May 22, 1908] 



SCIENCE 



813 



fully and inter-eompared and the neces- 

 sary corrections on standard instruments 

 frequently obtained. It was also recog- 

 nized that in magnetic survey work one of 

 the chief aims must be to multiply stations 

 in order to eliminate, as effectually as pos- 

 sible, station errors which frequently far 

 exceed observation errors. The time previ- 

 ously spent at a station was reduced from 



work subsequent to 1899 has distinctly ad- 

 vanced. 



Another feature enhancing the value of 

 the present maps is due to the fact that 

 in addition to the observations on land, 

 magnetic work at sea was inaugurated in 

 1903 on the Coast Survey vessels. There 

 were, hence, available the magnetic results 

 obtained by these vessels along our coasts 



two or three days to one day or two days 

 so as to permit doubling or tfebling the 

 number of stations per month. By 

 properly systematizing this work and de- 

 pending for certain corrections upon 

 magnetic observatories established at the 

 same time, it was found that the increased 

 speed in the execution of the work brought 

 about no sacrifice in accuracy. In fact, 

 owing to the care bestowed upon the in- 

 strumental errors as well as on other parts 

 of the work, the general accuracy of the 



during the four years 1903-1907, as also 

 the results obtained by the Carnegie In- 

 stitution of Washington, in Canada, 

 Mexico, Central America, and on the 

 Pacific Ocean. (Slide 3 showed the 

 progress to date of the ocean magnetic 

 work.) 



The present magnetic maps apply to the 

 date January 1, 1905. In order to be able 

 to refer all observations to this date, as 

 accurately as possible, special attention was 

 paid to the secular variations of the mag- 



