19 
for their other boundaries severally the vertical planes 
through ]3iAj3 and B x OB. But as each of the bounding 
planes of each of these plates cuts or crosses each of the 
bounding planes of the other, so likewise must each of the 
plates themselves cross or cut through the other, so again 
therefore must each of the slices of either plate cut across 
each of the slices of the other, and so therefore lastly, must 
each of the cylindric surfaces cross or cut the other. There- 
fore the vertical cylindric surfaces through ciGc and 
cut one another in some vertical straight line P lying within 
the vertical prismatic rod which is common to the plates 
bounded severally by the parallel vertical planes through 
a x Ga and A x OA and by the parallel vertical planes through 
j3i a/ 3 and Pi OP. 
‘‘Note on the Dimensions of Ships,” by J. P. Joule, 
D.C.L., LL.D, F.R.S. 
I have often thought that in practising the , art of ship- 
building men have too much neglected the study of the 
forms of the fish which make the waters their permanent 
habitation and are designed for the most part to attain the 
highest degree of velocity in the pursuit of their prey. 
No doubt the case of a ship partly, and that of a fish 
wholly immersed, are not strictly parallel, but they offer 
very many points for comparison of which we may avail 
ourselves. 
A fish makes use of its tail fin as the chief and nearly 
sole instrument of propulsion, and in the adoption of the 
screw propeller in preference to the old side wheels the 
steamers of the present day have secured a great advantage 
over the old forms. In the proportion of length to those of 
breadth and depth, however, although there has of late 
been some improvement, there would appear to be a lingering 
tendency to hold by the old mistaken idea that a ship was 
rather to be regarded as a wedge to cut the water than as 
