proceeded so that, any inconsistencies were found at once. This practice, 
however, was not always possible for the observers, with the result that 
such inconsistencies were sometimes not discovered until the measure¬ 
ments were worked up at the office and when it was impossible to return 
to check or verify them. After a few years more pains were taken to take 
the stream measurements at the same place so as to be able to compare 
the results in the various sections from year to year. Also it was attempt¬ 
ed to measure streams even though known to be seepage. These have 
definite characteristics that almost always enabled them to be recognized. 
They are indicated in the tables by parenthesis and are not included in 
the summations. 
The cause of discrepancies in different years is not always evident. It 
is to be expected that the results will disagree from year to year, but some¬ 
times the indications are that errors have entered, either from errors in 
measurement, from fluctuation in the stream or variations in the ditches. 
A considerable difficulty was found with the first measurement on a 
stream, to obtain the necessary detailed local knowledge. Sometimes the 
water commissioners themselves did not know the location of the head- 
gates or their wasteways and thus information was not reliable. If an 
error were committed in the measurement of a stream at any place, it 
would show too large a gain in one section and too small a gain in another, 
but the aggregate of the two would not be affected. While most of the 
measurements were made in the late fall, when storms were not frequent, 
yet sometimes there was interference from this cause. It was usually im¬ 
possible to wait until such unfavorable conditions passed away. The de¬ 
tailed measurements are therefore given complete. 
After the first few years, measurements were extended to the other 
tributaries of the Platte, and to the Arkansas, the Rio Grande and the 
Uncompahgre. 
Acknowledgements are due to the many who have aided in these mea¬ 
surements. In the earlier ones various ditch companies on the Platte and 
the Arkansas especially gave their aid. The water commissioners nearly 
always were glad to help for the information that aided them in the per¬ 
formance of their duties. Later, with more funds available, some allow¬ 
ance could be made for their time and rigs. The one who above all else 
has aided in the field work and in the reduction, has been Mr. Robert E. 
Trimble, of the Experiment Station, who began as my assistant in 1891 and 
continued throughout this series of nearly twenty years. To him fell 
nearly all of the field work. To his conscientious care the accuracy in 
detail is mostly due. 
ORDER OF ARRANGEMENT. 
The following is the order of arrangement of the measurements as 
given in Parts II. and III.: 
PART II. 
Cache La Poudre. 
PART III. 
Big Thompson. 
Little Thompson. 
St. Vrain. 
Left Hand. 
Boulder. 
Clear Creek. 
Ralston. 
South Platte. 
Arkansas. 
Fountain. 
Rio Grande. 
Conejos. 
Uncompahgre. 
