624 



NATURE 



[April 25, 1901 



removes the sand as quickly as it is brought, and further 

 growth is thus stopped. If the depth be further reduced, 

 e.g. during an ebbing tide, tlie ridges decay. In deep water the 

 height of the ridges is limited to that at which the velocity of 

 the stream can maintain an active eddy on the lee of the ridge, 

 and if the velocity be reduced the troughs silt up owing to the 

 sand falling into dead water. It is pro- 

 bable, also, that the amplitude is sometimes 

 restricted by the velocity of the current 

 surpassing the limit suitable to the fineness 

 or lightness of the material of which the 

 ridges are composed. For it is evident 

 that there is a difference between the pres- 

 sure upon the weather side and upon the 

 lee side of a ridge. I notice when wading 

 on sand banks ridged with tidal sand ripples 

 that the pressure of one's tread often causes 

 a bodily sliding of the sand from trough to 

 crest. The growth and even the mainten- 

 ance of the ripples demands a differential 

 movement of the material. There must 

 be a part to stand fast as well as a part to 

 be redistributed. Thus a moderate range 

 in the sizes of the particles of the loose 

 granular material is favourable to the for- 

 mation and retention of a wavy surface 

 through a considerable range of velocity 

 of the inducing fluid. It is evident that 

 two sets of crossing ridges cannot be 

 simultaneously produced by true current 

 action, and, in fact, crossing tidal sand 

 ripples are seldom seen. Sometimes, when 

 the direction of current has suddenly 

 changed at a particular time of tide, two 

 sets of ridges are successively formed. 

 The " tidal sand ripples " seem, in fact, 

 to be true current -lormed sand waves. 

 They are themselves rippled over with 

 * current mark," which is more properly a 

 ripple, in that (like capillary ripples) it only goes skin-deep 

 (Fig. 2). It seems to be due rather to the pulsations of a current 

 than to the current as such. I have often seen little sand ripples 

 in fairly deep rapid streams, which I sup- 

 pose would properly be called current 

 mark, facing obliquely or transversely to 

 the direction in which the sand was travel- 

 ling. This was near the shore where the 

 most marked pulsation was shorewards. 



I have described elsewhere ^ some little 

 dunes formed by the wind out of the very 

 fine and light sand of the dry Nile bed, when 

 the river was at its lowest. These dunes, 

 which in size and shape somewhat re- 

 sembled tidal sand ridges, were covered 

 with a beautiful tracery of little ripples. 

 In the formation of the seplian sand ripples 

 the heterogeneity of the sand plays a much 

 more important, the pulsation of the fluid 

 a much less important, part than in the 

 case of the current mark of streams. I 

 have one more observation on this point 

 to add to those already published. I 

 noticed, during strong winds lasting several 

 days, at Montrose, N.B., March 1900, that 

 the rippling of the remarkably uniform dry 

 drifting sand was very slow in beginning. 

 As soon, however, as there was a fair 

 supply of the coarser kind of sand grains 

 aggregated together, the rippling action 

 went on vigorously and rapidly. A mode- 

 rate range of sizes of material, e.g. fine 

 sand and coarse sand, is best for seolian 

 rippling. In the vicinity of big stones 

 the scour of the wind is too great. 

 During the occasions on which I visited 

 this little dune tract the strong breeze drifted the fine light 

 sand thickly near the surface, and in the afternoons there was 



1 "On Desert Sand Dunes Bordering the Nile Delta." — Geogr. Jonrn. 

 January 1900. 



a haze of flying sand extending 20-30 feet above the fore- 

 land. The conditions were, in fact, very similar to those with 

 which I have since become familiar in the case of formation of 

 tidal sand ripples. It seems highly probable that the fine, light 

 sand in this wind was in a condition similar to the "eddying 

 suspension " of water drift, proceeding in the manner described 



Fig. 3.— Rippled clouds. Photograph taken August 5, 1900, near Bjurnemouth, 5.15 p.m., looking S. 



by Osborne Reynolds, plus the rapid increase of slight in- 

 equalities forming waves, as above described. The same brisk 

 breezes which were forming dunes as regular trains of sand 



Fig 



The true aerial ripple mark. This is the negative of Fig. 3, and show? the ridges of still j 

 air between the whirling air of the clouds. 



waves, as high as the quantity of dry Nile sand permitted, were 

 not able to deal in the same free fashion with the coarse 

 quartz sand of the larger desert dunes west nd east of the 

 Delta. 



NO. 1643, VOL. 63] 



