer- 
 wald, is the flattened and subangular nature of its blocks, 
 some of which are of large size. Similar erratic deposits are 
 now forming in Baffin's Bay and the Western Atlantic. 3 
 
 Mr. C. S. Wilkinson, writing in 1881 about the occurrence 
 of glacial boulders in "the Hawkesbury Series," in New 
 South Wales, which belong to the Mezozoic age, says : " Here 
 we have huge and small angular masses of soft shales em- 
 bedded in pebble conglomerates and false-bedded sandstones. 
 With these sandstones (Hawkesbury) are interstratified argil- 
 laceous shales, and the boulders are of the same material 
 and contain the same fossils. It seems that during the 
 deposition of the Hawkesbury Series, the rapid and changing 
 currents which deposited the false-bedded sandstones, were at 
 intervals succeeded by quiet waters, from which the mud 
 forming the shales settled down, probably during winter 
 time when ground ice formed. Spring time following, the 
 ice broke up, and drifting about broke up some of the newly- 
 formed shales and mingled the shale fragments with pebbles 
 and sand brought by currents from the shores." 4 
 
 Ramsay says : " The Oolitic strata of the east of Scotland 
 contain such numbers of huge angular blocks, that their 
 possibly, though scarcely probably, glacial origin suggested 
 itself to my mind when I noted the facts during a journey 
 with Sir Roderick Murchison in 1859. The local character 
 of the blocks, chiefly but not entirely oolitic, is adverse to the 
 view ; but the smashed condition of many of the shells in the 
 interstratified oolitic clays is analogous to the state of the 
 shells in the upper drifts all over Britain." 5 
 
 In describing the breccias of Upper Oolitic age in the 
 north of Scotland, Professor Judd tells us that in certain of 
 
 2 Journ. Geol. Soc. vol. vi. 96, 97. 3 Phil. Mag. 4th ser. xxix. 290. 
 4 Geol. Mag. N.S. viii. 287. 5 Phil. Mag. 4th ser. xxix. 290. 
 
300 The Glacial Nightmare. 
 
 their beds they include numerous masses of foreign rock of 
 various sizes. The frequency of these included blocks appears 
 to increase as we go northwards till the upper Oolitic 
 strata are almost wholly made up of fragments of foreign 
 rocks, some of these being of enormous size, crowded together 
 in the greatest confusion and cemented by a sandy or cal- 
 careous matrix. The enclosed fragments of rock, as Hugh 
 Miller noticed, are clearly foreign, and contain palaeozoic fossils. 
 They are sometimes perfectly angular, and at others present 
 every degree of attrition, and are not unfrequently converted 
 into perfectly well-rounded pebbles. They are of all sizes, 
 from pebbles to masses of such prodigious dimensions as 
 altogether to startle the observer. ... At Allt-an-aird south 
 of the Green-Table point, is a mass composed of hard light- 
 coloured sandstones, alternating with indurated shales and 
 calcareous flagstones. This mass stands on end, its strata 
 being vertical, the exposed upper edge measures twenty 
 feet by ten feet. Another mass close by is probably 
 forty feet long and at least twenty thick, while in the exposed 
 section on the south of the Dunglass there may be seen several 
 large included blocks, one of which is ten feet long and four feet 
 thick. The blocks are nofc sorted but heaped together in the 
 wildest confusion, angular and subangular blocks, pebbles, 
 trunks of trees, stems of cycads, masses of coral, shells, shell 
 detritus, sand and mud. The greater number of the included 
 blocks are derived from the Caithness flagstones or middle 
 old red series. 6 
 
 In a paper published in the Quarterly Journal of the 
 Geological Society for 1837, by Mr. Godwin Austen, on a 
 boulder of granite in the chalk near Croydon, he says : 
 " This boulder, together with several large fragments of 
 greenstone, was fairly embedded in the solid chalk/' In 
 this paper he quotes the discovery by Major- General Portlock 
 of several blocks of black basalt isolated in the chalk of 
 Ireland ; 7 the similar discovery of flattened spheroids of 
 syenite in the chalk of Antrim in 1840 by Mr. Griffith, the 
 mention by Mantel, 8 of pebbles of quartz and some fragments 
 
 6 Quart. Journ. Geol. Soc. xxix. 187, etc. 
 
 7 See Report on Londonderry, p. 93, 94. 
 
 8 Geology of the S.E. of England, p. 78. 
 
Godwin A us ten . 30 1 
 
 of green schist iii the chalk, and the following statement of 
 Mr. Dixon : " Small pebbles and large rolled fragments of 
 sandstone and quartz rock are occasionally discovered in the 
 centre of the Upper Chalk. Mr. Coombe found one specimen 
 weighing near fourteen pounds, at Houghton, Sussex ; and I 
 have seen others from the same pit of two and three pounds 
 weight ; several also have been sent me by Mr. Catt from the 
 pits near Lewes." 
 
 Mr. G. Austen says he had an opportunity of examining 
 certain extraneous specimens from the chalk in the possession 
 of the Geological Society, and others belonging to Mr. Harris, 
 which were interesting as an addition to the crystalline rock 
 pebbles ; these were some small specimens which had the 
 appearance of being volcanic scoriae. Mr. Catt's collection 
 contained three specimens of rather soft marly, micaceous 
 sandstone of a greenish-grey colour, one pebble of opaque 
 quartz, two of transparent quartz, and one of coloured quartz ; 
 these were all small. There was also a much larger sub- 
 rhomboidal fragment of dark clay slate, with the angles 
 rounded, and lastly a block of fine-grained sandstone (quartzite) 
 weighing thirteen pounds fourteen ounces. All these frag- 
 ments had been derived from old sedimentary (palaeozoic) 
 strata. Mr. Gumming, of Devizes, had further shown him a 
 piece of thin fossil slate (very like common roofing slate) 
 embedded in a chalk pit. 
 
 Mr. Godwin Austen refers to the generally isolated character 
 of the finds of extraneous stones in the chalk. In the Croydon 
 find the granite boulder was accompanied by a collection of 
 blocks of smaller dimensions, composed of green stone, which 
 were much decomposed. 
 
 Speaking of the various extraneous stones found, he says 
 they are all water-worn, either in the form of shingle or 
 rounded boulders. He discusses Darwin's explanation of the 
 finding of a rounded block of greenstone on the outer coast of 
 a small atoll belonging to the Keeling group, namely that it 
 had become entangled in the roots of a tree and thus trans- 
 ported, a view which commended itself to Sir Charles Lyell 
 as explaining the extraneous stones in the chalk to which his 
 attention had been called by Mr. Catt. He himself would 
 rather invoke the agency of floating ice, and, in answer to 
 
3O2 The Glacial Nightmare. 
 
 Ly ell's objection about the tropical fauna of the chalk, he 
 says : " The presence of a sea is no indication as to whether 
 ice from polar regions may not occasionally be floated into it. 
 All that is requisite is that the area of water should have an 
 extension into the polar regions ; and then the liberated ice 
 must at all times have been distributed according to the same 
 laws as influence it now. Polar ice, it must be remembered, 
 is now occasionally floated down, and its load scattered, as low 
 as the Azores and Canaries, where its detritus becomes 
 embedded in deposits containing the forms of the South 
 Lusitanian fauna." He denies the possibility of any conceiv- 
 able mass of sea-weed having carried boulders of twelve and 
 fourteen pounds weight, and he adds that even if it be 
 assumed that sea plants could at any time have lifted the 
 largest boulders, and that a tangled mass of them could have 
 floated away at the same time all the smaller stones associated 
 with it, the difficulty as to the mass of loose sand still remains. 
 In regard to the kind of ice to which he attributes the portage 
 of the boulders, he says that the detritus borne by icebergs is 
 sub-aerial, and its form angular, and that he had never seen a 
 single angular fragment from the White chalk, nor from any 
 part of the Cretaceous series ; and he appeals to the coast ice of 
 the Polar lands, and says, " I know of no other agent but 
 that of coast ice by which a mass of incoherent materials could 
 have been held together as a characteristic portion of a sea 
 beach, and so deposited at a vast distance from the nearest 
 land. Such, I conclude, was the nature of the agency by 
 which the assemblage discovered in the chalk at Parley was 
 kept together and conveyed." ! 
 
 In a second paper, published in 1860 in the same Journal, 
 on the discovery of a mass of coal four feet square and from 
 four to ten inches thick, at a depth of 1 80 feet in the chalk of 
 Kent, the same author says : " Considering its friability, I 
 do not think that the agency of a floating tree could have been 
 engaged in its transport \ but looking at its flat angular form 
 it seems to me that its history may agree with what I have 
 already suggested with reference to the boulder in the chalk 
 at Croydon. We may suppose that during the Cretaceous 
 
 9 Quart. Journ. Geol. Soc. xiv. 252-265. 
 
J. Martin. 303 
 
 period some bituminous beds of the preceding Oolitic period 
 lay so as to be covered by water near the sea margin or along 
 some river bank, from which portions could be raised off 
 by ice and so drifted away until the ice was no longer able to 
 support its load." l 
 
 Eamsay says : " It may not be generally known that 
 Escher von der Linth is aware of boulders in the cretaceous 
 strata of the Alps." 5 
 
 Turning now to Tertiary times. 
 
 "The so-called flysch of Switzerland," says Dr. Croll, "has 
 by some geologists been assigned to Eocene times. In it are 
 found certain immense blocks, some of which consist of a 
 kind of granite not known to occur in situ in any part of the 
 Alps. Some of the blocks are ten feet and more in length, 
 and one at Halekeren, at the lake of Thun, is 105 feet in length, 
 ninety feet in breadth and forty-five in height. These 
 unmistakably indicate the presence of glaciers or floating ice. 
 This conclusion is further borne out by the fact that 'the 
 flysch ' is destitute of organic remains." 3 
 
 In a paper by M. J. Martin, he quotes, in support of his views 
 about inter- glacial climates, the fact that the flysch of Switzer- 
 land, which is generally barren of life, contains a vast number 
 of pebbles and blocks, notably at Sefey in the Canton of Yaud, 
 where enormous blocks of granite, gneiss, and protogine occur 
 very frequently in this deposit. At Halekerenthal, in the 
 Bernese Oberland, are others of enormous size, as much as 500 
 cubic metres in size. At Yberg in the Sihlthal great masses of 
 lias and of brown Jurassic rocks are mixed with blocks of granite. 
 It is similarly the case at Grabs, and in many other places. The 
 blocks of granite and of limestone are foreign to this district. 
 The same fact has been observed at Bolgen in Bavaria, and at 
 the northern foot of the Apennines in Upper Italy. In view of 
 these facts, he says, we are obliged to admit the work of ter- 
 tiary glaciers in many places, for no other explanation of the 
 facts is admissible. 4 
 
 In the paper already cited by Mr. Lydekker, he refers to the 
 lower Tertiaries near Leh in Ladakh, and says that in certain 
 
 1 Quart. Journ. Greol. Soc. xvi. 327. 
 
 2 Phil. Mag. 4th ser. xxix. 290. 3 Climate and Time 305, 306. 
 4 Bull. Soc. Geol. de France, 3rd ser. vol. ii. 271. 
 
304 The Glacial Nightmare. 
 
 soft and crumbly sandstones there are found embedded great 
 numbers of more or less rounded boulders of gneiss, often 
 attaining a diameter of several feet. Some of the isolated 
 boulders show the beds of the sandstone bending down below 
 them, and the polishing and smoothing of some of the others 
 seem suggestive of ice action. . . . " It is difficult to imagine 
 how these gneissic boulders could have attained their present 
 position in the midst of soft sandy strata without the aid of 
 ice as a means of transport." 5 
 
 In regard to Miocene times we have the observations of 
 Gastaldi and Kamsay about the Italian beds. The latter 
 communicated their results to Dr. Croll, who thus writes of 
 them : a Near Turin there is a series of hills, rising about 500 
 or 600 feet above the valleys, composed of beds of Miocene sand- 
 stone, marl, and gravel and loose conglomerate. These beds 
 have been carefully examined and described by Gastaldi. 6 
 The hill of the Luperga has been particularly noticed by him. 
 Many of the stones in these beds are striated in a manner 
 similar to those found in the true till or boulder clay of this 
 country. But what is most remarkable is the fact that large 
 erratic blocks of limestone, many of them from ten to fifteen 
 feet in diameter, are found in abundance in these beds. It 
 has been shown by Gastaldi that these blocks have all been 
 derived from the outer ridge of the Alps on the Italian side, 
 namely, from the range extending from Ivrea to the Lago 
 Maggiore, and consequently they must have travelled from 
 twenty to eighty miles. So abundant are they that ex- 
 tensive quarries have been opened in the 1 hill, for the 
 sake of procuring them. These facts prove not only the 
 existence of glaciers on the Alps during the Miocene period, 
 but of glaciers extending to the sea and breaking up into ice- 
 bergs ; the stratification of the beds amongst which the blocks 
 occur sufficiently indicating aqueous action and the former 
 presence of the sea/' 7 
 
 Professor Geikie, writing of these North Italian beds and also 
 quoting Gastaldi, says : " The slopes of the Moncaluri-Valenza 
 hills are sprinkled with boulders and large erratics of Alpine 
 
 5 Op. cit. 105. 
 
 6 See Memorie della Eeale Acad. delle Scienzi de Torino, 2nd ser. 
 vol. xx. 7 Climate and Time, 306, 307. ~- 
 
J. Martin. 305 
 
 rocks, which were at one time supposed to have been carried 
 across Piedmont by the ice of the Glacial epoch. But subse- 
 quent and more detailed observations have led Gastaldi to 
 the opinion that the blocks in question are merely the denuded 
 wreck of certain great beds of conglomerate belonging to the 
 Miocene formation. No one who has visited the ground is 
 likely to dispute this conclusion. One sees embedded in the 
 Miocene conglomerates large erratics of precisely the same 
 character as those that are lying loose on the hill slopes ; and 
 the conclusion seems irresistible that these latter are but 
 the relics of those portions of the conglomerate which the 
 denuding forces have carried away/' 8 
 
 M. J. Martin, in 1873, in discussing certain beds found in the 
 Chalonais in Burgundy, urged that they were of the same age 
 as others found scattered at Grosmont, Koumont, Magny, 
 etc. Similar beds on the plateaux of the Senonais and 
 the Gatinais were treated as of Miocene age by MM. Raulin 
 and Leymerie. They resemble again completely the so-called 
 poudingues of Nemours, which have been long classed as of the 
 date of the plastic clay. To the same system, again, he urges, 
 belong the clays with flints and with blocks of sandstone called 
 laderes in the department of Eure et Loire described by M. 
 Laugel and treated by him as of Miocene age. In this 
 view M. Hebert also concurred, and the section he gave 
 from Souance to Senonches shows these clays with flint 
 covered with lacustrine beds containing Limnasa longiscata 
 and Planorbis planatus. In addition to these facts he 
 quoted an excavation made in 1870 in the hamlet of Larrey 
 (Dijon) in the Miocene conglomerate, which was formed of 
 coraline and portlandean pebbles ; many were rolled, many 
 again polished and striated like glacial boulders. In this ex- 
 cavation masses of half a cubic metre in extent, with one side 
 angular and the other rubbed and striated, rested in the midst 
 of rounded pebbles, having a brilliant and soapy polish, with 
 fine lines scratched on them, in which it was not difficult to find 
 the particle of quartz which acted as a graver. The deposit 
 is considerable, opposite to and of the same age as the con- 
 glomerate with helix ramondi at the Dijon station. M. Jules 
 
 8 Great Ice Age, 437-439. 
 
306 The Glacial Nightmare. 
 
 Martin concluded definitely that these facts prove there were 
 two glacial epochs, one in Miocene and one in Quaternary 
 times ; the first one being characterized by the conglomerate 
 with polished and striated pebbles of Dijon, by the clay with 
 flints and with a puddingstone of flints in the Chalonais plain, 
 and by the blocks of puddingstone and of siliceous sandstone 
 of Grosmont, Roumont, Magny, etc. These Miocene debris, 
 both in the Seine valley and in that of the Saone, came from the 
 Morvan. 9 
 
 In 1866-1868 Mr. P. W. Stuart Menteath published two or 
 three memoirs on the evidence of a Miocene Glacial period in 
 the Pyrenees. In these papers the author classed certain con- 
 glomerates (poudingues) occurring near Pau, as of Miocene 
 age. These conglomerates he identified with those described 
 by M. Leymerie in the Haute Garonne under the name of Pou- 
 dingue de Palasson, and with the upper part of the similar beds 
 described by the Abbe Pouech in the department of the 
 Ariege. These beds, containing no fossils but having all the 
 appearance of a glacial formation, form, according to Mr. 
 Menteath, the boundary between the Miocene and Eocene 
 periods. He compares them with the Permian conglomerate 
 described by Ramsay. 
 
 M. Tardy, in a communication to the French Geological 
 Society in 1876, calls attention to large boulders of quartz in- 
 cluded in the Miocene beds at Bourg, and in the lacustrine marls 
 of Miocene age at Yarambon in La Bresse. He concludes 
 that they must have been transported by ice, and compares 
 them with the similar boulders found near Turin and the 
 Grand Chartreuse. 10 
 
 In regard to Pliocene times, the only evidence available is 
 that of the North Italian beds. According to Desor and Sto- 
 panni, clays containing unmistakable Pliocene shells, which 
 undoubtedly, according to Desor, lived and died where their 
 remains occur, are covered with morainic debris, including 
 striated stones ; and according to Stopanni these striated 
 stones are not confined to the moraines, but appear also in the 
 underlying Pliocene clay. "From their observations," says Mr. 
 Geikie, " Stopanni and Desor infer that during the formation 
 
 9 Bull. Soc. Geol. de France, 3rd ser. i. 390-397. 
 10 Id. 3rd ser. iv. 185, 186. 
 
Venetz. 
 
 307 
 
 of the marine Pliocene the old Adriatic sea covered all the 
 low grounds of Northern Italy, and extended into the deep 
 valleys of the Alps, down which crawled large glaciers, which 
 eventually entered the sea and shot their moraines upon its 
 bed." n Rutimeyer virtually came to the same conclusion. 
 Geikie, in summing up his views, says : " In the present state 
 of our knowledge no hard and fast line can be drawn between 
 these two periods, i.e. between Pliocene and Pleistocene times. 
 From early Pliocene times onwards we have proofs of a con- 
 tinuity of terrestrial conditions over a wide region in Europe. 
 And in the long lapse of time represented by the Pliocene and 
 Post-pliocene deposits a period during which considerable 
 changes in the mammalian fauna were taking place glaciers 
 now and again descended to the mouths of the great Alpine 
 valleys." 12 
 
 I have quoted a sufficient number of instances from 
 different geological horizons to show how those geologists 
 and they probably include the majority who believe in the 
 recurrence of glacial phenomena justify their conclusions. 
 If they appeal to astronomical and secular changes of climate, 
 they are of course bound to prove that the last great glacial 
 age was only one of many ; or, as Mr. Croll puts it, if the 
 glacial epoch resulted from these causes then such epochs 
 must have frequently supervened. The case needed more 
 than this, however ; it needed also evidence to show that the 
 glacial period itself was broken up into alternate zones of 
 mild and severe climatic conditions. Hence the theory of 
 inter- glacial periods. 
 
 Yenetz, who, like Charpentier, had not apparently heard of 
 ground-moraines, argued that a glacier cannot at the same 
 time scratch and polish rock surfaces and carpet them with 
 soft debris, and he consequently urged that the former 
 process was due to an earlier glacial action than the latter. 
 He also urged that, inasmuch as when the mountains were 
 covered with ice and snow they would be protected from 
 denudation, it is difficult to understand how, during the 
 maximum of the ice development, there could be angular 
 erratics deposited in the extreme terminal moraines ; and he 
 
 11 Great Ice Age, 441, 442. 12 Id. 445. 
 
 x 2 
 
308 The Glacial Nightmare. 
 
 consequently argued that where there are several successive 
 terminal moraines, they point to several stages of glaciation. He 
 also distinguished between two great moraine zones, an outer 
 and an inner, and attributed them to two distinct glaciations. 1 
 
 In regard to the distinction between the outer moraine 
 and the inner ones, Agassiz had noticed that the true 
 moraines in the Alps are not found at such a distance from 
 the radiating centres as the more evenly distributed debris 
 to which some geologists give the name of ground moraines ; 
 and he distinguished between two glacial epochs, one marked 
 by a widespread ice-sheet, evidenced by the ground- moraine, 
 and the other by a later glaciation to which the deposition of 
 the terminal moraines is to be attributed. This distinction 
 between an outer and inner moraine zone was adopted by 
 several later writers on the Alps. Thus Muhlberg, in two 
 memoirs on the erratics of the Aargau, published in 1869 and 
 1878, adopted this view. So did Bach in his description of 
 the erratic phenomena of Wurtemberg, published in 1869 
 Taramelli, in 1880, applied the same arguments to the Italian 
 phenomena in the Canton Ticino and its neighbourhood. H. 
 Hofer had, some years before, done the same in Carinthia. 
 These various writers had deduced a double glaciation from 
 the geographical distribution of the moraines. Another set 
 of writers came to the same conclusion from examining the 
 internal arrangement of the drift deposits, and professed to 
 find traces of temperate conditions in beds interstratified with 
 the true glacial beds. 
 
