946 



SCIENCE. 



[N. S. Vol. XV. No. 389. 



that terrestrial magnetism and electricity, 

 including the aurora, are directly affected 

 when sunspots appear, and that while the 

 evidence is less simple in the case of the 

 meteorological elements, temperature, pres- 

 sure, humidity and rainfall, there is a strong 

 probability that they too are somewhat af- 

 fected along with the sunspot frequency. 



While admitting the possibility that mag- 

 netic and electrical disturbances are the 

 causes of these meteorological changes atten- 

 tion was devoted only to the often made sug- 

 gestion of a variability of solar radiation as 

 an explanation of the supposed meteorolog- 

 ical effects. Lockyer's views were discussed, 

 and Halm's theory mentioned. It was pointed 

 out that there is a ready way of determining 

 whether changes of transmissibility in the 

 solar atmosphere exist as required by Halm's 

 theory. The great hindrance offered by the 

 earth's atmosphere to direct measures by the 

 actinometer of the variability of solar radia- 

 tion was pointed out, and the variations noted 

 in the results obtained at Montpellier since 

 1S83 were attributed to the influence of water 

 vapor. It was, however, pointed out that 

 spectral actinometry by the aid of the spec- 

 tro-bolometer might be more conclusive. 



Professor C. Abbe presented the next paper. 

 He said Professor A. Wolfer, who now suc- 

 ceeds Professor A. Wolf as Director of the 

 Federal Observatory at Zurich, has lately 

 revised the so-called Tables of Numbers ex- 

 pressing relative sunspot frequency, and has 

 communicated the results of this revision to 

 the Weather Bureau. By incorporating a 

 number of newly discovered observations, 

 especially a long series made at Kremsmun- 

 ster, and by revising all computations so as 

 to eliminate numerical errors. Professor Wol- 

 fer is now able to present a greatly improved 

 table of numbers for each month from 1740 

 to date, and a list of the dates of each maxi- 

 mum and minimum since the days of Galileo. 

 There is no evidence from this table of the 

 thirty-five-year period, but a slight periodicity 

 of fifty-five years is apparent. The intervals 

 from minimum to maximum are always 

 shorter than from maximum to minimum ; that 

 is to say, sunspots increase more rapidly than 



they decrease; moreover the intensity of a 

 maximum is greater in proportion as the rate 

 of formation of sunspots is greater. Professor 

 Newcomb's studies on the sunspot period re- 

 main unaltered by the revision. 



The communication will be published in 

 full in the next number of the Monthly 

 Weather Review. 



The 554th regular meeting was held May 

 24, 1902, Vice-President Marvin presiding. 



Mr. L. A. Bauer called attention to the re- 

 markable magnetic disturbances now occurring, 

 although this is near a sunspot minimum, and 

 to simultaneous disturbances recorded at 

 Cheltenham, Md., in Kansas and in Honolulu 

 on April 18, the date of the Guatamalan earth- 

 quake; and to other disturbances closely coin- 

 cident with the Martinique outburst. A fuller 

 report of this matter will appear in Science. 



The first regular paper was by Mr. G. K. 

 Gilbert, 'On the Mechanism of Volcanoes.' 

 The speaker said the matter to be presented 

 was timely rather than novel. He accepted 

 generally Major Button's views, but illustrat- 

 ed them by various instances from his own 

 observations. The first problem is why the 

 lava comes up : the primary force is gravi- 

 tation, and the column of lava must exert less 

 pressure in the depths than the neighboring 

 solid rocks ; accordingly the heavy, basic lavas, 

 as basalt, in order to become light enough to 

 rise must be highly heated, and are then very 

 fluid; while the lighter, acid lavas rise in a 

 very viscous condition, and flow slowly. The 

 flow ceases because the supply of material 

 lighter than the crust runs out. It is not yet 

 clear how the liquid pierces the crust. Erui>- 

 tions are of three kinds : dry, and then the 

 lava flows quietly out of the crater or fissures ; 

 or wet, and then the occluded water expands 

 into steam as the lava rises, thus forming a 

 porous mass, as pumice, and liberating dense 

 clouds ; or explosive, as at Krakatoa, under 

 conditions not well understood. Considerable 

 discussion followed the paper. 



Professor A. F. Zahm then read a paper on 

 'New Methods of Experimentation in Aerody- 

 namics,' outlining a portion of the researches 

 of Mr. Mattullath and himself at the Catholic 



