December 21, 1893] 



NATURE 



"i^lZ 



lordship will no doubt be glad to learn that so far as this De- 

 partment is concerned, scientific specimens sent by sample post, 

 and addressed to places abroad, will not be stopped in future ; 

 but I must state that this Department cannot guarantee the 

 delivery of such specimens abroad, inasmuch as they do not 

 come within the definition of sample packets as prescribed by 

 the Postal Union. ' I may add that within the last month I have, 

 on two occasions, sent specimens abroad by sample post wiih 

 perfectly satisfactory results. 



All naturalists will feel grateful to the Academy of Natural 

 Sciences of Philadelphia for agitating in this matter. But it is 

 to be regretted that the United States Postal Department should, 

 in another way, continue to maintain a barrier against cheap 

 transmission and interchange of specimens. The sample post 

 can, in any case, only be used for small packets, but larger 

 packages can now be sent to nearly all foreign countries by 

 parcel post, the introduction of which was an inestimable boon. 

 The United States Government stands almost alone in persist- 

 ently refusing to co-operate in this respect. It is not for 

 scientific men to inquire into what contracts that Government 

 may have entered into with private carrying companies, or how 

 far it may be influenced by hyper-protective susceptibilities ; 

 they can only regret the facts, and deplore the result. 



Lewi--ham, December 8. R. McLachlan. 



" Geology in Nubibus.'' — Mr. Deeley and Dr. Wallace. 



Mr. Deeley will not have anything to say to ice conveying 

 thrust as a solid body, which has been the sheet anchor of glacial 

 geology for many a decade. He also repudiates Dr. Wallace's 

 notion that regelation can in some way act as a compensating 

 element when crushing supervenes in ice, and thus enable it 

 under crushing pressure to convey thrust. So far so good. 



Mr. Deeley, however, bids me turn to ice acting as a viscous 

 body, a subject on which I have written a great deal in my 

 recent book, which he does not seem to have seen. 



There are two ways in which we can conceive a viscous body 

 lowing on a flat plain : (i) by pure fluid, or what is commonly 

 •called hydrostatical pressure, in- which the upper layers move 

 up and down, and the lower layers alone have a horizontal 

 motion ; C 2) by its particles rolling over each other. The former 

 ■depends, of course, entirely upon the difference of level of two 

 connected parts of the mass under consideration ; the latter 

 depends upon the slope of the upper surface of the fluid. 



I contend, as Forbes contended, that in the case of a body so 

 slightly fluid as ice, motion by hydrostatic pressure is practically 

 impossible. The consistency and mutual support of the parts 

 prevent the indefinite transmission of pressure in this way through 

 ice, and nowhere have I seen or heard that in detached masses 

 of a glacier cut off at either end by crevasses the ice rises in one 

 place, and sinks in another, or that the walls of these ice rifts or 

 the perpendicular ice walls in the arctic and antarctic regions 

 or in scarped icebergs bulge out below in the slightest degree, 

 as must happen if ice were to move in this method. 



Forbes" experiments and measurements and patient examina- 

 tion of the problem proved thst ice as a viscous body moves in 

 fact by its layers rolling over each other, and that this motion 

 is differential, being greatest at the surface and in the middle, 

 and least at the base and sides of a glacier. 



