30 



NATURE 



[November io, 1892 



Page 156. — For the above reason I withdraw the argument 

 about the cocoons of Rumia, although I believe that it still 

 holds if H. prasinana be substituted. 



Chapters x., xi. should be read in connection with the 

 experiments on Warning Colours since made by Mr. Beddard 

 and published in his volume, "Animal Coloration." 



Page 161. — The cockroach is not a good example. As Prof. 

 Weldon pointed out to me, there is no evidence that its un- 

 pleasant smell renders it unfit for food. The hive-bee would be 

 a better instance. 



Page 193, line 7 from bottom. — Fibrous should he fulvous. 



Page 203, line 6 from bottom. — For suited for read bearing. 



Page 208, line 13 from top. — Divert should be direct. 



Page 224. — I have since heard from Mr. Skertchly that he 

 did not intend the argument which I quote at the bottom of 

 the page to be taken seriously. 



Page 236. — Diadema bolina should be D. misippus, and it 

 and the Danais it mimics occur in three varieties, not in 

 two. I owe this to Col. Swinhoe ; the error was copied from 

 Trimen. E. B. P. 



The Geology of the Asiatic Loess. 



In the spring and early summer of this year I had the oppor- 

 tunity, in company with Mr. S. B. J. Skertchly, of examining 

 closely the loess deposits of Shantung, stretching from Chefoo to 

 Tsinan, the provincial capital. 



The investigation convinced us both that the original loess of 

 China must be regarded as a marine deposit. Subsequent to 

 the time of Mr. Skertchly's leaving the province, on June 17, I 

 was able to supplement these conclusions by the discovery of a 

 band of limestone rocks bored by pholades and crustaceans up 

 to a height of about iioo feet, above which line no indications 

 of late marine action were visible. The rocks in the locality 

 near Tsinan-fu are carboniferous limestones interbedded with 

 dioritic porphyries, and are still horizontal and unbroken for 

 some thousands of square miles, having received their present 

 contour in pre-loess ages. The dip for hundreds of square miles 

 in this locality seldom exceeds from 2° to 8°. These facts we 

 hope to make the subject of a joint memoir. 



The loess of China has, however, been traced almost con- 

 tinuously beyond the limits of the eighteen provinces to the foot 

 of the Pamirs. West of the Pamirs loess occurs in the valley 

 of the upper Oxus, probably in the Kizil Kum, and up to the 

 Caspian, and its marine origin requires us to believe in the sub- 

 mergence within late geologic time of the greater part of Central 

 Asia. Most geologists recoil at such a suggestion, and I am in 

 a small minority in accepting the view that the present distribu- 

 tion of ocean and continent is of very recent date. I may, how- 

 ever, in condonation of heterodox views, refer to the position 

 of the argument with regard to the alleged shifting of the terres- 

 trial axis of rotation, which has within the last few years entered 

 on a new phase. When some years ago I presented these views 

 to the Council of the Geological Society of London they were 

 scouted as utterly untenable. Since that time, while English 

 astronomers have held the view that practically the axis of rota- 

 tion has undergone, within the limits of observation, no change, 

 American astronomers have come to the conclusion that a secu- 

 lar movement is actually in progress. My own geological obser- 

 vations in Europe, North America, and Asia have led me to 

 infer that the North Pole has within recent geological time 

 shifted, and that a shift is in all probability in progress at the 

 present time along a line following approximately the direction 

 of the 70th meridian of west longitude. This shift is not to be 

 taken to involve a change in the direction in celestial space, but 

 is rather a rolling of the earth over its axis, the latter remaining 

 practically stationary. 



Dynamical causes sufficient to account for the change of 

 position of the terrestrial poles, and in consequence of the 

 parallels of latitude, seem to me to be at work. Prof. G. Dar- 

 win has calculated the probable change in the position of the 

 pole due to an elevation of the bed of the Pacific Ocean, but no 

 one has touched the converse effect of the change of the pole on 

 the relative levels of the oceans and continents. In addition to 

 the cause suggested in the possible elevation of large tracts of 

 continental land, there are other influences at work tending in 

 the same direction. The different distribution of the large masses 

 of ice around the poles, which probably varies within some- 

 what large limits, and the slow disturbance of equilibrium re- 



sulting from the growth of deltas and deep sea deposits, have 

 frequently been adduced. More important still is perhaps the 

 differential influence of tidal friction in retarding the rotation, 

 the effect of which must be sensibly unequal in the two hemi- 

 spheres north and south of the equator ; another cause may be 

 looked for in the action of aerial currents, the effect of which in 

 the northern hemisphere as containing greater masses of elevated 

 land must be greater. 



