870 REPORT— 1892. 



Hence I think that electrical windmills, at least for small installations, are likely 

 to assume this form, as they require little or no attention. It is obvious, however, 

 that mills of great size and strength of this form may be constructed, and, if neees- 

 sarv, several could be placed in any well-exposed position, each having its own 

 dynamo and set of accumulators. Regarding the electrical connections, only a 

 word is necessary. The only thing needed is that the circuit be broken when the 

 dynamo is running at less than the storing speed. This is easily managed by 

 having a governor attached to the dynamo shaft which makes and breaks contact 

 in a mercury pool at the required speed. I have also tried a foi'm of governor 

 which throws a greater or lers number of cells into the charging circuit as the 

 ■wind varies, and in this way the machine is always doing some work. 



G. Flexible MefaUic Tubing. Bij Gilbert R. Redgrave, As^oc.Inst.C.E. 



The author states that numerous etTorts have been made in recent times to 

 produce tubes of metal capable of being bent into any required form. The con- 

 ditions under which it becomes possible to obtain tlexibility in tubing are investi- 

 gated, and it is pointed out that this property depends upon a multiplicity of small 

 .segments or moving paits, securely jointed together, provision for movement being 

 made at the joints. In tubes of rigid metal various methods of forming these joints 

 are considered, and it is shown that success can best be obtained in tubes composed 

 of spirally wound strips or fillets of metal, which must be so disposed that the longi- 

 tudinal joint is not only sound, but is possessed of play or sliding power within 

 certain fixed limits. Owing to the great difficulty experienced in making a good 

 joint of metal against metal, all the earlier attempts to make a flexible joint de- 

 pended upon the presence of a strip of indiarubber inserted between the coils. In 

 con.sequence of the liability of indiarubber to deteriorate when employed in this 

 way, it became necessary to devise means to make a tight joint without rubber, 

 and this has now been accomplished. The new flexible metallic tubes are the in- 

 vention of Mons. E. Levavasseur, and he was led to the discovery by his successful 

 attempts to produce necklaces and bracelets of the well-known snake pattern. In 

 the first instance thise ornaments were manufactured by means of spirallv wound 

 strips of thin metal of two diHerent sections, so arranged that the joints produced by 

 the lower fillet were covered by a second fillet wound above it, thus, ^^r~L_ni_r~~. 

 At a later date Mons. Levavasseur manufactured similar objects by the use of a 

 single spirally wound fillet in the shape of the letter S, the coils being so disposed 

 that tht head of one S was clasped by the tail of the next, and from various modi- 

 fications in the form of the metal strip many beautiful and flexible chains were thus 

 produced, some of them being subsequently passed beneath a hydraulic press and 

 flattened into an oval section. It will readily be understood that all these tubes 

 were relatively very small in size, and made no pretence of being perfectly tight. 

 About six years ago, however, the inventor, on obser\ ing the clumsy contrivances 

 to obtain flexibility in the hose used in Paris for watering the roads, was induced 

 to turn his attention to the fabrication of pipes or tubes of larger dimensions, and 

 this he at first attempted to do by employing the S section already used lor 

 jewelleiT, with the addition of a thin strip of indiarubber inserted along the joint. 

 Some very SQCcessful tubes were made in this way, but finding that the rubber 

 "was liable to become rapidly worn out, and that in certain cases it soon oxidised 

 or became useless, Mons. Levavasseur determined to prepare a tube in whicli 

 rubber could be dispensed with. To do this an entirely new section was designed, 



which may be described as a , — ^ | | double channel. In coiling this strip the 



smaller notch fits into the larger one, and room is allowed for a certain amount 

 of play, so that a species of piston joint is obtained, as the small notch works 

 within the larger one in a very similar way to that in which the piston travels in 

 the cylinder. In tubes thus made, it becomes possible to omit the rubber and to 

 obtain a perfectly tight joint between the metal surfaces. The amount of play or 

 flexibility depends to some extent on the size of the pipe, the thickness of the 



