8 io 
Journal of Agricultural Research 
Vol. V, No. 17 
Table II gives the results of three determinations for each sample 
under each of the conditions tried. These penetrations were all taken 
with the same needle at different points on the surface of the sample. 
Reading from left to right, the first test was made at the center, the 
third 1 cm. from the edge of the dish, and the second halfway between 
the positions of the first and third tests. For the dish measuring 5.5 cm. 
in diameter, the first penetration was therefore taken 2.7 cm., the 
second about 1.9 cm., and the third about 1 cm. from the edge of the 
dish. 
It will be noted that the time elapsing between pouring the sample 
into the dish and determining its penetration varied from a total of 1 
hour to over 28 days; that the immersion in the water bath directly pre¬ 
ceding the test varied from 30 minutes to 1 % hours. Upon reviewing 
the results given in this table, it appears evident that, in general, for any 
given set of conditions preceding the immersion in the water bath, a 30- 
minute immersion in water gave less consistent check results than a cor¬ 
responding 1-hour or iX-hour immersion. Less difference is indicated 
between the i-hour and i^-hour immersions in water, but the balance of 
evidence appears to favor the latter period of time in so far as uniformity 
is concerned, even when negligible personal errors are taken into account. 
Thus, out of the 11 series of comparative tests of 1 hour and hours 
for all 12 materials, it will be found that in 61 cases the i%-hom immer¬ 
sion gave the most consistent results; in 21 cases the most consistent 
results were obtained with the i-hour immersion; and in 50 cases there 
is no preference so far as consistency in results was concerned. 
If the average of the three tests for any sample is taken for the i-hour 
air cooling and i-hour immersion in the bath, as compared with the 
30-minute air cooling and i-^bour immersion in the bath, it will be found 
that they practically coincide. The fact, however, that in the latter 
case there is less difference between the individual results indicates that 
the i^-hour immersion should have preference. 
Eliminating the 30-minute immersion in the bath before making the 
test, and considering only the i-hour and i>^-hour immersions in con¬ 
nection with short periods of prior cooling in air, Table III will be found 
to illustrate the differences above described. Here, comparing methods 
5 and 3, it will be seen that in seven cases the most consistent results 
were obtained by the i>^-hour immersion; in two cases the i-hour immer¬ 
sion produced the most consistent results; and in three cases there is no 
preference with regard to consistency in results. So far as rapidity in 
making the test is concerned, therefore, if a short-period air immersion 
is to be adopted, it would seem that 30 minutes in the air and hours 
in the bath prior to testing would be the most satisfactory minimum 
limits to adopt. 
