278 Experiments on the Strength of Wood. (APRIL, 
found to be the strongest, as the mean weight of the first and second 
= pa = 67°5, is greater than that of the third and fourth, 
66°55, or fifth and sixth, 58. 
A few days past | received the following ingenious experiment 
and information from Mr. George Smart, carpenter and builder on 
the Surrey side of Westminster Bridge, and well known for his 
invention of hollow masts, yards, booms, bowsprits, oars, and 
sweeps; and for his valuable application of Mr. Bramah’s press to 
compress wood, and prevent its shrinking when formed into staves 
for making gunpowder-barrels, canteens, &c. Neither should his 
peculiarly ingenious manner of forming saddle-trees be omitted. I 
beg to transmit you his own letter and drawing (Pl. LXV. Fig. 2): 
“* Description of a Frame for making Experiments on the Strength 
of Timber. 
A, A, two planks of fir, one the top, and the other the bottom, 
of the frame, into which are framed the two upright ends, B, B, 
with double tenans. The bottom tenans are pinned, and the 
shoulders, as they are called by carpenters, housed about half an 
inch into the bottom plank. ‘To give them more strength, the top 
plank, A, is morticed, to receive the top tenans and the wedges, 
C,C. ‘There is a hole cut in each end, to admit the thick ends of 
the lath, L, which, being put into its place, is kept down by the 
blocks, D, D. The wedges, C, C, being driven tight, the lath, L, 
having the shoulder at each end, cannot slip; and is so strained 
that on its horizontal position it will carry nearly as great a weight as 
it would suspend in a vertical direction. It is now nearly 17 years 
since I first put the lath into the frame; and, to all appearance, it 
is as strong as when first put in. The late ingenious Dr. Anderson 
was on a visit at my house when IJ tried the first experiment. He 
told me that Belidor affirmed, that timber made fast at both ends 
would carry one-third more than when it lay loose on its supports. 
On making the experiment, we had the lath without shoulders laid 
loose on the frame. It bent like a hoop before it broke with some- 
thing Jess than 11 lb. Next we put in the one with shoulders, and 
drove the wedges very tight: it sustained 240 lb. The Doctor was 
pleased, and much surprised that my experiment should differ so 
much from all those published by eminent men. This is accounted 
for by having shoulders on the ends of the lath, which, butting 
against the ends of the frame, cannot bend enough to allow the 
fibres of the wood to form a fulcrum to break one another. This 
experiment shows how necessary it is that all timbers should be 
carefully coaked down on plates of buildings, and not dovetailed, 
as is often done in carpentry. Much timber would be saved, and 
the building made stronger, by the foregoing plan of shoulders or 
coaking where it can be applied. The French, in ship-building, 
are particularly attentive to scoring and coaking wherever they can. 
5 
