626 THE STRENGTH 



This circumstance, which was shown too by 

 experiments 23 and 26, will enhance the value 

 of the latter, and of experiments 30 and 33; 

 especially as they are on a larger scale than any 

 other we have made. We have not in these 

 experiments been able to crush the top rib of 

 the beam; though in both 30 and 33, the 

 bottom rib was to the top one in a higher pro- 

 portion than had been used in experiment 19, 

 or in either of the former series of experiments ; 

 the ratio in experiment 33, being nearly that of 

 8 to 1 . The bottom rib, in both instances, was 

 rather too large for the thickness of the vertical 

 part, as was evident from that part having 

 failed the first. To have made the section then 

 of equal strength every where ; and conse- 

 quently to have disposed the metal in the most 

 economical manner possible (which has been 

 one principal view in these enquiries), the 

 vertical part of the beams should have been 

 rendered a little thicker ; and, as we have just 

 seen, perhaps, made uniform. The top rib too, 

 small as it was, would in both instances^ 

 possibly, have borne a reduction; only, that 

 it would have rendered the beam liable to have 

 twisted ; a tendency which shewed itself in 

 both experiments. 



From these experiments, and those commenc- 



