3^4 



NA TURE 



[February 15, 1894 



be wafted down a metallic tube three or more metres long by 

 a draught of air — a crucial test which I believe to have been 

 the first to make, though Mr. Bidwell does not affirm it. If 

 charged particles, dissociated particles, and the like are sucked 

 down there, I must insist that the said charged or dissociated 

 particle — whatever befall it — is a particle still ! If it is not 

 small enough, it will fail to stimulate condensation in any case or 

 theory. If it is small enough, it mzist produce nuclear condensa- 

 *ion per se, no matter how the particle is otherwise conditioned. 

 Aside from this, the electrical theory becomes more seriously 

 entangled by the faci that the jet itself is no mean generator of 

 electricity. I made a few tests on this point in the line of 

 Faraday's classical researches, and obtained marked charges from 

 the jet, increasing with its intensity. 



(5) I cannot claim much indulgence for my experiments in 

 air filtration, though I went through them laboriously enough. 

 But the work taught me, at least, the extreme elusiveness of 

 such a thing as "dust-free" air. Naturally I am biassed, 

 therefore, in regard to arguments, in the present case, based on 

 filtration. Suppose an oxyhydrogen flame burning in filtered 

 air is an active dust producer. The question at once arises on 

 what kind of hearth is the flame kindled : if it burns from a 

 glass tube, then sodium is probably volatilised ; if from a metallic 

 tube, then the metal is similarly in danger ; if from no base 

 at all, then where is the flame ? In what does the remarkable 

 activity of flames really consist ? Most reasonably, it seems to 

 me, in this, that particles therein entrapped are (as a rule) at 

 once volatilised, so that for each single particle we have now a 

 whole cloud of active nuclei precisely the kind wanted in § 2 ; i.e. 

 myriads of them, all in that extreme degree of tenuity which 

 best promotes condensation. In my work, glowing smokeless 

 charcoal often did better service than flames. Alkalies are here 

 ready for volatilisation. In general, hot flames are more active 

 than colder flames (Helmholtz), and they should be where dis- 

 gregation is needed. 



Fmally, a word with regard to red-hot platinum. " Here," 

 says Mr. Bidwell, "there can be no nuclei formed of 

 products of combustion, for there is no combastion, 

 simply ignition and incandescence." Is it possible that Mr. 

 Bidwell is not aware that red-hot platinum is particularly 

 remarkable for scattering small solid particles from 

 its surface? If so, he has narrowly escaped being overwhelmed 

 by the literature of the subject. Aside from this, what may be 

 the nature of platinum dissociated at red heat ? 



(6) To conclude : I cannot discern that the proof of anything 

 beyond condensation of supersaturated steam, induced by mere 

 "inert" nuclei, has yet been given. Nothing is even said of 

 atmospheric air, which indoors or out, stagnant or fresh, is al ways 

 active ; nor is it even hinted at that the efl'ect of dust is merely 

 an acceiitualion of the eff^ect produced by such air ; that dust- 

 stimulated condensation dift'ers merely in degree, by no means 

 in kind, from jet condensation in air. Air nominally purified 

 needs only a higher degree of supersaturation to evoke conden- 

 sation running through the whole gamut of colours. There 

 are no hard drawn lines. I wish there were. Indeed, 

 I wish the proof in question could or had been adduced, for, 

 together with my colleague Prof. F. H. Bigelow, I am well 

 aware of the important meteorological consequences to which 

 this result would lead. But before I can break away from the 

 time-honoured point of view, so safely trussed by the Kelvin 

 formula, the experimental evidences forthcoming must be as 

 rigorously " dust-foU " as the clear conscience with which I 

 am disposed to admit them. Carl Bakus. 



The Smithsonian Institution, Washington, D.C., U.S.A. 



The Origin of Lake Basins. 



The question of the origin of lake basins has again been 

 raised, and, unfortunately, there is tven now the same diamet- 

 rical opposition between the views of the glacialists and their 

 adversaries. Though the old lines of argumentation, perhaps, 

 have not been followed out with sufficient perseverance, new start- 

 ing-points certainly seem desirable. I should therefore like to 

 challenge the criticism in your columns on a demonstration of 

 the glacial origin of the greatest fjord basins in Norway. 



