44 



Eocpei'iments on the Adhesion of Iron Spikes, &fc. 



TABLE IV. 



Spikes driven to different depths. 



By comparing experiments 1 and 4 together, it will be found that 

 weight for weight, the flat spike had when driven 1.8 inches, an ad- 

 vantage of 42.3 per cent, over the square one ; and by a like com- 

 parison of experiments 2 and 5, it is evident the former had a supe- 

 riority of 37.7 per cent. As the spike when driven in only 1.9 

 inches had a much less proportion of its parallel faces exposed to 

 the reaction of the fibres and a greater proportion of the wedge- 

 shaped point, it is reasonable to expect that the retention would not 

 correspond precisely with the lengths inserted. It wqll be understood 

 that when we speak of cutting edges and the wedge-shaped portion 

 of spikes, whether square, flat, or cylindrical, the direction of the 

 cutting edges is always across the fibre or grain of the timber. ]t 

 must be evident that the wedge-shaped part may be so acute, as to 

 correspond nearly with two parallel faces, in which case, the tenden- 

 cy to retreat from the lateral pressures is small ; and the pressures 

 themselves, increasing from the point upwards to where the spike is 

 thickest, the total efficiency of a given length may be as great as that 

 of an equal length of the parallel faces, and even greater, provided 

 the thickness of the spike be so considerable as in driving it to pro- 

 duce much crushing and irregular folding of the fibres of the timber. 

 If, on the other hand, the edge be very blunt, the tendency to recoil 

 may be such as to diminish the adhesion, and in this case the effect 

 of the wedge shape is negative. In the other it may be positive.* 



* The following formula may represent the several experiments ; R=://d: c, in 

 which R is the observed retention ; Z= the length in inches of the part inserted ; 

 /= the force of retention on one inch of the parallel faces, and c= the difference 

 between the retention of a parallel portion of the spilce, and of an equal length of 

 the converging faces near the point. The sign of ambiguity arises from the cause 

 above explained. 