It is quite true that the rate of this motion on a flat plain 

 would depend theoretically on the slope of the upper surface of 

 the ice. It is established by experiment, however, that such 

 motion is very largely confined to the surface layers, and when we 

 approach the nether layers the motion quickly slackens, owing 

 to the internal friction and drag of the ice particles. Even on 

 inclined beds, glaciers have sometimes been found frozen to the 

 ground. The evidence of a large number of observers is con- 

 clusive, that as glaciers reach the level ground, the motion, 

 even of their upper layers, gradually stops. The masses 

 of ice that collect on the flat Siberian Tundras do not move at 

 all, nor do the thick horizontal ice beds examined by Dall in 

 Alaska. Argument, experiment, and observation are therefore 

 entirely against Mr. Deeley, upon whom the burden of proof 

 Tests. Perhaps he will explain what are the conditions under 

 which he conceives his ice sheets to hav« been formed, to have 

 been maintained, and to have moved. Mr. Wallace confesses 

 that he does not like to face these mechanical issues, which are 

 presupposed in all his reasoning. This is assuredly building on a 

 -quicksand, v^hich is not a profitable experiment. He cannot be 



serious, either, in arguing that because I believe in Charpentier's 

 view that the Alps were formerly higher, and consequently 

 nursed bigger glaciers, I am therefore committed to Ramsay's 

 extravagant notions, repudiated by nearly all explorers of 

 glaciers, that the lakes of Geneva and Lucerne were dug out by 

 ice. Charpentier's method, in such a case, would have prompted 

 him to first prove the capacity of ice to do the work, and most 

 people will agree that in a scientific argument this method is 

 alone fruitful. H. H. Howorth. 



30 Collingham Place, Earls Court. 



The Viscous Motion of Ice. 



Is not Sir H. Howorth wrong in assuming that there is no 

 transmission of hydrostatic pressure in ice ? Certainly Forbes 

 was of opinion that such transmission existed, and was necessary 

 to explain the remarkable parallelism between the motion of ice 

 and of vise, us fluids. It is a question of scale. Even a cup of 

 treacle will not flatten out indefinitely ; still less will a barrel of 

 pitch ; but I have no doubt a cubic mile of ice would flatten out, 

 but to what extent is a question for calculation, not for dogmatic 

 assertion. Unfortunately the first requisite of such calculations 

 is w-anting, as no determination of the coefficient of viscosity 

 exists. Canon Moseley's experiments are clearly out of court, 

 and in the interesting experiments of Mr. Coutts Trotter in 1883, 

 the length of the portion of ice which took part in the shearing 

 motion is not given. 



May I add that the paragraph in Sir H. Howorth's letter of 

 November 23, in answer to Mr. LaTouehe, is distinctly erron- 

 eous so far as our limited evidence goes. 



If Sir H. Howorth will draw to scale the observations of 

 Prof. Tyndall at the Tacul on the side of the Mer de Glace, or 

 those of Prof Forbes, given on page 554 of his own book, he 

 will see that while the velocity of the ice is greatest at the 

 surface, the viscous yielding or differential motion is greatest at 

 the bottom ; and the curve into which a vertical line in the ice 

 is thrown by the motion, is always convex towards the direction 

 of motion, is relatively flat above, and strongly curved towards 

 the base. This is exactly what we should expect on the viscous 

 hypothesis, and justifies the application of hydrodynamical 

 treatment to the problem, if only the necessary data were to 

 hand. 



19 The Boltons, S.W. John Tennant. 



December 12. 



Chemistry in Space. 



It may be of interest to your readers to know that the idea 

 of the arrangement of atoms in space, which is looked upon as 

 quite a modern one, is clearly put forth by Wollaston in his 

 paper entitled "On Super- Acid and Sub- Acid Salts" (Phil. 

 Trans, vol. xcviii. 1808, pp. 96-102). 



He discusses the constitution of the two oxalates of potash ; 

 and I make the following extracts, but must refer your readers 

 to the original paper for the full context. ..." when our views 

 are sufficiently extended, to enable us to reason with precision 

 concerning the proportions of elementary atoms, we shall find 

 the arithmetical relation alone will not be sufficient to explain 

 their mutual action, and that we shall be obliged to acquire a 

 geometrical conception of their relative arrangement in all the 

 Three dimensions of solid extension. . . . when the number of 

 one set of particles (combined with one particle), exceeds in the 

 proportion of four to one, then, on the contrary, a stable equili- 

 brium may again take place, if the four particles are situated at 

 the angles of the four equilateral triangles composing a regular 

 tetrahedron. ... It is perhaps too much to hope, that the 

 geometrical arrangement of primary particles will ever be per- 

 fectly known. " Thus Wollaston's conception of the combina- 

 tion of four particles with another is exactly the same as our 

 modern idea of the arrangement of four monovalent atoms (or 

 groups) in combination with a carbon atom. The same idea is 

 also developed somewhat later by Ampere in his "Letter to 

 Berthollet" {Annaks de Chimie, 90, p. 43-86, 1814), in which 

 he considers the molecules as forming various geometrical figures 

 dependent on the number of atoms contained therein. 



JOHX Caxnell Cain. 



The Owens College, Manchester, December 14. 



NO. 1260, VOL. 49] 



