R. W. Wood, Jr. — Combustion of Gas Jets. 481 



rapid condensation, considerable difficulty was experienced in 

 keeping the pressure uniform during the observations. The 

 tubes leading from the fiask to the manometer had to be kept 

 hot in order to prevent fluid ether instead of vapor from being 

 delivered at the orifice, which in this case was smaller, being 

 about the size of a small sewing needle. 



With coal gas this jet would stand a pressure of 20 cms without 

 going out. With ether vapor -J of a centimeter gave a flame 

 like a candle, while -| cm caused the flame to retreat about a 

 centimeter from the orifice and burn in mid- air. At T 7 ¥ cm 

 pressure the distance had widened to four centimeters, and the 

 vaporized ether could be seen rising as a cylinder the size of a 

 knitting needle surmounted by a hollow blue flame, fig. 4. At 

 l cm pressure the flame went out ! It behaved very much as 

 coal gas did when burnt in bad air, the flame showing no tend- 

 ency to form a tube under increased pressure. This seems to 

 be due to the fact that ether vapor requires a greater amount 

 of oxygen for its combustion. In pure oxygen it behaves very 

 much as coal gas does in air. Alcohol vapor comes between 

 the two, the flame standing a pressure of about 3 cms without 

 being extinguished, and showing a tendency to form a tube. 



The nature of these flames was studied with a revolving 

 mirror, but no interesting features were revealed except in the 

 case of the coal gas under high pressure. The action of the 

 rotating mirror is to spread out the flame in a broad band, 

 giving as it were an infinite number of instantaneous views, 

 placed side by side. If the upper parts of the jet move slower 

 than the lower, they will suffer lateral displacement, and the 

 flame images will be curved backwards. The flame shown in 

 fig. 2 C when examined in the mirror appeared as in fig. 5. 

 The serrated part of the flame represents the contracted por- 

 tion, and the alternate dark and light spaces show that this 

 part of the flame alternately ignites and goes out, with sufficient 

 speed, however, to keep the gas above in a state of continuous 

 combustion. The long narrow band below is the burning tube 

 of gas, and the bright teeth are rapid flashes of fire uniting 

 this with the burning mass above. Under greater pressure the 

 frequency of these flashes becomes reduced to such a degree 

 that they are appreciable to the eye. This is the case in flame 

 D (fig. 1). Here the upper portion alternately ignites and goes 

 out with a rapid snapping noise, the pulses varying from one 

 to ten a second. In the rotating mirror this flame presents the 

 appearance shown in fig. 6, the flashes being so slow that they 

 are completely isolated from each other by the mirror, each 

 one being seen as a broad flare of light. The line representing 

 the tube of fire is narrower now, the pressure being greater. 

 The slanting base of the " flare " is bright blue, above this is a 



