ABSTRACTS 



Authors of scientific papers are requested to see that abstracts, preferably 

 prepared and signed by themselves, are forwarded promptly to the editors. Each 

 of the scientific bureaus in Washington has a representative authorized to for- 

 ward such material to this journal, and abstracts of official publications should 

 be transmitted through the representative of the bureau in which they originate. 

 The abstracts should conform in length and general style to those appearing in 

 this issue. 



GEODESY. — Fourth general adjustment of the precise level net in the 

 United States and the resulting standard elevations. William 

 Bowie and H. G. Avers. Special Publication No. 18, U. S. 

 Coast and Geodetic Survey. Pp. 328, 4 pis., 1 map. 1914. 



This publication contains the results of all precise leveling, done 

 in the United States before the year 1912, which formed closed loops. 

 Several spur lines were not included. The elevations are referred 

 to mean sea level as the datum. The mean level of the sea at all 

 points on the open coasts is supposed to define a level or equipotential 

 surface. Deviations from this condition, if they exist, are not great 

 enough to be detected by leveling of even the highest quality. When 

 the mean sea levels at two tidal stations are connected the discrepancy 

 is of such an order that it may be due to the accumulated errors of 

 leveling. 



The great advantage of having mean sea level as the datum for 

 leveling is that the field work may start from many points; valuable 

 checks on the results are obtained; and the leveling of neighboring 

 countries may be joined to form a continuous net over a whole continent. 

 These conditions would not be possible if each country adopted an 

 arbitrary datum. 



The lines of precise leveling are so interlaced that they form a com- 

 plicated network, and the elevation of an interior point can be ob- 

 tained by a number of different routes from the sea coast. The whole 

 net was made consistent by an adjustment by the method of least 

 squares, which furnished the most probable values for each of the 

 points common to two or more lines. The elevations of the interme- 

 diate bench marks were then made to agree with the adjusted values of 

 these junction points. 



The present precise level net is considered to be of such strength and 

 extent that the elevations resulting from this fourth adjustment can 



255 



