E. Loomis— Observations of the U. S. Signal Service. 98 
than that given in the above table, which change would not 
materially affect the conclusions which I have drawn from the 
observations. 
now we attempt to represent by isobaric lines all the 
observations at the Signal Service stations for the month of 
July (employing for the mountain stations the values given in 
the last column of the preceding table), we find that all the 
observations are pretty well represented except those at the 
four stations above named, viz: Virginia City, Santa Fé, 
North Platte and Dodge City. 
If we assume the height of Virginia City above sea-level to 
be 5,694 feet, the corrected pressure becomes 29°74, which 
accords very well with the observations at the other stations. 
If we assume the height of Santa Fé above sea-level to be 7,000 
feet, the corrected pressure becomes 29°84, which also accords 
very well with the other observations. 
The results at North Platte and Dodge City appear to be 
about 0°30 inch too small, which is apparently due to an 
erroneous mode of reducing the observations to sea-level, as 
intimated on page 91. 
then drew the isobars which best represent all the Signal 
Service observations for the month of July, including the four 
Stations above named with the corrections which have been 
indicated. These curves exhibit an area of low pressure for 
the central part of the American continent similar to that which 
prevails in summer over nearly the whole of Asia, but far infe- 
rior in amount. 
was adopted for January for stations whose altitude is less 
than 1,200 feet. For stations whose altitude is greater than 
1,200 feet, the temperature at sea-level was determined in the 
manner described on page 90. The reduction to sea-level was 
then computed from the data thus furnished by using Dun- 
