Maech 13, 1903.] 



SCIENCE. 



425 



misprint. The statement that ' pumice and 

 bombs prove the existence of a molten magma 

 which rises well into the throat of the vol- 

 cano,' may be questioned, for ancient glassy 

 tuffs and pumice are abundant in the old 

 agglomerates of Martinique, and the bombs 

 are old rocks merely melted on the surface. 

 The estimate of amount of ash sediment dis- 

 charged, based in part on Russell's expression 

 of the cubical content of a steam cloud, is 

 full of fallacy. The argument is as follows: 

 If a single cloud has a capacity of four billion 

 cubic feet, is charged with one per cent, of 

 solid matter, and is regularly replaced every 

 five minutes by another cloud of the same 

 size, the total discharge of solid matter per 

 day is 11,520,000,000 cubic feet. This is one 

 and a half times the discharge of the Missis- 

 sippi River per year, and on this basis the 

 discharge of Pelee is greater than that of all 

 the rivers of the world combined, for the same 

 period of time. This argument is concluded 

 with the question, ' what becomes of the void 

 that is being formed in the interior ? ' The 

 defect in this sort of reasoning lies in the 

 assumiption that a primary eruption is con- 

 tinuous for days or even hours. There have 

 been a few moments of violent outburst at 

 certain intervals, which were undoubtedly ex- 

 plosions from great depth, and may be called 

 primary eruption. Secondary explosion con- 

 tinues for weeks in the intervals, and is occa- 

 sioned by the contact of superficial water and 

 hot deposits. Obviously such explosions are 

 only working over the same material, yet they 

 occasion tremendous puffs that rise many 

 thousand feet, and perfectly simulate deep- 

 seated processes. Professor Heilprin has 

 failed to discriminate primary and secondary 

 eruption when he talks of Mt. Pelee being ' in 

 a condition of forceful activity for upwards 

 of 200 days.' The reviewer questions whether 

 the volcano has been forcefully active from 

 great depths for that many minutes. There 

 have been eight or nine considerable eruptions, 

 and probably none of these lasted more than 

 five or ten minutes. There is no probability 

 of a void in the interior; there is a fissure 

 system, and with the removal of material from 

 the walls, there is probably collaijse that is 



compensated so gradually by subsidence over 

 a wide area, that it makes no appreciable ef- 

 fect even on the height of shore lines. 



As a whole the book is a good exposition in 

 popular style of the main facts connected with 

 the Caribbean eruptions of 1902. There are 

 not sufficient maps to make all geographical 

 matters clear, and there is a lack of diagram- 

 matic illustration, ' much needed to make in- 

 telligible certain explanatory or theoretical 

 statements. The scientific results of Pro- 

 fessor Heilprin's research would be more 

 easily grasped if they were tabulated; he will 

 doubtless compile tables in more technical 

 forms of publication. His summary of the 

 phenomena, and the description of events in 

 August which came under his immediate ob- 

 servation, will stand as records of pernianent 

 value to vulcanology. T. A. J., Jr. 



SCIENTIFIC JOURNALS AND ARTICLES. 



The opening (January) number of Volume 

 4 of the Transactions of the American Mathe- 

 matical Society contains the following papers : 

 ' Ortliocentric properties of the plane ?i-line,' 

 by Frank Morley; 'Definitions of a field by 

 independent postulates,' by L. E. Dickson; 

 ' Definitions of a linear associative algebra 

 by independent postulates,' by L. E. Dickson ; 

 'Two definitions of a commutative group by 

 sets of independent postulates,' by E. V. Hunt- 

 ington ; ' Definitions of a field (Korpev) by 

 sets of independent postulates,' by E. V. 

 Huntington ; ' On the invariants of differential 

 forms of degree higher than two,' by C. N. 

 Haskins ; ' Uber die Reducibilitat der Gruppen 

 linearer homogener Substitutionen,' by Alfred 

 Loewy ; ' The quartic curve as related to 

 conies,' by A. B. Coble ; ' The cogredient and 

 digredient theories of multiple binary forms,' 

 by Edward Kasner ; ' On the envelopes of the 

 axes of a system of conies passing through 

 three points,' by R. E. Allardice ; ' A Jordan 

 curve of positive area,' by W. E. Osgood. 



The December number of the Bulletin of 

 the American Mathematical Society con- 

 tains : ' Concerning the commutator subgroups 

 of groups whose orders are powers of primes,' 

 by W. B. Eite; 'Note on irregular determin- 

 ants,' by L. I. Hewes ; ' Note on the projec- 



