OF MONTE SOMMA AND VESUVIUS. 45 
evidence of the innermost or earliest structure and composition, ex- 
cept by the indirect method of examining the ejected materials, 
which we shall do at a later period. 
The splendid section of the Atrio del Cavallo, 320 metres in 
greatest depth, and above 5 kilometres in length, has been the great 
attraction, and studied by many eminent geologists, as Hamilton, 
Breislak, Von Buch, Monticelli, Scrope, Dufrenoy, Daubeny, Hum- 
boldt, not to speak of Phillips, Scacchi, Mallet, and others in more 
modernd ays. 
What we may see for ourselves, and what we learn from the 
writings of these careful observers, is that we have exposed in this great 
section a series of lavas and their derivatives, scoria, sand, ash, &c., 
interlaminated and piled one above the other in the order in which 
they were ejected. Now, as already remarked, if we compare the 
crater of Vesuvius in modern times with the prehistoric one of 
Somma, the exact resemblance of the two is sufficient to prove the 
similar conditions under which both were formed. From such facts 
we are led to conclude that the early eruptive phenomena were very 
nearly, if not exactly, like those going on before our eyes at the 
present day. 
The lavas of the hundreds of streams piled one above another 
and sectionized in the Atrio exhibit a great variety, within certain 
limits, as noticed by Von Buch*. Without any apparent order we 
find mixed together lavas with large leucite crystals a centimetre in 
diameter, and others in which this mineral is only discernible by 
the microscope. Some have a uniform grey matrix, which may be 
broken by crystals of either pyroxene or olivine, or of both together. 
In others, again, the three principal component minerals are largely 
developed together. The labours of Zirkel, Rosenbusch, Clifton 
Ward, Haughton, Hull, Fouque, and Levy, go to show that there is 
a great resemblance between these ancient and the modern historic 
lavas. 
There is a popular idea afloat that the ancient lavas are charac- 
terized by large crystals of leucite. Let us see the cause of this 
error. 
Most rock specimens that are chosen to represent these lavas are 
taken from those currents which show the crystalline components 
well developed ; and secondly, from age, this mineral has been partly 
kaolinized, which results in the change of colour from grey to 
white, so that the latter contrasts strongly with the dark matrix. 
On the other hand, although the crystals of leucite are of considerable 
size in recent lavas, as that of 1858, their colour so nearly resembles 
that of the rock mass that they are almost invisible. An examination 
of an extensive and fairly collected series of rock specimens, will show 
us that on the contrary the majority of the flows are fine-grained, 
especially with regard to the leucite crystals. The most constant 
of porphyritically enclosed crystals are those of pyroxene, much as 
in the lavas of 1631, 1794, and 1872. 
* “Description physique des iles Canaries, 1836. Traduit par C. Bou- 
langer, p. 339. ‘ 
