STRUCTURAL FEATURES OF SELENIUM 505 



opposite lower part of deposit on outside tube, it extends farther 

 down than the latter (less than ^^ cm.) and not nearly so far up. 

 Deposit on outside tube is [by this time] some 8 or 9 cms. long 

 [much longer than usual]."® Still another note, referring to the 

 tenth run, reads : "Zone 1 scrapes off harder than when deposit 

 is of character 'a,' and in the transparent region of zone 2 the de- 

 posit has appreciable thickness and resists the knife. Zone 3 [which 

 in this case appears to have been on the condensation tube itself] 

 is fairly below the [previous] deposit on the walls of the outside 

 tube. Zone 1 has nodules, on end and sides. "^ 



Examination of the material in the two tables shows no marked 

 effect of time on the character of the deposit, except of course to 

 deepen it. There seems to be some increase in the total length with 

 rising temperature. The two characters of the deposit in the E 

 series appear to be unrelated to both time and temperature, and it 

 is noticeable that the total length of the deposit does not depend 

 markedly upon which one of the two characters is present. Also, 

 the apparent transfer of zone 3 from the inside to the outside tube, 

 observed in a few cases, is unexplained, although it was noted to 

 occur, particularly in run E8, when the second character, b, was 

 present. Whether it occurred only when this character was present 

 is unknown. 



Since the vacuum for by far the greater part of the time of each 

 run was in all cases what may be termed "good," that is, a vacuum 

 high enough to insure molecular flow, it is to be supposed that any 

 variations of the vacuum over the small range of low pressures 

 present had no effect on the nature of the deposit. Whether the 

 relatively poor vacuum over a short part of each run before pressure 

 conditions permitting molecular flow were established, had anything 

 to do with determining which of the two general characters should 

 prevail in a run of the E series, is undetermined. The cause of 

 these two general characters does not appear to be cleared up by 

 the present data on the basis either of varied temperature, of chang- 

 ing time of run, or of poor vacuum. They remain for the time 

 being unexplained.^ 



«The point corresponding to this run has been found to he a little low on 

 the vapor pressure curve, perhaps because the material condensed on the 

 outside was not weighed. Condensation on the outside tube occurred in 

 some of the runs, but the amount so deposited was usually small. 



■Considerable light is thrown on this nodule structure appearing on the 

 outside surface of zone 1, by a study of the special thick deposit, discussion 

 of which has been omitted from the present paper, as already stated. 