 The first person who suggested that the glacial deposits 
 instead of being continuous are broken and intermittent, 
 was also Yenetz. He pointed out that the depot of bitu- 
 minous wood found on the southern flank of the great 
 glacial talus of Thonou, rests on one glacial bed and is 
 covered by another ; and, according to Morlot, he prepared a 
 map in which he showed the extent of the Rhone glacier 
 during its second phase. 2 
 
 In 1854 M. Morlot published a paper in the Bulletin de la 
 Societ e Yaudoise, on the Quaternary deposits of Switzerland. 
 
 1 See Penck, Yergletscherung der Deutschen Alpen, 212, 213. 
 
 2 Bull, de la Societe Yaudoise des Sciences Naturelles, iv. 45. 
 
Morlot. 309 
 
 This lie followed up by another, published in 1855, on a larger 
 scale, on the subdivision of the Quaternary deposits of the 
 same country. In these papers he urged that the beds called 
 ancient alluvium by M. Necker, in which the traces of the 
 mammoth and its contemporaries occur, and to which he 
 gives the name of diluvium, were deposited by the present 
 rivers and lakes, when the continent was at a much 
 lower level than it is now; that when they were laid down 
 the Alpine country was free from ice to a height of 3000 or 
 4000 feet above the sea level, and that it is probable the 
 ice zone was even more restricted, since the rivers flowing 
 from the glaciers would otherwise not have had the necessary 
 length and volume to deposit these beds. He went on to 
 urge that the deposit of this diluvium has been generally 
 recognized as anterior to the glacial epoch, since the erratic 
 deposits have been distinctly found in several places super- 
 imposed on these beds ; as near Geneva, where erratic 
 detritus to a depth of fifty feet overlies them, 100 feet above 
 the level of the lake. 
 
 This evidence, however, is not uniform. Thus at Tavel, 
 near Clarens, between Yevey and Villeneuve, he himself had 
 described an instance where a diluvial bed of seven feet thick 
 was imposed on a bed of unstratified blue clay forty -three feet 
 thick, enclosing rounded and striated Alpine boulders ; which 
 clay rested immediately on the molasse. M. Ischer after- 
 wards found a similar arrangement of beds near Berne. In 
 other cases the diluvium itself contains Alpine materials 
 which must have been carried by ice and not water, or they 
 would have filled up the lake of Geneva. 
 
 From these facts it follows, he urged, that a glacier must 
 have invaded the country before the diluvial epoch. During 
 that epoch it disappeared entirely, to return again afterwards. 
 The first invasion of the glacier was the one in which the 
 advance was the most extended and remarkable, and the 
 greater part of Switzerland must then have been covered 
 with ice. The absence of angular unrolled blocks at this 
 stage, as Mr. Chambers had remarked in Scotland and M. 
 Collomb in the Vosges, shows that few naked mountain 
 summits were then exposed to furnish them. This earlier 
 period, although very intense, was not of long duration, for it 
 
3io The Glacial Nightmare. 
 
 left no true moraines. This is especially noteworthy on the 
 Jura. 
 
 The second glacial epoch, on the contrary, must have been 
 prolonged, for it has left moraines which often rest on the 
 diluvium. These moraines are numerous, well-marked and 
 immense, but they are not found at the same distance from 
 the mountains as the debris of the first glacier. The Ehone 
 glacier, for instance, did not then extend beyond Geneva. 
 While the first glacier, near Yevey, passed over Mont Folly, 
 5863 feet above the sea level, and rounded its summit, the 
 second glacier did not rise above Avant and Lalliaz, 3263 
 feet above the sea level. The latter was therefore 2500 feet 
 below the former. The latter, in retreating, made several 
 halts, leaving moraines at each stage. These intermediate 
 moraines are well marked near Linth, where they have been 
 studied by M. Escher, and may be traced up to the present 
 glaciers, showing that the transition from this condition to 
 the present was gradual. 
 
 Following in the footsteps of Mr. Chambers in Scotland, 
 M. Morlot then proceeds to divide the so-called glacial beds 
 into two series, one marked by compact blue clay containing 
 boulders, the till of the Scotch geologists, and the other a 
 yellow or red friable clay like loess, but containing no shells, 
 in which the iron has been oxydized ; and he considers that the 
 latter is, to a large extent, the result of the so-called remanie- 
 ment of the former. Speaking generally, he attributes the blue 
 clay to the earlier glacial development, and the yellow or red 
 to the later, but adds that to this there are exceptions. 3 
 
 Near Yevey and Lausanne the clay deposited by the first 
 glacier is blue and compact, while that deposited by the 
 second one is friable and of a red colour. At Soleure, however, 
 which the second glacier did not reach, the clay of the first 
 glacier is also red and friable. He accounts for this by 
 the fact that in this case the deposit was made at the 
 termination of the glacier, where it was thin and accessible 
 to the air, which has oxydized the iron in the clay, thus giving 
 it its red colour. 
 
 The occasional occurrence of irregularly stratified rolled 
 
 vX 
 3 Id. 41, 42. 
 
Morlot, etc. 311 
 
 materials with, erratic detritus he explains as the result of 
 ice lakes formed by glacier dams. This kind of deposit was 
 named diluvium glaciare by M. Charpentier, and was com- 
 pared by M. Martin with the material of the asars, which also 
 shows the joint action of water and ice. Its great extent points* 
 he urges, to the second glacier having lasted a long time. 
 
 In recapitulating his conclusions, M. Morlot urges that in 
 Switzerland there are evidences of I. The first glacial epoch, 
 when Switzerland, Scandinavia, Scotland and the Vosges were 
 covered with ice sheets, whose debris contain 'no fossils and 
 are formed of till or compact blue clay with boulders. It pro- 
 bably lasted no long time. II. The diluvian period, when the 
 glaciers disappeared from the principal valleys of Switzerland, 
 as far as 8000 or 4000 feet above the present sea level. The 
 rivers then flowed at higher levels. It lasted a long time, pro- 
 bably as long as the present period. It was marked largely by a 
 similar fauna and flora as the present period. III. The second 
 glacial epoch, when the glaciers again invaded the Alpine 
 valleys, filled up the lakes, such as Geneva, but did not extend 
 as far as the first. To this period he attributes the loess, and 
 in certain localities the so-called diluvium glaciare. IV. Lastly 
 the modern epoch, which was a prolongation of the previous 
 one, was caused by the gradual shrinking of the glaciers, and 
 was marked by the disappearance of the mammoth. 4 
 
 M. Chavanne, in the discussion on M. Morlot's paper, said 
 he had noticed at Merges the glacial mud resting on the 
 molasse and covered in turn by diluvial beds ; and Dr. 
 Delaharpe said the same thing occurred at Charnblande 
 below Lausanne. M. Blanchet distinguishes three phases in 
 the glacial epoch : a period of the growth of the great glaciers, 
 a second period in which they were stationary, and a period 
 of their retreat, during which, for the most part, the lateral 
 moraines were formed. 5 M. Guyot, in his researches on the 
 Rhone glacier, also pointed out two phases in its history. 
 
 In 1857 M. Scipion Gras read a memoir before the French 
 Geological Society on the Quaternary period in the valley of 
 the Khone. In this elaborate memoir, he distinguishes five 
 
 4 Bull, de la Soc. Vaud. ir. 41, 45 ; Bibl. Universelle de Geneve 
 Archives xxix. 32-42. 
 
 5 Bull. Soc. Vaudoise, iv. 53. 
 
312 The Glacial Nightmare. 
 
 successive stages in the Quaternary period, while he says the 
 facts point to two distinct glacial epochs in the Alps ; one 
 marked by the deposition of what he calls the lower diluvium 
 with its angular blocks and striated pebbles, and the other 
 by the dispersal of erratic blocks which, he says, closes the 
 Quaternary age ; and he claims to have arrived at the same 
 conclusions as M. Morlot by another route. 6 
 
 In 1865 0. Heer published his " Urwelt der Schweiz," which 
 was translated into English by Mr. Heywood. In this work 
 he pointed out that the lignite beds at Diirnten in the canton 
 of Zurich, and at Utznach in the canton of St. Gallen, are 
 marked by plants not different from those of our time. These 
 lignites, he shows, are overlaid by great deposits of stratified 
 gravel and sand, upon the surface of which isolated blocks are 
 found. The pebbles contained in these beds have been 
 derived from the Alps. Heer accepted the views of Guyot 
 and Morlot about the double glaciation of Switzerland, and 
 he says, inter alia : " It is manifest that the lignites of 
 Diirnten and Utznach were formed before the stratified drift, 
 as they lie below it. ... The climate was then like that of 
 the present day. This condition of things must have lasted 
 some thousands of years in order to produce the thick 
 deposits of peat which the lignite originated. But gradually 
 the climate became colder, the glaciers advanced from the 
 Alps and changed the whole aspect of the country. Again 
 the temperature became warmer ; the glaciers slowly melted, 
 and retreated, after many oscillations, to their present limits. 
 . . . Two glacial epochs are indicated by the mode of dis- 
 persion of the boulders in the basin of the Rhone. ... In 
 Eastern Switzerland, until recently, the upper erratic forma- 
 tion only was known, which overlies the stratified pebble-beds. 
 Next at Wetzikon, Alpine blocks, showing all the signs 
 of glacial transport, have been found under the lignites. 
 This deposit of lignite, however, is of small extent ; but it 
 lies horizontally, and presents the same sequence of clay and 
 coal as at Diirnten ; so that it cannot be supposed that these 
 erratics have got under the lignite deposit by displacement 
 of the soil/'' In Morschweil on the Bodensee also, according to 
 
 6 Bull. Soc. Geol. de France, ser. ii. xiv. 241. 
 
Deicke and Geikie. 
 
 313 
 
 the investigations of Professor Deicke, erratics are overlaid 
 by lignites. Again, Deicke says that on the new road 
 near St. Gallen, erratic blocks occur over and below the 
 diluvium, and in the Lake district of Baden, the erratic blocks 
 lie under the erratics and partly immediately on the molasse 
 as at Stabringen. 
 
 Fortified by these facts, the conclusion may be adopted, says 
 Heer, that at two different periods glaciers invaded Switzer- 
 land, and that the lignite formation occurred in the interval 
 between them, thus representing an episode of several thousand 
 years in the long glacial epoch, the intercalated period having 
 been sufficiently long to diffuse over the low country a new 
 vegetable covering. 
 
 The stratified deposits of rolled pebbles must have taken 
 place at different periods ; first, after the epoch of the lignites, 
 which may be called inter-glacial ; secondly, during the 
 great extensions of the glaciers glacial drift ; and lastly, after 
 the second glacial extension. 7 
 
 A notable advance in the doctrine of interglacial climates 
 was made by Sir A. Geikie, who, in his well-known paper on 
 the Glacial drift of Scotland, read before the Glasgow 
 Geological Society in 1863, drew attention to the fact that in 
 certain places in Scotland, beds of peat and other traces of 
 similar land-surfaces have been found intercalated between 
 beds of boulder clay. He pointed out a notable case at 
 Cowbyres Bridge, about a mile and a quarter from Hawick. 
 "So far," he says, " as it was possible to ascertain the nature of 
 the vegetable remains in the peaty layer, they appeared to be 
 the rootlets of a kind of heath. The boulder clay which rested 
 upon these stratified beds coincided in every respect with that 
 below them, except perhaps in having not quite so many large 
 stones/' 8 
 
 He says he was much taken by surprise by the section, but 
 soon obtained other examples of the intercalation of stratified 
 materials in the tumultuous and unstratified till. These for 
 the most part did not contain traces of terrestrial animals or 
 plants. In another notable case, however, at the pit workings 
 
 7 Heer, Urzeit, etc., 531, 532 ; Trans, ii. 201, 202. 
 
 8 Trans. Geol. Soc. of Glasgow, i. 54, 55. 
 
The Glacial Nightmare. 
 
 of the coal field of Airdrie, in digging through the drift beds, 
 layers of peat were from time to time met with in certain 
 laminated clays, likewise decaying twigs and branches. In 
 specimens of these peaty layers which he examined, although 
 much decayed, the vegetable fibres were still distinct, and the 
 substance burnt with a dull lambent flame when put into the 
 fire." These beds were both underlaid and overlaid by till, 1 
 and Geikie argues from the facts that they possibly indicate 
 that a the boulder clay is not the result of one great catastrophe, 
 but of slow and silent, yet mighty, forces acting sometimes 
 with long pauses throughout a vast cycle of time." 2 He 
 then goes on to call attention to the discovery at Kilmaurs, 
 Ayrshire, in 1817, of two elephant's tusks in the boulder clay 
 itself, while at 17J feet from the surface other tusks occurred 
 in the same place ; and Dr. Scouler, who examined the spot 
 and saw some of the remains, assured Geikie they occurred in 
 a true boulder-bearing till. A second similar discovery was 
 made in 1820, during the excavation of the Union canal, 
 between Edinburgh and Falkirk, where an immense tusk is 
 reported to have occurred 15 or 20 feet from the surface, em- 
 bedded in the heart of the stiff boulder clay ; while a bone of an 
 elephant is also said to have been found at Chapelhall near 
 Airdrie, in laminated sand underlying till. Eemains of the elk 
 and reindeer are similarly quoted by Geikie as having been 
 found elsewhere in similar beds. And he concludes from the 
 evidence that " while Scotland was covered during the drift 
 period with a wide sheet of snow and ice. . . . there were 
 periods of varying climate; that sometimes many of the 
 valleys glowed with the vegetation of an Arctic summer, rills 
 of water trickled down their sides or dashed over their rocky 
 ledges, and herds of huge mammals pastured in their solitudes ; 
 that again the cold increased, and the ice, once more pushing 
 down the valleys, buried their beauty under one wide pall of 
 cheerless desolation. " 3 
 
 In a paper read on April 2, 1868, before the Geological 
 Society of Glasgow, Mr. Bennie gives the result of a large 
 series consisting of 248 borings, of an aggregate length of 
 21,245 feet, made for water, etc., in the surface deposits near 
 
 1 Id. 61-63. 2 Id. 66. 3 Id. 96. 
 
Bennie and Geikie. 315 
 
 Glasgow. He points out the numerous instances in which great 
 beds of water-formed materials, as sand and gravel, were laid 
 down upon such glacial debris as boulder clay, indicating a mild 
 interval, and how glacial conditions again prevailed, as shown 
 by clay being superimposed on them again. " The whole series 
 showing/' he says, " that the glacial period was not uniform 
 throughout its duration, but was broken up by warmer periods 
 during which water regulated the course of deposition. . . . 
 Between 50 and 60 bores were obtained, having sand, gravel 
 or clay, true water-drifted materials, interstratified with or 
 lying beneath boulder clay. Twenty-five of these had one bed 
 of sand or clay intercalated between two beds of boulder clay, 
 proving that one break occurred. Ten had two beds of sand 
 or clay between three beds of boulder clay. One bore had 
 three and two bores had four beds of sand thus interpolated, 
 while one bore had five such beds, a number sufficient to show 
 that the ice sheet was not continuous throughout the glacial 
 epoch, but disappeared forperiods so long that great beds of 
 water-made debris could be deposited in the interval." After 
 quoting some details, Mr. Bennie continues : " Such is the 
 evidence upon which we seek to establish an inter-glacial 
 character for the sands, clays and gravels, which we find 
 interstratified in these bores with what seems to be boulder 
 clay, and the conclusion clearly inferable from them is, that 
 there were several glacial periods ; or, if we consider the whole 
 as forming one great epoch, we must conclude that the glacial 
 conditions were not uniform throughout the whole time it 
 lasted, but were interrupted by warmer periods during which 
 sand, gravel and clay, the forms which water-carried drift 
 assumes, were the only sediment possible." 4 
 
 In 1871 Professor J. Geikie published a paper in the Geolo- 
 gical Magazine, on Changes of Climate during the Glacial epoch. 
 In this he discusses the evidence of the stratified beds included 
 occasionally in the till of Scotland. Among these he especially 
 notes the laminated brick clays resting, he says, in little 
 basin-shaped hollows of the till, and covered by it, in which 
 may sometimes be detected thin lines of peat. From these and 
 the fresh-water beds similarly intercalated, he argues that that 
 
 I 4 Trans. Geol. Soc. of Glasgow, vol. iii. 133, etc. 
 
3 1 6 The Glacial Nightmare. 
 
 section of the glacial epoch represented by the Scotch boulder 
 clay was not one long unbroken age of ice. It was certainly 
 interrupted by several intervening periods of less Arctic con- 
 ditions, during the prevalence of which the ice sheet must 
 gradually have melted away from the low grounds and given 
 place to streams and lakes and rivers. " At such periods a 
 vegetation like that of cold temperate regions clothed the 
 valleys with grasses and heaths and the hill sides with pines 
 and birch. Reindeer wandered across the country, whole 
 herds of the great ox and the mammoth frequented the grassy 
 vales. . . . Whether during any of our inter-glacial periods the 
 climate was ever mild enough to melt away all the ice and 
 snow from our Highland valleys, the record does not say. 
 What evidence we have, points to the existence of local 
 glaciers in our higher valleys and to moderate summers and 
 severe winters during such inter-glacial periods as we have 
 any certain records of. ... We must bear in mind that the 
 inter-glacial deposits are the veriest fragments. They have 
 been preserved only in sheltered hollows from the ravages of 
 the great ice plough and the interrupted and patchy portions 
 that remain are mere wrecks of what must once have been, in 
 the lower valleys, widespread and continuous deposits. Every 
 renewed descent of the glaciers upon the low ground would tend 
 to effect the removal of these accumulations, and it may well be 
 that of many inter-glacial periods, not a single representative 
 deposit now remains. ... So far then as the Scottish glacial 
 drifts are concerned, there is no evidence whatever to show 
 that the inter-glacial periods may not have been warm enough 
 to cause all the snow and ice to disappear from the country." 5 
 Again, Mr. Geikie says, " We shall probably never learn how 
 many great changes of climate took place during the accumu- 
 lation of the till. This arises from the fact that duriDg every 
 period of intensest cold, when the country was covered with a 
 more or less thick sheet of snow and ice, the loose materials 
 which in the preceding age had gathered in river valleys and 
 in lakes would almost inevitably be subjected to excessive 
 denudation. . . . yet in sheltered hollows we find sometimes 
 two, sometimes three and even four beds of till separated by 
 
 Geol. Mag. vol. viii. 551-553. 
 
Wood and Geikie. 317 
 
 intervening beds, deposits of gravel, sand and silt. We 
 cannot assume, however, that only four cold periods and three 
 intervening ages of milder conditions were comprised within 
 the first stage of the great glacial epoch. For aught that 
 we know there may have been many revolutions of climate. 6 
 He sums up the Scotch evidence thus : 
 
 1. The till of Scotland, with its intercalated beds, in- 
 dicates a vast lapse of time, during which there were 
 several great revolutions of climate ; how many we do not 
 know. 
 
 2. The beds of till point to intense Arctic conditions having 
 prevailed at the time of their deposition. 
 
 3. The beds intercalated with the till, consisting of deposits 
 of silt, clay and gravel, with land plants and mammalian 
 remains, and in some cases with marine shells, clearly show 
 that the intense Arctic cold which covered the country with 
 an ice-sheet, was interrupted not once but several times, by 
 long continuous ages of milder conditions. 
 
 4. The direct evidence shows these inter-glacial periods 
 were not characterized by warmer climates than that now 
 prevailing in the forest region of the high latitudes of North 
 America. 
 
