THE FLOW OF WATER IN WOOD-STAVE PIPE. 
91 
to flow, into the particular materials of which the pipes are constructed, except in 
sharply limited cases where the material is a meas- 
ure of the roughness. 
The writer considers the author's formula as good 
and as safe as any, if its limitations as shown by 
column 19, Table 2, be taken into account and if 
the velocity in the pipe under design is not much 
greater than in the data presented in Table 2. 
This suggests the difficulty in getting the engineer- 
ing public to recognize the average quality of an 
averaging formula. Thus the significance of the 
formulas on page 309, Trans. Amer. Soc. Civ. Engin., 
vol. 51 (1903), is very liable to be overlooked when 
presented in the form given on page 281, Eng. Rec, 
September 3, 1904. 
Unless these general interrelations of all pipe floiu 
data are specifically emphasized, it seems that a 
good opportunity will be lost. The author states 
on page 50, "However, in deriving the new for- 
mula, tests made on round, wood-stave pipe only 
were considered, in view of the proposed use of such 
a formula. " The inference that data on other pipes 
are not relevant will be drawn from this by the 
great body of readers. Thus the error so long per- 
petuated by hydraulic texts will be given apparent 
sanction. 
From a purely selfish viewpoint the writer would 
not be eager to see the author's paper broadened as 
suggested. It furnishes data to drive home matters 
about which there have been great uncertainty in 
the minds of a few, and general ignorance in the 
minds of the many who place a halo about text- 
book formulas. If the author does not, others will 
do it. 
The value of the paper as it stands is great. It is 
in the hope that the author may be persuaded to 
make the shift in viewpoint necessary to add much 
to the permanent value of his deductions that the 
writer ventures these comments. 
In The Cornell Civil Engineer of December, 1911, 
page 127, there are some data that may be found of 
interest in connection with the author's method of 
finding velocities by injecting a colored solution into 
the pipes and timing the interval from the instant 
of injection to midway between the first and last 
appearances of color downstream. 
Prof.Hoskins: The experimental work described 
in "The Flow of Water in Wood-Stave Pipe" is a 
valuable addition to knowledge on the subject, and 
the author's summary of conclusions appears to 
merit confidence as embodying present knowledge 
on the hydraulics of wood-stave pipe. The expo- ! 
nential formula adopted as the result of the discussion is probably as reliable a 
