498 TRANSACTIONS OF THE AMERICAN INSTITUTE. 



the batteries at Fort Darling, on James river, Va., and tlie gunboat Caron- 

 dolet and the rebel ram Arkansas, on the morning of 15th of July, 1862 ; 

 also the engagement between the gunboat Benton, commanded by the late 

 Captain Gwin, and the batteries at Haines' Bluflf, on the Yazoo river. 

 This vessel was struck twenty-nine times in this engagement. The first 

 named vessel's armor plates that were penetrated were three inches thick ; 

 the latter two vessels belonging to the Mississippi squadron, were protected 

 with armor plates two and a half inches thick. The second engagement (on 

 the morning July 15th) was witnessed by your memorialists, who exam- 

 amined the armor plates after the action. These plates were penetrated by 

 eight inch shot, the range being about the same as that when the Essex 

 passed the batteries at Port Hudson, September 1th, 1862, 



Now, if one inch of iron, and one inch of India rubber between the armor 

 plates and the timber (on the Essex) effectually resisted the penetration of 

 the before mentioned shots at short range, (of which fact there can be no 

 possibility of doubt,) when we take into consideration the width of the 

 river, position of the vessel, and the locality of the batteries from which 

 these shots were fired, as well as other evidences of the facts, what may be 

 expected of a casemate composed of from three to four sheets of iron, of one 

 inch in thickness, and the same number of vulcanized India rubber plates 

 interposed between them, the iron plates being clamped together by means 

 of flanges and grooves, (see accompanying drawings,) so as to afibrd a 

 mutual support, and to act as one continuous sheet. The probability is 

 that the power of resistance to penetration of this combination would only 

 be limited by the ability of the support or casemate to sustain the momen- 

 tum of the shot, and support the plates. 



The ability of such fastenings, as before mentioned, to prevent a vessel 

 from being swamped, going to pieces, or springing leaks during a heavy 

 sea, will depend only upon the depth of hold, width of beam, and the man- 

 ner in which the work is put together, and not merely upon bolts and rivets, 

 as has heretofore been the case. 



In order to further illustrate and explain this proposed plan, for exam- 

 ple, when we wish to drive a nail through a hard seasoned plank, we 

 naturally lay it upon and in close contact with a solid support, in order to 

 concentrate the effect of the blow upon the fibres of the wood at the point 

 of contact, and cut through them without incurring that resistance which 

 their fibres could exert to lessen the effect of the blow, if the elasticity of 

 the support was such as to allow the fibres to act, and diffuse the force of 

 the blow from the point of impact. So with the fibi'es of iron, for their re- 

 action is most powerful, if suUered to react. To prevent this in cutting or 

 punching a plate of iron, the most unyielding and solid support that can be 

 possibly devised is sought, in order to obtain the greatest effect of penetra- 

 tion. These are familiar instances, but they show conclusively that iron 

 plates should never be laid immediatel}' upon a solid support if we wish 

 them most effectually to resist the penetration of a cannon ball. It is true 

 that in punching a hole through an iron plate, we place it immediately 

 •over a hole or cavity; but we do this that the sharp edges of tbi^ 

 hole in conjunction with the sharp edges of the punch may act 

 as a pair of powerful shears to cut out the plug of iron, at the 



