WATER-BELLOWS. 



the tube where it becomes nine inches diameter,) ten vent 

 holes are bored through the fides of the tube ; they are cy- 

 lindrical, and two inches diameter ; their direftion is in- 

 clined, fo that they point downwards at about an angle of 

 45 degrees ; they are arranged at equal dillances round the 

 tube in two rows, the upper row having fix holes, and the 

 lower row four : it is through thefe holes that the air enters. 

 The tube is fupported in a vertical pofition by a framing, 

 and the lower end is introduced into a ftrong ton or caflc, 

 fix feet deep, and almoft as much in diameter, though it is 

 rather fmaller at top than at bottom. The tube defcends 

 through the head of the calk i8 inches, fo that it terminates 

 within 4^ feet of the bottom of the callc ; and a kind of 

 table made of a flat round ftone, or a plate of caft-iron, is 

 placed horizontally in the centre of the ca(k, at i8 inches 

 beneath the orifice of the tube, being fupported by a crofs 

 of wood, placed upon four legs, from the bottom of the 

 ca(l<. The cade is well clofed on all fides, particularly 

 round the tube, where it pafles through the head ; but there 

 is an air-pipe condufted away from the top of the caflc, to 

 convey the air to the furnace ; and from the bottom of the 

 caflv there is an opening, by which the water can pafs away. 

 The opening is regulated by a wooden (huttle, which pens 

 up the water to fuch a height within the cafl<, that the 

 opening through which the water iiTues will be always be- 

 neath the furface of the water, fo as to pres-ent the efcape 

 of the air by the fame paflage. 



The aftion of this machine is not fo eafy to explain as its 

 ftrufture, and it has at various times occupied much of the 

 attention of philofophers. Father Kircher was the firft 

 who defcribed the machine in his Mundus Subterraneus ; 

 but he did not fatisfaftorily explain the reafon of its aftion. 

 In the Memoires des Scavant etrangers, Barthes, the father, 

 has given a theory which is very defeftive ; and Dietricli 

 was of opinion that the air was produced by the decompofi- 

 tion of the water. 



M. Reaumur explains it thus : — The funnel of the tube 

 is always full of water, which ilfues rapidly through the 

 throat ; but finding immediately a larger place, the ftream 

 difperfes and fcatters into drops, becaufe it is no longer en- 

 clofed within a cyhndrical furface : it does not, therefore, 

 take any conftant figure, but the ftream is compofed of dif- 

 ferent fmall Itreams, or rather fucceffions of drops, which 

 are continually changing their pofition with refpeA to each 

 other. Now the intervals between thefe feparate ftreams or 

 drops are occupied by the air which is within the cavity of 

 the tube : fuppofe that between two ftreams feparated by 

 air a third comes to defcend, it will pulh the air before it 

 with all its force, and carry the air down to the cafk ; ana 

 this will be replaced by frefh air, entering at the vent-holes. 

 The irregular arrangement which the ftreams or drops take, 

 cither at their iffuing from the throat or in continuing their 

 fall, is fuch that few drops do not carry fome air down be- 

 fore them into the cafit : the water faUing upon the table 

 within the caflc dafties on all fides, and releafes the air which 

 rifes in the cadi, and iflTues through the air-pipe to the fur- 

 nace, whilft the water falls to the bottom of the cafk, and 

 efcapes gently through the fluice. 



A fingle trunk of the dimenfions juft defcribed is found 

 fufBcient to blow a forge or finery ; but for a fmelting fur- 

 nace, three are joined together, having a common trough of 

 fupply, and the air-pipes from the three callcs are joined to- 

 gether. M. Reaumur fuppofed that a greater height of the 

 tail would produce more air, becaufe it is longer expofed to 

 thofe changes of pofition in the different ftreams of water ; 

 b-ut he fuppofed that no adequate advantage would be 

 gained by an increafe of the diameter of the tube, becaufe 



3 



it would be more likely, in falling in a large body, to de- 

 fcend in a clofer column. 



The machines of the Pays de Foix are fomewhat dif- 

 ferently conftrufted : in thefe the water is conveyed into s. 

 refervoir, from the bottom of which a fquare trunk or tube 

 defcends to the refervoir or air-cheft, which is made very 

 long ; and the air-pipe proceeds from an elevated part of it, 

 to prevent the danger of fpray or fmall drops being carried 

 into the furnace. Inftead of a throat and the vent-holes, 

 the tube is made to divide into two branches, at the point 

 where it paffes through the bottom of the upper refervoir :. 

 thefe branches rife above the furface of the water in the re- 

 fervoir, fo that it cannot enter into them, but the water is 

 admitted at an opening between thefe two branches, fo that 

 in effeft the tube is divided into three, the centre being an 

 opening for the water to defcend, whilft the two outfide 

 branches admit the air to mix with the water and go 

 down. 



The editor of the Art des Forges fuppofes that the vent- 

 holes are ufelefs, but that the violent agitation of the water 

 in palling the throat, and daftiing upon the table within the 

 caflt, is fufficient to change the water into air. This is the 

 fame hypolhefis as that of Dietrich. 



Thefe various explanations rendered the fubjeft ftill more 

 obfcure ; and in 1 791, the Academy of Touloufe invited 

 philofophers to determine the caufe and the nature of the 

 ftream of air which is produced in thefe machines. M. 

 Venturi, profeffor of philofophy at Modena, gave the real 

 anfwer in an excellent paper on the principle of lateral com- 

 munication of motion in fluids. 



To explain the principle, this philofopher fuppofes a 

 number of equal balls to roll along in a horizontal trough, 

 in contaft with each otlier, with an uniform motion at the 

 rate of four balls in a fecond : fuppofe, on arriving at the 

 end of the trough, they fall fuddenly to a depth of 16 feet. 

 Now, from the laws of gravity, each ball will perform this 

 defcent in a fecond of time ; and as four balls fucceed each 

 other in each fecond, it follows that there will always be 

 four balls in the air at the fame time. The relative politions 

 of thefe will be as follows : the uppermoft ball will be one 

 foot from the point where they begin to fall, the fecond 

 four feet, the third nine feet, and the fourth fixteen feet. 

 This arifes from the acceleration which always takes place 

 in defcending bodies. A confideration of this circumftance 

 will give a proper idea of the difunion and fuccefGvc fepara- 

 tion of the particles which the accelerating force of gravity 

 produces in fluids, or in bodies which fall in a ftream. 



The rain-water flows out of gutters by a continual cur- 

 rent ; but during its fall, it feparates into portions in the 

 vertical direftion, and ftrikes the pavement with diftinft 

 blows. The water likewife divides, and is fcattered in the 

 horizontal direftion. The ftream which ifl^ues out of the 

 gutter may be one inch in diameter, and ftrike the pavement 

 over the fpace of one foot. The air which exifts between 

 the vertical and horizontal feparations of the water which 

 falls is impelled, and carried downwards. Other air fuc- 

 ceeds laterally ; and in this manner a current of air or wind 

 is produced round the place ftruck by the water. M. Ven- 

 turi went to the foot of the cafcades which fall from the 

 Glaciere of La Roche Melon on the naked rock at La No- 

 valefe, towards mount Cenis, and found the force of the 

 wind to be fuch as could fcarcely be withftood. If the 

 cafcade falls into a bafon of water, the air is carried to the 

 bottom, whence it rifes with violence, and difperfes the 

 water all round in the form of a mift. 



He formed one of thefe artificial blowing engines of a 

 fmall fize ; the vertical pipe was two inches in diameter, and 



four 



