VOLCANIC ERUPTIONS ON MARTINIQUE AND ST. VINCENT. 347 



Other pJit'Doinenti.-^-X tiiuil report on the recent :uul .still continuiiio- 

 eruptions of jNIont Pelec and Lti Soufriere must include the evidence 

 in ret'orence to the sounds o-enerated, the eartlKpiuke shocks, the areas 

 on which dust fell and its relation to the direction and force of air 

 currents, gravity waves in the air, intiuence of dust in the air on sun- 

 light, and, luost interesting of all, as well as the most novel, the magnetic 

 waves generatiHl, which were recorded almost instantaiu^ously at prac- 

 tically all the automatically recording magnetic stations of the world. 



TJte xfudy of the eartJi.s interior. — Perhaps the chief lesson taught 

 by the recent volcanic eruptions in the Antilles is the meagerness of 

 our knowletlge concerning the interior of the earth. In this more 

 pointedly perhaps than in related tields is the saying true that ''the 

 known is l>ut a small fraction of the unknown.'' In the study of the 

 earth's interior, the search for thc^ ultimate causes of volcanic tM-up- 

 tions, etc., a visit to an active volcano is most instructive and sug- 

 gestive, but such investigations should not end with the cessation of 

 the outbreaks. 



The manifestations wdiicli reach the earth's surface and on which 

 our judgment as to the condition of its interior nuist l)e cliietiy based 

 are movements in the rocks, (earthquakes, escape of heat, magnetic 

 changes, etc. While several of the phenomena referred to become 

 especially prominent during Nolcanic eruptions, they are not confined 

 to such occurrences or to the vicinity of Aolcanic vents, but may be 

 studied at all times and at any locality. Among the records which it 

 is desiral)le to ol)tain and from which some judgment in reference to 

 the condition of the earth's interior may be had are the occurrence of 

 earthquakes, tiieir character, direction of motion, location, l)oth geo- 

 graphical and vertical, of their centers and all else concerning them, 

 and changes in the uiagnetic condition of the earth. Observations in 

 these directions are highly desira])le in th(^ vicinity of volcanoes in 

 order that they may serve as danger signals, but may yield valual)le 

 returns when carried on at a distance from all centers of xolcanic 

 disturbance. 



In this connection 1 wish to suggest that the National Geographic 

 Society can make a substantial addition to our knowledge of the earth 

 by maintaining a magnetic and seismographic observatory. Let a 

 start be made by placing in our Hubbard Memorial Building the best 

 instruments of the nature just referred to that can l)e had, and extend 

 assistance to individuals, colleges, etc., at as many other localities as 

 practicable, in estal)lishing similar observatories. 



