324 
THE VOYAGE OF H.M.S. CHALLENGER. 
or (lie at sea, their soft parts, and even their bony structures, might be entirely removed 
in solution, while the stones and pebbles contained in their stomachs would remain as a 
part of the deposit. 
It has been pointed out that minute fragments of rocks, especially particles of quartz 
and other continental minerals, have been found in some of the deep-sea deposits at 
(Treat distances from the coasts of Africa and Australia. This abnormal distribution is 
to be accounted for by the great distance to which winds may carry dust from desert 
regions on the continental surfaces, as, for instance, the Sahara and the interior of 
Australia. 
It will thus be seen that the area to which continental debris may be transported 
over the floor of the ocean varies greatly in different localities. It is least along high 
and bold coasts in tropical and subtropical regions ; it is more extensive off the mouths 
of great rivers, off the coasts of desert regions, and in enclosed seas, but is most extensive 
towards the polar regions, where blocks of all sizes and kinds are widely distributed by 
icebergs and other kinds of floating ice. 
Minerals derived from the Disintegration of Continental Rocks . — ^An examination of 
terrigenous deposits shows that the prevailing minerals around continental shores are 
those that might be derived from the disintegration of emerged lands. The size of these 
minerals, as well as their abundance, is in direct relation with their greater or less 
distance from the coasts, except in iceberg regions. They have frequently a rolled aspect, 
their angles being softened, and they recall by all their peculiarities the same mineral 
species which constitute most of the geological layers making up the continental masses. 
Quartz plays the principal role. The normal position of these minerals is coincident with 
the distribution of terrigenous deposits, and if exceptionally they are found in pelagic 
deposits, they have been in these cases transported by icebergs, by atmospheric currents, 
or other agencies to which we have just referred in speaking of the distribution of con- 
tinental rocks. 
In some cases there are special characters which may serve as a guide in attempting 
to establish the terrigenous origin of these particles, but it must not be denied that this 
subject is surrounded with many difficulties. It is often difficult to determine the age 
of certain rocks by a study of their lithological composition ; in a much higher degree, 
therefore, is tlie determination of the isolated minerals which constitute these rocks a 
matter of great uncertainty. In all these cases the most certain guide is the mineralo- 
gical association witli the rock fragments in the deposits. There are some minerals which 
have not been recognised in recent eruptive rocks, or at least are extremely rare in these 
mas.ses, wliilc on the contrary they are extremely abundant in the rocks of the ancient 
eruptive series; tourmaline and muscovite arc examples. If minerals, about which 
there is uncertainty as to tlieir age and origin, be associated with fragmentary masses of 
cr}'stallinc and sedimentary rocks of the ancient series, we may conclude with very great 
