when driven into different species of Timber. 43 
drawing these spikes, the holes were almost perfectly square. This 
resulted from the position of the rings of annual growth and the 
greater elasticity in some directions than in others. It is probable 
that if the filed grooves in experiments 16 and 17 had been covered 
with a scale of oxide, as was the case with the plain spike used in 
experiment 18, the former would have given a result somewhat 
igher. 
When holes are drilled into stone blocks and afterwards plugged 
with timber to receive spikes in fastening on the chairs of edge rails, 
the method of experimenting just described finds an application, and 
it is probable that in such cases the grooved cylinder with a conical 
grooved point, may prove advantageous. 
A few experiments were made to determine the effect of driving 
to different depths, on the total amount of retention. For this pur- 
pose two different spikes were selected, viz. the square hand-wrought 
spike, the section of which was .405 x .402, and the wide flat one 
of which the section was .539X.288. They were respectively 
driven to a certain depth into unseasoned chestnut, and then subject- 
ed to a force just sufficient to start them. This force was noted, 
and the spike was immediately driven down one inch deeper than 
before, and the force again applied. All my experiments proved 
that when a spike is once started, the force required for its final ex- 
traction is much less than that which produced the first movement. 
This is readily accounted for on the principle that as the wedge- 
shaped point was from half an inch to an inch in length; and as 
this, on the starting back of the spike a very little distance, became 
mostly relieved from the pressure of the fibres, all that part of the « 
retention which had been due to the wedge-shaped portion of the 
spike was at once destroyed. The following table will show, how- 
ever, that the mere starting of the spike with parallel faces does not 
essentially diminish the retention, when again driven into the timber 
to a greater depth than before. But when a bar of iron is spiked 
upon wood, if the spike be driven down until the bar compresses the 
wood to a great degree, the recoil of the latter may become so great 
as to start back the spike a short distance after the last blow has been 
given. In this case a great diminution in the useful effect will be 
the consequence. This shows that a limit may exist to the force 
which we should apply in urging down spikes or bolts destined to 
fasten materials together. 
