Furthermore, R. A. Daly of Harvard had published his “Nature of volcanic action” on the basis of his summer at Kilauea in 1909. There was strong controversy against Brun on the water question, but the experts, including Day and Shepherd, came to the conclusion that lava eruption of the Kilauea type was actuated by such flaming gases as hydrogen, carbon monoxide, and sulfur; that these gases were in solution in some elemental form deep down in the earth; and that the chemistry of their emission heated the lava on its way up. The lava lakes were hotter at the top than at the bottom. We shall see that all lava partly solidifies at its own bottom and stays liquid above. The items of activity at Kilauea Volcano during the decade from 1911 to 1920 were marked fluctuation up and down in 1912–1913, with a notable low level in 1913, culminating in a strong earthquake in October. In 1914 the liquid lava came back into the bottom of Halemaumau pit, and in December Mauna Loa erupted in a fountain at its summit crater. The lava lakes of Kilauea grew bigger in 1915, and a triangular island appeared, lifting itself up from a shallow flat and even rotating or hinging horizontally. Its uplift was as a peaked escarpment of lava layers tilted in one direction, something very like Perret’s island of 1911. An affinity between Kilauea and Mauna Loa was obvious. In 1916 Mauna Loa completed its summit gushing by splitting open the mountain’s southwest rift and making a lava flow into ranch and forest lands of South Kona. But just as Mauna Loa activity ended, the entire Halemaumau bottom thirty miles away lowered dramatically during one day, leaving a deep seething puddle of melt, surrounded by roaring red hot avalanches. The coincidence, along with appropriate earthquakes, was unmistakable. Immediately after the lowering, the liquid lava of Halemaumau welled up border wall cracks and cascaded through the talus to form an oval pool in the bottom funnel of broken rock. The lava column rose 600 feet in the next six months and a lobate lake developed, its coves separated by sectors of overflow lava which lifted slowly into crags in the center. In 1917 the lakes and crags inside Halemaumau were less than 100 feet down, the lake shores became accessible for experiments with iron pipes, and the crags came into view from the Observatory, fully justifying the daily photograph for comparing changes of the distant pit. [Illustration: _11. Lava lake, showing bench, March 30, 1917_] [Illustration: _12. Halemaumau, showing lava lake and crags, December 8, 1916_] [Illustration: _13. Jaggar holding pipe for sounding lava lake, 1917. Cylinder on end of pipe holds Seger cones for measuring lava temperature_]