884 REPOET — 1893. 



■ proportion of gems to worthless pieces of mineral is not a percentage mere!}', but 

 of one to ruanj' tliousands, the utility of such a machine is obvious. 



The. concentrated gravel when washed is most carefully classified into sizes, 

 beginning, for diamonds, at one-sixteenth of an inch, and increasing by sixteenths 

 up to tive-eighths of an inch, or still further if desired. Each size of gravel is fed. 

 into a separator adapted to suit it. The separator has no moving parts — and takes 

 advantage, by means of a stream of water running through it, of the slight varia- 

 tion in specific gravity between the gems (3 5 to 4) and the worthless minerals 

 (2-5 to 3). It is possible to .separate such substances by immersing them in a pre- 

 pared solution of high specific gravity, just as pebbles and chips may be separated 

 in water ; but there are practical dithculties about such a process, and the gem- 

 separator described substitutes a moving current of water for the heavier solution, 

 with the advantage that the process is continuous, the separated materials beino- 

 depo.sited in their proper receptacles, those for the gems being guarded by locks. 

 The details of the machine are described, and a machine shown at work, 

 The operations of the rnachine are not confined to gems. The separation cf 

 any minerals from their gangue, provided always there is a slight difference in 

 specific gravity^ may be efi'ected, and the machine will work on broken material 

 in a dry or merely wetted state or on slimes run in with a stream of water. 



3. On some Experiments ivith Ventilating Ferns or Air-propellers. 

 By William Geokge Walker, M.Inst.M.E. 



These experiments have reference to those ventilating fans or air-propellers of 

 the screw-propeller type used for low-pressure ventilators. 



The primary object of the experiments was to test the efficiency of fans or air- 

 propellers, differing only from each other in the cross-section of their blades, which 

 section chiefly referred to_ the rearward or non-propelling face of the blades or 

 vanes. 



The first series of experiments were made with air-propellers 14 in. diameter 

 and 21 in. pitch, and of two, three, and six blades respectively. Each propeller 

 was tried at progressive revolutions, varying from 500 up to 2,000 per minute. 

 The blades were composed of sheet brass ^^ in. thick. 



The second series were made with the same propellers, but having a curved 

 convex protuberance fixed to the back surface of the blades, forming a section 

 which is a hollow plano-convex form, the convex surface constituting the non- 

 propelling surface, or backs of the blades. 



In all cases the efficiency of the blades was increased by the addition of the 

 convex surface ; in some cases the number of cubic feet per revolution was nearly 

 doubled, the power being the same in each case. 



Some further comparative experiments were made with fans of 2 feet in dia- 

 meter, and with blades of diflferent sectional form, viz., (I) flat blades, (2) curved 

 blades, (3) helical blades, (4) flat blades with round protuberance fixed at back,. 

 (5) curved blade with round protuberance fixed on, also other sections. 



The angle of the blades and the area were the same in each case. The experi- 

 ments most distinctly showed that a very great gain was obtained by the use of 

 the convex surface, and that the best results were obtained with a section of 

 concavo-convex form. 



_ The reason of the results may perhaps be explained from the ' Stream Line"" 

 principle. It seems that the convex surface at the back tends to fill up or destroy 

 the partial vacuum which exists at the back of each revolving blade. The existence 

 of an eddy in the wake of each blade must increase the rotary motion of the 

 air, caused by air passing through the propeller clinging or tending to rush into 

 the partial vacuum. 



4, On the Testing Machine and Experimental Steam Engine in the 



Engineering Laboratories of TJniversity College, Nottingham. 



By Prof. W. Robinson, 



