154 CARNEGIE INSTITUTION OF WASHINGTON. 



activity into 1915. In 1916, although the explosive activity had died 

 down, our work was continued on the hot springs about the base of the 

 mountain. 



The observations were followed by a year's work in the Laboratory, 

 the results of which have been prepared for publication, but will not 

 be in print in time to form a part of this year's report. 



During the past year opportunity has been offered, through the 

 courtesy of the National Geographic Society of Washington, to send 

 three men to the ''Valley of Ten Thousand Smokes," an extensive 

 active fumarole region adjacent to the volcano Katmai, on the Alaska 

 Peninsula. The recent eruption of Katmai was of the explosive type, 

 and was on a scale hardly equaled in historic times. It was accom- 

 panied by the formation of the "Valley" referred to, in which an un- 

 usual type of fumarole activity has been going on for several years. 

 The Laboratory furnished thermometric and gas-collecting apparatus to 

 the National Geographic Society expeditions of 1917 and 1918, and the 

 temperatures found were so high and the vapor composition so un- 

 usual that it was thought well to send a party this summer equipped 

 to do a certain amount of analytical and petrographic work in the 

 field, and to further investigate the fumarole phenomena as to tem- 

 peratures and flow of gases. A considerable amount of laboratory 

 work will be needed to supplement the observations. It is hoped that 

 the field and laboratory work will throw much additional light on the 

 phenomena of fumarolic and explosive volcanic activity. 



The observations at Kilauea early showed the close connection that 

 existed between the quantity of gas being given off from the lava lake 

 and the temperature and general activity. Efforts were made to 

 collect gases fresh from the lake and uncontaminated with air, a task 

 by no means easy. The unexpectedly large amounts of water met 

 with in the gases collected in 1912 interfered with the determination 

 of the quantitative composition of the gases as a whole, by condensing 

 in the pipes which led the gases out to the collecting-tube. The 

 composition of the "fixed" gases could be determined, but not their 

 relation to the water-vapor. In 1917 gas samples were obtained in 

 which the relation of water to the other constituents could be de- 

 termined. The analysis of these samples has been completed during 

 the past year.^ 



Perhaps the most striking thing about these analyses is the wide 

 variation in the composition of the gas samples, though collected at 

 about the same time and under fairly constant conditions as regards 

 the activity of the crater. It is evident that the gases are far from a 

 state of equilibrium when they issue into the atmosphere ; at the same 

 time the elements are quite extensively oxidized, although variable 

 amounts of unburned hydrogen, sulphur, and carbon monoxide do reach 

 the surface. Water is always present, and in surprisingly large amounts. 



' See abstract (35), below. 



