Chemistry and Physics. 71 



of the nitrogen and other flame gases this does not take place 

 nntil the top of the non-luminous zone is reached, where at a tem- 

 perature of a little over 1000° the decomposition occurs with an 

 increase of temperature, and the liberated carbon, being heated to 

 incandescence, gives the luminosity to the flame. The author 

 divides the flame into three zones. In the inner zone the tempera- 

 ture rises to 1000° to 1100° at the apex, and here the gaseous con- 

 stituents undergo various decompositions culminating in the pro- 

 duction of acetylene, and some hydrogen and carbon monoxide. 

 In the middle or luminous zone the temperature ranges from 1100° 

 to 1300°, and here the acetylene formed in the inner zone is 

 decomposed with the deposition of carbon, which at the moment 

 of separation is heated to incandescence by its own combustion as 

 well as by that of the hydrogen and carbon monoxide and so gives 

 luminosity to the flame. In the outer zone the cooling and dilu- 

 ting influence of the entering air render a thin layer non-lumin- 

 ous and finally extinguishes it. With reference to the loss of 

 luminosity in the Bunsen flame the author attributes it : (1) to 

 the chemical activity of the oxygen of the air which burns up the 

 hydrocarbons before they can form acetylene; (2) to the diluting 

 influence of the nitrogen which increases the temperature required 

 to form acetylene; (3) to the cooling effect of the air introduced. 

 — J. Chem. iSoc, lxi, 322, April, 1892. g. f. b. 



2. On the measurement of Osmotic Pressure. — It is well 

 known that the osmotic pressure of a salt solution calculated 

 from the electrolytic dissociation formula of Avrhenius does not 

 agree with that observed directly, but is always smaller. Thus 

 for potassium nitrate, the osmotic pressure as observed by Pfeffer 

 in a solution containing 08 gram in 100 grams, is 1304 mra , while 

 that calculated is 2530. For a one per cent solution of potassium 

 sulphate, the observed value is l758 mm , the calculated is 2480. 

 TAiiiiANJf has investigated the cause of this, and has pointed out 

 the fact that it is due to the permeability of the membrane of 

 copper ferrocyanide used in the experimental determinations. 

 Obviously if this membrane is not perfectly impermeable to the 

 salt employed (or to its ions) the observed osmotic pressure will 

 be less than it really is. Hence by determining the rate at which 

 the salt is diffused through the membrane a correction may be 

 obtained for the observed values. Moreover, according to the 

 author, the copper ferrocyanide membrane may be obtained in 

 two distinct forms. The first of these when fresh is transparent, 

 elastic and extremely thin and allows water to pass through it 

 freely. The second is opaque, dark brown in color, only slightly 

 elastic and much less permeable to water. This form is produced 

 when the solutions of copper sulphate and potassium ferrocyanide 

 remain in contact with each other for a long time on the sur- 

 face of a porous tile. The presence of sodium sulphate in the 

 copper sulphate solution facilitates its .formation. For the best 

 results the solutions should be more concentrated than those sug- 

 gested by Pfeffer ; say a normal solution of copper sulphate and 



