: 

Marcu 10, 1923] 
limestone resembling our Coralline crag, while towards 
the shore it becomes sandy. Ascending in the series, 
the sand increases and finally passes into conglomerates, 
which at the summit (90 m.) extend over a rocky 
platform bored by Lithodomus and encrusted with 
barnacles—to end against the foot of steep cliffs which 
are undercut and penetrated by sea caves. 
The Sicilian stage is sharply marked off from the 
Calabrian (Upper Pliocene) by a stratigraphical un- 
conformity and a fauna which is distinguished by the 
disappearance of many Pliocene mollusca, and the 
advent of many “cold ” species from the North Atlantic. 
These were brought probably by a cold marine current. 
At Reggio the Sicilian terrace has yielded an entire 
skeleton of Elephas antiquus. 
2. Mivazzian (Depéret). Coast-line at from 55 to 
60 m. The deposits of this stage are chiefly littoral 
with a fauna indicating a temperate climate, but 
warmer than that of the existing Mediterranean. 
3. TYRRHENIAN (Issel). Coast-line at 28 to 30 m. 
This includes the well-known Strombus beds (Strombus 
bubonius) which are found all round the Mediterranean. 
The fauna is characterised by ‘‘ warm” species, such 
as still live off the coast of Senegal and the Canary 
Islands. 
4. MonastTiriAn (Depéret). Coast-line 18 to 20 m. 
This-is named from the city of Monastir in Tunisia, 
adjacent to a locality very rich in fossils of the stage. 
The fauna is almost identical with the Tyrrhenian, but 
on the north coast of the Mediterranean contains no 
“warm ”’ species. 
Tue Four CoRRESPONDING RIVER TERRACES. 
General de Lamothe has shown that the four Quater- 
nary beaches or shore-lines of Algeria correspond with 
the four Quaternary terraces of the river Isser in 
Algeria, and Prof, Depéret has similarly identified the 
Quaternary river terraces of the Rhone with the ancient 
beaches of Provence. 
Thus the river terraces were determined by the base 
level of erosion, 7.e. in the first place by the position 
of the sea-level at the time of their formation. 
They are thus liberated from their supposed depend- 
ence on the four glacial episodes of Prof. Penck. This 
distinguished investigator had, as is well known, attri- 
buted the transport of the material of which they con- 
sist to the action of the comparatively feeble rivers 
which issued from the moraines of the glaciers at their 
full extension during a glacial episode,—a view scarcely 
inconsistent with paleontological evidence. 
Commont has correlated the four terraces of the 
Somme with those of the Rhine, and de Lamothe has 
correlated them with the four Quaternary sea-levels. 
_ But the three lower terraces of the Somme, 7.e. the first 
or Monastirian, second or Tyrrhenian, and third or 
Milazzian, all contain in their lowest deposits a warm 
fauna, which in the case of the lower two includes 
Hippopotamus—an animal which certainly was not 
swimming in the Somme at a time when Switzerland 
and the Baltic buckler were covered with ice and ice 
was floating in the English Channel. Messrs. Hinton 
and Kennard have further shown that a warm mam- 
malian fauna characterises the greater part of the 
terraces of the Thames. 
Thus both marine and river terraces unite in pro- 
NO. 2784, VOL. 111] 
NATURE 
a30 
claiming a warm climate and so far, apart from the 
Sicilian, we have encountered no signs of an Ice Age. 
CONNEXION OF THE TERRACES WITH THE MORAINES. 
We turn then to the moraines which afford evidence 
of the intercalation of glacial episodes in the otherwise 
genial climate of the Quaternary age. Prof. Depéret 
considers that he has proof of the association of the 
Milazzian terraces with the external moraine of the 
Rhone (Mindel), of the Tyrrhenian with the inter- 
mediate moraine (Riss), and of the Monastirian with 
the internal moraine (Wiirm). This association by no 
means implies synchronism, but it enables us to assign 
the several moraines to their respective stages. 
From the point of view we have now reached it will 
be perceived that the term “‘ Great Ice Age ” is a mis- 
nomer, and that instead of speaking of a glacial age 
interrupted by genial episodes, it would be far more 
in accordance with fact to speak of a genial age inter- 
rupted by glacial episodes. 
Since these glacial episodes were quite certainly 
intercalated it will naturally be asked why they are 
not more obviously represented in the fauna of the 
Quaternary age. The answer to this is that the 
remains of a cold fauna are by no means infrequent, 
but the gravels at the base of a terrace are not the 
place to look for them. It is to the slopes between 
the terraces that we should turn, for it is these which 
correspond with glacial episodes, and when, as generally 
happens, they are covered with léss we find in it the 
bones and teeth of such cold-loving species, as we 
might expect. 
It may further be pointed out that terraces, both 
marine and fluviatile, mark a stationary level of the 
sea when deposits were accumulating, in which the 
contemporary warm mammalia might easily be pre- 
served. 
In the intervals when the sea-level was changing 
the work of denudation ruled supreme and undisturbed 
deposits were formed but sparingly. Now and then 
no doubt the bones of some animal belonging to the 
cold fauna might escape destruction and find burial 
in a marine terrace along with the warm fauna proper 
to it, and thus possibly have arisen some of those 
perplexing anomalies of distribution with which we 
are only too familiar. 
REGRESSIONS. 
The movement of the sea-level does not appear to 
have been a simple fall from the Sicilian to the Milazzian 
Sicilian 
Tue Somme 
Milaszéan 
Tyrrhenian 

Ging Mindel Riss Wiirm | Neoléthic 
Peat - 24m 
Fic. 1.—Oscillations of the Quaternary sea-level. aé, Sicilian stage; dc, 
Milazzian; cd, Tyrrhenian; @ to Neolithic peat, Monastirian. The 
numbers indica ate the heights of the marine terraces in metres. 
coast-line and from the Milazzian to the Tyrrhenian ; 
there is evidence to show that it fluctuated (Fig. 1), first 
