10 H. J. Harman— Operations for obtaining [No. 1, 
of the portion. After velocities have been once assumed they have not 
been reconsidered or altered. 
In the case of the reduction of the observed discharge of the Dibang 
river, to its mean low-level discharge of the dry season, the rise of water 
was taken as 4 feet, but subsequently it was taken as 5 feet. 
The depths along the section of the Brahmaputra above Sadiya were 
all obtained on the same day. ‘The velocity observations at some of the 
stations were made when the river had risen 1 foot to 14 feet: reductions 
were therefore made to the observed velocities at these stations, 
(ec) The columns in the Synopsis Table, headed “ Mean Velocity of 
river at its mean low-level of the year,” and at “high flood level,” do 
not contain much informatjon unless they are compared with the sectional 
areas, and form of section. 
The sum of the discharges of the rivers forming the Brahmaputra river 
should equal the discharge of the Brahmaputra itself. The work done ful- 
fils this test sufficiently nearly to allow, I think, of the stated discharges 
being considered as sufficiently approximate for geographical purposes. 
The Subansiri measurements were made between the 25th and 28th of 
February 1878. The weather was fine for the work, the level of water 
constant ; very little wind disturbed the observations. The boatmen and 
others living at the site of the section stated that the water was very rarely 
seen at a lower level, and that during the observations it was at the dead-low 
of the dry season. 
Ninety soundings were made to obtain the section of the stream; at 
21 stations observations were taken for velocity and there were 510 record- 
ed passages of the floating instruments. The resulting discharge at lowest 
level of the year was found to be 16,945 cubic feet per second: the super- 
ficial area of the section 9,637 superficial feet, so that the mean velocity of 
the water was 1°7 feet per second. 
During the very high floods of the year the water is known to rise 22 
feet, but calculations have been made for ordinary high flood, taken at 16 
feet ; with this rise of water I have computed the discharge to be at least 
170,000 cubit feet per second ; area of section is 35,700 superficial feet, so 
that mean flood velocity is 4°7 feet per second. 
For extreme floods, 18,000 superficial feet must be added to the section 
area and 70,000 cubic feet to the discharge, which would bring the maxi- 
mum discharge up to 240,000 cubic feet per second, and sectional area to 
48,700 superficial feet. 
The site of the section was at Pathalipam village, 3 miles west of 
Gogah-muk ferry, and about 10 miles from the foot of the hills, from 
the gorge where the river issues out into the plains. 
Before issuing into the plains, the river flows gently for 9 miles in a 
hha! i ip 
