248 R. T. HILL — PELE AND THE WINDWARD ARCHIPELAGO 



them. The original source and nature of this matter before it acquired 

 the form and composition with which it made its exit, involving the 

 ultimate cause of vulcanism, is the great secret of the earth's interior 

 and largely a matter of scientific speculation, which will be more fully 

 discussed in the final pages of this paper. The chemical and physical 

 changes of this primitive matter in its progress from the earth's interior 

 to the surface was the main work of the volcano. 



The matter as it reached the surface and atmosphere not only repre- 

 sented the secondary products of the great interior work, but its expul- 

 sion at the surface was only one of the many stages of the volcano which 

 was at work long before surface eruptions and which will continue at 

 work long after they have ceased. These stages of the apparent work 

 are : 



1. The quiescent or fumarolic stage representing the normal condition 

 of the volcano in its periods of surface inactivity. 



2. The preparatory stage marked by increased fumarolic conditions, 

 slight local earthquakes, increasing temperature of the water, and other 

 conditions indicative of the ascent of the gases and magma in the vicin- 

 ity of the old conduits. 



3. The explosive or catastrophal stage marking the outbreak of the 

 magma at the surface and the clearing the debris which had previously 

 clogged the ancient vents. 



4. The active free working stage marked by a continuous series of 

 eruptions from the somma or summit crater. 



5. The decadence of force and closing of the vent again. 



The first or quiescent stage of the work of the Pele volcano is well 

 known. The ancient somma was probably filled with water, constitut- 

 ing a lake. Around the sides of the mountain at various places, notably 

 at Precheur, were a few warm springs indicative of the latent tempera- 

 ture beneath the surface. 



There is abundant evidence that for months prior to the actual dis- 

 charge of steam and lapilli the second stage was marked by the gradual 

 ascent of the magma, indicated and announced by increasing tempera- 

 ture of the waters of the lake in the old orifice, increased fumarolic 

 action, slight local tremors and noises within the mountain, swelling of 

 the streams, associated with mud and heat, and the final appearance of 

 clouds of steam and lapilli.* 



The third or catastrophal stage of the work of the volcano was specially 



* Professor Milne in Nature for November 27, 1902, notes that in Saint Vincent local earthquakes 

 had been on the increase for a year, and as far back as May, 1901, people were frightened away 

 from the northeast end of the Soufriere by rumblings and quakings, but there is no definite record 

 that Martinique was subject to any marked or unusual tremors preceding the first eruption, 

 except in the instances as noted by Mrs Prentice a few days before the catastrophe. 



