August 14, 1890] 



NATURE 



;79 



bodied in Mr. Longridge's "Treatise on the Application of 

 Wire to the Construction of Ordnance," 1884 (Spon) ; and again 

 in a paper in 1887, "Further Investigations regarding Wire 

 Gun Construction." 



Dr. Woodbridge, of America, claims to have originated the 

 system of strengthening guns with wire, in 1850; but to Mr. 

 Longridge belongs the credit of pointing out the proper mode 

 of winding on the wire with initial tension so adjusted as to 

 make the firing tension of the wire uniform for the maximum 

 proof powder pressure. 



Mr. Longridge's principle is applicable not only to engines of 

 destruction, but also to peaceful purposes, such as strengthening 

 the cylinders of hydraulic presses and lifts, and the copper pipes 

 of steam-engines ; for which a great, and, we hope, a profitable 

 future is in store. 



Keturning to the application of the principle to artillery, the 

 great object attained is the notable reduction in weight of the 

 gun — a matter of importance in siege artillery, where the weight 

 of the largest single piece of metal, the gun itself, is limited by 

 the difficulty of transport over bad roads and rough country. 

 By the use of Mr. Longridge's principle, the weight of a 

 howitzer can be reduced from five tons to three and a half— 

 quite sufficient to make all the difference between getting the 

 gun into position, or being compelled to leave it behind. 



It is also claimed as an advantage of the wire gun that the 

 construction will be found cheaper and more expeditious, when 

 once the appropriate machinery is erected ; and that this machi- 

 nery need not be nearly so elaborate and expensive as that re- 

 quired with the present system of construction with steel coils 

 shrunk on over each other. 



As we have seen in Part II., the appropriate initial state of 

 stress is, in the coil gun, dependent on such delicate fitting as 

 thousandths of an inch, and a slight irregularity in the texture 

 of the metal may be sufficient to completely modify the initial 

 stresses as designed. With the wire gun, on the other hand, 

 the wire can be coiled on to the inner tube from an equal parallel 

 coil of wire, and the appropriate tension given by means of a 

 certain weight running on the free part of the wire, and inci- 

 dentally testing the strength of the wire. Certain practical dif- 

 ficulties exist in securing the ends of the wire, and in providing 

 for longitudinal strength, which experience will doubtless soon 

 overcome. 



Besides Mr. Longridge's "Treatise," the most important is a 

 long article in the Revue d' Artillerie, on " Steel Wire Guns," by 

 Lieutenant G. Moch, since published as a separate book, and 

 also translated in the American " Notes on the Construction of 

 Ordnance," No. 48, 1888. 



Lieutenant Moch resumes Longridge's and Brooks's calcula- 

 tions, and presents the mathematical work in a more concise and 

 elegant form ; he applies his formulas to the design of the wire 

 guns, proposed in 1871 by Captain Schultz, who was unaware 

 of Mr. Longridge's previous work. 



We shall attempt here to present the essence of Lieutenant 

 Moch's article in a concise and geometrical form, depending on 

 the method and formulas of Parts I. and II., and illustrated by 

 the design of one of Schultz's guns ; referring the reader who 

 wishes to pursue the subject in all its practical details to Moch's 

 original article, and to Longridga's "Treatise." 



(44) Taking the cross-section of the gun across the powder- 

 chamber, as composed of the inner tube, a, the wire coil, B, 



/,0=20 



Fig. 9. 



and an outer jacket, c, then in the ideal state, the firing stresses 

 will be represented in Fig. 9, where the curve of circumferential 

 tension, t.^tt\ is a straight line in the wire coil b. 



The outer jacket, c, is merely required for protection of the 

 wire from damage by shot, so that it may be supposed fitted over 

 the wire without any appreciable shrinkage ; when the gun is at 

 rest, the jacket c will then be in a state of repose free from 

 stress ; but when the gun is fired, we may suppose the stresses 

 in c to be the powder-stresses (§ 12, p. 306), on the assumption 

 that the gun behaves as if homogeneous. 



We denote by r^ the internal radius of the tube, by r^ and r^ 

 the internal and external radii of the wire coil, and by r.^ the 

 external radius of the jacket, all measured in inches in our units. 



Then in the jacket c the curves t'.^t^ of circumferential ten- 

 sion and r.^p^ of radial pressure, representing firing stresses, will 

 be Barlow curves, the reflexions of each other in their medial 

 axis Ogt?:,. 



(45) The continuation of the Barlow curve r^p^ in the dotted 

 line up to/o will give graphically the powder pressure ;>o ; but 

 now the curve of firing radial pressure between rj and r^ will be 

 the broken curve /2/1A' of vdiich p\p^ in the tube A is the por- 

 tion of another Barlow curve, but of which /j/j in the coil B is 

 easily seen to be a portion of a hyperbola. 



For the curve of firing circumferential tension in the wire 

 being the straight line tO.,, the condition of equilibrium of any 

 cylindrical portion of the wire coil, bounded internally by the 

 radius r, requires that the rectangle r^t of circumferential resist- 



NO. 1085, VOL. 42] 



ance should be equal to the rectangle Op 

 in other words. 



rectangle O/2 ; or. 



the rectangle p^ = rectangle Op, 

 or 



the rectangle Oj/j = rectangle O.2P ; 



which proves that the curve p2p is a hyperbola, with OjO and 

 02^ as asymptotes. 



(46) The tangent at any point of this hyperbola — say at/j — 

 is drawn by joining the point p^ with points on OjO or Oo^ at 

 double the distance of/2 from O^^ or OjO, by a well-knowo 

 property of the hyperbola. 



But to draw the tangent at p.^ of the Barlow curve r^p^ we 

 must join p.^ with a point on O3O at a distance from Oj treble 

 the distance of/., from Oj^j. 



Similarly we can draw at/j the tangent to the hyperbola /j/i, 

 and the tangent to the Barlow curve pipo, when we know the 

 position of OiO^, the axis of this Barlow curve, pipQ. 



(47) The position of Oi^-i is fixed by the condition that the 

 curve of circumferential tension in the tube A is the reflexion of 

 the curve /i/o in 0,^i ; and the position of this curve of cir- 

 cumferential tension, /q/'j, is settled by the condition that the 

 rectangle O/o is equal to the sum of the areas of circumferential 

 resistance, bounded by i^^'^ in the jacket c, by the straight line 

 /j^'j in the wire coil B, and by the curve i(^i\ in the tube A. 



(48) It will be noticed in the diagram that, with the numbers 

 given there, the curve iQt\ lies to the left of the line r^r,, showing 



