620 



NA TURE 



[April 27, 1905 



less truncated as the verge is reached, and bearing little 

 evidence of attrition (Fig. 2). Mr. Molyneu.x is of 

 opinion that the cutting back of the falls is due to 

 the perpetual hammering action of the vast bodies of 

 water falling into, and down upon, the cracks between 

 the basalt columns, assisted by tne constant vibration of 

 the rock from the precipitated masses of water, and that 

 by this constantly exerted force the columns are rent 

 asunder and fall into the chasm, taking with them huge 

 and deep flakes of the precipice. At low water heaps of 

 these blocks, as yet angular and unreduced, may be seen 

 in the shallower ends of the chasm. 



Such is one phase of the erosion of the falls. .-Xnothor 

 power is at work below the water Une. The blocks that 



water, such parts as are protected by islands must be free 

 from such erosion. To-day there are three important 

 islands on the lip of the chasm, and more than fourteen 

 large ones in 4 miles of river above the falls. In the 

 channels between, there must be more prolonged submission 

 to moving currents, by which the cataracts at the ends 

 of the chasm are being deepened into sloping by-washes. 



The falls have checked the deepening of the Upper 

 Zambezi, and until they chisel the groove of the Grand 

 Canon back to the western edge of the basalt sheet, the 

 upper reaches must continue to run at a high altitude and 

 amid low-lying hills. This has prevented the Zambezi 

 becoming a navigable river throughout, and has also had 

 .•1 marke.l inlluence on the i,"- j;r:iphy of S,,utli Afri .,. 



through the jaws of thi 



Gorge. Danger Point 

 )m the Gcosrafhicat Jam 



Pr.olo. I'y Pcarf.ti. Bulawayo. 

 the left ; the promontory of the " knife eige" on the right. 



fall into the chasm disappear in the deeper waters at the 

 jaws of the gorge — yet, impelled by the rush of the current 

 in the confined walls, they must be grinding down and 

 perpetually deepening the cation, to emerge at the eastern 

 end as rounded pebbles and form the shingle beds of the 

 middle reaches. 



The e.xlraordinary zig-zags or acute angles in the 

 cation have always aroused comment, and the author 

 thinks that two main causes are responsible for them — 

 the position of islands that probably studded the river (as 

 now) and also the existence of master joints and fissures 

 in the basalt. On Hoaruka Island this action is ex- 

 eiTiplified in a striking manner, for a stream can be seen 

 falling down a crevice, that forms, peculiarly enough, 

 another acute angle with the chasm. 



Granted that the falls are due to the action of moving 



NO. 1852, VOL. 71] 



SEISMOLOGICAL NOTES. 

 "^rilE attraction of the moon has always been felt by 

 earthquake workers, whatever inay be its effect on 

 earthquakes themselves. 'I'he latest contributions to this 

 aspect of seismology are two papers in No. iS of the 

 Publications of the Earthquake Investigation Committee 

 in Japan. Prof. Omori deals with the lunar daily dis- 

 tribution, finding maxima of frequency between oh. and 

 5h., and again between lah. and i3h., reckoning from 

 the upper culmination. Dr. Imamura, dealing with the 

 synodic monthly variation in frequency, finds that this 

 shows an increase at the syzygies and quadratures : the 

 former is attributed to the combined effects of the attrac- 

 tion of the sun and the moon, while the latter is explained 

 by the fact that the time of high water at Tokio then 

 coincides with that of the diurnal maximiun of barometric 



