286 



SCIENCE 



[N. S. Vol. XXVIII. No. 713 



a marked increase in the lava output on Feb- 

 ruary 2 (B) — chart for 1906 — and an ex- 

 plosive maximum on February 7 (G), for an 

 account of which see the New York Sunday 

 Herald, April 1, 1906. The great Vesuvius 

 eruption culminated on April 8 (G) and the 

 reader is now referred to Fig. 3 which repre- 

 sents Mercalli's dynamic curve of the eruption 

 — D. C. — to which I have added the astronom- 

 ical data and the luni-solar and terrestrial 

 maxima curve extended horizontally to fit the 

 scale of Mercalli's diagram. Note how well 

 the progressive reactionary process, of which 

 we have spoken before, is exemplified in the 

 dynamic curve from April 4, working up 

 with increasing amplitude of vibration to the 

 great culmination on the eighth, and note the 

 general correspondence with the maxima curve 

 throughout the entire eruption. 



Assuming that we had had this luni-solar 

 and terrestrial maxima curve at the beginning 

 of 1906, it may be well to ask ourselves here 

 if we could have predicted the eruption? 

 Knowing the potential condition of Vesuvius 

 at the time, we should probably have expected 

 the eruption during the maxima combination 

 (C) preceded by (B), but, that failing to be 

 the crisis (although the activity was then 

 very great), I thinlv that we could have 

 predicted it for April 8 (G), preceded as 

 this was by a very strong terrestrial maxi- 

 mum. I have before pointed out that the 

 length of the line joining two different 

 maxima is a measure of the influence. In any 

 event, we could have been morally certain that 

 the eruption would have been precipitated by 

 the strong maxima which follow — in other 

 words, that the summer would not pass with- 

 out a great eruption. 



The San Francisco earthquake of April 18 

 showed the normal trifling lag behind the ter- 

 restrial maximum (H), but at Formosa an 

 earthquake occurred on April 14, and another 

 on March lY (F). Other severe earthquakes 

 occurred during the year in Iceland, Novem- 

 ber 8, 9 (N) and November 22 (0); Kopal 

 December 22 (P) ; Sicily September 13 (Jf ) ; 

 Calabria January 17 (A) and June 10 (/) ; 

 Central Asia August 13 (L) ; India March 10 



(E) and July 21 (Z) ; Columbia January 31 

 (B); West Indies February 21 (D) ; and 

 Saxony April 28 (7). 



On the curve for 1907 the five explosive 

 crises of the gi-eat eruption of Stromboli are 

 shown at (D, E, F, G, H), for a fuU account 

 of which, with dynamic curve, the reader is 

 referred to the Broohlyn Institute Museum, 

 Science Bulletin, Vol. 1, No. 7. Mauna Loa 

 was in eruption January 10 (4) and Jaggar 

 reports a violent disruptive explosion of 

 Bogosloff on September 1 (7). Severe earth- 

 quakes occurred at Jamaica, January 14 (4), 

 and March 25 {B) ; in Mexico, April 14 (7?) ; 

 Bitlis, March 31 (C) ; San Miguel (Azores) 

 April 2 (C) ; and in Calabria, October 23 (J). 



In order to look backward a little I have 

 selected 1902 as being rich in volcanic mani- 

 festations and have prepared a curve for that 

 year. The great explosive crises of Pelee oc- 

 curred on May 8, June 6, July 9, August 30 

 and December 16. Professor Lacroix writes 

 (" La montag-ne Pelee et ses eruptions ") " Le 

 8 Mai s'est produit le phenomene terrifiant 

 qui, en quelques minutes, et peut-etres moins, 

 a aneanti S. Pierre et ses 28,000 habitants" 

 (7»). Nuees ardentes 20 Mai {E) et 6 Juin 

 (F). "Une brusque recrudescence paroxys- 

 male le 9 Juillet " (G). "Le 24 Aout une 

 secousse de tremblement de terre est ressentie 

 dans toute I'ile " (5). " Un grand paroxysme 

 se prepare, il eclate le 30 Aout" (7). Nuee 

 ardente 16 Decembre {M). 



At St. Vincent: "A la fin d'Avril 1902 ils 

 (earthquakes) augmenterent, . . . le 29 Avril 

 il ne se produisit pas moins de 18 secousses 

 au Morne Rouge (G)." " Le 7 Mai eut lieu 

 la grande explosion (7))." " TJn nouveau 

 paroxysme s'est produit du 15-16 Octobre" 



(i). 



The Pelee paroxysm of July 9 does not cor- 

 respond with the curve. The writer would 

 call special attention to this, the only notable 

 exception in the entire series of events, in the 

 hope that some explanatory observation may 

 be forthcoming which shall give greater pre- 

 cision to the making of future curves. 



At Izalco (Salvador) there was a violent 

 eruption September 29 (K). 



