J. W. Judd — On Volcanos. 65 



by the more or less complete disengagement of their volatile con- 

 stituents ; and while in the former, cavities originate which are 

 occasionally lined with the most beautifully developed crystals of 

 the component minerals of the rock, — in the latter, a laminated struc- 

 ture is produced, the planes of which sometimes coincide with, but 

 not unfrequently cross, those produced by the devitrification of the 

 mass under pressure. 



But this attempt at a classification is far from exhausting the varieties 

 of the beautiful quartz-trachytes of Lipari. New forms are originated 

 through masses of obsidian being broken up and entangled in a stony 

 matrix, or by glassy streams enveloping stony or perlite fragments, 

 or, as is not unfrequently the case, by their catching up in their flow 

 angular fragments of lavas of different composition, and belonging 

 to earlier periods of eruption. Thus are originated the most singular 

 brecciated structures, and rocks of very peculiar and, at first sight, 

 puzzling character are produced. 



"When, however, these rocks are studied by the aid of the micro- 

 scope, new features of interest continually make their appearance, 

 only a very few of which it will be possible to notice in this place. 



In the most clear and translucent volcanic glasses which have yet 

 been examined, the beginnings of the process of devitrification can 

 always be detected. Minute acicular crystals of felspar (Belonites) 

 are seen, which, in a later stage of development, assume rectangular 

 forms and ruin-like terminations, and thus gradually approximate 

 to the ordinary characters of sanidine crystals. Other acicular or 

 filiform crystals of hornblende (Trichites) appear and combine into 

 radiating groups or tree-like masses of marvellous beauty. Where 

 these crystals reach the surface of a cavity in the lava, free develop- 

 ment of them often takes place, and we are enabled to study their 

 nature and characters with the greatest facility. 



Most frequently, however, the crystals unite in radiating masses, 

 giving rise to those globular concretions known as splicer ulites. In 

 some cases the formation of these sphasrulites has been determined 

 by the liberation, in the midst of the vitreous mass, of an infinitesimal 

 bubble of volatile matter. By the development of these crystalline 

 globules with such exquisitely beautiful concentric and radiated in- 

 ternal structures, the peculiar forms and distinctive opalescent lustre 

 of " perlite " is originated. 



Nowhere, perhaps, can better materials be found for illustrating 

 the development of these peculiarly interesting structures in vitreous 

 rocks than in Lipari. Some of the pearlstones of this island, as, for 

 instance, that of the lava-stream above Canneto, contain spheerulites 

 of the size of peas. To attempt anything like an adequate account 

 of the varieties assumed by the crystalline interiors and semi- vitreous 

 envelopes of these, would require numerous figures and an amount of 

 detailed description which would be out of place in these sketches. 



It is in the northern part of Lipari that we find the best examples 

 of the volcanic cones, craters, and. lava-streams of the latest period 

 of eruption in the Lipari Islands. 



Supposing a furnace containing many millions of tons of liquefied 



DECADE II. VOL. II. NO. II. 5 