 5. We are entitled to argue, from the few remaining fossils 
 yielded by them, that the climates of inter-glacial periods 
 were never positively warm. 7 
 
 Turning from Scotland to England, it has long been 
 argued that the drift beds of the Eastern Counties are 
 separable into a series. Mr. S. V. Wood, Jun., urged that 
 there are four distinct boulder clays in this district, namely, 
 (1) the Cromer clay, (2) the great chalky boulder clay, 
 (3) the purple clay of Holderness, (4) the Hessle clay. 
 The Cromer clay is underlaid by the forest bed and 
 overlaid, according to Wood and Geikie, by certain gravels 
 containing marine shells ; and here and there, as near Ipswich, 
 says Geikie, associated with beds of brick clay apparently of 
 fresh-water origin, in which fragments of wood, sticks and 
 logs have been found. The same author identifies this layer 
 with the brick earths of Brandon, containing palaeolithic 
 
 6 Id. 265-270. 7 Id. 29, 30. 
 
318 The Glacial Nightmare. 
 
 implements. " Thus," lie continues, "we have evidence to 
 snow that after the ice-sheet which laid down the Cromer 
 till had melted away, mild conditions of climate ensued. . . . 
 Overlying the Brandon beds, with their flint implements, 
 comes the great chalky boulder clay or bottom moraine of an 
 ice-sheet formed during the climax of glacial cold. It is this 
 boulder clay which has been traced south to the valley of the 
 Thames. Besting upon it occur sands, gravels, and loams 
 of much the same character as those that immediately overlie 
 the Cromer boulder clay. ... To these deposits succeeds a 
 third boulder clay, that which is known as the purple clay, 
 the presence of which points to a third advance of the ice- 
 sheet. Overlying the purple clay again, we encounter a series 
 of sands and gravels which in the valley of the Hurnber have 
 yielded remains of Pleistocene mammalia. . . . Lastly, these 
 beds are covered in their turn by a fourth sheet of boulder 
 clay, the Hessle boulder clay. . . . Thus we have evidence 
 in these English sections of no fewer than four glacial 
 epochs separated by intervening epochs of mild climatic 
 conditions." 8 
 
 In the north-west of England and the east of Ireland, the 
 drift series has been divided by Professor Hull and others 
 into three members, an upper and a lower boulder clay, 
 separated by beds of sand and gravel, in some cases contain- 
 ing marine debris, known as the middle sands. Mr. Geikie 
 would identify the lower boulder clay of Lancashire, Cheshire 
 and Ireland with the purple boulder clay of Yorkshire, the 
 middle sands with the Hessle estuarine beds, and the upper 
 boulder clay on both sides of the Irish Channel with the 
 Hessle boulder clay of Holderness ; and he proceeds to give 
 in detail the succession of climatic changes which these facts 
 imply, involving, he contends, a succession of glacial and mild 
 periods. 
 
 In Germany, Professor Penck has been a notable cham- 
 pion of inter-glacial beds. In his well-known memoir on 
 the erratic formations of Germany, he refers to a section 
 at Bixdorf in Brandenburg, where, he says, a bed con- 
 taining bones of Pleistocene mammals and shells separates 
 two distinct boulder clays, and he argues that this 
 8 Pre-historic Times, 261-265. 
 
Penck. 3 1 9 
 
 proves that the great Scandinavian glacier overran the 
 neighbourhood of Berlin at least twice, and that this 
 cannot be explained by a mere oscillation, since the 
 deposits in question occupy 400 square miles in Branden- 
 burg. The same author, after describing in some detail the 
 various instances of the occurrence of fresh-water deposits 
 in the neighbourhood of Potsdam, Baumgartenbriick, Glindov, 
 etc., urges that they point to an intercalation between two 
 beds of boulder clay, and to being relics of a mild period inter- 
 vening between two glacial ones. 1 As Penck puts these beds 
 of sand on a different horizon to the one before quoted, it im- 
 plies that in his view there is evidence of three cold periods 
 separated by two mild ones. In the province of Prussia proper, 
 Penck claims that there are also three boulder clays separated 
 in this case by beds of sand, etc., containing marine shells in 
 some cases mixed with fresh-water ones, pointing to there 
 having been there as in Brandenburg a threefold return of the 
 ice movement ; and this could not have been a mere oscillation 
 of the glacier, since it implies that during the intervals the 
 Baltic was free from ice, while it must have been filled with it 
 if the ice carried the Scandinavian boulders found in the several 
 clays. 2 Penck asserts that in Holstein there are also two 
 distinctly marked boulder clays, separated by a bed contain- 
 ing marine shells, and in one case a fresh-water one also. 3 
 Forchammer urged that there are at least four boulder clays 
 in Denmark, separated by intercalated beds of sand and loam. 
 While in Puggard's famous memoir on Moen, he describes 
 three sharply defined boulder clays, distinguished by their 
 stony contents. Penck has at some length professed to 
 describe the march of the ice on its three successive invasions 
 of that island. 4 At Faxo the same evidence is forthcoming 
 if we are to accept his reading of the section. Turning 
 to Scania in the south of Sweden, Penck quotes several 
 sections, and concludes that here also, as in Denmark and 
 North Germany, there is evidence that the great ice-sheet 
 invaded the country three times, having withdrawn from it 
 completely twice. 5 
 
 1 Zeitschriffc der Deutsches Geol. Gesellschaft, xxxi. 152-157. 
 
 2 Id. 161-167. 3 Id. 168-175. 4 Id. 175-179. 
 
 5 Id. 182. 
 
320 The Glacial Nightmare. 
 
 In summing' up his results he repeats emphatically that 
 North Germany was overwhelmed with ice not once, but three 
 times, separated by distinct inter-glacial intervals. 6 
 
 Tornebohm, in a letter to Professor Geikie, writing of the 
 two layers of till existing in Sweden, the lower of which is 
 generally darker in colour and contains fewer stones than 
 the upper, says : " The fact that the lower till exhibits such 
 marked evidence of denudation underneath the upper and 
 overlying mass, seems to point out that during the glacial 
 epoch there was a great interval of comparatively mild 
 climate, during which the ice retreated to the mountains." 7 
 Geikie also quotes the observations of Nathorst, who, 
 writing in 1873, 8 describes an inter-glacial bed containing 
 leaves of Salix polaris, Dryas octopetala, Limnea limosa, 
 Pisidium, Anodonta, and Cytheridea torosa as intercalated 
 between two boulder clays in Sweden and Denmark. Similar 
 intercalated beds containing similar plant and fresh-water 
 molluscan remains are named as occurring at Klagerup in 
 Scania. Holmstrom quotes this as evidence of an inter-glacial 
 period between the time when the great mer de glace which had 
 covered the whole country had retired and a later period when 
 some local glaciers crept down from the great mass of ice 
 that still lingered in the north, covering with morainic matter 
 the fresh-water clays which, during the interval, had accumu- 
 lated in pools upon the surface of the older till. 9 Mr. Geikie, 
 to whom I owe this passage, goes on to say that more 
 recently Mr. E. Erdmann, of the Geological Survey of 
 Sweden, has described the occurrence of inter-glacial land 
 and clay in the till or boulder clay of Scania. He believes 
 he has evidence to show that there were more inter-glacial 
 periods than one, for several inter-glacial deposits separated 
 by intervening masses of till occur one above another. In 
 one of the inter-glacial beds fresh-water shells were detected. 
 This bed was covered by seventy feet of boulder-clay. 1 
 
 Leaving the borders of the Baltic and going southwards, 
 Penck claims to have found in Saxony two boulder clays, 
 
 6 Id. 195. 7 Great Ice Age, 405, 406. 
 
 8 Kongl, Vetenskaps-Akademiens Forhandlingar 1873, Na^6. 
 
 9 Id. 1873, No. 1. * Great Ice Age, 407. ' 
 
Penck. 321 
 
 separated by flood and river gravel and sand, pointing to a 
 double glacial period there also. 2 
 
 Penck has especially examined the evidence in upper 
 Bavaria and the Tyrol. He devotes a chapter of his excel- 
 lent work on the " Glaciation of the German Alps," published 
 in 1882, to the red and white breccias which occur at Hot- 
 tingen near Innsbruck, near the village of Wallgau on the 
 Isar, and on the way from Weissenbach to the pass of Gafcht, 
 and contain plant remains. 
 
 These breccias, according to Penck, are distinctly overlaid 
 and underlaid by moraine matter, and he thence concludes 
 that there is conclusive evidence there of at least two and 
 perhaps three glacial periods, since it would not be possible 
 to explain the facts, namely the occurrence of these plants, at 
 the high level where they are found by a mere oscillation of 
 a glacier. 3 
 
 He then turns to the thin beds of coal, enclosed in con- 
 glomerates, found in the Algau on the banks of the Iller near 
 Sonthofen, and at an elevation of 900 metres above the 
 sea level. Penck aflirms of these^ coal-bearing conglomer- 
 ates that 4 they also are under and overlaid by morainic 
 debris, and therefore prove the existence of two sharply 
 denned glacial periods. To use his own words : " Im Iller 
 und Innthale sind zwei zeitlich scharf getrennte Verglets- 
 cherungen nachweisbar." * 
 
 He similarly concludes that the false bedded sands and 
 water- worn gravels, to which the name of Nagelfluh is given 
 by the German geologists and which intervene among the 
 morainic beds, similarly point to an intercalation of other 
 conditions between glacial epochs. 5 
 
 He further argues that in Bavaria as in the Aargau and in 
 upper Italy, the occurrence of outer moraines points to at least 
 three glacial periods, not merely local oscillations of glaciers, 
 but affecting the whole Alpine country. 6 In a subsequent 
 page he points out that it is only in the highlands of the 
 Alps of Scotland and of Norway where direct evidence of mani- 
 fold glaciation is forthcoming, and that when we get into the 
 lowlands this evidence fails us. 7 
 
 2 Id. 186-193. 3 Penck, op. cit. 228-251. - 4 Id. 251-262. 
 
 5 Id. 266-290. 6 Id. 312-bl3. 7 Id. 324. 
 
322 The Glacial Nightmare. 
 
 Lastly, we may shortly turn to the New World. In 1879 
 Mr. McGee published a paper in the Geological Magazine on 
 the Geology of the Mississippi Valley, in which he separates 
 the surface deposits of that region into six sections, which he 
 contends were formed during three distinct periods, probably 
 separated by considerable intervals. " It is believed also/' he 
 says, " that all are due either directly or indirectly to glacial 
 action, and hence that they afford evidence of three separate 
 glaciations." After describing these sections in detail, he con- 
 cludes that " three different glacial epochs have succeeded each 
 other at long intervals j and from analogy with the present as 
 well as from direct evidence, that a characteristic fauna and 
 flora spread over the laud during each of these intervals, whose 
 counterparts are to be sought for in the more recent marine 
 formations. . . . It may also be suggested that earlier glacial 
 eras have probably occurred at long intervals." 8 
 
 Professor Geikie has collected a large number of observations 
 from North America in regard to buried layers of peat and 
 trees, etc., which he claims as evidencing inter-glacial condi- 
 tions. The most satisfactory instances are from the State of 
 Ohio. 
 
 In that State, a number of years ago, Colonel James 
 Whittlesey published a section, obtained in digging an artesian 
 well at Columbia, in which a layer of blue clay enclosing a log 
 of wood was found covered by drift beds with boulders, and 
 underlaid by the stiff glacial clay known as hard pan. 9 
 
 Mr. Orton, in his account of Clermont County in the same 
 State, describes an ancient forest bed formed of carbonaceous 
 clay containing vegetable matter, leaves and wood, with 
 occasional beds of peat as intercalated between the blue 
 boulder clay or hard-pan and yellow clays with boulders. 1 
 
 Mr. Geikie says that the same observer has described as 
 occurring in the valley drifts of Ohio a forest bed seven feet 
 thick between two seams of ochreous gravel which he argues 
 grew in situ, during inter-glacial times. He is further of 
 opinion that " this inter-glacial stage must have had an im- 
 
 8 Geol. Mag. N. S. vol. vi. 354 and 418. 
 
 9 Smithsonian Contributions to Knowledge, xv. 
 
 1 Eeport of the Geol. Survey of Ohio, vol. i. Geology, 440r 
 
Winchell, 323 
 
 mensely long period for the accomplishment of the work which 
 we are obliged to refer to it." 2 
 
 In Minnesota Professor Winchell describes a bed of peat 
 occurring in Fillmore County, as underlying a gravelly clav, 
 " which contains boulders,, and which has the appearance of 
 being the same as the glacier deposit known as unmodified drift. 
 The same underlies it. Some portions of the clay above the 
 peat are reported to be blue, while the whole of that which 
 underlies it is of a blue colour. The country is heavily covered 
 with this drift clay, and some very large boulders of granite 
 lie near the wells that have met this peat." s 
 
 Mr. Geikie says that Professor Winchell had written to him 
 to the effect that in North- Western Ohio,which he had surveyed, 
 he found satisfactory proofs of the existence of two distinct 
 glacier deposits. The buried soil of Ohio, which extends 
 westwards through Indiana and Illinois, enters Iowa and even 
 appears in southern Minnesota, " is distinctly intercalated," he 
 says, " between glacier drift clays. . . . The two deposits 
 cannot be said to differ noticeably wherever I have examined 
 them, the overlying mass as being evidently of glacier origin as 
 the drift that underlies the soil-bed." 4 Mr. Geikie, commenting 
 on this, says, (< We must agree with Professor Winchell 
 that if the lower hard-pan be the morainic matter of a great 
 ice sheet, the similar deposit which overlies the forest bed in 
 Minnesota, etc., must have had a like origin ; in other words, 
 there was a return to glacial conditions after the forest bed 
 period, during which the ice sheet again invaded New England 
 and the North- Western States." 5 He further argues, from 
 the very small traces of erratic deposits in the so-called drift- 
 less area, that there were certain regions over which the older 
 ice sheet prevailed, but which the last ice flow did not cover. 6 
 
 The opinion of the best American geologists seems to be 
 very emphatically against a plurality of glacial periods, but 
 most of them claim that there were at least two. Thus, Mr. G. 
 K. Gilbert says, in America, where there is now great activity 
 in the investigation of glacial phenomena, the evidence of a 
 single interglacial period is cumulative and overwhelming, 
 while there is no indication of more than one. 
 
 " Op. cit. p. 428; Great Ice Age, 453. 3 Great Ice Age, 454. 
 
 4 Id. 462, 463. 5 Id. 463. Id. 465 
 
 Y 2 
 
324 The Glacial Nightmare. 
 
 Dr. Wright, in his most valuable book on the ice age in 
 North America, says, "I presume the majority of authorities 
 of most weight would prefer to speak of two glacial epochs in 
 America, believing that the ice-front withdrew to the Lauren- 
 tian highlands during the so-called interglacial period, and 
 thence re-advanced to the outer line of moraines which we 
 described in the chapter upon that subject." 7 
 
 This will suffice ; I have tried to state the case of those who 
 hold that there has been more than one glacial period as 
 fully and fairly as my space will allow. 
 
 This completes my survey of the historical development of 
 the glacial theory as taught by its more ardent advocates. It 
 is time that I should turn to the criticism which I propose 
 to apply to it. 
 
 7 Op. cit. 475. 
 
325 
 
 CHAPTER IX. 
 
 APPEALS TO TRANSCENDENTAL PHYSICS AND ASTRONOMY. 
 
 " Of late years, however, the opinion has been gaining ground among 
 our hammer-bearers that on this matter of cosmical climate they must, 
 after all, be content to follow the guidance of the astronomer and the 
 physicist, seeing that their own principles refuse, in this particular at 
 least, to yield as much assistance as could be desirable." J. GEIKIE, 
 " Great Ice Age," 94. 
 
 Appeal of G-eology to Physics and Astronomy Decrease in subterranean 
 heat : Lapperent, Laplace, Poisson, Hopkins, Thompson, Phillips, 
 Croll Different effects of subterranean heat on land and sea : 
 Frankland, Dawson, Croll Renoir's theory Variation in the 
 sun's heat: Vicaire, Balfour Stewart, Eenoir Hotter and colder 
 regions of space : Poisson and Hopkins Variation in the trans- 
 parency of space : Erman and Babinet Variation in the trans- 
 parency of the air Increased cold alone will not produce a glacial 
 epoch: Lecoq, Tyndall, Frankland, Croll, Woeikof Shrinkage of the 
 sun's disc : Blandet, D'Archiac, Lapperent, S. Reinach Shifting of 
 the earth's axis : I. If the earth is rigid : Alessandro degli Alessan- 
 dri, E. Hill, Jukes, Thompson, Airy, G. Darwin, Croll, J. Evans, 
 Twisden II. If the earth contains a molten centre: J. Evans, 
 Thompson, G. Darwin, Airy, E. Hill G-eological witness against a 
 shifting axis : Jukes, Haughton, Hutton, J. Geikie, Boucheporn, 
 Falsan Variation in the eccentricity of the earth's orbit : I. Effect 
 on mean temperature of the earth : Kepler, Croll and Carrick, 
 Herschell, Haughton, Meech and Croll II. Its effect on the 
 seasons : Kepler, Reclus, Lambert, Meech, Croll, Arago, Geikie 
 Variation in the obliquity of the ecliptic : Herschell, Laplace, 
 Stockwell, Ball, Drayson, Belt, Meech, Croll, G. Darwin Effects 
 of precession and nutation : Airy, Burr, Reclus, Meech, Humboldt, 
 Price General conclusion Croll, Geikie and Meech. 
 
 WE have now traced the growth and culmination of the 
 glacial theory as taught by its most distinguished advocates. 
 We have shown how they claim that the earth not long ago 
 was largely enshrouded in ice even down to the tropics. 
 That similar invasions have occurred over and over again in 
 geological time, and that it is very largely to this agency we 
 
326 The Glacial Nightmare. 
 
 must attribute the excavation of valleys and the formation of 
 lakes and fiords. 
 
 It is not unnatural that in days when science means, if 
 it means anything, the explanation of natural phenomena by 
 known and not by occult causes, the champions of the Glacial 
 theory as thus taught should have summoned the astronomer 
 and the physicist to their aid. The theory, as developed 
 by Agassiz and his scholars, postulates a portentous change 
 of climate operating over wide areas of the earth's surface ; 
 and those who tolerate no appeal involving the tamper- 
 ing with the known laws of Nature must needs find some 
 cause competent and adequate to so great a result. It 
 was natural, also, that the students of astronomy and of 
 physics, some of whom were fascinated by the theory 
 which was so pertinaciously pressed upon them, should have 
 made corresponding efforts to find a basis for the new 
 departure in geology. Nor, indeed, was the appeal entirely 
 a new one. Long before the glacial theory was heard of, 
 geologists had been brought face to face with the fact that in 
 the temperate regions of the earth remains of plants and 
 animals are forthcoming whose relatives now live only in the 
 tropics. When bories which belonged unmistakably to 
 elephants and rhinoceroses were found not only in Britain 
 and Geimany but under the perpetually frozen soil of Siberia, 
 and when, on the other hand, bones of the reindeer, the 
 glutton, and the polar bear were forthcoming from France and 
 England, it became impossible to doubt that in former times 
 the climate of the earth .must either have been different in 
 itself or differently distributed, and the physicist and 
 the astronomer were set to work to find an adequate cause. 
 The problem which now arose only differed from this older 
 problem in involving not only a more revolutionary change 
 but apparently an oscillation to and fro. To those who were 
 champions of Uniformity as well as of the Glacial theory it 
 was natural and necessary, perhaps, that they should 
 demand a cause which was not sporadic, abrupt and aimless, 
 like some single experiment made by nature, but one which 
 acted in a recurrent manner, and thus created a kind of 
 cosmic change of seasons at intervals during the long years 
 that are gone. 
 
Diminution of Subterranean Heat. 327 
 
 I propose to devote the present chapter to the consideration 
 of the various theories which have been forthcoming in response 
 to this demand, and to show how completely they fail in 
 meeting the conditions of the problem. I shall begin with a 
 cause which is terrestrial rather than astronomical, namely, 
 the decrease in the original heat of our planet. This 
 cause has been frequently urged, since men's minds were 
 impressed with the notion that the earth has been gradually 
 cooling from a molten mass, and have in consequence argued 
 that during the earlier geological periods its temperature was 
 very considerably higher than it is now. 
 
 M. Lapparent l remarks with truth that if we are to measure 
 the effect of subterranean heat by its effect in the case of 
 volcanoes we shall find that it is exceedingly local and limited, 
 since it does not suffice to melt the snow which so often 
 covers the flanks and even summits of volcanoes. This is 
 specially noteworthy in Iceland, which is a cradle of volcanic 
 energy and proves what a bad conductor the rocky fabric of 
 the earth is. If we remember also that where we can meet 
 with fairly complete sections the azoic gneiss and mica schist 
 which form the foundations of the earth's crust and underlie 
 the Cambrian and Silurian strata are found to be thousands 
 of feet thick, we have a measure of the very slight influence 
 the central heat had in constituting the climate of the earliest 
 geological horizons where traces of life occur. Nor is there 
 evidence forthcoming from the geological record either that 
 in palseozoic times the earth was substantially hotter than it 
 is now or that there has been any gradual process of change 
 in the later ages from a warmer to a cooler climate. Nor, 
 again, is it easy to see how, even if this were so, it would 
 in any way account for a glacial climate followed, as the 
 postulated glacial climate has been, by a temperate one, 
 unless it be argued, which it is not, that the subterranean 
 heat of the earth is subject to oscillations. 
 
