I can offer no explanation which can account for the extreme range in the results at this point, but that the final results are of value may be seen in the fact that the probable error of the final results and of a single observation for the altitude of this point deduced from comparisons with the record at Dodge City is almost the same as the same quantities deduced in the same way for the mouth of Tule Creek, a point where the range in the determination of the horary corrections was only half as great as at this camp. On the 30th and 31st of May great disturbance in the atmospheric pressure is reported at Dodge City. If these two days are omitted, the corrections become very nearly the same as those obtained for the general table, there being only an average difference of +0.005 inch between the two series in individual readings. I did not feel justified in throwing out these two days entirely, and they are incorporated in the work and included in the determination of the altitude.

Abnormal and monthly variations are not given, as the occupation of any one point was not lengthy enough to make any proper deduction therefrom. The barometrical pressure curve at Dodge City is given for the time of the operations of the survey, and also the curve of pressure of the barometers of the survey, reduced to the altitude of Dodge City. In comparing these two curves, differences may arise from the following causes:

1st. Errors of reading the instruments at the two stations.

2d. Imperfections in the instruments themselves.

3d. Atmospheric changes affecting one station and not the other. 4th. Errors in supposed differences of altitudes used in reduction. 5th. Errors in reduction.

Assuming the personal and instrumental errors constant, and that

observations, if sufficiently continuous at two stations, will wholly or partially eliminate the errors from the third source, comparisons between the curves will give us a means of comparing the probable errors arising from the fourth cause, and will give us a correction of more or less value, depending upon the length of time of the observations and the distance between the stations. I have found by this means the following corrections for the deduced altitudes of the principal points:

TABLE OF BAROMETRICAL READINGS AT DODGE CITY, UNCORRECTED FOR ALTITUDE, AND OF THE BAROMETRICAL READINGS TAKEN ON THE LINE OF THE SURVEY,

REDUCED TO THE ALTITUDE OF DODGE CITY.

Table of barometrical readings at Dodge City, &c.-Continued.

Table of barometrical readings at Dodge City, &c.—Continued.