Chemistry and Physics. 249 



example, England, such large annual changes and even greater 

 ones occur. Values of the magnetic declination secured four to 

 five decades ago, if referred to the present time with the aid of 

 secular change values as were expected, in accordance with the 

 experience in other parts of the United States, would be in error 

 one-third to one-half degree. 



Observations in other States near Louisiana show that this 

 change in the course of the secular variation is manifesting itself 

 in greater or less degree, according to locality, likewise in these 

 States. How permanent the present change may be, i. e., whether 

 it will continue for but a comparatively short period, so that 

 before long another reversal may be expected, after which east 

 declination will begin to diminish once more, can not be stated at 

 present. 



The magnetic survey of Louisiana has revealed other most 

 interesting features, as shown by the lines of equal magnetic 

 declination, dip, and horizontal intensity, drawn in conformity 

 with the observations. There were noticed marked relations with 

 well-known physiographic features. The curvatures and bend- 

 ings of the lines of equal magnetic declination appear to con- 

 form with courses of principal steams and shore lines of certain 

 bodies of water. Also a marked difference manifested itself in 

 the general direction of these lines in the middle of the southern 

 part of the State, just where there is a dividing line between the 

 newer and older geological formations. 



It is especially interesting that the irregularities in the distri- 

 bution of the earth's magnetism, as exhibited by the three sets of 

 lines, occur chiefly in the regions of the alluvial deposits brought 

 down by the Missisippi River. Owing to these irregularities the 

 compass needle is deflected from the direction it would ordinarily 

 have assumed by amounts varying from about o, l to 0°*5. They 

 are not local disturbances of such amount which ordinary instru- 

 ments would readily reveal, but they are of such a magnitude as 

 only approved instruments and methods would indisputably expose. 

 This point can not be emphasized too strongly for the sake of 

 geologists who undertake to discover relations between magnetic 

 disturbances and geological formations, employing crude instru- 

 ments, and using imperfect methods. 



Quite likely these irregularities are to be referred to small local 

 deposits of iron ore brought down from the upper States by the 

 Mississippi River." 



16. An Introduction to the Study of Spectrum Analysis ; by 

 W. Marshall Watts. Pp. vii, 325, 8vo. London and New 

 York, 1904 (Longmans, Green & Co.). — The first impression from 

 a hasty inspection of this work is the extraordinary variation in 

 the character and value of the numerous illustrations contained 

 in it. We recognize some cuts which are familiar in popular 

 works on spectroscopy from as early a date as 1870, and which 

 might have been omitted even then with profit to the reader. A 

 number of those illustrating the very primitive treatment of the 



