Vol. 58.] 10 THE PHYSICAL GEOGRAPHY OF THEIR AGE. 201 
neither this nor any other explanation that I could devise was 
satisfactory. So I consulted Prof. Garwood, in the hope that his 
wider experience of cold regions might supply the key to the 
problem. His reply was to this effect : 
‘ What can be seen in Spitsbergen will explain it. There the rocky mountain- 
slopes are greatly shattered by the frosts, and the material is always slipping 
down when the season permits. A talus is formed; this is covered up by snow 
in winter; when the fragments come showering down from above, they slide 
easily over this. Year by year this process is repeated, and so the fringe mainly 
grows on its outer edge. The fragments which fail to reach this, or fall when 
the scree is exposed, go to raise it; but these contributions are the less im- 
portant. A stone which has acquired considerable velocity before reaching 
the snow will travel for long distances before it comes to rest.’? 
This explanation seems to meet all difficulties, so that while more 
than one may be applicable to breccias which have not travelled far, 
er are very sporadic in character, we are justified in assuming wide 
fringes to indicate physical conditions such as now exist in Persia. 
The Keuper breccias suggest, though less forcibly, a recurrence 
of Rothliegende conditions, and the British Trias as a whole 
appears to be a ‘continental’ deposit.” If [ am right in regarding 
the Bunter Beds as the product of rivers far greater than any of our 
present British streams, and in denying the existence of Mercian 
Highlands of real importance, the sands and pebbles of the Bunter 
and the Lower Keuper may have been deposited on a lowland, which, 
notwithstanding, or perhaps because of, heavy precipitation on the 
mountains to the north and west, was itself arid, like many similar 
districts in Asia. Even the Keuper Marls, though deposited in an 
inland sea, do not necessarily imply a climate like that of the Dead 
Sea. At the Great Salt Lake the mean annual temperature is 51-98° 
Fahr., that of July being 76°64° and of January 25°88°: the rainfall, 
so far as I can ascertain, being about 10 or 11 inches. On the 
Caspian, the annual temperature at Baku is 61°88° Fahr., being 78°44° 
in July and 38°12° in January; while salt-lakes are abundant in 
Turkestan, where the winters are exceedingly cold. Sothe Dolomitic 
Conglomerate is also probably indicative of a continental climate, 
with cold winters and a rather limited rainfall. The Upper Oolite 
of Sutherland seems more likely to be the product of a mountainous 
' The velocity, he tells me, acquired by a toboggan in sliding obliquely 
down some 70 or 80 feet to the St. Moritzer See, will carry it across the snow- 
covered jce (about half a mile). 
2 Mr. Goodchild, in his paper on ‘Desert Conditions in Britain’ Trans. 
Geol. Soe. Edin. vol. vii (1896) p. 203, comes to much the same conclusion 
in regard to the ‘New Red Sandstones’; so I wish to say that the present 
paper was written without having read his, for I did not come across a 
reference to it until a few days before sending mine to the Geological Society. 
I should, however, hesitate to press one or two of his arguments: as, for 
example (p. 215), I would not say more than that the red colour of these breccias 
made it rather probable that they had adesert origin. In regard to the Torridon 
Sandstone, I may add that for some years I have been in the habit of pointing 
out to my students that subaérial ecnditions would be very favourable to its 
formation ; but I should not like to use astronger phrase. There isa suggestive 
paper bearing on the subject by Dr. W. Mackie, in the same volume of the 
‘Vrans. Geol. Soc. Edin., at p. 443. 
