Proceedings, &c, for 1885. 159 



the melted matters steam is generated of a high temperature and 

 very high pressure. And this steam may be considered as the 

 chief motor of most, if not all, volcanic eruptions. There is 

 greater probability of eruptions breaking out along such a fissure 

 than under any other circumstances, provided always that sufficient 

 water can percolate. As the volcanoes in the Sunda Straits have 

 been at rest for about two hundred years, we may now conclude 

 that the water supply during that time was insignificant, and only 

 materially increased in the last few years. In these years several 

 earthquakes took place along this fissure, which principally affected 

 the lighthouse at Java's first point. The most severe earthquake 

 took place on the 1st September, 1880; the top part of the light- 

 house broke off, and the rest had to be taken down. These earth- 

 quakes were no doubt the result of subterranean changes, and it may 

 be supposed that the fissure underwent several modifications, which 

 allowed the sea- water to penetrate in larger quantities. In the last 

 three years the pressure of the steam was evidently of sufficient force to 

 press the lava upwards out of the deeper-lying lava cavities into 

 the funnel of the crater of Krakatau, and the eruption took place 

 when at last the steam was forced through the lava into the funnel 

 of the crater and to the surface. A portion of the lava was at the 

 same time carried away, and ejected in the form of fine particles of 

 dust. The porous nature of the ejected matter — being nearly all 

 pumice stone — is, no doubt, to be attributed to the great force 

 with which the steam was blown through the lava. A more 

 elaborate account of the manner in which the eruption was brought 

 about will follow, as drawings are necessary to fully understand 

 the matter. It is, however, to be observed that the Krakatau 

 eruption has greatly modified our ideas in regard to the form and 

 extent of the subterranean cavities. Assuming that there is a 

 connection between the eruption and the increased activity of the 

 volcanoes of the Indian Archipelago since that time, the earth- 

 quakes in Australia coinciding with the eruptions (which at any 

 rate is a most remarkable coincidence), it follows that the cavities 

 are much more extensive than they are supposed to be by the 

 geologists of the present day. Krakatau was the only crater at 

 work. The eruptions of the 26th and 27th of August were accom- 

 panied by terrific reports and atmospheric vibrations. During those 

 days an incessant rumbling sound was heard, resembling that of 

 distant thunder, but the actual eruptions were accompanied by 

 sharp reports, resembling heavy cannon firing, whilst the most 

 terrific reports were of much shorter duration, and cannot be 

 compared to any other sound. The great distance at which the 

 reports on the 27th of August were heard exceeds all previous 

 experience. The reports were heard at Ceylon, in Burmah, 

 Manila, Doreh, and Geelvink Bay (New Guinea), and at Perth 

 (Western Australia), and at all places nearer to Krakatau. In order 



