8. Poivers — Notes on Hawaiian Petrology. 279 



from the first outbreak to the close of activity, which is 

 frequently a lava flow after the mountain has been cov- 

 ered by ash cones and perhaps buried under a layer of 

 ash. It is with the events toward the close of the life 

 of the volcano that this paper has dealt. The volcano 

 increases in height by extravasation from a summit 

 crater until the lava finds an easier path to the surface by 

 Assuring the mountain, thus pouring out flows and build- 

 ing cones on the sides while a sink is gradually formed at 

 the summit by successive inbreaks when the lava is with- 

 drawn down the feeding tube at the close of an eruption. 

 Mauna Loa and Kilauea have both been active through 

 historic time in building up the floor of the summit sink 

 as well as in pouring out flows through fissures down the 

 sides of the mountain. Increasing viscosity of the 

 magma may lead to the formation of domes of the Pelee 

 type, as in the historic record of Kilauea, and to lessened 

 activity with a longer period between great eruptions. 

 Finally, the gaseous lava which reaches the summit sink 

 may begin the formation of cones over the latter and at 

 last bury it. Mauna Kea and Hualalai may have 

 possessed summit sinks now buried, while the process of 

 burial was not completed at Haleakala. 



It has been seen that the cones on the sides of the 

 mountain have a tendency to a definite arrangment along 

 radial and occasionally along tangential lines. While one 

 group of these cones is formed near the close of the activ- 

 ity of the main volcano, as those of Hualalai, Mauna Kea, 

 and Haleakala, another group may be formed after the 

 close of the main activity and after the volcano has been 

 actively eroded for a long period of time, as on Niihau, 

 Kauai, East Oahu, East Molokai, and West Maui. Espe- 

 cially apparent in the case of the two westernmost islands 

 is the long interval between the close of the principal 

 activity and the revival of activity in cone-building. Simi- 

 larly, revival of activity may occur on Haleakala, Hua- 

 lalai, or Mauna Kea after these mountains have been 

 deeply dissected. That there may still be activity along 

 the line of the extinct volcanoes is indicated by the fact 

 that fish were reported to have died in large numbers off 

 Molokai in 1859; that in 1868 and 1877, concomitantly 

 with activity on Hawaii, the fish in the brackish water 

 fish-pond Nomilo, an old crater on the shore of Kauai, 

 were all killed; and by the fact that in 1868 one of the 

 earthquake epicenters appears to have been south of 

 Kauai. 



