254 H. S. WASHINGTON PERSISTENCE OF VENTS AT STROMBOLI 



dance of smoke, have sometimes rendered the exact relative location of 

 the vents somewhat difficult, if not impossible. It is here that the pres- 

 ence of the eastern and 

 western bounding ridges 

 proves to l^e invaluable 

 for orientation. On the 

 other hand, the plans 

 and sketches were made 

 b}^ different observers, 

 and independently of 

 any theory of persist- 

 ence in location of the 

 vents for any great 

 length of time. 



As the number of 

 illustrations is somewhat large, considerations of space necessitate a very 

 brief discussion of each and comparison with its successor, the one pre- 

 viously described, it being understood that the figures are presented in 

 an inverse chronological order. This order is adopted because the more 



0& 





\^ 



© 



Of 



Figure 3, — Plan of the Crater Terrace of StromhoU, 

 August, 1912 (Ferret) 



Figure 4. — Plan of Crater Terrace of Stromholi, May, 1906 (M'egner) 



modern plans are not only more reliable, but present the situations of the 

 vents more clearly than the earlier views, which often leave something to 

 be desired in these respects and can only be regarded as confirmatory of 

 the evidence furnished bv the later illustrations. 



