4 H. E. Gregory — A Geologic Beconnaissance 



30 miles apart. As would naturally be expected, there was a 

 decided disagreement between the latitudes as observed and 

 the latitudes as computed by triangulation, starting from any 

 given place, due to local deflections of the plumb line. The 

 eifect was to shorten the radii of curvatures of the meridians 

 crossing this upland and to make all latitudes observed near 

 its southern margin come out too far south and all near the 

 northern margin too far north. Owing to the extremely irreg- 

 ular character of the surface these plumb line deflections can 

 be only roughly estimated, so that it is to be expected that when 

 the Cuzco Valley map is connected to a more extended system 

 of triangulation, an error possibly as great as 4 or 5 seconds 

 may be found in the latitude. Mr. Bumstead reports: 



"The greatest differences between plumb line deflections were 

 found between two stations north and south of the great peak 

 of Salcantay which rises more than 15,000 feet above the canyons 

 on either side. There the difference of latitude by triangulation 

 is 15' 37".l and by observations 16' 13".5, a difference of 36".4, 

 only a small part of which can possibly be accounted for by uncer- 

 tainties in the observations." 



The astronomical latitude of the Cuzco depot as found from 

 meridian altitudes of 66 stars is— 13° 31' 29" 1".4, and the 

 geodetic latitude adopted as described by Mr. Bumstead is — 

 13° 31' 33". 5. 



In 1914 the Peruvian Government and the Harvard Obser- 

 vatory cooperated with the field party in a series of wireless 

 time signals. At about 7 o'clock each night 30 signals were 

 sent out' from the powerful station at San Cristobal at intervals 

 of about ten seconds. These were received in the field and at 

 the Observatory at Arequipa, where the exact times at which 

 the signals were sent were noted by the Observatory clock, cor- 

 rected by frequent observations of stars. Thus by a subsequent 

 comparison of notes the error of the field chronometer was 

 deduced. The success of this method of comparison is shown 

 by an average error probably of only ± s 0.03. 



Immediately after comparing the chronometer, local time 

 was observed by altitudes of east and west stars taken with 

 the transit previously described. 



The following table from Mr. Bumstead's report shows the 

 four values from which the longitude of the Cuzco depot was 

 derived : 



