February 12, 1915] 



SCIENCE 



253 



more in detail the effect of the eruptions of 

 1883, 1902 and 1912 on atmospheric tempera- 

 ture. Only volcanic eruptions of an esplosive 

 character had to be taken into special con- 

 sideration, because it is only when volcanic 

 dust has been projected in great quantity above 

 the ordinary elevation of the cirrus clouds, 

 that this dust could remain in suspension long 

 enough to be spread out all around the globe 

 by the winds of the stratosphere. 



In the case of the famous Krakatoa erup- 

 tion in 1883, the optical phenomena produced 

 by the volcanic dust veil have been observed 

 practically all over the world. The explosion 

 occurred on August 27, 1883. The main sky 

 phenomenon, produced by the dust, went 

 around the world in fifteen days from E. to W. 

 along the equator, spread out N. and S., was 

 observed in the Gulf of Mexico by the end of 

 September and all over the United States in 

 November. Curiously enough, the effect of 

 the Krakatoa dust veil on atmospheric tem- 

 perature seems to have attracted no special 

 attention. 



Besides the Krakatoa, other volcanoes have 

 been very active during the year 1883. St. 

 Augustin and Bogoslof of the Aleutian chain 

 of islands, as well as the Ometepe, may be 

 cited. 



The study of the temperature data of the 

 year 1902 is also of special interest, not only 

 because during that year the world's volcanic 

 activity was greatly intensified, but also be- 

 cause some of the explosive eruptions which 

 occurred undoubtedly produced a dust veil in 

 the higher layers of the atmosphere. 



Already in 1901 the outbursts of Mt. Colima, 

 Mexico, were more frequent and more intense 

 than during the preceding years. The same 

 in 1902 and even more so in 1903. 



On May 7, 1902, La Soufriere, St. Vincent, 

 was in violent eruption. The particular fea- 

 ture of this eruption was the enormous amount 

 of dust which was throvpn into the air and 

 distributed over a vast, somewhat elliptical 

 area. 



' On May 8, 1902, a sea of fire destroyed St. 

 Pierre, Martinique. The following violent 



eruptions of Mt. Pelee occurred on May 20 

 and 26, June 6, July 9 and August 30. 



The influence these eruptions may have had 

 on the thermal transparency of the higher 

 atmospheric layers is questionable. The ex- 

 cellent photographs taken by Lacroix show, 

 indeed, that the occasional blasts of incandes- 

 cent gases and ashes did not exceed an alti- 

 tude of 4,000 m. Only an extremely small 

 proportion of the projected pulverized ashes 

 could have reached the average altitude of the 

 cirrus clouds or even the stratosphere. This 

 may not have been the case in the violent erup- 

 tions of the Santa Maria volcano, in Guate- 

 mala. The eruptions began on October 24, 

 1902. 



The eruption of the Mua, on Sawaii of the 

 Samoa Islands, which occurred October 30, 

 1902, was not violent enough to be taken into 

 consideration. The same may be said about 

 the Isalco eruption in Salvador. On the con- 

 trary, the Tori-shima eruption of August 7 

 and 9, 1902, seems to have been very violent. 



There can be no doubt that during the year 

 1902 a considerable quantity of pulverized lava 

 must have been projected into the higher 

 layers of the atmosphere, above the clouds. 

 Bishop's ring was observed anew, as well as 

 extraordinary twilight phenomena, but a com- 

 parison is hardly possible with those which 

 were due to the Krakatoa eruption. One 

 single volcanic explosion, if sufficiently vio- 

 lent, may therefore obscure the stratosphere 

 very much more than a score of violent erup- 

 tions of a less explosive character. 



This seems to have been the case of the 

 Katmai eruption. Katmai volcano is in the 

 Aleutian range, Alaska, latitude 58° N., longi- 

 tude 155° W., approximately. On the after- 

 noon of June 6, 1912, it suddenly became ex- 

 plosively eruptive, continued in a state of great 

 activity for about three days, and was re- 

 ported to be still somewhat active at the end 

 of October, 1912. 



The fact that the Katmai eruption occurred 

 in a far northern latitude, and has not been 

 followed by similar volcanic outbreaks in 

 other parts of the world, is a most valuable 



