Some of the large shipbuilders have their own tanks, and so have the naval authorities of the great naval Powers. The one at Teddington was established through the munificence of a famous British shipbuilder, Mr Yarrow, who not only defrayed the cost of construction, but gave an endowment to assist in its upkeep. It is intended to serve the needs of the smaller builders who have not tanks of their own, and also for the investigation of matters of general interest to shipbuilders, and for such tests as that relating to the _Hawke_ and _Olympic_. In this last-named case, of course, two models were made, one to represent each ship, and they were towed along in such a way as to imitate very closely the movements of the ships at the time when they collided. It was as the result of these tests that the _Olympic_ was ordered to pay damages to the Admiralty, it being held that she was the cause of the accident. A very interesting investigation of this kind was recently carried out in the tank at the United States Navy Yard. The port of New York consists very largely of jetties projecting out from the banks of the river. With the growth of the Atlantic liner the old jetties had become too short, and questions arose as to the elongation of them. If it were done, how would it effect the current in the river, and the handling of shipping generally? If, on the other hand, it were not done, what would be the effect of the ships lying with their ends projecting out into the stream unprotected by a jetty. To determine these points the experimental tank was converted into a model of the New York Harbour, or at all events of that part in connection with which these questions arose. A false floor was put in, so as to make the depth exactly right in proportion to the width. Little model jetties were arranged to represent exactly the real ones, while against them were moored model vessels, so that the effect upon them could be observed as the model of the large vessel was towed past. In addition to this, special appliances were arranged for finding out what the disturbance might be which the movement of a giant liner produces under the surface as well as above it. For this purpose buoyant balls were employed, moored at various distances below the surface, from which thin rods projected upwards, the movement of which rendered visible the movements of the submerged balls and therefore the effects of the under-water currents.