120 ' BULLETIN OF THE 



becomes more marked. If the slit aud the Nicol be interchanged^ 

 the same results follow in the same order. {See page 148.) 



The experiment may be varied in a striking manner by substitu- 

 ing a double-image prism for the Nicol, when the two images may 

 be compared side by side. The experiments are somewhat trying 

 on account of the faintness of the light. The principal conditions 

 to be fulfilled are as follows : (1.) The sun is to be viewed through 

 the slit, the latter being placed as near as possible to the eye. (2.) 

 The double-image prism should be used in order to compare the 

 twO' images side by side. (8.) The slit should be from .01°"" to 

 .001°"° in width. (4.) The slit should be as nearly perfect aa 

 possible. 



The^e experiments seem to show, 1st, that light which passes 

 through a very narrow slit is polarized in a plane at right angles 

 to the slit ; 2d, that the shorter waves pass through more readily 

 than the longer. 



On this communication Mr. Newcomb remarked that there was 

 a temptation here to inquire whether the phenomena recited might 

 not have a bearing upon the unsolved problem of the amplitudes 

 of luminous vibrations. Mr. W. B. Taylor entertained the view 

 that these amplitudes were so small that even the narrow slit used 

 by Mr. Michelsou would bear too large a ratio to them to exercise 

 any appreciable affect. No explanation of the phenomena waa 

 suggested. 



Mr. H. M. Paul then remarked on 



EARTH TREMORS AS SHOWN BY ASTRONOMICAL OBSERVATIONS. 



[Abstract.] 



The observations referred to were made with a view of deter- 

 mining to how great an extent certain localities in the vicinity of 

 Washington city were liable to tremors during the passage of railway 

 trains, and they were undertaken in connection with the question 

 of the selection of a new site for the Naval Observatory. The mode 

 of inquiry was by observing with a 3;} inch telescope the effect of 

 such tremors upon a basin of mercury in a manner similar to 

 observations for determining the meridian alignment of transit 

 telescopes. In these cases the pole star was used. The magnifying 

 power was about 185. The mercury well was 14 inches by 10 

 inches, and the mercury amalgamated with tin, to within about two- 



