

22 



KXPKUIJIENTS WITJI lUMZEl) AIIJ. 

 PART IV.— SUPPLEMENTARY DATA. 



Color. 



Pale Violet 



Yellow 



Green 



Light Blue 



Dark Blue (.50!;= i). 



Violet 



Opa(iue 



Pale Violet 



Green-Yellow . . . 



Blue-Green 



Light Blue 



Blue(.57;;;=i). 



Violet 



Dark Violet 



Dull Yellow .... 



Volume 

 Ratio. 



.2.; 



■5- 



.80 



1. 00 



1.60 



2.80 



•30 

 •39 



.40 



•6S 

 1. 00 



1.27 



1.46 



4.75 



Bulk. 



417 

 1.92 



' 9- 



■•-•5 

 1. 00 



.62 



•36 



3-33 

 2.56 



2.50 



J-54 

 1. 00 



■79 

 .68 



L>iameler. 



1.61 

 1.24 

 1.24 

 1.08 

 1. 00 



■85 

 ■7' 



I.4V 

 1-37 

 1.36 



115 

 1. 00 



•93 



.ss 



.6c 



/• 



5 cm- 



24 



27" 



Series No. 



23 



i. Results Charted. — Some of these data are given in the accompanying 

 charts, figures 3 and 4, and the attempt is made to distinguish the different series 

 by different forms of dots. Part 2, made undei" the luost favoi-aide conditions and 

 giving the best residts, is marked by fidl dots; part 3 by open dots. 



In the construction of these data, the vohimes per minute (on the average 

 a])(>ut twice tlie actual numbei' of liters/min.) of saturated phosphorus emanation 

 aie made the ordinates, cori-esponding to the colors given by the abscissas. To 

 have a scale in the latter case I took the thickness d, of air plates giving, on 

 normal incidence and transmitted light, the identical color in case of Newton's in- 

 terferences. In figure 3 the data are on a large scale, but reach only as far as 

 opaque. In figure 4 the whole curve and remaining data aie constructed on a 

 smaller scale, L/nt., denoting liters per minute. The two curves of figure 3 refer 

 respectively to lower and higher temperatures 6, or to the 2d and 3d parts of table I. 

 The latter is decidedly the better, showing some definiteness of a locus, quasi-hyper- 

 bolic in charactei'. The low-temperature data aie diffuse, due, I think, to the 

 weak and variable ionizing activity of phosphorus at low temperatures. During 

 the long period of olisei'vation there must have been some corrosion of the jet. 

 This is probably one of the chief reasons why the dust contents (lit./min.), corre- 

 sponding to- the standard blue, vary. In the second place, the air of the room 

 does not retain a fixed degree of purity. 



5. T)i.'<ci(ss<ion. — In consulting these charts for practical 2)uriioses, curves ob- 

 tained in joining points of the same kind should alone be used, as the reasons for 

 the differences betw^een the series are naturally very complicated. In spite of all 

 care bestowed on the observations, they are insnfiicient to suggest the form of locus 

 with certainty. Their usefulness will therefore be confined to testing the theory 

 of the phenomenon when such a theory is forthcoming. In a general way the locu.s, 

 beginning with opaque and extending toward the right into the fainter secontl 