Another potential cause of shifting has never, that I am 

 aware of, been formulated. Although at present of comparatively 

 small influence, it must at various geological periods have been 

 of great importance. It leads on to dynamic considerations of 

 tidal energy beyond the compass of a letter to explain. The 

 relative part played by the sun and moon, as deduced from 

 gravitational formulae, does not quite agree with the observed 

 phenomena of our daily tides. It is believed by many that the 

 ordinary lunar tide, affecting mainly the oceanic envelope, is 

 complicated by the presence of a terrene tide largely influenced 

 by the sun, and that the earth does to an appreciable extent 

 yield twice in the twenty-four hours to the deforming force of 

 solar gravitation. So long as this oscillation takes place at 

 regularly recurrent intervals no energy is wasted. Should, how- 

 ever, a sudden snap occur, breaking the rhythm of the oscillation, 

 some energy is evidently spent, and this can only be made up from 

 the vis viva of rotation. Such snaps do occur occasionally ; the 

 regular oscillation is momentarily suspended, and the waters of 

 the ocean rush in to restore the equilibrium. This is the well- 

 known "tidal" wave that so frequently occurs in connection 

 with earthquakes. 



Such a snap on the equatorial line would simply retard the 

 rotational period generally. North or south of this line, as the 

 moments of rotation would be instantly unequal, the sphere 

 would roll over its axis of rotation, and a shift in the position 

 of the poles occur. The earth is not a perfectly rigid mass. 

 Were it as rigid as steel, the interior within a depth of 200 miles 

 would yet, under the pressure of gravitation, behave as a 

 liquid ; a shift in the pole would then be met either by a corre- 

 sponding shift in the equatorial protuberance, or a change in 

 the ocean level ; or, more probably, by a compound action of 

 both. In the latter case, to fulfil the conditions of equilibrium, 

 the ocean surface in the neighbourhood of the new equator 

 would rise, and if the shift were sufficiently great, would over- 

 flow the lowlands. If the equator, in the longitude of Central 

 Asia, had at any former time passed north of its present posi- 

 tion, and the rock masses of the Continent had not been elevated, 

 a mid-Asian sea must have resulted. The undisturbed posi- 

 tion of the carboniferous rocks, and the plain evidence that the 

 surface sculpturing of the rocks was of pre-loess age, show that 

 the process was unaccompanied by violent movements. 



The theory of the shift of the earth over its momentary axis 

 accounts better than any other for the geological condition of 

 polar lands, and I venture to state it again in brief, as on this 

 occasion the initiative has come from the astronomers, not the 

 geologists. Thos. W. Kingsmill. 



Shanghai, China, August 20. 



Note on Mr. Kingsmill's paper. 



I think it will be difficult for Mr. Kingsmill to adduce 

 evidence of geological changes large enough to produce any 

 considerable shifting of the position of the principal axes of the 

 earth, and accordingly I should feel sceptical as to a theory 

 which postulates that such change has been sufficient to explain 

 considerable changes of climate. 



With respect to a later part of the paper, I am entirely at 

 variance with his views. As far as I know "the relative part 

 played by the sun and moon " in producing oceanic tides is in 

 exact accordance with gravitational formulas. 



The existence of a terrene tide is a matter of speculation, but, 

 as the earth cannot be perfectly rigid, it must exist to some 

 extent. The amplitude of the lunar terrene tide must certainly 

 bear to that of the solar the same ratio that holds in the case of 

 oceanic tides, and there is no reason, that I know of, for attri- 

 buting a greater efficiency to solar action in the case of the 

 deformation of the solid portion of the earth. 



I am quite unable to follow the argument by which the 

 so-called "tidal" wave produced by earthquake shock is 

 supposed to produce a retardation of the earth's rotation. 



October 21. G. H. Darwin. 



NO. 1202, VOL. 47] 



