ON NA VAL ARCHITECTURE. 301 



correctly follow the line on the drawing as it passes under the indicating 

 point : thus the cutters cut a pair of water lines, simultaneously on the 

 two sides of the model. The model is then change. 1 in level and the 

 successive water lines are cut ; you see them here in a form analogous 

 to that which the ship builder adopts when he sets about making a 

 model. He gets his form by cutting out a series of pieces of board to 

 form the several water lines, the thickness of the boards representing 

 the interval between the lines, builds them together, and shapes off the 

 intermediate matter. 



Our particular material, paraffin, is most delightful to operate upon. 

 It is quite strong enough for all the purposes of the model, but it is 

 very easily cut with a knife, and may be operated upon with ordinary 

 carpenter's tools, with the advantage that from being without grain it 

 does not misguide the cutting edge. It has the pleasant property that 

 it does not tend to choke the tools, there being no stickiness about 

 it, in fact, with a smoothing plane a shaving of about one thousandth 

 part of an inch in thickness can be taken off quite cleanly. The 

 model, after bdng cut to the water lines, is shaped by suitable tools 

 to the finished form. That form is represented by the bottoms of 

 these cuts ; but, as the bottom of a cut when once it is reached no 

 longer exists as a guiding line, the operator when he had reached it 

 would be " out of soundings," and would not know whether he had 

 gone too deep or not ; to get over that difficulty, as soon as the water 

 lines are cut, a spur with short points, the alternate points being very 

 short, is run along each line, leaving a series of corresponding punctures, 

 and a little black powder is run into these punctures, so that when the 

 work is finished it can easily be seen whether the operator has gone 

 deep enough and not too deep. We have another test of the model's 

 correctness. Its displacement is carefully calculated beforehand, and 

 when it is finished its weight is taken. It is then loaded with enough 

 ballast to give it exactly the displacement that it ought to have. It is 

 placed in the water, and, by very accurate means for testing the immer- 

 sion, we see whether it comes to its proper plane of flotation. With a 

 model weighing 400 or 500 Ibs. we generally get the displacement right 

 within about two pounds, and if we were three or four pounds out we 

 should think there was some error, and should probably cut the model 

 afresh, or go over the calculations to see if there were any mistake. 



