OF SHIP CONSTRUCTION. 83 



rate measurements were taken. It was amazing what a difference in torsion would occur be- 

 tween the two ends of that structure, and so I started to investigate. It was a rather small 

 destroyer. We had it loaded to about a thousand tons at one time. I found a very stran<'-e 

 thing about torsional strength generally. Now we all know the enormous torsional strength 

 of a tube. Suppose we had a 6-inch tube, ys inch thickness, and, we will say, 20 feet long. 

 You understand that a torsional effort applied at one end could be taken off at the other end 

 of the tube with a very slight deflection. A very severe torsion would in fact show little 

 or no deflection. 



Now what do you suppose occurs if we should cut that tube longitudinally into two equal 

 parts? We simply would have two gutters. How does the torsional stiffness of the gutter 

 compaie with the torsional stiffness of the original tube? That interested me. We found 

 that the ratio is about 3,600 to 1 — that is all. As you know, an ordinary eaves gutter, laid 

 on the ground, of its own weight will show an angle of 90 degrees or more. Here we 

 have a ship — and a ship is nothing but a gutter — just about half of a tube, and so the con- 

 structor must look to it that he provides against abnormal fiber torsion. 



There is another point which has interested me considerably since we have constructed 

 two pallograpbs for the Construction Bureau. The author speaks of the natural period of 

 vibration of the hull. We have never found there is any such period from the use of a 

 pallograph. We find, when you change the period of application of the oscillating disturb- 

 ance, the ship will respond to almost any period before the nodes in the ship will change 

 very markedly. You can run her nodes from a position near the stem to a position forward 

 of the middle of the ship by simply changing the period, so I have an idea that a long, slen- 

 der structure, such as the Isherwood ship, will probably respond in some way to almost any 

 period that is applied to it, and in the case of some periods, when the nodal points are near 

 the middle of the ship, it will of course respond more readily than the vibrations of a higher 

 period. 



Mr. HiatT: — I would like to inquire if there are any shipbuilders present who can 

 give any further light on the amount of saving in weight they have found to exist in the 

 Isherwood system. 



The President : — Can anyone shed any further light on the subject of weight? No 

 doubt that question will be answered later in the written response which will be sent in on 

 the discussion. Mr. Flodin, have you any further comments to make in rejoinder? 



Mr. Flodin : — In regard to Mr. Hiatt's remarks as to the financial success of the ship, 

 the paper was not intended to reflect on the conservation of financiers. I am not much of 

 a gambler myself, but, on the other hand, I think a majority of you will agree with me 

 that there is a little too great conservatism among shipbuilders. 



Furthermore, Mr. Hiatt points out the weakening of the longitudinals where they are 

 cut at the bulkheads. Undoubtedly there is a weakening, but also, as I have already men- 

 tioned, I allowed the same reduction in strength for the longitudinals as in the case of the 

 shell. Furthermore, the classification societies now allow all longitudinals to be cut at the 

 butts of the shell platings; that is to say, the longitudinals are the same depth as the shell 

 plates, less the width of the butt lap, and are strapped at the shell butts. This construction 

 facilitates erecting, because the longitudinals may be shop-riveted to the shell plates. The 



