— 110 — 
What the actual transferred load is it is impossible to say, but it will be mak- 
ing an error on the side of safety if it be assumed that the load is equally divided 
between tho trusses ； any extra iron that may be thereby used in the vertical sway 
bracing will be well employed, iov it will be in a good place to resist vibration. 
Under this assumption let us investigate 
the stresses in the bracing. Let the nota- 
tion be as in Fig. 8, II and B! being the reac- 
tions due to tho weight W, distributed ac- 
cording to the law of the lever, so that 
R = W 
2 a + b 
Let G be the weight transferred by the brac- 
ing, then 
The stress in the vibration rod is therefore 
Fia .s. 
j i 
T= G&ec 0 = 
TV a sec q 
2 (a + b) 
The stress in J K is found by passing a piano to cut G IT, J K and J H, sup- 
posing that the only weight acting is ^äTb] E, and taking tho centre of moments 
at H. This gives 
f r Tn J ^ 2(a + b) _ JVa 
- 取 W ~ T ~ — 了 
Again taking the centre of moments at J and using the same cutting plane wo find 
tho stress in G II to be zero ； for the moment of the increase of weight at F is 
balanced by tho moment of tho increase of reaction at that point, making the resul- 
tant moment of the external forces zero, and the stresses in J II and J K t having no 
lover arms, tlieu. moments are zero, consequently the moment of the stress in G H 
is zero and the stress itself zero. 
To find the bending effect upon tlio post at K let us pass a plane cutting K F 
and J E, and take the centre of moments at K, then 
M =2^)l 2 ( a + b )]=Wa 
If h be the distance between centres of gravity of post channels and an intensity 
of four tons be employed the ffl*ea of one channel necessary to resist this bending 
will be 
A M Wa ， 
— 4 ん一 4/7 
But as this effect does not exist at the same time as the maximara load stress 
upon the post H F, it need be consiclered only when the post.is very lights 
