70 ANNUAL RECORD OF SCIENCE AND INDUSTRY. 



than this figure. The depths at which the daily and annual 

 variations respectively disappear are in the proportion of 1 

 to 19; or as the square roots of the lengths of the periods. 

 The amplitudes of the daily, annual, or other periodical 

 changes diminish in a geometrical ratio as the depths in- 

 crease arithmetically. The time required for the heat wave 

 to reach a depth of 7.3 meters at Edinburgh is nearly six 

 months. Therefore the maximum temperature at this depth 

 is synchronous with the minimum temperature at the sur- 

 face. Mr. Schott computes, from the best data available, 

 what should be the average invariable temperature at the 

 depth of 229 meters underneath Chicago, corresponding to 

 the depth from which water flows in a certain artesian well. 

 The actual temperature observed is only 12.8. The com- 

 puted tem2)erature is 16.2. The explanation of the discrep- 

 ancy can not be satisfactorily given, unless observations are 

 made upon the influence of the lake water, and the actual 

 rate of increase of temperature at that spot. Report Sec. 

 Smithsonian Institution^ 1874, 249. 



FOEMATIOX OF BASALT COLUMISrS. 



Mr. Robert Mallet republishes in full his views on the 

 origin and mechanism of the production of the prismatic or 

 columnar structure of basalt. He states that our knowledo'e 

 of the mechanism of the jointing of masses of rock is still 

 very far from being complete, but less complete is our knowl- 

 edge of that which constitutes the complicated systems of 

 joints shown in columnar basalt. The best-known writers 

 are generally at variance on this subject, although the most 

 philosophical views are those that have been defended by 

 Professor James Thomson, now of the University of Glas- 

 gow. According to Mr. Mallet's investigations, the primary 

 conditions for the production of straight prismatic structure 

 in basalt are, first, general homogeneity of the material; sec- 

 ond, a tolerable regularity in the general form of the mass ; 

 third, inequality in the rate of cooling in one or more direc- 

 tions. The more or less horizontal and tabular masses of 

 basalt as they occur in nature are usually cooled most rapid- 

 ly from their upper or lower faces ; or, if the basalt be found 

 filling a dike, then the rapid cooling has taken place on 

 the two opposite surfaces of the dike. When, in conse- 



