L. MECHANICS AND ENGINEERING. 495 



similar cylinders are then slipped successively over and into 

 each other until one is built up as thick and as long as may 

 be required. The flasks in question were made in four lay- 

 ers. The loose shells thus put together are fastened into a 

 solid cylinder by means of pure tin, which is melted and 

 worked in from the inside with the aid of gas blow-pipes. 

 The particular diameter given to these flasks is less than that 

 which would have been adopted had not Mr. Matthews had 

 facilities for making this size only, he having been for a num- 

 ber of years engaged in making similar soda-water reservoirs 

 of sheet steel soldered with tin. There would, however, be 

 no serious difficulty in the way of making such cylinders of 

 any required diameter. All of the flasks furnished by Mr. 

 Matthews were to bear 2000 pounds to the square inch, and, 

 in addition, an extra flask was ordered to be made exactly like 

 the other, and to be tested to destruction. This flask gave 

 way under a pressure of 3136 pounds. The rupture consisted 

 of the tearing of the sheets irregularly without regard to the 

 joints. The heads and the junctions of the heads to the 

 body were not affected. In every respect this form of holder 

 seems to combine strength with the assurance of freedom from 

 hidden flaws. Pamphlet by W. JV. JIM, Newport^ 1875. 



THE TIIRUST OF EMBANKMENTS. 



Professor Boussinesqhas communicated to the Royal Acad- 

 emy of Belgium a theoretical essay on the elastic equilibrium 

 of masses of powder, and on the pressure of earths devoid of 

 cohesion. These investigations bear directly upon the ques- 

 tion of the determination of the thickness necessary to be given 

 to a wall destined to sustain the pressure of an embankment 

 of earth. The problems in question were first resolved ap- 

 proximately by Coulomb in 1873, and very many distinguish- 

 ed inquirers have, since his time, occupied themselves with 

 this important subject. Most of these have assumed the fol- 

 lowing hypotheses : viz., that the wall, when it is overturned, 

 is borne down by a prism of earth, which with the wall slides 

 over the remaining earth and the lower portions of the wall in 

 a direction parallel to the plane of rupture. Coulomb's so- 

 lution of the question dispenses with the above hypothesis, 

 which is, in fact, generally inconsistent with actual experience, 

 but assumes that at the commencement of the rupture of the 