 Again, as Dr. Wright says, if this theory were tenable, the 
 Glacial period should not have culminated in the past, but we 
 should still be looking for its culmination in the future; for 
 both the earth and sun are cooling off. 2 
 
 1 Traite de Geologic, 1253-1251. 3 Ice Age in America, 407. 
 
328 The Glacial Nightmare. 
 
 Apart from these arguments, there remains a more crucial 
 one. The actual amount of heat which the surface of the 
 earth derives from subterranean sources has been calculated. 
 Laplace arguing from astronomical observation, made in the 
 time of Hipparchus, showed that during the last 2000 years 
 there has been no appreciable contraction of the earth by 
 cooling, for the length of the day has not been sensibly 
 shortened, not even to the amount of g-^^th of a second, so 
 that the contraction of the earth must have been inappreciably 
 small or none at all, as it could not take place without affecting 
 the length of the day. 
 
 Poisson calculated the actual amount of the superficial heat 
 due to the earth's cooling, subject to the conditions that the 
 power of transmitting heat is the same in every part of 
 it, and also that the temperature of the surrounding space 
 remains uniform. 
 
 The result is thus stated by Hopkins : " It is very small, 
 amounting only to about -g^th ^ a degree of Fahrenheit. It 
 must have been constantly diminishing for an immense period 
 of time, and has now approximated so near to its ultimate 
 limit, that if the earth's refrigeration should continue under 
 the same external conditions as at present, it would require, 
 as shown by Poisson, the enormous period of a hundred 
 thousand millions of years to reduce this small fraction to half 
 its actual value. Hence the mean superficial temperature of 
 the earth, and consequently the mean temperature of the 
 atmosphere in contact with the earth's surface may be con- 
 sidered as sensibly unalterable for all future time, provided 
 always that the sun, the earth's atmosphere, and the tempera- 
 ture of surrounding space shall remain unchanged. We are 
 here, however, more immediately concerned with past than 
 future changes. We should not be justified in supposing, 
 from what I have stated respecting the slow future variation 
 of the earth's superficial temperature, that it has been equally 
 slow for an equal period of past time; but still it is highly 
 probable that some millions of centuries must have elapsed 
 since the mean superficial temperature could have been 
 greater by a single degree than at present, from the operation 
 of the causes we are now discussing, assuming the permanence 
 of the conditions above-mentioned. However imperfect' our 
 
Variation in the conductivity of the Atmosphere. 329 
 
 geological chronology may be, these enormous periods of time 
 seem almost to preclude the possibility of applying this theory 
 to account for the changes of temperature of which we have 
 evidence in the more recent geological periods." 3 
 
 Sir William Thompson has shown that 10,000 years after 
 the solidification of the earth's crust first began, the amount 
 of heat which would traverse it could hardly affect the surface 
 temperature. He enlarges upon the subject in his usual pic- 
 turesque way : " Ten, twenty, thirty times the present rate of 
 augmentation of temperature downwards could not raise the 
 surface temperature of the earth and air in contact with it by 
 more than a small fraction of a degree Fahrenheit. The earth 
 might be a globe of white-hot iron covered with a crust of 
 rock 2000 feet thick, or there might be an ice-cold temperature 
 everywhere within thirty feet of the surface, yet the climate 
 could not on that account be sensibly different from what it 
 is or the soil be sensibly more or less genial than it is for 
 the roots of trees or smaller plants. Yet greater underground 
 heat is the hypothesis which has been most complacently dealt 
 with by geologists to account for the warmer climates of ancient 
 times." 4 
 
 Professor Phillips, while accepting the results which 
 Thompson, Poisson, Fourier and others had arrived at, and 
 the direct conclusion which they drew from it, went on to 
 argue that possibly in former times the atmosphere varied in 
 its capacity as a conductor of heat, and that consequently the 
 heat of the earth was retained longer or dissipated sooner as 
 the case might be. " The state of the atmospheric mantle," 
 he says, " which envelops the terraqueous globe, mitigates 
 solar heat and stellar radiation, and, like the clothing of a 
 steam cylinder, prevents excessive waste of the warmth 
 treasured within/' 
 
 To this argument Mr. Croll replies : "It is quite true, as 
 Professor Phillips suggests, that a diminution in the conduc- 
 tivity or in the diathermancy of the earth's atmosphere, and 
 an increase in its height, would increase the influence of 
 internal heat on the climate. But when we reflect that under 
 the present condition of the atmosphere the internal heat could 
 
 s Journal Koyal Geol. Soc. viii. 58, 59. 
 
 4 Geological Time, Trans. Geol. Soc., Glasgow, 1877, p. 250. 
 
330 The Glacial Nightmare. 
 
 not even sensibly affect the climate after the short period of 
 10,000 years from tha commencement of solidification of the 
 earth's surface, it appears very improbable that our atmosphere 
 could have ever been so far different from what it is at present, 
 that by means of it the internal heat could have produced and 
 maintained that high temperature of climate which is supposed 
 by some to have prevailed during the long Palasozoic ages of 
 the earth's history. And besides, the important fact is over- 
 looked that any change in the condition of the earth's atmosphere 
 which would prevent the dissipation of the earth's internal 
 heat into surrounding space, such as an increase in the quan- 
 tity of aqueous vapour contained in the atmosphere, would at 
 the same time tend to lower the temperature of the earth's 
 surface by diminishing the quantity of radiant heat reaching 
 the surface from without. Such a state of things would no 
 doubt equalize, to a certain extent, the extremes of summer 
 and winter temperature, but would not very sensibly increase 
 the mean annual temperature of the climate. In fact it would 
 rather have an opposite tendency." 5 
 
 Another modification of the theory that former climatic 
 changes were due to changes in the internal heat of the earth 
 is due to Professor Franklancl, who limits his explanation 
 to the Glacial period only. This he published in a 
 paper in the Philosophical Magazine for May, 1864, on the 
 "Physical Cause of the Glacial Epoch." "A satisfactory 
 theory," says Professor Frankland, "must take cogniz- 
 ance of the following points in the history of the Glacial 
 period: 1st. That its effects were felt over the entire 
 globe. 2nd. That it occurred, or at least terminated, at a 
 geologically recent period. 3rd. That it was preceded by 
 a period of indefinite duration, in which glacial action was 
 either altogether wanting, or was at least confined to regions 
 of considerable altitude. 4th. That during its continuance 
 atmospheric precipitation was much greater, and at one 
 period the height of the snow-line considerably less than at 
 present. 5th. That it was followed by a period extending to 
 the present time, when glacial action became again insignifi- 
 cant." G 
 
 5 Philosophical Mag. 4th ser. xxviii. 122, 123. 
 
 6 Phil. Mag. 4th ser. xxvii. 327. 
 
Frankland J s Theory. 331 
 
 The main position upon which Frankland based his theory 
 was that, in order to secure a sufficient supply of ice to con- 
 stitute a glacial epoch, we must, in the first place, have an 
 adequate amount of aqueous vapour in the atmosphere, and 
 this could only arise from the association of a greater amount 
 of heat with the waters of the ocean, and he concludes that 
 the sole cause of the phenomena of the glacial epoch was a 
 higher temperature of the ocean than that which obtains at 
 present. 
 
 This hypothesis rests chiefly upon the two following 
 propositions : 
 
 1st. That a higher oceanic temperature would give rise to 
 an increased evaporation, and consequently to an augmented 
 atmospheric precipitation. 
 
 2nd. That this increased atmospheric precipitation would 
 augment the average depth of permanent snow upon the ice 
 bearers, and would, within certain limits, depress the snow 
 line. 7 
 
 Having endeavoured to establish these two hypotheses, he 
 next endeavoured to discover a competent cause of a former 
 higher oceanic temperature ; and, having put aside all cosmical 
 causes as insufficient, he fell back upcn the internal heat of 
 the earth. Poisson showed that an increase of the surface 
 temperature of the earth to the extent of 10 Fahr. would 
 involve a descending rate of increase so rapid as to reach 
 200 Fahr. at a depth of only sixty feet, a physical condition 
 of the planet incompatible with the conditions of life in the 
 more recent geological epochs. Frankland, admitting this, 
 goes on to say that although it is true of the solid surface 
 of our planet, it is not necessarily so when extended to the 
 floor of the ocean ; and this (1) because of the conductivity of 
 water ; (2) its convection ; (3) its power of penetrating into any 
 chasms formed in the bed of the ocean ; (4) its specific heat. 
 These, combined, would continually lead to the abstraction of 
 the heat which the ocean floor acquired by the water above it, 
 which again would part witli it by increased evaporation. 
 Such evaporation, as Dalton showed, would increase largely 
 with its heat, and hence would arise the required amount of 
 
 7 Id. 328. 
 
332 The Glacial Nightmare. 
 
 vapour to be condensed into ice and snow, and hence an efficient 
 cause for the Glacial period. 
 
 This theory is certainly ingenious. 
 
 That the enhanced heat which the solid crust of the 
 earth received should be abstracted from it by the ocean 
 waters, which again parted with it by increased evaporation, 
 the whole effect being to make the sea warmer and the dry 
 earth cooler than at present, is a pretty scheme, but it fails at 
 all points. As Principal Dawson says : " It is contradicted by 
 the fossils of the period, which show that the seas were colder 
 than at present, and if it existed it would not produce the 
 results required, unless a preternatural arrest were at the same 
 time laid on the winds which spread the temperature of the 
 sea over the land. - The alleged facts observed in Norway, and 
 stated to support this view, are evidently nothing but the 
 results ordinarily observed in ranges of hills, one side of 
 which fronts cold sea water, and the other land warmed in 
 summer by the sun." 8 
 
 Mr. Crol] replied to Frankland in similar terms. " There is 
 no evidence," he says, " that the ocean was warmer during the 
 glacial epoch than at present ; but, on the contrary, we have 
 geological evidence to conclude that it must have been much 
 colder than at present. For example, on examining the 
 marine drift of that period, we find that it is decidedly of an 
 arctic character, indicating the low temperature of the seas 
 during that epoch. These beds show, for example, that our 
 British seas during that period contained in abundance 
 numerous species of shells which are now only to be found in 
 more northern latitudes. Croll enumerates eighteen Arctic 
 shells which no longer exist in the British seas, which occur 
 in the glacial drift of Scotland alone. 9 
 
 Besides, if this had been an efficient cause of climatic change 
 extending over the continuous history of the earth, we should 
 have had abundant evidence in the rocks of a gradual and 
 persistent change of temperature from one extreme towards 
 another, which we have not. Lastly, such a theory would in 
 no way account for the intercalation of a glacial period between 
 
 s Acadian Geology, 66, 67. 
 
 9 Phil. Mag. 4th. ser. xxviii. 124, 125. 
 
Renoir s Theory. 333 
 
 the temperate and semi-tropical conditions of Tertiary times 
 and those now subsisting. 
 
 The theory ignores the various astronomical and other facts 
 already quoted, and it is based on such postulates as that its 
 effects were felt all over the globe, which is contrary to 
 experience, inasmuch as the glacial phenomena are limited 
 both in time and in space. 
 
 These facts are conclusive, and I do not know that Professor 
 Frankland has a single disciple. 
 
 M. Renoir, having previously suggested two other theories 
 to account for the glacial climate and discarded them because 
 of their insufficiency, suggests a third, viz., that a very 
 intense glacial epoch coincided with the time when the 
 internal heat of the earth ceased to materially affect its 
 surface, and when at the same time the earth was so far 
 from the sun as to greatly mitigate its heating power. 
 Presently, in order to overcome the resistance of the ether in 
 space, the planets travelled more slowly and approached the 
 sun in a great spiral, which led to the melting of the ice 
 previously accumulated. I do not quote this theory to 
 seriously refute it, but only as an example of the way in 
 which some students of a science we are devoted to, are 
 prepared to throw away every kind of empirical test and of 
 scientific method in the pursuit of wild hypotheses of which 
 M. Falsan says, with this one in view : if n'etaient en definitive 
 que de simples jeux d'esprit." * 
 
 We may safely put aside a variation in the earth's internal 
 heat as an efficient cause of changes of climate. Let us now 
 turn to a variation in sources of heat outside the earth ; and 
 first to the sun. 
 
 The researches of astronomers have shown that certain of 
 the stars vary in luminosity, and go through a process 
 of accretion and diminution of light, and in certain instances 
 they have been noticed to break out into immense con- 
 flagration and then to subside again into their normal 
 condition. It has been suggested that the sun has passed 
 through similar phases and thus greatly affected terrestrial 
 climate. We can hardly conjecture a cause which would 
 
 La Periode Glaciaire, 193. 
 
334 The Glacial Nightmare. 
 
 account for the heafc of the sun being immensely enlarged, 
 except either a sudden addition of fuel to its furnace, which 
 would be witnessed by the perturbations which its increased 
 mass would cause among the planets, of which we have 
 no evidence, or a collision with some other mass of matter, 
 and of this we have no evidence either. These rapid and 
 sudden revolutions must be accompanied by the most wide- 
 spread catastrophes, and if the sun had ever been the victim 
 of one we should have had ample traces of it. The appeals 
 generally made, however, have been to milder changes. Such 
 appeals have been countenanced in England by Professor 
 Balfour Stewart, and in France by M. Renoir and M. Vicaire. 
 M. Yicaire 2 argues in favour of a diminution in the sun's heat, 
 which he attributes to a gradual exhaustion of the combustible 
 elements in the sun's photosphere ; but if this were so we 
 ought to have evidence of a gradual refrigeration of the 
 earth, which is not the case, nor can glacial epochs intervening 
 between temperate ones be thus explained. The other 
 inquirers referred to appeal to a change in the amount of 
 the sun's heat occasioned -by a prevalence or absence of sun 
 spots. In regard to this cause, which can be directly 
 experimented upon, the evidence is most unsatisfactory. The 
 recurrence of the spots has been discovered to observe a 
 regular periodicity, but their influence upon temperature seems 
 very small, if any. Wolff 3 showed that the succession of sun 
 spots occurs in regular periods of 11 '11 years, or nine cycles in 
 a century, and he argues that the years in which spots occur 
 are generally drier and more productive than others. Buys- 
 Ballot was also of opinion that the prevalence of sun spots 
 has a slight effect on climate. On the other hand, Counsellor 
 Schwab, after twenty-six years of observation, does not think 
 the spots exert any influence on the annual temperature. "A 
 writer in the Encyclopaedia Britannica, article Astronomy, 
 says that in 1827 the summer was wet and cold, the ther- 
 mometer at Paris rose only to 23 '7 Reaumur, and the sun 
 exhibited no spots ; whereas in the summer of 1807 the heat 
 was excessive, and the spots of vast magnitude. Warm 
 
 2 Bull. Soc. Geol. de France, ii. 211. 
 
 3 Comptes Eendus, xxxv. 704. 
 
Wanner and colder Regions in Space. 335 
 
 summers and winters of excessive rigour have happened in the 
 presence or absence of spots." ' 
 
 "We have no warrant, therefore, for supposing that an 
 increase or decrease of spots would greatly modify the earth's 
 climate, nor have we the slightest evidence that such a 
 variation on a great scale occurs in them at long intervals. 
 The theory is really without any basis of empirical fact. 
 
 Let us now move on. 
 
 It has long been known that the sun with its system of 
 planets has a proper motion of its own from one part of 
 space to another. Struve calculated that if it were seen 
 from the mean distance of stars of the first magnitude the 
 sun would appear to describe an arc of about Jrd of a second 
 in a year. If this motion, says Mr. Hopkins, were rectilinear 
 and uniform it would require a period of nearly 700,000 years 
 for the sun to move over a space equal to the distance of the 
 stars of the first magnitude from the earth; and he urges 
 that, whatever mistakes may underlie the hypothesis of Struve, 
 the sun's true motion would not be materially greater than 
 his estimate. Poisson argued that the result of this motion 
 might be to take our system into regions of space where a 
 large number of heat rays are forming. This view seems to 
 be based on a misapprehension. We know of no source of heat 
 external to the earth except solar and stellar heat. Inasmuch 
 as the fixed stars are situated at enormous distances from 
 each other or they would cause movements which we could 
 verify, this result would mean that the solar system had 
 approached a good deal nearer to some particular star. 
 ' This approximation of position to a single star," says Mr. 
 Hopkins, " could not take place consistently with the preserva- 
 tion not only of the remoter bodies of the solar system, but 
 even of the motion of the earth about the sun, according to 
 its present laws." 
 
 Suppose our sun should, approach a star within the present 
 distance of Neptune. That planet would no longer remain a 
 member of the solar system, and the motions of the other 
 planets would be disturbed in a degree which no one has ever 
 contemplated as probable since the existence of the solar 
 
 4 Meech, Smithsonian Contributions, ix. 35, 36. 
 
336 The Glacial Nightmare. 
 
 system. But, such a star, supposing- it to be no larger than 
 the sun, and to emit the same intensity of heat, would not 
 send to the earth much more than one thousandth part of the 
 heat which she derives from the sun, and would therefore 
 produce only a very small change of terrestrial temperature." 5 
 The argument just quoted applies to the case of the earth 
 approaching a source of heat and acquiring a correspondingly 
 high temperature, but the same reasoning applies to the 
 reverse problem. In order that it should be cooled and 
 acquire glacial conditions by a similar movement, it must 
 approach some reservoir of cold corresponding to a star as a 
 source of heafc, of which we know of none in nature, since it 
 would hardly affect the temperature of the earth at all if it 
 were to pass out of the sphere of radiation of all the fixed 
 stars, and yet retain its present proximity to the sun, which 
 supplies it virtually with all its external heat. 
 
 We may, therefore, put aside this cause as inadequate and 
 impossible. It has been urged by Erman and Babinet that 
 the amount of heat received by the earth may be affected by 
 certain parts of space being denser or more occupied by 
 meteoric dust. It is well known that the earth at certain 
 parts of its orbit encounters showers of meteoric dust and so- 
 called shooting stars, and Erman attributes to the effect of 
 these obscuring the sun, the cooler temperature which is said 
 to have been observed between the 7th and 10th of February 
 and the 8th and 13th of May. This view has been extended 
 so as to include cosmic periods, and it has been suggested that 
 a similar cause may have produced changes of climate, but 
 meteoric dust and showers of shooting stars are parts of the 
 solar system, miniature planets in fact, moving in definite and 
 defined orbits, and it is only when the earth traverses their 
 orbit that the phenomenon occurs. There is no evidence that 
 space is sporadically occupied by showers or by areas of this 
 kind unconnected with the solar system. 
 
 Others, again, have urged that space itself, even if it does not 
 contain areas filled with meteoric dust, may differ in trans- 
 parency - or density. If this were the case, and the earth had 
 traversed any considerable space of it, it must have retarded its 
 
 6 Journ. Geol. Soc., vol. viti. bl, 62. 
 
Heat as requisite as Cold for a Glacial Period. 337 
 
 movements, or it would certainly, continually or occasionally, 
 affect and obscure the light of the fixed stars, of neither 
 of which facts is there any evidence. So far as observed facts 
 enable us to form an opinion, it is virtually certain that space 
 is uniformly transparent, and does not contain areas marked 
 by special opacity. 
 
 A similar view has prevailed in regard to the earth's atmo- 
 sphere, namely, that it has varied in its capacity for trans- 
 mitting solar rays. Of this, again, we have no evidence 
 whatever, nor can we suggest a physical cause to account for 
 such a variation which should explain the vicissitudes of 
 geological climate or the necessities of animal and vegetable 
 life, since a cause like this can hardly have been a local one, 
 and must have prevailed over the whole atmospheric envelope. 
 I am not referring to possibilities of greater or less vapour in 
 the air, etc., which we shall consider in the next chapter, but 
 the presence in, or the absence from, the air of elements other 
 than vapour which would render it more or less opaque as the 
 case might be. 
 
 In regard to the theories previously discussed, there is a 
 further answer which seems conclusive. It was first 
 suggested, I believe, by M. Lecoq, and I will quote his neat 
 statement of it in his own words. " La principale cause 
 d 7 alimentation du glacier reside dans Tabondance du neve et 
 dans 1'etendue du cirque qui le re9oit. Les ressources du neve 
 n'existent que dans la neige qui tombe ; la neige ne peut se 
 former qu'aux depens de la vapeur elevee dans Patmosphere, 
 et celle-ci ne peut etre produite que par Faction de la chaleur 
 solaire." 6 
 
 Similar views were afterwards published and clearly stated 
 by Professor Tyndall. " Some eminent men," he says, "have 
 thought, and some still think, that the reduction of temperature, 
 during the glacier epoch, was due to a temporary diminution 
 of solar radiation; others have thought that, in its motion 
 through space, our system may have traversed regions of low 
 temperature, and that during its passage through these 
 regions, the ancient glaciers were produced. If I understand 
 the writings of the eminent men who have propounded and 
 
 Des Glaciers et des Climats, 1847, p. 553. 
 
