862 



NA TURE 



I ) 



a cli i ...-,, ;. ,-, as 

 ntimetre cube of the 

 ulicn working with 

 11,'ures given by 

 lys ottf cram of 



Conductivities of Aqueous Salt Solutions. 



In the I Olll-.'- n( Ug 



the liiHl Iwf) years is and 



conductivities of ci us. we 



have come across a miujik- i. i.m.n -• to 



us of interest anil imiwrtanir 



•|l,,. .,...,,(., conductiv '•' 

 usu.ii 1(1. refer 



solulu... '.^ :. toiind it d 

 concentrated solutions, to <.< 

 volumes of electrolyte tontai 



I, ' ■ the solution in (|ucsliiin < ont.nii a ^iMins of 



s.i loiisand f^r.un- oi w.itcr, ,niil i> of (lcnsit\- d, 



then til"' \olnni(' of solntmn ront.iinniL; one i;i.ini of 



waI'— ■ , . MullipKini; tlii> term |)\- ^, we 



I ( )( X Hi ' ■ ' 



(il- ni.iunitndr wlii'li \m- \\i!! denote by «', 



will. ! 'i.hn 1 1\ I' ' ' Tillies one cm. 



a|).in iiiiiiit <<'. ■( int.iuiiiig one 



gram ol w.iii :. At ini^h > < 'Ik. entrations, the \iscosity 

 of tli<' Miliition is .1 tai tor w liich cannot lie nejjlected. 

 Makn; iilol assumption, /.(. lliat conductivity 



and i: jiroportional, uc arn\-e at a corrected 



COndlii i ! \ ; i \ , 



iooo+;r 



». corr . = J- K1J . 



looorf 



Multiph- by looo, and we have the coiulut ti\ ity 

 nndir tlie same conditions due to an amount of solu- 

 tion ( (intainm^ lOoo grams of water. If now this be 

 plotted a>4amst v M or the ucii^ht molar concentra- 

 tion (inols iH-r thoiis.md grains of water), the result is 

 a cuiAc wliu h, m the cases of potassium, sodium, and 

 litliium ( hloiides, becomes nearly linear after x/M has 

 exceeded o-5-i, and remains so up to the limit to which 

 we have so far carried our measurements {x/M = 3 to 5, 

 depending on the elcctrnlvtc). Wo have plotted the 

 conductivity, densit\ . ami \is(()sit\ data of other 

 observers for certain otiier solutions, and have 

 obtained similar results. 



Expressed verbally — if a thousand grams of water 

 are put between electrodes one cm. 

 apart and one of the salts in question 

 gradually added, the increase in con- 

 ductance of the cell (corrected for 

 viscosity change) brought about by 

 dissolving, say, an extra one-tenth 

 gram equivalent of salt is nearly 

 independent of the concentration 

 of the solution when a certain limit 

 of concentration has been exceeded. 

 The slope of the curve is given by t 



s 

 {1000 +x)Kri 2 



A curve of the same slope is got if | 

 the molar fraction of the salt in the i 

 solution is plotted against the con- 5 

 ductance (corrected for viscosity I 

 and measured between electrodes 

 one cm. apart) of a mol of solu- 

 tion : that is, within the limits 

 mentioned, the corrected conduct- 

 ance of a solution is practically 

 proportional to the ratio salt mole- 

 cules/total molecules. 



This relation, obtained by considering, not, as is 

 customary, the conductance of a fixed weight of salt 

 to which increasing amounts of water are added, but 

 the conductance of a fixed weight of water to which 

 increasing amounts of salt are added, would appear 



NO. 2824, VOL. 112] 



to have considerable implications in 

 for the theon,' of stronjr clectrolyu-. 

 be considered, ami tin- ciata more full 



baci 



not 



any 



t»"- 



this preliminary note as a result 



1 paper published some months 



Elcktroch. 29, 163, 1923) and 



I, as its title did not indicate 

 '1 this paper 



{i.e. the 



tllC WCJgi 



aqueous i 



\'er\- high contciilratic): 



so far measured). He In 



cent, lithium chloride and j 



respectively, after which th- 



tial difference between his meinou 01 j)io( 



is that we take into our conductance c,\ 



density of the solution. If his curve - 



this way, the changes in direction at t h 



tions become far less marked, and thei 



moderate concentrations, whilst bet ■ 



less linear. < losel\- :iiinT'ovini;tte ill t\ • 



by us for 



I\im;'> ( oile!-e, ^ii,:!id W .< .2.. 

 .\o\c'nil)er z^. 



itv acain'^t 



irs 

 : he 



in 

 ra- 



>re 



.at 



The London Fogs of November 2,=.- 27. 1923. 



The accompanying grapli s the hourly 



variation in the quantity o! purit>-, that 



is, sooty matter, in "' i i.ondon ' \Sestminster), 



which was chieflN hie for the recent smoke 



fogs. '"'" - s V dearly that the maximum 



dens; ■_; occurred ab<:)ut mid-day, and 



there ...... .. ..iw... ...V icase in density which coi"- i''-' 



I 



T.Mcf 



Fig. I. — Suspended impurity in the air > 



witii the period of lighting up of tiies iu the morning. 

 The tisjures from which the curve is plotted were 

 obtained from mv au: oorder, designed for 



the .\d\ i.-^ory Commit; ,,:uospheric Pollution. 



When a large number of days are averaged^nd 



