420 NATURAL SCIENCE [December 



are important. A summary of the work of transporting agents con- 

 cludes this portion of the work. 



The fifth and last part of the hook is devoted to the consideration 

 of the regolith, a term used to include the soil which covers the solid 

 rocks like a blanket (p>/yo?), an incoherent mass of varying thickness 

 composed of materials essentially the same as those which make up 

 the rocks themselves, but in greatly varying conditions of mechanical 

 aggregation and chemical combination. The materials of the regolith 

 are either sedentary or transported, the first class being either residual 

 like wacke, laterite, terra rossa, &c, or cumulose like peat. Residual 

 deposits are generally unstratified and may possess characteristics 

 which they have ' inherited ' from hundreds or even thousands of feet of 

 rock which have totally disappeared. Transported deposits are col- 

 luvial (scree and avalanche debris), alluvial and aeolian, or glacial. 

 Alluvia vary in character and to a striking extent in fertility according 

 to whether the river bearing them drains a dry or a damp country ; in 

 the former case the undecomposed silicates will be rich in plant food. 

 Where saline deposits are absent, indeed, the soils of very dry 

 countries are of an extremely fertile character and only need rain to 

 be covered with luxuriant vegetation, and this is especially the case 

 because percolating rain has not been present to concentrate the clay 

 into a ' hard-pan ' below the surface. 



The averages of several hundred analyses of soils in dry and humid 

 regions show that the soluble constituents of the soil in arid tracts 

 amount to 30 and in wet tracts to 15 per cent., the greatest differences 

 being in the quantity of lime, magnesia, and alkalies. 



Richthofen's aeolian hypothesis is considered satisfactory when 

 applied to the Chinese loess, but not to that of America, which latter 

 the author supposes to be the result of streams draining from the ice 

 of the glacial period. 



An important organic agent in the formation of marine deposits in 

 inlets is found in the eel-grass, which grows vigorously as soon as a 

 little mud has formed on the bottom, and then by deadening all cur- 

 rents promotes very rapid deposition of fine salt. 



In dealing with the ' rock-flour ' brought down by glaciers, Pro- 

 fessor Merrill alludes to the small amount of actual clay present. It 

 has been proved by Mr E. Dickson that what there is of this material 

 is ground up felspar and not kaolin. 



To illustrate the wonderful degree of comminution reached in soils, 

 it is pointed out that the total surface area of the grains in a cubic 

 feet of soil amounts on an average to 50,000 square feet. On areas 

 like this the operation of organic acids, plants, animals, and water 

 must be enormous. Ants working in such soil effect as much change 

 as the earthworms studied by Darwin, and Shaler's calculation that 

 the former bring a layer of soil one-fifth of an inch thick to the surface 

 every year finds a close parallel in the figures obtained by Darwin. 



Professor Merrill's book is an admirable introduction to a com- 

 plicated and difficult subject. It is packed with facts, not perhaps 

 arranged in the best possible manner, and illuminated with a wonderful 

 series of analyses which contain a fund of valuable information. The 

 author is modest in his deductions, not eager to generalise, and only 

 adopting or advocating a conclusion after full discussion. It is to be 