338 The Glacial Nightmare. 
 
 advocated the above hypotheses, all of them seem to have 
 overlooked the fact that the enormous extension of glaciers 
 in bygone ages demonstrates, just as rigidly the operation of 
 heat as well as cold. Cold alone will not produce glaciers. 
 You may have the bitterest north-east winds in London 
 throughout the winter, without a single flake of snow. Cold 
 must have the fitting object to operate upon, and this object 
 the aqueous vapour of the air, is the direct product of heat. 
 The latent heat of aqueous vapour at the temperature of its 
 production in the tropics is about 1000 Fahr. ; for the latent 
 heat augments as the temperature of evaporation descends. A 
 pound of water, then, vaporized at the equator, has absorbed 
 1000 times the quantity of heat which would raise a pound 
 of the liquid one degree in temperature. But the quantity of 
 heat that would raise a pound of water one degree, would 
 raise a pound of cast iron ten degrees : hence, simply to con- 
 vert a pound of the water of the equatorial ocean into vapour, 
 would require a quantity of heat sufficient to impart to a pound 
 of cast iron 10,000 degrees of temperature. But the fusing 
 point of cast iron is 2000 Fahr., therefore for every pound of 
 vapour produced, a quantity of heat has been expended by 
 the sun, sufficient to raise five pounds of cast iron to its 
 melting point. Imagine, then, every one of those ancient 
 glaciers with its mass of ice quintupled ; and imagine the 
 place of the mass so augmented, to be taken by an equal 
 weight of cast iron raised to the white heat of fusion, we 
 shall then have the exact expression of the solar action, in- 
 volved in the production of the ancient glaciers. Substitute 
 the hot iron for the cold ice our speculations would instantly 
 be directed to account for the high temperature of the glacial 
 epoch, and a complete reversal of some of the hypotheses above 
 quoted would probably ensue. 
 
 It is perfectly manifest, that by weakening the sun's action, 
 either through a defect of emission, or by the steeping of 
 the entire solar system in space of a low temperature, we 
 should be cutting off the glaciers at their sources. Vast 
 masses of mountain ice indicate infallibly the existence of 
 commensurate masses of atmospheric vapour, and a propor- 
 tionately vast action on the part of the sun. In a distilling 
 apparatus, if you required to augment the quantity distilled, 
 
Heat as requisite as Cold for a Glacial Period. 339 
 
 you would not surely attempt to obtain the low temperature 
 necessary to condensation by taking the fire from under your 
 boiler ; but this, if I understand them aright, is what has been 
 done by those philosophers who have sought to produce the 
 ancient glaciers by diminishing the sun's heat. It is quite 
 manifest that the thing most needed to produce the glaciers 
 is an improved condenser ; we cannot afford to lose an iota of 
 solar action ; we need, if anything, more vapour, but we need 
 a condenser so powerful that this vapour, instead of falling in 
 liquid showers to the earth, shall be so far reduced in tem- 
 perature as to descend in snow." 7 
 
 Professor Frankland puts Tyndall's view in quite as 
 picturesque and convincing a manner. " The formation of 
 glaciers," he says, " is a true process of distillation, requiring 
 heat as well as cold for its due performance. The produce of 
 a still would be diminished, not increased, by an absolute 
 reduction of temperature. A greater differentiation of tem- 
 perature is required to stimulate the operation into greater 
 activity. . . . Professor Tyndall proves conclusively that all 
 such theories assume a condition of things which would cut 
 off the glaciers at their source, by diminishing the evaporation 
 upon which their existence essentially depends." 8 
 
 Croll urges that what Professor Tyndall says shows plainly 
 that the glacial epoch was not brought about by our earth 
 passing through a cold part of space. A general reduction of 
 temperature over the whole globe certainly would not produce 
 a glacial epoch. Suppose the sun were extinguished and our 
 globe exposed to the temperature of stellar space (239 F.) ; 
 this would certainly freeze the ocean solid from its surface to 
 its bottom, but it would not cover the land with ice. 9 
 
 Woeikof says : " A cursory glance at the present condition 
 of our globe shows us that cold alone will not produce 
 permanent snow and glaciers when vapour of water is 
 deficient. There are no permanent snow or glaciers in the 
 Verkhojansk Mountains in North-east Siberia, yet at the foot of 
 them the mean annual temperature is below 4 F. and that of 
 
 7 Heat as a mode of Motion, 188, 189. 
 
 8 Phil. Mag. 4th ser. xxvii. 325. 
 
 9 Climate and Time, 79, 
 
 Z 2 
 
34-O The Glacial Nightmare. 
 
 January below 56 F. The reason is that the snow-fall is 
 but small, and thus the snow is easily melted in summer. 
 In New Zealand, on the contrary, owing to the enormous snow- 
 fall in the mountains, glaciers descend to about 700 feet above 
 sea level on the west side (lat. 43 S.) . At this height the mean 
 annual temperature mnst be about 50 F. Snowfall and frost 
 " are of rare occurrence, even in winter/' * 
 
 These arguments seem to me to be conclusive. 
 
 In the theories already discussed the Supposed influence 
 upon the earth's climate was distributed over its whole sur- 
 face ; we will now turn to a cause which would, if effective, 
 have a differential effect in different areas. The view here 
 criticized was published by M. Blandet, 3 and adopted by 
 D'Archiac, Lapperent and others. According to it the sun is 
 gradually but definitely shrinking in size, and it is by the 
 contraction and concentration of its sphere that the greater 
 part of its heat is generated. Consequently, the apparent 
 diameter of the sun was formerly much greater. In con- 
 sequence of the great distance separating us from the sun and 
 of its small apparent disc, the rays which reach the earth 
 from it form a virtually parallel bundle which illuminate a 
 hemisphere at a time and no more. If its apparent diameter 
 were increased to 47, other things remaining alike, the solar 
 rays, instead of forming a cylinder, would form a cone, whose 
 base would be the solar disc. The result would be that if one 
 limit of illumination was one pole, the other limit would be 
 the parallel of 43 on the opposite hemisphere. Thus a con- 
 siderably larger space than one hemisphere would be con- 
 tinually illuminated, and nowhere on the earth would the 
 nights be of forty-eight hours. 
 
 In such a case the sun's density would be very much less and 
 its texture would be largely nebulous afnd it would give for each 
 unit of surface less heat and light than it does now; but 
 being much larger in size, and the earth being much nearer 
 its periphery, the actual amount of heat received might be 
 the same, while the climate produced would be one of much 
 greater uniformity. 8 
 
 1 Nature, xxv. 423. 
 
 2 Bull. Soc. Geol. de France, 2nd ser. xxv. 777. 
 
 3 Lapperent, p. 34, 35, and 1257. 
 
The Shifting of the Earth's Axis. 341 
 
 Such is the theory in question. 
 
 It is quite true if Laplace's nebular theory be sustained, 
 that at some time in the very remote geological history of the 
 earth such a condition of the sun as is here pictured must 
 have ensued. It is also true that the evidence of the earlier 
 geological periods points to a greater uniformity of climate 
 then than now, but it is a long way from these statements to 
 the conclusion that in recent geological times the sun has 
 shrunk in this portentous way ; besides such a cause might 
 account for a continuous and regular change of climate in 
 some particular direction ; but, as M. S. Reinach says, how 
 are we to explain by it how the so-called glacial epoch came to 
 an end ? Any theory which accounts for the beginning of the 
 glacial epoch must also explain its close. Nor do I know of 
 any astronomer who would countenance the notion that the 
 sun has shrunk so rapidly in geological times from normal 
 causes as to justify such a postulate, which is indeed merely an 
 example of the perverse ingenuity of science rather than of its 
 sober logic. 
 
 Let us pass on. 
 
 It has been a favourite speculation ever since men 
 realized that the remains of tropical animals have been found 
 in temperate and arctic latitudes, while similar remains of 
 animals of an arctic type have occurred far to the south of 
 their present habitat, that the Poles of the earth, and conse- 
 quently its Equator, have shifted their position, thus shifting 
 the position of the areas marked by extremes of heat and cold 
 on the earth. 
 
 The notion is at least as old as the fifteenth century, and 
 may be found in a work by Alessandro degli Alessandri, 
 entitled Lies Geniales, 4 and became a favourite theory with 
 the naturalists who had to explain how an elephant could 
 ever have lived in Europe. 
 
 The shifting of the poles and the equator, according to these 
 philosophers, was due to a shifting of the earth's axis within 
 the earth itself so that it pierced it through different points 
 and in a different direction. The suggestion that a body like 
 the earth, moving at the rate it does, could, without a very 
 
 4 Lib. v. ch. ix. 
 
342 The Glacial Nightmare. 
 
 serious dislocation, alter the direction of its axis under any 
 circumstances, seems fantastic to the mathematician. As 
 Mr. Hill says : " Mathematicians may seem to geologists almost 
 churlish in their unwillingness to admit a change in the earth's 
 axis. Geologists scarcely know how much is involved in 
 what they ask. They do not seem to realize the vastness 
 of the earth's size, or the enormous quantity of her motion. 
 When a mass of matter is in rotation about an axis, it cannot 
 be made to rotate about a new one except by external force. 
 Internal changes cannot alter the axis, only the distribution 
 of the matter and motion about it. If the mass began to 
 revolve about a new axis, every particle would begin to move 
 in a new direction. What is to cause this ? When a cannon- 
 ball strikes an iron plate obliquely, the shock may deflect it 
 into a new direction. The earth's equator is moving faster 
 than a cannon-ball. Where is the force that could deflect every 
 portion cf it, and every particle of the earth into new directions 
 of motion ? " 5 
 
 This argument applies to every possible case, and notably 
 to the case where the earth is postulated to be a regular sphere, 
 every axis of which is an axis of figure ; but the case becomes 
 much more difficult, or rather more impossible to realize, 
 when we' consider that the earth is not a perfect sphere. It 
 is, in fact, a spheroid, with an equatorial diameter larger by 
 many miles than its polar diameter, and it thus has a huge 
 solid ring round it. If the earth has cooled down from a 
 liquid mass, and has rotated at its present speed ever since, 
 the oblateness of its shape must have been originally as great 
 as it is now, and it must have had ab inittOj a protuberant 
 shell bulging out beyond the form of a true sphere, till it 
 reached to the extent of 13 J miles, or nearly 700,000 feet, 
 about its equator. " It is very difficult," says Professor 
 Jukes, " to see what force, internal or external, could have 
 given to a globe thus weighted and balanced all round, such 
 a permanent tilt as would cause it to spin on any other than 
 its shortest diameter, or could so alter its form as to make any 
 other diameter shorter than its original axis of rotation." ( 
 
 We know of no cause competent to alter the position of 
 
 5 aeol. Mag. 1878, 265. 6 Athenaeum, Sept 8, 1860, 323. 
 
The Shifting of the Earth's Axis. 343 
 
 the axis, except the removal of a large quantity of matter 
 from one part of the earth to the other, so as to alter its 
 centre of gravity, and in this way to cause a gap between its 
 new axis of figure and its old axis of revolution. If this 
 were done on a great scale, and rapidly, it would no doubt 
 cause a change in the latter axis, and, as Professor Haughton 
 says, the earth would begin to wabble, and it would continue 
 to wabble as a top does when going to sleep, until the 
 two axes had again coincided. Meanwhile, however, the 
 crust of the earth would be tremendously shattered and 
 dislocated, and, as Mr. Twisden has shown, two vast tide 
 waves would sweep the earth, submerging the equator every 
 150 days to a depth of six miles or more. Of this we have no 
 evidence at all. Suppose, on the other hand, that instead of 
 this change having been rapid, in accordance with the 
 recognized theory of uniformity, the change were very 
 slow and the " result of expansion and contraction, up- 
 heaval and depression, denudation and deposition." These 
 forces, no doubt, are slowly altering the distribution of solid 
 matter in the earth, and it then becomes a question of calcula- 
 tion what amount of such changes would have an appreciable 
 effect in changing the position of the axis. 
 
 Sir Win. Thompson says : " As to changes of the earth's 
 axis, I need not repeat the statement of dynamical principles 
 which I gave with experimental illustrations three years ago ; 
 but may remind you of the chief result, which is that, for 
 steady rotation, the axis round which the earth revolves must 
 be a ' principal axis of inertia/ that is to say, such an axis 
 that the centrifugal forces called into play by the rotation 
 balance one another. The vast transpositions of matter at the 
 earth's surface, or else distortions of the whole solid mass, 
 which must have taken place to alter the axis sufficiently to 
 produce sensible changes of climate in any region must be 
 considered and shown to be possible or probable before any 
 hypothesis accounting for change of climate by alterations of 
 the axis can be admitted." 
 
 Professor Airy analyzed the question very completely. 
 " To begin with the case most favourable for the production 
 of a large effect," he says, "I will suppose the earth to be a 
 perfectly rigid spheroid, which has in distant ages and under 
 
344 The Glacial Nightmare. 
 
 proper circumstances assumed the form of equilibrium, and 
 whose axis of rotation coincides with its axis of figure and 
 is a principal axis. And I will suppose a mountain mass to 
 be elevated by something like a gaseous explosion (the 
 hollow being filled up by the influx of neighbouring matter) 
 in latitudes not very different from 45. It is evident that the 
 principal effect of the elevation would be that the axis of 
 rotation would no longer be a principal axis, that the 
 principal axis of largest moment would now be a little 
 way beyond the present axis of rotation, and that the two 
 other principal axes whose moments formerly were equal 
 would now have moments very slightly different. The effects 
 which follow are scarcely affected by this last modification. 7 
 It is well known that, under these circumstances, the axis 
 of rotation would wander in the solid earth. But it would 
 not wander indefinitely ; its pole would describe on the 
 earth's surface an ellipse not sensibly different from a circle 
 whose centre would be the pole of the new principal axis, and 
 after a certain time it would return to its former position. 
 The greatest change, therefore, in the terrestrial position of 
 the earth's pole would be double the distance of the new 
 principal axis from the former principal axis/' 8 
 
 " A very slight investigation/' Professor Airy continues, 
 " suffices to show that the angular change of position of the 
 principal axis will depend upon the proportion which the 
 increased moment of inertia resulting from all the changes 
 made by the elevation bears to the excess of the moment of 
 inertia round the polar axis above the moment of inertia 
 round an equatorial axis. This latter excess is about half the 
 moment of inertia of the equatorial protuberance; that is half 
 the moment of inertia of a mass of matter 25,000 miles long, 
 6000 miles broad, and 13 miles deep. And what mountain 
 mass can compare sensibly with this ? Even if a mountain 
 mass contained one-thousandth part of this matter (which I 
 apprehend is very far above the fact) the shift of the earth's 
 pole would be only two or three miles ; and this, though it 
 would greatly surprise astronomers and might sensibly affect 
 the depth of water in harbours, would produce no such 
 changes of climate as those which it is desired to explain/' 
 
 7 Athenaeum, Sept. 22, 1860, p. 384. Id. 
 
The Shifting of the Earths Axis. 345 
 
 Professor George Darwin has entered into more detailed 
 calculations. He takes for instance the case of an ocean bed 
 15,000 feet deep elevated into a continent 1100 feet high, 
 which is a little above the average, and then calculates the 
 amount of deformation that would result. Taking a continent 
 the size of Africa elevated in this way and a corresponding 
 area depressed in the most effective shapes and positions. 
 This would alter the position of the pole less than two degrees. 
 " Moreover," says Mr. Hill, " this is obtained on the supposition 
 that the elevation and depression are obtained by the actual 
 removal of matter from the surface of the one area, and its 
 deposition on the other. If we adopt the easier supposition 
 that elevation is produced by the expansion of subjacent 
 strata; depression by their contraction, the resulting change 
 is diminished to a considerable, possibly to a very great 
 extent. If, for instance, the expansion or contraction take 
 place between the depths of ten and fifty miles, the shift of 
 the pole will be diminished to less than a sixtieth part of its 
 former amount. If the supposed areas occupy areas other 
 than the most effective, then changes must be still further 
 reduced, and may even vanish altogether." 9 
 
 Dr. Croll, writing on the change here postulated, says : 
 " There probably never was an upheaval of such magnitude 
 in the history of our earth. And to produce a deflection of 
 3 1 7' (a deflection which would hardly sensibly affect climate) 
 no less than one-tenth of the entire surface of the earth 
 would require to be elevated to the height of 10,000 feet. 
 A continent ten times the size of Europe elevated two miles 
 would do little more than bring London to the latitude of 
 Edinburgh, or Edinburgh to the latitude of London. He 
 must be a sanguine geologist indeed who can expect to 
 account for the glaciation of this country, or for the former- 
 absence of ice around the poles, by this means. We know 
 perfectly well that since the Glacial epoch there have been 
 no changes in the physical geography of the earth sufficient 
 to deflect the pole half a dozen miles, far less half a dozen 
 degrees. It does not help the matter much to assume a 
 distortion of the whole solid mass of the globe. This, it is 
 true, would give a few degrees additional deflection of the 
 G-eol. Mag. 1878, 264, 265. 
 
346 The Glacial Nightmare. 
 
 pole ; but that such a distortion actually took place is more 
 opposed to geology and physics than even the elevation of 
 a continent ten times the size of Europe to a height of two 
 miles." 10 
 
 The fact is, when some geologists approach the realm of 
 physics, they seem as if they were entitled to make unlimited 
 drafts upon time and force. When Dr. J. Evans suggested 
 that a certain elevation of the earth's crust might shift the pole 
 15 or 20, Professor Twisden subjected his premises and 
 conclusions to mathematical induction and showed that such 
 a movement instead of moving the pole 15 or 20 would only 
 move it about 10', and that to obtain a movement of 1 
 the zone of land in question must be raised five miles, while 
 to move the pole 20 it would require the removal of a sixth 
 part of the equatorial bulge ; and that even if a transfer of 
 this enormous amount of matter took place it would not of 
 necessity produce any effect, and might only produce a small 
 effect on the axis of figure. 1 
 
 While these arguments have been generally held to be 
 conclusive against a change in the position of the earth's axis 
 in a solid globe, it has been in recent years suggested, and 
 has been especially urged by Dr. John Evans, that the case 
 might be different if the earth were a shell filled with molten 
 matter. To this argument, it seems to me, there are two 
 objections. First the postulate itself seems to be inadmissible. 
 
 Sir Wm. Thompson says : " That the earth cannot, as many 
 geologists suppose, be a liquid mass enclosed in only a thin 
 shell of solidified .matter, is demonstrated by the phenomena 
 of precession and nutation. Mr. Hopkins . . . applied 
 mathematical analysis to investigate the rotation of rigid 
 ellipsoidal shells enclosing liquids, and arrived at the con- 
 clusion that the solid crust of the earth must be not less than 
 800 or 1000 miles thick. ... It has always seemed to me 
 that Mr. Hopkins might have pressed his argument further, 
 and have concluded that no continuous liquid vesicle at all 
 approaching to the dimensions of a spheroid 6000 miles in 
 diameter can possibly exist in the earth's interior without 
 rendering the phenomena of precession and nutation very 
 
 10 Discussions on Climate and Cosmology, p. 5. 
 1 Journ. Geol. Soc. xxxiv, 41. ; 
 
The Shifting of the Earths Axis. 347 
 
 sensibly different from what they are . . . unless the average 
 substance of the earth is more rigid than steel, its figure must 
 yield to the distorting forces of the moon and sun not in- 
 comparably less than it would if it were fluid ... if the 
 rigidity of the earth on the whole were only as much as that 
 of steel and iron, the earth as a whole would yield as much 
 to the tide-producing influences of the sun and moon as it 
 would if it had no rigidity at all ; and it would yield by more 
 than three-fourths of the fluid yielding if its rigidity were no 
 more than that of glass. Such a deformation as this would 
 be quite undiscoverable by any direct or astronomical ob- 
 servations ; but if it existed, it would largely influence the 
 actual phenomena of the tides and of precession and nuta- 
 tion." In regard to the latter effect, Sir William says that 
 the only ground on which it could be otherwise would be on 
 fche assumption that there is an enormous liquid vesicle, or a 
 solid nucleus separated by a liquid layer from the outer crust 
 in the interior, and that the loss of precessional effective 
 moment of inertia, owing to this portion not being carried 
 round in the precessional movement is almost exactly com- 
 pensated by a diminution of the generating couple in very 
 nearly the same proportion by elastic yielding ; " and he urges 
 as very improbable that the defect of moment of inertia due 
 to fluidity in the earth's interior bears approximately the 
 same ratio to the whole moment of inertia as the actual elastic 
 yielding bears to the yielding which would take place if the 
 earth were perfectly fluid. 
 
