1122 
Journal of Agricultural Research 
Vol. V, No. 94 
needle and further rejections made of those which vary more than one point from 
that obtained with the standard needle, on a sample having a penetration of approxi¬ 
mately 60. 
The committee further stated that it did not think it advisable to rec¬ 
ommend at the present time a standard needle for reference, deferring 
such action until the next annual meeting of the society. Until such 
recommendation is made needles furnished with penetration machines 
are to be considered standard. 
The above-recommended practice is representative of the method for 
selecting needles which has been followed in the Office of Public Roads 
and Rural Engineering, and the standard used for comparison and selec¬ 
tion was a needle originally supplied with the penetration machine in 
use. It has been, however, not uncommon practice in certain labora¬ 
tories to purchase a package of No. 2 needles and to use them on the 
assumption that they possessed the requisite dimensions and shape. In 
an effort to prove the fallacy of such an assumption the authors have 
taken an enlarged photograph of a package of Roberts's No. 2 needles, an 
examination of which will serve to make clear the ordinary variations 
in point, shape, and taper (PI. LXXXII, fig. 1). 
These variations are more clearly shown through a consideration of 
the results obtained in selecting needles to be used for routine testing in 
the office. Several packages were first sorted with the aid of a magnify¬ 
ing glass and micrometer caliper, and a selection made of those whose 
shape and size appeared to be identical with the shape and size of the 
standard. From a lot of 72 needles, only 12 were thus selected. From 
these 12, those were selected for use which gave practically identical 
results in the penetrometer with a so-called standard needle on two 
samples of oil asphalt. The results of these tests are given in Table I, 
from which it may be seen that only 5 of the 12 needles fulfilled the 
requirements. Needles that failed to give accepted values on the harder 
materials were not tried on the softer. 
Inasmuch as only 5 needles out of 72 proved acceptable, it may be seen 
what results would follow from the indiscriminate use of No. 2 needles as 
such. 
It is further to be noted particularly that there is no existing single 
standard with which comparison can be made, owing to the fact that 
there is no means of accurately defining or gauging the type of needle in 
use. The work herein described was undertaken for the purpose of devis¬ 
ing, if possible, a needle which would give results practically identical with 
results now obtained in using the so-called standard needles, and which 
could be accurately described and duplicated at any time. 
The standard needle on file in the office is 1.8 inches in length, with a 
diameter of 0.040 inch for a length of 1 inch from the eye. The remainder 
of the needle tapers in a parabolic curve to a sharp point. The simpler 
