82 AN ANALYSIS OF THE ISHERWOOD SYSTEM 



versally used to advantage. I should not use that system myself on the Lakes, for instance. 

 The reason has to do with repairs. If you run Lake steamers in the Canal, or the big 600- 

 footers, carrying 15,000 tons or more — if you run them in shallow water and constantly rub 

 the bottoms, you will give the Isherwood system very extensive damage, not in any one 

 place, but running quite a distance along the bottom, and this is true particularly in the canals. 

 You cannot use them in the canals, because in the Canadian canals, for which I have built a 

 number of vessels, the bilges are constantly touching the banks and the sides of the ships 

 are in contact with the locks because the locks restrict the dimensions of the boat, and the 

 shipbuilder in trying to improve the dimensions has made the boat narrower ; you will have 

 long scores between the longitudinals and constant trouble. I would not recommend the Ish- 

 erwood system for that kind of work. Mr. Isherwood may have different views, but that 

 is the idea I have formed from my experience. 



I should not recommend this construction in connection with very large passenger steam- 

 ers, because you have a large number of decks which have to go in anyway, and therefore you 

 have ample longitudinal strength in the association of these decks with longitudinal bulk- 

 head, all forward and aft, and there is no reason why you should "Isherwood" these steamers. 



I saw a reference in this paper to the Storstad. The Storstad, as you know, sank the 

 Empress of Ireland in the St. Lawrence River some years ago with a terrible loss of life. I 

 had occasion to investigate the collision for the owners of the Storstad, particularly with a 

 view to finding out what happened, judging by the damage done to the Storstad. The Em- 

 press of Ireland was under water, and we checked up a good deal of what happened by ex- 

 amining the damage to the Storstad. The one thing I want to bring to your notice is this — 

 the Storstad had rammed the Empress of Ireland a heavy blow at an angle around 40 de- 

 grees, glancing past the Empress of Ireland and cutting a broad hole in the Empress of 

 Ireland down through all the decks. The Storstad had only damaged about 12 feet of her 

 own length out of a total length of 425 feet ; in other words, the longitudinal framing stood 

 out against this ramming strain and resisted it in a way that no transverse framing would 

 have done, the Storstad not being damaged in the collision bulkhead and not up to it. 



So, as regards damage in collision, the one boat gets off well — if it is the rammer. As 

 to the other fellow of course, well, he ought to get out of the way, I suppose. 



That brings up the question of repairs. There is very little repairing to be done in a 

 case of that kind. The bow of the Storstad was cut off by running a torch around the hull, 

 the piece dropped off, and a new bow put in. I believe that in a transverse ship there would 

 have been a concertina effect and that the damage would have extended back to the bulk- 

 head, and that very much more serious damage would have occurred. 



Mr. E. a. Sperry, Member: — I want to call attention to one caution that, it seems to 

 me, ought to be exercised with reference to the construction of longitudinal framed ships, 

 and that is with reference to the torsional factor. The foresquare modulus must receive its 

 principal backbone from decking. Now in all modern cargo ships, where the hatches have 

 to be such a large proportion of the total decking, you have to go carefully with regard to 

 any weakening of the ship against excessive torsion. This was called to my attention when 

 we had the stabilizer on the Worden. There was a very unusual condition. The gyro was 

 used on one occasion to roll the boat and rolled it tremendously in perfectly still water. We 

 thought we noticed quite a difference in rolling on the two ends of the ship', and so we erected 

 battens and with a dock nearby — not too close to interfere with the wave action — very accu- 



