29t 



THE CIVIL ENGINEER AND ARCHITECTS JOURNAL 



[OOTOBBB, 



(ihtained from coals, it is well tliat the ninde of niakinf; it should 

 he ireiierally known. 



To those wlio are unacquainted with the chemical composition of 

 water, it may seem stranse that water should he rendered the 

 source of fire ; but to most of our readers it must he well known 

 that water is composed of liydroeen, the most inflanunalde of 

 bodies, and of oxyfren, which, when in the form of pas, is the 

 most active supporter of combustion. A ]dan of ohtaininir hydro- 

 jien c:as from water, by passing: steam throufrh a hot tube containing- 

 iron has been long known. The ratiniin/i' of that process is, that 

 the steam when in contact with heated iron becomes decomposed, 

 the oxypen imitine: with the iron to form an oxide of that metal, 

 and the hydrogen is liberated in the form of gas. This, indeed, 

 is the best mode of obtaining hydrogen gas in a state of purity ; 

 but for the purpose of illumination, such gas is of no value. 

 The flame, though emitting great heat, is scarcely visible. The 

 illuminating power of coal-gas depends on the carbon it contains ; 

 and the more carbon is contained in carburetted hydrogen, the 

 greater is its illuminating power. It is owing to the great pro- 

 portion of carbon in turpentine, that it aflF(u-ds such a brilliant 

 light in the "camphine" lamjis, the only difficulty in thel)urning 

 of that substance being to produce perfect combustion ; otherwise 

 the abundance of carbon causes volumes of dense smoke. To 

 render the water-gas illuminating, it is necessary, therefore, to 

 combine with it a portion of carbon ; and it is this part of the 

 process in which the principal novelty of the invention consists. 

 The apparatus employed in the manufacture of the gas is exhi- 

 bited at work at the Polytechnic Institution. It consists of a 

 furnace, in which are three long iron retorts placed perpendicularly. 

 Two of these are nearly filled with coke and old iron chains, or 

 pieces of iron. Water is admitted into the first of these, and 

 being converted into steam, it is then decomposed by the iron, 

 and the hydrogen gas which is liberated, absorbs at the same time 

 some portion of carbon from the heated coke. The gas and resi- 

 dual steam are then passed into the second retort, where a similar 

 process of dec(miposition and of further carbonization takes place ; 

 and it then issues into the third retort, where it is brought into 

 contact with heated tar, and absorbs from it a large portion of 

 carbon. The carburetted gas is then forced through some vertical 

 tubes, to permit the deposition of superfluous tar, and is conducted 

 into the gasometer ready for use. The illuminating power of this 

 (las is estimated to exceed that of ordinary coal-gas, 25 per cent. ; and 

 its freedom from sulphur and other impurities renders it far prefer- 

 able to coal-gas. Respecting the economy of the process. Dr. Ryan 

 says nothing ; and we believe that it was on this point that the 

 invention failed to be practically useful when first introduced. 

 The cost of the fuel to heat the retorts, of the iron to decompose 

 the steam, and of the tar to carbonise the gas would, we fear, 

 amount to more than the cost of coal, in most parts of England, 

 for making the ordinary kind of coal-gas. In many circumstances, 

 however, we conceive this mode of generating illuminating gas 

 may be advantageous, especially when the purity of the gas con- 

 sumed is an object of importance. 



ON ISOMETRICAL RERSPECTIVE. 

 By R. G. Clark. 



The object of this article is to communicate an easy method, by 

 construction, for determining the transverse and conjugate 

 diameters of an ellipse touching the sides of an oblique parallelo- 

 gram, being the isometrical projection of a circle inscribed in a 

 square. 



Draw the isometrical parallelogram A B C D, and its two 

 diagonals A C, and B D ; bisect O C in m, also bisect A tn in N, 

 and then with the radius A N and centre N, describe a semicircle 

 cutting O B, produced in R. Again, with centre O, radius O R, 

 describe a semicircle a R c cutting A C in a and c ; then will o c be 

 the transverse diameter required of the ellipse. In like manner, 

 bisect O D in .?, and B *• in r, and with the centre r, and radius )• B, 

 describe a semicircle cutting O C in i ; again, with centre O and 

 radius O t, describe an arc cutting D B in 6 and d, then will bd be 

 the conjugate diameter required. 