 Again he says : " At the surface and for many miles below 
 the surface, the rigidity is certainly very much less than that 
 of iron ; and therefore at great depths the rigidity must be 
 enormously greater than at the surface. That both the 
 rigidity and the resistance to compression should be much 
 greater several hundred miles down than at the surface, seems 
 a natural consequence of the enormous pressure experienced 
 at those great depths by the matter of the earth." : 
 
 Elsewhere the same distinguished physicist says : " What- 
 ever be its age, we may be quite sure the earth is solid in its 
 interior; not, I admit, throughout its whole volume, for 
 there certainly are spaces in volcanic regions occupied by 
 2 Phil. Trans. 153, pp. 573-581. 
 
348 The Glacial Nightmare, 
 
 liquid lava; but whatever portion of the whole mass is liquid, 
 whether the waters of the ocean or melted matter in the 
 interior, these portions are small in comparison with the 
 whole j and we must utterly reject any geological hypothesis 
 which, whether for explaining underground heat, or ancient 
 upheavals and subsidences of the solid crust, or earthquakes, 
 or existing volcanoes, assumes the solid earth to be a shell of 
 30 or 100, or 500 or 1000 kilometres thickness, resting on an 
 interior liquid mass. This conclusion was first arrived at by 
 Hopkins, who may therefore be called the discoverer of the 
 earth's solidity. He was led to it by a consideration of the 
 phenomena of precession and nutation, and gave it as shown 
 to be highly probable, if not absolutely demonstrated, by his 
 confessedly imperfect and tentative investigation. But a 
 rigorous application of the perfect hydrodynamical equation 
 leads still more decidedly to the same conclusion/'' 
 
 Sir Wm. Thompson has shown that if the earth were a 
 liquid mass covered with a solid crust, " the solid crust would 
 yield so freely to the deforming influence of sun and moon, 
 that it would simply carry the waters of the ocean up and 
 down with it, and there would be no sensible tidal rise and 
 fall of water relatively to it. The state of the case," he 
 continues, " is shortly this : The hypothesis of a perfectly 
 rigid crust containing liquid, violates physics by assuming 
 preternaturally rigid matter, and violates dynamical astronomy 
 in the solar semiannual and lunar fortnightly nutations ; but 
 tidal theory has nothing to say against it. On the other hand, 
 the tides decide against any crust flexible enough to perform 
 the nutations correctly with a liquid interior, or as flexible 
 as the crust must be unless of preternaturally rigid matter." 3 
 
 The same position has been maintained with great learning 
 by Mr. George Darwin, who most emphatically supports the 
 views of Thompson, and proves mathematically that far from 
 the earth having a fluid nucleus, its effective rigidity is at 
 least as great as that of steel. 4 
 
 Driven from this position, the advocates of an interior 
 fluidity to the earth have argued that they do not mean that 
 the whole of the interior of the earth is fluid, but that the 
 earth in fact consists of a solid shell enclosing a solid nucleus, 
 
 8 Report Brit. Assoc. 1876, pp. 6 7. 4 Nature, xxvii. 23. 
 
The Shifting of the Earths Axis. 349 
 
 between which and itself there is a fluid zone. In regard to 
 this notion I cannot realize its possibility. Under what 
 physical conditions it is possible for the centre of the 
 earth to be a huge solid sphere floating in a liquid zone 
 which separates it from the outer crust I do not know. Nor 
 would that meet Mr. G. Darwin's objection in regard to rigidity. 
 But granting that in some transcendental manner it is 
 possible to realize that the earth instead of being a rigid solid 
 is really a shell enclosing a fluid mass, it does not seem to 
 me that the problem is any more advanced. 
 
 Professor Airy says : " Let us suppose that the earth 
 is not absolutely rigid, but that there is susceptibility 
 to change of form, either from that degree of yielding 
 or fracture to which most solid substances are liable, or 
 from the hydrostatic pressure of internal fluid. This, as I 
 conceive, puts an end to all supposition of change of axis. 
 The first day's whirl would again make the axis of 
 rotation to be a principal axis and the position of the axis 
 is then permanent." 5 
 
 I cannot myself see how the mechanical difiiculty is to be 
 overcome. A shell filled with molten matter would behave in 
 regard to the shifting of its axis, so far as I understand 
 mechanical principles, as a solid sphere would, every portion of 
 it would be revolving with the same speed and momentum; 
 and it seems to me that it would be just as difficult to cause 
 a shell to shift round on a nucleus when both are revolving 
 rapidly at the same speed on the same axis and in the same 
 direction, as it would to change the direction of an axis through 
 a solid sphere. Nor can I realize a liquid molten nucleus 
 sharply defined from a solid shell. There would be an inter- 
 mediate viscous zone, a gradation from one to the other, and 
 this would make the slipping of the shell over the nucleus as 
 difficult as if the earth were a continuously rigid body. 
 
 Again it seems to have been forgotten that the earth is not 
 a sphere. 
 
 Whether the earth be solid throughout or be a hollow 
 shell, the protuberant belt of thirteen miles thick about its 
 equator which has been there, so far as we know, since its 
 
 5 Athenaeum, Sept. 22, I860, p. 384. 
 
35 The Glacial Nightmare. 
 
 history began, stands completely in the way of our placing 
 the earth's axis of rotation anywhere else than where it is 
 without a portentous dislocation of which we have no 
 evidence. It is no use writing about the way in which a 
 shell can slip over its nucleus unless we can at the same time 
 explain how the belt of solid matter about the old equator 
 can be shifted so as to accommodate itself to a new equator. 
 The argument seems to me to completely ignore the real 
 conditions of the problem. 
 
 Mr. E. Hill well says : " The mathematician asks the 
 geologists whether they have realized the earth's size. Its 
 deviation from a sphere is trifling in amount. In what 
 amount ? In figure doubtless. Were its section drawn on 
 this page, a microscope would be required to distinguish the 
 curve from a circle. But in quantity the deviation is not 
 trifling. The height of the highest mountain and the depth 
 of the deepest sea together do not equal its extent. All 
 the mountain chains added together would not perceptibly 
 increase the volume of the equatorial protuberance. The 
 mass of all the continents reinforced by that of all the seas 
 would not be the fifth part of it. To change the earth's 
 shape, this vast protuberance must be shifted or masked. 
 Where is the power that can shift it, the elevation that can 
 mask it ? What are our puny upheavals and subsidences of 
 an ocean here and a continent there, compared with a girdle 
 of matter thirteen miles in thickness ? Remove 10,000 feet 
 of rock from the surface of one half the earth and spread 
 it over the other half. You could not thereby bring the pole 
 halfway to the present arctic circle. Sufficient changes in 
 the earth's surface will undoubtedly shift the pole. But will 
 geologists grant the changes that would be sufficient ? " 6 
 
 Apart from all these arguments drawn from astronomical and 
 physical considerations and which make the postulate of a 
 shifting of the earth's axis impossible, we have the best 
 geological reasons against it. Thus Professor Jukes writes : 
 " We have no reason to look to any other spot on the globe 
 than the present north pole as the centre of the cold climate 
 during this Glacial or Pleistocene period. Neither has any 
 
 6 Geol. Mag. 1878, 266, 
 
The Shifting of the Earths Axis. 351 
 
 one yet ventured to point to any other region of tie globe as 
 having been possibly its arctic region during any previous 
 geological period, basing his argument on the fossils of that 
 region having a more arctic character than the contem- 
 poraneous fossils of surrounding countries." 7 
 
 Again, Dr. Haughton, in a remarkable paper read before 
 the Royal Society, brought together some very interesting 
 evidence, based entirely on geological grounds, to show that 
 the north pole cannot have shifted its place since the 
 Secondary period. He says, inter alia, " McClintock found 
 tropical shells of the Jurassic period in the Parry Islands, in 
 lat. 76 20'. Among them were remains of a reptile 
 allied to the Indian gavial. The triassic beds of Spitzbergen, 
 lat. 79, have afforded species of Nautilus, Ammonites, 
 Ceratites and Halobia, closely allied to, if not identical with 
 those of the St. Cassian beds of South Austria. In the 
 neighbourhood of Cook's Inlet in Alaska, lat. 60 N., 
 shells characteristic of the triassic and Jurassic periods have 
 been found ; and similar fossils are found along the Pacific 
 Coast of North America. 
 
 " It is not possible to explain the occurrence of tropical 
 animals in the above-mentioned localities by any change in 
 the position of the earth's axis, even if so great an amount 
 of change as would be required were possible. This state- 
 ment can be proved as follows. Let a great circle be drawn 
 joining Spitzbergen with Cook's Inlet, Alaska. This 
 circle will pass nearly through the north pole. In order to 
 explain the tropical climate of these two localities, and also 
 of the Parry Islands, the pole must be displaced at right 
 angles to the great circle joining Spitzbergen and Alaska 
 along the meridian long. 117 B., nearly that of Pekin. 
 The present difference of latitude between New Orleans 
 and Spitsbergen is 45, so that in order to make the arctic 
 regions tropical we must move the north pole 45 on the 
 meridian of Pekin, bringing it within 300 miles to the north 
 of that city. Hence it follows that during the triassic period 
 Pekin lay under the north pole, covered by the polar ice-cap. 
 Let us now consider what the south pole was doing. It had 
 
 7 Athensenm, Sept. 22, 1860, 333. 
 
352 The Glacial Nightmare. 
 
 moved on the opposite meridian, and had readied the mouth 
 of the Rio ^egro, on the east coast of Patagonia, about 1000 
 miles to the south-south-east of Valparaiso and the Chilian 
 Andes. Jurassic strata have been found in the Chilian 
 Andes at 34 S., containing the tropical, ammonites biplex, 
 which is found also in Alaska, 60 N., and in Europe. This 
 locality lies within 700 miles of the necessary position of the 
 south pole and cannot have enjoyed a tropical climate. The 
 proposed alteration of the north pole is consistent with the 
 occurrence of tropical animals in the Parry Islands, in Spitz- 
 bergen, and in Alaska ; while the proposed alteration of the 
 south pole would permit tropical animals to exist in New 
 Zealand and New Caledonia ; but the occurrence of Jurassic 
 ammonites within 700 miles of the south pole is fatal to the 
 proposed shifting of the axis of rotation, even if that were 
 allowable, to the extent required/' 
 
 Turning to the Tertiary age, he shows that in Greenland 
 Disco, lat. 70, Grinnell Land, lat. 81 44', Spitzbergen, 
 west coast, lat. 77, and Alaska and the Mackenzie river, 
 lat. 70 to 6, have furnished remains of miocene plants 
 evidencing a climate such as is now enjoyed by the northern 
 parts of Italy, and indicate a mean annual temperature of 
 48, where the mean annual temperature is now as low as zero. 
 <( It can be shown," says Dr. Haughton, " by a similar method 
 to that employed for the triassic and Jurassic periods, that 
 the north pole was practically in the same place during the 
 miocene period that it now occupies. If we join the 
 Mackenzie river and Spitzbergen by the arc of a great 
 circle, the north pole must be moved at right angles to this 
 arc, away from Greenland through 30 in orde^ to give all 
 these northern localities a Lombardic climate. Tre direction 
 in which the pole must be moved is on the meridian of 
 Nagasaki, and it reaches a point close to Yakutsk, within 800 
 miles of the peninsula of Kamtschatka and the Island of 
 Saghalien off the Amur. Here we meet with a difficulty 
 similar to that offered by the south pole in the triassic period. 
 The Island of Saghalien and the peninsula of Kamtschatka 
 contain miocene coal beds, requiring at least a sub-tropical 
 climate, which would be impossible under the supposed cir- 
 cumstances. Also the islands of Yesso, Nagasaki and Kiusiu, 
 
Evidence against a Shifting Axis. 353 
 
 somewhat farther off, contain similar coal beds. It is very 
 remarkable that, while there exist so many proofs of a warm 
 climate near the south pole in former geological periods, 
 there is no evidence of cooler climates having ever existed in 
 the tropics." 8 
 
 Elsewhere Professor Haughton says : " My contention in 
 brief is that the tertiary plant remains indicating a climate 
 similar to that of Lombardy are so situated round the north 
 pole that no possible change in the position of that pole 
 (even were such permitted by mechanical considerations) 
 would give them the climatic conditions as to temperature 
 which they require. ... I claim to have surrounded the 
 north pole with such a network of Lombardic plants, re- 
 quiring Lombardic heat, but not Lombardic light, as to 
 render the escape of the pole from its present position as 
 difficult as that of l a rat in a trap surrounded by terriers.' *' 
 
 To quote evidence from the other hemisphere. Captain 
 Hutton, than whom it would not be easy to quote a better 
 authority, says : " There is no evidence whatever of a glacial 
 period having occurred in New Zealand, although if it had 
 occurred, there is every reason to expect that it would have 
 left sufficiently clear traces behind it. I will also add that 
 as New Zealand is nearly antipodal to Great Britain, any 
 change of climate in one place, caused by a change in the 
 position of the earth's axes of rotation, would also necessitate 
 a similar change in the other place. " * 
 
 Lastly, in regard to the evidence of so-called glacial beds, I 
 shall endeavour to show presently, thaix while Western Europe 
 and Eastern America were largely loaded with ice, Northern 
 Asia and North- Western America had a sufficiently mild 
 temperature to enable forests to grow and forest animals to 
 thrive there, right up to the arctic circle. This means that 
 the shifting of the north pole, if such shifting is to be avail- 
 able as a cause of the drift phenomena of Europe and East- 
 ern America, must have been towards the region of the North 
 Atlantic rather than towards that of the North Pacific. 
 
 The difficulties of the position are further intensified in the 
 
 Nature, xviii. 267, 268. 9 Geol. Mag. 1879, 92, 93. 
 
 1 Geol. Mag. 1875, 583. 
 
354 The Glacial Nightmare. 
 
 presence of the demands of the ultra-glacialists. Thus 
 Mr. J. Geikie, in arguing against the theory of a change of 
 axis, says: "The glacial epoch was interrupted by more 
 than one warm interglacial period. It is not therefore only 
 one change in the position of the poles, but several that we 
 should require, and all these stupendous changes so improb- 
 able in themselves, must be supposed to have taken place at 
 quite a recent date, and within a comparatively short space 
 of time." 2 It is not every glacialist who has been so modest. 
 M. Boucheporn 8 would explain the European phenomena by 
 shifting the pole to some point in the Baltic north of Prussia 
 and Poland. The American drift he would explain by another 
 shift, and he does not scruple to invoke other changes of the 
 same kind to meet the various exigencies of the problem. 
 " Mais," says M. Falsan, " il se heurta a tant de difficulty's 
 qu''il ne put faire accepter son systeme." * 
 
 I will now turn to those recurrent astronomical causes to 
 which appeals have so continually been made in modern times 
 to explain the glacial period; and first let us consider a 
 cause which, if subject to variation, may affect climate in two 
 ways, first, in increasing or diminishing the total heat 
 received during a year by the earth ; and, secondly, in 
 altering its distribution between the seasons, namely, a varia- 
 tion in the eccentricity of the earth's orbit. The ancients 
 believed that the earth moves round the sun in a circular 
 orbit. If it moved in a circular orbit, and its axis were 
 always at right angles to the plane of its orbit, and if it 
 always moved at the same rate, there would be no seasons, 
 day and night would be equal everywhere, and every place 
 on the earth would have the same climate which it has at the 
 two equinoxes all the year round. It was Kepler who first 
 proved that the orbit is elliptical and that the sun occupies 
 one of the foci, so that at one portion of the year the earth 
 is nearer to the sun than at another. When at its greatest 
 distance, or, to use astronomical language, when it is at its 
 aphelion it is about 93,750,000 miles distant from the sun, 
 when nearest, that is at its perihelion, it is about 90,259,000. 
 
 2 Great Ice Age, 85, note. 3 Eecherches geologiques, i. 18, 
 * Falsan, op cit. 195. 
 
Effect of Eccentricity on Climate. 355 
 
 The mean distance, as estimated by astronomers since the 
 corrections of Encke, Hansen, Foucault and Hind, is 
 91,839,000 miles, 5 and the amount of eccentricity of the 
 orbit is now about a seventeen-thousandth of its major axis. 
 
 It has long been known, however, that the earth's orbit is not 
 constant, but that it is altered in position and extent by the 
 attraction of the planets, which cause it at some times to be more 
 eccentric than at others. The variation in the eccentricity of the 
 earth's orbit, dependent as it is on a very complicated series 
 of planetary and other attractions, involves a difficult problem 
 and proves to be very irregular. On the first day of this century 
 the eccentricity was '016792, a century later this will have 
 diminished by '0000416. It has been thus diminishing for 
 13,000 years, and will go on diminishing for 24,000 years 
 more, when it will reach its minimum of '0033. It will then 
 begin to increase, and go on doing so for 70,000 years, and 
 then attain to '0211. In another 100,000 years it will reach 
 046. 
 
 Having established that the earth's orbit varies in eccen- 
 tricity, let us now see how this affects climate. First, in 
 regard to the quantity of heat received by the earth as a 
 whole, it has been proved, from the known laws of heat and 
 the properties of the ellipse, that the total amount of heat 
 received by the earth in a year varies in inverse proportion 
 to the length of the minor axis of its orbit ; and inasmuch 
 as the minor axis is longest when the orbit is nearest a circle 
 and shortest when it is most eccentric, it follows that the 
 more eccentric the earth's orbit is the more heat does it 
 receive from the sun in a year. 
 
 Herschell calculated the reciprocal proportions of the two 
 elements, and concluded that if the eccentricity were as great 
 as a fourth of the major axis, it would not raise the mean 
 temperature of the earth more than three per cent. The 
 amount of this variation takes in the whole range of planetary 
 eccentricities from that of Pallas and Juno downwards. 6 
 
 Professor Haughton thus states the case : f ' If H denotes 
 the general amount of sun heat received by, the earth when 
 her orbit is circular, the annual amount of sun-heat received 
 when her orbit is most eccentric will be H (1 0'00333) ; that 
 
 6 Eeclus, The Earth, 6. 6 Trans. Geol. Soe. iii. 
 
 A a 2 
 
356 The Glacial Nightmare. 
 
 is to say, the total variation of annual sun -heat received by 
 the earth due to variation of eccentricity of orbit is only -g-g-g-rd 
 part of the total heat received, or in other words, if the sun 
 at present be sufficient in one year to melt 100 feet of ice 
 at the Equator, it would melt 3J inches less when the ec- 
 centricity of the orbit was at its maximum. This variation 
 of sun-heat received by the earth is quite inadequate to affect 
 geological climates sensibly."" 7 
 
 Mr. Meech says that 100,000 years ago, according to 
 Leverrier's tables, the eccentricity was "0473 as compared 
 with "0168 now. The fraction of 365*24 days, counting the 
 days at twelve hours each in respect to solar illumination, 
 amounts to between four and five hours of sunshine in a year, 
 and by so small a quantity only has the sun's annual intensity, 
 during 100,000 years past, ever exceeded the yearly value at 
 the present time. Nor can it depart from its present annual 
 value by more than the equivalent of five hours of average 
 sunshine a year, for 100,000 years to come. 
 
 Leverrier's superior limit of variation '0777 may have been 
 reached at some very remote but unknown period. At that 
 epoch the annual intensity is computed, as before, to have 
 exceeded the intensity of the present by thirteen hours of 
 sunshine in a year. On the other hand, the inferior limit of 
 eccentricity being near to zero, indicates only four minutes of 
 average sunshine in a year less than the present annual 
 amount. Between these two extreme limits, all annual 
 variations of the solar intensity, whether past or future, must 
 be included. 8 In this view Croll, who has examined the problem 
 with some care, also concurs. We may take it, therefore, that 
 increasing the eccentricity of the earth's orbit will only very 
 slightly affect the earth's mean climate, and that whatever 
 effect is produced is in the direction of increasing and not 
 decreasing its temperature ; and, as a matter of fact, inas- 
 much as the earth's orbit has been becoming more circular, 
 its mean temperature has also for a long time been slowly 
 declining. 
 Let us now see if any more can be made of the second 
 
 7 Trans. Roy. Ir. Acad. Oct. 1881, pp. 50, 51. 
 Smith. Contr. ix. 36. 
 
Effect of Eccentricity on Climate. 357 
 
 possible result of altering the eccentricity, namely in so far as 
 it affects the seasons. 
 
