RECENT PROGRESS IN GEODESY. 267 
\ 
observations either in a desert or in a country where the 
comforts of life are unattainable. Still another condition 
must not be forgotten. In a country where earthquakes are 
frequent we may have sudden changes in the direction of 
the plumb line. Observations which are to bring out varia¬ 
tions in the latitude, whether they be continuous and com¬ 
paratively rapid, or of long duration, must be free from local 
disturbances of this kind; and then the weather conditions 
must be thought of. At some places one would not get three 
good nights in a month. No latitude station would be pos¬ 
sible in such a neighborhood. The location of the stations 
is to be decided therefore in this way. A single pair of stars 
observed at three stations gives the required data. The posi¬ 
tion of the pole is determined from two coordinates, and on 
the position of the station as regards longitude depends the 
precision of the determined quantities. The weight of the 
result in a mathematical sense is a function of the longitude 
of the stations, and if we have any number of stations on 
the same parallel, we may estimate in numerical values 
the precision of the determination of the position of the 
pole. Any other parallel and any other combination of sta¬ 
tions can be similarly expressed, and our task is simply to 
find from all the different possible combinations of stations 
on separate parallels that one which gives a maximum value 
for the precision. With this end in view, no less than eight¬ 
een combinations, of which fifteen are in the northern hem¬ 
isphere, have been examined by Professor Albrecht. From 
a mathematical point of view only nine were acceptable, and 
from these nine it was necessary to choose four which were 
physically practicable and which at the same time offered 
the necessary social conditions. These stations have not yet 
been definitely decided upon, but they will presumably be on 
the parallel of latitude 39° 8', and, besides, two will almost 
certainly be in the United States. The others will be, one in 
Japan and the other in Italy. Of the American stations six 
have been critically examined, viz., Ukiah, in California; 
Dover, in Delaware ; Gaithersburg and Annapolis Junction, 
in Maryland, and Round Hill and Leesburg, in Virginia. 
39—Bull. Phil. Soe., Wash., Vol. 13. 