The above may be demonstrated thus : — Because the sides of the 

 oblique parallelogram respectively touch the curve, they are tan- 

 gents to it. By the properties of the Conic Sections (see Dr. 

 Hymer's elegant treatise), we have, AOX MO = a O- ; but 

 MO =: ra O, therefore a O- = A O X m O. Also, by the property 

 of the Circle, we have a O^ ^ O R-. A similar mode of proof 



applies to the conjugate diameter. After the diameters are thus 

 determined, the curve can be easily trammelled in tlie usual way. 

 It would also be well to state an easy rule, by calculation, founded 

 on the above construction. 



ot 



Rule. — Multiply the diameter of the circle by l-22t for the 

 transverse diameter, and by -707 for the conjugate diameter. 



E.r. — Given the diameter of a circular turn-table = 14' 5", 

 to find the transverse and conjugate diameters of its isometrical 

 representation. 



Here, by the rule, 14-5 x 1'224 = 17-748 transverse diameter. 

 M.-5 X -707= 10-25 conjugate diameter. 



The previous method of construction, however simple, I have 

 not before met with in any work on isometrical perspective. The 

 rule by calculation is easily deduced from the construction, mak- 

 ing the isometrical diameter, or the given diameter, of the circle 

 equal to unity : Thus, because the isometrical angle O A B = SO'', 

 therefore BO = i A C. Hence OA = V(l" — '25) =-866; 

 therefore, « O- = A O X M O = SGR X -433 = -3749. By 

 extracting the square root of each side, we have n O = '612; con- 

 sequently, the transverse diameter ac z= 1-224. Also, B O = 

 ^ B C = -5 ; therefore, BOx|BO = 6 0=; hence i O = -3535 ; 

 consequently, the conjugate diameter 6rf = -707. These are the 

 numbers as given in the rule. I have not seen this rule in Jopling's 

 treatise, but there is a table given of diameters, with the same 

 figures to the Diameter 1. It will be observed, that all the lines 

 that are in the figure are not required in the construction, but 

 only the two diagonals : the other lines are only drawn to assist in 

 the demonstration. The method given by Professor Farish in his 

 paper on Immetrical Perspective, in Gregory's '- Mathematics for 

 Practical Men," is very tedious, both by construction and calcula- 

 tion. 



EXPERIMENTS ON CEMENT. 



A good deal of attention has recently been directed to the merits 

 of a cement called " Portland " cement, manufactured by Messrs. 

 Aspden and Robins, of Northfleet ; and on INIonday, the 18th ult., 

 a numerous body of architects, builders. &c., assembled at the 

 town premises of these gentlemen, in Great Scotland-yard, to 

 witness a number of experiments with the cement, both alone 

 and in combination with sand, in different proportions ; the follow- 

 ing are some of the trials made : — 



Best Stock Bricks Cemented against the Wall. 



Experiment 1. — 17 stock bricks were cemented together with 

 romnn cement (all cement) and projected before the face of a wall, 

 as fig. 1. They broke down with 7 lb. placed on the end. 



Fig. 1. 



Experiment 2. — 11 stock bricks, cemented together with 1 sand 

 and 1 roman cement, broke down with 7 lb. placed on the end. 



Experiment 3. — 38 bricks, cemented with neat, patent portland 

 cement, broke down with 141b. placed on the end. 



Experiment 4. — 30 bricks, cemented together with 1 portland 

 cement and 1 sand, broke down with 15 lb. at end. 



Experiment 5. — 22 bricks cemented together with 1 portland 

 cement and 2 sand, broke down with 168 lb. at end. 



Experiment 6. — 25 bricks, with 1 portland cement and 4 sand, 

 broke down with 56 lb. at end. 



