SPECIAL FORMS OF STRUCTURE. 
509 
passes gradually from the soft state to the hard dolerite, 
greenstone, or other trap rock from which it has been de¬ 
rived. 
Agglomerate ,—In the neighborhood of volcanic vents, we 
frequently observe accumulations of angular fragments of 
rocks formed during eruptions by the explosive action of 
steam, which shatters the subjacent stony formations, and 
' hurls them up into the air. They then fall in showers around 
the cone or crater, or may be spread for some distance over 
the surrounding country. The fragments consist usually of 
different varieties of scoriaceous and compact lavas; but 
other kinds of rock, such as granite or even fossiliferous 
limestones, may be intermixed ; * in short, any substance 
through which the expansive gases have forced their way. 
The dispersion of such materials may be aided by the wind, 
as it varies in direction or intensity, and by the slope of the 
cone down which they roll, or by floods of rain, which often 
accompany eruptions. But if the power of running water, 
or of the waves and currents of the sea, be sufficient to carry 
the fragments to a distance, it can scarcely fail to wear off 
their angles, and the formation then becomes a conglomerate. 
If occasionally globular pieces of scoriae abound in an ag¬ 
glomerate, they may not owe their round form to attrition. 
When all the angular fragments are of volcanic rocks the 
mass is usually termed a volcanic breccia. 
Laterite is a red or brick-like rock composed of silicate of 
alumina and oxide of iron. The red layers called “ ochre 
beds,” dividing the lavas of the Giant’s Causeway, are late- 
rites. These were found by Delesse to be trap impregnated 
with the red oxide of iron, and in part reduced to kaolin. 
When still more decomposed, they were found to be clay col¬ 
ored by red ochre. As two of the lavas of the Giant’s Cause¬ 
way are parted by a bed of lignite, it is not improbable that 
the layers of laterite seen in the Antrim cliffs resulted from 
atmospheric decomposition. In Madeira and the Canary Isl¬ 
ands streams of lava of subaerial origin are often divided by 
red bands of laterite, j)robably ancient soils formed by the 
decomposition of the surfaces of lava-currents, many of these 
soils having been colored red in the atmosphere by oxide of 
iron, others burnt into a red brick by the overflowing of 
heated lavas. These red bands are sometimes prismatic, the 
small prisms being at right angles to the sheets of lava. 
Red clay or red marl, formed as above stated by the disin¬ 
tegration of lava, scoriae, or tuff, has often accumulated to a 
great thickness in the valleys of Madeira, being washed into 
them by alluvial action; and some of the thick beds of late- 