 Kepler not only showed that the earth's orbit is elliptical, 
 but that the earth moves faster as it gets nearer to the sun 
 and slower as it recedes from it. And this according 
 to a beautiful and symmetrical law which may be thus 
 stated: If a line be supposed to connect the earth in 
 motion with the sun at rest, which is called in astronomy a 
 radius vector, this line will describe or sweep over that 
 portion of the whole area or surface of the ellipse which is 
 included between its consecutive positions ; and the motion 
 of the earth will be such that equal areas are thus swept over 
 by the revolving radius vector in equal times, in whatever 
 part of the circumference of the ellipse the sun may be 
 moving. 9 
 
 This translated into ordinary language means that the 
 earth does not move round the sun at the same rate, but 
 moves faster in that part of its orbit where it is nearer than 
 it does in that which is farther from the sun. In fact, says 
 Reclus, from the 20th of March to the 22nd of September 
 (i.e. between the two equinoxes), during the spring and 
 summer of the northern hemisphere, the earth takes 186 
 days to travel over the first and largest half of its orbit ; while 
 during the winter period, from the 22nd of September to the 
 20th of March, only 179 days are required to accomplish the 
 second half of its journey. 1 
 
 It follows from this that if we excluded the question of the 
 obliquity of the earth's axis, and limited ourselves strictly to 
 the effects of eccentricity, tnere would be no distinction be- 
 tween the two hemispheres, but summer and winter would 
 take place all over the world at the same time. And if the 
 eccentricity were to remain as it is now, the world's summer 
 would always be shorter than its winter by about seven days. 
 When the earth's orbit was most nearly circular the differ- 
 ence would almost disappear ; when, on the other hand, its 
 eccentricity was greatest, there would be a difference of 
 about thirty-six days between the lengths of the seasons. 
 It was Lambert who, in his " Pyrometrie," published in 1779, 
 
 9 Herschell, Outlines, 209. * Eeclus, 11. 
 
3 5 8 The Glacial Nig htma re . 
 
 first combined Kepler's law last quoted and the known 
 qualities of heat into a law of much value in the discussion of 
 climate, namely that the amount of heat received by the earth 
 from the sun is proportional to the area described by its 
 radius vector, or, as Sir John Herschell states the case, the 
 momentary supply of heat received by the earth varies in 
 exact proportion to the angular velocity; and from this it 
 follows that equal amounts of heat are received from the sun 
 in passing over equal angles of it, wherever in the ellipse 
 those angles may be. Thus, says Lambert, it may come to 
 pass in the case of comets, that in traversing the 180 degrees 
 of longitude while near the sun, they may receive as much light 
 and heat in a few days or weeks as they shall in traversing 
 the remaining part of their orbits during many centuries of 
 years. Of all the planets, he adds, Mercury has the greatest 
 eccentricity, and this planet receives from the sun, during the 
 32| days while it is nearest him, as much light and heat 
 as it receives during the 53-^ days while it is farthest 
 from him. 2 
 
 Mr. Meech after discussing the mathematical issues 
 involved in this position concludes thus : f< During each of the 
 four astronomic seasons of spring, summer, autumn and 
 winter, the intensities received from the sun are precisely 
 equal. For in each season the earth passes over three signs 
 of the zodiac, or a quadrant of longitude. In the year 1855, 
 for example, spring in the northern hemisphere, commencing 
 at the vernal equinox, March 20th, lasted eighty-nine days; 
 summer, beginning at the summer solstice, June 21st, con- 
 tinued ninety-three days ; autumn, commencing at the 
 equinox, September 23rd, continued ninety-three days ; and 
 winter, beginning at the winter solstice, December 22nd, lasted 
 ninety days; yet, notwithstanding "their unequal lengths, the 
 amounts of heat and light which the whole earth received were 
 equal in the several periods. 3 Not only so, but the amount of 
 heat received by each hemisphere during the year is precisely 
 the same. Croll has described this result lucidly : <f What- 
 ever may be the eccentricity of the earth's orbit, the two 
 hemispheres must receive equal quantities of heat per annum ; 
 
 2 Haugh ton, Trans. Hoy. Ir. Acad. xxviii. 49. 
 
 3 Smithsonian Contributions, ix. 12. 
 
Heat received by the Earth. 359 
 
 for proximity to the SUD is exactly compensated by the effect 
 of swifter motion. For example, whatever extra heat the 
 southern hemisphere may at present receive from the sun 
 during its summer months, owing to its greater proximity, 
 is exactly compensated by a corresponding loss arising from 
 the shortness of the season ; and, on the other hand, whatever 
 deficiency of heat we in the northern hemisphere may at 
 present have during our summer half-year, in consequence of 
 the earth's distance from the sun, is also exactly compensated 
 by a corresponding gain, owing to the length of the season. 
 
 Arago puts it in another way. The sun is not always 
 equally distant from the earth ; at present its least distance is 
 observed in the early days of January, and the greatest in the 
 early days of July ; but a time will come when the minimum will 
 occur in July and the maximum in January. Would the two 
 summers under these circumstances differ sensibly ? At first 
 sight everyone would probably answer in the affirmative, since 
 there is a difference in the distance of the earth amounting to a 
 thirtieth of the whole. " Let, however, the consideration of 
 the velocities be introduced into the problem, and the result 
 will be on the side opposite to that we originally imagined ; 
 the part of the orbit where the sun is found nearest the earth 
 is at the same time the point where the luminary moves most 
 rapidly along. . . . The hypothesis we have adopted would 
 give on account of the lesser distance, a springtide and 
 summer hotter than they are in our days ; but on account of 
 the greater rapidity, the sum of the two seasons would be 
 shorter by about seven days. Thus then, all things 
 considered, the compensation is mathematically exact." 4 We 
 have got to this point, then, that the earth receives precisely 
 the same amount of heat during each of the four astronomical 
 seasons, and that each hemisphere also receives a precisely 
 equal amount of heat during the year, whatever the variation 
 in eccentricity of the orbit may be. The only substantial 
 effect upon climate, as measured by the seasons, caused by 
 such a variation, is to make the seasons longer or shorter, 
 that is to say, to distribute the heat over a greater or less 
 number or days, and thus to make the seasons on the one 
 
 4 Op. cit. 221-224 
 
360 The Glacial Nightmare. 
 
 hand more contrasted, or on the other more equable; but 
 inasmuch as precisely the same amount of heat in each 
 season is received, whatever be the eccentricity, it will be 
 seen that the total direct effect, as measured by the number 
 of heat-rays received by the earth, and by each hemisphere 
 annually, is nil. 
 
 Under any circumstances this change could not produce 
 such a climatic effect as is required by the Glacial theory. 
 The tempering of the relative severities of summer heat 
 and winter cold might be a great element in affecting 
 the distribution of vegetable and animal life, but not in 
 affecting the amount of cold or heat on the earth as indexed 
 by such a measure as ice. It takes the same number of thermal 
 degrees to melt the same quantity of ice ; and if we apply the 
 same amount it does not matter how we apply it, whether 
 quickly or slowly, the result must be the same. The alterna- 
 tion of very hot summers and very cold winters will have pre- 
 cisely the same effect as that of moderate summers and 
 moderate winters, if the amount of actual heat received and 
 the amount of ice to be melted is the same in both cases, and 
 if it only differs in the length of time over which it is distri- 
 buted. 
 
 In regard to the daily average of heat received, it must, on 
 the other hand, be remembered that while the increase of the 
 eccentricity of the earth's orbit would increase the disparity 
 between summer and winter, it, on the other hand, would, 
 as we have seen, tend to increase the total mean heat of the 
 earth. As the eccentricity enlarged, therefore, there would 
 be a larger quantity of heat to distribute, which would tend 
 to raise the temperature of each day in winter, and thus 
 partly compensate for the sun's heat being distributed over 
 a longer time. 
 
 Treating the issue apart from meteorology, and simply as a 
 question of the effective heating-power of the sun as applied 
 to the earth, it seems plain that, except the slight increase in 
 the mean annual temperature already referred to, the altera- 
 tion in the eccentricity of the earth's orbit can have little or 
 no effect upon climate. Arago says : " The changes of form 
 and of the position of the terrestrial orbit are mathematically 
 
Inefficiency of varying Eccentricity > &c. 361 
 
 inoperative, or at most their influence is so mimite that it is 
 not indicated by the most delicate instruments. " 5 
 
 Croll speaks equally emphatically. " All the conclusions 
 arrived at by astronomers/' he says, (i.e. the conclusions 
 adverse to an appeal to increased eccentricity as a cause of 
 glacial climate) " are perfectly legitimate so far as the direct 
 effects of eccentricity are concerned; and it was quite 
 natural, and in fact proper, to conclude that there was nothing 
 in the mere increase of eccentricity that could produce a 
 glacial epoch. How unnatural would it have been to have 
 concluded that an increase in the quantity of heat received 
 from the sun should lower the temperature and cover the 
 country with snow and ice ! " 
 
 Mr. J. Geikie speaks exactly in the same way. He 
 approves of the argument of the astronomers that the earth 
 receives the same total amount of heat in the aphelion as in 
 the perihelion part of its orbit, and says that this is true how- 
 ever much we may vary the eccentricity, and that the only 
 effect of increased eccentricity is to make the earth's climate 
 as a whole warmer. 
 
 We will now examine another astronomical cause to which 
 a good deal has been assigned in the shape of climatic 
 change, namely, the change in the obliquity of the earth's 
 axis. 
 
 The degree of obliquity of the earth's axis determines, of 
 course, the area over which the heat received from the sun is 
 distributed. 
 
 If the earth revolved at right angles to the plane of the 
 ecliptic, the two hemispheres would have the same seasons at 
 the same time; each would have its summer when the earth 
 was nearest the sun, and its winter when it was farthest 
 away ; and all places on the same parallel of latitude, either 
 north or south of the equator, would receive an equal quantity 
 of light and heat in each twenty- four hours. The portion of 
 the globe lighted by the sun would invariably extend from 
 one pole to the other, and in both hemispheres the days and 
 nights would always be twelve hours long. 
 
 5 Arago, op. cit. 221-224. 
 
362 The Glacial Nightmare. 
 
 Lastly, and most important for our purpose, inasmuch as 
 the sun would in this case be always over the equator, the 
 equatorial region would receive the greatest quantity of heat 
 possible while the two poles would be in a condition of most 
 wintry cold. 
 
 If the earth's axis, instead of being at right angles to the 
 plane of the ecliptic, were to be moved so as to incline some- 
 what to it, it is obvious that the sun would not be always 
 over the equator, but would travel slightly into each hemi- 
 sphere, and pro tanto mitigate the climate of that hemisphere, 
 and cause the other to be more rigorous, which effect would 
 be reversed every six months ; and the more the axis was 
 inclined, the more would this effect ensue, causing a diminu- 
 tion of equatorial heat and an enhancement of the heat of 
 each pole during the summer, and the reversal of this 
 process during the winter. 
 
 As a matter of fact the earth's axis and the plane of the 
 ecliptic are now inclined to each other at an angle of about 
 23 degrees. To the observer of the heavens the result is that 
 the sun appears to cross the equator twice a year, in March 
 and September, at the equinoxes, when alone there is equal 
 day and night everywhere. After crossing the equator in 
 March it seems to move northwards until it reaches its farthest 
 limit in June, where it remains stationary a short time, whence 
 the point is known as the solstice (solis statio) f or rather the 
 summer solstice. It then moves southwards, crosses the 
 equator, and in December reaches its farthest limit south- 
 ward, which is known as the winter solstice. 
 
 This apparent movement of the sun makes it pass in turn 
 directly over every part of the zone bounded on either side 
 by the so-called Tropics of Cancer and Capricorn, and the 
 distance of these tropics from the equator, measures, in fact, the 
 precise degree of obliquity of the ecliptic, and the contrast, 
 other things being equal, between summer and winter in each 
 hemisphere. If this obliquity be subject to variation, it will 
 vary the distance to which the sun will travel every half- 
 year, either north or south of the equator. The greater the 
 inclination the greater the distance it will travel, that is to 
 say, the greater will be the distance between the equator .and 
 the two tropics and vice versa. 
 
Variation in the Obliquity of the Ecliptic. 363 
 
 As a matter of fact, the obliquity of the ecliptic is subject 
 to variation, due to the manifold action of the planets. 
 
 " The amount of this variation in the plane of the ecliptic," 
 says Herschell, " is about 48" per century, and has long been 
 recognized by astronomers, by an increase of the latitudes of 
 all stars in certain situations and their diminution in the 
 opposite regions. Its effect is to bring the ecliptic by so much 
 per annum nearer to coincidence with the equator, to cause 
 the two tropics to approach the equator, and the arctic and 
 antarctic circles to approach their respective poles at the 
 rate of forty-five feet annually ; the result being to make 
 terrestial climates more uniform. This is the present effect, 
 but the movement is not continuous but cyclical, and the 
 result is an oscillatory movement, by which, after many years, 
 the obliquity returns to its former position, thus causing a 
 certain cyclical variation of climate. Laplace was the first to 
 measure the amount of the change, and he calculated thatthe 
 oscillation was about 1 22' 34" on each side of 23 28', 
 which was the obliquity in the year 1801. Mr. Stockwell 6 
 has recalculated the position and arrived at virtually the same 
 result as Laplace, and shown that the limits of the variation 
 of obliquity are from 24 35' 58" to 21 58' 36", whence, he 
 says, it follows that the greatest and least declinations of the 
 sun at the solstices can never differ from each other to any 
 greater extent than 2 37' 22". 7 
 
 6 Smithsonian Contributions to Knowledge, vol. xvii. 
 
 7 It is true that Colonel Drayson, in a paper entitled " The Last 
 Glacial Epoch/' claims to have showa that astronomers like Laplace, 
 Leverrier, Herschell, Airy, Stockwell, and others are mistaken, and that 
 the obliquity has once been much greater, so great in fact as to bring 
 the arctic circle down to lat. 54 34' 13" N. ; but it is quite plain 
 that he has been mistaken, and Mr. Belt explains how. His main 
 argument is that observations extending over the last 400 years show 
 that the pole of the earth, instead of describing a small circle round 
 
 alone, the curve performed by the earth's axis would be a circle; but 
 inasmuch as this curve is affected by nutation, and its outline is 
 accordingly sinuous, Mr. Belt says that Colonel Drayson has mis- 
 taken for the segment of a larger circle the motion of the terrestrial 
 pole down the trough caused by one of the sinuosities in question 
 as it dips below the mean distance from the centre. (Belt, Quart. 
 Journ. of Science, 2nd ser. iv. 444). And we may rest assured that the 
 long roll of distinguished astronomers who have examined the pro- 
 blem have not in fact been mistaken in their induction. 
 
364 The Glacial Nightmare. 
 
 Mr. Belt, who relies upon increased obliquity to explain his 
 glacial epoch, urges that while the astronomical forces as 
 measured by Laplace and others undoubtedly limit the range 
 of variation in the inclination of the earth's axis to the amount 
 already referred to, yet that the conditions would be greatly 
 modified if the earth's crust were sensibly augmented in 
 certain places, and he calls on astronomers to reconsider the 
 problem in view of such a change having taken place. 
 AdditioDal data, he urges, respecting the exact figure of the 
 earth have accumulated since the problem was last treated, 
 and this has gone to show that the earth's contour is more 
 irregular than was supposed. 8 
 
 The challenge has not been left unanswered. " Professor 
 George Darwin " has, in fact, shown that ' ' supposing the whole 
 equatorial regions up to latitude 45 north and south were sea, 
 and the water to the depth of 2000 feet were placed on the 
 polar regions in the form of ice, and this is the most favour- 
 able redistribution of weight possible for producing a change 
 of obliquity, it would not shift the arctic circle by so much 
 as an inch." We may take it, therefore, that the calcula- 
 tion of Laplace fixes the extreme possible variation in the 
 obliquity of the ecliptic very closely, and we will proceed to 
 examine the question on that hypothesis. 
 
 We will first consider it as an isolated one, in so far as it 
 is affected by varying obliquity only, and apart altogether 
 from other elements. 
 
 There are two ways in which the obliquity of the earth's 
 orbit affects climate : namely, first, in regulating the relative 
 quantities of heat received by either hemisphere in summer 
 and in winter ; and secondly, in regulating the distribution of 
 that heat over the surface of the hemisphere. 
 
 Sir Robert Ball, in his recent publication, entitled "The 
 Cause of the Ice Age," claims to be the first tc have shown 
 that, with the present obliquity, " of the total amount of heat 
 received from the sun on a hemisphere of the earth in the 
 course of a year 63 per cent, is received during the summer, 
 and 37 per cent, is received during the winter." 1 Sir 
 
 Belt, op. cit. 463. 9 Crol), Discussions on Climatology 4. 
 
 1 Op. cit. 90. 
 
Effect of varying Obliquity upon Climate. 365 
 
 Robert Ball accepts the rigid limitation in the extent of the 
 variation of the obliquity proved by Laplace, and he allows 
 that within any reasonable period of geological time this limit 
 has not been approached. He accordingly treats the present 
 obliquity as virtually constant, and says of the numbers above 
 quoted, that " they are both functions of a single element, 
 namely, the obliquity of the ecliptic. As this fluctuates but 
 little, at least within the periods that are required for recent 
 ice ages, the numbers we have given are regarded as sensibly 
 constant throughout every phase through which the earth's 
 orbit has passed within glacial times." Being a constant, 
 it may, it seems to me, be completely disregarded in this 
 discussion, nor can I see how Sir Robert Ball connects it 
 with an ice age. Whatever potency it has is being exerted 
 now just as much as it would be then. If it were an efficient 
 cause of an ice age, we ought to be passing through one now. 
 We may therefore disregard this law, as an element which 
 remains the same, and which was as much present in the 
 supposed interglacial mild periods as in the glacial severe 
 ones, and which may be discarded when we are considering 
 the effects of change and flux of climate. 
 
 Let us pass on, therefore, to the second effect of the in- 
 clination of the earth's axis; namely, the variation in the 
 area over which the heat received in each hemisphere is 
 distributed. It is clear that as the sun leaves the equator 
 and approaches one pole, it must mitigate its climate for the 
 time being, while it increases the severity of the other from 
 which it moves away ; and that the farther we make the sun 
 travel north and south, that is, the more we increase the 
 obliquity, the more pronounced will be the effect in 
 question, until, when the utmost limit of obliquity is 
 reached and the earth's axis lies in the same plane as that of 
 the earth's orbit, each hemisphere, with its pole for a focus, 
 will have six months of continual sunshine and six months 
 of continual darkness, and thus have a very hot summer 
 alternating with a very cold winter. This seems plain 
 enough, although Mr. Belt, in his well-known paper, seems to 
 have entirely gone astray on the point, and argued as if when 
 
 1 Id. 121. 
 
366 The Glacial Nightmare. 
 
 there was no obliquity, perpetual spring would prevail at the 
 poles, while he attributes the present severity in the Arctic 
 and Antarctic regions to the obliquity of the axis ; overlooking 
 the fact that the poles could not suffer more deprivation of 
 sunshine than when the sun was farthest removed from them, 
 namelv, when it was continuously over the equator. 
 
 Colonel Drayson, who admitted that the effect of altering 
 the inclination of the axis would be to intensify the contrast 
 between summer and winter in each hemisphere, nevertheless 
 seemed to suppose that, if we could so change the angle of 
 obliquity as to bring the Arctic circle into the Temperate 
 Zone, we should cause an arctic climate to prevail in the 
 temperate regions, and thus do much to solve the question of 
 the cause of the glacial epoch. 
 