On the western coast of Norway we have the well-known 

 series of great fjords, generally of enormous depth, about 400 

 fathoms in Hardangerfjord, 600 in Sognefjord, and 300 in Nord- 

 fjord, to take only the greatest. The heads of the innermost 



NO. 1268, VOL. 49I 



branches are nowhere at a greater distance than some twenty 

 miles from the watershed of the country, and the necks between 

 them nowhere as much as forty miles. When for some reason, 

 change of climate or rise of land, the snow began to gather 

 on the neighbouring plateau, the highest in Scandinavia, glaciers 

 would creep down the steep slopes and valleys, and immediately 

 get to the deep fjords. But here the glacier ends must needs be 

 carried away or dissolved as fast as they came on. By the neces- 

 sarily slow growth of the narrow neves, it is impossible that the 

 glaciers could advance at once with so great dimensions that 

 the fjords were not able to master them successively. Over the 

 narrow necks between them an ice cap certainly might push 

 farther out, but as there is less than twenty miles to draining 

 outlets on either side, this cap could attain no considerable 

 thickness, and bring no great ice flow westwards from the high 

 land behind. On the lower foreland, near the coast-line, there 

 is nowhere sufficient gathering basin for a great neve, and the 

 small fjord branches could easily drain their eventual surplus. 

 My opinion is, then, that no great inland ice could possibly ad- 

 vance farther west in Norway than to the close set row of fjord- 

 heads. 



It may be said that the whole country in the great Ice Age 

 was so^much elevated that the fjords were only dry valleys. 

 But no amount of elevation could ever drain fjord troughs gen- 

 erally 300 to 600 fathoms deep, and only loo to 150 wide at the 

 brim. And an adequate differential lift of the inner side of the 

 troughs would give the old palaeozoic mountains in Norway the 

 height of the youjigest mountain ranges, and is on the face of it 

 impossible. It may be further said that ice from the high land 

 behind the watershed might have contributed to the supply, 

 and that the glaciers then would be large enough. But even in this 

 case they must needs commence as small ice tongues, which 

 would he cut off successively ; and the boulders show that no 

 transport from any distance behind the watershed took place 

 during the great Ice Age. 



The ideal have put forward is capable of maintaining itself by its 

 own power, but its position is greatly strengthened by direct facts. 

 It can be demonstrated that a great inland ice has tried to advance 

 from the high land westward beyond the fjord heads, and signally 

 failed. This was the case in the last Ice Age, of which we can 

 trace all the prominent signs. As the ice cap was not able to 

 build itself up to any great dimensions near the western preci- 

 pices to the fjords, its greatest height was piled up farther to the 

 south-east, and the ice shed was drawn in the direction of the 

 eastern margin, down the eastern valleys. From this time we find 

 boulders transported up the eastern slope from a distance up to 

 eighty miles trom the watersheds, and these boulders can be 

 followed in great heights in the western valleys only to the 

 fjord-heads, when they suddenly drop down to the old sea 

 beach. This shows, beyond all doubt, that the glaciers from 

 the second inland ice which far away to the south-east laid up 

 the upper till in Prussia, on the north-western side, only reached 

 the fjord-bottoms, and were not able to fill the fjords. We 

 have in this an empirical proof of my idea. An inland ice is 

 really not able to advance beyond a close set row of deep 

 depressions as the Norwegian fjords. 



But yet we have unquestionable proof that the exterior part 

 of the west coast was extremely glaciated. Just in the mouth of 

 the great Sognefjord itself we have the Sulen Isles with rock 

 scorings and eastern erratics (but none from behind the water- 

 shed) up to 1800 feet above sea level, 5400 feet above the fjord 

 bottom. How can these boulders have been transported across 

 this abyss when no inland ice could ever have advanced beyond 

 the fjord head, and no local glaciers from the peninsula on 

 either side could ever have crossed the deep channels, or piled 

 itself up to sufficient height here only a few miles from the 

 steep slope to the Atlantic Ocean basin ? I cannot see any way 

 to account for these facts other than by the supposition that the 

 Jjords or depressions of the same kind and depth did not exist 

 when the first great inland ice was forming, but were quite 

 completed when the second (and last) commenced. Ergo, the 

 Norwegian fjords are of early glacial or interglacial origin. 



At this point of my reasoning comes in the conclusive series 

 of arguments which puts any other origin than glacial erosion 

 quite out of question for this peculiar flat troughs in old solid azoic 

 and palaeozoic country. I wish to lay especial stress upon the fact 

 that Norwegian geologists for many years have laid great weight 

 upon the really marvellous circumstance of lake-distribution 

 only in glaciated districts, as one of the best of the many in- 

 direct proofs of the glacial origin of rock basins. I should think. 



