382 
Nebraska to latitude 384° in southern Kan- 
sas, a distance of about 300 miles along the 
meridian. Ninebases have been measured 
along this are in addition to the Salina Base 
on the 39th parallel arc, which serves also 
to control the lengths on a portion of the 
98th meridian are. But four more bases 
are necessary for this arc, one at the Rio 
Grande and three in the Dakotas. 
During the year 1900 two triangulation 
parties were in operation, one working 
northward in Nebraska and the other 
southward in Kansas. A base party of 
ten officers and men, which arrived on the 
working ground on July 16, 1900, in Ne- 
braska, had by January 23, 1901, standard- 
ized the base apparatus twice, at the begin- 
ning and end of the season, and had meas- 
ured the nine primary bases referred to 
above. The probable error of each base is 
less than one part in a million. If this 
feat of measuring nine bases as well as 
standardizing the apparatus in but little 
more than six months, while holding the 
accuracy up to the best standards of the 
past, is considered with reference to the 
moderate size of the party and the time 
which has been required for former pri- 
mary base measurements, it will be seen 
that there has been no lack of progressive- 
ness along this line. 
One of the events of the year has been 
the connection of the gravity measures in 
the United States with those in Europe by 
swinging a set of the half-second pendulums, 
which serve to determine the relative 
values of gravity, at the base station at 
Washington and at the European stations 
at London, Paris and Potsdam, at which 
the more important European absolute 
measures have been made. The result of 
this expedition is to import, at a very small 
cost in time and money, the expensive and 
laborious determinations of the absolute 
value of gravity which have been made in 
Europe. 
SCIENCE. 
[N.S. Vou. XIII. No. 323. 
During the last two years the instru- 
ment and the methods used in precise level- 
ing of the Coast and Geodetic Survey have 
been radically changed with a view to in- 
creasing both the accuracy and the rapidity 
of the work.* The evidence as to the accu- 
racy of the new work is rapidly accumulat- 
ing and so far fully justifies the changes 
made. The lines of leveling are being 
rapidly extended, a total of 1750 miles 
having been run during the last two years. 
The most marked progress, however, in 
the matter of leveling has been the adjust- 
ment of the level net covering the eastern 
half of the United States. More than 
13,000 miles of precise leveling had been 
run by various organizations in the United 
States. But until within a year the results 
had not been correlated. To obtain the 
results it was necessary to search through 
scores of volumes, and even when this had 
been done it was found that the results had 
been published as if each line or group of 
lines was entirely independent of the others, 
whereas in fact the connections existed for 
treating the whole as a sing!e net upon one 
basis. The adjustment of this net has now 
been made. The elevations and descrip- 
tions of the four thousand permanent 
bench marks connected with the net, and 
the principal items of information in regard 
to each of the lines, have been published in 
a single volume.} 
During the year the report upon the 
transcontinental triangulation, which marks 
an epoch in the history of geodesy in the 
United States, has been published.{ The 
computation of the eastern oblique arc, ex- 
tending from Maine to Louisiana has been 
* See Proceedings of the American Society of Civil 
Engineers, November, 1900, pp. 1113-1161. 
+See Appendix 8 of the Coast and Geodetic Sur- 
vey Report for 1898-99, pp. 347-886. 
{The Transcontinental Triangulation, Special 
Publication No 4, of the Coast and Geodetic Survey, 
Washington, 1900, 4to, 871 pages. 
