﻿960 



Dr. L. Silberstein and Mr. Trivelli on the 



The fine spherical grains of Svedberg could have only at 

 the utmost a point contact, and this might not have been 

 intimate enough. It is even credible that in view of the 

 Brownian motion of these minute bodies there was actually 

 no permanent contact between them, as becomes very likely 

 from Svedberg's remark on page 185, that "even over such 

 a small distance as 1 micron no noticeable transport of silver 

 ions takes place/' 



At any rate, we have found in our case the property of 

 clumps as units well verified. Without attempting to 

 reproduce in this place all our available evidence *, we may 

 support and illustrate the said principle by the following- 

 data. Fig. 1 represents the frequency curve and, in the 



Fig. 1. 



u 150 600 — 



ft. 100 400 

 o E 



";i 75^300- 



£ 5CS200 



W12C 



3154 GRAINS MEASURED. 



5.71 . I0 8 GRA!NS PER SQ.CH. PLATES H. 



X= 1.04 Ji 2 CALC. X = M!U 2 QBS. 

 o = S!ZE-FREGUENCY PER 1000 SR Al US. 03S 

 A = S!ZE-FREQUENCY PER 1000 GRA1HS.C A5.C 



U = 120e-° 8, 0<-o.e$ 2 



B=SI2E-AREA PER SQ.CM. PLATE. 



YsCtte-oBKx-oeS 2 c _ _120]5_ 



toco 



J^. 



inset, a microgram of a sample of grains of the aforesaid 

 W-12-C emulsion. This emulsion was spread over the glass 

 plate in a single layer so as to obtain the maximum number 

 of grains per unit area with the least possible overlapping. 

 The emulsion, after the coating, was kept in its liquid state 

 long enough to enable the majority of the grains to settle 

 with their flat faces on the surface of the glass. Under 

 these circumstances they, and especially the larger grains, 

 form numerous clumps of from 2 up to 33 grains, as will be 



* Discussed in a paper just sent to Phot. Journ. by Trivelli, Righter, 

 and Sheppard. [This paper has since been published in Phot. Journal 

 for September 1922, p. 407.] 



