368 • Geology of 



hornblende, with some small per centage of carbonate of lime. It 

 contains also remains of land shells and moa bones, the latter generally 

 surrounded by marly concretions. The eminent German traveller and 

 geologist, Baron von Kichthofen, has thrown a great deal of light upon 

 the mode of its formation, through his researches on the nature of 

 loess deposits in China, where they cover districts of enormous extent,, 

 and reach a thickness of 500 to 1500 feet, measured in a vertical 

 direction. He has shown in his last publications, that the loess in 

 China could only be of subserial origin deposited by agencies, w r hich at 

 the rjresent time are still at work in forming that rock. Atmospheric 

 currents, together with the growth of grass and other vegetation 

 during an untold number of years, are the principal agencies by which 

 the loess has been deposited. In the first instance, rain-water running 

 down the more or less steep slopes of the country carries with it fine 

 particles, which are partly retained by tbe grass or amongst its roots, 

 whilst the wind blowing across the land takes up a great amount of fine 

 sediment, afterwards also partly caught and retained by the grass. How- 

 ever, a third and most important agent is to be found in the roots of the 

 plants themselves gradually decaying, and thus raising the ground. 

 There is a peculiar vertical capillary texture observable in the true 

 loess, deriving doubtless its origin from the decaying of the number- 

 less rootlets during many past generations of grasses, to be also 

 noticed in numerous localities in Banks' Peninsula. Thus von 

 Kichthofen correctly styles the loess beds a grave-yard of innumerable 

 generations of grasses. Of course I do not wish it to be understood 

 that all beds of the nature of loess have been formed in that way. 

 Many have been deposited in lakes and lagoons, others by rivers over- 

 flowing in heavy freshets the low ground along their banks ; but the 

 general character and position of the principal loess (or loam) beds in 

 this province prove clearly that they have been formed by the modus 

 operandi pointed out by von Kichthofen. There is, however, one 

 difference which I wish to point out, and that is the absence in the 

 Canterbury beds of the peculiar small marly nodules so common on 

 the Khine, the Danube, and China, where they are named loess babies, 

 little loess men, and stone ginger — if we do not consider the large 

 marly concretions surrounding the moa bones their equivalents. The 

 remarkable regular concretions assuming so many curious forms 

 obtained in the gorge of the Kakaia and near the junction of the 

 Acheron, have been formed in argillaceous beds of lacustrine origin. 

 They resemble the so-called Morpholites of Ehrenberg found at 

 Denderah in Egypt, or the Marlekor of Sweden, both having been 



