276 
GEOLOGY: W. M. DAVIS 
particularly those occupied by serpentine rocks, which are characterized by 
rolling highlands of moderate relief at altitudes of 600, 800 or 1000 feet. These 
seem to be elevated peneplain areas; they are trenched by relatively steep- 
sided valleys, and are adjoined either by surmounting residual mountains, 
presumably formed of more resistant rocks, or by the lower hills and lowlands 
of the southwestern coast where weaker rocks prevail. The erosion of the 
subdued southwestern lowlands and of the narrow valleys in the highland areas 
has evidently been accomplished after the partial peneplanation of the island 
and its subsequent elevation, and during the same period of higher stand that 
witnessed the cutting of the sea cliffs along the northeastern coast, previous 
to the recent submergence by which the shoreline was embayed. The amount 
of the recent submergence may well have been from 600 feet or more; the pre- 
vious upheaval of the northeastern side of the island, before its cliffs were cut, 
was probably at least twice as great, for the sea-cliffs today, in spite of being 
partly submerged, not infrequently still show 600 or 1000 feet of their height 
above water. The absence of all consideration of these inferences in the 
geological accounts of New Caledonia affords a striking illustration of the con- 
trast between the older geological philosophy that based its theories only on 
the structure of rocks and their mineral and fossil contents, and the newer 
philosophy of geology which broadens the older one by adding thereto a reason- 
able consideration of surface forms and their evolution. 
During my relayed trip around the island on three trading steamers, sup- 
plemented by local sail-boat excursions, in June and July, 1914, the rolling 
highlands were recognized as elevated peneplains at many points on both coasts. 
Where their vegetation is scanty, as is often the case, the soils of the highland 
slopes are laid bare in rain-washed gulleys which disclose their varied colors, 
dark or black at the surface and usually a strange mixture of vivid reds and 
ochres beneath. At certain points the open workings of the highland laterite 
mines were seen, and at one harbor where a steamer touched for an afternoon 
I had time to climb the slopes and inspect the excavations. The residual 
nature of the deposits was manifest enough. The boulders, referred to in 
Miller's paper as lying at the bottom of the loose deposits and as affording a 
rim or coating that is scraped off and added to the ore pile, are perhaps partly 
concretionary in origin, but some of them appeared to be incompletely decom- 
posed rock kernels lying almost in place within a matrix of more disintegrated 
material. It is significant that the abundant hill-side detritus is not worked; 
ore of paying richness and quantity seems to be limited to the highland areas. 
It is further significant that, as is usual in such residual deposits, analyses of 
the highland ore deposits show a much higher percentage of nickel and cobalt 
than is found in the underlying serpentines. 
In view of all this it seems reasonable to infer that the ore deposits are the 
result of surface enrichment by leaching and concentration during the later 
stages of the above-mentioned cycle of peneplanation, and that they have been 
undergoing removal rather than further accumulation and enrichment in the 