 Croll points out, with his usual perspicuity, the fallacy of 
 this notion. " The polar regions/' he says, " owe their cold 
 not to the obliquity of the ecliptic, but to their distance from 
 the equator. Increased obliquity in part tempers their 
 climate ; it diminishes the heat received in the equatorial 
 regions, and increases that about the poles. With the present 
 obliquity the proportion is 100 to 42'47. If the obliquity 
 were increased to 35, the amount at the equator would be 
 reduced to 94'93 parts, and that at the poles increased to 
 59'81, being an increase at the poles of nearly one half. At 
 latitude 60, the present quantity is 57 parts, but this 
 would increase to 63 under similar conditions. If the 
 Arctic circle were brought down to England, the tem- 
 perature of these islands would be enhanced and not dimin- 
 ished. The winters would be colder, but the summers would 
 be hotter in a much greater degree, and the greater 
 the obliquity the more would the increase of summer 
 heat exceed its decrease in winter. If the obliquity were 
 increased until it reached its absolute limit of 90, when 
 the earth's axis would coincide with the plane of the ecliptic, 
 the Arctic circle would then extend to the equator. Would 
 this," says Croll, " produce a glacial period? A square foot 
 at the pole would receive as much heat as a square foot at the 
 equator if the sun remained on the equator the entire year, 
 and the total amount received would be as the ratio of half 
 the circumference of a circle to the diameter, i.e., 1*5708 to 1, 
 
Effect of varying Obliquity upon Climate. 367 
 
 or half as much heat again ; and all this heat received by the 
 poles would be concentrated into six months, so that they 
 would be receiving twice as much heat as is received by the 
 equator at present in the same time, and more than three 
 times what it received then. There would be continuous sun 
 there for six months without the ground having the 
 opportunity of cooling for a single hour. Nothing living on 
 the earth could exist in the polar regions under so fearful a 
 temperature as would then prevail in the summer months. 
 How absurd to suppose that this would tend to pro- 
 duce a glacial epoch. Not only would every particle of 
 ice in ,the polar regions be dissipated, but the very seas 
 around the pole would be at boiling-point. Colonel Drayson, 
 in fact, accepts this conclusion, and his glacial period is an 
 alternation of great cold in winter and great heat in summer j 
 but alternations of this kind would not produce a glacial epoch 
 nor accumulate land-ice and icebergs. Great cold in a single 
 winter would involve a deep snowfall, which would disappear 
 in the following summer ; land-ice of any thickness requires 
 the accumulated snows of centuries for its production/' This 
 reasoning of Croll's seems conclusive. 
 
 In the previous discussion, I have endeavoured to isolate 
 the direct effect of increased obliquity. 
 
 In calculating this effect as a practical issue, we must 
 remember that the question cannot in practice be entirely 
 detached, but that the result of changed obliquity is com- 
 plicated with the result of changed eccentricity in the orbit 
 and changed velocity in the earth's motion, or, as Mr. Meech 
 puts it, the variation in the intensity of the sun in different 
 latitudes is a function of the eccentricity and the obli- 
 quity. 
 
 Taking this into consideration, the problem of measuring 
 the actual result can be approximately solved. 
 
 Mr. Meech has calculated and given tables of the sun's 
 present annual intensity in different latitudes. The general 
 result he arrived at is, that if we assign twelve thermal months 
 to the equator, at the Tropic of Cancer or Capricorn there will 
 be but eleven months ; in the latitude of New Orleans there 
 will be ten and a half, in the latitude of Philadelphia nine 
 and a half, at London eight, at the polar circle six, and at the 
 
368 The Glacial Nightmare. 
 
 north or south pole five. 3 Taking these figures as a basis 
 '0187 as the eccentricity of the earth's orbit 10,000 years 
 ago, and taking the formula for obliquity from Struve 
 and Peters' tables, he arrived at the result that the annual 
 intensity within the torrid zone then averaged one thermal day 
 and a half less than now. From latitude 35 to lat. 50 
 it was virtually the same as at present. Above 50 the annual 
 intensity was then greater in an increasing rate towards the 
 pole at which point it was between seven and eight thermal 
 days greater than at the present time ; in other words, the 
 poles, both north and south, 10,000 years ago received 
 twenty rays of solar heat in a year where they now receive 
 but nineteen. Owing to change of obliquity of the ecliptic 
 the sun may be compared to a swinging lamp ; at the former 
 period it apparently moved farther to the north and to the 
 south, passing more rapidly over the intermediate space. 
 
 In this calculation the difference of obliquity allowed for is 
 1 15', which is very near Laplace's maximum variation of 
 1 22' 34" above or below the present obliquity. It thence 
 follows that since the suu and earth were put in their present 
 relation to each other, the annual intensity upon the temperate 
 zone has never varied ; between the tropics it has not varied 
 from its present annual amount by more than about a^o^h part, 
 and is now very slightly increasing. The most marked vari- 
 ation has been at the poles, where the variation amounts to 
 four times that at the equator. In its annual amount, the 
 polar cold is now very slowly increasing from century to 
 century, which effect must continue so long as the obliquity 
 of the ecliptic is diminishing. 4 Croll has followed up Mr. 
 Meech' s calculations somewhat farther. He says, according 
 to Mr. Meech, if 365*24 thermal days represent the present 
 total annual quantity of heat received at the equator from the 
 sun, 151 '59 thermal days would represent the quantity 
 received at the poles. Adopting the same mode of 
 calculation, it follows that when the obliquity was at its 
 maximum, the quantity of heat received at the equator 
 would be 363*51 thermal days, and at the poles 160*04 
 thermal days. The equator would therefore receive 1*73 
 
 Op. cit. 30. 4 Meech, op. cit. 38. 
 
Changeof Obliquity cannot produce Glacial Climate. 369 
 
 thermal days less heat and the poles 8'45 thermal days 
 more heat than at present. 5 When the obliquity was at its 
 maximum, therefore, the Poles would be receiving nineteen 
 rays for every eighteen they are receiving at present, or 
 nearly as much heat as latitude 76 is now getting. This, 
 all other things being equal, would mean a rise in the mean 
 annual temperature equal to 14 or 15, provided, of course, 
 says Dr. Croll, that this extra heat was employed wholly in 
 raising the temperature, which would be the case if the polar 
 regions were free from ice and snow, but as they are covered 
 with ice and snow, the extra heat would simply melt it, and 
 the ice-covered surface could never rise above 32 ; the actual 
 result would be the melting of an eighteenth more snow and 
 ice than at present. 
 
 We seem driven by these results to the conclusion that, 
 even if it were possible to alter the obliquity to any 
 amount, no increase or decrease, however great, could possibly 
 account for a glacial climate or a warm period. The change 
 in the obliquity would no doubt increase the contrast between 
 summer and winter in certain latitudes, but the increased cold 
 of one season would be compensated by the increased warmth 
 of the next, and the ice made in the winter would be melted 
 in the summer. We see an example of this in the planet 
 Mars, in whose case the obliquity of the ecliptic reaches an 
 angle of 28 42'. The result is that in Mars there are very 
 hot summers and very cold winters, and we can actually 
 watch how in the summer the snows which have covered 
 almost a third of the planet's disc in the preceding winter 
 disappear, this taking place in each hemisphere alternately. 6 
 In order to obtain glacial climates or the reverse, we must 
 have some cause operating over several seasons, and not 
 merely a succession of severely contrasted seasons, in which 
 one exactly undoes the work of the other. 
 
 Dr. Croll sums up the case against the appeal to altera- 
 tions in the obliquity of the ecliptic thus : This theory 
 appears to be beset by a twofold objection : (1) it can be 
 shown from celestial mechanics that the variation in the 
 obliquity of the ecliptic must always have been so small that 
 
 5 Climate and Time, 399. 6 See Le Hon, L'homme fossile 340. 
 
 B b 
 
370 The Glacial Nightmare. 
 
 it could not materially affect the climatic conditions of the 
 globe ; and (2) even admitting that the obliquity could change 
 to an indefinite extent,, it can be shown that no increase or 
 decrease, however great, could possibly account for either the 
 glacial epoch or a warm, temperate condition of climate in 
 polar regions. 7 
 
 Elsewhere he speaks of the absurdity of the supposition 
 that an increase of obliquity can possibly account for a Glacial 
 period. 8 
 
 Let us, in conclusion, turn to another element in the 
 earth's motion which is also liable to variation, namely, the 
 well-known motion of the earth's axis due to precession of the 
 equinoxes and nutation. In explaining this difficult question 
 I shall avail myself of Professor Airy's lectures and of a paper 
 by Mr. Burr in vol. i. of the " Intellectual Observer." 
 
 Hipparchus when observing, as was the fashion among 
 ancient astronomers, the passage of certain stars across the 
 mouth of a long shaft or well at the spring equinox, found on 
 comparing his observations with those of Aristillus and 
 Timocharis, who lived 150 years before, that the stars did not 
 cross the field of vision at the same time, but a little earlier 
 every year. After closely observing for thirty years he 
 found that they moved westward about a degree in every 
 seventy-five years, with the result that the solar year is 
 20 minutes 22'9 seconds shorter than the sidereal year. As 
 the result observed was to carry the equinox to meet the sun, 
 the movement was called the precession or acceleration of 
 the equinoxes. It was reserved for Newton to explain this 
 motion by means of his gravitation theory : " He arrived at 
 the theoretical conclusion that the earth is not a perfect 
 sphere, but flattened at the poles, which subsequent accurate 
 measures showed to be correct. The equatorial diameter is 
 twenty-six miles greater than the polar, producing a ring or 
 belt of matter thirteen miles high at the equator. As, accord- 
 ing to the law of gravitation, every particle of matter in the 
 universe attracts every other particle of matter wherever 
 situated, with a force directly proportioned to their mass, 
 and varying inversely as the square of the distance, the sun 
 
 1 Climate and Time, 8. 8 Id. 416. 
 
Effect of Precession and Nutation. 
 
 and moon, both large masses of matter and comparatively 
 near the earth, act powerfully on it ; but as there is more 
 matter at the equator than elsewhere, and this projecting 
 matter is nearer to the sun and moon than the matter at the 
 centre of the earth, the projecting portion is more strongly 
 attracted than the centre, and by its endeavour, in conformity 
 with this attraction to approach the sun and moon, tilts the 
 equator a little out of its position, so as to bring the earth's 
 axis more nearly perpendicular to the equator. . . . This part 
 of the earth cannot move alone. The earth's axis which may 
 be imagined as passing through it like a bar of iron from 
 pole to pole, and stretching to its vanishing points in the 
 heavens must move too; and this effect also was early noticed, 
 viz., that the place of the pole in the sky did move, and that 
 it was clearly the direction of the pole that changed and not 
 the stars themselves." 
 
 " If the path of the pole be traced out among the stars, it 
 will be found that this extremity of the earth's axis moves in 
 an arc of 23^ radius round the pole of the ecliptic." 9 The 
 immediate effect is best seen, says Sir John Herschell, in the 
 apparent position of the Pole-star. At the time of the con- 
 struction of the earliest catalogues it was 12 from the pole ; 
 it is now only 1 24', and will approach yet nearer to within 
 half a degree ; after which it will again recede and slowly give 
 place to others, which will succeed in its companionship to 
 the pole. After a lapse of about 12,000 years, the star A Lyrse, 
 the brightest in the northern hemisphere, will occupy the 
 remarkable position of a Pole-star, approaching to within about 
 5 of the pole. 1 It will be seen thafc the result of this motion 
 is that the earth's axis, in moving about its central point 
 (which is fixed) as a pivot, generates^ double cone about the 
 axis of the ecliptic, and thus each half of it imitates the gyra- 
 tions of a child's top. Prom this cause there is no variation in 
 the angle at which the earth's axis is inclined to the ecliptic ; 
 the angle remains the same, but the axis itself points in a 
 different direction. 
 
 The rate of motion is very slow. It has taken 2000 years 
 to move 33, and if it were not interfered with by other 
 
 9 Op. cit. 358. 1 Outlines, etc., 191. 
 
 B b 2 
 
372 The Glacial Nightmare. 
 
 motions, to which I shall presently turn, it would be about 
 25,868 in going completely round. 
 
 The motion of precession here described is complicated by 
 another motion called nutation. This was discovered by 
 Bradley, who, while engaged in trying to find the parallax 
 of a star in Draco, noticed that it had a proper motion 
 which was not explainable by precession only, as pre- 
 viously understood. It was then seen that, inasmuch as 
 precession is caused by the efforts of the sun and moon to 
 pull the equatorial projection towards themselves, and that 
 as the sun is not constantly vertical over the equator itself 
 but moves from one tropic to the other, thus causing the 
 seasons the effect of this attraction in the way of dis- 
 turbing the position of the equator itself must be greatest 
 at the two solstices, when the sun is farthest from the equator 
 and nothing at all at the two equinoctial points, when it is 
 immediately vertical over the equator, where its pulling will 
 have no tendency to tilt the axis. 
 
 Precession, as affected by the sun, is therefore greater in 
 summer and in winter than at other times, and this irregu- 
 larity constitutes Solar Nutation. A similar irregularity 
 is caused by the moon every month, when it is north or south 
 of the equator, and this forms a part of Lunar Nutation. 
 
 Another part of it, which is more important, is also more 
 complicated. The ecliptic or orbit of the moon is always at 
 an angle of 5 degrees to that of the earth. The two points 
 where the orbits intersect, called nodes, are not fixed, but in 
 consequence of the sun's attraction they are constantly moving, 
 and go completely round the circuit of the orbit in about 
 nineteen years. On the other hand, since the earth's equator 
 is at an angle of 23^- to that of its ecliptic or orbit, it follows 
 that the inclination of the moon's orbit to the earth's equator is 
 constantly changing. For half the nineteen years it is very 
 little inclined to the earth's equator, and for the remaining 
 half is much inclined to it. In the former case, precession 
 goes on very slowly, and in the latter very rapidly ; and 
 this irregularity in the rate constitutes the great part of 
 Lunar Nutation. 
 
 The tilting of the axis from side to side of its normal 
 position is modified by the earth's rotation into motion, in a 
 
Effect of Precession and Nutation. 373 
 
 small ellipse, having a major axis of 1 8", which, is completed in 
 about nineteen years. The effect, if the rotation existed alone, 
 would be the motion of the earth's pole in this small ellipse ; 
 but as precession is going on at the same time, and carrying 
 the pole of the earth around the pole of the ecliptic in a circle 
 of 23i radius, the two motions become compounded, and the 
 result is an undulation of the precession curve, the small 
 ellipses of nutation never being completed, but the motion of 
 the pole being turned into an undulatory circle instead of a 
 uniform sweep round the ecliptic pole. 
 
 The effect of the combination of precession and nutation 
 with the alteration in the obliquity of the earth's axis is, 
 according to careful calculations, to increase the annual 
 advance of the equinox from 50"'2 to 62", and thus to reduce 
 the cycle of 25,800 years due to precession to one of 21,000 
 years. 
 
 The result of the three combined motions has been tersely 
 stated by Mr. Meech in another way. He says : " The 
 perigee advances in longitude 11 '8" annually; by which 
 the instant when the earth is nearest the sun will date 
 about five minutes in time later every year. The time of 
 perihelion, which now falls in January, will at length occur 
 in February, and ultimately return to the southern hemi- 
 sphere the advantage which we now possess." * 
 
 The effect of the two motions was also clearly stated by 
 Reclus. " Every year the exact moment of the March equinox 
 anticipates by about 20 minutes the time at which the 
 corresponding equinox fell in the year preceding. Each 
 revolution of the earth round the sun brings a fresh 
 advance of 2 minutes in the determination of the equinox ; 
 and as during the long course of ages the axis of the earth 
 does not intermit in this swaying motion, the time must come 
 when the conditions of the seasons will be entirely changed. 
 The relative position of the two hemispheres towards the sun 
 will be reversed, and the long winter and short summer which 
 characterized one will be transferred to the other." l 
 
 In the year 4000 B.C. the perihelion, or point of nearest 
 approach of the earth to the sun, coincided with the autumnal 
 
 2 Smiths. Contrib. ix. 38. " The Earth," 12. 
 
374 The Glacial Nightmare. 
 
 equinox of our hemisphere or with the 22nd of September, 
 while in the year 1250 it coincided with the winter solstice. 4 
 
 The effect and the only effect therefore of precession and 
 nutation, when combined with obliquity of the axis, is 
 to reverse the position of the two hemispheres towards the 
 sun every 10,500 years, and thus to reverse the conditions 
 which prevail in them. 
 
 The ultimate result will be of course that sometimes 
 our summer will take place when the earth is nearest 
 to the sun, and sometimes when it is farthest from the 
 sun. 
 
 Humboldt, it seems to me, stated the concrete result 
 quite conclusively long ago. <f lt might be supposed at 
 the first glance," he says, " that the occurrence of the 
 perihelion at an opposite time of the year (instead of the 
 winter, as is now the case, in summer) must necessarily pro- 
 duce great climatic variations, but on the above supposition, 
 the sun will no longer remain seven days longer in the northern 
 hemisphere ; no longer, as is now the case, traverse that part 
 of the ecliptic from the autumnal to the vernal equinox in a 
 space of time which is one week shorter than that in which it 
 traverses the other half of its orbit, from the vernal to the 
 autumnal equinox. The difference of temperature, which is 
 considered as the consequence to be apprehended from the 
 turning of the major axis, will on the whole disappear, princi- 
 pally from the circumstance that the point of our planet's 
 orbit in which it is nearest to the sun is at the same time 
 always that over which it passes with the greatest velocity/' * 
 This seems conclusive. 
 
 If the combination here mentioned were a really effective 
 factor in altering climate, we ought assuredly to find more 
 evidence of it in the southern hemisphere. As Mr. E. L. 
 Price says : " At the present time an aphelion winter and a 
 perihelion summer both occur in the southern hemisphere, 
 and we hear of no abnormal cold or increase of ice. Certainly 
 South America is not invaded by ice, and our ships are going 
 round Cape Horn as usual. It is not known that glaciers on 
 the Andes have grown higher or longer, while it is known 
 
 4 Lapperent, 33. 
 
 6 Humboldt' s Cosmos, JBohn's edition, iv. 458-9. 
 
Effect of Combination of Astronomical Variations. 375 
 
 that by the rise of Norway of but three feet in a century, the 
 mountain glaciers have grown larger by three hundred 
 feet." 6 
 
 So much for the combination of precession and nutation 
 with an obliquity of the ecliptic. 
 
 If, however, we add the element of a varying eccentricity 
 the case becomes a little different. As we have seen, the 
 effect of precession is to reverse the position of the two 
 hemispheres every 10,500 years. This cycle is much shorter 
 than the grand cycles in which the periods of greatest and 
 least eccentricity occur, so that several of the former may 
 occur during one of the latter. The result of this is 
 that sometimes our summer in the northern hemisphere will 
 coincide in time with the greatest possible distance which the 
 earth can get away from the sun, and sometimes our winter 
 will similarly occur. To use astronomical language, either our 
 summer or our winter may occur when the earth is in peri- 
 helion, and at the time of the earth's greatest eccentricity ; the 
 difference between this condition of things and their occur- 
 ring when the earth is in aphelion with the same eccentricity 
 has been supposed to offer a potent cause of climatic change. 
 
 In order to test the problem Mr. Meech has taken the most 
 extreme care. He says : " Let us then look back to that 
 primeval epoch when the earth was in aphelion at midsummer 
 and the eccentricity at its maximum value assigned by 
 Leverrier near to "0777." He then calculates that this would 
 diminish the midsummer intensity by about 9, and increase 
 the midwinter intensity by 3 or 4, the temperature of spring 
 and autumn being nearly unchanged, which he says does not 
 seem adequate to the geological effects required. 7 
 
 This exhausts the possible astronomical changes which can 
 affect climate, and it seems to follow that under no circumstances 
 known to us could astronomical causes produce such a result 
 as the glacial period. Croll, a very acute person and a great 
 champion of ultra-glacial views, says it would be absurd to 
 assert that they could. 8 Mr. J. Geikie, the other great English 
 champion of similar views, agrees with him. " Each of these 
 arguments," he says, "is strictly consonant with well- 
 
 6 The Glacial Epochs, 8. 7 Smiths. Contrib. ix. 40. 
 
 8 Climate and Time, 13. 
 
376 The Glacial Nightmare. 
 
 ascertained fact, and cannot possibly be gainsaid ; and there- 
 fore it may be at once admitted that purely astronomical 
 causes alone will not account for that wonderful alternation 
 of extreme climates to which the geological record bears 
 witness/' 9 
 
 A more neutral person, an astronomer and mathematician, 
 Mr. Meech, sums up his lucid and admirable examination 
 of the problem in these words : f< The preceding discussion 
 seems to prove rigidly that, under the present system of 
 physical astronomy, the sun's intensity could never have been 
 materially different from what is manifested upon the earth 
 at the present day. The causes of notable geological changes 
 must be other than the relative position of the sun and earth 
 under their present laws of motion." 10 
 
 9 Great Ice Age, 103. 10 Op. cit. 41. 
 
RETURN EARTH SCIENCES LIBRARY 
 
 TO * 230 Earth Sciences Bldg. 642-2997 
 
 1 MONT 
 
 ALL BOOKS MAY BE RECALLED AFTER 7 DAYS 
 Books needed for class reserve are subject to immediate recall 
 
 DUE AS STAMPED BELOW 
 
 MAY 1 5 ?nnfi 
 
 - o C 
 
 ~ ~ 
 
 - n ~ 
 
 UNIVERSITY OF CALIFORNIA, BERKELEY 
 FORM NO. DD8, 7m, 12/80 BERKELEY, CA 94720 
 
 s 
 

 
 k