326 Johji Casper Branner 



floating one hundred feet was noted. Wooden rods twelve 

 feet in length were used as floats. The,se rods were two inches 

 square at one end, from which they tapered the whole length of 

 the rod to a sharp point at the other. They were weighted so 

 as to float upright and to leave the pointed end about two feet 

 out of the water to serve as a signal. Where the rods could 

 not be used on account of shallow water, a surface float with a 

 weight attached by a cord was substituted. From the data 

 thus obtained the volume of the river was deduced. Sets of 

 samples of the water were taken along the cross-section at the 

 time of the velocity observation, each set being in three parts, 

 one each from the surface, mid-depth, and three feet from the 

 bottom. In collecting the sample from the bottom, in order 

 to avoid taking it from the liquid mud usually present next to 

 the bottom, the collecting apparatus was so arranged that the 

 sample was taken three feet from the actual bed of the stream. 

 In order to avoid the possible mingling of the water from low- 

 er depths with that above, and to insure that the samples fairly 

 represented the part of the stream from which it was taken, 

 an open glass tube holding one liter was used for a collecting 

 vessel. This was so arranged as to close securely b}' means 

 of two rubber balls. When a sample was to be taken, the 

 stoppers were caught back, leaving the ends of the tube en- 

 tirely unobstructed ; the tube was then sunk by means of a 

 rod, care being taken to keep its axis parallel with the current 

 of the stream. By means of a gage the depth to which it 

 was desired to sink it was determined. When the vessel 

 reached the desired point, a jerk of the string released the 

 rubber balls, which closed the ends of the tube and confined 

 in it a representative of the part of the stream from which it 

 was taken. The samples were always taken at the time the 

 volume of the stream was being measured. They were placed 

 in separate, clean bottles for examination. 



In order to determine the amount of matter carried in me- 

 chanical suspension these samples were all taken to the labo- 

 ratory and filtered until the water was perfectly clear ; the fil- 

 ter containing the suspended matter was then dried, and 

 weighed at the temperature at which it had previousl}' been 



