ANDERSON : THE VOLCANOES OF ICELAND. 
167 
of the lava, none had ever reached the crater ; but he was quite 
willing to make the attempt. We therefore started next morning, 
keeping at first some distance from and then close alongside the lava. 
At last we found it necessary, in order to get to a set of cinder heaps 
which promised a passable road, to cross the main stream, and had 
some difficulty in getting our clever little nags across ; but persever- 
ance prevailed, and going further on we encamped in the evening at 
the last patch of grass at the edge of the desert. Next morning, 
after a hard frosty night, the weather proved good, in fact, the only 
good day for many days, and by riding as far as possible into the 
desert and then leaving the horses and going forward on foot, the 
craters, the objects of such a long journey, were at last reached. 
They extend in a line for several miles along a great fissure, which is 
still in many places clearly visible. At the lower end are two or 
three dwarf craters, then the two or three main orifices from which 
most of the lava has poured out in billows of fire, now solid and 
black, it is true, but retaining their shape perfectly ; and, further on, 
several others from which the gases and steam evidently chiefly 
escaped. 
The higher craters, from which the steam and vapours escaped, 
are roundish or oval ; and the fissures can still be seen along their 
bottoms in places of a width of several feet. Traces of it are also 
visible going under the heaps of scoriae which separate adjacent 
craters, and here constitute their walls. The outer slopes of the 
craters are gentle, the inner nearly precipitous, this conformation 
being apparently due to the scoriae having been ejected in a pasty 
condition, so that they stuck where they fell ; and thus, while those 
which fell again directly into the fissure would be blown out again, 
those which fell out of the direct line attached themselves and did 
not roll back to fill up the vent, as we so often see in ash cones. 
These craters also illustrate most strikingly the fact that water, 
except as running streams, has scarcely any eroding power. Though 
they have been erupted over 100 years their edges are as sharp and 
perfect as the day they were formed, the explanation being that the 
scoriae are so porous that the rain as it falls, and the snow as it melts, 
instantly soaks in, and never appears on the surface as a stream. 
