“> 
o 1 er: 
1879.] Discharge of large Rivers in Assam. 9 
(2) The arithmetical mean of the times noted at a station of the pas- 
sages of a certain floating instrument, was taken as the mean time of pas- 
sage of that instrument. The record of times was kept to half seconds, 
and it is rarely that discrepancies of over 1 second from the mean occur. 
These discrepancies were due, first to errors of observation and 2ndly to 
considerable divergence of the floating instrument from the general line taken. 
Having obtained at a station the mean time of passage of each kind of 
instrument, and having regard to the number and quality of the observa- 
tions, a value was assumed as representing the mean time of passage of the 
water at that station. 
For the adjacent station, a mean time of passage was obtained in a 
similar manner. Then for the whole included section, a value was assumed 
as representing the mean time of passage, and this value was employed in 
the computation. 
In such a case as the following: at station (a) the velocity of the 
water is high, and the swift water extends nearly to the adjacent station c, 
and shows a well defined margin-line of the current. At station ¢ the 
velocity is low. 
In such cases the section from station (a) to station (c) is sub-divided 
into 2 parts ; the swift water portion is dealt with by itself, the remaining 
part by itself. 
The method adopted of obtaining the data for computation gives, I 
think, the results required, and without much labour. 
(ec) The Flood Discharges are merely probable values. They were 
obtained in the following way. The section lines were laid down on the 
Revenue Survey large-scale maps of the rivers. An examination was made 
on the ground to see what changes had taken place since the maps were 
made. All vertical heights on these flood sections (shown on charts of 
river sections) were estimated. The flood section was then divided into 
portions, and an assumed velocity was assigned to each portion, the value 
being chiefly determined by the section of the portion, its position, and from 
what local information I could gather. The main-stream high flood velo- 
city has been taken at 7 feet per second. To obtain the total Flood Dis- 
charge it was necessary to increase the velocities of each portion of the 
dry season section. After once assuming values for the flood velocities of 
the different portions, no alterations were made subsequently. 
(d) Some of the rivers measured, were above the mean low-level of 
the dry season. To obtain the discharge of the river at its mean low-level 
of the dry season, the following method was adopted : the area of each por- 
tion of the section, due to rise, was computed and subtracted from the 
observed area of that portion ; then, to the diminished area of the portion, 
a mean velocity was given which was less than the observed mean velocity 
2 
PO OS 
