252 A. deF, Palmer^ Jr. — Condensation in the Steam Jet. 



where the abscissas are times in seconds and the ordinates 

 values of D-3 in millimeters, the pressure being S'S*^™ and the 

 temperature of the surrounding air 25-5° C. The other series 

 gave very similar results, and, while the time of exposure was 

 too great to allow an accurate determination of the period in 

 any case, the periodic nature of the phenomenon is established 

 beyond doubt. If a series of photographs could be made by 

 extremely short exposures rapidly following one another, it 

 would probably be possible to accurately determine the ampli- 

 tude and period of the function connecting D with the time^ 

 and hence also the dependence of these quantities on the actu- 

 ating pressure of the jet. An attempt was made to accomplish 

 this by utilizing the rapid discharges of a large Leyden jar bat- 

 tery as a means of illumination, but in no case was the light 

 sufficient to produce measurable negatives. 



Provided the temperature and dust content of the surround- 

 ing air remain constant, the velocities Y^ of the ascending 

 vapor at any point of the jet and Y^ of the descending con- 

 densation may be expressed by the formulae 



V,=/,(H,P) 

 and V,=/,(I),P) 



where H indicates height above the nozzle and the other letters 

 retain the meaning already ascribed to them. If 



H = D = Constant 



and if P is constant it is obvious that Yi=Y2, but if 



H = D = a; 



Y2 will be greater or less than Y^ according as a? is a decreasing 

 or an increasing quantity. 



The foregoing observations indicate that, the pressure being 

 constant, a? is a periodic but not a simple harmonic function of 

 the time, and it is possible that the phenomena may be graph- 

 ically represented as in fig. 6, where the abscissas indicate time 

 and the ordinates values of x. At the origin Y^ is zero. The 

 steam, issuing from the nozzle, ascends with a gradually 

 decreasing velocity and increasing degree of supersaturation 

 until condensation commences, either spontaneously or induced 

 by the entrance of dust particles from the surrounding air. 

 The curve AB is thus traced, the velocity Y^ at each point 

 being equal to hx/ht and represented by the tangent to the 

 curve. Beyond B the velocity which determines the value of 

 X is obviously the resultant of Y^ and Y^. At first Y, is very 

 much the greater, and the representative point rapidly descends 

 along some such curve as BC. During this operation the sue- 



