202 



SCIENCK 



[N. S. Vol. XI. No. 267. 



with dust be taken as one. To determine 

 the constant of integration it is to be ob- 

 served that when a; = 0, or A; = 0, there is 

 no absorption ; but the current of air must 

 nevertheless be kept up passing either 

 through the absorption tube or not, to re- 

 tain a colored held in the color tube. Now 

 saturated dusty air would be objectionable 

 seeing that an opaque field is produced 

 unsuitable for measurement. Hence even 

 when a; ^ 0, or A; = 0, the influx current is 

 unsaturated, and we may suppose this 

 minimum current to be obtained through a 

 fixed initial length of tubing, .t„, of the same 

 kind. Thus the tube virtually begins at 

 — a;„, and equation (1) becomes on integra- 

 tion 



n= e rv ('^j 



If the observations be so made that the 

 dust density issuing from the length x of 

 tubing is constant, we have for a given 

 color tube, n = n' and 



X +x^ _x' + x„ 



(3) 



For different tubes under the same con- 

 ditions, 



k/k' = ^^ /^. 

 X + x„ / x' + Xg 



Hence, ignoring the undeterminable con- 

 stant, one may write briefly 



2k = vr/(x + xJ. (4) 



With this preliminarj^ theory as a point 

 of departure, I have been making an ex- 

 tended series of observations on the con- 

 densation of supersaturated steam obtained 

 from jets, and following the general method 

 of color tubes described in my memoir * on 

 the subject published by the U. S. Weather 

 Bureau in 1895. By passing measured 

 quantities of air ( F litres per minute), sat- 

 urated with the emanation of phosphorus, 



* A Report on the Condensation of Atmospberio 

 Moisture ; Bulletin, No. 12, pp. 104, U. S. Weather 

 Bureau, Washington, 1895. 



through different lengths of absorption 

 tubing,* and regulating the air current by a 

 graduated stop-cock till a definite color 

 (full blue) appears in the field of the color 

 tube, the condition n = constant is fulfilled, 

 on provision that no change has occurred in 

 the action of the tube during the interval. 

 As Fis the influx of dusty air per minute, 

 1000V60=:rr'D, and as 2k = vrl{x + x^), 



^^2^ F 



r ' x + x^' 



where V/{x + a;J is given by the observa- 

 tions of volume in terms of length of ab- 

 sorption tube for a fixed blue in the field of 

 the color tube. If V„ be the minimum vol- 

 ume, i. e., the influx volume of dusty air 

 corresponding to a; = 0, A: = 2.65/r ■ VJx„ ; 

 hence, finally, 



2-65 F- F, 



k = 



(5) 



Aside from the difficulty of observing 

 subject to color criteria, equation (3) is 

 well borne out by the data obtained, to the 

 extent that F is a linear function of x. F,,, 

 however, does not always coincide with ob- 

 servation, and a certain additional tube 

 length must in certain cases be added to 

 compensate for the dissipation (eddies?) 

 encountered on entering the absorption 

 tube. But this is non-essential. 



A brief summary of the present results is 

 given in the following table, which shows 

 the material of the absorption tubes, their 

 diameter, the variations of length (a;) and 

 volume ( F) occurring in each series of ex- 

 periments (usually 10 or 20 in number). 

 The pressure under which steam issues 

 from the jet is shown under p (centims. of 

 mercury), and the temperature (°C.) of the 

 air flowing into the color tube, under d. 

 Within reasonable limits, discussed else- 

 where, the color tube is not sensitive to 



* Ordinary tubing used for absorption purposes, 

 i. e., to catch the dust particles moving laterally out 

 of the dust-laden air current. 



