TRANSACTIONS OF THE SECTIONS. 83 



express the total heats possessed by equal weights (we may say by 1 granune) of 

 dilFereat elements by the atomic weights of the corresponding elements, I obtained 

 the numbers tabulated in the last column, in which we at once observe a remarkable 

 coincidence of the numbers for the elements cadmium, tin, and lead. Besides, it 

 deserves to be noticed that the second figure is in all cases either 0, 1, 2 or 8 ; so 

 tliat if we retained only the first figure, with the second as a decimal; the numbers 

 obtained woidd not differ very much from whole numbers. 



As the experimental data which I have used, although the best obtainable, leave 

 much to be desired, and as from the recent experiments of Weber it appears that 

 the specific heat varies with the temperatm-e, I will not venture upon any fui'ther 

 remarks, but confine myself to drawing attention to the fact that between the total 

 heats possessed by different bodies in comparable conditions, there seem to exist 

 numerical relationships which possibly may come out more clearly when more data 

 are be known. 



On cm improved form of Filter Pump. By T. E. TnoEPE, F.B.S.E, 



In the St. Petersburg Correspondence of the ' Berichte dor Ueutschen Chemischen 

 Gesellschaft ' (No. 7, 1872), Dr. Mendelejeff described a new filter pump, constructed 

 by Urn. Jagno, of Moscow. It consists of a tube about 1 metre in length, and from 

 8-10 millimetres in diameter, to the side of which, at about 3-5 centimetres from 

 the upper end, is affixed a side tube about 5 centimetres in length. The upper 

 end of the vertical tube is cut slantwise, and is connected by means of a strong 

 caoutchouc tube with a stopcock in connexion with the water-supplj'. In the hori- 

 zontal side tube is also fixed a caoutchouc tube at least 1 centimetre in outside 

 diameter, the walls of which must be not less than 2-3 millimetres in tliickness. 

 The end of this caoutchouc tube is pushed within the horizontal side tube, and ends 

 in a Bunsen's valve ; i. e. a piece of glass rod is inserted into the end, and the tube 

 is cut by a single blow on a chisel. The edges of the slit are thus sharp, and on 

 outward pressure being applied to the tube, they readily and completely adhere, 

 making a perfectly air-tight conjunction. The other end of the caoutchouc tube is 

 connected with the vessel to be exhausted. On allowing water to flow through the 

 vertical tube, the caoutchouc tube rapidly pulsates from the opening and shutting of 

 the valve. Energetic suction is thus set up ; and it is easy by the fall of water 

 through the 1 metre to obtain a vacuum equivalent to 700 millims. of mercury. The 

 working of the apparatus obviously depends upon the principle of the hydraulic 

 ram ; it is readily set up at a small cost, and will doubtless take the place of the 

 Bunsen filter pump, as it obviates the necessity of a fall of 30 feet. There are a 

 few disadvantages connected with the use of the caoutchouc valve above described : 

 owing to the diminution of its elasticity by long-continued working, its efficacy 

 diminishes after a time ; it not only fails to bring about rapid exhaustion, but it 

 permits of the back-fiow of the water so soon as the conjunction of its edges ceases 

 to be perfect. To obviate these inconveniences, the author has devised an improved 

 form of valve. At the end of the side tube is a funnel-shaped cone of metal, pierced 

 near its apex with a number of holes ; into the cone is fitted a sheet of unvulcanized 

 caoutchouc, shaped like a filter ; this presses against the sides of the cone, and 

 ett'ectually prevents the entrance of air or water from without. The' slightest 

 pressure from within is sufficient to disturb the adhesion of the caoutchouc and cone, 

 and to allow of the ti-ansniission of air through the holes. 



This form of valve is of a more permanent character than the other, and allows 

 of a more rapid exhaustion. In the new form of the instrument a manometer is 

 attached to the side tube to ascertain the degree of exhaustion ; and by a screw and 

 spring the rate of exhaustion can be regulated with the utmost nicety. Further, 

 bv means of a clamp arrangement, the vacuum within the pump can be maintained 

 without disturbing the screw, if it should be suddenly necessary to_ disconnect the 

 caoutchouc tube from the piece of apparatus to bo evacuated. This form of filter 

 pump has the great advantage of portability over the original one of Bunsen ; it 

 may be constructed in such a manner that it can be readily transported to any part 

 of the laboratory ; and it necessitates no alteration in the existing arrangements of 

 pipes and fittings. 



