20 BULLETIN 376, U. S. DEPARTMENT OF AGRICULTURE. 
ATTACHMENT OF PIEZOMETERS. 
As it is not often practicable to secure permission to make several 
holes at each, manometer in a wood pipe used commercially, the 
writer accepted the discussion of Profs. Marx, Wing, and Hoskins con- 
cerning the position of the point of attachment to the pipe and whether 
different results will be given by multiple attachment than by attach- 
ment at a single point. 1 Their first conclusion is : 
When the pressure in the given cross section of the pipe everywhere exceeds that 
of the atmosphere an open piezometer will stand at the same height at whatever point 
of the cross section it be attached, and whether it communicates with the pipe at 
one point or at several. 
As a rule taps were not made on the top of the pipe, as the writer 
judged that more air bubbles would be in the water at this part of 
the pipe than at some lower point. Care was exercised in choosing 
the reach, so that guages could be set at each end of it, where the 
positions of the two taps would be such that the same relationship to 
velocity would hold. All taps were made on tangents. The posi- 
tion on the circumference was chosen in the neutral zone vv'here the 
influence of curves would be a minimum. For instance, if the pipe 
was straight in horizontal alignment, but curved vertically, the taps 
were made in the side of the pipe. In experiments on pipe No. 52, 
where the pipe followed a chosen gradient but was curved horizon- 
tally, taps were made near the top of the pipe. (PL V, fig. 3.) 
The essential requirement in a piezometer connection is to exclude 
all positive or negative influence of velocity head. The hole through 
the pipe must be normal to the pipe and as clean cut as possible on 
the inside. If splinters are pushed off the inner surface, then either 
positive or negative influence from the velocity head must act on 
the column in the gauge, whereas the pressure head alone is desired. 
The gauges were attached to the pipe in a manner slightly modi- 
fied from that used by Noble 2 and later by Moritz. 3 They bored a 
hole for the nipple with a wood bit until the tip of the bit pierced 
the inner surface. In the experiments described in this paper a 
seven-sixteenth-inch wood bit was used to make a hole about 1 
inch deep. Then a twist drill one-eighth inch in diameter was 
twisted by hand until the inner surface of the pipe was cleanly 
pierced (fig. 3) . Experiments made with both systems showed the 
holes made by the last method to be more nearly free from splinters 
which might affect the gauge tube by velocity head. 
i Trans. Amer. Soc. Civ. Engin., 40 (1S98), p. 526. 
2 Id., 49 (1902), p. 119. 
3 Id., 74 (1911), p. 411. 
